Nephrotoxicity Related to New Therapeutic Compounds

Abstract

Toxic nephropathy is an important cause of reversible renal injury. This article focuses on the nephrotoxicity of several new therapeutic compounds. Selective cyclooxygenase-2 inhibitor is associated with sodium retention, hypertension, ankle edema, and acute renal failure. The incidence of renal complication is similar to conventional nonsteroidal anti-inflammatory drugs. Bisphosphonates, especially when used in high dose for prolonged duration, can cause toxic acute tubular necrosis and renal failure. Pamidronate is also associated with a specific form of collapsing focal segmental glomerulosclerosis similar to one found in patients with human immunodeficiency virus (HIV) infection. Acyclic nucleoside phosphonate, a new group of antiviral agents, can cause Fanconi-like syndrome and progressive renal impairment. On the other hand, indinavir, a potent protease inhibitor for the treatment of HIV infection, can cause crystalluria, renal stone, acute tubular obstruction and chronic interstitial nephritis. Intravenous immune globulin and hydroxyethyl starch, a new plasma expander, are associated with acute renal failure with characteristic renal histology known as osmotic nephrosis. In short, physicians should be cautious about possible renal toxicity during the use of any new therapeutic agents.

• Acute renal failure
• NSAID
• Pamidronate

INTRODUCTION

Drug-related nephrotoxicity is a common cause of acute renal failure in clinical practice. As many as one-third of all the cases of acute renal failure are caused by toxin-related acute tubular necrosis, and another 10% of cases by drug-induced interstitial nephritis.[1] This article focuses on the nephrotoxicity of several new therapeutic compounds.

NEPHROPATHY RELATED TO SELECTIVE CYCLOOXYGENASE-2 INHIBITOR

The therapeutic action and toxicity of nonsteroidal anti-inflammatory drugs (NSAIDs) are based on the inhibition of the enzyme cyclooxygenase (COX) that catalyzes the committed step in the synthesis of prostaglandins.[2] Cyclooxygenase exists as two unique isoforms (that is, COX-1 and COX-2), encoded by genes on chromosomes 9 and 1, respectively. The antagonism of inflammation and pain by NSAIDs is believed to result from the inhibition of COX-2, while the antagonism of gastric mucosal defense and platelet aggregation are the results of COX-1 inhibition.[2] Recent clinical trials established selective COX-2 inhibitors as effective analgesic and anti-inflammatory agents with a lower incidence of major gastrointestinal side effects as compared to conventional NSAIDs.[2]

Now it is known that the biological role of COX-2 extends beyond inflammation and pain. The intrarenal distribution of COX-2 in rats, dogs, and elderly humans includes the macula densa, thick ascending limbs of loop of Henle, and papillary interstitial cells.[2] The COX-2 in normal human kidney may enhance sodium excretion, regulate renin release, and mediate tubulo-glomerular feedback.[2]

While the renal problems associated with the use of NSAIDs are well known, the renal effect of selective COX-2 inhibitors are increasingly recognized (Table 1). Recently, Whelton et al.[8] reviewed 50 studies of 13,000 patients treated with celecoxib for an average of 12 weeks. The incidence of adverse renal events was 4.3% (in comparison, the incidence of adverse renal event involving NSAIDs and placebo were 4.1% and 2.5%, respectively). The profile of adverse renal events was alsosimilar between celecoxib and other NSAIDs, with edema in 2.1% cases, new-onset hypertension in 0.8%, and worsened preexisting hypertension in 0.6%.[8] In 5000 patients treated for up to 2 years, the incidence of adverse renal events was as high as 14.4%.[8]

Table 1. Studies with Different Cyclooxygenase-2 Inhibitors.

Reference	Subjects	Drugs	Duration	Results
Rossat[3]	healthy salt-depleted (n=40)	celecoxib vs. naprosyn, placebo	7 days	Na retention as naprosyn; transient reduction in GFR and RBF in high dose
Whelton[4]	healthy elderly (n=29)	celecoxib vs. naprosyn	5 days	reduced urine Na as naprosyn; reduce GFR
Whelton[5]	CKD patients (n=75)	celecoxib vs. naprosyn, placebo	5 days	no change in GFR
Castella-Lawson[6]	elderly on high-salt diet (n=36)	rofecoxib vs. indomethacin, placebo	2 weeks	reduced urine Na as indomethacin; no change in GFR
Swan[7]	elderly on low-salt diet (n=75)	rofecoxib vs. indomethacin, placebo	5 days	reduced urine Na as indomethacin; reduce GFR

Na, sodium; GFR, glomerular filtration rate; RBF, renal blood flow.

In a prospective study, Chan et al.[9] examined 287 patients (mostly with osteoarthritis) who had a history of Helicobacter-related gastrointestinal bleeding. The patients were randomized to receive either celecoxib or diclofenac (plus omeprazole) and followed for 6 months. In this study, adverse renal event developed in 24.3% and 30.8% of patients receiving celecoxib and diclofenac, respectively.[9] Renal failure (defined as a progressive rise in serum creatinine to above 200 µmol/L) and hypertension developed in 5.6% and 13.6% patients, respectively, of the celecoxib group, both similar to the diclofenac group.[9]

In short, the body of clinical evidence strongly suggests that selective COX-2 inhibitors are not “renal-safe.” Nevertheless, because selective COX-2 inhibitors have been on the market for less than a decade, their long-term effects on the kidney remain to be defined. It is interesting to see whether the chronic use of selective COX-2 inhibitors, alone or with other analgesics, gives rise to analgesic nephropathy, and whether the chronic use of selective COX-2 inhibitors affects the progression of chronic kidney disease in general.

BISPHOSPHONATE-RELATED NEPHROPATHY

Bisphosphonate binds phosphate crystals within bone matrix, resulting in disruption of osteoclastic activity. It is commonly used for malignancy-related hypercalcemia, osteolytic metastasis, and Paget's disease. Markowitz et al.[10] recently reported seven patients, all human immunodeficiency virus (HIV) negative, who developed collapsing focal segmental glomerulosclerosis (FSGS) during pamidronate treatment of malignancy (mostly multiple myeloma). Renal biopsy specimens typically show glomeruli collapse of the glomerular basement membranes, hyperplasia of visceral epithelial cells, and severe tubulointerstitial damage, including tubular microcyst formation.[10] There is characteristically global reduction and segmental loss of podocyte staining for synaptopodin. In this series, it was noted that at the recommended dose of 90 mg intravenously monthly, renal toxicity was infrequent.[10] However, higher doses for prolonged duration may produce nephrotoxicity.

In addition to collapsing FSGS, renal failure secondary to toxic acute tubular necrosis (ATN) may be seen following exposure to a variety of bisphosphonates.[11] Bisphosphonate-induced ATN occurs in 2% to 10% of patients. Bisphosphonates are actively secreted by renal tubular cells, which probably cause direct tubular toxicity. Classically, renal biopsy specimens show toxic necrosis of tubular cells with no inflammation.[11] In contrast to the toxic ATN caused by other agents, bisphosphonate-induced ATN typically presents with a gradual increase in serum creatinine over months, and the recovery is often slow after stopping the drug. It has been suggested that the risk of ATN may be reduced by infusing the drug slowly (for example, over 30 minutes rather than 10 minutes as recommended).[11]

NEPHROPATHY RELATED TO ACYCLIC NUCLEOSIDE PHOSPHONATES

The acyclic nucleoside phosphonates cidofovir, adefovir, and tenofovir have proved to be effective invitro, in animal models and clinical studies against a wide variety of DNA virus and retrovirus infections.[12] Renal toxicity of adefovir dipivoxil is the most well established.[13] When administered intravenously, 98% of adefovir is excreted unchanged in urine. A substantial proportion of patients develop nephrotoxicity, which limits the dosage, and 35% of all cases require cessation of treatment after one year.[13] Clinically, glycosuria is the earliest feature, followed by Fanconi-like syndrome and progressive renal impairment. It is now established that acyclic nucleoside phosphonates has a direct toxic effect on replication of mitochondrial DNA for the synthesis of cytochrome C oxidase.[13]

The mitochondrial toxicity of acyclic nucleoside phosphonates is particularly prominent in proximal renal tubular cells because of their high metabolic activity and the presence of human renal organic anion transporter-1 (hOATI), which is capable of transporting the acyclic nucleoside phosphonates.[14] Ho et al.[14] found that the acyclic nucleoside phosphonates adefovir and cidofovir were approximately 400- to 500-fold more cytotoxic in Chinese hamster ovary (CHO) cells expressing hOATI as compared to control cell line. More importantly, the cytotoxicity of both drugs in CHO cells expressing hOATI was markedly reduced in the presence of probenecid, and hOATI inhibitor.[14] In addition to adequate hydration, it has been suggested that probenecid prophylaxis before and after acyclic nucleoside phosphonates treatment may reduce the risk of nephropathy.[14] This group of drug should be used with caution in patients with preexisting renal impairment, and dosage adjustment may be necessary.

CRYSTAL-RELATED NEPHROPATHY

Indinavir is a potent protease inhibitor widely used in combination with reverse-transcriptase inhibitors to treat HIV disease. Individuals treated with indinavir are prone to develop urinary complications, including renal colic, renal calculi, lower urinary tract symptoms, and indinavir crystalluria.[15] In general, 10% to 20% of patients receiving indinavir have asymptomatic crystalluria, and 4% have loin pain and hematuria[16] the pathogenesis of indinavir nephropathy is generally attributed to intraluminal formation of indinavir crystal, which causes tubular obstruction, resulting in interstitial inflammation, granuloma formation, and chronic interstitial fibrosis.[16]

Herman et al.[17] recently performed a retrospective cohort study of 781 patients, based on two large HIV centers in London, to determine the risk factors for the development of indinavir-associated renal complications. An overall indinavir nephropathy incidence of 7.3% was identified. Concomitant acyclovir doubled the risk of indinavir nephropathy and we, therefore, recommend careful monitoring when prescribing acyclovir with indinavir. Other important risk factors of indinavir nephropathy are low urine pH, low urine flow rate, and preexisting liver dysfunction.[15]

Many commonly used medications are associated with the production of crystals that are insoluble in human urine.[18] Important examples include acyclovir, foscarnet, sulfonamide, ciprofloxacin, methotrexate, triamterene, primidone, and high dose vitamin C.[18] Intratubular precipitation of these crystals can lead to acute renal insufficiency. Other medications cause crystalluria only (for example, aspirin, xylitol, ampicillin, and cephalexin). In this respect, urine microscopic examination often gives useful diagnostic clues.

OSMOTIC NEPHROSIS

Over the past two decades, intravenous immune globulin (IVIG) therapy has gained widespread use for a variety of clinical disorders. Acute renal failure following IVIG administration was first reported in 1987 and occurs in 1% to 10% of cases.[19] Up to now, around 100 cases of acute, usually transient, IVIG-related renal failure have been reported. Although the cause of IVIG-associated acute renal failure is unknown, it may be related to the stabilizing agent (for example, sucrose) used in the IVIG preparation.[20] The development and resolution of acute renal failure is typically rapid, but is some cases recovery may be delayed and require dialysis. Renal histology typically shows isometric vacuolization of proximal tubular cells.[20] In affected patients, recurrence of acute renal failure may be avoided by selection of a preparation with a different stabilizing agent.[19]

The IVIG-associated nephropathy is the prototype of osmotic nephrosis, which has also been reported after mannitol or dextran therapy. Recently, hydroxyethyl starch is increasingly used for volume restoration in kidney donors and patients with septic shock. Nevertheless, a recent randomized study found that the use of this agent is an independent risk factor for acute renal failure in patients with severe sepsis.[21] At least in isolated cases, renal biopsy specimen demonstrated major osmotic nephrosis-like lesions at the proximal tubular level.[22]

CONCLUSIONS

With the advent of new therapeutic agents, toxin-induced renal injury is increasingly being recognized. Physicians should be cautious about possible renal toxicity during the use of any of the new therapeutic agents. It is hoped that with improving understanding inthe pathophysiology of toxin-induced renal injury, newer agents with less renal toxicity can be developed.

ACKNOWLEDGMENT

This article is based on a talk on the sample title by Dr. C.C. Szeto at the International Society of Nephrology 2004 Conference on the Prevention of Progression of Kidney Disease, Hong Kong, June 2004.