VH mutations

I well remember a meeting in Italy when VH gene mutations were the hot topic. I was in the camp that believed that CLL was frozen in time as a pre-germinal center tumor, with unmutated VH genes. I guess this was based on no stronger evidence than that CLL cells had IgD molecules on their surface. IgD is an unusual form of immunoglobulin, almost only found as a surface molecule and usually at an early stage of differentiation. Several centers had been sequencing the DNA of the immunoglobulin chains in small numbers of patients. Probably the greatest number had been sequenced by Tom Kipps from San Diego, and as an admirer of Tom, I was plaesed to see that his findings fitted with my prejudices.

Guillaume Dighiero from Paris and Harry Schroeder from the US had assembled all the published case reports of CLLs with VH gene sequences. They found 76 and much to my surprise half of them had mutated VH genes. I scrutinised these cases carefully and was very dubious. Several of the mutated cases had been reported by Nick Chiorazzi from North Shore, Long Island, and had surface IgG rather than IgM+D. This is unusual in CLL and since class switching from M to G is a germinal center task it was natural that such cases should be mutated, another germinal center task. Other cases were clearly not CLL, being CD5 negative and still others had serum paraproteins, an unusual finding in CLL.

The way we got into VH gene sequencing was like this. Martin Glennie had been raising anti-idiotype antibodies against lymphomas as part of our lymphoma therapy protocol. One of these antibodies, 9G4, proved of no value as a therapy because it cross reacted with other antibodies. The patient happened to make a cold agglutinin and Freda Stevenson had the idea that 9G4 might cross-react with all cold aglutinins.

Cold agglutinins had been an interest of mine since I was a juior doctor. I should expplain that cold agglutinins are IgM antibodies that are present in everybody's blood at low concentrations, that bind to red blood cells in the cold and cause them to stick together. In certain circumstances high levels of cold agglutinins are found in the blood - after infection with mycoplasma pneumoniae or EB virus, and in some elderly people where the antibody is monoclonal, and in some lymphomas where it is made by the tumor.

So we set about collecting all the cold agglutinin specimens we could find. We produced cell lines that secreted cold agglutinins. It turned out to be true. All cold agglutinins reacted with 9G4. We then sent the cell lines to America to have the immunoglobulin genes sequenced. This demonstrated that all cold agglutinins used the same VH gene. It was then known as V4-21 but has now been renamed as V4-34. It soon became clear that we could not continue this research with someone doing the sequencing for us so we had to learn to do it ourselves. Doing it by hand is a tedious time-consuming business so I was able to raise some money to buy a machine to automate it.

Freda Stevenson and David Oscier looked at 20 CLLs and sure enough found that both mutated and unmutated V genes were present. The striking finding was that the mutated cases had the del 13q14 chromosomal abnormality while the unmutated one had trisomy 12. This made us suspicious that the more benign cases were mutated, and to test this I collected another 64 cases. Now that we had large numbers, there was no doubt about it; the mutated cases had an average survival of 25 years and the unmutated cases only 8 years. I presented these data in 1997 and 1998 at the British Society for Haematology, the IWCLL and the ASH meetings. Nick Chiorazzi heard the presentation at the IWCLL. He had sequences on 64 cases and correlated their survival with the VH gene status and found exactly the same as we had. He also found that CD38 expression gave very similar results. We decided to publish the papers together as being mutually confirmatory.

Following this paper the value of VH gene mutational status has been confirmed several times by the French, Germans, and Swedes in particular.