# Controlled feeding trials to separate ultra-processing from calories

Source: https://onco.cc/ideas/idea-nl-upf-controlled-feeding-trial/  
OnCo record `idea-nl-upf-controlled-feeding-trial` (Idea). Data CC BY-NC 4.0, attribute "Data from OnCo (onco.cc)"; commercial use needs a licence.

## TL;DR

We do not know whether ultra-processed food raises cancer risk because it makes people fat, or because of something in the food itself. Feeding volunteers matched diets for a few weeks and measuring cancer-relevant biology can tell the two apart.

## Summary

The NIH inpatient trial by Hall (2019) showed people ate about 500 kcal/day more on an ultra-processed diet matched for macronutrients, supporting the adiposity route. Whether additives, emulsifiers, nitrites, sweeteners and packaging chemicals have direct effects on the gut microbiome, inflammation and DNA damage independent of energy intake is untested in humans. Controlled feeding with cancer-relevant intermediate endpoints is the only way to answer it before cohorts with 20-year follow-up mature, and it informs whether policy should target processing per se or energy density.

## Fields

- Kind: Idea
- Last checked: 2026-09-08
- Hypothesis: An ultra-processed diet matched for energy, macronutrients, fibre and weight change to a minimally processed diet produces measurable adverse changes in gut microbiome diversity, faecal genotoxicity, systemic inflammation and insulin over four weeks, indicating mechanisms beyond adiposity.
- Rationale: Cohorts cannot disentangle processing from energy intake; mechanistic feeding trials in humans are feasible, affordable and have precedent (Hall 2019). Results would sharpen or refute the case for UPF-specific regulation.
- Proposed test: Randomised crossover feeding study (n about 60) with domiciled or fully provided diets, weight-stable design, endpoints including shotgun metagenomics, faecal water genotoxicity assays, CRP, IL-6, HOMA-IR and colonic biopsy proliferation markers in a subset.
- Maturity: preclinical-evidence
- Actor: research

## Sources

- Bottleneck evidence (Prevention we already have is not deployed): Islami et al., Proportion of cancer attributable to modifiable risk factors (CA 2018): https://doi.org/10.3322/caac.21440

## Connected records

- cancers: [Colorectal cancer](https://onco.cc/cancers/colorectal/)
- fronts: [Diet, Exercise & Lifestyle](https://onco.cc/fronts/nutrition-lifestyle/), [Prevention & Risk](https://onco.cc/fronts/prevention/)
- technologies: [Dietary fibre and the gut microbiome for immunotherapy response](https://onco.cc/technologies/dietary-fibre-microbiome-io/), [Red and processed meat reduction](https://onco.cc/technologies/red-processed-meat-reduction/), [Ultra-processed food and sugar-sweetened drinks](https://onco.cc/technologies/ultra-processed-food-ssb/)
- terms: [Glycaemic index and glycaemic load](https://onco.cc/terms/glycaemic-index/), [Gut microbiome diversity and composition](https://onco.cc/terms/gut-microbiome-diversity/), [Ultra-processed food (NOVA group 4)](https://onco.cc/terms/ultra-processed-food/)
- bottlenecks: [Misinformation and unproven therapies](https://onco.cc/bottlenecks/b-misinformation/), [Prevention we already have is not deployed](https://onco.cc/bottlenecks/b-prevention-adoption/)
- key papers: [Proportion and number of cancer cases and deaths attributable to potentially modifiable risk factors in the United States](https://onco.cc/key-papers/paper-islami-ca-cancer-j-clin/)

---
JSON: https://onco.cc/api/v1/entities/idea-nl-upf-controlled-feeding-trial.json