AUTHOR=Pu Jian-Yuan , Xu Wei , Zhu Qian , Sun Wei-Ping , Hu Jie-Jun , Cai Dong , Zhang Jun-Yong , Gong Jian-Ping , Xiong Bin , Zhong Guo-Chao 

TITLE=Prediagnosis ultra-processed food consumption and prognosis of patients with colorectal, lung, prostate, or breast cancer: a large prospective multicenter study

JOURNAL=Frontiers in Nutrition

VOLUME=10

YEAR=2023

URL=https://www.frontiersin.org/journals/nutrition/articles/10.3389/fnut.2023.1258242

DOI=10.3389/fnut.2023.1258242

ISSN=2296-861X

ABSTRACT=<sec id="sec1"><title>Background and aims</title><p>Whether ultra-processed food consumption is associated with cancer prognosis remains unknown. We aimed to test whether prediagnosis ultra-processed food consumption is positively associated with all-cause and cancer-specific mortality in patients with colorectal, lung, prostate, or breast cancer.</p></sec><sec id="sec2"><title>Methods</title><p>This study included 1,100 colorectal cancer patients, 1750 lung cancer patients, 4,336 prostate cancer patients, and 2,443 breast cancer patients. Ultra-processed foods were assessed using the NOVA classification before the diagnosis of the first cancer. Multivariable Cox regression was used to calculate hazard ratio (HR) and 95% confidence interval (CI) for all-cause and cancer-specific mortality.</p></sec><sec id="sec3"><title>Results</title><p>High ultra-processed food consumption before cancer diagnosis was significantly associated with an increased risk of all-cause mortality in lung (HR<sub>quartile 4 vs. 1</sub>: 1.18; 95% CI: 0.98, 1.40; <italic>P</italic><sub>trend</sub> = 0.021) and prostate (HR<sub>quartile 4 vs. 1</sub>: 1.18; 95% CI: 1.00, 1.39; <italic>P</italic><sub>trend</sub> = 0.017) cancer patients in a nonlinear dose–response manner (all <italic>P</italic><sub>nonlinearity</sub> &lt; 0.05), whereas no significant results were found for other associations of interest. Subgroup analyses additionally revealed a significantly positive association with colorectal cancer-specific mortality among colorectal cancer patients in stages I and II but not among those in stages III and IV (<italic>P</italic><sub>interaction</sub> = 0.006), and with prostate cancer-specific mortality among prostate cancer patients with body mass index &lt;25 but not among those with body mass index ≥25 (<italic>P</italic><sub>interaction</sub> = 0.001).</p></sec><sec id="sec4"><title>Conclusion</title><p>Our study suggests that reducing ultra-processed food consumption before cancer diagnosis may improve the overall survival of patients with lung or prostate cancer, and the cancer-specific survival of certain subgroups of patients with colorectal or prostate cancer.</p></sec>