Mechanism of Action

Cidofovir suppresses cytomegalovirus (CMV) replication by selective inhibition of viral DNA synthesis. Biochemical data support selective inhibition of CMV DNA polymerase by cidofovir diphosphate, the active intracellular metabolite of cidofovir. Cidofovir diphosphate inhibits herpesvirus polymerases at concentrations that are 8- to 600-fold lower than those needed to inhibit human cellular DNA polymerases alpha, beta, and gamma1, 2, 3. Incorporation of cidofovir into the growing viral DNA chain results in reductions in the rate of viral DNA synthesis.

In Vitro Susceptibility

Cidofovir is active in vitro against a variety of laboratory and clinical isolates of CMV and other herpesviruses (Table 1). Controlled clinical studies of efficacy have been limited to patients with AIDS and CMV retinitis.

Resistance

CMV isolates with reduced susceptibility to cidofovir have been selected in vitro in the presence of high concentrations of cidofovir4. IC50 values for selected resistant isolates ranged from 7–15 µM.

There are insufficient data at this time to assess the frequency or the clinical significance of the development of resistant isolates following VISTIDE administration to patients.

The possibility of viral resistance should be considered for patients who show a poor clinical response or experience recurrent retinitis progression during therapy.

Cross Resistance

Cidofovir-resistant isolates selected in vitro following exposure to increasing concentrations of cidofovir were assessed for susceptibility to ganciclovir and foscarnet4. All were cross resistant to ganciclovir, but remained susceptible to foscarnet. Ganciclovir- or ganciclovir/foscarnet-resistant isolates that are cross resistant to cidofovir have been obtained from drug naive patients and from patients following ganciclovir or ganciclovir/ foscarnet therapy. To date, the majority of ganciclovir-resistant isolates are UL97 gene product (phosphokinase) mutants and remain susceptible to cidofovir5. Reduced susceptibility to cidofovir, however, has been reported for DNA polymerase mutants of CMV which are resistant to ganciclovir6–9. To date, all clinical isolates which exhibit high level resistance to ganciclovir, due to mutations in both the DNA polymerase and UL97 genes, have been shown to be cross resistant to cidofovir. Cidofovir is active against some, but not all, CMV isolates which are resistant to foscarnet10–12. The incidence of foscarnet-resistant isolates that are resistant to cidofovir is not known.

A few triple-drug resistant isolates have been described. Genotypic analysis of two of these triple-resistant isolates revealed several point mutations in the CMV DNA polymerase gene. The clinical significance of the development of these cross-resistant isolates is not known.

PHARMACOKINETICS

VISTIDE must be administered with probenecid. The pharmacokinetics of cidofovir, administered both without and with probenecid, are described below.

The pharmacokinetics of cidofovir without probenecid were evaluated in 27 HIV-infected patients with or without asymptomatic CMV infection. Dose-independent pharmacokinetics were demonstrated after one hr infusions of 1.0 (n = 5), 3.0 (n = 10), 5.0 (n = 2) and 10.0 (n = 8) mg/kg (See Table 2 for pharmacokinetic parameters). There was no evidence of cidofovir accumulation after 4 weeks of repeated administration of 3 mg/kg/week (n = 5) without probenecid. In patients with normal renal function, approximately 80 to 100% of the VISTIDE dose was recovered unchanged in urine within 24 hr (n = 27). The renal clearance of cidofovir was greater than creatinine clearance, indicating renal tubular secretion contributes to the elimination of cidofovir.

The pharmacokinetics of cidofovir administered with probenecid were evaluated in 12 HIV-infected patients with or without asymptomatic CMV infection and 10 patients with relapsing CMV retinitis. Dose-independent pharmacokinetics were observed for cidofovir, administered with probenecid, after one hr infusions of 3.0 (n = 12), 5.0 (n = 6), and 7.5 (n = 4) mg/kg (See Table 2). Approximately 70 to 85% of the VISTIDE dose administered with concomitant probenecid was excreted as unchanged drug within 24 hr. When VISTIDE was administered with probenecid, the renal clearance of cidofovir was reduced to a level consistent with creatinine clearance, suggesting that probenecid blocks active renal tubular secretion of cidofovir.

In vitro, cidofovir was less than 6% bound to plasma or serum proteins over the cidofovir concentration range 0.25 to 25 µg/mL. CSF concentrations of cidofovir following intravenous infusion of VISTIDE 5 mg/kg with concomitant probenecid and intravenous hydration were undetectable (< 0.1 µg/mL, assay detection threshold) at 15 minutes after the end of a 1 hr infusion in one patient whose corresponding serum concentration was 8.7 µg/mL.

DRUG-DRUG INTERACTIONS

SPECIAL POPULATIONS

Delayed Versus Immediate Therapy (Study 105)

In stage 1 of this open-label trial, conducted by the Studies of the Ocular Complications of AIDS (SOCA) Clinical Research Group, 29 previously untreated patients with peripheral CMV retinitis were randomized to either immediate treatment with VISTIDE (5 mg/kg once a week for 2 weeks, then 3 mg/kg every other week) or to have VISTIDE delayed until progression of CMV retinitis13. In stage 2 of this trial, an additional 35 previously untreated patients with peripheral CMV retinitis were randomized to either immediate treatment with VISTIDE (5 mg/kg once a week for 2 weeks, then 5 mg/kg every other week), immediate treatment with VISTIDE (5 mg/kg once a week for 2 weeks, then 3 mg/kg every other week), or to have VISTIDE delayed until progression of CMV retinitis. Of the 64 patients in this study, 12 were randomized to 5 mg/kg maintenance therapy, 26 to 3 mg/kg maintenance therapy, and 26 to delayed therapy. Of the 12 patients enrolled in the 5 mg/kg maintenance group, 5 patients progressed, 5 patients discontinued therapy and 2 patients had no progression at study completion. Based on masked readings of retinal photographs, the median [95% confidence interval (CI)] time to retinitis progression was not reached (25, not reached) for the 5 mg/kg maintenance group. Median (95% CI) time to the alternative endpoint of retinitis progression or study drug discontinuation was 44 days (24, 207) for the 5 mg/kg maintenance group. Patients receiving 5 mg/kg maintenance had delayed time to retinitis progression compared to patients receiving 3 mg/kg maintenance or deferred therapy.

Delayed Versus Immediate Therapy (Study 106)

In an open-label trial, 48 previously untreated patients with peripheral CMV retinitis were randomized to either immediate treatment with VISTIDE (5 mg/kg once a week for 2 weeks, then 5 mg/kg every other week), or to have VISTIDE delayed until progression of CMV retinitis14. Patient baseline characteristics and disposition are shown in Table 3. Of 25 and 23 patients in the immediate and delayed groups respectively, 23 and 21 were evaluable for retinitis progression as determined by retinal photography. Based on masked readings of retinal photographs, the median [95% confidence interval (CI)] times to retinitis progression were 120 days (40, 134) and 22 days (10, 27) for the immediate and delayed therapy groups, respectively. This difference was statistically significant. However, because of the limited number of patients remaining on treatment over time (3 of 25 patients received VISTIDE for 120 days or longer), the median time to progression for the immediate therapy group was difficult to precisely estimate. Median (95% CI) times to the alternative endpoint of retinitis progression or study drug discontinuation (including adverse events, withdrawn consent, and systemic CMV disease) were 52 days (37, 85) and 22 days (13, 27) for the immediate and delayed therapy groups, respectively. This difference was statistically significant. Time to progression estimates from this study may not be directly comparable to estimates reported for other therapies.

Dose-response study of VISTIDE (Study 107)

In an open-label trial, 100 patients with relapsing CMV retinitis were randomized to receive 5 mg/kg once a week for 2 weeks and then either 5 mg/kg (n = 49) or 3 mg/kg (n = 51) every other week. Enrolled patients had been diagnosed with CMV retinitis an average of 390 days prior to randomization and had received a median of 3.8 prior courses of systemic CMV therapy. Eighty-four of the 100 patients were considered evaluable for progression by serial retinal photographs (43 randomized to 5 mg/kg and 41 randomized to 3 mg/kg). Twenty-six and 21 patients discontinued therapy due to either an adverse event, intercurrent illness, excluded medication, or withdrawn consent in the 5 mg/kg and 3 mg/kg groups, respectively. Thirty-eight of the 100 randomized patients had progressed according to masked assessment of serial retinal photographs (13 randomized to 5 mg/kg and 25 randomized to 3 mg/kg). Using retinal photographs, the median (95% CI) times to retinitis progression for the 5 mg/kg and 3 mg/kg groups were 115 days (70, not reached) and 49 days (35, 52), respectively. This difference was statistically significant. Similar to Study 106, the median time to retinitis progression for the 5 mg/kg group was difficult to precisely estimate due to the limited number of patients remaining on treatment over time (4 of the 49 patients in the 5 mg/kg group were treated for 115 days or longer). Median (95% CI) times to the alternative endpoint of retinitis progression or study drug discontinuation were 49 days (38, 63) and 35 days (27, 39) for the 5 mg/kg and 3 mg/kg groups, respectively. This difference was statistically significant.

Nephrotoxicity

Dose-dependent nephrotoxicity is the major dose-limiting toxicity related to VISTIDE administration. Cases of acute renal failure resulting in dialysis and/or contributing to death have occurred with as few as one or two doses of VISTIDE. Renal function (serum creatinine and urine protein) must be monitored within 48 hours prior to each dose of VISTIDE. Dose adjustment or discontinuation is required for changes in renal function (serum creatinine and/or urine protein) while on therapy. Proteinuria, as measured by urinalysis in a clinical laboratory, may be an early indicator of VISTIDE-related nephrotoxicity. Continued administration of VISTIDE may lead to additional proximal tubular cell injury, which may result in glycosuria, decreases in serum phosphate, uric acid, and bicarbonate, elevations in serum creatinine, and/or acute renal failure, in some cases, resulting in the need for dialysis. Patients with these adverse events occurring concurrently and meeting a criteria of Fanconi's syndrome have been reported. Renal function that did not return to baseline after drug discontinuation has been observed in clinical studies of VISTIDE.

Intravenous normal saline hydration and oral probenecid must accompany each VISTIDE infusion. Probenecid is known to interact with the metabolism or renal tubular excretion of many drugs (see PRECAUTIONS). The safety of VISTIDE has not been evaluated in patients receiving other known potentially nephrotoxic agents, such as intravenous aminoglycosides (e.g., tobramycin, gentamicin, and amikacin), amphotericin B, foscarnet, intravenous pentamidine, vancomycin, and non-steroidal anti-inflammatory agents (see DOSAGE AND ADMINISTRATION).

Preexisting Renal Impairment

Initiation of therapy with VISTIDE is contraindicated in patients with a baseline serum creatinine > 1.5 mg/dL, a creatinine clearance ≤ 55 mL/min, or a urine protein ≥ 100 mg/dL (equivalent to ≥ 2+ proteinuria).

Hematological Toxicity

Neutropenia may occur during VISTIDE therapy. Neutrophil count should be monitored while receiving VISTIDE therapy.

Decreased Intraocular Pressure/Ocular Hypotony

Decreased intraocular pressure may occur during VISTIDE therapy, and in some instances has been associated with decreased visual acuity. Intraocular pressure should be monitored during VISTIDE therapy.

Metabolic Acidosis

Decreased serum bicarbonate associated with proximal tubule injury and renal wasting syndrome (including Fanconi's syndrome) have been reported in patients receiving VISTIDE (see ADVERSE REACTIONS). Cases of metabolic acidosis in association with liver dysfunction and pancreatitis resulting in death have been reported in patients receiving VISTIDE.

General

Due to the potential for increased nephrotoxicity, doses greater than the recommended dose must not be administered and the frequency or rate of administration must not be exceeded (see DOSAGE AND ADMINISTRATION).

VISTIDE is formulated for intravenous infusion only and must not be administered by intraocular injection. Administration of VISTIDE by infusion must be accompanied by oral probenecid and intravenous saline prehydration (see DOSAGE AND ADMINISTRATION).

Uveitis/Iritis

Uveitis or iritis was reported in clinical trials and during postmarketing in patients receiving VISTIDE therapy. Treatment with topical corticosteroids with or without topical cycloplegic agents should be considered. Patients should be monitored for signs and symptoms of uveitis/iritis during VISTIDE therapy.

Information for Patients

Patients should be advised that VISTIDE is not a cure for CMV retinitis, and that they may continue to experience progression of retinitis during and following treatment. Patients receiving VISTIDE should be advised to have regular follow-up ophthalmologic examinations. Patients may also experience other manifestations of CMV disease despite VISTIDE therapy.

HIV-infected patients may continue taking antiretroviral therapy, but those taking zidovudine should be advised to temporarily discontinue zidovudine administration or decrease their zidovudine dose by 50%, on days of VISTIDE administration only, because probenecid reduces metabolic clearance of zidovudine.

Patients should be informed of the major toxicity of VISTIDE, namely renal impairment, and that dose modification, including reduction, interruption, and possibly discontinuation, may be required. Close monitoring of renal function (routine urinalysis and serum creatinine) while on therapy should be emphasized.

The importance of completing a full course of probenecid with each VISTIDE dose should be emphasized. Patients should be warned of potential adverse events caused by probenecid (e.g., headache, nausea, vomiting, and hypersensitivity reactions). Hypersensitivity/allergic reactions may include rash, fever, chills and anaphylaxis. Administration of probenecid after a meal or use of antiemetics may decrease the nausea. Prophylactic or therapeutic antihistamines and/or acetaminophen can be used to ameliorate hypersensitivity reactions.

Patients should be advised that cidofovir causes tumors, primarily mammary adenocarcinomas, in rats. VISTIDE should be considered a potential carcinogen in humans (See Carcinogenesis, Mutagenesis, & Impairment of Fertility). Women should be advised of the limited enrollment of women in clinical trials of VISTIDE.

Patients should be advised that VISTIDE caused reduced testes weight and hypospermia in animals. Such changes may occur in humans and cause infertility. Women of childbearing potential should be advised that cidofovir is embryotoxic in animals and should not be used during pregnancy. Women of childbearing potential should be advised to use effective contraception during and for 1 month following treatment with VISTIDE. Men should be advised to practice barrier contraceptive methods during and for 3 months after treatment with VISTIDE.

Drug Interactions

Carcinogenesis, Mutagenesis, & Impairment of Fertility

Chronic, two-year carcinogenicity studies in rats and mice have not been carried out to evaluate the carcinogenic potential of cidofovir. However, a 26-week toxicology study evaluating once weekly subscapular subcutaneous injections of cidofovir in rats was terminated at 19 weeks because of the induction, in females, of palpable masses, the first of which was detected after six doses. The masses were diagnosed as mammary adenocarcinomas which developed at doses as low as 0.6 mg/kg/week, equivalent to 0.04 times the human systemic exposure at the recommended intravenous VISTIDE dose based on AUC comparisons.

In a 26-week intravenous toxicology study in which rats received 0.6, 3, or 15 mg/kg cidofovir once weekly, a significant increase in mammary adenocarcinomas in female rats as well as a significant incidence of Zymbal's gland carcinomas in male and female rats were seen at the high dose but not at the lower two doses. The high dose was equivalent to 1.1 times the human systemic exposure at the recommended dose of VISTIDE, based on comparisons of AUC measurements. In light of the results of these studies, cidofovir should be considered to be a carcinogen in rats as well as a potential carcinogen in humans.

Cynomolgus monkeys received intravenous cidofovir, alone and in conjunction with concomitant oral probenecid, intravenously once weekly for 52 weeks at doses resulting in exposures of approximately 0.7 times the human systemic exposure at the recommended dose of VISTIDE. No tumors were detected. However, the study was not designed as a carcinogenicity study due to the small number of animals at each dose and the short duration of treatment.

No mutagenic response was observed in microbial mutagenicity assays involving Salmonella typhimurium (Ames) and Escherichia coli in the presence and absence of metabolic activation. An increase in micronucleated polychromatic erythrocytes in vivo was seen in mice receiving ≥ 2000 mg/kg, a dosage approximately 65-fold higher than the maximum recommended clinical intravenous VISTIDE dose based on body surface area estimations. Cidofovir induced chromosomal aberrations in human peripheral blood lymphocytes in vitro without metabolic activation. At the 4 cidofovir levels tested, the percentage of damaged metaphases and number of aberrations per cell increased in a concentration-dependent manner.

Studies showed that cidofovir caused inhibition of spermatogenesis in rats and monkeys. However, no adverse effects on fertility or reproduction were seen following once weekly intravenous injections of cidofovir in male rats for 13 consecutive weeks at doses up to 15 mg/kg/week (equivalent to 1.1 times the recommended human dose based on AUC comparisons). Female rats dosed intravenously once weekly at 1.2 mg/kg/week (equivalent to 0.09 times the recommended human dose based on AUC) or higher, for up to 6 weeks prior to mating and for 2 weeks post mating had decreased litter sizes and live births per litter and increased early resorptions per litter. Peri- and post-natal development studies in which female rats received subcutaneous injections of cidofovir once daily at doses up to 1.0 mg/kg/day from day 7 of gestation through day 21 postpartum (approximately 5 weeks) resulted in no adverse effects on viability, growth, behavior, sexual maturation or reproductive capacity in the offspring.

Pregnancy

Nursing Mothers

It is not known whether cidofovir is excreted in human milk. Since many drugs are excreted in human milk and because of the potential for adverse reactions as well as the potential for tumorigenicity shown for cidofovir in animal studies, VISTIDE should not be administered to nursing mothers. The U.S. Public Health Service Centers for Disease Control and Prevention advises HIV-infected women not to breast-feed to avoid postnatal transmission of HIV to a child who may not yet be infected.

Pediatric Use

Safety and effectiveness in children have not been studied. The use of VISTIDE in children with AIDS warrants extreme caution due to the risk of long-term carcinogenicity and reproductive toxicity. Administration of VISTIDE to children should be undertaken only after careful evaluation and only if the potential benefits of treatment outweigh the risks.

Geriatric Use

No studies of the safety or efficacy of VISTIDE in patients over the age of 60 have been conducted. Since elderly individuals frequently have reduced glomerular filtration, particular attention should be paid to assessing renal function before and during VISTIDE administration (see DOSAGE AND ADMINISTRATION).

Reporting of Adverse Reactions

Malignancies or serious adverse reactions that occur in patients who have received VISTIDE should be reported to Gilead in writing to the Director of Clinical Research, Gilead Sciences, Inc., 333 Lakeside Drive, Foster City, CA 94404 or by calling 1-800-GILEAD-5 (445-3235), or to FDA MedWatch 1-800-FDA-1088/fax 1-800-FDA-0178.

Dosage

THE RECOMMENDED DOSAGE, FREQUENCY, OR INFUSION RATE MUST NOT BE EXCEEDED. VISTIDE MUST BE DILUTED IN 100 MILLILITERS 0.9% (NORMAL) SALINE PRIOR TO ADMINISTRATION. TO MINIMIZE POTENTIAL NEPHROTOXICITY, PROBENECID AND INTRAVENOUS SALINE PREHYDRATION MUST BE ADMINISTERED WITH EACH VISTIDE INFUSION.

Induction Treatment

The recommended induction dose of VISTIDE for patients with a serum creatinine of ≤ 1.5 mg/dL, a calculated creatinine clearance > 55 mL/min, and a urine protein < 100 mg/dL (equivalent to < 2+ proteinuria) is 5 mg/kg body weight (given as an intravenous infusion at a constant rate over 1 hr) administered once weekly for two consecutive weeks. Because serum creatinine in patients with advanced AIDS and CMV retinitis may not provide a complete picture of the patient's underlying renal status, it is important to utilize the Cockcroft-Gault formula to more precisely estimate creatinine clearance (CrCl). As creatinine clearance is dependent on serum creatinine and patient weight, it is necessary to calculate clearance prior to initiation of VISTIDE. CrCl (mL/min) should be calculated according to the following formula:

Maintenance Treatment

The recommended maintenance dose of VISTIDE is 5 mg/kg body weight (given as an intravenous infusion at a constant rate over 1 hr), administered once every 2 weeks.

Dose Adjustment

Probenecid

Probenecid must be administered orally with each VISTIDE dose. Two grams must be administered 3 hr prior to the VISTIDE dose and one gram administered at 2 and again at 8 hr after completion of the 1 hr VISTIDE infusion (for a total of 4 grams).

Ingestion of food prior to each dose of probenecid may reduce drug-related nausea and vomiting. Administration of an antiemetic may reduce the potential for nausea associated with probenecid ingestion. In patients who develop allergic or hypersensitivity symptoms to probenecid, the use of an appropriate prophylactic or therapeutic antihistamine and/or acetaminophen should be considered (see CONTRAINDICATIONS).

Hydration

Patients must receive at least one liter of 0.9% (normal) saline solution intravenously with each infusion of VISTIDE. The saline solution should be infused over a 1–2 hr period immediately before the VISTIDE infusion. Patients who can tolerate the additional fluid load should receive a second liter. If administered, the second liter of saline should be initiated either at the start of the VISTIDE infusion or immediately afterwards, and infused over a 1 to 3 hr period.

Method of Preparation and Administration

Inspect vials visually for particulate matter and discoloration prior to administration. If particulate matter or discoloration is observed, the vial should not be used. With a syringe, extract the appropriate volume of VISTIDE from the vial and transfer the dose to an infusion bag containing 100 mL 0.9% (normal) saline solution. Infuse the entire volume intravenously into the patient at a constant rate over a 1 hr period. Use of a standard infusion pump for administration is recommended.

It is recommended that VISTIDE infusion admixtures be administered within 24 hr of preparation and that refrigerator or freezer storage not be used to extend this 24 hr limit.

If admixtures are not intended for immediate use, they may be stored under refrigeration (2–8°C) for no more than 24 hr. Refrigerated admixtures should be allowed to equilibrate to room temperature prior to use.

The chemical stability of VISTIDE admixtures was demonstrated in polyvinyl chloride composition and ethylene/propylene copolymer composition commercial infusion bags, and in glass bottles. No data are available to support the addition of other drugs or supplements to the cidofovir admixture for concurrent administration.

VISTIDE is supplied in single-use vials. Partially used vials should be discarded (see Handling and Disposal).

Compatibility with Ringer's solution, Lactated Ringer's solution or bacteriostatic infusion fluids has not been evaluated.

Handling and Disposal

Due to the mutagenic properties of cidofovir, adequate precautions including the use of appropriate safety equipment are recommended for the preparation, administration, and disposal of VISTIDE. The National Institutes of Health presently recommends that such agents be prepared in a Class II laminar flow biological safety cabinet and that personnel preparing drugs of this class wear surgical gloves and a closed front surgical-type gown with knit cuffs. If VISTIDE contacts the skin, wash membranes and flush thoroughly with water. Excess VISTIDE and all other materials used in the admixture preparation and administration should be placed in a leak-proof, puncture-proof container. The recommended method of disposal is high temperature incineration.

Patient Monitoring

Serum creatinine and urine protein must be monitored within 48 hours prior to each dose. White blood cell counts with differential should be monitored prior to each dose. In patients with proteinuria, intravenous hydration should be administered and the test repeated. Intraocular pressure, visual acuity and ocular symptoms should be monitored periodically.

VISTIDE should be stored at controlled room temperature 20–25 °C (68–77 °F).