Studies Back Neoadjuvant Strategies for Melanoma

Jeffrey S. Weber, MD, PhD


March 24, 2021

This transcript has been edited for clarity.

Hello. I'm Dr Jeffrey Weber. I'm a medical oncologist at the Laura and Isaac Perlmutter Cancer Center here in New York City. Today we're going to talk about neoadjuvant therapy.

The February issue of Nature Medicine included two articles on neoadjuvant therapy. In the first, Alex Menzies and colleagues from Australia summarized a compilation of neoadjuvant trials and discussed the clinical data. This summary included 192 patients with resectable stage III, palpable stage IIIB, or IIIC melanoma; 141 received neoadjuvant immuno-oncology drugs, and 51 received targeted therapy.

The pathologic complete response (pCR) rate, which in many other cancers is a very important clinical indicator, was 47% for those in the targeted-therapies group vs 33% for the immunotherapy group overall. But in the combination immunotherapy group the rate was 43%. For those with pCR, the 2-year relapse-free survival was an excellent 89%, compared with 50% for those without a pCR.

If you look at overall survival at 2 years, it was 95% for the pCR group vs 83%for those without a pCR. For those who had any pathologic response, which includes pCR, near CR (90% of the tumor regressed), or pathologic partial response (50% of the tumor regressed), the 2-year relapse-free survival was 96% with immuno-oncology drugs, which is pretty darn good. For the targeted therapies, the overall 2-year survival for those with pCR was an excellent 91% compared with 99% for those with pCR who got immuno-oncology drugs. Clearly, having a pCR is incredibly important clinically and pCR looks better with immuno-oncology drugs than with targeted drugs.

In all, 13 of the 17 who relapsed after targeted neoadjuvant therapy had recurrences in the brain compared with three of 15 who received immuno-oncology drugs. That seems like a big difference to me. The median follow-up of the entire group is only 18.8 months, so this is modest. Seven of the patients who received immunotherapy progressed before they could get to surgery.

On average, patients received 6 weeks of neoadjuvant therapy, although this was a mixed bag; some patients in the immuno-oncology group only received treatment for3 weeks, meaning a single dose of pembrolizumab. Most of the others got two cycles of immuno-oncology drugs for a total of 6 weeks. So this was a somewhat heterogeneous group. Overall survival at 2 years was 87% — some targeted, some immunotherapy — which is not much different from what Ascierto and colleagues found in the CheckMate 238 data on survival at 4 years.

Looking at the combination immunotherapy survival, 96% at 2 years is quite impressive, compared with 76% for monotherapy. This suggests that you'd probably rather get combination than monotherapy, if you had a RECIST (Response Evaluation Criteria in Solid Tumors) response on scanning before going to surgery at the end of your neoadjuvant therapy. Almost all of the patients who didn't have a RECIST response had no pathologic response. But almost all those who had a RECIST CR had either a pCR or a near pCR.

Again, it seems that immuno-oncology drugs are useful for those with any pathologic response — pCR, near pCR, or pathologic partial 50% response. All of those patients had an association with relapse-free survival, whereas only the targeted pCRs had an association with relapse-free survival. The authors concluded that achieving a pCR probably should become a new endpoint for drug assessment and approvals, which I believe is a reasonable position. This follow-up is modest, but they definitely are correct in saying that these data are grist for the mill to conduct randomized studies of neoadjuvant vs adjuvant therapies.

The second, smaller study also looked at the neoadjuvant spectrum. In fact, some of the patients in this second study, by Rozeman and colleagues, were included within the overall group in the Menzies trial. Rozeman and colleagues conducted a study of he OpACIN and the OpACIN-neo studies. OpACIN included about 20 patients and nine of them received combination immuno-oncology drugs (ipilimumab 3 mg/nivolumab 1mg) and were included in the current article; 86 patients were in the OpACIN-neo study. OpACIN-neo was a three-arm study of neoadjuvant immunotherapy, with two cycles of ipilimumab 3 mg/nivolumab 1 mg; two cycles of ipilimumab 1 mg/nivolumab 3 mg; or sequential nivolumab and ipilimumab for a total of four injections — essentially two combination cycles, all spaced over 6 weeks, followed by surgery.

The OpACIN study had an impressive 4 years of follow-up. In the seven out of nine patients who had a pathologic response of any kind, there were no relapses, but these were small numbers. OpACIN-neo had 2 years of follow-up and 86 patients, with an 84% relapse-free survival. If you had a pathologic response, your relapse-free survival was 97% after 2 years, whether you had a pCR, near pCR, or a pathologic partial response, but only a 36% relapse-free survival in 2 years if you had no pathologic response (less than a pathologic partial response of 50%).

The authors looked at a number of biomarkers and at tumor mutational burden (TMB), as well as at the interferon gamma signature that we're now familiar with and that is associated with outcome with a number of immuno-oncology drugs. As you would expect, at baseline, a high TMB and high gamma interferon signature were associated with both relapse-free survival and a 100% pathologic response of any kind. Impressive data.

For the whole population of OpACIN neo, 2-year survival was 95%, with no real difference among the arms. Toxicity was much lower in arm B (ipilimumab 1 mg/nivolumab 3 mg) with about 27% grade 3, 4, or 5 immunotherapy-related adverse events. But 21% of the patients overall needed hormone replacement, which is a fairly high number.

Again, with the immuno-oncology drugs, baseline TMB was associated with event-free survival in any way, shape, or form: 93% with the high TMB vs 56% at the median TMB. Both markers were associated with pathologic response. As you would expect, those patients with a low TMB, low gamma interferon signature, had a poor survival and a poor level of pathologic response. The gamma interferon signature contained the expected genes that were expressed, including CCXL 9, 10, and 11 gamma interferons, STAT1, CCR5, and IDO1, as well as HLA-DRA.

Only one of 71 patients with a pathologic response of any kind relapsed in that trial vs 16 of 23 who did not have a pathologic complete, near complete, or partial response. The authors' conclusions are quite nuanced and reasonable. These data certainly support conducting randomized controlled trials of neoadjuvant or neoadjuvant/adjuvant vs adjuvant therapy alone.

Please feel free to call in with questions. This is Dr Jeffrey Weber. Thank you very much.

Jeffrey Weber, MD, PhD, is deputy director of the Laura and Isaac Perlmutter Cancer Center at NYU Langone Medical Center and co-director of its Melanoma Research Program. His research, which has been continuously funded by the NCI for more than 20 years, focuses on experimental therapeutics and drug development, particularly in immunotherapy.

Follow Medscape on Facebook, Twitter, Instagram, and YouTube


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.