Reduced Intestinal Inflammation With Lumacaftor/Ivacaftor in Adolescents With Cystic Fibrosis

Abstract

A chronic intestinal inflammation may occur in patients with cystic fibrosis (CF), while no therapeutic management is proposed. Although Lumacaftor/Ivacaftor is well-known to modulate the defective cystic fibrosis transmembrane conductance regulator (CFTR) protein in lungs, no data are available on the impact of this treatment on CF intestinal disorders. We, therefore, investigated the evolution of intestinal inflammation after initiation of Lumacaftor/Ivacaftor in CF adolescents (median of follow-up: 336 days [IQR: 278;435]). Median fecal calprotectin concentrations decreased significantly after Lumacaftor/Ivacaftor initiation (102 μg/g [IQR: 69–210]) compared with the baseline (713 μg/g (IQR:148–852), P = 0.001). To our knowledge, this study showed for the first time that CF-related intestinal inflammation is improved by Lumacaftor/Ivacaftor treatment.

What Is Known/What Is New

What Is Known

• Chronic intestinal inflammation can occur in cystic fibrosis children.
• No management is currently available to treat intestinal inflammation in cystic fibrosis.

What Is New

• Intestinal inflammation decreases in the majority of patients after initiation of Lumacaftor/Ivacaftor therapy.
• Intestinal inflammation after Lumacaftor/Ivacaftor initiation may evolve independently of the respiratory function.
• Fecal calprotectin level should be an important outcome to consider in the evaluation of therapies with Cystic Fibrosis Transmembrane Conductance Regulator modulators.

An infographic is available for this article at: https://links.lww.com/MPG/B896.

During the past decade, improvement in the management of people with cystic fibrosis (CF) has led to an ageing patient population. Although CF morbidity and mortality still mainly depend on respiratory function, impact of the disease on other organs currently represents new challenges.

It is now established that chronic intestinal inflammation may affect patients with CF, even in the absence of digestive symptoms (1,2). Fecal calprotectin, a biomarker reflecting intestinal inflammation, is significantly higher in children with CF compared with healthy controls and has been correlated with the severity of the genetic disease, as well as with intestinal dysbiosis (3,4). Such chronic intestinal inflammation may have a short- and long-term impact on morbidity and mortality in patients with CF, with consequences on nutritional status, recognized to be a predictor of lung function deterioration and overall survival (5). Dhaliwal et al showed that fecal calprotectin levels were inversely correlated with weight and height z-scores in CF children (6). In addition, a CF murine model study showed that reduction of intestinal inflammation improved weight growth (7).

Moreover, it has recently been shown that patients with CF are at major risk of small bowel and colon cancers, which occur more frequently (standardized incidence ratio, 18.9; 95% confidence interval [CI]: 9.4–38.3) and 10.9 [95% CI: 8.4–14.11], respectively) and earlier (median age at diagnosis, 32 ± 9 vs 68 ± 11.1 years, respectively) in people with CF compared with the general population (8). The mechanisms responsible for the increased risk of gastrointestinal cancer in patient with CF are unclear. The loss, however, of cystic fibrosis transmembrane conductance regulator (CFTR) gene function in the intestinal tract in mice is associated with intestinal tumor formation, suggesting to consider the CTFR as a tumor suppressor gene (9). CF-associated intestinal dysbiosis could also be linked to the development of gastrointestinal cancer. For example, Fusobacterium and Escherichia coli, which have been associated to colorectal cancer, are over-represented in intestinal microbiota of children with CF compared with healthy controls (10,11). Moreover, chronic intestinal inflammation is an established risk factor for colorectal cancer, correlated with the intensity and duration of the inflammation in Inflammatory Bowel Diseases (IBD) (12,13). Taken together, it can be assumed that chronic intestinal inflammation, occurring from childhood, may represent a risk factor for intestinal malignancy in CF (4,14,15).

Consequently, the management of intestinal inflammation must be integrated into CF care. Currently, no specific anti-inflammatory treatments are proposed to patients with CF. On the one hand, although some studies indicate efficacy regarding probiotics’ ability to reduce intestinal inflammation, they are not recommended in CF because of inter-individual variability and a limited level of evidence (16). On the other hand, the modulation of the defective CFTR protein may be another potential therapeutic option. Ivacaftor, a CFTR potentiator for CF patients with class III mutations, such as G551D, is effective in reducing intestinal inflammation but this treatment only involves 4% of CF patients (17). The new Lumacaftor/Ivacaftor combination is indicated for F508del homozygous patients, representing 40% of the whole Caucasian CF population. Although Lumacaftor/Ivacaftor is well-known to modulate the defective CFTR protein in lungs, limited data are available on the impact of this treatment on CF intestinal disorders. In the present study, we investigated intestinal inflammation following the initiation of Lumacaftor/Ivacaftor in CF adolescents, by monitoring fecal calprotectin concentrations.

MATERIALS AND METHODS

Patient Cohort

To conduct this pilot study, we retrospectively gathered data from the medical records of adolescents with CF between 12 and 18 years of age, followed at the Bordeaux or Toulouse University Hospitals who started Lumacaftor/Ivacaftor between January 2016 and July 2018 and who had fecal calprotectin assay at the time of treatment initiation or in the previous month. Fecal calprotectin assays before and more than 3 months after initiation of Lumacaftor/Ivacaftor were collected, along with concomitant clinical and laboratory data, such as weight, height, body mass index (BMI), sweat test, fecal elastase, and lung function from each patient's medical records. Pancreatic insufficiency was defined as a fecal elastase <200 μg/g.

At the time of the patient follow-up period, Lumacaftor/Ivacaftor had already received marketing authorization in France and was prescribed to the patient as part of routine care. The retrospective data collected and analyzed were obtained exclusively from patients’ medical records and no specific interventions were performed for this study. In this context, according to the law in force, the approval of an ethics committee was not required. Patients were informed about this study and gave their consent for the use of their data.

Measurements of Fecal Calprotectin

The fecal calprotectin assays used were performed at the time of treatment initiation or during the previous month for the baseline levels, and during an annual check-up for postinitiation levels. If more than 1 fecal calprotectin assay was available after 3 months of treatment, only the first assay was considered. No pulmonary exacerbation was reported in the medical records at the time of the different calprotectin measurements. In both participating study centers, concentrations of fecal calprotectin were measured by a quantitative enzyme immunoassay (fCAL; Bühlmann, Schönenbuch, Switzerland), as per the manufacturer's instructions. The result was initially interpreted quantitatively. Then, we applied a cut-off of 250 μg/g, recently validated to predict endoscopic lesions in IBD, in order to identify patients with a significantly increased calprotectin concentration (4,18).

Statistical Analysis

A nonparametric Wilcoxon matched-pairs test was used to compare quantitative variables before and after the onset of treatment. Correlations were calculated using the Spearman method. Statistical analysis was performed using the R studio program (version 1.1.463 for Windows) and plotted using GraphPad Prism (version 5.01 for Windows); a P value <0.05 was considered indicative of statistical significance.

RESULTS

Baseline Patient Characteristics

Of the 32 adolescents who initiated Lumacaftor/Ivacaftor during the study period, 15 of them had a fecal calprotectin performed in the month before initiation of treatment and were included in the present study. Their characteristics at baseline are summarized in Table 1. All patients had exocrine pancreatic insufficiency. Lumacaftor/Ivacaftor treatment was initiated between 12 and 16 years of age (median: 12 years, interquartile range [IQR]: 12–14.5). The median concentration of calprotectin was 713 μg/g (IQR: 148–852); 9 patients (60%) showed a concentration above 250 μg/g.

TABLE 1 - Summary of patient characteristics at baseline and during reassessment after Lumacaftor/Ivacaftor administration

				Baseline						Reassessment post-Lumacaftor/Ivacaftor
Patient	Site	Age	Sex	FC (μg/g)	Weight z-score	Height z-score	BMI z-score	ppFEV1	Delay (days)	Δ FC (μg/g)	Δ Weight z-score	Δ Height z-score	Δ BMI z-score	Δ ppFEV1	Δ SC (mmol/L)	EPF
1	T	12	M	154	−0.65	−0.18	−0.68	101	210	−133	−0.52	−0.38	−0.58	−15	N/A	PI
2	T	12	M	49	−0.09	0.33	−0.29	104	381	40	0.45	0.15	0.41	3	−5	PI
3	B	12	F	78	−1.18	0.21	−1.98	72	308	−18	0.65	−0.19	1.16	13	−7	PI
4	B	12	M	903	−1	−0.18	−1.28	71	916	−653	0.21	−0.19	0.26	3	N/A	N/A
5	B	12	M	142	−2.41	−2.05	−1.56	63	659	112	−0.71	−0.19	−1.11	−1	N/A	PI
6	T	12	M	352	−1.33	−1.85	−0.26	71	336	−250	0.22	0.14	0.06	12	−11	PI
7	B	12	M	903	0.04	0.29	0.02	89	294	−869	0.25	0.15	0.08	−3	N/A	N/A
8	T	12	M	73	−1.35	−0.86	−1.16	88	366	−3	0.33	0.3	0.17	−5	−27	PI
9	T	12	F	189	−1.57	−1.34	−1.24	107	201	−121	0.62	0.36	0.6	7	N/A	PI
10	B	13	M	1800	−1.47	−1.27	−1.02	70	756	−1629	0.67	0.38	0.46	23	−2	PI
11	T	14	F	754	−0.16	−0.98	0.28	109	490	−624	−0.61	−0.35	−0.35	3	−17	PI
12	T	15	M	713	−0.87	−0.99	−0.53	97	103	−582	0.56	0.25	0.53	−12	−16	PI
13	T	15	M	1457	−1.54	−1.49	−0.98	90	368	−1110	−0.12	0.02	0.31	14	−15	PI
14	B	15	F	719	−1.51	−0.91	1.22	65	267	−644	0.01	0.13	−2.4	7	N/A	PI
15	B	16	M	801	−1.94	−1.68	−1.35	98	290	−352	0.01	0.15	0	−16	−17	PI
Median (IQR)		12 (12–14.5)		713 (148–852)	−1,3 (−1.5 to −0.8)	−1 (−1.4 to −0.2)	−0.1 (−1.3 to −0.3)	89 (71–99.5)	336 (279–436)	−352 (−649 to −70)	0.2 (−0.1 to 0.5)	0.1 (−0.2 to 0.2)	0.2 (−0.2 to 0.4)	3 (−4 to 9.5)	−15 (−17 to −7)

Δ = evolution following initiation of Lumacaftor/Ivacaftor. B = Bordeaux; BMI = body mass index; EPF = exocrine pancreatic function; F = females; FC = fecal calprotectin; M = males; N/A = not available; PI = pancreatic insufficient; ppFEV₁ = percent predicted forced expiratory volume in 1 second; SC = sweat chloride test; T = Toulouse.

The median z-score for weight, height, and BMI were −1.3 (IQR: −1.5 to −0.8), −1 (IQR: −1.4 to −0.2) and −1 (IQR: −1.3 to −0.2), respectively. The median percent of predicted Forced Expiratory Volume in 1 second (ppFEV₁) was 89% (IQR: 71–99.5). At inclusion, these parameters were not correlated with calprotectin concentrations (Table 1).

Outcome After Initiation of Lumacaftor/Ivacaftor

The reassessment visit was conducted at a median time of 336 days (IQR: 278–435) after treatment onset. The evolution of patients is summarized in Table 1. Compared with baseline, fecal calprotectin concentrations decreased significantly (7-fold) after at least 3 months of treatment with Lumacaftor/Ivacaftor (median [IQR]: 713 μg/g (148–852) vs 102 μg/g (69–210), respectively, P = 0.001] (Fig. 1A). Calprotectin levels, however, remained above 250 μg/g in 4 patients (27%) and increased in 2 patients (13%) (Fig. 1B). We did not observe any significant difference in outcome regarding weight, height, and BMI z-score, or ppFEV₁. Changes in sweat test and calprotectin were not correlated. Noteworthy, all patients remained pancreatic-insufficient (missing data for 2 patients) (Table 1).

FIGURE 1: Overall (A) and per patient (B) evolution in fecal calprotectin concentrations after initiation of Lumacaftor/Ivacaftor. The error bars on the boxplot represent the maximum and minimum values.

DISCUSSION

Intestinal inflammation is now widely accepted as an integral part of CF disease and its management is becoming a major topic. Limited data are, however, available on the impact of CFTR modulators on this digestive tract disorder. In this pilot study, we showed for the first time that modulation of CFTR function by Lumacaftor/Ivacaftor was associated with a significant decrease of intestinal inflammation, using fecal calprotectin concentration as a surrogate marker. To note, this decrease of intestinal inflammation was not correlated with respiratory function changes, suggesting that CF-related digestive disorders may evolve independently from pulmonary disease, as previously reported (15). Interestingly, fecal calprotectin levels increased on Lumacaftor/Ivacaftor in 2 patients. It is possible that some patients are unresponsive to treatment at the gastrointestinal level, suggesting an interindividual variability in the treatment response, as it has been reported for pulmonary function (19). Intercurrent factors, such as postantibiotic dysbiosis, may also occur and contribute to intestinal inflammation. Studies are needed to better characterize and understand persistent intestinal inflammation on Lumacaftor/Ivacaftor in these patients.

Probably because of a limited number of patients in this study, ppFEV₁ as well as nutritional and anthropometric measurements were not significantly improved by Lumacaftor/Ivacaftor. Moreover, we were unable to identify correlations between intestinal inflammation and nutritional status at baseline and at follow-up, even if it has been described in the literature (6). It should be reminded that nutritional status of patients in CF is multifactorial and an integral part of the CF care, which probably did not enable us to observe a link between the evolution of intestinal inflammation and nutritional status.

This pilot study has several limitations. Intestinal inflammation has been assessed by fecal calprotectin with a cut-off of 250 μg/g. It was reported that calprotectin levels can vary according to the immunoassay used, limiting comparability between studies (20). Thus, the same method was used for all dosages in our study, which allows for better inter- and intra-individual comparability. Furthermore, the findings of this study need to be confirmed using a larger sample size. The evolution of fecal calprotectin over time in CF adolescents has been poorly studied. Ellemunter et al (15) conducted a longitudinal assessment of fecal calprotectin concentrations in CF patients over a period of up to 12 years and found stable concentrations between 10 and 20 years. A control group could be used to better discern the part of treatment in the evolution of intestinal inflammation. Lumacaftor/Ivacaftor, however, has marketing authorization for F508del homozygous patients older than 12 years of age, and it is therefore, impossible to match the control group on both age and mutations. Future studies should also investigate the evolution of digestive symptoms on Lumacaftor/Ivacaftor treatment.

The association between gut microbiota composition and intestinal inflammation has already been mentioned in the literature (4,12,17). Published studies suggest the link between inflammation, dysbiosis and cancer risk in CF alimentary tract disorders (14). Larger scale longitudinal studies are warranted to follow the simultaneous evolution of microbiota and inflammation when Lumacaftor/Ivacaftor is administered to CF patients. In addition, in vitro and in vivo studies will be required to decipher the pathophysiological role of Lumacaftor/Ivacaftor on the digestive mucosa. In the long-term, registry analysis will be valuable in fully documenting the impact of protein therapies on CF digestive cancer risk.

CONCLUSIONS

This study showed for the first time that modulation of CFTR function by Lumacaftor/Ivacaftor was associated with a significant decrease of intestinal inflammation in a group of adolescents with CF. The short- and long-term consequences of such a decrease in intestinal inflammation have yet to be determined, especially regarding the risk of bowel malignancy in CF.

In conclusion, reduced intestinal inflammation was seen with Lumacaftor/Ivacaftor treatment in a group of adolescents with CF. The short- and long-term consequences of such a decrease in intestinal inflammation have yet to be determined, especially regarding the risk of bowel malignancy in CF.