WO2013029116A1

WO2013029116A1 - Method for predicting treatment responsiveness

Info

Publication number: WO2013029116A1
Application number: PCT/AU2012/001034
Authority: WO
Inventors: Stephen M. Jane; Charbel DARIDO
Original assignee: Monash University
Priority date: 2011-08-31
Filing date: 2012-08-31
Publication date: 2013-03-07

Abstract

The invention relates to a method for predicting if a squamous cell neoplasm in a subject will be susceptible to treatment with a PI3K inhibitor, a method for selecting a subject with a squamous cell neoplasm for PI3K inhibitor treatment, and a method for treating a squamous cell neoplasm in a subject, and to kits related to said methods.

Description

METHOD FOR PREDICTING TREATMENT RESPONSIVENESS

FIELD

The present invention relates to methods for predicting if a squamous cell neoplasm will be susceptible to treatment with phosphatidylinositol 3-kinase inhibitors and use of

phosphatidylinositol 3-kinase inhibitors in methods of treating susceptible squamous cell neoplasms.

BACKGROUND

Squamous cell neoplasms are among the most common cancers in humans and other animals. They usually arise from the outer layers of skin and mucous membrane cells, known as ectodermal or endodermal cells. For this reason, they can develop in a large number of organs and tissues, including skin, lips, mouth, esophagus, urinary bladder, prostate, lung, vagina and cervix, among others. Approximately ninety percent of cases of head and neck cancer (cancer of the mouth, nasal cavity, nasopharynx, throat and associated structures) are due to squamous cell neoplasms.

As they grow, squamous cell neoplasms can spread to lymph nodes (under the armpits, in the groin or in the neck, for example) or via the bloodstream to other parts of the body. For this reason squamous cell neoplasms should be treated promptly and the removal should be carried out by a specialized medical practitioner trained in the complete surgical removal of the cancer. Surgical removal usually involves removal of the cancer under local anesthetic, or if the cancer is larger, it may need to be removed with specialized technique involving a skin graft placed over the wound. Radiotherapy, which uses X-rays to kill the cancer cells, is an alternative in selected patients, after assessment by a dermatologist. If the cancer is fou nd to have spread to lymph nodes, these may also need to be removed, or treated with radiotherapy to delay further spread.

Despite the common name, squamous cell neoplasms arising in different body sites can show tremendous differences in their presenting signs and symptoms, natural history, prognosis, and response to treatment. Due to their primarily superficial nature, squamous cell neoplasms are often treated surgically. However, such treatment may leave behind some cancerous cells which may proliferate and spread over time. Also, the remnant cells may give rise to more aggressive and invasive forms of the cancer.

Despite their prevalence, the molecular basis of squamous cell neoplasms remains poorly understood. Studies in human tissue models have provided evidence that activation of Ras signaling, in concert with inhibition of nuclear factor kappa-light-chain-enhancer of activated B cell (NF- κΒ) function, is sufficient for malignant transformation of keratinocytes. Activated Ras stimulates multiple effectors including the Raf/MEK/ERK pathway, the phosphatidylinositol 3-kinase

(PI3K)/AKT/mTOR pathway, and the guanine nucleotide exchange factors. Each of these has been clearly implicated in the pathogenesis of keratinocyte hyperproliferation and dedifferentiation, and/or overt squamous cell neoplasms in mice, although data from primary h uman samples is less compelling, or completely lacking. In human squamous cell neoplasms, Ras activation is most frequently due to mutations in codon 12, 13 or 61 of one of the three Ras genes. However, mutations in Ras isoforms are found in only 22% of squamous cell neoplasms. An alternate mechanism for activation of PI3K/AKT/mTO signaling in squamous cell neoplasm is through loss of expression of the phosphatase and tensin homolog (PTEN) tumor suppressor gene. Loss of PTEN has been implicated in squamous cell neoplasm in both humans and mice, but, despite this, somatic mutations, gene deletions and promoter hypermethylation of PTEN have not been detected, suggesting that other mechanisms of inactivating the gene may be involved in squamous cell neoplasm pathogenesis.

Since activation of the PI3K pathway is a recurrent feature in many human neoplasms, PI3K inhibitors have been identified as promising candidates for new cancer treatments. However, there is a high degree of variability in the susceptibility of squamous cell neoplasms to treatment with PI3K inhibitors. Consequently, there is a need for a reliable method for predicting whether a squamous cell neoplasm will be susceptible to treatment with a PI3K inhibitor.

SUMMARY

A first aspect provides a method for predicting if a squamous cell neoplasm in a subject will be susceptible to treatment with a phosphatidyl inositol 3-kinase (PI3K) inhibitor, said method comprising:

a) determining the level of micro-RNA-21 (miRNA-21) and/or Grainy head-like 3 (Grhl3) in tissue from the squamous cell neoplasm; and

b) comparing the level of miRNA-21 and/or Grhl3 in tissue from the squamous cell neoplasm to a control level of miRNA-21 and/or Grhl3;

wherein an increased level of miRNA-21 and/or a decreased level of Grhl3 in tissue from the squamous cell neoplasm, compared to the control level, indicates that the squamous cell neoplasm will be susceptible to treatment with a PI3K inhibitor. A second aspect provides a method for selecting a subject with a squamous cell neoplasm for PI3K inhibitor treatment, said method comprising:

a) determining the level of miRNA-21 and/or Grhl3 in tissue from the squamous cell neoplasm;

b) comparing the level of miRNA-21 and/or Grhl3 in tissue from the squamous cell neoplasm to a control level of miRNA-21 and/or Grhl3; and

c) selecting the subject for PI3K inhibitor treatment if the tissue from the squamous cell neoplasm has an increased level of miRNA-21 and/or a decreased level of Grhl3 as compared to the control. The methods of the first and second aspects may be carried out in vitro.

A third aspect provides a method for treating a squamous cell neoplasm in a subject, said method comprising: a) determining the level of miRNA-21 and/or Grhl3 in tissue from the squamous cell neoplasm;

c) selecting the subject for PI3K inhibitor treatment if the squamous cell neoplasm has an increased level of miRNA-21 and/or a decreased level of Grhl3 as compared to the control level; and

d) administering a PI3K inhibitor to a subject selected by c). An alternative form of the third aspect involves use of a PI3K inhibitor for treating a squamous cell neoplasm in a subject, wherein the squamous cell neoplasm has an increased level of miRNA-21 and/or a decreased level of Grhl3, compared to a control level.

Another alternative form of the third aspect involves use of a PI3K inhibitor in the manufacture of a medicament for treating a squamous cell neoplasm in a subject, wherein the squamous cell neoplasm has an increased level of miRNA-21 and/or a decreased level of Grhl3, compared to a control level.

A fourth aspect provides a method for preventing a squamous cell neoplasm in a subject, said method comprising applying a topical composition comprising inositol to skin of the subject.

An alternative form of the fourth aspect involves use of a topical composition comprising inositol for preventing a squamous cell neoplasm in a subject.

Another alternative form of the fourth aspect involves use of a topical composition comprising inositol in the manufacture of a medicament for preventing a squamous cell neoplasm in a subject.

A fifth aspect provides a kit for detecting if a squamous cell neoplasm in a subject will be susceptible to treatment with a PI3K inhibitor, said kit comprising:

a) reagents for determining the level of miRNA-21 and/or Grhl3 in tissue from the squamous cell neoplasm; and

b) instructions for determining whether the squamous cell neoplasm has an increased level of miRNA-21 and/or a decreased level of Grhl3 in tissue from the squamous cell neoplasm.

A sixth aspect provides a kit for treating a squamous cell neoplasm in a subject, said kit comprising: a) reagents for determining the level of miRNA-21 and/or Grhl3 in tissue from the squamous cell neoplasm;

b) a PI3K inhibitor; and

c) instructions for determining whether the squamous cell neoplasm has an increased level of miRNA-21 and/or a decreased level of Grhl3 in tissue from the squamous cell neoplasm.

In an embodiment of the fifth and sixth aspects, the reagents for determining the level of miRNA-21 and/or Grhl3 comprise one or more of Q-RT-PCR reagents, array reagents, Northern blot reagents, Western blot reagents, and ELISA reagents. DETAILED DESCRIPTION

The work described herein demonstrates that miRNA-21 and Grhl3 are predictive biomarkers and that elevated miRNA-21 and/or decreased Grhl3 in tissue from the squamous cel l neoplasm, compared to the control level, indicates that the squamous cell neoplasm will be susceptible to treatment with a PI3K inhibitor.

The inventors have established that tissue-specific control of PTEN expression plays a pivotal role in the suppression of squamous cell neoplasms. PTEN expression is largely controlled by the transcription factor Grainy head-like 3 (Grhl3), a member of a highly conserved family of transcription factors critical for epidermal development and homeostasis across a wide range of species. The inventors found that deletion of Grhl3 in adult epidermis evokes loss of expression of PTEN, a direct GRHL3 target, resulting in activation of PI3K/(AKT)/mammalian target of rapamycin (mTOR) signaling and inducing aggressive squamous cell neoplasms. The integral function of Grhl3, coupled with its role in maintaining the balance between keratinocyte differentiation and proliferation defines Grhl3 as a critical innate surveillant to prevent skin cancer. In this regard, its function is similar to inhibitor of nuclear factor kappa-B kinase subunit alpha (IKKct), which also induces keratinocyte terminal differentiation and prevents squamous cell neoplasms. However, in contrast to ΙΚΚα-deficient tumors that exhibit epidermal growth factor receptor (EGFR)-induced extracellular signal-regulated kinase (ERK) activation, loss of Grhi3 results in exclusive upregulation of PI3K/AKT/mTOR signaling, with a complete loss of p-ERK expression and no change in the levels of p-EGFR. Restoration of PTEN expression completely abrogates squamous cell neoplasm formation. Importantly, the inventors have shown that human squamous cell neoplasms exhibit reduced levels of GRHL3 and PTEN, and that this is associated with increased expression of miRNA-21, which targets both GRHL3 and PTEN.

It is known that miRNA-21 targets PTEN in human cancers, but the inventors have shown that it also targets Grhl3. This synchronous targeting of GRHL3 and PTEN by miRNA-21 establishes a proto- oncogenic network with amplification of PI3K/AKT/mTOR signaling and induction of squamous cell carcinoma in humans. This is the first example of coordinate miRNA-mediated regulation of both a transcription factor and its direct target gene.

Through examination of a large number of primary human squamous cell neoplasms, the inventors found that expression of both GRHL3 and PTEN are reduced through upregulation of miR-21, explaining the long-standing paradox of loss of PTEN expression in the absence of genetic or epigenetic alterations to the gene. The co-ordinate targeting of both tumor suppressors by miR-21 provides an example of miRNA-dependent amplification of signaling cascades that are evident in both normal and cancerous tissues. In this setting, the synchronous regulation of the pathway inhibitor (PTEN) and its transcriptional regulator (GRHL3) by a solitary miR (miR-21) establishes a proto-oncogenic network involving enhanced PI3K/AKT/mTOR signaling in these tumors. This appears to be the first example of coordinate miRNA-mediated regulation of both transcription factor and its direct target gene for signaling amplification.

This data establishes that reduced levels of Grhl3 and/or PTEN and/or increased levels of miRNA-21 can be used to accurately predict which squamous cell neoplasms will be susceptible to treatment with compounds that inhibit PI3K/AKT/mTO signaling, or administration of Grhl3 and/or PTEN. The data also defines the Grhl3 gene as a potent suppressor of squamous cell carcinoma of the skin in mammals, acting through the direct transcriptional regulation of PTEN. Supporting this conclusion are phylogenetic, biochemical, and expression data coupled with functional rescue studies in vitro in a human system, and in vivo in the mouse.

As used herein, "subject" means any organism susceptible to developing a squamous cell neoplasm. Preferably, the subject is a mammal. The mammal may be a human, or may be a domestic, zoo, or companion animal. While it is particularly contemplated that the methods of the invention are suitable for humans, they are also applicable to veterinary treatment, including treatment of companion animals such as dogs and cats, and domestic animals such as horses, cattle and sheep, or zoo animals such as felids, canids, bovids, and ungulates and for use on laboratory anaimals including rats, mice, monkeys and apes. More preferably, the subject is a human.

As used herein "tissue" includes any tissue containing epithelial cells, such as skin tissue or mucous membrane tissue. "Tissue from a squamous cell neoplasm" as used herein means one or more cells from a squamous cell neoplasm. The cells could be isolated from the neoplasm by methods known in the art. Such methods include fine needle aspiration, core needle biopsy, vacuum assisted biopsy, large core biopsy, open surgical biopsy, shave biopsy, punch biopsy and elliptical biopsy. In other embodiments, the tissue includes tumor blood or lymphatic vessel tissue and the miRNA-21 and/or Grhl3 is measured in vessel tissue. In still other embodiments, the tissue includes fluid from the tumor and the miRNA-21 and/or Grhl3 is measured in the fluid. In still other embodiments, the tissue comprises blood, and the miRNA-21 and/or Grhl3 is measured in the blood, or in plasma or serum from the blood.

As used herein, "squamous cell neoplasm" includes a number of morphological sub-types and variants, including Bowen's disease, Marjolin's ulcer, Erythroplasia of Queyrat, head and neck cancer, papillary carcinoma, verrucous squamous cell carcinoma, papillary squamous cell carcinoma, cervical squamous cell carcinoma, squamous cell carcinoma, large cell keratinizing squamous cell carcinoma, small cell keratinizing squamous cell carcinoma, small cell squamous cell carcinoma, adenoid/pseudoglandular squamous cell carcinoma, intraepidermal squamous cell carcinoma, lymphoepithelial carcinoma, basaloid squamous cell carcinoma, clear cell squamous cell carcinoma, keratoacanthoma, signet ring cell squamous cell carcinoma and spindle cell squamous cell carcinoma. "Head and neck cancer" as used herein includes mouth cavity, nasal cavity, sinus, lip, tongue, salivary gland, nasopharynx, larynx and throat cancer. As used herein, "PI3K inhibitor" includes any compound capable of inhibiting PI3K/AKT/mTOR signaling. Examples of such compounds include A66 (C₁₇H23 502S2; CAS No.: 1166227-08-2); AS 252424 (5-[l-[5-(4-Fluoro-2-hydroxy-phenyl)-furan-2-yl]-meth-(Z)-ylidene]-thiazolidine-2,4-dione); AS-605240 (5-(6-Quinoxalinylmethylene)-2,4-thiazolidine-2,4-dione; C₁₂H₇N₃0₂S; CAS No.: 648450- 29-7); AZD6482 (C₂₂H₂₄N₄0₄; CAS No.: 1173900-33-8); BAG956 (2-methyl-2-[4-(2-methyl-8-pyridin-3- ylethynyl-imidazo[4,5-c]quinolin-l-yl)-phenyl]-propionitrile); BBD130 (2-Methyl-2-[4-(3-methyl-2- oxo-8-pyridin-3-ylethynyl-2,3-dihydro-imidazo[4,5-c]quinolin-l-yl)-phenyl]-propionitrile); BEZ235 (2- Methyl-2-[4-(3-methyl-2-oxo-8-quinolin-3-yl-2,3-dihydro-imidazo[4,5-c]quinolin-l-yl)-phenyl]- propionitrile; C₃oH₂3N₅0; CAS No.: 915019-65-7); BKM-120 (C₁₈H₂₁F₃N₅0₂; CAS No.: 1202777-78-3); CALlOl (C₂₂H₁₈FN₇0; CAS No.: 870281-82-6); D-87503 (C₁₇H₁₅N₅OS; CAS No.: 800394-83-6); D-106669 (C₁₇H₁₈N₅0; CAS No.: 938444-93-0); Deguelin ((7aS,13aS)-9,10-Dimethoxy-3,3-dimethyl-13,13a- dihydro-3H,7aH-pyrano[2,3-c;6,5-f']dichromen-7-one); demethoxyviridin; GDC-0941 (2-(lH-lndazol- 4-yl)-6-(4-methanesulfonyl-piperazin-l-yl methyl )-4-morpholin-4-yl-thieno[3,2-d]pyrimidine bismesylate; C₂₃H₂₇N₇0₃S₂; CAS No.: 957054-30-7); GSK1059615 (GSK615; 5-[[4-(4-Pyridinyl)-6- quinolinyl]methylene]-2,4-thiazolidenedione; C₁₈H N₃0₂S; CAS No.: 958852-01-1); GSK2126458 (GSK212; C ₁₇F₂N₅0₃S; CAS No.: 1086062-66-9); IC87114 (C₂₂H₁₉N₇0; CAS No.: 371242-69-2); KU- 55933 (2-Morpholin-4-yl-6-thianthren-l-yl-pyran-4-one); LY294002 (2-(4-Morpholinyl)-8-phenyl-4H- l-benzopyran-4-one; C₁₉H₁₇N0₃; CAS No.: 154447-36-6); 3-Methyladenine (3-Methyl-3H-purin-6- amine; C₅H₇N₅; CAS No.: 5142-23-4); MK-2206 (8-(4-(l-aminocyclobutyl)phenyl)-9-phenyl-8_;9- dihydro-[l,2,4]triazolo[3,4-f] [l,6]naphthyridin-3(2H)-one); myricetin (3,5,7-trihydroxy-2-(3,4,5- trihydroxyphenyl)-4H-l-benzopyran-4-one; C₁₅H₁₀O₈; CAS No.: 529-44-2); NU 7026 (2-(4- Morpholinyl)-4H-naphthol [l,2-b]pyran-4-one; C₁₇H₁₅N0₃; CAS No.: 154447-35-5); NU 7441 (8- Dibenzothiophen-4-yl-2-morpholin-4-yl-chromen-4-one); OSU-03012 (2-Amino-N-[4-[5-(2- phenanthrenyl)-3-(trifluoromethyl)-lH-pyrazol-l-yl] phenyl]-acetamide; C₂₆H₁₉F₃N₄0; CAS No.: 742112-33-0); Perifosine (l,l-dimethylpiperidinium-4-yl octadecyl phosphate); PI 103 (3-[4-(4- Morpholinylpyrido[3',2':4,5]furo[3,2-d]py midi n-2-yl]phenol hydrochloride; C₁₉H₁₆N₄0₃; CAS No.: 371935-74-9); PI828 (2-(4-Morpholinyl)-8-(4-aminopheny)l-4H-l-benzopyran-4-one; C₁₉H₁₈N₂0₃; CAS No.: 942289-87-4); PIK-293 (C₂₂H₁₉N₇0; CAS No.: 900185-01-5); PIK-294 (C₂₈H₂₃N₇0₂; CAS No.:

900185-02-6); PIK75 (N'-[(lE)-(6-bromoimidazo[l_;2-a]pyridin-3-yl)methylene]-N_;2-dimethyl-5- nitrobenzenesulfonohydrazide hydrochloride); PIK90 (N-(7,8-Dimethoxy-2,3-dihydro-imidazo[l,2- c]quinazolin-5-yl)-nicotinamide; C₁₈H₁₇N₅0₃; CAS No.: 677338-12-4); PIK93 (C₁₄H₁₆CIN₃0₄S₂; CAS No.: 593960-11-3); PKI-587 (C₃₂H₄₁N₉0₄; CAS No.: 1197160-78-3); PP-121 (l-Cyclopentyl-3-(lH- pyrrolo[2_;3-b]pyridin-5-yl)-lH-pyrazolo[3_;4-c ]pyrimidin-4-amine; C₁₇H₁₇N₇; CAS No.: 1092788-83-4); PX-866 ([(3aR_;6f_;9S_;9aR_;10/?_;llaS)-6-[[bis(prop-2-enyl)amino]methylidene]-5-hydroxy-9-

(methoxymethyl)-9a_/lla-dimethyl-l_;4_;7-trioxo-2_;3_;3a_;9_;10 1-hexahydroindeno[4_;5-/7]isochromen- 10-yl] acetate; C₂₉H₃₅N0₈; CAS No.: 502632-66-8); quercetin (sophoretin; C₁₅H₁₀O₇; CAS No.: 117-39- 5); SF1126 ((3S)-4-[[(lS)-l-carboxy-2-hydroxyethyl]amino]-3-[[2-[[(2S)-5- (diaminomethylideneamino)-2-[[4-oxo-4-[[4-(4-oxo-8-phenylchromen-2-yl)morpholin-4-ium-4- yl] methoxy] butanoyl]amino]pentanoyl]amino]acetyl]

amino]-4-oxobutanoic acid acetate; C₄₁H₅₂N₈0₁₆); tandutinib (1-piperazinecarboxamide, 4-[6- methoxy-7-[3-(l-piperidinyl)propoxy]-4-quinazolinyl]-N-[4-(l-methylethoxy)phenyl]-); tetrodotoxin citrate; TG100-115 (C₁₈H₁₄N₅0₂; CAS No.: 677297-51-7); TGX-115 (8-(2-Methylphenoxy)-2-(4- morphonilyl)-4(lH)-quinolinone; C₂₀H₂oN₂0₃; CAS No.: 351071-62-0); TGX-221 (7-Methyl-2-(4- morpholinyl)-9-[l-(phenylamino)ethyl]-4H-pyrido-[l,2-a] pyrimidin-4-one; C₂₁H₂₄N₄0₂; CAS No.: 663619-89-4); thioperamide maleate; WHI-P 154 (2-Bromo-4-[(6,7-dimethoxy-4- quinazolinyl)amino]phenol; C₁₅H₁₄BrN₃0₃; CAS No.: 211555-04-3); wortmannin; XL147 (C₂₁H₁₅N₅0₂S₂; CAS No.: 956958-53-5); XL765 (C₃₁H₂₉N₅0₆S; CAS No.: 1123889-87-1); ZSTK474 (C₁₉H₂₁F₂N₇0_2; CAS No.: 475110-96-4).

The determining of the level of mi NA-21 and/or Grhl3 may be achieved by any methods known in the art for determining the level of nucleic acids and/or proteins. Methods for determining the level of a nucleic acid include both direct and indirect methods, such as, quantitative real time polymerase chain reaction (Q-RT-PCR), array and Northern blot. Methods for determining the level of a protein include Western blot and ELISA.

The measuring of miRNA-21 and/or Grhl3 can occur after a cancer diagnosis has been made and prior to in initiation of a standard of care cancer therapy (e.g., surgery and/or chemotherapy). In some embodiments, the measuring of miRNA-21 and/or Grhl3 occurs after a cancer has become resistant to a standard of care therapy. These embodiments are not mutually exclusive.

To assess the relative level of miRNA-21 and/or Grhl3, the level of miRNA-21 and/or Grhl3 in tissue can be subjected to one or more of various comparisons. In general, it can be compared to: (a) miRNA-21 and/or Grhl3 level(s) in normal tissue from the organ in which the cancer originated; (b) miRNA-21 and/or Grhl3 levels in a collection of comparable cancer tissue samples; (c) miRNA-21 and/or Grhl3 level in a collection of normal tissue samples; or (d) miRNA-21 and/or Grhl3 level in an arbitrary standard.

The identifying or selecting step of the screening methods described herein optionally comprises comparing the measurement of miRNA-21 and/or Grhl3 to a reference measurement of miRNA-21 and/or Grhl3, and scoring the miRNA-21 and/or Grhl3 measurement from the tissue as elevated or decreased based on statistical analysis or a ratio relative to the reference measurement. In some embodiments, the reference measurement comprises at least one of the following (a) a measurement of miRNA-21 and/or Grhl3 from healthy tissue of the subject of the same tissue type as the neoplastic tissue; (b) a database containing multiple miRNA-21 and/or Grhl3 measurements from healthy or cancerous tissues from other subjects; or (c) a reference val ue calculated from multiple miRNA-21 and/or Grhl3 measurements from healthy or cancerous tissues from other subjects, optionally further including statistical distribution information for the multiple measurements, such as standard deviation.

In some embodiments, a miRNA-21 measurement of at least 1.0 standard deviation greater than a median miRNA-21 measurement in corresponding healthy tissue is scored as elevated miRNA-21. In other embodiments, a miRNA-21 measurement that is statistically significantly greater than miRNA- 21 measurements in corresponding healthy tissue, with a p-value less than 0.1, or less than 0.05, or less than 0.01, or less than 0.005, or less than 0.001 is scored as elevated miRNA-21.

In some embodiments, a Grhl3 measurement of at least 1.0 standard deviation less than a median Grhl3 measurement in corresponding healthy tissue is scored as decreased Grhl3. In other embodiments, a Grhl3 measurement that is statistically significantly less than Grhl3 measurements in corresponding healthy tissue, with a p-value less than 0.1, or less than 0.05, or less than 0.01, or less than 0.005, or less than 0.001 is scored as decreased Grhl3.

In some variations, the level of miRNA-21 and/or Grhl3 in a subject is compared to a predetermined "cut-off" concentration of miRNA-21 and/or Grhl3 that has been determined from observations to represent an predictive measure of miRNA-21 and/or Grhl3 that is predictive of efficacy of PI3K therapy. Determination of a suitable cut-off is made using, e.g., statistical analysis of miRNA-21 and/or Grhl3 level data collected from multiple healthy and/or cancer patients. If a "cut-off value is employed, the cut-off concentration preferably is statistically determined to have optimal discriminating value for subjects who benefit from PI3K therapy (e.g., to have maximum sensitivity and specificity). It will be appreciated that statistical analysis of a dataset will permit clinicians to make informed decisions based on concentrations other than the optimal discriminating concentration (e.g., above or below the optimal discriminating concentration).

Considerations regarding the probability of success of PI3K-targeted therapy based on mi NA-21 and/or Grhl3 measurement, versus the probability of success of alternative therapies based on any available clinical data, can guide the selection of an appropriate cut-off concentration of miRNA-21 and/or Grhl3 for making a treatment decision. In one embodiment, the level of Grhl3 in the squamous cell neoplasm is decreased by 60% or more as compared to the control. Alternatively, the level of Grhl3 in the squamous cell neoplasm may be decreased by more than 61%, 71%, 72%, 73%, 74%, 75%, 76%, 77%, 78%, 79%, 80%, 81%, 82%, 83%, 84%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% as compared to the control.

In some embodiments, the screening methods described herein further comprise measuring the expression of at least one additional marker, such as PTEN in the tissue. Standard multivariate statistical analysis tools are used to optimize the predictive value of miRNA-21 and/or Grhl3 in combination with one or more of these additional markers.

As used herein "susceptible to treatment with a PI3K inhibitor" means that the PI3K inhibitor is effective against such a cell. The cell may cease to grow and/or proliferate or may die following treatment with a PI3K inhibitor. As used herein "treating a squamous cell neoplasm" means inhibiting the growth and proliferation of neoplastic cells, and/or causing the death of neoplastic cells. Preferably, the treatment involves the administration of a therapeutic amount of the therapeutic compound. A therapeutic amount of a therapeutic compound refers to an amount of the compound that is sufficient to inhibit, halt or eradicate the condition being treated when the compound is administered alone or in conjunction with another agent. The treatment may involve the co-administration of more than one therapeutic compound. Co-administration may be simultaneous or sequential.

"Treating" or "treatment" refers to both therapeutic treatment and prophylactic or preventative measures, wherein the aim is to prevent, ameliorate or slow down (lessen) squamous cell neoplasm. "Preventing", "prevention", "preventative" or "prophylactic" refers to keeping from occurring, or to hinder, defend from, or protect from the occurrence of a condition, disease, disorder, or phenotype, including an abnormality or symptom. A subject in need of prevention may be prone to develop the condition.

The term "ameliorate" or "amelioration" refers to a decrease, reduction or elimination of a condition, disease, disorder, or phenotype, including an abnormality or symptom. A subject in need of treatment may already have the condition, or may be prone to have the condition or may be in whom the condition is to be prevented. As used herein, "administering" refers to contacting a subject with a compound. Administering may be achieved by any means by which the inhibitor can be delivered to the site to be treated. Suitable types of administration include both systemic and localized forms of administration, such as intravenously, intraperitoneally, intranasally, transdermally, topically, via implantation, subcutaneously, parentally, intramuscularly, orally and via adsorption.

The invention also provides a kit comprising reagents for carrying out the methods described herein. Such a kit may include reagents selected from Q-RT-PCR reagents, array reagents, Northern blot reagents, Western blot reagents and ELISA reagents necessary to carry out the described methods. The kit may also include instructions for carrying out the methods.

"Q-RT-PCR reagents" as used herein means the reagents required to carry out a Q-RT-PCR reaction for analysis of a tissue. Such reagents include primers, deoxynucleotides, buffers and enzymes. "Array reagents" as used herein means the reagents required to carry out an array reaction for analysis of a tissue. Such reagents include primers, deoxynucleotides, buffers and enzymes.

Inositol induces the expression of Grhl3 which in turn induces the expression of PTEN. The inventors have shown that miRNA-21 and Grhl3 are predictive biomarkers and that elevated miRNA- 21 and/or decreased Grhl3 in tissue from the squamous cell neoplasm, compared to the control level, indicates that the squamous cell neoplasm will be susceptible to treatment with a PI3K inhibitor.

The inventors propose that inducing expression of GRHL3 using inositol, can cause suppression of squamous cell neoplasms.

"Topical composition" as used herein means a composition that is specifically formulated to be applied to the skin. A topical composition may comprise a therapeutic compound that is to be delivered to the skin. Topical compositions include transdermal compositions, liquids, lotions, creams, gels, ointments, powders and sprays.

It is to be understood that, if any prior art publication is referred to herein, such reference does not constitute an admission that the publication forms a part of the common general knowledge in the art, in Australia or any other country.

In the claims which follow and in the preceding description of the invention, except where the context requires otherwise due to express language or necessary implication, the word "comprise" or variations such as "comprises" or "comprising" is used in an inclusive sense, i.e. to specify the presence of the stated features but not to precl ude the presence or addition of further features in various embodiments of the invention.

It must also be noted that, as used in the subject specification, the singular forms "a", "an" and "the" include plural aspects unless the context clearly dictates otherwise. It will be apparent to the person skilled in the art that while the invention has been described in some detail for the purposes of clarity and understanding, various modifications and alterations to the embodiments and methods described herein may be made without departing from the scope of the inventive concept disclosed in this specification.

BRIEF DESCRIPTION OF THE FIGURES

The invention will now be described with reference to the following non-limiting examples, in which:

Figure 1 A - G shows hyperproliferation of Grhl3-nu\\ keratinocytes (A and B) Histology and PCNA IHC on skin from E18.5 wild type (WT) and Gr )\3^~'^~ (KO) embryos.

E - epidermis; D - dermis; hf - hair follicle. (C, D and E) Cell numbers, appearance, and soft agar colony numbers of cultured keratinocytes from WT and KO E18.5 embryos. For keratinocyte cultures, 2.6 X 10⁴ cells were seeded at Day 0. For soft agar, 3.4 X 10⁵ cells were seeded at Day 0. Arrow - pseudo-tumor. The differences in cell number were statistically significant (p<0.03) using a Student's t-test. (F) PCR analysis of deletion in genomic DNA from epidermis derived from a Grhl3 ^/~/K14Cre+ E18.5 embryo and a 15-week-old adult mouse. The undeleted (flox) band of 425 bp, and the deleted (Δ) band of 282 bp are indicated.

(G) Q-PCR of Grhl3 expression levels in epidermis from 15-week-old wild type and Grhl3^A/~/K14Cre+ mice. Error bars represent the standard deviation (+/^_ SD). Figure 2 A - H shows Grhl3 functions as a tumor suppressor in mice

(A, B and C) Tumor incidence, number, and malignant potential in DMBA/TPA-treated WT and Grhl3^^~/K14Cre+ mice.

(D) Macroscopic appearance of DMBA/TPA-treated WT and Grhl3^A'^~/K14Cre+ mice at 15 weeks.

(E) Histology of squamous cell carcinoma at low (left panel) and high magnification from DMBA/TPA- treated Grhl3^A'^~/K14Cre+ mice.

(F) Grhl3 expression levels by Q-RT-PCR in normal skin (NAD), papillomas (pap) and squamous cell carcinomas from DMBA/TPA-treated WT and Grhl3^A/~/K14Cre+ mice. The differences in expression between wild type and Grhl3^A/~/K14Cre+-denved tissues were significant (p<0.001) using a Student's t-test.

(G) Tumor incidence in WT and Grhl3 ^/~/K14Cre+ mice treated with TPA alone.

(H) Macroscopic and microscopic appearance of spontaneous snout tumor in Grhl3^A/~/K14Cre+ mice.

Figure 3 A - F shows PTEN is a direct transcriptional target gene of GRHL3

(A) Alignment of the promoter regions of PTEN genes from the indicated species. The GRHL3 DNA consensus sequence is enlarged and bolded.

(B) EMSA of recombinant (r) GRHL3 binding to the PTEN promoter probe. A 100-fold molar excess of unlabelled cold competitor probes (PTEN or Grhl3 consensus) were added in the indicated lanes. The migration of the specific GRHL3/DNA complex is arrowed.

(C) ChIP analysis of endogenous GRHL3 on the PTEN promoter. Chromatin from the human keratinocyte line (HaCAT) was immunoprecipitated using antisera to GRH L3, and amplified with PTEN primers. Pre-immune sera (IgG) and the muscle-specific MyoD promoter were used as negative controls, and the input chromatin is shown.

(D) Q-RT-PCR (upper panel) and immunoblot (lower panel) of PTEN expression in wild type (WT) and Grhl3^~ (KO) E18.5 skin. For Q-RT-PCR, bars represent standard errors, and HPRT served as a reference. The difference in expression was statistically significant (p<0.02) using a Student's t-test. For the immunoblot, actin served as the loading control.

(E) Immunoblots of epidermal lysates from WT and KO E18.5 embryos probed with the stated antibodies. Actin and HSP70 served as loading controls.

(F) IHC analysis of skin from WT and KO E18.5 embryos using the stated antibod ies. Arrow in the lower right panel indicates p-AKT staining in the basal layer.

Figure 4 A - E shows PTEN is the critical GRHL3 target gene in dysregulated cell growth and PI3K/AKT activation

(A) I ncreased susceptibility of PTEN^{+ ~}/Grhl3^{+ ~} mice to DMBA/TPA-induced squamous cell carcinoma. (B) Relative PTEN expression.

(C) Macroscopic appearance of multiple aggressive squamous cell carcinoma in DMBA/TPA-treated

(D) Immunoblots using the stated antibodies of lysates from GRHL3-kd and scrambled control (Scr) HaCAT cells grown in serum, or transduced with MSCV-based retroviral supernatants carrying wild type PTEN or the C124S phosphatase dead PTEN mutant as indicated. GAPDH served as the loading control.

(E) Growth kinetics of the cell lines listed in (C). I X 10⁵ cells for each line were seeded at Day 0. The differences between the GRHL3-kd or GRHL3-kd + C124S mutant and the GRH L3-kd + wild type PTEN or Scr control cells at Day 8 was significant (p<0.004) using the Students t-test.

Figure 5 A - E shows skin and tumors from Grhl3^A/~/K14Cre+ mice exhibit PI3K pathway activation (A) Immunoblot of lysates from epidermis, papillomas, and squamous cell carcinomas from wild type and Grhl3 ^/~/K14Cre+ mice probed with PTEN and p-S6 antibodies. Actin served as the loading control.

(B) IHC analysis of squamous cell carcinoma from wild type and Grhl3^A/~/K14Cre+ mice using AKT and p-AKT antibodies.

(C) Analysis of H-Ras mutations in tumors from wild type and Grhl3^A/~/K14Cre+ mice.

(D) Immunoblot of lysates from epidermis, papillomas, and squamous cell carcinomas from wild type and Grhl3^A/~/K14Cre+ mice probed with ERK1/2 and p-ERKl/2 antibodies. The actin panel in (A) served as the loading control.

(E) IHC analysis of squamous cell carcinoma from wild type and Grhl3^A/~/K14Cre+ mice using ERK1/2 and p-ERKl/2 antibodies.

Figure 6 A - H shows miR-21-induced loss of GR L3 and PTEN expression in human squamous cell carcinoma

(A) Quantification of GRHL3 and PTEN expression levels by Q-RT-PCR in human squamous cell carcinomas isolated by LCM, normalised to expression in the adjacent non-tumor containing tissue. (B) Heat map of the relative expression of miRNAs predicted to target GRHL3 in two human squamous cell carcinoma and adjacent normal epidermis. The fold change in tumor (T) versus normal (N ) is shown.

(C) Alignment between miR-21 and the 3'UTR of human GRHL3.

(D) Quantification of human miR-21 expression levels relative to U6 by Q-RT-PCR in 10 squamous cell carcinoma and their matched controls.

(E) GRHL3 and (F) PTEN expression quantitated by Q-RT-PCR and normalised to β-actin in HaCaT cells overexpressing miR-21 or a scrambled control.

(G) Immunoblots of lysates from control and miR-21 overexpressing HaCaT cells probed with the stated antibodies. Actin and HSP70 served as the loading controls.

(H) Luciferase activity of the wild type (3'UTR) or mutant (M ut21) GRHL3 3'UTR reporter construct transfected into HEK293 cells in the presence and absence of expression vectors carrying miR-21 or its antagomir (miRZip21) as indicated. Expression of β-Galactosidase from a cotransfected reporter construct was used as control for transfection efficiency.

Figure 7 shows that the mi -21/GRHL3/PTEN proto-oncogenic network is evident in human SCC cell lines from different tissue origins. Q-PCR was performed on RNA from the listed SCC cell lines from head and neck (SCC4 and SCC9), skin (SCC 13) and cervix (A431) using primers for miR-21, Grhl3 and Pten.

Figure 8 shows inhibition of PI3K/mTOR pathway and reduced cellular proliferation in human SCC cell lines treated with BEZ235. (A) The listed cell lines were treated with BEZ235 at the stated concentrations for 48 hours and cell lysates prepared and analysed by immunoblot for PI3K signaling. (B) Growth curves for the same lines treated for varying times with differing

concentrations of BEZ235.

Figure 9 shows prevention of SCC development in BEZ235 treated Grhl3/K14Cre+ mice. (A) Schematic of BEZ235 and DMBA/TPA treatment. (B) Percentage of mice without tumors in vehicle or BEZ235 treated mice exposed to the DMBA/TPA protocol. (C) Tumor scoring of mice treated with BEZ235 and DM BA/TPA protocol.

EXAMPLES

Example 1 - Experimental Procedures

Generation of Experimental Animals

For construction of the conditional Grhl3 targeting vector, a fragment for the 3' homology arm extending 4 kb 3' from exon 4 from the mouse Grhl3 genomic locus isolated from a 129/SV/J genomic library (Ting et al., (2003) Nat. Med. 9, 1513-1519) was PCR amplified. A 4.9 kb PCR fragment extending from 5' of exon 2 to 3' of exon 4 was generated with a loxP sequence and Hind\\\ site at its 3' end, a Kpn\ site at its 5' end, and flanked by Xba\ sites. This was cloned into TOPO 2.1 and sequenced. The "loxP" arm was then subcloned into TOPO 2.1 containing the 3' arm as an Xba\ fragment and the two arms were released as a Kpn\ fragment and subcloned into the Frt-PGKNeo- Frt-loxP vector (Meyers et al., (1998) PNAS 95, 13513-13518). The 5.5 kb 5' homology arm was PCR amplified with flanking SacW sites and a 5' Bam \ site introduced, sequenced and subcloned as a SacW fragment to generate the final construct. The vector was linearized with Not\ and electroporated into W9.5 embryonic stem cells. G418-resistant clones in which the targeting vector had recombined with the endogenous Grhl3 gene were identified using Bam \ digested genomic DNA probed with the 5' probe, which distinguished between the endogenous (8.4 kb) and targeted (6.5 kb) alleles. Recombination was confirmed with a 3' probe using a Hind\\ \ digest that distinguished between the endogenous (15 kb) and targeted (4.8 kb) alleles. Two ES cell clones were used to generate mouse lines on a C57BL/6 background. Grhl3^{fl +} (where // is the undeleted "floxed" allele) heterozygous mice were identified by hybridising SomHI-digested genomic DNA with the 5' probe. M ice heterozygous for the targeted allele were crossed with B6FLPE transgenic mice to excise the neor cassette, and these animals were used for all subsequent experiments. Grhl3^+/~ mice were crossed with K14Cre+ transgenic mice (Jonkers et al., (2001) Nat. Genet. 29, 418-425) and the resultant Grhl3^+/~/K14Cre+ animals were crossed with Grhl3^m mice to provide the Grhl3^hh/K14Cre+ experimental animals (where Δ is the deleted floxed allele). PTEN^{+ ~} mice were kindly provided by Dr Tak Mak, and intercrossed with Grhl3^{+ ~} mice to generate compound heterozygotes. Genotyping of Mice

Mice were genotyped by PCR using genomic DNA prepared from tail biopsies or embryonic tissues.

Products of 634 bp were generated from the wild type and 801 bp from the targeted Grhl3 alleles.

Deletion efficiency was also assessed. The size of the undeleted {flox) band was 425 bp and the deleted (Δ) band was 282 bp. The primer sequences utilized in the genotyping and deletion efficiency assessment were:

Genotyping sense, 5'- CTTGCTTGAGACCATGCCAT-3'

Genotyping antisense, 5'- TG CACCCATATCCACATG CA-3'

Deletion efficiency sense (undeleted and deleted), 5'- GCAGATATCCATCACACTGG-3'

Deletion efficiency antisense (undeleted), 5'- TGCACCCATATCCACATGCA-3'

Deletion efficiency antisense (deleted), 5'- TATCAGGGAAGAGCAGAGAC-3'

PCR conditions were 94°C for 2 minutes followed by 35 cycles of 94°C for 30 seconds, 55°C for 30 seconds and 72°C for 1 minute with a final 5 minutes extension at 72°C.

Tumour I nduction and Skin Barrier Analysis in Mice

Tumours were induced in wild type, Grhl3^+/~/PTEN^+/+, PTEN^+/~/Grhl3^+/+, PTEN^+/~/Grhl3^+/~, and Grhl3^A/~ /K14Cre+ mice through the application of 25 μg DMBA (Sigma-Aldrich) in acetone to a shaved area on the back followed one week later by twice-weekly application of TPA (7.6 nmol) in 150 μΙ of acetone for up to 30 weeks. In some experiments TPA was applied without DMBA initiation. Skin barrier analysis was performed as previously described (Hardman et al., (1998) Development 125, 1541-1552).

Isolation and Culture of Mouse Keratinocytes and Cell Lines

Mouse dorsal skin collected from wild type and Gr \3^~f~ E18.5 embryos was treated with 2 mg/ml Dispase (Invitrogen) and incubated overnight at 4°C. The epidermis was separated from the dermis and digested in 0.25% trypsin for 10 minutes. Primary mouse keratinocytes were cultured as previously described (Lichti et al., (2008) Nat. Protoc. 3, 799-810), in a calcium and serum-free keratinocyte medium (Lonza, cat. no. CC-3108), supplemented with a "Bulletkit" (Lonza) containing epinephrine, transferrin, bovine pituitary extract, EGF, insulin, hydrocortisone and antibiotic, as per the manufacturer's instructions. For soft agar culture of keratinocytes, 3.4 X 10⁵ cells from epidermis from E18.5 wild type and Grhl3 embryos were mixed with agarose to form a top layer of 0.3% agarose. This layer was plated into a 60 mm plate on top of a solid layer of 0.5% agarose. After 8 days, colonies were stained with a 0.005% crystal violet solution, and each colony containing more than 10 cells was counted. The human keratinocyte cell line, HaCAT was obtained from ATCC and cultured in Dulbecco's modified Eagle's medium (DMEM), supplemented with 10% FCS, 4 mM L-

Glutamine, and 100 μg/ml penicillin/streptomycin solution. Cell size in the epidermal basal layer was determined from 24 independent sections derived from six E18.5 wild type and 6r/i/3^{_ ~} embryos using the AxioVisionLE software from Zeiss. The difference in size was statistically significant (p<0.008) using a Student's t-test.

RNA preparation and Q-RT-PCR

For gene expression analysis, normal skin from E18.5 wild type and Grh ^'1' embryos, and adult wild type and Grhl3^A/~/K14Cre+ mice, and papillomas and squamous cell carcinoma from wild type and

Grhl3 ^h/K14Cre+ mice were homogenised in TRIzol (Invitrogen) and RNA extracted according to the manufacturer's instructions. Q-RT-PCR was carried out as described previously (Ting et al., (2003) supra) with the following primer pairs:

Mouse Grhl3 sense 5' -AAGGAAGATGTCGAATGAACTTG-3'

Mouse Grhl3 antisense 5'-TCGTCCTCATTACTGTAGGGAAA-3'

Human GRHL3 sense 5 '-G CAAGCG AGG AATCTTAGTCAA-3 '

Human GRHL3 antisense 5'-ACGTGGTTGCTGTAATGCTGA-3'

Mouse PTEN sense 5'- AGGCACAAGAGGCCCTAGAT-3'

Mouse PTEN antisense 5'- CTGACTGGGAATTGTGACTCC-3'

Human PTEN sense 5'-GGCACCGCATATTAAAACGTA-3'

Human PTEN antisense 5'- ATG CCATTTTTCCATTTCCA-3'

Mouse Hprt sense 5 '-GCTGGTG AAAAG GACCTCT-3¹

Mouse Hprt antisense 5'-CACAGGACTAGAACACCTGC-3'

Human HPRT sense 5'-ATGGACAGGACTGAACGTCT - 3'

Human HPRT antisense 5'-CTTGCGACCTTGACCATCTT - 3'

Human actin sense 5'-CTGGAACGGTGAAGGTGACA-3'

Human actin antisense 5'-AAGGGACTTCCTGTAACAATGCA-3'

A Student's t-test was used to determine statistical difference in expression levels with p values <

0.05 considered significant, and results were analysed using Prism (GraphPad). The error bars in all expression analyses represent the standard error of the mean. ChIP and EMSA

ChIP was performed as described previously (Wilanowski et al., (2008) EMBO J. 27, 886-897), with anti-GRHL3 antibodies (Aviva Systems Biology, San Diego, CA USA) and primers amplifying the PTEN promoter as detailed below. EMSAs were performed using recombinant mouse GRHL3 protein, and oligonucleotides containing the conserved GRHL3 binding site in the PTEN promoter as detailed below. The DNA consensus-binding site for GRHL3 was used as a cold competitor (Ting et al., (2005) Science 308, 411-413).

PTEN promoter (ChIP) sense 5'-CACCAGTTTGGGGACTCTCT-3'

PTEN promoter (ChIP) antisense 5'-GAACCCCAACCCTTCCTG-3'

PTEN promoter (EMSA) sense only 5'-GGGGCTGCTTGTGTAACCAGCTCCCCAGGCGC-3' Immunoblot Analysis and IHC

For immunoblot analysis, skin from E18.5 wild type and GrhB^'1' embryos, and adult skin, papillomas and squamous cell carcinoma from wild type and Grhl3^A,~/K14Cre+ mice were lysed in RIPA buffer containing protease inhibitors at 4°C overnight. Insoluble material was removed by centrifugation. Forty micrograms of protein was run in each lane of a denaturing 10% SDS-PAGE gel, and subsequently transferred to PVDF membrane. For blocking and antibody dil ution, 5% milk powder (Diploma) in PBS with 0.1% Tween-20 was used. For IHC, tissues were collected and fixed in 4% PFA overnight, embedded in paraffin and sectioned at 8 μιη onto Superfrost-Plus Slides and processed as per standard protocols using DAB staining. Antibodies used for immunoblotting and

immunohistochemistry were: PTEN (#9559), AKT (#9272), phospho-AKT (#9271), PDK1 (#3062), phospho-PDKl (#3061), 4EBP1 (#9644), phospho-4EBPl (#9451), S6 (#2217), phospho-S6 (#2215), ERK (#9102), phospho-ERK (#9101), GAPDH (#2118) (Cell Signalling), Actin (sc-1616) and HSP70 (sc- 24) (Santa Cruz), PCNA (PC10, DAKO), and GRH L3 (Aviva Systems Biology). Secondary antibodies included donkey antirabbit IgG, donkey anti-goat HRP, and sheep anti-mouse IgG conjugated to horseradish peroxidase (H RP) (Amersham Biosciences). The ECL Detection kit was used with HyperfilmTM (Amersham Biosciences) for all immunoblots. shRNAs, Lentiviral and Retroviral Infection

The shRNA target oligonucleotides for Grhl3 and the scrambled (Scr) control were cloned into the pSUPER. retro. neo+GFP vector using the BglW and Hind\\\ sites. The target sequences and generation of viruses are detailed as previously described (Caddy et al., (2010) Dev. Cell 19, 138-147). HaCAT cells were trans-infected over a 24-hour period and GFP-positive cells were selected by FACS, and cultured for 8 days, or harvested for preparation of lysates. Knockdown of GRHL3 expression was confirmed by immunoblot using antibodies to GRH L3 and GAPDH. For expression of wild type PTEN, or the phosphatase dead C124S PTEN mutant in GRHL3-kd cells, the respective cDNAs were cloned into MSCV-DSRed using the Bam \ and Xho\ sites. PTEN expression was confirmed in DSRed+ cells by immunoblot using PTEN antibody, and GAPDH as a loading control. Laser Capture Microdissection (LCM)

Surgical specimens of squamous cell carcinoma resected from patients (incl uding tumour and adjacent normal epidermis) were embedded in OCT medium and stored at -80°C. The sections were stained with Histogene LCM frozen staining kit just before commencing LCM. The cryosections (8 μιη) were microdissected using a Veritas™ microdissection instrument (Arcturus) according to the standard protocol. Tumour tissues and normal tissues were captured onto Capsure macro LCM caps. RNA extraction and amplification were performed according to the manufacturer's instructions. The reagents for staining, RNA extraction and RNA amplification were obtained from Arcturus. Q-RT-PCR was performed as detailed above. Mutational Analysis of H-ras

PCR primers amplifying codons 12, 13, and 61 were designed on the basis of genomic DNA sequences for H-ras. M utations were detected by sequencing as previously described (Ise et al., (2000) Oncogene 19, 2951-2956). The mutations were further confirmed by restriction fragment length polymorphism (RFLP) analysis as previously detailed (Jaworski et al., (2005) Oncogene 24, 1290-1295).

RNA Isolation and microRNA Array

For microRNA analysis, total RNA was isolated from HaCaT cells, squamous cell carcinoma and normal tissue samples using Trizol (Invitrogen). The differential microRNA expression in two normal skin and two squamous cell carcinoma samples were analysed using Genechip miRNA array (Affymetrix, CA, USA) and the data were displayed using the Partek GS software. To validate the results of microarray, RNA isolated from 10 squamous cell carcinoma and matched normal samples was reverse transcribed using the QuantiMir RT kit (System Biosciences, CA, USA) according to the manufacturer's instructions. The forward primer of miR-21 was designed based on the mature microRNA sequence and custom made from Geneworks. Q-PCR was performed using Power Sybr Green Master Mix from Applied Biosystems (Foster City, CA). To over-express miR-21 in HaCaT cells, the lentiviral vector-based microRNA precursor construct was used (System Biosciences).

Luciferase assays

The 3'UTR region of human GRHL3 containing the predicted site for miR-21 was subcloned into the pM IR-REPORT Luciferase vector and co-transfected with the pMIR-REPORT β-Galactosidase control plasmid (Applied Biosystems). The mutant construct (Mut21) was generated by deleting 10 bp (419 -428) from the miR-21 site in the 3'UTR of GRHL3. HEK293T cells were cultured in 24-well plates, and each well transfected with 0.5 μg of either pM\R-GRHL3'UTR or pM\R-GRHL3'UTR-Mut21 Luciferase with the pMIR^-Galactosidase vector. After 15h, the miR-21 precursor construct was coinfected with and without the lentivector-based antagomir to miR-21 "miRZip-21" (System Biosciences). Firefly Luciferase was measured 48h after transfection/infection using the Dual-light Chemiluminescent Reporter Gene Assay System (Applied Biosystems) as per the manufacturer's instructions and normalized to β-Galactosidase activity to control for differences in transfection efficiency.

Example 2 - Grhl3 Deletion During Embryogenesis Causes Epidermal Keratinocyte

Hyperproliferation

Initial experiments focused on the proliferative potential of keratinocytes in the Grhl3-nu\\ mice. The epidermis in embryonic day (E) 18.5 Grh ^'1' animals was markedly thickened (Figure 1A), and expression of the proliferative marker PCNA was expanded compared to wild type epidermis, with mitotic cells extending into the suprabasal layers (Figure IB). This hyperproliferation was cell intrinsic, as keratinocytes cultured from E18.5 Gr/7/3^_/~embryos grew more rapidly than the wild type controls (Figure 1C), and displayed loss of cell-cell contact inhibition, forming heaped up pseudotumors in the culture dish (Figure ID). They also displayed increased colony numbers in soft agar (Figure IE), strengthening the possibility that Grhl3 could play a tumor suppressor role in skin cancer in adult mice. To investigate this, mice carrying a conditionally targetable Grhl3 allele were generated, with loxP sites flanking exons 2 and 4 of the gene. Mice homozygous for the floxed alleles (Grhl3^{fl fl}) were healthy and fertile, and when crossed with Grhl3^{+ ~} mice carrying a B6-Cre transgene expressed at the two-cell stage of development, generated Grhl3^A/~/B6-Cre+ mice (where Δ is the deleted floxed allele) that phenocopied the Grhl3-nu\\ animals (data not shown). To delete Grhl3 in the skin, Grhl3^{fl ~} mice were crossed with a line carrying a keratin (K) 14-driven Cre transgene. Although patchy deletion in the epidermis has been reported with this line as early as E13.5 (using a ROSA cross), analysis of the epidermis in E18.5 Grhl3^fl/~/K14Cre+ embryos revealed less than 20% deletion and all E18.5 Grhl3^{fl ~}/K14Cre+ embryos displayed normal skin barrier formation (not shown). High levels of deletion (>95%) of the Grhl3^fl allele were only detected after birth (from PI onwards), and consistent with this, expression of Grhl3 in the deleted skin was markedly reduced (Figures IF and 1G). Interestingly, the Grhl3^A/~/K14Cre+ mice displayed no defect in skin barrier function at P7 (data not shown), and appeared normal at 8 weeks, indicating that although Grhl3 was essential for establishment of the barrier in utero, it was not essential for its maintenance after birth.

Example 3 - Mice with Grhl3 Deletion in Keratinocytes Display Enhanced Susceptibility to

Chemical-Induced and Spontaneous squamous cell carcinoma

To examine the role of Grhl3 in skin tumorigenesis, a well-established chemical carcinogenesis protocol was used on cohorts of Grhl3^A/~/K14Cre+ mice and wild type controls (n= 18 in each group), in which tumors were initiated by topical application of 7, 12-dimethylbenz[a]anthracene (DMBA), and promoted by twice-weekly 12-0-tetradecanoylphorbol-13-acetate (TPA). Papilloma formation was observed within 4 weeks in the Grhl3^A/~/K14Cre+ animals, but not until 13 weeks in any controls, by which time 100% of the mutant mice had developed tumors (Figure 2A). The rapidity of tumor formation was striking; particularly as the background strain of the mutant animals (C57BI/6) is notoriously resistant to chemical-induced tumor formation. Total tumor numbers were also markedly higher in the mutant versus wild type mice (Figure 2B), and many of the papillomas progressed on to form squamous cell carcinomas (Figure 2C), a rare event in the wild type mice. The squamous cell carcinomas grew rapidly and were often multiple (Figures 2D), and histologically were poorly differentiated with numerous mitoses (Figure 2E). The epidermis adjacent to the tumors was markedly thickened compared to wild type mice, and resembled the epidermis from Grhl3-nu\\ embryos. Consistent with this, perturbed expression of the terminal differentiation markers filaggrin and involucrin, and the keratins 5, 6 and 10 in the Grhl3-de\eted adult skin also mirrored the changes observed in the Grhl3-nu\\ embryos. Expression of Grhl3 was markedly reduced in treated skin prior to the onset of tumors, and completely absent in both papillomas and squamous cell carcinomas from the Grhl3^A/~/K14Cre+ mice (Figure 2F). Tumors were also observed with high frequency in Grhl3 ^h/K14Cre+ animals treated with TPA alone, suggesting that loss of Grhl3 is sufficient for disease initiation (Figure 2G). With ageing, 100% of the untreated Grhl3^A,~/K14Cre+ mice developed epidermal hyperplasia and spontaneous tumors that were predominantly squamous papillomas or carcinomas affecting the snout and neck (Figure 2H). Identical findings were obtained when mouse mammary tumor virus (MMTV)-Cre transgenic mice were used to delete Grhl3 in keratinocytes (data not shown). These snout and neck tumors in both Grhl3^A/~ lines phenocopied changes observed in mice with a keratinocyte-specific deficiency of the tu mor suppressor, PTEN.

Example 4 - PTEN is a Direct Transcriptional Target Gene of GRHL3

The DNA consensus-binding site for human GRHL3 and its Drosophila homolog, grh has been conserved across 700 million years of evolution. On this basis, it was predicted that GRHL3 target genes should contain this motif conserved across a wide range of species. In view of this, a customised dataset of genomic regions located within lOkb of gene transcriptional start sites that are conserved in placental mammals with the GRHL3 consensus (and highly related sequences) was interrogated, an approach that had successfully been employed to identify a key GRHL3 target in wound repair. A highly conserved site was identified in the promoter region of the PTEN gene (Figure 3A), and confirmed specific binding of GRH L3 to this site in vitro in electrophoretic mobility shift assays (EMSA) (Figure 3B), and in vivo by chromatin immunoprecipitation (ChIP) (Figure 3C). Consistent with PTEN being a direct target of GRHL3, expression of the gene was markedly reduced in the skin of Grhl3-nu\\ E18.5 embryos at both RNA and protein level (Figure 3D). Loss of PTEN leads to accumulation of PIP3, and as a consequence, increased activity of the serine/threonine kinases PDK1 and AKT, with resultant activation of the mTOR kinase complex 1 (mTORCl). This leads to activation of S6K1 and phosphorylation of 4EBP1 and ribosomal protein S6, which provide a robust readout of mTORCl signalling. Analysis of these downstream effectors of PI3K signalling in epidermis from Grhl3-nu\\ embryos revealed increased levels of PDK1, S6 and 4EBP1, as well as their phosphorylated forms, p-PDKl, p-S6 and p-4EBPl (Figure 3E). This increase in abundance and activity of these proteins reflects a high level of mTORCl signalling. The levels of the catalytic pllO isoform and the p85 regulatory subunit of the PI3K protein, which lie upstream of PTEN, were not altered. PTEN levels were markedly reduced in the basal layer of the epidermis, and co-incident with this, the levels of p-AKT were increased in this layer (Figure 3F). The size of the cells in the basal layer was also increased, a feature common to cells from m ice lacking PTEN.

Example 5 - Expression of PTEN Rescues the Tumorigenic Phenotypes Induced by Loss of GRHL3 To determine whether Grhl3 and PTEN interacted epistatically, mice carrying heterozygous deletions of the two genes were intercrossed to generate compound heterozygotes (Grh ^^'/PTEN^^'), and these animals were compared with Grhl3^{+ ~}/PTEN^{+ +} and Grh\3^{+ +}/PTEN^{+ ~} controls in the chemical- induced tumor model (Figure 4A). Although mice heterozygous for PTEN alone

display increased susceptibility to DMBA/TPA-induced tumors, the additional loss of a single Grhl3 allele markedly enhanced the development of multiple aggressive squamous cell carcinoma (Figure 4B, C). Conversely, the restoration of PTEN to wild type status in the Grhl3^+/~/PTEN^+/+ mice, completely safeguarded these animals against tumor formation. Taken together, this data suggests that PTEN is the critical downstream target of Grhl3 for prevention of squamous cell carcinoma. To confirm this in a human model, a human keratinocyte cell line (HaCAT) was generated in which the expression of GRHL3 had been knocked down using a specific shRNA containing lentivirus (GRHL3- kd) (Figure 4D). A line transduced with a scrambled control shRNA (Scr) served as the control. The level of PTEN was markedly reduced in the GRHL3-kd line compared to control, and p-AKT and p-S6 levels were increased in these cells in both the presence (Figure 4D), and absence of serum. This latter result was consistent with the resistance to growth factor and nutrient withdrawal observed with constitutive PI3K activation in tumor cell lines. The GRHL3-kd cells also proliferated more rapidly than the Scr control cells (Figure 4D). Re-introduction of wild type PTEN into the GRHL3-kd line restored the inhibition of phosphorylation of AKT and S6 (Figure 4D), and led to suppression of cell growth (Figure 4E). In contrast, introduction of the C124S phosphatase dead PTEN mutant, to equivalent levels to that achieved with the wild type protein, failed to inhibit AKT and S6 phosphorylation (Figure 4D), and also failed to suppress the enhanced proliferation of the GRH L3-kd cells (Figure 4E). These findings suggest that PTEN is the critical downstream target of GRHL3 in tumorigenesis.

Example 6 - Activation of PI 3 K/ AKT and Repression of Ras/MAPK/ERK Signalling in Gr 7/3-deficient Tumors

PTEN levels and PI3K signalling were compared in unaffected skin, papillomas, and squamous cell carcinomas derived from both wild type and Grhl3^A/~/K14Cre+ mice (Figures 5A and B). Loss of GRHL3 was associated with a reduction in PTEN in unaffected skin and almost complete loss in papillomas and squamous cell carcinomas that was not observed in the wild type skin or tumors. This was accompanied by increased levels of p-S6 and p-AKT in the tumors from Grhl3^A/~/K14Cre+ mice. Although p-S6 levels were also increased in the tumors from wild type mice, p-AKT expression was weak suggesting that the increase in p-S6 was due to other pathways feeding into the mTORCl pathway below the level of AKT. Previous studies have suggested that activation of H-ras and complete loss of PTEN are mutually exclusive in skin carcinomas. The present findings confirm this, as no mutations were detected in codons 12, 13, or 61 in the H-Ras gene in tumors derived from Grhl3^A/~/K14Cre+ mice, whereas 16% of tumors from the wild type mice carried mutations in these codons (Figure 5C). Similar frequencies of H-Ras mutations have been detected in series of human squamous cell carcinoma. An almost complete absence of p-ERKl/2 was observed in tumors from the Grhl3^A/~/K14Cre+ mice despite levels of ERK1/2 that were comparable to unaffected skin, and skin from wild type controls (Figures 5D and E). This finding was comparable to the reduction in p- ERK1/2 in the epidermis of Grhl3-nu\\ embryos. In contrast, the levels of p-ERKl/2 in tumors from wild type mice were markedly elevated (Figures 5D and E). Example 7 - A miR-21-dependent proto-oncogene network targets GRHL3 and PTEN

To establish the relevance of the mouse findings in the human system, the expression of GRHL3 was analyzed by Q-RT-PCR in 37 consecutive human squamous cell carcinomas, and the adjacent non- tumor-affected epidermis isolated by laser capture microdissection (LCM). In almost all cases, GRHL3 levels were markedly reduced in the tumors compared to the adjacent epidermis, with expression reduced by more than 90% in over half the samples (Figure 6A). A coordinate reduction in PTEN expression was also observed in these tumors (Figure 6A). Sequence analysis of the GRHL3 coding exons and splice donor and acceptor sites failed to detect any m utations in the tumors, and the methylation status of the CpG islands in the GRHL3 promoter was unchanged in tumors compared to normal skin (data not shown). An alternate mechanism for the reduction in GRHL3 expression in the tumors could be through overexpression of a specific microRNA (miRNA), which has been shown in some contexts to function as an oncogene by targeting tumor suppressors. To examine this, a miRNA array was interrogated with total RNA derived from two human squamous cell carcinomas with undetectable levels of GRHL3, and their matched normal skin controls (Figure 6B). The analysis focused on miRNAs that were predicted to target GRHL3 using the mirWIP database. Of this set, miR-21 exhibited the greatest differential between normal and tumor tissue, and its sequence aligned with nucleotides 414-436 in the GRHL3 3'UTR (Figure 6C). Interestingly, miR-21 has previously been shown to function as an oncogene by targeting PTEN. An additional 10 squamous cell carcinomas and their matched controls were examined and they were found to demonstrate a greater than 6-fold difference in miR-21 expression between the two groups (Figure 6D). This data confirmed that enforced expression of miR-21 could inhibit GRHL3 mRNA expression in the human HaCaT cell line (Figure 6E), in keeping with recent reports indicating that mRNA destabilization usually comprised the major component of miR-dependent gene repression. Expression of PTEN mRNA was also reduced in this line, consistent with the effect on GRHL3, and with direct targeting of PTEN by miR-21 (Figure 6F). GRHL3 and PTEN protein levels were also markedly reduced in the miR- 21 expressing HaCaT cells (Figure 6G), and this was accompanied enhanced cell growth. It was established that GRHL3 is a direct target of miR-21, using a luciferase reporter linked to the GRHL3 3'UTR (Figure 6H). Co-transfection of this construct with a miR-21 expression vector into HEK293 cells resulted in a marked reduction in luciferase activity compared to a scrambled control sequence, and this was reversed when an antagomir of miR-21 (miRZip21) was also transfected. Deletion of the miR-21 binding site in the GRHL3 3'UTR completely abrogated its effect (Figure 6H), indicating that GRHL3 is a direct target of miR-21.

Analysis of the perturbed miR-21/Grhl3/PTEN axis was extended into SCC cell lines derived from other tissues. SC9 (head and neck), and A431 (cervix) both demonstrate the same signature as primary tumors and the skin SCC line, SCC13. In contrast SCC4 (head and neck) does not display the characteristic changes indicating that only a subset of head and neck tumors may be responsive to the treatment strategies outlined above (Figure 7).

Example 8 - Elevated levels of miRNA-21 and reduced levels of Grhl3 are predictive for squamous cell carcinomas that will be susceptible to PI3K inhibitors

To validate the ability of miRNA-21 and/or Grhl3 levels to be predictive for susceptibility of squamous cell neoplasms to treatment with a PI3K in hibitor, samples of cells from squamous cell neoplasms having increased levels of miRNA-21 and/or reduced levels of Grhl3 are compared to samples of cells from squamous cell neoplasms having no statistically significant difference in the level of miRNA-21 and/or Grhl3 as compared to a control level.

Differential responsiveness to treatment in this context is illustrated in Figure 8. Three cell lines with the defined perturbation of the miR-21/Grhl3/Pten axis all demonstrate reduced growth with exposure to low concentrations of the dual inhibitor BEZ235. In contrast, SCC4 which lacks the gene expression signature only displays reduced growth at the highest concentrations of BEZ235 indicative of differential susceptibility of SCC depending on miR-21/Grhl3/PTEN levels.

Other SCC cell lines (derived from oral and skin SCC patients) such as SCC-15, SCC-25, SCC-68 and CAL-27 will be grown in specific culture media. The dual pll0-PI3K and mTORCl/2 inhibitory activity of NVP-BEZ235 will be assessed in these SCC cells. Dose-response experiments are predicted to show that NVP-BEZ235 is able to inhibit the phosphorylation of AKT, S6, and 4EBP1 in these cells.

Moreover, time course experiments will reveal if long-term exposure to low concentrations of NVP- BEZ235 will result in sustained inhibition of p-AKT, p-S6 and p-EBPl levels. HaCaT cell line will serve as a reference for the basal activation of PI3K/AKT cascade. Reduction of cellular proliferation in the SCC cells will be analyzed and associated with Gl arrest and induction of apoptosis. Markers of apoptosis such as cleavage products of caspase-3 and PARP will be detected by Western blot.

Viability and cell proliferation will be determined in comparison between cells with high miRNA-21 and/or reduced levels of Grhl3 (eg. SCC-9, SCC-13, SCC-15, SCC-25, SCC-68 and CAL-27), and cells having no statistically significant difference in the level of miRNA-21 and/or Grhl3 (eg. SCC-4) as compared to a control level (HaCaT cells). Under the tested experimental conditions, we establish a differential effect of NVP-BEZ235 according to the concentration and time used, which will confirm that increased levels of miRNA-21 and/or reduced levels of Grhl3 can serve to accurately predict which squamous cell neoplasms will be susceptible to treatment with the PI3K/mTOR inhibitor.

Different compounds will be also tested for their potential therapeutic properties in the SCC cells, including Everolimus and BKM-120 as well as others PI3K inhibitors. Example 9 - Administration of PI3K inhibitors protects against squamous cell neoplasm formation in genetically predisposed subjects

To examine the potential prophylactic action of the PI3K inhibitors, 2 groups of 30 Grhl3^t'^/~/K14Cre+ mice each (initiated with DMBA) were treated with vehicle alone or with NVP-BEZ235, coincident with the topical application of TPA (twice a week until the end of the experiment). The tolerable reasonable dose of inhibitor (35 mg/kg) was administered five times per week as a micro-emulsion diluted in distilled deionized water by oral gavage. Treatment was for 15 weeks. Mice were monitored twice weekly for tumour number, tumour size and progression. All animals in which a tumor progresses to more than 1cm X 1cm area were immediately euthanized by cervical dislocation and their tissues examined histologically. After 15 weeks of DM BA/TPA painting, the cohort of mice treated with NVP-BEZ235 was protected against tumor and SCC development. However, Grhl3^A/~ /K14Cre+ control animals treated with vehicle alone have developed tumors after 5 weeks of treatment. This experiment indicates that administration of PI3K inhibitors protects against squamous cell neoplasm formation in the genetically predisposed subjects (Grhl3^A/~/K14Cre+). Treatment for shorter periods (as little as four weeks) also conferred absolute protection from development of SCC, and the duration of treatment was also directly proportional to susceptibility to benign papillomatous skin lesions with mice treated for 8 and 12 weeks exhibiting fewer papillomas than mice treated for 4 weeks (Figure 9),

Other compounds and their respective controls will be tested and administered by oral gavage (eg. Everolimus at lOmg/kg and BKM-120 at 50mg/kg).

Example 10 - Topical administration of inositol protects against squamous cell neoplasm formation in genetically predisposed subjects

Grhl3 induction by Inositol was found to efficiently prevent neural tube defects and Inositol has shown anti-cancer characteristics in skin cancer following DMBA/TPA treatment. The effect of Inositol administration (or related compound) will be tested for the prevention of SCC development in adult mice through Grhl3 induction . All mice will be fed with Inositol-free chow and randomly assigned into two groups: control and Inositol-treated groups. We will use groups of 30 susceptible mice including Grhl3"^h/K14Cre+, Pten^+/", Grhl3^l'^/~/K14Cre+ Pten^+/", Grhl3^+/~ Pten^{+ "}, CT/CT and CT/CT Pten^+/~. Curly tail CT/CT mice (Grhl3 hypermorphic) are expected to be Inositol-responsive, whether the Grhl3 ^h/K14Cre+ mice will be Inositol-resistant. Within each group 15 mice will be treated topically with a standard moisturizing cream, whereas the second group of 15 mice will be applied the same cream containing 2% myo-lnositol or related compounds (eg. lnsP6). Bioactive Inositol will be administered concurrently with the DMBA/TPA carcinogens treatment for a period of 15 weeks. Mice will be monitored twice weekly for tumour number, tumour size and progression. All animals in which a tumor progresses to more than 1cm X 1cm area will be immediately euthanized by cervical dislocation and their tissues will be examined histologically. Topical administration of Inositol will show its important capacity as a preventive agent against skin SCC and also demonstrate that it is possible to propose topical use of Inositol-rich cream as a prophylactic application in genetically predisposed subjects (low Grhl3).

Claims

A method for predicting if a squamous cell neoplasm in a subject will be susceptible to treatment with a PI3K inhibitor, said method comprising:

a) determining the level of miRNA-21 and/or Grhl3 in tissue from the squamous cell neoplasm; and

wherein an increased level of miRNA-21 and/or a decreased level of Grhl3 in tissue from the squamous cell neoplasm, compared to the control level, indicates that the squamous cell neoplasm will be susceptible to treatment with a PI3K inhibitor.

A method for selecting a subject with a squamous cell neoplasm for PI3K inhibitor treatment, said method comprising:

c) selecting the subject for PI3K inhibitor treatment if the tissue from the squamous cell neoplasm has an increased level of miRNA-21 and/or a decreased level of Grhl3 as compared to the control.

A method for treating a squamous cell neoplasm in a subject, said method comprising: a) determining the level of miRNA-21 and/or Grhl3 in tissue from the squamous cell neoplasm;

d) administering a PI3K inhibitor to a subject selected by c).

The method of claim 1, wherein the method is carried out in vitro.

The method of claim 1, wherein the control level of miRNA-21 and/or Grhl3 is determined by measuring the level of miRNA-21 and/or Grhl3 in non-neoplastic tissue from the subject.

The method of claim 1, wherein the squamous cel l neoplasm is a squamous cell carcinoma or a head and neck cancer.

The method of claim 6, wherein the head and neck cancer is selected from the group consisting of mouth cavity, nasal cavity, sinus, lip, tongue, salivary gland, nasopharynx, larynx and throat cancer.

The method of claim 1, wherein the level of miRNA-21 is determined by Q-RT-PCR or array. The method of claim 1, wherein the level of Grhl3 is determined by Q-RT-PCR, Northern blot, or Western blot.

The method of claim 1, wherein the subject is a mammal. The method of claim 10, wherein the mammal is a human.

The method of claim 1, wherein the squamous cell neoplasm has a Grhl3 and /or PTEN genotype with no mutations.

The method of claim 9, wherein the level of Grhl3 in the squamous cell neoplasm is decreased by more than 90% as compared to the control.

The method of claim 1, wherein the tissue is skin tissue or mucous membrane tissue.

The method of claim 1, wherein the PI3K inhibitor is selected from the group consisting of A66 (C₁₇H₂₃N₅0₂S_2; CAS No.: 1166227-08-2); AS 252424 (5-[l-[5-(4-Fluoro-2-hydroxy-phenyl)- furan-2-yl]-meth-(Z)-ylidene]-thiazolidine-2,4-dione); AS-605240 (5-(6- Quinoxalinylmethylene)-2,4-thiazolidine-2,4-dione; C₁₂H₇N₃0₂S; CAS No.: 648450-29-7); AZD6482 (C₂₂H₂₄N₄0₄; CAS No.: 1173900-33-8); BAG956 (2-methyl-2-[4-(2-methyl-8-pyridin- 3-ylethynyl-imidazo[4,5-c]quinolin-l-yl)-phenyl]-propionitrile); BBD130 (2-Methyl-2-[4-(3- methyl-2-oxo-8-pyridin-3-ylethynyl-2,3-dihydro-imidazo[4,5-c]quinolin-l-yl)-phenyl]- propionitrile); BEZ235 (2-Methyl-2-[4-(3-methyl-2-oxo-8-quinolin-3-yl-2,3-dihydro- imidazo[4,5-c]quinolin-l-yl)-phenyl]-propionitrile; C₃oH ₅0; CAS No.: 915019-65-7); BKM- 120 (C₁₈H₂₁F₃N₆0₂; CAS No.: 1202777-78-3); CAL101 (C₂₂H₁₈FN₇0; CAS No.: 870281-82-6); D- 87503 (C₁₇H₁₅N₅OS; CAS No.: 800394-83-6); D-106669 (C₁₇H₁₈N₆0; CAS No.: 938444-93-0); Deguelin ((7aS_/13aS)-9,10-Dimethoxy-3,3-dimethyl-13_;13a-dihydro-3H,7aH-pyrano[2,3- c;6,5-f]dichromen-7-one); demethoxyviridin; GDC-0941 (2-(lH-lndazol-4-yl)-6-(4- methanesulfonyl-piperazin-l-ylmethyl)-4-morpholin-4-yl-thieno[3,2-d]pyrimidine bismesylate; C₂₃H₂₇N₇0₃S₂; CAS No.: 957054-30-7); GSK1059615 (GSK615; 5-[[4-(4-Pyridinyl)- 6-quinolinyl]methylene]-2,4-thiazolidenedione; C₁₈H N₃0₂S; CAS No.: 958852-01-1);

GSK2126458 (GSK212; C₂₅H₁₇F₂N₅0₃S; CAS No.: 1086062-66-9); IC87114 (C₂₂H₁₉N₇0; CAS No.: 371242-69-2); KU-55933 (2-Morpholin-4-yl-6-thianthren-l-yl-pyran-4-one); LY294002 (2-(4- Morpholinyl)-8-phenyl-4H-l-benzopyran-4-one; C₁₉H₁₇N0₃; CAS No.: 154447-36-6); 3- Methyladenine (3-Methyl-3H-purin-6-amine; C₅H₇N₅; CAS No.: 5142-23-4); MK-2206 (8-(4-(l- aminocyclobutyl)phenyl)-9-phenyl-8,9-dihydro-[l,2,4]triazolo[3,4-f][l,6]naphthyridin-3(2H)- one); myricetin (3,5,7-trihydroxy-2-(3,4,5-trihydroxyphenyl)-4H-l-benzopyran-4-one;

C₁₅H₁₀O₈; CAS No.: 529-44-2); NU 7026 (2-(4-Morpholinyl)-4H-naphthol[l,2-b]pyran-4-one; C₁₇H₁₅N0₃; CAS No.: 154447-35-5); NU 7441 (8-Dibenzothiophen-4-yl-2-morpholin-4-yl- chromen-4-one); OSU-03012 (2-Amino-N-[4-[5-(2-phenanthrenyl)-3-(trifluoromethyl)-lH- pyrazol-l-yl]phenyl]-acetamide; C₂₅H₁₉F₃N₄0; CAS No.: 742112-33-0); Perifosine (1,1- dimethylpiperidinium-4-yl octadecyl phosphate); PI103 (3-[4-(4-

Morpholinylpyrido[3',2':4,5]furo[3,2-d]pyrimidi n-2-yl]phenol hydrochloride; C₁₉H₁₆N₄0₃; CAS No.: 371935-74-9); PI828 (2-(4-Morpholinyl)-8-(4-aminopheny)l-4H-l-benzopyran-4- one; C₁₉H₁₈N₂0₃; CAS No.: 942289-87-4); PIK-293 (C₂₂H₁₉N₇0; CAS No.: 900185-01-5); PIK-294 (C₂₈H₂₃N₇0₂; CAS No.: 900185-02-6); PIK75 (N'-[(lE)-(6-bromoimidazo[l_;2-a]pyridin-3- yl)methylene]-N,2-dimethyl-5-nitrobenzenesulfonohydrazide hydrochloride); PIK90 (N-(7,8- Dimethoxy-2,3-dihydro-imidazo[l,2-c]quinazolin-5-yl (-nicotinamide; C₁₈H₁₇N₅03; CAS No.: 677338-12-4); PIK93 (C₁₄H₁₅CIN₃0₄S₂; CAS No.: 593960-11-3); PKI-587 (C₃₂H₄₁N₉0₄; CAS No.: 1197160-78-3); PP-121 (l-Cyclopentyl-3-(lH-pyrrolo[2,3-fa]pyridin-5-yl)-lH-pyrazolo[3_;4- c ]pyrimidin-4-amine; C₁₇H₁₇N₇; CAS No.: 1092788-83-4); PX-866 ([(3aR_;6f_;9S_;9aR_;10R_;llaS)- 6-[[bis(prop-2-enyl)amino]methylidene]-5-hydroxy-9-(methoxymethyl)-9a_;lla-dimethyl- l^ -trioxo^^^a^ O^ll-hexahydroindeno^S-^lisochromen-lO-yl] acetate; C₂₉H₃₅N0₈; CAS No.: 502632-66-8); quercetin (sophoretin; C₁₅H₁₀O₇; CAS No.: 117-39-5); SF1126 ((3S)-4- [[(lS)-l-carboxy-2-hydroxyethyl]amino]-3-[[2-[[(2S)-5-(diaminomethylideneamino)-2-[[4- oxo-4-[[4-(4-oxo-8-phenylchromen-2-yl)morpholin-4-ium-4- yl]methoxy]butanoyl]amino]pentanoyl]amino]acetyl]

amino]-4-oxobutanoic acid acetate; C ₁H₅₂N₈0₁₆); tandutinib (1-piperazinecarboxamide, 4- [6-methoxy-7-[3-(l-piperidinyl)propoxy]-4-quinazolinyl]-N-[4-(l-methylethoxy)phenyl]-); tetrodotoxin citrate; TG100-115 (C₁₈H₁₄N₅0₂; CAS No.: 677297-51-7); TGX-115 (8-(2- Methylphenoxy)-2-(4-morphonilyl)-4(lH)-quinolinone; C₂₀H2oN₂0₃; CAS No.: 351071-62-0); TGX-221 (7-Methyl-2-(4-morpholinyl)-9-[l-(phenylamino)ethyl]-4H-pyrido-[l_;2-a]pyrimidin- 4-one; C₂₁H₂₄N₄0₂; CAS No.: 663619-89-4); thioperamide maleate; WHI-P 154 (2-Bromo-4- [(6,7-dimethoxy-4-quinazolinyl)amino]phenol; C₁₅H₁₄BrN₃0₃; CAS No.: 211555-04-3);

wortmannin; XL147 (C₂₁H₁₅N₅0₂S₂; CAS No.: 956958-53-5); XL765 (C₃₁H₂₉N₅0₅S; CAS No.: 1123889-87-1); ZSTK474 (C₁₉H₂₁F₂N₇0_2; CAS No.: 475110-96-4).

A kit detecting if a squamous cell neoplasm in a subject will be susceptible to treatment with a PI3K inhibitor, said kit comprising:

A kit for treating a squamous cell neoplasm in a subject, said kit comprising:

a) reagents for determining the level of miRNA-21 and/or Grhl3 in tissue from the squamous cell neoplasm;

b) a PI3K inhibitor; and

The kit of claim 16 or claim 17, wherein the reagents for determining the level of miRNA-21 and/or Grhl3 comprise one or more of Q-RT-PCR reagents, array reagents, Northern blot reagents, Western blot reagents, and ELISA reagents.

19. A method for preventing a squamous cell neoplasm in a subject, said method comprising applying a topical composition comprising inositol to skin of the subject.