The Predictive Value of the Tumor-Stroma Ratio for Neoadjuvant Endocrine Therapy in Hormone Receptor-Positive Breast...

Approximately 75% of breast cancers are hormone receptor-positive (HR+). Endocrine therapy may be administered to improve survival. Neoadjuvant therapy may enable breast- and axilla-conserving surgery, but identifying who will benefit remains challenging. This study evaluates the predictive value of the tumor-stroma ratio (TSR) for response to neoadjuvant endocrine therapy (NET) and examines the accuracy of MRI in predicting (near) pathological complete response (pCR). Women diagnosed between 2014 and 2024 with invasive HR+, human epidermal growth factor 2-negative breast cancer who received NET in two Dutch hospitals were included. A total of 208 women (215 tumors) were included, of whom 81 (37.7%) had a stroma-low tumor and 134 (62.3%) stroma-high. Almost 1 out of 4 stroma-low cases achieved (near) pCR after NET. In both univariable (OR 2.85, 95% CI 1.32-6.15, p = 0.008) and multivariable analysis (OR 3.70, 95% CI 1.54-8.90, p = 0.003), stroma-low tumors showed an increased likelihood of achieving (near) pCR compared to stroma-high tumors. In node-positive patients (n = 117), univariable analysis presented higher odds of pCR in the axilla after NET in the stroma-low group (OR 4.41, 95% CI 1.08-18.08, p = 0.039), although this was not significant in multivariable analysis (p = 0.060). The MRI correctly reported (near) pCR in 12 (75%) stroma-low cases and 8 (35%) stroma-high. Stroma-low patients significantly more often achieved (near) pCR of the tumor and axilla after NET compared to stroma-high patients. The accuracy of MRI to predict (near) pCR was higher in the stroma-low group compared to the stroma-high group.