To compare the newest CMV retinitis drug, cidofovir, with a regimen of the ganciclovir intraocular device plus oral ganciclovir with respect to efficacy in preventing vision loss.

To compare a treatment regimen that incorporates highly active local therapy (ganciclovir device) with a treatment regimen that does not.

Cytomegalovirus (CMV) is among the most frequently encountered opportunistic infections in patients with AIDS. In the era of prophylaxis for pneumocystic pneumonia, CMV disease is estimated to affect 45 percent of patients with AIDS sometime between the diagnosis of AIDS and death. Retinitis has been estimated to account for up to 85 percent of CMV disease in these patients, making CMV retinitis the most common ocular infection encountered. CMV retinitis is a relatively late-stage manifestation, associated with cluster of differentiation 4 (CD4) + T-cell counts < 100 cells/µL and often < 50 cells/µL.

All currently available treatments for CMV suppress viral replication but do not eliminate the virus from the body. Discontinuation of therapy is associated with a prompt relapse of the retinitis. Despite the use of chronic suppressive therapy, relapse of the retinitis generally occurs, at least with systemically administered anti-CMV drugs.

The first two treatments approved for CMV retinitis were intravenous ganciclovir and intravenous foscarnet. Both are given by daily intravenous infusions and therefore require central venous catheters. The development of newer treatments has focused not only on efficacious treatments, but also on treatments that do not require central venous catheters. Available treatments now include oral ganciclovir, the ganciclovir intraocular device, and intravenous cidofovir.

In vitro data suggest that combination therapies are synergistic in inhibiting viral replication; these therapies include a foscarnet-ganciclovir combination and a cidofovir-ganciclovir combination. In the SOCA--CMV Retinitis Retreatment Trial, the combination of intravenous ganciclovir and foscarnet was more effective than either drug alone for the treatment of relapsed retinitis. Therefore, the combination of intermittent intravenous cidofovir and daily oral ganciclovir may be an attractive therapy for relapsed disease because it may provide synergy for controlling both ocular and visceral disease while not necessitating either a central venous catheter or an intraocular surgical procedure.

The Ganciclovir-Cidofovir CMV Retinitis Trial (GCCRT) is a randomized, multicenter clinical trial. Patients will be assigned to receive one of two regimens: (1) ganciclovir intraocular device plus oral ganciclovir or (2) intravenous cidofovir. The intraocular device will be surgically implanted at baseline and again every 6 to 8 months in eyes with CMV retinitis. Oral ganciclovir is taken at a dose of 1 gram three times daily. Cidofovir will be administered intravenously at 5 mg/kg once weekly for 2 consecutive weeks and once every 2 weeks thereafter. If disease progression occurs in patients receiving cidofovir, patients will be given reinduction therapy, and oral ganciclovir at a dose of 1 gram three times per day will be added to the treatment. If patients assigned to cidofovir are unable to tolerate that regimen, an alternative systemic regimen will be recommended.

Study outcome variables include a decrease of three or more lines from baseline in best corrected visual acuity and rate of visual field loss. The study will also assess other variables including mortality, blood CMV and HIV load, quality of life, and medical costs.

Treatment assignment will not be masked to either patients or clinicians; however, reading of fundus photographs to determine both change in retinal involvement and progression will be masked.

Inclusion criteria:

• Age 13 years or older
• Diagnosis of AIDS according to current Centers for Disease Control and Prevention (CDC) definition
• Diagnosis of active CMV retinitis by a SOCA-certified ophthalmologist (involvement of any zone or amount of retina is allowed)
• Best corrected visual acuity of 20/100 or better in at least one eye
• At least one lesion 750 cells/µL or greater
• Platelet count 50,000 cells/µL or greater
• Willingness and ability, with the assistance of a caregiver if necessary to comply with treatment and follow up procedures
• Willingness of all men and women of childbearing potential to practice adequate birth control to prevent pregnancies during the study and for 3 months afterwards
• Collection of all baseline data within 5 days prior to randomization
• Signed consent statement

Exclusion criteria:

• Media opacities that preclude visualization of the fundus of all otherwise eligible eyes
• Treatment for CMV retinitis with the ganciclovir intraocular implant within 9 months of study entry
• Medical problems or drug or alcohol abuse sufficient to hinder adherence to treatment or follow up procedures
• Unwillingness to refrain from breast-feeding during the study and for 3 months afterwards