RESUMEN
BACKGROUND: Understanding co-occurrence patterns and prognostic implications of immune-related adverse events is crucial for immunotherapy management. However, previous studies have been limited by sample size and generalisability. In this study, we leveraged a multi-institutional cohort and a population-level database to investigate co-occurrence patterns of and survival outcomes after multi-organ immune-related adverse events among recipients of immune checkpoint inhibitors. METHODS: In this retrospective study, we identified individuals who received immune checkpoint inhibitors between May 31, 2015, and June 29, 2022, from the Massachusetts General Hospital, Brigham and Women's Hospital, and Dana-Farber Cancer Institute (Boston, MA, USA; MGBD cohort), and between April 30, 2010, and Oct 11, 2021, from the independent US population-based TriNetX network. We identified recipients from all datasets using medication codes and names of seven common immune checkpoint inhibitors, and patients were excluded from our analysis if they had incomplete information (eg, diagnosis and medication records) or if they initiated immune checkpoint inhibitor therapy after Oct 11, 2021. Eligible patients from the MGBD cohort were then propensity score matched with recipients of immune checkpoint inhibitors from the TriNetX database (1:2) based on demographic, cancer, and immune checkpoint inhibitor characteristics to facilitate cohort comparability. We applied immune-related adverse event identification rules to identify patients who did and did not have immune-related adverse events in the matched cohorts. To reduce the likelihood of false positives, patients diagnosed with suspected immune-related adverse events within 3 months after chemotherapy were excluded. We performed pairwise correlation analyses, non-negative matrix factorisation, and hierarchical clustering to identify co-occurrence patterns in the MGBD cohort. We conducted landmark overall survival analyses for patient clusters based on predominant immune-related adverse event factors and calculated accompanying hazard ratios (HRs) and 95% CIs, focusing on the 6-month landmark time for primary analyses. We validated our findings using the TriNetX cohort. FINDINGS: We identified 15 246 recipients of immune checkpoint inhibitors from MGBD and 50 503 from TriNetX, of whom 13 086 from MGBD and 26 172 from TriNetX were included in our propensity score-matched cohort. Median follow-up durations were 317 days (IQR 113-712) in patients from MGBD and 249 days (91-616) in patients from TriNetX. After applying immune-related adverse event identification rules, 8704 recipients of immune checkpoint inhibitors were retained from MGBD, of whom 3284 (37·7%) had and 5420 (62·3%) did not have immune-related adverse events, and 18 162 recipients were retained from TriNetX, of whom 5538 (30·5%) had and 12 624 (69·5%) did not have immune-related adverse events. In both cohorts, positive pairwise correlations of immune-related adverse events were commonly observed. Co-occurring immune-related adverse events were decomposed into seven factors across organs, revealing seven distinct patient clusters (endocrine, cutaneous, respiratory, gastrointestinal, hepatic, musculoskeletal, and neurological). In the MGBD cohort, the patient clusters that predominantly had endocrine (HR 0·53 [95% CI 0·40-0·70], p<0·0001) and cutaneous (0·61 [0·46-0·81], p=0·0007) immune-related adverse events had favourable overall survival outcomes at the 6-month landmark timepoint, while the other clusters either had unfavourable (respiratory: 1·60 [1·25-2·03], p=0·0001) or neutral survival outcomes (gastrointestinal: 0·86 [0·67-1·10], p=0·23; musculoskeletal: 0·97 [0·78-1·21], p=0·78; hepatic: 1·20 [0·91-1·59], p=0·19; and neurological: 1·30 [0·97-1·74], p=0·074). Similar results were found in the TriNetX cohort (endocrine: HR 0·75 [95% CI 0·60-0·93], p=0·0078; cutaneous: 0·62 [0·48-0·82], p=0·0007; respiratory: 1·21 [1·00-1·46], p=0·044), except for the neurological cluster having unfavourable (rather than neutral) survival outcomes (1·30 [1·06-1·59], p=0·013). INTERPRETATION: Reliably identifying the immune-related adverse event cluster to which a patient belongs can provide valuable clinical information for prognosticating outcomes of immunotherapy. These insights can be leveraged to counsel patients on the clinical impact of their individual constellation of immune-related adverse events and ultimately develop more personalised surveillance and mitigation strategies. FUNDING: US National Institutes of Health.
Asunto(s)
Inhibidores de Puntos de Control Inmunológico , Neoplasias , Humanos , Inhibidores de Puntos de Control Inmunológico/efectos adversos , Estudios Retrospectivos , Femenino , Masculino , Persona de Mediana Edad , Anciano , Neoplasias/tratamiento farmacológico , Neoplasias/inmunologíaRESUMEN
BACKGROUND: Cutaneous immune-related adverse events (cirAEs) are the most common toxicities to occur in the setting of immune checkpoint inhibitor (ICI) therapy. Identifying patients who are at increased risk of developing cirAEs may improve quality of life and outcomes. OBJECTIVES: To investigate the influence of cancer type and histology on the development of cirAEs in the setting of ICI therapy and survival outcomes. METHODS: This retrospective cohort study included patients recruited between 1 December 2011 and 30 October 2020. They received ICI from 2011 to 2020 with follow-up of outcomes through October 2021. We identified 3668 recipients of ICI therapy who were seen at Massachusetts General Brigham and Dana-Farber. Of these, 669 developed cirAEs. Records that were incomplete or categories of insufficient sample size were excluded from the study cohort. Multivariate Cox proportional hazards models were used to investigate the impact of cancer organ system and histology on cirAE development, after adjusting for demographics, Charlson Comorbidity Index, ICI type, cancer stage at ICI initiation, and year of ICI initiation. Time-varying Cox proportional hazards modelling was used to examine the impact of cirAE development on mortality. RESULTS: Compared with other nonepithelial cancers (neuroendocrine, leukaemia, lymphoma, myeloma, sarcoma and central nervous system malignancies), cutaneous squamous cell carcinoma [cSCC; hazard ratio (HR) 3.57, P < 0.001], melanoma (HR 2.09, P < 0.001), head and neck adenocarcinoma (HR 2.13, P = 0.009), genitourinary transitional cell carcinoma (HR 2.15, P < 0.001) and genitourinary adenocarcinoma (HR 1.53, P = 0.037) were at significantly higher risk of cirAEs in multivariate analyses. The increased risk of cirAEs translated into an adjusted survival benefit for melanoma (HR 0.37, P < 0.001) and cSCC (HR 0.51, P = 0.011). CONCLUSIONS: The highest rate of cirAEs and subsequent survival benefits were observed in cutaneous malignancies treated with ICI therapies. This study improves our understanding of patients who are at highest risk of developing cirAEs and would, therefore, benefit from appropriate counselling and closer monitoring by their oncologists and dermatologists throughout their ICI therapy. Limitations include its retrospective nature and cohort from one geography.
Cutaneous immune-related adverse events (cirAEs) are the most common complications to occur for oncology patients treated with immune checkpoint inhibitors (ICIs). cirAEs can lead to increased use of healthcare resources and significant morbidity. Identifying patients who are at increased risk of developing cirAEs may improve quality of life and outcomes. In this study, we aimed to investigate the influence of cancer organ system and histology on the development of cirAEs and survival outcomes. To do this, we included a cohort of patients retrospectively between 1 December 2011 and 30 October 2020. We identified 3668 ICI recipients who were seen at Massachusetts General Brigham and Dana-Farber in Boston, Massachusetts. Of these, 669 people developed cirAEs. Multivariate Cox proportional hazards models were used to investigate the impact of cancer organ system and histology on cirAE development, after adjusting for demographics, Charlson Comorbidity Index, ICI type, cancer stage at ICI start, and year of ICI initiation. Time-varying Cox proportional hazards modelling was used to examine the impact of cirAE development on mortality. We found that, compared with other nonepithelial cancers, patients with cutaneous squamous cell carcinoma (cSCC) and melanoma were at significantly higher risk of cirAEs. The increased risk of cirAEs translated into an adjusted survival benefit for melanoma and cSCC. This study improves our understanding of patients who are at highest risk of developing cirAEs those with melanoma and cSCC. Therefore, many patients could benefit from appropriate counselling and close monitoring by their oncologists and dermatologists throughout ICI therapy.
Asunto(s)
Inhibidores de Puntos de Control Inmunológico , Humanos , Masculino , Femenino , Estudios Retrospectivos , Persona de Mediana Edad , Anciano , Inhibidores de Puntos de Control Inmunológico/efectos adversos , Neoplasias/tratamiento farmacológico , Neoplasias/patología , Neoplasias/mortalidad , Neoplasias/inmunología , Neoplasias/terapia , Erupciones por Medicamentos/etiología , Erupciones por Medicamentos/patología , Erupciones por Medicamentos/epidemiología , Neoplasias Cutáneas/patología , Neoplasias Cutáneas/mortalidad , Neoplasias Cutáneas/inmunología , Neoplasias Cutáneas/tratamiento farmacológico , AdultoRESUMEN
BACKGROUND: The recent expansion of immunotherapy for stage IIB/IIC melanoma highlights a growing clinical need to identify patients at high risk of metastatic recurrence and, therefore, most likely to benefit from this therapeutic modality. OBJECTIVE: To develop time-to-event risk prediction models for melanoma metastatic recurrence. METHODS: Patients diagnosed with stage I/II primary cutaneous melanoma between 2000 and 2020 at Mass General Brigham and Dana-Farber Cancer Institute were included. Melanoma recurrence date and type were determined by chart review. Thirty clinicopathologic factors were extracted from electronic health records. Three types of time-to-event machine-learning models were evaluated internally and externally in the distant versus locoregional/nonrecurrence prediction. RESULTS: This study included 954 melanomas (155 distant, 163 locoregional, and 636 1:2 matched nonrecurrences). Distant recurrences were associated with worse survival compared to locoregional/nonrecurrences (HR: 6.21, P < .001) and to locoregional recurrences only (HR: 5.79, P < .001). The Gradient Boosting Survival model achieved the best performance (concordance index: 0.816; time-dependent AUC: 0.842; Brier score: 0.103) in the external validation. LIMITATIONS: Retrospective nature and cohort from one geography. CONCLUSIONS: These results suggest that time-to-event machine-learning models can reliably predict the metastatic recurrence from localized melanoma and help identify high-risk patients who are most likely to benefit from immunotherapy.
Asunto(s)
Melanoma , Neoplasias Cutáneas , Humanos , Melanoma/patología , Neoplasias Cutáneas/patología , Estudios Retrospectivos , Recurrencia Local de Neoplasia/epidemiología , Recurrencia Local de Neoplasia/patologíaRESUMEN
BACKGROUND: Diverse microbiome communities drive biogeochemical processes and evolution of animals in their ecosystems. Many microbiome projects have demonstrated the power of using metagenomics to understand the structures and factors influencing the function of the microbiomes in their environments. In order to characterize the effects from microbiome composition for human health, diseases, and even ecosystems, one must first understand the relationship of microbes and their environment in different samples. Running machine learning model with metagenomic sequencing data is encouraged for this purpose, but it is not an easy task to make an appropriate machine learning model for all diverse metagenomic datasets. RESULTS: We introduce MegaR, an R Shiny package and web application, to build an unbiased machine learning model effortlessly with interactive visual analysis. The MegaR employs taxonomic profiles from either whole metagenome sequencing or 16S rRNA sequencing data to develop machine learning models and classify the samples into two or more categories. It provides various options for model fine tuning throughout the analysis pipeline such as data processing, multiple machine learning techniques, model validation, and unknown sample prediction that can be used to achieve the highest prediction accuracy possible for any given dataset while still maintaining a user-friendly experience. CONCLUSIONS: Metagenomic sample classification and phenotype prediction is important particularly when it applies to a diagnostic method for identifying and predicting microbe-related human diseases. MegaR provides various interactive visualizations for user to build an accurate machine-learning model without difficulty. Unknown sample prediction with a properly trained model using MegaR will enhance researchers to identify the sample property in a fast turnaround time.
Asunto(s)
Aprendizaje Automático , Metagenoma , Metagenómica , Humanos , Fenotipo , ARN Ribosómico 16S/genéticaRESUMEN
Ribosomal protein (RP) genes encode structural components of ribosomes, the cellular machinery for protein synthesis. A single functional copy has been maintained in most of 78-80 RP families in animals due to evolutionary constraints imposed by gene dosage balance. Some fungal species have maintained duplicate copies in most RP families. The mechanisms by which the RP genes were duplicated and maintained and their functional significance are poorly understood. To address these questions, we identified all RP genes from 295 fungi and inferred the timing and nature of gene duplication events for all RP families. We found that massive duplications of RP genes have independently occurred by different mechanisms in three distantly related lineages: budding yeasts, fission yeasts, and Mucoromycota. The RP gene duplicates in budding yeasts and Mucoromycota were mainly created by whole genome duplication events. However, duplicate RP genes in fission yeasts were likely generated by retroposition, which is unexpected considering their dosage sensitivity. The sequences of most RP paralogs have been homogenized by repeated gene conversion in each species, demonstrating parallel concerted evolution, which might have facilitated the retention of their duplicates. Transcriptomic data suggest that the duplication and retention of RP genes increased their transcript abundance. Physiological data indicate that increased ribosome biogenesis allowed these organisms to rapidly consume sugars through fermentation while maintaining high growth rates, providing selective advantages to these species in sugar-rich environments.
Asunto(s)
Hongos/metabolismo , Duplicación de Gen , Proteínas Ribosómicas/genética , Evolución Molecular , Proteínas Fúngicas/genética , Hongos/clasificación , Hongos/genética , Conversión Génica , Dosificación de Gen , Filogenia , Especificidad de la EspecieAsunto(s)
Antígeno B7-H1 , Enfermedades del Tejido Conjuntivo , Inhibidores de Puntos de Control Inmunológico , Receptor de Muerte Celular Programada 1 , Humanos , Femenino , Masculino , Persona de Mediana Edad , Receptor de Muerte Celular Programada 1/antagonistas & inhibidores , Enfermedades del Tejido Conjuntivo/tratamiento farmacológico , Enfermedades del Tejido Conjuntivo/mortalidad , Enfermedades del Tejido Conjuntivo/inmunología , Antígeno B7-H1/antagonistas & inhibidores , Anciano , Inhibidores de Puntos de Control Inmunológico/uso terapéutico , Inhibidores de Puntos de Control Inmunológico/efectos adversos , Enfermedades Autoinmunes/tratamiento farmacológico , Enfermedades Autoinmunes/mortalidad , Adulto , Estudios de CohortesAsunto(s)
Neoplasias , Piel , Humanos , Neoplasias/tratamiento farmacológico , Administración CutáneaRESUMEN
BACKGROUND: Chromosomal rearrangements have been shown to facilitate speciation through creating a barrier of gene flow. However, it is not known whether heterogeneous rates of chromosomal rearrangement at the genome scale contributed to the huge disparity of species richness among different groups of organisms, which is one of the most remarkable and pervasive patterns on Earth. The largest fungal phylum Ascomycota is an ideal study system to address this question because it comprises three subphyla (Saccharomycotina, Taphrinomycotina, and Pezizomycotina) whose species numbers differ by two orders of magnitude (59,000, 1000, and 150 respectively). RESULTS: We quantified rates of genome rearrangement for 71 Ascomycota species that have well-assembled genomes. The rates of inter-species genome rearrangement, which were inferred based on the divergence rates of gene order, are positively correlated with species richness at both ranks of subphylum and class in Ascomycota. This finding is further supported by our quantification of intra-species rearrangement rates based on paired-end genome sequencing data of 216 strains from three representative species, suggesting a difference of intrinsic genome instability among Ascomycota lineages. Our data also show that different rates of imbalanced rearrangements, such as deletions, are a major contributor to the heterogenous rearrangement rates. CONCLUSIONS: Various lines of evidence in this study support that a higher rate of rearrangement at the genome scale might have accelerated the speciation process and increased species richness during the evolution of Ascomycota species. Our findings provide a plausible explanation for the species disparity among Ascomycota lineages, which will be valuable to unravel the underlying causes for the huge disparity of species richness in various taxonomic groups.
Asunto(s)
Ascomicetos/genética , Genoma Fúngico/genética , Elementos Transponibles de ADN/genética , Evolución Molecular , Orden Génico , Filogenia , Duplicaciones Segmentarias en el Genoma/genéticaAsunto(s)
Melanoma , Ganglio Linfático Centinela , Neoplasias Cutáneas , Humanos , Biopsia del Ganglio Linfático Centinela , Melanoma/diagnóstico , Melanoma/cirugía , Melanoma/patología , Neoplasias Cutáneas/cirugía , Neoplasias Cutáneas/patología , Ganglios Linfáticos/patología , Ganglio Linfático Centinela/patología , Escisión del Ganglio LinfáticoRESUMEN
Prurigo nodularis (PN) is a chronic inflammatory skin disease that is associated with variability in peripheral blood eosinophil levels and response to T-helper 2 targeted therapies (Th2). Our objective was to determine whether circulating immune profiles with respect to type 2 inflammation differ by race and peripheral blood eosinophil count. Plasma from 56 PN patients and 13 matched healthy controls was assayed for 54 inflammatory biomarkers. We compared biomarker levels between PN and HCs, among PN patients based on absolute eosinophil count, and across racial groups in PN. Eleven biomarkers were elevated in PN versus HCs including interleukin (IL)-12/IL-23p40, tumor necrosis factor-alpha (TNF-α), Thymic stromal lymphopoietin (TSLP), and macrophage-derived chemokine (MDC/CCL22). Additionally, PN patients with AEC > 0.3 K cells/µL had higher Th2 markers (eotaxin, eotaxin-3, TSLP, MCP-4/CCL13), and African American PN patients had lower eosinophils, eotaxin, and eotaxin-3 versus Caucasian and Asian PN patients (p < 0.05 for all). Dupilumab responders had higher AEC (p < 0.01), were more likely to be Caucasian (p = 0.02) or Asian (p = 0.05) compared to African Americans, and more often had a history of atopy (p = 0.08). This study suggests that blood AEC > 0.3 K and Asian and Caucasian races are associated with Th2 skewed circulating immune profiles and response to Th2 targeted therapies.
Asunto(s)
Citocinas , Prurigo , Humanos , Quimiocina CCL26 , Prurigo/tratamiento farmacológico , Linfopoyetina del Estroma Tímico , Inflamación , BiomarcadoresRESUMEN
Importance: Objectively determining disease progression in craniofacial morphea (CM) is challenging, as clinical findings of disease activity are often lacking. Objective: To evaluate the utility of 3-dimensional (3D) stereophotogrammetry in detecting disease progression in CM over time. Design, Setting, and Participants: This prospective cohort study included 27 pediatric and adult patients with CM from 2 hospitals in Boston (Boston Children's Hospital and Brigham & Women's Hospital) consecutively enrolled from April 1, 2019, to March 1, 2023. Review of 3D stereophotogrammetry images and data analysis occurred from March 1 to April 1, 2023. Main Outcomes and Measures: Clinical and 3D stereophotogrammetry assessments were performed at 2- to 12-month intervals, depending on the clinical context. The 3D stereophotogrammetry images were then qualitatively rated as demonstrating no progression or definitive progression by an expert (board-certified plastic craniofacial surgeon) and nonexpert (board-certified dermatologist) in 3D stereophotogrammetry. In addition, κ coefficients were calculated for interrater reliability. Results: Of 27 patients with CM (19 female; median age, 14 [range, 5-40] years) and 3D stereophotogrammetry images obtained from a minimum of 2 time points (median, 4 [range, 2-10] images) spaced a median of 3 (range, 2-12) months apart, 10 experienced progression of their disease based on clinical assessments performed during the study period. In all cases in which clinical progression was favored, blinded qualitative assessment of 3D stereophotogrammetry images also favored progression with substantial interrater reliability (κ = 0.80 [95% CI, 0.61-0.99]). Furthermore, review of 3D stereophotogrammetry detected occult progression of asymmetry not noted on clinical examination in 3 additional patients. Conclusions and Relevance: In this prospective cohort study, blinded assessment of sequential 3D stereophotogrammetry images in patients with CM not only corroborated clinical assessment of disease progression but also detected occult progression of facial asymmetry not appreciable on clinical examination alone. Therefore, 3D stereophotogrammetry may serve as a useful adjunct to clinical examination of patients with CM over time. Future investigations are warranted to validate 3D stereophotogrammetry as an outcome measure in CM.
Asunto(s)
Esclerodermia Localizada , Adulto , Humanos , Femenino , Niño , Adolescente , Reproducibilidad de los Resultados , Estudios Prospectivos , Esclerodermia Localizada/diagnóstico por imagen , Imagenología Tridimensional/métodos , Fotogrametría/métodos , Progresión de la EnfermedadRESUMEN
Prurigo nodularis (PN) is a chronic inflammatory skin disease that disproportionately affects African Americans and is characterized by pruritic skin nodules of unknown etiology. Little is known about genetic alterations in PN pathogenesis, especially relating to somatic events which are often implicated in inflammatory conditions. We thus performed whole-exome sequencing on 54 lesional and nonlesional skin biopsies from 17 PN patients and 10 atopic dermatitis (AD) patients for comparison. Somatic mutational analysis revealed that PN lesional skin harbors pervasive somatic mutations in fibrotic, neurotropic, and cancer-associated genes. Nonsynonymous mutations were most frequent in NOTCH1 and the Notch signaling pathway, a regulator of cellular proliferation and tissue fibrosis, and NOTCH1 mutations were absent in AD. Somatic copy-number analysis, combined with expression data, showed that recurrently deleted and downregulated genes in PN lesional skin are associated with axonal guidance and extension. Follow-up immunofluorescence validation demonstrated increased NOTCH1 expression in PN lesional skin fibroblasts and increased Notch signaling in PN lesional dermis. Finally, multi-center data revealed a significantly increased risk of NOTCH1-associated diseases in PN patients. In characterizing the somatic landscape of PN, we uncover novel insights into its pathophysiology and identify a role for dysregulated Notch signaling in PN.
RESUMEN
Prurigo nodularis (PN) is an understudied inflammatory skin disease characterized by pruritic, hyperkeratotic nodules. Identifying the genetic factors underlying PN could help to better understand its etiology and guide the development of therapies. In this study, we developed a polygenic risk score that predicts a diagnosis of PN (OR = 1.41, P = 1.6 × 10-5) in two independent and continentally distinct populations. We also performed GWASs, which uncovered genetic variants associated with PN, including one near PLCB4 (rs6039266: OR = 3.15, P = 4.8 × 10-8) and others near TXNRD1 (rs34217906: OR = 1.71, P = 6.4 × 10-7; rs7134193: OR = 1.57, P = 1.1 × 10-6). Finally, we discovered that Black patients have over a two-times greater genetic risk of developing PN (OR = 2.63, P = 7.8 × 10-4). Combining the polygenic risk score and self-reported race together was significantly predictive of PN (OR = 1.32, P = 4.7 × 10-3). Strikingly, this association was more significant with race than after adjusting for genetic ancestry. Because race is a sociocultural construct and not a genetically bound category, our findings suggest that genetics, environmental influence, and social determinants of health likely affect the development of PN and may contribute to clinically observed racial disparities.
Asunto(s)
Dermatitis , Prurigo , Humanos , Población Negra , Dermatitis/etnología , Dermatitis/genética , Predisposición Genética a la Enfermedad , Prurigo/etnología , Prurigo/genética , Factores de RiesgoRESUMEN
Syphilitic osteitis is one of the rare and often under-reported complications of early syphilis. Recognizing this entity is important as it may mimic other conditions like multiple myeloma, lymphoma, or metastatic malignancies. Misdiagnosis and delayed management can lead to irreversible destructive lesions. We herein report a case of calvaria syphilitic osteitis that was initially investigated for possible lymphoma and later diagnosed as secondary syphilis.
RESUMEN
Hemophagocytic lymphohistiocytosis (HLH) is a life-threatening condition characterized by an exaggerated but dysregulated immune response resulting in hyperinflammation, with a potential for progression to multiple organ dysfunction and failure. Infectious diseases, inflammatory disorders, malignancies and immunodeficiency syndromes are known triggers of HLH in adults. The present study reported the case of a middle-aged man with HLH triggered by B-cell lymphoma who was successfully treated with dexamethasone; etoposide, prednisone, vincristine, cyclophosphamide, hydroxy-doxorubicin and rituximab chemotherapy; and multiple intrathecal methotrexate with a good outcome.
RESUMEN
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.