Safety and Tolerability of Intravenous Fish Oil Lipid Emulsion in Children Undergoing Hematopoietic Cell Transplantation

Parenteral nutrition (PN) is often necessary for children undergoing allogeneic allogeneic bone marrow, cord, or peripheral blood stem cell (from any donor, including haploidentical donor) hematopoietic cell transplantation (HCT) to avoid malnutrition and growth failure due to the gastrointestinal toxicity associated with preparative myeloablative conditioning regimens. Intravenous lipid emulsion is a required component of PN to provide energy and essential fatty acids when oral or enteral intake is nil or minimal. Soybean oil lipid emulsion (SOLE) is the standard lipid emulsion used in PN. Complications such as metabolic instability, cholestasis, hypertriglyceridemia, and anorexia, which are common during HCT, may also be attributable to the SOLE component of PN. Fish oil based lipid emulsion (FOLE) is used in certain populations and is associated with improved outcomes, specifically in infants with intestinal failure and PN associated cholestasis. It has been reported that supplementation with omega-3 fatty acids from fish oil emulsion may preserve lean body mass, improve quality of life, and decrease length of hospital stay. Provision of FOLE to children undergoing HCT is a novel and innovative strategy to potentially mitigate the metabolic and inflammatory side effects of HCT and its treatment.

HYPOTHESIS: The provision of FOLE is safe and well tolerated in children following HCT. Our aims are to determine the safety profile (as measured by the number of grade 2 or higher adverse events) and tolerability of intravenous FOLE (as measured by serum fatty acid profiles) among children undergoing HCT compared to children receiving SOLE.

HYPOTHESIS: The provision of FOLE is associated with reduced incidence of hypertriglyceridemia in children following HCT, compared to children receiving SOLE and our aim is to determine the number (and percentage) of subjects with fasting hypertriglyceridemia during the trial. To explore trends in associations with other relevant outcomes, other exploratory outcomes (inflammatory biomarkers, body composition measurements, and quality of life indicators) will be assessed in both groups of children following HCT.

METHODS: This is a randomized, controlled, unblinded pilot study to test the safety and tolerability of FOLE, compared to standard SOLE, in children hospitalized for HCT with an underlying hematologic malignancy or myelodysplasic syndrome.

Children cared for in the Dana-Farber Cancer Institute/Boston Children's Cancer and Blood Disorders Center and who will be admitted to BCH for allogeneic HCT will be enrolled (10 subjects in each arm). Patients will be included if they meet all of the entry criteria and none of the exclusion criteria.

All enrolled subjects will receive standard transplant and medical care per HCT guidelines and protocols. A baseline study visit will occur between 14 and 7 days prior to transplant day (day 0). Regardless of enteral intake, at day 0, subjects will be randomized to either control or experimental lipid infusion. The experimental group will receive FOLE and the control group will receive SOLE beginning at day 0 of HCT. All patients, in both the control and experimental groups, will receive standard nutritional care as follows:

• The energy goal for each patient will be the measured resting energy expenditure REE, or the estimated energy expenditure determined and calculated by the clinical dietitian, to be provided by oral, enteral, and/or parenteral energy intake. Intravenous lipids will provide approximately 10 kcal/kg/day. Energy intake from other sources (oral intake, enteral feeding, PN) will be titrated to meet the individualized energy goal, accounting for the contribution from intravenous lipids.
• PN (i.e., dextrose/amino acids) will be initiated if oral or enteral energy intake declines to less than estimated or measured (REE) for >3-5 days, but no earlier than "day 0" of the transplant course.
• Amino acids in PN will be provided according to standard, age-adjusted estimations, minus any contributions from oral or enteral protein intake.
• Lipids will be provided to all enrolled subjects at 1 g/kg/day. On the first day of the lipid infusion, lipids will be infused over 18 hours through a central or peripheral venous line per standard procedures.
• A lipid panel will be obtained 4 hours after the 1st lipid infusion is completed and while the subject is nil per os (NPO) for at least 2 hours.
• Should the subject have hypertriglyceridemia >500 mg/dL with associated abdominal pain, an amylase/lipase level will be drawn to screen for pancreatitis.
• If required (or requested) and after tolerating day 1 of the initial 18-hour lipid infusion, the lipid infusion rate may be changed and lipids may be infused over a range of 8 to 20 hours if the rate does not exceed the maximal lipid infusion rate
• The control group will receive standard SOLE, a 20% lipid solution (Intralipid®, Baxter Healthcare, Deerfield, IL, USA).
• The experimental group will receive FOLE (Omegaven®, Fresenius Kabi AG, Bad Homburg v.d.h., Germany), a 10% lipid solution. The experimental product, Omegaven®, will be purchased from the manufacturer, stored in BCH Research Pharmacy, and labeled appropriately. The experimental product will not be altered in any way from its original manufactured state and may be infused in the same manner as standard SOLE.
• After engraftment occurs (3 consecutive days with absolute neutrophil count >500), lipids will be discontinued at day 30, if clinically indicated, or when the patient is ready for hospital discharge, whichever comes first.
• Lipid rates of 0.15 g fat/kg/hour have been associated with fat overload syndrome; all rates are below this threshold.

Standardized demographic and clinical assessments of all study participants will be performed via medical record review concomitant with the hospitalization and selected clinic visits. Data will include medical record number, underlying diagnosis, type of donor, date of transplant, date of birth, gender, conditioning and prophylactic medication regimens. Results of routine laboratory tests (serum chemistries, blood cell counts) will be recorded twice weekly. Essential fatty acid profiles, pediatric lipid panels, and inflammatory markers will be assessed at specified intervals.

Body weight will be measured by an electronic digital scale accurate to 0.1 kg (same scale used at all visits). Standing height will be measured by stadiometer to the nearest 0.1 cm (same stadiometer used at all visits). Oral and enteral nutrient intake during the inpatient admission period of the study will be evaluated through standard nursing documentation and calorie counts. Daily calorie counts are routine practice in the HCT unit. Nutrient intake will be tabulated from 24-hour recall at outpatient visits. The total enteral and oral intake of calories, protein, carbohydrate, and fat will be calculated by nutrient analysis software (ESHA Food Processor, ESHA Research, Inc., Salem, OR). Parenteral intake will be calculated by the BCH pharmacy's specifications for PN, lipid emulsions, and intravenous fluid. The Pediatric Quality of Life Inventory (Peds QL 4.0) Generic Core and SCT Modules are reliable, validated tools for measuring health related quality of life in children and adolescents ages 2 to 18 with items specific to children who have had HCT. Separate parent and age-appropriate child questionnaires that address similar questions regarding quality of life will be administered at baseline, day +30, and day +100. Air-displacement plethysmography (ADP) is a non-invasive, quick and accurate measure of body density. ADP (BOD POD, Cosmed USA, Concord CA) will be used at specified intervals to estimate body composition (percent body fat and percent lean mass) in subjects.

Individual subjects will be withdrawn from the study if they develop:

1. uncontrolled bleeding attributed to the use of Omegaven
2. persistent fasting hypertriglyceridemia (triglyceride level > 500 mg/dl on 3 separate occasions, obtained 4 hours after the lipid infusion has been completed and while the subject is NPO for at least 2 hours)
3. persistent essential fatty acid deficiency (triene:tetraene >0.2 for 2 consecutive measurements)
4. intolerance or allergy to the assigned lipid emulsion
5. parent or guardian request
6. principal investigator believes withdrawal is in the best medical interest of the patient
7. baseline fasting (NPO for 2 hours) triglyceride level > 500 mg/dL on 2 consecutive measurements

Study discontinuation will occur:

1. upon recommendation of the data and safety monitoring board
2. two or more serious adverse events, occurring across different subjects, possibly or probably related to study participation.

Adverse events for each study arm will be tabulated and categorized by grade. Incidence of adverse events will be presented graphically in a dot plot with each adverse event listed on the vertical axis, incidence displayed as a percentage on the horizontal axis, with different markers for each study. The primary endpoint, the number of adverse events that are grade 2 or higher, will be summarized as mean ± SD. Comparison of the primary endpoint across study arms will be assessed by the two one-sided tests procedure.

This consists of testing:

H1o: μFOLE - μSOLE ≤ θ1 vs H1a: μFOLE - μSOLE > θ1 H2o: μFOLE - μSOLE ≥ θ2 vs H2a: μFOLE - μSOLE < θ2. Equivalence will be declared only if both H1o and H2o are rejected, each at α = 0.05. This is equivalent to testing that the 1 - 2α confidence interval for μFOLE - μSOLE is contained within the equivalence interval [θ1, θ2]. If the distribution of adverse events is more skewed than expected, then a lognormal distribution will be assumed.