A multicentre cohort study assessing the utility of routine blood tests as adjuncts to identify complete responders in rectal cancer following neoadjuvant chemoradiotherapy

Abstract

Purpose: Management of rectal cancer with a complete clinical response (cCR) to neoadjuvant chemoradiotherapy (NACRT) is controversial. Some advocate "watch and wait" programmes and organ-preserving surgery. Central to these strategies is the ability to accurately preoperatively distinguish cCR from residual disease (RD). We sought to identify if post-NACRT (preoperative) inflammatory markers act as an adjunct to MRI and endoscopy findings for distinguishing cCR from RD in rectal cancer.

Methods: Patients from three specialist rectal cancer centres were screened for inclusion (2010-2015). For inclusion, patients were required to have completed NACRT, had a post-NACRT MRI (to assess mrTRG) and proceeded to total mesorectal excision (TME). Endoluminal response was assessed on endoscopy at 6-8 weeks post-NACRT. Pathological response to therapy was calculated using a three-point tumour regression grade system (TRG1-3). Neutrophil-lymphocyte ratio (NLR), platelet-lymphocyte ratio (PLR), serum albumin (SAL), CEA and CA19-9 levels post-NACRT (preoperatively) were recorded. Variables were compared between those who had RD on post-operative pathology and those with ypCR. Statistical analysis was performed using SPSS (version 21).

Results: Six hundred forty-six patients were screened, of which 422 were suitable for inclusion. A cCR rate of 25.5% (n = 123) was observed. Sixty patients who achieved cCR were excluded from final analysis as they underwent organ-preserving surgery (local excision) leaving 63 ypCR patients compared to 359 with RD. On multivariate analysis, combining cCR on MRI and endoscopy with NLR < 5 demonstrated the greatest odds of ypCR on final histological assessment [OR 6.503 (1.594-11.652]) p < 0.001]. This method had the best diagnostic accuracy (AUC = 0.962 95% CI 0.936-0.987), compared to MRI (AUC = 0.711 95% CI 0.650-0.773) or endoscopy (AUC = 0.857 95% CI 0.811-0.902) alone or used together (AUC = 0.926 95% CI 0.892-0.961).

Conclusion: Combining post-NACRT inflammatory markers with restaging MRI and endoscopy findings adds another avenue to aid distinguishing RD from cCR in rectal cancer.

Keywords: MRI; Neoadjuvant chemoradiotherapy; Neutrophil–lymphocyte ratio; Pathological response; Rectal cancer.