Does Neoadjuvant Immunotherapy Pathologic Complete Response Predict Better Outcomes in Melanoma?

Jeffrey S. Weber, MD, PhD


April 16, 2019

This transcript has been edited for clarity.

Hello. I'm Dr Jeffrey Weber. I am a medical oncologist and the deputy director at the Laura and Isaac Perlmutter Cancer Center in New York City at the NYU Langone Medical Center.

Today I'd like to tell you about a couple of articles that appeared recently in Nature Medicine, which generally is a fairly basic publication, but these have significant clinical implications. There were two rather similar articles that made a very interesting point about the concept of neoadjuvant immunotherapy and melanoma.

The first was a randomized study that included only 20 patients—10 in each arm—with stage IIIB/IIIC resected melanoma, who received either neoadjuvant ipilimumab and nivolumab at the standard doses of 3 mg/kg and 1 mg/kg, respectively, for two courses.[1] Then they had surgery and received two additional courses—that was the so-called neoadjuvant arm—versus the adjuvant arm, where all patients had surgery and then received four courses of the same therapy.

The interesting aspect of this study is that there was significant toxicity. Overall, nine out of 10 patients in each arm had grade 3/4 immune-related adverse events and they had great difficulty completing the therapy. All patients still got to surgery, but at the end of the day, this was a trial that had serious toxicity and it ended early.

However, that's not the interesting aspect. What's interesting is that there was a significant response rate in the neoadjuvant arm, and seven of the nine evaluable patients had significant or major clinical responses, and two of those patients had pathologic complete responses. Again, of those who had clinical responses, which were generally major responses with more than 90% clearing of the lymph nodes, none of those patients have relapsed with almost 2 years of follow-up.

Of those who had no response, one in each arm died, and one of those who died in the neoadjuvant arm, of course, was a patient who did not respond. Interestingly, an additional patient had severe and significant side effects and also died. Nonetheless, the fact that only two patients died and only one patient relapsed in the neoadjuvant arm versus four patients in the adjuvant arm gave some thought that the neoadjuvant approach may be superior.

Looking at the correlative marker studies, there was a significant increase in the clonality in the responders, or the benefiters, compared with the nonbenefiters. The patients who did well with neoadjuvant therapy tended to have both an increased level of clonality and a better clonality at baseline. Overall, these are very interesting data suggesting that the clonality of the T cells infiltrating the tumors may be beneficial.

The authors also looked at a number of other factors and found, as you would expect, that things like PD-L1 and tumor mutational burden were also associated with doing better with neoadjuvant immunotherapy. This trial has actually been continued with what we call flip doses of ipilimumab and nivolumab, and that's a report for another day.

In the same issue of Nature Medicine, there was a report of 23 patients, which was in a slightly different format.[2] This study included patients with stage IIIB, IIIC, and IV resected melanoma—mostly IIIB and IIIC.

The authors varied a neoadjuvant approach, giving either four doses of nivolumab over 8 weeks—every 2 weeks at the standard dose of 240 mg—followed by surgical resection and then 6 months of adjuvant nivolumab, which is half the standard FDA-approved regimen, versus getting two cycles of ipilimumab and nivolumab pretreatment—that's the neoadjuvant ipilimumab/nivolumab arm—at the standard FDA-approved doses, also followed by 6 months of adjuvant nivolumab.

Again, the major issue was significant toxicity. There was a relatively high level of toxicity in the arm in which patients received the ipilimumab and nivolumab, but the response rate that could be measured by RECIST was actually 73%, which is very impressive. The pathologic complete response rate in the ipilimumab/nivolumab neoadjuvant arm was 45%. That would be compared with a 25% response rate only and a very low pathologic complete response rate in the group that received nivolumab only.

This trial was stopped early because two out of the 12 patients who were treated in the nivolumab-alone arm couldn't have surgery because they progressed too quickly. Nonetheless, the progression-free survival, where they're looking at the RECIST response at the time of surgery, 8 weeks after starting, showed that you had a much better progression-free survival rate in the combination ipilimumab/nivolumab arm compared with nivolumab alone. Survival was superior and virtually all of the patients who got ipilimumab/nivolumab have remained alive, whereas several of the patients have already died in the nivolumab-only arm.

As you would expect, PD-L1 staining in the primary was associated with doing well. The tumor mutational burden and the clonality were associated with doing well with neoadjuvant combination immunotherapy. In addition, the increase in the clonality over time was also clearly associated with benefit.

Although the trial stopped early because too many patients progressed and couldn't have surgery in the nivolumab-alone arm, these authors suggested that neoadjuvant combination immunotherapy may be a superior regimen.

The authors also concluded that if the achievement of a pathologic complete response happens with limited follow-up—in this particular trial, it was only about 15 months—those patients will probably do very well. Furthermore, these patients may not need additional therapy or may need only a small amount of additional therapy, in which case the authors gave only 6 months of adjuvant nivolumab.

It's not clear that neoadjuvant therapy is better than adjuvant therapy. The major issue is whether the achievement of a pathologic complete response in melanoma using neoadjuvant combination immunotherapy would predict a superior outcome, meaning you didn't have to give them any other additional treatment.

These are important questions that need to be addressed in a randomized, phase 3 trial.

Again, this is Dr Jeffrey Weber reporting. Please feel free to write in with comments and questions. Thanks very much for your attention.

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.