Changes in Merkel Cell Oncoprotein Antibodies (AMERK) After Radiation Therapy (RT) in Curatively Treated Merkel Cell Carcinoma (MCC) and Association With Recurrence

Abstract

Serum antibodies to the Merkel oncoprotein (AMERK) are expressed in approximately 50% of patients (pts) with MCC and can be used to monitor for recurrence. This study aimed to describe changes in AMERK levels after RT for MCC and identify patterns that predict disease recurrence.Pts with MCC were identified retrospectively if they had curative-intent postoperative or definitive RT at a single institution from 2010-2020. Pts treated with RT at other institutions were excluded due to inability to assess RT quality, previously shown to impact MCC outcomes. All patients had a pre-RT AMERK and at least one post-RT AMERK (median no.5, range 1-10). Changes in AMERK were described and compared using Wilcoxon rank sum test for two-sample non-parametric comparisons and the Chi-square test for recurrence association.Of 202 MCC pts treated at our institution, 61 curatively treated pts had a pre-RT AMERK test performed. AMERK was detectable in 27 of the 61 patients (44%) with a median value of 2950 (range 127 to 115,000 [upper limit of assay]). Detectable pre-RT AMERK was associated with decreased MCC recurrence compared to pts with undetectable pre-RT AMERK (OR 0.35, 95% CI 0.12,1.04). The 27 pts with positive pre-RT AMERK had Stages I (n = 5), II (n = 1), and III (n = 21) from sites of the extremity (n = 14), head/neck (n = 7), unknown primary (n = 4), and trunk (n = 2). Pts with gross disease at RT (n = 11) had higher median pre-RT AMERK than those without gross disease at RT (n = 16) (median 8550 vs. 1840, P = 0.06). With a median follow-up of 20 months (range 2.3-61.3), there were 7 recurrences, all were distant. All AMERK-positive pts were alive at last follow-up except for the single immunosuppressed pt who recurred and died of MCC. A ≥50% AMERK reduction by 6-months after RT had a sensitivity of 95% (95% CI 74-100) for detecting disease-free status and a negative predictive value of 83% (95% CI 41-97%), reflecting the probability of recurrence if < 50% AMERK at 6-months. The sensitivity and negative predictive values of ≥50% 6-month AMERK reduction both increased to 100% respectively when excluding pre-RT borderline AMERK values (75-150). One pt who recurred after ≥50% AMERK reduction from an initial > 115,00 value, had initiated high-dose corticosteroids for a comorbid condition. Out of 8 recurrence-free patients with AMERK levels available more than 2-years post-RT, 50% (4 of 8) had detectable AMERK (range 239-1960, > 2 years post-RT).Reduction in AMERK post-RT can be used to predict MCC recurrence, which rarely occurs with a sustained ≥50% reduction by 6-months post-RT. AMERK may be persistently detectable in recurrence-free pts beyond 2 years. Further research is needed to determine the impact of corticosteroid initiation on AMERK levels.