Predictive value of ctDNA-based MRD combined with PET-CT and gastroscopy for treatment efficacy in locally advanced esophageal squamous cell carcinoma: A phase II exploratory clinical trial.

e16119 Background: Based on the SANO trial establishing active surveillance non-inferiority after clinical complete response (cCR) in oesophageal cancer, and prior evidence of prognostic value of ctDNA (circulating tumor DNA) in locally advanced esophageal squamous cell carcinoma (ESCC), this phase II exploratory trial aims to evaluate the combined predictive value of ctDNA, PET-CT, and gastroscopy for treatment efficacy in cCR patients after neoadjuvant therapy. Methods: Patients (pts) with locally advanced or oligometastatic ESCC (cT1-4a N+ M0/M1 limited to supraclavicular lymph node metastasis only) were enrolled. Treatment comprised induction chemotherapy (albumin-bound paclitaxel/cisplatin/capecitabine, 3 cycles) followed by concurrent radiotherapy (50 Gy) with 2 cycles of anlotinib (10 mg, p.o., qd, d1-14, q3w) and camrelizumab (200 mg, iv, d1, q3w). Post-treatment response was assessed at 4-6 weeks. A post-radiotherapy surveillance regimen integrating ctDNA-based MRD (minimal residual disease) detection, PET-CT, and gastroscopy was implemented. Patients with no evidence of disease (negative for MRD M, PET-CT P, and gastroscopy G) were classified as MPGN. Any positive finding among these three modalities were classified as MPGP. Subsequent surgery considered residual disease and patient preference. The primary endpoint was the 2-year event-free survival (EFS) rate. Secondary endpoints included pathological complete response (pCR) rate, 2-year overall survival (OS) rate, and safety. Results: As of January 14, 2026, 40 pts were enrolled (median age 63 years). Stage distribution: 47.5% III, 25.0% IVA, 20.0% IVB. After a median follow-up of 17.5 months (IQR: 11.2, 33.5), the 2-year EFS rate was 55.6% (95% CI, 38.9-79.3), with a median EFS of 34.9 months (95% CI: 21.1-NA). Of the 16 patients (40.0%) undergoing radical surgery post-neoadjuvant therapy, pCR and R0 rates were 43.8% (95% CI: 19.8-70.1) and 93.8% (95% CI: 69.8-99.8), respectively. A primary tumor ΔSUVmax reduction of ≥65% post-radiotherapy correlated with improved EFS (p = 0.0052) and OS (p = 0.0050). Patients were classified as MPGN (n = 9) or MPGP (n = 28). The 2-year EFS rate was 85.7% (95% CI: 63.3-100.0) in MPGN vs. 47.1% (95% CI: 27.3-81.1) in MPGP; 2-year OS rate was 100.0% vs. 62.1% (95% CI: 42.3-91.3), respectively. Any-grade TEAEs occurred in 92.5% of patients, most commonly esophagitis (70.0%), odynophagia (60.0%), leukopenia (32.5%), and fatigue (22.5%). Conclusions: Integrated post-treatment assessment using ctDNA-based MRD, PET-CT, and endoscopy shows predictive potential for survival in locally advanced ESCC following induction chemotherapy and concurrent radiotherapy with anlotinib and camrelizumab. These results require prospective validation in larger cohorts with longer follow-up. Clinical trial information: ChiCTR2200056920.