Effect of CFTR Modulators on Anthropometric Parameters in Individuals with Cystic Fibrosis: An Evidence Analysis Center Systematic Review

There is a strong positive association between nutrition status and lung function in cystic fibrosis (CF). Improvements in clinical care have increased longevity for individuals with CF, and it is unknown how cystic fibrosis transmembrane regulator (CFTR) modulation therapy affects nutrition status over time. The objective of this systematic review of the literature was to examine anthropometric (height, weight, and body mass index [BMI; calculated as kg/m2]) and body composition outcomes of CFTR modulation therapy. A literature search of Medline (Ovid), Embase, and CINAHL (EBSCO) databases was conducted for randomized controlled trials examining the effect of CFTR modulation therapy on anthropometric and body composition parameters, published in peer-reviewed journals from January 2002 until May 2018. Articles were screened, data were synthesized qualitatively, and evidence quality was graded by a team of content experts and systematic review method- ologists. Significant weight gain with ivacaftor was noted in children and adults with at least 1 copy of G551D mutation. In adults with at least 1 copy of R117H the effect of ivacaftor on BMI was not significant. Effects on BMI were mixed in adults with class II mutations taking ivacaftor with lumacaftor. There was no significant change in BMI in children homozygous for F508del who took ivacaftor with tezacaftor. Elexacaftor tezacaftor-ivacaftor increased BMI and body weight in individuals 12 years of age and older who were hetero- or homozygous for the F508del mutation. The effect of CFTR modulation therapy on anthropometric parameters depends on the genetic mutation and the type of modulation therapy used.

More research is needed to understand the long-term clinical impact of these drugs on nutritional status, including body composition and the role of dietary intake fibrosis transmembrane regulator (CFTR) gene, leading to dysregulation of chloride and bicarbonate transport, thick fluid secretions, and severe down- stream impacts on the respiratory and gastrointestinal systems.1 Nearly 85% to 90% of individuals with CF develop exocrine pancreatic insufficiency, result- ing in malabsorption of fat; carbohy- drates; protein; and fat-soluble vitamins A, D, E, and K.2 Malnutrition, and subsequent poor growth in chil- dren, is a common problem in CF due to malabsorption, increased work of breathing, chronic inflammation, and chronic lung infection. Nutritional status in both children and adults is closely associated with lung function in CF, with better growth and higher weight, body mass index (BMI; calculated as kg/m2) and fat-free mass being associ- ated with better lung function.3,4 In addition, malnutrition and low BMI are associated with poorer survival in chil- dren and adults with CF.5 For these rea- sons, a high-calorie, unrestricted fat diet is currently recommended for children and adults with CF.3 The degree of pancreatic insufficiency, malnutrition, and poor growth often depends on the severity of the CFTR mutation.6

There are more than 2,000 variants of the CFTR mutation, with approximately 350 mutations known to cause CF.7 The mutations are classified based on the mechanism by which the defective gene disrupts normal CFTR function.1 Class I mutations impair the production of the CFTR protein completely. In class II mutations, the CFTR protein is misfolded and unable to reach the cell surface. The most common CF mutation, F508del (a class II mutation), falls into this category, and 88% of individuals with CF have at least 1 copy of this mutation.7 Class III mutations (“gating mutations”) prevent the CFTR protein channel from opening once it reaches the cell surface. Class IV muta- tions interfere with conductance of chlo- ride ions,1,8 and class V mutations produce insufficient quantities of CFTR protein. CFTR mutation classes I, II, and III represent the more severe disease phe- notypes.6 During the past decade, several drugs have been developed that specif- ically target the underlying defective CFTR protein. These drugs are known as CFTR modulators, and their mechanism of action varies based on the CFTR defect. Ivacaftor (a potentiator) acts on cell surface CFTR protein channels allowing for improved chloride transport, and lumacaftor and tezacaftor (correctors) improve the structure of the CFTR pro- tein and its trafficking to the cell sur- face.9 As some mutations have multiple defects, CFTR modulators are often used in combination with each other. CFTR modulator therapy has been shown to be effective in improving clinical outcomes, such as increased FEV1% predicted (percent of predicted forced expiratory volume in 1 second), decreased sweat chloride levels, and decreased risk of pulmonary exacerbation.

Although CFTR modulators have the potential to improve anthropometric (height, weight, and BMI) and body composition parameters, these outcomes have not been the primary focus of CFTR modulator therapy trials. There is uncer- tainty regarding their effects on nutri- tional status, whether these effects are dependent on CFTR mutation, and whether effects are modified by habitual diet. Therefore, we conducted a system- atic review (SR) to determine the effects of various CFTR modulator therapies, compared to placebo/control, on anthro- pometric and body composition param- eters in children and adults with CF, including how diet modifies these effects.This SR was conducted following the methods of the Evidence Analysis Center at the Academy of Nutrition and Dietetics (Academy),16 in accordance with Preferred Reporting Items for Systematic Reviews and Meta-Ana- lyses17 was prospectively registered at PROSPERO (CRD42018097373; http://www.crd. york.ac.uk/PROSPE RO/display_ record.php?ID¼CRD42018097373).18 More details on methods can be found on the Evidence Analysis Library website.19Six registered dietitian nutritionists were selected by the Academy’s Evidence-Based Practice Committee to conduct this SR. The SR team also included 2 patient advocates, a project manager, SR methodologists, a medical librarian, a lead analyst and evidence analysts.Eligibility Criteria, Literature Search, and Study SelectionResearch questions were framed in the standard PICO format to include the population (P), intervention or exposure (I), comparison or control (C), and outcome (O). In order to be included, studies must have examined partici- pants diagnosed with CF as the target population, the intervention must have been CFTR modulation therapy, studies were required to have a comparison group (study design was limited to randomized controlled trials [RCTs] only) and report on at least 1 outcome of interest (weight and growth parameters or body composition) as appropriate for the age group examined.

For pediatric participants (0-17 years of age), accept- able measures were percentiles or z scores for BMI, weight, or length/height. For adults, acceptable measures were weight and BMI. Body composition measures of interest included fat mass, fat-free mass (or lean body mass), or total body water measured by skinfold measurements, mid-upper arm circumference, bioelectrical impedance analysis, dual-energy x-ray absorpti- ometry, air displacement plethysmog- raphy, deuterium oxide, or the 4- compartment model.Additional inclusion criteria were publication in a peer-reviewed journal and in the English language due to resource constraints, and publication from January 2002 until the search date of May 4, 2018. The inclusion date of 2002 was chosen to remain consistent with other current evidence-based nutrition guidelines for people with CF,20,21 Trials were required to have at least 10 partici- pants in each study arm. Participant age, study duration, and setting were unre- stricted. Medline (Ovid), Embase, and CINAHL (EBSCO) databases were searched using a combination of CFTR modulation therapy terms, and weight and growth or body composition pa- rameters, and RCT methods. The search plan used for the Medline (Ovid) database can be found in Figure 1 (available at www.jandonline.org). Relevant SRs were hand-searched for articles that may have been missed in the databases search. Ar- ticles were uploaded to a Rayyan data- base,22 and titles/abstracts were screened for eligibility criteria by the project man- ager and lead analyst. The full texts of potentially included articles were screened for eligibility criteria by 2 blin- ded workgroup members, and discrep- ancies were discussed until the workgroup reached consensus. All arti- cles not meeting inclusion criteria were excluded. Workgroup members were asked to review the list of included arti- cles to determine whether any relevant articles had been missed by the system- atic search (Figure 2).

Included articles were assigned to trained evidence analysts who extracted data using the Academy’s Data Extraction Tool.23 Data extracted included bibliographic information, study inclusion criteria, source of funding, participant demographics, and comorbid conditions, participant attri- tion, details of the intervention (dose and duration), and results reported for outcomes of interest. Analysts also rated the quality of the study by eval- uating risk of bias using the Academy’s Quality Criteria Checklist. Risk of bias domains assessed included reporting, selection, attrition, performance, and detection bias. Articles were rated as positive quality (low risk of bias), neutral quality (demonstrates some risk of bias), or negative quality (dem- onstrates significant risk of bias). Data extraction was reviewed by the lead analyst, who also completed a blinded, second evaluation of study quality. Any discrepancies in study quality between the first 2 reviewers were addressed in a third review.Using extracted data, the project manager and lead analyst compiled drafts of tabular and narrative evidence summaries. Evidence summaries were reviewed in detail by 3 workgroup members. When multiple articles re- ported on an individual study, data were examined closely in order to avoid reporting any findings in dupli- cate. When authors reported an outcome of interest, but the data were incomplete, corresponding authors were contacted directly and additional data were provided by Vertex Pharmaceuticals Inc. Meta-analysis was attempted if studies/populations were similar enough to warrant pooling data and complete data could be obtained.

Most studies did not report complete data for outcomes of interest, and complete data could not be obtained from authors/funders and therefore could not be included in meta-analysis. In the evidence summary for ivacaftor with lumacaftor, the workgroup elected not to report meta-analysis due to heterogeneity in study populations (homozygous vs heterozygous F508del mutation) and findings. Evidence summaries were utilized to create conclusion statements directly answering the PICO question based on available evidence. For each outcome, conclusion statements were graded using the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) method, which includes consideration of risk of bias across included studies, number of studies included and samples sizes, consistency of findings across studies, effect size, and generalizability.24 Conclusion statements were graded according to the certainty of evidence ascertained in the GRADE Summary of Findings Table. High certainty of evi- dence (GRADE) translated into a GradeI conclusion statements indicating good/strong evidence (Academy); moderate certainty evidence (GRADE) translated into Grade II conclusions statements indicating fair evidence (Academy); and low or very low cer- tainty evidence (GRADE) translated into Grade III conclusion statements, indicating limited/weak evidence (Academy).23,24 All workgroup mem- bers voted and unanimously approved conclusion statements and grades.Updated Search of the Literature This SR supported development of an evidence-based nutrition practice guideline for individuals with CF. CFTR modulation therapy evolved even dur- ing development of this guideline, and triple-combination therapy (elex- acaftor-tezacaftor-ivacaftor) became available for use in 2019. Therefore, the SR methodologist supplemented the original search with a search of the PubMed database for RCTs investi- gating drug therapies in individuals with CF published since the original cutoff date of May 5, 2018 until February 20, 2020 (Figure 1; available at www.jandonline.org).

A total of 625 original articles were identified and results from 16 articles representing 13 RCTs were synthesized narratively. However, meta-analysis was not possible or appropriate for any of the interventions examined. All RCTs demonstrated low risk of bias. All included studies were industry spon- sored. CFTR modulation therapies examined included ivacaftor (n¼6), ivacaftor with lumacaftor (n¼4), iva- caftor with tezacaftor (n¼1), and elex- acaftor with tezacaftor and ivacaftor (n¼2). The targeted class of CFTR mu- tation varied by RCT. Authors had cor- respondence with Vertex Pharmaceuticals Inc in regard to obtaining more data for outcomes of interest and received additional data on 2 included studies.Six RCTS (represented by 9 articles)10-12,26-31 examined the effect of ivacaftor CFTR modulation therapy on weight/growth parameters and/or body composition in participants with CF (Table 1).10-12,26-31 All trials utilized the same dose of ivacaftor: 150 mg every 12 hours or twice per day. While the ENVISION trial targeted pediatric participants,10,27-29 Edgeworth and colleagues31 included adult partici- pants only. The remaining studies included both pediatric and adult participants. Sample sizes ranged from 2031 to 167 (STRIVE) partici- pants,12,27-29 and trial durations ranged from 8 weeks (KONNECTION)11 to 48 weeks (ENVISION and STRIVE).10,27-29Study sites were exclusively in economically developed countries. The ENVISION and STRIVE trials and the trial by Edgeworth and colleagues tar- geted participants with at least 1 copy of the G551D mutation (class III), while the KONNECTION11 trial targeted CF participants with non-G5511D gating mutations (class III). Finally, the KON- DUCT trial targeted participants with the R117H mutation (class IV) and the DISCOVER trial targeted participants who were homozygous for the F508del mutation (class II).

Weight, growth parameters, and body composition were secondary outcomes reported in these phase III trials. There was no in- formation available to clarify whether the outcomes of CFTR modulation therapy were modified by dietary intake.In pediatric participants ≤20 years of age with at least 1 copy of the G551D mutation (class III), 48 weeks of 150 mg ivacaftor twice daily increased weight- for-age (WFA) and BMI-for-age z scores by 0.35 and 0.39, respectively, compared to placebo (mean baseline BMI z score was e0.199 and WFA z score was e0.292).27 The same dose may increase BMI-for-age z score after8 weeks in 6- to 17-year-olds with other class III mutations, but evidence was limited.11 In adults with CF with at least 1 copy of the G551D mutation (class III) and with optimal3 or low mean BMI at baseline, 4 to 48 weeks of treatment with 150 mg ivacaftor twice daily increased weight and BMI by a mean of 2.9 kg and 0.58 to 1.2, respectively.10,11,31 In participants ≤20 years of age who were homozygous for the F508del mutation (class II), consistent with a lack of efficacy on lung function, 16 weeks of 150 mg ivacaftor twice daily had no statistically significant effect on WFA or BMI z scores compared to placebo.26 In adults with CF who had class IV mutations and optimal mean BMI, 150 mg of iva- caftor twice daily for 24 weeks had no statistically significant effect on BMI compared to placebo.30 The effect of ivacaftor on body composition in pe- diatric participants with class III and class II mutations was not described. Only 1 RCT examined changes in body composition in adults with 1 copy of G551D (class III mutations), and found that there was no statistically signifi- cant effect of CFTR modulation therapy on fat-free mass over 4 weeks.31A summary of workgroup conclu- sions for ivacaftor RCTs are presented in Table 2. In general, moderate/fair evidence described that in children and adults with at least 1 copy of the G551D mutation (class III), ivacaftor (150 mg twice daily) increases BMI. Evidence certainty is low/weak for other class III mutations. In pediatric patients homozygous for the F508del mutation (class II) or adult patients with at least 1 copy of the R117H mu- tation (class IV), ivacaftor may not have any significant effect on BMI, although the certainty in the evidence was low/ weak.One RCT examined the effect of iva- caftor with lumacaftor on BMI-for-age z score in children with CF, ages 6 to 11 years who were homozygous for the F508del CFTR mutation.15 Participants with CF (59% female, mean SD age¼8.8 1.6 years) were randomized to receive either 200 mg lumacaftor with 250 mg ivacaftor every 12 hours (q12h) (n¼96) or placebo (n¼97) for 24 weeks. Mean standard deviation BMI- for-age z score at baseline was e0.1 0.8 in the intervention group and e0.1 0.9 in the placebo group.

There was no difference in BMI-for-age z score change between groups at 24 weeks. Body composition and caloric intake were not measured.Two RCTs with low risk of bias (positive quality) examined the effect of ivacaftor with lumacaftor on BMI in adults and are summarized in Table 1. Rowe and colleagues32 studied 126 participants with CF who were het- erozygous for the F508del CFTR muta- tion (48% female, ≥18 years, mean standard deviation age¼29.9 8.4 years) and randomized to receive either 400 mg lumacaftor q12h plus 250 mg ivacaftor q12h (n=62) or placebo (n=63); 119 completed the 8-week study. Consis- tent with a lack of clinical efficacy on lung function, there were no between- group differences in body weight or BMI changes during the trial.32 The TRAFFIC and TRANSPORT trials studied adult participants with CF who were homozygous for the F508del CFTR mutation.14,25 Participants were ran- domized to 1 of 3 groups for 24 weeks, including 600 mg/day lumacaftor plus250 mg ivacaftor q12h, 400 mg lumacaftor plus 250 mg ivacaftor q12h, or placebo. Nutritional status at baseline was not reported for adult participants alone. The pharmaceu- tical company funding the trial pro- vided information on the effect of treatment on BMI for adult partici- pants only (≥18 years of age). Compared to the placebo group (n=271), during a course of 24 weeks, those receiving 600 mg/day luma- caftor and 250 mg q12h ivacaftor had an increased least squares mean dif- ference BMI of 0.24 (95% CI 0.09-0.40)kg/m2 (P=0.002) and the group receiving 400 mg lumacaftor q12h and250 mg ivacaftor q12h had an increased BMI of 0.21 (95% CI 0.05- 0.37).14,25 The effect of treatment on body composition was not described in either study, nor was the role of diet in modifying study outcomes.A summary of workgroup conclu- sions for ivacaftor with lumacaftor RCTs are presented in Table 2.

Moderate/fair certainty evidence described that for children ages 6 to 11 years with CF homozygous for the F508del muta- tion and a mean BMI-for-age z score in the healthy range, there was no effect of 200 mg lumacaftor with 250 mg ivacaftor twice daily on BMI-for-age z score after 24 weeks. With moderate/ fair certainty, among adults with CF heterozygous for the F508del mutation,8 weeks of 400 mg lumacaftor with250 mg ivacaftor twice daily has no effect on BMI, which is consistent with a lack of clinical effect on lung function. In adult participants homozygous for F508del, this regimen given for 24 weeks appeared to increase BMI, although baseline nutritional status in this group was not reported. Effect of treatment on body composition was not described, nor was the role of diet, in either study. Ivacaftor with TezacaftorIn participants with CF, 1 RCT with low risk of bias (positive quality) reported on the effect of ivacaftor with teza- caftor on weight/growth parameters, summarized in Table 1. A total of 156 participants 12 to 20 years of age who were homozygous for the F508del CFTR mutation (class II) were randomized to receive either 100 mg of tezacaftor once daily with 150 mg of ivacaftor twice daily or matched placebo for 24 weeks.13 Baseline nutritional status was not available for the pediatric population only. There were no signif- icant differences in BMI-for-age z score change between groups.

Body compo- sition and dietary intake were not measured.13 Workgroup conclusions are summarized in Table 2. With a low/ weak certainty in the evidence, in pe- diatric participant homozygous for the F508del CFTR mutation, tezacaftor (100 mg) plus ivacaftor (150 mg) for 24 weeks did not increase BMI z scores, although the role of baseline nutri- tional status is unknown.The updated search of the PubMed database identified 13 RCTs published from May 2018 until February 20, 2020. Two RCTs met inclusion criteria, both of which investigated the effects of elexacaftor-tezacaftor-ivacaftor in in- dividuals ≥12 years of age (Table 3).33,34 Sample sizes ranged from 107 to 403 participants and study du- rations ranged from 4 to 24 weeks. In the study by Middleton and col- leagues,33 triple-combination therapy for 24 weeks was compared to a placebo in individuals heterozygous for the F508del mutation, while in Heijer- man and colleagues,34 triple- combination therapy for 4 weeks was compared to dual combination therapy (tezacaftor plus ivacaftor) in individuals homozygous for the F508del mutation. BMI and body weight were increased significantly in individuals receiving elexacaftor-tezacaftor- ivacaftor compared to those receiving a placebo or dual combination therapy.

Results of this SR indicate that the ef- fects of CFTR modulators on increasing weight and growth in individuals with CF are likely dependent on the therapy formulation (single vs combination therapy) and the targeted CFTR muta- tion. Ivacaftor alone may increase weight and growth parameters, with clinically significant magnitudes, in individuals with class III (gating) mu- tations.10-12,27-29 As further evidence, a 3-year follow-up observational study found sustained benefits on z scores for BMI- and WFA in of pediatric and adult patients with a G551D mutation who were treated with ivacaftor in clinical trials, compared to matched registry- based controls who were homozygous for the F508del mutation.35 Other observational and retrospective studies support a beneficial effect of ivacaftor on anthropometric parameters among patients with a G551D mutation.36-38 Ivacaftor alone did not improve growth parameters in patients homo- zygous for the F508del CFTR muta- tion.26 However, this would be expected, as it is known be ineffective in restoring CFTR function in this group and is, thus, not indicated for individuals homozygous for the F508del mutation. The effect of ivacaftor alone on people with other mutations, such as R117H,30 is not clear given the limited number of RCTs.

Ivacaftor in combination with lumacaftor is currently only indicated for individuals with CF who are homozy- gous for the F508del CFTR mutation.39 Available RCTs showed clinically sig- nificant increases in weight parameters in adult, but not pediatric, participants with this genotype.14,15,25 An open- label phase III trial did find significant increases in z scores for BMI-, weight-, and stature-for-age among 2- to 5- year-old children, although the lack of a control group limits interpretation.40 Two recent meta-analyses also found statistically significant increases in BMI with ivacaftor+lumacaftor treatment among participants homozygous for the class II F508del CFTR mutation,41,42 although they did not attempt to differentiate between pediatric and adult populations. Consistent with findings of a lack of significant impact on FEV1% predicted, growth parame- ters did not improve in individuals with CF receiving ivacaftor+lumacaftor who are heterozygous for the F508del mutation.The combination of ivacaftor and tezacaftor in individuals with CF ho- mozygous for the F508del CFTR with at least 1 copy of the G551D mutation (class III), 48 wk of 150 mg ivacaftor twice daily increased WFA and BMI-for-age z scores by 0.35 and 0.39, respectively, compared to placebo, when mean baseline WFA z score was e0.292 and BMI z score was e0.199. The same dose may increase BMI-for-age z score after 8 wk in 6- to 17-year-olds with other gating mutations, but evidence was limited. Effect of ivacaftor on body composition in pediatric participants was not described, nor was the role of diet.In adults with CF with at least 1 copy of the G551D mutation (class III) and with optimal or low mean BMI at baseline, 4-48 wk of treatment with 150 mg ivacaftor twice daily increased weight and BMI by a mean of 2.9 kg and 0.58-1.2, respectively.

There was no statistically significant effect of CFTR modulation therapy on fat-free mass. Effect of dietary intake on the relationship of interest was not described mutation did not influence growth parameters, despite significant im- provements in lung function.13 Another randomized, controlled trial of iva- caftor+tezacaftor in individuals 12 years and older who were heterozy- gous for the F508del mutation did not report significant changes in BMI in the group as a whole over 8 weeks,43 although BMI-for-age z score or per- centiles were not reported for pediatric participants, excluding the study from this SR. In 2019, the US Food and Drug Administration approved an updated combination therapy, ivacaftor+tezacaftor+elexacaftor, which improved both lung function and BMI (as well as BMI-for-age z scores) in in- dividuals who are homozygous or het- erozygous for the F508del mutation.33,34 As individuals with 1 or 2 copies of the F508del mutation constitute >85% of the CF population in the United States,44 there is high po- tential for the majority of individuals with CF to be treated with this triple combination therapy. More research is needed to fully understand the effects of combination therapies on nutrition status in all classes of CFTR mutations. The specific mechanisms for improved BMI in some patients receiving CFTR modulators have not been elucidated. CFTR is localized throughout epithelial cell membranes present in multiple organ systems, thus CFTR correction by modulator therapy likely impacts the gastrointestinal system in addition to the lung epithelium. In a study by Stallings and colleagues45 published after the search dates for this SR, ivacaftor treatment for 3 months decreased resting energy expenditure, improved fat absorption, and reduced gut inflammation in 23 children and adults with class III gating mutations, providing possible mechanisms for weight gain in the setting of ivacaftor use. Improvements with ivacaftor treatment were more robust in partic- ipants with pancreatic insufficiency.45 In addition, results of open-label trials described improved pancreatic function in 2- to 5-year-old children with gating mutations after treatment with ivacaftor.

An increase in dietary fat intake and total calories was noted in Italian and North American people with CF with gating mutations who took ivacaftor for 3 months, and this increase was correlated with weight gain.48 As it is recommended that CFTR modulators are taken with a fat-containing food, this may also partially contribute to improvements in weight. While malnutrition, poor growth, and low weight have been common issues for people living with CF,3 CFTR modulation therapy has the potential to change the nutritional landscape in CF. Although some patients may still struggle with underweight and sub- optimal nutrition status, even with highly effective modulator therapy, others may experience excessive weight gain. A CF Foundation Patient Registry analysis between the years of 1998 and 2017 found an increase in the proportion of overweight and obesity in the US CF population over 2 de- cades.49 A recent cross-sectional study of a US adult CF center also found that 25% of adults with CF were overweight and 6.6% were obese.50 As malnutrition declines and overweight and obesity increase in people with CF, there is potential for new CFTR modulators to accelerate the development of over- weight and obesity in this population; such trends should continue to be monitored in this population.

Although BMI is more often clinically used as the indicator of nutritional status in individuals with CF, it does not differentiate between specific body composition components (eg, fat mass vs fat-free mass). Fat-free mass is more strongly associated with lung function in CF compared to BMI.4 Body compo- sition was only examined in 1 eligible RCT for this SR,31 which found no sta- tistically significant change in fat-free mass assessed by bioelectrical imped- ance analysis among adults treated with ivacaftor for 28 days. In contrast, Stallings and colleagues45 reported significant increases in both fat-free mass and fat mass via dual-energy x- ray absorptiometry after ivacaftor treatment for 3 months. It is possible that changes in body composition with ivacaftor are evident over a longer duration of treatment. Furthermore, bioelectrical impedance analysis may over- or underestimate fat mass and fat-free mass in people with CF and this may have contributed to differences in findings on body composition changes with CFTR modulation therapy.51,52 The effects of other CFTR modulators on body composition changes are not known and deserve further investigation.The role of diet in the relationship between modulators and nutritional status was not explored in any of the articles included in this review. His- torically, a high-calorie, high-fat, and high-protein diet has been prescribed to people with CF due to increased caloric expenditure related to chronic lung disease and inflammation, impaired nutrient utilization related to exocrine pancreatic insufficiency, and better survival outcomes in patients who were prescribed a high-fat diet with heavy pancreatic enzyme use.53 Limited evidence suggests that caloric absorption is improved with iva- caftor45; however, best practices for dietary prescriptions and adjustment of pancreatic enzyme replacement therapy in the setting of CFTR modu- lation therapy are not known. While some people with CF may continue to require high-calorie diets historically prescribed, others may require alter- ations to the prescription for energy and nutrient intake based on individual weight/growth trends once CFTR modulation therapy is initiated. Ulti- mately, clinicians must encourage pa- tients to maintain their CF therapies, including pancreatic enzyme replace- ment therapy, and also take an indi- vidualized approach to dietary intake monitoring and recommendations as the landscape of nutrition care in CF changes in the setting of CFTR modulators.

To our knowledge, this is the first study to systematically review the ef- fects of CFTR modulators, in isolation and in combination, focusing on BMI and other growth parameters in pedi- atric and adult patients with CF. How- ever, the small number of studies available for analysis limited the ability to generate evidence-based conclusion statements with high levels of certainty. Another limitation was that BMI and other growth parameters were reported as secondary outcomes in all included studies; thus, studies may not have had adequate power to detect outcomes. Studies did not report changes in the proportions of participants who were, for example, underweight and normal weight, but rather reported mean values for weight parameters. Given the widespread approval for clinical care in many countries, it is unlikely that there will be additional placebo-controlled trials of already established modula- tors, thus the body of work available for future systematic literature reviews examining nutrition and body composition effects will likely be limited to nonrandomized studies. BMI in pediatric populations highly varies with age and sex,54 and in some studies that included pediatric ages, BMI was inappropriately reported without con- verting to z scores or percentiles, thus limiting the ability to draw conclusions in these age ranges. Whether baseline nutritional status influenced the effects of CFTR modulators, or whether diet played a role in the relationships examined, could not be determined based on the limited data available. Finally, the small number of studies and heterogeneous populations (eg, age and mutations) limited the ability to perform meta-analyses.

The efficacy of CFTR modulators on improving nutritional status in in- dividuals with CF was highly depen- dent on the therapy formulation (single vs combination therapy) and the CFTR mutation of the targeted population. As new, highly effective CFTR modulators are developed, the potential for improvement in growth and nutrition status coupled with increasing longevity presents new challenges and opportunities for interdisciplinary teamwork and partnership in nutrition care for people with CF. Expanding the body of research on how specific CFTR modulators affect nutrition status, and on best clinical practice to adapt to these effects is necessary to determine optimal nutritional strategies in this population, and will allow for improved care of people with CF as Elexacaftor they age.