Treatment for CMV Retinitis (Systemic):

Intravenous formulations of ganciclovir (5 mg/kg IV twice daily) and foscarnet (90 mg/kg IV two times daily) are approved for the treatment of CMV retinitis in immune-compromised individuals. Higher doses of ganciclovir (5 mg/kg IV bid) or foscarnet (90 mg/kg IV bid) are used initially for 2-3 weeks as induction therapy followed by maintenance treatment at lower doses. Ganciclovir is preferred treatment by most practitioners. The major side effect associated with ganciclovir is bone-marrow suppression, which may prevent or limit concomitant use of AZT. Ionized hypocalcemia, reversible renal insufficiency, and penile ulcers (mainly in uncircumcised men) are associated with foscarnet treatment.

Foscarnet appears equivalent to ganciclovir in limiting the spread of CMV retinitis; however, it may confer a survival benefit over ganciclovir in patients with advanced HIV, presumably because of its intrinsic anti-HIV activity. The NEI/ACTG study for ocular complications of AIDS (Meinert et al.) randomized 234 patients with CMV retinitis to receive ganciclovir (107 patients) or foscarnet (127 patients). The trial was suspended after 19 months when a significant mortality benefit associated with foscarnet was observed. Although both drugs were equally effective in halting progression of CMV retinitis, the median survival was 12.6 months in the foscarnet group and 8.5 months in the ganciclovir group. This survival benefit could not be attributed to differences in antiretroviral therapy between the two groups. However, foscarnet was associated with substantial toxicity: 22/39 treatment switches from foscarnet to ganciclovir were attributed to foscarnet toxicity, while only 1/14 switches from ganciclovir to foscarnet was attributed to ganciclovir toxicity.

Ganciclovir has been demonstrated to delay progression of peripheral CMV retinitis in AIDS patients (Peripheral CMV retinitis is not immediately sight threatening). Spector et al. randomized 22 patients to deferred treatment and 13 to immediate treatment with ganciclovir (5 mg/kg bid for 14 days followed by 5 mg/kg qd for 14 weeks). Patients in the deferred group were offered ganciclovir if retinitis progressed, 10/13 in the immediate group, and 20/22 in the deferred group had progressive CMV retinitis. The median time to progression was 13.5 days for the deferred group versus 49.5 days for the immediate treatment group (P = 0.001).

The results of a clinical trial in patients who required retreatment of CMV after maintenance therapy has failed were recently reported (Jabs et al.). The study (ACTG 228) compared ganciclovir to foscarnet to a foscarnet/ganciclovir combination in recently treated patients who reported CMV disease progression. A total of 271 patients were randomized to one of three arms; foscarnet (90 mg/kg q 12h x 14 days with 120 mg/kg qd maintenance), ganciclovir (5.0 to 7.5 mg/kg q 12h x 14 days with 10 mg/kg qd maintenance), or a combination of foscarnet (90 mg/kg q 12h x 14 days with 90 mg/kg qd maintenance) and ganciclovir (5 mg/kg qd x 14 days with 5 mg/kg qd maintenance therapy). CMV retinitis progression was reported at 4.8 months in the combination group, 2.1 months in those who received ganciclovir alone, and 1.6 months in those who received foscarnet alone (P = 0.00002). 'Those who reported disease progression in the monotherapy groups were switched to the alternate monotherapy treatment; those who progressed in the combination group received higher doses of both drugs. For patients originally receiving either ganciclovir or foscarnet monotherapy, a median of 1.6 relapses were reported. There was no significant difference between switching to the alternate treatment and remaining on the initial regimen. However rates of visual field loss were significantly lower in the combination group (16- degrees/month compared to 18- and 31- degrees/month) (P = 0.003). Quality of life scores reported were significantly better in the ganciclovir monotherapy group.

MSL-109, a human anti-CMV monoclonal antibody, is being studied in combination with either foscarnet or ganciclovir for the treatment of retinitis (ACTG 266). This phase II double-blind, placebo-controlled trial will randomize patients to one of two doses of MSL-109 or placebo to be administered concomitantly with standard induction therapy. A second study is being conducted by the Studies of the Ocular Complications of AIDS (SOCA) of the National Eye Institute (NEI) with the ACTG (ACTG 294). This trial compares MSL-109 (60 mg IV once every two weeks) to placebo in 167 patients receiving standard therapy for either newly diagnosed or relapsed CMV retinitis.

An intraocular ganciclovir implant has been reported to delay progression of peripheral CMV retinitis. A New Drug Application (NDA) has been submitted to the Food and Drug Administration and has been reviewed by an advisory committee to the FDA, which has recommended approval.

The Chiron Ganciclovir Implant Study Group (Kupperman et al.) has reported the results from a randomized trial examining two release rates of an intraocular ganciclovir implant (1µg/h and 2 µg/h) versus standard dose intravenous ganciclovir. A total of 188 patients with symptomatic CMV retinitis were evenly divided to receive either of the two intraocular ganciclovir doses or IV ganciclovir. While no significant differences in disease progression were reported between the two intraocular dose groups, significant delays in disease progression were reported after 182 days in the 2 µg/h ganciclovir implant group and 72 days in the ganciclovir group (P = 0.000 1). The rate of other eye involvement was 40% in the implant group and 16% in the IV group (NS). Extraocular disease occurred in 15% of the patients in the implant groups combined. There was a slight decrease in visual acuity in the immediate post-operative period for implant patients, although visual acuity returned to preoperative levels within 24 weeks. No statistical differences in survival were reported. Incidences of endophthalmitis, retinal detachment, and vitreous hemorrhage were higher in the implant groups. However, due to the use of G-CSF required by neutropenic patients receiving ganciclovir infusions, the incidences of catheter related sepsis and severe nausea was higher in the IV group.

Martin et al. randomized twenty six patients (30 eyes) to either immediate treatment with a 1 µg/h intraocular ganciclovir implant or deferreded treatment None of the patients enrolled had received prior treatment for CMV retinitis. The median time to progression of retinitis was 15 days in the deferred group (n = 16) versus 226 days in the immediate treatment group (n = 14)(P < .00001). In the entire study population, biopsy- proven visceral CMV disease developed in 8/26 (31%) patients and 67% of the 26 patients developed retinitis in the fellow eye. Seven patients (18%) suffered a treatment-induced retinal detachment, with five occurring between days 30 and 64. There were 2 cases of persistent hypotony reported.