HER2-low has emerged as a new predictive biomarker in metastatic breast cancer. However, its prognostic value in early-stage carcinomas needs to be revisited. We aimed to evaluate the association of HER2-low carcinomas with PAM50 risk groups combined with clinicopathological variables in early breast cancer.

We conducted a retrospective analysis of 332 patients with early-stage breast cancer that underwent PAM50 signature analysis between 2015 and 2021at Hospital Universitario 12 de Octubre (Madrid, Spain). Clinical and pathological variables were collected from medical records. After adjusting for potential confounders, we estimated Odds Ratio (OR) and 95% confidence interval for high-risk PAM50 subgroup, comparing HER2-low versus HER2-zero carcinomas by multivariable logistic regression. P values below 0.05 were deemed statistically significant.

192 (57%) patients were classified as HER2-low carcinomas. Median follow-up was 34 months. Adjusted OR for high-risk PAM50 when comparing HER2-low versus HER2-zero carcinomas was 1.31 (95% CI: 0.75–2.30, p = 0.33). The multivariable model detected significant associations for Ki-67% (≥20% vs. <20%: OR = 4.03, 95% CI: 2.15–7.56, p < 0.001), T staging category (T2/T3 vs. T1: OR = 3.44, 95% CI: 1.96–6.04, p < 0.001), progesterone receptor (PR ≥ 20% vs. <20%: OR = 0.44, 95% CI: 0.23–0.83, p = 0.01), nodal staging category (N+ vs. N0: OR = 3.8, 95% CI: 1.89–7.62, p < 0.001) and histological grade (grade 2 vs. 1: OR = 2.41, 95% CI: 1.01–5.73, p = 0.04; grade 3 vs 1: OR = 5.40, 95%CI: 1.98–14.60, p = 0.001).

In this early-stage breast cancer cohort, HER2-low was not associated with a high-risk PAM50 compared to HER2-zero carcinomas. Ki-67 ≥ 20%, T2/T3, histological grade 2/3, N+ and PR<20% were significantly associated to a high-risk PAM50.