Transanal local excision reduces surgical burden without evidence of increased recurrence in lower rectal GIST: a multicenter inverse probability of treatment weighting analysis.

Abstract

Background: Lower rectal gastrointestinal stromal tumors (rGISTs) pose a surgical challenge, as complete resection must be balanced against anorectal function preservation. Transanal local excision (taLE) may offer a less invasive alternative to transabdominal laparoscopic radical resection (tlRR), but comparative evidence, particularly for larger tumors, is limited.

Methods: We conducted a multicenter retrospective cohort study of patients with primary lower rGIST (lower tumor margin ≤ 5 cm from the anal verge; tumor size ≥ 2 cm and < 7 cm) undergoing taLE or tlRR (April 2015-April 2023). Inverse probability of treatment weighting (IPTW) balanced baseline characteristics, and doubly robust models adjusted residual imbalance. Perioperative outcomes, recurrence-free survival (RFS), and a prespecified ≥ 5 cm subgroup analysis were evaluated.

Results: Among 101 patients (taLE, n = 74; tlRR, n = 27; median follow-up, 40 months), IPTW-adjusted analyses showed significantly reduced surgical burden with taLE: shorter operative time (- 118.4 min, P < 0.001), lower postoperative pain (- 2.5, P < 0.001), earlier drain removal (- 5.1 days, P = 0.008), no permanent stoma (0/74 vs. 5/27), and lower odds of temporary stoma (OR 0.05, P < 0.001) and reoperation (OR 0.02, P = 0.009). These advantages remained largely consistent in the ≥ 5 cm subgroup. Only four recurrences occurred, with no significant RFS difference (log-rank P = 0.43); power was limited.

Conclusions: taLE was associated with substantially reduced surgical burden in selected patients with lower rGIST, and no significant difference in recurrence-free survival was observed during available follow-up. However, with only four events, the study was underpowered to establish oncologic equivalence. The taLE-based organ-preserving pathway may represent a promising option pending confirmation in larger cohorts.