Search

Colorectal cancer (CRC) is the third most common cancer worldwide. Increasing evidence suggests that CRC should be considered a heterogeneous disease, with multiple differences between proximal and distal tumours. The immune system may, depending on the context, promote or inhibit tumour growth, and different immune cell subsets have been found to be associated with impaired or improved prognosis in CRC. The major aim of this thesis was to investigate the prognostic impact of different immune cell signatures in CRC, with particular reference to primary tumour location, and, furthermore, to perform a characterization of immune cell signatures in relation to anthropometric factors.The study cohort for Papers I-III consists of all 626 cases of CRC in the prospective, population-based cohort Malmö Diet and Cancer Study (MDCS) from 1991 up until December 31, 2008, of which tumours from 557 cases were available for tissue microarray construction, including 201 (36.2%) right-sided and 145 (26.1%) left-sided colon cancers, and 209 (37.7%) rectal cancers. Immunohistochemistry was applied to assess the density of tumour-infiltrating immune cells. For Paper IV, the analyses were restricted to the 584 cases included in the European Prospective Investigation into Cancer (EPIC) cohort, of which the MDCS forms part. Anthropometric measurements were taken at baseline. Cox proportional hazards regression models were applied to study the hazard ratios for survival, and the risk of CRC with particular immune cell compositions.Paper I shows that dense infiltration of B cells is an independent favourable prognostic factor in CRC. Paper II demonstrates that high infiltration of cytotoxic T cells is an independent favourable prognostic factor only in right-sided colon cancer, whereas high infiltration of regulatory T cells is an independent prognostic factor only in rectal cancer. Moreover, re-analysis of the data from paper I revealed that the prognostic impact of B cells is only evident in right-sided tumours.Paper III demonstrates that high expression of programmed cell deaht ligand 1 (PD-L1) on immune cells is an independent favourable prognostic factor only in patients with right-sided and left-sided colon cancer. Paper IV shows that obesity, indicated by several anthropometric factors, is associated with risk of CRC with high infiltration of B cells and cytotoxic T cells but with low infiltration of regulatory T cells in both sexes, albeit with weaker associations in women. Moreover, the results show that obesity is associated with risk of CRC with low PD-L1 expression on immune cells in men, but with high PD-L1 expression on immune cells in women.These results show that the prognostic impact of tumour-infiltrating immune cells in CRC differs according to primary tumour location. It is also demonstrated that obesity might influence the immune microenvironment of CRC. In summary, the findings indicate that primary tumour location, anthropometric factors, and sex are all important factors to include in future studies on the tumour microenvironment of CRC.

Additional file 1: Supplementary Table 1. Baseline characteristics of participants with available information on NSAID use and colorectal cancer family history. Supplementary Table 2. TRIPOD Checklist - Prediction Model Development and Validation. Supplementary Table 3. Factors considered for inclusion in the LiFeCRC score. Supplementary Table 4. LifeCRC model selection characteristics across pre-defined risk categories in the derivation and validation samples. Supplementary Table 5. Added predictive performance for age and lifestyle-based (LiFeCRC) colorectal cancer risk prediction models. Results are stratified by age groups. Supplementary Table 6. Sensitivity analysis by subgroups.

Additional file 2: Supplementary Figure 1. Workflow of risk model development and validation. Supplementary Figure 2. Survival function of average predictor values of the derivation cohort. Supplementary Figure 3. Discrimination and relative variable importance based on Random Survival Forest models for colorectal cancer prediction. Supplementary Figure 4. Random Survival Forest colorectal cancer full model calibration. Supplementary Figure 5. Model performance comparison the LiFeCRC score and a colorectal cancer risk model including only age. (a) Calibration plot of predicted 10-year colorectal cancer risk for a model that included only age and the LiFeCRC score model with additional lifestyle predictors (waist circumference, body height, daily alcohol consumption, smoking, physical activity, and daily intake of vegetables, dairy products and red meat). (b) Decision curves illustrating net benefit of prediction models for a range of colorectal cancer risk thresholds, used to decide about further treatment or intervention. Decisions curves are shown for different models: none treatment, all treatment, treatment based on the age-model, treatment based on the LiFeCRC model. Supplementary Figure 6. Predicted 10-year absolute risk of colorectal cancer for a healthy and unhealthy lifestyle. Risk across different age-groups and a constant body height of 166 cm. Unhealthy lifestyle: waist circumference of 100 cm, high daily alcohol consumption, smoker, physically inactive, 80 g daily vegetable intake, 70 g daily dairy products intake, 60 g daily processed meat intake, and 90 g daily sugar and confectionary intake. Healthy lifestyle: waist circumference of 70 cm, low daily alcohol consumption, non-smoker, physically active, 430 g daily vegetable intake, 630 g daily dairy products intake, 0 g daily processed meat intake, and 5 g daily sugar and confectionary intake. Supplementary Figure 7. Full model performance including NSAID use and family history of colorectal cancer.

BACKGROUND & AIMS: Colorectal cancer located at different anatomical subsites may have distinct etiologies and risk factors. Previous studies that have examined this hypothesis have yielded inconsistent results, possibly because most studies have been of insufficient size to identify heterogeneous associations with precision. METHODS: In the European Prospective Investigation into Cancer and Nutrition study, we used multivariable joint Cox proportional hazards models, which accounted for tumors at different anatomical sites (proximal colon, distal colon, and rectum) as competing risks, to examine the relationships between 14 established/suspected lifestyle, anthropometric, and reproductive/menstrual risk factors with colorectal cancer risk. Heterogeneity across sites was tested using Wald tests. RESULTS: After a median of 14.9 years of follow-up of 521,330 men and women, 6291 colorectal cancer cases occurred. Physical activity was related inversely to proximal colon and distal colon cancer, but not to rectal cancer (P heterogeneity = .03). Height was associated positively with proximal and distal colon cancer only, but not rectal cancer (P heterogeneity = .0001). For men, but not women, heterogeneous relationships were observed for body mass index (P heterogeneity = .008) and waist circumference (P heterogeneity = .03), with weaker positive associations found for rectal cancer, compared with proximal and distal colon cancer. Current smoking was associated with a greater risk of rectal and proximal colon cancer, but not distal colon cancer (P heterogeneity = .05). No heterogeneity by anatomical site was found for alcohol consumption, diabetes, nonsteroidal anti-inflammatory drug use, and reproductive/menstrual factors. CONCLUSIONS: The relationships between physical activity, anthropometry, and smoking with colorectal cancer risk differed by subsite, supporting the hypothesis that tumors in different anatomical regions may have distinct etiologies.

Additional file 1: Supplementary Table 1. Baseline characteristics of participants with available information on NSAID use and colorectal cancer family history. Supplementary Table 2. TRIPOD Checklist - Prediction Model Development and Validation. Supplementary Table 3. Factors considered for inclusion in the LiFeCRC score. Supplementary Table 4. LifeCRC model selection characteristics across pre-defined risk categories in the derivation and validation samples. Supplementary Table 5. Added predictive performance for age and lifestyle-based (LiFeCRC) colorectal cancer risk prediction models. Results are stratified by age groups. Supplementary Table 6. Sensitivity analysis by subgroups.

Background & Aims - Colorectal cancer located at different anatomical subsites may have distinct etiologies and risk factors. Previous studies that have examined this hypothesis have yielded inconsistent results, possibly because most studies have been of insufficient size to identify heterogeneous associations with precision. Methods - In the European Prospective Investigation into Cancer and Nutrition study, we used multivariable joint Cox proportional hazards models, which accounted for tumors at different anatomical sites (proximal colon, distal colon, and rectum) as competing risks, to examine the relationships between 14 established/suspected lifestyle, anthropometric, and reproductive/menstrual risk factors with colorectal cancer risk. Heterogeneity across sites was tested using Wald tests. Results - After a median of 14.9 years of follow-up of 521,330 men and women, 6291 colorectal cancer cases occurred. Physical activity was related inversely to proximal colon and distal colon cancer, but not to rectal cancer (P heterogeneity = .03). Height was associated positively with proximal and distal colon cancer only, but not rectal cancer (P heterogeneity = .0001). For men, but not women, heterogeneous relationships were observed for body mass index (P heterogeneity = .008) and waist circumference (P heterogeneity = .03), with weaker positive associations found for rectal cancer, compared with proximal and distal colon cancer. Current smoking was associated with a greater risk of rectal and proximal colon cancer, but not distal colon cancer (P heterogeneity = .05). No heterogeneity by anatomical site was found for alcohol consumption, diabetes, nonsteroidal anti-inflammatory drug use, and reproductive/menstrual factors. Conclusions - The relationships between physical activity, anthropometry, and smoking with colorectal cancer risk differed by subsite, supporting the hypothesis that tumors in different anatomical regions may have distinct etiologies.