This application is for analyses of data generated by the Cytomegalovirus (CMV) Retinitis and Viral Resistance (CRVR) Study. CMV retinitis is a common opportunistic infection among patients with AIDS. Untreated CMV retinitis is a progressive infection, the end result of which is destruction of the retina and blindness. Unless immune reconstitution occurs as a consequence of highly active antiretroviral therapy (HAART), chronic, suppressive therapy is required to prevent relapse of the disease. Resistant CMV (due to mutations in the UL97 or UL54 genes) has been reported to occur in approximately25% of patients by 9 months of anti-CMV therapy. Resistant CMV in the blood or urine is associated with substantially increased risks of retinitis progression, loss of retinal area, and loss of visual acuity. The genome of CMV detected in the eye is nearly always identical to that detected in the blood or urine. Many relapses of retinitis are due to limited intraocular drug penetration rather than resistance, and current practice typically is to re-induce patients with the same drug rather than switch drugs, an approach which is ineffective for resistant CMV. If resistant CMV is identified, treatment is changed to an alternative drug. Current methods for identifying patients who harbor resistant CMV require culturing and testing isolates for resistance, a process which may take 10 weeks. In order for resistance testing to achieve clinical utility, more rapid methods for identifying patients who harbor resistant CMV are needed; two candidates are: 1) CMV viral load, and 2) polymerase chain reaction (PCR) amplification of CMV DNA from blood specimens and sequencing it for mutations conferring ganciclovir resistance. The CRVR Study is a prospective study of 309 patients with AIDS and CMV retinitis for the occurrence of resistant CMV isolated from either the blood or urine. We will use data generated from the CRVR Study to: 1) evaluate CMV viral load as a means to rapidly identify patients who harbor resistant CMV; 2) evaluate PCR and sequencing of the CMV UL97 gene from blood specimens as a means to rapidly identify patients who harbor CMV resistant to ganciclovir; and 3) to evaluate the effect of HAART on the incidence of resistant CMV. Our hypotheses are that: 1) increases in CMV viral load will be a marker for patients who harbor resistant CMV; 2) screening of blood directly for CMV UL97 mutations will identify patients who harbor ganciclovir-resistant CMV; and 3) that the rate of resistance will be decreased among patients who developed CMV retinitis and then were started on HAART but similar to that from the pre-HAART era among patients who develop CMV retinitis after already having been treated with HAART. Three publications are planned: 1) """"""""CMV viral load as a predictor of resistant CMV among patients with CMV retinitis""""""""; 2) """"""""PCR amplification and sequencing of the UL97 gene from blood specimens to identify resistant CMV among patients with CMV retinitis""""""""; and 3) """"""""The effect of HAART on the incidence of resistant CMV among patients with CMV retinitis.""""""""
