Liquid and solid fructose intake and colorectal cancer mortality by physical activity in U.S. prospective cohorts

Background : We recently found that liquid fructose promotes colorectal cancer (CRC) metastasis in preclinical models, including human cancer cell lines and mice (Feng et al., 2025). In humans, fructose is consumed in both liquid and solid forms (Li et al., 2023), but whether these sources relate differently to CRC mortality is unknown. Additionally, physical activity may modify the metabolic effects of fructose (Tappy & Rosset, 2019). Methods : We assessed pre- and post-diagnostic liquid and solid fructose intakes using quadrennial food frequency questionnaires in the Nurses' Health Study (1984-2018) and Health Professionals Follow-Up Study (1990-2018). Nutrient intakes were energy-adjusted using the residual method. Post-diagnostic fructose intake was assessed at least six months but no more than four years after diagnosis. Liquid fructose was derived from sugar-sweetened beverages and juice, and solid fructose was calculated as total minus liquid fructose. Cox proportional hazards models were used to estimate hazard ratios (HRs) for fructose intake and CRC-specific mortality, stratified by pre-diagnostic physical activity (<11.5 vs. ≥11.5 metabolic equivalent of task [MET]-hrs/wk) and adjusted for demographics, tumor characteristics, lifestyle factors, and diet. Results : Among 724 CRC patients with low pre-diagnostic physical activity (<11.5 MET-hrs/wk), we observed 96 CRC-specific deaths over 7,412 person-years. Higher post-diagnostic, but not pre-diagnostic, liquid and solid fructose intakes were associated with higher CRC-specific mortality. Post-diagnostic liquid fructose showed the strongest association with CRC mortality (Multivariable [MV]-adjusted HR per 5 g/day, 1.18; 95% CI, 1.04-1.34; P-value, 0.011), whereas post-diagnostic solid fructose showed a more modest association (MV-adjusted HR per 20 g/day, 1.04; 95% CI, 1.01-1.06; P-value, 0.002). Among 705 patients (6,611 person-years; 95 CRC-specific deaths) with high pre-diagnostic physical activity (≥11.5 MET-hrs/wk), neither liquid nor solid fructose intake, pre- or post-diagnostic, was associated with CRC-specific mortality. These findings showed that the positive associations between fructose intake and CRC mortality were observed for individuals with low physical activity, confined to post-diagnostic fructose intake, and stronger for liquid than for solid fructose. Conclusions : Higher post-diagnostic fructose intake, especially from liquid sources, was associated with increased CRC-specific mortality among patients with low physical activity. These findings suggest a potential interaction between physical activity and fructose intake in the context of CRC mortality and highlight the importance of food choices in CRC survivorship.