Adjuvant pembrolizumab extends RFS in high-risk melanoma
Click Here to Manage Email Alerts
Pembrolizumab extended RFS compared with high-dose interferon or ipilimumab as adjuvant therapy for patients with high-risk melanoma, according to randomized phase 3 study results presented during the virtual ASCO Annual Meeting.
Pembrolizumab (Keytruda, Merck) also appeared better tolerated.
Results did not show a significant difference in OS between treatment groups; however, researchers emphasized longer follow-up is needed and that crossover may have diluted the true OS effect.
“Our study is now the third study to show a solid RFS benefit for adjuvant anti-PD-1 in high-risk resected melanoma,” researcher Pauline Funchain, MD, assistant professor of medicine at Cleveland Clinic Lerner College of Medicine and a HemOnc Today Next Gen Innovator, told Healio. “Regardless of what long-term follow-up shows about OS — and I suspect adjuvant immunotherapy will show an OS benefit eventually — the delay or absence of recurrence is a really big deal to patients and their experience walking the cancer journey.”
Pembrolizumab is an anti-PD-1 antibody approved for the treatment of several cancer types. In February 2019, the FDA approved the agent as adjuvant treatment for patients with melanoma who have lymph node involvement after complete resection.
Funchain and colleagues conducted the intergroup S1404 trial to determine whether 1 year of adjuvant treatment with pembrolizumab extended survival compared with high-dose ipilimumab (Yervoy, Bristol Myers Squibb) — a CTLA-4 antibody — or high-dose interferon, the two regimens FDA-approved for adjuvant therapy of high-risk resected melanoma at the time of study design.
“Immunotherapy worked well in the metastatic setting, and this study set out to understand if immunotherapy would be effective in the adjuvant setting — both in terms of RFS and OS,” Funchain said.
The study included 1,303 adults with resected stage IIIA (11%), IIIB (49%), IIIC (34%) and IV (6%) melanoma enrolled between October 2015 and August 2017.
All patients had complete staging and adequate surgery rendering them disease free, and those with sentinel node-positive disease underwent completion lymph node dissection.
Exclusion criteria included central nervous system metastasis and prior treatment with ipilimumab, PD-1 blockade or interferon.
Researchers assigned 648 patients to 200 mg pembrolizumab via IV every 3 weeks for 52 weeks.
Investigators assigned the other 655 patients to one of two control regimens. One control regimen consisted of interferon alfa-2b dosed at 20 MU/m2 via IV on days 1-5 of weeks 1 to 4, followed by 10 MU/m2 via subcutaneous administration on days 1, 3 and 5 of weeks 5 to 52. The other control group received 10 mg/kg ipilimumab via IV every 3 weeks for four doses, then every 12 weeks for up to 3 years.
Treatment groups were balanced with regard to sex, age and disease stage.
Primary endpoints included RFS and OS among all patients, as well as OS among patients with PD-L1-positive biopsies at baseline.
Per study protocol, investigators performed the final analysis 3.5 years after the last patient was randomly assigned to treatment, by which time 512 RFS events and 199 OS events had occurred.
Results showed a statistically significant improvement in RFS among pembrolizumab-treated patients compared with the pooled control group (HR = 0.74; 99.62% CI, 0.57-0.95).
“Immunotherapy works by getting the host immune system to identify tumor,” Funchain said. “If you start with a clinical setting where there is little to no tumor, then the opportunity for immune recognition is lower. In that way, an RFS magnitude lower than that which we see in the metastatic setting makes sense.”
Researchers observed no significant difference in OS between pembrolizumab and the other therapies in the overall study population (HR = 0.84; 96.3%, 0.62-1.13) or among the 1,070 patients with PD-L1-positive baseline biopsies (HR = 0.88; 97.8% CI, 0.6-1.29).
However, the prespecified analysis was set for when the last patient reached 3.5 years of follow-up.
“This is a relatively short window for OS observations,” Funchain said. “Median OS from metastatic disease currently runs between 3 to 5 years and counting, so OS from an adjuvant study will need to run at least 5 [years] — and preferably 10 years — to be definitive.”
In addition, study protocol allowed for crossover. As observed in prior studies in oncology, this can dilute the true OS effect of first-line therapy.
“Multiple subsequent therapies were available and continue to expand and improve post-adjuvant OS, further diluting any effect of OS that adjuvant pembrolizumab may confer,” Funchain said.
Investigators observed no unexpected adverse events, and pembrolizumab appeared relatively well-tolerated compared with the other two therapies, Funchain said.
Analysis of grade 3 or higher adverse events showed patients assigned pembrolizumab appeared less likely than those assigned ipilimumab or high-dose interferon to experience any adverse event (31.7% vs. 57.5% vs. 71.9%), any adverse event related to treatment (19.5% vs. 49.2% vs. 71.2%) or an adverse events leading to discontinuation (16.9% vs. 64.8% vs. 24.7%).
The most common grade 3 or higher toxicities in the pembrolizumab group included elevated alanine transaminase (3%), diarrhea (2.8%), elevated aspartate transaminase (2.2%) and colitis (2%).
“For patients who are hesitant about adjuvant immunotherapy — whether due to concern over side effects, medical comorbidities, financial toxicity or a whole host of other personal concerns — this study shows no statistically significant short-term detriment to OS, which perhaps may afford a bit of leeway and flexibility in terms of skipping adjuvant immunotherapy within a shared decision-making conversation,” Funchain told Healio.
Perspective
Back to Top
Alexander M. Menzies, BSc(Med), MBBS(Hons), FRACP, PhD
In the era prior to effective therapy, patients had a high risk for recurrence and death after surgical resection. We now know adjuvant therapy prevents recurrence. Whether it is targeted therapy or PD-1-bsaed immunotherapy, it roughly halves the risk for recurrence and is now standard therapy across the world. But does adjuvant immunotherapy improve OS?
Importantly, PD-1 immunotherapy has a durable survival benefit in the metastatic setting. We may be curing patients in that setting, so it’s still not clear whether we should be treating people now in the adjuvant setting or wait until recurrence.
The problem with adjuvant therapy is some patients will not recur anyway, some will recur despite treatment, and a small group — about one in five patients — will derive benefit. However, all patients are treated and all have risks for toxicity. In addition, 1 year of drug therapy is expensive, and it requires regular clinic visits and infusions.
We need to move forward with adjuvant immunotherapy but we need more data on the use of adjuvant PD-1, and we need to know the effects of what built-in crossover trial design may have on OS.
The S14014 trial used OS as a primary endpoint, which was impressive, but the per-protocol final analysis took place at a preplanned 3.5 years after the last patient was randomly assigned, which is quite early. Only 199 survival events had occurred, and this limited the power of the study to detect an OS difference.
We saw an improvement in RFS as expected, as well as a trend toward but still statistically insignificant improvement in OS with pembrolizumab over standard of care. We saw a similar result in a preplanned PD-L1-positive subgroup.
Subsequent treatment has the potential to impact OS. The percentage of patients who received PD-1 as part of subsequent treatment was higher in the standard-of-care group than the pembrolizumab group (47% vs. 30%). But large numbers of patients — 31% in the pembrolizumab group and 30% in the standard-of-care group — don’t have any post-recurrence treatment reported. I think that impacts results of this study ... so this still doesn’t completely answer the question of post-recurrence treatment on OS.
The University of Sydney
Collapse
Grossmann KF, et al. Abstract 9501. Presented at: ASCO Annual Meeting (virtual meeting); June 4-8, 2021.
Read more about
Click Here to Manage Email Alerts