Projects

HLA Expression and Cancer

HLA and Cancer Questionnaire

April 2005 Update

Several significant advances were made during the 13^th Histocompatibility Workshop HLA and Cancer component. Monoclonal antibodies were developed for use in detection of HLA class 1molecules on tumour tissue, which were validated and sent to participating laboratories for useon fixed tissue. These monoclonals were directed at epitopes present on all class 1 molecules, to HLA-A2 and also to beta-2-microglobulin. This was a significant step towards standardising reagents for use in expression studies. A standard scoring system for semi-quantitating the degree of class 1 expression on tumour cells was also devised, and used by all laboratories when submitting their results. While tumours totally lacking in class 1 expression or those with apparently normal expression are easily recognised, the scoring system was designed to classify those tumours with heterogeneous patterns of expression.

One of the controversial issues to emerge during the 13^th workshop was the use of formalin fixed tissue over frozen tissue. While formalin fixed tissue is more readily available and permits large retrospective studies, many monoclonal reagents do not react satisfactorily with fixed tissue, and hence there is a restriction on the number of antibodies that can be used. There is also a question concerning the quality of DNA and RNA that can be prepared from fixed tissue, although techniques have been published which claim high success rates. These issues need to be addressed during the next phase of collaborative work.

Another issue is the availability of reagents that are HLA allele specific. Since a significant number of tumours have been shown to demonstrate specific allele dropouts, the lack of specific reagents for this purpose is still a significant deficiency in our testing armory. This again is largely a reflection of the fact that fixed tissue has been used for many of these studies. It is almost certain that the number of cases demonstrating single allele dropouts has been grossly underestimated in previous studies for this reason. A concerted effort needs to be made to establish a bank of allele specific monoclonal antibodies for this purpose, particularly for those that are known to be restriction elements for defined peptides used in vaccinations, which of course will vary depending on the tumour being studied. Again the use of fresh tissue will enhance the number of reagents available for use.

With the foundation established at the 13^thWorkshop we would like to take this component into the next workshop phase with an integrated and planned approach to studying expression, mechanisms underlying changes in expression, andclinical relevance particularly with respect to tumour spread and response to peptide therapy.

Expression

The reading of stained sections for Class 1 expression requires standardisation across participating laboratories. To this end two quality control exercises will be established as part of ongoing collaborative studies. The first will involve the the exchange ofboth stained paraffin and frozen section slides on a range of tumours and scoring between laboratories compared. The sections will also be assessed by an independent pathologist in order to ascertain that only tumour cells are being assessed with respect to percentage of cells staining. The second exercise will involve the exchange of a limited number of paraffin and frozen sections for staining with workshop monoclonal reagents in the individual laboratories. A standard immunocytochemistry technique will be used, and the results compared between laboratories.

There will be opportunities for participating laboratories to compare alternative methods of measuring expression. Alternative methods include isoelectric focussing which was used to a limited extent in the current workshop, and infra-red spectroscopy (IRS) which can be used to measure molecular changes which occur when antibodies bind to a cell surface. The advantage of IRS is that it can be used in a quantitative way to measure accurately the percentage of tumour cells that are expressing the relevant molecule. This method therefore has the potential to eliminate subjective assessment, which is an inherent feature of the immunocytochemistry technique. It is essential however in the context of the continuing collaborative work that alternative methods of expression are conducted in parallel with immunocytochemistry in order to fully assess the validity and usefulness of alternative methods.

Mechanisms Underlying Alterations in HLA Expression

All tumours studied for HLA class 1 expression should be investigated for molecular mechanisms underlying any observed changes using standard methodology. DNA used for this purpose will be either from fresh frozen tumour tissue isolated by microdissection and subsequent DNA extraction, or from tumour cell lines. The use of fixed tissue for molecular studies can be investigated in a comparative study.

Methods used should include:
-   Loss of heterozygosity using microsatellite markers
-   Sequencing of individual genes for mutations including HLA A,B,C,tapasin, Tap, calnexin, calreticulin, and beta-2-microglobulin where expression studies indicate these may be involved.
-   Study of DNA methylation in tumour cell lines using 5-aza-2’deoxycytidine as a de-methylating agent.
The role of NK cell killing of HLA class 1 negative tumour cells and their effectiveness in combating tumour growth is not well understood. Collaboration with the KIR gene group in defining KIR alleles and their corresponding receptors in individuals with reduced HLA class 1 expression will be a first step towards understanding the role of these genes in tumour immunity. Any study of KIR genes in this context will also require information on the expression of the non-classical class 1 genes, HLA E and HLA-G.

Tumours to be studied

Future studies will need to restrict the number of tumours studied to a small workable group (5-6). This will encourage the collection of large numbers of patients for each tumour rather than small numbers of patients in a large number of tumour types, enabling a more comprehensive analysis particularly in relation to clinical course. Working groups of interested researchers will be formed for each tumour type to organise testing and clinical data collection.

Clinical Data

An extensive clinical questionnaire was prepared for the 13thWorkshop that will be maintained for future studies but will be expanded to include detailed information on immunotherapy. It is critically important for the science to progress that as many patients as possible who are undergoing immunotherapy approaches, which rely on class 1 expression of the target tumour cell, be included for study. The solid groundwork has been laid but the next step is to take that knowledge we have gathered to the clinic. An important step in that process is to demonstrate the usefulness of measuring Class 1 expression in predicting an individual patient's response to immunotherapeutic approaches.

Project Co-ordinator:

Brian Tait
Victorian Transplantation and Immunogenetic Service
The Royal Melbourne Hospital, Parkville, Victoria, AUSTRALIA 3050
Email: bdtait@arcbs.redcross.org.au

BACK