Prognostic value of circulating tumor DNA (ctDNA) testing in patients (pts) with rectal cancer after neoadjuvant therapy (NAT) and surgery.

Abstract

212 Background: ctDNA is a highly prognostic biomarker for pts with GI malignancies. However, its utility in perioperative treatment decision-making is not established. We evaluated the association of post-NAT and post-surgical ctDNA status with outcomes in rectal cancer pts. Methods: We analyzed 1,656 longitudinal plasma samples from 257 rectal cancer pts (clinical stages I: 7%; II: 23%; III: 65%; IV: 5%). A personalized, tumor-informed assay (Signatera, Natera, Inc.) was used for ctDNA detection and quantification. In this cohort, 169 (66%) pts received NAT followed by surgery, 26 (10%) received surgery alone, and 62 (24%) had nonoperative management (NOM) after NAT. NAT included total neoadjuvant therapy (N=183), chemoradiation (N=33), and chemotherapy (N=15). Post-NAT time point (tp) was taken after completion of NAT and before surgery (surgery cohort) or between 2-16 weeks post completion of NAT (NOM cohort). Event-free survival (EFS) was defined as the interval from end of NAT to the date of clinical progression (NOM cohort). Disease-free survival (DFS) was defined as the interval from surgery to radiological recurrence. Results: The median follow-up was 619 (0-2,535) days. Pre-treatment ctDNA results were available for 50/257 pts; 94% (47/50) of whom were ctDNA+. Post-NAT tp was available for 87 pts, of whom 29.9% (26/87) were ctDNA+ and were less likely to achieve a clinical response to NAT [complete response: 8% (ctDNA+) vs. 51% (ctDNA-), p=0.0017]. Of the 87 pts with post-NAT tp, 48 underwent surgery, 39.5% (19/48) of whom were ctDNA+ post NAT and had worse pathological response [TRG score >2 (ctDNA+) vs. 0-2 (ctDNA-), p<0.001]. The remaining 39/87 pts underwent NOM, of whom 21.9% (7/39) were ctDNA+ and had significantly worse EFS compared to ctDNA- pts (HR: 7.6, 95% CI: 1.9-31, p<0.001). Among the pts who underwent surgery + NAT (N=195), ctDNA was measured in the MRD window (2-12 weeks post-surgery, prior to adjuvant therapy) for 128. Of these, 14.8% (19/128) were ctDNA+ (MRD+). Clinical recurrence was observed in 79% (15/19) of MRD+ pts and in 11% (12/109) of MRD- pts. Notably, of the MRD- pts who clinically recurred, 83% (10/12) did convert to ctDNA+ on subsequent blood draws. MRD+ pts had significantly worse DFS compared to MRD- pts (HR: 18, 95% CI: 7.9-39, p<0.0001). Similar results were observed regardless of whether the pts received NAT (DFS; NAT: HR=13, 95% CI: 5.8-31, p<0.0001; no NAT: HR=12, 95% CI:5.29-7957, p<0.001). Conclusions: Post NAT ctDNA status is associated with clinical/pathological response and survival outcomes. As such, ctDNA can potentially guide treatment decision-making in rectal cancer pts. The post-NAT tp was prognostic of recurrence in pts who had NOM. In pts receiving surgery, ctDNA status in the MRD window was highly prognostic of outcomes. Further studies are warranted to evaluate ctDNA-guided management in the post NAT and NOM settings.