1.1 Highly effective CFTR modulator therapies

CF is caused by mutations in the CFTR gene, which encodes for an anion channel on the apical surface of epithelial cell membranes. CFTR dysfunction causes impaired ion and fluid transport, impacting multiple organ systems.⁴ Multiple mutations in CFTR exist, which can lead to variable functional defects in the CFTR protein. Currently, six classes of mutation are described. In class I mutations, truncated RNA is degraded before CFTR protein production. Class II mutations cause protein trafficking defects-- this class includes the most common CF causing mutation, F508Del mutation. In class III, or “gating” mutations, CFTR is trafficked to the apical surface of the cell membrane, but the ion channel is defective. Class IV–VI mutations in CFTR lead to protein that is present at the cellular surface but in insufficient quantities or with decreased conduction.⁵ Currently approved CFTR modulators are small molecules that act as either correctors, which improve protein folding and allow trafficking to the cell membrane, or potentiators, which improve the function of the ion channel at the cellular surface.⁶ Ivacaftor is a CFTR potentiator, and is approved for PwCF with 97 different CFTR mutations ages 1 month and above.^{3, 7} Class I and II mutations are not eligible for ivacaftor monotherapy, as protein is not present at the cell surface for potentiation. CFTR correctors can be used in combination with a potentiator for class II mutations (including F508Del).⁶ While multiple corrector-potentiator combination therapies exist, elexacaftor-tezacaftor-ivacaftor (ETI) has demonstrated superior efficacy and thus is considered to be a highly effective modulator therapy (HEMT), along with ivacaftor monotherapy, depending on an individual's mutations.⁸ With the FDA approval of ETI for PwCF ages 6 and above, nearly 90% of people with CF are now eligible for an HEMT.¹

With improved longevity and overall health in CF, nutritional status remains paramount for the achievement of improved overall health and quality of life. However, the definition of “optimal nutritional status” must evolve to meet the current needs of people with CF. It is essential to not only clarify and redefine the relationships between nutritional status and pulmonary function but also to understand the impact of HEMT on metabolic and nutritional outcomes.

1.2 Anthropometrics and body composition in CF

Nutritional status is the balance between an individual's dietary intake of nutrients, and nutrient needs for growth, reproduction, normal cellular function, health, and activities of daily living. Adequate nutritional status is indicated by multiple metrics including calorie and nutrient intake, assessment of nutrition biomarkers, and body composition. In clinical practice, the primary indicator of nutritional status in people with CF is body mass index (BMI), a calculated metric of weight relative to height, although weight-for-age is also associated with pulmonary outcomes.^{1, 9-11} The link between BMI and clinical outcomes has guided the nutritional management of people with CF, with a goal BMI of ≥50th percentile for age in children, ≥22 kg/m² in adult females, and ≥23 kg/m² in adult males. CF guidelines have advocated for achievement of these BMI goals through a high calorie (110%–120% of daily recommended intake) and high fat (35%–40% calories from fat) diet, and supplementation of pancreatic enzymes and fat-soluble vitamins.¹¹

While BMI is clinically used as a surrogate for fat-free mass (FFM), in reality, BMI correlates positively with both fat mass and FFM. The distinction between BMI and FFM is of great importance in CF, as lean mass is a stronger correlate of lung function (percent predicted of forced expiratory volume in 1 min, FEV₁pp) as compared to BMI and fat mass.^{12, 13} Unfortunately, individuals with CF have been shown to have lower FFM than healthy controls without CF, while maintaining similar amounts of fat mass.^14-17 Thus, BMI has a low sensitivity for detecting FFM depletion in CF.¹⁴ Lower FFM has been found to be associated with lower FEV₁pp and increased frequency of pulmonary exacerbations in CF, even when BMI is within a healthy range.^{12, 13, 17, 18} Further, whereas BMI positively correlates with lung function in CF, two independent studies have shown percent body fat to inversely correlate with lung function in people with CF.^{17, 18} Thus, it is likely that results indicating better lung function in people with CF who are overweight or obese based on BMI are likely driven by lean mass, as opposed to adiposity.^{19, 20}

1.3 HEMT and anthropometrics

HEMT have been associated with increases in weight and BMI in people with CF.²¹ In a phase 3, randomized, double-blind, placebo-controlled trial designed to evaluate the efficacy of ivacaftor in 161 people with CF > 12 years of age with at least one copy of the G551D gating mutation, subjects in the ivacaftor group had gained on average 2.7 kg more weight than placebo after 48 weeks of treatment.²² Similar results were seen in a randomized, double-blind, placebo-controlled trial of youth ages 6–11 years with G551D mutations, treated with ivacaftor for 24 weeks.²³ In both adult and pediatric studies, differences between treatment and placebo groups were seen by Day 15.²⁴ In an observational study of young children ages 4–24 months with gating mutations, changes in weight-for-age Z-scores were again seen at both 6 and 12 weeks as compared to baseline.²⁵ Additionally, in pediatric patients, ivacaftor has been associated with increases in height Z-score and height velocity.^26-28

Use of ETI has also been associated with significant increases in weight and BMI compared to placebo.^{8, 29, 30} In a phase 3, randomized, double-blind, placebo-controlled trial designed to test the safety and efficacy of ETI in subjects 12 years and older with F508Del CFTR mutations, BMI was higher at 24-week follow-up as compared to baseline with a mean treatment difference of 1.04 in the treatment group.⁸ In a phase 3, randomized, double-blind, active-controlled study in which the triple combination therapy (ETI) was compared to dual combination therapy (tezacaftor-ivacaftor) in subjects homozygous for F508Del mutation, treatment with ETI was associated with a least squares mean increase in BMI of 0.6 kg/m² and a least squares mean body weight increase of 1.6 kg compared to tezacaftor-ivacaftor after 4 weeks.²⁹ In a prospective observational study of 487 people with CF ages 12 and older, BMI increased by a mean of 0.42 kg/m² in adults, and BMI Z-score increased by a mean of 0.11 in youth.³¹ An open-label, phase 3 study of ETI demonstrated increases in weight Z-score and BMI Z-score at 24 weeks of therapy, compared to baseline, which were maintained over a 96-week open-label extension.^{32, 33} While the rate of weight and BMI increase slowed during the extension period, there was a slight increase in height Z-score score during the extension period.³³

1.4 HEMT and body composition

Changes in body composition have also been seen with HEMT. Two different longitudinal studies of ivacaftor have shown similar increases in fat mass associated with treatment, but only one of the two studies found changes in FFM.^{34, 35} Interestingly, in a single-center, double-blind placebo-controlled 28-day crossover study of ivacaftor in adults, followed by a 5-month open-label extension, changes in FFM occurred earlier in the course of treatment (between baseline and Day 28), and changes in fat mass occurred between 28 days and 5 months.³⁶ Additionally, after 1 year of lumacaftor-ivacaftor in adults, increases in fat mass without changes in FFM were seen.³⁷

Although several are pending, fewer published studies are available assessing changes in body composition following treatment with ETI. In a small, single-center pilot study of adolescents and adults, 2 years of ETI treatment was associated with increased lean body mass, % fat Z-score, and fat mass/height² Z-score.³⁸

2.1 Energy balance

Energy expenditure is higher in individuals with CF as compared to the general population, with baseline needs of youth with CF best approximated using the estimated energy expenditure formula at the active level.^{39, 40} In CF, resting energy expenditure (REE) is associated with pancreatic function, but the relationship with pulmonary function is less clear.^{10, 41} While some studies have found that REE does not change during pulmonary exacerbation, or correlates with pulmonary function, others have found higher REE before treatment of pseudomonas aeruginosa infection that decreased after therapy.^42-44

HEMT may impact REE in people with CF. Stallings et al. found higher weight and lower REE in a cohort of 23 youth and adults (ages 5–61) with gating mutations after 3 months of treatment with ivacaftor, without any changes in energy intake.³⁴ Conversely, in another prospective study of 18 youth and young adults with gating mutations treated with ivacaftor, no significant differences were seen in REE between baseline and 6, 12, and 24 months in the cohort as a whole. Interestingly, the three subjects with the highest REE at baseline had the largest declines at 24-month follow-up suggesting the possibility that the degree of change in REE may be related to severity of baseline hypermetabolism.³⁵ In 15 children ages 4–24 months, sleeping energy expenditure was not different between baseline and 12-week follow-up following ivacaftor therapy, but was normal at both time points and energy intake was higher at 12 weeks compared to baseline.²⁵ In adults treated with ETI, both energy expenditure and intake were lower at follow up as compared to baseline.³⁰

2.2 Intestinal pH

In CF, abnormal pancreatic bicarbonate secretion leads to delayed alkalization of intestinal contents in the small bowel, and thus decreased lipase activity in the duodenum, as a pH greater than 4 is required for lipase to remain active.^{45, 46} In a small cohort of individuals treated with ETI, the time to reach and sustain a duodenal pH of >5.5 was lower after 1 month of ivacaftor as compared to baseline.⁴⁷ Studies examining changes in intestinal pH and motility following ETI are pending.⁴⁸

2.3 Pancreatic function

Exocrine pancreatic insufficiency is seen in approximately 85% of individuals with CF by age 12 months, and changes in pancreatic function with HEMT have been shown, thus far, to be highly age-dependent.⁴⁹ In a CF ferret model, treatment with ivacaftor of pregnant mothers led to pancreatic sufficiency in offspring.⁵⁰ In humans, ETI exposure in utero for infants with CF has been associated with lower immunoreactive trypsinogen values at birth.⁵¹ Studies of ivacaftor in infants and young children have shown higher fecal elastase and lower immunoreactive trypsinogen (trending toward normal values) after treatment with ivacaftor.^{27, 28} In children greater than 6 years of age with pancreatic insufficiency, enrolled in a 96-week observational study of ETI, there was a small increase in fecal elastase seen over the observational period, yet only one subject in the study attained an elastase of >200.³³ In adults, an increase in coefficient of fat absorption has been seen after 3 months of ivacaftor.³⁴

Changes in surrogate markers of nutrient absorption, such as improvements in micronutrient levels, have been observed following HEMT. In adults >18 years of age, higher levels of fat-soluble vitamin levels were shown after treatment with ETI as compared to baseline.⁵² In a separate study, serum 25-hydroxy vitamin D levels (the primary indicator of vitamin D status) significantly increased following 1 year of ETI therapy.⁵³ Such improvements in fat-soluble vitamin levels are likely a consequence of increased absorption.

2.4 Intestinal microbiome

CF is associated with reduced diversity of intestinal bacteria, and intestinal inflammation which are not directly correlated with pancreatic function.^{54, 55} In addition to alterations in the bacterial composition of the gut, the outcomes of metabolic functions performed by the gut microbiome have also been shown to be different in CF as compared to healthy controls.⁵⁶

Limited treatment effect on the gut microbiome has been seen with ivacaftor. While Ooi et al. found an increase in the relative abundance of Akkermansia, a mucin-degrading bacteria associated with improved gut health, after 6 months of ivacaftor, no changes in alpha or beta diversity were seen.⁵⁷ Pope et al. did not observe any changes in the intestinal microbiome in 12 participants with pancreatic sufficient CF and an R117H mutation, treated with ivacaftor. Of note, these participants were simultaneously enrolled in a study examining the effect of antibiotic therapies, which may have also impacted the microbiome during that time.⁵⁸ Ronan et al. examined the effects of ivacaftor on individuals with CF and G551D mutations over 1 year and also did not find any significant change in the composition of the GI microbiome with treatment.⁵⁹ In a study of the fecal microbiome and metabolome in 18 individuals with gating mutations, ages 6–61 years, at baseline and after 3 months of treatment with ivacaftor, no changes were seen in the microbiome, yet modest differences in the fecal lipidome were appreciated between time points.⁶⁰ Despite minimal changes in the intestinal microbiome, inflammation as measured by fecal calprotectin has been found to be lower after ivacaftor treatment as compared to baseline.^{34, 57}

3.1 Prevalence of overweight and obesity

Overweight and obesity are emerging problems in CF, predating the widespread availability of HEMT, with 17% of individuals with CF meeting the criteria for overweight and 7% meeting the criteria for obesity in 2019.¹ Additionally, the median weight percentile has increased with each new birth cohort since 1992.¹ Between 2000 and 2019, there was a >300% increase in overweight status and >400% increase in obesity in CF.⁶¹ In a registry analysis of people with CF ages 2 and above between 2000 and 2019, overweight and obesity in CF were associated with less severe CFTR mutations (class IV and V), better pulmonary function, older age, living in a zip code with median income less than $20,000, use of any CFTR modulator, and pancreatic sufficiency.⁶¹ Notably, this analysis predated the availability of ETI. Given the known impact of HEMT on weight and nutritional status, it is likely the trend toward increased prevalence of overweight and obesity in CF will be accelerated with the widespread use of this therapy. In a 5-year clinical effectiveness study of ivacaftor, the prevalence of overweight and obesity in adult participants increased from 16% overweight and 8% obese at baseline to 25% overweight and 11% obese at 5.5 years.⁶²

While it is imperative to recognize and address overnutrition in CF, prevention of undernutrition has historically been the hallmark of CF nutrition care and remains an essential consideration. Not all individuals with CF are eligible for HEMT, nor do all individuals demonstrate robust increases in weight with HEMT use. In the same 5-year clinical effectiveness study of long-term study of ivacaftor, the prevalence of underweight in pediatric patients treated with ivacaftor also increased from 4% to 6%, while in adults, prevalence of underweight decreased from 11% to 4%.⁶²

3.2 Relationship between pulmonary function and BMI in the era of HEMT

Over the past several years, data from the CF patient registry demonstrate that not only have both pulmonary function and BMI improved, but at any given BMI percentile, pulmonary function is better. For example, in 2016, a BMI at the 50th percentile for age and sex was associated with an FEV₁pp around the 90th percentile, while in 2020, this same BMI was associated with an FEV₁pp above the 95th percentile.¹

The degree of incremental improvement in FEV₁pp as BMIs increase into the ranges of overweight and obesity is a topic that warrants further investigation. A recently published meta-analysis suggested individuals with CF who were overweight and obese had better lung function than those who were normal weight.¹⁹ Notably, all of the studies reported in the meta-analysis were conducted before the widespread availability of HEMT, although other modulators, such as lumacaftor-ivacaftor and tezacaftor-ivacaftor, were available for individuals with F508Del mutations, which do have some impact on weight trajectory and pulmonary function. Further, obese lung transplant patients (not specifically CF) have been found to have decreased survival compared to normal weight.^63-65 Additional knowledge regarding the relationship between BMI and FEV₁pp, particularly at higher BMIs, will strengthen evidence-based guidelines for optimizing nutrition status in people with CF, across all ages and body sizes.

4.1 Hypertension and cardiovascular disease

In a single-center, retrospective study of 134 adults with CF, the proportion of patients meeting the criteria for hypertension increased from 35% to 63% following approximately 1 year of ETI therapy, while the proportion of people who were overweight increased from 19.4% to 31.3% and the proportion of people who were obese increased from 7.5% to 9.7%.⁷⁰ Increases in blood pressure following ETI may reflect the adiposity increases and/or may be influenced by correction or reduction in salt losses.

A recent study designed to investigate the epidemiology and outcomes of cardiac events in people with CF, using large patient databases, found a significantly higher risk of cardiac disease in CF compared to matched healthy counterparts, similar to other chronic inflammatory diseases.⁷¹ The prevalence of hypertension was low in CF overall, but higher in individuals with CF who had major adverse cardiac events (MACEs) than those who did not. In one of the databases examined, there was also a higher prevalence in obesity in those with CF and MACE than those without.⁷¹ A recent case series of six people with CF who developed acute myocardial infarction did not demonstrate a clear pattern of consistent risk factors for coronary artery disease.⁷² The current literature demonstrates not only that individuals with CF are at risk of macrovascular complications, but also the need for better understanding of how CF, obesity and other related factors modulate risk.

4.2 Dyslipidemia

Dyslipidemia has been described in individuals with CF with lower levels of total, LDL, and HDL cholesterol, as well as hypertriglyceridemia.⁷³ Additionally, total cholesterol and triglycerides are associated with increasing age and BMI in people with CF.^{73, 74} Total cholesterol levels, as well as HDL and LDL-cholesterol, have been shown to increase following ETI therapy in two single-center studies of adults with CF and one single-center study of children and adolescents.^{70, 75} Possible mechanisms for increased lipid levels by ETI may be direct effects on cholesterol metabolism and trafficking, reduction in inflammation and/or oxidative stress, and/or improvements in dietary fat absorption.^{34, 76, 77} While ETI may correct hypocholesterolemia, the long-term effects of excess adiposity or pro-atherogenic diet on CVD risk in the setting of HEMT in people with CF are yet unknown. As life span increases in CF, modifiable factors influencing CVD risk should be a priority area of investigation.

4.3 Glucose intolerance and CF-related diabetes

CF-related diabetes (CFRD) is the most common comorbidity in people with CF, with over 40% of adults >20 years of age ultimately developing CFRD.⁷⁸ The genesis of CFRD is incompletely understood, but involves underlying defects in beta cell function and resulting insulin insufficiency. It is not clear why some people with CF develop CFRD while others do not. Proposed mechanisms include direct effects of CFTR loss on beta cell size and function, bystander damage and inflammation extending from exocrine pancreas damage to the endocrine pancreas, and defects in the incretin hormone axis. Genome-wide association studies have also indicated overlap between risk variants for type 2 diabetes mellitus and CFRD.⁷⁹ As the CF population ages, insulin resistance is becoming more recognized as a possible contributor to CFRD.⁸⁰ Although controlled clinical trials are needed, adiposity and/or diet may influence insulin resistance, insulin secretion, and ultimately diabetes risk in people with CF, as they do in the general population.^{67, 81-83} In a small single-center study, visceral adipose tissue (a known risk factor for insulin resistance) was found to be higher in adults with CF compared to BMI-matched controls without CF.⁶⁷ Visceral adipose tissue, in turn, was positively associated with added sugar intake and fasting blood glucose in people with CF.⁶⁷ Further analysis of this cohort indicated that unsaturated fatty acids, lower glycemic load, and plant protein intake correlated with better fasted glycemic outcomes.⁸¹ The relationships between diet, adiposity, and glucose intolerance or CFRD risk may become further evident with increasing prevalence of overweight and obesity and/or increasing age.

4.4 Hepatic steatosis and metabolic dysfunction-associated steatotic liver disease

Hepatic steatosis—the accumulation of fat in the liver—is highly common in CF, with more recent single-center reports of 15%–36% prevalence, depending on the diagnosis definition and method used.^84-87 In a small cross-sectional study, lumacaftor/ivacaftor therapy was associated with less hepatic steatosis, although the mechanism of action is not clear.⁸⁴ Causes of hepatic steatosis in people with CF have not been elucidated, although its presence has historically been considered a benign condition in this population and progression to cirrhosis has not been observed.⁸⁸ Early studies in CF linked hepatic steatosis to malnutrition.⁸⁹ On the contrary, a recent study found a positive association between hepatic steatosis and BMI, with prevalence higher in those with CF who were overweight (47%), reminiscent of metabolic dysfunction-associated steatotic liver disease (MASLD) in the general overweight and obese population.⁸⁵ In MASLD, hepatic steatosis is a metabolic consequence of overnutrition and adipose tissue mishandling of excess fatty acids into the liver, leading to hepatic lipotoxicity, inflammation, and fibrosis.⁹⁰ While MASLD is known to be associated with glucose intolerance and type 2 diabetes mellitus, hepatic steatosis in people with CF was not shown to correlate with CFRD status in one small retrospective study.^{85, 90} Lifestyle factors, such as excess calories, poor quality diet, and sedentary activity, are risk factors for MASLD; and lifestyle interventions are effective at reducing steatosis in the general population.⁹⁰ Whether these lifestyle factors influence hepatic steatosis in CF is not known. However, detailed studies on the causes and consequences of hepatic steatosis in CF are warranted as overnutrition, rather than undernutrition, becomes a concern in the CF population.

4.5 Colon cancer

With increasing life expectancy, an increase in age-related cancers among the CF population is expected. However, the relative rate of bowel cancer (ileum and colon) in adults with CF over age 30 has been reported to be over six times higher than that of people without CF.⁹¹ Hypothesized causes of elevated risk in CF include the CFTR mutation itself, chronic intestinal inflammation, chronic intestinal cell turnover, bile acid dysregulation, nutrient deficiencies (such as vitamins D and E), and immunosuppression (in individuals who have undergone transplant).⁹² In the general population, type 2 diabetes and obesity are risk factors for colorectal cancer, likely driven by insulin resistance and hyperinsulinemia as a growth factor, as well as inflammation.^93-95 Elevated visceral adiposity also increases risk, independently of BMI.^{96, 97} Of note, CFRD was previously shown to correlate with increased risk of pre-cancerous adenomas in people with CF, and with increased risk of overall gastrointestinal tract cancer.^{98, 99} Modifiable lifestyle factors such as a high-fat, refined carbohydrate, low-fiber Western-style diet, play a large role in colorectal cancer risk in the general population, although the role of diet on cancer risk in people with CF has not been determined.^{100, 101} Given the historic tendency of people with CF to consume high total fat, high-saturated fat, and low fiber diets, the role of these modifiable factors on colon cancer risk should be prioritized in this post-modulator era.^{67, 69, 81}

4.6 Obstructive sleep apnea

OSA, nocturnal hypoxemia, and excessive daytime sleepiness are common comorbidities in both children and CF, traditionally attributed to lower and upper airway disease.^102-104 In the general population, obesity is a well-recognized risk factor for OSA.¹⁰⁵ The impact of weight on OSA in CF is a subject of ongoing investigation as the prevalence of overweight/obesity increases in this population.^{105, 106} A study of youth and adults with CF found that adults with overweight and obesity had a higher apnea-hypopnea index than normal and underweight subjects.¹⁰⁷ Conversely, in a German cohort of 52 adults with CF (15% with overweight/obesity) who underwent polysomnography, BMI was not different between subjects with and without diagnoses of OSA.¹⁰⁴ HEMT may also alter the landscape of sleep-disordered breathing in CF. In a small study of adults with CF and severe lung disease treated with ETI, a significant improvement in nocturnal oxygenation and respiratory rate was seen between baseline and after 3 months of ETI, and sustained over a 12-month follow-up period.¹⁰⁸