Editor's Mailbox

VOL 112 / NO 3 / SEPTEMBER 2002 / POSTGRADUATE MEDICINE

Type 2 diabetes therapy and liver damage: the debate continues

The article "Type 2 Diabetes Therapy" by Adam B. Mayerson, MD, and Silvio E. Inzucchi, MD (March, page 83), states that "the currently available thiazolidinediones, rosiglitazone maleate (Avandia) and pioglitazone hydrochloride (Actos), are not known to contribute to idiosyncratic hepatocellular injury." Rosiglitazone and pioglitazone are the successors to troglitazone (Rezulin), which was the prototypical agent in the thiazolidinedione class of drugs approved by the US Food and Drug Administration (FDA) in 1997. Rosiglitazone was not associated with hepatotoxicity in premarketing trials (1). However, hepatotoxicity has been reported in at least two diabetic patients treated with rosiglitazone. Both patients were taking other medications or had comorbidities that may have contributed to the hepatic injury (2,3), but rosiglitazone appeared to be the most likely culprit.

During premarketing drug trials, 0.26% of patients taking pioglitazone (and a similar percentage of those taking placebo) had an asymptomatic, reversible increase in alanine aminotransferase levels to more than three times the upper limit of normal. Hepatotoxicity was not observed or reported with pioglitazone in the early postmarketing period (4). Three cases of pioglitazone-associated hepatotoxicity were reported in the medical literature in recent months. The cases represent a spectrum of hepatocellular injury, including hepatitis after 7 months of therapy (5), fulminant hepatic failure (6), and mixed reversible hepatocellular-cholestatic liver injury after 6 months of therapy (7).

It is not known whether these case reports represent a causal relationship with, an association with, or a direct link to therapy. Recommended periodic monitoring of liver enzyme levels, especially in the first year of therapy, may identify potential idiosyncratic reactions.

Until more long-term data on safety are available, current clinical evidence suggests that rosiglitazone and pioglitazone do not share the hepatotoxic profile of troglitazone. The differences may be due to troglitazone's structure and its unique tocopherol side chain (8). The frequency of hepatic injury with the two currently available thiazolidinediones and its risk factors remain unknown and need to be further elucidated by adequate postmarketing trials.

Farhad Zangeneh, MD
Ananda Basu, MD
Rochester, Minnesota

References

1. Salzman A, Patel J. Rosiglitazone is not associated with hepatotoxicity. (Abstr) Diabetes 1999;48:63A
2. Forman LM, Simmons DA, Diamond RH. Hepatic failure in a patient taking rosiglitazone. Ann Intern Med 2000;132(2):118-21
3. Al-Salman J, Arjomand H, Kemp DG, et al. Hepatocellular injury in a patient receiving rosiglitazone: a case report. Ann Intern Med 2000;132(2):121-4
4. Pioglitazone product information. Physicians' desk reference. 54th ed. Montvale, NJ: Medical Economics, 2000:3088-92
5. Maeda K. Hepatocellular injury in a patient receiving pioglitazone. (Letter) Ann Intern Med 2001;135(4):306
6. Chase MP, Yarze JC. Pioglitazone-associated fulminant hepatic failure. Am J Gastroenterol 2002;97(2):502-3
7. May LD, Lefkowitch JH, Kram MT, et al. Mixed hepatocellular-cholestatic liver injury after pioglitazone therapy. Ann Intern Med 2002;136(6):449-52
8. Scheen AJ. Thiazolidinediones and liver toxicity. Diabetes Metab 2001;27(3):305-13

Dr Mayerson and Dr Inzucchi's response:

Drs Zangeneh and Basu address continued concerns about the potential for hepatocellular injury with the currently available thiazolidinediones. Several case reports have described patients treated with either rosiglitazone (1-6) or pioglitazone (7-9) who had moderate to severe elevations of liver enzyme levels. In each case, however, a cause-and-effect relationship was only conjectured by the investigators. In most of these cases, questions could be raised about such a relationship. For example, in some situations the patient may have recently taken other potentially hepatotoxic agents or may have had other medical conditions to which the liver injury could have been justifiably ascribed (1,2). In other cases, the pattern of liver enzyme abnormalities (3,5) or the tempo (2,7) was inconsistent with previous well-described reports of troglitazone-induced hepatic injury.

As pointed out by Drs Zangeneh and Basu, premarketing trials of rosiglitazone and pioglitazone, involving several thousand subjects, demonstrated a minuscule frequency of significant liver enzyme abnormalities, equivalent to that seen in placebo-treated subjects (10,11). This was in marked contrast to the threefold increased risk of such abnormalities noted in the premarketing trials of troglitazone (12).

The hepatic safety of both rosiglitazone and pioglitazone has been confirmed during postmarketing experience by several investigators (13-17). Tens of millions of prescriptions for these agents have been dispensed, and the incidence of liver enzyme abnormalities is exceedingly low. Indeed, the rates are likely the same as, or perhaps even lower than, what would have been expected in this group of patients, who often have steatohepatitis.

Before the availability of thiazolidinediones, Jick and colleagues (18) found the incidence rate of abnormal liver function in a group of 44,406 type 2 diabetic patients in the United Kingdom to be 53.2 per 10,000 person-years. Of the 605 patients with liver disease in this survey, 57 cases had no clear underlying cause, yielding an incidence of 5.0 per 10,000 person-years. While most of these liver injuries were mild and resolved without complications, three cases of liver failure occurred. If these data were extrapolated to the several million patients who have been exposed to rosiglitazone or pioglitazone in the United States, several hundred cases of severe liver injury already would have occurred. The relative paucity of hepatic problems with the current thiazolidinediones might reflect the tendency for these drugs to decrease liver fat content in patients with type 2 diabetes (19).

Finally, in virtually all of the aforementioned case reports (1,2,4-9), the diagnosis of liver injury was made after the patient presented for evaluation of a symptom of liver disease (eg, jaundice, nausea, malaise), not because of routine monitoring of liver enzyme levels. Thus, one might question the utility of the current monitoring program, which adds considerably to the cost and complexity of prescribing this already expensive class of medication.

Admittedly, it is possible that the mild liver enzyme abnormalities in thiazolidinedione-treated patients (which were not reported to the FDA nor communicated to any medical journal) may have been detected by the monitoring program before more severe abnormalities occurred. However, given the exceedingly low frequency of severe liver enzyme abnormalities thus far reported with either rosiglitazone or pioglitazone, it may be time for the FDA to at least revisit the appropriateness and cost-effectiveness of the current program for monitoring thiazolidinedione therapy.

References

1. Forman LM, Simmons DA, Diamond RH. Hepatic failure in a patient taking rosiglitazone. Ann Intern Med 2000;132(2):118-21
2. Al-Salman J, Arjomand H, Kemp DG, et al. Hepatocellular injury in a patient receiving rosiglitazone: a case report. Ann Intern Med 2000;132(2):121-4
3. Hachey DM, O'Neil MP, Force RW. Isolated elevation of alkaline phosphatase level associated with rosiglitazone. (Letter) Ann Intern Med 2000;133(4):752
4. Ravinuthala RS, Nori U. Rosiglitazone toxicity. (Letter) Ann Intern Med 2000;133(8):658
5. Dhawan MK, Castillo RA, Agrawal R, et al. Rosiglitazone induced granulomatous hepatitis. (Abstr) Am J Gastroenterol 2001;96(Suppl 1):S191
6. Gouda HE, Khan A, Schwartz J, et al. Liver failure in a patient treated with long-term rosiglitazone therapy. Am J Med 2001;111(7):584-5
7. Maeda K. Hepatocellular injury in a patient receiving pioglitazone. (Letter) Ann Intern Med 2001;135(4):306
8. Chase MP, Yarze JC. Pioglitazone-associated fulminant hepatic failure. Am J Gastroenterol 2002;97(2):502-3
9. May LD, Lefkowitch JH, Kram MT, et al. Mixed hepatocellular-cholestatic liver injury after pioglitazone therapy. Ann Intern Med 2002;136(6):449-52
10. Avandia package insert. Philadelphia: GlaxoSmithKline, 2002
11. Actos package insert. Lincolnshire, Ill: Takeda Pharmaceuticals North America, 2002
12. Rezulin package insert. Secaucus, NJ: Parke-Davis, 1999
13. Lebovitz HE, Salzman A. Rosiglitazone liver safety update. (Abstr) Diabetes 2000;49(Suppl 1):A39
14. Zawadski JK, Green L, Graham DJ. Thiazolidinedione-associated 15-month postmarketing hepatotoxicity. Abstract P2-567. Annual meeting of the Endocrine Society, June 2001, Denver
15. Lebovitz HE, Kreider M, Freed MI. Evaluation of liver function in type 2 diabetic patients during clinical trials: evidence that rosiglitazone does not cause hepatic dysfunction. Diabetes Care 2002;25(5):815-21
16. Alsheikh-Ali AA, Abourjaily HM, Karas RH. Risk of adverse events with concomitant use of atorvastatin or simvastatin and glucose-lowering drugs (thiazolidinediones, metformin, sulfonylurea, insulin, and acarbose). Am J Cardiol 2002;89(11):1308-10
17. Khan M, St Peter J, Yu D, et al. Rosiglitazone in combination with HMG-CoA reductase inhibitors does not increase the incidence of liver enzyme abnormalities in patients with type 2 diabetes. (Abstr) Diabetes 2002;51(Suppl 2):A103
18. Jick SS, Stender M, Myers MW. Frequency of liver disease in type 2 diabetic patients treated with oral antidiabetic agents. Diabetes Care 1999;22(12):2067-71
19. Mayerson AB, Hundal RS, Dufour S, et al. The effects of rosiglitazone on insulin sensitivity, lipolysis, and hepatic and skeletal muscle triglyceride content in patients with type 2 diabetes. Diabetes 2002;51(3):797-802

Please send technical questions related to the Web site to Ann Harste