IMPORTANCE Merkel cell carcinoma (MCC) often behaves aggressively; however, disease-recurrence data are not captured in national databases, and it is unclear what proportion of patients with MCC experience a recurrence (estimates vary from 27%-77%). Stage-specific recurrence data that includes time from diagnosis would provide more precise prognostic information and contribute to risk-appropriate clinical surveillance. OBJECTIVE To estimate risk of stage-specific MCC recurrence and mortality over time since diagnosis. DESIGN, SETTING, AND PARTICIPANTS This prospective cohort study included 618 patients with MCC who were prospectively enrolled in a Seattle-based data repository between 2003 and 2019. Of these patients, 223 experienced a recurrence of MCC. Data analysis was performed July 2019 to November 2021. MAIN OUTCOMES AND MEASURESStage-specific recurrence and survival, as well as cumulative incidence and Kaplan-Meier analyses. RESULTS Among the 618 patients included in the analysis (median [range] age, 69 years; 227 [37%] female), the 5-year recurrence rate for MCC was 40%. Risk of recurrence in the first year was high (11% for patients with pathologic stage I, 33% for pathologic stage IIA/IIB, 30% for pathologic stage IIIA, 45% for pathologic stage IIIB, and 58% for pathologic stage IV), with 95% of recurrences occurring within the first 3 years. Median follow-up among living patients was 4.3 years. Beyond stage, 4 factors were associated with increased recurrence risk in univariable analyses: immunosuppression (hazard ratio [HR], 2.4; 95% CI, 1.7-3.3; P < .001), male sex (HR, 1.9; 95% CI, 1.4-2.5; P < .001), known primary lesion among patients with clinically detectable nodal disease (HR, 2.3; 95% CI, 1.4-4.0; P = .001), and older age (HR, 1.1; 95% CI, 1.0-1.3; P = .06 for each 10-year increase). Among 187 deaths in the cohort, 121 (65%) were due to MCC. The MCC-specific survival rate was strongly stage dependent (95% at 5 years for patients with pathologic stage I vs 41% for pathologic stage IV). Among patients presenting with stage I to II MCC, a local recurrence (17 arising within/adjacent to the primary tumor scar) did not appreciably diminish survival compared with patients who had no recurrence (85% vs 88% MCC-specific survival at 5 years). CONCLUSIONS AND RELEVANCEIn this cohort study, the MCC recurrence rate (approximately 40%) was notably different than that reported for invasive melanoma (approximately 19%), squamous cell carcinoma (approximately 5%-9%), or basal cell carcinoma (approximately 1%-2%) following definitive therapy. Because more than 90% of MCC recurrences arise within 3 years, it is appropriate to adjust surveillance intensity accordingly. Stage-and time-specific recurrence data can assist in appropriately focusing surveillance resources on patients and time intervals in which recurrence risk is highest.

No abstract

No abstract

Prognostic analysis for early-stage (stage I/II) melanomas is of paramount importance for customized surveillance and treatment plans. Since immune checkpoint inhibitors have recently been approved for stage IIB and IIC melanomas, prognostic tools to identify patients at high risk of recurrence have become even more critical. This study aims to assess the effectiveness of machine-learning algorithms in predicting melanoma recurrence using clinical and histopathologic features from Electronic Health Records (EHRs). We collected 1720 early-stage melanomas: 1172 from the Mass General Brigham healthcare system (MGB) and 548 from the Dana-Farber Cancer Institute (DFCI). We extracted 36 clinicopathologic features and used them to predict the recurrence risk with supervised machine-learning algorithms. Models were evaluated internally and externally: (1) five-fold cross-validation of the MGB cohort; (2) the MGB cohort for training and the DFCI cohort for testing independently. In the internal and external validations, respectively, we achieved a recurrence classification performance of AUC: 0.845 and 0.812, and a time-to-event prediction performance of time-dependent AUC: 0.853 and 0.820. Breslow tumor thickness and mitotic rate were identified as the most predictive features. Our results suggest that machine-learning algorithms can extract predictive signals from clinicopathologic features for early-stage melanoma recurrence prediction, which will enable the identification of patients that may benefit from adjuvant immunotherapy.

No abstract

No abstract

No abstract

No abstract