Chronic irAEs After Adjuvant Anti–PD-1 Therapy More Common Than Previously Recognized in High-Risk Resected Melanoma

Douglas Johnson, MD, MSCI

Chronic immune-related adverse effects (irAE) arising from adjuvant anti–PD-1 therapy in patients with high-risk melanoma are more common than previously recognized, and often persisted with longer follow-up, according to findings from a multicenter cohort study published in JAMA Oncology.

Chronic irAEs, which are defined as those that persist after 12 weeks of anti–PD-1 discontinuation, occurred in 43.2% of patients enrolled to the trial (n = 167/387). Although the majority of these chronic irAEs were grades 1 or 2 (96.4%; n = 161/167), most of the effects persisted until last available follow-up (85.6%; n = 143/167).

“It is certainly important to recognize that anti–PD-1 therapy, regardless of this study and others, certainly does improve outcomes for patients with melanoma and is something that I will certainly still be offering my patients” lead study author Douglas Johnson, MD, MSCI, an associate professor of Medicine at Vanderbilt University Medical Center, and clinical director of Melanoma at Vanderbilt-Ingram Cancer Center, told OncLive. “There is a subset of patients with stage III melanoma who have a very low risk of recurrence, and for those patients, [these data] may make us reevaluate how to [treat] them, and perhaps [revisit] how [we] make risk/benefit determinations in those settings.”

Moreover, irAEs such as endocrinopathies (83.0%; n = 73/88), arthritis (48.9%; n = 22/45), xerostomia (52.9%; n = 9/17), neurotoxicities (73.3%; n = 11/15) and ocular effects (62.5%; n = 5/8), were the most likely to become chronic, while effects impacting visceral organs like the liver, colon, lungs, and kidneys, had much lower rates of becoming chronic.

“We tend to think of specific organs that are being affected by these toxicities,” Johnson said. “The big ones that come to mind are the colon, the lungs, and the liver, but the toxicities in those organs seem to really get better; it was toxicities in the smaller organs that seemed to more commonly develop into chronic [effects].”

Adjuvant therapy with anti–PD-1 agents has been found to result in improved relapse-free survival (RFS) in patients with resected, advanced-stage melanoma. However, the long-term effects of this treatment approach remain uncharacterized. To address this knowledge gap, investigators launched this retrospective, multicenter cohort study to determine the incidence, length, spectrum and associations of chronic irAEs that arise from adjuvant anti–PD-1 treatment in patients with stage III and IV melanoma. The study was conducted from 2015 to 2020 at 8 medical centers across the United States and Australia.

The median age for patients enrolled on the study was 63 years (range, 17-88) and the majority were male (60.7%; n = 235/387). Additionally, 74.9% (n = 290) had preexisting comorbidities, 85.8% (n = 332) had cutaneous primary sites, and 51.2% (n = 198) had BRAF/NRAS wild-type variants. Moreover, 33.1% (n = 128) of patients had resected stage IIIB melanoma, while 39.5% (n = 153) had resected stage IIIC disease. Around 84% (n = 326) of patients had received prior treatment with nivolumab (Opdivo), and 15.8% (n = 61) of patients received pembrolizumab (Keytruda), with a median treatment duration of 306 days (range, 1-1049).

Reasons for treatment discontinuation included therapy completion for 50% (n = 193), irAEs for 25.3% (n = 98) of patients, and disease progression for 20.9% (n = 81) of patients. Moreover, median RFS and overall survival had not been reached. At a median follow-up of 529 days, the majority of patients (93%) were alive. Notably, patients with observed acute and chronic irAEs were found to have superior RFS compared with those who did not experience irAEs.

Of all 387 patients included on the trial, 69.0% (n = 279) experienced acute irAEs; 44.2% (n = 171) of these effects were grade 2 or higher, while 13.4% (n = 52) were grades 3 to 5 in severity. Two patients died from irAEs; 1 patient had myocarditis, while the other had Guillain-Barré–like syndrome.

The most common acute irAEs observed were dermatitis/pruritus (25.8%), thyroiditis/hypothyroid (16.3%), arthralgias (10.6%), colitis/diarrhea (9.8%), hepatitis (6.2%), pneumonitis (4.4%), and xerostomia (4.1%). About 28% of patients required treatment with glucocorticoids; 55.0% needed this for grade 2 irAEs, while 45.0% needed them for grade 3 to 5 effects.

In those who developed chronic irAEs, systemic glucocorticoids were needed in 32.9% (n = 55) of patients. The most common chronic irAEs observed were hypothyroidism (14.0%), arthralgias (5.7%), dermatitis/ pruritus (6.6%), adrenal insufficiency (3.1%), and xerostomia (2.3%).

Age and gender were not found to be linked with the development of chronic irAEs (63.0 years vs 63.0 years in those who did not develop chronic irAEs, P = .67 and male vs female, P = .31, respectively). Moreover, the median time of onset of acute irAEs was not determined to be associated with chronicity (P = .95). With the exception of endocrinopathies, those who were given glucocorticoids for acute effects did not go on to develop chronic effects more often than those who did not receive the agents (P = .15).

The investigators concluded that the risks of chronic irAEs should be integrated into treatment decision-making for this patient population.

“In terms of future directions, we might potentially be able to intervene early and identify these effects before they established themselves to become chronic,” Johnson noted.

Reference

Patrinely JR, Johnson R, Lawless AR, et al. Chronic immune-related adverse events following adjuvant anti–PD-1 therapy for high-risk resected melanoma. JAMA Oncology. Published online March 25, 2021. doi:10.1001/jamaoncol.2021.0051