Baseline and post-treatment biomarkers of resistance to anti-PD-1 therapy in acral and mucosal melanoma: an observational study.

BACKGROUND: Immunotherapies targeting programmed cell death-1 (PD-1) and its ligands have improved clinical outcomes for advanced melanoma. However, many tumors exhibit primary resistance or acquire secondary resistance after an initial positive response. The mechanisms of resistance are not well understood, and no validated predictive biomarkers are available. This exploratory study aimed to characterize baseline differences and molecular changes arising during treatment in acral and mucosal melanomas that exhibited primary or secondary resistance to anti-PD-1 monotherapy. METHODS: This was an observational retrospective study of 124 patients who had been treated for metastatic acral or mucosal melanoma with anti-PD-1 monotherapy. Tumor samples were collected at baseline (all patients) and post-treatment (resistant tumors only) and were assayed by immunohistochemistry, whole-exome sequencing, and RNA sequencing. RESULTS: At baseline, more non-progressor than resistant tumors exhibited expression of PD-L1, immune cell infiltration, and high tumor mutational burden (TMB); baseline PD-L1 expression was also more common in secondary-resistant than in primary-resistant tumors as well as in late versus early secondary-resistant tumors. Non-progressor tumors also had higher median baseline expression of an 18-gene T cell-inflamed gene expression profile (TcellinfGEP). Among resistant tumors, the proportion of PD-L1-positive melanomas and the expression of the TcellinfGEP mRNA signature increased during treatment, while the expression of mRNA signatures related to WNT and INFA1 signaling decreased. There was evidence for greater changes from baseline in secondary-resistant versus primary-resistant tumors for some markers, including expression of RAS-related and WNT-related mRNA signatures and density of CD11c+ and FOXP3+ T cells. Greater changes in CD11c+ cell density were observed in early compared with late secondary-resistant tumors. CONCLUSIONS: Our findings suggest that TcellinfGEP and PD-L1 expression, TMB, immune cell infiltration, and RAS and WNT signaling warrant further investigation as potential mechanisms and/or biomarkers of anti-PD-1 therapy resistance in acral and mucosal melanomas. Confirmation of these findings in larger populations is needed.

Bleeding and spotting with the levonorgestrel 13.5 mg intrauterine system: the impact of insertion timing.

OBJECTIVE: To assess the impact of early versus late menstrual cycle insertion on bleeding/spotting in the 90 days following levonorgestrel (LNG) 13.5 mg intrauterine system (IUS) insertion. STUDY DESIGN: In this observational study, participants received a LNG 13.5 mg IUS and provided 90 days of bleeding/spotting data by answering the following daily text: "Have you had no flow (0), spotting (1), or bleeding (2) today?" We dichotomized insertion timing as early (days 1-7 from last menstrual period) and late (remainder of menstrual cycle) and compared bleeding/spotting between the two groups in the 90- and 30-day reference periods. We used multivariate regression methods to study associations between cycle day at insertion, parity, historical bleeding, recent hormonal contraceptive use and bleeding/spotting. RESULTS: In the 90-day dichotomous analysis (n=125), we found no differences in the number of days of bleeding/spotting, bleeding or spotting between the early and late insertion groups. In the 30-day dichotomous analysis (n=131), early insertion was associated with fewer days of bleeding than late insertion (5±3 vs. 7±4 days, p<.01). Recent hormonal contraceptive users experienced fewer days of bleeding than new users (5±4 vs. 7±3 days, p<.01). In the 90- and 30-day regression models, earlier insertion was associated with fewer days of bleeding (p=.02, p=.02). Recent contraceptive use was associated with fewer days of bleeding/spotting (90-day, p=.03) and fewer days of bleeding (30-day, p<.01). Nulliparity was associated with spotting (30-day, p=.04). CONCLUSIONS: Early cycle insertion does not impact 90-day bleeding/spotting. Early cycle insertion and recent hormonal contraceptive use decrease 30-day bleeding. IMPLICATIONS: The LNG 13.5 mg IUS may be inserted throughout the menstrual cycle with small differences in bleeding patterns in the 30 but not the 90 days following insertion. Shared decision making should determine timing of insertion.