Renal toxicity and chemotherapy in children with cancer

Abstract

The clinical use of antineoplastic drugs can be limited by different drug‐induced toxicities. Of these, renal dysfunction may be one of the most troublesome in that it can be cumulative and in general is only partially reversible with the discontinuation of the treatment. Renal toxicity may be manifested as a reduction of the glomerular filtration rate, electrolyte imbalances, or acute renal failure. Careful assessment of renal function has to be performed taking into account that the impairment of renal function is initially silent and only later may be clinically dramatic. When clinically indicated, the reduction or, in cases of severe nephrotoxicity, the suspension of chemotherapy should be considered to avoid the progressive deterioration of the compromised glomerular and/or tubular function.

Introduction

Over recent decades, chemotherapy treatment of childhood malignant tumours has resulted in a significant increase in survival. Due to the use of radiotherapy and polychemotherapy, five‐year survival in affected children and adolescents is significantly improved, and up to 80% of patients can be considered cured 1.

In consideration of a normal life expectation for these patients, it is of crucial importance to reduce the onset of long‐lasting side effects associated with chemotherapy administration 2. Among the most important side effects, concerning the subsequent development of the child and potential morbidity, is the deterioration of kidney function, resulting from the use of nephrotoxic antineoplastic drugs, widely used due to their efficacy in many paediatric tumours.

Several factors appear to increase the patient risk for nephrotoxicity. Innate drug toxicity, host factors such as age or pre‐existing renal damage, anticancer treatments (including surgery and/or radiotherapy), and renal handling of the drugs can enhance the risk for kidney impairment 3. Data from the Childhood Cancer Survivor Study report that severe renal disease (grade ≥ 3) is rare, being present in 0.5–0.8% of long‐term survivors (follow‐up >5 years), but with a significant increased risk when compared with their siblings (0.2%; relative risk: 8.1; 95% CI: 2.9–23.1) 4, 5.

A recent systematic review on early and late renal adverse effects after nephrotoxic treatment in children with cancer based on data from 57 studies reports that the prevalence of renal dysfunctions after treatment with cisplatin, carboplatin, ifosfamide, radiation therapy involving the kidney region and/or nephrectomy ranged from 0% to 84%. This variation is influenced by the heterogeneity of patients' for disease, age and treatment received, but especially by the large variation both in the methods used to investigate the renal function and in the length of follow‐up 6. Data from the UK Renal Registry Report show that 1.9% of renal failures in children were due to malignancy. Furthermore, data on survivors of childhood cancer report that 0.5% had developed renal failure with a relative risk to siblings of 8.9 (95% CI: 2.2–36.6). The incidence rate of nephrotoxicity in such studies has varied widely, with investigations reporting no renal toxicity (e.g., in initial studies of standard dose carboplatin) to those with a very high incidence of renal damage (e.g., up to 100% of hypomagnesaemia in children receiving cisplatin) (Table 1) 7, 8.

Table 1.

Antineoplastic agents' nephrotoxicity

Drug	Incidence rate (%)
• Cisplatin	10–80
• lfosfamide	1.4–30
• Methotrexate	1.8–12
• Nitrosoureas (high dose)	<10
• Carboplatin (high dose)	0–25
Rare
• Actinomycin D
• Anthracyclines
• Cyclophosphamide
• Gemcitabine
• Melphalan
• Vincristine

Indeed, their use can cause changes in the glomerulus and several sections of the tubule, and lead to severe kidney damage. Acute damage can be of such an extent as to require the suspension of chemotherapy treatment; on the other hand, while not representing an immediate threat to survival, chronic administration can be a significant cause of morbidity in adults. About 30% of children treated with ifosfamide can experience severe tubular damage, and 30–60% of those treated with cisplatin develop varying grades of glomerular or tubular damage 7.

Assessment of renal function

Chemotherapy‐related kidney damage can affect several components of the nephron. Until recently, chemotherapy‐related nephrotoxicity was often evaluated by applying World Health Organization recommendations, which included measuring urine nitrogen or plasma creatinine, and the grade of proteinuria 9. However, these evaluations provide an incomplete picture of renal damage, particularly in young children, where plasma creatinine values are a somewhat insensitive indicator of compromised renal function. Since chemotherapy‐related damage can affect all parts of the nephron, it is useful to examine the three major aspects of the excretory apparatus.

Glomerular function is assessed by means of the glomerular filtration rate (GFR), by calculating creatinine clearance using the Schwartz formula. The original formula was based on the alkaline‐picrate method. However, the availability of the isotope dilution mass spectroscopy‐traceable international standard for creatinine calibration prompted Schwartz et al. to revise the formula in 2009 with enzymatic creatinine results 10. The new Schwartz formula was defined as 39.1 × (height [m]/serum creatinine [mg dl⁻¹])^0.516 × (1.8/cystatin C serum concentration [mg l⁻¹]^0.294 × (30/BUN [mg dl⁻¹])^0.169 × (1099 males) × (height of child [m]/1.4)^0.1888. However, the latter can overestimate the extent of glomerular filtration in patients with reduced muscle mass, where it would be more appropriate to assess the clearance of ⁵¹Cr‐labelled ethylenediaminetetraacetic acid (⁵¹Cr‐EDTA). Indeed, the isotope binds to small molecules that are neither filtered nor reabsorbed 11, 12. Considering the difficulty of execution and the costs, this test should only be performed prior to starting the first course of chemotherapy, and in cases with significantly altered clearance values, calculated using the Schwartz formula. However, some authors report a lack of precision with that method. By using different patient demographics other than length or height (as in the Schwartz formulae), such as weight and height or age and gender, the sensitivity was about 60% to accurately identify a 30% relative decrease in GFR 13.

Proximal tubule function is assessed by measuring the concentration of calcium, magnesium, phosphates and urates, and glucose in the blood and in urine: by determining the excretion of low molecular weight proteins in the urine, including α₁‐microglobulin and β₂‐microglobulin 14, 15. Moreover, it is important to verify the presence of urinary bicarbonates in acidotic patients because the proximal renal tubular acidosis caused by nephrotoxic drugs in the context of Fanconi's syndrome originates from the inability to reabsorb bicarbonate normally in the proximal tubule.

Tubular function is estimated by calculating the ratio of the tubular reabsorption of phosphate (TRP) and renal tubular threshold for phosphate (TmPO₄/GFR). TRP (%) is defined as: 1 − (urine phosphorus concentration/urine creatinine concentration) × (serum creatinine concentration/serum phosphorus concentration) × 100. TRP < 85% is defined as TRP dysfunction.

TmPO₄/GFR is the quotient of maximal rate of tubular phosphate reabsorption and the glomerular filtration rate. For children aged 2–15 years, references ranges were 1.15–2.6 mmol l⁻¹ (21–47 mg dl⁻¹) according to http://www.baspath.co.uk/test_directory/tindex/TmPGFR.htm 16.

Distal tubule function is investigated by evaluating the concentration and acidification capacity in a urine sample collected in the morning. Urinary osmolality >600 mOsm l⁻¹ and a urinary pH <5.4 indicate a normal distal tubule concentration and acidification capacity 17. Should the pH be greater than the value indicated, the ability of the kidneys to maintain acid–base equilibrium should be assessed, starting from the evaluation of blood bicarbonates in order to exclude proximal or distal tubular acidosis. In any case, in patients with decreased GFR, tubular function may be more variable due to renal fibrosis. Interstitial fibrosis may be due to the presence of scarred tubules that are fibrosed, and/or the secretion of matrix that fills in between the nephrons 18.

Altered renal calcium and phosphorus metabolism results in changes in bone metabolism that can be assessed by assaying for alkaline phosphatase and bone isoenzyme activities. The increased elimination of certain enzymes of tubular origin can be indicative of kidney damage; indeed, enzymes such as N‐acetyl‐D‐glucosaminidase (NAG), alanine aminopeptidase (AAP), lactate dehydrogenase (LDH), are present at different sites within the tubule; the increased secretion of these enzymes reflects tubular damage, even if it is does not allow identification of the site of damage.

When present, proteinuria should be distinguished as being glomerular or tubular; the former, consisting predominantly of higher molecular weight proteins, is due to increased glomerular permeability, whereby the quantity of protein filtered exceeds the maximum tubular reabsorption capacity. On the other hand, tubular proteinuria can be the consequence of proximal tubule damage, with reduced absorption of low molecular weight proteins from the ultrafiltrate, or increased protein excretion due to tubular cell damage; the percentage of albumin is generally low with microglobulins being particularly present. The albumin/β₂‐M ratio makes it possible to distinguish between the two forms 19.

At the time of diagnosis of neoplastic disease, it would be useful to perform at least one 24 h urine collection in order to determine electrolytes, phosphates, osmolarity and creatinine clearance, besides the plasma levels of electrolytes, creatinine and urine nitrogen, while application of a urine stick may be useful in monitoring kidney function at home and/or in the interval between one cycle and another. Urine collection from paediatric patients presents many challenges. For infants and young children, a special urine collection bag can be adhered to the skin surrounding the urethral area. The bag will need to be changed frequently to collect all of the urine, and each bag will need to be emptied into the special container. Another option is the use of a catheter which can be inserted into the bladder and left there for 24 h to obtain urine.

Since chemotherapy‐related toxicity can lead to altered renal function even months after the suspension of chemotherapy, renal function tests should be repeated periodically. The Dutch (LATER), UK (UKCCSG) and US (COG) guidelines agree that patients receiving ifosfamide [>16 000 mg m⁻²], platinum drugs [>450 mg m⁻²], renal radiotherapy including total body irradiation (TBI) or nephrectomy may be defined as high‐risk patients. In these cases intensive renal screening by measurement urinalysis (for proteinuria), blood pressure measurement, surveillance for both glomerular and tubular impairment, including measurement of serum creatinine, electrolytes, magnesium (for patients receiving a platinum drug), phosphate and bicarbonate (for patients receiving ifosfamide) is recommended 20, 21, 22. The clinical characteristics of ifosfamide‐induced renal injury are proximal tubular wasting of glucose, phosphate, bicarbonate, sodium, potassium, and amino acids; proteinuria; and decreased glomerular filtration rate. These renal effects related to the use of ifosfamide‐based chemotherapy appear to be dose‐dependent. Electrolyte abnormalities such as hypomagnesaemia and hypokalaemia are commonly reported adverse effects of cisplatin and carboplatin, in addition to increased serum creatinine and uraemia. Kidney surgery and radiation therapy may cause renal impairments, such as hyperfiltration or radiation nephropathy, characterized by hypertension, decline in GFR and proteinuria 23. The detection of any changes in these levels requires the execution of additional instrumental assessments, such as renal ultrasound and bone scintigraphy, guided by the findings of the screening investigations and the patient's specific medical circumstances.

Ifosfamide‐induced nephrotoxicity

Ifosfamide, a cyclophosphamide analogue, is an alkylating agent with significant activity with regard to a wide variety of tumours, in both adults and children. The two compounds differ from one another by the position of the 2‐chloroethyl group 24, 25.

Ifosfamide was first introduced into clinical practice in 1970, but the incidence of haemorrhagic cystitis, sometimes severe, with significantly compromised bladder function, rapidly restricted its use. Subsequently, a compound was identified that was capable of detoxifying the metabolites responsible for this complication without altering its anti‐cancer efficacy, 2‐mercaptoethanesulphonate (MESNA) 26. Initial data relating to ifosfamide‐induced nephrotoxicity dates from 1972, when increased plasma creatinine values were reported for the first time in adults who had been administered the drug at high doses 27. The incidence of ifosfamide‐induced chronic renal damage in children varies widely, being included in numerous case series between 1.4% and 30% 28, 29, 30.

The mechanism underlying the toxicity has still not been entirely clarified. The differences in the metabolism of ifosfamide and cyclophosphamide suggests that the nephrotoxicity of the former may be correlated with the production of high quantities of chloroacetaldehyde, which is not found in treatment with cyclophosphamide 31. At the plasma and urinary concentrations present, the metabolite, which is highly toxic for epithelial cells, would form the basis for renal toxicity.

The toxicity can affect any of the nephron segments, individually or in combination, even though the proximal renal tubule appears to be the site that is principally affected by the toxic effects. The extent of renal damage is highly variable, ranging from sub‐clinical forms that can only be detected by thorough diagnostic testing, such as increased urinary AAP excretion, to intermediate forms showing biochemical changes, such as glycosuria, phosphaturia and aminoaciduria, with no clinically obvious consequences, up to clinically significant forms of glomerular and tubular damage that manifest as Fanconi's syndrome, renal tubular acidosis and diabetes insipidus 7. Altered reabsorption of phosphates is one of the more common manifestations of ifosfamide nephrotoxicity and it is indicative of severely compromised tubules 32, 33. In hypophosphataemia patients, the measurement of the ratio TmPO₄/GFR is preferred to other indices of the renal handling of phosphates and it is the most significant parameter, in that it is directly correlated with the risk of compromised bone metabolism. Hypophosphataemic rickets is the most obvious consequence of altered bone metabolism secondary to chronic renal damage, and even though calciuria and hypocalcaemia are infrequent, cases of reduced vitamin D levels are described in active forms (Table 2) 34.

Table 2.

Clinical and laboratory manifestations of ifosfamide‐induced nephrotoxicity

Glomerular toxicity
• Acute and chronic renal failure, that can result in the suspension of therapy and limit the use of other potentially nephrotoxic drugs.
Proximal tubular toxicity
• Fanconi's syndrome, which can include:
1. phosphaturia and hypophosphataemia;
2. bicarbonaturia and proximal tubular acidosis;
3. kaliuria and hypokalaemia;
4. calciuria and hypocalcaemia;
5. magnesiuria and hypomagnesaemia.
Distal tubular toxicity
• Nephrogenic diabetes insipidus
• Distal tubular acidosis

Distal tubule involvement causes reduced urinary concentration and, in the most severe cases, nephrogenic diabetes insipidus. Tubular acidosis, both of proximal and distal origin, can result 35. Histological tests can show the presence of tubular atrophy and necrosis and interstitial lymphocytic infiltration, an indicator of inflammatory reaction secondary to tubular damage 36.

Among the various risk factors examined, age, cumulative drug dose, treatment in association with cisplatin, the presence of pre‐existing renal damage, prior nephrectomy or the presence of renal malformations prior to initiation of chemotherapy in patients with syndromes such as Denys–Drash or WAGR, appear to be the most significant in the onset of ifosfamide‐induced renal damage. Numerous studies report a higher incidence of proximal tubular damage in children under the age of 5, even though severe levels of ifosfamide‐induced nephrotoxicity have been reported in all age bands. One study conducted in Canada between 1984 and 1989 highlighted how patients with severe renal damage would have a lower mean age (78.1 ± 64.1 months) with respect to patients where said damage had not been shown (104 ± 67 months); both groups had undergone the same number of chemotherapy cycles 37. Regarding the cumulative dose of drug administered, some studies show patients who have received total drug doses >60 g m⁻² to be at greater risk, even though significant damage has been reported for cumulative doses between 12 and 60 g m⁻² 34, 38, 39, 40, 41. On the other hand, severe forms of toxicity with the development of Fanconi's syndrome appear to be correlated with both age (<30 months), and the cumulative dose of ifosfamide (>60 mg m⁻²). Nephrectomy represents a further significant risk factor for the development of chemotherapy‐related renal damage as a result of reduced total renal parenchyma. One study conducted in order to identify the most significant risk factor for the development of Fanconi's syndrome has highlighted that, by comparing nephrectomized and non‐nephrectomized patients, the relative risk for nephrectomized patients to develop renal Fanconi's syndrome is 11.4 42; the explanation may be found in the increased load in the residual nephrons. Furthermore, damage may be aggravated by radiotherapy to the kidneys; the latter results in the onset of interstitial fibrosis, with a reduction in glomerular filtration and increased load for the individual nephrons (Table 3) 43.

Table 3.

Major risk factors in the onset of ifosfamide‐induced renal damage

• Age <5 years

• Total dose of drug administered (>60 g m −2 )

• Association with cisplatin

• Pre‐existing renal damage

• Prior nephrectomy

• Pre‐existing renal malformations

Despite the variability of ifosfamide metabolism with various treatment schemes having been documented, there is no clear evidence that the various methods of intravenous administration influence the risk of nephrotoxicity. Increased risk in the case of additional treatment with cisplatin appears to have been verified, even though the latter causes renal damage through a different mechanism, predominantly resulting in reduced glomerular filtration and loss of magnesium 29, 33, 40, 44, 45, 46. In addition, aminoglicosides, frequently used with neutropenic patients, do not seem to represent an exacerbating factor in the development of ifosfamide‐related renal damage 42. However, there is significant individual variability in the onset and severity of renal involvement, as, in some cases, these manifest acutely from the first treatment cycle, while in others, chronic damage arises gradually and becomes evident even several months after completion of therapy 33, 40, 47, 48.

The degree of reversibility and the long‐term development of kidney damage following treatment with ifosfamide remain uncertain. In any case, the possibility of recovery is dependent on the extent of damage, and is more likely the less renal function is compromised. Partial or complete recovery of renal function between chemotherapy cycles or following its suspension is reported 49, 50, 51. Other studies appear to confirm that tubular damage can progress even several months after cessation of therapy 29. On the other hand, progression to chronic renal failure has been clearly documented in adults 44, 45, 46, 47, 48, 49, 50, 51, 52.

Tubular damage appears to be the primary event subsequently leading to compromised glomerular function. In addition, tubular leakage of amino acids is the most sensitive marker for nephrotoxicity and the predictive parameter of subsequent renal damage, generally preceding defects in the reabsorption of phosphates, glycosuria, damage to sodium absorption, renal tubular acidosis and reduced creatinine clearance.

Careful monitoring of tubular function, by evaluation of the excretion fraction of electrolytes and the presence of glycosuria, phosphaturia and aminoaciduria and by measuring blood electrolyte levels in the initial stages of treatment and prior to each chemotherapy cycle, becomes fundamentally important in order to identify the patients at greatest risk of progression to chronic renal damage. Indeed, in the presence of changes in renal function, the need to correct any changes in electrolytes that may be the consequence of secondary tubulopathy (including those predisposing to the onset of Fanconi's syndrome) should be assessed, and in the most severe cases, the option to modify or suspend chemotherapy treatment should be considered, together with avoiding administration in association with other potentially nephrotoxic chemotherapy agents (e.g. cisplatin), and in cases of severe nephrotoxicity, the suspension of ifosfamide and the continuation of therapy with cyclophosphamides is envisaged. Cyclophosphamide, an isomer of ifosfamide, is not thought to cause nephrotoxicity, possibly because of different pharmacology, but clinical studies to confirm, especially at very long‐term follow‐up, are lacking.

The risk of ifosfamide renal damage may be reduced by adopting some limitations in its use such as cumulative doses over 80 g m⁻², children younger than 5 years, and patients with a previous exposure to cisplatin or with pre‐existing renal damage. In addition, children should be monitored regularly even after the completion of treatment with a specific surveillance programme, including blood pressure measurement, baseline serum creatinine along with potassium, bicarbonate, phosphate and calcium, dipstick analysis for proteinuria, and monitoring of growth in order to detect any potential dysfunction or alteration at an early stage.

Cisplatin‐induced nephrotoxicity

In 1964, Rosenberg reported the biological properties of cisplatin for the first time while testing the effects of electrical fields on bacterial growth, and subsequently demonstrated its anticancer properties in laboratory animals 53, 54. The drug was introduced for the treatment of cancer for the first time in 1971 55, 56, 57. This is a neutral complex with two reversibly bound hydrochloride groups and two inert amines in the cis configuration. Several hypotheses have attempted to explain the cytotoxicity and selectivity of the anti‐cancer action of cisplatin; at the molecular level, the ability of the drug to bind DNA has been underlined 58. About 90% of the cisplatin administered is tightly bound to plasma proteins within 2 h of administration, and only the unbound platinum appears to exert anti‐cancer activity 59. Pharmacokinetics studies indicate that the drug has a triphasic half‐life lasting 30 min, 60 min and 24 h, respectively 60.

The nephrotoxicity associated with the use of cisplatin in childhood cancer therapy has been demonstrated for some time, while recent studies have described new aspects of the tubular damage, including increased elimination of low molecular weight peptides and reduced excretion of Tamm‐Horsfall protein 61, 62. Tamm‐Horsfall protein is a glycoprotein which is secreted in the thick ascending loop of Henle of the kidney. After renal tubular damage, the secretion of THP is reduced. Early data on cisplatin‐induced nephrotoxicity indicated how 25% of patients showed an increase in blood nitrogen levels following 1 or 2 weeks of treatment 63.

The cellular mechanism underlying cisplatin‐induced nephrotoxicity has not been entirely clarified, even though several mechanisms have been proposed. The accumulation of toxic platinum metabolites inside tubule cells seems to be one of the most important factors 57.

Among the potential levels of damage, protein binding, enzyme inhibition, mitochondrial damage, damage to DNA, RNA or both have been proposed from time to time 64, 65. More recent studies have demonstrated a reduction in sulphydryl groups on mitochondrial proteins and calcium uptake, while other studies have hypothesized that the sulphydryl groups bound to the platinum enter inside the tubule cells through an active transport mechanism with subsequent cellular distribution of the platinum metabolites 66, 67.

Compromised renal function during treatment with cisplatin can be acute or chronic. Acute damage arises within 24–48 h of administration of the drug, particularly in the presence of inadequate hydration. In this phase, there is a particular reduction in glomerular filtration with acute kidney injury due to tubular necrosis. The most significant damage occurs at the loop of Henle, the distal tubule and the collecting ducts, and manifests clinically as altered levels of electrolytes or full‐blown renal failure; in many cases, suspension of treatment is accompanied by recovery of glomerular function. Chronic manifestations consist of progressive reduction in renal function: tubular atrophy, interstitial fibrosis and chronic inflammatory reactions with lymphocytes and plasma cells are observed morphologically 68.

Hypomagnesaemia is the most frequently reported electrolyte disturbance during therapy with cisplatin, and if severe may be responsible for tetany or convulsions. Increased renal magnesium excretion appears to be caused by a defect in its transport in juxtamedullary nephrons or collector ducts 69. When the kidney function is not altered, only 25% of the glomerular magnesium filtrate is reabsorbed at the proximal tubule, while 50–60% is reabsorbed in the ascending tract of the loop of Henle 70. Increased magnesium clearance has been observed in treated patients, with reduced blood levels. Altered magnesium homeostasis appears to be dependent on the cumulative dose of drug administered; in particular, hypomagnesaemia is apparent for cumulative doses of at least 300 mg m⁻². However, reduced plasma magnesium can also happen much earlier, and occur immediately after conclusion of the chemotherapy cycle, particularly with repetition of the cycles 71. Hence, in the intervals between chemotherapy cycles, it is essential to frequently monitor (at least weekly) the blood magnesium concentration in patients undergoing therapy with cisplatin; in the follow‐up of patients no longer receiving therapy, the periodicity of the follow‐up checks should be established according to the electrolyte changes present.

Increased potassium elimination has also been documented in patients treated with cisplatin; reduced tubular reabsorption would appear to be due to the increased loss of sodium, potassium and water at the most distal segment of the nephron, resulting in Na⁺‐dependent potassium secretion 72, 73. In any case, hypomagnesaemia could also contribute to the reduced serum potassium levels. In addition, vomiting, induced by administration of the drug, represents an additional mechanism exacerbating the hypokalaemia. Besides the electrolyte disturbances, increased elimination of amino acids, β₂‐microglobulin, NAG, albumin and IgG have also been documented. From data published in the literature, it emerges that in the initial phases of administration of the drug, proteinuria is essentially of tubular origin; later, glomerular involvement occurs, with increased elimination of albumin and IgG 74. In some cases, the onset of hypertension has also been reported 75.

The association with nephrotoxic drugs appears to be an additional risk factor, and since ifosfamide‐associated renal damage differs from cisplatin‐induced toxicity, numerous studies agree with the statement that the two drugs in association can exacerbate ifosfamide‐induced tubular toxicity 57, 76.

From the time when cisplatin became one of the most potent anti‐cancer agents used, the application of measures that could prevent or reduce kidney toxicity became important. The most commonly used preventive measures are hydration prior to the start of treatment and the infusion of mannitol along with administration of the drug, thus reducing the urinary concentration of the drug and hence nephrotoxicity 77, 78. Preliminary studies have reported the use of agents with protective activities with regard to cisplatin, such as para‐aminobenzoic acid, methimazole, chlorpromazine, L‐arginine, and triamcinolone 79, 80, 81, 82, 83; however, many of these studies have been conducted on animals in vivo and their efficacy in reducing cisplatin‐associated nephrotoxicity in humans has yet to be definitively shown.

Various attempts have been made to develop platinum analogues with reduced toxicity. Carboplatin represents one of the products from this search. Carboplatin nephrotoxicity is similar in nature in terms of causing glomerular impairment and hypomagnesaemia but less common (especially for glomerular toxicity) and usually much less severe than cisplatin‐induced renal toxicity 84.

Appropriate hydration with isotonic saline solution is the adopted strategy to reduce the incidence of cisplatin‐induced nephrotoxicity. Recent clinical guidelines have established that the addition of diuretics, such as mannitol and furosemide, is no more nephroprotective than the use of hydration alone 77 and that only patients receiving high‐dose cisplatin (e.g., 100 mg m⁻²) benefit from forced diuresis 85, 86, 87, 88. Magnesium deficiency may enhance the incidence and severity of cisplatin nephrotoxicity, thus continuous magnesium supplementation should be administered routinely to children receiving cisplatin‐based treatment 57.

Many pharmacological agents (i.e., N‐acetyl‐cysteine, α‐tocopherol, lipoic acid, sodium thiosulfate salicylate, ebselen, capsaicin, D‐methionine, amifostine) have been investigated in order to reduce the cisplatin‐related nephrotoxicity but currently no specific agent has been recognized as a preventive measure due to the lack of clear efficacy 89.

Nephrotoxicity due to other chemotherapy agents

Even though ifosfamide and cisplatin are the chemotherapy agents most often responsible for kidney damage, the use of other anti‐cancer drugs is also associated with compromised glomerular function, including high‐dose methotrexate and nitrosoureas.

High‐dose methotrexate is among the most important drugs for the treatment of certain forms of cancer, particularly leukaemias, lymphomas and osteosarcoma; it represents a structural analogue of folic acid and acts as an antagonist of this vitamin. Folate plays a central role in metabolic reactions involving the transfer of one carbon unit, essential in replicating cells for methionine, adenine, purine and deoxythymidylic acid biosynthesis.

Methotrexate toxicity acts particularly on the haematopoietic system and the intestinal mucosa, and renal involvement is frequent in high‐dose regimens. Renal damage only occurs with a sudden increase in plasma creatinine and ureic nitrogen levels, and is believed to be due to the precipitation of methotrexate and its less soluble metabolites (7‐OH‐MTX and DAMPA) in acid urines 90, 91. However, normal serum creatinine levels do not allow exclusion of the onset of side effects, while increased urinary excretion of NAG would be indicative of prolonged elimination of the drug, and hence toxicity 92. It appears that the drug and its metabolites also exercise their toxic action at the renal tubule level. The damage induced by the drug results in its reduced elimination, precisely at the renal level; this leads to high plasma concentrations of the drug, which can make rescue with folic acid insufficient, intensifying the other side effects, particularly myelosuppression, mucositis, hepatotoxicity and dermatitis 93, 94, 95, 96. MTX‐related acute nephrotoxicity seems to be completely reversible, because after long‐term follow‐up, MTX‐treated cancer survivors did not manifest glomerular or tubular dysfunction. Large cohort studies have failed to demonstrate a significant association between high‐dose methotrexate and chronic nephrotoxicity 97, 98. The application of hyperhydration and alkalinization of the urine using NaHCO₃ in order to prevent the precipitation of toxic products reduces the incidence of toxic effects, without any impact on the pharmacokinetics of the drug.

Nitrosoureas (carmustine, lomustine) are alkylating agents, and their cytotoxic effect is operated through an alkylation mechanism that inhibits DNA replication, RNA synthesis and protein synthesis 99, 100. Their anti‐cancer activity requires a biotransformation that occurs through the non‐enzymatic cleavage of the molecule, releasing derivatives possessing alkylating activity.

Although renal damage is not one of the toxic effects most frequently associated with treatment with nitrosoureas, in rare cases, and for very high cumulative doses (>1200 mg m⁻²), cases of acute renal failure have been reported; renal damage can even appear 1–2 years after the suspension of treatment 101, 102.

Conclusions

Chemotherapy‐associated nephrotoxicity represents a particularly significant side effect, particularly in the treatment of childhood cancers, with regard to the patient's potential life expectancy and increased survival. Careful evaluation of renal function appears to be fundamental, considering that compromised renal function is frequently clinically silent in the early stages, and in consideration of the preventive measures to be implemented. In particular, the reduction and/or modification of the chemotherapy regimen should be considered where there are changes in laboratory results indicating compromised glomerular and/or tubular function, even if only initial. The concomitant administration of potentially nephrotoxic antibiotics should also be avoided. Agents which cause renal damage tend to be synergistic in their toxic effects. It is therefore important to identify those antibiotics that have potential nephrotoxic effects so that their dosage can be based on the patient's renal function and all factors that may potentiate the toxicity can be corrected. These preventive strategies are generally followed by clinicians, although clinical studies to confirm these data are lacking.

In patients where chemotherapy‐induced renal damage has occurred, renal function should be closely monitored prior to each cycle, during the treatment and even after the completion of treatment.

Nomenclature of targets and ligands

Key ligands in this article are hyperlinked to corresponding entries in http://www.guidetopharmacology.org, the common portal for data from the IUPHAR/BPS Guide to PHARMACOLOGY 103.

Competing Interests

There are no competing interests to declare.

Ruggiero, A. , Ferrara, P. , Attinà, G. , Rizzo, D. , and Riccardi, R. (2017) Renal toxicity and chemotherapy in children with cancer. Br J Clin Pharmacol, 83: 2605–2614. doi: 10.1111/bcp.13388.