31 October 2022 in npj Precision Oncology
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.
11 January 2024 in American Journal of Clinical Pathology
Cowden syndrome (CS) is a multisystem disease with an elevated lifetime risk of internal malignancy. We aim to assess the role of PTEN immunostain as a screening test for CS in a variety of common CS-associated neoplasms, with a particular focus on cutaneous tumors. We retrospectively searched for patients meeting criteria for CS and/or demonstrating germline PTEN mutation from 2008 to 2022. We then performed PTEN immunostains on tumors of these patients as well as control cases. Our study included 30 patients with CS who had a total of 25 CS-associated malignancies (13 thyroid, 8 breast, and 4 endometrial carcinomas). Specifically, there were 11 patients with biopsy-confirmed CS-associated cutaneous neoplasms, including 1 patient with multiple trichilemmomas and 3 with multiple sclerotic fibromas. In total, 45 CS-associated tumors (6 trichilemmomas, 7 sclerotic fibromas, 5 thyroid carcinomas, 18 adenomatous thyroid nodules, 6 breast carcinomas, and 3 endometrial carcinomas) and 31 non-CS cases (9 trichilemmomas, 5 sclerotic fibromas, 8 adenomatous thyroid nodules, and 3 thyroid, 3 breast, and 3 endometrial carcinomas) were available for PTEN immunohistochemical staining. PTEN expression was lost in 43 (96%) of 45 CS-associated lesions and retained in 30 (97%) of 31 sporadic tumors. The overall sensitivity and specificity of PTEN loss of expression as a screening test for CS were 96% and 97%, respectively. PTEN immunohistochemistry on CS-associated tumors, especially trichilemmomas, can serve as a readily accessible and cost-effective screening test for CS.
11 March 2026 in ESMO Open
Anti-programmed cell death protein 1 (PD-1) therapy is the cornerstone for managing advanced cutaneous squamous-cell carcinoma (CSCC). Nevertheless, many patients experience treatment failure. Limited data exist regarding outcomes following anti-PD-1 failure. This study investigates progression patterns and clinical outcomes in CSCC patients post-anti-PD-1 therapy. We conducted a retrospective analysis of CSCC patients treated with anti-PD-1 at the Dana-Farber Cancer Institute. We evaluated clinicopathological features and outcomes. Overall survival (OS) and event-free survival (EFS) were estimated using the Kaplan-Meier method. CSCC-specific mortality was estimated using cumulative incidence. Multivariate regression was used to investigate prognostic factors. Among 238 patients receiving immunotherapy, 72 exhibited anti-PD-1 failure with a median age of 72 years; 22% were female and 29% were immunosuppressed. Median follow-up was 23 months [95% confidence interval (CI) 19-37 months] and median duration of immunotherapy was 3 months (range 1-44 months). Progression after anti-PD-1 failure occurred as local (21%), locoregional (37%), or distant metastatic (42%). Primary resistance was observed in 62.5%, while 37.5% developed secondary resistance. Patients with primary resistance exhibited a significantly lower tumor mutational burden (TMB) (P = 0.03). Among 61 patients receiving subsequent treatment, cetuximab-based therapy and local treatments each were administered in 31%. Complete and partial responses were achieved in 5% and 25%, respectively, while 6% had stable disease, 28% progressed, and 36% were non-assessable. Median OS was 44.1 months (95% CI 17.4 months-not achieved) and median EFS2 (time from starting subsequent treatment post-anti-PD-1 until recurrence, progression, or death) was 7.1 months (95% CI 3.2-12.2 months). CSCC-specific death was 50% at 5 years (95% CI 30% to 67%). Prior chemotherapy was associated with poorer OS (hazard ratio 2.87, 95% CI 1.03-7.98, P = 0.04). In this cohort of patients with CSCC, distant metastatic and locoregional progression were the predominant patterns following anti-PD-1 failure, with lower TMB linked to primary resistance. Prior chemotherapy was associated with poorer OS. Select patients benefited from subsequent treatments, including cetuximab and local therapy.
5 March 2023 in Clinical Biochemistry
10 April 2025 in Laboratory Investigation
2 February 2026 in British Journal of Dermatology
Using a large multi-institutional cohort of patients with thin melanomas, our study identified that while predictive of improved prognosis in later stage disease, the presence of tumour-infiltrating lymphocytes is associated with poorer outcomes in patients with thin melanomas. Increasing mitotic rate, Charlson Comorbidity Index, absence of radial growth phase and self-pay/no insurance are also significantly associated with poorer outcomes, while tumour thickness, historically one of the most important prognostic features, is not independently associated with recurrence or mortality. This study elucidates novel risk factors that could prognosticate recurrence or mortality among thin melanomas and can be used to enable more personalized surveillance and therapeutic strategies.