Valganciclovir: A New Treatment for Cytomegalovirus Retinitis

In March 2001 the U.S. Food and Drug Administration (FDA) approved valganciclovir (Valcyte) as a new treatment for cytomegalovirus (CMV) retinitis. CMV retinitis is an inflammation of the retina, characterized by loss of visual acuity (sharpness) or "blind spots." A herpesvirus, CMV can infect almost any part of the body, but in persons with HIV it can devastate the eyes. If left untreated, CMV retinitis can lead to blindness.

History of CMV

Prior to the HIV/AIDS epidemic (i.e., before 1981), CMV retinitis was a rare disease occurring only among individuals with primary (inherited) immunodeficiency syndromes, autoimmune disorders, and those who had undergone organ transplants or immunosuppressive cancer chemotherapy. In people with HIV, retinitis is by far the most common manifestation of CMV. (In persons who have had bone marrow transplants, in contrast, CMV pneumonitis [lung inflammation] occurs much more frequently than retinitis.)

From the beginning of the HIV epidemic through 1996, estimates of the incidence of CMV disease in HIV-positive individuals ranged from 10% to 40%. In the current era of highly active antiretroviral therapy (HAART), the CMV prevalence rate among those who have access to HAART is less than 5%. CMV disease generally occurs late in the course of HIV disease and is associated with very low CD4 cell counts. The average CD4 cell count in persons with newly diagnosed CMV retinitis is below 30 cells/mm3. The main symptoms of CMV retinitis include "floaters" (moving spots), blurred vision, "blind spots" (missing portions of the visual field), and visions of flashing lights or sparks. Even subtle visual changes, such as a minor loss of peripheral vision, can signal the development of CMV retinitis.

Treatments for CMV retinitis are suppressive rather than curative, that is, they inhibit CMV progression but do not eradicate the virus. Resistance to the antiviral drugs previously approved by the FDA for CMV retinitis -- intravenous (IV) ganciclovir (Cytovene), foscarnet (Foscavir), and cidofovir (Vistide) -- is common. Until recently, no oral drug indicated for CMV induction therapy existed. Induction therapy is used to control the active spread of CMV and usually requires higher and/or more frequent dosing than is required after the disease has been brought under control. Once the spread of CMV is controlled, a lower and/or less frequent dose (known as maintenance therapy) is used to prevent CMV reactivation. In severely immunosuppressed people with CMV retinitis, lifelong maintenance therapy is recommended to keep the infection in a quiescent, or inactive, state. However, many people taking HAART have been able to stop preventive therapy. Recently published guidelines from the U.S. Public Health Service (USPHS) and the Infectious Diseases Society of America (IDSA) indicate that "discontinuation of [secondary CMV] prophylaxis may be considered in patients with a sustained (e.g., greater than three- to six-month) increase in CD4 cell count to greater than 100-150 cells/mm3 on HAART."

Pivotal CMV Study: Valganciclovir

Valganciclovir is the oral prodrug of IV ganciclovir. A prodrug is an inactive form of a drug that is converted within the body to a usable, or active, form. Valganciclovir is thus converted into ganciclovir when taken into the body as a pill. In 1997 the FDA declined to approve valganciclovir solely on the basis of pharmacokinetic data (information about drug absorption, metabolism, and elimination) demonstrating that WATCH a 900 mg daily dose of valganciclovir yielded drug levels comparable to standard daily IV ganciclovir at a dose of 5 mg/kg (as measured by area under the curve, or AUC, an expression of the total drug concentration in the blood plasma over time). Instead, the FDA required Hoffmann-La Roche, the manufacturer of valganciclovir, to conduct a randomized, controlled clinical trial demonstrating equivalence between valganciclovir and IV ganciclovir for CMV retinitis induction therapy.

Roche WV 15376 was a registrational study by Hoffmann-La Roche that evenly randomized 160 persons with CMV retinitis to receive either valganciclovir or IV ganciclovir. (A registrational, or pivotal, study is designed specifically -- with input from the FDA -- to produce data to be considered by the FDA for the drug's approval.) Valganciclovir was taken at a dose of 900 mg twice daily for three weeks, followed by 900 mg once daily for one week; IV ganciclovir was taken at a dose of 5 mg/kg twice daily for three weeks, followed by 5 mg/kg once daily for one week. After the four-week induction phase, all participants received maintenance therapy with open-label valganciclovir (900 mg daily). The primary study endpoint was CMV retinitis progression within four weeks of initiating treatment. Progression was measured using fundus photography, a specialized form of medical imaging used to diagnose abnormalities in the eye. Progression was defined as expansion or enlargement of an existing retinal lesion greater than or equal to 750 mm or the appearance of new lesions greater than or equal to 750 mm in diameter.

Statistically, the study was not a traditional head-to-head comparison (comparing two drugs with each other), but a noninferiority study (proving that one medication is not inferior to another). The study was powered to show that valganciclovir was not 10% less effective than IV ganciclovir. Both study arms were similar in terms of baseline demographics and disease status: approximately 90% were men, approximately 70% were taking HAART, the median CD4 cell count was 23 cells/mm3, the median HIV viral load was approximately 4,000 copies/mL, approximately 24% had zone 1 retinitis, and approximately 25% had bilateral retinitis (i.e., in both eyes). [Ed. note: Assessments of retinal abnormalities are described using a grading protocol defined in terms of zones that specify the location of lesions. The diameter of the optic nerve (the optic "head") is used as a unit of measurement such that "zone" is a unique measure for a given individual; specifically, zone refers to the number of optic head distances from the center. Zone 1 retinitis refers to lesions within the optic nerve; zones 2 and 3 refer to lesions located two and three diameter-lengths away from the optic nerve, respectively, and so on.]

Study Results

Of 146 participants who completed the four-week induction phase, seven of 73 participants (10%) in the valganciclovir arm experienced retinitis progression, compared with seven of 73 (10%) in the IV ganciclovir arm. No progression was seen in 64 and 63 participants, respectively, in the valganciclovir and IV ganciclovir arms. Hoffmann-La Roche also carried out an open-label safety study, Roche WV 15705, involving approximately 200 participants. No significant differences in adverse events or hematological (blood) toxicities were noted between the valganciclovir and IV ganciclovir arms.

Safety Data

Safety data for the Roche WV 15376 study were available for 158 participants. Adverse events were similar for valganciclovir and IV ganciclovir: headache (9% vs. 5%, respectively), diarrhea (16% vs. 10%), fever (13% vs. 11%), nausea (8% vs. 14%), and vomiting (11% vs. 6%). The administration of both IV ganciclovir and IV foscarnet requires long-term catheters, or plastic tubes, placed in large central veins (usually underneath the collarbone); catheters are prone infection. Not surprisingly, the IV ganciclovir group had significantly more infection: 2% for valganciclovir vs. 11% IV ganciclovir. There were no significant differences in hematological abnormalities between the valganciclovir and ganciclovir arms: absolute neutrophil count less than 750 cells/mm3 (21% vs. 19%), hemoglobin 6.5-8.0 g/dL (5% vs. 3%), and low platelet counts of 25,000-50,000 cells/mm3 0% vs. 1%). [Ed. note: Neutrophils are a type of immune system white blood cell that protects against bacterial infections. Hemoglobin is a measure of red blood cell oxygen-carrying capacity; a low hemoglobin level is an indication of anemia. Platelets are cell fragments necessary for normal blood clotting.]

The safety profile of valganciclovir is almost identical to that of IV ganciclovir. The most serious toxicity of both valganciclovir and IV ganciclovir is neutropenia (low neutrophil count), which leaves those taking the drug prone to infection. In the Roche WV 15376 study, approximately 20% participants experienced a grade 3 or 4 adverse event by week 4. [Ed. note: In clinical trials, symptoms are graded on basis of severity and frequency. Grade 1 side effects are mild and transient, grade 2 side effects are moderate or persistent, grade 3 adverse effects are severe, and grade 4 side effects are life-threatening.]


Valganciclovir has been shown to provide systemic drug levels (AUC) comparable to IV ganciclovir. Hoffmann-La Roche fulfilled the FDA Antiviral Drugs Division's approval requirements by carrying out a randomized, controlled Phase study of valganciclovir, which showed that valganciclovir was not inferior to IV ganciclovir for induction therapy for AIDS-related CMV retinitis. The recommended dosage of valganciclovir for individuals with active CMV retinitis is 900 (two 450 mg tablets) taken with food twice daily for 21 days. Following induction treatment, the recommended dosage is 900 mg (two 450 mg tablets) taken with food once daily.

Hoffmann-La Roche should be commended for testing valganciclovir against the "gold standard" treatment, IV ganciclovir; other available CMV antiviral drugs were approved using the quick and easy immediate vs. deferred trial design (in which some participants are given the drug at the start of the trial and compared with other participants who receive no treatment until later in the study). Valganciclovir is the first drug that has been tested against IV ganciclovir for induction therapy for CMV retinitis since the Studies of Ocular Complications of AIDS (SOCA) Research Group tested IV foscarnet vs. IV ganciclovir in 1990.

In the 13 years since the FDA approved IV ganciclovir, the prevalence of AIDS-related CMV retinitis has declined dramatically in industrialized countries, due in large part to the use of HAART. Nevertheless, having an effective oral agent to treat CMV -- freeing those who need treatment from the risks and discomfort of IV therapy -- will completely alter the clinical management of this infection.

A similar article by Michael Marco appears on TAG's Web site:

Michael Marco is director of the Infections and Oncology Project of the New York-based Treatment Action Group (TAG).

CMV Factors and Diagnosis

Factors most associated with CMV disease are a CD4 cell count below 50 cells/mm3 and at least one other prior opportunistic infection (OI). Recent studies have shown that anyone who has detectable CMV levels using the PCR (polymerase chain reaction) viral load test, or who is either CMV antigen or culture positive, is at higher risk for developing CMV disease; however, a measurable CMV level by itself does not always correlate with active infection or symptomatic disease.

The diagnosis of CMV retinitis is usually confirmed by an ophthalmologist, a physician who specializes in diseases of the eye. The ophthalmologist applies chemicals to dilate the eye, making it easier to look inside the eyeball for evidence of retinal lesions that may be caused by CMV. The virus also commonly attacks the gastrointestinal tract (stomach and intestines) and the nervous system (brain and spinal cord); the diagnosis of CMV disease in other parts of the body is usually done by performing a biopsy of tissue from the suspected organ.

For more information about CMV see:

Back to the SFAF BETA Winter 2002 contents page.