Long-Term Follow-Up Demonstrates Improved Outcomes With Two-Site Peptide Vaccination in Melanoma
Clinical Summary:
- Design/Population: This long-term follow-up of a multicenter melanoma vaccine trial evaluated peptide vaccination with Montanide ISA-51 with or without granulocyte-macrophage colony-stimulating factor (GM-CSF). The study also compared administration of the vaccine at one injection site versus two injection sites.
- Key Outcomes: Adding GM-CSF did not improve immune responses or long-term clinical outcomes. In contrast, administering the vaccine at two injection sites was associated with significantly improved recurrence-free survival and a trend toward improved overall survival compared with a single injection site.
- Clinical Relevance: These findings suggest that vaccine administration strategy may influence long-term clinical outcomes and support further optimization of peptide vaccine approaches for melanoma.
Craig Slingluff, MD, University of Virginia, Charlottesville, Virginia, discusses long-term follow-up from a phase 2 melanoma vaccine trial evaluating peptide vaccination with or without granulocyte-macrophage colony-stimulating factor (GM-CSF) and the impact of vaccine administration strategy on clinical outcomes.
The study demonstrated that adding GM-CSF did not improve vaccine efficacy, whereas administering peptide vaccines at two injection sites resulted in significantly improved recurrence-free survival compared with a single injection site. Dr Slingluff also discusses the implications of these findings for the design of future cancer vaccine trials and the evolving role of neoadjuvant immunotherapy and vaccine-based strategies in melanoma.
Transcript:
Hi, my name is Craig Slingluff. I'm a surgeon and researcher at the University of Virginia where my research focuses on cancer immunology, especially cancer vaccines, mostly in patients with melanoma.
This recent study was a long-term follow-up of a clinical trial we did more than 10 years ago, where the basic focus was on a cancer vaccine approach for melanoma using peptides to induce CD8 T-cells against melanoma antigens. The question we asked in the original study, which was published in 2009, was whether adding granulocyte-macrophage colony-stimulating factor (GM-CSF) to the vaccine adjuvant improved the immune response compared with not adding it. All the vaccines also included incomplete Freund's adjuvant, called Montanide ISA-51, so it was with or without GM-CSF added to that.
That was really our main question because we and others had been using GM-CSF based on a lot of studies suggesting that it was helpful in priming an immune response initially. This continues to be an important question because GM-CSF is still used with a number of vaccine approaches. It's also used in oncolytic virus approaches for melanoma and other cancers.
We were surprised to find in that initial study that including GM-CSF dramatically reduced the circulating CD8 T-cell immune response to the vaccine compared with not including it, just using incomplete Freund's adjuvant alone. That study also asked a question that, at the time, we thought was not that exciting, which was whether it's better to vaccinate in 2 sites versus 1. So, giving the same vaccine in 1 site with each dose or dividing that same dose into 2 sites each time. We were interested in knowing that just for some clinical trial design questions and thought that wouldn't be that interesting, but it turns out to be interesting.
In the initial data we found much lower immune responses by adding GM-CSF. We didn't see a significant improvement in immune response by vaccinating in 2 sites versus 1, but there was a trend toward that.
We continued to follow these patients, who were at 5 centers, and then reported this longer-term follow-up, with a median follow-up of 5.6 years, but some patients were followed for as long as 20 years.
There we found, first with regard to the GM-CSF, that there was no improvement in survival or recurrence-free survival by adding GM-CSF. If anything, there was a slight diminution in the overall clinical outcome with GM-CSF, but it certainly wasn't significant. There was clearly nothing to suggest a benefit of GM-CSF in the vaccines.
Then the other question about 2-site vaccination versus 1—surprisingly, we saw significantly improved recurrence-free survival by vaccinating in 2 sites compared with 1, and a trend that looks quite promising for better overall survival, with a P value of .08 for 2-site vaccination versus 1.
There was also a real leveling of the curves. To me, this is very provocative. People don't think a lot about vaccine sites and how to divide the dose. Sometimes trials will give 2 different vaccines in 2 different sites or divide the peptides into 2 different sites. Sometimes they rotate from 1 site to another, so a different site each time. We found that giving the vaccines at the same visit each time but dividing the dose into 2 different sites each time, seems to be better and, surprisingly, leads to better clinical outcomes. The clinical outcome improvement was significant even in multivariable analyses, where really that was the main determinant of outcome.
The overall CD8 T-cell response again trended better with 2-site vaccination, but not significantly. So it's interesting to think about what may be causing this difference. It may well be a different quality of the T-cell response, which we'd be interested in digging into further in the future.
These data stand to us as being valuable, and in our current trials and future trials we will divide peptide vaccine doses into 2 sites each time, and we will not be using GM-CSF, as we have not been for a long time.
Source:
Ninmer EK, Zhu H, Sarkar A, et al. Impact of GM-CSF and two-site vaccination on clinical outcomes after multipeptide vaccination for melanoma: Long-term analysis of a randomized phase II trial. Clin Cancer Res. Published online: June 15, 2026. doi: 10.1158/1078-0432.CCR-25-4606
