Exocrine pancreatic insufficiency (PI) is present from birth in approximately 85% of patients with cystic fibrosis (CF). It is generally accepted practice to initiate pancreatic enzyme replacement therapy (PERT) immediately after the diagnosis of PI to allow optimal absorption of nutrients for adequate growth.
Current recommendations state that enzymes should be administered to all infants with CF fed infant formula and solid foods containing macronutrients (1,2). The capsule(s) should be opened and the beads placed on a spoon containing a small amount of applesauce, infant formula, or fruit puree and provided directly before the feed (1–3). FitzSimmons et al (4) advises that enzyme dosing should begin with 1000 U lipase /kg/meal, and should not exceed 10,000 U lipase /kg/day to decrease the risk of fibrosing colonopathy. Little is known about safe and effective dosing of pancreatic enzyme therapy in infants and young children (5). There are no controlled datasets that allow true discrimination between effective doses of pancreatic enzymes in this age group. Several consensus documents advise certain dosages of pancreatic enzymes for treatment of infants, but the recommendations are based solely on anecdotal reports (2).
The objectives of the present study were to evaluate efficacy, safety and palatability of different doses of Pancrease MT (McNeil Consumer Healthcare, Ft Washington, PA) in infants with CF younger than 30 months old.
PATIENTS AND METHODS
The present study was a phase II randomized, investigator-blinded, parallel-group pilot study in infants with CF with PI. The study design included a run-in period (days 1–5) and a randomization period (days 6–11). Study visits were scheduled on day 1 (visit 1), day 6 (visit 2), and at the end of the study on day 11 (visit 3). At visit 1, parents were instructed to administer 500 U lipase/kg/meal, with a maximum of 5 doses per day, for 5 full days on an outpatient basis. Current pancreatic enzymes were stopped. The 500 U lipase/kg dose was chosen for baseline (and for 1 of the treatment groups) because it could, in theory, be effective in reducing steatorrhea. This low dose is deemed appropriate given the lack of clinical data supporting higher doses and given the short-term duration of this trial. Because pancreatic enzyme supplementation is a vital therapy in the treatment of children with PI, it was chosen not to completely interrupt PERT for a given number of days. At visit 2, children were randomly assigned (according to a computer-generated randomization schedule) to 1 of the 4 parallel treatment groups in a 1:1:1:1 ratio as follows: 500, 1000, 1500, or 2000 U (Ph. EUR) lipase/kg/meal. All of the dosages are represented in presentation as European Pharma Units, which are equivalent to 0.8 USP units of lipase activity. To preserve the blind for the treating physicians, an unblinded research pharmacist at the study site packaged the study medication. A carton was dispensed to the parent or guardian at visits 1 and 2, and contained 6 bottles of study medication. Each bottle contained 5 capsules. Each capsule contained the meal-specific dose required for each dose group and the appropriate number of microtablets for the subject's weight in kilograms with a maximum of 10,000 U lipase/kg/day.
Parents were instructed to record in a diary the date, time, and amount of all of the food and formula consumed by the subject. The date and time of study medication were also recorded in the diary.
During the 2 study periods, the patients had to maintain a defined diet containing 125% to 150% of the recommended daily allowance appropriate for patients with CF. No sample size estimation was performed for this study. The number of subjects included was based on the anticipated availability of subjects in this age group during the planned study period at the planned study sites.
Infants with CF, ages 6 to 30 months, with a diagnosis of CF confirmed by genetic testing or abnormal sweat testing in combination with clinical signs consistent with a diagnosis of CF were included. All of the subjects had a documented history of abnormal coefficient of fecal fat absorption (CFA) or lower than 15 μg fecal elastase per gram of stool, confirming a diagnosis of CF-related PI. All of the subjects were stable on current pancreatic enzyme therapy. Exclusion criteria were hypersensitivity to porcine products; herbal supplements or concomitant corticosteroid therapy; use of nasogastric tube feeding for supplemental enteral nutrition; exacerbation of chronic lung infections; and clinically significant vomiting, malnutrition, severe dehydration or uncorrected electrolyte disorders. Patients were recruited from 5 university medical centers in Belgium and the Netherlands; each center included 4 patients, except for 1 center that included 2 patients.
Written parental informed consent was obtained. The study protocol was approved by local independent ethics committees or institutional review boards of each participating institution and was conducted in accordance with the Declaration of Helsinki.
To evaluate preliminary efficacy, the primary endpoint was the change in CFA measured from baseline (visit 2, randomization) to the end of the study period (visit 3). Stools for quantification of fecal fat were collected during the last 72 hours of the first period (baseline period) and of this randomization period (end of study period). The daily fat intake during the study was recorded and calculated. CFA of fat was determined as follows:
The percentage of carbon dioxide (13CO2) expired (cumulative % 13C) by a 13C MTG (13C mixed triglyceride) substrate breath test was calculated to be a surrogate marker of lipase activity. The 13C MTG breath test was performed twice: first during the third day of the stool collection after the run-in phase (at baseline, day 6) and again after the randomized phase of the study (day 11) under supervision of a breath test technician.
The test meal was prepared just before ingestion. A volume of water corresponding to the subject's normal intake was heated to 60°C and mixed with 1 g of polyethylene glycol to solubilize 100 mg 1,3-distearyl,2-[13C-carboxyl] octanoyl glycerol. The solution was shaken and an adequate quantity of formula powder was added to obtain a 15% formula concentration. The contents were stirred and cooled to room temperature. Fifty percent of the energy content of this test meal was provided by fat. Breath samples were collected in the following manner: a nasal prong (6 French Argyle) was placed in the nasopharynx and 10-mL end-tidal breath was sampled slowly during expirations with a syringe. The breath samples were stored in vacutainers. The sampling occurred twice before the meal and at 15-minute intervals for a period of 6 hours after the meal. During the test the infant/toddler was kept as quiet as possible and no other examinations were allowed. The breath test was started after a fast of at least 3 hours, and the next feeding was allowed 3 hours after the first breath sample. Vacutainers were transported to the laboratory in which 13C enrichment of expired air was determined with an isotope-ratio mass spectrometer (IRMS, ABCA, Cheshire, UK). Results were expressed as the percentage of 13CO2 recovery per hour and as the cumulative value after 6 hours. Only subjects with adequate breath samples were included for analysis, that is, subjects with 2 breath test samples that contained sufficient CO2 for analysis. Weight changes during the study period were calculated.
Parents made a daily assessment of palatability (ease of swallowing) of the microtablets by the infant/child using a 4-point scale (0 = poor, 1 = fair, 2 = good, 3 = excellent) in responding to the following question: “How easy to swallow do you feel the study medication is?”
At each visit, safety assessments consisted of routine physical examinations, including weight, height, head circumference, and measurements of vital signs (blood pressure, axillary temperature, and pulse). Any adverse symptom or sign that a subject experienced during the study was recorded as an adverse event. An assessment of the seriousness of these events was also recorded.
Demographic characteristics including age, sex, ethnicity, and race were summarized. Descriptive statistics (n, mean, median, minimum, and maximum) were calculated for all of the efficacy parameters for each treatment group. For the safety analysis, the number and percentage of subjects with specific adverse events were tabulated by treatment group before and after the randomization, using Medical Dictionary for Regulatory Activities (MedDRA) terms.
The populations analyzed for this study were enrolled subjects, randomized subjects, and subjects who completed the study. All of the safety analyses were based on the enrolled subjects (subjects who met the study inclusion criteria and whose parents signed the consent form). All of the enrolled subjects who completed the run-in period and were randomized to study treatment were included in the randomized subject population. Efficacy analyses were based on all of the randomized subjects who completed the study. Treatment compliance was evaluated by checking the study medication bottles (including empty bottles) at each visit. The measurement of compliance during a study period was defined as:
Patient Demographics and Dosing
A total of 18 subjects were enrolled; 16 completed the study. One parent pair withdrew consent before randomization and 1 withdrew consent after their child had been randomized to the 1000 U lipase/kg/meal group (Fig. 1).
The median age of the 17 randomized children was 16.5 months (min 5.8, max 29.6). There were 12 girls (12/18) (Table 1). Mean compliance for each treatment group during the randomization period was 89% to 99%.
The median daily dose was 2200 USP (1600–2500) U lipase/kg with a mean daily fat intake of 4 g/kg during the run-in period (n = 17). Table 1 shows the median daily dosages during the experimental period in the 4 groups. The average daily intake of energies, fat, and protein was comparable across time and among treatment groups.
During the run-in period, the median daily fat intake was 42 g/day (22–84); during the randomized period the median daily fat intake was between 30 (range 20–36) and 41 (range 27–49) g/day in the different groups. The changes in median weight from screening to the end of the study were 0.05 kg (range −0.1 to 0.2) in the 500-U group, 0.30 kg (range –0.1 to 0.7) in the 1000-U group, –0.05 kg (range −0.2 to 0.1) in the 1500-U group, and 0.15 kg (range −0.3 to 0.5) in the 2000-U group. None of the reported changes in daily fat intake and weight were considered clinically relevant. At the time of randomization, the median CFA of fat was 93% (range 91%–96%) in the 500-U group, 90% (range 85%–95%) in the 1000-U group, 83% (range 67%–93%) in the 1500-U group, and 92% (range 90%–96%) in the 2000-U group. The changes in mean CFA of fat were –2% in the 500-U group, 1% in the 1000-U group, –1% in the 1500-U group, and –2% in the 2000-U group. Of the 16 patients who performed both breath tests, 12 (3 in each treatment group) provided adequate breath samples for analysis. Inadequate breath sampling resulted from technical problems. During the run-in period the median cumulative % 13C was 11 (range −8 to 59). After randomization, the median cumulative % 13C was 18 (range 14–23) in the 500-U, 14 (range −1 to 17) in the 1000-U, 10 (range 10–27) in the 1500-U and 3 (range 1–49) in the 2000-U groups.
The median palatability scores were 2.8 during the run-in period (range 0–3) and 2.6 (range 0.3–2.8) during the randomization period. Mean scores ≥2 were observed in the 500-U, 1000-U, and 1500-U groups. The highest dose group scored a median of 1.8.
Three adverse events were reported in the 18 enrolled patients (17%) during the open-label run-in period and 4 events during the randomized period. During the run-in period complaints were diarrhea (1), vomiting (1), and rhinitis (1). In the 500 U/kg/meal treatment group complaints were abdominal pain (1), abnormal stools (1), and increased bowel movements (1). One patient randomized to the 1000 U/kg/meal treatment group experienced constipation. In the 2000 U/kg/meal treatment group, there was 1 event of vomiting and another event of rhinitis.
The intensity and nature of adverse events were similar for all of the doses of Pancrease MT administered during the study. No subjects withdrew from the study because of adverse events.
We investigated the efficacy, palatability, and safety of pancrelipase microtablets in infants and toddlers. Based on ethical reasons, we refrained from the theoretically optimal design, the placebo-controlled trial. We considered that even temporary complete cessation of PERT would harm the patient in an undue manner. Rather, we chose for a low-dose baseline dosage during a run-in period, followed by randomization toward 4 different PERT dosages. The randomized allocation to treatment was chosen to reduce potential selection and observer bias in this study and to minimize the chances of uneven distribution on factors that could influence the results or prognosis. The 500 U lipase/kg dose chosen for run-in and 1 of the treatment groups were associated with a CFA of 89%. A CFA of 89% is high compared to CFA at diagnosis of PI, but lower than obtained in healthy controls (normal: 95% in infants and 97% in children) (6). After randomization, subjects maintained a CFA close to 90% to the end of the study. This observation suggests that pancrelipase microtablets in a dosage higher than 500 U lipase/kg/meal does not increase the CFA in these patients. The classical 72-hour fecal fat collection with calculation of a CFA is based on assessment of fat absorption as the difference between fat intake and fat output in the stool (1). Fecal fat reflects the result of various factors contributing to fat absorption (pancreas, hepatobiliary, and mucosal disorders), and thus is not by itself specific enough to evaluate exocrine pancreatic function. In CF, it is known that nonpancreatic factors can contribute to the mechanism of fat malabsorption (7). The 13C MTG breath test is an alternative test to evaluate exocrine or supplemented lipase activity (8). The 13C MTG breath test correlates with duodenal lipolytic activity because of both residual endogenous and exogenous pancreas activity (9,10). Age-specific test meals and breath-sampling techniques for the MTG breath test have been defined with normal values for the pediatric population. Reference values in healthy children with corresponding age for cumulative % 13C exhalation after performing a 13C MTG breath test in comparable conditions are between 14% and 44% (11).
In this study, we measured a large variation in cumulative % 13C with the median cumulative % 13C in the 500- and 1000-U groups in the low-normal range and the median cumulative % 13C in the 1500- and 2000-U groups below the normal range.
The large variation in our results could be caused by a wide range of underlying residual endogenous lipase activity and perhaps insufficient breath sampling. However, the within-subject variability and interindividual variation of the 13C MTG breath test in our results are consistent with previous reports, especially in conditions of mild fat malabsorption (7,8,12).
Palatability varied from poor to excellent in the different groups. Overall, mean palatability was scored as fair to good by the parents in each of the treatment groups.
Pancrease MT did not cause serious adverse events in this small study population. Gastrointestinal disorders were the most frequently reported adverse events, consistent with the known gastrointestinal effects of PERT. The intensity and nature of adverse events were similar for all doses of Pancrease MT administered during the study. No subjects withdrew from the study because of adverse events.
In conclusion, treatment with Pancrease MT at a dosage of 500 U lipase/kg/meal resulted in a CFA of approximately 89% in most pediatric subjects ages 6 to 30 months with PI caused by CF. Pancrease MT doses from 500 to 2000 USP units lipase/kg/meal were well tolerated and mean palatability was scored as fair to good, but neither of the higher dose regimens increased the coefficient of fat absorption in these patients. A randomized trial in a larger patient group with longer follow-up is needed to confirm these results.
The authors gratefully thank all of the parents of participating patients and dieticians and nursing staff of the various institutions for their assistance.
1. Borowitz D, Baker RD, Stallings V. Consensus report on nutrition for pediatric patients with cystic fibrosis. J Pediatr Gastroenterol Nutr 2002; 35:246–259.
2. Sinaasappel M, Stern M, Littlewood J, et al
. Nutrition in patients with cystic fibrosis: a European consensus. J Cyst Fibros 2002; 1:51–75.
3. Borowitz DS, Grand RJ, Durie PR. Use of pancreatic enzyme supplements for patients with cystic fibrosis in the context of fibrosing colonopathy. Consensus Committee. J Pediatr 1995; 127:681–684.
4. FitzSimmons SC, Burkhart GA, Borowitz D, et al
. High-dose pancreatic-enzyme supplements and fibrosing colonopathy in children with cystic fibrosis. N Engl J Med 1997; 336:1283–1289.
5. Munck A, Duhamel JF, Lamireau T, et al
. Pancreatic enzyme replacement therapy for young cystic fibrosis patients. J Cyst Fibros 2009; 8:14–18.
6. Littlewood JM, Wolfe SP, Conway SP. Diagnosis and treatment of intestinal malabsorption in cystic fibrosis. Pediatr Pulmonol 2006; 41:35–49.
7. Kalivianakis M, Verkade HJ, Stellaard F, et al
. The 13
C-mixed triglyceride breath test in healthy adults: determinants of the 13
response. Eur J Clin Invest 1997; 27:434–442.
8. Herzog DC, Delvin EE, Albert C, et al
C-labeled mixed triglyceride breath test (13
C MTG-BT) in healthy children and children with cystic fibrosis (CF) under pancreatic enzyme replacement therapy (PERT): a pilot study. Clin Biochem 2008; 41:1489–1492.
9. Weaver LT, Amarri S, Swart GR. 13
C mixed triglyceride breath test. Gut 1998; 43(suppl 3):S13–S19.
10. Vantrappen GR, Rutgeerts PJ, Ghoos YF, et al
. Mixed triglyceride breath test: a noninvasive test of pancreatic lipase activity in the duodenum. Gastroenterology 1989; 96:1126–1134.
11. Dijk-van Aalst K, Van Den Driessche M, Der Schoor S, et al
C mixed triglyceride breath test: a noninvasive method to assess lipase activity in children. J Pediatr Gastroenterol Nutr 2001; 32:579–585.
12. Kalivianakis M, Elstrodt J, Havinga R, et al
. Validation in an animal model of the carbon 13-labeled mixed triglyceride breath test for the detection of intestinal fat malabsorption. J Pediatr 1999; 135:444–450.