Colorectal Cancer Prognosis And Mortality Risk Improved By Eating Healthy Diet
Overview
The relationship between diet and colorectal cancer prognosis remains poorly understood, and there are no specific lifestyle recommendations available. A comprehensive search for randomized controlled trials (RCTs) and longitudinal observational studies examining post-diagnosis dietary factors, supplement use, and colorectal cancer survival outcomes was conducted using PubMed and Embase, covering studies up to February 28, 2022. The evidence was evaluated through random-effects dose-response meta-analyses where applicable (with a minimum of three studies), and reviewed by the CUP Global independent Expert Committee on Cancer Survivorship and Expert Panel. The analysis included five RCTs and 35 observational studies, encompassing 30,242 cases, over 8,700 all-cause deaths, 2,100 colorectal cancer-specific deaths, and 3,700 instances of progression, recurrence, or disease-free events.
Meta-analyses were conducted on various dietary factors, including whole grains, nuts/peanuts, red and processed meats, dairy products, sugary drinks, artificially sweetened beverages, coffee, alcohol, dietary glycemic load/index, insulin load/index, marine omega-3 polyunsaturated fatty acids, supplemental calcium, circulating 25-hydroxyvitamin D (25[OH]D), and their associations with all-cause mortality. Additionally, the associations between alcohol, supplemental calcium, circulating 25(OH)D, and colorectal cancer-specific mortality, as well as circulating 25(OH)D and recurrence/disease-free survival, were analyzed.
The overall quality of evidence was graded as ‘limited.’ There was ‘limited—suggestive’ evidence indicating inverse associations between healthy dietary/lifestyle patterns (including diets rich in plant-based foods), whole grains, total coffee (both caffeinated and decaffeinated), and all-cause mortality. Conversely, there were positive associations between unhealthy dietary patterns, sugary drinks, and all-cause mortality. Other exposure-outcome relationships were deemed to have ‘limited—no conclusion’ evidence.
To establish concrete lifestyle recommendations for colorectal cancer survivors, further high-quality cohort studies and meticulously designed RCTs are necessary.
Introduction
Colorectal cancer ranks as the third most commonly diagnosed cancer after lung and breast cancer and is the second leading cause of cancer-related deaths worldwide. In 2020, over 1.9 million cases and more than 900,000 deaths were attributed to colorectal cancer globally. Projections suggest that by 2040, these numbers will rise to 3.2 million new cases and 1.6 million deaths annually. Factors contributing to this increase include the adoption of Western diets and lifestyles, population growth, and longer life expectancies.
In high-income countries, the five-year relative survival rate for colorectal cancer is around 60%, whereas in lower-income countries, it averages around 40%. Improvements in detection and treatment have resulted in more survivors, especially in developed nations. In 2020, there were approximately 5.2 million individuals living within five years of a colorectal cancer diagnosis. However, extended survival often comes with increased co-morbidities, such as cardiovascular disease, which is the leading cause of non-cancer deaths among these patients. Survivors also face risks of recurrence, metastasis, or developing second primary cancers, posing significant global health and financial challenges.
Preventive strategies focusing on lifestyle changes may enhance cancer survivorship. Numerous lifestyle factors have been linked to colorectal cancer development. The World Cancer Research Fund/American Institute for Cancer Research (WCRF/AICR) Third Expert Report indicates strong evidence that consuming red meat increases colorectal cancer risk, while whole grains, dietary fiber, dairy products, and calcium supplements reduce it. There is also strong evidence that processed meat and alcohol consumption elevate the risk. Evidence on other dietary factors remains limited.
A recent umbrella review reinforced the link between alcohol intake and increased colorectal cancer risk and between the intake of calcium, whole grains, and dairy products and a reduced risk. However, the impact of dietary factors on survival is not well understood, highlighting the need for targeted dietary recommendations for colorectal cancer survivors. Currently, survivors are advised to follow general guidelines for cancer and chronic disease prevention.
Recent meta-analyses of observational studies have explored associations between post-diagnosis dietary habits and colorectal cancer survival, though they generally include few studies. These analyses revealed inverse associations between post-diagnosis intake of whole grains and coffee, adherence to American Cancer Society recommendations, and all-cause mortality, as well as between a prudent dietary pattern, calcium supplementation, and colorectal cancer-specific mortality. Conversely, unhealthy dietary patterns were linked to higher all-cause and colorectal cancer-specific mortality, and the Dietary Inflammatory Index was associated with increased all-cause mortality.
A recent meta-analysis on circulating 25(OH)D concentrations reported inverse associations with all-cause and colorectal cancer-specific mortality, though it included a mix of pre- and post-diagnosis assessments. Another meta-analysis of two randomized controlled trials on vitamin D3 supplementation showed a favorable effect on colorectal cancer progression or death. Most meta-analyses to date have focused on categorical analyses without exploring non-linear relationships.
As part of the WCRF Global Cancer Update Programme (CUP Global), comprehensive systematic literature reviews and meta-analyses were conducted to evaluate the associations of post-diagnosis dietary factors, supplement use, and colorectal cancer outcomes. The findings were independently interpreted and graded by the CUP Global Expert Committee and Expert Panel. This paper presents evidence on post-diagnosis dietary factors and colorectal cancer outcomes, while evidence on body fatness, physical activity, and an overall summary are discussed in accompanying papers.
Method
The systematic review followed a pre-published CUP Global protocol, with detailed methods and the PRISMA checklist available in Supplementary Text S1 and Table S1. PubMed and Embase databases were searched up to February 28, 2022, for studies meeting these criteria: (1) RCTs lasting at least six months, longitudinal observational studies, or pooled analyses of these designs; (2) involving at least 100 participants; (3) examining dietary exposures assessed at or after diagnosis. Dietary exposures included patterns, foods, circulating 25(OH)D, beverages, macro- and micronutrients, and dietary supplements. For simplicity, these exposures are referred to as ‘post-diagnosis’ throughout the manuscript.
Colorectal cancer outcomes examined were all-cause mortality, cause-specific mortality, progression/recurrence/disease-free survival, and any second primary cancers. In cases of multiple publications from the same study, the publication with the most outcome events was used. Studies focusing solely on nutrient-based patterns or lacking information on foods and beverages contributing to the dietary pattern were excluded.
Data on study and participant characteristics and results for each exposure-outcome association were extracted into the CUP Global database. RCTs and observational studies were reviewed separately, and relevant publications were meta-analyzed when at least three studies were available for a given exposure, or descriptively synthesized when meta-analysis was not possible.
RCTs and publications related to WCRF/AICR Cancer Prevention Recommendations were descriptively synthesized even with fewer than three studies. The results for each exposure-outcome association were summarized in text, and relative risks (RRs) comparing extreme exposure categories were presented in forest plots. Dietary and lifestyle patterns were categorized as ‘healthy’ or ‘unhealthy’ to identify any trends in associations.
Linear dose-response meta-analyses calculated summary RRs and 95% confidence intervals (CIs) from multivariable adjusted results for continuous exposures or exposures with at least three categories. A Der Simonian–Laird random-effects model was used to assess heterogeneity, which was evaluated using tau2 and I2 statistics, with 95% predictive intervals (PIs) estimating the range of values for a new study. Heterogeneity sources were explored with at least three studies in predefined subgroups (cancer subsite, subtype, exposure assessment timeframe relative to cancer treatment, and risk of bias domains). Small-study effects and publication bias were examined using Egger’s test and funnel plots for datasets with 10 or more studies. Non-linear dose-response meta-analysis was conducted using restricted cubic splines when at least five studies with three or more exposure categories were available, comparing linear and non-linear models using the likelihood ratio test. Sensitivity analyses, including leave-one-out analysis and exclusion of locally advanced and metastatic patients, were performed as appropriate.
The Cochrane risk-of-bias tool (RoB 2) was used for RCTs, while a modified version of the Risk of Bias for Nutrition Observational Studies (RoB-Nobs) tool was employed for observational studies. The RoB-Nobs tool, initially developed by the USDA Nutrition Evidence Systematic Review, was further refined by the Imperial College London review team for cancer survivorship studies. The tool assesses seven domains: confounding, participant selection, exposure classification, deviations from intended exposures, missing data, outcome measurement, and selective reporting. Studies not included in meta-analyses were descriptively assessed for potential bias sources (selection bias, exposure and outcome assessment information bias, and residual confounding).
A second reviewer verified study selection, data extraction, and risk of bias assessment, with any disagreements resolved by consensus with a third reviewer.
The CUP Global independent Expert Committee on Cancer Survivorship and the WCRF International Expert Panel interpreted the findings. The Expert Committee provided preliminary conclusions, which were finalized by the Expert Panel independently of the ICL review team. Evidence quality was graded using predefined criteria, assessing the quantity, consistency, magnitude, and precision of summary estimates, existence of dose-response relationships, risk of bias, study design, generalizability, and mechanistic plausibility, resulting in classifications of ‘strong’ (convincing, probable, substantial effect on risk unlikely) or ‘limited’ (limited-suggestive or limited-no conclusion).
Result
The systematic review included 69 publications, examining dietary factors and colorectal cancer (CRC) outcomes. The studies comprised five randomized controlled trials (RCTs) and 35 observational studies, involving 30,242 CRC cases and significant mortality and progression data. Most studies were from North America and Europe, focusing primarily on CRC patients of white ethnicity. The studies varied in stages of cancer, with a range from 100 to 2910 participants.
Randomised Controlled Trial (RCT) Findings
- Vitamin D3: High-dose versus regular-dose vitamin D3 showed no overall survival benefit but suggested improved progression-free survival.
- Vitamin C: Supplementation worsened both overall and progression-free survival.
- Omega-3 Fatty Acids: Enriched supplements resulted in worse survival outcomes.
- Nutritional Counselling: Led to longer CRC-specific survival and fewer recurrences compared to high-protein supplements or usual diet.
Observational Study Findings
Dietary Patterns
- Healthy Patterns: Generally showed an inverse association with all-cause mortality, though some estimates crossed the null.
- Unhealthy Patterns: Mostly indicated positive associations with all-cause mortality.
Specific Dietary Exposures
- Nuts and Peanuts: No clear association with all-cause mortality.
- Red and Processed Meat: No significant association with all-cause or CRC-specific mortality.
- Dairy Products: Little evidence for an association with all-cause mortality.
- Sugary and Artificially Sweetened Drinks: Sugary drinks associated with higher all-cause mortality; artificially sweetened drinks with lower mortality.
- Coffee: Lower all-cause mortality observed with both caffeinated and decaffeinated coffee intake.
- Alcohol: Suggestive of an inverse association with all-cause mortality but no association with CRC-specific mortality.
- Glycemic Load and Index, Insulin Load and Index:** Positive associations with all-cause mortality, with some estimates crossing the null.
Other Nutrients
- Marine n-3 PUFA: Marginal inverse association with all-cause mortality.
- Vitamin D (25(OH)D): Inverse associations with all-cause and CRC-specific mortality.
- Calcium: Marginal inverse association with supplemental calcium and all-cause mortality.
Risk of Bias
- RCTs often had a high risk of bias due to blinding issues.
- Observational studies faced challenges with incomplete adjustments for confounders and selection bias.
Overall, the review suggests that certain dietary patterns and specific nutrients may influence survival outcomes in CRC patients, though further research with rigorous methodologies is needed to confirm these associations.
Conclusion
The systematic review analyzed evidence from randomized controlled trials (RCTs) and longitudinal observational studies to understand the impact of post-diagnosis dietary factors and supplement use on colorectal cancer survival. Twenty-one associations were meta-analyzed, while others, including dietary and lifestyle patterns, were descriptively reviewed. The quality of evidence was independently graded.
Various post-diagnosis dietary and lifestyle patterns were examined, categorized broadly into ‘healthy’ and ‘unhealthy’ patterns. ‘Healthy’ patterns, characterized by high intakes of fruits, vegetables, whole grains, nuts, and legumes, and low intakes of red and processed meats, were generally inversely associated with all-cause mortality. Conversely, ‘unhealthy’ patterns, which included high intakes of refined grains, red and processed meats, sugary drinks, and low coffee consumption, were positively associated with all-cause mortality. The associations could be due to the cumulative or synergistic effects of individual components.
Studies incorporating ‘healthy body weight’ and high physical activity in dietary patterns reported inverse associations with all-cause mortality, similar to those based solely on dietary components. However, the association between healthy patterns and lower mortality risk might be driven by either dietary and lifestyle components or primarily by physical activity. A unique finding was an inverted J-shaped relationship between body mass index (BMI) and all-cause mortality, with the lowest risk at 28 kg/m², differing from the conventional healthy BMI of 18.5–24.9 kg/m².
Biological mechanisms linking dietary patterns and cancer progression are not well understood. Adherence to healthy dietary patterns has been associated with reduced markers of inflammation, oxidative stress, and insulin resistance. Western diets, conversely, are linked to higher levels of inflammation and hyperinsulinemia. Colorectal cancer patients consuming Western diets and high glycaemic loads may experience elevated insulin levels, potentially facilitating tumor recurrence and higher mortality risks. Molecular epidemiology studies are needed to clarify these mechanisms and the impact of dietary factors on cancer recurrence and survival.
Higher intakes of whole grains and coffee were associated with lower all-cause mortality risk. Whole grains, rich in dietary fiber, might reduce cancer risk by decreasing exposure to intestinal carcinogens and promoting beneficial gut microbiota. Coffee, with its antioxidant, anti-inflammatory, insulin-sensitizing, and anti-tumor properties, was linked to reduced all-cause mortality, regardless of caffeine content. The inverse association was particularly noted among stage III cancer survivors.
Higher sugary drink intake was associated with increased all-cause mortality, likely due to effects on energy metabolism, insulin resistance, lipid metabolism, inflammation, and immune function. The potential inverse association between artificially sweetened beverages and mortality warrants further investigation due to limited study numbers.
Marginal inverse associations were observed for higher post-diagnosis alcohol consumption and circulating 25(OH)D levels, likely influenced by reverse causation. Studies rated as having moderate or serious bias did not support stronger conclusions for these associations despite the anticancer properties of vitamin D.
Other dietary factors, such as nuts, red and processed meat, dairy products, marine n-3 PUFAs, dietary glycaemic load, and insulin indices, showed null or inconsistent results. The review highlighted the need for more detailed studies on dietary supplements’ types, dosages, and interactions with cancer treatments to provide clearer conclusions.
Cancer survival studies face inherent methodological challenges, including selection bias and reverse causation. Future studies should use standardized dietary assessments, perform time-varying analyses, and include diverse populations to improve the robustness of findings.
In conclusion, there is ‘limited-suggestive’ evidence that healthy dietary patterns, whole grain intake, and coffee consumption post-diagnosis are associated with lower all-cause mortality, while unhealthy dietary patterns and sugary drinks are linked to higher mortality risk. The current evidence, though not strong enough for definitive guidelines, can inform future research and potentially contribute to colorectal cancer survivors’ lifestyle recommendations. Further well-designed observational studies, mechanistic research, and RCTs are needed to strengthen the evidence base and guide clinical recommendations.
Oncology Related Tools
Other
Latest Research
