The real cost of iron chelation therapy

Before the advent of iron chelation, many chronically transfused patients faced a multitude of complications of iron overload including cardiomyopathy, liver cirrhosis, endocrinopathy, and premature death. The first iron chelator, desferrioxamine, was licensed by the FDA in 1968. Desferrioxamine significantly reduced iron overload1 and, if initiated early, prevented a number of these life-threatening complications.2 Even when started in patients with established iron overload, some benefit was observed.2 Survival of patients with thalassemia improved dramatically.3 The costs associated with desferrioxamine treatment were also substantial, however. Because of its short half-life, desferrioxamine must be administered by a continuous infusion for 8 to 12 hours 5 to 7 days a week. Infusions are painful and cumbersome and can significantly limit lifestyle choices. Therefore, compliance with this regimen is often poor. Poor compliance in turn leads to development or exacerbation of iron overload complications and mounting health-care costs. Desferrioxamine itself is expensive and its cost is further compounded by the necessity to purchase infusional equipment. Because desferrioxamine has a number of serious side effects, patients require regular monitoring with audiology, ophthalmoscopy, and skeletal surveys. In addition, iron status needs to be assessed regularly by measuring ferritin levels (a surrogate measure of hepatic iron content) or by directly assessing iron content by liver biopsy, superconducting susceptometry (SQUID), or a new MRI “Ferriscan” technology. All of these monitoring modalities come at a cost and are not readily available in many institutions. For example, SQUID is available in only four places in the world. The bottom line is further increased by travel costs and lost wages of patients or patients' caregivers. A better iron chelator was clearly needed. In 1987 deferiprone, a three-times-daily oral chelator, was first used in humans. It was relatively inexpensive and trials have shown the drug to be of particular benefit in preventing cardiac complications of iron overload.4 Concerns over serious side effects, however, such as severe agranulocytosis, halted its further study in North America. The drug is now used in Europe and Asia but remains unlicensed in North America. It is available only through the FDA Treatment Use Program in the United States and the Special Access Program in Canada. Recently deferasirox, a once-daily oral chelator, was shown to be effective in trial settings5 and received its approval from FDA in 2005 and Health Canada in 2006. This new drug is significantly more expensive than desferrioxamine and its side-effect profile includes cytopenias and renal impairment. Despite the vast amount of literature on iron chelation therapy (ICT), there is clearly a paucity of literature dealing with its economic and psychosocial impact. The two articles appearing in this issue of TRANSFUSION attempt to fill this void. In one article, Payne and colleagues6 examine treatment with desferrioxamine in a “naturalistic setting.” The authors studied drug efficacy, cost, and impact on quality of life. Retrospective chart review revealed that, despite years of therapy, mean serum ferritin levels remained unacceptably high placing the patients at risk for iron overload complications.6 Unfortunately, the authors did not collect data on the more robust measurements of iron content such as liver biopsy or elaborate on the rate of serious complications including cardiomyopathy or premature death. Although minor side effects were frequent, serious side effects were rare. Payne and colleagues6 also systematically examined the costs associated with this type of ICT and concluded that they were “substantial.” To assess patients' perception of health as well as emotional, physical, and mental well-being, Payne and colleagues utilized a standardized questionnaire. The utilized questionnaire is a validated and widely used tool, although it is generic. Arguably, the ideal tool should be specific for the major features of conditions requiring iron chelation and ICT itself. Not surprisingly, chelated patients scored significantly lower than their age matched normal controls in a number of categories. The study by Payne and colleagues was unable to discern the impact of iron chelation versus the impact of living with a chronic disease. The existing literature on impact of ICT on the quality of life is limited but consistent: parenteral ICT has a significant negative impact on the quality of life.7 The article by Delea and associates8 addresses compliance and cost-effectiveness of ICT. Although desferrioxamine boasted more than 90 percent compliance in trial settings, the real-life compliance lagged behind at 59 to 78 percent.8 Their literature review also showed that compliance with deferiprone, an oral formulation, was substantially better than with the parenteral drug. Among predictors of noncompliance was age. Reduced compliance was marginally more frequent in patients aged 10 to 18 years compared to the other age groups. In our practice, we frequently see a drop in compliance as children graduate to an adult hemoglobinopathy program. The reasons behind this phenomenon are probably multiple but invariably include inability to reconcile increased time commitments for education, social life, and work with a rigid schedule of ICT. The cost of poor compliance is dire, because it is associated with significant increase in the risk of cardiac disease, endocrinopathy, and death. One of the reviewed studies showed that patients with a mean of 225 to 300 infusions of desferrioxamine annually had approximately 95 percent survival to age 20 years and 90 percent survival to age 30 years. In contrast, those with a mean of 0 to 75 infusions annually had only 20 percent survival to age 20 years and 0 percent survival to age 30 years!8 To demonstrate the monetary costs of noncompliance, the authors review their own two recent studies. They describe a Markov model to estimate lifetime incidence and cost of iron-related complications given “typical” versus “adequate” compliance with desferrioxamine. Typical compliance led to a mounting tally of complications and a hefty price of hospitalizations and treatment. The model did not consider cost to a society in terms of lost productivity due to disease or premature death, but one can imagine this cost to be quite substantial. A similar model was used to compare cost-effectiveness of desferrioxamine and deferasirox in terms of quality adjusted life-years from the standpoint of a US health-care system. To the best of our knowledge, this study is the first publication to examine cost-effectiveness of any iron chelator. The model predicted modest savings in terms of cost per quality-adjusted life-year and lifetime dollar costs with an oral versus infusional ICT. Studies by Payne and coworkers and Delea and coworkers have been sponsored by Novartis, the manufacturer of desferrioxamine and the new oral chelator, deferasirox. Both studies try to convince the reader that existing therapy has suboptimal real-life efficacy due to poor compliance, is expensive, and may have negative impact on the patients' quality of life. Both studies express hope that a novel iron chelator will be “more tolerable and less burdensome” and “could result in improved clinical and economic outcomes.” We may not agree with all of the authors' conclusions; however, we view deferasirox as a major advance in the field of ICT and hope that it will ease the burden of our patients. Finally, these two studies highlight another important fact: the cost of a treatment should not be considered synonymous with the cost of a drug. The real cost of treatment should take into account a number of factors including disease burden, cost of treatment, cost of monitoring therapy, risk of complications, cost of treating complications, quality of life, compliance, and consequences of noncompliance as well as social and societal impact. We thank Dr. B. Saxon (University of Toronto) for his helpful comments and suggestions.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.