Stevenson, Mary L.

ImportanceThere exists substantial heterogeneity in outcomes within T stages for patients with cutaneous squamous cell carcinoma (cSCC).ObjectiveTo determine whether a customized generative pretrained transformer model, trained on a comprehensive dataset with more than 1 trillion parameters and equipped with relevant focused context and retrieval augmented generation (RAG), could excel in aggregating and interpreting vast quantities of data to develop a novel class-based risk stratification system that outperforms the current standards.Design, Setting, and ParticipantsTo build the RAG knowledge base, a systematic review of the literature was conducted that addressed risk factors for poor outcomes in cSCC. Using the RAG-enabled generative pretrained transformer (GPT) model, we developed a novel class-based risk stratification system that assigned point values for risk factors, culminating in a GPT-based prognostication system called the artificial intelligence–derived risk score (AIRIS). The system’s performance was validated on a combined prospective and retrospective cohort of 2379 primary cSCC tumors (1996-2023) with at least 36 months of follow-up, against Brigham and Women’s Hospital (BWH) and American Joint Committee on Cancer Staging Manual, eighth edition (AJCC8) systems in stratifying risk for locoregional recurrence (LR), nodal metastasis (NM), distant metastasis (DM), and disease-specific death (DSD).Main Outcomes and MeasuresPerformance metrics evaluated included distinctiveness, homogeneity, and monotonicity, as defined by the AJCC8, as well as sensitivity, specificity, positive predictive value, negative predictive value, accuracy, the area under the receiver operating characteristic curve, and concordance.ResultsThe median age at diagnosis was 73 (IQR, 64-81) years, with 38.5% female patients and 61.5% male patients. The AIRIS prognostication system demonstrated superior sensitivity across all outcomes (LR, 49.1%; NM, 73.7%; DM, 82.5%; and DSD, 72.2%) and the highest area under the receiver operating characteristic curve values (LR, 0.69; NM, 0.81; DM, 0.85; and DSD, 0.80), indicating significantly enhanced discriminative capability compared with the BWH and AJCC8 systems. While all systems were comparably distinctive, the AIRIS prognostication system consistently demonstrated the lowest proportion of tumors exhibiting poor outcomes in low-risk categories, suggesting its improved homogeneity and monotonicity.Conclusions and RelevanceThe results of this diagnostic study suggest that the AIRIS system outperforms the existing BWH and AJCC8 prognostication systems, potentially providing a more effective tool for predicting poor outcomes in cSCC. This study illustrates the potential of large language models in refining prognostic tools, offering implications for treating patients with cancer.

ImportanceCutaneous squamous cell carcinoma (CSCC) is the second most common malignant disease in the US. Although it typically carries a good prognosis, a subset of CSCCs are highly aggressive, carrying regional and distant metastatic potential. Due to its high incidence, this aggressive subset is responsible for considerable mortality, with an overall annual mortality estimated to equal or even surpass melanoma. Despite this morbidity, CSCC is excluded from national cancer registries, making it difficult to study its epidemiology and outcomes. Therefore, the bulk of the CSCC literature is composed of single-center and multi-institutional retrospective cohort analyses. Given variations in reporting measures and analyses in these studies, interpretability between studies and the ability to pool results are limited.ObjectiveTo define standardized reporting measures for retrospective CSCC studies.FindingsAn expert panel was convened to determine standardized guidelines for recording and analyzing retrospective CSCC data. A total of 13 dermatologists and dermatologic surgeons with more than 5 years of posttraining experience and considerable experience with performing CSCC outcomes research were recruited to the panel. Consensus recommendations were achieved for CSCC retrospective study reporting measures, definitions, and analyses.Conclusions and RelevanceThe recommendations in this report present the potential to standardize future CSCC retrospective studies. With such standardization, future work may have greater interstudy interpretability and allow for pooled analyses.

BACKGROUND The management of skin cancers has evolved with the development of Mohs micrographic surgery and a greater emphasis on surgical training within dermatology. It is unclear whether these changes have translated into innovations and contributions to the reconstructive literature. OBJECTIVE To assess contributions from each medical specialty to the cutaneous head and neck oncologic reconstructive literature. METHODS The authors conducted a systematic review of the head and neck reconstructive literature from 2000 through 2015 based on a priori search terms relating to suture technique, linear closure, advancement, rotation, transposition and interpolation flaps, and identified the specialty of the senior authors. RESULTS The authors identified 74,871 articles, of which 1,319 were relevant. Under suture technique articles, the senior authors were primarily dermatologists (58.2%) and plastic surgeons (20.3%). Under linear closure, the authors were dermatologists (48.1%), plastic surgeons (22.2%), and otolaryngologists (20.4%). Under advancement and rotation flaps, the senior authors were plastic surgeons (40.5%, 38.9%), dermatologists (38.1%, 34.2%), and otolaryngologists (14.4%, 21.6%). Under transposition and interpolation flaps, the senior authors were plastic surgeons (47.3%, 39.4%), dermatologists (32.3%, 27.0%), and otolaryngologists (15.3%, 23.4%). CONCLUSION The primary specialties contributing to the cutaneous head and neck reconstructive literature are plastic surgery, dermatology, and otolaryngology.