Background: Rectal cancer is common and frequently treated with neoadjuvant radiotherapy prior to surgery to reduce the risk of tumour recurrence. However, the therapeutic benefits and side effects of radiotherapy can vary between patients, and there are currently no validated biomarkers to predict treatment response. Tumour cell density (TCD) and tumour-infiltrating lymphocyte (TIL) density are proven prognostic biomarkers in colorectal cancer; however, their utility in predicting radiotherapy response remains unclear. We assessed the prognostic and predictive value of TCD and TIL density in rectal cancer patients treated with radiotherapy. Methods: TCD was quantified using a manual point-counting method in 253 pre-treatment biopsies and across the entire tumour area of 569 resection specimens from the MRC CR07 clinical trial, which randomised patients to either neoadjuvant short-course radiotherapy (SCRT) or straight to surgery (control). TIL density was measured in 102 biopsies and matched resection specimens (73 SCRT, 29 control) across different tumour areas using deep learning-based cell detection in MIM (HeteroGenius Ltd., Leeds, UK). Cutoffs for low/high-TCD and TIL density were both pre-defined and derived from survival data using the survminer R package. Survival analyses were performed to evaluate the predictive and prognostic value of TCD/TIL in relation to overall and cancer-specific survival. Results: TCD in the resection specimens was lower in the SCRT group (19.9%, IQR 12.9–26.7%) than the control group (34.3%, IQR 27.7–40.5%, p < 0.001). In control resections, low-TCD was associated with a higher risk of all-cause mortality (HR 2.20, 95% CI 1.41–3.44, p < 0.001) and cancer-related death (HR 2.69, 95% CI 1.41–5.13, p = 0.0026). In contrast, after SCRT, low resection TCD was associated with a reduced risk of death (HR 0.63, 95% CI 0.40–0.98, p = 0.04). In the SCRT group, low biopsy TCD prior to radiotherapy was associated with a reduced risk of cancer-related death (HR 0.34, 95% CI 0.13–0.89, p = 0.028). Across both trial arms, TIL density was higher in pre-treatment biopsies than resections (2492 vs. 1304/mm2, p < 0.001). Low biopsy TIL density was associated with an increased risk of all-cause mortality (HR 2.43, 95% CI 1.24–4.76, p = 0.01). The SCRT group had lower TIL density in the resection compared with controls (1210 vs. 1615/mm2, p < 0.001), and low resection TIL density across the whole tumour area was associated with a higher risk of death (HR 2.55, 95% CI 1.11–5.87, p = 0.027). Conclusions: Our findings support the role of TCD and TIL density as quantitative biomarkers in rectal cancer patients. TCD can be used to assess the degree of response to radiotherapy, and contrasting survival associations are observed between straight-to-surgery and SCRT-treated patients. This study raises the possibility of using TCD as both a prognostic and predictive biomarker. TIL density failed to show predictive value but demonstrated expected prognostic associations.
CORE (COnnecting REpositories)
CORE (COnnecting REpositories)