A personalized adjuvant neoantigen peptide vaccine, PGV-001, was successfully synthesized and administered to patients across a wide range of malignancies who had a greater than 30% chance of disease recurrence.
A personalized adjuvant neoantigen peptide vaccine, PGV-001, was successfully synthesized and administered to patients across a wide range of malignancies who had a greater than 30% chance of disease recurrence, according to Thomas Urban Marron, MD, PhD. To further improve responses, the vaccine is now being combined with checkpoint blockade.1
In the trial, which was presented during the 2021 AACR Annual Meeting, 13 patients received PGV-001; this included 6 patients with head and neck cancer, 2 with non–small cell lung cancer, 1 with urothelial carcinoma, 1 with breast cancer, and 3 with multiple myeloma.1 After a mean follow-up of 925 days, 4 patients did not have any evidence of disease, 4 were receiving subsequent lines of therapy, and 4 died. Notably, 2 of the patients who died did so due to documented disease recurrence. The vaccine was also found to be well tolerated, and an initial analysis of patient samples confirmed immunogenicity.
“The neoantigen space is a very exciting space. We can use [this approach] to build on the success that we have seen in cancer immunotherapy, as far as all the checkpoint-blocking antibodies where we do see significant benefit in a subset of patients,” Marron said. “[However], we still have a lot of room to grow. Trials like this, and the OpenVac pipeline that we have developed, really allow us to better understand the optimal way in which we can vaccinate patients against their own cancer.”
In an interview with OncLive®, Marron, an assistant director of early phase and immunotherapy clinical trials at the Tisch Cancer Institute, and an assistant professor of medicine at the Icahn School of Medicine at Mount Sinai, further discussed the potential of the PGV-001 neoantigen peptide vaccine, ongoing trials examining this approach in different cancers, and the next steps for research.
Marron: Neoantigen vaccines are exciting. One of the real barriers to progress in the field of cancer immunotherapy is that most immunotherapies really rely on there being a pre-existing immune response against a patient's tumor. A patient's immune system already has to have been primed to recognize tumor antigen. If that is not the case, particularly for patients who have relatively few neoantigens, so a low mutational burden, that most likely is why we see much lower response rates in that population.
To improve that [you can] either add on agents such as ipilimumab [Yervoy] which can potentially help with the priming of the T-cell response, or you can use these personalized vaccine approaches. Many personalized vaccine approaches have been [evaluated] in patients with metastatic cancer. This trial is unique in that it is one of the few that has really looked at the use of neoantigen vaccines in the adjuvant setting, particularly in a heterogeneous mixture of cancer types.
The goal of [the treatment] is cure, but we know there is a very high likelihood that [the disease] will come back after whatever adjuvant therapy is standard, because of the presence of micrometastatic disease. [This] is really an ideal setting to test a vaccine like this. It is also an ideal setting [because] these vaccines might be able to elicit a de novo immune response. The thought is that there is an inverse relationship between tumor burden and efficacy of a variety of immunotherapies. Patients [in whom] the residual disease is really just microscopic, are [those] who may derive the most benefit from either a vaccine or immunotherapies that are being looked at.
We take patients who are undergoing curative-intent surgery—we actually had 3 patients undergoing autologous stem cell transplant who had myeloma. Then, after any standard adjuvant therapies, such as those with lung cancer who were getting platinum doublet chemotherapy, we then administer the vaccine to prime the T-cell response and keep them in remission.
We enrolled patients [with a] cutoff of at least a 30% chance of recurrence, [which] was based on tumor stage and a variety of histologic characteristics. Obviously, for multiple myeloma, you have a near 100% chance of recurrence after autologous stem cell transplant. [However], we similarly see very high rates of recurrence in those who have head neck cancer who are undergoing a second surgery, which was a good proportion of [the study population]; we had 6 patients with head and neck cancer, most of [whom] had recurrent disease and were undergoing a subsequent surgery. [We] also [enrolled] patients who had advanced lung cancer. Nearly all these patients received an adjuvant therapy after their surgery or transplant, [which was followed by] the vaccine.
All these patients had tumor available for whole-exome sequencing and bulk-RNA sequencing. We also had germline data from a blood draw. We were able to identify mutations unique to the tumor and also to confirm expression, or presumed confirmation of expression, by looking at the RNA. We used a computational pipeline called OpenVax, [which was] developed at Mount Sinai, and uses a combination of expression and neoantigen data, as well as patients’ unique human leukocyte antigen to identify the optimal neoantigens to include within the vaccine.
We are in the process of demonstrating the immunogenicity that we are successfully eliciting with the vaccine in this first trial of PGV-001. [The] same pipeline has been used to identify neoantigens for 2 additional trials that are ongoing. One of [these trials] is looking at patients with metastatic urothelial carcinoma; here, the vaccine is given in conjunction with atezolizumab [Tecentriq]. A combination of checkpoint blockade and the vaccine is probably the ideal combination. An adjuvant trial [is also being done] in glioblastoma. Here, tumor-treating fields are used at the same time that the vaccine is administered. Those are both exciting subsequent trials that are further validating the same pipeline.
Our primary objectives were really, first of all, to determine the safety and tolerability of the vaccine. [We saw that the vaccine] was very well tolerated. Patients had minimal injection[-associated] adverse effects that were typically related to the poly-ICLC, which is the viral mimic that activates the immune system. The second primary objective was to determine the feasibility of creating and administering the vaccine, and we were pretty successful.
Fifteen patients were enrolled and we synthesized the vaccine [for these participants]. We administered it to 13 [patients] because 1 patient actually had bad biology and his lung cancer progressed through adjuvant platinum doublet [chemotherapy] and recurred very quickly. The other patient was still in remission, but she opted for a clinical trial closer to home. [Of the] 13 patients who were started on the vaccine, 11 received all 10 vaccines, while the other 2 [patients] received 7 and 9 vaccines, respectively. It was a feasible approach, but obviously many of visits were involved. We achieved both primary objectives.
The third primary objective is really to determine the immunogenicity of the vaccine. We have only looked at a few patients and [we have] some data from the first patient that we were able to analyze. We [observed a] pretty robust induction of both the CD4 and CD8 T-cell response.
We have blood samples that were collected throughout [the trial] and we looked at baseline ex vivo T cells, week 8, and week 27. We see that after the first 6 vaccines, we are barely able to see any sort of T-cell response in the peripheral blood. [However], after the full 10 vaccines at the 6-month mark, we see a potent induction of T-cell immunity. Every vaccine out there is using different numbers of boosters. As far as working in vivo, we are [almost] working in the dark, but this really underscores the need to use many boosters to get the biggest bang for our buck.
The trial that is combining [the vaccine] with a checkpoint blockade is very interesting. We had 2 patients with head and neck cancer who had progressive disease after completion of the vaccine. They both went on to receive single-agent immunotherapy and they had amazing responses; that might have been the response they would have achieved anyway. One of them had actually received prior immunotherapy, interestingly enough.
It really underscored the fact that a combination of priming a T-cell response with a vaccine, and then further activating the immune system with checkpoint-blocking antibodies is a really nice combination. I look forward to seeing the results from our trial in bladder cancer, where patients are receiving atezolizumab at the same time as they are receiving these vaccines. It will be very interesting to evaluate the lymphocyte phenotype in the lab at various time points in that population vs the population we have treated, where there was not any checkpoint blockade.
By studying the immunogenicity and the schedule that we have in this trial, and other trials like it, we can learn a lot about, and potentially develop, further personalized vaccines. Personalized vaccines are obviously very attractive because you get a polyclonal response to tumor antigens, so you do not have as many escape mechanisms available to the tumor.
We can also use these same pipelines and these same study designs with shared neoantigens. Many people are looking at KRAS vaccines, given the high rates of KRAS G12C/G12D mutations in lung and pancreatic cancers, as well as EGFR vaccines. Using a relatively small number of shared neoantigens, we could provide vaccines to a pretty large [number] of patients with cancer.