WO1997013875A1

WO1997013875A1 - Methods for predicting the outcome of persistent hbv infection and the outcome of cytokine therapy

Info

Publication number: WO1997013875A1
Application number: PCT/GB1996/002519
Authority: WO
Inventors: Mark Richard Thursz; Howard Christopher Thomas; Adrian Vivian Sinton Hill; Dimitris Mantafounis
Original assignee: Imperial College Of Science, Technology & Medicine
Priority date: 1995-10-13
Filing date: 1996-10-14
Publication date: 1997-04-17
Also published as: EP0862652A1; AU7310496A

Abstract

Methods for determining the outcome of cytokine therapy in subjects suffering from chronic virus infections are provided, as well as methods for predicting the outcome of persistent virus infections, eg persistent HBV infections. Kits for use in such methods are also provided.

Description

METHODS FOR PREDICTING THE OUTCOME OF PERSISTENT HBV INFECTION AND THE OUTCOME OF CYTOKINE THERAPY

The present invention relates to methods of predicting the outcome of persistent HBV infection in a subject, as well as the outcome of cytokine therapy, particulariy interferon therapy, in patients suffering from Chronic Hepatitis infection, particularly chronic HBV infection.

Approximately 300 million people worlwide are persistently infected with hepatitis B virus (HBV) . In people exposed to the virus after the neonatal period the majority recover and only 5-10% will develop a persistent infection. Up to 50% of these are likely to die prematurely from end stage liver disease or hepatocelluiar carcinoma which are complications of HBV infection (Beasly et al , J. Infec . Dis . , 2:1129-33

(1981) ) . Viral elimination may be induced by treatment with interferon alpha or lymphoblastoid interferon. However, the response rate for this therapy is limited, eg only around 40% in the case of chronic HBV.

In addition, interferon therapy can also be unpleasant as the majority of patients suffer influenza-type symptoms. More severe side effects include suppression of bone marrow, abnormalities of thyroid function and endogenous depression. Interferon treatment is also expensive and thus there are pressures to rationalise the use of such treatment within the healthcare community.

Twin studies have established that the outcome of HBV infection, transient or persistent infection, is, in part, determined by genetic susceptibility (Lin et al, Anticancer Research, 9:737 (1989)) . We have previously shown that the MHC class II allele DRB1*1302 is associated with resistance to persistent mfecton (Thursz et al , N. Eng . J. Med . , 3321065 (1995)) and the codon 52 variant of the mannose binding protein gene with susceptibility to persistent HBV infection (Thomas et al , Lancet (1996) in press) .

There is a bi-allelic polymorphism of the promoter for the TNF alpha gene. The mutation is located at position -308 icldLive to che transcription initiation sequence of the TNF alpha gene (Mcguire, et al , Nature, 371:508- 510 (1994)) . The rare TNF-2 allele is associated with higher constitutive and inducible transcription of the TNF alpha gene. The TNF-2 allele is found to be in linkage disequilibrium with the B8-DR3 haplotype m

Caucasian subjects. At position -308, the rarer allele, known as TNF2, is associated with increased TNF production (Wilson et al , Eur . J . Immunol . , 24(l) :191-5

(1994) ) . This allele and the -238 variant whose functional significance is unknown, have both ceen associated with adverse outcomes from other infectious diseases such as P. alciparum malaria.

TNF mhiDits the expression of HBV in HBV transgenic mice (Daniels et al , Lancet , 335:875 (1990) ; Guilhot et al, J. virol . , 67 (12) :7444-9 (1993) ) and TNF production is higher m patients with chronic HBV infection who respond successfully to interferon therapy This phenomenon occurs towards the end of a period of treatment and is therefore not of use in identifying, prior to treatment, patientε likely to respond to interferon.

Furthermore, a study of factors which determine the response to interferon in Caucasians with chronic HBV infection, identified an association with the MHC haplotype A1-B8-DR3 (Scully, et al , Hepatology, 12:1111- 1117 (1990) wnich is m strong linkage disequilibrium with the TNF2 allele m Caucasian populations.. Of 11 patients wich this haplotype, 10 responded to interferon whilst only 1 did not. This one patient was subsequently found to have a lymphoma which may have impaired his immune function and thus impaired his ability to eradicate the virus even with the help of interferon.

There thus exists a clinical need for a method or predictive test which would allow assessment of subjects suffering from a chronic hepatitis virus infection whicn, in turn, would allow cytokine, eg interferon, treatment to be targeted to those patients with a reasonable chance of responding.

We have now found that all patients who carry the TNF-2 allele respond to interferon treatment. However, for patients nomozygous for the TNF-1 allele it must oe assumed that only about 40% will repond to interferon treatment .

Polymorphism of any cytokine or cytokine promoter, including Alpha interferon subtypes, gamma interferon, IL2 , IL4, IL5, IL6 , IL10 and IL12, could also be expected to influence outcome, not only of hepatitis B infection, but also hepatitis C, hepatitis G, human papilloma virus, human immunodeficiency virus and other persistent virus infections. Thus, sensible targeting of cytokine therapy should be possible.

We nave also established that the presence of the TNF-2 allele is also itself linked with the development of persistent HBV --nfection.

Thus, in a first aspect, the present invention provides a method for assessing and/or predicting the probable outcome of treating a subject suffering from a persistent virus infection with a cytokine comprising the step of determining whether the subject carries one or more alleles associated with an improved prooability of a therapeutic response when treated with said cytokine.

In one embodiment of this aspect of the invention, the persistent virus infection is a hepatitis virus infection, particularly hepatitis B. In the case of the latter the method comprises determining whether the subject carries the TNF-2 allele.

In a second aspect the present invention provides a method for assessing and/or predicting the probable outcome of hepatitis virus infection m a subject comprising the step of determining whether the subject carries the TNF-2 allele.

In one embodiment of this aspect of the invention, the hepatitis virus infection is hepatitis B virus (HBV) infection, and in particular the method allows a determination of whether the subject s more likely to develop a persistent HBV infection.

The preferred method of carrying out the determination for both these aspects of the invention is to analyse a sample of the subject's DNA. Such a sample can conveniently be obtained from a biological sample, eg a blood samσle. Suitably, the DNA obtained from the biological sample will be amplified using techniques well .no n to those skilled n the art, eg PCR techniques. For example, the TNF alpha gene region, and more particularly, the TNF alpha promoter region, can be amplified. Such techniques will involve the use of at least one pair of suitable primers. Suitable primers can be chosen on the basis of the DNA sequence of the cytokine gene in question , eg the TNF alpha gene.

In the context of the present invention, TNF alpha gene region can mean the whole of the TNF alpha gene or, alternatively, a part thereof. Clearly, however, if only a part is amplified it should include the promoter for the gene since this is where the TNF-2 allele will be found , if it is present .

In a third aspect, therefore, the present invention provides a kit for use in a method for assessing and/or predicting the probable outcome of treating a subject suffering from a persistent virus infection with a cytokine, which comprises at least one pair of primers suitable for PCR amplification of at least a portion of the gene coding for the cytokine.

In a fourth aspect, therefore, the present invention provides a kit for use n a method for assessing and/or predicting the probable outcome of hepatitis virus infection in a subject, which comprises at least one pair of primers suitable for PCR amplification of at least a portion of the gene coding for the TNF alpha.

Preferreα features of each aspect of the invention are as for each other aspect, mutatis mu tandis . The invention will now be described by reference to the following examples, which should not be construed as in any way limiting the invention.

Example 1

A group of 14 patients suffering from chronic hepatitis B virus infection were studied. All the patients had been undergoing treatment with either alpha interferon or lymphoblastoid interferon for at least three months.

Samples of peripheral blood were collected for each patient and DNA extracted using a Nucleon II kit (Scott labs) . DNA was apmplified using PCR and TNF alpha promoter alleles were detected using dot blot hybridisation using the method described by McGuire et al , supra .

Of the 14 patients, 11 were responding to the interferon treatment. Of these 11, 5 were homozygote for the common TNF-1 allele and 6 were either heterozygote or homozygote for the TNF-2 allele Of the three patients wno failed to respond to the treatment, none carried the TNF-2 allele (ie they were all homozygous for the TNF-1 allele) .

Thus, t would appear that the TNF-2 allele confers an increased probability of response to interferon.

Example 2

A large case-control study was performed on a population m the GamJDia. Participants were all west Africans living in the western coastal region of the Gambia in the area surrounding the capital, Banjul. Two different groups were recruited for the study between 1988 and 1990. In the first group, children up to 10 years were recruited at the Royai Victoria Hospital, Banjul and the Medical Research Council MRC) Hospital, Fa ara wnere they had been seen for a variety of conditions unrelated to HBV. The second group were adult healthy male blood donors. Both populations nad previously been studied as part of a case-control study of susceptibility to malaria. Ethical permission for this study was granted by the MRC Gambian Government joint Ethics Committee. Consent for phlebotomy was obtained from each child's parent or guardian.

Participants were divided into groups according to the result of serological tests for HBV. People who had never been ex^posed to HBV, were HBV core antibody (anti-HBc) negative. People who had spontaneously recovered from HBV infection were IgG anti-HBc positive and HBV surface antigen ⁽HBsAg⁾ negative. People with persistent HBV infection were IgG anti-HBc and HBsAg positive. People with acute HBV infection, who may not have had sufficient opportunity to eliminate HBsAg _\HBsAg positive and IgM anti-HBc positive) were excluded from the analysis. ₁07 people wno haα been vaccinated against HBV (_<10%) were classified m the uninfected group and were therefore not included in the analysis of the TNF allele frequencies. The five individuals with HIV antibodies (_<1%) were also excluded.

Seroloσical Testing

Plasma samples were taken from all participants and stored at -20°C. Anti-HBc (total) , anti-HBc (IgM) , HBsAg status and anti-HBs concentration were dtermined cy ELISA according to the manufacturers instructions (Boehringer Mannheim, Munich, Germany) . HIV antibodies were dtermined by Wellcozyme ELISA (Wellcome, Beckenham, UK) Positive results were confirmed by western blot .

TNF promoter -308 variant genotyping

The polymerase cnam reaction was used to amplify a 519 base pair fragment of the promoter from position -502 to +17. The primer sequences were: 5' -CAAACACAGGCCTCAGGACTC-3 ' 3 ' -AGGGAGCGTCTGCTGGCTG-5'

About lOOng (2μl) of genomic DNA which had been extracted frcm peropheral blood was added to 25ul of reaction mix which contained 0. lμM of each primer. lOOμM of each dNTP, 67mM tris-HCl, 16mM (NH -SO., 2mM MgCl₂ and 0.01% Tween- 20.

The reaction mix was heated to 95°C for 10 minutes then 1.5U Taq DNA polymerase was added. The reaction mix was then given 35 cycles of 95°C for 1 minute, 60°C for 1 minute and 72°C for 1 minute m a Perkm-Elmer 9600 Thermocycler with heated lid. The PCR product was finally left at 72°C for 10 minutes.

The PCR product was denatured with 0.4M NaOH, dotted on to nylon membranes (Amersham High-Bond N) using a vacuum manifold and fixed for 1 minute with UV light. Allele- specific oligonucleotides were end labelled with digoxigenm-ddUT? and hybridized to the membrane with 3M TMAC (tetramethyiammonium chloride) solution at 55°C for 1 hour. Excess probe was removed by washing m 2 x SSPE/0.1% SDS at room temperature and subsequently in 3M TMAC at 58°C. The membranes were treated with anti- digoxigenm Fab fragments and, after washing to remove excess antibody, they were incubated with Lumigen PPD and exposed to X-ray film for 15-30 minutes. Allele type was scored by two independent observers. The accuracy of this method was confirmed by DNA sequencing.

-308 variant oligonucleotide probes: wild type 5 ' -AGGGGCATGGGGACGGG-3 ' -308 variant 5 ' -AGGGGCATGAGGACGGG-3 '

TNFα promoter -238 variant genotyping

The -238g allele was typed by amplification refractory mutation system PCR using a conserved primer (5'- AAGACCCCCCTCGGAATCA-3 ' ) together with sequence speciric primers for either the -238g allele (5'- AGACCCCCCTCGGAATCG-3' ) or the 238a allele (5'- AAGACCCCCCTCGGAATCA-3' ) to generate a 459bp product. As a positive control each reaction mixture contained the following primers:

5' -CCAAAGATTCAGGTTTACTCACG-3 ' ; and

5' -ACTTAACTATCTTGGGCTGTGAC-3 ' to amplify a 266bp fragment of the human c.2 microglobulm gene. Negative controls, using water instead of template, were included on each plate. Each 25αi reaction mixture contained lOOng genomic DNA, 67mM Tris HCl, 16mM ammonium sulphate, 2mM magnesium chloride, lOOμM of each dNTP, 0.5 units of Taq polymerase (Biolme) , 0. lμM of each TNF primer and 0.2μM of each c.2 microglobulm primer. The mixture was incubated at 95°C for 5 mmutes followed by 5 cycles of (1 minute at 95°C, 1 minute at 67°C, 1 minute at 72°C) , then 25 cycles of (1 minute at 95°C, 1 minute at 62°C, 1 minute at 72°C) and finally 10 mmutes at 72°C. The products were resolved on a 2% agarose gel stained with ethidium oromide and visualised under UV light. Statistical analysis

Univariate analysis was performed by comparing the proportion of suojects with each allelic variant between cases ⁽HBsAg positive) and controls (HBsAg negative) using the crude odds ratio, exact 95% confidence intervals and p values calculated using the χ² test. Mult variate analysis was performed using a logistic regression model to allow for the interaction of the TNF alleles with HLA class II alleles and with other possible confounding variables εuch as age, sex, ares of les deuce and ethnic background, each term was added to the model in a stepwise method and selected for inclusion when it resulted m a significant improvement of the model with a statistical significance of p>0.05

Results

Position -308 variants

Table 1 Allele frequencies for the TNFα promoter -308 variants supjects with persistent HBV 'HBsAg positive anti-HBc dgM) negative) and with self limiting HBV infection (anti-HBc (total) positive, HBsAg negative) .

-308 -308 Sample

Heterozygotes Homozygotes Size

(%) (%)

Self limiting 24.5 3.6 332 infection

Persistent

HBV 32.7 3.5 171 infection

HBeAg positive 34.3 4.0 99

HBeAg negative 36.1 2.7 36

In the persistently infected group there were 6 (3.5%) homozygotes for TNF2 and 56 (32.7%) heterozygotes, compared to 12 (3.6%) homozygotes and 80 (24.1%) neterozygotes the transiently infected group. In analysing these data it was necessary to stratify for HLA DRB1*1302, which is associated with HBV clearance and is m linkage disequilibrium with the TNF2 allele this population. After stratification, the odds ratio for persistent HBV infection in subjects with the TNF2 allele was 1.56 (95% confidence interval 1.03-2.38) When the children w th severe malaria (associated with TNF2) are excluded from the analysis the odds ratio for persistent infection for subjects with TNF2 is 2.03 (95% C.I. 1.02- 4.05) , p=0.04. In the logistic regression model, terms were included for TNFα -308 genotype, ethnic background, area of residence, the presence of cerebral malaria, the presence of DRB1*1302 and the presence of HLA-DQ8. The variables TNFα -238 genotype, age and sex were rejected because they did not result in a significant improvement in the model. Using this model the associateion of TNF2 with persistent HBV infection was confirmed with p=0.032.

Position -238 variants

Table 2 : Allele frequencies for the TNFα promoter -238 variants in subjects with persistent HBV (HBsAg positive anti-HBc (IgM) negative) and with self limiting HBV infection (anti-HBc (total) positive, HBsAg negative) .

-238 -238 Sample

Heterozygotes Homozygotes Size

(%) (%)

Self limiting 12.0 1.1 194 infection

Persistent

HBV 9.8 3.3 βl infection

HBeAg positive 9.6 3.2 31

HBeAg negative 0 5

⁰

The frequency of these alleles were similar in the transient and persistently infected subjects. Discussion

These results demonstrate that the -308 allelic variant (TNF2) but not the -238 variant, is associated with incresed susceptibility to persistent HBV infection with a relative risk of approximately 2. This finding is at first sight surprising as previous studies have indicated that TNF has antiviral properties and it would be anticipated that alleles associated with increased TNF production (TNF2) would be associated with enhanced viral clearance. However, a possible explanation can De found if recent observations that activation induced cell death in mature lymphocytes can be mediated through the TNF receptor as well as through the better known apoptosis receptor, Fas are taken into account. These studies describe inhibition of virus-specific CD8+ T cell expansion mediated by TNF suggesting that disregulation of TNF production may actually produce tolerance to viral antigens (Zheng et al , Na ture , 377:348-51 (1995)) . These in-vitro phenomena may contribute to the failure of patients with persistent HBV infection to mount an adequate CD8+ cytolytic response comparable to that observed in acute infection followed by recovery (Bertoletti et al , PNAS USA , 88 (23) : 10445-9 (1991)) .

Claims

CLAIMS :

1 A method for assessing and/or predicting the propable outcome of treating a subject suffering from a persistent virus infection with a cytokine comprising the step of determining whether the subject carries one or more alleles associated with an improved probability of a therapeutic response when treated with said cytokine.

2. A method for assessing diid/ox predicting the probable outcome of hepatitis virus infection in a subject comprising the step of determining whether the subject carries the TNF-2 allele.

3. A method as claimed in claim 1 wherem the persistent virus infection is hepatitis B infection, hepatitis C infection, hepatitis G infection, human papilloma virus infection or human immunodeficiency virus infection.

4 A methoα as claimed in claim 3 wherein the persistent virus infection is a hepatitis virus infection.

S. A method as claimed in claim 4 wherein the hepatitis virus infection is a chronic hepatitis B virus infection.

6. A method as claimed in any one of claims 1 to 5 wnerem the cytokine is lymphoblastoid interferon, Alpha interferon, including subtypes thereof, gamma interferon, IL2, IL4, IL5, IL6 , ILIO or IL12.

~" A method as claimed in claim 6 wherein the cytokine is lymphoblastoid interferon or Alpha interferon.

8. A method as claimed in claim 7 wnerem it is determined wnether the subject carries the TNF-2 allele.

9. A method as claimed in claim 2 wherein the hepatitis virus infection is a hepatitis B (HBV) virus infection.

10. A method as claimed m claim 9 wherein the method is for determining whether the subject is more likely to develop a persistent HBV infection.

11. A method as claimed m any one of claims 1 to 10 wherein the determination is carried out using DNA obtamed from a biological sample.

12. A method as claimed in claim 11 wnerem the biological sample is a blood sample .

13. A kit for use in a method for assessing and/or predicting the probable outcome of treating a subject suffering from a persistent virus infection with a cytokine, wnicn comprises at least one pair of primers suitable for PCR amplification of at least a portion of the gene coding for the cytokine.

14. A kit as claimed in claim 13 modified by any one or more of the features of any one or more of claims 3 to 8.

15. A kit for use in a method for assessing and/or predicting the probable outcome of hepatitis virus infection m a subject, which comprises at least one pair of primers suitable for PCR amplification of at least a portion of the gene coding for the TNF alpha.

16. A kit as claimed in claim 15 modified by any one or more of the features of claim 9 or claim 10.

17. A kit as claimed in any one of claims 13 to 16 which includes one or more cf the following pairs of primers :

-.

5' -CAAACACAGGCCTCAGGACTC-3' 3' -AGGGAGCGTCTGCTGGCTG-5' ;

5' -AAGACCCCCCTCGGAATCA-3' with either 1) 5' -AGACCCCCCTCGGAATCG- 3 ' or ii) 5' -AAGACCCCCCTCGGAATCA-3' ; or

5' -CCAAAGATTCAGGTTTACTCACG-3 ' 5' -ACTTAACTATCTTGGGCTGTGAC-3' .