The relationship between tumor cell density in the pretreatment biopsy and survival after chemotherapy in OE02 trial esophageal cancer patients.

Abstract

49 Background: Neoadjuvant chemotherapy (chemo) followed by surgery is the standard treatment for UK patients with resectable oesophageal cancer [OeC] since the OE02 trial (Allum et al. J. Clin. Oncol, 2009), yet some patients receive no benefit from chemo. We investigated whether tumour cell density (TCD) in the pre-chemo biopsy can predict patient benefit from chemo among OE02 patients. Methods: Hematoxylin and eosin stained diagnostic biopsy slides from 281 OE02 trial OeC patients reflective of the whole trial population based on patient characteristics and survival (70% adenocarcinoma, surgery alone: n=141; chemo followed by surgery: n=140) were digitized. TCD (percentage of epithelial tumor cells/viable tumor mass) was quantified by point counting and cut offs for overall survival analyses were identified. The relationship between TCD and clinicopathological data and its predictive value (heterogeneity of treatment effect) were assessed. Results: TCD ranged between 16-99% with 85% of cases in the 25-75% range. TCD was initially divided into tertiles. The treatment hazard ratio (HR) was examined within each tertile showing evidence of heterogeneity with clear benefit from chemo seen only in the middle tertile. Cut offs were refined through examination of treatment HR within TCD quintiles. Treatment HRs were 1.25 (95%confidence interval 0.66, 2.35), 1.94 (1.39, 2.71) and 0.65 (0.36, 1.18), respectively, for TCD <40% (n=46), 40-70% (n=183) and >70% (n=52), suggesting that only patients with 40-70% TCD benefit from neoadjuvant chemo (HR>1 favours chemo, heterogeneity p=0.006). Low Mandard tumor regression grade (1-3 versus 4-5) was related to a lower biopsy TCD (median: 47% versus 56%, p=0.0402). Conclusions: TCD appears to be able to predict survival benefit from neoadjuvant chemo in OeC patients and is associated with tumour regression. Our result requires validation in an independent dataset to confirm its potential clinical utility. If validated, TCD measurement in the diagnostic biopsy could be used to determine which OeC patients would benefit from neoadjuvant chemo versus those who would benefit most from management with surgery alone.