While many studies are addressing low ERBB2 expression in invasive breast cancer, none have dealt with DCIS. Yet DCIS is raising, patients (pts) increasingly undergo surgical and often systemic therapy to ward off recurrences, which although low are mostly invasive. Unfortunately, no markers are currently available to identify pts who will or will not relapse. With this premise, we aimed to investigate whether low ERBB2 expression is associated with distinct clinico-pathological characteristics and prognosis among pts with DCIS.
This cohort study was conducted using data from a prospectively maintained institutional database on all consecutive 7645 pts undergoing breast surgery between 2012 and 2020 at our Institute. Primary DCIS was classified by immunohistochemistry (IHC) as ERBB2-0 if IHC scored 0, ERBB2-low if IHC was 1+ or 2+; ISH was not considered as it is not routinely used in DCIS diagnostics; and ERBB2-overexpressed if IHC was 3+.
A total of 375 DCIS pts were analyzed; median age was 54 (27-88) years; primary tumor size≤ 2 cm in 59.7% of pts, grade III in 32.8%, estrogen receptor (ER) positive in 80.3%. Breast conserving surgery was performed in 70.7% of pts, adjuvant endocrine and radiotherapy in 14.40% and 38.40%, respectively; 52.54% of DCIS presented low ERBB2 expression. Differentiated grade (p<0.001), Ki67<20 (p<0.001), and ER-positive (p<0.001) were most common among ERBB2-low. The rate of ERBB2 low was as low as 6.60% and 1.52% in ER-negative (0-9%) and -intermediate (10-49%) and increased to 39.09% in ER-high (50%-95%) and 52.79 % in ER-very high (ER > 95%) tumors (p<0.001). At a median follow-up of 39 (IQR 16-65) months, the incidence for local relapse was 0.075 with no significant differences by ERBB2 expression. Among 20 evaluable relapses with paired primary and recurrent tumors, 6 presented a discordant ERBB2 status. The increase and decrease in ERBB2 expression occurred equally.
Low ERBB2 expression in DCIS is associated with features of reduced aggressiveness. Importantly, receptor conversion may happen prompting retesting in recurrent cases. The data are consistent with those observed in invasive breast cancer.
The authors.
Has not received any funding.
G. Pruneri: Other, Personal, Advisory Board: Roche, Bayer, AstraZeneca; Other, Personal, Speaker’s Bureau: Roche, Bayer, AstraZeneca. S. Di Cosimo: Other, Personal, Speaker’s Bureau: AstraZeneca; Other, Personal, Advisory Board: Pierre-Fabre; Other, Personal, Other: IQVIA, MEDSIR. All other authors have declared no conflicts of interest.