Background: Patients with HER2+ breast cancer who have residual disease after neoadjuvant anti-HER2 plus chemotherapy have a high risk of recurrence and benefit from adjuvant trastuzumab emtansine (T-DM1). We hypothesize that endocrine-responsive residual tumors after neoadjuvant treatments may have good outcomes among patients receiving only adjuvant endocrine therapy plus trastuzumab. Using paired pre- and post-treatment samples from CALGB 40601 and other neoadjuvant cohorts that did not include adjuvant T-DM1, we investigated survival by pretreatment and residual disease ESR1 and PgR gene expression in CALGB 40601, and by estrogen receptor (ER) and progesterone receptor (PR) immunohistochemistry (IHC) in the other cohorts. We considered endocrine-responsive tumors those with ER and/or PR expression by gene expression or IHC.

Methods: CALGB 40601 is a randomized neoadjuvant trial of single vs. dual HER2-targeting (trastuzumab and/or lapatinib added to paclitaxel). We only included those patients who had not suffered from disease progression or death during their preoperative treatments. In total, 77 patients with paired pretreatment and residual disease tumors were profiled by mRNA sequencing and studied. Cutoffs for ESR1 and PgR mRNA expression mimicking clinical positivity were obtained using 265 pretreatment CALGB 40601 tumors (and 1,045 TCGA samples for ESR1). We also examined ER and PR IHC in paired tumors from 202 patients treated at 4 different collaborating institutions; all had residual disease after neoadjuvant HER2 targeting plus chemotherapy. We considered ER- or PR-positive as ≥10% positively staining cells. The primary endpoint was EFS, defined as the time from randomization to event in CALGB 40601 and from the first systemic therapy to event in the 4-institution validation cohort.

Results: In 77 patients from CALGB 40601 with paired (pretreatment/residual disease) specimens, 38 (49.3%) had ESR1+/ESR1+ tumors. The EFS was superior in the ESR1+/ESR1+ (n=38) group than in the remaining others (the log-rank test, p=0.011). The 5- and 7-year EFS rates for the ESR1+/ESR1+ (n=38) were 92.1% and 89.2%, whereas the rates were uniformly < 70% in the others. In particular, the 5-year EFS rate among 11 patients with ESR1+/ESR1- tumors was 61.4%. This remained significant in multivariable analysis with clinical stage and treatment arm; the hazard ratio (HR) for EFS in ESR1+/ESR1+ versus all others was 0.29 (95% CI, 0.09-0.90). In ESR1+/ESR1+ tumors, 5-year EFS rates were high for those whose residual disease and also being PgR+ (n=32) or PAM50 LumA or Normal-like (n=34) (93.8% and 97.1%, respectively). In the institutional validation cohort, pretreatment /residual disease ER(+)/ER(+) tumors (n=113) had superior 3-year EFS versus all others (p=0.010). At a median follow-up of 35.9 months, the 3-year EFS rates for ER(+)/ER(+) and all other groups were 96.6% and 82.6%, respectively. This remained strongly significant in multivariable analysis with clinical stage; the HR for EFS in ER(+)/ER(+) versus all other groups was 0.28 (95% CI, 0.09-0.88). Among 46 who also had PR+ in the residual disease, the 3-year EFS was 100.0%.

Conclusions: HER2+ patients with ER+ pretreatment and ER+ residual disease after neoadjuvant chemotherapy + HER2-targeting have a very good survival outcome despite not receiving additional anti-HER2 targeting with T-DM1. This may provide a simple mechanism to better tailor therapy within residual disease patients using serial ER measurements.

Citation Format: Sung Gwe Ahn, Aranzazu Fernandez-Martinez, Mei-Yin C. Polley, Soong June Bae, Seho Park, Núria Chic, Fara Brasó-Maristany, Benedetta Conte, Valentina Guarneri, Maria Vittoria Dieci, Gaia Griguolo, PierFranco Conte, Joon Jeong, Aleix Prat, Lisa A. Carey, Charles M Perou. A good prognosis of endocrine-dependent tumors among residual invasive cancer after anti-HER2 therapy: CALGB 40601 (Alliance) and validation studies [abstract]. In: Proceedings of the 2020 San Antonio Breast Cancer Virtual Symposium; 2020 Dec 8-11; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2021;81(4 Suppl):Abstract nr PS10-02.