Abstract
Background:
Sweat chloride (SC) concentrations in people with cystic fibrosis (PwCF) reflect relative CF transmembrane conductance regulator (CFTR) protein function, the primary CF defect. Populations with greater SC concentrations tend to have lesser CFTR function and more severe disease courses. CFTR modulator treatment can improve CFTR function within specific CF genotypes and is commonly associated with reduced SC concentration. However, SC concentrations do not necessarily fall to concentrations seen in the unaffected population, suggesting potential for better CFTR treatment outcomes. We characterized post-modulator SC concentration variability among CHEC-SC study participants by genotype and modulator.
Methods:
PwCF receiving commercially approved modulators for ≥90 days were enrolled for a single SC measurement. Clinical data were obtained from chart review and the CF Foundation Patient Registry (CFFPR). Variability of post-modulator SC concentrations was assessed by cumulative SC concentration frequencies.
Results:
Post-modulator SC concentrations (n=3787) were collected from 3131 PwCF; most (n=1769, 47%) were after elexacaftor/tezacaftor/ivacaftor (ETI) treatment. Modulator use was associated with lower SC distributions, with post-ETI concentrations the lowest on average. Most post-ETI SC concentrations were <60 mmol/L (79%); 26% were <30 mmol/L. Post-ETI distributions varied by genotype. All genotypes containing at least one F508del allele had individuals with post-ETI SC ≥60mmol/L, with the largest proportion being F508del/minimal function (31%).
Conclusions:
Post-modulator SC concentration heterogeneity was observed among all genotypes and modulators, including ETI. The presence of PwCF with post-modulator SC concentrations within the CF diagnostic range suggests room for additional treatment-associated CFTR restoration in this population.
Keywords: CFTR modulators, sweat chloride, CFTR function, epidemiology
Introduction
Cystic fibrosis (CF) is caused by mutations in the cystic fibrosis transmembrane conductance regulator (CFTR) gene resulting in the absence or dysfunction of CFTR protein, an anion channel in epithelial cells lining the respiratory tract, gastrointestinal system, reproductive system, and sweat glands. (1) In the sweat glands, reduced CFTR activity limits reabsorption of chloride, resulting in abnormally high sweat chloride (SC) concentrations. A CF diagnosis is confirmed by elevated SC concentrations ≥ 60 mmol/L. (2) Individuals with two severe CFTR variants typically have SC concentrations ≥ 80 ammol/L, indicative of minimal CFTR activity. In contrast, individuals with partial CFTR dysfunction, often associated with residual function gene variants, may have SC concentrations between 60–79 mmol/L or below the diagnostic threshold of 60 mmol/L; residual function genotypes and lower SC concentrations are associated with less severe CF disease phenotypes and improved survival. (3)
The development of CFTR modulators has clarified the relationship between SC and CFTR protein activity. These oral therapeutics increase CFTR protein activity, cause rapid and sustained SC reductions, and improve clinical outcomes in people with CF (PwCF) with specific responsive gene variants. In clinical trials of PwCF with one or two F508del-CFTR variants treated with elexacaftor/tezacaftor/ivacaftor (ETI), the most effective modulator based on clinical outcome, a substantial portion of participants had post- ETI SC concentrations <60 mmol/L, the diagnostic threshold for CF, or even below 30 mmol/L, the range typical for people without CF, including CFTR gene variant carriers. (4, 5) However, there was heterogeneity in sweat chloride reduction, including PwCF whose sweat chloride concentrations remained above the diagnostic threshold, suggesting persistent CFTR dysfunction. Clinical outcomes also varied but whether post-ETI SC concentrations are associated with these outcomes is unclear. (6)
CHEC-SC (Characterizing CFTR Modulated Changes in Sweat Chloride and their Association with Clinical Outcomes; NCT03350828) is an ongoing multicenter cohort study of SC in PwCF in the United States receiving commercially approved CFTR modulators. (7) In phase 1, CHEC-SC enrolled individuals receiving ivacaftor, lumacaftor/ivacaftor (lum/iva), or tezacaftor/ivacaftor (tez/iva). (8) In phase 2, CHEC-SC started enrolling individuals on ETI, including some who previously enrolled on earlier modulators. We sought to estimate the proportion of the CHEC-SC cohort that may benefit from additional CFTR restoration by examining the heterogeneity of post-modulator SC concentrations.
Methods
The CHEC-SC study design has been previously reported. (7) Eligible participants taking a commercially approved CFTR modulator for ≥90 days were enrolled across 51 sites in the CF Therapeutics Development Network (TDN). Sweat was collected from each participant at a single study visit, as well as clinical data including genotype, spirometry, and prior CFTR modulator use. SC measurements were performed by local clinical laboratories. Diagnostic and other historic pre-modulator SC concentrations were recorded from sources as described below. Previously enrolled participants who switched to a different commercially approved CFTR modulator were eligible to re-enroll in the study after ≥90 days on the new modulator. The study was approved by Advarra’s Institutional Review Board. All analyses were performed using R Statistical Software (v4.2.2). (9)
Pre-modulator SC data were pulled from multiple sources; electronic medical record data captured during the CHEC study visit, other TDN studies with participant overlap and data sharing agreements (GOAL, PROMISE, PROSPECT), and the CF Foundation’s Patient Registry (CFFPR). The source used to calculate pre-modulator SC was determined hierarchically based on which of the above sources had available data. CHEC study visit data were used whenever possible. If CHEC data were not available, pre-modulator SC was pulled from GOAL, PROMISE, and/or PROSPECT if available. SC data from the CFFPR were only used if results were not available from other sources.
Pre- and post- modulator SC, as well as demographic and clinical data, were descriptively summarized across modulators. Genotype groups were also descriptively summarized across modulators. SC distributions were summarized by cumulative distribution functions (CDF). (10) Cumulative distributions were estimated empirically using the ecdf function and visualized using the ggplot2 package. (11) An area-proportional Euler diagram was created using the eulerr R package to visualize CHEC enrollments and re-enrollments on different modulators. (12) Associations between select baseline demographics and post-ETI SC concentrations were investigated in F508del homozygous participants via regression analysis of post-ETI SC concentrations. Covariates in the regression model included sex at birth, race (White, Non-White/More Than One Race, and Unknown or Not Reported), ethnicity (Hispanic/Latino, Non-Hispanic/Latino), age group at the time of sweat collection (6–11, 12–17, 18–25, and >25 years), pre-modulator SC concentration quartile (<94, 94 to 102, 102 to 110, and ≥110 mmol/L), and post-ETI percent predicted forced expiratory volume in 1 second (ppFEV1) group (<40, 40 to <70, 70 to <100, and ≥100). Post-ETI ppFEV1 groups were determined using pre-defined thresholds. Confidence intervals obtained from the regression analysis were not adjusted for multiple testing due to the exploratory nature of the analysis.
R117H poly T status was determined based on centralized source document review. Post-modulator ppFEV1 was calculated using Global Lung Function Initiative (GLI) equations. (13) If a participant re-enrolled in the study while receiving a different modulator, sex at birth and race/ethnicity designations collected at the first enrollment were used to calculate ppFEV1 at each subsequent enrollment.
SC concentrations <10 mmol/L were imputed with a value of 9 mmol/L and were averaged across replicates (e.g., right and left arm) when available. Pre-modulator SC was calculated by averaging first across replicates and then across collection dates for all available historic SC collected from a given source prior to the participant’s first reported modulator use.
Results
Study population
A total of 3787 SC concentrations were collected among 3131 PwCF enrolled in CHEC-SC from January 2018 to April 2022 across 51 TDN sites, with 47% of SC measurements (n=1769) obtained from PwCF receiving ETI, 14% receiving ivacaftor (n=536), 21% receiving lum/iva (n=785), and 18% receiving tez/iva (n=697). Out of 3131 PwCF enrolled in CHEC, 2598 (83%) had pre-modulator SC gathered at the CHEC study visit. An additional 181 individuals (5.8%) had their pre-modulator SC pulled from GOAL, PROMISE, PROSPECT, or CFFPR data. No participants had pre-modulator SC <10 mmol/L and 38 participants had at least one post-modulator SC <10 mmol/L. A description of the study cohort stratified by modulator is provided in Table 1. A diagram describing eligible and enrolled participants is available in Figure S1 (online supplement). Recruitment of participants in the younger age groups for ivacaftor and lum/iva as compared to tez/iva and ETI was enabled by label extensions to these lower ages during the study period.
Table 1.
CHEC-SC participant demographics by CFTR modulator treatment at enrollment.
| Modulator at CHEC-SC Enrollment |
||||
|---|---|---|---|---|
| IVA (N=536) | LUM/IVA (N=785) | TEZ/IVA (N=697) | ELE/TEZ/IVA (N=1769) | |
|
| ||||
| Female Sex, n (%) | 273 (50.9%) | 368 (46.9%) | 364 (52.2%) | 849 (48.0%) |
| Age at sweat collection (years) | ||||
| Mean (SD) | 18.8 (15.4) | 16.7 (9.9) | 21.7 (10.6) | 23.4 (12.4) |
| Category, n (%) | ||||
| 0–1 years | 11 (2.1%) | 0 (0%) | 0 (0%) | 0 (0%) |
| 2–5 years | 82 (15.3%) | 91 (11.6%) | 0 (0%) | 0 (0%) |
| 6–11 years | 127 (23.7%) | 190 (24.2%) | 65 (9.3%) | 200 (11.3%) |
| 12–17 years | 108 (20.1%) | 218 (27.8%) | 251 (36.0%) | 547 (30.9%) |
| 18–25 years | 96 (17.9%) | 159 (20.3%) | 219 (31.4%) | 460 (26.0%) |
| >=26 years | 112 (20.9%) | 127 (16.2%) | 162 (23.2%) | 562 (31.8%) |
| Race, n (%) | ||||
| White | 511 (95.3%) | 763 (97.2%) | 675 (96.8%) | 1695 (95.8%) |
| Black or African American | 8 (1.5%) | 5 (0.6%) | 3 (0.4%) | 22 (1.2%) |
| Asian | 1 (0.2%) | 3 (0.4%) | 2 (0.3%) | 3 (0.2%) |
| American Indian or Alaska Native | 1 (0.2%) | 0 (0%) | 0 (0%) | 2 (0.1%) |
| Native Hawaiian or Other Pacific Islander | 0 (0%) | 1 (0.1%) | 1 (0.1%) | 1 (0.1%) |
| More than One Race | 5 (0.9%) | 1 (0.9%) | 10 (1.4%) | 25 (1.4%) |
| Unknown or Not Reported | 10 (1.9%) | 6 (0.8%) | 6 (0.9%) | 21 (1.2%) |
| Ethnicity, n (%) | ||||
| Hispanic or Latino | 60 (11.2%) | 32 (4.1%) | 28 (4.0%) | 100 (5.7%) |
| Not Hispanic or Latino | 476 (88.8%) | 752 (95.8%) | 668 (95.8%) | 1669 (94.3%) |
| Not Reported/Unknown | 0 (0%) | 1 (0.1%) | 1 (0.1%) | 0 (0%) |
| Diagnostic sweat chloride | ||||
| Mean (SD) | 84.0 (26.7) | 101.7 (12.8) | 98.8 (17.1) | 98.8 (18.4) |
| Median (Min, Max) | 91.0 (19.0, 176.0) | 101.5 (43.0, 165.0) | 100.3 (33.0, 165.0) | 101.0 (17.2, 165.0) |
| Category, n (%) | ||||
| 110+ mmol/L | 74 (13.8%) | 152 (19.4%) | 139 (19.9%) | 374 (21.1%) |
| 100 to <110 mmol/L | 90 (16.8%) | 241 (30.7%) | 178 (25.5%) | 466 (26.3%) |
| 80 to <100 mmol/L | 145 (27.1%) | 261 (33.2%) | 225 (32.3%) | 535 (30.2%) |
| 60 to <80 mmol/L | 87 (16.2%) | 20 (2.5%) | 36 (5.2%) | 117 (6.6%) |
| 30 to <60 mmol/L | 84 (15.7%) | 1 (0.1%) | 22 (3.2%) | 50 (2.8%) |
| <30 mmol/L | 19 (3.5%) | 0 (0%) | 0 (0%) | 10 (0.6%) |
| Missing | 37 (6.9%) | 110 (14.0%) | 97 (13.9%) | 217 (12.3%) |
| Post-modulator sweat chloride | ||||
| Mean (SD) | 53.0 (26.5) | 80.6 (16.8) | 87.5 (18.7) | 44.6 (21.3) |
| Median (Min, Max) | 48.5 (9.0, 130 5) | 80.5 (27.0, 133.5) | 91.0 (10.0, 123.0) | 42.0 (9.0, 129.0) |
| Category, n (%) | ||||
| 110+ mmol/L | 10 (1.9%) | 29 (3.7%) | 45 (6.5%) | 8 (0.5%) |
| 100 to <110 mmol/L | 18 (3.4%) | 63 (8.0%) | 120 (17.2%) | 24 (1.4%) |
| 80 to <100 mmol/L | 77 (14.4%) | 326 (41.5%) | 361 (51.8%) | 95 (5.4%) |
| 60 to <80 mmol/L | 87 (16.2%) | 282 (35.9%) | 123 (17.6%) | 243 (13.7%) |
| 30 to <60 mmol/L | 231 (43.1%) | 84 (10.7%) | 26 (3.7%) | 939 (53.1%) |
| <30 mmol/L | 113 (21.1%) | 1 (0.1%) | 22 (3.2%) | 460 (26.0%) |
| Post-modulator ppFEVl | ||||
| Mean (SD) | 88.4 (24.6) | 82.5 (21.8) | 76.7 (25.9) | 88.3 (23.7) |
| Median (Min, Max) | 93.2 (23.8, 137.1) | 87.6 (23.1, 129.9) | 81.4 (9.6, 132.5) | 93.3 (17.5, 139.5) |
| Category, n (%) | ||||
| 100+ | 179 (33.4%) | 145 (18.5%) | 132 (18.9%) | 618 (34.9%) |
| 70 to <100 | 185 (34.5%) | 378 (48.2%) | 306 (43.9%) | 735 (41.5%) |
| 40 to <70 | 71 (13.2%) | 149 (19.0%) | 167 (24.0%) | 310 (17.5%) |
| <40 | 27 (5.0%) | 34 (4.3%) | 82 (11.8%) | 73 (4.1%) |
| Missing | 74 (13.8%) | 79 (10.1%) | 10 (1.4%) | 33 (1.9%) |
Re-enrollment
A total of 588 (18.8%) of PwCF enrolled in CHEC-SC re-enrolled on a different modulator, with 68 (2%) of all CHEC-SC participants enrolling three times. Combinations of modulators among re-enrollees are shown in Figure 1, which captures modulator use associated with CHEC study visits. It does not capture the use of other modulators taken prior to or post CHEC enrollment unless those modulators were associated with other CHEC study visits. For participants who enrolled in CHEC while receiving ETI, 500 (28.3%) had enrolled at least once previously while taking a different modulator.
Figure 1. Area-proportional diagrams of CFTR modulator enrollments among CHEC-SC enrollees.
Plots highlight numbers of individual CHEC-enrollees for which sweat chloride data are available for only one versus for multiple CFTR modulators during CHEC-SC. Individual participants could re-enroll in CHEC-SC if their modulator prescription changed.
Genotype groupings
Genotype groupings within the study cohort, stratified by modulator, are summarized in Table 2. Out of 3787 SC concentrations obtained in CHEC-SC, 2412 (63.7%) were collected from F508del homozygous participants and 1167 (30.8%) were collected from F508del heterozygous participants. The remaining 208 (5.5%) of SC were obtained from participants without an F508del mutation. Out of the 1769 participants enrolled on ETI, 1002 (56.6%) were F508del homozygous and 734 (41.5%) were F508del heterozygous. Among F508del heterozygous participants receiving ETI, the majority had a minimal function mutation as their second variant (24.7% of participants on ETI). Out of the participants receiving ETI who did not carry an F508del variant (1.9%), all 33 carried at least one CFTR variants currently included in the ETI label.
Table 2.
CHEC-SC participant CFTR genotypes by modulator treatment at enrollment.
| IVA (N=536) | LUM/IVA (N=785) | TEZ/IVA (N=697) | ELE/TEZ/IVA (N=1769) | |
|---|---|---|---|---|
|
| ||||
| F508del with F508del | - | 784 (99.9%) | 626 (89.8%) | 1002 (56.6%) |
| F508del with | 379 (70.7%) | 1 (0.1%) | 53 (7.6%) | 734 (41.5%) |
| G551D | 172 (32.1%) | - | - | 62 (3.5%) |
| Other gating mutation | 24 (4.5%) | - | - | 8 (0.5%) |
| R117H | 63 (11.8%) | - | 1 (0.1%) | 32 (1.8%) |
| Splice mutation | 61 (11.4%) | - | 29 (4.2%) | 46 (2.6%) |
| Missense mutation | 56 (10.4%) | - | 17 (2.4%) | 63 (3.6%) |
| Minimal function mutation | 2 (0.4%) | 1 (0.1%) | 1 (0.1%) | 437 (24.7%) |
| Other mutation | 1 (0.2%) | - | 5 (0.7%) | 86 (4.9%) |
| G551D with | 71 (13.2%) | - | - | 9 (0.5%) |
| G551D | 1 (0.2%) | - | - | - |
| Other gating mutation | - | - | - | - |
| R117H | 3 (0.6%) | - | - | - |
| Splice mutation | 2 (0.4%) | - | - | - |
| Missense mutation | 7 (1.3%) | - | - | - |
| Minimal function mutation | 53 (9.9%) | - | - | 8 (0.5%) |
| Other mutation | 5 (0.9%) | - | - | 1 (0.1%) |
| Other gating mutation with | 11 (2.1%) | - | - | 2 (0.1%) |
| Other gating mu. ation | 1 (0.2%) | - | - | - |
| R117H | - | - | - | - |
| Splice mutation | 1 (0.2%) | - | - | - |
| Missense mutation | - | - | - | 1 (0.1%) |
| Minimal function mutation | 5 (0.9%) | - | - | - |
| Other mutation | 4 (0.7%) | - | - | 1 (0.1%) |
| R117H with | 25 (4.7%) | - | 2 (0.3%) | 6 (0.3%) |
| R117H | - | - | - | - |
| Splice mutation | 2 (0.4%) | - | 2 (0.3%) | 1 (0.1%) |
| Missense mutation | 4 (0.7%) | - | - | - |
| Minimal function mutation | 17 (3.2%) | - | - | 4 (0.2%) |
| Other mutation | 2 (0.4%) | - | - | 1 (0.1%) |
| Splice with | 25 (4.7%) | - | 12 (1.7%) | - |
| Splice mutation | 2 (0.4%) | - | 1 (0.1%) | - |
| Missense mutation | 1 (0.2%) | - | - | - |
| Minimal function mutation | 22 (4.1%) | - | 9 (1.3%) | - |
| Other mutation | - | - | 2 (0.3%) | - |
| Missense mutation with | 22 (4.1%) | - | 4 (0.6%) | 8 (0.5%) |
| Missense mutation | 4 (0.7%) | - | 1 (0.1%) | - |
| Minimal function mutation | 15 (2.8%) | - | 3 (0.4%) | 5 (0.3%) |
| Other mutation | 3 (0.6%) | - | - | 3 (0.2%) |
| Minimal function mutation with | 3 (0.6%) | - | - | 8 (0.5%) |
| Minimal function mutation | - | - | - | 4 (0.2%) |
| Other mutation | 3 (0.6%) | - | - | 4 (0.2%) |
| Other mutations | - | - | - | - |
Most frequent ten mutations by category. Full list of mutations by type can be found in the Supplemental Table S2.
Other gating mutations include: G1244E, G178R, G551S, S1251N, S1255P, S549N, S549R; Splice mutations include: 2789+5G->A, 3272–26A->G, 3849+10kbC->T, 711+3A->G E831X; Missense mutations include: A455E, D1152H, L206W, P67L, R117C, R347H, R352Q, S492F, S945 L, T338I; Minimal function mutations include: 1717–1G->A, 2184insA, 3659delC, 621+1G->T, G542X, I507 del, N1303K, R347P, R553X, W1282X
IVA = ivacaftor; LUM = lumacaftor; TEZ = tezacaftor; ELE = elexacaftor.
Cumulative distribution
Post-modulator cumulative SC distributions were shifted left (indicating higher probability of lower concentrations) compared with pre-modulator SC distributions (median SC 100 mmol/L [IQR: 90, 108.5]; Figure 2). Of the four modulators, the post-ETI SC distribution was shifted furthest to the left (median SC 42 mmol/L [IQR: 29, 57]) and the post-tez/iva SC distribution was shifted the least (median SC 91 mmol/L [IQR: 80, 99.5]). The median pre-modulator SC concentration for participants receiving ETI was 101 mmol/L [IQR:92, 109] (Figure 3). For all genotypes, cumulative post-ETI SC distributions were shifted to the left compared with pre-modulator distributions (Figures 3 and 4). The most common genotypes receiving ETI were F508del homozygous (n=1002; median post-ETI SC 39.8 mmol/L [IQR: 29, 53.4]) and F508del with a minimal function mutation (n=437; median post-ETI SC 51 mmol/L [IQR: 37.5, 63.5]).
Figure 2. Cumulative distribution of CHEC-SC participant sweat chloride concentrations by CFTR modulator.
Sweat chloride was obtained prior to modulator use and at CHEC enrollment (post-modulator). For each modulator, the vertical axis represents the percentage of participants with sweat chloride (mmol/L) measured at or below the corresponding x-axis value. Individuals who re-enrolled in CHEC on a different modulator were included once for each applicable distribution.
Figure 3. Cumulative distribution of CHEC-SC participant sweat chloride concentrations on ELE/TEZ/IVA by genotype.
Sweat chloride was obtained prior to modulator use and at CHEC enrollment (post-modulator). For each genotype, the vertical axis represents the percentage of participants with sweat chloride (mmol/L) measured at or below the corresponding x-axis value.
Figure 4. Cumulative distribution of CHEC-SC participant sweat chloride concentrations on ELE/TEZ/IVA by genotype, pre- and post-modulator.
Sweat chloride was obtained prior to modulator use and at CHEC enrollment (post-modulator). For each modulator, the vertical axis represents the percentage of participants with sweat chloride (mmol/L) measured at or below the corresponding x-axis value.
Sweat chloride categories
Among participants receiving ETI, 26% of SC concentrations were <30mmol/L, 79% were <60 mmol /L, and 7% were ≥80 mmol/L (Table 1), while 21% of SC concentrations were <30 mmol/L, 64% were <60 mmol/L, and 22.7% were ≥80 mmol/L among participants on ivacaftor. Fewer participants on lum/iva or tez/iva had SC concentrations <60 mmol/L (11% and 7%, respectively). Among participants on ETI, the observed average pre-modulator SC was greater for those with post-ETI SC ≥60 mmol/L than for those with post-ETI SC ≤ 30 mmol/L. In addition, 26% of male participants on ETI had post-modulator SC ≥60 mmol/L compared to 15% of female participants on ETI (Table S1). Lastly, 31% of participants with F508del/Minimal function variants had post-ETI SC ≥ 60 mmol/L compared to 18% of F508del homozygous participants and ~10% of F508del heterozygotes with a gating variant, R117H, or missense variant. (Tables S2, Figures 3 and 4). Distributions of SC by modulator are shown in Figure S2.
Association between post-ETI SC and participant characteristics among F508del homozygotes
The association between baseline characteristics and post-ETI SC concentration in F508del homozygous participants is presented in Figure S3. Based on the results of the regression analysis, post-ETI SC was 7.4 mmol/L less in females than males (95% CI = (−10.1, −4.7)). Additionally, post-ETI SC was 6.0 mmol/L less in participants between 6–11 years of age (95% CI = (−11.2, −0.8)) and 6.0 mmol/L greater in participants between 18–25 years of age (95% CI = (2.5, 9.6)), relative to participants 26 years of age and older. Lastly, post-ETI SC was 6.4 mmol/L less in subjects with pre-modulator SC <94 mmol/L (95% CI = (−10.2, −2.51)) and 4.3 mmol/L less in those with pre-modulator SC 94 to <102 mmol/L (95% CI = (−8.2, −0.4)), relative to participants with pre-modulator SC >110 mmol/L. A 10.8 mmol/L difference in post-ETI SC was observed in participants identifying as non-white or more than one race (n = 978), relative to white participants (n = 17), (95% CI = (−0.1, 21.7)). The adjusted R-squared from the full model was 0.09, suggesting that less 10% of the variability in post-ETI SC concentration was explained by baseline characteristics included in the regression analysis.
Conclusions
The CHEC-SC study is a large epidemiologic study that has captured SC concentrations in over 3,000 individuals with CF in the U.S. treated with a CFTR modulator, including 1769 individuals treated with ETI. Importantly, we observed large variability in post-modulator SC across individuals, with approximately 20% of PwCF having post-ETI SC concentrations in the CF diagnostic range, consistent with persistent CFTR protein dysfunction. As in the phase 1 CHEC-SC analysis (8), heterogeneity of post-modulator SC concentrations was driven largely by modulator, and by extension, genotypes for which those modulators were approved. However, within subgroups sharing the same genotype and modulator treatment, we observed substantial post-modulator SC variability, suggesting a range from incomplete to near-complete CFTR restoration. It remains to be determined whether additional CFTR restoration is possible and whether additional restoration would improve clinical outcomes for those with poorer post-modulator SC outcomes. Among participants with at least one F508del allele, F508del heterozygotes carrying minimal function variants were most likely to have post-ETI SC concentrations remaining in the CF diagnostic range (31%), while F508del heterozygotes carrying a gating or missense variant or R117H were the least likely, with ~90% having SC outcomes below 60mmol/L.
Although CFTR modulators, especially ETI, have been life-altering for many PwCF, some individuals do not experience a robust clinical outcome to treatment despite having modulator-responsive genetic variants. Proposed reasons for this include reduced adherence, poor drug absorption, drug-drug interactions through CYP-3A inhibitors or other mechanisms, undetected complex alleles, or the presence of modifier genes. (6) Sweat chloride provides one way to measure the impact of CFTR modulation on its target – the CFTR protein – and may inform our understanding of non-response.
In studies performed prior to the development of modulators, sweat chloride concentrations were found to correlate with genotype, in vitro predictions of CFTR function, clinical phenotypes (e.g., pancreatic insufficiency), and survival. (3, 14) Despite these relationships, diagnostic SC generally has not been used as a prognosticator as there is substantial individual variation and genotype has provided similar clinically relevant information (e.g., predicting pancreatic insufficiency), with the exception of variants like the R117H which have variable penetrance. (15) In the post-modulator era, in which SC concentrations are more broadly distributed within genotype groups, post-modulator SC concentrations have the potential to provide additional prognostic information for individuals and populations but should be interpreted cautiously, as earlier pre-modulator observations of an association between SC concentration and survival have yet to be repeated in PwCF receiving modulators. (3, 16) Investigations of the relationships between individual SC concentrations and short and long-term clinical outcomes remain active (17, 18) and are the goal of the CHEC-SC study. Importantly, PwCF and the CFTR variant N1303K have been shown to respond clinically to ETI without evidence of a corresponding reduction in SC concentration. (19)
We performed a regression analysis on F508del homozygous participants receiving ETI to assess the association between select baseline demographic data and post-modulator SC concentrations. Only modest associations were observed between post-ETI SC and sex, age group, and baseline SC concentration. However, limited associations were observed between post-ETI SC and race, ethnicity, and lung disease state. In this population, the model accounted for <10% of variability in post-modulator SC concentration, suggesting that considerable heterogeneity remains unexplained.
Our study was limited by an inability to measure variation in modulator exposure due to differences in adherence, drug-drug interactions, or dose adjustments. However, our results presumably reflect real world post-modulator SC heterogeneity occurring across treated CF populations today. Investigations into the relationship between serum drug levels, sweat chloride, and clinical outcomes will be important in unraveling causes of post-modulator SC heterogeneity. The use of local laboratories for sweat chloride analysis and reliance on historic SC measurements likely introduced some variability in our results, although our study design is supported by previously published preliminary data. (7) As in our prior report, the CHEC cohort remains limited in its recruitment of non-white participants, reflecting both a disproportionate lack of modulator eligibility among minority races and ethnicities across the US CF population and the continued institutional and cultural barriers to enrollment of those historically under-represented in clinical research. (20) Finally, we relied on clinical CF genotype reports. Absence of full CFTR sequencing likely contributed to misidentification of some CHEC-SC participants with complex variant alleles.
Despite these limitations, this observational study of PwCF eligible for CFTR modulators suggests that a substantial proportion of treated individuals continue to have evidence of CFTR dysfunction and may continue to experience detrimental effects despite modulator treatment. A key question is whether previously defined SC diagnostic cut-points (<60 mmol/L, 30–59 mmol/L, and <30 mmol/L) are the most informative to describe risk for post-modulator disease progression. As in those with N1303K, sweat chloride may not always reflect CFTR function and disease modification in other organ systems. Even in those PwCF with a post-modulator SC in the 30–59mmol/L range, there may yet be room for greater CFTR restoration, and potentially better future clinical outcomes and alteration of disease course. Conversely, as the relationship between SC concentration and CFTR activity is logarithmic, the relationship between SC change and CFTR restoration in the intermediate range may have more variability than with very low levels of CFTR activity, where small improvements lead to large changes in SC. (14, 21) In addition, minimum levels of CFTR activity necessary to prevent or reverse disease likely differ by organ system (pancreas vs. lungs, reproductive system), age, and disease stage. (22, 23)
Summary
We have described SC concentration distributions among PwCF taking the CFTR modulator ETI, with mean post-ETI SC concentrations lesser than those observed for previously approved modulators across all eligible genotypes. We have also described a subpopulation with higher post-modulator SC concentrations consistent with persistently reduced CFTR activity for which improved therapies – or optimization of current therapy (e.g., dose adjustment, modification of concurrent medications, improved drug absorption) may further mitigate CF disease progression.
Supplementary Material
Highlights.
Using data from the CHEC-SC study of modulator-induced sweat chloride (SC) concentrations, we describe the proportion of individuals with post-modulator SC above the diagnostic threshold for CF (≥ 60 mmol/L).
Across all FDA-approved modulators, elexacaftor/tezacaftor/ivacaftor (ETI) was associated with lower SC, with SC < 60 mmol/L in approximately 80% of individuals on ETI.
Among individuals with F508/minimal function variants or F508del/F508del, 31% and 18% had post-modulator SC ≥ 60 mmol/L respectively.
Higher post-modulator SC concentrations in a substantial number of individuals suggest that additional CFTR modulation may be achievable.
Acknowledgments
Special thanks to CF participants and families of CF children who participated in the study, and whose dedication to research made the trial possible. We would also like to thank the Cystic Fibrosis Foundation for supporting this study through the CF Therapeutics Development Network.
Funding Source:
ETZ was supported by the Cystic Fibrosis Foundation ZEMANI17K0 and NIH/NCATS Colorado CTSA Grant Number UL1 TR002535. NMH was supported by the Cystic Fibrosis Foundation HAMBLE17K0 and National Institutes of Health (NIH) grants P30 DK 089507 and UL1 TR002319. MWK was supported by the Cystic Fibrosis Foundation and NIH grant UL1TR002548.
Abbreviations:
- CFTR
cystic fibrosis transmembrane conductance regulator
- SC
sweat chloride
- ETI
elexacaftor/texacaftor/ivacaftor
Appendix: CHEC-SC Study Group Authors
TDN Coordinating Center: Julia Young, Katherine Odem-Davis, Susan Heideke, Jill VanDalfsen, Irene Bondick, Mark Warden, Melita Romasco, Kathy Seidel, Anna Mead, Iris Emerman, Greg Asmus, Diane Cook, Arthur Baines, Nicole Mayer-Hamblett
Participating Sites (Site Investigators (SI) and Research Coordinators (RC)):
Children’s Hospital Medical Center of Akron, Akron, OH – SI: Gregory Omlor; RC: Michelle Powers; University of Michigan, Michigan Medicine, Ann Arbor, MI – SI: Shijing Jia; RC: Jessica Carey; Emory University, Atlanta, GA – SI: Arlene Stecenko; RC: Eric Hunter; Children’s Healthcare of Atlanta and Emory University, Atlanta, GA – SI: Kevin Kirchner; RC: Joy Dangerfield; Johns Hopkins University, Baltimore, MD – SI: Natalie West; RC: Britany Zeglin; The Children’s Hospital Alabama, University of Alabama at Birmingham, Birmingham, AL – SI: Steven Rowe; RC: Jonathan Bergeron; Boston Children’s Hospital, Boston, MA – SI: Gregory Sawicki; RC: Robert Fowler, Monica Ulles; Massachusetts General Hospital, Boston, MA- SI: Lael Yonker; RC: Margot Hardcastle; The Cystic Fibrosis Center of Western New York, Buffalo, NY – SI: Carla Frederick; RC: Nadine Caci; University of North Carolina at Chapel Hill, Chapel Hill, NC – SI: George Retsch-Bogart; RC: Robin Johnson; Medical University of South Carolina, Charleston, NC – SI: Sylvia Szentpetery; RC: Shelia Parnell; Ann & Robert H; Lurie Children’s Hospital of Chicago, Chicago, IL – SI: Susanna McColley; RC: Larissa Rugg; University of Chicago, Chicago, IL – SI: Edward Naureckas; RC: Spring Maleckar; Cincinnati Children’s Hospital Medical Center, Cincinnati, OH – SI: Gary McPhail; RC: Rachel Dyke; Rainbow Babies and Children’s Hospital/University Hospitals Cleveland Medical Center, Cleveland, OH – SI: Erica Roesch; RC: Tia Rone; Nationwide Children’s Hospital, Columbus, OH – SI: Karen McCoy; RC: Terri Johnson; University of Texas Southwestern, Dallas, TX – SI: Raksha Jain; RC: Terri Visnick-Chang; University of Texas Southwestern / Children’s Health, Dallas, TX – SI: Meghana Sathe; RC: Mary Klosterman; National Jewish Health, Denver, CO – SI: Jennifer Taylor-Cousar; RC: Connor Balkissoon; Children’s Hospital Colorado, Aurora, CO – SI: Edith Zemanick; RC: Mary Cross; Children’s Hospital of Michigan, Detroit, MI – SI: Ibrahim Abdulhamid; RC: Debra Driscoll; University of Florida, Gainesville, FL – SI: Jorge Lascano; RC: Noni Graham; Helen DeVos Children’s Hospital, Grand Rapids, MI – SI: Susan Millard; RC: Heather Mulroy; Hershey Medical Center Pennsylvania State University, Hershey, PA – SI: Gavin Graff; RC: Diane Kitch; Baylor College of Medicine, Houston, TX – SI: Fadel Ruiz; RC: Omalee Lopez; Riley Hospital for Children, Indianapolis, IN – SI: Don Sanders; RC: Bre Bohart; University of Kansas Medical Center, Kansas City, KS – SI: Joel Mermis; RC: Lawrence Scott; Children’s Mercy Kansas City, Kansas City, MO – SI: Alvin Singh; RC: Miguel Bahena-Garcia; Dartmouth Hitchcock Medical Center, Lebanon, NH – SI: Alix Ashare; RC: Barbara Rodgers; University of Kentucky, Lexington, KY – SI: Jamshed Kanga; RC: Christina Payne; University of Arkansas for Medical Sciences, Little Rock, AR – SI: Larry Johnson; RC: Kathleen Hicks; Children’s Hospital of Los Angeles, Los Angeles, CA – SI: Carmen Reyes; RC: Daniel Quevedo; University of Wisconsin, Madison, WI – SI: Michael Rock; RC: Melanie Nelson; University of Tennessee CF Care and Research Center, Memphis, TN – SI: Tonia Gardner; RC: Catherine Horobetz; Children’s Hospital of Wisconsin, Milwaukee, WI – SI: Nicole Brueck; RC: Laura Roth; The Minnesota Cystic Fibrosis Center, Minneapolis, MN – SI: Joanne Billings; RC: Mary Bailey (Lynch) and Brooke Noren; West Virginia University – Morgantown, Morgantown, WV – SI: Kathryn Moffett; RC: Tammy Clark; Vanderbilt Children’s Hospital, Nashville, TN – SI: Rebekah Brown; RC: Brijesh Patel; Yale University School of Medicine, New Haven, CT – SI: Marie Egan; RC: Catalina Guzman; The Nemours Children’s Clinic – Orlando, Orlando, FL – SI: Floyd Livingston; RC: Sherry Wooldridge; Nemours Children’s Health – Pensacola, Pensacola, FL – SI: Okan Elidemir; RC: Holly Turner; Children’s Hospital of Philadelphia, Philadelphia, PA – SI: Clement Ren; RC: Jean Malpass; Children’s Hospital of Pittsburgh of UPMC, Pittsburgh, PA – SI: Joseph Pilewski; RC: Matt Butoryak; Oregon Health Sciences University, Portland, OR – SI: Aaron Trimble; RC: Brendan Klein; Primary Children’s Cystic Fibrosis Center, Salt Lake City, UT – SI: Fadi Asfour; RC: Judy Jensen; University of California San Diego, San Diego, CA – SI: Kathryn Akong; RC: Jenna Mielke; University of Washington Medical Center, Seattle, WA – SI: Moira Aitken; RC: Teresa Gambol; Seattle Children’s Hospital, Seattle, WA – SI: Ronald Gibson; RC: Sharon McNamara; SSM Health Cardinal Glennon Children’s Hospital, St. Louis, MO – SI: Gary Albers; RC: Freda Branch; St. Louis Children’s Hospital, St. Louis, MO – SI: Daniel Rosenbluth; RC: Sarah Smith; Toledo Children’s Hospital, Toledo, OH – SI: Bruce Barnett; RC: Kelly Hoot; New York Medical College at Westchester Medical Center, Valhalla, NY – SI: Allen Dozor; RC: Armando Ramirez
Footnotes
Conflict of Interest Statement:
ETZ reports personal consulting fees for advisory board participation and grant support to her institution for clinical trial participation from Vertex Pharmaceuticals. She has received personal consulting fees and grant support from the Cystic Fibrosis Foundation and grant support from the National Institutes of Health.
NMH serves as a consultant through her institution in her role as Executive Director of the CF Therapeutics Development Network Coordinating Center (CF TDNCC) and has received personal consulting fees from Vertex Pharmaceuticals and Enterprise Therapeutics. She has received grant funding from the Cystic Fibrosis Foundation (CFF) and National Institutes of Health (NIH).
DRV has received personal consulting fees from AbbVie, Albumedix, AN2, Aradigm, Armata, Arrevus, BiomX, Calithera, Chiesi USA, Cipla, Clarametyx, Corbus, CFF, cystetic Medicines, Eloxx, Enbiotix, Eveo, Galephar, Horizon, IBF, ICON clinical sciences, Ionis, Kala, Merck, Microbion, NDA, Protalix, PTC, Pyros, Pulmocide, Recida, Respirion, Savara, Vast, VRTX, and Zambon.
MWK has received personal consulting fees and grant support to his institution for clinical trial participation from AzurRx/FirstWave Biopharma, Insmed, Laurent, and Vertex. He has received personal consulting fees from Abbvie, cystetic Medicines, EnBiotix, Mylan, Nabriva, PBM BC Holdings, and Sionna. He has received grant support from the Cystic Fibrosis Foundation and the National Institutes of Health.
MNW has no conflicts to disclose.
MM has no conflicts to disclose.
IE has no conflicts to disclose.
CR has no conflicts to disclose.
JY has no conflicts to disclose.
KOD has no conflicts to disclose.
Declaration of interests
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Credit Author Statement: ETZ: Writing – original draft, conceptualization, investigation, funding acquisition; NMH: Writing- review and editing, visualization, conceptualization, investigation, methodology, supervision, funding acquisition; IE: formal analysis, methodology, software, validation, writing -initial draft, research, KOD: formal analysis, methodology, software, validation, writing, research, MM, MW: formal analysis, methodology, software, validation, writing, DV, CR: writing – review and editing, conceptualization, investigation, JY: Writing – review and editing, methodology, research, MWK : Writing – review and editing, conceptualization, investigation. All authors drafted the manuscript or revised it critically for important intellectual content. All authors approved the final version of the manuscript.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Bibliography
- 1.Stoltz DA, Meyerholz DK, Welsh MJ. Origins of cystic fibrosis lung disease. N Engl J Med. 2015;372(16):1574–5. [DOI] [PubMed] [Google Scholar]
- 2.Farrell PM, White TB, Ren CL, Hempstead SE, Accurso F, Derichs N, et al. Diagnosis of Cystic Fibrosis: Consensus Guidelines from the Cystic Fibrosis Foundation. J Pediatr. 2017;181S:S4–S15 e1. [DOI] [PubMed] [Google Scholar]
- 3.McKone EF, Velentgas P, Swenson AJ, Goss CH. Association of sweat chloride concentration at time of diagnosis and CFTR genotype with mortality and cystic fibrosis phenotype. J Cyst Fibros. 2015;14(5):580–6. [DOI] [PubMed] [Google Scholar]
- 4.Heijerman HGM, McKone EF, Downey DG, Van Braeckel E, Rowe SM, Tullis E, et al. Efficacy and safety of the elexacaftor plus tezacaftor plus ivacaftor combination regimen in people with cystic fibrosis homozygous for the F508del mutation: a double-blind, randomised, phase 3 trial. Lancet. 2019;394(10212):1940–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Middleton PG, Mall MA, Drevinek P, Lands LC, McKone EF, Polineni D, et al. Elexacaftor-Tezacaftor-Ivacaftor for Cystic Fibrosis with a Single Phe508del Allele. N Engl J Med. 2019;381(19):1809–19. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Castellani C. When triple therapy is not working: A reverse iceberg perspective. J Cyst Fibros. 2023;22(3):367–9. [DOI] [PubMed] [Google Scholar]
- 7.Zemanick ET, Konstan MW, VanDevanter DR, Rowe SM, Clancy JP, Odem-Davis K, et al. Measuring the impact of CFTR modulation on sweat chloride in cystic fibrosis: Rationale and design of the CHEC-SC study. J Cyst Fibros. 2021;20(6):965–71. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Mayer-Hamblett N, Zemanick ET, Odem-Davis K, VanDevanter D, Warden M, Rowe SM, et al. Characterizing CFTR modulated sweat chloride response across the cf population: Initial results from the CHEC-SC study. J Cyst Fibros. 2023;22(1):79–88. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.R Core Team. R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing; 2023. [Google Scholar]
- 10.Wasserman L. All of Nonparametric Statistics. New York, NY: Springer-Verlag; 2006. [Google Scholar]
- 11.Wickham H. ggplot2: Elegant Graphics for Data Analysis. New York, NY: Springer-Verlag New York; 2016. [Google Scholar]
- 12.Larsson J. eulerr: Area-Proportional Euler and Venn Diagrams with Ellipses. 2022.
- 13.Quanjer PH, Stanojevic S, Cole TJ, Baur X, Hall GL, Culver BH, et al. Multi-ethnic reference values for spirometry for the 3–95-yr age range: the global lung function 2012 equations. Eur Respir J. 2012;40(6):1324–43. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.McCague AF, Raraigh KS, Pellicore MJ, Davis-Marcisak EF, Evans TA, Han ST, et al. Correlating Cystic Fibrosis Transmembrane Conductance Regulator Function with Clinical Features to Inform Precision Treatment of Cystic Fibrosis. Am J Respir Crit Care Med. 2019;199(9):1116–26. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Espel JC, Palac HL, Bharat A, Cullina J, Prickett M, Sala M, et al. The relationship between sweat chloride levels and mortality in cystic fibrosis varies by individual genotype. J Cyst Fibros. 2018;17(1):34–42. [DOI] [PubMed] [Google Scholar]
- 16.Durmowicz AG, Witzmann KA, Rosebraugh CJ, Chowdhury BA. Change in sweat chloride as a clinical end point in cystic fibrosis clinical trials: the ivacaftor experience. Chest. 2013;143(1):14–8. [DOI] [PubMed] [Google Scholar]
- 17.Fidler MC, Beusmans J, Panorchan P, Van Goor F. Correlation of sweat chloride and percent predicted FEV(1) in cystic fibrosis patients treated with ivacaftor. J Cyst Fibros. 2017;16(1):41–4. [DOI] [PubMed] [Google Scholar]
- 18.Nichols DP, Paynter AC, Heltshe SL, Donaldson SH, Frederick CA, Freedman SD, et al. Clinical Effectiveness of Elexacaftor/Tezacaftor/Ivacaftor in People with Cystic Fibrosis: A Clinical Trial. Am J Respir Crit Care Med. 2022;205(5):529–39. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Sadras I, Kerem E, Livnat G, Sarouk I, Breuer O, Reiter J, et al. Clinical and functional efficacy of elexacaftor/tezacaftor/ivacaftor in people with cystic fibrosis carrying the N1303K mutation. J Cyst Fibros. 2023. [DOI] [PubMed] [Google Scholar]
- 20.McGarry ME, McColley SA. Cystic fibrosis patients of minority race and ethnicity less likely eligible for CFTR modulators based on CFTR genotype. Pediatr Pulmonol. 2021;56(6):1496–503. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Wine JJ. How the sweat gland reveals levels of CFTR activity. J Cyst Fibros. 2022;21(3):396–406. [DOI] [PubMed] [Google Scholar]
- 22.Castellani C, De Boeck K, De Wachter E, Sermet-Gaudelus I, Simmonds NJ, Southern KW, et al. ECFS standards of care on CFTR-related disorders: Updated diagnostic criteria. Journal of Cystic Fibrosis. 2022;21(6):908–21. [DOI] [PubMed] [Google Scholar]
- 23.Sermet-Gaudelus I, Girodon E, Vermeulen F, Solomon GM, Melotti P, Graeber SY, et al. ECFS standards of care on CFTR-related disorders: Diagnostic criteria of CFTR dysfunction. J Cyst Fibros. 2022;21(6):922–36. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.