Association between ultra-processed foods and risk of cancer: a systematic review and meta-analysis

Background

Despite increasing evidence that has shown the association of ultra-processed foods (UPFs) with cancer risk, the results remain inconclusive. We, therefore, conducted the meta-analysis to clarify the association by including recently published studies.

Methods

A comprehensive search was conducted in PubMed, Embase, and Web of Science to identify all relevant studies from inception to January 2023. To pool data, fixed-effects or random-effects models were used where appropriate. Subgroup analyses, sensitivity analyses, and publication bias tests were performed.

Results

A total of 13 studies (4 cohort studies and 9 case-control studies) were included in the analysis, with a total of 625,738 participants. The highest UPFs consumption was associated with increased risk of colorectal cancer (OR = 1.23, 95% CI: 1.10-1.38), colon cancer (OR = 1.25, 95% CI: 1.14-1.36), and breast cancer (OR = 1.10, 95% CI: 1.00-1.20) but not rectal cancer (OR = 1.18, 95% CI: 0.97-1.43) and prostate cancer (OR = 1.03, 95% CI: 0.93-1.12). In addition, the subgroup analyses showed that a positive association between UPFs consumption and colorectal cancer was observed among men (OR = 1.31, 95% CI: 1.15-1.50), whereas no significant association was observed among women (OR = 1.10, 95% CI: 0.94-1.29).

Conclusion

The present meta-analysis suggests that high UPFs consumption is associated with a significantly increased risk of certain site-specific cancers, especially the digestive tract and some hormone-related cancers. However, further rigorously designed prospective and experimental studies are needed to better understand causal pathways.

Ultra-processed food consumption is associated with increase in fat mass and decrease in lean mass in Brazilian women: A cohort study

Objective

To investigate the association between ultra-processed food consumption at 23–25 years of age and measurements of body composition–fat mass, fat mass distribution and lean mass at 37–39 years of age in Brazilian adults.

Methods

1978/1979 birth cohort study conducted with healthy adults from Ribeirão Preto, São Paulo, Brazil. A total of 1,021 individuals participated in the fat mass analysis (measured by air displacement plethysmography) and 815 in the lean mass analysis and fat mass distribution (assessed by dual energy X-ray absorptiometry). Food consumption was evaluated by a food frequency questionnaire. Food items were grouped according to the level of processing as per the NOVA classification. Ultra-processed food consumption was expressed as a percentage of total daily intake (g/day). Linear regression models were used to estimate the effect of ultra-processed food consumption (g/day) on body mass index, body fat percentage, fat mass index, android fat, gynoid fat, android-gynoid fat ratio, lean mass percentage, lean mass index and appendicular lean mass index. Marginal plots were produced to visualize interactions.

Results

The mean daily ultra-processed food consumption in grams was 35.8% (813.3 g). There was an association between ultra-processed food consumption and increase in body mass index, body fat percentage, fat mass index, android fat and gynoid fat and decrease in lean mass percentage, only in women.

Conclusion

A high ultra-processed food consumption is associated with a long-term increase in fat mass and a decrease in lean mass in adult women.

Is there any putative mediatory role of inflammatory markers on the association between ultra-processed foods and resting metabolic rate?

The resting metabolic rate (RMR) represents the largest component of total daily energy expenditure. The sale of ultra-processed foods (UPF) is increasing globally; however, UPF can have many adverse effects, including increasing inflammatory markers and altering RMRs. This cross-sectional study included 285 healthy overweight and obese women. Anthropometric measurements were evaluated using a bioelectrical impedance analyzer InBody 770 scanner. High-sensitivity C-reactive protein (hs-CRP), plasminogen activator-1 (PAI-1), monocyte chemoattractant protein (MCP-1), and interleukin-1 beta (IL-1β) blood levels were measured after a 12-h fasting. Indirect calorimetry was used to evaluate the RMR by using the Weir equation, and RMR deviation (RMR estimated - RMR actual), RMR per body mass index (BMI), and free fat mass (FFM) were estimated. A validated food frequency questionnaire (FFQ) was used, and seven groups of UPFs were extracted based on the NOVA method. A negative association between the RMR [β = -0.159, 95% confidence interval (CI): -0.471, -0.052, P = 0.044], RMR per BMI (β = -0.014, 95% CI: -0.025, -0.006, P = 0.036), and RMR per FFM (β = -0.241, 95% CI: -0.006, -0.000, P = 0.041) using the NOVA score was observed after adjusting for confounders. This association disappeared after inclusion of each inflammatory marker. All the markers may inversely mediate the relationship between the mentioned variables and the NOVA score. hs-CRP and MCP-1 also had a negative effect on the relationship between the NOVA score and RMR deviation. Finally, UPF intake is likely related with the RMR, mediated through changes in the production of hs-CRP, PAI-1, MCP-1, and IL-1β.