Researchers at the Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China, have developed and validated a clinical prognostic model incorporating age, tumour (T), nodal (N) stage, and chemotherapy to predict survival outcomes in patients with early-stage medullary breast cancer (medBC).
The study included 650 patients (<65 years of age, 78.2 percent) with early-stage medBC from the Surveillance, Epidemiology, and End Results (SEER) database. They were grouped 2:1 into two cohorts: the training cohort (patients diagnosed between 2011 and 2015; n=449; White, 69.5 percent; Asian or Pacific Islander, 6.5 percent) used to develop the predictive model, and the internal validation cohort (patients diagnosed between 2016 and 2021; n=201; White, 63.7 percent; Asian or Pacific Islander, 7.5 percent). In addition, 188 medBC patients (<65 years of age, 93.6 percent; Asian, 100 percent) treated at the National Cancer Center, China, between 1999 and 2024 formed the external validation cohort. [Hong Kong Med J 2026;32:330-340]
Baseline characteristics were similar across the study groups. Most patients had early T and N stages. About half (55.2 percent) of patients in the SEER cohort and 71.3 percent in the real-world cohort had triple-negative breast cancer (TNBC). Treatments received by patients in the respective cohorts were surgery (98.5 and 100 percent), radiotherapy (51.2 and 49.5 percent), and systemic chemotherapy (74.8 and 88.3 percent). The median Ki-67 index for patients in the real-world cohort was 70 percent (range, 0–90 percent). Complete Ki-67 data were available for 100/188 patients, of whom 53.0 percent had a Ki-67 index >70 percent.
A nomogram was developed in the training cohort based on the four (out of 13 evaluated) clinicopathological variables, which were found to be associated with OS in univariate and multivariable analyses. Among these four prognostic variables, T stage (T3–T4) was the strongest predictor of inferior survival outcomes, followed by age (≥65 years), N stage (N2–N3), and chemotherapy (none). The overall C-index was 0.76 (95 percent confidence interval [CI], 0.69–0.83), and the area under the curve (AUC) values for predicting 1-, 3-, 5- and 10-year OS were 0.83, 0.76, 0.75, and 0.76, respectively.
In the internal validation cohort, the C-index was 0.86; AUC values were 0.78 and 0.83 for 3- and 5-year OS, respectively. One- and 10-year OS data were not assessed due to limited number of patients.
In the external validation cohort, the C-index was 0.90; AUC values were 0.85, 0.85, and 0.84 for 3-, 5-, and 10-year OS, respectively.
The low-risk group had significantly better OS than the high-risk group across training and validation cohorts (p<0.05).
Separate analyses found no significant association between Ki-67 index and OS, although high Ki-67 index is typically associated with medBC. Patients with a family history of breast cancer had shorter OS than those without (239.8 vs 284.3 months), but the difference was not statistically significant (p=0.10).
“This clinical model may facilitate identification of patients at higher risk of poor survival outcomes and support optimized clinical management of early-stage medBC,” the authors concluded.