Randomized Phase I/II Clinical Trial of a Melanoma Helper Peptide Vaccine with or without Systemic Agonistic Anti-CD27 Antibody (Varlilumab).

PURPOSE: This randomized phase I/II clinical trial (NCT03617328) was designed to test whether administration of systemic agonistic anti-CD27 antibody (varlilumab) concurrent with a melanoma vaccine is safe and enhances vaccine immunogenicity.

PATIENTS AND

METHODS

Adults with definitively treated, high-risk melanoma were randomized to receive a shared antigen vaccine targeting CD4+ T cells (6 melanoma helper peptides; 6MHP) either with (arm A) or without (arm B) systemic varlilumab over 11 weeks. A final vaccine was given at week 25 to assess memory response.

RESULTS

Thirty-three participants were treated at two centers. After enrolling 17 participants, the dose-limiting toxicity (DLT) rate with vaccine alone required protocol revision to limit local toxicity. Overall, DLT rates for arms A and B were 6% (1/17) and 31% (5/16), respectively. Twenty (61%) participants had CD4+ T-cell responses ex vivo. Persistent responses were detected in two (2/14, 14%) on arm A and one (1/16, 6%) on arm B. Memory response was detected in two (2/13, 15%) on arm A and four (4/16, 25%) on arm B. On arm A, there was a significant reduction in circulating CD4+ T cells. Four-year disease-free survival rates for arms A and B were 20% [95% confidence interval (CI), 6%-67%] and 69% (95% CI, 49%-96%), respectively.

CONCLUSIONS

No synergistic toxicity was observed with combination treatment. Similar durability in immunogenicity was found with or without varlilumab. Depletion of circulating CD4+ T cells with varlilumab may have abrogated benefits of inducing tumor-cognate CD4+ T cells with vaccination. Induction of memory responses supports further work to optimize shared antigen vaccines in combination with other immunotherapies.

SIGNIFICANCE: Combination treatment with 6MHP vaccination and varlilumab was safe but did not improve CD4+ T-cell response rates. Circulating CD4+ T cells were reduced with varlilumab, and clinical outcome was improved without varlilumab. This study demonstrates induction of memory responses after shared antigen vaccination; however, CD27 agonism did not enhance durability of response. Depletion of CD4+ T cells after varlilumab may have interfered with the beneficial effects of vaccination.