Neoadjuvant chemoradiotherapy (CRT) followed by total mesorectal excision (TME) is the standard treatment for locally-advanced rectal cancer. There is interest in de-escalating local therapy after a clinical complete response (cCR) to CRT. We hypothesized that a watch-and-wait (WW) strategy offers comparable cancer-specific survival (CSS), superior quality-adjusted survival, and reduced cost compared to upfront TME.We developed a decision-analytic model to compare WW, low anterior resection (LAR), and abdominoperineal resection (APR) for patients achieving a cCR to CRT. Rates of local regrowth, pelvic recurrence, and distant metastasis were derived from series comparing WW to TME after pathologic complete response. Lifetime incremental costs and quality-adjusted life-years (QALYs) were calculated between strategies, and sensitivity analyses were performed to study model uncertainty.The base case 5-year CSS was 93.5% (95% confidence interval [CI] 91.5 to 94.9%) on a WW program, compared to 95.9% (95% CI 93.6 to 97.4%) after upfront TME. WW was dominant relative to LAR, with cost savings of $28,500 (95% CI $22,200 to $39,000) and incremental QALY of 0.527 (95% CI 0.138 to 1.125). WW was also dominant relative to APR, with cost savings of $32,100 (95% CI $21,800 to $49,200) and incremental QALY of 0.601 (95% CI 0.213 to 1.208). WW remained dominant in sensitivity analysis unless the rate of surgical salvage fell to 73.0%.Using current multi-institutional recurrence estimates, we observed comparable CSS, superior quality-adjusted survival, and decreased costs with WW compared to upfront TME. Upfront TME was preferred when surgical salvage rates were low.
View details for DOI 10.1093/jnci/djaa003
View details for PubMedID 31930400