HER2-positive status is an independent predictor for coexisting invasion of ductal carcinoma in situ of the breast presenting extensive DCIS component.

DCIS of the breast with coexisting invasion is commonly seen, and no consensus on any biomarker capable of discriminating this subgroup has been reached yet. We retrospectively examined the receptor status and the histological grade in Chinese DCIS patients to identify any independent predictor in order to discriminate a subgroup with coexisting invasion from pure DCIS patients. A consecutive Chinese DCIS patient cohort registered at a single institution was included for ER, PR, and HER2 status, as well as for evaluation of the histological grade. Patients with invasion foci >1cm in diameter were excluded. The HER2 gene amplification status was further examined by FISH when the IHC result was HER2 (2+). Molecular subtypes were also profiled. Age, histological grade, ER, PR, and HER2 status were included in association analyses. In total, 183 patients were included. A hundred and forty patients had pure DCIS, and 43 patients had DCIS with invasion. The luminal A subtype accounted for 49.7% of all cases, the HER2-positive subtype for 27.9%, and only 10.4% and 12.0% represented the luminal B and basal-like subtypes, respectively. Univariate analyses showed that histological Grade 2, Grade 3, and HER2-positive status were associated with DCIS with invasion, odds ratios 5.1 (P = 0.017), 5.2 (P = 0.01) and 3.34 (P = 0.001), respectively. However, only the HER2-positive status was of statistical significance in the multivariate logistic regression analyses after adjustment for other markers, odds ratio 3.8 (95%CI 1.4-10, P = 0.008). The 43 cases with invasion were further stratified into extensive or small DCIS components according to the percentage of DCIS to total tumor area using 25% as the cutoff point. Multinomial logistic regression with pure DCIS cases as reference showed that the HER2-positive status was associated only with the group showing an extensive DCIS component, odds ratio 6.2 (95%CI 1.8-21, P = 0.003), but not with the group having a small DCIS component. Our study demonstrates that HER2-positive status is an independent predictor for DCIS, with invasion presenting an extensive DCIS component, and favors the hypothesis that HER2 overexpression or gene amplification is involved in the transition from DCIS to invasive disease.