Orlistat Therapy for Children With Type 1 Hyperlipoproteinemia: A Randomized Clinical Trial

Abstract

Context

Patients with type 1 hyperlipoproteinemia (T1HLP), a rare genetic disorder, have extreme chylomicronemia and recurrent episodes of acute pancreatitis. Currently, the only therapeutic option is to consume an extremely low-fat diet because the triglyceride-lowering medications are not efficacious.

Objective

To determine the efficacy of orlistat, a gastric and pancreatic lipase inhibitor, in reducing serum triglyceride levels in patients with T1HLP.

Design and Setting

We conducted a randomized, open-label, clinical trial with a four-period, two-sequence (“orlistat” and “off orlistat” for 3 months), crossover study design.

Patients

Two unrelated young Asian Indian males (11 and 9 years old) with T1HLP due to homozygous large GPIHBP1 deletions were enrolled at the UT Southwestern Medical Center. The patients were randomized to receive 3 months of orlistat or no therapy (off), then crossed over to the other arm, and this sequence was then repeated. Fasting serum triglyceride levels, fat-soluble vitamins, and gastrointestinal side effects were assessed.

Results

Compared with the two off periods, orlistat therapy reduced serum triglycerides by 53.3% and 53.0% in patient 1 and 45.8% and 62.2% in patient 2. There was no deficiency of fat-soluble vitamin levels, and their growth continued. There were no serious adverse effects of orlistat; patient 1 had a mild increase in passage of gas and bloating, and patient 2 had constipation with mild stool leakage.

Conclusion

Orlistat is safe and highly efficacious in lowering serum triglycerides in children with T1HLP and should be the first-line therapy in conjunction with an extremely low-fat diet.

Patni et al. conducted a randomized clinical trial of orlistat in two children with type 1 hyperlipoproteinemia and showed that it was well tolerated and lowered serum triglycerides by more than 50%.

Type 1 hyperlipoproteinemia (T1HLP), also known as familial chylomicronemia syndrome, is a rare, autosomal recessive condition characterized by extreme hypertriglyceridemia due to accumulation of chylomicrons resulting in recurrent episodes of acute pancreatitis (1, 2). In most patients, T1HLP is due to lipoprotein lipase (LPL) deficiency, but biallelic mutations in lipase maturation factor 1 (LMF1), apolipoprotein CII (APOC2), apolipoprotein A-5 (APOA5), and glycosyl-phosphatidylinositol–anchored high-density lipoprotein–binding protein 1 (GPIHBP1) have been reported as well (3). Patients with T1HLP have eruptive or tuberous xanthomas, recurrent pancreatitis, lipemia retinalis, and hepatosplenomegaly. Visceral xanthomas and pancreatic atrophy have also been reported (4). Most important, acute pancreatitis is often a cause for increased morbidity and even mortality in these patients. Treatment of these patients poses a challenge because the current medications for hypertriglyceridemia, such as fibrates, niacin, and ω-3 fatty acids (docosahexaenoic and eicosapentaenoic acids), are ineffective (2, 5). The only effective therapy is an extremely low-fat diet with <10% to 15% of the total energy as fat (2, 6).

Because the basic defect in T1HLP is the reduced clearance of chylomicrons due to impaired lipolysis of triglycerides (TGs), reduction in dietary fat by reducing chylomicron formation can lower serum TGs. However, despite strict dietary adherence, some patients continue to have severe hypertriglyceridemia and recurrent acute pancreatitis. Orlistat, a gastric and pancreatic lipase inhibitor, can lower dietary fat absorption by 30% (7) and is currently approved for weight reduction for obese children age 12 years and older (8). Orlistat, by lowering dietary fat absorption, may reduce serum TG levels in patients with T1HLP by decreasing the substrate available for chylomicron formation. Norgren et al. (9) have demonstrated safety of orlistat in obese children ages 7 to 12 years. They found no serious adverse effects (including liver toxicity and vitamin deficiencies) and no negative effects of orlistat on psychological or physical well-being or linear growth (9). Because there are no previous controlled trials of orlistat therapy in patients with T1HLP, we determined the efficacy and safety of orlistat in two young patients with T1HLP using a randomized, crossover, open-label trial.

Patients and Methods

The protocol was approved by the UT Southwestern Medical Center Institutional Review Board. The participants and parents signed a written informed consent.

Patients

Patients were considered for study if they were ≥8 years old with T1HLP and had fasting serum TG levels >1000 mg/dL. Patients were excluded if they had any secondary causes of hypertriglyceridemia; liver disease; severe anemia; illicit drug use; major surgery in the past 3 months; congestive heart failure; serum creatinine >2.5 mg/dL; cancer within the last 5 years; gastrointestinal surgery in the past; current therapy with anticoagulants, digoxin, antiarrhythmics, or cyclosporine; chronic malabsorption syndromes; cholestasis; acute illnesses such as acute pancreatitis in the last 8 weeks; or pregnancy or lactation.

Study design

Because T1HLP is a rare disease, we designed an alternating, repeated crossover trial as per Lagakos (10). A randomized, open-label, crossover trial with four periods and two sequences (“orlistat” and “off” for 3 months each) was conducted. A double-blind, placebo-controlled trial was not planned as symptomatic side effects of orlistat described in previous studies would make it ineffective. The study design allowed for patients to act as their own controls and thus increase the power of the study. The total duration of each patient’s participation was 1 year. Primary end point of the study was the mean fasting serum TG levels. Primary safety end points of the study included serum fat-soluble vitamin levels and gastrointestinal side effects (oily stools, anal leakage, diarrhea, and flatus with discharge).

Materials

For our study, we used 60-mg orlistat capsules (Alli; GlaxoSmithKline, Pittsburg, PA). We used a graded system for dosing of orlistat based on the weight of the child (Supplemental Table 1).

Study procedures

Patients were counseled to consume a diet with <15% of total energy as fat and a multivitamin supplement daily containing 3500 IU vitamin A (71% as retinol and 29% as β-carotene), 400 IU (as cholecalciferol) vitamin D, 30 IU (as dl-α-tocopheryl acetate) vitamin E, and 25 μg (as phytonadione) vitamin K throughout the study. They were instructed to take vitamins at least 2 hours apart from orlistat dosing. The parents were instructed on how to record in detail all the foods and drinks consumed over 3 days (2 weekdays and 1 weekend day) in the food record books provided during the baseline and the four study periods. The 3-day food records were analyzed for nutrients using the SuperTracker database, which is based on the Food and Nutrient Database for Dietary Studies and the Food Patterns Equivalents Database, both from the US Department of Agriculture Food Surveys Research Group. The patients were then randomized to receive 3 months of orlistat therapy or no therapy (off period), and this alternating sequence was repeated once. At the end of each 3-month study period, fasting blood samples were drawn for 3 consecutive days for serum TGs (the primary end-point variable). We also measured fat-soluble vitamins, complete blood counts, serum chemistry, liver function tests (alanine aminotransferase, aspartate aminotransferase, bilirubin, alkaline phosphatase, γ-glutamyl transferase), glucose, hemoglobin A1C, creatine phosphokinase, total protein, albumin, uric acid and lactate dehydrogenase, and thyroid function tests. All patients completed a gastrointestinal symptoms questionnaire during the visits.

Statistical analyses

The randomization sequence was generated using SAS PROC PLAN (SAS Institute, Cary, NC). Data were summarized with descriptive statistics and graphics for each subject, and adverse events were cautiously recorded.

Results

Two unrelated Asian Indian boys (11 and 9 years old) with T1HLP were enrolled. Patient 1 had a homozygous deletion spanning 54,623 bp on chromosome 8 (4), and patient 2 harbored a homozygous deletion spanning 17,499 bp on chromosome 8, with both deletions encompassing GPIHBP1 gene (11). Baseline characteristics of study patients are described in Table 1.

Table 1.

Baseline Features of Patients

Characteristic	Patient 1	Patient 2
Mutation	Homozygous 54,623-bp del encompassing GPIHBP1 gene (4)	Homozygous 17,499-bp del encompassing GPIHBP1 gene (11)
Age, y	11	9
Race	Asian Indian	Asian Indian
Age at diagnosis, mo	1	2
Sex	Male	Male
Height, m	1.45	1.24
Weight, kg	30.9	21.2
BMI (percentile for age)	4	10
Puberty status	Prepubertal	Prepubertal
Highest serum TGs, mg/dL	2274	37,248

Abbreviation: BMI, body mass index.

Compared with the two off periods, orlistat therapy reduced mean fasting serum TGs by 53.3% and 53.0% in patient 1 and 45.8% and 62.2% in patient 2 (Table 2). Orlistat therapy also reduced mean fasting serum cholesterol by 35% and 34% in patient 1 and 37% and 45% in patient 2. Interestingly, there was slight reduction in high-density lipoprotein cholesterol levels noted with orlistat therapy in both patients. Three-day food record analyses during the study showed that dietary fat intake was 10% to 17% of the total daily energy in patient 1 and 12% to 17% in patient 2 (Table 2). There was no deficiency of fat-soluble vitamin levels during orlistat therapy periods, although some fat-soluble vitamin levels could not be measured due to lipemia, mostly during some off-orlistat periods (Table 2).

Table 2.

Mean Fasting Serum Lipids, Lipoproteins, Fat-Soluble Vitamin Levels, and Dietary Fat Intake During the Study

Variables	Patient 1					Patient 2
	Baseline	Month 3: Orlistat	Month 6: Off	Month 9: Orlistat	Month 12: Off	Baseline	Month 3: Off	Month 6: Orlistat	Month 9: Off	Month 12: Orlistat
Fasting TGsa, mg/dL	1611	705	1511	758	1614	1737	1857	1007	3121	1181
Total cholesterola, mg/dL	254	176	272	159	243	242	257	161	340	185
HDL cholesterola, mg/dL	22	14	22	19	22	14	15	10	16	12
Vitamin A, µg/dL	51	43	75	41	54	50	NM	49	NM	45
α-Tocopherol, mg/L	39	27	58	28	33	46	NM	27	NM	28
25-Hydroxyvitamin D, ng/dL	12	24	16	26	22	19	NM	27	NM	40
Vitamin K, pg/mL	NM	>5000	>5000	>5000	NM	4296	NM	1934	NM	NM
Total energy, calories	1856	1787	1835	1445	1911	NA	1855	1640	1569	1776
Fat intake, % of total calories	22	12	15	10	17	NA	14	17	12	17

Weight-based orlistat dose for patient 1 was 60 mg with the first and second meal and 120 mg with the third meal, whereas patient 2 was given 60 mg three times daily with meals.

Abbreviations: HDL, high-density lipoprotein; NA, not available; NM, not measurable due to lipemia.

^aMeans of three consecutive daily fasting values are shown. Normal range is 33–129 mg/dL for serum TGs; 125–170 mg/dL for total cholesterol; 38–76 mg/dL for HDL cholesterol; 24–49 µg/dL for vitamin A; 6.2–14.3 mg/L for α-tocopherol; 20–100 ng/dL for 25-hydroxyvitamin D; and 80–1160 pg/mL for vitamin K.

Both patients completed the study without any serious adverse effects. There were no episodes of acute pancreatitis during the study. Both patients continued to gain height and weight during the trial (Fig. 1). Serum chemistries, complete blood counts, liver function tests, kidney function tests, plasma glucose, and thyroid function tests stayed stable throughout the study (Supplemental Tables 2 and 3). Patient 1 had a mild increase in passage of gas and bloating, and patient 2 had constipation with mild stool leakage on rare occasions during orlistat therapy.

Figure 1.

Height and weight growth in patients during the study. The shaded areas indicate the treatment periods, the solid symbols represent the visit after being on orlistat, and the unfilled symbols represent the baseline and the visit after being off orlistat for 3 months, respectively. Both patients continued to grow well in height and weight during the study.

Discussion

To our knowledge, our study is the first randomized, controlled trial of orlistat in patients with T1HLP reporting >50% reduction in fasting serum TGs over a period of 3 months. Compared with mean serum TG values of 1511 to 3121 mg/dL during the off periods, mean serum TG values during the orlistat therapy periods were 705 to 1181 mg/dL, markedly reducing the risk of acute pancreatitis in both patients with T1HLP. A previous open-label study reported serum TG lowering with orlistat in two siblings (10 and 12 years old) with LPL deficiency (12); however, the precise efficacy could not be determined because the baseline data in one patient were obtained during acute pancreatitis, and in the other, there was a large discrepancy between serum TGs on orlistat at home (504 mg/dL) and during the study visit (2180 mg/dL). In a 34-year-old man with presumed familial chylomicronemia, orlistat therapy reduced fasting serum TGs by 33%, but the patient was consuming 20 g of medium-chain TGs in addition to 50 g of other dietary fats (13).

Previous studies of orlistat use in adolescents and children have reported a relatively high incidence of minor gastrointestinal side effects, mostly related to fat malabsorption (14, 15). It is likely that these gastrointestinal side effects are minimal when combined with an extremely low-fat diet in patients with T1HLP. Both our patients tolerated orlistat well with minimal side effects.

In the last 5 years, several other investigational approaches have been reported to lower serum TGs in patients with T1HLP, including gene therapy, and inhibition of microsomal TG transfer protein and apolipoprotein C3. Alipogene tiparvovec was the first gene therapy approved in Europe for LPL deficiency in 2012. It is an adeno-associated virus gene therapy that causes expression of the naturally occurring p.S447* variant of the human LPL gene, which has higher-than-normal LPL enzymatic activity. It has to be administered by multiple intramuscular injections (>40) in the legs given at a single visit, under spinal anesthesia. Although alipogene tiparvovec initially lowered serum TGs, the effect was transient as patients developed antibodies to the capsid proteins (16, 17). Alipogene tiparvovec was subsequently withdrawn from the market in April 2017 (18).

Lomitapide inhibits microsomal TG transfer protein, which is required for the production of chylomicrons in enterocytes and for the production of very-low-density lipoproteins in the hepatocytes. It was reported to help prevent episodes of acute pancreatitis in a 44-year-old woman with LPL deficiency but resulted in cirrhosis after 12 to 13 years (19).

Volanesorsen, an antisense inhibitor of apolipoprotein C3 (an inhibitor of LPL) synthesis, lowered serum TGs by 56% to 86% in three adults with LPL deficiency in an open-label study (20). Preliminary results for the phase 3 randomized, double-blind, placebo-controlled study showed that Volanesorsen lowered serum TGs by 77% and 50% in 33 patients with T1HLP at 13 weeks and 52 weeks, respectively (21). However, beside injection site reactions in 11.8% of the patients, Volanesorsen resulted in severe thrombocytopenia and early termination of the study in five patients, two of whom had blood platelet counts <25,000/µL (21). Volanesorsen is not yet approved by the US Food and Drug Administration. Furthermore, lomitapide and Volanesorsen have not been studied in children to date.

We conclude that orlistat is safe and highly efficacious in lowering serum TGs in children with T1HLP and should be the first-line therapy in conjunction with an extremely low-fat diet and fat-soluble vitamin supplementation. Longer-term clinical trials with a larger number of patients, however, are warranted to determine the efficacy and safety of orlistat with prolonged use in patients with T1HLP for preventing acute pancreatitis.

Glossary

Abbreviations:

LPL

lipoprotein lipase

T1HLP

type 1 hyperlipoproteinemia

TG

triglyceride