Idiotype vaccines - an update

The idea that the B cell receptor might be used as a target for immunotherapy belongs to George and Freda Stevenson who published a paper in Nature in 1975. They called this target idiotype because it was the same for every tumor cell but different for every other normal B lymphocyte. The following year they made an antibody which I used to treat a patient with CLL. Subsequently Ron Levy at Stanford made monoclonal antibodies against idiotype and successfully treated patients with non-Hodgkin's lymphoma with them. In 1981 Ron Levy and George Stevenson were jointly awarded the first Armand Hammer Prize for Cancer Research for this work.

It soon became apparent that the labor involved in making made to measure anti-idiotypic monoclonal antibodies for lymphoma patients was so great, that it was never seriously feasible - even in America, and certainly not in the NHS. The focus turned from monoclonal antibodies to idiotype vaccines. Work in mice showed that it was possible to produce immune responses against idiotype that were therapeutically active. Ron Levy managed to produce vaccines from the patients' tumors and carried out a trial in lymphoma patients. In a paper published in 1997 in Blood he showed that if an antibody response was generated by a patient against the idiotype vaccine, then such patients lived for significantly longer than patients who were unable to generate an antibody response. This looked like good news, but an alternative explanation, other than that the vaccination was doing the patients good, was that this was a way of separating good risk patients from bad risk patients - if you were capable of mounting an immune response you were in a better risk bucket than if you couldn't

Nevertheless, this result was so encouraging that no fewer than three randomized clinical trials were undertaken to see if vaccinated patients lived for longer than non-vaccinated patients. AT least two of these trials have now reported (though not yet published). The bottom line is that it is certainly true that if you make antibody against idiotype you live longer than those who can't make antibody, but overall there was no difference in survival between the two arms.

Last week at the Annual LRF lecture, I heard Ron Levy give an update on these trials. The idiotype protein is strapped to a carrier protein called Keyhole Limpet Hemacyanin (KLH) Rather than go to the expense of measuring anti-idiotype (a separate assay for each patient), you can pick out your good responders by measuring the response to KLH. Sure enough the vaccinated patients who made an antibody response to KLH did better than those who did not - so if you want to use it as a screening test choose immune response to KLH rather than immune response to idiotype.

Of course the control patients were given KLH and some made a response to it. The important question is, "Do they get an improved survival over control patients without an immune response to KLH?"

The answer is, they do not. Therefore, if you are capable of generating an immune response, then vaccination against idiotype does improve survival - at least that is the new hypothesis. Because this was a subset analysis, and the original trial was not designed in this way, the statistics are not valid. The results can only be used for hypothesis generating. A new randomized trial restricting the entrants to those capable of making an immune response is about to start.

I emphasize that this is in follicular lymphoma. Although a trial in CLL has been under way, CLL patients have a poor ability to generate an immune response and are therefore not good candidates for such a trial.

Another trial mentioned by Ron Levy concerned the use of CpG - an immune stimulant that stimulates antigen presentation via the Toll-Like Receptor 9 on dendritic cells. It appears that this process can be stimulated by a mixture of monoclonal antibodies - an agonistic anti OX40 antibody and an antagonistic anti-CTLA4 antibody.