Fasting-mimicking diet causes hepatic and blood markers changes indicating reduced biological age and disease risk

Abstract

In mice, periodic cycles of a fasting mimicking diet (FMD) protect normal cells while killing damaged cells including cancer and autoimmune cells, reduce inflammation, promote multi-system regeneration, and extend longevity. Here, we performed secondary and exploratory analysis of blood samples from a randomized clinical trial (NCT02158897) and show that 3 FMD cycles in adult study participants are associated with reduced insulin resistance and other pre-diabetes markers, lower hepatic fat (as determined by magnetic resonance imaging) and increased lymphoid to myeloid ratio: an indicator of immune system age. Based on a validated measure of biological age predictive of morbidity and mortality, 3 FMD cycles were associated with a decrease of 2.5 years in median biological age, independent of weight loss. Nearly identical findings resulted from a second clinical study (NCT04150159). Together these results provide initial support for beneficial effects of the FMD on multiple cardiometabolic risk factors and biomarkers of biological age.

Fig. 1. CONSORT diagram.

Consolidated Standards of Reporting Trials (CONSORT) diagram of 102 contacted study participants of which 100 were enrolled into the study two arms. Arm 1 (n = 48), the control group, maintained their normal caloric intake for a 3-month monitoring period. Data were collected at enrollment and again after 3 months. Participants in arm 2 (n = 52) started the FMD after randomization. The FMD was provided for 5 days per month for three consecutive cycles. Data were collected at enrollment, at the completion of the first FMD cycle but before resuming normal dietary intake, and also on average 5 days after study participants resumed their normal diet after the final FMD cycle. After the initial 3-month period, study participants in arm 1 also started the FMD. An optional follow-up visit in the clinic for analysis was offered to all participants about 3 months after the completion of the third FMD cycle. From Wei, Min et al. “Fasting-mimicking diet and markers/risk factors for aging, diabetes, cancer, and cardiovascular disease.” Science translational medicine vol. 9377 (2017): eaai8700. doi:10.1126/scitranslmed.aai8700. Reprinted with permission from AAAS.

Fig. 2. Fasting-mimicking decreases the hepatic fat fraction and diabetes risk markers, and increases the lymphoid to myeloid ratio.

A Body mass index BMI; N = 15 biologically independent samples. p = 0.0002; Baseline (BL)= gray bar, 3 fasting-mimicking diet (FMD) cycles= white bar. B Total fat in grams; N = 15 biologically independent samples measured by DEXA. p = 0.0021. C Subcutaneous adipose tissue (SAT) in liters; N = 15 biologically independent samples. p = 0.0084. D Visceral adipose tissue (VAT) in liters; N = 15 biologically independent samples. p = 0.0017. E Visceral adipose tissue (VAT) in liters for study participants affected by overweight and obesity; N = 10. p = 0.0030. F Hepatic fat fraction (HFF) in percent for all study participants (N = 15, p = 0.0495) and G HFF in study participants affected by overweight and obesity (N = 10, p = 0.0465) and (H) for study participants with fatty liver described as >5% HFF (N = 5, p = 0.0197). I Pancreatic fat fraction (PFF) for N = 15 study participants. p = 0.2778. J Serum glucose in pre-diabetic study participants at baseline and after 3 FMD cycles; N = 13. p = < 0.0001. K Homeostatic model assessment to measure insulin resistance (HOMA-IR) in pre-diabetic study participants; N = 11. p = 0.0455. L Hemoglobin A1c (HbA1c) levels in pre-diabetic study participants; N = 3. p = 0.0315. M The lymphoid to myeloid ratio in study participants that maintained a normal diet; N = 27. p = 0.9723 Baseline (BL)= gray bar, 3 month follow-up (3mo)= white bar. N Following 3 monthly cycles of a fasting-mimicking diet, the lymphoid to myeloid ratio was increased for all participants that completed the trial; N = 61. p = 0.0049. O The lymphoid to myeloid ratio at baseline for study participants older than 40 years at baseline (BL > 40; N = 34; dark gray bar), after completion of the 1st FMD cycle (p = 0.9585; 1FMD light gray bar) and 3–5 days of resuming normal food intake after the 3rd FMD cycle (p = 0.0356, 3 FMD white bar). Mean ± SD. Paired two-tailed t-test. *p < 0.05, **p < 0.01, ***p < 0.001. C–I measured by MRI. Source data are provided as a Source Data file.

Fig. 3. FMD reduces biological age between enrollment and completion of trial independent of weight loss.

A On average biological age decreased by approximately 1.5 years (green line) between baseline and after three FMD cycles; however, the median decrease was nearly 2.5 years (red line), while the modal decrease was just over 3.5 years (yellow line) in the 52 study participants. B Biological age at baseline and three month follow-up did not change for n = 16 control study participants. C The level of change in biological age was inversely related to the baseline difference in biological relative to chronological age, suggesting that those who had the highest aging rates at baseline showed the greatest improvements. Points are color coded such that red denotes study participants with accelerated biological ages who showed improvements following three cycles of FMD (N = 11), blue denotes study participants with decelerated biological ages who showed improvements following three cycles of FMD (N = 25), yellow denotes study participants with accelerated biological ages who showed worsening biological age following three cycles of FMD (N = 1), and green denotes study participants with decelerated biological ages who showed worsening biological age following three cycles of FMD (N = 15). D Change in biological age between baseline and completion of three cycles of FMD is not significantly associated with declines in BMI during this time (r = −0.075, p = 0.35). E This is even more apparent when limiting the sample to those who did experience reductions in BMI following 3 cycles of FMD. C–E Median-based biweight midcorrelation was used to test for association. Significance was assessed with unadjusted two-sided p-values. F Predicted median life expectancy based on chronological and biological ages was increased following three cycles of FMD. The Kaplan–Meier plot depicts predicted survival assuming an age of death equivalent to predicted median life expectancy based on each study participants biological and chronological age at Baseline and following 3 cycles of FMD. Source data are provided as a Source Data file.