Peptide vaccines possess several advantages over mRNA vaccines but are generally less effective at inducing antitumor immunity. The bottlenecks limiting peptide vaccine efficacy could be elucidated by tracking and comparing vaccine-induced T-lymphocytes in successful and unsuccessful cases. Here we have applied our recent database of neoantigen-specific T cell receptors (TCRs) to profile tumor-specific T cells following vaccination with a neoantigen peptide vaccine and to correlate this with the response. Mice were vaccinated prophylactically with p30 peptide encoding B16 melanoma neoantigen (K739N mutation in Kif18b gene). The B16F0 melanoma in the vaccinated mice was additionally treated by a CTLA-4 checkpoint blockade. T cells from the tumors, tumor-draining lymph nodes (tdLNs) and vaccine depots were isolated, phenotyped, sorted by subsets and sequenced for TCR repertoires. The vaccine induced the accumulation of tumor-specific CD4+ Th cells in the tdLNs, while in the tumors these cells were present and their frequencies were not changed by the vaccine. These cells also accumulated at the vaccine depots, where they were phenotypically skewed by the vaccine components; however, these effects were minor due to approximately 50-fold lower cell quantities compared to the tdLNs. Only some of the p30-specific Th cells showed tumoricidal activity, as revealed by the reverse correlation of their frequencies in the tdLNs with the tumor size. The CTLA-4 blockade did not affect the tumor growth or the frequencies of tumor-specific cells but did stimulate Th cell motility. Thus, we have shown that tumor-specific Th clones accumulate and/or expand in the tdLNs, which correlates with tumor suppression but only for some of these clones. Tumor infiltration by these clones is not correlated with the growth rate.

Background: Some peptide anticancer vaccines elicit a strong T-cell memory response but fail to suppress tumor growth. To gain insight into tumor resistance, we compared two peptide vaccines, p20 and p30, against B16 melanoma, with both exhibiting good in vitro T-cell responses but different tumor suppression abilities. Methods: We compared activation markers and repertoires of T-lymphocytes from tumor-draining (dLN) and non-draining (ndLN) lymph nodes for the two peptide vaccines. Results: We showed that the p30 vaccine had better tumor control as opposed to p20. p20 vaccine induced better in vitro T-cell responsiveness but failed to suppress tumor growth. Efficient antitumor vaccination is associated with a higher clonality of cytotoxic T-cells (CTLs) in dLNs compared with ndLNs and the convergence of most of the enriched clones. With the inefficient p20 vaccine, the most expanded and converged were clones of the bystander T-cells without an LN preference. Conclusions: Here, we show that the clonality and convergence of the T-cell response are the hallmarks of efficient antitumor vaccination. The high individual and methodological dependencies of these parameters can be avoided by comparing dLNs and ndLNs.