Longitudinal liquid biopsy profiling in breast cancer patient: Clinical utility and guidelines

Background :The clinical impact of repeated liquid biopsies in metastatic breast cancer have not been fully established. While circulating tumor DNA (ctDNA) profiling provides actionable insights, its value after an initially positive test is uncertain. We evaluated the clinical relevance of serial ctDNA analyses according to the initial biopsy result. Methods :Patients from the prospective STING trial (NCT04932525) who underwent ≥2 plasma-derived ctDNA next-generation sequencing assays were analyzed. Somatic variants were classified according to the ESMO Scale for Clinical Actionability of Molecular Targets (ESCAT). Mutational burden, clonal evolution, and emergent actionable alterations were compared between the first (BL1) and subsequent (≥BL2) biopsies. Clinical trial and off-label therapy access were recorded. Results :Among 152 patients, 77% had hormone receptor-positive and 19% triple-negative disease. The number of liquid biopsies ranged from 2 to 6 (median = 2). Fifty-nine patients (38.8%) had a non-contributive first biopsy; 38 became contributive at BL2 and 3 at BL3, while 14 remained non-contributive. The total number of mutations did not differ significantly between BL1 and later biopsies (p = 0.274) but showed a strong correlation (Spearman ρ = 0.65). Newly emergent variants mostly involved clonal hematopoiesis-related ( TP53, DNMT3A, TET2 ) or resistance-associated genes ( RB1, BR CA). Eighty-six patients accessed a clinical trial or off-label targeted therapy; 68 (79%) were guided by alterations already detected at BL1 and 14 (9%) by emerging mutations. Only 10 patients developed new ESCAT I-III alterations, including 7 with ESR1 mutations; others involved ERBB2, PTEN, and somatic BRCA2. Four patients acquired ESCAT IV alterations. Conclusions :Serial liquid biopsies are most relevant in patients with an initially non-contributive assay, with a conversion rate of 64%, as repetition frequently yields actionable findings. In contrast, repeating full-panel sequencing after a contributive biopsy provides limited added value, except for ESR1 mutation monitoring after aromatase inhibitor exposure, which could be performed using lower-cost targeted assays. These results underline that the choice of re-biopsy should balance potential clinical impact and cost-effectiveness.