WO2010129941A1 - Correlation of hpv e6 and e7 expression with progression of cervical disease - Google Patents

Abstract

Compositions and methods for diagnosing cervical disease are provided. The methods include measuring the expression level of a high-risk HPV E6 and/or E7 RNA in a body sample from a subject, wherein the level of HPV E6 and/or E7 mRNA positively correlates with cervical disease progression. Threshold expression levels of HPV E6 and E7 mRNA are further provided that distinguish normal cervical samples from diseased cells or tissue and samples that are indicative of low-grade cervical disease from those that are indicative of high-grade disease. Further, genes with expression levels that are relatively invariant between normal and various cervical disease states have been identified. Such genes can serve as suitable endogenous controls for the normalization of target gene expression levels in cervical samples. Kits comprising reagents for practicing the methods of the invention are further provided.

Description

CORRELATION OF HPV E6 AND E7 EXPRESSION WITH PROGRESSION OF

CERVICAL DISEASE

REFERENCE TO A SEQUENCE LISTING SUBMITTED AS A TEXT FILE VIA EFS-WEB

The official copy of the sequence listing is submitted concurrently with the specification as a text file via EFS-Web, in compliance with the American Standard Code for Information Interchange (ASCII), with a file name of 388607SEQLIST.txt, a creation date of May 10, 2010, and a size of 21 KB. The sequence listing filed via EFS-Web is part of the specification and is hereby incorporated in its entirety by reference herein.

FIELD OF THE INVENTION

The present invention relates to methods for detecting cervical disease.

BACKGROUND OF THE INVENTION

Carcinoma of the cervix is the second most common neoplasm in women, accounting for approximately 12% of all female cancers and causing approximately

250,000 deaths per year (Baldwin et al. (2003) Nature Reviews Cancer 3:1-10). In many developing countries where mass screening programs are not available, the clinical problem is more serious. Cervical cancer in these countries is the number one cause of cancer deaths in women. The majority of cases of cervical cancer represent squamous cell carcinoma, although adenocarcinoma is also seen. Cervical cancer can be prevented by population screening as it evolves through well-defined noninvasive intraepithelial stages, which can be distinguished morphologically (Williams et al. (1998) Proc. Natl. Acad. Sci. USA 95 : 14932-14937). While it is not understood how normal cells become transformed, the concept of a continuous spectrum of histopatho logical change from normal, stratified epithelium through cervical intraepithelial neoplasia (CIN) to invasive cancer has been widely accepted for years. The precursor to cervical cancer is dysplasia, also known in the art as CIN or squamous intraepithelial lesions (SIL). Squamous intraepithelial abnormalities may be classified by using the three-tiered (CIN) or two-tiered (Bethesda) system. Under the Bethesda system, low-grade squamous intraepithelial lesions (LSIL), corresponding to CINI and HPV infection, generally represent productive HPV infections with a relatively low risk of progression to invasive disease. High-grade squamous intraepithelial lesions (HSIL), corresponding to CINII and CINIII in the three-tiered system, show a higher risk of progression to cervical cancer than do LSIL, although both LSIL and HSIL are viewed as potential precursors of malignancy. Patient samples may also be classified as ASCUS (atypical squamous cells of unknown significance) or AGUS (atypical glandular cells of unknown significance) under this system.

A strong association of cervical cancer and infection by high-risk types of human papilloma virus (HPV), such as types 16, 18, and 31, has been established. In fact, a large body of epidemiological and molecular biological evidence has established HPV infection as a causative factor in cervical cancer. Moreover, HPV is found in 85% or more of the cases of high-grade cervical disease. However, HPV infection is very common, possibly occurring in 5-15% of women over the age of 30, but few HPV-positive women will ever develop high-grade cervical disease or cancer. The presence of HPV alone is indicative only of infection, not of high-grade cervical disease, and, therefore, testing for HPV infection alone results in many false positives. See, for example, Wright et al. (2004) Obstet. Gynecol. 103:304-309.

Current literature suggests that HPV infects the basal stem cells within the underlying tissue of the uterine-cervix. Differentiation of the stem cells into mature keratinocytes, with resulting migration of the cells to the stratified cervical epithelium, is associated with HPV viral replication and re-infection of cells. During this viral replication process, a number of cellular changes occur that include cell-cycle deregulation, active proliferation, DNA replication, transcriptional activation and genomic instability (Crum (2000) Modern Pathology 13:243-251; Middleton et al. (2003) J. Virol. 77:10186-10201; Pett et al. (2004) Cancer Res. 64:1359-1368).

Most HPV infections are transient in nature, with the viral infection resolving itself within a 12-month period. For those individuals who develop persistent infections with one or more oncogenic subtypes of HPV, there is a risk for the development of neoplasia in comparison to patients without an HPV infection. Given the importance of HPV in the development of cervical neoplasia, the clinical detection of HPV has become an important diagnostic tool in the identification of patients at risk for cervical neoplasia development. The clinical utility of HPV-based screening for cervical disease is in its negative predictive value. An HPV negative result in combination with a history of normal Pap smears is an excellent indicator of a disease-free condition and a low risk of cervical neoplasia development during the subsequent 1-3 years. However, a positive HPV result is not diagnostic of cervical disease; rather it is an indication of infection. Although the majority of HPV infections is transient and will spontaneously clear within a 12-month period, a persistent infection with a high-risk HPV viral subtype indicates a higher risk for the development of cervical neoplasia.

Cyto logical examination of Papanicolaou-stained cervical smears (Pap smears) currently is the method of choice for detecting cervical cancer. The Pap test is a subjective method that has remained substantially unchanged for 60 years. There are several concerns, however, regarding its performance. The reported sensitivity of a single Pap test (the proportion of disease positives that are test-positive) is low and shows wide variation (30-87%). The specificity of a single Pap test (the proportion of disease negatives that are test-negative) might be as low as 86% in a screening population and considerably lower in the ASCUS PLUS population for the determination of underlying high-grade disease. See, Baldwin et al., supra. A significant percentage of Pap smears characterized as LSIL or CINI are actually positive for high-grade lesions. Furthermore, up to 10% of Pap smears are classified as ASCUS (atypical squamous cells of undetermined significance), i.e., it is not possible to make a clear categorization as normal, moderate or severe lesion, or tumor. However, experience shows that up to 10% of this ASCUS population has high- grade lesions, which are consequently overlooked. See, for example, Manos et al. (1999) JAMA 281 :1605-1610.

Infection with high-risk human papillomavirus (hrHPV) is known to be directly associated with the development of cervical cancer (Bosch et al., 2002; Schiffman et al., 1993; Walboomers et al., 1999). The most prevalent hrHPV types for the progression to cervical cancer are HPV 16, 18, 31, 33, and 45 which account for 80% of the infected population (Kraus et al., 2006; Munoz et al., 2003). Historically, DNA testing has been the predominant method used to detect HPV infection. There is now significant evidence that testing for oncogenic HPV DNA is more sensitive and has a higher negative predictive value for the detection of high grade, cervical intraepithelial neoplasia (CIN2+) as compared to cervical cytology (Clavel et al., 2001; Manos et al., 1999; Molden et al., 2005a; Molden et al., 2005b). This increased sensitivity, however, is accompanied by a decrease in specificity and positive predictive value as many HPV infections are transient and will not result in significant cervical disease. Thus, a method for diagnosing cervical disease, particularly high-grade disease, that is independent of or works in conjunction with conventional Pap smears and molecular testing for high-risk HPV infection is needed. Such a method should be able to specifically identify high-grade cervical disease that is present in all patient populations, including those cases classified as LSIL or CINI by Pap staining that are actually positive for high-grade lesions (i.e., "false negatives"). Therefore, there is a need in the art for specific, reliable diagnostic methods that are capable of detecting high-grade cervical disease and of differentiating high-grade disease from early-stage HPV infection and mild dysplasia.

SUMMARY OF THE INVENTION

Compositions and methods are provided for diagnosing cervical disease. The methods include measuring the expression level of a high-risk HPV E6 and/or E7 RNA in a body sample from a subject, wherein the level of HPV E6 and/or E7 mRNA positively correlates with cervical disease progression. Methods for reliable and sensitive quantitation of HPV E6 and E7 RNA levels in body samples, such as cervical cytology and biopsy samples, are provided. Threshold expression levels of HPV E6 and E7 mRNA are presented herein that distinguish normal cervical samples from diseased cells or tissue and samples that are indicative of low-grade cervical disease from those that are indicative of high-grade disease. Further, genes with expression levels that are relatively invariant between normal and various cervical disease states have been identified. Such genes can serve as suitable endogenous controls for the normalization of target gene expression levels in cervical samples. Kits comprising reagents for practicing the methods of the invention are further provided. The methods of the invention can be used alone or in combination with traditional gynecological diagnostic techniques that analyze morphological characteristics or HPV infection status. In this manner, measurement of the expression level of HPV E6 and E7 RNA can reduce the high false-negative rate of the Pap test and may facilitate mass automated screening. BRIEF DESCRIPTION OF THE DRAWINGS

Figures IA and IB provide real-time RT-PCR data comparing threshold cycle values (C_T) of various housekeeping genes versus specimen type in cervical biopsy samples. ACTB: beta actin; TBP: TATA box binding protein; GAPDH: glyceraldehyde- 3-phosphate dehydrogenase; 18S: 18S ribosomal RNA; PPIA: peptidylprolyl isomerase A (cylophilin A); YWHAZ: tyrosine 3-monooxygenase/tryptophan 5-monooxygenase activation protein, zeta polypeptide; GUSB: beta-glucuronidase; HMBS: hydroxymethylbilane synthase; RPLPO: large ribosomal protein, PO; IPO8: importin 8; PGKl : phosphoglycerate kinase 1; POLR2A: polymerase (RNA) II (DNA directed) polypeptide A, 220 kDa; TFRC: transferrin receptor; UBC: ubiquitin C; B2M: beta-2- microglobulin; HPRTl : hypoxanthine phosphoribosyltransferase 1; WNL: within normal limits; CIN: cervical intraepithelial neoplasia; SCC: squamous cell carcinoma.

Figures 2A and 2B provide real-time RT-PCR data comparing C_T values of various housekeeping genes versus specimen type in cervical cytology samples. NILM: negative for intraepithelial lesion and malignancy; LSIL: low-grade squamous intraepithelial lesions; HSIL: high-grade squamous intraepithelial lesions.

Figures 3A-3C provide real-time RT-PCR data comparing C_T values of HPV E6, HPV E7, and the highest of HPV E6 or E7 mRNA levels, respectively, encoded by high- risk HPV types in cervical biopsy specimens. The C_T values of oncogenic mRNA molecules encoded by 15 high risk HPV types were determined by real-time RT-PCR using RNA isolated from each of 96 cervical biopsy specimens. Biopsy diagnoses are denoted on the x-axis and the C_T value of oncogene mRNA are indicated on the y-axis. C_T values that had corresponding standard deviations of greater than 1 were removed from the analysis. Open circles represent biopsy specimens containing high risk HPV DNA and expressing the cognate HPV mRNA. Closed circles denote biopsy specimens that do not contain detectable high risk HPV DNA. Three specimens were not included in this analysis due to insufficient RNA quality.

Figures 4A-4C provide real-time RT-PCR data comparing absolute copy number values of HPV E6, HPV E7, and the highest of HPV E6 or E7 mRNA levels, respectively, encoded by high-risk HPV types in cervical biopsy specimens. The absolute quantities of oncogenic mRNA molecules encoded by high risk HPV types were determined by realtime RT-PCR using RNA isolated from each of 96 cervical biopsy specimens. Biopsy diagnoses are denoted on the x-axis and the number of copies of oncogene mRNA are indicated on the y-axis. Specimens with 0 copies of HPV mRNA were assigned an absolute copy number of 0.01 in order to facilitate graphing on a loglO scale. Open circles represent cytology specimens containing high risk HPV DNA and expressing the cognate HPV mRNA. Closed circles denote cytology specimens that do not contain detectable high risk HPV DNA. Three specimens were not included in this analysis due to insufficient RNA quality.

Figures 5A-5C provide real-time RT-PCR data comparing normalized target values of HPV E6, HPV E7, and the highest of HPV E6 or E7 mRNA levels, respectively, encoded by high-risk HPV types in cervical biopsy specimens that have been normalized to GUSB. The normalized target value (NTV) values of oncogenic mRNA molecules encoded by high risk HPV types were determined by real-time RT-PCR using RNA isolated from each of 96 cervical biopsy specimens. Biopsy diagnoses are denoted on the x-axis and the NTV of oncogene mRNA are indicated on the y-axis. The NTV is generated by a ratio of copy number for the HPV transcript to the copy number of the endogenous control transcript multiplied by 1000 to facilitate graphing. Open circles represent biopsy specimens containing high risk HPV DNA and expressing the cognate HPV mRNA. Closed circles denote cytology specimens that do not contain detectable high risk HPV DNA.. Three specimens were not included in this analysis due to insufficient RNA quality.

Figures 6A-6C provide real-time RT-PCR data comparing target values, normalized to POLR2A of HPV E6, HPV E7, and the highest of HPV E6 or E7 mRNA levels, respectively, encoded by high-risk HPV types in cervical biopsy specimens. The normalized target values (NTV) of oncogenic mRNA molecules encoded by high risk HPV types were determined by real-time RT-PCR using RNA isolated from each of 96 cervical biopsy specimens. Biopsy diagnoses are denoted on the x-axis and the NTV of oncogene mRNA are indicated on the y-axis. The NTV is generated by a ratio of copy number for HPV transcripts to the copy number of the endogenous control transcript multiplied by 1000 to facilitate graphing. Open circles represent biopsy specimens containing high risk HPV DNA and expressing the cognate HPV mRNA. Closed circles denote cytology specimens that do not contain detectable high risk HPV DNA. Three specimens were not included in this analysis due to insufficient RNA quality.

Figures 7A-7C provide real-time RT-PCR data comparing comparative CT values of HPV E6, HPV E7, and the highest of HPV E6 or E7 mRNA levels, respectively, encoded by high-risk HPV types in cervical biopsy specimens that have been normalized to GUSB. The comparative values of oncogenic mRNA molecules encoded by high risk HPV types were determined by real-time RT-PCR using RNA isolated from each of 96 cervical biopsy specimens. Biopsy diagnoses are denoted on the x-axis and the fold expression of oncogene mRNA are indicated on the y-axis. Open circles represent biopsy specimens containing high risk HPV DNA and expressing the cognate HPV mRNA. Closed circles denote cytology specimens that do not contain detectable high risk HPV DNA. Three specimens were not included in this analysis due to insufficient RNA quality. Figures 8A-8C provide real-time RT-PCR data comparing comparative CT values of HPV E6, HPV E7, and the highest of HPV E6 or E7 mRNA levels, respectively, encoded by high-risk HPV types in cervical biopsy specimens that have been normalized to POLR2A. The comparative values of oncogenic mRNA molecules encoded by high risk HPV types were determined by real-time RT-PCR using RNA isolated from each of 96 cervical biopsy specimens. Biopsy diagnoses are denoted on the x-axis and the fold expression of oncogene mRNA are indicated on the y-axis. Open circles represent biopsy specimens containing high risk HPV DNA and expressing the cognate HPV mRNA. Closed circles denote cytology specimens that do not contain detectable high risk HPV DNA.. Three specimens were not included in this analysis due to insufficient RNA quality.

Figures 9A and 9B provide real-time RT-PCR data comparing C_T values of HPV E6, and the highest of HPV E6 or E7 mRNA levels, respectively, encoded by high-risk HPV types in cervical cytology specimens. HPV E6/E7 mRNA molecules were amplified by real-time RT-PCR using RNA isolated from each of 96 cervical cytology specimens. Cytological diagnoses are denoted on the x-axis and the raw CT values derived from amplification reactions are indicated on the y-axis. CT value is inversely proportional to mRNA level, with higher mRNA levels resulting in lower CT values. The absence of any detectable mRNA is indicated by a CT value of 40. Open circles represent cytology specimens containing high risk HPV DNA and producing CT values <40 for the cognate HPVE6 (_^4) or HPV E6/E7 mRNA (B). Closed circles denote cytology specimens that do not contain detectable high risk HPV DNA. Open squares represent specimens that contain high risk HPV DNA but generate CT values equal to 40 for HPV E6 (A) or HPV E6/E7 mRNA (B). Cross marks represent specimens in which the lowest viral mRNA CT value did not correspond to the specimen genotype. Six specimens were not included in this analysis due to insufficient RNA quality. Samples with biopsy information are designated as follows; crosses represent CINl, diamonds indicate CIN2, closed squares are CIN3 and inverted triangles indicate SCC. Figure 10 provides real-time RT-PCR data comparing C_T values of HPV E7 encoded by HPV 16 in cervical cytology specimens. HPV 16 E7 mRNA molecules were amplified by real-time RT-PCR using RNA isolated from HPV 16 DNA-positive cervical cytology specimens and cytology specimens negative for any detectable high-risk HPV DNA. Cytological diagnoses are denoted on the x-axis and the raw CT values derived from amplification reactions are indicated on the y-axis. CT value is inversely proportional to mRNA level, with higher mRNA levels resulting in lower CT values. The absence of any detectable mRNA is indicated by a CT value of 40. Open circles represent cytology specimens containing HPV 16 viral DNA and producing CT values below 40 for HPV 16 E7 mRNA. Closed circles denote cytology specimens that do not contain detectable high risk HPV DNA. Open squares represent specimens that contain HPV 16 DNA but generate CT values equal to 40 for HPV 16 E7 mRNA. Samples with biopsy information are designated as follows; crosses represent CINl, diamonds indicate CIN2, closed squares are CIN3 and inverted triangles indicate SCC. Figures 1 IA and 1 IB provide real-time RT-PCR data comparing absolute copy number values of HPV E6, and the highest of HPV E6 or E7 mRNA levels, respectively, encoded by high-risk HPV types in cervical cytology specimens. The absolute quantities of HPV E6 mRNA molecules encoded by high risk HPV types were determined by realtime RT-PCR using RNA isolated from each of 96 cervical cytology specimens. Absolute quantities of HPV E7 mRNA were determined for HPV Type 16 only. Cytological diagnoses are denoted on the x-axis and the number of copies of HPV E6 mRNA (_^4) or HPV E6/E7 mRNA (B) are indicated on the y-axis. Specimens with 0 copies of HPV mRNA were assigned an absolute copy number of 0.01 in order to facilitate graphing on a loglO scale. Open circles represent cytology specimens containing high risk HPV DNA and the corresponding HPV E6 (A) or HPV E6/E7 (B) mRNA. Closed circles denote cytology specimens that do not contain detectable high risk HPV DNA. Open squares represent specimens that contain high risk HPV DNA but not quantifiable HPV E6 (_^4) or HPV E6/E7 (B) mRNA. Six specimens were not included in this analysis due to insufficient RNA quality. Samples with biopsy information are designated as follows; crosses represent CINl, diamonds indicate CIN2, closed squares are CIN3 and inverted triangles indicate SCC.

Figure 12 provides real-time RT-PCR data comparing absolute copy number values of HPV E7 encoded by HPV 16 in cervical cytology specimens. The absolute quantity of HPV 16 E7 mRNA molecules was determined by real-time RT-PCR using RNA isolated from HPV 16 DNA-positive cervical cytology specimens and cytology specimens negative for any detectable high-risk HPV DNA. Cytological diagnoses are denoted on the x-axis and the number of copies of HPV 16 E7 mRNA is indicated on the y- axis. Specimens with 0 copies of HPV 16 E7 mRNA were assigned an absolute copy number of 0.01 in order to facilitate graphing on a loglO scale. Open circles represent cytology specimens containing HPV 16 viral DNA and expressing HPV 16 E7 mRNA. Open squares represent specimens that contain HPV 16 DNA without detectable HPV 16 E7 mRNA. Closed circles indicate high-risk HPV DNA-negative specimens. Samples with biopsy information are designated as follows; crosses represent CINl, diamonds indicate CIN2, closed squares are CIN3 and inverted triangles indicate SCC.

Figure 13 provides real-time RT-PCR data comparing normalized target values of HPV E6 encoded by high-risk HPV types in cervical cytology specimens. Expression levels of HPV E6 transcripts normalized to the GUSB or POLR2A endogenous controls were determined by real-time RT-PCR using RNA isolated from each of 96 cervical cytology specimens. Cytological diagnoses are denoted on the x-axis and the ratio of target transcript copies over endogenous control transcript copies (GUSB or POLR2A) are denoted on the y axis. These values were multiplied by 1000 to facilitate graphing. Specimens exhibiting no copies of HPV mRNA were assigned a fold-expression value of 0.01 in order to facilitate graphing on a loglO scale. Open circles represent cytology specimens containing HPV 16 viral DNA and expressing the corresponding E6 mRNA.

Open squares indicate specimens that contain HR HPV DNA but which are not expressing mRNA. Closed circles denote cytology specimens that do not contain detectable HR HPV DNA. Six specimens were not included in this analysis due to inadequate RNA quality. Samples with biopsy information are coded as follows; crosses are CINl, diamonds are CIN2, filled squares are CIN3 and triangles are SCC.

Figure 14 provides real-time RT-PCR data comparing normalized target values of the highest of HPV E6 or E7 encoded by high-risk HPV types in cervical cytology specimens. Expression levels of HPV E6 and E7 transcripts normalized to GUSB and POLR2A were determined by real-time RT-PCR using RNA isolated from each of 96 cervical cytology specimens. Cytological diagnoses are denoted on the x-axis and the ratio of target transcript copies over endogenous control transcript copies are denoted on the y axis. These values were multiplied by 1000 to facilitate graphing. Specimens exhibiting no expression of HR HPV mRNA were assigned a fold-expression value of 0.01 to facilitate graphing on a log 10 scale. Open circles represent cytology specimens containing HR HPV viral DNA and expressing the corresponding HPV E6 or E7 mRNA. Open squares indicate specimens containing HR HPV DNA, but no detectable RNA. Closed circles denote cytology specimens that do not contain detectable HR HPV DNA. Six specimens were not included in this analysis due to inadequate RNA quality. Samples with biopsy information are coded as follows; crosses are CINl, diamonds are CIN2, filled squares are CIN3 and triangles are SCC.

Figure 15 provides real-time RT-PCR data comparing normalized target values of HPV E 7 encoded by HPV 16 in cervical cytology specimens. Expression levels of HPV 16 E7 transcripts normalized to the GUSB endogenous control were determined by real-time RT-PCR using RNA isolated from each of 65 cervical cytology specimens. Cytological diagnoses are denoted on the x-axis and the ratio of target transcript copies over endogenous control transcript copies (GUSB or POLR2A) are denoted on the y axis. These values were multiplied by 1000 to facilitate graphing. Specimens exhibiting no expression of HPV mRNA were assigned a fold-expression value of 0.01 in order to facilitate graphing on a log 10 scale. Open circles represent cytology specimens containing HPV 16 viral DNA and expressing HPV 16 E7 mRNA. Closed circles denote cytology specimens that do not contain detectable HR HPV DNA. Samples with a dominant virus other than HPV 16 were excluded from this analysis as were samples with inadequate RNA quality. Samples with biopsy information are coded as follows; crosses are CINl, diamonds are CIN2, filled squares are CIN3 and triangles are SCC.

Figure 16 provides real-time RT-PCR data comparing the relative standard curve- derived values of HPV E6 encoded by high-risk HPV types in cervical cytology specimens. The relative levels of HPV E6 transcripts were determined by real-time RT- PCR using RNA isolated from each of 96 cervical cytology specimens. Cytological diagnoses are denoted on the x-axis and over-expression relative to the average of the NILM specimens is denoted on the y axis. The relative standard curve is derived from a ratio of the normalized target value of the HPV transcript to the normalized target value of the average of the normals. Samples were normalized to either GUSB or POLR2A. Specimens exhibiting no expression of HPV mRNA were assigned a value of 0.01 in order to facilitate graphing on a loglO scale. Open circles represent cytology specimens containing HR HPV viral DNA and the respective E6 mRNA. Open squares indicate specimens that containing HR HPV DNA, but with no detectable RNA. Closed circles denote cytology specimens that do not contain detectable HR HPV DNA. Six specimens were not included in this analysis due to inadequate RNA quality. Samples with biopsy information are coded as follows; crosses are CINl, diamonds are CIN2, filled squares are CIN3 and triangles are SCC.

Figure 17 provides real-time RT-PCR data comparing the relative-standard curve- derived values of the highest of HPV E6 or E7 encoded by high-risk HPV types in cervical cytology specimens. The relative mRNA levels of HPV E6 and E7 transcripts were determined by real-time RT-PCR using RNA isolated from each of 96 cervical cytology specimens. Cyto logical diagnoses are denoted on the x-axis and over-expression relative to the average of the NILM specimens is denoted on the y axis. Samples were normalized to GUSB or POLR2A. Specimens exhibiting no expression of HR HPV mRNA were assigned a value of 0.01 to facilitate graphing on a loglO scale. Open circles represent cytology specimens containing HR HPV viral DNA and the corresponding HR HPV E6 or E7 mRNA. Open squares indicate specimens that contain HR HPV DNA, but with no detectable RNA. Closed circles denote cytology specimens that do not contain detectable HR HPV DNA. Samples with inadequate RNA quality were excluded from this analysis. Samples with biopsy information are coded as follows; crosses are CINl, diamonds are CIN2, squares are CIN3 and triangles are SCC.

Figure 18 provides real-time RT-PCR data comparing the relative standard curve- derived values of HPV E7 encoded by HPV 16 in cervical cytology specimens. The relative levels of HPV 16 E7 transcripts were determined by real-time RT-PCR using RNA isolated from a subset of cervical cytology specimens. Cytological diagnoses are denoted on the x-axis and over-expression relative to the average of the NILM specimens is denoted on the y axis. Samples were normalized to either GUSB or POLR2A. Specimens exhibiting no expression of HPV mRNA were assigned a value of 0.01 in order to facilitate graphing on a log 10 scale. Open circles represent cytology specimens containing HPV 16 viral DNA and expressing HPV 16 E7 mRNA. Closed circles denote cytology specimens that do not contain detectable HR HPV DNA. Samples with a dominant virus other than HPV 16 were excluded from this analysis as were samples with inadequate RNA quality. Biopsy information for these cytology specimens is coded as follows; crosses are CINl, diamonds are CIN2, filled squares are CIN3 and triangles are SCC.

Figures 19A and 19B provide real-time RT-PCR data comparing the comparative C_T values of HPV E6 or the highest of HPV E6 or E7 encoded by high-risk HPV types in cervical cytology specimens normalized to GUSB. Relative HPV E6 and E7 mRNA expression levels from high risk HPV types were determined by real-time RT-PCR using RNA isolated from each of 96 cervical cytology specimens. Cytological diagnoses are denoted on the x-axis and the fold over-expression compared to the average expression in the NILM specimens is indicated on the y-axis. Specimens exhibiting less than 2-fold over-expression were assigned a fold expression value of 0.01 in order to facilitate graphing on a log 10 scale. Expression of viral transcripts was normalized to the expression of the endogenous control GUSB for each specimen. Open circles represent cytology specimens containing high risk HPV DNA and the cognate HPV E6 (A) or HPV E6/E7 (B) mRNA Closed circles denote cytology specimens that do not contain detectable high risk HPV DNA. Open squares represent specimens that contain high risk HPV DNA but not HPV E6 (_^4) or HPV E6/E7 mRNA (B). Six specimens were not included in this analysis due to insufficient RNA quality. Samples with biopsy information are designated as follows; crosses represent CINl, diamonds indicate CIN2, closed squares are CIN3 and inverted triangles indicate SCC.

Figure 20 provides real-time RT-PCR data comparing the comparative C_T values of HPV E7 encoded by HPV16 in cervical cytology specimens, normalized to GUSB. The relative levels of HPV 16 E7 mRNA molecules were determined by real-time RT-PCR using RNA isolated from HPV 16 DNA-positive and HPV DNA-negative cytology specimens. Cytological diagnoses are denoted on the x-axis and the fold over-expression of HPV 16 E7 mRNA relative to the average expression in the NILM samples is indicated on the y-axis. Specimens exhibiting less than 2-fold over-expression of HPV mRNA were assigned a fold-expression value of 0.01 in order to facilitate graphing on a loglO scale. Open circles represent cytology specimens containing HPV 16 viral DNA and quantifiable HPV 16 E7 mRNA. Closed circles denote cytology specimens that do not contain detectable high risk HPV DNA. Open squares represent specimens that contain HPV 16 DNA but no detectable HPV 16 E7 mRNA. Samples with biopsy information are designated as follows; crosses represent CINl, diamonds indicate CIN2, closed squares are CIN3 and inverted triangles indicate SCC.

Figures 21 A and 2 IB provide real-time RT-PCR data comparing the comparative C_x values of HPV E6 or the highest of HPV E6 or E7 encoded by high-risk HPV types in cervical cytology specimens normalized to POLR2A. Relative HPV E6 and E7 mRNA expression levels from high risk HPV types were determined by real-time RT-PCR using RNA isolated from each of 96 cervical cytology specimens. Cytological diagnoses are denoted on the x-axis and the fold over-expression compared to the average expression in the NILM specimens is indicated on the y-axis. Specimens exhibiting less than 2-fold over-expression were assigned a fold expression value of 0.01 in order to facilitate graphing on a log 10 scale. Expression of viral transcripts was normalized to the expression of the endogenous control POLR2A for each specimen. Open circles represent cytology specimens containing high risk HPV DNA and the cognate HPV E6 (A) or HPV E6/E7 (B) mRNA Closed circles denote cytology specimens that do not contain detectable high risk HPV DNA. Open squares represent specimens that contain high risk HPV DNA but not HPV E6 (_^4) or HPV E6/E7 mRNA (B). Six specimens were not included in this analysis due to insufficient RNA quality. Samples with biopsy information are designated as follows; crosses represent CINl, diamonds indicate CIN2, closed squares are CIN3 and inverted triangles indicate SCC.

Figure 22 provides real-time RT-PCR data comparing the comparative C_T values of HPV E7 encoded by HPV 16 in cervical cytology specimens, normalized to POLR2A. The relative levels of HPVl 6 E7 mRNA molecules were determined by real-time RT-PCR using RNA isolated from HPV 16 DNA-positive and HPV DNA-negative cytology specimens. Cytological diagnoses are denoted on the x-axis and the fold over-expression of HPV 16 E7 mRNA relative to the average expression in the NILM samples is indicated on the y-axis. Specimens exhibiting less than 2-fold over-expression of HPV mRNA were assigned a fold-expression value of 0.01 in order to facilitate graphing on a loglO scale. Open circles represent cytology specimens containing HPV 16 viral DNA and quantifiable HPV 16 E7 mRNA. Closed circles denote cytology specimens that do not contain detectable high risk HPV DNA. Open squares represent specimens that contain HPV 16 DNA but no detectable HPV 16 E7 mRNA. Samples with biopsy information are designated as follows; crosses represent CINl, diamonds indicate CIN2, closed squares are CIN3 and inverted triangles indicate SCC.

DETAILED DESCRIPTION OF THE INVENTION

The present invention provides compositions and methods for diagnosing cervical disease and distinguishing patients with high-grade disease or those that are likely to develop high-grade disease from those with low-grade disease that will most likely not progress.

By "cervical disease" is intended any form of dysplastic (mild, moderate, severe) or neoplastic disease of the cervix, including cervical intraepithelial neoplasia I (CINI), CINII, CINIII, low-grade squamous intraepithelial lesions (LSIL), high-grade squamous intraepithelial lesions (HSIL), squamous cell carcinoma (SCC), carcinoma in situ, and adenocarcinoma.

Dysplasia is conventionally defined in morphological terms by a loss of normal orientation of epithelial cells, accompanied by alterations in cellular and nuclear size, shape, and staining characteristics. Dysplasia is graded according to the degree of the cellular abnormalities (i.e., mild, moderate, severe) and is widely accepted to be an intermediate stage in the progression from normal tissue to neoplasia, as evidenced by the identification of pre -malignant dysplastic conditions such as CIN.

By "high-grade cervical disease" is intended those conditions classified by colposcopy as premalignant pathology, malignant pathology, moderate to severe dysplasia, and cervical cancer. Underlying high-grade cervical disease includes histological identification of CINII, CINIII, HSIL, carcinoma in situ, adenocarcinoma, and cancer (FIGO stages I-IV). In contrast, "low-grade cervical disease" includes those conditions classified as CINI or LSIL and also refers to ASCUS or AGUS samples. By "cervical cancer" is intended any cancer or cancerous lesion associated with cervical tissue or cervical cells. The methods of the present invention permit the identification of cervical disease, which includes moderate to severe dysplasia and cervical cancer (i.e., CINII conditions and above), based on the expression levels of HPV E6 and/or E7 mRNA that are expressed at a higher level in cervical disease. In some embodiments, HPV E6 and/or E7 expression levels can be used to diagnose high-grade cervical disease and to distinguish high-grade disease from low-grade disease.

"Diagnosing cervical disease" is intended to include, for example, diagnosing or detecting the presence of cervical disease, monitoring the progression of the disease, and identifying or detecting cells or samples that are indicative of cervical disease. The terms diagnosing, detecting, and identifying cervical disease are used interchangeably herein. As discussed above, a significant percentage of patients presenting with Pap smears classified as normal, CINI, or ASCUS actually have lesions characteristic of high- grade cervical disease. Thus, the methods of the present invention permit the identification of high-grade cervical disease in all patient populations, including these "false negative" patients, and facilitate the detection of rare abnormal cells in a patient sample. The diagnosis can be made independent of cell morphology, although the methods of the invention can also be used in conjunction with conventional diagnostic techniques, e.g., Pap test, molecular testing for high-risk types of HPV, etc. The methods of the present invention can also serve as a molecular diagnostic tool for the grading of cervical tissue samples (e.g., biopsy samples) as low-grade or high-grade disease, which is less subjective than methods that assess morphological and histological characteristics.

HPV types have been divided into high and low-risk categories based on their association with cervical cancer and precancerous lesions. Low-risk HPV types include types 6, 11, 42, 43, 44 and are not associated with an increased risk of cervical cancer. In contrast, high-risk HPV types, including types 16, 18, 26, 31, 33, 35, 39, 45, 51, 52, 53, 56, 58, 59, 66, 68, 73, and 82, are those that are strongly associated with cervical cancer and squamous intraepithelial lesions. See, for example, Munoz et al. (2003) N Engl J Med 348:518-527; and Wright et al. (2004) Obstet. Gynecol. 103:304-309, each of which is herein incorporated by reference in its entirety. In fact, over 99% of cervical cancers are associated with high-risk HPV infection. Persistent high-risk HPV infection leads to the disruption of the cell cycle and mitotic checkpoints in cervical cells through the action of HPV genes E2, E6, and E7. In particular, HPV E7 causes an increase in cyclin E and the subsequent release of the transcription factor E2F from the retinoblastoma (Rb) protein. The released E2F transcription factor then triggers the transcription of a variety of S-phase genes, including topoisomerase II alpha (Topo2A), MCM proteins, cyclins El and E2, and pl4arf, resulting in loss of cell cycle control. HPV E2 further stimulates overexpression of S-phase genes such as p21^waf11 by activating the Sp-I transcription factor. The cell cycle disruption caused by persistent HPV infection can lead to mild cervical dysplasia that may then progress to moderate or severe dysplasia and eventually to cervical cancer in some cases.

HPV infection within cervical keratinocytes results in a number of alterations that disrupt the activities within the cell cycle. The E6 and E7 oncoproteins of the high-risk HPV subtypes have been implicated in a number of cellular processes related to increased proliferation and neoplastic transformation of the infected keratinocytes. The E6 protein has been implicated in two critical processes. The first is the degradation of the p53 tumor suppressor protein through ubiquitin-mediated proteolysis. Removal of functional p53 eliminates a major cell cycle checkpoint responsible for DNA repair prior to entry into DNA replication and mitosis (Duensing and Munger (2003) Prog Cell Cycle Res. 5:383- 391). In addition, E6 has been shown to interact with the c-myc protein and is responsible for direct transcriptional activation of the hTERT gene with subsequent expression of telomerase (McMurray and McCance (2003) J Virol. 77:9852-9861; Veldman et al. (2003) Proc Natl Acad Sci U.S.A. 100: 8211-8216). Activation of telomerase is a key step in cancer biology responsible for the maintenance of telomere length on replicating chromosomes and this enzyme ensures functionally intact chromosomes during cellular immortalization.

Various splice variants of HPV E6 are expressed by high-risk HPV types, including but not limited to HPV 16 and HPVl 8. In some embodiments of the methods of the invention, the expression level of the unspliced HPV 16 E6 transcript is assessed (the coding sequence of which is provided in SEQ ID NO: 79) or the level of expression of the E6*I transcript is assessed (the coding sequence of which is provided in SEQ ID NO: 81). In yet other embodiments, the level of both the unspliced E6 transcript and the E6*I transcript is detected using a primer pair that can amplify either transcript. In still other embodiments, a primer pair is used that can amplify all E6 transcripts. The HPV oncoprotein E7 is known to contribute to cellular proliferation through two independent mechanisms. The first is the inactivation of the TGF -beta tumor suppressor pathway responsible for cell cycle arrest at the Gl phase through direct interaction of E7 with the Smad proteins (Smad 2, 3 and 4), thereby inhibiting their ability to bind to DNA (Lee et al. (2002) J Biol Chem. 277:38557-38564). Likewise, E7 is known to specifically interact with the Rb tumor suppressor protein. Within the Gl phase of the cell cycle, Rb complexes the E2F transcription factor and prevents E2F from activating gene transcription. At the Gl /S boundary, the Rb protein is phosphorylated with release of the E2F transcription factor - thereby initiating E2F gene transcription and entry into the S phase of the cell cycle. The HPV E7 oncoprotein abrogates this control mechanism by directly binding with Rb and displacing E2F from the complex. This results in E2F driven gene transcription independent of normal cell cycle control (Duensing and Munger (2003) Prog Cell Cycle Res. 5:383-391; Duensing and Munger (2004) Int J Cancer 109:157-162; Clarke and Chetty (2001) Gynecol Oncol. 82:238-246). This release of E2F uncouples gene transcription from cell cycle control and results in prolonged and aberrant transcription of S-phase genes responsible for DNA synthesis and cellular proliferation. In addition, the combined actions of both E6 and E 7 have been shown to contribute to centrosome abnormalities and the subsequent genomic instability in cervical neoplasia (Duensing and Munger (2004) Int J Cancer 109:157-162). The methods disclosed herein provide superior detection of high-grade cervical disease in comparison to PAP smears and/or HPV infection testing and allow for molecular grading and detection of disease with biopsy samples, which is less subjective than pathological methods based on morphological and histological characteristics. In particular aspects of the invention, the sensitivity and specificity of the present methods are equal to or greater than that of conventional Pap smears. As used herein, "specificity" refers to the level at which a method of the invention can accurately identify samples that have been confirmed as NIL by colposcopy (i.e., true negatives). That is, specificity is the proportion of disease negatives that are test-negative. In a clinical study, specificity is calculated by dividing the number of true negatives by the sum of true negatives and false positives. By "sensitivity" is intended the level at which a method of the invention can accurately identify samples that have been colposcopy-confϊrmed as positive for high- grade cervical disease (i.e., true positives). Thus, sensitivity is the proportion of disease positives that are test-positive. Sensitivity is calculated in a clinical study by dividing the number of true positives by the sum of true positives and false negatives. In some embodiments, the sensitivity of the disclosed methods for the detection of high-grade cervical disease is at least about 70%, at least about 80%, or at least about 90, 91, 92, 93, 94, 95, 96, 97, 98, 99% or more. Furthermore, the specificity of the present methods is at least about 70%, at least about 80%, or at least about 90, 91, 92, 93, 94, 95, 96, 97, 98, 99% or more. The term "positive predictive value" or "PPV" refers to the probability that a patient has cervical disease, in general, or high-grade cervical disease when restricted to those patients who are classified as positive using a method of the invention. PPV is calculated in a clinical study by dividing the number of true positives by the sum of true positives and false positives. In some embodiments, the PPV of a method of the invention for diagnosing cervical disease or high-grade cervical disease is at least about 40%, while maintaining a sensitivity of at least about 90%, more particularly at least about 95%. The "negative predictive value" or "NPV" of a test is the probability that the patient will not have the disease when restricted to all patients who test negative. NPV is calculated in a clinical study by dividing the number of true negatives by the sum of true negatives and false negatives.

Quantitative analysis of expression levels of HPV E6 and/or E7 permits the detection of cervical disease and in some embodiments, the differentiation of samples indicative of underlying high-grade cervical disease from samples that are indicative of benign proliferation, early-stage HPV infection, or mild dysplasia. By "early-stage HPV infection" is intended HPV infection that has not progressed to cervical dysplasia. As used herein, "mild dysplasia" refers to LSIL and CINI where no high-grade lesion is present. Some embodiments of the methods of the invention also distinguish cells indicative of high-grade disease from normal cells, immature metaplastic cells, and other cells that are not indicative of clinical disease. In this manner, some embodiments of the methods of the invention permit the accurate identification of high-grade cervical disease, even in cases mistakenly classified as normal, CINI, LSIL, or ASCUS by traditional Pap testing (i.e., "false negatives"). In some embodiments, the methods for diagnosing cervical disease, in general, or high-grade cervical disease are performed as a reflex to an abnormal or atypical Pap smear. That is, the methods of the invention may be performed in response to a patient having an abnormal or atypical Pap smear result. In other aspects of the invention, the methods are performed as a primary screening test for cervical disease or high-grade cervical disease in the general population of women, just as the conventional Pap test is performed currently. The methods of the invention may also be used to grade biopsy specimens or to detect cervical disease in biopsy samples as a more accurate alternative to analysis of morphological or histological characteristics.

The diagnostic methods of the invention comprise collecting a body sample (e.g., cervical sample) from a subject and analyzing the expression levels of HPV E6 and/or E7 in the sample.

By "body sample" is intended any sampling of cells, tissues, or bodily fluids in which expression of a HPV E6 or E7 mRNA can be detected. Examples of such body samples include but are not limited to blood, lymph, urine, gynecological fluids, biopsies, and smears. Body samples may be obtained from a patient by a variety of techniques including, for example, by scraping or swabbing an area or by using a needle to aspirate bodily fluids. Methods for collecting various body samples are well known in the art. The terms "subject" and "patient," which are used interchangeably herein, refer to a human. The subject may have had a Pap smear prior to assessing the expression levels of a high-risk HPV E6 and/or E7, which may or may not have been abnormal or atypical. The subject may or may not have had HPV genotyping to assess the presence of HPV DNA from a high-risk HPV type(s).

In particular embodiments, the body sample comprises cervical cells, as cervical tissue samples or as cervical cells in suspension, particularly in a liquid-based preparation. In one embodiment, cervical samples are collected according to liquid-based cytology specimen preparation guidelines such as, for example, the SurePath® (TriPath Imaging, Inc.) or the ThinPrep® preparation (CYTYC, Inc.). Body samples may be transferred to a glass slide for viewing under magnification. Fixative and staining solutions may be applied to the cells on the glass slide for preserving the specimen and for facilitating examination. In other embodiments, the body sample comprises formalin-fixed paraffin- embedded tissue samples, such as from a biopsy. The expression level of HPV E6 and/or E7 in the body samples is determined.

Many expression detection methods use isolated RNA. Any RNA isolation technique that does not select against the isolation of mRNA can be utilized for the purification of RNA from cervical cells (see, e.g., Ausubel et al, ed., (1987-1999) Current Protocols in Molecular Biology (John Wiley & Sons, New York). Additionally, large numbers of tissue samples can readily be processed using techniques well known to those of skill in the art, such as, for example, the single-step RNA isolation process of Chomczynski (1989, U.S. Pat. No. 4,843,155). In some embodiments, total RNA is extracted using a modified protocol from Epicentre, as described by Murphy et al. (2009) J Virol Mehods 156(1-2): 138-44, which is herein incorporated by reference in its entirety. The isolated RNA can then be reverse transcribed into complementary DNA

(cDNA) using any method known in the art. In some embodiments, isolated RNA is reverse transcribed using the High capacity cDNA Reverse Transcription Kit from Applied Biosystems (Foster City, CA, USA). In some of these embodiments, 1 μg RNA is reverse transcribed in a 20 μl reaction mix. In order to detect potential DNA contamination, a reaction mix is prepared that lacks the reverse transcriptase enzyme (i.e., no RT control).

In those embodiments wherein the expression level of HPV E6 or E7 is assessed with amplification techniques, such as real-time polymerase chain reaction (PCR), the reverse transcription reaction and the PCR can be performed in the same reaction mixture or the two reactions can be performed as two separate steps. In particular embodiments of the invention, the reverse transcription reaction is performed first, followed by amplification via PCR in a separate reaction mixture.

In particular embodiments, the diagnostic methods of the invention comprise collecting a body sample from a patient and performing real-time PCR analysis (e.g., TaqMan®) on reverse-transcribed RNA isolated from the sample to measure the expression levels of HPV E6 and/or E7. Such methods typically utilize pairs of oligonucleotide primers that are specific for the E6 or E7 of interest. Methods for designing oligonucleotide primers specific for a known sequence are well known in the art. Real-time PCR permits the detection of PCR products at earlier stages of the amplification reaction. Specifically, in real-time PCR the quantitation of PCR products relies on the few cycles where the amount of nucleic acid material amplifies logarithmically until a plateau is reached. During the exponential phase, the amount of target nucleic acid material should be doubling every cycle, and there is no bias due to limiting reagents. Methods and instrumentation for performing real-time PCR are well known in the art. See, for example, Bustin (2000) J. Molec. Endocrinol. 25:169-193; Freeman et al. (1999) Biotechniques 112:124-125; Halford (1999) Nat. Biotechnol. 17:835; and Heid et al. (1996) Genome Res. 6(10):986-994, all of which are herein incorporated by reference in their entirety. Many different dyes and probes are available for monitoring PCR and detecting

PCR products, more particularly real-time PCR products. In particular aspects of the invention, a 5' nuclease assay is used to monitor PCR, particularly real-time PCR (e.g., TaqMan®), and to detect PCR amplification products of HPV E6 and E7 RNA. In such 5' nuclease assays, a fluorogenic/quencher oligonucleotide probe (e.g., a TaqMan® probe) is added to the PCR reagent mix. The fluorogenic/quencher oligonucleotide probe (e.g., TaqMan® probe) comprises a high-energy fluorescent reporter dye at the 5' end (e.g., FAM) and a low-energy quencher dye at the 3' end (e.g., TAMRA or a non-fluorescent quencher). When the fluorogenic/quencher oligonucleotide probe (e.g., TaqMan® probe) is intact, the reporter dye's fluorescent emission is suppressed by the close proximity of the quencher. The fluorogenic/quencher oligonucleotide probe (e.g., TaqMan® probe) is further designed to anneal to a specific sequence of the HPV E6 or E7 between forward and reverse primers, and, therefore, the probe binds to the E6 or E7 nucleic acid material in the path of the polymerase. PCR amplification results in cleavage and release of the reporter dye from the quencher-containing probe by the nuclease activity of the polymerase. Thus, the fluorescence signal generated from the released reporter dye is proportional to the amount of the PCR product. Methods and instrumentation (e.g., ABI Prism 7700 Detector; Perkin Elmer/ Applied Biosytems Division) for performing real-time PCR using a variety of probes are well known in the art. Moreover, methods for designing appropriate probes for real-time PCR are generally known in the art and commercially available.

Sequences of the genes encoding E6 and E7 transcripts from high-risk HPV types are known in the art as are the sequences of the transcripts, from which appropriate primers and probes for real-time PCR can be designed. The complete genome of HPV 16, HPV18, HPV31, HPV33, and HPV45 is provided in Genbank Accession Nos: K02718,

X05015, EF422123, EF422127, and EF202167, respectively. The coding sequence for the HPV 16 E6 unspliced transcript is provided in GenBank Accession No. AF003019 (and is set forth in SEQ ID NO: 79). The HPV16 E6*I splice variant coding sequence is provided in GenBank Ace. No. D00735.1 (and is set forth in SEQ ID NO: 81). The coding sequences for HPV18, HPV31, HPV33, and HPV45 E6 are provided in GenBank Ace.

Nos. A06328, EF422123.1, EF422127, and Y13218, respectively (and are set forth in SEQ ID NOs: 82, 83, 84, and 85, respectively). Coding sequences for HPV16 and HPV18 are provided in GenBank Ace. Nos. 80 and 82, respectively (and are set forth in SEQ ID NOs: 80 and 82, respectively). In those embodiments wherein the expression level of HPV 16 E6 is being assessed, the level of any one of the various HPV 16 E6 transcripts that may result from alternative splicing can be assessed, or the levels of various combinations of the transcripts, or all of the transcripts can be assessed through the use of a primer pair (and in some embodiments, a probe) that specifically amplify one transcript, more than one transcript, or all transcripts. Such primer/probe pairs can be found in Table 15 and include primers having the nucleotide sequences set forth in SEQ ID NOs: 4 and 5 and corresponding probe having the nucleotide sequence set forth in SEQ ID NO: 6 to detect unspliced HPV 16 E6, and primers having the nucleotide sequences set forth in SEQ ID NOs: 7 and 8 and corresponding probe having the nucleotide sequence set forth in SEQ ID NO: 9 to detect unspliced HPV 16 E6 and the HPV 16 E6*I splice variant. In some embodiments, real-time PCR primers and probes are designed so as to amplify and detect all of the HPV 16 E6 transcripts. In some of these embodiments, the primers have the nucleotide sequences set forth in SEQ ID NOs: 1 and 2 and the corresponding probe has the nucleotide sequence set forth in SEQ ID NO: 3. In those embodiments wherein the expression level of multiple splice variants of the E6 or E 7 gene from a high-risk HPV type are assessed independently or simultaneously through the use of a primer/probe set that amplifies and detects all transcripts, the variant, variant combination, or the simultaneous detection of all transcripts having the highest level of expression is used to diagnose cervical disease or to distinguish low-grade from high-grade disease by comparing the patient's highest E6 and/or E7 expression level to a provided threshold amount.

In other embodiments wherein the expression level of HPV 18 E6 is measured, primers having the nucleotide sequences set forth in SEQ ID NOs: 13 and 14 and a corresponding probe having the nucleotide sequence set forth in SEQ ID NO: 15 can be used. For detection of HPV31 E6, primers having the nucleotide sequences set forth in SEQ ID NOs: 19 and 20 and a probe having the nucleotide sequence set forth in SEQ ID NO: 21 may be used in some embodiments. HPV33 E6 expression levels can be assessed with primers having the nucleotide sequences set forth in SEQ ID NOs: 22 and 23 and a probe having the sequence set forth in SEQ ID NO: 24. In other embodiments, HPV45 E6 expression can be measured using primers having the sequences set forth in SEQ ID NOs: 25 and 26 and a probe having the sequence set forth in SEQ ID NO: 27.

In those embodiments wherein E7 expression levels are determined, HPV 16 E7 expression can be measured with primers having the nucleotide sequences set forth in SEQ ID NOs: 10 and 11 and a probe having the sequence set forth in SEQ ID NO: 12. For the detection of HPV 18 E7, primers having the nucleotide sequences set forth in SEQ ID NOs: 16 and 17 and a probe having the nucleotide sequence set forth in SEQ ID NO: 18 may be used in some embodiments.

Some subjects may be infected with multiple high-risk HPV types. For these subjects, the expression level of HPV E6 and/or E7 of each of the various viral types can be assessed and the HPV E6 and/or E7 transcript that is expressed at the highest level can be used to diagnose cervical disease.

In some embodiments, the level of expression of HPV E6 alone in a patient sample is determined. The HPV E6 expression level of the patient sample can then be compared to threshold levels provided herein specific for HPV E6 expression levels to diagnose cervical disease. In other embodiments, the expression level of HPV E7 is solely determined. In these embodiments, the HPV E7 expression level of the patient sample can be compared to threshold levels provided herein that are specific for HPV E7 expression levels to diagnose cervical disease. In yet other embodiments, the expression of both HPV E6 and E7 is determined. In some of these embodiments, the highest of the HPV E6 or HPV E7 expression level can be compared to threshold levels provided herein that are specific for a combination of HPV E6 and E7 expression levels to diagnose cervical disease. Provided herein are threshold levels of HPV E6 and/or E7 expression that can be used to determine if a sample is indicative of the presence or likelihood of developing cervical disease. Threshold levels are also provided that can distinguish the presence or likelihood of developing low-grade vs. high-grade cervical disease. As used herein, "threshold" or "threshold amount" refers to the cut-off point of E6, E7, or the highest of E6 and E7 expression levels above which is indicative of a subject having or likely to develop cervical disease or in some embodiments, high-grade cervical disease. Thresholds are provided that distinguish normal samples from those that are diseased. Samples with E6, E7, or the highest of E6 and E7 expression levels that fall below this threshold are considered normal, whereas those samples that are above this threshold amount are indicative of the presence or increased likelihood of developing cervical disease.

Thresholds are also provided that distinguish low-grade cervical disease from high-grade disease. Samples with E6, E7, or the highest of E6 and E7 expression levels that fall below this threshold are indicative of those that are low-grade and that most likely will not progress, whereas those samples that are above this threshold amount are indicative of those that are high-grade or are likely to progress to high-grade cervical disease.

In some embodiments, the correlation between E6 and/or E7 expression levels and cervical cancer progression is independent of the integration status of the HPV viral genome into the host genome. In these embodiments, the level of expression of E6 and/or E7 can be used to diagnose cervical disease using the thresholds provided herein even if the virus has not integrated into the host genome.

The expression level of a gene measured with real-time PCR can be determined and defined using various analytical methods, including but not limited to a threshold cycle (C_T), absolute copy number derived from a standard curve, a normalized target value wherein the absolute copy number of the target is normalized to the absolute copy number of an endogenous control gene, fold-change of the normalized target value over the normalized target value of normal samples, and the delta-delta C_T (ΔΔC_T) method.

Threshold levels that distinguish normal from diseased tissue and low-grade from high-grade disease are provided for each method of defining the level of E6 and E7 expression in cytology or tissue (e.g., biopsy) samples. It should be noted that those patients from which samples were derived and E6 and/or E7 expression levels measured therefrom that have E6 and/or E7 expression levels that are at or near the threshold levels (e.g., within 1%, 2%, 3%, 4%, 5%, 6%, 7%, 8%, 9%, 10%, 11%, 12%, 13%, 14%, 15%) should be monitored or the measurement of the E6 and/or E7 expression level should be repeated with the same or another body sample.

As used herein, the terms "threshold cycle" or "cycle threshold," which are used interchangeably herein and are abbreviated as C_T, refer to the amplification cycle number at which exponential amplification of a target is first detected. There is a linear relationship between ( \ value and the starling concentration of DNΛ in the reaction mixture. The C r value is negatively correlated with target expression level. A negative correlation defines a relationship between two variables (e.g., C_T value and target expression level), wherein a change in one variable in one direction results in a change in the second variable in the opposite direction. Thus, as the C₁ value decreases, tbe target expression level increases and as the Cj value increases, the target expression level decreases.

For cervical cytology samples, a HPV E6 C₁, HPV E7 Ci, or the lowest of a HPV E6 C s or HPV E7 Ci below about 38 is indicative of the presence of cervical disease in the subject front which the cytology sample is derived or an increased likelihood of the subject developing cervical disease when compared to a control subject. Λs used herein, the terms "control subject'^", "norraal .subject"', and "norma! patienf refer to a subject that has not been diagnosed as having cervical disease or docs not have a predisposition for cervical disease. A HPV F6 C₁ value, a HPV E7 C₇ value, or the lowest of a HPV E6 C₁ value or a HPV E7 C_T value of about }5 to about 3X (including, but not limited to about 35, 35.1, 35.2, 35.3, 35/1, 35.5, 35.6, 35.7, 35.8, 35.936. 36.1. 36.2. 36.3. 36/1, 36.5, 36.6, 36.7, 36.8, 36.9, 37, 37.1, 37.2, 37.3, 37.4, 37,5, 37,6, 37.^", 37.8, 37.9, and 38) in a cervical cytology sample is indicative of the presence or increased likelihood of developing low-grade cervical disease, whereas a HPV F6 C_T value, a HPV E7 C_T value, or the lowest of a HPV E6 C₁ value or a HPV E7 Gj value of less than about 35 is indicative of the presence or increased likelihood of developing high-grade cervical disease.

For cervical biopsy samples, a HPV E6 Oj, HPV E7 C₁, or the lowest of HPV E6 Cj or ) IPV E7 C_r below about .]i< is indicative of the presence of cervical disease in the subject from which the biopsy sample is derived or an increased likelihood of the subject developing cervical disease when compared to a control subject. A HPV lib C^'χ value, a HPV E7 Ci value, or the lowest of a HPV E 6 C₁ value or a HPV E7 C₁ value of about 35 to about 38 (including, but not limited to about .15, 35.1 , 35.2, 35.3, 35.4, 35.5, 35.6, 35.7, 35.8, 35.9, 36. 36.1, 36.2, 36.3. 36.4. 36.5, 3o.6, 3o.^~, 3b.8, 3b.9, 37, 37.1, 37.2, 37.3, 37.4, 37.5, 37.6, 37.8, 37.9, and 38) in a cervical biopsy sample is indicative of tbc presence or increased likelihood of developing low-grade cervical disease, whereas a HPV YP- C _I value, a HPV F7 C_r value, or the lowest of a ΪIPV R6 C _; value or a HPV VJ C₁ value of lest_* than about 35 is indicative of the presence or increased likelihood of developing high-grade cervical disease.

Through the use of a standard curve, the expression level of 1 IPV Iϋ6 or E7 can be expressed as an absolute copy number. By "absolute copy number" is intended the estimated amount of nucleic acid starting material in a sample based on the comparison of the Ci value of the target transcript in an experimental sample to a standard curve that plots the Cr values of the target transcript in PCR reaction mixtures that have been provided a known amount of nucleic acid starting material against the amount of the starting material. The nucleic acid starting material can be provided by adding into the reaction mixture, for example, a DNA plasmid that encodes the transcripts.

For cervical cytology samples, a HPV E6 copy number, a HPV E7 copy number, or the highest of a HPV E6 copy number or a HPV E7 copy number above about 1 is indicative of the presence of cervical disease in the subject from which the cytology sample is derived or an increased likelihood of the subject developing cervical disease when compared to a control subject. Λ HPV E6 copy number, a HPV E7 copy number, or the highest of a HPV E6 copy number or a HPV E7 copy number of about 1 Io about 10 (including, but not limited to about 1 , 1 .5, 2, 2,5, 3, 3.5, 4, 4.5, 5, 5.5, 6, 6.5, 7, ^".5, 8, 8.5, 9. 9.5. and 10) in a cervical cytology sample is indicative of the presence or increased likelihood of developing low-grade cervical disease, whereas a HPV E6 copy number, a HPV E7 copy number, or the highest of a HPV E6 copy number or a HPV E7 copy number of greater than about 10 is indicative of the presence or increased likelihood of developing high-grade cervical disease.

For cervical biopsy samples, a HPV E6 copy number, or the highest of UPV F6 copy number or HPV H^" copy number above about 10 is indicative of the presence of cervical disease in the subject from which the biopsy sample is derived or an increased likelihood of the subject developing cervical disease when compared to a control subject. A HPV E6 copy number, or the highest of HPV E 6 copy number or HPV E 7 copy number of about 10 to about 50 (including, but not limited to about 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, '-10, 41, 42, 43, 44, 45, 46, 47, 48, 49, and 50) in a cervical biopsy .sample is indicative of the presence or increased likelihood of developing low-grade cervical disease, whereas a HPV Eό copy number, or the highest of HPV E6 copy number or I IPV F7 copy number of greater than about 50 is indicative of the presence or increased likelihood of developing high-grade cervical disease.

Cervical biopsy samples having a l ϊPV Ml copy number above about 1 is indicative of the presence of cervical disease in the subject from which the biopsy sample is derived or an increased likelihood of the subject developing cervical disease when compared to a control subject. A HPV E7 copy number of about 1 to about 10 (including, but not limited to about 1, 1.5, 2, 2.5, 3, 3.5, 4, 4,5, 5. 5.5, 6, 6.5, 7, 7.5, 8, 8.5, 9, 9,5, and 10 ) in a cervical biopsy sample is indicative of the presence or increased likelihood of developing low-grade cervical disease, whereas a HPV F7 copy number of greater than about 10 is indicative of the presence or increased likelihood of developing high-grade cervical disease.

Absolute copy numbers of HPV E6 or E7 can be normalized against the absolute copy number of an endogenous control gene to derive a normalized target value. The terms "normalized target value" and "normalized copy number" are used interchangeably herein and refer to the absolute target copy number derived from a standard curve as described above that is divided by the absolute copy number of an endogenous control that was derived from an endogenous control standard curve as described above. In some embodiments, the normalized target value can be multiplied by 1000 in order to avoid presenting the data as very small numbers. Unless noted otherwise, all of the normalized copy numbers presented herein and thresholds for the same have been multiplied by 1000. In order to normalize HPV E6 and E7 mRNA expression levels, it is necessary to first identify a suitable endogenous control gene that can reliably serve as a marker for RNA quality and has an expression profile that remains relatively constant across normal and diseased cervical tissues and across various cervical disease states (e.g., CINI, CINII, CINIII, SCC, LSIL, HSIL). Provided herein (see Example 1) is an expression analysis of a panel of various housekeeping genes in cytology and biopsy samples that identifies those housekeeping genes that can be used as an endogenous control gene to normalize HPV E6 or E7 mRNA levels for the purposes of diagnosing cervical disease in a body sample (e.g., cytology, biopsy specimens). Therefore, methods are provided for normalizing the expression level of a target gene in a cervical sample comprising determining the expression level of the target gene and an endogenous control gene, and dividing or subtracting the endogenous control gene expression level from the target gene expression level, wherein the endogenous control gene is one that is listed below. Further methods are provided for diagnosing cervical disease in a subject, wherein expression levels of HPV E6 or E 7 mRNA from a high-risk HPV type in a sample from the subject are determined and normalized against a suitable endogenous control gene provided herein to produce a normalized value of HPV E6 or E7 mRNA. An increase in the normalized value of HPV E6 or E7 mRNA in the subject compared to a normal subject is indicative of the presence of cervical disease or an increased likelihood of the subject developing cervical disease when compared to the normal subject.

For cytology samples, HPV E6 and E7 mRNA levels can be normalized using any one of the following endogenous control genes: peptidylprolyl isomerase A (PPIA),, beta- 2-microglobulin (B2M), phosphoglycerate kinase 1 (PGKl), large ribosomal protein (RPLPO), ubiquitin C (UBC), beta-glucuronidase (GUSB), and polymerase (RNA) II (DNA directed) polypeptide A (POLR2A). In particular embodiments, HPV E6 and E7 mRNA levels are normalized to GUSB and in other embodiments, POLR2A is used to normalize E6 and E7 expression levels.

For biopsy samples, HPV E6 and E7 mRNA levels can be normalized to any one of the following endogenous control genes: tyrosine 3-monooxygenase/tryptophan 5- monooxygenase activation protein, zeta polypeptide (YWHAZ), hydroxymethylbilane synthase (HMBS), transferrin receptor (TFRC), peptidylprolyl isomerase A (PPIA), importin 8 (IPO8), beta-2 -microglobulin (B2M), phosphoglycerate kinase 1 (PGKl), large ribosomal protein (RPLPO), ubiquitin C (UBC), beta-glucuronidase (GUSB), and polymerase (RNA) II (DNA directed) polypeptide A (POLR2A). In particular embodiments, HPV E6 and E7 mRNA levels are normalized to GUSB and in other embodiments, POLR2A is used to normalize E6 and E7 expression levels.

For cervical cytology samples wherein the target transcript expression level has been normalized by the expression of GUSB and wherein the normalized target value has been multiplied by 1000, a HPV E6 normalized copy number or a HPV E7 normalized copy number above about 1 is indicative of the presence of cervical disease in the subject from which the cytology sample is derived or an increased likelihood of the subject developing cervical disease when compared to a control subject. A HPV E6 normalized copy number (normalized to GUSB) or a HPV E7 normalized copy number (normalized to GUSB) of about 1 to about 20 (including, but not limited to about 1 , 2, 3, 4, 5, 6, 7, K, 9, 10, 11, 12, 13, 14, 15. 16, 17, 18, 19, and 2Oj in a cervical cytology sample is indicative of the presence or increased likelihood of developing low-grade cervical disease, whereas a HPV £6 normalized copy number (normalized to GlJSB) or a HPV El normalized copy number (normalized to GUSB) of greater than about 20 is indicative of the presence or increased likelihood of developing high- grade cervical disease.

Cervical cytology samples, wherein the target transcript expression level has been normalized by the expression of GUSB and wherein the normalized target value has been multiplied by 1000, and for which the highest of HPV E6 normalized copy number or HPV E7 normalized copy number above about 1 is indicative of the presence of cervical disease in the subject from which the cytology sample is derived or an increased likelihood of the subject developing cervical disease when compared to a corstro! subject. The highest of HPV Eό normalized copy number (normalized to GUSB) or HPV E7 normalized copy number (normalized to GUSB) of about 1 to about 100 (including, but not limited to about 1, 2. 3, 4, 5, ό, 7, 8, 9, 10. 1 1 , 12, 13, 14, 15, 16, 17, IS. 19, 20, 25, 30, 35, 40, 45, 50, 55, 60, 65, 70, 75, 80, 85, 90, 95, and 100) in a cervical cytology sample is indicative of the presence or increased likelihood of developing low-grade cervical disease, whereas the highest of HPV E6 normalized copy number (normalized to GUSBj or HPV E7 normalized copy number (normalized to CHJSB) of greater than about 100 is indicative of the presence or increased likelihood of developing high-grade cervical disease.

For cervical biopsy samples wherein the target transcript expression level has been normalized by the expression of GUSB and wherein the normalized target value has been multiplied by 1000, a HPV E6 normalized copy number, or the highest of HPV Eo normalized copy number or HPV E 7 normalized copy number above about 1 is indicative of the presence of cervical disease in the subject from which the biopsy sample is derived or an increased likelihood of the subject developing cervical disease when compared to a control subject. A HPV E6 normalized copy number (normalized to GUSBj, or the highest of HPV £6 normalized copy number (normalized to GUSB ) or HPV E7 normalized copy number (normalized to GUSB) of about 1 to about 20 (including, but not limited to about 1, 2, 3, 4, 5, 6, 7, 8, 9, 10. 1 1 , 12, 13, 14, 15, 16, 17, IS. 19, and 20) in a cervical biopsy sample is indicative of the presence or increased likelihood of developing low-grade cervical disease, whereas a HPV E6 normalized copy number (normalized to GUSBj, or the highest of HPV E6 normalized copy number (normalized to GUSBj or HPV E7 normalized copy number (normalized to GUSB) of greater than about 20 is indicative of the presence or increased likelihood of developing high-grade cervical disease.

Cervical biopsy samples, wherein the target transcript expression level has been normalized by the expression of GUSB and wherein the normalized target value has been multiplied by 1000, havirsg a F-IPV E7 normalized copy number above abou? 0,1 is indicative of the presence of cervical disease in the subject from which the biopsy sample is derived or an increased likelihood of the subject developing cervical disease when compared to a control subject. A HPV E7 normalized copy number (normalized to GUSBj of about 0.1 to about 5 (including, but not limited to about 0.1, 0.2, 0.3, GA 0.5, OA 0,7, 0.8, 0», 1 , 1 .5, 2, 2.5, 3, 3.5, 4, 4.5, and 5) in a cervical biopsy sample is indicative of the presence or increased likelihood of developing low-grade cervical disease, whereas a HPV E7 normalized copy number (normalized Io GUSB) of greater than about 5 is indicative of the presence or increased likelihood of developing high-grade cervical disease.

For cervical cytology samples wherein the target transcript expression level has been normalized by the expression of POLR2A and wherein the normalized target value has been multiplied by 1000, a HPV E6 normalized copy number or a HPV E7 normalized copy number above about 1 is indicative of the presence of cervical disease in the subject from which the cytology sample is derived or an increased likelihood of the subject developing cervical disease when compared to a control subject. A HPV Eb normalized copy number (normalized to POLR2A) or a HPV E7 normalized copy number (normalized to POLR2A) of about 1 to about 10 (including, but not limited to about I ₅ 1 .5, 2. 2.5. 3, 3.5. 4, 4.5, 5, 5.5, 6, 6.5, 1. 7.5, 8, %.5. ^Q, 9.5, and 10) in a cervical cytology sample is indicative of the presence or increased likelihood of developing low-grade cervical disease, whereas a HPV E6 normalized copy number (normalized to POLR2A) or a HPV E7 normalized copy number (normalized to POLR2A) of greater than about 10 is indicative of the presence or increased likelihood of developing high-grade cervical disease. Cervical cytology samples, wherein the target transcript expression level has been normalized by the expression of POLR2A and wherein the normalized target value has been multiplied by 1000, and for which the highest of HPV E6 normalized copy number or HPV E 7 normalized copy number above about 1 is indicative of the presence of cervical disease m ^" the subject from which the cytology sample is derived or an increased likelihood of the subject developing cervical disease when compared to a control subject. The highest of ) (PV E6 normalized copy number S normalized to POLR2A) or ) (PV E7 normalized copy number (normalized to POLR2A) of about 1 to about 20 (including, but not limited to 1 , 1 .5, 2. 2.5, 3, 3.5. 4, 4.5, 5, 5.5, 6, 6.5, 7, 7.5, 8. 8.5. 9, ^l>.5, 10, 10.5, 1 1. 1 1.5, 12, 12.5, 13, 13.5, 14, 14.5, 15, 15.5, 16. 16.5. 17, 17.5, 18, 18.5, K>, 19.5, and 20) in a cervical cytology sample is indicative of the presence or increased likelihood of developing low-grade cervical disease, whereat_* the highest of HPV Eό normalized copy number (normalized to POLR2A) or HPV E7 normalized copy number (normalized to POLR2A) of greater than about 20 Ls indicative of the presence or increased likelihood of developing high-grade conical disease.

For cervical biopsy samples wherein the target transcript expression level has been normalized by the expression of POLR2A and wherein the normalized target value has been multiplied by 1000, a HPV E6 normalized copy number, a HPV E7 normalized copy number, or the highest of HPV Eb normalized copy number or HPV E7 normalized copy number above about 1 is indicative of the presence of cervical disease in the subject from which the biopy sample is derived or an increased likelihood of the subject developing cervical disease when compared to a control subject. A HPV E6 normalized copy number (normalized to POLR2A), a HPV I-¹7 norma Ii/ ed copy number (normalized to POΪJ12Λ ), or the highest of HPV £6 normalized copy number (normalized to POLR2Λ ) or HPV E7 normalized copy number (normalized to POLR2A) of about 1 to about 10 (including, but not limited to about 1 , 1.5, 2, 2.5, 3, 3.5, 4, 4.5, 5, 5.5, 6, 6.5. 7, 7.5, 8, 8.5, 9, 9.5, and 10} in a cen'fcal biopsy sample is indicative of fhe presence or increased likelihood of developing low-grade cervical disease, whereas a 1 IPV E6 normalized copy number (normalized to POLR2A), a HPV E7 normalized copy number (normalized to PQLR2A), or the highest of 1 IPV E6 normalized copy number (normalized to POLR2A) or I IPV VJ normalized copy number (normalized to POLR2A) of greater than about 10 is indicative of the presence or increased likelihood of developing high-grade cervical disease.

The normalized target copy number can be expressed as a fold-change of the normalized copy number of HPV E6 or HPV E7 of the patient sample relative to the normalized copy number HPV E6 or HPV E7 of a normal patient sample. This method of analyzing the data is referred to herein as the relative standard curve method and it involves dividing the normalized target (e.g., HPV E6, E7) value by the normalized target value for a normal patient sample (or an average of normal patient samples). The normalized target value can be derived using any of the suitable endogenous control genes described herein, including but not limited to GUSB and POLR2A.

In some embodiments, the normalized copy number of the normal patient sample is actually an average normalized copy number derived from a population of normal patient samples. The average normalized copy number can be derived from a population of normal patients samples of various sizes, including but not limited to, about 2, 5, 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 200, 300, 400, 500, 600, 700, 800, 900, 1000, 2000, 3000, 4000, 5000, 6000, 7000, 8000, 9000, 10,000 or greater number of patients. The larger the population, the more precise the average normalized copy number is expected to be. HPV E6 and E7 mRNA levels can be normalized to any of the suitable endogenous control genes disclosed herein, including but not limited to GUSB and POLR2A.

For cervical cytology samples wherein the target transcript expression level has been normalized by the expression of GUSB and is expressed as a fold-change relative to a normal patient sample, a fold-change of normalized HPV E6 copy number relative to the normalized HPV E6 copy number of a norma] patient body .sample, or the highest of a fold-change of normalized HPV E6 copy number relative to the normalized HPV E6 copy number of a normal patient body sample or a fold-change of normalized HPV E7 copy number relative to the normalized HPV E7 copy number of a normal patient body sample above about 1 is indicative of the presence of cervical disease its the subject from which the cytology sample is derived or a - n increased likelihood of the subject developing cervical disease when compared to a control subject. A fold-change of normalized HPV E6 copy number relative to the normalized HPV Eo copy number of a normal patient body sample (all normalized to GUSB), or the highest of a fold-change of normalized HPV E6 copy number relative to the normalized HPV Eb copy number of a normal patient body sample (all normalized to GUSB) or a fold-change of normalized HPV E7 copy number relah^'ve to the normalized HPV E7 copy number of a normal patient body sample (alS normalized to GUSB) of about 1 to about 10 (including, but not limited to about 1. 1.5. 2, 2,5, 3. 3.5, 4, 4.5, 5, 5.5, 6,

7, 7,5, 8, 8.5, 9, 9.5, and K)) in a cervical cytology sample is indicative of the presence or increased likelihood of developing low-grade cervical disease, whereas fold-change of normalized HPV E6 copy number relative to the normalized HPV Eb copy number of a normal patient body sample (all normalized to GUSBj, or the highest of a fold-change of normalized HPV E6 copy number relative to the normalized HPV E6 copy number of a normal patient body .sample (all normalized to GUSBj or a fold-change of normalized HPV E7 copy number relative to the normalized HPV V l copy number of a normal patient body sample (all normalized to Gl ^'SB) of greater than about 10 is indicative of the presence or increased likelihood of developing high-grade cervical disease. Cervical cytology samples, wherein the target transcript expression level has been normalized by the expression of GUSB and is expressed as a fold-change relative to a normal patient sample, having a fold-change of normalized HPV E7 copy number relative to the normalized HPV E7 copy number of a normal patient body sample above about 1 is indicative of the presence of cervical disease in the subject from which the cytology sample is derived or an increased likelihood of the subject developing cervical disease when compared Io a control subject. A fold-change of normalized HPV E7 copy number relative to the normalized HPV hi copy number of a normal patient body sample (all normalized to GUSR) of about I to about 4 (including, but not limited tø about 1, 1. 1, 1.2, 1.3, I A 1.5, l .(_\ 1.7, 1.8, 1.9, 2, 2.1, 2.2, 2.3, 2.4, 2.5, 2.6, 2.7, 2.S, 2.9. 3, 3. 1 , 3.2, 3.3, 3,4, 3,5, 3.(i, 3.7, 3.8, 3.^<\ and '■]) in a cervical cytology sample is indicative of the presence or increased likelihood of developing low-grade cervical disease, whereas a fold-change of normalized HPV E7 copy number relative to the normalized HPV E7 copy number of a normal patient body sample (all normalized to GUSB) of greater than about 4 is indicative of the presence or increased likelihood of developing high-grade cervical disease. For cervical cytology samples wherein the target transcript expression level has been normalized by the expression of POLR2A and is expressed as a fold-change relative to a normal patient sample, a fold-change of normalized HPV E6 copy number relative to the normalized HPV Kό copy number of a norma. I patient body sample, or the highest of a fold-change of normalized HPV E6 copy number relative to the normalized HPV E6 copy number of a normal patient body sample or a fold-change of normalized HPV E7 copy number relative to the normalized HPV E 7 copy number of a normal patient body sample above about 1 is indicative of me presence of cervical disease in the subject from which the cytology sample is derived or an increased likelihood of the subject developing cervical disease when compared to a control subject. A fold-change of normalized HPV E6 copy number relative to the normalized HPV E6 copy number of a normal patient body sample (ail normalized to POLR2A), or the highest of a fold-change of normalized HPV E6 copy number relative to the norma.il/ed 11 PV Vi? copy number of a normal patient body- sample (all normalized to POLR2A) or a fold-change of normalized HPV E7 copy number relative to the normalized 1 (PV E7 copy number of a normal patient body sample (all normalized to POLRzAj of about 1 to about 10 (including, but not limited to about L Ϊ.5. 2, 2.5, 3, 3.5, 4, 4.5, 5, 5.5, 6, 5,5, 7, 7.5, 8, 8.5, 9, °>.5, and 10) in a cervical cytology sample is indicative of the presence or increased likelihood of developing low-grade cervical disease, whereas a fold-change of normalized HPV E6 copy number relative Io the normalized HPV Eό copy number of a normal patient body sample (all normalized to POI R2A), or the highest of a fold-change of normalized HPV E6 copy number relative to the normalized 11 PV I-¹6 copy number of a normal patient body .sample (all normalized to P0LR2A) or a fold-change of normalized HPV E7 copy number relative to the normalized ) IPV E7 copy number of a normal patient body sample (all normalized to POLR2Λ ) of greater than about 10 is indicative of the presence or increased likelihood of developing high-grade cervical disease.

Cervical cytology samples, wherein the target transcript expression level has been normalized by the expression of POLR2A and is expressed as a fold-change relative to a normal patient sample, having a fold-change of normalized HPV E7 copy number relative to the normalized HPV E7 copy number of a normal patient body sample above about 1 is indicative of the presence of cervical disease in the subject from which the cytology sample is derived or an increased likelihood of the subject developing cervical disease when compared to a control subject. A fold-change of normalized HPV E7 copy number relative to the normalized HPV E7 copy number of a normal patient body sample Sail normalized to POLR2A) of about 1 Io abouf 5 (including, but not limited to about 1 , j ,1 ,

1.2, 1.3, 1.4, 1.5. 1.6. 1.7, 1.8, 1.9, 2. 2. 1. 2.2, 2.3, 2.4, 2.5, 2.b, 2.7, 2.8, 2.9, 3, 3.1, 3.2,

3.3, 3.4, 3.5, 3.6, 3.7, 3.8, 3.9, 4, 4.1 „ 4.2, 4.3, 4 A 4.5, 4,*\ 4,7, 4.H, 4», and 5} in a cervical cytology sample is indicative of the presence or increased likelihood of developing low-grade cervical disease, whereas a fold-change of normalized HPV E7 copy number relative to the normalized ! IPV (^■"^" copy number of a normal patient body sample (ail normalized to POLR2A) of greater than about 5 is indicative of the presence or increased likelihood of developing high-grade cervical disease.

The "delta-delta C_T" or "ΔΔC_T" is derived by subtracting an endogenous control C_T value for a particular sample from the target gene (e.g., HPV E6, E7) C_T, which is referred to as the delta C_T, and then the delta C_T of the sample is subtracted by the delta C_T of a normal patient sample (or an average of normal patient samples) to arrive at the delta- delta C_T. For this method, the number 2 was raised to the power of the negative delta- delta C_T and the data is expressed as the fold-change of each target gene relative to the normal patient body sample (or average of normal patient samples) to derive what is referred to herein as the "comparative C_T value". In some embodiments, the delta C_T of the normal patient sample is actually an average delta C_T derived from a population of normal patient samples. The average delta C_T can be derived from a population of normal patients samples of various sizes, including but not limited to, 2, 5, 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 200, 300, 400, 500, 600, 700, 800, 900, 1000, 2000, 3000, 4000, 5000, 6000, 7000, 8000, 9000, 10,000 or greater. The larger the population, the more precise the average delta C_T is expected to be.

Any of the suitable endogenous controls described heroin can be used to derive the delta-delta C_T and comparative C^'χ value, including but not limited to, GUSB and POLR2A.

For cervical cytology samples, a HPV E6 comparative C_T value (relative to GUSB). a HPV E7 comparative C_T value (relative to GIJSB), or the highest of a HPV E6 comparative C_T value Crelah^'ve to GUSB) or a HPV E7 comparative C_T value (relative to GUSB) above about 2 is indicative of the presence of cervical disease in the subject from which the cytology sample is derived or an increased likelihood of the subject developing cervical disease when compared to a control subject. A HPV E6 comparative C_T value (relative to GUSB), a HPV E7 comparative C_T value (relative to GUSB), or the highest of a HPV E6 comparative C_T value ( relative to GU S B) or a HPV E7 comparative C_T value (relative to GUSB) of about 2 to about 5 (including, but not limited to about 2, 2.L 2.2, 2,3, 2,4, 2.5, 2.6, 2.7, 2.8, 2.9, 3, 3.1 , 3.2, 3.3, 3.4, 3.5, 3.6, 3.7, 3.8, 3.9. 4, 4. 1, 4.2, 4.3,

4.4, 4.5, 4.6, 4.7. 4.S. 4.9, and 5 ) in a cervical cytology sample is indicative of the presence or increased likelihood of developing low-grade cervical disease, Λvhereas a WPV E6 comparative C_T value (relative to CHJSB), a HPV E7 comparative C_T value (relative to GUSBj, or the highest of a HPV E6 comparative C_x value (relative to GUSB) or a HPV E7 comparative C_x value (relative to CJUSB) of greater than about 5 is indicative of the presence or increased likelihood of developing high-grade cervical disease.

For cervical biopsy samples, a HPV E6 comparative C_T value (relative to GUSB), or the highest of a HPV Eb comparative C_x value (relative to GUSB) or HPV E7 comparative C_T value (relative to GUSB) above about 2 is indicative of the presence of cervical disease in the subject from which the biopsy sample is derived or an increased likelihood of the subject developing cervical disease when compared to a control subject. A HPV E6 comparative C_T value (relative to C]USB), or the highest of a HPV E6 comparative C_T value (relative to GUSB) or HPV E7 comparative C_T value (relative to CJUSB) of about 2 to about 15 (including, but not limited to about 2, 2.5, 3, 3.5, 4, 4.5, 5, 5.5, 6. 6.5, 7, 7.5, 8, 8.5, 9, 9.5, 10, 10.5, 11, 11.5, 12, 12.5, 13, 13.5, 14, 1-1.5, and 15 ) in a cervical biopsy sample is indicative of the presence or increased likelihood of developing low-grade cervical disease, whereas a HPV E6 comparative C_T value {relative to GUSB). or the highest of a WPV E6 comparative C_T value (relative to GUSB) or HPV E7 comparative C_T value (relative to GUSB) of greater than about 15 is indicative of the presence or increased likelihood of developing high-grade cervical disease.

Cervical biopsy samples having a HPV E7 comparative C_T value (relative to GUSB) above about 2 is indicative of the presence of cervical disease in the subject from which the biopsy sample is derived or an increased likelihood of the subject developing cervical disease when compared to a control subject. A HPV E7 comparative C_T value

(relative Io GIJSB) of about 1 to about K) (including, but not limited to about 2, 2.5, 3. 3.5, 4, 4.5, 5, 5.5, 6, 6.5, 7, ^~ .5. 8, 8.5, 9, 9.5, and 10) in a cervical biopsy sample is indicative of the presence or increased likelihood of developing low-grade cervical disease, whereas a HPV £7 comparative C_T value i relative to GUSB) of greater than about 10 is indicative of the presence or increased likelihood of developing high-grade cervical disease.

For cervical cytology samples, a HPV E6 comparative C_T value (relative to POLR2A) above about 2 is indicative of the presence of cervical disease in the subject from which the cytology sample is derived or an increased likelihood of the subject developing cervical disease when compared to a control subject. A HPV E6 comparative C_T value (relative to POLR2A) of about 2 io about 5 (including, but not limited to about 2.

2. 1 , 2.2, 2.3, 2.4. 2.5. 2.6, 2.7, 2.8, 2.9, 3. 3.1, 3.2, 3.3, 3.4, 3.5, 3.Cx 3.7, 3.8, 3.9, 4, 4.1,

4.2, 4.3, 4.4, 4.5, 4.6, 4.7, 4.8, 4.9, and 5) in a cervical cytology sample is indicative of the presence or increased likelihood of developing low-grade cervical disease, whereas a HPV E6 comparative C_T value (relative to PQLR2Aj of greater than about 5 is indicative of the presence or increased likelihood of developing high-grade cervical disease.

Cervical cytology samples having a HPV E7 comparative C_T value (relative to POLR2A), or the highest of a HPV E6 comparative C_x value (relative to POLR2A) or a HPV E7 comparative C_T value (relative to POLR2A ) above about 2 is indicative of the presence of cervical disease in the subject from which the cytology sample is derived or an increased likelihood of the subject developing cervical disease when compared to a control subject, A HPV E7 comparative C_x value (relative to POLR2A), or the highest of a HPV E6 comparative C_T value (relative to POLR2A) or a HPV XiI comparative C_T value (relative to POLR2A) of about 2 to about 10 (including, but not limited to about 2, 2.5, 3, 3,5, 4, 4.5, 5, 5.5, 6, 6.5, 7, 7.5, 8, 8,5, 9, 9.5, and 10) in a cervical cytology sample is indicative of the presence or increased likelihood of developing low-grade cervical disea.se, whereas a ) (PV E7 comparative C_T value (relative to POLR2Λ ), or the highest of a HPV Eb comparative C_T value (relative to POLR2Λ) or a HPV hi comparative C_T value (relative to PQ IJR 2 A) of greater than about 10 is indicative of the presence or increased likelihood of developing high-grade cervical disease.

For cervical biopsy samples, a HPV E6 comparative C_T value (relative to POI R2A,s, or the highest of a ) (PV Eό comparative C_T value (relative to P< )LR2Λ ) or HPV E 7 comparative C_T value (relative to P0LR2A) above about 2 is indicative of the presence of cervical disease in the subject from which the biopsy sample is derived or an increased likelihood of the subject developing cervical disease when compared to a control subject. A HPV E6 comparative C_T value (relative to POI R 2A), or the highest of a I IPV iό comparative C_T value (relative to POLR2A) or HPV E7 comparative C_T value (relative to POI .R2A) of about 2 to about 20 (including, but not limited to about 2, 2,5, 3. 3.5, 4, 4.5, 5. 5.5, 6, 6.5, 7, 7.5, 8, 8.5, 9, 9.5, 10. 10.5, 1 1, 11.5, 12, 12.5. 13, 13.5, 14, 14.5, 15, 15.5, 16, lti.5, 17, 17.5, 18, 18.5, 19, 19.5, and 2Oj in a cervical biopsy sample is indicative of the presence or increased likelihood of developing low-grade cervical disease, whereas a HPV E6 comparative C_T value (relative to PGLR2A), or the highest of a I IPV M) comparative C_T value ( relative to PCX R2A) or 11 PV V ^~ comparative C_T value (relative to POLR2A) of greater than about 20 is indicative of the presence or increased likelihood of developing high-grade cervical disease.

Cervical biopsy samples having a HPV E7 comparative C_T value (relative to POLR2A) above about 2 is indicative of the presence of cervical disease in the subject from which the biopsy sample is derived or an increased likelihood of the subject developing cervical disease when compared to a control stsbject. A HPV E7 comparative C_T value (relative to POLR2A) of about 2 to about 15 (including, but not limited to about 2, 2.5, 3, 3.5, 4, 4.5, 5, 5.5. 6, 6.5, 7, 7.5, 8, 8.5, 9, 9.5, 10, 10.5, 11, 11.5, 12, 12.5, 13, 13.5, 14, 14.5, and 15) in a cervical biopsy sample is indicative of the presence or increased likelihood of developing low-grade cervical disease, whereas a HPV E7 comparative C_T value (relative to POI R2A) of greater than about 15 is indicative of fhe presence or increased likelihood of developing high-grade cervical disease.

As the methods of the invention are capable of distinguishing normal from diseased tissue and low-grade from high-grade disease, the methods can be used during surgical procedures or following a surgical procedure to determine the boundaries of a tumor that is to be or has been resected. Kits for practicing the methods of the invention are further provided. By "kit" is intended any manufacture (e.g., a package or a container) comprising at least one reagent, e.g., nucleic acid primer/probe set, etc. for specifically detecting the expression of a high- risk HPV E6 or E7. In some embodiments, the kit comprises nucleic acid primer pairs and probe sets for the measurement of HPV E6 expression levels in HPV 16, 18, 31, 33, or 45 virus type. In some of these embodiments, the primers for the amplification of HPV 16 E6 detect unspliced HPV 16 E6 and have the sequences set forth in SEQ ID NOs: 1 and 2 and the probe for the detection of the amplification product has the sequence set forth in SEQ ID NO: 3. In other embodiments, HPV 16 E6 primers amplify HPV 16 E6 unspliced + I and have the sequences set forth in SEQ ID NOs: 7 and 8 and the detection probe has the sequence set forth in SEQ ID NO: 9. In still other embodiments, the kit comprises primers and a probe that can amplify and detect all HPV 16 E6 transcripts. In these embodiments, the primers can have the nucleotide sequences set forth in SEQ ID NOs: 1 and 2 and the probe has the sequence set forth in SEQ ID NO: 3. A kit for the detection of HPV 18 E6 can have primers with the sequences set forth in SEQ ID NOs: 13 and 14 and a probe sequence of SEQ ID NO: 15. A kit for the detection of HPV 18 E7 can have primers with the sequences set forth in SEQ ID NOs: 16 and 17 and a probe sequence of SEQ ID NO: 18. A kit for the detection of HPV31 E6 can have primers with the sequences set forth in SEQ ID NOs: 19 and 20 and a probe sequence of SEQ ID N0:21. A kit for the detection of HPV33 E6 can have primers with the sequences set forth in SEQ ID NOs: 22 and 23 and a probe sequence of SEQ ID NO:24. A kit for the detection of HPV45 E6 can have primers with the sequences set forth in SEQ ID NOs: 25 and 26 and a probe sequence of SEQ ID NO:27.

The kit may comprise primers and corresponding probes for the detection of an endogenous control gene, including any of the suitable endogenous control genes disclosed elsewhere herein (e.g., GUSB, P0LR2A). In some embodiments, the kit may comprise a negative control sample, such as a normal patient body sample. In particular embodiments, the kit may comprise literature reciting the E6/E7 mRNA threshold values (expressed as C_τ values, absolute copy numbers, normalized target values, fold-change of the normalized target value over the normalized target value of normal samples, or as a comparative C_T value) that can be used to compare experimental sample E6/E7 mRNA expression levels against to arrive at a diagnosis.

The kit may be promoted, distributed, or sold as a unit for performing the methods of the present invention. Additionally, the kits may contain a package insert describing the kit and methods for its use. Any or all of the kit reagents may be provided within containers that protect them from the external environment, such as in sealed containers or pouches. Positive and/or negative controls may be included in the kits to validate the activity and correct usage of reagents employed in accordance with the invention. One of skill in the art will appreciate that any or all steps in the methods of the invention could be implemented by personnel or, alternatively, performed in an automated fashion. Thus, the steps of body sample preparation and detection of HPV E6, HPV E7, and/or endogenous control gene expression (e.g., by TaqMan analysis®) may be automated. It is to be noted that the term "a" or "an" entity refers to one or more of that entity; for example, "a virus" is understood to represent one or more viruses. As such, the terms "a" (or "an"), "one or more," and "at least one" can be used interchangeably herein. Throughout this specification and the embodiments, the words "comprise," "comprises," and "comprising" are used in a non-exclusive sense, except where the context requires otherwise.

As used herein, the term "about," when referring to a value is meant to encompass variations of, in some embodiments ± 50%, in some embodiments ± 20%, in some embodiments ± 10%, in some embodiments ± 5%, in some embodiments ± 1%, in some embodiments ± 0.5%, and in some embodiments ± 0.1% from the specified amount, as such variations are appropriate to perform the disclosed methods or employ the disclosed compositions.

Further, when an amount, concentration, or other value or parameter is given as either a range, preferred range, or a list of upper preferable values and lower preferable values, this is to be understood as specifically disclosing all ranges formed from any pair of any upper range limit or preferred value and any lower range limit or preferred value, regardless of whether ranges are separately disclosed. Where a range of numerical values is recited herein, unless otherwise stated, the range is intended to include the endpoints thereof, and all integers and fractions within the range. It is not intended that the scope of the presently disclosed subject matter be limited to the specific values recited when defining a range.

Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of skill in the art to which the invention pertains. Although any methods and materials similar herein can be used in the practice or testing of the present invention, the preferred methods and materials are described herein. The following examples are offered by way of illustration and not by way of limitation:

EXPERIMENTAL

Example 1 : Analysis of housekeeping genes.

In order to develop a robust assay to evaluate the expression of various HPV genes, effective endogenous control genes for RNA expression in the target tissues (i.e., cervical cytology and biopsy samples) needed to be identified. The expression level of sixteen housekeeping genes in a representative set of cervical cytology and biopsy specimens was evaluated to determine which housekeeping genes would be beneficial for the normalization of cervical cytology and biopsy tissues. The list of housekeeping genes that were evaluated is provided in Table 1.

Table 1. Housekeeping genes that were assessed in cervical cytology and biopsy expression analysis.

Three main criteria were used in the selection process. The endogenous control gene should not be significantly altered between disease states. That is, the expression level should be relatively constant between normal to cancerous states and within the range of these samples. The endogenous control gene should also serve as a measure of RNA quality. If a gene used for normalization is too abundant or more stable than the transcript of interest, it will not accurately normalize the transcript of interest. Finally, an endogenous control primer/probe set should be specific for RNA and should not amplify or detect DNA.

The threshold cycle (C_T) of each endogenous control gene for the biopsy samples across each diagnosis is shown in Figure 1. There were four individual biopsy samples for which most of the housekeeping genes (except 18S and POLR2A) have a high C_T signal, indicating a poor quality sample. Some of the housekeeping genes (e.g., 18S and POLR2A) are expressed in all of the biopsy samples. Other housekeeping genes (e.g., ACTB, GAPDH, HPRTl, TBP) have high C_x levels in a number of samples, whereas those samples express other housekeeping genes. This suggests that ACTB, GAPDH, HPRTl, TBP transcripts are not very abundant or less stable than others and would not serve as true indicators of the RNA quality of a sample. The ACTB, TBP, GAPDH, and HPRTl genes were eliminated as useful endogenous control genes for the E6/E7 quantitative assays for this reason and because the expression of these genes was variable between the samples and between and within the disease states. To test for the specificity of each endogenous control assay for RNA, human genomic DNA was also tested on the endogenous control TLDA card. It is important to determine if the assay detects genomic DNA considering the RNA for most quantitative HPVE6/E7 assays will be derived from fixed specimens that could be contaminated with DNA and HPV does not have introns. The assays designed for 18S, GAPDH, and PPIA all detected human genomic DNA at significant levels (Ct levels less than 37), whereas the other assays do not. Further analysis of these genes revealed potential difficulties in designing assays that can reliably distinguish RNA from DNA in these transcripts. The gene for 18S does not contain introns. Therefore, distinguishing 18S RNA from DNA is not possible without some other independent method to demonstrate the absence of genomic DNA within the sample. Williams et al. (2008) Cancer Letters 271 (2008) 81-84 demonstrated that many ACTB and GAPDH pseudogenes can be detected with trace amounts of DNA. In the case of GAPDH, a number of processed GAPDH pseudogenes are 90-95% homologous to the gene. These processed pseudogenes can be transcribed, which would further confuse the analysis of RNA levels (Benham et al. EMBO 3 (1984) 2635-2640, Liu et al. BMCGenomics 10 (2009) 480-491). Thus, 18S, ACTB, and

GAPDH were not chosen for use as an endogenous control. Further, PPIA was not chosen for the quantitative HPV E6 and E7 assays described in Example 2, as the Applied Biosystems® designed assay detected genomic DNA. This transcript could be used, however, to normalize HPV E6 and E7 levels with a properly designed assay that doesn't detect genomic DNA.

Several genes (GUSB, PPIA, YWHAZ, HMBS, RPLPO, IPO8, PGKl, POLR2A, TFRC, UBC, and B2M) fit the selection criteria and although any of these genes could serve as endogenous control genes for normalization of HPV E6 or E7 levels in cervical biopsy samples, only GUSB and POLR2A were chosen to serve as endogenous controls for the quantitative E6/E7 assays described in Example 2.

A similar strategy was employed for the cytology samples. The housekeeping genes 18S and ACTB were not considered as potential normalization genes for these samples for the reasons stated above. The C_T levels of each endogenous control gene for the cytology samples across each diagnosis are shown in Figure 2. The housekeeping genes HMBS, HPRTl, TBP, TFRC, IPO8, and YWHAZ were eliminated as useful endogenous control genes for the E6/E7 quantitative assays because these genes were expressed at low levels, as indicated by high C_T levels. Several genes (GUSB, PPIA, RPLPO, PGKl, POLR2A, UBC, and B2M) fit the selection criteria and although any of these genes could serve as endogenous controls for normalizing E6 and E7 levels in cervical cytology samples, only GUSB and POLR2A were chosen to serve as endogenous controls for the quantitative E6/E7 assays described in Example 2.

Methods Appropriate endogenous control genes were chosen for additional experiments by testing a representative set of biopsy and cytology samples using the Endogenous Control "Taqman® Low-Density Array" (TLDA) Card from Applied Biosystems® (Foster City, CA, USA). The TLDA Card can be used to assess the expression levels of the following 16 endogenous control genes: ACTB, HPRTl, TBP, GAPDH, 18S, PPIA, YWHAZ, GUSB, HMBS, RPLPO, IPO8, PGKl, POLR2A, TFRC, UBC, and B2M. The biopsy and cytology samples were prepared as described in Example 2. 120 ng of reverse-transcribed RNA was loaded into each card reservoir and run on an Applied Biosystems® 7900 realtime PCR instrument according to the manufacturer's directions. For the biopsy specimens, 22 normal (within normal limits; WNL), 3 normal adjacent (NAT), 5 cervicitis, 5 cervical intraepithelial neoplasia I (CINI), 1 CINII, 27 CINIII, 26 squamous cell carcinoma (SCC), and 1 adenocarcinoma sample were tested, for a total of 88 samples. For the cytology samples, 20 negative for intraepithelial lesion and malignancy (NILM), 22 low-grade squamous intraepithelial lesions (LSIL), and 22 high-grade squamous intraepithelial lesions (HSIL) samples were tested, for a total of 64 samples.

Example 2. Quantitative HPV E6/E7 real-time PCR assay. Biopsy Samples

Nintey-six formalin- fixed paraffin-embedded cervical biopsy samples were obtained from commercial sources and RNA was isolated therefrom. The samples comprised 6 within normal limits (WNL), 6 normal adjacent tissue (NAT), 31 CINI, 26 CINIII, and 27 squamous cell carcinoma samples. Threshold cycle determination

For each biopsy specimen, the threshold cycle (C_T) of HPV E6 mRNA molecules, corresponding to five high-risk HPV types (HPV 16, 18, 31, 33 and 45), was determined by real-time PCR amplification of reverse-transcribed RNA followed by analysis with a manual threshold of 0.2. The level of HPV 16 and HPV 18 E7 mRNA was also quantified in the same manner. C_T values of all the analyzed HPV E6 and E7 transcripts are presented in Table 2. The lowest C_T value of HPV E6, E7 or the combination of E6 or E7 mRNA determined for a given specimen is plotted in Figure 3A, 3B, or 3C, respectively, as a function of biopsy diagnosis. For a subset of the CIN III and SCC specimens that did not express E6 or E7 mRNA from one of these high risk virus types, reflex quantification was performed to test for expression of mRNA from additional high risk viruses present in the specimens as determined by HPV DNA detection and typing. Samples that had a GUSB C_T signal of 40 or greater were removed from the analysis. Signals having a standard deviation of greater than or equal to one were considered invalid. Unlike absolute quantification methods, C_T values can not account for the presence of contaminating DNA. Therefore, samples having C_T values in those reactions that were not reverse transcribed (no RT controls) that are equal to the C_T value of the experimental sample with RT were considered invalid and removed from the analysis.

In every CINIII and SCC biopsy sample that was positive for HPV E6 mRNA, the HPV E6 mRNA transcript that had the highest level of expression, as indicated by the lowest C_τ, corresponded to an HPV DNA type present in the specimen (open circles, Figure 3A). In addition, two of the six WNL specimens contained one or more high-risk HPV type, and had HPV E6 C_T values of less than 36. The other four WNL samples were negative for high risk HPV DNA and had HPV E6 C_T values of 39 or greater. Thirteen CIN I samples were positive for high-risk HPV DNA, but only four of them had HPV E6 C_T values of equal to or less than 37. One CIN I sample that was negative for high-risk HPV DNA had a HPV E6 C_x of 37.61, but all other CINI samples had a HPV E6 C_x of greater than 38 (closed circles, Figure 3A). Five of the six NAT samples had HPV E6 C_T values of less or equal to 35. The remaining NAT sample had a HPV E6 C_T value of 37.68.

In general, the data indicate that the C_T values for HPV E6 mRNA decreases with increasing severity of biopsy grade, with a mean C_T value of 34.88 for WNL specimens, 34.88 for NAT, 38.91 for CIN I, 28.14 for CIN III, and 24.40 for SCC. Threshold levels of HPV E6 C_T values that distinguish cervical disease grade were determined from these data to be: a C_T value of greater than 38 copies for normal samples, a C_T value of 35 to 38 for low grade cervical disease, and a C_T value of less than or equal to 34 for high-grade disease. These threshold levels were selected in order to maximize the diagnosis rate of CIN III and SCC while minimizing the false positive detection rate for normal tissue. None of the CIN III or SCC specimens were negative with this criteria. Two of the NATs were classified as high grade and four were low grade. Three CIN I samples were high grade, two were considered low grade, and the rest were normal. With this threshold, two of the WNL were considered low grade and the rest of these samples were normal.

Analysis of HPVE 7 mRNA for HPV 16 and 18 in these tissues revealed a similar trend as the HPV E6 data although the determined thresholds were different. As can be seen in Figure 3B, five of six normal (WNL) tissue samples had C_T values greater than or equal to 37 for E7. The remaining WNL sample had a C_T value of 36. Three of the six NAT had C_T values less than 35 while 3 had C_T values greater than 37 (Figure 3B). One of the 29 CIN I specimens had a C_T value of less than 35, five had C_T values ranging from 37 to 38, while the remaining samples had C_T values greater than 39. All of the HPV 16 or 18 DNA-positive CIN III and SCC samples expressed E7 mRNA. There was one CIN III specimen that had a C_T of 38. C_T values of E7 mRNA in CIN III and SCC samples ranged from 38-25.7 and 35-20, respectively. Mean HPV E7 CT values were 38.84 for WNL specimens, 35.82 for NAT, 39.44 for CIN I, 30.83 for CIN III, and 25.11 for SCC samples. Threshold levels of E7 C_T levels that distinguish disease states are as follows: a C_T value greater than 38 indicates normal cervix, a C_T value between 35 and 38 indicates low grade cervical disease, and a C_T less than 35 is indicative of high grade disease. Using this criteria, one CIN III would be considered normal.

Finally, an analysis of the lowest C_T value of either E6 or E7 mRNA was performed to determine thresholds for the detection of disease. Mean levels of the C_T values for the various diseases were as follows: 38.48 for WNL, 35.18 for NAT, 38.80 for CIN I, 28.12 for CIN III and 24.23 for SCC. Using a threshold C_x value of less than 35 corresponding to high-grade disease, a C_T value ranging between 35 and 38 indicating low grade disease, and a C_T value greater than 38 corresponding to normal cervix, all of the CIN III and cancer specimens were detected. Three CIN I samples and three NATs were classified as high-grade using these threshold levels, while two WNLs 3 CIN I, and 3 NATs were classified as low-grade. The other samples were considered normal tissue by this criteria. These results indicate that E6 and E7 C_T levels decrease with the progression of cervical disease and that with the use of defined threshold levels, one can distinguish diseased tissue from normal and grades of the disease.

Absolute Copy Determination

For each biopsy specimen, the absolute quantity of HPV E6 mRNA molecules, corresponding to five high-risk HPV types (HPV 16, 18, 31, 33 and 45), was determined by real-time PCR amplification of reverse-transcribed RNA followed by extrapolation of copy number from a standard curve. The level of HPV 16 and HPV 18 E7 mRNA was also quantified in the same manner. The absolute copy numbers of all the analyzed HPV E6 and E7 transcripts are provided in Table 3. The highest absolute copy number of HPV E6, E7, or the combination of E6 or E7 mRNA determined for a given specimen is plotted in Figures 4A, 4B, and 4C, respectively, as a function of biopsy diagnosis. For a subset of the CIN III and SCC specimens that did not express E6 or E7 mRNA from one of these high-risk virus types, reflex quantification was performed to test for expression of mRNA from additional high risk viruses present in the specimens as determined by HPV DNA detection and typing. Samples that had a GUSB C_T signal of 40 or greater were removed from the analysis. In every CINIII and SCC biopsy sample that was positive for HPV E6 mRNA, the

HPV E6 transcript with the highest level of expression corresponded to an HPV DNA type present in the specimen (open circles, Figure 4). In addition, two of the six WNL specimens contained one or more high-risk HPV type, and at least one copy of a corresponding type of HPV E6 mRNA. Thirteen CIN I samples were positive for high-risk HPV DNA, but neither HPV E6 mRNA could be detected at a threshold of >1 copy in 50 ng reverse-transcribed RNA (Fig. 4A) in seven of these samples. Four within normal limits (WNL) and 16 CIN I specimens were negative for high-risk HPV DNA (closed circles, Fig. 4). Six of sixteen of the HPV DNA-negative CIN I samples had E6 mRNA copy values of >1, but less than 4. Four of the six samples comprising normal adjacent tissue (NAT) to a tumor contained high-risk HPV DNA. Five of the six NAT samples had E6 mRNA levels of greater than 10 copies, while four of the six had greater than 30 copies of E6 mRNA (Figure 4A, Table 3).

In general, the data indicate that HPVE6 mRNA copy number increases with increasing severity of biopsy grade, with a mean mRNA copy number of 2.7 for WNL specimens, 43.61 for NAT, 1.9 for CIN I, 6367 for CIN III, and 96703 for SCC. Threshold levels of HPV E6 mRNA copy number that distinguishes cervical disease grade were determined from these data to be <10 copies for normal cervix, 10-50 copies for low grade cervical disease, and >50 copies for high grade disease (Figure 4). These threshold levels were selected to maximize the diagnosis of CIN III and SCC while minimizing the false positive detection rate for normal tissue. None of the CIN III or SCC specimens were negative with this criteria. Three of the NATs were classified as high-grade, two were low-grade, and one was normal. With these threshold levels, one WNL and one CIN I sample were considered low-grade and the rest of these samples were normal. Analysis of HPVE 7 mRNA for HPV 16 and 18 in these tissues revealed a similar trend as the HPV E6 data, although the determined thresholds were different. As can be seen in Figure 4B, all normal tissues had less than 10 copies of E7, with most below 3 copies, and one at 7.6 copies of E7 mRNA. All six NAT samples had at least 1 copy of E7 mRNA, three had values over 10, while two had greater than 30 copies. 11 of the 29 CIN I specimens expressed at least one copy of E7 mRNA, although 8 of these did not contain the corresponding DNA virus. Two of the CIN I tissues had greater than 5 copies of E7 mRNA, and one had 36 copies. All but one of the HPV 16 or 18 DNA-positive CIN III and SCC samples expressed E7 mRNA. CIN III and SCC specimens that did not contain HPV 16 or 18 E7 mRNA were infected with and expressing a different HPV virus (Fig. 4B). Excluding the one negative CIN III sample, the range of expression of E7 mRNA in CIN III and SCC samples were 19.52-6093 and 13-168096 copies, respectively. Mean mRNA copy numbers were 2.4 for WNL specimens, 15.4 for NAT, 2.6 for CIN I, 1088 for CIN III, and 46220 for SCC. Threshold levels were set at <1 copies for normal cervix, 1-10 copies for low grade cervical disease and >10 copies for high grade disease. Using these criteria, one CIN III sample would be considered false negatives, although this specimen did contain E6 mRNA.

Finally, an analysis of the highest copy number of either E6 or E7 mRNA was performed to determine thresholds for the detection of disease. Mean levels of the highest transcript levels were 10.19 for WNL, 45.66 for NAT, 15.52 for CIN I, 6371 for CIN III and 98966 for SCC. Using a threshold of greater than 50 copies for high grade disease, all of the CIN III and cancer specimens were detected. Two CIN I samples and three NAT samples were also classified as high-grade while two WNLs and 3 NATs were classified as low-grade. It is notable that, although the pathologist classified NAT as normal by morphological criteria, tissues were positive for expression of HPVE6 and E7 mRNA. The other samples were considered normal tissue by these criteria. These results indicate that E6 and E7 copy number increase with the progression of cervical disease and that absolute quantification of E6 and E7 mRNA copy numbers can distinguish diseased tissue from normal and grades of disease. Table 3. Absolute copy numbers of HPV E6 and E7 in cervical biopsy samples.

Ul K>

Ul

Ul

4-

Normalized target value determination GUSB

For a subset of biopsy specimens, the absolute copy number of HPV E6 mRNA molecules, corresponding to five high-risk HPV types (HPV 16, 18, 31, 33 and 45), was determined by real-time PCR amplification of reverse-transcribed RNA followed by extrapolation of copy number from a standard curve. The levels of HPV 16 and HPV 18 E7 mRNA were also quantified in the same manner. The absolute value of GUSB was determined in an identical fashion from a separate standard curve for GUSB. The copy numbers determined for HPV E6 or E7 were normalized to the copies for the endogenous control gene (see methods) to produce a normalized target value (NTV). This value provides a copy number that is adjusted to the level of the endogenous control gene, enabling normalization of RNA quality and quantity. The NTV values of all the analyzed HPV E6 and E7 transcripts are provided in Table 4. The highest normalized target value of HPV E6, E7, or the combination of E6 and E7 mRNA determined for a given specimen is plotted in Figures 5A, 5B, and 5C, respectively, as a function of biopsy diagnosis. For a subset of the CIN III and SCC specimens that did not express E6 or E7 mRNA from one of these high-risk virus types, reflex quantification was performed to test for expression of mRNA from additional high risk viruses present in the specimens as determined by HPV DNA detection and typing. Samples that had a GUSB C_T signal of 40 or greater were removed from the analysis.

In every CIN III and SCC biopsy sample that was positive for HPV E6 mRNA, the HPV E6 transcript with the highest level of expression, normalized to GUSB, corresponded to an HPV DNA type present in the specimen (open circles, Figure 5). All of the CIN III and SCC samples tested in this analysis were positive for E6 expression, with an NTV value ranging from 27 to over 46000 and 197 to over 12000 for CIN III and SCC, respectively. Four of twelve CIN I samples had a normalized value of greater than one, two of the samples had a normalized target value of 2.33 and 4.95, respectively. One of the two NAT samples tested in this manner, had a NTV of 44 and was HPV DNA- positive. In general, the data indicate that HPVE6 NTV levels increase with increasing severity of biopsy grade, with a median normalized target value of 22 for NAT, 0.98 for CIN I, 5181 for CIN III, and 3929 for SCC. Threshold levels of HPV E6 mRNA NTV that distinguish various cervical disease grades were determined from these data to be less than 1 NTV for normal, between 1-20 for low-grade cervical disease, and greater than 20 for high-grade disease (Figure 5A). These threshold levels were selected to maximize the diagnosis of CIN III and SCC, while minimizing the false positive detection rate for normal tissue. None of the CIN III or SCC specimens were negative with these criteria. One NAT sample was classified as high-grade and one was normal. With this threshold, four CIN I samples were considered low-grade and the rest of these samples were normal.

Analysis of HPVE 7 mRNA for HPV 16 and 18 in these tissues revealed a similar trend as the HPV E6 data, although the determined thresholds were different. As can be seen in Figure 5B, one NAT sample had a NTV of 19.13, while another was less than 0.1. Two of the 7 CIN I specimens had a NTV of greater than 1, while the others were less than 1. AU of the HPV 16 or 18 DNA-positive CIN III and SCC samples expressed E7 mRNA. CIN III and SCC specimens that did not contain HPV 16 or 18 E7 mRNA were infected with and expressing a different HPV virus (Table 4). The range of normalized target values for E7 mRNA in CIN III and SCC samples was 23-740 and 6.33 to almost 6000, respectively. Mean HPV E7 normalized values were 9.6 for NAT, 0.51 for CIN I, 231 for CIN III, and 1669 for SCC. Threshold levels were set at less than 0.1 NTV for normal cervix, 0.1-5 for low-grade cervical disease, and greater than 5 for high-grade disease. Using these criteria, no CIN III or SCC samples were false negative, and one NAT sample would be detected as having high grade disease.

Finally, an analysis of the highest NTV value of either E6 or E7 mRNA was performed to determine threshold levels for the detection of disease. Mean levels of the highest transcript normalized value were 22.28 for NAT, 9.0 for CIN I, 5194 for CIN III, and 4048 for SCC specimens (Figure 5C). Using a threshold of greater than 20 NTV for high-grade disease, all of the CIN III and cancer specimens were detected. One CIN I and one NAT specimen were also classified as high-grade. A threshold between 1-20 NTV units defined low-grade disease, which identified five CIN I samples. A threshold lower than 1 defined normal samples, which identified five CIN I samples and one NAT sample. These results indicate that E6 and E7 NTV values increase with high-grade disease and the determined threshold levels can distinguish diseased tissue from normal and grades of disease.

POLR2A

For a subset of biopsy specimens, the normalized target value (NTV) of HPV E6 mRNA molecules, corresponding to five high-risk HPV types (HPV 16, 18, 31, 33 and 45), was determined by real-time PCR amplification of reverse-transcribed RNA followed by extrapolation of copy number from a standard curve. The level of HPV 16 and HPV 18 E7 mRNA were also quantified in the same manner. The absolute value of POLR2A was determined in an identical fashion from a separate standard curve for POLR2A. The copies determined for HPV E6 or E7 were normalized to the copies for the particular endogenous control gene (see methods) to produce a normalized target value (NTV). This value provides a copy number that is adjusted to the level of the endogenous control gene, enabling normalization of RNA quality and quantity. The highest normalized target value of HPV E6, E7, or the combination of E6 and E7 mRNA determined for a given specimen is plotted in Figures 6 A, 6B, and 6C as a function of biopsy diagnosis, respectively. For a subset of the CIN III and SCC specimens that did not express E6 or E7 mRNA from one of these high-risk virus types, reflex quantification was performed to test for expression of mRNA from additional high-risk viruses present in the specimens as determined by HPV DNA detection and typing. Samples that had a POLR2A Ct signal of 40 or greater were removed from the analysis. In every CIN III and SCC biopsy sample that was positive for HPV E6 mRNA, the

HPV E6 transcript with the highest level of expression, normalized to POLR2A, corresponded to an HPV DNA type present in the specimen (figure 6A). All of the CIN III and SCC samples tested in this analysis were positive for E6 expression, with an NTV ranging from 11 to over 14482 and 153 to over 21010 for CIN III and SCC, respectively. Two of eleven CIN I samples had a normalized value of greater than one, with the highest of the two having a value of 1.38. One of the two NAT samples analyzed in this manner had a NTV of 4.55 and was HPV DNA positive for the same virus type as the expression.

In general, the data indicate that HPVE6 NTV levels increase with increasing severity of biopsy grade, with a median normalized target value of 2.28 for NAT, 0.34 for CIN I, 1851 for CIN III, and 4659 for SCC. Threshold levels of HPV E6 mRNA NTV that distinguish various cervical disease grades were determined from these data to be less than 1 NTV for normal, between 1-10 for low-grade cervical disease and greater than 10 for high-grade disease. These threshold levels were selected to maximize the diagnosis of CIN III and SCC, while minimizing the false positive detection rate for normal tissue. None of the CIN III or SCC specimens were negative with these criteria. One NAT was classified as low-grade and one was normal. With this threshold, two CIN I samples were considered low-grade and the rest of these samples were normal.

Analysis of HPVE 7 mRNA for HPV 16 and 18 in these tissues revealed a similar trend as the HPV E6 data, although the determined thresholds were different. As can be seen in Figure 6B, one NAT sample had at NTV of 1.98, while the other had a value ten times less with a value of 0.12. Every CIN I specimen had a NTV of less than 1. All of the HPV 16 or 18 DNA positive CIN III and SCC samples expressed E7 mRNA to varying degrees. CIN III and SCC specimens that did not contain HPV16 or 18 E7 mRNA were infected with and expressing a different HPV virus (Table 4). The range of the normalized target values for E7 mRNA in CIN III and SCC samples was 15.77 to 412.97 and 2.43 to 7315.46, respectively. Mean HPV E7 normalized values were 1.05 for NAT, 0.19 for CIN I, 131.8 for CIN III, and 2073 for SCC. Threshold levels were set at less than 1 NTV for normal cervix, 1-10 for low-grade cervical disease and greater than 10 for high-grade disease. Using these criteria, one SCC samples was misclassified as low grade. One NAT was classified as low-grade disease with this particular threshold, but no CIN I in this category.

Analysis of the highest NTV of either E6 or E7 mRNA was performed to determine threshold levels for the detection of disease. Mean levels of the highest transcript normalized value were 2.6 for NAT, 3.65 for CIN I, 1855 for CIN III, and 4746 for SCC specimens (Figure 6C). Using a threshold of greater than 10 NTV for high grade disease, all of the CIN III and cancer specimens were detected. One CIN I specimen was classified as high-grade. A threshold between 1-10 NTV units defined low-grade disease, which identified two CIN I samples and one NAT sample. A threshold lower than 1 defined normal samples, which identified the 9 remaining CIN I samples and one NAT sample. These results indicate that E6 and E7 NTV increases with high-grade disease and the determined threshold levels can distinguish diseased tissue from normal and grades of disease.

Table 4. Normalized target values of HPV E6 and E7 in cervical biopsy samples.

o

Comparative C_T

For each biopsy specimen, the relative quantities of HPV E6 mRNA molecules, corresponding to five high-risk HPV types (HPV 16, 18, 31, 33 and 45), were determined using the Comparative CT method of mRNA quantification. Relative levels of HPV 16 and 18 E7 mRNA were also determined in the same manner. Transcript levels were normalized to the endogenous control transcript of either GUSB or POLR2A for each specimen, and then the normalized values were calibrated relative to the average expression of the normal (WNL) specimens. The comparative C_T values of all the analyzed HPV E6 and E7 transcripts, normalized to GUSB or POLR2A, are provided in Tables 5 and 6, respectively. Absolute quantification methods allow for the discrimination between absolute signal due to mRNA amplification and artifactual signal resulting from amplification of contaminating genomic DNA by subtraction of the absolute value of nucleic acid quantified in the "minus-RT" control reaction. The comparative CT method of relative quantification is not readily amenable to this type of compensation. Therefore, samples producing equivalent CT values in the experimental reaction and the "minus-RT' control reaction, were considered unacceptable and removed from the analysis. Additionally, relative expression values less than 2-fold above the average WNL expression levels were considered invalid and were assigned a value of zero. The fold over-expression of HPV E6 or E7 mRNA relative to the mean WNL expression for a given specimen is plotted in Figures 7 and 8 as a function of biopsy grade.

GUSB

E6 mRNA only

Relative expression values obtained for the HPV E6 transcript alone are plotted in Figure 7A. Selection criteria for expression values were established as described above. All CIN III and SCC specimens were determined to contain HPV E6 mRNA at levels > 15 -fold higher than the average expression of the WNL's. The NAT category had four specimens that had high risk HPV DNA and expressed HPVE6 mRNA above two fold of the normal tissues, the remaining samples were negative for expression of E6 mRNA. Three samples with a CIN I designation, containing high risk HPV DNA, were above 15 fold greater than the average of the WNL tissues for HPVE6 expression (Figure 7A). Two CIN I samples expressed E6 transcripts at 2-15 fold while the remaining 24 had normal levels of E6 message. Two of the WNL samples had values over ten fold of the normal tissues while the rest were similar to WNL. The relative HPV E6 expression amounts for the biopsy categories increased with increasing severity of diagnosis with mean fold over-expression values of 8.25 for WNL, 11.11 for NAT, 41.50 for CIN I, 301.20 for CIN III, and 2845 for SCC. Threshold relative expression values for HPV E6 mRNA indicative of high grade cervical disease were determined from these data by finding a cut-off point that included the highest number of CIN III and SCC samples in the category (>15). Low grade disease was determined by excluding the majority of high grade samples from CIN I (2- 15-fold). A cut-off for the absence of cervical disease was assigned by determining the threshold below which a maximum number of WNL specimens fell with minimal inclusion of CIN I samples (<2- fold). Every CIN III and cancer sample is expressing greater than 15 fold more E6 mRNA than the average of the normal tissues. Three of the CIN I biopsy tissues have E6 mRNA levels greater than 15 fold to the normal tissues and thus would be categorized as high grade. These three samples had the corresponding HR-HPV DNA genotype. Two additional CIN I samples express E6 mRNA at moderate levels (between 2 and 15 fold) and were classified as low grade. The other CIN I samples had E6 mRNA amounts similar to WNL and were considered to be normal. One HPV DNA positive NAT and two WNL tissues had E6 transcript levels greater than 15 fold to normal tissue and are considered high grade samples. One normal tissue and three NAT had E6 message levels greater than five fold relative to the average of the normal tissues. These samples would be low grade dysplasia. The remaining samples for both categories were expressing normal levels of E6 mRNA.

E7 mRNA only

HPV 16 and 18 E7 mRNA was assessed in biopsy samples and analyzed with the comparative Ct method using GUSB as the housekeeping gene for normalization.

Relative expression values obtained for the HPV E7 transcript alone are plotted in Figure 7B. Selection criteria for expression values were established as described above. All but two CIN III and one SCC specimen were determined to contain HPV E7 mRNA at levels > 10-fold higher than the average expression of the WNL 's (Figure 7B). Although these three high-grade specimens did not express above normal levels of E7 mRNA, they did express very high amounts of E6 message (figure 7A). The NAT category had four specimens that contained high risk HPV DNA and expressed HPVE7 mRNA above two fold of the normal tissues; the remaining samples were negative for expression of E7 mRNA. Two samples with a CIN I designation, containing high risk HPV DNA, were above two fold greater than the average of the WNL tissues for HPVE7 expression. The remaining 27 had normal amounts of E7 message. Two of the WNL samples had values over five fold of the normal tissues while the rest were within the levels for normal tissues.

The relative HPV E7 expression levels for the biopsy categories increased with increasing severity of diagnosis with mean fold over-expression values of 6.017 for WNL, 6.3 for NAT, 0.83 for CIN I, 176 for CIN III, and 2130 for SCC. Threshold relative expression values for HPV E6 mRNA indicative of high grade cervical disease were determined from these data by finding a cut-off point that included the highest number of CIN III and SCC samples in the category (>10). Low grade disease was determined by excluding the majority of high grade samples from CIN I (2-10 fold). A cut-off for the absence of cervical disease was assigned by determining the threshold below which a maximum number of WNL specimens fell with minimal inclusion of CIN I samples (<2- fold). Using these criteria, the majority of CIN III and cancer samples were classified as high-grade. Three of these samples would be classified as normal with these E7 limits but would be high grade with the inclusion of HPVE6 mRNA. There were 6 cancers and 1 CIN III samples containing higher E7 than E6 mRNA. This observation did not alter the classifications of the samples but it illustrates the importance of all these transcripts. One CIN I would be classified as high grade, one as low grade and the remaining samples as normal tissue. Two of the NATs are categorized as high grade, two as low grade, and the rest have values similar to normal tissue. Finally, the two HPV DNA positive WNL samples are expressing greater than five fold levels of E7 mRNA, one would be high grade and the other low grade. The remaining samples are expressing levels similar to normal tissue.

Combined E6 and E7 mRNA

Each CIN III or SCC biopsy specimen was determined to contain at least 1 high- risk HPV DNA type. All CIN III and SCC specimens were also determined to contain a concordant high-risk type HPV E6 or E7 mRNA at levels > 20-fold higher than the average expression of the WNL' s (Figure 7C). There were no high grade samples negative for expression of E6 or E7 mRNA. Four HPV DNA positive NAT samples had HPV E6 or E7 message amounts greater than 5 fold relative to normal tissue. Two of the hrHPV DNA negative NAT specimens had amounts similar to normal tissue (Figure 7C). In the CIN I group, 6 of 29 specimens displayed 2-fold or better over-expression of a corresponding HPV E6/E7 mRNA. Five out of the six CIN I samples that had greater than two fold expression were also positive for the same hrHPV DNA type, one was hrHPV DNA negative. 23 additional CIN I samples did not contain elevated relative levels of HPV E6 or E 7 mRNA. Two of the WNL biopsy cases were hrHPV DNA-positive and these had levels of E6 or E7 mRNA of the same viral type of greater than 15 fold. The remaining WNL samples were expressing HPV mRNA levels at less than 2 fold.

The relative expression levels for the biopsy categories increased with increasing severity of diagnosis with mean fold over-expression values of 8.25 for WNL, 11.35 for NAT, 41.81 for CIN I, 303.5 for CIN III, and 3230 for SCC specimens. Threshold relative expression levels for HPV E6/E7 mRNA indicative of high grade cervical disease were determined from these data by finding a cut-off point that included the highest number of CIN III samples in the category while excluding the majority of CIN I and WNL specimens as possible (> 15-fold). A cut-off for low grade cervical disease was established at 2-15 fold which would include CIN I samples but exclude WNL. Finally, a threshold below 2 was established for normal tissue. Using these criteria, all CIN III and SCC, three CIN I, one NAT and two WNL would be classified with high grade disease. No CIN III or SCC were within the low grade category, while 3 CIN I and 3 NAT were within this category. The E6 or E7 message levels similar to WNL only included those that were WNL or CIN I.

POLR2A

E6 mRNA only

Relative expression values obtained for the HPV E6 transcript alone with normalization to POLR2A are plotted in Figure 8A. Selection criteria for expression values were established as described above. Again, all CIN III and SCC specimens were determined to contain HPV E6 mRNA at levels > 40-fold higher than the average expression of the WNL's. The NAT category had four specimens that had high risk HPV DNA and expressed HPVE6 mRNA above ten fold of the normal tissues, the remaining samples were expressing levels of HPV E6 message of 4 and 1 fold respectively. Three samples with a CIN I designation, containing high risk HPV DNA, were above 10 fold greater than the average of the WNL tissues for HPVE6 expression. Two CIN I samples expressed E6 transcripts at a level in between 2-10 fold while the remaining 24 had lower than normal levels of E6 message. Two of the WNL samples had values over two fold of the normal tissues while the rest were within normal levels. The relative HPV E6 expression levels for the biopsy categories increased with increasing severity of diagnosis with mean fold over-expression values of 2.48 for WNL, 16.82 for NAT, 35.71 for CIN I, 675.10 for CIN III, and 18787 for SCC. Threshold relative expression values for HPV E6 mRNA indicative of high grade cervical disease were determined from these data by finding a cut-off point that included the highest number of CIN III and SCC samples in the category (>20). Low grade disease was determined by excluding the majority of high grade samples from CIN I (2-20-fold). A cut-off for the absence of cervical disease was assigned by determining the threshold below which a maximum number of WNL specimens fell with minimal inclusion of CIN I samples (<2-fold). Every CIN III and cancer sample is expressing greater than 20 fold more E6 mRNA than the average of the normal tissues. Two of the CIN I biopsy tissues have E6 mRNA levels greater than 20 fold to the normal tissues and thus would be categorized as high grade. These samples had the corresponding hrHPV DNA genotype. Three additional CIN I samples express E6 mRNA at moderate levels (between two and twenty fold) and were classified as low grade, two of which had HPV DNA corresponding to the message expressed. The other CIN I samples had E6 mRNA amounts comparable to normal tissue. Two HPV DNA positive NAT had E6 transcript levels greater than 20 fold to normal tissue and were considered high grade samples with these classifications. There were three tissues adjacent to the tumor that had levels between 2 and 20 and were considered low grade. One NAT had levels of E6 message corresponding to normal levels. Low grade disease was also observed in two normal tissues according to this threshold. The remaining samples for the WNL category were expressing normal levels of E6 mRNA.

E7 mRNA only

HPV 16 and 18 E7 mRNA was assessed in biopsy samples and analyzed with the comparative Ct method using POLR2A as the housekeeping gene for normalization. Relative expression values obtained for the HPV E7 transcript alone are plotted in Figure 8B. Selection criteria for expression values were established as described above. All but one CIN III and one SCC specimen were determined to contain HPV E7 mRNA at levels >20-fold higher than the average expression of the WNL's. Although the two high-grade specimens did not express E7 mRNA at similar levels to the other samples, they did express very high amounts of E6 message (Figure 8A). One cancer specimen had 4 fold expression of E7 transcript. The NAT category had four specimens that contained high risk HPV DNA and expressed HPVE7 mRNA above five fold of the normal tissues, the remaining samples had similar levels of E7 expression to normal tissue. Two samples with a CIN I designation, containing high risk HPV DNA, were above three fold greater than the average of the WNL tissues for HPVE7 expression. The remaining 27 had similar to normal levels of E7 message. One WNL sample had values over five fold while the remaining were comparable to normal tissues.

The relative HPV E7 expression levels for the biopsy categories increased with increasing severity of diagnosis with mean fold over-expression values of 1.4 for WNL, 10.04 for NAT, 0.71 for CIN I, 451 for CIN III, and 15266 for SCC. Threshold relative expression values for HPV E6 mRNA indicative of high grade cervical disease were determined from these data by finding a cut-off point that included the highest number of CIN III and SCC samples in the category (>15). Low grade disease was determined by excluding the majority of high grade samples from CIN I (2-15 fold). A cut-off for the absence of cervical disease was assigned by determining the threshold below which a maximum number of WNL specimens fell with minimal inclusion of CIN I samples (<2- fold). Using these criteria, the majority of CIN III and cancer samples were classified as high-grade. One cancer was classified as low grade and one CIN III was normal with these limits set for E7 but would be high grade for HPVE6 mRNA expression. There were 6 cancers and 1 CIN III samples containing higher E7 than E6 mRNA. No CIN I specimens would be classified as high grade. Two CIN I samples would be categorized as low grade and the remaining samples as normal tissue. Two of the NATs are categorized as high grade, two low grade, and the rest have values similar to normal tissue. Finally, one WNL sample is expressing greater than five fold or low grade levels of E7 mRNA, while the remaining are expressing levels similar to normal tissue.

Combined E6 and E7 mRNA

Each CIN III or SCC biopsy specimen was determined to contain at least 1 high- risk HPV DNA type. All CIN III and SCC specimens were also determined to contain a concordant hrHPV type and express E6 or E7 mRNA at levels > 70-fold higher than the average expression of the WNL's (Figure 8C). There were no high-grade samples samples negative for expression of E6 or E7 mRNA. In the CIN I group, 4 of 29 specimens displayed 5-fold or better over-expression of a corresponding HPV E6/E7 mRNA, two of which were expressing at levels similar to high grade disease. All four of these samples were expressing their cognate HPV DNA virus. Two additional CIN I samples have levels of either message at levels greater than two fold over the average of the normal samples. The remaining CIN I samples had expression comparable to normal tissues. NAT expressed E6 or E7 message significantly (over 10 fold) in four HPV DNA positive samples, the other two had levels of E6/E7 message of 4 and 1, respectively. Two of the WNL biopsy cases had levels of E6 or E7 mRNA greater than 2.5 fold relative to normal tissue. The remaining WNL samples were expressing HPV E6 or E7 mRNA levels at less than 2 fold.

The relative expression levels for the biopsy categories increased with increasing severity of diagnosis with mean fold over-expression values of 2.5 for WNL, 17.10 for NAT, 35.8 for CIN I, 711.3 for CIN III, and 21407 for SCC specimens. Threshold relative expression levels for HPV E6/E7 mRNA indicative of high grade cervical disease were determined from these data by finding a cut-off point that included the highest number of CIN III samples in the category while excluding the majority of CIN I and WNL specimens as possible (>20-fold). A cut-off for low grade cervical disease was established at 1-20 fold which would include CIN I samples but exclude WNL. Finally, a threshold below 2 was established for normal tissue. Using these criteria, all CIN III and SCC, two CIN I, two NAT and no WNL tissues would be classified with high grade disease. No CIN III or SCC fell into the low grade category, while 4 CIN I, 3 NAT, and 2 WNL were within this category. The samples classified as normal included those that were WNL or CIN I.

VO

O

-J

K)

4-

Ul

Cytology

Threshold cycle determination

The C_T values of all the analyzed HPV E6 and E7 transcripts are provided in Table 7.

HPV E6 mRNA

For each cytology specimen, E6 mRNA corresponding to five high-risk HPV types (HPV 16, 18, 31, 33 and 45), was amplified by real-time PCR with reverse-transcribed RNA. The lowest raw C_T value obtained for HPV E6 mRNA from a given specimen is plotted in Figure 9A as a function of cytological diagnosis. A subset of the LSIL and HSIL specimens were determined not to contain E6 mRNA from one of these high risk virus types. For these such samples, reflex amplification was performed to test for the presence of E6 mRNA from additional high risk viruses present in the specimens as determined by HPV DNA genotyping. Replicate C_T values associated with standard deviations ≥l.O were considered invalid. Unlike absolute quantification methods, C_T values cannot be readily adjusted to reflect the presence of any contaminating DNA that is detected in the "minus-RT" reaction. Therefore, samples producing equivalent C_T values in both the experimental reaction and the "minus-RT' control reaction, were considered invalid and removed from the analysis. Twenty- four of the 29 HSIL cytology samples produced C_T values <40 indicative of HPV E6 mRNA expression in these specimens. For all 24, the HPV type giving rise to the lowest C_T value correlated with an HPV DNA type present in the sample. Five HSIL specimens that were genotyped as positive for at least 1 of 15 high risk HPV DNA types did not contain measurable E6 mRNA. Fourteen of the 30 LSIL cytology samples displayed quantifiable HPV E6 mRNA (CT<40) corresponding to a high-risk DNA type present in the specimens. Eleven LSIL cases were high-risk DNA positive but E6 mRNA negative and five cases were negative for high-risk HPV DNA.

HPV 16 E6 amplification generated the lowest C_T value for only one of the 5 HPV 16 DNA-positive NILM samples. One of the other HPV16-positive NILM specimens produced a C_T value equal to 40, indicative of no HPV16 E6 mRNA expression. The lowest C_T values observed for the remaining 3 HPV 16 DNA-positive NILM samples were derived from viruses not detected in the specimens by genotyping. One other NILM specimen contained high-risk HPV DNA (Types 31 and 45), but E6 mRNA was not detected in this specimen. C_T values <40 for HPV E6 mRNA were detected in 4 HPV DNA-negative NILM samples, while 20 NILM specimens were negative for both HPV DNA and E6 mRNA (C_τ=40).

Mean HPV E6 CT values for the cytology diagnosis groups were Cχ=39.26 for NILM, C_τ=36.84 for LSIL and C_τ=33.71 for HSIL. As the cytological diagnosis for these cervical specimens became more severe, the Cx values for E6 amplification decreased, reflecting an increase in the levels of E6 mRNA. Threshold Cx values for HPV E6 mRNA amplification indicative of high grade cervical disease were determined from these data by finding a cut-off point that included the highest number of HSIL samples in the category while excluding the majority of NILM specimens and as many LSIL samples as possible (Cχ<35). A cut-off for normal cervix was assigned by determining the threshold below which a maximum number of NILM specimens fell with minimal inclusion of LSIL samples (C_τ>38). The range of values indicative of low grade disease was thus defined by the normal and high grade thresholds (Cχ=35-38). Application of these Cx value thresholds to the data set would place 22 of the 29

NILM cases into the normal category. Six NILMs would be classified as low grade disease and 1 would fall into the high grade disease class. The copy number thresholds would identify 17 LSIL specimens as normal, 4 as low grade disease and 9 as high grade. Notably this would include the correct classification of the LSIL case with negative clinical follow-up as normal based upon the absence of detectable HPV E6 mRNA. By this classification, 1 LSIL specimen with CIN 11+ disease would be correctly classified as high grade, however two confirmed cases of CIN II would now fall into the low grade disease group. HPV E6 mRNA quantities in the HSIL specimens would place 5 samples in the normal class, based upon Cx values >38 for E6 mRNA. One of these specimens was confirmed by biopsy to be CIN I. Two other HSIL cases designated as normal produced RNA of borderline quality based upon their level of endogenous control transcript which may have negatively affected test performance on these samples. Post- analysis reflex testing on a third HSIL classified as normal demonstrated the presence of HPV E6 mRNA specific for types 35, 51 , 52 that had not been previously tested for. Five HSIL samples would be identified as low grade disease (including one confirmed case of CIN I and the remaining 19 would be categorized as high grade disease. Eight of these "high grade" are confirmed CIN 11+ by biopsy and 1 is a confirmed case of CIN I. Importantly, this particular CIN I case was upstaged to CIS by adjudication of the cytology slide. Combined E6 and E7 mRNA

HPV E6 mRNA was quantified for each of the five high-risk HPV types in every cytology specimen (plus additional types identified for reflex testing). The second major oncogenic transcript encoded by HPV, E7, was also analyzed, but exclusively for the

HPV 16 virus type. The lowest CT value for HPV E6 or HPV 16 E7 mRNA amplification for a given specimen is plotted in Figure 9B as a function of cytological diagnosis. Biopsy follow-up information is provided when available.

Twenty-four HSIL samples contained detectable E6 or E7 transcript (Cχ<40) corresponding to an HPV DNA type present in the sample. Five HSIL specimens that were genotyped as positive for high risk HPV DNA did not contain measurable E6 mRNA as indicated by Cx values equal to 40. Seventeen LSIL samples were positive for high-risk HPV DNA and produced Cx values less than 40 for HPV E6 or E7 mRNA corresponding to a high-risk DNA type present in the specimens. Eight LSIL cases were high-risk DNA positive but E6/E7 mRNA negative as specified by Cx values less than 40. Four HPV16 DNA-positive NILM specimens produced Cx values below 40. Three of these NILM's contained detectable HPVl 6 E7 transcript but not HPV 16 E6 (compare to 14(A).

Overall the data indicate that the Cx value for HPVE6/E7 mRNA amplification decreases with increasing severity of cytological diagnosis, with a mean Cx value of 39.07 for NILM specimens, 35.58 for LSIL and 32.74 for HSIL. Threshold C_x values for HPV E6/E7 mRNA amplification indicative of high grade cervical disease were determined from these data by finding a cut-off point that included the highest number of HSIL samples in the category while excluding the majority of NILM specimens and as many LSIL samples as possible (Cχ<35). A cut-off for absence of cervical disease was assigned by determining the threshold below which a maximum number of NILM specimens fell with minimal inclusion of LSIL samples (Cχ>38). The range of values indicative of low grade disease was thus defined by the normal and high grade thresholds (Cχ=36-38). Application of these Cx value thresholds to the data set would classify 22 of the NILM cases as normal, 6 as low grade and 1 as high grade cervical disease. Fourteen of the LSIL specimens would fall into the normal category, again including the biopsy-confirmed case with no evidence of dysplasia. Four LSIL specimens would be classified as low grade and two of these four are confirmed cases of CIN II. Twelve of the LSIL specimens would fall into the high grade disease group including one case of confirmed CIN II. The HPV E6/E7 Cx values and thresholds described here would categorize 5 of the HSIL cases as normal (one of which is CIN I by biopsy), 3 as low grade and 23 as high grade cervical disease. Two of the HSIL cases designated as normal produced RNA of borderline quality based upon their levels of endogenous control transcript which may have affected the performance of the test on these specific samples. Post-analysis reflex testing on a third HSIL classified as normal demonstrated the presence of HPV E6 mRNA specific for types 35, 51, 52 that had not been previously tested for. The HSIL/low grade group contains one case of confirmed CIN I and the HSIL/high grade specimens include 8 cases of confirmed CIN 11+ disease and 1 cases of CIN I. Importantly, this particular CIN I case was upstaged to CIS by adjudication of the cytology slide.

Independent analysis of HPV 16 E7 mRNA

The lowest C_x values for HPV 16 E7 mRNA are plotted for HPV DNA-negative and HPV 16 DNA-positive specimens in Figure 10. Eleven HPV 16 DNA-positive HSIL specimens generated C_x values below 40, indicative of HPV 16 E7 mRNA expression. The remaining two DNA-positive samples did not contain detectable HPV 16 E7 mRNA and thus produced C_x values equal to 40. Seven DNA-positive LSIL and 4 DNA-positve NILM also generated C_x values <40 for HPV 16 E7 mRNA amplification. Five LSIL specimens were HPV 16 DNA positive but mRNA negative. The C_x values for HPV 16 E7 mRNA amplification displayed a decreasing trend with increasing severity of cyto logical diagnosis. Mean C_x values of 39.44 for NILM, 36.38 for LSIL, and 32.96 for HSIL were observed.

Threshold C_x values for HPV 16 E7 mRNA amplification indicative of high grade cervical disease were determined from these data by finding a cut-off point that included the highest number of HSIL samples in the category while excluding the majority of NILM specimens and as many LSIL samples as possible (C_x<35). A cut-off for absence of cervical disease was assigned by determining the threshold below which a maximum number of NILM specimens fell with minimal inclusion of LSIL samples (C_x>38). The range of values indicative of low grade disease was thus defined by the normal and high grade thresholds (C_x=36-38). Six of the biopsy-confirmed specimens were HPV16 DNA positive and included in this particular analysis. Application of these C_x value thresholds to the NILM sample set would classify 25 normal, 3 low grade and 1 high grade case. The LSIL specimens would be classified as 11 normal (including one confirmed case with no evidence of dysplasia), 1 low grade and 5 high grade (including 1 confirmed case of CIN2). Application of these C_x value thresholds to the HSIL sample set would classify 2 specimens as normal, 2 as low grade and 9 as high grade. The two HSIL specimens categorized as low grade were confirmed cases of CIN III. It is important to note however that both of these particular specimens contained > 300 copies of HPV E6 mRNA and were correctly classified into the high grade group in alternative analyses. The remaining HSIL specimens classified as high grade disease contained 2 cases of confirmed CIN 11+ disease.

OO

OO

OO

OO 4-

Absolute Copy Determination

The absolute copies of all the analyzed HPV E6 and E7 transcripts are provided in Table 8.

HPV E6 mRNA

For each cytology specimen, the absolute quantities of HPV E6 mRNA molecules, corresponding to five high-risk HPV types (HPV 16, 18, 31, 33 and 45), were determined by real-time PCR amplification of reverse-transcribed RNA followed by interpolation from a standard curve. For a subset of the LSIL and HSIL specimens that did not express E6 mRNA from one of these high risk virus types, reflex quantification was performed to test for expression of mRNA from additional high risk viruses present in the specimens (as determined by HPV DNA detection and genotyping). In the event that HPV E6 was amplified in the control "minus-RT" reactions as a result of contaminating genomic DNA, these values were subtracted from any mRNA quantified in the test sample to ensure the accurate quantification of only HPV E6 messenger RNA. The highest absolute copy number for HPV E6 mRNA determined for a given specimen is plotted in Figure 1 IA as a function of cyto logical diagnosis.

All 29 HSIL specimens were positive for at least one of 15 high-risk HPV types. Twenty- four of these 29 HSIL samples contained detectable E6 transcript corresponding to an HPV DNA type present in the sample. Five HSIL specimens that were genotyped as positive for high-risk HPV DNA did not contain measurable E6 mRNA. Clinical follow- up information was available for 11 of the HSIL cytology specimens and is indicated on the graph. Biopsy indentified 3 cases of CIN I, 4 CIN II, 3 CIN III and 1 case of SCC as identified by the HSIL cytology. In the set of 30 LSIL cytology specimens, 25 were positive for high-risk HPV

DNA and 5 were negative for all virus types tested. Fourteen of the 30 LSIL cytology samples displayed quantifiable HPV E6 mRNA corresponding to a high-risk DNA type present in the specimens. Eleven LSIL cases were high-risk DNA positive but E6 mRNA negative. Clinical follow-up was available for 4 of the LSIL samples and identified 3 cases of CIN II and 1 case without evidence of cervical disease.

Six of the 30 NILM specimens contained high-risk HPV DNA. Only lof them exhibited quantifiable HPV E6 mRNA corresponding to an HPV virus type detected in the same specimen. The remaining 5 high-risk HPV DNA-positive NILM samples did not contain quantifiable HPV E6 mRNA. Low levels of HPV E6 mRNA were detected in 3 of the HPV DNA-negative NILM samples. This may reflect the background signal present in the mRNA quantification assay or could be explained by low level mRNA expression from virus below the lower limit of detection for the DNA detection and genotyping procedure. The data indicate again that the absolute HPV E6 copy number increases as the severity of cytological diagnosis increases, with mean E6 copy numbers of 2.48 for NILM, 118.10 for LSIL, and 832.20 for HSIL specimens. Threshold levels of HPV E6 mRNA copy number indicative of high grade cervical disease grade were determined from these data by finding a cut-off point that included the highest number of HSIL samples in the category while excluding the majority of NILM specimens and as many LSIL samples as possible (>10 copies). A cut-off for normal cervix was assigned by determining the threshold below which a maximum number of NILM specimens fell with minimal inclusion of LSIL samples (<1 copy). The range of values indicative of low grade disease was thus defined by the normal and high grade thresholds (1-10 copies). Application of these absolute copy number thresholds to the data set would place

25 of the 29 NILM cases into the normal category. One NILM would be classified as low grade disease and 3 would fall into the high grade disease class. The copy number thresholds would identify 16 LSIL specimens as normal, 1 as low grade disease and 13 as high grade. Notably this would include the correct classification of the LSIL case with negative clinical follow-up as normal based upon the absence of detectable HPV E6 mRNA. In addition, three LSIL/CIN II cases would be identified as high grade based upon the quantity of HPV E6 mRNA detected.

HPV E6 mRNA quantities in the HSIL specimens would place 5 samples into the normal class, based upon the absence of detectable E6 mRNA. One of these specimens was confirmed by biopsy to be CIN I . Two other HSIL cases designated as normal produced RNA of borderline quality based upon their levels of endogenous control transcript and this may have negatively affected their performance in the assay Post- analysis reflex testing on a third HSIL classified as normal demonstrated the presence of HPV E6 mRNA specific for types 35, 51 , 52 that had not been previously tested for. One HSIL would be identified as low grade disease and the remaining 23 would be categorized as high grade disease. Eight of these HSIL/high grade samples are confirmed CIN 11+ by biopsy and 2 are confirmed CIN I. Pap slides from both of the HSIL/CIN I cases classified as high grade were re-evaluated by a board certified cytotechnologist and the cytological diagnosis was up-graded from HSIL to Carcinoma in situ (CIS) for one of them. Notably, the case that was re-classified as CIS was determined to contain over 12,000 copies of HPV E6 mRNA.

Combined E6 and E7 mRNA

HPV E6 mRNA was quantified for each of the five high-risk HPV types in every cytology specimen (plus additional types identified for reflex testing). The second major oncogenic transcript encoded by HPV, E7, was also analyzed, but exclusively for the HPV 16 virus type. The highest absolute copy number for HPV E6 or HPV 16 E7 mRNA determined for a given specimen is plotted in Figure 1 IB as a function of cytological diagnosis.

Twenty-four HSIL samples contained detectable E6 or E7 transcript corresponding to an HPV DNA type present in the sample. Five HSIL specimens that were genotyped as positive for high risk HPV DNA did not contain measurable E6/E7 mRNA. The clinical follow-up information available for 11 of the HSIL cases is indicated on the graph.

Eighteen LSIL samples were positive for high-risk HPV DNA and displayed quantifiable HPV E6 or E7 mRNA corresponding to a high-risk DNA type present in the specimens. With the inclusion of HPV 16 E7 transcript quantification, only seven LSIL cases were high-risk DNA positive but E6/E7 mRNA negative as compared to eleven observed with HPV E6 analysis alone. Four of the HPV 16 DNA-positive NILM specimens were determined to contain more than one copy of HPVl 6 E7 mRNA, though only 1 of these 4 exhibited quantifiable HPV 16 E6 mRNA (see H(A)). The quantification of HPV E6/E7 mRNA in DNA-negative NILM specimens also increased by two upon inclusion of HPV16 E7. Overall the data indicate that HPVE6/E7 mRNA copy number increases with increasing severity of cytological diagnosis, with a mean mRNA copy number of 3.73 for NILM specimens, 300.3 for LSIL and 1271 for HSIL. Threshold levels of HPV E6/E7 mRNA copy number indicative of high grade cervical disease were determined from these data by finding a cut-off point that included the highest number of HSIL samples in the category while excluding the majority of NILM specimens and as many LSIL samples as possible (>10 copies). A cut-off for absence of cervical disease was assigned by determining the threshold below which a maximum number of NILM specimens fell with minimal inclusion of LSIL samples (<10 copies). The range of values indicative of low grade disease was thus defined by the normal and high grade thresholds (1-10 copies). Application of these absolute copy number thresholds to the data set would classify 20 of the NILM cases as normal, 5 as low grade and 4 as high grade cervical disease. Twelve of the LSIL specimens would fall into the normal category, again including the biopsy- confirmed case with no evidence of dysplasia. Three LSIL specimens would be classified as low grade and 15 as high grade. This LSIL/high grade group includes three cases of confirmed CIN II. The HPV E6/E7 mRNA quantities and thresholds described here would categorize 5 of the HSIL cases as normal (one of which is CIN I by biopsy), 1 as low grade and 23 as high grade cervical disease. Two HSIL cases designated as normal produced RNA of borderline quality based upon their levels of endogenous control transcript which may have affected performance in the RT-PCR assay. Post-analysis reflex testing on a third HSIL classified as normal demonstrated the presence of HPV E6 mRNA specific for types 35, 51 , 52 that had not been previously tested for. The HSIL/high grade specimens include 8 cases of confirmed CIN 11+ disease and 2 cases of CIN I. Importantly, one of these CIN I cases was upstaged to CIS by adjudication of the cytology slide.

Independent analysis of HPV 16 E7 mRNA

The absolute quantities of HPV 16 E7 mRNAs are plotted for HPV DNA-negative and HPV16 DNA-positive specimens in Figure 12. The test set included: 13 HPV16 DNA-positive HSIL; 12 HPV 16 DNA positive LSIL and 5 HPV DNA-negative LSIL; and 5 HPV 16 DNA-positive plus 24 HPV DNA-negative NILM specimens. Eleven HPV 16 DNA-positive HSIL specimens expressed more than 1 copy of HPV 16 E7 mRNA and two HSIL specimens contained HPV 16 DNA but did not express detectable HPV 16 E6 mRNA. In the HPV 16 DNA-positive LSIL group, five samples contained no detectable HPV 16 mRNA and seven specimens expressed quantifiable HPV 16 E7 mRNA. None of the 5 HPV DNA-negative LSIL samples contained measurable HPV 16 E7 mRNA. Four of 5 HPV 16 DNA positive NILM specimens contained HPV 16 E7 mRNA. Two of the 25 HPV DNA-negative NILM displayed low copy numbers of HPV 16 E7 mRNA.

These data suggest that the absolute HPV 16 E7 copy number increases as the severity of cytological diagnosis increases, with mean E7 copy numbers of: 1.71 for

NILM; 349 for LSIL; and 1102 for HSIL specimens containing HPV 16 DNA. Threshold levels of HPV 16 E7 mRNA copy number indicative of high grade cervical disease grade were determined from these data by finding a cut-off point that included the highest number of HSIL samples in the category while excluding the majority of NILM specimens and as many LSIL samples as possible (>10 copies). A cut-off for normal cervix was assigned by determining the threshold below which a maximum number of NILM specimens fell with minimal inclusion of LSIL samples (<1 copies). The range of values indicative of low grade disease was thus defined by the normal and high grade thresholds (1-10 copies). Application of these absolute copy number thresholds to the NILM sample set would classify 23 normal, 4 low grade and 2 high grade cases. The LSIL specimens would be classified as 10 normal (including one confirmed case with no evidence of dysplasia), 1 low grade and 6 high grade (including 1 confirmed case of CIN2). Application of these absolute copy number thresholds to the HSIL sample set would classify 2 specimens as normal, 1 as low grade and 10 as high grade. The HSIL specimen categorized as low grade was a confirmed case of CIN III. It is important to note however that this particular specimen did contain over 400 copies of HPV E6 mRNA and was correctly classified into the high grade group in the previous analysis. The remaining HSIL specimens classified as high grade disease contained 3 cases of confirmed CIN 11+ disease.

VO

VO

VO

Normalized target value determination

The hrHPV E6 and E7 expression data set was analyzed with the normalized target value (NTV) method, in which the absolute quantity of an mRNA is divided by the absolute quantity of the endogenous control transcript for that particular sample. For this method, the absolute quantity of endogenous control mRNA for each specimen was interpolated from a standard curve. The normalized target value is an intermediate step in the relative standard curve analysis method. For the purpose of graphing on a logio scale, all NTVs have been multiplied by 1000 and specimens that had values of zero were assigned the value of 0.01. Data for both GUSB and POLR2A normalization are described together except where indicated. The normalized target values of all the analyzed HPV E6 and E7 transcripts are provided in Table 9.

E6 mRNA only

The NTVs of just the HPV E6 transcripts are plotted in Figure 13. Twenty- five of the twenty-nine HSIL cytology samples had a positive NTV for an HPV transcript corresponding to an HPV DNA type present in the sample. Four HSIL specimens that were genotyped as positive for at least 1 of 15 high risk HPV DNA types had NTVs of zero. Thirteen of the 31 LSIL cytology samples displayed positive NTVs for mRNAs that corresponded to a high-risk DNA type present in the specimens. Fourteen LSIL cases were high-risk DNA positive but had NTVs of zero and four cases were negative for high- risk HPV DNA. Only 2 of the 30 NILM samples exhibited a NTVs of greater than one for an HPV E6 mRNA corresponding to an HPV virus type detected in the specimen. Notably, both were positive for HPV type 16. Two NILM specimens negative for hrHPV DNA exhibited some over-expression of HPV E6 mRNA, both of which were HPVl 8. The data indicate again that the normalized target value increases as the severity of cytological diagnosis increases, with mean NTVs of 5.2 for NILM specimens, 743.7 for LSIL and 2622.8 for HSIL when normalized to GUSB and NTVs of 0.6, 115.9, and 249.4 for NILM, LSIL and HSIL respectively, when normalized to POLR2A. Threshold levels of HPV E6 mRNA copy number indicative of high grade cervical disease grade were determined from these data by finding a cut-off point that included the highest number of HSIL samples in the category while excluding the majority of NILM specimens and as many LSIL samples as possible (NTV>20 for GUSB). A cut-off for normal cervix was assigned by determining the threshold below which a maximum number of NILM specimens fell with minimal inclusion of LSIL samples (NTV<1 for GUSB). The range of values indicative of low grade disease was thus defined by the normal and high grade thresholds (NTV=l-20 for GUSB). NTV cutoffs for POLR2A normalization were <1 for normal, 1-10 for low grade, and >10 for high grade.

Application of these NTV thresholds to the data set would place 26 of the 30 NILM cases into the normal category for GUSB and POLR2A. When normalized by GUSB, two NILM specimens would be classified as low grade disease while 2 would fall into the high grade disease class. For POLR2A normalization, four NILM specimens would be classified as low grade. The copy number thresholds would identify 18 LSIL specimens as normal and 13 as high grade. Notably this would include the correct classification of the LSIL case with negative clinical follow-up as normal based upon the absence of detectable HPV E6 mRNA. In addition, three LSIL/CIN II cases would be identified as high grade based upon the quantity of HPV E6 mRNA detected.

HPV E6 mRNA quantities in the HSIL specimens would place 6 samples into the normal class, based upon the absence of detectable E6 mRNA. One of these specimens was confirmed by biopsy to be CIN I. Two other HSIL cases designated as normal produced RNA of borderline quality based upon their levels of endogenous control transcript and this may have negatively affected their performance in the assay. Post- analysis reflex testing on a third HSIL classified as normal demonstrated the presence of HPV E6 mRNA specific for types 35, 51 , 52 that had not been previously tested for. One HSIL would be identified as low grade disease and the remaining 23 would be categorized as high grade disease. Eight of these HSIL/high grade samples are confirmed CIN 11+ by biopsy and 2 are confirmed CIN I. Pap slides from both of the HSIL/CIN I cases classified as high grade were re-evaluated by a board certified cytotechnologist and the cytological diagnosis was up-graded from HSIL to carcinoma in situ (CIS) for one of them. Notably, the case that was re-classified as CIS was determined to have an NTA of 28,622.5.

Combined E6 and E7 mRNA

The highest NTV for HPV E6 or E7 mRNA determined for a given specimen is plotted in Figure 14 as a function of cytological diagnosis. As for the other methods of analysis discussed thus far, in every case of HPV E6/E7 mRNA positive LSIL and HSIL cytology, the highest NTV corresponded to an HPV DNA type present in the specimen. Of the six NILM specimens positive for hrHPV DNA, four displayed an NTV greater than one. 10 hrHPV DNA positive LSIL specimens had NTV values for E6 and E7 of zero, as did five hrHPV DNA positive HSIL specimens. Of these five HSIL specimens, two had poor quality RNA as indicated by GUSB CTs above 36. Twenty-four NILM and 4 LSIL specimens were negative for high-risk HPV DNA. All 4 of the HPV DNA-negative LSIL samples also had NTV values for E6 and E7 of zero. NTV values of greater than one were detected in two of the DNA-negative NILM specimens.

Overall the data indicate that normalized threshold values for HPVE6/E7 mRNA increase with increasing severity of cyto logical diagnosis, with mean NTVs of 10.3 for NILM specimens, 1619.0 for LSIL and 3982.4 for HSIL when normalized to GUSB and 0.9, 192.5, and 482.2 fold over-expression for NILM, LSIL and HSIL respectively, when normalized to POLR2A. Threshold levels of HPV E6/E7 mRNA copy number indicative of high grade cervical disease grade were determined from these data by finding a cut-off point that included the highest number of HSIL samples in the category while excluding the majority of NILM specimens and as many LSIL samples as possible (NTV>100 for GUSB). A cut-off for normal cervix was assigned by determining the threshold below which a maximum number of NILM specimens fell with minimal inclusion of LSIL samples (NTV<1 for GUSB). The range of values indicative of low grade disease was thus defined by the normal and high grade thresholds (NTV = 1 to 100 for GUSB). NTV cutoffs for POLR2A normalization were <1 fold for normal, 1-20 fold for low grade, and >20 fold for high grade. Application of these GUSB NTV thresholds to the data set would classify 24 of the

NILM cases as normal, 5 as low grade and 1 as high grade cervical disease. For the POLR2A specific threshold, the one high grade NILM specimen is converted to low grade. For both GUSB and POLR2A normalization, fourteen of the LSIL specimens would fall into the normal category, again including the biopsy-confirmed case with no evidence of dysplasia. Six LSIL specimens would be classified as low grade and 14 as high grade. This LSIL/low grade group contains one and the high grade group includes two cases of confirmed CIN II. The HPV E6/E7 mRNA quantities and thresholds described here would categorize 5 of the HSIL cases as normal (one of which is CIN I by biopsy), 1 as low grade and 23 as high grade cervical disease. For the POLR2A specific threshold, one high grade NILM specimen is converted to low grade. Two HSIL cases designated as normal produced RNA of borderline quality based upon their levels of endogenous control transcript which may have affected performance in the RT-PCR assay. Post-analysis reflex testing on a third HSIL classified as normal demonstrated the presence of HPV E6 mRNA specific for types 35, 51, 52 that had not been previously tested for. The HSIL/high grade specimens include 8 cases of confirmed CIN 11+ disease and 2 cases of CIN I. Importantly, one of these CIN I cases was upstaged to CIS by adjudication of the cytology slide.

Independent analysis of HPV 16 E7 mRNA

The normalized target value for HPV 16 E 7 mRNA from HPV 16 DNA positive and DNA-negative specimens were quantified (Figure 15). Eleven of the 13 HPV 16 DNA- positive HSIL specimens possessed NTV values of greater than one for HPV 16 E7 mRNA. Seven HPV 16 positive LSIL specimens possessed NTVs of greater than one for

HPV 16 E7 mRNA. Four HPV 16 positive LSIL samples contained no detectable HPV 16 E7 mRNA, and of these, one expressed HPV 16 E6 but not E7 mRNA. None of the 4 HPV DNA-negative LSIL samples contained measurable HPV 16 E7 mRNA. Four HPV 16 DNA positive NILM specimens possessed positive NTVs HPV 16 E7 mRNA. None of the 25 HPV DNA-negative specimens demonstrated positive NTVs for HPV 16 E7 mRNA. These data suggest that the NTVs of HPV E7 mRNA increases as the severity of cytological diagnosis increases, with a mean NTV for HPVl 6 E7 of 9.1 for NILM specimens, 1835.6 for LSIL and 3747.6 for HSIL when normalized to GUSB and mean NTVs of 0.7, 174.1, and 606.8 for NILM, LSIL and HSIL respectively, when normalized to POLR2A. Threshold levels of HPV 16 E7 mRNA copy number indicative of high grade cervical disease grade were determined from these data by finding a cut-off point that included the highest number of HSIL samples in the category while excluding the majority of NILM specimens and as many LSIL samples as possible (>20 fold). A cut-off for absence of disease was assigned by determining the threshold below which a maximum number of NILM specimens fell with minimal inclusion of LSIL samples (<1 fold). The range of values indicative of low grade disease was thus defined by the normal and high grade thresholds (1-20 fold) for GUSB normalization. Cutoffs for POLR2A normalization were <1 fold for normal, 1-10 fold for low grade, and >10 fold for high grade.

Application of thresholds to the NILM sample set would classify 25 normal, 1 low grade and 3 high grade cases for GUSB normalization and 25 normal, and 4 low grade cases for POLR2A. The LSIL specimens normalized with GUSB would be classified as 9 normal (including one confirmed case with no evidence of dysplasia) and 7 high grade (including 1 confirmed case of CIN2). For POLR2A normalization, five specimens would be classified as high grade, two as low and nine as normal. Application of the GUSB threshold to the HSIL sample set would classify 2 specimens as normal, and 11 as high grade. For POLR2A, one high grade HSIL sample would be converted to low grade. This converted HSIL specimen was a confirmed case of CIN III. It is important to note however that this particular specimen possessed an NTV of 445.6 for E6 mRNA when normalized by PORL2A and was correctly classified into the high grade group in the previous analysis. The remaining HSIL specimens classified as high grade disease contained 3 cases of confirmed CIN 11+ disease.

O O

O

O

O 4-

Relative Standard Curve Analysis

The hrHPV E6 and E7 expression data set was analyzed with the relative standard curve method, which compares HPV mRNA levels (the target value) of a particular specimen to a calibrator sample or group of samples. In this analysis, the calibrator was the average normalized target value for all of the NILM specimens. For this method, the absolute quantity of endogenous control mRNA for each specimen was also interpolated from a standard curve. Then, the absolute quantity of HPV E6 mRNA for each specimen was divided by the absolute quantity of the endogenous control gene for that same specimen to derive the normalized target value. To determine expression levels for each sample relative to a calibrator, the normalized target values for each specimen was divided by the average normalized target value for all of the NILM specimens determined with the respective HPV assay. Data for both GUSB and POLR2A normalization are described together except where indicated. The relative standard curve analysis of all the analyzed HPV E6 and E7 transcripts, normalized to GUSB or POLR2A, are provided in Tables 10 and 11 , respectively.

E6 mRNA only

Because the presence of HPV E7 mRNA was quantified solely for the HPV 16 virus type, fold over-expression of just the HPV E6 transcript as determined by the relative standard curve method is plotted in Figure 16. Twenty of the 24 HSIL cytology samples over-expressed an HPV transcript corresponding to an HPV DNA type present in the sample. Five HSIL specimens that were genotyped as positive for at least 1 of 15 high risk HPV DNA types did not contain measurable E6 mRNA. Eight of the 26 LSIL cytology samples displayed quantifiable HPV E6 mRNA corresponding to a high-risk DNA type present in the specimens. Thirteen LSIL cases were high-risk DNA positive but did not over-express E6 mRNA and four cases were negative for high-risk HPV DNA. Only 2 of the 30 NILM samples exhibited over-expression of an HPV E6 mRNA corresponding to an HPV virus type detected in the same specimen. Notably, both were positive for HPV type 16. Two specimens negative for hrHPV DNA exhibited some over-expression of HPV E6 mRNA, both of which were HPVl 8.

The data indicate again that the absolute HPV E6 copy number increases as the severity of cytological diagnosis increases, with mean fold over-expression of 1.4 for NILM specimens, 140.9 for LSIL and 1086.7 for HSIL when normalized to GUSB, and 0.9, 111.9, and 452.6 mean fold over-expression for NILM, LSIL and HSIL respectively, when normalized to POLR2A. Threshold levels of HPV E6 mRNA copy number indicative of high grade cervical disease grade were determined from these data by finding a cut-off point that included the highest number of HSIL samples in the category while excluding the majority of NILM specimens and as many LSIL samples as possible (>10 fold). A cut-off for normal cervix was assigned by determining the threshold below which a maximum number of NILM specimens fell with minimal inclusion of LSIL samples (<1 fold). The range of values indicative of low grade disease was thus defined by the normal and high grade thresholds (1-10 fold) for both GUSB and POLR2A normalization.

With this criteria for both GUSB and POLR2A, 12 of 19 overexpressing HSIL samples were considered high grade. Within the over expressing HSIL samples, there was biopsy information for eight of them. Six of them were either CIN 2 or CIN 3, where two of them were classified as CIN I. Adjudication of the cytology result with a board certified cytotechnologist indicated that the CIN I HSIL actually was Carcinoma in situ. Interestingly, this sample expressed a very high level of mRNA. There were five HSILs that did not expressEβ or E7 mRNA, including one CIN I sample. Of these, two possessed poor quality RNA as determined by GUSB CTs above 36. Seven samples with LSIL cytology as normalized with GUSB and eight as normalized by POLR2A expressed E6 or E7 message at greater than 10 fold levels. Two of these samples had CIN 2 on biopsy. There were two NILM samples, one with hrHPV DNA, that expressed levels greater than ten fold and would be classified with high grade disease. One CIN2+ HSIL sample was classified as low grade by the threshold set. This sample did express significant levels of HPV 16 E7 however, and would thus be detected by the E7 assay. One LSIL specimen and four NILM samples as normalized by GUSB expressed E6 or E7 transcripts at amounts between 1-10 fold. One of the LSIL samples was CIN2+ on biopsy. The rest of the 26 NILM and 14 LSIL samples were negative.

Combined E6 and E7 mRNA

The highest fold over-expression of HPV E6 or E7 mRNA determined for a given specimen is plotted in Figure 17 as a function of cyto logical diagnosis. Five HSIL and 5 LSIL specimens expressed HR RNA that was detected as a part of reflex testing, and were not included in this analysis.

As for the other methods of analysis discussed thus far, in every case of HPV E6/E7 mRNA positive LSIL and HSIL cytology, the highest expressing HPV E6/E7 mRNA corresponded to an HPV DNA type present in the specimen. Of the six NILM specimens positive for hrHPV DNA, only four showed significant over-expression of mRNA relative to the average of NILM specimens. Eight hrHPV DNA positive LSIL specimens did not over-express any hrHPV RNA, while five hrHPV DNA positive HSIL specimens did not. Of these five HSIL specimens, two had poor quality RNA as indicated by GUSB CTs above 36. Twenty- four NILM and 4 LSIL specimens were negative for high-risk HPV DNA. All 4 of the HPV DNA-negative LSIL samples were also negative for over-expression of high-risk HPV E6/E7 mRNA. Significant over-expression of HPV transcript could be detected in two of the DNA negative NILM specimens.

Overall the data indicate that HPVE6/E7 mRNA over-expression increases with increasing severity of cytological diagnosis, with mean over-expression of 2.1 for NILM specimens, 254.8 for LSIL and 1284.7 for HSIL when normalized to GUSB and 1.5, 305.0, and 894.2 mean fold over-expression for NILM, LSIL and HSIL respectively, when normalized to POLR2A. Threshold levels of HPV E6/E7 mRNA copy number indicative of high grade cervical disease grade were determined from these data by finding a cut-off point that included the highest number of HSIL samples in the category while excluding the majority of NILM specimens and as many LSIL samples as possible (>10 fold). A cut-off for normal cervix was assigned by determining the threshold below which a maximum number of NILM specimens fell with minimal inclusion of LSIL samples (<1). The range of values indicative of low grade disease was thus defined by the normal and high grade thresholds (1 to 10 fold) for GUSB normalization. Cutoffs for POLR2A normalization were <1 fold for normal, 1-10 fold for low grade, and >10 fold for high grade.

With this criteria, 17 of 19 over expressing HSIL samples were considered high grade for both GUSB and POLR2A normalization. Within the over expressing HSIL samples, there was biopsy information for eight of them. Six of them were either CIN 2 or CIN 3, where two of them were classified as CIN I. Adjudication of the cytology result with a board certified cytotechnologist indicated that one of the CIN I HSIL specimens actually was Carcinoma in situ. Interestingly, this sample expressed a very high level of mRNA. There were five HSILs that did not express E6 or E7 mRNA, including two CIN I samples. Of these, two possessed poor quality RNA as determined by GUSB CTs above 36. For both GUSB and POLR2A, 9 samples with LSIL cytology expressed E6 or E7 message at greater than 10 fold levels. Two of these samples had CIN 2 on biopsy. There were two NILM samples for GUSB and POLR2A, one with hrHPV DNA, that expressed levels greater than 10 fold and would be classified with high grade disease. Two LSIL specimens and two NILM sample expressed E6 or E7 transcripts at amounts between 1-10 fold. For POLR2A, one NILM specimen classified as normal with GUSB was converted to low grade. Also, one of the LSIL samples was CIN2+ on biopsy. The rest of the 26 NILM and 14 LSIL samples were negative.

Independent analysis of HPV 16 E7 mRNA

HPV 16 E7 mRNA from HPV DNA negative and positive specimens were quantified for relative expression by the relative standard curve (Figure 18). The test set included: 13 HPV 16 DNA-positive HSIL; 12 HPV 16 DNA positive LSIL and 4 HPV DNA-negative LSIL; and 4 HPV 16 DNA-positive plus 24 HPV DNA-negative NILM specimens.

Eleven of the 13 HPV 16 DNA-positive HSIL specimens contained detectable levels of HPV 16 E7 mRNA. An additional HPV 16 positive HSIL contained an additional high-risk HPV virus that was expressing high levels of oncogene mRNA, and one other expressed only HPV 16 E6 mRNA and no E7.

Seven HPV 16 positive LSIL specimens expressed quantifiable HPV 16 E7 mRNA. Four HPV 16 positive LSIL samples contained no detectable HPV 16 E7 mRNA. None of the 4 HPV DNA-negative LSIL samples contained measurable HPV 16 E7 mRNA. Four HPV 16 DNA positive NILM specimens contained HPV 16 E7 mRNA. None of the 25 HPV DNA-negative specimens demonstrated HPV 16 E7 mRNA expression relative to the average of the NILM specimens.

These data suggest that the expression of HPV E7 mRNA relative to the average of the NILMs increases as the severity of cyto logical diagnosis increases, with E7 mean fold over-expression of: 1.0 for NILM; 197.4 for LSIL; and 403.0 for HSIL specimens containing HPV16 DNA when normalized to GUSB, and 1.0, 252.5, and 879.9 mean fold over-expression for NILM, LSIL and HSIL respectively, when normalized to POLR2A.. Threshold levels of HPV 16 E7 mRNA copy number indicative of high grade cervical disease grade were determined from these data by finding a cut-off point that included the highest number of HSIL samples in the category while excluding the majority of NILM specimens and as many LSIL samples as possible (>4 fold). A cut-off for absence of disease was assigned by determining the threshold below which a maximum number of NILM specimens fell with minimal inclusion of LSIL samples (<1 fold). The range of values indicative of low grade disease was thus defined by the normal and high grade thresholds (1-4 fold) for GUSB normalization. Cutoffs for P0LR2A normalization were <1 fold for normal, 1-5 fold for low grade, and >5 fold for high grade.

With this criteria, all over expressing HSIL samples were considered high grade for GUSB normalization, while one sample dropped to low grade with POLR2A. Within the over expressing HSIL samples, there was biopsy information for four of them. All three of them were either CIN 2 or CIN 3 with biopsy. There were two HPV 16 DNA positive HSILs that did not expressEβ or E7 mRNA. One of these is expressing HV 16 E6 transcript at significant levels however, and would be detected by the E6 assay. Five samples with LSIL cytology expressed E6 or E7 message at greater than 10 fold levels. One of these samples had CIN 2 on biopsy. There was one HPV 16 DNA positive NILM samples that expressed levels greater than ten fold and would be classified with high grade disease. One CIN3+ HSIL sample was classified as low grade by the threshold set. Two LSIL specimens and two NILM sample expressed E6 or E7 transcripts at amounts between 1-4 fold when normalized with GUSB. One NILM classified as normal was converted to low grade upon normalization with POLR2A. The rest of the 26 NILM and 11 LSIL samples were negative.

Table 10. Relative standard curve values of HPV E6 and E7 in cervical cytology samples relative to GUSB.

W

Table 11. Relative standard curve values of HPV E6 and E7 in cervical cytology samples relative to P0LR2A.

OO

VO

O

Comparative C_T

The comparative C_T values of all the analyzed HPV E6 and E7 transcripts, normalized to GUSB or POLR2A, are provided in Tables 12 and 13, respectively.

GUSB

E6 mRNA only

For each cytology specimen, the relative quantities of HPV E6 mRNA molecules, corresponding to five high-risk HPV types (HPV 16, 18, 31, 33 and 45), were determined using the Comparative CT method of mRNA quantification. Transcript levels were normalized to the endogenous control transcript of GUSB for each specimen, and then the normalized values were calibrated relative to the average expression of the NILM specimens. Absolute quantification methods readily allow for the discrimination between absolute signal due to mRNA amplification and artifactual signal resulting from amplification of contaminating genomic DNA by subtraction of the absolute value of nucleic acid quantified in the "minus-RT" control reaction. The comparative C_T method of relative quantification is not readily amenable to this type of compensation. Therefore, samples producing equivalent C_T values in both the experimental reaction and the "minus- RT' control reaction, were considered invalid and removed from the analysis. Additionally, relative expression values less than 2-fold above the mean NILM expression level were considered invalid and were assigned a value of 0.

Relative expression values obtained for the HPV E6 transcript alone are plotted in Figure 19A. Sixteen HSIL specimens were determined to contain HPV E6 mRNA at levels > 2-fold higher than the average expression of the NILM 's. None of the remaining 7 HSIL samples expressed at least 2-fold more E6 or E7 mRNA as compared to the NILM diagnosis group. One of these 7 HPV E6 niRNA-negative HISL specimens did however display increased relative HPV E7 transcript. In the LSIL group, 7 specimens contained high-risk HPV DNA and also displayed 2-fold or better over-expression of a corresponding HPV E6 mRNA. Fourteen additional LSIL samples containing HPV DNA did not contain elevated relative levels of HPV E6 mRNA. The remaining 5 LSIL specimens were completely negative for all high-risk HPV DNA types tested and only one of them displayed elevated HPV E6 mRNA. Just 6 of the NILM cytology cases were HPV DNA-positive. Two of these samples over-expressed HPV E6 transcript of a type concordant with the sample genotype and the other four did not. One HPV DNA negative NILM sample contained E6 transcript at greater than 2 fold levels. The relative HPV E6 expression levels for the three cytology categories increased with increasing severity of diagnosis with mean fold over-expression values of 0.54 for NILM, 23.62 for LSIL and 92.81 for HSIL specimens. Threshold relative expression values for HPV E6 mRNA indicative of high grade cervical disease were determined from these data by finding a cut-off point that included the highest number of HSIL samples in the category while excluding the majority of NILM specimens and as many LSIL samples as possible (>5-fold). Inclusion of biopsy information also assisted the determination of this threshold. A cut-off for the absence of cervical disease was assigned by determining the threshold below which a maximum number of NILM specimens fell with minimal inclusion of LSIL samples (<2-fold). The range of values indicative of low grade disease was thus defined by the normal and high grade thresholds (2-5 -fold). With these criteria, 16 of 17 high- fold expression HSIL samples were considered high grade. Within these HSIL cases, six had accompanying biopsy information. Five specimens were either CIN II or CIN III, where one of them was classified as CIN I. Adjudication of the cytology result by a board certified cytotechnologist indicated that the CIN I HSIL actually was

Carcinoma in situ. Interestingly, this sample contained a very high level of mRNA. There were six HSIL specimens that did not contain HPV E6 or E7 mRNA, including two CIN I samples. Of these, two possessed poor quality RNA as determined by their endogenous control transcript levels. Four specimens with LSIL cytology expressed E6 or E7 message at greater than 5 fold levels. Two of these were identified as CIN II on biopsy. There were two NILM samples, one with high-risk HPV DNA, that contained HPV E6/E7 levels greater than five fold and would be classified with high grade disease. One CIN 11+ HSIL sample was classified as low grade by the threshold set. This sample did contain significant levels of HPV 16 E7 however, and would thus be detected by the E7 assay. Four LSIL specimens and one NILM sample expressed E6 or E7 transcripts at amounts between 2-4 fold. One of the LSIL samples was CIN2+ on biopsy. The rest of the 26 NILM and 14 LSIL samples were negative.

Combined E6 and E7 mRNA For each cytology specimen, the relative quantities of HPV E6 mRNA molecules, corresponding to five high-risk HPV types (HPV 16, 18, 31, 33 and 45), were determined using the comparative C_T method of mRNA quantification. Relative levels of HPV 16 E7 mRNA were also determined in the same manner. The fold over-expression of HPV E6 or E7 mRNA relative to the mean NILM expression for a given specimen is plotted in Figure 19B as a function of cyto logical diagnosis.

Every HSIL cytology specimen was determined to contain at least 1 high-risk HPV DNA type. Seventeen HSIL specimens were also determined to contain a concordant high-risk type HPV E6 or E7 mRNA at levels > 2-fold higher than the average expression of the NILM 's. None of the remaining 6 HSIL samples expressed at least 2-fold more E6 or E7 mRNA as compared to the NILM diagnosis group. Two of these niRNA-negative specimens may possess poor RNA quality as indicated by GUSB CTs above 36. In the LSIL group, 11 specimens contained high-risk HPV DNA and also displayed 2-fold or better over-expression of a corresponding HPV E6/E7 mRNA. Ten additional LSIL samples containing HPV DNA did not contain elevated relative levels of HPV E6 or E7 mRNA. The remaining 5 LSIL specimens were completely negative for all high-risk HPV DNA types tested and only one of them displayed elevated E6/E7 mRNA. Just 6 of the NILM cytology cases were HPV DNA-positive. Three of these samples over-expressed an E6 or E7 transcript greater than two-fold to the average of the NILMs of a type concordant with the sample genotype and the other three did not. One additional NILM specimen, negative for high-risk HPV DNA, contained 2-fold more HPV E6 or E7 message than the average of the normal samples. The relative expression levels for the three cytology categories increased with increasing severity of diagnosis with mean fold over-expression values of 1.218 for NILM, 118.3 for LSIL and 218.7 for HSIL specimens. Threshold relative expression levels for HPV E6/E7 mRNA indicative of high grade cervical disease were determined from these data by finding a cut-off point that included the highest number of HSIL samples in the category while excluding the majority of NILM specimens and as many LSIL samples as possible (>5-fold). Inclusion of biopsy information also assisted the determination of this threshold. A cut-off for the absence of cervical disease was assigned by determining the threshold below which a maximum number of NILM specimens fell with minimal inclusion of LSIL samples (<2-fold). The range of values indicative of low grade disease was thereby defined by the normal and high grade thresholds (2-5-fold). With these criteria, all 17 over-expressing HSIL samples were considered high grade. Within the high grade HSIL samples, there was biopsy information for seven cases. Six were either CIN II or CIN III, and one was classified as CIN I. Adjudication of the cytology result by a board certified cytotechnologist indicated that the CIN I HSIL actually was Carcinoma in situ. Interestingly, this sample contained a very high level of HPV mRNA. Of the six HSIL specimens that did not express E6 or E7 mRNA, two were classified as CIN I upon biopsy. These two CINI samples are different from the two DNA- positive RNA-negative HSIL's with poor quality mRNA mentioned above. Eight samples with LSIL cytology contained HPV E6 or E7 message at greater than 5 fold levels. Two of these samples were classified as CIN II on biopsy. There were three NILM samples, two with high-risk HPV DNA, that had 5 -fold elevated levels of HPV mRNA and would be classified with high grade disease. No HSIL sample was classified as low grade by the threshold set. Four LSIL specimens and one NILM sample contained E6 or E7 transcripts at amounts between 2-4 fold. One of the LSIL samples was CIN 11+ on biopsy. The rest of the 26 NILM and 14 LSIL samples were negative

Independent analysis of HPV 16 E7 mRNA

HPV E7 mRNA was analyzed independently for HPV type 16 and the relative expression levels of HPV 16 E7 mRNA are plotted for HPV DNA-negative and HPV 16 DNA-positive specimens in Figure 20. The test set included: 13 HPV16 DNA-positive HSIL; 12 HPV 16 DNA positive LSIL and 5 HPV DNA-negative LSIL; and 4 HPV 16 DNA-positive plus 25 HPV DNA-negative NILM specimens. Eleven HPV 16 DNA- positive HSIL specimens exhibited >2-fold relative levels of HPV 16 E7 mRNA. Two HSIL specimens contained HPV 16 DNA but did not contain increased relative levels of HPV 16 E7 mRNA. Six of the HPV 16 DNA-positive LSIL specimens had elevated levels of HPV 16 E7 mRNA. The other 6 HPV 16 DNA-positive LSIL samples did not contain >2-fold increased levels of HPV 16 E7 mRNA. AU 5 of the HPV DNA-negative LSIL samples also showed no increase in HPV 16 E7 mRNA as compared to the average NILM levels. Three of the 5 HPV 16 DNA-positive NILM specimens demonstrated relative over- expression of HPV 16 E7 mRNA and the other 2 did not. None of the HPV 16 DNA- negative NILM specimen displayed increased relative expression of HPV 16 E7 mRNA.

The relative HPV 16 E7 mRNA expression levels for the three cytology categories increased with increasing severity of diagnosis with mean fold over-expression values of 1.08 for NILM, 147.2 for LSIL and 253.1 for HSIL specimens. Threshold relative expression values for HPV 16 E7 mRNA indicative of high grade cervical disease were determined from these data by finding a cut-off point that included the highest number of HSIL samples in the category while excluding the majority of NILM specimens and as many LSIL samples as possible (>5-fold). A cut-off for absence of disease was assigned by determining the threshold below which a maximum number of NILM specimens fell with minimal inclusion of LSIL samples (<2-fold). The range of values indicative of low grade disease was thereby defined by the normal and high grade thresholds (2-5-fold). With these criteria, 10 of 11 HPV 16 E7 mRNA-positive HSIL samples were considered high grade. Within these HSIL samples, there was biopsy information for three cases. All three of them were either CIN II or CIN III with biopsy. There were two HPV 16 DNA positive HSILs that did not express HPV E6 or E7 mRNA. One of these contains HV 16 E6 transcript at significant levels however, and would be detected by the E6 assay. Five samples with LSIL cytology expressed E6 or E7 message at greater than 5 fold levels. One of these samples was CIN II by biopsy. There were two HPV 16 DNA positive NILM samples that expressed levels greater than five fold and would be classified with high grade disease. One CIN III HSIL sample was classified as low grade by the threshold set. One LSIL specimen and one NILM sample contained E6 or E7 transcripts at amounts between 2-4 fold. The rest of the 26 NILM and 11 LSIL samples were negative.

POLR2A

E6 mRNA only

For each cytology specimen, the relative quantities of HPV E6 mRNA molecules, corresponding to five high-risk HPV types (HPV 16, 18, 31, 33 and 45), were determined using the comparative C_T method of mRNA quantification. Transcript levels were normalized to the endogenous control transcript of POLR2A for each specimen, and then the normalized values were calibrated relative to the average expression of the NILM specimens. Absolute quantification methods allow for the discrimination between absolute signal due to mRNA amplification and artifactual signal resulting from amplification of contaminating genomic DNA by subtraction of the absolute value of nucleic acid quantified in the "minus-RT" control reaction. The comparative C_T method of relative quantification is not readily amenable to this type of compensation. Therefore, samples producing equivalent C_T values in both the experimental reaction and the "minus- RT' control reaction, were considered unacceptable and removed from the analysis. Additionally, relative expression values less than 2-fold above the average NILM expression levels were considered invalid and were assigned a value of 0. Relative expression values obtained for the HPV E6 transcript are plotted in Figure 2 IA. Selection criteria for expression values were established as described above. Eighteen HSIL specimens were determined to contain HPV E6 mRNA at levels > 2-fold higher than the average expression of the NILM's. None of the remaining 5 HSIL samples expressed at least 2-fold more E6 mRNA as compared to the NILM diagnosis group. In the LSIL group, 8 specimens contained high-risk HPV DNA and also displayed 2-fold or better over-expression of a corresponding HPV E6 mRNA. Thirteen additional LSIL samples containing HPV DNA did not contain elevated relative levels of HPV E6 mRNA. The remaining 5 LSIL specimens were completely negative for all high-risk HPV DNA types tested and none of them displayed elevated E6 mRNA. Just 6 of the NILM cytology cases were HPV DNA-positive. Two of these samples over-expressed E6 transcript of a type concordant with the sample genotype and the other four did not. Two of the high-risk HPV DNA-negative NILM specimens gave rise to HPV E6 or E7 signals above the 2-fold limit.

The relative HPV E6 expression levels for the three cytology categories increased with increasing severity of diagnosis with mean fold over-expression values of 0.53 for NILM, 8.23 for LSIL and 143.0 for HSIL specimens. Threshold relative expression values for HPV E6 mRNA indicative of high grade cervical disease were determined from these data by finding a cut-off point that included the highest number of HSIL samples in the category while excluding the majority of NILM specimens and as many LSIL samples as possible (>5-fold). A cut-off for the absence of cervical disease was assigned by determining the threshold below which a maximum number of NILM specimens fell with minimal inclusion of LSIL samples (<2-fold). The range of values indicative of low grade disease was thus defined by the normal and high grade thresholds (2-5-fold).

Based upon these criteria and the fold-increase in HPV E6 mRNA in the specimens, 17 HSIL samples would have been classified as high grade, including all six confirmed CIN2 + cases. One CINl case would have been designated as high grade as well, however this sample contains one of the highest levels HPV E6 mRNA in the study set. This specimen is likely to have been misclassified as a CINl, as cyto logical adjudication reclassified this sample as carcinoma in situ. Therefore, it is likely that E6 and E7 expression analysis correctly identified this sample as high grade whereas biopsy did not. Five HSIL specimens would have been identified as normal based on these thresholds, however 1 of these cases is a confirmed CINl

The thresholds would have designated 6 LSIL specimens, including 2 CIN2 cases as high grade disease, and 2 LSIL specimens as low grade disease, includingl CIN2 case. This down-graded LSIL specimen does however contain high levels of HPV 16 E7 transcript, and could have been detected by the HPV 16 E7 assay. The remaining LSIL samples would all have been classified as low grade disease. Two of the NILM specimens were classified as low grade and two were identified as high grade. The remaining 16 NILM samples would have been designated as normal.

Combined E6 and E7 mRNA

The fold over-expression of HPV E6 or E7 mRNA relative to the mean NILM expression for a given specimen is plotted in Figure 2 IB as a function of cyto logical diagnosis. Every HSIL cytology specimen was determined to contain at least 1 high-risk HPV DNA type. Eighteen HSIL specimens were also determined to contain a concordant high-risk type HPV E6 or E7 mRNA at levels > 2-fold higher than the average expression of the NILM 's. None of the remaining 5 HSIL samples expressed at least 2-fold more E6 or E7 mRNA as compared to the NILM diagnosis group. In the LSIL group, 10 specimens contained high-risk HPV DNA and also displayed 2-fold or better over- expression of a corresponding HPV E6/E7 mRNA. Eleven additional LSIL samples containing HPV DNA did not contain elevated relative levels of HPV E6 or E7 mRNA. The remaining 5 LSIL specimens were completely negative for all high-risk HPV DNA types tested and none of them displayed elevated E6/E7 mRNA. Just 6 of the NILM cytology cases were HPV DNA-positive. Four of these samples over-expressed an E6 or E7 transcript of a type concordant with the sample genotype and the other two did not. Two of the high-risk HPV DNA-negative NILM specimens gave rise to HPV E6 or E7 signals above the 2-fold limit.

The relative expression levels for the three cytology categories increased with increasing severity of diagnosis with mean fold over-expression values of 1.01 for NILM, 186.6 for LSIL and 450.1 for HSIL specimens. Threshold relative expression levels for HPV E6/E7 mRNA indicative of high grade cervical disease were determined from these data by finding a cut-off point that included the highest number of HSIL samples in the category while excluding the majority of NILM specimens and as many LSIL samples as possible (> 10-fold). A cut-off for the absence of cervical disease was assigned by determining the threshold below which a maximum number of NILM specimens fell with minimal inclusion of LSIL samples (<2-fold). The range of values indicative of low grade disease was thereby defined by the normal and high grade thresholds (2- 10-fold).

Biopsy data was available for five LSIL and eleven HSIL specimens. In the HSIL sample set, all CIN3 specimens fall above the cutoff for high grade disease of 10 fold over-expression. Only one CIN2 in the LSIL sample set, with a fold over-expression of 4.77, lies below this cut off. This sample contains high levels of HPV16 E7 transcript however, and would have been detected by the E7 assay. Of the three CINl samples in this data set, one contained no detectable increase in HPV E6/E7 mRNA, and one lies below the cutoff for high grade disease. The third CINl is one of the highest expressers of all of the HSIL specimens. This specimen is likely to have been misclassifϊed as a CINl, as cytological adjudication reclassifϊed this sample as carcinoma in situ. Therefore, it is likely that E6 and E7 expression analysis correctly identified this sample as high grade whereas biopsy did not.

Independent analysis of HPV 16 E7 mRNA

The relative expression levels of HPV 16 E7 mRNA are plotted for HPV DNA- negative and HPV16 DNA-positive specimens in Figure 22. The test set included: 13 HPV 16 DNA-positive HSIL; 12 HPV 16 DNA positive LSIL and 5 HPV DNA-negative LSIL; and 4 HPV 16 DNA-positive plus 24 HPV DNA-negative NILM specimens. Twelve HPV 16 DNA-positive HSIL specimens exhibited >2-fold relative levels of HPV 16 E6 mRNA as indicated by open circles. Only one HPV16-positive HSIL specimen contained HPV 16 DNA but did not contain an increased relative level of HPV 16 E6 mRNA.

Eleven HPV 16 DNA-positive HSIL specimens exhibited >2-fold relative levels of HPV 16 E7 mRNA as indicated by open circles. Two HSIL specimens contained HPV 16 DNA but did not contain increased relative levels of HPV 16 E7 mRNA. Five of the HPV 16 DNA-positive LSIL specimens also included elevated levels of HPV 16 E6 mRNA. The other 7 HPV 16 DNA-positive LSIL samples did not contain >2-fold increased levels of HPV 16 E7 mRNA. AU 5 of the HPV DNA-negative LSIL samples also showed no increase in HPV 16 E7 mRNA as compared to the average NILM levels. Four of the 5 HPV 16 DNA-positive NILM specimens demonstrated relative over- expression of HPV 16 E7 mRNA and one other did not. None of the HPV 16 DNA- negative NILM specimen displayed increased relative expression of HPV 16 E7 mRNA. The relative HPV 16 E7 mRNA expression levels for the three cytology categories increased with increasing severity of diagnosis with mean fold over-expression values of 0.80 for NILM, 163.6 for LSIL and 594.1 for HSIL specimens. Relative expression threshold values for HPV 16 E7 mRNA indicative of high grade cervical disease were determined from these data by finding a cut-off point that included the highest number of HSIL samples in the category while excluding the majority of NILM specimens and as many LSIL samples as possible (>10-fold). A cut-off for absence of disease was assigned by determining the threshold below which a maximum number of NILM specimens fell with minimal inclusion of LSIL samples (<2-fold). The range of values indicative of low grade disease was thereby defined by the normal and high grade thresholds (2- 10-fold). Biopsy data was available for one LSIL and four HSIL specimens in this sample set. When normalized by GUSB, all CIN2 and CIN3 specimens demonstrated an over- expression of at least 30 fold, and therefore fall above the cutoff for high grade disease. There was one exception however, a CIN3 in the HSIL data set that demonstrated 2.95 fold over-expression. This sample expressed significant levels of the HPV 16 E6 transcript however, and would have been detected using the E6 transcript cutoff.

00

Materials and Methods Specimens

Anonymized residual SurePath™ cervical cytology and formalin- fixed paraffin- embedded (FFPE) specimens were acquired through different vendors from geographically diverse locations under Institutional Review Board approval. Residual vials were obtained 2-5 weeks (average storage time was 3 weeks) following collection and storage at room temperature. 32 NILM, 32 LSIL, and 32 HSIL cervical cytology samples (96 total) were processed for these experiments. 12 Normal, 31 CINI, 26 CINIII, and 27 squamous cell carcinoma biopsy samples (96 total) were processed for these experiments.

Follow-up biopsies were performed on a select number of cytology samples. Results can be found in Table 14.

Table 14. Follow-up biopsy information on select number of cervical cytology samples.

High-risk HPV DNA genotyping

HPV genotype was determined by real-time PCR amplification of DNA isolated from cervical biopsy and cytology specimens using HPV type-specific primers and probes. Primers and 6-F AM-MGB probes targeting the E6/E7 genomic regions were designed with Primer Express Software version 3.0 (Table 16) and purchased from Applied Biosystems (Foster City, CA). Fifteen high risk HPV types (16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 68, 73, and 82) and 2 low risk types (6 and 11) were analyzed. Duplicate reactions containing 900 nM each primer, 250 nM probe and DNA at a final concentration of 83 ng/uL were performed for each virus type. Amplification was performed on a 7900HT Real Time PCR System (Applied Biosystems). The amplification conditions were: 2 minutes at 50⁰C, 10 minutes at 95°C, and a two-step cycle of 95°C for 15 seconds and 60⁰C for 60 seconds, for a total of 40 cycles. DNA adequacy was determined by amplification of B-globin and RNaseP genomic sequences. Specimens were called positive for a specific virus type if both of the duplicate amplification reactions resulted in C_x values <40.

RNA Isolation

Total RNA was isolated from both biopsy and cytology specimens using modified versions of the MasterPure™ RNA purification Kit (Epicentre® Biotechnologies, Madison, WI, USA), as described by Murphy et al. (2009) J Virol Methods 156 (1-2): 138-44. Briefly, the FFPE protocol was modified as follows. Ten 10 μm sections cut on a microtome were combined with 500 μl xylene (EMD, Darmstadt, Germany), mixed vigorously for 10 seconds, and centrifuged for five minutes at full speed. The xylene was removed and the sections were washed two times with 300 μl 100% ethanol (Sigma- Aldrich, St. Louis, MO, USA), and allowed to air dry. 60 μl of 50 μg/μl Proteinase K was diluted into 540 μl of Tissue and Cell Lysis Solution, added to the pellet and incubated over night at 6O⁰C. The mixture was then incubated on ice for 3-5 minutes. To precipitate total protein, 300 μl of MPC Protein Precipitation Reagent was added to the sample followed by mixing and centrifugation at 10,000 xg for 10 minutes. One volume of isopropyl alcohol (Sigma- Aldrich) was added and total nucleic acid was precipitated by mixing and centrifugation at 4⁰C and 10,000 xg for 10 minutes. Samples were then washed one time with 500 μl 70% ethanol and allowed to air dry. To recover a total nucleic-acid fraction, the pellet was resuspended in 40 μl DEPC water (EMD) and 5 μl transferred to 25 μl of DEPC water. The remaining 35 uL of sample was then treated with DNase and purified as described in the manufacturer's protocol. Purified RNA was resuspended in 35 μl DEPC water. Following RNA isolation, a final DNase treatment was performed using Baseline-ZERO™ DNase (Epicentre® Biotechnologies) according to the manufacturer's directions.

The cytology protocol was modified as follows. Samples were removed from the residual vial to a 15 ml centrifuge tube. The vial was then washed with 4 ml SurePath® (TriPath Imaging, Inc.) preservative and the wash was also added to the centrifuge tube. The cells were pelleted at 1,800 xg, washed with Tris-Buffered water, and pelleted at 2,200 xg. The remaining protocol was performed exactly as described above for the biopsy samples, beginning with the proteinase K treatment.

RT-PCR

RNA was reverse transcribed using the High capacity cDNA Reverse Transcription Kit from Applied Biosystems (Foster City, CA, USA). 1 μg RNA per 20 ul reaction mix was treated using the manufacturer's recommended reaction conditions. Real Time PCR was performed on an Applied Biosystems 7900 RT-PCR instrument using Express qPCR Supermix with Rox from Invitrogen (Carlsbad, California, USA).

Primers and probes were designed to detect either the E6 or E7 transcript of HPV 16 and 18 and only the E6 transcript for HPV31, 33 and 45. Primers and probes are listed in Table 15. All assays were tested for optimal efficiency. The housekeeping genes GUSB and POLR2A were detected with inventoried assays from Applied Biosystems (Hs99999908_ml and HsOOl 72187_ml respectively). The GUSB assay crosses exon boundary 11-12 in GUSB GenBank Accession No. NM_000181.3 and the POLR2A assay crosses exon boundary 1-2 in POLR2A GenBank Accession No. NM 000937.3. 50 ng/well of cDNA in a 12 μl reaction mix was processed using standard cycling parameters as described in the manufacturer's protocol. Control reactions containing non-reverse transcribed RNA as a template were performed for each specimen. These "minus-RT" reactions were used to discriminate between amplification of HPV genomic DNA and the HPV E6 and E7 mRNA's. This is particularly important because one of the mRNA products encoded by the HPV E6 gene does not contain introns and cannot otherwise be distinguished from genomic DNA.

The readout for a real time PCR assay is the threshold cycle or Ci. This is the PCR cycle where a .statistically .significant increase in PCR product concentration Ls first detected. There is a linear relationship between C_f value and the starting concentration of O\A in the reaction mixture.

Table 15. Primers and probes used in quantitative real-time PCR assays.

4-

Table 1 o. Primers and probes used for viral genoτyping.

Data Analysis

Data was analyzed in Microsoft Excel (Redmond, WΛ, USΛ) and reported in five different ways: raw Cx values, comparative CV absolute quantification, relative standard curve method and the normalized target value. Data points in which the standard deviation between replicate amplification reactions was >j ,0 were considered invalid and were not used in further anaylsis. Additionally, if the i\ value of the "'mimus-R P' control reaction was equal to or less than the C_t value obtained for the corresponding test reaction, the data point was assigned a <\ value of 40 and as a result an raRΛΛ quantity of zero. For the comparative C] method, the endogenous control C] for a particular sample was subtracted from the target gene C\ to gel tbe delta C_T. The average delta CT of me normal samples from the experiment was then subtracted from the delta C₁ for each target gerse to gel tbe delta-delta C₁. The number 2 was raised to the power of the negative delta-delta Cj to get the fold over-expression of each target gene relative to the average of the normal specimens (WXL for biopsy and NILM for cytology).

For absolute quantification, standard curves for each assay were created using five plasmids encoding the E6 and E7 genes for all five virus types tested (Genscript, Piscataway, NJ, USA) as well as two plasmids encoding either the GUSB or POLR2A genes (Open Biosystems, now part of Thermo Scientific, Huntsville, AL. USA ). Standard concentrations were from 10 to 1 million copies per reaction in six steps, and the concentration of a particular target or endogenous control transcript was interpolated from the equation for tbe line of each respective standard curve as deiemdrsed by linear regression, If amplification products were quantified from tbe '"minus-RI'^"1 controls due to the presence of contaminating genomic DNA. these values were subtracted from the copy number determined for the experimental reactions, thus ensuring tbe specific quantification of HPV E 6 or E 7 messenger RNA.

For tbe relative standard curve method, the standard curve-derived copy number of the target transcript (after subtraction of signal derived from the '"minus- RT reaction ) was divided by the copy number of the endogenous control transcript to get tbe normalized target value (N T V). "1 his value was then divided by the average normalized target value for all of the normal specimens to derive the fold-difference in target expression relative to the average of the normal samples.

The normalized target value from the relative standard curve method was also reported separately. All publications and patent applications mentioned in the specification are indicative of the level of those skilled in the art to which this invention pertains. All publications and patent applications are herein incorporated by reference to the same extent as if each individual publication or patent application was specifically and individually indicated to be incorporated by reference.

Many modifications and other embodiments of the inventions set forth herein will come to mind to one skilled in the art to which these inventions pertain having the benefit of the teachings presented in the foregoing descriptions and the associated drawings. Therefore, it is to be understood that the inventions are not to be limited to the specific embodiments disclosed and that modifications and other embodiments are intended to be included within the scope of the appended claims. Although specific terms are employed herein, they are used in a generic and descriptive sense only and not for purposes of limitation.

Claims

THAT WHICH IS CLAIMED:

1. A method for diagnosing cervical disease in a subject comprising obtaining a body sample from said subject and determining expression levels of a human papillomavirus (HPV) E6 mRNA, E7 mRNA, or both HPV E6 and HPV E7 mRNA from a high-risk HPV type, wherein the expression level of said HPV E6 mRNA above a first threshold amount; the expression level of said HPV E7 mRNA above a second threshold amount; or the highest expression level of either said HPV E6 mRNA or said HPV E7 mRNA above a third threshold amount is indicative of the presence of cervical disease in said subject or an increased likelihood of said subject developing cervical disease when compared to a control subject.

2. The method of claim 1 , wherein determining expression levels of said HPV E6, HPV E7, or both HPV E6 and HPV E7 mRNA comprises real-time polymerase chain reaction.

3. The method of claim 2, wherein the expression level of HPV E6 or E7 mRNA is defined as a threshold cycle (C_T), and wherein said C_T is negatively correlated with HPV E6 and HPV E7 expression level.

4. The method of claim 3, wherein said body sample comprises a cervical cytology sample.

5. The method of claim 4, wherein a HPV E6 C_x, a HPV E7 C_x, or the lowest of a HPV E6 C_x or a HPV E7 C_x below about 38 is indicative of the presence of cervical disease in said subject or an increased likelihood of said subject developing cervical disease when compared to a control subject.

6. The method of claim 5, wherein a HPV E6 C_x, a HPV E7 C_x, or the lowest of a HPV E6 C_x or a HPV E7 C_x of about 35 to about 38 is indicative of the presence of low-grade cervical disease in said subject or an increased likelihood of said subject developing low-grade cervical disease.

7. The method of claim 5, wherein a HPV E6 C_T, a HPV E7 C_T, or the lowest of a HPV E6 C_T or a HPV E7 C_T below about 56 is indicative of the presence of high- grade cervical disease in said subject or an increased likelihood of said subject developing high-grade cervical disease.

8. The method of claim 3, wherein said body sample comprises a cervical biopsy sample.

9. The method of claim 8, wherein a HPV E6 C_x, a HPV E7 C_x, or the lowest of a HPV E6 C_T or a HPV E7 C_T below about 38 is indicative of the presence of cervical disease in said subject or an increased likelihood of said subject developing cervical disease when compared to a control subject.

10. The method of claim 9, wherein a HPV E6 C_x, a HPV E7 C_x, or the lowest of a HPV E6 C_x or a HPV E7 C_x of about 35 to about 38 is indicative of the presence of low-grade cervical disease in said subject or an increased likelihood of said subject developing low-grade cervical disease.

11. The method of claim 9, wherein a HPV E6 C_x, a HPV E7 C_x, or the lowest of a HPV E6 C_x or a HPV E7 C_x below about 35 is indicative of the presence of high- grade cervical disease in said subject or an increased likelihood of said subject developing high-grade cervical disease.

12. The method of claim 2, wherein the expression level of HPV E6 or E7 mRNA is defined as an absolute copy number.

13. The method of claim 12, wherein said body sample comprises a cervical cytology sample.

14. The method of claim 13, wherein said first threshold amount, said second threshold amount, or said third threshold amount is an absolute copy number of about 1.

15. The method of claim 14, wherein a HPV E6 absolute copy number, a HPV E7 absolute copy number, or the highest of a HPV E6 absolute copy number or a HPV E7 absolute copy number of about 1 to about 10 is indicative of the presence of low-grade cervical disease in said subject or an increased likelihood of said subject developing low- grade cervical disease.

16. The method of claim 14, wherein a HPV E6 absolute copy number, a HPV E7 absolute copy number, or the highest of a HPV E6 absolute copy number or a HPV E7 absolute copy number above about 10 is indicative of the presence of high-grade cervical disease in said subject or an increased likelihood of said subject developing high-grade cervical disease.

17. The method of claim 12, wherein said body sample comprises a cervical biopsy sample.

18. The method of claim 17, wherein said first threshold amount or said third threshold amount is an absolute copy number of about 10.

19. The method of claim 18, wherein a HPV E6 absolute copy number, or the highest of a HPV E6 absolute copy number or a HPV E7 absolute copy number of about

10 to about 50 is indicative of the presence of low-grade cervical disease in said subject or an increased likelihood of said subject developing low-grade cervical disease.

20. The method of claim 18, wherein a HPV E6 absolute copy number, or the highest of a HPV E6 absolute copy number or a HPV E7 absolute copy number above about 50 is indicative of the presence of high-grade cervical disease in said subject or an increased likelihood of said subject developing high-grade cervical disease.

21. The method of claim 17, wherein said second threshold amount is an absolute copy number of about 1.

22. The method of claim 21 , wherein a HPV E7 absolute copy number of about 1 to about 10 is indicative of the presence of low-grade cervical disease in said subject or an increased likelihood of said subject developing low-grade cervical disease.

23. The method of claim 21 , wherein a HPV E7 absolute copy number above about 10 is indicative of the presence of high-grade cervical disease in said subject or an increased likelihood of said subject developing high-grade cervical disease.

24. The method of claim 2, wherein the expression level of HPV E6 or E7 mRNA is defined as a normalized copy number, wherein the normalized copy number is an absolute copy number of HPV E6 or E7 mRNA divided by an absolute copy number of an endogenous control gene, multiplied by 1000.

25. The method of claim 24, wherein said body sample comprises a cervical cytology sample.

26. The method of claim 25, wherein said endogenous control gene is selected from the group consisting of peptidylprolyl isomerase A (PPIA), beta-2 -microglobulin

(B2M), phosphoglycerate kinase 1 (PGKl), large ribosomal protein (RPLPO), ubiquitin C (UBC), beta-glucuronidase (GUSB), and polymerase (RNA) II (DNA directed) polypeptide A (POLR2A).

27. The method of claim 26, wherein said endogenous control gene is GUSB.

28. The method of claim 27, wherein said first threshold amount or said second threshold amount is a normalized copy number of about 1.

29. The method of claim 28, wherein a HPV E6 normalized copy number or a

HPV E7 normalized copy number of about 1 to about 20 is indicative of the presence of low-grade cervical disease in said subject or an increased likelihood of said subject developing low-grade cervical disease.

30. The method of claim 28, wherein a HPV E6 normalized copy number or a

HPV E7 normalized copy number above about 20 is indicative of the presence of high- grade cervical disease in said subject or an increased likelihood of said subject developing high-grade cervical disease.

31. The method of claim 27, wherein said third threshold amount is a normalized copy number of about 1.

32. The method of claim 31 , wherein the highest of a HPV E6 normalized copy number or a HPV E7 normalized copy number of about 1 to about 100 is indicative of the presence of low-grade cervical disease in said subject or an increased likelihood of said subject developing low-grade cervical disease.

33. The method of claim 31 , wherein the highest of a HPV E6 normalized copy number or a HPV E7 normalized copy number above about 100 is indicative of the presence of high-grade cervical disease in said subject or an increased likelihood of said subject developing high-grade cervical disease.

34. The method of claim 26, wherein said endogenous control gene is POLR2A.

35. The method of claim 34, wherein said first threshold amount or said second threshold amount is a normalized copy number of about 1.

36. The method of claim 35, wherein a HPV E6 normalized copy number or a

HPV E7 normalized copy number of about 1 to about 10 is indicative of the presence of low-grade cervical disease in said subject or an increased likelihood of said subject developing low-grade cervical disease.

37. The method of claim 35, wherein a HPV E6 normalized copy number or a

HPV E7 normalized copy number above about 10 is indicative of the presence of high- grade cervical disease in said subject or an increased likelihood of said subject developing high-grade cervical disease.

38. The method of claim 34, wherein said third threshold amount is a normalized copy number of about 1.

39. The method of claim 38, wherein the highest of a HPV E6 normalized copy number or a HPV E7 normalized copy number of about 1 to about 20 is indicative of the presence of low-grade cervical disease in said subject or an increased likelihood of said subject developing low-grade cervical disease.

40. The method of claim 38, wherein the highest of a HPV E6 normalized copy number or a HPV E7 normalized copy number above about 20 is indicative of the presence of high-grade cervical disease in said subject or an increased likelihood of said subject developing high-grade cervical disease.

41. The method of claim 24, wherein said body sample comprises a cervical biopsy sample.

42. The method of claim 41, wherein said endogenous control gene is selected from the group consisting of tyrosine 3-monooxygenase/tryptophan 5-monooxygenase activation protein, zeta polypeptide (YWHAZ), hydroxymethylbilane synthase (HMBS), transferrin receptor (TFRC), peptidylprolyl isomerase A (PPIA), importin 8 (IPO8), beta- 2-microglobulin (B2M), phosphoglycerate kinase 1 (PGKl), large ribosomal protein (RPLPO), ubiquitin C (UBC), beta-glucuronidase (GUSB), and polymerase (RNA) II (DNA directed) polypeptide A (POLR2A).

43. The method of claim 42, wherein said endogenous control gene is GUSB.

44. The method of claim 43, wherein said first threshold amount or said third threshold amount is a normalized copy number of about 1.

45. The method of claim 44, wherein a HPV E6 normalized copy number, or the highest of a HPV E6 normalized copy number or a HPV E7 normalized copy number of about 1 to about 20 is indicative of the presence of low-grade cervical disease in said subject or an increased likelihood of said subject developing low-grade cervical disease.

46. The method of claim 44, wherein a HPV E6 normalized copy number, or the highest of a HPV E6 normalized copy number or a HPV E7 normalized copy number above about 20 is indicative of the presence of high-grade cervical disease in said subject or an increased likelihood of said subject developing high-grade cervical disease.

47. The method of claim 43, wherein said second threshold amount is a normalized copy number of about 0.1.

48. The method of claim 47, wherein a HPV E7 normalized copy number of about 0.1 to about 5 is indicative of the presence of low-grade cervical disease in said subject or an increased likelihood of said subject developing low-grade cervical disease.

49. The method of claim 47, wherein a HPV E7 normalized copy number above about 5 is indicative of the presence of high-grade cervical disease in said subject or an increased likelihood of said subject developing high-grade cervical disease.

50. The method of claim 42, wherein said endogenous control gene is POLR2A.

51. The method of claim 50, wherein said first threshold amount, said second threshold amount, or said third threshold amount is a normalized copy number of about 1.

52. The method of claim 51 , wherein a HPV E6 normalized copy number, a HPV E7 normalized copy number, or the highest of a HPV E6 normalized copy number or a HPV E7 normalized copy number of about 1 to about 10 is indicative of the presence of low-grade cervical disease in said subject or an increased likelihood of said subject developing low-grade cervical disease.

53. The method of claim 51 , wherein a HPV E6 normalized copy number, a HPV E7 normalized copy number, or the highest of a HPV E6 normalized copy number or a HPV E7 normalized copy number above about 10 is indicative of the presence of high- grade cervical disease in said subject or an increased likelihood of said subject developing high-grade cervical disease.

54. The method of claim 2, wherein the expression level of HPV E6 or E7 mRNA in said body sample is expressed as a fold-change of a normalized HPV E6 or E7 copy number of said body sample relative to a normalized HPV E6 or E7 copy number of a normal patient body sample, wherein the normalized HPV E6 or E7 copy number is derived by dividing a HPV E6 absolute copy number or HPV E7 absolute copy number by an endogenous control absolute copy number, and wherein the fold-change of the normalized HPV E6 or E7 copy number of said body sample relative to the normalized HPV E6 or E7 copy number of a normal patient body sample is derived by dividing the normalized HPV E6 or E7 copy number of said body sample by the normalized HPV E6 or E7 copy number of a normal patient body sample.

55. The method of claim 54, wherein said normalized HPV E6 or E7 copy number of a normal patient body sample is an average normalized HPV E6 or E7 copy number of body samples from a population of normal patients.

56. The method of claim 54 or 55, wherein said body sample comprises a cervical cytology sample.

57. The method of claim 56, wherein said endogenous control gene is selected from the group consisting of peptidylprolyl isomerase A (PPIA), beta-2 -microglobulin

(B2M), phosphoglycerate kinase 1 (PGKl), large ribosomal protein (RPLPO), ubiquitin C (UBC), beta-glucuronidase (GUSB), and polymerase (RNA) II (DNA directed) polypeptide A (POLR2A).

58. The method of claim 57, wherein said endogenous control gene is GUSB.

59. The method of claim 58, wherein said first threshold amount is a fold- change of a normalized HPV E6 copy number of said body sample relative to a normalized HPV E6 copy number of a normal patient body sample of about 1.

60. The method of claim 59, wherein said fold-change of a normalized HPV E6 copy number of said body sample relative to a normalized HPV E6 copy number of a normal patient body sample of about 1 to about 10 is indicative of the presence of low- grade cervical disease in said subject or an increased likelihood of said subject developing low-grade cervical disease.

61. The method of claim 59, wherein said fold-change of a normalized HPV E6 copy number of said body sample relative to a normalized HPV E6 copy number of a normal patient body sample above about 10 is indicative of the presence of high-grade cervical disease in said subject or an increased likelihood of said subject developing high- grade cervical disease.

62. The method of claim 58, wherein said second threshold amount is a fold- change of a normalized HPV E7 copy number of said body sample relative to a normalized HPV E7 copy number of a normal patient body sample of about 1.

63. The method of claim 62, wherein said fold-change of a normalized HPV E7 copy number of said body sample relative to a normalized HPV E7 copy number of a normal patient body sample of about 1 to about 4 is indicative of the presence of low- grade cervical disease in said subject or an increased likelihood of said subject developing low-grade cervical disease.

64. The method of claim 62, wherein said fold-change of a normalized HPV E7 copy number of said body sample relative to a normalized HPV E7 copy number of a normal patient body sample above about 4 is indicative of the presence of high-grade cervical disease in said subject or an increased likelihood of said subject developing high- grade cervical disease.

65. The method of claim 58, wherein said third threshold amount is the highest of a fold-change of a normalized HPV E6 copy number of said body sample relative to a normalized HPV E6 copy number of a normal patient body sample or a fold-change of a normalized HPV E7 copy number of said body sample relative to a normalized HPV E7 copy number of a normal patient body sample of about 1.

66. The method of claim 65, wherein said fold-change of a normalized HPV E6 copy number of said body sample relative to a normalized HPV E6 copy number of a normal patient body sample or a fold-change of a normalized HPV E7 copy number of said body sample relative to a normalized HPV E7 copy number of a normal patient body sample of about 1 to about 10 is indicative of the presence of low-grade cervical disease in said subject or an increased likelihood of said subject developing low-grade cervical disease.

67. The method of claim 65, wherein said fold-change of a normalized HPV E6 copy number of said body sample relative to a normalized HPV E6 copy number of a normal patient body sample or a fold-change of a normalized HPV E7 copy number of said body sample relative to a normalized HPV E7 copy number of a normal patient body sample above about 10 is indicative of the presence of high-grade cervical disease in said subject or an increased likelihood of said subject developing high-grade cervical disease.

68. The method of claim 57, wherein said endogenous control gene is POLR2A.

69. The method of claim 68, wherein said first threshold amount is a fold- change of a normalized HPV E6 copy number of said body sample relative to a normalized HPV E6 copy number of a normal patient body sample of about 1.

70. The method of claim 69, wherein said fold-change of a normalized HPV E6 copy number of said body sample relative to a normalized HPV E6 copy number of a normal patient body sample of about 1 to about 10 is indicative of the presence of low- grade cervical disease in said subject or an increased likelihood of said subject developing low-grade cervical disease.

71. The method of claim 69, wherein said fold-change of a normalized HPV E6 copy number of said body sample relative to a normalized HPV E6 copy number of a normal patient body sample above about 10 is indicative of the presence of high-grade cervical disease in said subject or an increased likelihood of said subject developing high- grade cervical disease.

72. The method of claim 68, wherein said second threshold amount is a fold- change of a normalized HPV E7 copy number of said body sample relative to a normalized HPV E7 copy number of a normal patient body sample of about 1.

73. The method of claim 72, wherein said fold-change of a normalized HPV E7 copy number of said body sample relative to a normalized HPV E7 copy number of a normal patient body sample of about 1 to about 5 is indicative of the presence of low- grade cervical disease in said subject or an increased likelihood of said subject developing low-grade cervical disease.

74. The method of claim 72, wherein said fold-change of a normalized HPV E7 copy number of said body sample relative to a normalized HPV E7 copy number of a normal patient body sample above about 5 is indicative of the presence of high-grade cervical disease in said subject or an increased likelihood of said subject developing high- grade cervical disease.

75. The method of claim 68, wherein said third threshold amount is the highest of a fold-change of a normalized HPV E6 copy number of said body sample relative to a normalized HPV E6 copy number of a normal patient body sample or a fold-change of a normalized HPV E7 copy number of said body sample relative to a normalized HPV E7 copy number of a normal patient body sample of about 1.

76. The method of claim 75, wherein the highest of a fold-change of a normalized HPV E6 copy number of said body sample relative to a normalized HPV E6 copy number of a normal patient body sample or a fold-change of a normalized HPV E7 copy number of said body sample relative to a normalized HPV E7 copy number of a normal patient body sample of about 1 to about 10 is indicative of the presence of low- grade cervical disease in said subject or an increased likelihood of said subject developing low-grade cervical disease.

77. The method of claim 75, wherein the highest of a fold-change of a normalized HPV E6 copy number of said body sample relative to a normalized HPV E6 copy number of a normal patient body sample or a fold-change of a normalized HPV E7 copy number of said body sample relative to a normalized HPV E7 copy number of a normal patient body sample above about 10 is indicative of the presence of high-grade cervical disease in said subject or an increased likelihood of said subject developing high- grade cervical disease.

78. The method of claim 2, wherein the expression level of HPV E6 or E7 mRNA is defined as a comparative C_T value, wherein said comparative C_T value is 2^{" ΔΔCT} , wherein ΔΔC_T has been derived by subtracting a ΔC_T of a normal patient body sample from a ΔC_T of said body sample, wherein the ΔC_T of said body sample has been derived by subtracting an endogenous control C_T of said body sample from a HPV E6 or E7 C_T of said body sample, and wherein the ΔC_T of a normal patient body sample has been derived by subtracting an endogenous control C_T of said normal patient body sample from a HPV E6 or E7 C_T of said normal patient body sample.

79. The method of claim 78, wherein said ΔC_T of a normal patient body sample is an average ΔC_T from a population of normal patients.

80. The method of claim 78 or 79, wherein said body sample comprises a cervical cytology sample.

81. The method of claim 80, wherein said endogenous control gene is selected from the group consisting of peptidylprolyl isomerase A (PPIA), beta-2 -microglobulin

(B2M), phosphoglycerate kinase 1 (PGKl), large ribosomal protein (RPLPO), ubiquitin C (UBC), beta-glucuronidase (GUSB), and polymerase (RNA) II (DNA directed) polypeptide A (POLR2A).

82. The method of claim 81, wherein said endogenous control gene is GUSB.

83. The method of claim 82, wherein said first threshold amount, said second threshold amount, or said third threshold amount is a comparative C_T value of about 2.

84. The method of claim 83, wherein a HPV E6 comparative C_T value, a HPV E7 comparative C_T value, or the highest of a HPV E6 comparative C_T value or a HPV E7 comparative C_T value of about 2 to about 5 is indicative of the presence of low-grade cervical disease in said subject or an increased likelihood of said subject developing low- grade cervical disease.

85. The method of claim 83, wherein a HPV E6 comparative C_T value, a HPV E7 comparative C_T value, or the highest of a HPV E6 comparative C_T value or a HPV E7 comparative C_T value above about 5 is indicative of the presence of high-grade cervical disease in said subject or an increased likelihood of said subject developing high-grade cervical disease.

86. The method of claim 81 , wherein said endogenous control gene is POLR2A.

87. The method of claim 86, wherein said first threshold amount is a comparative C_T value of about 2.

88. The method of claim 87, wherein a HPV E6 comparative C_T value of about 2 to about 5 is indicative of the presence of low-grade cervical disease in said subject or an increased likelihood of said subject developing low-grade cervical disease.

89. The method of claim 87, wherein a HPV E6 comparative C_T value above about 5 is indicative of the presence of high-grade cervical disease in said subject or an increased likelihood of said subject developing high-grade cervical disease.

90. The method of claim 86, wherein said second threshold amount or said third threshold amount is a comparative C_T value of about 2.

91. The method of claim 90, wherein a HPV E7 comparative C_T value, or the highest of a HPV E6 comparative C_T value or a HPV E7 comparative C_T value of about 2 to about 10 is indicative of the presence of low-grade cervical disease in said subject or an increased likelihood of said subject developing low-grade cervical disease.

92. The method of claim 90, wherein a HPV E7 comparative C_T value, or the highest of a HPV E6 comparative C_T value or a HPV E7 comparative C_T value about about 10 is indicative of the presence of high-grade cervical disease in said subject or an increased likelihood of said subject developing high-grade cervical disease.

93. The method of claim 78 or 79, wherein said body sample comprises a biopsy sample.

94. The method of claim 93, wherein said endogenous control gene is selected from the group consisting of tyrosine 3-monooxygenase/tryptophan 5-monooxygenase activation protein, zeta polypeptide (YWHAZ), hydroxymethylbilane synthase (HMBS), transferrin receptor (TFRC), peptidylprolyl isomerase A (PPIA), importin 8 (IPO8), beta- 2-microglobulin (B2M), phosphoglycerate kinase 1 (PGKl), large ribosomal protein (RPLPO), ubiquitin C (UBC), beta-glucuronidase (GUSB), and polymerase (RNA) II (DNA directed) polypeptide A (POLR2 A).

95. The method of claim 94, wherein said endogenous control gene is GUSB.

96. The method of claim 95, wherein said first threshold amount or said third threshold amount is a comparative C_T value of about 2.

97. The method of claim 96, wherein a HPV E6 comparative C_T value, or the highest of a HPV E6 comparative C_T value or a HPV E7 comparative C_T value of about 2 to about 15 is indicative of the presence of low-grade cervical disease in said subject or an increased likelihood of said subject developing low-grade cervical disease.

98. The method of claim 96, wherein a HPV E6 comparative C_T value, or the highest of a HPV E6 comparative C_T value or a HPV E7 comparative C_T value about about 15 is indicative of the presence of high-grade cervical disease in said subject or an increased likelihood of said subject developing high-grade cervical disease.

99. The method of claim 95, wherein said second threshold amount is a comparative C_T value of about 2.

100. The method of claim 99, wherein a HPV E7 comparative C_T value of about 2 to about 10 is indicative of the presence of low-grade cervical disease in said subject or an increased likelihood of said subject developing low-grade cervical disease.

101. The method of claim 99, wherein a HPV E7 comparative C_T value above about 10 is indicative of the presence of high-grade cervical disease in said subject or an increased likelihood of said subject developing high-grade cervical disease.

102. The method of claim 94, wherein said endogenous control gene is POLR2A.

103. The method of claim 102, wherein said first threshold amount or said third threshold amount is a comparative C_T value of about 2.

104. The method of claim 103, wherein a HPV E6 comparative C_T value, or the highest of a HPV E6 comparative C_T value or a HPV E7 comparative C_T value of about 2 to about 20 is indicative of the presence of low-grade cervical disease in said subject or an increased likelihood of said subject developing low-grade cervical disease.

105. The method of claim 103, wherein a HPV E6 comparative C_T value, or the highest of a HPV E6 comparative C_T value or a HPV E7 comparative C_T value above about 20 is indicative of the presence of high-grade cervical disease in said subject or an increased likelihood of said subject developing high-grade cervical disease.

106. The method of claim 102, wherein said second threshold amount is a comparative C_T value of about 2.

107. The method of claim 106, wherein a HPV E7 comparative C_T value of about 2 to about 15 is indicative of the presence of low-grade cervical disease in said subject or an increased likelihood of said subject developing low-grade cervical disease.

108. The method of claim 106, wherein a HPV E7 comparative C_T value above about 10 is indicative of the presence of high-grade cervical disease in said subject or an increased likelihood of said subject developing high-grade cervical disease.

109. A method for normalizing a target gene expression level in a cervical sample comprising determining a target gene expression level and an endogenous control gene expression level in a cervical sample, and dividing or subtracting the endogenous control gene expression level from the target gene expression level, wherein said endogenous control gene is selected from the group consisting of tyrosine 3- monooxygenase/tryptophan 5-monooxygenase activation protein, zeta polypeptide (YWHAZ), hydroxymethylbilane synthase (HMBS), transferrin receptor (TFRC), peptidylprolyl isomerase A (PPIA), importin 8 (IPO8), beta-2-microglobulin (B2M), phosphoglycerate kinase 1 (PGKl), large ribosomal protein (RPLPO), ubiquitin C (UBC), beta-glucuronidase (GUSB), and polymerase (RNA) II (DNA directed) polypeptide A (POLR2A).

110. A method for diagnosing cervical disease in a subject comprising obtaining a body sample from said subject, determining a HPV E6 or HPV E 7 expression level from a high-risk HPV type in said body sample, wherein the HPV E6 or HPV E7 expression level is normalized against an endogenous control gene expression level to produce a normalized HPV E6 or HPV E7 value, wherein an increase in the normalized HPV E6 or E7 value in said subject compared to a normal subject is indicative of the presence of cervical disease or an increased likelihood of said subject developing cervical disease when compared to said normal subject, wherein the endogenous control gene is selected from the group consisting of tyrosine 3-monooxygenase/tryptophan 5-monooxygenase activation protein, zeta polypeptide (YWHAZ), hydroxymethylbilane synthase (HMBS), transferrin receptor (TFRC), peptidylprolyl isomerase A (PPIA), importin 8 (IPO8), beta- 2-microglobulin (B2M), phosphoglycerate kinase 1 (PGKl), large ribosomal protein (RPLPO), ubiquitin C (UBC), beta-glucuronidase (GUSB), and polymerase (RNA) II (DNA directed) polypeptide A (POLR2A).

111. The method of claim 109 or 110, wherein said endogenous control gene is POLR2A.

112. The method of claim 109 or 110, wherein said endogenous control gene is GUSB.

113. The method of any one of claims 109- 112, wherein determining the HPV E6 or E7 expression level and said endogenous control gene expression level comprises real-time polymerase chain reaction.

114. The method of any one of claims 2-108 and 113, wherein said real-time quantitative polymerase chain reaction uses a fluorogenic/quencher oligonucleotide probe.

115. The method of any one of claims 1-114, wherein said high-risk HPV type is selected from the group consisting of HPV 16, HPV 18, HPV 26, HPV 31, HPV 33, HPV 35, HPV 39, HPV 45, HPV 51, HPV 52, HPV 53, HPV 56, HPV 58, HPV 59, HPV 68, HPV 73, and HPV 82.

116. The method of claim 115, wherein said high-risk HPV type is selected from the group consisting of HPV 16, HPV 18, HPV 31, HPV 33, and HPV 45.

117. The method of claim 116, wherein said high-risk HPV type is selected from the group consisting of HPV 16 and HPV 18.

118. The method of claim 117, wherein the expression level of HPV 16 E6 is determined using real-time quantitative polymerase chain reaction and a primer pair and fluorogenic/quencher oligonucleotide probe used to amplify and detect HPV 16 E6 having the nucleotide sequences selected from the group consisting of: a) a primer pair having the nucleotide sequences set forth in SEQ ID NOs: 1 and 2 and a fluorogenic/quencher oligonucleotide probe having the nucleotide sequence set forth in SEQ ID NO: 3; b) a primer pair having the nucleotide sequences set forth in SEQ ID NOs: 4 and 5 and a fluorogenic/quencher oligonucleotide probe having the nucleotide sequence set forth in SEQ ID NO: 6; and, c) a primer pair having the nucleotide sequences set forth in SEQ ID NOs: 7 and 8 and a fluorogenic/quencher oligonucleotide probe having the nucleotide sequence set forth in SEQ ID NO: 9.

119. The method of claim 117, wherein the expression level of HPV16 E7 is determined using real-time quantitative polymerase chain reaction and a primer pair having the nucleotide sequences set forth in SEQ ID NOs: 10 and 11 and a fluorogenic/quencher oligonucleotide probe having the nucleotide sequence set forth in SEQ ID NO: 12.

120. The method of claim 117, wherein the expression level of HPV18 E6 is determined using real-time quantitative polymerase chain reaction and a primer pair having the nucleotide sequences set forth in SEQ ID NOs: 13 and 14 and a fluorogenic/quencher oligonucleotide probe having the nucleotide sequence set forth in SEQ ID NO: 15.

121. The method of claim 117, wherein the expression level of HPV 18 E7 is determined using real-time quantitative polymerase chain reaction and a primer pair having the nucleotide sequences set forth in SEQ ID NOs: 16 and 17 and a fluorogenic/quencher oligonucleotide probe having the nucleotide sequence set forth in SEQ ID NO: 18.

122. The method of claim 116, wherein the expression level of HPV31 E6 is determined using real-time quantitative polymerase chain reaction and a primer pair having the nucleotide sequences set forth in SEQ ID NOs: 19 and 20 and a fluorogenic/quencher oligonucleotide probe having the nucleotide sequence set forth in SEQ ID NO: 21.

123. The method of claim 116, wherein the expression level of HPV33 E6 is determined using real-time quantitative polymerase chain reaction and a primer pair having the nucleotide sequences set forth in SEQ ID NOs: 22 and 23 and a fluorogenic/quencher oligonucleotide probe having the nucleotide sequence set forth in SEQ ID NO: 24.

124. The method of claim 116, wherein the expression level of HPV45 E6 is determined using real-time quantitative polymerase chain reaction and a primer pair having the nucleotide sequences set forth in SEQ ID NOs: 25 and 26 and a fluorogenic/quencher oligonucleotide probe having the nucleotide sequence set forth in SEQ ID NO: 27.

125. A kit for determining HPV E6 or HPV E7 expression level in a body sample, said kit comprising a primer pair and a fluorogenic/quencher oligonucleotide probe selected from the group consisting of: a) a primer pair and a fluorogenic/quencher oligonucleotide probe to amplify and detect HPV 16 E6 having the nucleotide sequences selected from the group consisting of: i) a primer pair having the nucleotide sequences set forth in

SEQ ID NOs: 1 and 2 and a fluorogenic probe having the nucleotide sequence set forth in SEQ ID NO: 3; ii) a primer pair having the nucleotide sequences set forth in

SEQ ID NOs: 4 and 5 and a fluorogenic probe having the nucleotide sequence set forth in SEQ ID NO: 6; and, iii) a primer pair having the nucleotide sequences set forth in SEQ ID NOs: 7 and 8 and a fluorogenic probe having the nucleotide sequence set forth in SEQ ID NO: 9; b) a primer pair having the nucleotide sequences set forth in SEQ ID NOs: 10 and 11 and a fluorogenic/quencher oligonucleotide probe having the nucleotide sequence set forth in SEQ ID NO: 12 for amplifying and detecting HPV 16 E7; c) a primer pair having the nucleotide sequences set forth in SEQ ID

NOs: 13 and 14 and a fluorogenic/quencher oligonucleotide probe having the nucleotide sequence set forth in SEQ ID NO: 15 for amplifying and detecting HPV 18 E6; d) a primer pair having the nucleotide sequences set forth in SEQ ID NOs: 16 and 17 and a fluorogenic/quencher oligonucleotide probe having the nucleotide sequence set forth in SEQ ID NO: 18 for amplifying and detecting HPVl 8 E7; e) a primer pair having the nucleotide sequences set forth in SEQ ID NOs: 19 and 20 and a fluorogenic/quencher oligonucleotide probe having the nucleotide sequence set forth in SEQ ID NO: 21 for amplifying and detecting HPV31 E6; f) a primer pair having the nucleotide sequences set forth in SEQ ID NOs: 22 and 23 and a fluorogenic/quencher oligonucleotide probe having the nucleotide sequence set forth in SEQ ID NO: 24 for amplifying and detecting HPV33 E6; and, g) a primer pair having the nucleotide sequences set forth in SEQ ID NOs: 25 and 26 and a fluorogenic/quencher oligonucleotide probe having the nucleotide sequence set forth in SEQ ID NO: 27 for amplifying and detecting HPV45 E6.