Prognostic model for three-year postoperative local recurrence in cutaneous squamous cell carcinoma: a Chinese
Yuanhong Liu1, Suzheng Zheng1, Hao Song1
1Hospital for Skin Diseases, Institute of Dermatology, Chinese Academy of Medical Sciences & Peking Union Medical College, Nanjing, China.
Background:
Cutaneous squamous cell carcinoma (cSCC) carries a risk of postoperative local recurrence, with nearly 90% of events occurring within 3 years; however, prognostic tools tailored to Asian populations remain limited. This study aimed to identify independent risk factors for postoperative local recurrence and develop a nomogram to predict individualized 1-, 2-, and 3-year recurrence probabilities in patients with cSCC.
Methods:
Clinicopathological data from 603 patients with cSCC who underwent surgical treatment alone at four Chinese centers were retrospectively analyzed. Patients were randomly divided into a training cohort and a validation cohort at a ratio of 7:3. Univariable and multivariable Cox proportional hazards regression analyses were performed in the training cohort to identify independent predictors of postoperative local recurrence. A nomogram was then constructed based on these independent risk factors. The prognostic performance of the model was evaluated using time-dependent receiver operating characteristic curves, calibration curves, and decision curve analysis. Finally, the total nomogram score was calculated for each patient, and all patients were categorized into low-, intermediate-, and high-risk groups based on the optimal cutoff values. Kaplan-Meier analysis was then performed to compare recurrence-free survival among the different risk groups.
Results:
A total of 603 patients were included, with 422 in the training cohort and 181 in the validation cohort. During follow-up, 77 recurrence events were observed. Multivariable Cox regression identified age, tumor size, tumor thickness, histologic differentiation, regional stage, and AJCC stage as independent predictors of recurrence. These predictors were incorporated into a nomogram for individualized recurrence risk estimation. The model demonstrated good discrimination, with area under the curve values ranging from 0.759 to 0.869, along with strong calibration and favorable clinical utility in decision curve analysis. Based on the total nomogram score, patients were stratified into low-risk (<99.0), intermediate-risk (99.0-143.9), and high-risk (≥144.0) groups, with significant differences in recurrence-free survival among the three groups (log-rank p < 0.0001).
Conclusions:
This nomogram provides a practical tool for individualized prognostic assessment and may support individualized clinical decision-making for 1-, 2-, and 3-year outcomes 1.
