Breast cancers that display heterogeneity of HER2 expression may represent a distinct subset of HER2-positive breast cancer that is associated with lower rates of pathologic complete response, according to a phase II study from Dana-Farber Cancer Institute. This trial evaluated HER2 heterogeneity as a predictor of response to neoadjuvant treatment with trastuzumab emtansine (T-DM1) plus pertuzumab. HER2 heterogeneity was defined as either the presence of HER2 positivity in between 5% and 50% of tumor cells or the presence of an area of tumor testing HER2-negative.
“Intratumor HER2 heterogeneity assessed by routine pathology evaluation is a strong predictor of pathologic complete response to a dual HER2-targeted therapy regimen. If validated, this may need to be considered in selecting patients for HER2-targeted regimens without chemotherapy in the curative setting…. Patients with heterogeneous tumors may need additional chemotherapy in addition to classic anti-HER2 therapy,” said Otto Metzger Filho, MD, who presented the results at the 2019 ASCO Annual Meeting.1
“When we think about HER2 positive disease at the tumor level, we expect that the majority of cells would be HER2-positive, but a fraction are actually heterogeneous,” he said. Within a breast tumor, there may be at least two distinct subclones with different levels of HER2 amplification. Retrospective studies have found this HER2 heterogeneity to vary from 10% to 30%, depending on the definition and the population being studied, and to be associated with inferior outcomes.
“Investigating the impact of HER2 heterogeneity on response to therapy is an important step while we try to de-escalate chemotherapy and rely on HER2-targeted therapy.”— Otto Metzger Filho, MD
Tweet this quote
“Investigating the impact of HER2 heterogeneity on response to therapy is an important step while we try to de-escalate chemotherapy and rely on HER2-targeted therapy,” Dr. Metzger Filho said. “To my knowledge, this is the first clinical trial designed and conducted to prospectively evaluate this.”
Heterogeneity Associated With Lack of Response
The single-arm phase II study was designed to evaluate HER2 heterogeneity and its relationship to pathologic complete response. The study enrolled 164 patients with stage II and III breast cancer with centrally confirmed HER2-positive disease treated neoadjuvantly with T-DM1 plus pertuzumab for 6 cycles (completed by 95%). Heterogeneity was evaluated on ultrasound-guided core biopsy samples from two different sites within each tumor.
Assessment of HER2 heterogeneity was centrally performed by pathologists blinded to the treatment outcome. The results were analyzed using a stratified test to prevent confounding between hormone receptor status and pathologic complete response. Pathologic complete response was defined by residual cancer burden class 0.
The median tumor size was 2.8 cm; 68% of patients had hormone receptor–positive disease; and 84% had stage II disease. By immunohistochemistry (IHC), 24% were HER2 2+ and 74% were HER2 3+.
HER2 Heterogeneity in 10%
HER2 heterogeneity was identified in 10% of cases, of whom 81% were hormone receptor–positive and 19% were hormone receptor–negative. In the nonheterogeneous subset of 141 patients, pathologic complete responses (residual cancer burden class 0) were achieved by 55%, whereas none were observed in the 16 patients with HER2 heterogeneity, Dr. Metzger Filho reported.
“The study met its primary endpoint by demonstrating a significant association between HER2 heterogeneity and pathologic complete response, adjusted by hormone receptor status,” he said.
Similarly, with residual cancer burden class 0 or 1 as the outcome, pathologic complete response was achieved by 67% of the nonheterogeneous subgroup and 25% of the heterogeneous group, (P = .004). In an exploratory analysis of outcome by strength of immunohistochemistry staining, response rates were 56% in HER2 3+ tumors and 27% in 2+ tumors (P = .002). When the analysis excluded cases that were classified as HER2 heterogeneous, the difference in pathologic complete response rates between HER2 immunohistochemistry 3+ (58%) and 2+ (40%) was less pronounced (P = .10), he added.
“The association between heterogeneity and pathologic complete response remains significant when adjusted by both hormone receptor status and HER2 IHC status,” Dr. Metzger Filho concluded.
Digital Spatial Profiling
USING DIGITAL spatial profiling, the investigators have found differences in protein HER2 level between two biopsies of the same tumor, illustrating intratumor heterogeneity of HER2 expression by a different technique.
“Looking across 13 heterogeneous cases, we can visualize that the heterogeneity is more diverse than we could expect with the classic pathologic evaluation … and that the HER2 protein level in the HER2 heterogeneous cases is lower than in the nonheterogeneous cases, regardless of pathologic complete response,” he said. The findings suggest a means of identifying a subset of HER2-positive patients who may need chemotherapy or novel treatment approaches, Dr. Metzger Filho said. ■
DISCLOSURE: Dr. Metzger Filho has received honoraria from Grupo Oncoclinicas and Roche Brasil; has received institutional research funding from AbbVie, Cascadian Therapeutics, Eisai, Pfizer, Roche/Genentech, and Susan G. Komen for the Cure; and has received travel/accommodations/expenses from Grupo Oncoclinicas.
1. Metzger Filho O, Viale G, Trippa L, et al: HER2 heterogeneity as a predictor of response to neoadjuvant T-DM1 plus pertuzumab: Results from a prospective clinical trial. 2019 ASCO Annual Meeting. Abstract 502. Presented June 3, 2019.
Mark Pegram, MD
Mark Pegram, MD, the Susy Yuan-Huey Hung Professor of Medical Oncology and Director of the Stanford Breast Oncology Program, said the Dana-Farber study “independently confirms that HER2 heterogeneity is a distinct clinical entity with lower levels of HER2 expression and...