Prognostic performance of the Japanese Breast Cancer Society pathological response classification after neoadjuvant...

Pathological response after neoadjuvant chemotherapy is strongly associated with prognosis in breast cancer. However, further prognostic stratification is needed among patients with residual invasive disease because their outcomes vary widely. The six-tier Japanese Breast Cancer Society (JBCS) pathological treatment effect classification provides a graded assessment of histological response in the primary breast tumor, but its prognostic validity has not been sufficiently established in a large cohort.

We retrospectively analyzed 1,003 patients with invasive breast cancer who underwent surgery after neoadjuvant chemotherapy at two institutions between 2006 and 2022. Distant recurrence-free survival (DRFS) was evaluated according to the six-tier JBCS classification. Multivariable Cox models incorporating either the six-tier JBCS classification or ypT were adjusted for age, breast cancer subtype, and ypN and compared using the corrected Akaike information criterion (AICc) and Bayesian information criterion (BIC). Exploratory analyses also evaluated simplified classifications.

The six-tier JBCS classification significantly stratified DRFS (log-rank P < 0.001). Five-year DRFS increased from 60.6% for Grade 0 to 94.4% for Grade 3, although several intermediate categories overlapped. The classification remained independently associated with DRFS (global P < 0.001). However, the ypT model showed better fit (AICc, 3052.00 vs. 3065.71; BIC, 3105.70 vs. 3124.28). Exploratory simplification analyses retained significant prognostic stratification, although none showed better model performance than ypT.

The six-tier JBCS classification provides clinically meaningful prognostic information after neoadjuvant chemotherapy. However, ypT showed better model fit, and overlap among intermediate categories suggests scope for refinement.