New Research Highlights Clinical Significance of HER2-Low Breast Disease

Human epidermal growth factor receptor 2 (HER2)–low is a recent novel classification of breast cancer, in which the cancerous cells express lower levels of the HER2 protein than sufficient to classify patients as having HER2–positive disease, with immunohistochemistry (IHC) scores typically coming in at 1+/2+ (borderline) and in situ hybridization results indicating nonamplified status.

With figures ranging from 50% to 60% of all breast cancers—and some sources saying this subtype accounts for at least 55% of breast cancers—research is continuing to increasingly focus on the disease subtype, as evidenced by a pair of abstracts presented at this year’s San Antonio Breast Cancer Symposium by teams from the University of Texas MD Anderson Cancer Center.

HER2-Low Early-Stage Triple-Negative Breast Cancer¹

In this cross-sectional study, which was a subanalysis of patients with stage I to III triple-negative breast cancer (TNBC) enrolled in the ARTEMIS trial from 2015 to 2021, the researchers explored tumor response among 367 patients receiving anthracycline-based neoadjuvant therapy (NAT), comprising those with HER2-low status (n = 149; median [IQR] age, 51 [44-60] years) and HER2-zero status (n = 218; median [IQR] age, 52 [43-60] years). Tumor tissue analyzed was extracted prior to NAT initiation.

Most patients had T2 (59% in both groups) or T1 (21%, HER2-low; 27%, HER2-zero) tumors (P = .44), negative nodal status (56% and 64%, respectively; P = .10), invasive ductal carcinoma (IDC) histology (95% and 89%; P = .001), histologic grade of 3 (85% in both groups; P = .88), Ki67 levels above 35% (76% and 87%; P = .017), and an androgen receptor (AR) level below 10% (64% and 79%; P = .001). The study authors summed that HER2-low tumors were less likely to have metaplastic histology (P = .001), more likely to be AR positive (P = .01), and were associated with lower Ki67 expression (P = .017).

“Little is known about the biological significance of low HER-2 expression in patients with stage I-III TNBC receiving NAT,” the authors wrote. They used DEseq2 for differential gene expression and pathway analysis, Hallmark gene sets for gene set enrichment analysis, and CIBERSORT for deconvolution analysis.

Tumor-infiltrating lymphocyte levels and PD-L1 expression did not differ between the patient groups—nor was race/ethnicity seen to be a determining factor—and “deconvolution analyses revealed no significant difference between predicted proportions of immune cell subpopulations between HER2-low and HER2-zero tumors,” the authors wrote.

However, upregulation of several genes was seen among patients with HER2-low vs HER2-zero tumors: ACSM1, which is involved in fatty acid metabolism, and DHRS2 and UGT2B28, which are involved in steroid hormone metabolism. In addition, increased expression of EREG among patients with HER2-low tumors and downregulation of genes involved in immune response pathways in patients with HER2-zero tumors were linked to non–partial complete responses (pCRs) among the study population. The 3 most significantly downregulated pathways among the patients with HER2-zero tumors, looking at those who achieved a pCR vs not, were Hallmark allograft rejection (false discovery rate [FDR] q = 0.007), interferon gamma response (FDR q = 0.002), and interferon alpha response pathways (FDR q = 0.007).

Rates of pCR did not significantly differ between the groups, the authors noted, with 46% of patients with HER2-zero tumors and 40% of those with HER2-low tumors (P = .34) achieving this outcome.

The authors concluded that “biological differences between HER2-zero and HER2-low tumors exist and may influence future personalized treatment for patients with early-stage TNBC.”

HER2-Low Metastatic Breast Cancer²

Echoing the sentiments of the authors above, in that it is important to better understand the HER2-low disease phenotype, the authors of this abstract conducted a retrospective review of 4256 female patients with HER2-negative metastatic breast cancer (mBC) treated at MD Anderson between 2006 and 2019. Most (n = 3053) had stage I to III (early-stage) recurring mBC and 1203, de novo disease. For this analysis, HER2-low disease had an IHC score of 1+/2+ and negative results following fluorescence in situ hybridization (FISH) and HER-2 negative disease had an IHC score of 0. The outcomes of interest were overall survival (OS) and disease-free survival (DFS).

HER2-low disease was seen among 59.3% each of the early-stage recurrent and de novo patients.

Overall, following univariate logistic analysis, pCR was significantly associated with negative (10.2%) vs positive (2.7%) estrogen receptor (ER) status, negative (8.6%) vs positive (2.0%) progesterone receptor (PR) status, and negative lymphatic vascular invasion (9.2% vs 2.0%). Multivariate regression analysis produced similar results for ER status (P = .0124), PR status (P = .0439), and lymphatic vascular invasion (P < .0001).

Following multivariate logistic regression analysis, patients with recurrent disease were shown to be more likely to have HER2-low disease if they were White (odds ratio [OR], Black vs White patients, 0.885; P = .31; Hispanic vs White patients, 0.756; P = .04 ; Other race vs White, 0.670; P = .01), had a higher nuclear grade (OR, II vs I, 1.870; P = .006; III vs I, 1.887; P = .005), and had ER-positive status (OR, 1.940; P < .0001). These patients had a median OS of 5.3 years—1985 patients died—with patients with HER-2 low disease having 5.4 years and HER2-zero disease, 4.8 years.

Following multivariate Cox regression analysis, significant associations were seen between longer OS and HER2-low status overall (HR, 0.87; P = .008), after adjusting for age, race, stage, nu clear grade, lymphatic vascular invasion, and ER/PR status, and among the cohort who received neoadjuvant chemotherapy (HR, 0.87; P = .04) after adjusting for stage, nuclear grade, ER/PR status, and pCR. Median DFS was 1.9 years, with similar 5-years DFS rates seen in the HER2-low (14.9%) and HER2-zero (14.0%) cohorts. Factors significantly associated with DFS were age, stage, histology, nuclear grade, and ER/PR status.

Among those with de novo disease, associations were seen between HER2-low status and higher nuclear grade (OR, II vs I, 1.838; P = .008; III vs I, 1.856; P = .007) and ER-positive status (OR, 1.933; P < .0001). Their median OS was 3.2 years—767 patients died—with fewer patients with HER2-low disease being alive at the 5-year mark vs those with HER2-zero disease: 28.6% vs 32.5% (HR, 0.834; P = .026). Factors significantly associated with OS were race, histology, nuclear grade, ER/PR status, and HER2-low status (HR, 0.834; P = .0260).

The authors noted that their real-world findings “help to establish the prevalence of HER2-low and outcomes for this selective cohort,” with HER2-low status significantly associated with nuclear grade and ER positivity overall and among patients with mBC, with longer OS and DFS vs patients with HER2-zero disease.

References

1. Yam C, Korkut A, Kong E, et al. Clinical and Molecular characteristics of HER2-low/zero, early stage triple negative breast cancer. Presented at: SABCS; December 6-10, 2022; San Antonio, TX. Abstract HER2-01.

2. Raghavendra AS, Liu DD, Mouabbi JA, Tripathy D. Prevalence of HER2-low among stage I-III, metastatic breast cancer patients and their outcomes by HER2 status. Presented at: SABCS; December 6-10, 2022; San Antonio, TX. Abstract HER2-04.