Dana-Farber Cancer Institute's Dr. David Reardon reports on data showing a phase 1 trial looking at a personalized vaccine plus pembrolizumab sparked tumor-targeting T cells in glioblastoma and was linked to longer survival in immune responders.
Globoblastoma remains one of the most challenging cancers in oncology today. It is the most common that affects adult patients and unfortunately remains incurable and essentially uniformly fatal. It also is a cancer that has proven to be particularly refractory to immunotherapy treatments. We had previously studied our personalized peptide vaccine approach developed at the Dana-Farber by my colleague, Catherine Wu. Um, called Neovax in patients, in an initial study for patients with glioblastoma where we demonstrated that this vaccine approach can stimulate systemic immune responses specific to targets unique to each patient, individual tumor. Um, however, these immune responses did wane over time and, uh, tumor associated T cells from these patients ultimately, Became exhausted and lost their responsiveness. We designed the current trial to try to overcome that limitation, and what we did with this trial for newly diagnosed glioblastoma patients was to combine our neo antigen vaccine called Neovax with pembrolizumab, the PD1 inhibitor, um, uh, treatment. And we evaluated 4 cohorts of patients on this trial for newly diagnosed patients. For the first time, we combined this vaccine with standard temozolamide chemotherapy for patients with an MGMT methylated subclass of tumor. These are patients who have been demonstrated to have a likelihood of benefit from chemotherapy. And we also included 3 cohorts with the MGMT unmethylated subtype, and this subtype are patients who are known to have very little to no benefit from chemotherapy. And here we substituted the vaccine and pembrolizumab for chemotherapy in these patients. And what we did that was unique and important from a scientific point of view is that we varied the timing. Of the initiation of the vaccine and the initiation of the PD1 inhibitor pembrolizumab, to see if the timing mattered. If we, one cohort we gave the vaccine prior to initiated initiation of PD1 blockade with pembrolizumab, a second cohort, we did the opposite where we started the pembrolizumab prior to vaccine priming, and then a third cohort, we, we did a bit of a hybrid where we initially gave a single dose and then continued it after vaccine priming. And our results demonstrated that um across all cohorts of patients, um, significant responses to the vaccinated peptides were achieved for all patients, including the cohort who received immunotherapy, where we were concerned immunotherapy could abrogate the responsiveness to immunotherapy treatment. And interestingly, in this subset of patients, we saw a significant improvement in their overall survival relative to control, historical control patients treated at the Dana-Farber, who received standard of care, chemotherapy and radiation therapy. Amongst the patients who, we varied the timing in the 3 cohorts of the two therapies together, where no chemotherapy was administered. We saw all of those cohorts generate good immune responses, but interestingly, the patients who received uh the uh pembrolizumab prior to or during the initiation of vaccine treatment seemed to have the best survival and best outcome. Amongst all patients, we saw good responses that were maintained. Immunologically over time, up to 11 to 2 years of follow-up treatment, those patients who made good responses were able to maintain them over time, and there seemed to be a correlation that the more responsiveness, the more immune responses we measured, the better survival and outcome. Finally, we use state of the art, um, detailed analyses of, uh, the, uh, immune reactive T cells from patients. And identified genes as well as gene enrichment um uh programs that were associated with improved outcome, as a strategy to try to identify which patients are more likely to benefit from this approach and which may not. Finally, uh, we looked at the um gene expression subtype of glioblastoma associated with these patients, and interestingly, we saw the subtype that predicts, A poorer response to standard radiation and chemotherapy, to paradoxically have the best outcome, uh associated with our neoantigen vaccine and uh pembrolizumab immunotherapy. So the key takeaways from our ASCO 2026 presentation are that a personalized cancer vaccine can show promise. Uh, Further promise in combination with immune checkpoint blockade for glioblastoma, uh, patients. Um, we need to do more detailed analysis, but our, um, preliminary results identify factors associated with each patient's tumor that may predict a higher likelihood of therapeutic benefit. We also observed that chemotherapy did not, um, uh, lessen the likelihood of uh therapeutic benefit for patients. And we have um ongoing initial ongoing further studies to try to improve the outcome uh with subsequent combination approaches, uh, utilizing this type of strategy as part of the, um, the next phase of, of studies for these patients.
