Breast cancer researchers at The University of Texas MD Anderson Cancer Center, Houston, are beginning to select patients with HER2-positive breast cancer who might forgo surgery after neoadjuvant chemotherapy. While this fine-tuning is currently intended to inform clinical trials—in particular, to select patients for an exceptional-responder trial—it could, in the future, guide patient selection for de-escalation of therapy, they said.

Susie Sun, MD

“Evidence is growing that certain patients are exceptional responders to neoadjuvant chemotherapy. This suggests that when properly identified, certain of these women may someday be candidates for nonoperative treatment,” said lead researcher Susie Sun, MD, a breast surgical oncology fellow at MD Anderson.

At the 2019 American Society of Breast Surgeons Annual Meeting, Dr. Sun reported the results of a study comparing the clinicopathologic characteristics of HER2-positive patients who had no evidence of residual cancer on the pathology report with patients who did demonstrate residual cancer.¹ While not exactly pinpointing which patients can safely forgo surgery, it clearly showed which women should not forgo this modality because of a higher occurrence of residual disease following neoadjuvant therapy: those with hormone receptor–positive disease, those with ductal carcinoma in situ (DCIS) on the initial biopsy, and those with inconclusive results on multimodality imaging. It also confirmed that multimodality imaging is not reliable in identifying pathologic complete response. “Imaging is associated with a high rate of false-negatives,” Dr. Sun reported.

“At MD Anderson, it is not our practice to use multimodality imaging to select these patients, because other studies from our group have proven it to be unreliable. Multimodality imaging combined with multiple image-guided core biopsies of the tumor bed, however, is a highly reliable means of identifying patients with true pathologic complete response,” she said.

“Certain patients are exceptional responders to neoadjuvant chemotherapy [and] may someday be candidates for nonoperative treatment.”

— Susie Sun, MD

Tweet this quote

Study Details

The study examined 280 patients with HER2-positive breast cancer (cT1–T2, cN0–N1) treated neoadjuvantly with HER2-targeted regimens followed by resection. Patients received a taxane-based regimen with either trastuzumab alone (n = 255) or trastuzumab plus pertuzumab (n = 25). The multimodality imaging consisted of mammography plus ultrasound; 13% of patients also underwent magnetic resonance imaging (MRI).

Multivariate analyses were performed to determine predictors of residual disease. Importantly, the study assessed eradication of both the invasive and DCIS components, evaluated the frequency of DCIS co-occurring with invasive cancer, and determined its impact on residual disease after systemic neoadjuvant treatment.

“For currently accruing trials that evaluate the elimination of surgery for exceptional responders, the eradication of both invasive and DCIS components is necessary, because DCIS may serve as a nidus for carcinoma in the future,” Dr. Sun said.

Of the 280 patients, 37.5% achieved a pathologic complete response for both invasive disease and DCIS (ypT0), whereas 55.4% had a pathologic complete response for invasive disease. Residual DCIS (ypTis) remained present in 17.9% of patients. “Residual DCIS alone is an important consideration, as these patients would not be optimal candidates for an omission-of-surgery trial,” she commented.

In evaluating the effect of treatment on the DCIS component, the researchers found DCIS on the initial biopsy in 129 patients (46.1%); neoadjuvant chemotherapy eradicated it in 46 (35.7%). The presence of DCIS on biopsy was associated with a higher proportion of patients with residual disease compared to those without DCIS (69% vs 57%; P = .04), she reported.

Hormone receptor status was also important. Among patients with hormone receptor–negative disease, 60 (49.2%) achieved a pathologic response (ypT0N0). This number fell to 42 patients (26.6%) among the hormone receptor–positive cohort. On multivariate analysis, hormone receptor–positive tumors were 2.7 times more likely (P < .0001) to contain residual disease in the breast and/or nodes, compared to hormone receptor–negative tumors, after neoadjuvant therapy.

While postchemotherapy multimodality imaging had a sensitivity of 97.1% for identifying residual disease in the breast and nodes, its negative predictive value was 70.6%, she noted.

“Similar to previous studies, multimodality imaging using mammogram, ultrasound, and MRI was not reliable in identifying patients with a pathologic complete response,” Dr. Sun said. “Imaging was least reliable in evaluating lymph node status, with a negative predictive value of only 40.0%. In the breast and nodes, the negative predictive value was still low—70.6%—showing that imaging alone is unreliable for selecting patients for enrollment in clinical trials for HER2-positive breast cancer.”

“Image-guided percutaneous biopsy is required to safely select patients for inclusion in ongoing and future elimination-of-surgery clinical trials,” she emphasized. ■

DISCLOSURE: Dr. Sun reported no conflicts of interest.

REFERENCE

1. Sun S, van la Parra R, Rauch G, et al: Patient selection for non-operative management of HER2-positive invasive breast cancer after neoadjuvant systemic therapy. 2019 American Society of Breast Surgeons Annual Meeting. Abstract 575089. Presented May 2, 2019.