Serial assessment of cell-free circulating tumor DNA (ctDNA) to assess treatment effect and minimal residual disease during neoadjuvant and adjuvant therapy in colorectal cancer

Abstract

Introduction: Circulating tumor DNA (ctDNA) has demonstrated the potential to identify patients with minimal or “molecular” residual disease (MRD) who have an elevated risk of cancer recurrence following completion of definitive therapy with surgery. While neoadjuvant therapy may help mitigate this risk, patients who have post-resection MRD following neoadjuvant therapy may benefit from additional adjuvant therapy and/or close surveillance. Longitudinal ctDNA assessment after completion of each therapeutic intervention may provide additional insight into dynamic changes in ctDNA and better inform these therapy decisions. Methods: A total of 43 patients underwent neoadjuvant therapy and resection for Stage 0-IV colorectal cancer (CRC). Cell-free DNA (cfDNA) was extracted (median 27ng) from serial plasma samples collected a median of 31 days post-completion of neoadjuvant therapy (41/43 patients), post-resection (42/43), and post-completion of adjuvant therapy (13/16). Patients were followed until disease recurrence or death. cfDNA was analyzed utilizing the LUNAR assay (Guardant Health) which incorporated a genomic and epigenomic ctDNA assessment to report a “ctDNA detected” or “ctDNA not detected” result at a lower limit of detection of 0.01% variant allele fraction. This single blood sample cfDNA assay utilized a variant classifier to differentiate tumor-derived from non-tumor derived cfDNA alterations in the absence of other genomic DNA (e.g. tissue sequencing or peripheral blood mononuclear cells [PBMC]). Persistent ctDNA was defined as “ctDNA detected” following completion of therapy. Cleared ctDNA was defined as “ctDNA detected” followed by “ctDNA not detected” after completion of standard therapy. Negative ctDNA was defined as “ctDNA not detected” at all time points. Results: In total, 16/41 (39%) post-neoadjuvant, 9/42 (21%) post-resection, and 3/13 (23%) post-adjuvant samples showed detection of ctDNA. Ten of 11 patients (91%) with persistent ctDNA after surgery (n=8) or adjuvant therapy (n=3) recurred at a median (mTTR) of 150 days (range 35 - 440). One patient with persistent ctDNA who was recurrence-free had less than 6 months of follow-up (161 days). Excluding the patient with minimal follow-up, persistent ctDNA had a positive predictive value for recurrence of 100%. In 6 patients where ctDNA cleared after surgery, 3 patients (50%) recurred at a mTTR of 333 days (range 84 - 452). Median follow-up in the recurrencefree patients was 768 days (range 627 - 803). In 26 patients with negative ctDNA after surgery (n=13) or adjuvant therapy (n=14), 7 patients recurred (27%) with a mTTR not reached. In the 7 patients who recurred, mTTR was 425 days (range 128 - 517). Median follow-up in the 19 recurrence-free patients was 673 days (range 300 - 824). Patients with persistent ctDNA were significantly more likely to experience disease recurrence (P