The Cost of Prevention and Treatment of CMV Retinitis
The past several years have seen a greatly expanded arsenal of treatments approved by the Food and Drug Administration for the treatment of CMV retinitis.In addition to intravenous foscarnet (Foscavirþ, Astra) and intravenous ganciclovir (Cytovene IVþ, Roche), oral ganciclovir (Cytoveneþ, Roche) and a sustained-release ganciclovir intraocular device (Vitrasert®, Chiron) have been approved for the treatment of CMV retinitis. Cidofovir (Gilead) has been recommended for approval by an FDA advisory panel and should be commercially available soon. In addition, the off-label use of ganciclovir and foscarnet for repeated intravitreous injections has proven effective in uncontrolled studies. Finally, the availability of oral ganciclovir and findings from the Syntex (Roche)-sponsored prophylaxis study have raised hopes that CMV retinitis may become, like Pneumocystis carinii pneumonia, a preventable disease.
The costs associated with CMV retinitis are by any account substantial, driven by the expense of both the drugs themselves and the drugs used to treat drug toxicity, the need for indefinite maintenance therapy, the home care associated with intravenous therapy, the frequency of the disease (up to 40% of patients with AIDS develop CMV retinitis), and the desire on the part of patients and care providers to treat CMV retinitis aggressively because of the value placed on preservation of eyesight.
Much is known about the efficacy of different drugs and different drug dosages, but little information is available to care providers about the cost of therapy.As more HIV-infected patients are moved into managed-care settings, and as states try to cope with budget-busting but effective advances in other areas of HIV care (such as protease inhibitors), it is inevitable that the costs associated with CMV retinitis will come under greater scrutiny.
It is impossible to come up with a single estimate of the cost of treatment of CMV retinitis. Any cost analysis has to make assumptions which may not take into account regional differences, including the cost of home care, reimbursement for intravitreous injections, and medical standards of care (e.g., frequency of laboratory monitoring, ophthalmic examinations, and retinal photographs). The jointly sponsored Chiron-Roche study, which has randomized patients to treatment with intravenous ganciclovir, Vitrasert®, or Vitrasert® plus oral ganciclovir will help decide the relative efficacies of these different regimens and may yield information on cost. Cidofovir has not been compared to either ganciclovir or foscarnet in clinical trials.
Any cost analysis of CMV retinitis treatment must make assumptions about the cost of anti-CMV drugs, drugs to treat side effects of anti-CMV therapy, patient survival, and the incidence of complications such as retinal detachment and line sepsis.With these issues in mind, it is estimated that the cost of treatment per patient per year is between $30,000 and $100,000. The average wholesale price of foscarnet is two to three times that of ganciclovir. Surprisingly, oral ganciclovir is actually more expensive than intravenous ganciclovir. Systemic side effects (including leukopenia) are similar with oral and intravenous ganciclovir in published studies. However, the use of the oral drug reduces the costs associated with IV paraphernalia, line infection, and home health care, so that treatment with oral ganciclovir is comparable in cost to that with intravenous ganciclovir. It must be kept in mind that induction and reinduction therapy preceding oral ganciclovir maintenance therapy must be done with intravenous ganciclovir, thus substantially increasing the cost of "oral therapy."
Local therapy has the potential to dramatically reduce the cost of treating CMV retinitis, because the need for laboratory tests and physical examinations as well as IV paraphernalia is significantly reduced.Separate uncontrolled studies by Canadian and Australian researchers calculated that repeated intravitreous injections of ganciclovir as maintenance therapy were roughly a third as expensive as intravenous ganciclovir. In our experience, however, the need to repeat the injections weekly makes this regimen impractical for most patients.
Vitrasert® is estimated to cost $4000 (average sale price to the hospital). Surgical, anesthetic, and operating room fees add roughly $2000 per surgery (slightly more than the cost of a Hickman catheter placement). If one assumes from published data that 50% of patients have unilateral retinitis at the time of diagnosis, that the implant is replaced every six months, and that median survival after the diagnosis of CMV retinitis is one year, it can be estimated that, on average, a patient will require three implants over his/her lifetime. Estimates of cost savings with Vitrasert® are based on clinical studies indicating a much lower rate of relapsed CMV retinitis (thus avoiding costly intravenous reinduction therapy) and the other advantages of local therapy listed above. The long-term cost advantage of Vitrasert® is obviously reduced if the patient is concurrently treated with oral ganciclovir to prevent extraocular disease.
The cost benefits of prophylaxis against CMV retinitis in AIDS patients with oral ganciclovir remain unproven. At a cost of approximately $15,000 per patient per year for the cost of the drug alone, oral ganciclovir for prevention of CMV retinitis is in many ways the opposite extreme of the low-cost, highly efficacious use of oral drugs to prevent PCP. The detection of CMV DNA in blood by PCR may someday help identify patients at particularly high risk of CMV retinitis, allowing for more selective and cost-effective use of oral prophylaxis.
Of course, the cost of CMV retinitis care is only part of the issue.Determining the quality of life associated with different therapies and the value of preservation of vision is enormously difficult, if not impossible. For example, what is the benefit of aggressive surgical and medical therapy of advanced unilateral CMV retinitis which has resulted in extensive retinal destruction and limited vision, when oral ganciclovir alone might be sufficient to prevent retinitis in the other eye, but not to preserve whatever vision is left in the infected eye? One hopes that such decisions will be left to the patient and the physician, but economic forces may eventually dictate otherwise. For now it is reasonable to make the following generalizations:
This article was provided by Johns Hopkins AIDS Service. It is a part of the publication Hopkins HIV Report.