Development and validation of nomograms for predicting survival outcomes in patients with T1-2N1 breast cancer to identify those who could not benefit from postmastectomy radiotherapy.

Purpose: In this study, we aimed to develop and validate nomograms for predicting the survival outcomes in patients with T1-2N1 breast cancer to identify the patients who could not benefit from postmastectomy radiotherapy (PMRT). Methods: Data from 10191 patients with T1-2N1 breast cancer were extracted from the Surveillance, Epidemiology, and End Results (SEER) database. Of them, 6542 patients who had not received PMRT formed the training set. Concurrently, we retrospectively enrolled 419 patients from the Affiliated Hospital of North Sichuan Medical College (NSMC), and 286 patients who did not undergo PMRT formed the external validation set. The least absolute shrinkage and selection operator (LASSO) and multivariate Cox regression analyses were used for selecting prognostic factors in the training set. Using the selected factors, two prognostic nomograms were constructed. The nomograms' performance was assessed using the concordance index (C-index), calibration curves, decision curve analysis (DCA), and risk subgroup classification. The stabilized inverse probability of treatment weights (IPTWs) was used to balance the baseline characteristics of the different risk groups. Finally, the survival outcomes and effectiveness of PMRT after IPTW adjustment were evaluated using adjusted Kaplan-Meier curves and Cox regression models. Results: The 8-year overall survival (OS) and breast cancer-specific survival (BCSS) rates for the SEER cohort were 84.3% and 90.1%, with a median follow-up time of 76 months, while those for the NSMC cohort were 84.1% and 86.9%, with a median follow-up time of 73 months. Moreover, significant differences were observed in the survival curves for the different risk subgroups (P < 0.001) in both SEER and NSMC cohorts. The subgroup analysis after adjustment by IPTW revealed that PMRT was significantly associated with improved OS and BCSS in the intermediate- (hazard ratio [HR] = 0.72, 95% confidence interval [CI]: 0.59-0.88, P=0.001; HR = 0.77, 95% CI: 0.62-0.95, P = 0.015) and high- (HR=0.66, 95% CI: 0.52-0.83, P<0.001; HR=0.74, 95% CI: 0.56-0.99, P=0.039) risk groups. However, PMRT had no significant effects on patients in the low-risk groups. Conclusion: According to the prognostic nomogram, we performed risk subgroup classification and found that patients in the low-risk group did not benefit from PMRT.