Operational classification of cutaneous squamous cell carcinomas based on unsupervised clustering of real cases by experts.

BACKGROUND: There is currently no staging system for cutaneous squamous cell carcinoma (cSCC) that is adapted to decision-making and universally used. Experts have unconscious ability to simplify the heterogeneity of clinical situations into a few relevant groups to drive their therapeutic decisions. Therefore, we have used unsupervised clustering of real cases by experts to generate an operational classification of cSCCs, an approach that was successful for basal cell carcinomas. OBJECTIVE: To generate a consensual and operational classification of cSCCs. METHOD: Unsupervised independent clustering of 248 cases of cSCCs considered difficult-to-treat. Eighteen international experts from different specialties classified these cases into what they considered homogeneous clusters useful for management, each with freedom regarding clustering criteria. Convergences and divergences between clustering were analysed using a similarity matrix, the K-mean approach and the average silhouette method. Mathematical modelling was used to look for the best consensual clustering. The operability of the derived classification was validated on 23 new practitioners. RESULTS: Despite the high heterogeneity of the clinical cases, a mathematical consensus was observed. It was best represented by a partition into five clusters, which appeared a posteriori to describe different clinical scenarios. Applicability of this classification was shown by a good concordance (94%) in the allocation of cases between the new practitioners and the 18 experts. An additional group of easy-to-treat cSCC was included, resulting in a six-group final classification: easy-to-treat/complex to treat due to tumour and/or patient characteristics/multiple/locally advanced/regional disease/visceral metastases. CONCLUSION: Given the methodology based on the convergence of unguided intuitive clustering of cases by experts, this new classification is relevant for clinical practice. It does not compete with staging systems, but they may complement each other, whether the objective is to select the best therapeutic approach in tumour boards or to design homogeneous groups for trials.

Nomogram for predicting survival after first-line anti-PD-1-based immunotherapy in unresectable stage IV melanoma: a multicenter international study.

BACKGROUND: The introduction of anti-programmed cell death protein 1 (PD-1) immunotherapy has revolutionized the treatment landscape for melanoma, enhancing both response rates and survival outcomes in patients with advanced stages of the disease. Despite these remarkable advances, a noteworthy subset of patients (40%-60%) does not derive advantage from this therapeutic approach. This study aims to identify key predictive factors and create a user-friendly predictive nomogram for stage IV melanoma patients receiving first-line anti-PD-1-based immunotherapy, improving treatment decisions. MATERIALS AND METHODS: In this retrospective study, we included patients with unresectable stage IV melanoma who received first-line treatment with either anti-PD-1 monotherapy or anti-PD-1 plus anti-cytotoxic T-lymphocyte associated protein 4 between 2014 and 2018. We documented clinicopathological features and blood markers upon therapy initiation. By employing the random survival forest model and backward variable selection of the Cox model, we identified variables associated with progression-free survival (PFS) after the first-line anti-PD-1-based treatment. We developed and validated a predictive nomogram for PFS utilizing the identified variables. We assessed calibration and discrimination performance metrics as part of the evaluation process. RESULTS: The study involved 719 patients, divided into a training cohort of 405 (56%) patients and a validation cohort of 314 (44%) patients. We combined findings from the random survival forest and the Cox model to create a nomogram that incorporates the following factors: lactate dehydrogenase (LDH), S100, melanoma subtype, neutrophil-to-lymphocyte ratio (NLR), body mass index, type of immune checkpoint inhibitor, and presence of liver or brain metastasis. The resultant model had a C-index of 0.67 in the training cohort and 0.66 in the validation cohort. Performance remained in different patient subgroups. Calibration analysis revealed a favorable correlation between predicted and actual PFS rates. CONCLUSIONS: We developed and validated a predictive nomogram for long-term PFS in patients with unresectable stage IV melanoma undergoing first-line anti-PD-1-based immunotherapy.