Background: Pathological response is a critical factor in predicting long-term survival of patients with esophageal cancer after preoperative therapy. However, the validity of using pathological response as a surrogate for overall survival (OS) for esophageal cancer has not yet been established. In this study, a literature-based meta-analysis was conducted to evaluate pathological response as a proxy endpoint for survival in esophageal cancer.
Methods: Three databases were systematically searched to identify relevant studies investigating neoadjuvant treatment for esophageal cancer. The correlation between pathological complete response (pCR) and OS were assessed using a weighted multiple regression analysis at the trial level, and the coefficient of determination (R2) was calculated. The research design and histological subtypes were considered in the performance of subgroup analysis.
Results: In this meta-analysis, a total of 40 trials, comprising 43 comparisons and 55,344 patients were qualified. The surrogacy between pCR and OS was moderate (R2 = 0.238 in direct comparison, R2 = 0.500 for pCR reciprocals, R2 = 0.541 in log settings). pCR could not serve as an ideal surrogate endpoint in randomized controlled trials (RCTs) (R2 = 0.511 in direct comparison, R2 = 0.460 for pCR reciprocals, R2 = 0.523 in log settings). A strong correlation was observed in studies comparing neoadjuvant chemoradiotherapy and neoadjuvant chemotherapy (R2 = 0.595 in direct comparison, R2 = 0.840 for pCR reciprocals, R2 = 0.800 in log settings).
Conclusions: A lack of surrogacy of pathological response for long-term survival at trial level is established in this study. Hence, caution should be exercised when using pCR as the primary endpoint in neoadjuvant studies for esophageal cancer.
Keywords: Esophageal cancer; Meta-analysis; Pathological response; Surrogate endpoint; Systematic review; pCR.
© 2023. Society of Surgical Oncology.