NEW YORK – Results from a Phase Ib clinical trial published Thursday in the journal Cell demonstrated the safety and preliminary efficacy of combined treatment with the personalized, neoantigen-based vaccine NEO-PV-01 and the anti-PD-1 checkpoint inhibitor nivolumab (Bristol Myers Squibb's Opdivo) for patients with advanced melanoma, non-small cell lung cancer, or bladder cancer.
BioNTech subsidiary Neon Therapeutics designed NEO-PV-01 to target neoepitopes encoded by somatic mutations in patients' tumors. Each personalized iteration of the vaccine evaluated in this trial consists of up to 20 unique peptides selected using bioinformatics algorithms following whole-exome and RNA sequencing of patients' tumor samples.
In the Phase Ib study, patients received nivolumab for 12 weeks while their unique vaccine was manufactured. Patients did not have to meet a PD-L1 expression threshold in order to receive the immunotherapy. Ultimately, 60 patients received NEO-PV-01, including 27 patients with melanoma, 18 with non-small cell lung cancer, and 15 with urothelial carcinoma of the bladder, urethra, ureter, or renal pelvis. Patients continued receiving nivolumab both during and after the vaccine administration.
In this cohort, eight withdrew from the trial due to disease progression, adverse events, or other non-specified reasons, leaving 51 patients across the three cancer types having completed the full course of treatment.
The primary endpoints of the trial were to assess the safety and tolerability of the combination of NEO-PV-01 and nivolumab. The authors of the Cell study, led by Patrick Ott of the Dana-Farber Cancer Institute, noted that the combination demonstrated a favorable safety profile. Only two patients experienced serious adverse events.
Ott and colleagues also assessed secondary endpoints including objective response rates, progression-free survival, and overall survival. The objective response rates among patients with melanoma, NSCLC, and bladder cancer, respectively, were 59 percent, 29 percent, and 27 percent. The median progression-free survival for these three histology groups was 23.5, 8.5, and 5.8 months, respectively, and the one-year overall survival rates were 96 percent, 83 percent, and 67 percent. The median overall survival was not reached for patients with melanoma or NSCLC, but was 20.7 months for patients with bladder cancer.
However, the authors cautioned that because of the single-arm design of the trial they couldn't attribute patients' responses to the vaccine alone, since they couldn't rule out nivolumab's role in bolstering patients' immune responses.
That said, the investigators explored indicators of patients' responses specifically to the vaccine. In a post-hoc analysis, they measured the percentage of all NEO-PV-01 vaccinating peptides that elicited interferon-gamma responses in samples of patients' peripheral blood at three time points: pre-treatment, pre-vaccine, and post-vaccine.
Using the IFN Gamma EliSpot Assay, they found that among patients who had sufficient samples at all three time points, significant immune responses to multiple vaccinating peptides were observed, including 52 percent responses to peptides in patients with melanoma, 47 percent in patients with NSCLC, and 52 percent in patients with bladder cancer. An average of 42 percent and 24 percent of vaccinating peptides generated CD4-positive and CD8 T-cell responses, respectively.
The researchers also evaluated immune responses in patients' samples 52 weeks after the start of nivolumab treatment, and detected persistent immune responses for 58 percent of neoepitopes. "These data demonstrate that vaccination with NEO-PV-01 results in the generation of T-cell responses that are specific and durable," wrote the authors.
The researchers also demonstrated the ability of patients' CD4-positive and CD8-positive T cells to migrate to the solid tumors and kill cancer cells by looking for the expression of CD107a, a marker for T-cell degranulation. Across all three cohorts of patients, the overall surface expression of CD107a was detected in the presence of 58 percent of the epitopes after vaccination.
To home in on whether these vaccine-induced cytolytic T-cell responses were actually resulting in anti-tumor responses, the researchers selected three melanoma patients whose cancer had not progressed nine months after treatment and performed T-cell receptor sequencing in both the peripheral blood and serial tumor biopsies. On the basis of their findings through this analysis, the authors wrote, "these data demonstrate trafficking of neoantigen-specific CD4-positive and CD8-positive T cells into the tumor."
Finally, Ott and colleagues looked into whether the neoantigen-specific T cells could release additional epitopes to generate a broader neoantigen-specific immune response. When they screened several patients' samples for T-cell responses against epitopes that weren't included in the vaccine, they found peripheral immune responses against a handful of these neoepitopes and noted that these responses had not been present before administration of nivolumab or the vaccine. This, they wrote, suggested that "the presence of epitope spread could be a surrogate marker for tumor cell killing by neoantigen-specific T cells induced by vaccination."
They highlighted as particularly important that this epitope spread was correlated with patients' progression-free survival across all three tumor types, which further indicated that "it may only be necessary to target a subset of neoantigens expressed by the tumor to generate a broad immune response against the expressed neoantigens."
Despite these exploratory indications of immune response to NEO-PV-01, Ott and colleagues concluded that a confirmatory randomized trial comparing neoantigen-directed therapy plus anti-PD-1 treatment against just anti-PD-1 therapy will be needed to truly parse out the specific contribution of the personalized neoantigen vaccine.