Pathologic Complete Response and Survival in Rectal Cancer: A Systematic Review and Meta-Analysis

Abstract

Importance: Pathologic complete response (pCR) is increasingly used as a surrogate end point for survival in randomized clinical trials (RCTs) in the field of gastrointestinal oncology. Although this approach has been endorsed by the US Food and Drug Administration, it is still novel and incompletely verified.

Objective: To evaluate the trial-level association between pCR and survival in rectal cancer RCTs examining the effectiveness of neoadjuvant therapy.

Data sources: A meta-analysis was conducted to identify eligible RCTs in PubMed, EMBASE, and Cochrane databases published between database inception to January 3, 2024.

Study selection: RCTs evaluating neoadjuvant therapies in patients with rectal cancer who underwent subsequent surgical resection and reported pCR, overall survival (OS), and disease-free survival (DFS) were included.

Data extraction and synthesis: Two investigators extracted the data, which were verified by a third author. The risk of bias and the certainty of the evidence were evaluated using the Cochrane Risk of Bias Tool, version 2, and the Grading of Recommendations Assessment, Development, and Evaluation tool, respectively. Adjusted odds and hazard ratios with their 95% CIs were extracted. Weighted linear regression was used to assess the correlation between pCR and both OS and DFS.

Main outcomes and measures: Trial-level correlation between pCR and both OS and DFS.

Results: Twenty-five RCTs, including 11 882 patients, met the inclusion criteria. On meta-regression analysis, pCR was not correlated with OS (β = 0.37; 95% CI, -0.98 to 1.71; P = .57). Similarly, pCR was not correlated with DFS (β = -0.84; 95% CI, -2.55 to 0.87; P = .32). Two studies (8%) used in the meta-analysis had high risk of bias. On sensitivity analysis excluding these studies, pCR was still not associated with OS or DFS. Subgroup analysis of RCTs using total neoadjuvant therapies was not possible due to limited sample size.

Conclusions and relevance: In this systematic review and meta-analysis of RCTs comparing neoadjuvant therapies in rectal cancer, there was no trial-level association between pCR and survival. These results suggest that the use of pCR as a surrogate end point for survival should be reexamined and used more cautiously.

Figure 1.. Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) Study Flow Diagram

pCR indicates pathologic complete response.

Figure 2.. Meta-Regression Comparing Pathologic Complete Response (pCR) and Overall Survival (OS)

Meta-regression to evaluate study-level association between treatment effects on pCR and OS. Treatment effects are expressed as odds ratios (ORs) for pCR and log hazard ratios (HRs) for OS. Each circle represents a comparison of the experimental group with the control group, with the size of the circles representing the weight of the comparison, directly proportional to the number of patients in each study.

Figure 3.. Meta-Regression Comparing Pathological Complete Response (pCR) and Disease-Free Survival (DFS)

Meta-regression to evaluate study-level association between treatment effects on pCR and DFS. Treatment effects are expressed as odds ratios (ORs) for pCR and log hazard ratios (HRs) for DFS. Each circle represents a comparison of the experimental group with the control group, with the size of the circles representing the weight of the comparison, directly proportional to the number of patients in each study.

Figure 4.. Risk of Bias