Trial Information

Microsatellite Instability in Anal Squamous Cell Carcinomas of HIV-Positive Versus HIV-Negative Patients

Background:

In the highly active antiretroviral therapy (HAART) era, the incidence of many cancers
remains higher in the HIV-positive than in the general population. This is not only true for
AIDS-defining malignancies, such as Kaposi's sarcoma or lymphoma, but also for other
cancers, which comprise 58% of all neoplasms in a recently described US cohort [1]. This
trend has generated renewed interest in understanding the molecular mechanisms of
carcinogenesis in the immune suppressed patient. Potential explanations include prolonged
survival of patients with HIV on HAART, high incidence of co-infection with oncogenic
viruses, and exposure to risks such as smoking and alcohol [2].

Squamous cell carcinoma of the anus (SCCA) requires integration of human papillomavirus
(HPV) DNA into anal canal cell chromosomes. HIV-positive men are at increased risk for
developing anal cancer. A Swiss study linking the Swiss HIV cohort study and Swiss cantonal
cancer registries has shown greatly elevated SIRs for anal cancer (SIR = 33.4, 95% CI =
10.5-78.6) [3]. Clearly, HPV-HIV coinfection plays a major role in SCCA development, but
whether this contribution is associated with distinct molecular pathway of carcinogenesis,
remains hypothetical [4].

Anal cancer occurs earlier in HIV-positive individuals (mean age 37 years) compared with
HIV-negative men (58 years) and HIV-negative women (65 years). This difference of two
decades in age of onset supports the hypothesis that the biology of SCCA differs between
HIV-positive and HIV-negative patients [5]. The effect of HIV infection on the natural
history of anal HPV infection is poorly understood, but most experts agree on two points: 1)
95% of HIV-positive homosexual men are co-infected with HPV; and 2) HIV infection favors
persistence of HPV infection within the ano-genital tract (uterine cervix, vagina, and anal
canal).

Two major and mutually exclusive types of genomic instability are involved in cancer
progression. The first, known as chromosomal instability (CIN), results from a series of
genetic changes, including activation of oncogenes and inactivation of tumor-suppressor
genes such as p53 and APC. The second, known as microsatellite instability (MSI), results
from somatic (acquired) inactivation of DNA mismatch repair genes (MMR), such as MLH1 (by
hypermethylation of its promoter), typically leading to mutation of genes with coding
microsatellites, such as BAT 25, BAT26 and CAT25, transforming growth factor receptor II
(TGF-RII) and BAX [6].

MSI reflects genomewide instability (resulting from a deficient MMR system), and occurs at
increased frequency in HIV-associated cervical intraepithelial and lung neoplasia. In both
cases, the frequency of MSI was six fold greater in the HIV-associated tumors, compared with
tumors of HIV-indeterminate patients [7]. Others authors have also reported a high rate of
MSI in Kaposi's sarcomas and aggressive lymphomas obtained from HIV-infected patients,
whereas there is no evidence of similar instability in lesions from HIV-negative patients
[8]. The recent observation that the MSI phenotype is restricted to HIV-related lymphomas
suggests that a cardinal feature of cancer, such as genetic instability, can be highly
influenced by host immunity [9].

MSI status of SCCA in HIV-positive patients has never been investigated, but we have
performed a CIN analysis of tumor specimens from 18 HIV- and 10 HIV+ patients diagnosed with
SCCA in two US institutions [10]. Tumors in HIV-negative patients were more likely to
present CIN than were tumors in HIV-positive patients (24.1% versus 6.6%, P<0.001).
Consistent loss of heterozygosity (LOH) on chromosomes 17p, 18q, 5q, and 11q was observed in
HIV-negative patients with SCCA. By contrast, allelic loss at 17p, 5q, and 18q was virtually
absent in tumors of HIV-positive individuals, indicating that immune suppression promote
SCCA progression through an alternate pathway to CIN.

Own Research in the Field:

The applicant is a colorectal surgeon, directly involved in the treatment of patients with
anorectal malignancies since 2001. He has collected preliminary data showing specific
patterns of molecular aberrations in HIV-related SCCA. These findings have been published in
peer-reviewed journals, with the consequence that the applicant is regularly solicited by
Editors to write expert opinions and state of the art reviews in the field.

The Swiss cohort of HIV-positive patients will provide a unique opportunity to conduct this
research, because of the large number of individuals already enrolled in the existing
database, and the existing expertise in the field developed by Professor Hirschel and his
team. A large population of patients is indeed a sine qua none condition to start this
project due to the relative rarity of these tumors.

Last but not least, the molecular aspect of this research will be conducted in the
laboratory of Dr Pierre Hutter, PhD, who is a leader in our country for the detection of
hereditary non-polyposis colorectal cancer (HNPCC). Thus, Dr Hutter laboratory will provide
an excellent environment to perform the molecular analyses described herein.

Study Hypotheses:

We hypothesize that, in the HIV population, microsatellite instability, rather than
chromosomal instability, is the favoured pathway for rapid progression of anal
intra-epithelial neoplasia towards invasive SCCA. The second hypothesis is that, by
comparison, MSI is either absent or at least very rare in SCCA of HIV-negative individuals.

Study Aims and Objectives

The primary objective of this study is to assess microsatellite instability status in tumor
specimens of 15 HIV-positive patients from the Swiss Cohort who were diagnosed with SCCA; an
interim analysis will then be performed to determine the incidence of MSI in SCCA of
HIV-positive patients. If a majority of tumours exhibit MSI in this population, we will
proceed with the second part of the study, which objective is to perform the same analyses
in similar tumors of 15 HIV-negative patients, and thus to compare the molecular profiles of
SCCA in both populations. By doing so, the potential role of immune suppression in
carcinogenesis progression might be established.

Study Design and Plan

Patients:

Patients from the Swiss HIV cohort with a biopsy-proven diagnosis of squamous cell carcinoma
of the anus will be identified using the existing database. We choose to analyse at least 15
samples, depending on availability, between 3 SHCS centers. Tumors from HIV-negative
patients who underwent surgery for SCCA will be retrieved from the Department of Pathology
of University Hospital Geneva, pending review of the protocol by the local Ethics Committee.

Intervention:

All tumor samples will be reassessed for confirmation of diagnosis by a pathologist. For
each patient, paraffin blocks with both tumor tissue and normal mucosa will be selected for
DNA extraction and polymerase chain reaction at microsatellite targets. The tissues were
routinely collected from the operating room, fixed in buffered formalin, embedded in
paraffin, and stored for a variable number of months prior to selection for analysis.
Paraffin-embedded blocks will be cut with a microtome into 20-µm thick sections. Using a
sterile scalpel blade, areas of normal (non-tumor) and cancer tissue will be microdissected
under a dissecting microscope utilizing a haematoxylin and eosin-stained section as a guide.
The specimens will be deparaffinised in toluol, purified with absolute ethanol, and
centrifuged at 14,000 rpm. Previous experience in Dr Hutter's laboratory has demonstrated
that it is indeed technically possible to perform this type of molecular analysis in
paraffin-embedded tumour specimen.

We will use the reference panel of microsatellite primers recommended for colorectal cancer
specimens to determine the presence of microsatellite instability in SCCA. These include the
microsatellite markers BAT25, BAT26, as well as CAT25.

Study Evaluations:

Initial evaluation of data will be performed after analysis of the first 15 tumours
specimen. If MSI is detected in the HIV-positive population, further investigation will be
conducted in tumors of HIV-negative patients. Data analysis will be subject to progress
reports conducted with all investigators at regular intervals. We chose to restrict, at
least initially, our analysis to 15 tumours for three reasons:

1. This is a relatively rare cancer, and it is unlikely that we will be able to retrieve
more than 25 SCCA patients from the Swiss HIV cohort.

2. Our previous study was conducted with 18 tumours specimen in each group, and the
differences between groups were statistically significant.

3. This is a pilot exploratory study, and the working hypothesis may prove wrong; it is
therefore critical to limit the initial costs in accordance with the study budget and
availability of Dr Hutter's expertise. If our hypothesis is supported by the initial
results, we may proceed with additional investigations and a more elaborated research
project with adequate funding resources.¨ For these reasons, we will restrict our
research to three SHCS centers. Zurich, Geneva and Lausanne were chosen essentially
based on existing relationship between the visceral surgery units.

Data Analysis:

MSI status of tumors will be assessed by Pierre Hutter. The results will then be correlated
with the epidemiological data from the Swiss cohort, and correlated with the patients`
clinical characteristics, including viral load, CD4+ count, as well as clinical outcome. The
results will also be correlated with treatment outcome and cancer survival.

Time Frame:

Tumor specimen will be retrieved from the Departments of Pathology in Lausanne, Geneva and
Zurich in the beginning of 2009. The MSI analysis is a straightforward procedure, which
takes about 24 hours in the laboratory of Dr Hutter. Thus, the analysis of the 15 specimen
in the HIV-positive population will be completed by the end of March 2009. Further decision
regarding subsequent analyses in the HIV-negative population will then be subjected to
decision, according to the results obtained in these initial tumors.

Status of Ethical Approval:

Patients enrolled in the Swiss HIV cohort study initially gave their consent for
participating in this type of study. Almost all patients at the HIV centres consented to
take part in genetic research in order to identify specific factors that might result in
differences in the course of the disease. Obviously, all analyses will be performed on tumor
specimen, and these will not be traceable to the patients' identities. Thus, all genetic
testing will be totally anonymous. By contrast, further approval by the local Ethics
Committee in Geneva will be sought, should the analysis be performed on tumor specimen from
HIV-negative patients.

Study budget:

The initial analysis of 15 HIV-positive patients will be performed free of charge by Pierre
Hutter. The cost for molecular study is estimated at 250.-CHF per case. Thus, in case the
analysis of MSI in this group leads to promising results, it is estimated that the cost of
analysis in 12-15 HIV-negative patients will cost around 3,000.-CHF. Funding for this type
of research will be available through the Fonds de Service of Department of Surgery at
University Hospital Geneva.