Patricia F. Fitzpatrick Dimond Ph.D. Technical Editor of Clinical OMICs President of BioInsight Communications
New results presented at ASH will impact production of autologous immunotherapies.
New findings presented at this year’s American Society of Hematology meeting support the use of active immunotherapies in blood cancers. Research is revealing new information about the nature of immunoglobulin cancer antigens and how the immune system reacts to cancer.
The data offers clues for future vaccine design and development, including which patients are most likely to respond. The big story here is that all idiotype vaccines may require manufacturing using an idiotype grafted onto an IgM isotype and not an IgG isotype.
Biovest’s BiovaxID is pegged to be the next therapeutic cancer vaccine to gain approval. Like the only marketed cancer vaccine, Dendreon’s Provenge, it is an autologous treatment. Provenge is indicated for asymptomatic, hormone-resistant prostate cancer, while BiovaxID is being evaluated for the blood cancer non-Hodgkin B-cell lymphoma. Biovest’s lead program is in follicular lymphoma (FL).
The latest clinical data from Biovest came earlier this year at median follow-up of 56.6 months. Stephen Schuster, M.D., lead investigator of the study, of the University of Pennsylvania’s Abramson Cancer Center, said that median time to relapse, or disease-free survival (DFS), after randomization was 44.2 months for those in the BiovaxID arm versus 30.6 months for the control arm. The investigators concluded that vaccination after a chemotherapy-induced remission of ≥6 months prolongs remission duration in patients with FL.
The demonstrated efficacy of BiovaxID begs the question of why two other Phase III trials of nearly identical candidates—Favrille’s Specifid and Genitope’s MyVax—failed to reach their primary endpoints.
Comparing BiovaxID, Specifid, and MyVax
BiovaxID consists of an individual patient’s tumor’s complete idiotype protein, an immunoglobulin expressed on the surfaces of B-cell lymphoma cells. Individually manufactured from a tissue biopsy obtained from the patient’s tumor, the vaccine selectively targets only cancerous B cells. The final vaccine is administered as a subcutaneous injection along with granulocyte-macrophage colony-stimulating factor (GM-CSF) and keyhole limpet hemocyanin (KLH), which together enhance the potency of the immune response induced by BiovaxID.
Like BiovaxID, both were anti-idiotype vaccines conjugated to KLH and delivered with GM-CSF. Favrille reported in May 2008 that its Phase III trial testing Specifid failed to show a stastistically significant improvement in time to disease progression. That same year Genitope announced that the primary analysis of Phase III MyVax data showed no statistically significant difference in progression-free survival between those receiving MyVax and patients receiving the control substance.
Genitope did, however, say that analysis of a prespecified endpoint showed a statistically significant difference in PFS between patients who mounted a positive immune response to the tumor-specific idiotype and those who did not.
Some believe BiovaxID’s success may be linked to the fact that in its Phase III study, all vaccinated patients were in a state of complete remission. The other candidates’ protocols, in contrast, allowed patients in partial remission to remain in the study, a factor that may have hindered efficacy.
Finding that Immunoglobulin Isotypes Affect Efficacy
Dr. Schuster also pointed out that the real problem may relate to the way the recombinant vector was designed. “Both Favrille’s Specifid, expressed in an insect cell system, and Genitope’s MyVax, produced in CHO cells, were produced as recombinant proteins,” he said. “There is the potential that with recombinants, the three-dimensional conformation of the idiotype protein won’t be identical to the protein produced by the patient’s tumor.”
In the case of Genitope’s vaccine, the variable region of the patient’s idiotype was cloned into an IgG heavy-chain immunoglobulin. “Not all patients with FL produce IgG isotypes,” he explained.
“One possibility has been that the antibody isotype affects the body’s ability to mount an anti-idiotype immune response,” Dr. Schuster continued. “For example, IgG-bearing idiotype may be less immunogenic than IgM bearing idiotype.”
He further noted that data from the BiovaxID trial helped address this issue. “Because our vaccine was derived from the patient’s tumor, we had the idiotype and the isotype specific to the tumor. The recombinant studies all used IgG, one fundamental difference, while basically, in our trial, we used whatever a patient had.”
Dr. Schuster explained that in the BiovaxID trial, immunological and biochemical characterization showed that 55 patients’ tumors produced IgG isotype and 60 produced IgM isotype. “Among the IgG idiotypes, 40 got the IgG vaccine, and among IgM, 35 got the IgM vaccine. We could therefore look at disease-free survival in isotype-matched controls.
“Patients that got an IgM vaccine when compared to IgM controls showed a highly different disease-free survival, and basically, everybody in the study got an isotype-matched vaccine,” he added.
The investigators reported that in the IgM isotype group, they observed that patients who were treated with isotype-matched BiovaxID had significantly longer DFS: 52.9 months versus 28.7 months for patients who received the control vaccine. In contrast, in the IgG isotype group, there was no difference in median DFS between the patients who received isotype-matched BiovaxID and those with IgG isotype tumors who received control vaccine.
Understanding IgM and IgG
Dr. Schuster says that his colleagues at The University of Texas M.D. Anderson Center will run assays on the plasma from trial patients to see whether there are higher levels of immune responses to IgM. “This is one potential explanation for why this study worked,” he said. His research team concluded that vaccine isotype may affect clinical outcome and explain differing results between the BiovaxID trial and other vaccine trials.
The investigators said that the reason why IgM-Id but not IgG-Id was effective is unclear. Mouse and in vitro studies have shown that Ids that have switched to IgG are tolerogenic whereas Ids of their IgM progenitors were highly immunogenic. Further, the scientists noted, the Fc of IgG has MHC Class II T-cell epitopes that activate regulatory T cells and tip the immune response toward tolerance rather than immunogenicity.
GEN asked Dr. Schuster what implications this information had for the production of autologous idiotype vaccines. “We saw no benefit to the vaccine in patients who had IgG tumors and got IgG vaccines. All of the benefit in the study was among IgM patients who got IgM vaccines.
“My thought is not to not offer a vaccine to patients with IgG tumors, but to offer IgM vaccines to everyone.” This could be accomplished by cloning a patient’s tumor idiotype into an IgM expression vector.
Biovest commented, “This startling finding establishes that careful initial selection of heavy-chain isotype in idiotype vaccine manufacture may ultimately be a key predictor of clinical benefit.” This data needs to be confirmed by further characterization of immune responses among vaccine responders. If this pans out to be true, development of isotype cancer vaccines will need to focus on the potential impact of immunoglobulin isotypes on vaccine efficacy.
Patricia F. Dimond, Ph.D. ([email protected]), is a principal at BioInsight Consulting.