Context
Receiving a personalized forecast for lip squamous cell carcinoma (LSCC) is crucial. This study aimed to develop predictive models for postoperative prognosis prediction in LSCC patients.
Methods
Two medical institutes provided 136 postoperative LSCC patients diagnosed between April 2015 and July 2022. The patients were randomly assigned to training and validation groups at a 7:3 ratio. Based on the results of univariate and multivariate Cox regression analyses, two nomograms were made to predict overall survival (OS) and progression-free survival (PFS) in LSCC patients who had surgery.
Results
According to univariate and multivariate Cox regression analyses, a lower pathological grade, being older than 70 years, having a higher T-stage, positive lymph node metastasis (LNM), perineural invasion (PNI), vascular invasion (VI), and positive surgical margin were all independent predictors of a lower OS. The independent predictors of worse PFS included higher pathological grade, age ≥ 70 years, higher T-stage, LNM, PNI, and positive surgical margins. Two nomograms were created based on the data above to forecast the 3- and 5-year OS and PFS rates in patients with LSCC. In the training group, the C-indices for the OS and PFS nomograms were 0.865 and 0.801, respectively, whereas those in the validation group were 0.915 and 0.815, respectively. The calibration curves demonstrated a reasonable degree of agreement between the observed survival rates and the predictions.
Conclusions
Curve analysis revealed superior performance of the nomogram compared to other related predictors. Kaplan-Meier curves showed statistical distinctions between the high-, medium-, and low-risk groups. In summary, With a good clinically personalized applicability value, this retrospective study developed and validated two nomograms for predicting OS and PFS in patients with postoperative LSCC.