Tenofovir-related Kidney Toxicity Linked to High Drug Concentrations, May Not Always Be Reversible

Tenofovir-related Kidney Toxicity Linked to High Drug Concentrations, May Not Always Be Reversible

SUMMARY: Three recently published studies shed further light on kidney (renal) toxicity associated with tenofovir (Viread, also in the Truvada and Atripla coformulations), which is approved for treatment of both HIV and hepatitis B. According to a research letter in the April 24, 2010 issue of AIDS, kidney function abnormalities may result from higher than expected drug concentrations. A related study found that kidney function may not return to normal after discontinuing tenofovir, while a third report indicated that kidney-related biomarker changes were observed in a small number of children with prolonged tenofovir use.

By Liz Highleyman

The issue of kidney toxicity caused by tenofovir remains controversial due to conflicting study data. Kidney impairment was not observed in pivotal trials that led to the drug's approval, but those studies excluded people with pre-existing kidney damage. Several real-world clinical studies have found an increased likelihood of kidney toxicity in a small proportion (typically around 1%-2%) of people taking tenofovir, especially those with other risk factors such as older age and high blood pressure; other studies, however, have not seen this association.

Diagram showing the basic physiologic mechanisms of the kidney (Source: Wikipedia)

Tenofovir Concentration

Sonia Rodriguez-Novoa from Hospital Carlos III in Madrid, Spain, and colleagues looked at the relationship between tenofovir exposure and kidney tubular dysfunction.

The kidneys perform 2 primary functions; the glomerulus filters out various substances from the blood, while the tubules reabsorb water. Tenofovir is processed and excreted in the kidneys by a combination of glomerular filtration and active tubular secretion. Tenofovir seldom affects glomerular function (as indicated by glomerular filtration rate), the study authors noted as background, but abnormalities in kidney tubular function appear to be more common.

Variations or polymorphisms in genes encoding transporter proteins involved in tenofovir elimination might help explain individual differences in kidney toxicity, they added, since slowed elimination might lead to drug overexposure in the kidneys.

The researchers first conducted a larger study of kidney tubule function, determined by 24-hour urine monitoring, among 284 HIV patients; 154 were on tenofovir-containing antiretroviral therapy (ART), 49 were on ART without tenofovir, and 81 were treatment-naive. They found that significantly more participants taking tenofovir exhibited kidney tubular dysfunction compared with the other 2 groups (22%, 6%, and 12%, respectively; P < 0.05).

In the present sub-study, the investigators prospectively analyzed 92 tenofovir recipients. Kidney tubular dysfunction was determined on the basis of glucose, amino acids/protein, and beta-2-microglobulin in the urine, altered reabsorption of phosphorus, and abnormal uric acid excretion (any 2 manifestations). The median duration of tenofovir exposure was 33 months.

Results

	19% of participants met criteria for kidney tubular dysfunction, while 81% had normal tubular function.
	The 2 groups were well-matched with regard to sex, age, weight, high blood pressure, hepatitis C coinfection, length of tenofovir exposure, and use of kidney-toxic drugs; however, those with kidney dysfunction were more likely to have diabetes.
	Creatinine clearance (an indicator of glomerular function) was normal and comparable between the 2 groups.
	The median tenofovir plasma trough (lowest between doses) concentration was significantly higher in patients with kidney tubular dysfunction compared to those with normal function (182 vs 106 ng/mL, respectively; P = 0.001).
	A tenofovir plasma concentration of 160 ng/mL was the best threshold for kidney tubular dysfunction risk, with 61% sensitivity and 80% specificity.
	In a multivariate analysis including all relevant risk factors, only tenofovir plasma levels > 160 ng/mL were associated with kidney tubular dysfunction (odds ratio [OR] 4.8, or nearly 5-fold higher risk).
	Multivariate analysis also showed that female sex (OR 71) and the ratio of body weight to plasma creatinine (OR -1.19) were independently associated with higher tenofovir plasma levels.

"This dose-dependent effect further supports an involvement of tenofovir in tubular dysfunction," the researchers concluded.

"If tubular damage persists for long periods in the absence of renal insufficiency, laboratory and clinical manifestations other than those associated with kidney failure may develop ([e.g.,] premature osteoporosis due to bone mineral loss) under long-term tenofovir therapy," they elaborated.

"The mechanism of kidney tubular damage by tenofovir is probably related [to] its renal clearance, supporting that higher tenofovir plasma levels may lead directly to a greater accumulation of tenofovir in the renal tubular cells and, consequently, to kidney toxicity," they continued.

This concentration-dependent effect is important, they suggested, because tenofovir dose reduction might be an option, especially for people with chronic hepatitis B.

Incomplete Reversibility

In the second study, published in the February 19, 2010 advance online edition of the Journal of Acquired Immune Deficiency Syndromes, Karen Wever and Andrew Carr from St. Vincent's Hospital in Sydney and colleagues assessed the reversibility of tenofovir-related kidney toxicity.

Some studies have shown that kidney toxicity resolves fairly quickly after stopping tenofovir, but this has often been based on short-term follow-up or looked only at creatinine clearance, a less sensitive measure of kidney function.

This study included 24 HIV positive men who discontinued tenofovir due to kidney function impairment, defined as an estimated glomerular filtration rate (eGFR) below 90 using the Modified Diet in Renal Disease (MDRD) equation; lower numbers (in mL/min/1.73 m) indicate poorer glomerular function, and 60 is the usual threshold for chronic kidney disease.

The median duration of tenofovir use was 30 months. The investigators retrospectively analyzed changes in eGFR during a median follow-up period of 13 months after stopping tenofovir.

Results

	Most participants had some degree of kidney impairment before they started tenofovir, with a median eGFR of 74.
	At the time of tenofovir discontinuation, the median eGFR had fallen to 51.
	Median eGRF increased to 58 during follow-up, but did not return to the pre-tenofovir level.
	Only 10 patients (42%) returned to their pre-tenofovir eGFR.
	Results were similar using an alternative estimation method, the Cockcroft-Gault equation.
	Results were also similar after excluding patients who had shorter follow-up after stopping tenofovir.

"In this population, tenofovir-related renal toxicity was not always fully reversible," the study authors concluded.

"Greater eGFR improvement was significantly associated with more rapid decline in eGFR on tenofovir therapy," while a slower decrease over time predicted persistent impairment, they noted. Improvement was also more likely among people who took tenofovir with a protease inhibitor (as opposed to a NNRTI), with a trend for shorter duration of tenofovir exposure.

Based on these findings, the researchers suggested that a decline in eGFR -- even if gradual and not falling below 60 -- "may merit discontinuation of tenofovir to avoid permanent renal dysfunction."

Kidney Function in Children

Few studies have looked at kidney function impairment in HIV positive children.

As reported in the February 20, 2010 issue of AIDS, Ali Judd from the Medical Research Council Clinical Trials Unit in London and colleagues investigated the link between tenofovir use and abnormal kidney function in a large cohort of children on ART.

This nested case-control study included 456 ART-exposed children (age 2-18) in the Collaborative HIV Paediatric Study, which represents about 95% of HIV positive children in the UK and Ireland.

The researchers analyzed serum (but not urine) biochemistry data collected during 2002-2008 at 7 hospitals. Case patients had either confirmed hypophosphatemia (elevated phosphate) or an eGFR less than 60. Each case patient was matched with 3 control children seen at the same hospital.

Results

	20 of 456 children (4.4%) had hypophosphatemia and 1 had eGFR less than 60.
	10 of 20 case patients (50%) had taken tenofovir-containing ART for a median of 18 months as part of second-line or salvage therapy, compared with 11 of 60 control subjects (18%).
	The incidence of hypophosphatemia was 4.3 per 100 person-years among tenofovir recipients, compared with 0.9 per 100 person-years among those not exposed to tenofovir.
	In a multivariable analysis, only tenofovir exposure during the previous 6 months was associated with hypophosphatemia (odds ratio 4.81; P = 0.01).
	Among 6 of 10 children with hypophosphatemia who had at least 4 subsequent measurements, phosphate values returned to normal after tenofovir was discontinued.
	Among 4 children with 3 or fewer subsequent measurements, phosphate values rose but remained below normal.

"Hypophosphatemia was uncommon (4%), but was associated with prolonged tenofovir use, and was generally reversible following tenofovir withdrawal," the study authors concluded.

These findings "highlight the importance of continuing to monitor longer-term renal function, in particular tubular function, especially in those taking tenofovir," they added. "Further studies assessing urine biochemistry measures, which more accurately indicate renal tubular damage, are required."

Investigator affiliations: