Dr. Kwak's Reply
January 12, 2000
Karl L. Schwartz
1305 Mason Ave.
Statton Island, NY 10306
Dear Mr. Schwartz:
I want to thank you for your thoughtful letter of December 10, 1999, regarding our pivotal, phase III clinical trial of lymphoma vaccination. You have carefully considered a number of issues. However, I am concerned about several inaccuracies in the interpretation of available clinical data and perceptions which may not be entirely valid. Accordingly, I appreciate the opportunity to clarify several important points.
I would like to begin by making two global statements, which will provide the proper context for the detailed questions you and others may have. First, cancer vaccine development, in general, must be focused on answering two, independent questions in the proper sequence. The first question is a scientific one, whether one can even immunize against an inherently weak, self-tumor antigen. This is a critical question, as clearly the patient’s immune system has failed already to recognize antigens presented by the tumor cell and to eliminate them. Answering this scientific question is the goal of most Phase I and II cancer vaccine clinical trials. In the case of the lymphoma idiotype vaccine we have been developing, it has taken some ten years to answer this question in a definitive manner, first by the demonstration of specific antibody responses in vaccinated patients (NEJM 327:1209, 1992; Blood 89:3129, 1997) and now more recently, by the demonstration of specific T-cell responses in vaccinated patients (Nature Medicine 5:1171, 1999). The second, medical question, then, is whether the vaccine treatment can produce meaningful clinical benefit. The demonstration of molecular remissions perhaps gives us a glimpse into the answer to this question, but it should in no way be interpreted as the definitive answer. The only way to truly address this question of meaningful clinical benefit is with a randomized, controlled clinical trial. I should point out that whenever we have published on our results at the NCI, we have intentionally refrained from publishing survival curves, as a precaution against overinterpretation of the data and the risk that some would draw premature conclusions about meaningful clinical benefit. Your letter refers to the rights of patients to have access to a proven vaccine. However, until question #2 has been answered in a controlled, randomized clinical trial, the vaccines being tested must be regarded as investigational, similar to other experimental agents which at this time look promising, but are not yet proven or licensed for general clinical use.
Secondly, my clinical colleagues and I are confident that our commitment to serve the best medical interests of the patients we evaluate and treat is not compromised by our mandate to perform clinical trials. The field of oncology has a long history of promising treatments that have been accepted as effective based on preliminary results, only to be proven otherwise after further analysis (stem cell transplantation in breast cancer is a recent example). Unfortunate consequences are that patients may be given false hope, and the development of other, potentially beneficial treatments are impeded. The fact is that progress in medicine comes slowly and most often occurs in incremental steps. Therefore, the only way to make real progress in the long run is to invest in controlled clinical trials that are scientifically rigorously designed and carried out. In addition, this may require clinical research which is performed independent of outside influences, oftentimes introduced by industrial support. Accordingly, please understand that we have carefully focused the Phase III clinical trial on a small subset of non-Hodgkin’s lymphoma types (follicular small cleaved cell and follicular mixed) and on a subset of previously untreated patients who are in complete remission when they receive the vaccine.
Allow me now to respond to several of your detailed questions:
Example 1 – A central hypothesis, based in part on the lack of effectiveness of idiotype vaccines in experimental animals which had previously established tumor masses, is that the patients whom the vaccine is most likely to benefit are those who are in a true minimal disease state, defined in the current protocol as complete clinical remission (CR). Furthermore, we believe that the past interpretation of clinical anti-tumor responses, measured by shrinkage of small, residual tumor masses in partial remission (PR) patients in a disease such as lymphoma, which already naturally waxes and wanes, is inherently problematic. Thus, at the current time, we feel that patients who do not achieve a CR from chemotherapy are best served by offering them treatment on protocols with other, non-vaccine experimental agents.
Example 2 – The current NCI Phase III clinical trial is unique in its design in at least two aspects. First, while conventional randomized, Phase III trials administer only the best available standard therapy to the control group, the control group in this trial will not only be receiving the best available standard therapy for follicular lymphoma (chemotherapy alone), they will also be receiving two additional investigational agents, each of which alone has substantial potential for anti-tumor effects. Specificially, keyhole limpet hemocyanin (KLH) is a non-specific immune enhancer which has shown activity against an early type of bladder cancer. Similarly, Phase II clinical trials are currently being planned administering GM-CSF alone (without any vaccine) as treatment for cancers of the skin (melanoma), as well as other solid tumor types. Indeed, we feel strongly that the recent addition of GM-CSF to our vaccine formulation enabled the definitive answer to question #1 above. Thus, the success in inducing T-cell responses distinguishes the Id-KLH vaccine formulation containing GM-CSF from those tested previously with other adjuvants.
Secondly, two-thirds of the patients will be randomized to receive the vaccine treatment, instead of 50%, which is customary. This modification was made purely for the purpose of making enrollment on this trial more attractive to patients, who would otherwise decline enrollment because of an inaccurate perception that they could be randomized to treatment with a placebo. It does not in any way reflect any subjective bias the investigators have regarding which treatment arm will prove to be superior.
I should also point out that based on growing interest in the scientific community in the development of vaccines against lymphomas, we expect additional Phase I or II single arm clinical trials (in which all patients receive the experimental vaccine), seeking to reproduce the results with the vaccine formulations reported (e.g., Id-KLH + GM-CSF), to be initiated in the near future by other academic centers in the U.S.
Example 3 – In contrast with other forms of immunotherapy (e.g., monoclonal antibodies), cancer vaccines require an intact immune system in the host. The reason for this requirement is that vaccines work indirectly, by activating host immune T cells and B cells to do the actual job of potentially neutralizing tumor cells. Thus, it follows that the patients who are most likely to benefit from this type of immunotherapy are those whose immune systems have not been previously subjected to multiple rounds of potentially immunosuppressive chemotherapy treatments. Successive courses of chemotherapy have a cumulative negative effect on the immune system, particularly for T cells. Most other available therapies against lymphoma would not be expected to have this limitation.
Example 4 – Based in large part on observations in experimental animals which showed clearly the lack of effectiveness of idiotype vaccines against measurable tumor burdens, again, a central hypothesis has been that the patients most likely to benefit from vaccine therapy are those whose initial tumor burdens (at diagnosis) have been reduced to minimal disease. Even in a chemotherapy-induced CR (absence of clinically apparent tumor), we know that there are microscopic tumor cells still present and that the disease has therefore not been completely eradicated. Achieving even this state of minimal disease thus usually requires some form of conventional therapy before the vaccine is administered. The potential drawback of immunosuppression from chemotherapy is outweighed by this benefit. However, theoretically, if radiation therapy, surgery, or monoclonal antibody therapy could reliably produce the same proportion of CR’s, they could be used as alternatives instead.
In conclusion, I want to thank you for this opportunity to clarify the important issues surrounding cancer vaccine development, in general, and the current Phase III clinical trial in particular. I appreciate your efforts and those of groups like yours to disseminate this important information to eligible patients, who need to be informed of this potential opportunity. In addition, you should be greatly encouraged by the enormous research effort directed at improving upon current treatment results for all types of lymphomas. There are many new experimental agents (e.g., novel chemotherapeutic agents, monoclonal antibodies and their derivative toxins, radioimmunotherapy, antisense therapy). being tested in various phases of clinical development, many of which are being tested in various lymphoma types and in previously treated patients. Until the clinical activity of any one of these has been definitively demonstrated in a pivotal, Phase III clinical trial, they must all be regarded equally as promising, yet experimental and unproven.
Finally, together, through the combined efforts of individual investigators, systematically testing various promising agents in high quality clinical trials, with the support of patients, their families, and their primary care oncologists, I am confident we will continue to make the progress which is so sorely needed.
Sincerely,
Larry W. Kwak, M.D., Ph.D.
Principal Investigator