Fig. 1: Study workflow and ctDNA status analysis.

a Schematic representation of the study design and timeline of blood sample collection. Plasma samples were collected at six time points: pre-therapy (pre-NAC), post-NAC (pre-op), and 1, 3, 6, and 12 months post-surgery. NAC neoadjuvant chemotherapy, op surgery. b Overview of patient enrollment. Six patients were retrospectively enrolled in cohort 1, and 34 patients were prospectively enrolled in cohort 2. The combined analysis included a total of 40 patients. c Changes in the number of tumor-informed ctDNAs with VAF ≥ 1% in cohort 1 patients with and without recurrence. Patients without recurrence (n = 3) exhibited a decrease in ctDNA, whereas those with recurrence (n = 3) showed an increase after surgery. VAF variant allele frequency, RF recurrence-free, R recurrence; Student’s t test P = 0.05. d Changes in the number of tumor-informed ctDNAs with VAF ≥ 1% in cohort 2 patients with and without tumor progression. Patients without tumor progression (n = 25) exhibited a decrease in ctDNA, whereas those with tumor progression (n = 9) showed an increase after surgery. VAF variant allele frequency, PF progression-free, P progression; Student’s t test P = 0.007. e Longitudinal ctDNA status after therapy and clinical landmarks for cohort 1 patients. NAC neoadjuvant chemotherapy, adj adjuvant chemotherapy. f Longitudinal ctDNA status after therapy and clinical landmarks for cohort 2 patients. NAC neoadjuvant chemotherapy, adj adjuvant chemotherapy. g Kaplan–Meier curve for RFS in cohort 1, stratified by ctDNA status at the initial postsurgical assessment. RFS recurrence-free survival; log-rank P = 0.034. h Kaplan–Meier curve for PFS in cohort 2, stratified by ctDNA status at the initial postsurgical assessment. PFS progression-free survival; log-rank P = 0.025.