Estimation of Subtype‐Specific Tumour Growth Rate Distributions and Interval Cancer Proportions in Breast Cancer Using Biology‐Inspired Natural History Models

Le résumé fourni par la source

Breast cancer progression varies across molecular subtypes, influencing detection rates and prognosis. Differences in the proportion of interval cancers (PIC)-cancers detected symptomatically between scheduled screening rounds-have been observed across subtypes, likely reflecting heterogeneity in tumour growth dynamics. We applied a continuous-growth natural history model to a cohort of 7818 breast cancer patients diagnosed between 2007 and 2020 in Sweden. Using subtype-specific estimates, we quantified the distributions of interval cancers within the 2-year period between screening rounds, differentiating by tumour sizes of missed interval cancers at their last negative screen. Additionally, the fitted model enabled estimation of subtype-specific tumour mean doubling times (MDT) and the average duration from tumour onset to symptomatic diagnosis. Triple-negative breast cancers (TNBC) exhibited the fastest growth (MDT: 126 days), followed by Luminal B-like (203 days) and Luminal A-like (332 days). Corresponding mean asymptomatic times (MAT) were 4.5, 7.6, and 11.8 years, respectively. Observed PICs among regular attenders were 48.9% (TNBC), 32.3% (Luminal B-like), and 22.0% (Luminal A-like), closely aligning with model-based expectations. Approximately 17% of TNBC interval cancers were estimated to have had diameters ≤ 1 mm at the last negative screen, versus 13% for Luminal B-like and 8% for Luminal A-like tumours. Our model integrates tumour growth into the detection processes to estimate subtype-specific tumour dynamics and their impact on the PIC. The results support the hypothesis that higher PICs in more aggressive subtypes are driven by faster tumour growth and shorter asymptomatic periods. Our results may inform screening strategies tailored to patients' risk.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.