Ifosfamide nephrotoxicity

INTRODUCTION — Ifosfamide is a synthetic structural isomer of cyclophosphamide that has been approved for concurrent use with other drugs in the treatment of metastatic germ-cell testicular cancer and some (especially pediatric) sarcomas.

Nephrotoxicity due to direct tubular injury is a prominent complication of ifosfamide therapy; glomerular toxicity may also occur. Issues related to ifosfamide nephrotoxicity will be reviewed here. The nephrotoxicity of other chemotherapeutic agents is discussed separately:

●(See "Nephrotoxicity of chemotherapy and other cytotoxic agents".)

●(See "Cisplatin nephrotoxicity".)

●(See "Therapeutic use and toxicity of high-dose methotrexate", section on 'Renal toxicity'.)

PATHOGENESIS — Many in vitro studies suggest that the metabolite chloroacetaldehyde is directly toxic to the tubular cells, rather than the parent drug or another metabolite acrolein [1-4]. This relationship could explain why cyclophosphamide, although structurally similar to ifosfamide, has virtually no nephrotoxicity. At equivalent doses, the rate of chloroacetaldehyde generation with ifosfamide is much greater than with cyclophosphamide, although there is considerable variability from patient to patient in the amount of chloroacetaldehyde produced after ifosfamide administration [1]. Other toxic metabolites specific to ifosfamide but not cyclophosphamide, such as isophosphoramide mustard, may contribute to nephrotoxicity, although in vitro studies have given conflicting results [5]. On the other hand, acrolein is believed to be the major cause of the bladder toxicity (hemorrhagic cystitis) seen with both ifosfamide and cyclophosphamide.

The mechanism of injury remains unproven but may be cellular oxidative stress, leading to energy depletion via mitochondrial damage, a mechanism similar to that observed with the mitochondrial cytopathies [4,6]. Disruption of cell membrane function is another possible mechanism of toxicity.

Mesna, which is used concurrently with ifosfamide to prevent bladder toxicity, might also be expected to reduce kidney damage due to chloroacetaldehyde or other nephrotoxic metabolites of ifosfamide. However, mesna has a complex intrarenal pharmacology and association with intratubular glutathione. Thus, it has not been possible to demonstrate a clear role for mesna in prevention of ifosfamide nephrotoxicity [1].

CLINICAL MANIFESTATIONS — Ifosfamide-induced kidney injury is primarily manifested by the onset of one or more of the following signs of tubular dysfunction [1,7-12]:

●Hypophosphatemia induced by decreased proximal tubular phosphate reabsorption; when prolonged, this can lead to rickets in children and osteomalacia in adults [7-11,13].

●More generalized impairment of proximal tubular function (Fanconi syndrome) as manifested by renal glucosuria, bicarbonaturia, aminoaciduria, tubular proteinuria (demonstrated by a marked increase in urinary excretion of low-molecular-weight proteins such as beta-2-microglobulin, but not of albumin), and potassium wasting (which may cause severe hypokalemia) [1,7-9,11-13].

●Metabolic acidosis with a normal anion gap (hyperchloremic) acidosis due to type 1 (distal) or type 2 (proximal) renal tubular acidosis [13].

●Polyuria due to nephrogenic diabetes insipidus (ie, resistance to antidiuretic hormone), which appears to be relatively rare [8,10,13,14]; when polyuria does occur, it is more often an appropriate response to isotonic saline hydration therapy given during chemotherapy administration. (See 'Etiology of polyuria' below.)

The time of onset of tubular dysfunction is variable. In a report of 27 patients with manifestations of tubular dysfunction, the diagnosis was made during ifosfamide therapy in 9 patients and after the cessation of therapy in 18 patients at a median of 13 months [7]. In addition to tubular dysfunction, ifosfamide therapy can lead to a reduction in glomerular filtration rate (GFR) [8,9]. In most patients, the decline in GFR is mild (usually no more than 20 to 30 percent below the baseline level) unless ifosfamide is given in combination with another nephrotoxin such as cisplatin [8,9,13]. Ifosfamide treatment has also been reported to lead to albuminuria in long-term childhood cancer survivors [15]. (See 'Risk factors' below and 'Long-term prognosis' below.)

The clinical manifestations of ifosfamide nephrotoxicity are different from the major manifestations of nephrotoxicity induced by cisplatin. Cisplatin primarily affects the S3 segment of the proximal tubule and the distal nephron, leading to acute kidney injury and hypomagnesemia due to urinary magnesium wasting. Glucosuria and aminoaciduria can occasionally occur. (See "Cisplatin nephrotoxicity".)

Etiology of polyuria — As noted in the preceding section, ifosfamide can induce nephrogenic diabetes insipidus. However, it is an uncommon complication that typically occurs in patients who have other signs of tubular toxicity, such as glucosuria, aminoaciduria, and hypophosphatemia [14].

A more frequent problem is the marked polyuria that may complicate the administration of high-dose ifosfamide-based chemotherapy protocols. The cause of the polyuria in almost all such cases is a sodium diuresis (not a dilute water diuresis as in diabetes insipidus) that results from the volume expansion induced by saline therapy given to maintain a high urine output during chemotherapy administration, particularly if ifosfamide, cisplatin, or carboplatin is part of the regimen. Thus, the diuresis is appropriate and resolves if sodium intake is reduced and excretion of the excess fluid is permitted. In addition, the urine osmolality tends to be similar to that of plasma, not dilute as in diabetes insipidus. (See "Evaluation of patients with polyuria", section on 'Solute (osmotic) diuresis'.)

A confounding factor is the presence of orthostatic (postural) hypotension. Although this finding may suggest possible volume depletion, it may also be due to chemotherapy-induced autonomic neuropathy. Volume expansion may minimize orthostasis by maximizing the intravascular volume, but it also leads to persistence of the polyuria.

RISK FACTORS — The risk factors for ifosfamide nephrotoxicity are cumulative dose, age under four to five years, and prior or concomitant cisplatin therapy. Prior nephrectomy, presumably as a consequence of reduced renal excretory capacity, has also been shown to increase the risk of ifosfamide nephrotoxicity in children with cancer.

Cumulative ifosfamide dose — Clinically significant nephrotoxicity has been described in children, mostly at a total dose above 60 g/m² [7-9,11,13,16], with severe nephrotoxicity primarily occurring at a total dose of 120 g/m² or more [8]. The signs of nephrotoxicity are initially transient but become persistent in some patients as the cumulative dose increases.

The high frequency of transient nephrotoxicity was demonstrated in a prospective study of 23 children and young adults who were treated with high-dose ifosfamide (9 g/m² given over five days for 11 cycles at three- to six-week intervals) [11]. Acute tubular dysfunction occurred in all patients and was characterized by increases in the excretion of beta-2-microglobulin, glucose, phosphate, and amino acids. These abnormalities generally disappeared by day 0 of the next cycle. Only three patients (13 percent) developed persistent Fanconi-like syndrome or requirement for oral electrolyte (potassium and phosphate) supplementation. The cumulative dose was 70 g/m² in one and over 90 g/m² in the other two.

The differences in risk of persistent nephrotoxicity according to cumulative dose were illustrated in a series of 76 children and young adults in which grading criteria were used to define the severity of ifosfamide nephrotoxicity [8]. The criteria included the glomerular filtration rate (GFR), the renal threshold for phosphate excretion (to evaluate for phosphaturia), the serum bicarbonate concentration (to evaluate for metabolic acidosis), and the early morning osmolality (to evaluate urinary concentrating ability).

The following findings with respect to moderate and severe nephrotoxicity were noted at a median follow-up of six months:

●At a cumulative ifosfamide dose below 84 g/m², there were no cases of severe nephrotoxicity, and 5 of 45 (11 percent) had moderate nephrotoxicity.

●At a cumulative ifosfamide dose 84 to 119 g/m², 1 of 16 had severe nephrotoxicity, and 4 of 16 had moderate nephrotoxicity (6 and 25 percent, respectively).

●At a cumulative ifosfamide dose 120 g/m² or more, 5 of 15 had severe nephrotoxicity, and 6 of 15 had moderate nephrotoxicity (33 and 40 percent, respectively).

Three large cohort studies have confirmed an association between higher cumulative ifosfamide doses and an increased risk of glomerular impairment, measured by reduced estimated GFR [15,17] and by an increased risk of stage 3 to 5 chronic kidney disease [18].

Lower doses are safer but may occasionally be associated with persistent tubular injury. At a median follow-up of 19 months in a series of 593 children and adolescents with sarcoma who were treated with ifosfamide, the incidence of tubulopathy (defined as at least two of hypophosphatemia, glucosuria, and proteinuria on at least two examinations) was 0.4 and 6.5 percent in those exposed to an ifosfamide dose of ≤24 g/m² and between 24 to 60 g/m², respectively [7].

Young children — Children under four years of age at the time of treatment may be at increased risk of ifosfamide nephrotoxicity [7,16]. In a multivariate analysis of the above study of 593 children and adolescents, children who were less than four years of age at diagnosis had an 8.7-fold higher rate of tubulopathy, compared with an 18.6 increased risk with ifosfamide doses ≥60 g/m² [7]. A possible mechanism is increased activity in young children of CYP3A, the enzyme believed to be responsible for the generation of chloroacetaldehyde, which appears to mediate ifosfamide nephrotoxicity [19]. However, some very long-term studies have failed to show a relationship between young age and a higher risk of persistent nephrotoxicity [9,20]. (See 'Pathogenesis' above.)

Prior nephrectomy — A study of 120 children, 10 of whom had previously undergone nephrectomy as part of their anti-cancer treatment, found that nephrectomized children were much more likely to develop ifosfamide nephrotoxicity [21]. Subclinical nephrotoxicity (defined by phosphaturia and aminoaciduria) and clinically evident nephrotoxicity (Fanconi syndrome) each occurred in three patients (30 percent). In contrast, in the remaining 110 children, subclinical and clinically evident nephrotoxicity occurred in 4 and 12 patients, respectively (4 and 11 percent). As suggested above, this is presumably due to reduced renal excretory capacity caused by previous nephrectomy. Although not investigated specifically in the published literature, it is likely that children with other causes of impaired kidney function will also have a higher risk of developing nephrotoxicity after ifosfamide treatment.

Prior or concomitant cisplatin therapy — Concurrent administration of ifosfamide and cisplatin may have additive nephrotoxicity [16,21-23], and ifosfamide nephrotoxicity can occur at lower cumulative doses in patients with appreciable cisplatin exposure [22]. This potentiating effect may not occur in patients treated with low doses of ifosfamide (≤24 g/m²) [7]. (See "Cisplatin nephrotoxicity".)

Carboplatin is a less toxic analog of cisplatin, and it is not clear if concurrent carboplatin therapy is a risk factor for ifosfamide nephrotoxicity [7]. (See "Cisplatin nephrotoxicity", section on 'Use of cisplatin analogs'.)

PREVENTION — The major method to prevent ifosfamide nephrotoxicity is to limit the cumulative dose. Other preventative strategies, such as coadministration of mesna or N-acetylcysteine (NAC), have been evaluated, but efficacy is unproven.

Limit ifosfamide dose — As noted above, the risk of nephrotoxicity is low at cumulative ifosfamide doses of 60 g/m² or less and, when toxicity does occur, it is usually mild to moderate in severity and more likely to be transient [8,9,16,20]. Limiting the cumulative ifosfamide dose may be especially important in patients with additional risk factors for toxicity, such as young children (age under four years), those who have undergone prior nephrectomy or have other causes of kidney dysfunction, and those treated with cisplatin. (See 'Risk factors' above.)

Therapies of unproven efficacy

Mesna — Mesna is a synthetic sulfhydryl compound that can detoxify ifosfamide metabolites. The main clinical benefit of mesna is to prevent the accumulation of the metabolite acrolein, which causes hemorrhagic cystitis. As a result, all patients treated with ifosfamide in modern chemotherapy regimens also receive mesna. (See "Chemotherapy and radiation-related hemorrhagic cystitis in cancer patients", section on 'Mesna'.)

Whether the coadministration of mesna also reduces kidney dysfunction remains unproven [1,24]. In experimental studies, prior in vivo mesna administration did not prevent in vitro nephrotoxicity induced by chloroacetaldehyde, the metabolite that appears to be responsible for ifosfamide nephrotoxicity [24]. It is unclear whether alternative mesna administration schedules are more likely to prevent ifosfamide nephrotoxicity. (See 'Pathogenesis' above.)

N-acetylcysteine — N-acetylcysteine (NAC), which is used in the treatment of acetaminophen (paracetamol) intoxication, has been effective in an animal model for the prevention of ifosfamide nephrotoxicity [25,26]. Clinical studies in humans have not been reported.

LONG-TERM PROGNOSIS — The long-term impact of ifosfamide-induced kidney injury in children has been addressed in a small number of studies [7-9,13,20,27]. The best long-term data following moderate ifosfamide exposure were provided in a cross-sectional study of 183 childhood cancer survivors who had been treated with a median cumulative ifosfamide dose of 54 g/m², with only 17 receiving more than 60 g/m²; none of the patients received concurrent therapy with cisplatin or carboplatin [9].

At an average follow-up of 10 years, 90 percent had normal tubular function, and 79 percent had a normal estimated glomerular filtration rate (eGFR), defined as greater than 90 mL/min per 1.73 m². Almost all of the abnormal findings were mild:

●All but one of the 39 patients with a reduced eGFR had a value between 60 and 89 mL/min per 1.73 m²; one patient had an eGFR of 50 mL/min per 1.73 m²

●Two patients had mild hypomagnesemia

●Two patients had hypophosphatemia (although 24 percent had a reduced threshold for phosphate excretion)

●One patient had a low serum bicarbonate concentration (19 mEq/L)

●Significant renal glucosuria was present in eight patients

Using a combined nephrotoxicity score in 81 evaluable patients, 63 percent had no nephrotoxicity, 36 percent had mild nephrotoxicity, and one percent had moderate nephrotoxicity. No patients had severe nephrotoxicity. No patient required long-term treatment with bicarbonate, phosphate, calcium, or vitamin D. In multivariate analysis, the length of follow-up and older age at diagnosis were independent predictors of a reduction in GFR. Thus, the authors concluded that ongoing evaluation for progressive nephrotoxicity is warranted.

The risk of moderate to severe chronic nephrotoxicity is higher at cumulative doses ≥84 g/m². A smaller study of 25 children with cancer evaluated serially up until 10 years after treatment with a median cumulative dose of 106 g/m² showed persistence of glomerular impairment, with 13 percent having a GFR <60 mL/min/1.73m² at 10 years, but resolution of clinically significant tubular toxicity in all affected patients within the same time frame [20]. (See 'Cumulative ifosfamide dose' above.)

There are few data on long-term kidney function in adults who have received ifosfamide. The best data come from a retrospective cohort of 259 adults who received ifosfamide chemotherapy [28]. The mean estimated GFR fell from 82 to 67 mL/min per 1.73 m² after five years; most of this reduction occurred during the course of chemotherapy (likely reflecting acute kidney injury), although kidney function continued to decline thereafter. No patient developed end-stage kidney disease (ESKD).

SUMMARY AND RECOMMENDATIONS

●General principles – Ifosfamide is a synthetic structural isomer of cyclophosphamide that is used to treat metastatic germ-cell testicular cancer and some (especially pediatric) sarcomas. Nephrotoxicity due to direct tubular injury is a prominent complication of ifosfamide therapy. (See 'Introduction' above.)

●Clinical manifestations – Kidney injury is manifested by signs of tubular dysfunction including glucosuria, aminoaciduria, tubular proteinuria (ie, increased urine excretion of low-molecular-weight proteins such as beta-2-microglobulin, but not of albumin), and rarely, polyuria due to nephrogenic diabetes insipidus. With repeated treatment and hence higher cumulative doses, patients may develop hypophosphatemia due to decreased proximal phosphate reabsorption, hypokalemia due to potassium wasting, and a normal anion gap hyperchloremic metabolic acidosis.

In addition to tubular dysfunction, ifosfamide can lead to a reduction in glomerular filtration rate (GFR) and to albuminuria (reflecting glomerular proteinuria). The reduction in GFR is generally mild unless ifosfamide is given in combination with another nephrotoxin such as cisplatin. (See 'Clinical manifestations' above.)

●Risk factors – Risk factors for ifosfamide nephrotoxicity include higher cumulative dose and, to a lesser extent, younger age at treatment, prior nephrectomy (and probably other causes of kidney dysfunction), and prior or concomitant cisplatin therapy. Clinically significant nephrotoxicity is more likely to occur at a total dose above 60 g/m² and severe nephrotoxicity above 120 g/m². (See 'Risk factors' above.)

●Prevention – The cornerstone of prevention of ifosfamide nephrotoxicity is to limit the cumulative dose. Other modalities, such as mesna and N-acetylcysteine (NAC), have not been proven to be effective. (See 'Prevention' above.)

●Prognosis – Limited studies suggest that the long-term prognosis of moderate ifosfamide nephrotoxicity is good with the majority of patients recovering normal tubular function and maintaining normal or only mildly reduced GFR during the first 10 years after completing treatment. (See 'Long-term prognosis' above.)