Watch-and-Wait after Neoadjuvant PD-1 Blockade in Nonmetastatic dMMR/MSI-H Lower Rectal Cancer: A Phase 2 Trial (BASKET)

Neoadjuvant PD-1 blockade produces substantial tumor regression in mismatch repair-deficient or microsatellite instability-high (dMMR/MSI-H) locally advanced lower rectal cancer (LRC), but response assessment and the safety of watch-and-wait (W&W) remain uncertain. BASKET was an open-label, multicenter, single-arm, phase 2 study conducted at two Chinese centers (ClinicalTrials.gov, NCT04643041). Patients with non-metastatic dMMR/MSI-H LRC received six cycles of neoadjuvant sintilimab (200 mg every 3 weeks), followed by multimodal assessment including endoscopy, digital rectal examination, CEA, MRI, CT, and transrectal ultrasound-guided biopsy. Patients meeting protocol-defined clinical complete response (cCR) criteria entered W&W surveillance. Of 15 treated patients, 13 completed treatment and response assessment. All 13 per-protocol patients achieved cCR and entered W&W (13/13, 100%), corresponding to 13/15 (86.7%) in the intention-to-treat (ITT) population. Only 1 of 13 per-protocol patients (7.7%) achieved radiologic complete response on MRI; persistent radiographic abnormalities were adjudicated using the other response modalities. At a median follow-up of 37.7 months (range, 0.7-56.2) in the ITT population, one death occurred 0.7 months after treatment initiation, and no local recurrence or distant metastasis. The estimated 1-year disease-free and overall survival rates were both 93.3% (95% CI, 68.1%-99.8%). Among the 13 per-protocol patients no local recurrence, distant metastasis, or death was observed. No grade 3 or higher immune-related adverse events or biopsy-related complications occurred. These findings suggest that W&W may be feasible in carefully selected patients following rigorous multimodal response assessment and with close surveillance; however, the incremental value of biopsy and the oncologic equivalence of W&W to surgery remain unproven.