Evaluation of the Effectiveness of In-line Immobilized Lipase Cartridge in Enterally Fed Patients With Cystic Fibrosis

Abstract

Background: 

Pancreatic insufficiency occurs in most patients with cystic fibrosis (CF) contributing to malnutrition. In the United States, 3600 patients with CF require enteral feeding (EF). Oral pancreatic enzymes are commonly used with EF, despite not being designed or approved for this use. An immobilized lipase cartridge (ILC) for extracorporeal digestion of enteral feedings was developed. The sponsor provided it to patients via a structured program, which we evaluated to assess the effectiveness of the ILC on nutritional status.

Methods: 

The program provided the ILC to patients prescribed the device while reimbursement efforts were ongoing. Baseline anthropometric data were obtained and subsequent measurements of height, weight, and body mass index (BMI) were collected at 6 and 12 months.

Results: 

Inclusion criteria were met by 100 patients (age = 0--45 years). Over 12 months of use in patients >2 years of age (n = 93), there were significant improvements seen in height and weight z-scores with improvement trend seen in BMI. The frequency of achieving the 50th percentile increased steadily for weight and BMI from baseline to 12 months but not for height.

Conclusions: 

This evaluation of a program to assist patient access to ILC demonstrates that better growth is possible over standard of care. The association of ILC use with significant improvements in anthropometric parameters over a 12-month period in people with CF demonstrates the effectiveness of ILC as rational enzyme therapy during enteral feedings.

What Is Known/What Is New

What Is Known

• Exocrine pancreatic insufficiency is common in cystic fibrosis.
• Fat malabsorption related to exocrine pancreatic insufficiency negatively impacts nutritional status and pulmonary function.
• Traditional pancreatic enzyme replacement therapy is not designed for use with enteral feedings.

What Is New

• Following use of immobilized lipase cartridge out to 12 months, patients experienced significantly steady increase in height and weight and steady improvement trend in body mass index.
• Fifty percentage of patients achieved ≥50th percentile for body mass index at 12 months, compared with 37.1% at baseline.
• Immobilized lipase cartridge use demonstrated statistically significant improvements in growth in patients with cystic fibrosis requiring enteral feedings.

Exocrine pancreatic insufficiency (EPI) occurs in 85% to 90% of patients with cystic fibrosis (CF) and leads to the inability to absorb long-chain, calorie-dense fats and fatty acids (FA), such as arachidonic acid, eicosapentaenoic acid (EPA), and docosahexaenoic acid (DHA) ⁽¹⁾. Fat malabsorption negatively effects growth in children and BMI in adults; both of which are important to maintain pulmonary function ^(2,3). As a result of caloric deficiencies, approximately 3600 patients with CF in the United States require enteral feeding ⁽⁴⁾.

Pancreatic enzyme replacement therapy (PERT), provided as capsules taken orally before every meal and snack, is the standard of care to support digestion in patients with CF who have EPI ⁽⁵⁾. Given the lack of any alternatives, PERT has been used with continuous enteral feedings, despite not being designed or approved for use in this setting. The inconsistency in clinical efficacy that exists with PERT products used during enteral feeding is multifactorial and includes patient compliance, product formulation, frequency, and mode of delivery ⁽¹⁾. In fact, prospective studies supporting approval and use of oral PERTs excluded patients on enteral nutrition ⁽⁶⁾. Current treatment practices to crush tablets and capsules or open pancreatic enzyme capsules are not consistent with the labeling for the majority of PERTs. Only 1 FDA-approved PERT (PERTZYE, Cheisi, Inc.) has a package insert with instructions on how to deliver the contents of the lowest dose capsule (4000 USP lipase unit) through gastrostomy tubes 14 French or larger ⁽⁷⁾. It is estimated that the duration of action of oral PERTs is about 45 to 60 minutes following administration ⁽⁸⁾, and therefore rational dosing to match enzyme to substrate would require multiple doses administered throughout continuous overnight feedings to be effective. This administration schedule is often abandoned for an inadequate but more tolerable dosing schedule that includes PERT before overnight feedings and may include PERT in the middle of the feeding and/or after the feeding.

Two recently published studies demonstrated the safety, tolerability, and effect on FA absorption of a new enzyme strategy to aid fat digestion with continuous enteral feedings ^(9,10), a single use digestive cartridge containing immobilized lipase (RELiZORB; Alcresta Therapeutics, Newton, MA). The cartridge connects in-line with an enteral feeding set. As enteral formula flows through the cartridge, immobilized lipase enzyme hydrolyzes intact triglyceride fats within the formula into more absorbable forms, whereas the lipase is retained within the cartridge.

Although increased levels of omega-3 FAs in both red blood cells and in plasma and a decrease in the frequency and severity of some symptoms of malabsorption was observed with immobilized lipase cartridge (ILC) use, changes in anthropometric measures were not systematically reported ^(9,10). Payors have been reluctant to cover the cost of this ILC because of the lack of long-term clinical data ⁽¹¹⁾. A reimbursement program was designed to provide free access to the ILC in those patients for whom it was prescribed while awaiting insurance coverage. Although initially intended to run for 90 days, eligibility was extended for 12 months. Evaluation of the data collected during the course of this program provided an opportunity to evaluate extended and consistent use of the ILC on clinical parameters of nutritional status including BMI, height, and weight. This real-world data provides unique insight into the clinical performance and outcomes of the ILC ^(12,13).

MATERIALS AND METHODS

The ILC sponsor (Alcresta Therapeutics) initiated a program in late 2016 to provide assistance to patients/caregivers to gain access to the ILC and cover out-of-pocket costs for patients who were prescribed the device while efforts to obtain reimbursement were ongoing with applicable payers. The initial third-party pharmacy hub did not collect clinical data; we report on a revised program that used a different third party that began collecting data in 2018. Patients were eligible for the program if they had a diagnosis of CF, utilized enteral nutrition via gastrostomy tube because of poor nutritional status related to CF, and had been receiving oral PERTs for enteral feeding before starting ILC use. The program originally anticipated providing the ILC free of charge for 90 days. Due to lengthy appeal processes, the program was extended to 12 months. Patients/caregivers who enrolled in the study signed a consent form that expressly authorized the sponsor to “collect information related to RELiZORB treatment to assist in the coordination of care, efforts to obtain reimbursement for RELiZORB, and to demonstrate the safety and efficacy of RELiZORB.” As a program evaluation, Institutional Review Board oversight and Informed Consent were not required ⁽¹⁴⁾.

Upon enrollment, patients received either an in-clinic demonstration and/or verbal instructions on the use of the ILC and supporting patient education materials including a toll-free phone number for technical assistance. Patients continued use of whichever formula their clinician prescribed. PERT use for meals and snacks was unchanged.

Baseline patient referral data included a unique external subject ID, number of shipments/boxes, disease/diagnosis code, date of birth, sex, prescriber, date of first and last digestive cartridge shipment, formula volume (mL/hour), height (cm), weight (kg), and body mass index (BMI, kg/m²) at date of enrollment. Data on height, weight, and BMI at all subsequent follow-up intervals was collected.

The baseline value is the closest measurement to the start of the treatment period with the ILC, from −2 months to the day of first shipment. The closest subject measurement within the follow-up time window for 6 months (+/− 2 months) and 12 months (+/− 2 months) was captured. To be included in the analyses, patients need to receive ≥6 months of the ILC and were not included if they went >45 days without a shipment being requested. This was based on an estimate that a real-world regimen with moderately good adherence would require 30 cartridges over 45 days, an estimated use of 4 nights/week. Patients/caregivers were required to request additional cartridges. In 2019, near the end of the program, the hub called patients to ask if they needed another shipment. Only patients with a baseline measurement were included in the program analysis presented.

To compare growth in CF patients using enteral feedings in the absence of ILC, we examined data in the CF Foundation Patient Registry (CFFPR) during the calendar year 2014, before the availability of the ILC (CFFPR-2014); ILC use was not recorded in the CFFPR until 2018 and could have confounded a contemporaneous comparison. The overwhelming majority of patients included in this analysis were noted to be pancreatic-insufficient (PI) based on documentation of PERT use (for detailed methods see Table 3).

Weight, height, and BMI were normalized to age- and sex-specific z-scores and percentiles of healthy individuals ages 0 to 24 months of age using the World Health Organization Child Growth Standards ⁽¹⁵⁾ and for those 2 to 20 years of age using data from the Centers for Disease Control and Prevention growth charts ⁽¹⁶⁾. The CF Foundation recommends that all patients should be evaluated sequentially for weight and height changes. For children and adolescents ages 2 to 20 years, it is recommended that BMI percentiles should be maintained at or above the 50th percentile. In this analysis, we assessed changes for height and weight gain and evaluated BMI relative to these goals.

Statistical Plan

Summary statistics for continuous variables included the number of nonmissing observations, the mean, standard deviation (SD), median, 25th and 75th percentile, minimum and maximum. Frequency distributions counts and percentages for categorical variables are provided. A mixed model repeated measures (MMRM) analysis of variance model was used to estimate the least-squares (LS) mean and standard error (SE) of anthropometric measurements (height, weight, BMI) and their z-scores and percentiles relative to baseline anthropometric measures. Differences in LS means comparing months 0 versus 6, 0 versus 12, and 6 versus 12 are presented. Statistical significance was set at P ≤ 0.05. SAS Software (SAS Institute, Cary, NC) was used for all data management and statistical analysis.

RESULTS

Between the start of the program in June 2018 and its termination in October 2019, 178 patients were enrolled. The 100 participants (49% girls) who had at least 1 anthropometric measurement available at baseline and subsequent measurements at 6 and 12 months of use and are included in this analysis. Figure 1 describes analysis criteria. Ages ranged from 0 to 45 years with a mean (SD) age of 9.8 (7.0) years (Table 1). Patients received care at 15 different CFF-accredited care center programs.

FIGURE 1: Subjects enrolled and data analyzed.

TABLE 1 - Demographic characteristics in subjects with anthropometric measures at baseline

Parameter		All (0 to 45y)	Ages 0 to <2 y	Ages 2 to <5 y	Ages 5 to <12 y	Ages 12 to <18 y	Ages >18 y
Participants Enrolled	N	100	7	18	49	21	5
Sex
Male	N	51	3	10	22	14	2
Female	N	49	4	8	27	7	3
Weight, kg	Mean (SD)	28.9 (15.8)	8.2 (2.4)	13.8 (1.9)	25.9 (6.4)	49.0 (9.0)	57.3 (16.6)
Height, cm	Mean (SD)	125.18 (2.9)	69.2 (9.3)	94.3 (5.3)	125.6 (12.1)	160.1 (9.4)	163.6 (6.5)
BMI, kg/m2	Mean (SD)	16.9 (2.7)	N.A.	15.5 (1.4)	16.1 (1.6)	19.0 (2.9)	21.2 (5.1)
BMI Percentile	Mean (SD)	40.8 (27.3)	N.A.	41.2 (31.3)	43.3 (24.7)	36.4 (29.5)	N.A.
Weight for length percentile (WHO)	Mean (SD)	N.A.	61.4 (24.5)	N.A.	N.A.	N.A.	N.A.

BMI = body mass index; SD = standard deviation; WHO = World Health Organization.

Upon entry into the program, the rate of enteral feedings was recorded. The distribution of mean enteral formula volume per hour by age is in Supplemental Table 1; https://links.lww.com/MPG/C45. Shipments were made not more than once a month (30 ILC cartridges per box; 1 cartridge supports 500 mL of enteral formula). There was no reconciliation of actual ILC use or remaining patient supply at the end of the program. The mean number (SD) of boxes shipped was 12.8 (7.46) over 12 months; the mean number of cartridges shipped was 384.6 (223.8), and the mean estimated number of cartridges used per month was 33.5 (20.33).

Ninety-five percentage of participants were <18 years of age. As the majority of subjects were in their active growth years, normalized data were critical. There were significant improvements in height and weight z-scores in the first 6 months of use (P < 0.001 and P = 0.002, respectively) and over 12 months of use (P = 0.036; P = 0.017, respectively). The rate of increase in anthropometric measures for patients 2 to 18 years of age (Table 2) was more pronounced in the first 6 months of ILC use (height z-score LS mean [SE]: 0.17 (0.058), P = 0.005; weight: 0.20 [0.079], P = 0.014) and stabilized or had modest gains from 6 to 12 months of use (height: −0.05 [0.45]; weight −0.01 [0.077] neither statistically significant). These findings were supported by a comparison to the historical control group (Table 3). Compared with 1868 patients in the CFFPR-2014 group (mean age = 10.24 years), the ILC patients (mean age = 9.8 years) had statistically significant increases in overall height and weight z-score at 6 months (mean treatment difference = 0.20 and 0.24, P = 0.001 and 0.048, respectively). At 12 months, parity was seen for all anthropometric parameters. Age group comparison to historic controls is shown in Supplemental Table 2, https://links.lww.com/MPG/C38.

TABLE 2 - Growth measurement summary and change over time at months 0, 6, and 12 postdigestive cartridge start in 93 subjects ages 2 to 18 years with anthropometric measures at baseline

Growth measurement summary				Change over time
	Month 0	Month 6	Month 12	0 to 6 m		0 to 12 m		6–12 m
Parameter	LS mean (SE)			LS mean (SE)	P value	LS mean (SE)	P value	LS Mean (SE)	P value
Height (cm)	129.40 (2.602)	133.11 (2.480)	135.68 (2.489)	3.71 (0.357)	<0.001	6.29 (0.390)	<0.001	2.58 (0.244)	<0.001
Height z-score	−0.92 (0.110)	−0.75 (0.107)	−0.80 (0.112)	0.17 (0.058)	0.005	0.12 (0.063)	ns	−0.05 (0.45)	ns
Height percentile	26.52 (2.544)	30.00 (2.680)	28.83 (2.789)	3.47 (1.26)	0.001	2.31 (1.463)	ns	−1.16 (1.200)	ns
Weight, kg	30.45 (1.586)	32.23 (1.488)	34.32 (1.734)	1.78 (0.334)	<0.001	3.87 (0.422)	<0.001	2.09 (0.466)	<0.001
Weight z-score	−0.84 (0.105)	−0.64 (0.106)	−0.63 (0.104)	0.20 (0.079)	0.014	0.20 (0.093)	0.033	−0.01 (0.077)	ns
Weight percentile	27.71 (2.424)	31.62 (2.597)	31.72 (2.837)	3.91 (1.618)	0.018	4.02 (2.492)	ns	0.10 (2.070)	ns

SE = standard error.

TABLE 3 - Comparison of change in height, weight and body mass index z-scores for immobilized lipase cartridge to CF Foundation Patient Registry Historical Controls

		Historical controls				ILC
										Six month treatment difference		12-month treatment difference
Age group	Parameter	n	First measurement mean	Last measurement mean	Difference within 1 y	n	Baseline (±2 mo)	6 mo (±2 mo)	12 mo (±2 mo)	Mean @ 6 mo	P value	Mean @ 12 mo	P value
Overall
	Height z-score	1868	−1.00	−0.98	0.02	100	−0.97	−0.77	−0.84	0.20	0.001	0.13	0.070
	Weight z-score	1868	−0.98	−0.89	0.09	100	−0.88	−0.64	−0.66	0.24	0.048	0.22	0.147
	BMI z-score	1868	−0.47	−0.37	0.09	100	−0.33	−.21	−0.09	0.12	0.728	0.23	0.226

Patients who had consented to have data entered into the CF Foundation Patient Registry (CFFPR) and were receiving enteral feedings at any time during 2014 were eligible to be in the historical control group. Out of the 1868 patients, 1847 were identified as pancreatic insufficient based on the documented need for pancreatic enzyme therapy (PERT). To be included, patients had to have at least 2 measurements of height and weight at least 6 months apart and not be receiving treatment with ivacaftor. The duration of enteral feedings and the method of nutrient digestion was unknown. ILC = immobilized lipase cartridge.

Four of the 7 patients with baseline data between the ages of 0 and 24 months were in a different age category by the 6 and 12-month follow-up dates, therefore, the numbers were too small in this group to categorize weight-for-length growth patterns over the year of use. In the 88 patients 2 to 18 years of age, there was a trend toward significant increase in BMI z-scores (P = 0.059) (Supplemental Table 3, https://links.lww.com/MPG/C39).

The CF Foundation recommends specific goals only for BMI, but we analyzed the frequency of achieving the 50th percentile for height and weight as well as BMI in the 88 patients ages 2 to 18 years. This parameter increased steadily for weight and BMI from baseline to 12 months but not for height. Height percentiles reached 50% in 20.2%, 23.6%, and 21.1% of patients at 0, 6, and 12 months, respectively. Weight percentiles reached 50% in 18%, 25.5%, and 28.9% of patients at 0, 6, and 12, respectively. BMI reached 50% in 37.1%, 49.1%, and 50.0% in patients at 0, 6, and 12 months, respectively.

DISCUSSION

Malnutrition remains a significant issue for a subset of patients with CF and is associated with worse pulmonary outcomes, quality of life, and survival ^(5,17–20). For many of these patients with refractory malnutrition and poor growth, enteral feeds through gastrostomy tubes are recommended. There have been improvements in the overall nutritional status of the US CF population over the past decades and the use of enteral nutrition has contributed to some of these trends. In fact, the baseline BMIs for the patients reported here is relatively high, indicating that their use of enteral nutrition before this program had been of some benefit. Nonetheless, the fact that we demonstrate additional improvements in the highly clinically relevant nutritional parameters, height, weight, and BMI z-scores over a 12-month period, indicate that better growth is possible with improved nutrient absorption. We also saw improvements in height and weight z-scores in the first 6 months of use relative to a historical cohort. These observations regarding the effectiveness of the ILC support this novel approach to pairing digestive enzymes with substrate during nocturnal enteral feedings.

There are several possible explanations for the observed increases in anthropometric parameters. Most clinicians recommend that PERT capsules be taken orally at the beginning and the end of nocturnal enteral feedings despite a lack of evidence for this strategy ⁽²⁾. In preclinical studies, the ILC used and evaluated in this program was determined to digest up to >90% of long-chain fats in a variety of formulas ⁽²¹⁾. At 9.2 calories per gram, every additional volume of lipids completely digested and available for absorption will lead to weight gain, likely explaining the improvements in weight z-score we observed. We noted significant increases in height z-scores within the ILC patients and in comparison to the CFFPR-2014 group. Although increases in height in individuals with CF treated with highly effective modulators suggest that stunting may be a manifestation of CF ⁽²²⁾, better nourished patients with CF tend to have better height growth, which in turn is associated with improved pulmonary outcomes ⁽¹⁷⁾. There was more of improvement in growth parameters in the first 6 months of ILC utilization compared with the later 6 months. This can be attributed to catch-up growth with better nutrient absorption as well as possible normal variation of pulsatile growth seen in children ^(23–25). Micronutrients as well as macronutrients may affect growth. Fatty acid imbalances can play a critical role in the production of eicosanoid inflammatory mediators and may contribute to manifestations of CF ^(20,26,27). Studies evaluating FA supplementation have demonstrated a decrease in pulmonary exacerbations, antibiotic use, improved fat-free mass, and grip strength associated with increases in DHA and EPA levels ^(28,29). Previously, use of this ILC demonstrated improvements in EPA and DHA profiles ^(9,10). In addition, the continuous action of lipase upon triglycerides in enteral formulas decreases undigested nutrients in the gut lumen ⁽²⁹⁾. Not only are malabsorbed fats unavailable as an energy source, they likely contribute to symptoms, such as bloating, nausea, and loose stools. In a more severely malnourished cohort of patients with CF who required enteral feedings, short-term use of the ILC led to improvements in gastrointestinal symptoms and preservation of appetite and breakfast consumption ⁽⁹⁾; improvements in symptoms were also seen with longer term use ⁽¹⁰⁾. Although not measured in this program, it is possible that a reduction in these unintended consequences of nocturnal enteral feedings may have contributed to improved use and increased nutrient intake.

Malnutrition represents a significant economic burden in the United States. Indeed, malnourished patients in the general population account for up to a 300% increase in hospital costs ⁽³⁰⁾. They are at greater risk of complications and readmissions; 4 to 6 days longer hospital length of stay ^(30,31) and a 54% higher likelihood of hospital readmission within 30 days ⁽³²⁾. In patients with CF, malnutrition during childhood is associated with increased pulmonary exacerbations throughout life ^(17,20). When PERTs are opened and beads used in whole or crushed in an attempt to match enzyme and substrate, there is an increased risk of clogged feeding tubes ⁽³³⁾, thus contributing to increased healthcare costs. This program was created to improve the nutritional status and quality of life of individuals with CF whose poor nutritional status was being managed with enteral feedings. Program evaluations, such as this are designed to provide postmarketing data to inform future care and reimbursement decisions. Although not designed as a study of medical economics, the results of this program provide evidence supporting the efficacy of this device and its coverage by third party payors.

This was not a traditional research study but rather the information reported is part of a program evaluation ⁽¹⁴⁾. As such, there are limitations to interpretation of the data. Real-world use of the ILC was a criterion for entry into the program, thus no control group was recruited. A comparator group, however, was analyzed from the CFFPR. Although imperfect, this comparator group analysis supported the salutary effect of ILC on height and weight growth. Of the 178 patients who entered the program, 78 (44%) had missing data after baseline, and were therefore, excluded from analysis. This could introduce selection bias. It is possible that those who did not have complete data may have been less likely to adhere or alternatively, may have had improved growth, and therefore did not continue enteral feedings. Data, however, included in the supplement for all patients in the program indicates the same trends of improvement in both weight and height. Data before consent was not captured in this program, and therefore we are unable to comment on growth velocity before baseline. Similarly, we do not have information regarding CFTR mutations or concomitant medication use and do not know whether subjects started CFTR modulator drugs during the observation period. The modulator ivacaftor has been demonstrated to lead to significant increases in weight and height ^(21,33) and approximately 4% of individuals with CF are eligible to receive this treatment. Therefore, at most only 4% of subjects in the study would possibly have been on ivacaftor, minimally impacting results noted.

Lumacaftor/ivacaftor and tezacaftor/ivacaftor were available for many patients during the period of data collection; however, the impact of these drugs on weight and growth is far less than observed with ivacaftor in patients with the G551D CFTR mutation ^(21,33,34). Although an estimated 90% of people with CF will be eligible to receive the highly efficacious modulator combination of elexacaftor, tezacaftor, and icavaftor ⁽³⁵⁾, it was not approved by the US Food and Drug Administration until October of 2019 and only for those over 12 years of age. Three quarters of the patients in this program were under the age of 12 years, so for those reasons modulator use with this triple combination is unlikely to bias these data.

CONCLUSIONS

In summary, this program evaluation demonstrates significant long-term clinical growth improvements in people with CF who use an ILC to enable digestion of enteral feedings. When the CFF enteral nutritional guidelines were written in 2016, no clear recommendations with regard to use of PERT during enteral feeds were included given the lack of available data ⁽²⁾. Poor nutrition significantly impacts important clinical outcomes for people with CF and use of the ILC was associated with improved height and weight, adding to previous data showing that it improves fatty acid absorption and quality of life. These results support the notion that ILC be considered the new standard of care in patients with CF and EPI to provide rational enzyme therapy during enteral feedings. Future studies providing prospective case-controlled comparisons of the benefits of ILC in enterally fed patients with CF would add to the current literature.