Unravelling the Impact of Diet on Cardiovascular Health in Heterozygous Familial Hypercholesterolemia.

The research combines a dietary RCT (objectives 1, 2) and a cohort investigation (objective 3), both including adults with genetically-defined HeFH.

The dietary intervention will be undertaken as a crossover RCT at the Institute on nutrition and functional foods (INAF, Laval University) in Quebec City. Participants (men and women with genetically-defined HeFH) will consume in a random order a fully-controlled whole food plant-based diet low in red and processed meats and high in plant foods, reflecting Canada's Food Guide principles (WFPB diet in further sections), and a standard North American diet (SAD diet; i.e. control diet), reflecting current dietary intakes of French-Canadians, for 4 weeks each, and separated by a 4-week washout period.

Once enrolled, participants will enter a 4-week run-in period at the beginning of which they will stop taking their lipid-lowering medication for the study duration (16 weeks). Other medications (e.g. anti-hypertensive drugs, oral contraceptives) will be allowed, but doses will have to remain constant during the study. At the end of the run-in, participants will provide a blood sample and complete a self-administered, validated web-based food-frequency questionnaire. Participants will be randomly assigned to either the WFPB-SAD or SAD-WFPB diet sequence, using a computer-assisted program. The randomization (1:1) will be stratified by sex (female vs male) and LDL-receptor (LDLR) genotype (receptor-negative vs others) in 8 blocks of 8 subjects. The two diets will be consumed for 4 weeks and separated by a 2- to 4-week washout period.

During diet phases, all meals and foods will be provided to ensure optimal diet control using a 7-day cyclic menu. The WFPB diet emphasizes low intakes of red and processed meats and high intakes of whole plant-based foods with water as the drink of choice, as per Canada's Food Guide. The SAD diet reflects current dietary intakes of the French-Canadian adult population in terms of foods, nutrients, and diet quality, as characterized in recent surveys conducted in an age- and a sex-representative sample of adults from the Province of Quebec. An experienced food technician will prepare the diets. Each food and ingredient will be weighed with a precision of ±0.1 g.

During the diet phases, participants will visit the INAF clinical research unit three times weekly (Monday, Wednesday, Friday) to collect their meals and had the option to eat on-site under supervision. At that time, they will also be given their evening meals and the next day's packaged breakfast to take home. Weekend meals will be provided at the Friday visits. A checklist will be provided to all participants to identify the consumed foods. Alcohol, vitamin supplements, and natural health products will not be allowed. Tea and coffee (black) will be allowed (≤2 servings/day without daily fluctuations). Participants will be instructed to maintain their usual physical activity level. At the end of each diet, fasting plasma samples will be collected in ethylenediaminetetraacetic acid (EDTA) tubes. Parts of the samples will be stored at -80℃ for future ancillary analyses. Participants' anthropometry, body composition (DEXA scan), and blood pressure will be measured. To incorporate gender aspects associated with the intervention in knowledge transfer activities, subjects' appreciation of the diets and appetite sensation will be assessed using visual analog scales filled after each meal on day 7 and 28 of each diet.

Objective 1: To investigate the impact of a whole food plant-based diet low in red and processed meats and high in plant foods in place of a standard North American diet on LDL-C concentrations - the primary treatment target in HeFH management. This objective will output a convincing demonstration of the cholesterol-lowering effects of diet in adults with HeFH, supporting the effectiveness of dietary interventions in HeFH management. Investigators will compare post-diet LDL-C levels (primary outcome) using mixed models for repeated measures.

The following analyses will complement those on LDL-C in order to obtain a comprehensive perspective on the impact of diet on CVD risk factors and CVD risk. We will also compare post diet levels in: ApoB; Total cholesterol; Lp(a)-corrected LDL-C; HDL-cholesterol; Triglycerides; Non-HDL-C; ApoA1; Lp(a); C-reactive protein; Fasting plasma glucose; Fasting insulin; HbA1c.

All these parameters will be measured from the plasma samples collected at the end of each diet. We will also compare post-diet systolic and diastolic blood pressure, and compare post-diet ten-year risk of CVD event, using the validated FH risk score. The same statistical approaches as those used for LDL-C will be used for these analyses.

To maximize the study contribution, the following exploratory analyses will be conducted from the per-protocol dataset, to minimize the risk of confounding due to protocol deviations. All of these should be considered exploratory outcomes.

• Investigators will assess whether diet could influence HeFH detection, given that none of the participants were using cholesterol-lowering medication during the dietary intervention, by comparing the proportion of participants whose post-diets LDL-C levels will exceed the thresholds for FH investigation defined in the Simplified Canadian Definition for FH (≥4.50 mmol/L for 18 to 40 years, and >5.00 mmol/L for over 40 years). Post-diet proportions will be compared with McNemar's test.
• Investigators will assess the impact of the diets on fasting appetite hormones (total GLP-1 and peptide YY), PCSK9 and noncholesterol sterols, using linear mixed models, as described for LDL-C.
• To identify the determinants of the LDL-C response to the experimental diets, investigators will evaluate a series of potential determinants using simple linear regression models. The tested determinants will include:

i. unmodifiable factors (sex, age, LDLR genotype); ii. lifestyle-related factors (BMI, as a surrogate of energy balance51, and mean calorie intake); iii. cardiometabolic health surrogates (CRP, HOMA-IR, total fasting GLP-1, systolic blood pressure); iv. and cholesterol and lipoprotein-metabolism related factors (lathosterol/cholesterol ratio, campesterol/cholesterol ratio, and levels of PCSK9, Lp(a) and LDL-C).

For all potential determinants, investigators will use values obtained at the end of the control diet to simulate the metabolic state of individuals before adopting a healthy diet. To determine which factors are the most strongly associated with the LDL-C response to the diet, investigators will fit all possible reduced models. The model with highest adjusted R2 - the model with the best balance between goodness-of-fit and complexity - will be selected for presentation. The LDL-C response will be defined as the percent difference between post diet levels and will be treated as the dependent variable. Each potential predictor will be treated as the independent variable. These analyses will be repeated on ApoB response.

• To support the translation of the findings into clinical practice, investigators will compare the impact of the diets on perceived appetite sensations and meal appreciation, at the end of each diet, using linear mixed models. Diet will be treated as a fixed effect, while subjects will be treated as a random effect. However, we will use an a priori sex-specific approach, motivated by findings from a previous study that reported marked differences between males and females in appetite sensations and diet appreciation following a Mediterranean diet. This sex-specific analysis is intended to maximize the potential for clinical translation, regardless of sex. Thus, the models will include an interaction term. In addition, the models will be adjusted for meal, weekday, daily energy intake, age, BMI, and diet sequence.
• Investigators will evaluate changes in appetite sensations and meal appreciation between the first and last weeks of each dietary intervention, to understand how repeated exposure to the diets influences these parameters and support the translation of our findings into clinical practice. Again, an a priori sex-specific approach will be used for the aforementioned reasons. For these analyses, assessment period (first week vs last week) will be treated as a fixed effect and participants will be treated as a random effect. To determine the difference between the dietary interventions, an interaction term (assessment period * diet) will be added to the models. Two different models will be fitted according to sex. To assess how sex impacts time-related changes between the two diets, linear mixed models will be used. These models will consider the changes in appetite sensations from the first to the final week of the interventions as the dependent variable. Additionally, an interaction term between sex and diet will be included in the models to examine how the time-related effect of diet on appetite sensations might vary by sex. All these models will be adjusted for meal, weekday, daily energy intake, age, BMI, and sequence.

Objective 2: To assess the impact of a diet low in red and processed meats and high in plant foods in place of a standard North American diet on the plasma metabolome. This objective serves a dual purpose as it will: 1. inform on the dietary impact on systemic metabolism, beyond its impact on LDL-C levels, further supporting the importance of healthy dietary habits in HeFH; and 2. allow the identification of an unbiased HeFH-specific plasma metabolomic signature of a diet low in red and processed meats and high in plant foods to be used as an objective proxy of diet adherence. To document post-diet differences in plasma metabolome (secondary outcome) and identify the metabolomic signature of the WFPB diet, investigators will derive a reproducible multi-metabolite model discriminating the post-WFPB/post-SAD metabolomic profiles.

Objective 3: To evaluate the association between the plasma metabolomic signature of a diet low in red and processed meats and high in plant foods and cardiovascular disease risk. The metabolomic signature will allow to infer how objective adherence to a healthy diet is associated with cardiovascular disease risk in a distinct, large cohort of free-living adults with HeFH. This will provide an innovative demonstration of the role of diet in long-term cardiovascular disease prevention in adults with HeFH, further supporting the importance of a healthy diet in this high-risk population. To evaluate the association between the plasma metabolomic signature of the healthy WFPB diet and cardiovascular disease risk in adults with HeFH, investigators will leverage a distinct cross-sectional cohort of adults with HeFH. Upon recruitment, participants underwent a complete clinical assessment and fasting blood samples were collected in EDTA tubes. Investigators will fit and standardize the WFPB multi-metabolite model (signature) to the cohort metabolomic data. The relationship between the WFPB multi-metabolite model (independent variable) and cardiovascular disease risk factors (plasma lipids, blood pressure, plasma glucose, insulin, HbA1c) will be assessed using linear regressions.

Results will fill key gaps on the short- and long-term influence of diet on cardiovascular health in HeFH.

Inclusion criteria

• Genetically-defined HeFH
• Premenopausal women with a regular menstrual cycle for >3 months
• Postmenopausal women (without hormone replacement)

Exclusion criteria

• Any cardiovascular disease-risk exacerbating conditions (age >60 years, homozygous FH, personal history of cardiovascular disease, diabetes/anti-diabetic drug use, severe obesity, unstable body weight for >3 months, uncontrolled hypertension, genetic hypertriglyceridemia),
• Allergies/aversions to components of the experimental diets
• Any condition that would interfere with optimal participation in the intervention.