Oncological outcomes of local excision compared with radical surgery after neoadjuvant chemoradiotherapy for rectal cancer: a systematic review and meta-analysis

Abstract

Background Low rectal cancer is conventionally managed with neoadjuvant chemoradiotherapy (CRT) followed by radical surgery (RS). In patients who refuse a stoma or are unfit for RS, an alternative approach may be the use of pre-op CRT and local excision (LE) where tumours are responsive. The aim of this systematic review is to determine whether differences exist in local recurrence (LR), overall survival (OS) and disease-free (DFS) survival between patients treated with CRT+LE and CRT+RS. Methods A literature search was performed using MEDLINE/ PubMed/Ovid databases and Google Scholar between 1946 and 2013. Studies comparing outcome following LE and RS post-CRT were included. A pooled analysis was carried out using the Mantel-Haenszel statistical (random effects) model to identify differences in LR, OS and DFS between CRT+LE and CRT+RS. Results Eight studies were suitable for pooled analyses of LR whereas five and four studies were analysed for OS and DFS, respectively. When RS was used as the reference group, LR rate was higher in the LE group. However, this was nonsignificant (odds ratio (OR) 1.29, confidence interval (CI) 0.72–2.31, p=0.40). Similarly, no difference was observed in 10-year OS (OR 0.96, CI 0.38–2.43, p=0.93) or 5-year DFS (OR 1.04, CI 0.61–1.76, p=0.89). There was evidence of publication bias in studies used for DFS. Subgroup analysis of above outcomes in T3/any N stage cancers showed no difference in LE versus RS. Conclusion In the current evidence synthesis, there was no statistical difference in the LR, OS and DFS rates observed between patients treated with LE and RS for rectal cancer post-CRT. LE post-CRT may represent a viable alternative to RS for some patients wishing to avoid RS. However, further randomised studies are required to confirm these results.