Long-term follow-up showed pTVG-HP nearly doubled median OS compared with placebo in biochemically recurrent prostate cancer.
The DNA vaccine pTVG-HP (MVI-816), which encodes prostatic acid phosphatase (PAP), was associated with significantly longer overall survival (OS) than placebo in patients with PSA-recurrent, non-metastatic, castration-sensitive prostate cancer, according to long-term follow-up data from a randomized phase 2 trial (NCT01341652) published in Clinical Genitourinary Cancer.¹
The median OS was 13.4 years in patients who were randomly assigned to pTVG-HP (n = 48) vs 8.6 years in those randomized to placebo (n = 49; HR, 0.42; 95% CI, 0.20-0.90). Findings were consistent in a subset of 59 patients treated at a single site with complete long-term follow-up data on survival and subsequent treatment, in whom the adjusted HR for OS was 0.60 (95% CI, 0.36-1.00) after accounting for baseline characteristics and time-varying exposure to androgen deprivation therapy (ADT) and subsequent treatment.
The randomized, placebo-controlled phase 2 trial was conducted between 2011 and 2016 at 3 academic centers: University of Wisconsin Carbone Cancer Center, University of California San Francisco, and Johns Hopkins University.¹ Eligible patients had PSA-recurrent, non-metastatic, castration-sensitive prostate cancer with a pre-treatment PSA doubling time (DT) of less than 12 months and no metastases by conventional imaging (CT and bone scan).
Patients were randomly assigned 1:1 to 100 μg of pTVG-HP plus 200 μg of GM-CSF adjuvant or 200 of μg GM-CSF alone, administered intradermally every 2 weeks for 3 months, then every 3 months for up to 2 years total. Randomization was stratified by pre-treatment PSA DT (0-3, 3-6, or 6-12 months), initial Gleason score (≤7 or >7), and baseline PSA (≤10 ng/mL or >10 ng/mL).
The trial's original primary end point, 2-year metastasis-free survival (MFS), was not met, with rates of 41.8% in the pTVG-HP arm vs 42.3% in the placebo arm (P = .97).² The trial was reopened at the University of Wisconsin site in 2025 to collect longer-term OS and subsequent-treatment data; comparable data could not be obtained from the other 2 sites beyond approximately 4 years from randomization, and median follow-up for the long-term OS analysis was 4.1 years (range, 0.1-13.8+ years).¹
The median MFS among the 59 patients with complete long-term follow-up data, was 23.9 months with pTVG-HP vs 18.1 months with placebo (HR, 0.80; 95% CI, 0.41-1.55). Of the 56 patients in this cohort who received subsequent ADT, median time from randomization to ADT initiation was 2.2 years with pTVG-HP vs 1.4 years with placebo (HR, 0.70; 95% CI, 0.41-1.20). Among 46 patients who went on to receive therapy for castration-resistant disease after ADT, median time from ADT initiation to next therapy was 4.7 years with pTVG-HP vs 2.1 years with placebo (HR, 0.62; 95% CI, 0.34-1.12). No patients in the cohort received a second-generation androgen receptor pathway inhibitor concurrently with ADT for castration-sensitive disease, reflecting the treatment landscape at the time the trial was conducted.
The study authors noted that the survival benefit observed with pTVG-HP could not have been detected using the trial's original MFS end point, and that the findings are consistent with prior data from other prostate cancer vaccines, including sipuleucel-T (Provenge), suggesting cancer vaccines may affect longer-term survival more than shorter-term progression measures. Reported limitations included that the OS analysis was unplanned and the trial was not prospectively powered to detect a survival difference, that long-term data could not be obtained from all original trial sites, introducing potential selection bias, and that heterogeneity in subsequent therapies for castration-resistant disease may have influenced the OS difference between arms.
A randomized trial evaluating pTVG-HP in combination with PD-1 blockade for OS in metastatic castration-resistant prostate cancer is being planned, according to the authors, who also suggested that earlier-stage settings with shorter, more feasible end points, such as the neoadjuvant setting, may be better suited to evaluating single-agent vaccine activity going forward.
