A meta-analysis did not find any correlation between pathologic complete response and overall survival or disease-free survival in patients with rectal cancer.
No trial-level association between pathologic complete response (pCR) and survival was observed in a systematic review and meta-analysis of randomized controlled trials that compared neoadjuvant therapies in rectal cancer, according to a study (CRD42022354557) published in JAMA Network Open.
All included studies reported pCR, which was identified as no residual tumor or lymph node metastasis, though 2 studies were excluded due to significant heterogeneity. The pooled pCR odds ratio (OR) was 1.46 (95% CI, 1.11-1.92). Overall survival (OS) was reported in 23 studies, of which 1 was excluded due to significant heterogeneity. The pooled OS HR was 0.98 (95% CI, 0.92-1.05).
pCR was plotted against OS, and on weighted regression analysis, no correlation was found between pCR and OS (β = 0.37; 95% CI, −0.98 to 1.71; P = .57).
Disease-free survival (DFS) was reported in 21 trials, and after excluding 2 due to significant heterogeneity, 19 were included in the analysis. The pooled DFS HR was 0.94 (95% CI, 0.88-1.01). On weighted regression analysis, pCR was not correlated with DFS (β = −0.84; 95% CI, −2.55 to 0.87; P = .32).
Additionally, 10 studies (40%) enrolled patients who did not undergo curative resection in survival analyses, thus leading to different sample sizes when comparing pCR and survival estimates. The cohort accounted for less than 15% of the included patients per study. After excluding these studies, pCR continued not to correlate with OS or DFS.
A subgroup analysis that assessed usage of proportional HRs in survival analysis or methods to overcome nonproportionality. After excluding 11 studies (48%) that had nonoverlapping survival curves or cited methods to overcome nonproportional HRs when Kaplan-Meier curves overlapped, pCR remained uncorrelated with OS or DFS.
Another analysis excluded 7 trials that evaluated neoadjuvant radiation alone in historical cohorts and still found that pCR was not correlated with OS or DFS.
Regarding the risk of bias, it was found to be low in 18 trials, with some concerns in 6 trials, and high in 4 trials. When the 4 trials with a high risk of bias were excluded, similar results were found for pCRs correlation with OS (β = 0.40; 95% CI, −1.00 to 1.81; P = .55) and pCRs correlation with DFS (β = −0.79; 95% CI = −2.63 to 1.04; P = .38). In a sensitivity analysis that excluded the studies contributing to significant heterogeneity, pCR was not associated with OS or DFS.
“Our trial-level analysis did not reveal a correlation between pCR and DFS or OS in rectal cancer RCTs with a high level of evidence,” wrote lead study author Kavin Sugumar, MD, a general surgery resident at Tulane University School of Medicine, with coauthors in the publication. “Our study’s findings suggest a recommendation against using pCR as a surrogate end point [SEP] for neoadjuvant therapies in rectal cancer until conclusive trial-level evidence of its association with long-term outcomes is firmly established.”
The final meta-analysis included 25 randomized controlled trials with a total of 11,882 patients. To be eligible, studies had to be randomized controlled trials with 2 or more arms, include patients with rectal cancer who underwent neoadjuvant therapy before surgery in both arms, report the pCR rate of both arms in the form of absolute numbers or ORs or with data sufficient enough to calculate them, and report survival data presented as HRs or data sufficient enough to calculate them.
Trials were excluded if they included patients who received immunotherapy, included patients with metastatic cancer, or did not have results published at the time of the literature search, from June 6, 2022, to January 3, 2024.
The trial’s primary end point was the correlation between pCR and OS. A secondary end point was the correlation between pCR and DFS.
References
Sugumar K, Lie JJ, Stucky CC, et al. Pathologic complete response and survival in rectal cancer: a systematic review and meta-analysis. JAMA Netw Open. 2025;8(7):e2521197. doi:10.1001/jamanetworkopen.2025.21197