Abstract
<jats:p>Background: Acral lentiginous melanoma (ALM) is the predominant melanoma subtype in East Asian populations, accounting for roughly 50 to 58% of cases, compared with 2 to 3% in populations of European ancestry. ALM is genomically and biologically distinct from sun-exposed cutaneous melanoma, and East Asian and acral patients were markedly under-represented in the pivotal antiPD1 registration trials. At the time this study was designed, no prospective trial had evaluated a checkpoint inhibitor specifically in ALM. We conducted a phase II trial to estimate the activity of pembrolizumab in this population. Methods. In this single centre, single arm, open label phase II trial, adults with metastatic or locoregionally advanced inoperable ALM who were naive to antiPD1 or antiPDL1 therapy received pembrolizumab 200 mg intravenously every 3 weeks until progression, unacceptable toxicity, or withdrawal. The primary endpoint was objective response rate (ORR) by RECIST 1.1. Secondary endpoints included duration of response (DoR), clinical benefit rate (CBR), progression-free survival (PFS), overall survival (OS), and safety (CTCAE v4.0). A Simon minimax two-stage design (P0=0.10, P1=0.30, power=80%) planned enrolment of up to 28 patients. Results. Between February 2017 and June 2019, 9 patients were enrolled before recruitment was halted for slow accrual, the interval availability of reimbursed pembrolizumab, and a low observed response signal. Median age was 72 years (range 48 to 78); 6 (67%) were male; all had ECOG performance status 0 and metastatic disease; 7 (78%) had received prior therapy. One patient achieved a partial response (ORR 11.1%, 95% CI 0.0 to 31.6%), with a DoR of 19 months; 3 had stable disease and 4 progressed. CBR (response or stable disease greater than or equal to 12 weeks) was 44.4% (95% CI 12.0 to 76.9). At a median follow-up of 7.6 months, median PFS was 3.4 months (95% CI 1.4 to 21.3) and median OS was 7.6 months (95% CI 2.0 to 34.3). Two grade 3 adverse events occurred, both assessed as unrelated to study drug; no treatment-related grade 3 or above events were recorded. In an exploratory analysis, an LDH to upper limit of normal ratio >1.5 was associated with worse OS (median 4.3 vs 26.5 months; HR 4.58, 95% CI 0.82 to 25.7; log-rank p=0.06). Conclusions. Recruitment was constrained by disease rarity and a shifting reimbursement landscape, and the trial closed before completing stage 1. Within these limitations, single-agent pembrolizumab showed only modest activity in advanced ALM, consistent with the limited efficacy subsequently reported in larger contemporary acral melanoma cohorts. The exploratory association between elevated LDH ratio and poorer survival warrants prospective evaluation.</jats:p>