Nab-paclitaxel weekly versus dose-dense solvent-based paclitaxel followed by dose-dense epirubicin plus cyclophosphamide in high-risk HR+/HER2− early breast cancer: results from the neoadjuvant part of the WSG-ADAPT-HR+/HER2− trial

•16-week neoadjuvant nab-paclitaxel induces higher pCR than sb-paclitaxel–epirubicin/cyclophosphamide in high-risk HR+/HER2− breast cancer.•First prospective phase III trial showed that higher RS is predictive for pCR.•Associations of higher RS and ET non-response with pCR are moderated by menopausal status (and/or ET agent used).•pCR mitigates the unfavorable impact of higher RS on dDFS.•Further trials should investigate therapy de-escalation in patients with RS >25, ET response, and lower clinical risk. BackgroundIn high-risk hormone receptor-positive/human epidermal growth factor receptor 2-negative (HR+/HER2−) early breast cancer (EBC), nanoparticle albumin-bound (nab)-paclitaxel showed promising efficacy versus solvent-based (sb)-paclitaxel in neoadjuvant trials; however, optimal patient and therapy selection remains a topic of ongoing research. Here, we investigate the potential of Oncotype DX® recurrence score (RS) and endocrine therapy (ET) response (low post-endocrine Ki67) for therapy selection.Patients and methodsWithin the WSG-ADAPT trial (NCT01779206), high-risk HR+/HER2− EBC patients were randomized to (neo)adjuvant 4× sb-paclitaxel 175 mg/m2 q2w or 8× nab-paclitaxel 125 mg/m2 q1w, followed by 4× epirubicin + cyclophosphamide (90 mg + 600 mg) q2w; inclusion criteria: (i) cN0-1, RS 12-25, and post-ET Ki67 >10%; (ii) cN0-1 with RS >25. Patients with cN2-3 or (G3, baseline Ki67 ≥40%, and tumor size >1 cm) were allowed to be included without RS and/or ET response testing. Associations of key factors with pathological complete response (pCR) (primary) and survival (secondary) endpoints were analyzed using statistical mediation and moderation models.ResultsEight hundred and sixty-four patients received neoadjuvant nab-paclitaxel (n = 437) or sb-paclitaxel (n = 427); nab-paclitaxel was superior for pCR (20.8% versus 12.9%, P = 0.002). pCR was higher for RS >25 versus RS ≤25 (16.0% versus 8.4%, P = 0.021) and for ET non-response versus ET response (15.1% versus 6.0%, P = 0.027); no factors were predictive for the relative efficacy of nab-paclitaxel versus sb-paclitaxel. Patients with pCR had longer distant disease-free survival [dDFS; hazard ratio 0.42, 95% confidence interval (CI) 0.20-0.91, P = 0.024]. Despite favorable prognostic association of RS >25 versus RS ≤25 with pCR (odds ratio 3.11, 95% CI 1.71-5.63, P ≤ 0.001), higher RS was unfavorably associated with dDFS (hazard ratio 1.03, 95% CI 1.01-1.05, P = 0.010).ConclusionsIn high-risk HR+/HER2− EBC, neoadjuvant nab-paclitaxel q1w appears superior to sb-paclitaxel q2w regarding pCR. Combining RS and ET response assessment appears to select patients with highest pCR rates. The disadvantage of higher RS for dDFS is reduced in patients with pCR. These are the first results from a large neoadjuvant randomized trial supporting the use of RS to help select patients for neoadjuvant chemotherapy in high-risk HR+/HER2− EBC. In high-risk hormone receptor-positive/human epidermal growth factor receptor 2-negative (HR+/HER2−) early breast cancer (EBC), nanoparticle albumin-bound (nab)-paclitaxel showed promising efficacy versus solvent-based (sb)-paclitaxel in neoadjuvant trials; however, optimal patient and therapy selection remains a topic of ongoing research. Here, we investigate the potential of Oncotype DX® recurrence score (RS) and endocrine therapy (ET) response (low post-endocrine Ki67) for therapy selection. Within the WSG-ADAPT trial (NCT01779206), high-risk HR+/HER2− EBC patients were randomized to (neo)adjuvant 4× sb-paclitaxel 175 mg/m2 q2w or 8× nab-paclitaxel 125 mg/m2 q1w, followed by 4× epirubicin + cyclophosphamide (90 mg + 600 mg) q2w; inclusion criteria: (i) cN0-1, RS 12-25, and post-ET Ki67 >10%; (ii) cN0-1 with RS >25. Patients with cN2-3 or (G3, baseline Ki67 ≥40%, and tumor size >1 cm) were allowed to be included without RS and/or ET response testing. Associations of key factors with pathological complete response (pCR) (primary) and survival (secondary) endpoints were analyzed using statistical mediation and moderation models. Eight hundred and sixty-four patients received neoadjuvant nab-paclitaxel (n = 437) or sb-paclitaxel (n = 427); nab-paclitaxel was superior for pCR (20.8% versus 12.9%, P = 0.002). pCR was higher for RS >25 versus RS ≤25 (16.0% versus 8.4%, P = 0.021) and for ET non-response versus ET response (15.1% versus 6.0%, P = 0.027); no factors were predictive for the relative efficacy of nab-paclitaxel versus sb-paclitaxel. Patients with pCR had longer distant disease-free survival [dDFS; hazard ratio 0.42, 95% confidence interval (CI) 0.20-0.91, P = 0.024]. Despite favorable prognostic association of RS >25 versus RS ≤25 with pCR (odds ratio 3.11, 95% CI 1.71-5.63, P ≤ 0.001), higher RS was unfavorably associated with dDFS (hazard ratio 1.03, 95% CI 1.01-1.05, P = 0.010). In high-risk HR+/HER2− EBC, neoadjuvant nab-paclitaxel q1w appears superior to sb-paclitaxel q2w regarding pCR. Combining RS and ET response assessment appears to select patients with highest pCR rates. The disadvantage of higher RS for dDFS is reduced in patients with pCR. These are the first results from a large neoadjuvant randomized trial supporting the use of RS to help select patients for neoadjuvant chemotherapy in high-risk HR+/HER2− EBC.