Treatment and survival of de novo HER2-positive metastatic breast cancer across regional cancer networks in the Netherlands: a cohort study
Abstract
Background Human epidermal growth factor receptor 2 (HER2)-targeted therapy has improved survival in HER2-positive metastatic breast cancer (MBC). Real-world practice may differ across oncologic networks, even within countries with uniform treatment guidelines. We assessed regional variation in first-line treatment and overall survival (OS) among patients with de novo HER2-positive MBC in the Netherlands in 2013-2023. Patients and methods Data from the Netherlands Cancer Registry were analysed. The registry includes seven regions, anonymised as A-G. Crude regional proportions of first-line HER2-targeted therapy were described. Inverse probability of treatment weighting (IPTW) was applied, with region as the exposure. Risk ratios for receiving HER2-targeted therapy were estimated using IPTW-adjusted Poisson regression. OS was analysed using the Kaplan–Meier method and IPTW-adjusted Cox regression, additionally adjusting for baseline covariates and first-line treatment. Results Among 2158 patients, use of first-line HER2-targeted therapy ranged from 70% (region F) to 88% (region C). After adjustment, no regional differences were observed (P ≥ 0.05), except for patients in region C, who were more likely to receive HER2-targeted therapy versus those in region A (risk ratio 1.09, 95% confidence interval, 1.02-1.16, P = 0.01). Observed median OS ranged from 42.8 months (region A) to 74.9 months (region B). After adjustment, no statistically significant regional differences in OS were observed. Regions B and E showed borderline longer OS versus region A (hazard ratio 0.77 and 0.77, P = 0.06 and 0.07, respectively). Conclusions No large regional differences in first-line HER2-targeted therapy were observed among de novo HER2-positive MBC patients in the Netherlands. The observed regional OS difference could largely be explained by differences in baseline patient characteristics.