Flavocoxid is a proprietary blend of purified, plantderived

Transcription

1 Annals of Internal Medicine Original Research Acute Liver Injury due to Flavocoxid (Limbrel), a Medical Food for Osteoarthritis A Case Series Naga Chalasani, MD; Raj Vuppalanchi, MD; Victor Navarro, MD; Robert Fontana, MD; Herbert Bonkovsky, MD; Huiman Barnhart, PhD; David E. Kleiner, MD; and Jay H. Hoofnagle, MD, on behalf of the Drug-Induced Liver Injury Network Background: Flavocoxid is a prescription medical food that is used to treat osteoarthritis. It is a proprietary blend of 2 flavonoids, baicalin and catechins, which are derived from the botanicals Scutellaria baicalensis and Acacia catechu, respectively. Objective: To describe characteristics of patients with acute liver injury suspected of being caused by flavocoxid. Design: Case series. Setting: Drug-Induced Liver Injury Network Prospective Study ongoing at multiple academic medical centers since Patients: Four adults with liver injury. Measurements: Clinical characteristics, liver biochemistry values, and outcomes. Results: Among 877 patients enrolled in the prospective study, 4 had liver injury suspected to have been caused by flavocoxid. All were women; ages ranged from 57 to 68 years. All developed symptoms and signs of liver injury within 1 to 3 months after initiating flavocoxid. Liver injury was characterized by marked elevations in levels of alanine aminotransferase (mean peak, 1268 U/L; range, 741 to 1540 U/L), alkaline phosphatase (mean peak, 510 U/L; range, 286 to 770 U/L), and serum bilirubin (mean peak, mol/l [9.4 mg/dl]; range, 34.2 to 356 mol/l [2.0 to 20.8 mg/dl]). Liver biochemistry values decreased to the normal range within 3 to 12 weeks after flavocoxid was stopped, and all patients recovered without experiencing acute liver failure or chronic liver injury. Causality was adjudicated as highly likely in 3 patients and as possible in 1 patient. Limitation: The frequency and mechanism of liver injury could not be assessed. Conclusion: Flavocoxid can cause clinically significant liver injury, which seems to resolve within weeks after cessation. Primary Funding Source: National Institute of Diabetes and Digestive and Kidney Diseases. Ann Intern Med. 2012;156: For author affiliations, see end of text. Flavocoxid is a proprietary blend of purified, plantderived bioflavonoids, including baicalin and catechins, that is marketed as a medical food used for therapy for arthritis (1). It became available for treatment of chronic osteoarthritis in the United States in It is available only by prescription in tablets of 250 mg and 500 mg under the brand name Limbrel (Primus Pharmaceuticals, Scottsdale, Arizona) (2). A medical food, as defined in section 5(b) of the Orphan Drug Act (21 U.S.C. 360ee [b] [3]), is a food which is formulated to be consumed or administered enterally under the supervision of a physician and which is intended for the specific dietary management of a disease or condition for which distinctive nutritional requirements, based on recognized scientific principles, are established by medical evaluation (3). Medical foods differ from dietary supplements in that they are not designed for consumption by healthy individuals but as treatments for specific medical conditions. According to the product labeling, flavocoxid has few adverse effects, and double-blind, controlled trials have shown that its adverse event profile is similar to that of placebo (2). Although mild elevations in serum liver enzyme levels have been reported (4, 5), no reports of clinically apparent liver injury linked to flavocoxid have been published. Herein, we report 4 cases of liver injury due to flavocoxid that were found in a prospective study of patients with suspected drug-induced liver injury. METHODS The Drug-Induced Liver Injury Network (DILIN) is a consortium of 8 clinical centers and 1 data coordinating center. The network was established in 2004 to conduct a prospective observational study of individuals suspected to have liver injury due to prescription agents or herbal supplements (6). The network enrolls patients with suspected drug-induced liver injury and collects clinical, biochemical, and histologic data along with serum, urine, and DNA for mechanistic studies. We evaluated each participant for competing or coexisting causes of liver injury and adjudicated the liver injury event by using a predefined severity scale (6). We performed a structured causality assessment that was based on See also: Print Editors Notes Editorial comment Summary for Patients....I-38 Web-Only Appendix Appendix Tables Conversion of graphics into slides 2012 American College of Physicians 857

2 Original Research Acute Liver Injury due to Flavocoxid Context Flavocoxid is a proprietary prescription medical food used to treat osteoarthritis. Contribution This case series describes 4 adults who developed symptoms and signs of liver injury within 3 months after initiating flavocoxid. Manifestations included marked elevations in alanine aminotransferase, alkaline phosphatase, and serum bilirubin levels that resolved within 3 to 12 weeks after flavocoxid was stopped. Caution The frequency of liver injury among persons receiving flavocoxid is not known. Implication Liver injury is a proven adverse effect of flavocoxid. The Editors expert consensus and the Roussel Uclaf Causality Assessment Method (RUCAM) (7). Causality scores ranged from 1 (definite) to 5 (unlikely). The institutional review boards at all participating centers approved the observational study. All participants in the study provided written informed consent. Role of the Funding Source Investigators from the National Institute of Diabetes and Digestive and Kidney Diseases helped design the DILIN Prospective Study and helped develop the current report. RESULTS Between 2004 and 2010, a total of 877 patients with suspected drug-induced liver injury were enrolled in the DILIN Prospective Study. Four patients had liver injury that was attributed, at least in part, to use of flavocoxid. (Detailed case histories of the 4 patients are provided in the Appendix, available at All 4 patients were women, aged 57 to 68 years, who were treated with flavocoxid in doses of 250 to 500 mg twice daily for arthritis or musculoskeletal pain symptoms (Table). Within 1 to 3 months after starting flavocoxid, patients developed signs and symptoms of liver injury that included jaundice (n 3), pruritus (n 3), abdominal pain (n 3), fever (n 2), and rash (n 1). Initial total serum bilirubin levels ranging from 32.5 to mol/l (1.9 to 11.9 mg/dl) were accompanied by marked elevations in alanine aminotransferase (ALT) levels (range, 741 to 1375 U/L) and mild to moderate increases in alkaline phosphatase (ALP) levels (range, 286 to 530 U/L). The peak serum bilirubin level, which ranged from 34.3 to 356 mol/l (2.0 to 20.8 mg/dl), was reached 2 to 13 days after initial presentation. Liver biopsies in 2 patients showed moderate acute hepatitis. Results of tests for acute viral hepatitis (including IgM antibody to hepatitis E virus) (8) and chronic viral hepatitis and imaging studies for biliary obstruction were negative. No patients were hospitalized, although 2 underwent liver biopsy. No patients developed prolongation of the prothrombin time or signs or symptoms of hepatic failure. Each patient s liver injury began to resolve within a few days after stopping flavocoxid, and serum enzyme levels decreased to normal within 1 to 3 months of onset. No patients had evidence of chronic or ongoing liver injury or were rechallenged with flavocoxid. Causality was assessed as very likely due to flavocoxid in 3 patients and as possibly due to flavocoxid in 1 patient. The possible case was complicated by exposure to several other agents that might have caused the injury, including pregabalin, duloxetine, and tizanidine. Three cases were ranked as moderate severity (score, 2 [jaundiced but not hospitalized]) and 1 as mild severity (score, 1 [serum enzyme elevations without jaundice]). Table. Demographic and Clinical Features, Laboratory Test Results, Time to Improvement, Causality Assessment, and Severity Scores for 4 Patients With Flavocoxid Hepatotoxicity Enrolled in the DILIN Prospective Study Patient Sex Age, y Latency, d ALT ALP Ratio (R Value)* Peak ALT Level, U/L Peak ALP Level, U/L Peak Bilirubin Level, mol/l (mg/dl) Time to Improve >50% From Peak, d ALT Total Bilirubin ALP RUCAM Score DILIN Causality Score 1 Female (7.3) Female (2.0) Female (20.8) NA NA NA Female (7.5) DILIN Severity Score ALP alkaline phosphatase; ALT alanine aminotransferase; DILIN Drug-Induced Liver Injury Network; NA not available; RUCAM Roussel Uclaf Causality Assessment Method. * Levels of ALT and ALP are both expressed as multiples of the upper limit of normal. A ratio below 2 indicates cholestatic liver injury, a ratio above 5 indicates hepatocellular liver injury, and a ratio between 2 and 5 indicates a mixed pattern. For patient 3, time to improve 50% from peak could not be calculated because the results of laboratory tests between day 9 after enrollment and her 6-mo visit were not available. At day 9, her ALT level was 789 U/L, ALP level was 455 U/L, and bilirubin level was mol/l (19.3 mg/dl). At her 6-mo visit, her liver biochemistry values normalized (ALT level, 12 U/L; ALP level, 57 U/L; bilirubin level, 8.6 mol/l [0.5 mg/dl]). See the Appendix (available at for the patient s case history June 2012 Annals of Internal Medicine Volume 156 Number 12

3 Acute Liver Injury due to Flavocoxid Original Research DISCUSSION In a large study of patients with suspected druginduced liver injury, we found 4 patients with reversible liver injury highly likely or possibly caused by flavocoxid. In an English-language MEDLINE search of the literature done in February 2012, we found no specific reports of liver injury due to flavocoxid, although clinical trials of the agent had reported instances of hepatotoxicity (4, 5, 9). One 12-week randomized trial that compared flavocoxid (500 mg twice daily) with naproxen (500 mg twice daily) in 220 patients with osteoarthritis reported 16 patients with mild elevations in serum aminotransferase levels (flavocoxid, 11; naproxen, 5; P 0.08) and 12 with elevated bilirubin levels (flavocoxid, 5; naproxen, 7; P 0.18) (5). Another open-label study of flavocoxid (500 mg twice daily) for 60 days in 1067 individuals with osteoarthritis reported liver test abnormalities in only 1 patient (0.1%) but did not provide details of the abnormalities (9). Postmarketing surveillance of flavocoxid done by Primus Pharmaceuticals showed 31 hepatic adverse events among users, which suggests a possible incidence of about 0.011% of flavocoxid-induced liver injury (1). The reported adverse events included 22 instances of elevated liver enzyme levels, 6 instances of jaundice, and 3 cases of hepatitis. Clinical and biochemical features of 8 of the cases of clinically apparent liver disease provided by the manufacturer (Levy RM. Personal communication.) are shown in Appendix Table 1 (available at The pattern of liver injury was hepatocellular in 2 cases, cholestatic in 1 case, mixed hepatocellular-cholestatic in 3 cases, and unknown in 2 cases. Two patients had eosinophilia and fever, suggesting drug hypersensitivity. No patients developed signs or symptoms of acute liver failure, and injury improved in all cases after flavocoxid was stopped. One patient restarted flavocoxid on her own and rapidly redeveloped liver injury. Our report provides convincing evidence that flavocoxid is capable of causing clinically apparent, acute liver injury. In all 4 cases, we excluded other common diagnoses and the injury resolved promptly after flavocoxid was stopped. The clinical signature of flavocoxid-induced liver injury seems to be a mild to moderate, mixed hepatocellular-cholestatic hepatitis that arises 2 to 12 weeks after starting the medication and resolves rapidly once it is stopped. No reported cases were associated with signs of liver failure or death, and we found no evidence of residual or chronic injury in any patients. A proportion of the clinically apparent cases had mild evidence of hypersensitivity early in the course of illness as marked by low-grade fever, rash, and eosinophilia, but these features were not prominent and not present in all cases. A limitation of our report is that the frequency, risk factors, and mechanisms of clinically significant liver injury due to flavocoxid could not be assessed. We speculate that hypersensitivity may play a role because immunoallergic features occurred in some patients, and flavocoxid is also associated with hypersensitivity pneumonitis (1, 10). The primary ingredients of flavocoxid are baicalin and catechins, which are concentrated and standardized to greater than 90% purity. Baicalin is a free B-ring flavonoid derived from the root of Scutellaria baicalensis (Chinese skullcap). Experimental data suggest that it has hepatoprotective activity against various hepatotoxic insults (11, 12). Catechin is a flavan flavonoid. It is the stereoisomer epicatechin and is derived from the bark of Acacia catechu (Mimosa catechu). The hepatotoxic potential in humans of green tea extracts that contain catechins is well-recognized (13, 14). Two catechins contained in green tea are epigallocatechin gallate and epicatechin gallate. The former is generally believed to be responsible for the hepatotoxicity caused by green tea extracts (15); however, epicatechin is the flavonoid that is common to both green tea and flavocoxid, raising suspicion that it may be the ingredient that causes liver injury. Flavocoxid differs from dietary supplements in that it was approved and is marketed as a medical food. However, like dietary supplements, medical foods do not require formal premarketing safety and efficacy studies (16). Given its botanical ingredients, variation in the concentration of substances due to various factors, such as harvesting conditions, seasonal variations, and geographic origin, is possible. Unpredictable or unregulated concentration of such polyphenolic substances as catechins may set the stage for toxicity. In summary, flavocoxid, a medical food used to treat osteoarthritis, is capable of causing acute liver injury that typically manifests as a mixed hepatocellular-cholestatic hepatitis arising after 1 to 3 months of use and is rapidly reversible after stopping the medication. From Indiana University School of Medicine, Indianapolis, Indiana; Temple University School of Medicine, Philadelphia, Pennsylvania; University of Michigan, Ann Arbor, Michigan; Carolinas Medical Center, Charlotte, North Carolina; Duke Clinical Research Institute, Durham, North Carolina; and the National Cancer Institute and National Institute of Diabetes and Digestive and Kidney Diseases, Bethesda, Maryland. Acknowledgment: The authors thank Primus Pharmaceuticals and Dr. Robert Levy, Chief Medical Officer, for providing the case histories reported to them of 8 patients with suspected flavocoxid-induced liver injury. Financial Support: The DILIN is supported by the National Institute of Diabetes and Digestive and Kidney Diseases ( This research was supported in part by the Intramural Research Program of the National Cancer Institute, National Institutes of Health. Potential Conflicts of Interest: Disclosures can be viewed at M Reproducible Research Statement: The authors are willing to provide the narratives and laboratory data of the cases reported in this paper to the interested public June 2012 Annals of Internal Medicine Volume 156 Number

4 Original Research Acute Liver Injury due to Flavocoxid Requests for Single Reprints: Naga Chalasani, MD, Division of Gastroenterology and Hepatology, Indiana University School of Medicine, 1050 Wishard Boulevard, RG 4100, Indianapolis, IN 46202; , Current author addresses and author contributions are available at References 1. Primus Pharmaceuticals: Limbrel Web site. Accessed at on 3 March Primus Pharmaceuticals. Limbrel Product Information. Scottsdale, AZ: Primus Pharmaceuticals; Accessed at on 3 March U.S. Food and Drug Administration. Medical Foods. Silver Spring, MD: U.S. Department of Health and Human Services; Accessed at /food/foodsafety/product-specificinformation/medicalfoods on 3 March Morgan SL, Baggott JE, Moreland L, Desmond R, Kendrach AC. The safety of flavocoxid, a medical food, in the dietary management of knee osteoarthritis. J Med Food. 2009;12: [PMID: ] 5. Levy RM, Khokhlov A, Kopenkin S, Bart B, Ermolova T, Kantemirova R, et al. Efficacy and safety of flavocoxid, a novel therapeutic, compared with naproxen: a randomized multicenter controlled trial in subjects with osteoarthritis of the knee. Adv Ther. 2010;27: [PMID: ] 6. Fontana RJ, Watkins PB, Bonkovsky HL, Chalasani N, Davern T, Serrano J, et al; DILIN Study Group. Drug-Induced Liver Injury Network (DILIN) prospective study: rationale, design and conduct. Drug Saf. 2009;32: [PMID: ] 7. Rockey DC, Seeff LB, Rochon J, Freston J, Chalasani N, Bonacini M, et al; US Drug-Induced Liver Injury Network. Causality assessment in drug-induced liver injury using a structured expert opinion process: comparison to the Roussel- Uclaf causality assessment method. Hepatology. 2010;51: [PMID: ] 8. Davern TJ, Chalasani N, Fontana RJ, Hayashi PH, Protiva P, Kleiner DE, et al; Drug-Induced Liver Injury Network (DILIN). Acute hepatitis E infection accounts for some cases of suspected drug-induced liver injury. Gastroenterology. 2011;141: e1-9. [PMID: ] 9. Pillai L, Burnett BP, Levy RM; GOAL Study Cooperative Group. GOAL: multicenter, open-label, post-marketing study of flavocoxid, a novel dual pathway inhibitor anti-inflammatory agent of botanical origin. Curr Med Res Opin. 2010; 26: [PMID: ] 10. Levy RM. Flavocoxid and hypersensitivity pneumonitis [Letter]. Chest. 2011;140:827-8; author reply 828. [PMID: ] 11. Qiao H, Han H, Hong D, Ren Z, Chen Y, Zhou C. Protective effects of baicalin on carbon tetrachloride induced liver injury by activating PPAR and inhibiting TGF 1. Pharm Biol. 2011;49: [PMID: ] 12. Guo HX, Liu DH, Ma Y, Liu JF, Wang Y, Du ZY, et al. Long-term baicalin administration ameliorates metabolic disorders and hepatic steatosis in rats given a high-fat diet. Acta Pharmacol Sin. 2009;30: [PMID: ] 13. Bonkovsky HL. Hepatotoxicity associated with supplements containing Chinese green tea (Camellia sinensis) [Letter]. Ann Intern Med. 2006;144: [PMID: ] 14. Mazzanti G, Menniti-Ippolito F, Moro PA, Cassetti F, Raschetti R, Santuccio C, et al. Hepatotoxicity from green tea: a review of the literature and two unpublished cases. Eur J Clin Pharmacol. 2009;65: [PMID: ] 15. Schmidt M, Schmitz HJ, Baumgart A, Guédon D, Netsch MI, Kreuter MH, et al. Toxicity of green tea extracts and their constituents in rat hepatocytes in primary culture. Food Chem Toxicol. 2005;43: [PMID: ] 16. U.S. Food and Drug Administration. Guidance for Industry: Frequently Asked Questions About Medical Foods. Silver Spring, MD: U.S. Department of Health and Human Services; Accessed at ComplianceRegulatoryInformation/GuidanceDocuments/MedicalFoods/ucm htm on 2 January ACP CHAPTER MEETINGS For information on upcoming ACP chapter meetings, including scientific programs and registration forms, please visit /meetings/chapter June 2012 Annals of Internal Medicine Volume 156 Number 12

5 Annals of Internal Medicine Current Author Addresses: Drs. Chalasani and Vuppalanchi: Division of Gastroenterology and Hepatology, Indiana University School of Medicine, 1050 Wishard Boulevard, RG 4100, Indianapolis, IN Dr. Navarro: Thomas Jefferson University, Main Building, Suite 480, 132 South 10th Street, Philadelphia, PA Dr. Fontana: University of Michigan, Alfred Taubman Health Care Center, 1500 East Medical Center Drive, Floor 3, Reception D, Room 3326, Ann Arbor, MI Dr. Bonkovsky: Carolinas HealthCare System, Suite 201, Cannon Research Center, 1542 Garden Terrace, Charlotte, NC Dr. Barnhart: Duke University, Duke Box 3850, Durham, NC Dr. Kleiner: National Cancer Institute, Building 10, Magnuson CC, Room 2B44, 10 Center Drive, Bethesda, MD Dr. Hoofnagle: National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health, Building 31, Room 9A27, Bethesda, MD Author Contributions: Conception and design: N. Chalasani, R. Vuppalanchi, H. Bonkovsky, J.H. Hoofnagle. Analysis and interpretation of the data: N. Chalasani, V. Navarro, R. Fontana, H. Bonkovsky, H. Barnhart, D.E. Kleiner, J.H. Hoofnagle. Drafting of the article: N. Chalasani, R. Fontana, H. Bonkovsky, H. Barnhart. Critical revision of the article for important intellectual content: N. Chalasani, V. Navarro, H. Bonkovsky, D.E. Kleiner, J.H. Hoofnagle. Final approval of the article: N. Chalasani, V. Navarro, R. Fontana, H. Bonkovsky, H. Barnhart, D.E. Kleiner, J.H. Hoofnagle. Provision of study materials or patients: N. Chalasani, R. Vuppalanchi, V. Navarro, R. Fontana, H. Bonkovsky. Statistical expertise: H. Barnhart. Obtaining of funding: N. Chalasani, H. Bonkovsky, H. Barnhart. Administrative, technical, or logistic support: J.H. Hoofnagle. Collection and assembly of data: N. Chalasani, R. Vuppalanchi, R. Fontana, D.E. Kleiner, J.H. Hoofnagle. APPENDIX: CASE HISTORIES OF 4 PATIENTS ENROLLED IN THE DILIN PROSPECTIVE STUDY Patient 1 A 58-year-old, non-hispanic white woman presented to her physician with worsening fatigue, low-grade fever, pruritus, and dark urine approximately 10 weeks after starting flavocoxid at a dose of 500 mg twice daily for long-standing spinal stenosis and osteoarthritis. Her medical history included dyslipidemia, coronary artery disease, and mild renal insufficiency. She reported no medicinal allergies or history of adverse drug reactions. Her concomitant prescription and over-the-counter (OTC) medicines included aspirin, tramadol, omeprazole, ezetimibe, 1% oxiconazole cream, flaxseed oil, zolpidem, glucosamine, multivitamins, and calcium plus vitamin D. The findings from her physical examination were unremarkable. Laboratory tests at presentation included a normal complete blood count without eosinophilia. Her serum ALT level was 1105 U/L, aspartate aminotransferase (AST) level was 1198 U/L, ALP level was 487 U/L, and total bilirubin level was 42.7 mol/l (2.5 mg/dl). Results of these tests were normal on routine testing 2 months before and 1 month after flavocoxid was started (Appendix Table 2). Results of tests for acute and chronic viral hepatitis, autoantibodies, and serum immunoglobulins were unrevealing. Abdominal imaging showed no abnormalities. A liver biopsy was interpreted as compatible with drug-induced liver injury, with focal necrosis and inflammation, portal lymphocytic inflammatory infiltrates with some eosinophils, and plasma cells. She was instructed to discontinue flavocoxid. Her serum enzyme levels increased over the next 10 days, and her serum bilirubin level increased to mol/l (7.3 mg/dl). Thereafter, her condition improved and her liver abnormalities resolved, decreasing to the normal range by 12 weeks after presentation. She was not rechallenged with flavocoxid but resumed all other medicines without any problems. In the formal assessment of causality, the DILIN investigators judged the liver injury as very likely due to flavocoxid (score, 2). The RUCAM score was 8 (probable), and the severity was scored as moderate (score, 2 [jaundice without hospitalization or signs of hepatic failure]). Patient 2 A 57-year-old, non-hispanic white woman presented to her physician with several days history of severe generalized pruritus, rash, and dark urine approximately 9 weeks after starting flavocoxid in a dose of 250 mg twice daily for osteoarthritis. She was known to have normal liver test results on routine testing 3 months before starting treatment. Her medical history was also significant for gastroesophageal reflux disease. Her concomitant prescription and OTC medicines included long-term use of pantoprazole (40 mg, as needed), FemHRT (Warner Chilcott, Dublin, Ireland) (1 mg norethindrone acetate and 5 mcg ethinyl estradiol), fluticasone nasal spray, aspirin, calcium with vitamin D, glucosamine, and a multivitamin. She was a nonsmoker and consumed 1 to 2 glasses of wine 3 times per week. She reported no medicinal allergies. On presentation, the findings from her physical examination were unremarkable except for a faint maculopapular rash. Laboratory tests revealed an ALT level of 741 U/L, AST level of 233 U/L, ALP level of 286 U/L, and total bilirubin level of 32.4 mol/l (1.9 mg/dl). A complete blood count was normal without eosinophilia. Results of tests for viral hepatitis were negative, and abdominal imaging showed no competing cause. Antimitochondrial antibodies were undetectable and serum globulin levels were normal, but other autoantibodies were not measured. Her primary care physician instructed her to stop flavocoxid and prescribed a short course of oral prednisone (10 mg/d, tapered over 9 days). She continued her other long-term prescription and OTC medicines. After stopping flavocoxid, her symptoms improved rapidly and her liver abnormalities resolved over the ensuing 4 weeks (Appendix Table 3). In the formal causality assessment, the DILIN investigators judged the liver injury as very likely due to flavocoxid (score, 2). The RUCAM score was 8 (probable), and the severity was scored as moderate (score, 2 [jaundice without hospitalization or signs of hepatic failure]). Patient 3 A 60-year-old, non-hispanic white woman presented to her physician with several days history of dark urine, itching, and jaundice approximately 12 weeks after starting flavocoxid (500 mg twice daily) and 7 weeks after starting pregabalin (75 mg 3 times daily) for fibromyalgia. Her medical history was also significant for hypertension, depression, gastroesophageal reflux disease, deep venous thrombosis, and benign kidney tumor. Her concomitant prescription medicines included zolpidem, duloxetine, fentanyl, oxycodone with acetaminophen, lansoprazole, ondanse June 2012 Annals of Internal Medicine Volume 156 Number 12 W-297

6 Appendix Table 1. Summary of Cases of Potential Liver Injury due to Limbrel Reported to Primus Pharmaceuticals DILIN Causality Score Time to Resolution, wk Biliary Tract Work-up Result Hepatitis Serology Result Total Bilirubin Level, mol/l (mg/dl) ALP Level, U/L Symptoms ALT Level, U/L Patient Age, y Sex Latency, wk 1 60 Male 4 to 5 Itching, dark urine (3.1) Negative NA 4 to Female 3 Fever (8.7) Negative NA Female 16 to 20 None 330 NA NA NA NA Female 4 Jaundice ( 6) NA Negative Female NA Fever 432 NA 23.9 (1.4) NA NA NA 1 Fever, eosinophilia, positive rechallenge (7.1) NA Negative 4 3 Elevated lipase levels 6 67 Female 10 Nausea, abdominal pain, itching, jaundice 7 NA NA NA NA (3.3) NA NA NA 4 Hospitalized 8 63 Female 4 to 5 Dark urine, fatigue (4.5) Negative Negative 4 to 5 2 First reported instance of jaundice; no rechallenge ALP alkaline phosphatase; ALT alanine aminotransferase; DILIN Drug-Induced Liver Injury Network; NA not available. tron, and tizanidine. She did not report using any OTC medicinal or herbal products. She was a nonsmoker and reported no alcohol consumption. Her drug allergies included penicillin, cephalosporins, sulfa, and sumatriptan. The findings from her physical examination were unremarkable. Her laboratory tests revealed an ALT level of 1178 U/L, AST level of 1108 U/L, ALP level of 426 U/L, and total bilirubin level of mol/l (11.9 mg/dl). A complete blood count was normal without eosinophilia. Results of tests for viral hepatitis and serum autoantibodies and abdominal imaging showed no competing cause. She was instructed to discontinue both flavocoxid and pregabalin while continuing the other medicines. A liver biopsy revealed expanded portal tracts with dense chronic inflammatory cell infiltrate containing numerous plasma cells along with interface hepatitis and periportal collapse. The lobule showed marked unrest with patchy foci of inflammation and collapse. Over the next few weeks, her symptoms resolved, and her liver biochemistry values improved and were normal 6 weeks later (Appendix Table 4). In the formal assessment of causality, the DILIN investigators considered the clinical course as highly likely due to drug-induced liver injury (score, 2) but had difficulty in specifically attributing the liver injury to flavocoxid or pregabalin. The compound-specific causality score was 4 (possible) for flavocoxid and 3 (probable) for pregabalin. The RUCAM score for flavocoxid was 2 (unlikely), and the severity was scored as moderate (score, 2 [jaundiced but not hospitalized]). Patient 4 A 67-year-old, non-hispanic white woman presented to her physician with fatigue and dark urine 4 weeks after starting flavocoxid (500 mg twice daily) for long-standing osteoarthritis. Her other medical problems included severe obesity, atrial fibrillation, hypertension, dyslipidemia, gastroesophageal reflux disease, and the irritable bowel syndrome. Her concomitant prescription and OTC medicines included aspirin, furosemide, hydrochlorothiazide, levothyroxine, metoprolol, omeprazole, potassium, simvastatin, warfarin, calcium, fish oil, and linum seed oil. She reported no alcohol consumption. Her medicinal allergies included erythromycin, morphine, and tetracycline. The findings from her physical examination were unremarkable except for obesity. Laboratory tests at presentation revealed an ALT level of 1375 U/L, AST level of 1050 U/L, ALP level of 770 U/L, and total bilirubin level of 71.8 mol/l (4.2 mg/dl). A complete blood count was normal without eosinophilia. Results of tests for viral hepatitis were negative. Although anti smooth muscle and mitochondrial antibodies were undetectable, antinuclear antibody was positive at 1:1280 dilution, whereas total serum globulin levels were normal. Abdominal imaging and a subsequent endoscopic retrograde cholangiopancreatography showed no evidence of biliary obstruction. A liver biopsy was not done. She was instructed to discontinue flavocoxid. She remained jaundiced for almost a month, but her condition gradually improved, and liver test results had all decreased to the normal range by 12 weeks after stopping flavocoxid (Appendix Table 5). In the formal causality assessment, the DILIN investigators judged the liver injury as very likely due to flavocoxid (score, 2). The RUCAM score was 9 (highly probable), and the severity was scored as moderate (score, 2 [jaundiced but not hospitalized]). W June 2012 Annals of Internal Medicine Volume 156 Number 12

7 Appendix Table 2. Time Course of Liver Biochemistry Results Before and After Starting Flavocoxid in Patient 1 Time From Starting Flavocoxid Time From Onset ALT Level, U/L ALP Level, U/L Bilirubin Level, mol/l (mg/dl) Before starting Before onset (0.6) Routine testing 1 mo Before onset (0.4) 3 mo (2.5) Symptomatic 4 d (3.4) 10 d (7.3) Liver biopsy 16 d (3.9) 26 d (2.1) 4 mo 30 d (1.3) Asymptomatic 5 wk (0.8) 6 wk (0.8) 6 mo 12 wk (0.5) 24 wk (0.2) 32 wk (0.8) Normal values ( 1.2) ALP alkaline phosphatase; ALT alanine aminotransferase. Appendix Table 3. Time Course of Liver Biochemistry Results Before and After Starting Flavocoxid in Patient 2 Time From Starting Flavocoxid Time After Stopping ALT Level, U/L ALP Level, U/L Bilirubin Level, mol/l (mg/dl) Before starting Before stopping (1.0) 9 wk (1.9) Itching and rash 2 d (2.0) Prednisone, 10 mg daily 11 wk 2 wk (0.9) Prednisone stopped 13 wk 4 wk (0.6) Asymptomatic Normal values ( 1.2) ALP alkaline phosphatase; ALT alanine aminotransferase. Appendix Table 4. Time Course of Liver Biochemistry Results Before and After Starting Flavocoxid in Patient 3 Time From Starting Flavocoxid Time After Stopping ALT Level, U/L ALP Level, U/L Bilirubin Level, mol/l (mg/dl) 12 wk (11.9) Nausea, jaundice 3 d (16.5) 13 wk 7 d (20.8) 10 d (19.3) 10 mo 6 mo (0.5) Normal values ( 1.2) ALP alkaline phosphatase; ALT alanine aminotransferase June 2012 Annals of Internal Medicine Volume 156 Number 12 W-299

8 Appendix Table 5. Time Course of Liver Biochemistry Results Before and After Starting Flavocoxid in Patient 4 Time From Starting Flavocoxid Time After Stopping ALT Level, U/L ALP Level, U/L Bilirubin Level, mol/l (mg/dl) Before starting Before stopping 19 4 wk (4.2) Dark urine and fatigue 2 d (4.2) 5 wk 6 d (6.6) Ultrasound 8 d (7.1) ERCP 6 wk 13 d (7.5) 15 d (6.7) 7 wk 20 d (3.7) 2 mo 4 wk (2.4) 4 mo 12 wk (0.6) Asymptomatic 10 mo 9 mo (0.4) Normal values ( 1.2) ALP alkaline phosphatase; ALT alanine aminotransferase; ERCP endoscopic retrograde cholangiopancreatography. W June 2012 Annals of Internal Medicine Volume 156 Number 12