Low risk of serious kidney toxicity on tenofovir, but mild renal dysfunction cause for concern

Five different cohort studies presented this week at the Fifteenth International AIDS Conference in Bangkok have found that the risk of kidney toxicity on Gilead’s nucleotide analogue, tenofovir (Viread), is extremely rare in individuals who initiate the drug with normal renal function. However, a sixth study, using more sensitive measurements than the other five, found that tenofovir is associated with mild renal dysfunction, which may subsequently make the kidneys more susceptible when tenofovir is co-administered with nephrotoxic drugs.

Similar renal profile to d4T

The longest study to report on the renal safety of tenofovir so far is Gilead’s three year, or 144 week, 903 study from Staszewski and colleagues. This was a Phase III, multicentre, double-blind, active-controlled trial in antiretroviral-naïve patients. Each arm consisted of 296 individuals who were randomised to receive either tenofovir or d4T plus 3TC and efavirenz. At baseline, median serum creatinine levels were less than 1.5 mg/dL, serum phosphorus at or greater than 2.2 mg/dL and calculated creatinine clearance was at or greater than 60 mL/min. No significant differences in renal parameters were seen between the tenofovir and d4T arms to 144 weeks and no-one in the tenofovir arm developed serum creatinine levels higher than 2.1 mg/dL (Grade 2), and those that had Grade 2 toxicity did so between baseline and week 48. In contrast, there were two cases of Grade 3 toxicity in the d4T arm.

No patient developed Fanconi’s Syndrome or renal tubulopathy to 144 weeks and no patient discontinued therapy due to renal toxicity in the tenofovir arm, leading the investigators to conclude that the renal safety profile was similar in patients receiving tenofovir or d4T alongside 3TC and efavirenz.

Only 0.2% renal abnormalities

The largest study to report on the safety and efficacy of tenofovir so far comes from the combined data of the European/Australian Viread expanded access programme, reported here by Gallias and colleagues. Data were presented on 5686 adults “with adequate renal function”, receiving open-label tenofovir in combination with other antiretrovirals for a median of just under six months. There were a total of 332 study discontinuations, of which 101 (1.8%) were caused by an adverse event. Of these, only 0.2% were caused by renal abnormalities. Data was available on serum creatinine and phosphate levels in 2821 patients; Grade 3 or 4 elevations in serum creatinine and reductions in serum phosphate were observed in 0.3% and 0.6% of patients, respectively.

Renal toxicity rare

Scott and colleagues reviewed 447 patient charts of individuals in two large Los Angeles HIV clinics who had been on tenofovir for at least twelve weeks and found that Grade 1 and 2 renal toxicity was rare (1.3%) with no clinical symptoms seen. There were no Grade 3 or 4 increases in serum creatinine clearance seen within 24 weeks of starting tenofovir, and five of the six cases Grade 1 or 2 increased creatinine clearance were due to other medical events not associated with tenofovir. Of the 13 patients (2.9%) with baseline creatinine clearance greater than 1.5 mg/dL, six saw a decrease in creatinine clearance whilst on tenofovir, five remained stable, one progressed from Grade 1 to 2 due to ddI-associated lactic acidosis, and, remarkably, one patient with underlying HIV nephropathy improved whilst on tenofovir. Multivariate analysis, adjusting for reason for tenofovir, demographics, baseline CD4 and viral load, found that there was a correlation (no p value given) between detectable viral load at baseline and increased in serum creatinine clearance.

Clinical experience with renal failure

A German retrospective follow-up study from Jaegel-Guedes and colleagues at an HIV clinic in Munich examined patient files for abnormalities in renal function, including clinical events, in 206 patients (16% of whom were female) on tenofovir for a median of 12.5 months. At baseline, one patient had a history of renal impairment; additionally 28% were on potentially nephrotoxic drugs, and another 42% started potentially nephrotoxic drugs during follow-up. In total, renal failure occurred in two out of 25 patients with pre-existing renal dysfunction, who experienced acute renal failure after two and five months on tenofovir, respectively. The first patient’s renal failure was related to co-administration of nephrotoxic drugs. The second patient had a history of renal failure associated with Burkitt's lymphoma and suffered from progressive disease while on chemotherapy.

African Americans at no higher risk

A study of African Americans, who historically have higher rates of renal disease, found no increase in the risk of renal toxicity from tenofovir when compared with AZT, and changes in serum creatinine clearance were similar to results seen in other clinical trials. Lewis and colleagues compared 50 treatment naïve patients receiving tenofovir for 48 weeks or longer with 46 patients receiving AZT at a Houston, Texas HIV clinic. At each time point there were no significant differences in changes from baseline in serum creatinine and creatinine clearance. Nothing higher than Grade 1 elevations in creatinine clearance were seen in the tenofovir group, whereas one patient in the AZT group had a Grade 2 elevation.

Mild renal dysfunction

However, when Stefan Mauss and colleagues from the Centre for HIV and Hepatogastroenterology in Dusseldorf, Germany performed a cross-sectional study in 74 individuals on tenofovir and 84 patients on other antiretrovirals, they found that tenofovir was associated with mild renal dysfunction when measured with sensitive methods.

Levels of creatinine, cystatin and electrolytes were measured in both serum and urine collected over 24 hours. Renal clearance was calculated on the basis of cystatin clearance (normal was defined as >80 ml/min) which is more sensitive than measuring creatinine. The patients on tenofovir had significantly (p<0.01) lower mean cystatin clearance (87±21 ml/min) compared with those not on tenofovir (96±20 ml/min), although it was still within the normal range in all but 28 (38%) individuals on tenofovir and 21 (25%) not on tenofovir (p=0.08).

Additionally, they found that the patients on tenofovir had a higher mean protein content in urine than those not on tenofovir (118±114 mg/d vs. 96±58 mg/d, p<0.05). In total 26 (35%) patients on tenofovir vs. 15 (18%) not on tenofovir experienced tubular proteinuria, defined as excess protein in the urine greater than 130 mg/d with or without mild glomerular damage (p<0.02), which reversed completely in three patients after tenofovir discontinuation. No differences in electrolytes and creatinine were seen and no-one suffered from the more serious Fanconi Syndrome.

In multivariate analysis no association was found between renal function or proteinuria and the time on all antiretrovirals, time on tenofovir or prior use of indinavir, leading the investigators to conclude that tenofovir is associated with mild renal dysfunction when measured with sensitive methods, and this may make the kidneys more vulnerable when nephrotoxic drugs are co-administered.

Nephrotoxic drugs include:

• Amphotericin B
• Tetracycline
• Aciclovir
• Pentamidine
• NSAIDs (including aspirin and ibuprofen)

Since many different nephrotoxic drugs could be taken at the same time as tenofovir, the researchers suggest that more sensitive measurements of renal function should be included in the future design of trials using tenofovir.