January 9 2004
by Chaya Venkat
Genitope Corporation May Try Vaccine in CLL
We have been invited to attend the Genitope Advocacy Summit meeting on January 22-23. Until now, Genitope's efforts have been focused on developing a patient-specific vaccine for Non-Hodgkin's' lymphoma, follicular lymphoma to be precise. It looks like they are getting ready to initiate a clinical trial for CLL patients. Better late than never. P.C. and I look forward to attending this meeting and will report back to you our impressions. In the meanwhile, I thought I would give you a heads-up on what this technology is all about, get your juices flowing as it were. The first section describes how idiotype vaccines work. The second section has some of the details of Genitope's NHL clinical trials, and the last section has a list of questions that I will try to clarify at our meeting.
Idiotype Vaccines and How They Work
Any of you guys out there rich enough to afford a fully custom made suit or a pair of custom made shoes? (And if you are that rich, where is your generous donation for Project Alpha?) Neither P. C. nor I fall into that category, but I have heard it said that off-the-rack mass-produced products are not even in the same ball park. Well, idiotype vaccines are custom built from the molecule up, each batch of vaccine built to be just right for that special patient and no other.
This is how it works. Cancers such as CLL (and NHL) are called "clonal" because each cancer cell that has accumulated in your body grew out of that first cancer cell that survived and had babies. That first cancer cell is the great grandmamma of each and every CLL cell that came after. The idiosyncrasies of that first malignant cell are reflected in all its daughters, generation after generation. Specifically, we want to look at the immunoglobins ("Ig") expressed by the CLL cells. These protein markers are specific for a given patient. (Don't make the mistake I made when I first started on my learning curve: the word "idiotype" does NOT come from the word "idiot"! The word just means specific to a given patient.) The idea is to get a sample of your particular CLL cells, examine the cells to identify the Ig protein marker that is common to all of your CLL cells, and then through a very complex process grow large quantities of that idiotype protein.
This is not all that different from the approach taken in vaccinating people against diseases like polio, for example. A small snippet of the actual polio virus is presented to the body, by means of an injection. This little snippet of the virus is not dangerous in itself, but it sets off alarm bells through out the body. Dendritic cells and other antigen presenting cells carry the offending piece of protein to every corner of the body, make sure all the fighting troops of the immune system are aware of it, recognize it, and remember what it looks like. The analogy that comes to my mind is one of putting up a "wanted" poster in all police stations and post offices. It helps if the mug shot looks a lot like the criminal, gets wide publicity, and each cop on the beat is encouraged to remember the facial features. If this is done right, next time when the ugly dude shows up in town, the cops recognize him right away and are ready to neutralize him.
Cancer vaccines face some obvious problems. The cancer grew in the first place because the snippets of protein on the surface of the cancer cell are not sufficiently ugly to get the immune system riled up, the face of the criminal is just too ordinary. If they were as ugly (the correct word is "immunogenic", meaning getting an immune response) as the polio virus, for example, the cancer cells would never have a chance to grow in the first place, they would be attacked and killed before they became a problem and you would not be wasting your time reading this article. Something has to be done to make the immune system sit up and take notice, the equivalent of blackening the teeth of the guy in the "wanted" mug shot, sketch in an ugly scar across the cheek, etc. You get my drift.
The other piece of the puzzle is that cancer patients often do not have a sufficient number of effective antigen presenting cells, the cells that carry the mug shot of the wanted man to each police station and post office. Without getting the word out to the troops, the wanted man can be ugly as sin, but he is going to slip through the gaping holes in the security net. Then there is the question of memory, how long the cops on the beat remember the face of the wanted man. That too depends on the effectiveness of the antigen presenting cells, to some degree.
Last but not least, once the ugly face of the cancer cell is given wide publicity, and easily recognized by cops of the immune system, it boils down to the number of cops available to make the arrest, are there enough law enforcement officials to take care of each and every wanted criminal?
Bottom line, most vaccine approaches need to address these four issues:
Compared to Follicular Lymphoma (this was the target of many of the vaccine efforts in NHL), CLL is a tougher nut to crack. For starters, CLL cells have very few of the immunoglobin molecules on their surface, whereas FL cells are literally festooned with them. Manufacture of the idiotype vaccine for CLL is that much harder, since the researchers have to search for the few Ig sticking out on the surface of CLL cells. Making the idiotype vaccines is a long and complex process. Different companies use different approaches to making the vaccine. But it is now generally accepted that vaccine approaches in NHL work better when (1) the snippet of the cancer is presented along with KLH (keyhole limpet hemocyanin) something that makes the cancer snippet look so horrible and weird that the immune system has no option but to take notice (2) the vaccine is administered along with GM-CSF (granulocyte macrophage colony stimulating factor) to increase the number of effective fighting troops (3) patients go into the vaccination after they have been through therapy of one sort or another, so that the bulk of the cancer is already gotten rid of. This is important to avoid the outnumbered armies point we made above.
It is fortunate that CLL patients do not progress as rapidly as most other cancers. A vaccine manufacturing procedure that takes several months does not sound so good for cancer patients with aggressive disease and overall survival measured in a few months, the patient is likely to be dead before the medicine is ready for him. Not so for us lucky folks to get "the good kind of cancer". But the very indolent nature of CLL can be a double edged sword, it is that much harder to prove effectiveness of the vaccine. Would the patient have lived just as long with or without the vaccine? It costs a lot of money to do rigorous statistics to prove these things, especially if a large number of patients have to be followed for a long time to make the call.
A number of idiotype vaccine clinical trials are currently underway, including the pivotal Phase III trial, conducted at the NCI. Use of active immune therapy among the treatment options for lymphoma and indolent leukemia may well depend on the outcome of this as well as the Genitope-sponsored Phase III study being conducted at Stanford University and 33 additional sites in the U.S. and Canada. These trials are conceptually similar the differences are in the manner in which the vaccine is manufactured. The original NCI trial uses hybridoma technique-based vaccine production, which is both expensive and not always successful in manufacturing vaccine for each patient. Some of the later trials, including Genitope's Phase III study, use molecular biology means for vaccine production, using a technique called "PCR" or polymerase chain reaction. This is considered an improvement since the costs come down and there is a better chance that vaccines can be made for each patient. Genitope's 2002 ASCO presentation claims that there have been no production failures in tumor specimens that express surface immunoglobulin.
Here is the reason for all the excitement. The vaccine trial at Stanford began way back in 1988 and it recruited 41 patients. They were a mixed bunch, some had several rounds of prior chemotherapy, some were chemo naive. After tumor de-bulking, all 41 patients were vaccinated with their custom-made vaccines. 20 out of this group of 41 patients demonstrated a vaccine response (you know what that is, if you have ever been vaccinated for other diseases, here is where it helps if you actually feel some discomfort, pain or swelling at the site of the injection suggesting that your body is actually responding to the vaccination). The time to disease progression and overall survival are vastly different. While the median time to progression was less than 16 months for patients that did not respond to the vaccine, for the patients that did have an immune response to the vaccine the time to disease progressions was a whopping 95 months! Median survival for the non-responders was 7 years. For the responders, median survival has not yet been reached!
These statistics of the Stanford study make the therapeutic benefits of their vaccine quite clear, but it is interesting to note that even some of the responders to the vaccine did have eventual disease progression. With any cancer therapy, there is always the question of the cancer cells learning to evade the immune surveillance, of finding ways to circumvent the security net. Sequential analysis of the cancer cells used to make the vaccine and the samples after patients relapsed showed that there was Ig remained genetically similar. Not identical in all ways, but similar. There were "point mutations", which may have been due to natural progression of the disease, or a mechanism that the cancer has discovered to evade surveillance. To carry our "wanted" analogy, it is possible that as time passes the criminal ages and his face gradually changes and acquires new wrinkles, he may loose his hair and need to use reading glasses etc, making the mug shot not quite as accurate and the identification a little more difficult. A more sinister explanation for a failure to identify the criminal would be that the bad guy figured out a way to do cosmetic surgery on himself and was therefore better able to disguise himself from law enforcement. So far, the consensus seems to be that the idiotype has not changed over time for this group of patients, with their particular disease.
Genitope's Phase-III clinical trials for NHL
You can learn about the technology in general by going to http://www.genitope.com/pb.pdf .
Below is some information about their Phase -III clinical trial for NHL. You can get a lot more background at their website by clicking on http://www.genitope.com/tr2003.html .
Phase 3 trial of a specific recombinant idiotype conjugated to KLH with GM-CSF compared with nonspecific immunotherapy, KLH with GM-CSF, in patients with follicular NHL (Protocol: G2000-03).
This is a multi-center, randomized and double-blind Phase-III study. In case you are not familiar with these terms, patients will be assigned randomly to either the experimental or control group, twice as many patients funneled into the experimental group. Neither the patients nor the researchers conducting the study will know if a given patient is in the experimental or control group. The logic of this approach is to prevent results getting biased in favor of the experimental therapy. This is a large scale study. I believe they plan to enroll as many as 700 patients (thanks, Donna), at different centers across the country. Large scale trials of this type are almost always needed before the drug can become commercial. Obviously, one of the downsides for patients participating in such late stage trials is that they do not control whether or not they will actually get the vaccine therapy. In this trial there is a 33% chance they will be in the control group. On the plus side, enough work has been done in the earlier stages of the trials that researchers have probably worked out the details of dosage, toxicity and method of administration and based on the earlier results you have a better chance of evaluating risk versus reward in participating in the trial.
In keeping with many vaccine approaches, the current trial too focuses on the effectiveness of the vaccine on patients with very low tumor load. In order to make sure this is indeed the case, untreated patients with follicular NHL (stage III or IV) will first undergo chemotherapy with eight cycles of CVP (cyclophosphamide, vincristine and prednisone), and those who respond with sufficient tumor load reduction will then receive a series of immunizations with their personalized idiotype vaccine, combined with KLH and GM-CSF. Patients in the control group will get KLH with GM-CSF, but not the idiotype vaccine.
Some questions to resolve:
Based on their approach thus far, I have some questions that I would like to address to them. Please write and let me know if any of you have questions that you want asked; after all P.C. and I will be attending as your representatives.
As you can see, it is an exciting development, and I have a laundry list of questions. Stay tuned, we are finally getting started on serious idiotype vaccine trials for CLL, and I for one am pleased to see this development.
You can contribute to support our efforts.