Tenofovir renal safety

The renal safety of tenofovir has been a topic of interest since early reports (involving >30 cases) of significant renal toxicity with renal tubular damage and/or acute renal failure, with evidence of distinct renal pathological changes characterized by prominent proximal tubular injury [17]. In the large Gilead 903 study, in which participants had normal renal function at baseline, no evidence of significant renal toxicity could be identified over a 3-year treatment period using serum creatinine measures [18]. However, subsequent studies analyzed in a recent meta-analysis have demonstrated reproducible but modest reductions in glomerular filtration rate (of approximately 4 mL/min), with no discernible influence of tenofovir therapy on rates of acute or chronic renal injury or chronic renal failure, or of heavy proteinuria (>2 g/day), at the population level [19].

The overall prevalence of renal impairment or significant proteinuria in tenofovir-treated patients appears to have remained low at approximately 3% in both HIV-infected [20] and hepatitis B virus (HBV)-infected [21] populations, and attempts to identify useful clinical risk factors have not proved conclusive. Pre-existing renal impairment is known to be associated with risk of acute renal injury associated with tenofovir, and there is some evidence that alternative or additional explanations for renal impairment are frequent among tenofovir-treated patients with declining renal function [22]. However, current or previous renal impairment does not necessarily preclude the use of tenofovir in patients with limited treatment options [23], as long as renal function is appropriately monitored and proper dose adjustment is employed.

In this regard, simple measurements of serum creatinine levels have a limited capacity for capturing medication effects on renal function, so that the use of calculated methods such as the Cockcroft–Gault or CKD-EPI equation that adjust for body weight and gender are more useful for monitoring purposes. Assessment and monitoring of urinalysis for glycosuria and proteinuria, with quantification of urinary protein excretion (urinary protein/creatinine and microalbumin/creatinine ratio) in those with evidence of proteinuria, is also recommended, in light of the predilection for tenofovir to affect proximal tubular function. There have also been case reports of osteomalacia caused by phosphate wasting due to tenofovir-induced proximal renal tubular dysfunction. In addition to having proteinuria and/or glycosuria, patients with this problem may have hypophosphatemia and/or an elevated fractional excretion of phosphate.

Lactic acidosis, hyperlactatemia, and acute hepatic steatosis

Lactic acidosis is probably the most severe clinical manifestation of mitochondrial dysfunction, in which loss of mitochondrial oxidative function leads to increased reliance on “anaerobic” metabolism and the inevitable accumulation of lactate. In the setting of NRTI therapy, there is now an appreciation of a spectrum of clinical disease associated with elevated systemic lactate levels [24]. At one end of this spectrum is a relatively common syndrome of mild, asymptomatic, nonprogressive hyperlactatemia (generally >2.5 mmol/L), which appears to represent a “compensated” homeostatic system in which elevated lactate production is balanced by effective mechanisms of lactate clearance. While the degree of hyperlactatemia appears to be greater in the presence of stavudine or didanosine therapy than of zidovudine or abacavir, this syndrome appears to be benign irrespective of the choice of NRTI therapy.

An “intermediate” syndrome has also been described, characterized by symptomatic hyperlactatemia or hepatic steatosis without systemic acidosis, which is almost uniformly associated with stavudine therapy (incidence ~13/1,000 person-years). In this setting, lactate levels and symptoms can be controlled following modification of NRTI therapy.

The severe life-threatening hyperlactatemia syndromes, with lactic acidosis and hepatic steatosis, are relatively uncommon (1–2/1,000 person-years). A critical aspect of lactic acidosis is its unpredictability, as it typically occurs in patients who have been on stable NRTI regimens for months or even years, and is not heralded by increased lactate levels before the development of the fulminant syndrome. Host risk factors for NRTI-associated lactic acidosis include concurrent liver disease, female gender, and obesity, and while the majority of reported cases in recent years have involved stavudine-based ART, cases involving zidovudine therapy also occur. The cornerstone of management of this condition is early recognition of the clinical manifestations: abdominal symptoms including nausea, vomiting, anorexia, abdominal pain and distention, fatigue, with biochemical evidence of hepatocellular liver damage and lactate level generally >5 mmol/L. These clinical and laboratory abnormalities should lead to prompt cessation of NRTI therapy. A more recently identified clinical syndrome accompanying lactic acidosis is progressive, severe neuromuscular weakness mimicking the Guillain–Barré syndrome. Overall, the mortality associated with NRTI-associated lactic acidosis remains high (approximately 50%).

It is unfortunate to note that while this treatment complication has become exceedingly rare in developed countries with limited use of stavudine (or zidovudine), there are increasing reports of severe and fatal hyperlactatemia syndromes from resource-limited settings where stavudine is still frequently prescribed [25]. It is hoped that the implementation of updated treatment guidelines, along with increased recognition of the high costs associated with managing this and other drug toxicity syndromes associated with these thymidine analogue NRTI drugs [26], will soon change this situation.

Pancreatitis

The differential diagnosis for pancreatitis occurring in an HIV-infected individual includes direct didanosine and stavudine toxicity, severe PI-induced hypertriglyceridemia, and the effects of other drugs such as pentamidine. A diagnosis of pancreatitis includes clinical symptoms of abdominal pain combined with elevated serum amylase and/or lipase levels, thus excluding cases of isolated hyperamylasemia that have been associated with moderate to severe immune deficiency where the relationship between this biochemical abnormality and pancreatitis is uncertain.

The most comprehensive data concerning risk factors for NRTI-associated pancreatitis comes from the Johns Hopkins AIDS Service cohort (n = 2,613 cases) [27], with supporting data from the ACTG 5025 study. In these analyses, didanosine and stavudine were associated with roughly equivalent risk of pancreatitis, with estimated incidence rates of 0.8 and 1.1 cases per 100 person-years, respectively. Combining these drugs increased the risk approximately twofold, while concurrent didanosine and hydroxyurea use increased the relative risk approximately eightfold, including several fatal cases. More recently, cases of pancreatitis have been reported with concurrent use of didanosine and tenofovir, including with reduced doses of didanosine (250 mg/day), presumably due to the ability of tenofovir to “boost” didanosine effects in vivo [28].

Non-cirrhotic portal hypertension

A relatively rare syndrome (estimated prevalence ~8/10,000 patient-years) [29] of progressive portal hypertension associated with the development of ascites and varices has been noted among antiretroviral-treated patients. This syndrome is characterized by exuberant nodular regenerative hyperplasia and microvascular portal venopathy pathologically, and has been associated with didanosine exposure (odds ratio ~2), either alone or in combination with stavudine [29, 30], although cases have been described in the absence of exposure to these NRTIs [29]. Given the prominent role of venous thrombosis in the hepatic pathology, it is also proposed that additional factors such as enhanced microbial translocation from the gut and prothrombotic conditions may contribute to disease predisposition [31].

Neuropathy

The prevalence and long-term clinical impact of neuropathy is likely to be underestimated, as the symptoms are of gradual onset, and clinical signs of nerve damage are not always assessed in clinical practice. There is a definite contribution of HIV disease per se to the pathogenesis of a form of distal sensory neuropathy that is clinically and electrophysiologically indistinguishable from so-called “toxic neuropathy” from antiretroviral drugs [32]. It is therefore difficult to determine the relative contributions of disease- and drug-associated factors in the syndrome, as these effects are likely to be synergistic. The clinical syndrome common to both HIV-associated and toxic sensory polyneuropathy is dominated by peripheral pain and dysesthesia, with rare motor involvement.

In an international cohort study in which the overall prevalence of symptomatic peripheral neuropathy was approximately 50%, exposure to stavudine (odds ratio, OR 7.7) or didanosine (OR 3.2) were dominant risk factors, along with age >40 years (OR 2.9) [33].

Reversal of established neuropathy appears to be a slow process that is dependent on cessation of the offending NRTI, and there is limited evidence for treatment strategies other than topical capsaicin 8% [34], although several other agents such as gabapentin and tricyclic antidepressants have been used. As with any clinical neuropathy, a search for contributing factors such as diabetes, excessive alcohol consumption, and vitamin deficiencies (e.g. thiamine, B₁₂, and folate) should be undertaken in patients who have neuropathy.