Omegaven for Compassionate Use in the Treatment of Parenteral Nutrition-Associated Liver Disease

1. The overall purpose of the treatment is to offer alternative treatment to children who developed parenteral nutrition-associated liver disease (PNALD) and have not responded positively to currently available medical therapies. PNALD develops in newborns dependent on parenteral nutrition (PN) and are unable to tolerate adequate enteral feedings to support fluid and nutritional fluids; although PN is necessary and life sustaining, it can result in severe liver disease. The investigators want to see if a form of intravenous fat mixture, Omegaven, helps in the treatment of children with PNALD. Experimental data has shown that the intravenous fat mixture (Intralipid) that is currently being used as a part of newborn PN, may be contributing to the liver disease. Omegaven is different fat mixture manufactured and used in Europe in adults. It is different in that it is made from highly refined fish oil, while Intralipid is made from soybeans. Case control studies only, and not randomized clinical trials, show that Omegaven may reverse or treat PNALD. Since Omegaven is not currently approved for use in the United States, treatment IND was obtained from Food and Drug Administration (FDA).
2. Feeding intolerance is a common pathway of a broad array of medical and surgical disease states encountered in the neonatal population. Consequently, a high percentage of patients in neonatal intensive care units (NICU) need to rely on PN if they are to survive and grow. Through decades of research the development of PN has been a great success in that its use has maintained countless patients in sufficient nutritional status to recover from medical or surgical conditions. However, one complication of PN that remains poorly understood and that still causes considerable morbidity and mortality is PNALD. There are a wide variety of case scenarios for PNALD. There are scenarios in which newborns suffer irrecoverable bowel loss from some process (e.g., necrotizing enterocolitis totalis, mid gut volvulus) and consequently require long term PN as they await bowel transplant; as a result they almost always also require liver transplant secondary to PNALD. This further delays their surgery as they wait for two organs instead of one (Chungfat, Dixler et al. 2007). More commonly, cases involve patients who have prolonged bowel dysfunction that requires long term use of PN but who slowly recover and gradually transition to enteral feeds. For these infants the race between bowel recovery and PNALD progression often decides whether the patient lives or dies.
3. Although the etiology of PNALD is poorly understood and likely to be multi-factorial (Steinbach, Clark et al. 2008), recent animal research and human case studies suggest that altering the dose and the composition of the lipid component of PN might alter the course of PNALD (Alwayn, Gura et al. 2005; Gura, Parsons et al. 2005; Gura, Duggan et al. 2006; Gura, Lee et al. 2008). The lipid products used for PN in the United States are derived from soy oil; the brand used at Cardinal Glennon Children's Hospital is Intralipid®. Soy-based PN lipids are rich in omega-6 FA and also contain small amounts of omega-3 FA in the form of linolenic acid precursor. The potential toxicity of omega-6 fatty acids as it relates to PNALD has been recently revealed (Diamond, Sterescu et al. 2008; de Meijer, Gura et al 2009). This FA composition is sufficient to prevent essential FA deficiency (EFAD), historically measured as the ratio of specific omega-6 FA in the blood (Gura, Lee et al 2009).

In a published article (Gura, Duggan et al. 2006) reporting 2 cases in depth and in a subsequent paper (Gura, Lee et al. 2008) reporting 18 patients with established PNALD who required continued PN, investigators replaced the standard soy based lipid component of PN with a fish oil derived parenteral lipid (Omegaven™) that contains a high percentage (30 to 60%) of omega-3 FA. This therapy, which is still experimental in the United States, was associated with a gradual but ultimately profound correction of PNALD as indicated by laboratory markers such as conjugated bilirubin and circulating liver enzymes; the median time to reduce serum direct bilirubin from >2mg/dl to <2mg/dl was 9.4 weeks. No significant adverse effects of Omegaven™ in the doses utilized were identified.

The current study is to identify infants with PNALD of moderate severity or worse, defined as two consecutive measurements of direct bilirubin of 3 mg/dL after being on PN for > 3 weeks and who are anticipated to require PN for another 3 weeks, which puts them at risk of further progression of PNALD. For inclusion, other causes of liver disease must be ruled out. Upon meeting entry criteria, infant families will be offered, through an Institutional Review Board (IRB) approved consent process, the same treatment protocol for their infants that was successfully applied in the above cited clinical cases and detailed below. Monitoring for side effects, complications and disease progression or regression will be instituted. Therapy will be stopped when direct bilirubin is < 1 mg/dL for 2 weekly measurements or when the patient is tolerating enough enteral nutrition so that PN is no longer needed. Clinical and laboratory monitoring will be continued for 2 weeks post Omegaven™ infusion and as indicated by events thereafter. Additional endpoint criteria are: patients receiving transplants; AST/ALT, GGTP, C-reactive protein and albumin, mortality, infection rates and types, and frequency of red blood cell transfusions; if infant worsens with treatment or does not tolerate treatment with Omegaven, specifically for the adverse event of severe uncontrolled bleeding as listed in the risk section.