AKI is a common occurrence in hospitalized patients with cancer and, when it occurs, it is associated with higher costs, increased hospital length of stay, and increased morbidity and mortality. In some cases, AKI may lead to changes in chemotherapy regimens that may lessen the chances of disease remission, or AKI may exclude patients from clinical trials. In the largest study on this topic, 37,267 incident patients with cancer in Denmark had follow-up over a 7-year period in the early 2000s. The 1-year risk of AKI, as defined by RIFLE risk (>50% increase in serum creatinine), was 17.5%. The 1-year risk for more severe AKI, the RIFLE injury (>100% increase in serum creatinine), and failure (>200% increase in serum creatinine of >4 mg/dL and requiring dialysis) risk categories was 8.8% and 4.5%, respectively. The 5-year risk for the RIFLE risk, injury, and failure AKI categories was even higher, at 27%, 14.6%, and 7.6%, respectively. Critically ill patients with cancer have a higher incidence of AKI and AKI requiring dialysis than critically ill patients without cancer. ^, In those patients with cancer who suffer an episode of AKI, the mortality is elevated, such that the mortality was 13.6% in those without AKI and progressively increased with higher RIFLE stage AKI (risk, 49%; injury, 62.3%; failure, 86.8%). In patients with hematologic malignancies who were undergoing induction therapy, the mortality in those with RIFLE risk AKI was 13.6%, whereas the mortality in those with no AKI was 3.8% over an 8-week period. Importantly, those patients who required dialysis during the period of induction therapy had a mortality of 61.7%. It is important to note that more contemporary studies may demonstrate that outcomes in patients with cancer and AKI are not universally poor, with one study showing that 82% of critically ill patients with cancer with AKI completely recovered kidney function, whereas partial recovery was observed in 12%, and chronic renal replacement therapy (RRT) was required in only 6% of patients. The prognosis of AKI greatly depends on the underlying functional and premorbid conditions of the patient, as well as the overall state of the malignancy. Thus individualized decisions regarding the appropriateness of RRT in these patients is warranted.

A specific group of patients with a high incidence of AKI is those undergoing hematopoietic stem cell transplantation (HSCT). Various studies have shown an incidence of AKI ranging from 23% to 73% according to the different cutoff points used to define AKI. Furthermore, the incidence of AKI varies according to the type of HSCT, with most data supporting the finding that myeloablative allogenic transplantation is associated with a higher incidence of AKI.

The risk factors for AKI in patients with cancer are multiple and depend on the specific type of cancer. These can be broadly divided into those that are modifiable and those that are not ( Table 57.1 ). Certain cancers appear to carry a higher AKI risk than others, with renal cell carcinoma, hepatocellular carcinoma, multiple myeloma, and lymphoma having higher rates of AKI. As an example, patients undergoing total (radical) nephrectomy for renal cell carcinoma have an incidence of AKI as high as 33.7%. Even partial nephrectomy to spare nephron loss is associated with AKI, albeit less commonly. Patients with acute lymphoma or leukemia undergoing induction chemotherapy are also at especially high risk to develop AKI, primarily due to tumor lysis syndrome and drug nephrotoxicity. In a series of 537 patients undergoing induction therapy for acute myelogenous leukemia or high-risk myelodysplastic syndrome, 36% developed AKI. Whenever possible, modifiable risk factors should be addressed in an attempt to lower the incidence of AKI.

The causes of AKI in patients with solid organ cancers are similar to those of the general population, with an overrepresentation of obstructive nephropathy (especially with prostate, ovarian, and cervical cancer), chemotherapy-associated nephrotoxicity, sepsis-associated ischemic acute tubular necrosis in the setting of neutropenia, and AKI due to hypercalcemia. In patients with hematologic malignancies, the causes of AKI are also similar to those of the general population, with some notable unique causes, such as AKI associated with multiple myeloma (e.g., cast nephropathy, light- or heavy-chain−associated glomerulonephritis or associated with hypercalcemia), tumor lysis syndrome, and tumor infiltration of the kidneys associated with lymphoma and leukemia. In those patients receiving a stem cell transplant, there are other situation-specific causes of AKI including hepatic venoocclusive disease and cytokine release syndrome.

Prerenal causes of AKI are common in patients with malignancies and often result from poor oral intake, as well as chemotherapy-induced nausea, vomiting, and diarrhea. Hypercalcemia and its natriuretic effect may also lead to volume depletion and prerenal azotemia. In addition, patients with cancer may be prescribed common medications, such as diuretics, angiotensin-converting enzyme inhibitors, angiotensin receptor blockers, or nonsteroidal anti-inflammatory drugs that can affect renal autoregulation and exacerbate prerenal AKI. Prerenal causes are so common in this patient population that judicious trials of intravenous (IV) fluids are reasonable in most patients presenting with AKI.

Many of the causes of AKI seen in patients with cancer are not unique and are covered in other chapters. However, certain intrarenal and postrenal causes are seen almost exclusively in the setting of malignancy and are covered here and shown in Table 57.2 . AKI associated with glomerular disorders, chemotherapeutic agents, HSCT, and hypercalcemia and seen in the setting of renal cell carcinoma are covered in their respective sections in this chapter.

Tumor Lysis Syndrome

Tumor lysis syndrome (TLS) is encountered in patients with bulky, rapidly growing and chemosensitive malignancies, such as high-grade lymphomas (e.g., Burkitt lymphoma), leukemia, or other cancers with large cellular burdens. ^, Typically, TLS is seen after the administration of chemotherapy but can occur spontaneously as well. TLS has been defined by the Cairo-Bishop criteria ( Table 57.3 ) and is biochemically typified by the findings of numerous electrolyte disorders (e.g., hyperkalemia, hyperphosphatemia, hypocalcemia, hyperuricemia) that result from the release of intracellular substances after cell lysis. ^, Importantly, TLS may lead to sudden death and seizures due to the electrolyte disturbances encountered in these patients. The mechanism of AKI in these patients is at least partially due to uric acid nephropathy but may also be influenced by cytokine release and nephrocalcinosis, the latter due to elevated serum concentrations of phosphate and calcium.

Table 57.3

Cairo-Bishop Criteria for Tumor Lysis Syndrome

From Cairo MS, Bishop M. Tumor lysis syndrome: new therapeutic strategies and classification. Br J Hematol . 2004;127(1):3–11.

Type of Criteria	Features
Laboratory criteria for tumor lysis syndrome	Requires 2 or more of the following criteria within 3 days to or 7 days after initiation of chemotherapy: • Uric acid level ≥8 mg/dL or 25% increase from baseline • Potassium level ≥6.0 mEq/L or 25% increase from baseline • Phosphorus level ≥6.5 mg/dL for children • Phosphorus level ≥4.5 mg/dL for adults or 25% increase from baseline • Calcium level ≤7 mg/dL or 25% decrease from baseline
Clinical criteria for tumor lysis syndrome	Laboratory tumor lysis syndrome plus one or more of the following criteria: • Creatinine >1.5 times upper limit of age-adjusted reference range • Cardiac dysrhythmia or sudden death • Seizure

Uric acid nephropathy and AKI are the result of the precipitation of insoluble uric acid in the renal tubules that occurs when purine nucleotides are released from dying cancer cells. These nucleotides are metabolized by xanthine oxidase into uric acid, which is subsequently filtered in the glomerulus and concentrated in the renal tubules. This forms an insoluble precipitate, leading to a combination of intratubular obstruction, vasoconstriction, and inflammation, culminating in a fall of the glomerular filtration rate and AKI. Understanding this pathophysiology leads to rational protocols for the prevention of TLS ( Fig. 57.1 ). While most of the literature supports crystalopathy-related AKI as the primary mechanism of injury, one study has postulated a novel pathophysiologic model through which AKI is associated with endothelial dysfunction from high levels of extracellular histones. This animal study proved that extracellular histones are released in vast amounts during TLS, causing a profound endothelial injury in the mouse model. The mechanisms of histone-mediated damage implicate endothelial cell activation mediated by Toll-like receptor 4. In addition, TLS has now been associated with novel agents, such as the B cell lymphoma 2 inhibitor venetoclax, the monoclonal antibody obinutuzumab, and even immune checkpoint inhibitors, but data on such associations are limited to case reports and case series.

Approach to the management of tumor lysis syndrome.

ECG, Electrocardiogram; GFR, glomerular filtration rate; IV, intravenous; SPS, sodium polystyrene sulfonate.

From Rosner MH, Perazella MA. Acute kidney injury in patients with cancer. N Engl J Med. 2017;376(18):1770–1781.

The first step in a preventive strategy rests on identifying patients at risk for TLS, which should include any patient with a large, rapidly growing and chemosensitive tumor or those patients who, before treatment, may already be manifesting signs of tumor lysis with an associated electrolyte disturbance. Low-risk patients should be treated with oral fluids and avoidance of other nephrotoxins; high-risk patients should receive IV 0.9% saline to maintain adequate glomerular filtration and tubular flow rates that facilitate rapid clearance and dilution of uric acid, as well as the clearance of potassium and phosphate. The prophylactic use of a xanthine oxidase inhibitor (allopurinol or febuxostat) is recommended because these drugs will block de novo formation of uric acid and limit further increases in the levels of this compound. For patients who already have elevated uric acid levels, treatment with recombinant urate oxidase (rasburicase) is beneficial for rapidly lowering uric acid levels and is recommended. Rasburicase converts uric acid to a soluble and readily excreted compound, allantoin. It should be cautioned that rasburicase also leads to the production of hydrogen peroxide, which can cause methemoglobinemia and hemolytic anemia in patients with glucose-6-phosphate dehydrogenase (G6PD) deficiency, and at-risk patients should be tested for this condition before administration.

For those patients manifesting signs of TLS, close monitoring of serum electrolyte levels and clinical status is required (see Fig. 57.1 ) and electrolyte disorders require rapid therapy. Patients should receive aggressive IV hydration, a xanthine oxidase inhibitor, and rasburicase. For patients presenting with established AKI, emergent hemodialysis to correct electrolyte disorders may be required. Urine alkalinization, which increases uric acid solubility in the renal tubules, is not recommended because it can increase the likelihood of calcium phosphate precipitation.

Light-Chain Cast Nephropathy and Acute Kidney Injury Associated With Myeloma

Patients with myeloma have an incidence of AKI that can be as high as 50%. The most common cause of AKI in these patients is cast nephropathy, followed by light chain–related proximal tubulopathy, various glomerular diseases (including light-chain deposition disease and primary [AL] amyloidosis), hypercalcemia, hyperuricemia, and other contributing factors ( Fig. 57.2 ).

Approach to distinguish primary membranous nephropathy from cancer-associated membranous nephropathy.

EM, electron microscopy; GBM, glomerular basement membrane; IF, immunofluorescence; Ig, immunoglobulin; LM, light microscopy; MN, membranous nephropathy; NELL-1, neural epidermal growth factor–like 1 protein; PCDH7, protocadherin 7; PLA2R, phospholipase A2 receptor; SEMA 3B, semaphorin-3B; THSD7A, thrombospondin type I domain-containing 7A.

Cast nephropathy is due to the filtration of large amounts of myeloma-produced free light chains into the renal tubules, which bind to Tamm-Horsfall protein (also known as uromodulin) to form insoluble aggregates. These intratubular precipitates lead to intrarenal obstruction and local inflammation, which may be further exacerbated by the proximal tubular uptake of pathogenic light chains, which also leads to activation of inflammatory pathways. Sensitive assays for serum free light chains, serum and urine protein electrophoresis, and immunofixation facilitate the diagnosis of myeloma-related AKI. Typically, patients with myeloma-related glomerular diseases will present with high levels of albuminuria (>2 g/day) indicative of the glomerular injury. Patients with cast nephropathy will demonstrate high levels of urine and serum light chains and lower levels of albuminuria. Of note, 15% of patients with AKI in the setting of myeloma may have a cause of AKI completely unrelated to myeloma; thus kidney biopsy should be performed in unclear cases.

Therapy for cast nephropathy focuses on a combination of hydration and ensuring sufficient renal tubular flow, correction of hypercalcemia, treatment of other precipitating or aggravating factors, and chemotherapy to lower the myeloma production of pathogenic light chains. Chemotherapeutic regimens for myeloma generally include a proteasome inhibitor, such as bortezomib, which does not require dosing adjustments for renal function. Regimens including bortezomib (usually in combination with dexamethasone and another agent, such as thalidomide or melphalan) have led to renal response rates as high as 72% and dialysis discontinuation in 57% of patients. Other effective agents include thalidomide, lenalidomide, pomalidomide, and carfilzomib. A great deal of controversy remains surrounding the benefit of removing light chains by therapeutic plasma exchange or high-flux hemodialysis; currently, these methods cannot be routinely recommended.

As discussed earlier, the prognosis of critically ill patients with cancer and AKI is poor, with mortality rates approaching 60% to 70% if RRT is needed. Thus clinicians may be faced with difficult decisions when offering invasive intensive therapy for such a patient population and may appropriately opt for more conservative and palliative approaches. However, it is important to individualize these decisions because select patients (e.g., those with fewer comorbid conditions, higher functional status, and cancers that are responsive to therapy) may benefit from more intensive therapy. Therefore decisions regarding the initiation of dialysis therapy require input from the entire care team, as well as the patient and family. These decisions should consider the reversibility of the AKI, long-term cancer prognosis, quality of life before critical illness, and the patient’s preferences.

As has been observed with AKI, CKD is also noted to be a possible complication of numerous cancers and their associated therapy. There are several explanations for this observation. An important driver of this untoward complication is the fact that preexisting and often unrecognized CKD is common in patients with various types of malignancy. In addition to underlying kidney disease, incomplete recovery of kidney function after multiple episodes of AKI, nephrectomy for kidney cancer with a reduction in nephron mass, and exposure to several courses of nephrotoxic medications can ultimately lead to the development of glomerulosclerosis and tubulointerstitial fibrosis, all of which either exacerbate underlying kidney disease or promote the development of new CKD.

In regard to preexisting CKD in patients with cancer, the Renal Insufficiency and Anticancer Medications (IRMA)-1 and-2 studies clearly illustrate the extent of this problem. ^, In the two IRMA studies, 52.9% (IRMA-1) and 50.2% (IRMA-2) of patients with an active malignancy had an estimated glomerular filtration rate (eGFR) <90 mL/min/1.73 m ² , respectively. ^, In this group of patients, the prevalence of stage G3 CKD was 12% (IRMA-1) and 11.8% (IRMA-2), respectively. Stage G4 CKD was rare in these two studies (0.9% and 0.7%, respectively). In an Australian cohort with various cancers ( n = 4077) using a prospective population cohort design, eGFR was <60 mL/min/1.73 m ² in 30% of patients and <45 mL/min/1.73 m ² in 8.3%. In a cohort of Chinese patients with underlying cancer observed from 2000 to 2004, stage G3 or more severe CKD (defined as an eGFR < 60 mL/min/1.73 m ² ) was noted in 12.8% (1051 of 8223). The relatively high prevalence of CKD in patients with cancer has been confirmed in other studies, irrespective of the type of malignancy.

Similar to patients with malignancy who develop AKI, CKD in patients with malignancy experience an increased risk of death compared with those without kidney disease. Higher mortality is not uniformly noted; however, the association of CKD with mortality appears to be more pronounced with certain cancer types and definitively related to CKD severity. Patients with cancer and stage G3 CKD had an adjusted hazard ratio (HR) for cancer-specific mortality of 1.12 (95% confidence interval [CI], 1.01–1.26). In contrast, patients with stage G4 CKD had an adjusted HR for cancer-specific mortality of 1.75 (95% CI, 1.32–2.32). As described in a population-based cohort study, stage G3 CKD (eGFR <30 to <60 mL/min/1.73 m ² ) was a significant risk factor for death from cancer. In this study, the adjusted HR for cancer-specific mortality was 1.18 (95% CI, 1.08–1.29) for each 10 mL/min/1.73 m ² decrease in eGFR. It is notable that mortality risk was highest for patients with breast and urinary tract cancers. Patients with kidney failure receiving hemodialysis who underwent radiofrequency ablation for hepatocellular cancer (HCC) were identified using the Japanese Diagnosis Procedure Combination database. Four patients without kidney failure were matched for each patient with kidney failure based on age, sex, treatment hospital, and treatment year. In-hospital mortality was more than 7-fold higher in the 437 patients with kidney failure compared with the 1345 matched controls (1.1% vs. 0.15%, respectively; odds ratio [OR] = 7.77; P <.001).

In addition to CKD identified by eGFR measurement, the presence of proteinuria was also associated with mortality in patients with cancer. A retrospective study of 9465 patients with newly diagnosed cancer from January to December 2010 observed an increase in all-cause mortality in patients with eGFR <60 mL/min/1.73 m ² ) and proteinuria. As seen in other studies, CKD was not associated with mortality in all type of cancers. For example, an eGFR <60 mL/min/1.73 m ² was independently associated with death from hematologic (adjusted HR, 2.93; 95% CI, 1.36–6.31) and gynecologic cancers (adjusted HR, 2.82; 95% CI, 1.19–6.70), but not for other cancers. Furthermore, a study in patients with colorectal cancer ( n = 3379) noted that eGFR <60 mL/min/1.73 m ² was not associated with mortality. In addition, a prospective cohort study of patients with cancer that examined the risk of death as a secondary endpoint did not show an association between CKD and mortality. Thus it appears that underlying CKD in patients with cancer is associated with mortality in some, but not all types of malignancy.

As has been previously reviewed, CKD is a well-recognized complication of various malignancies and their therapies. However, in addition to the observed increase in CKD prevalence in patients with cancer, both CKD and ESKD appear to be risk factors for the development of a number of malignancies. ^, Thus the relationship between the kidney and cancer can be viewed as bidirectional. The association between severity of kidney disease and the risk of incident cancer was examined over 6,000,420 person-years of follow-up. During this time, 76,809 incident cancers were identified in 72,875 persons. After adjustment for time-updated confounders, eGFR was inversely associated with the risk of kidney cancer. The adjusted HR was 2.28 (95% CI, 1.78–2.92) for an eGFR <30 mL/min/1.73 m ² . The authors also observed an increased risk of urothelial cancer at an eGFR <30 mL/min/1.73 m ² but no significant associations between eGFR and other types of cancer. Several observational studies have suggested an increased cancer risk in patients with kidney failure receiving maintenance dialysis RRT ( Table 57.4 ). A study from Taiwan published in 2015 demonstrated higher risks of cancer in patients with ESKD (HR, 3.43) with higher risks specifically observed for oral, colorectal, liver, blood, breast, renal, upper urinary tract, and bladder cancer than those seen in the general population. In one study, the observed increased risk for renal parenchymal cancer in patients with ESKD was related to the development of acquired renal cystic disease, which increases with time on dialysis. A U.S. study that examined data from 1996 to 2009 observed a higher incidence of cancer in patients with ESKD. The cumulative 5-year incidence of any cancer was 9.48% compared with a 5-year cumulative incidence of cancer in U.S. transplant recipients of 4.4%.

The increased mortality observed as a complication in certain patients with cancer with CKD compared with those without kidney disease has several plausible explanations. The presence of preexisting CKD may limit the use of effective agents or promote underdosing of anticancer regimens that might otherwise be curative. ^, In addition, underlying CKD may alter the bioavailability and/or safety profile of some anticancer drugs, which can potentially lead to different and suboptimal treatment choices. The development of AKI superimposed on CKD, which is a common complication, can lead to the cessation of effective chemotherapeutic regimens, allowing unhindered and potentially rapid tumor growth. ^, Furthermore, CKD and ESKD may also be associated with impaired immune surveillance, which would allow cancers to grow and metastasize, especially when combined with insufficient drug dosing. This risk is further increased when these patients receive a kidney transplant, where immunosuppressive regimens (and other factors) enhance the risk for recurrent or de novo cancers that have been associated with mortality. ^, It is also possible that the adverse renal effects of the anticancer drug used leads to acute and/or progressive kidney injury or to worsening of preexisting CKD, which leads to increased all-cause mortality unrelated to cancer. ^, Finally, a nihilistic view of patients with CKD or ESKD who develop cancer may lead to undertreatment of potentially curable or modifiable malignancies.

Patients with cancer require frequent assessment of kidney function to ensure proper dosing of chemotherapeutic agents. ^{, ,} In addition, close monitoring during ongoing anticancer therapy is critical for timely recognition of drug-induced nephrotoxicity (either AKI or progression of underlying CKD). ^, The Kidney Disease: Improving Global Outcomes (KDIGO) guidelines have recommended that the glomerular filtration rate (GFR) should be the standard measure to evaluate kidney function using the most accurate method available in the individual cancer patient. This is of particular importance because the incidence of CKD in patients with cancer is as high as 50%. Lack of appreciation for the reduced GFR in patients with CKD can lead to erroneous, harmful, and potentially life-threatening dosing of medications that are cleared predominantly by the kidneys. However, the optimal method for assessing GFR in patients with cancer is currently unknown. Several small studies have examined the accuracy and precision of the traditional eGFR equation, including the four-variable Modification of Diet in Renal Disease, Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI), and two equations derived specifically in patients with cancer (the Wright and Martin formulas). ^, The previously described bedside eGFR equations all performed similarly well when compared with GFR determined by technetium-99m diethylenetriamine pentaacetic acid ( ^Tc99m DTPA) clearance and when used for selecting the appropriate drug dosage for carboplatin. It is worth noting that all GFR estimating equations significantly underdosed carboplatin, highlighting the need for continued research to determine the most accurate methodology for assessing GFR in this growing patient population.

The Onco-GFR study by Silva and colleagues used measured GFR using chromium-51-labeled ethylenediamine tetraacetic acid ( ⁵¹ Cr-EDTA) clearance in 1200 patients with solid tumors to test six GFR estimating equations. They reported both the bias (median of the differences between measured GFR and estimated GFR) and accuracy (1 minus the percentage of GFR estimates within 30% of measured GFR in mL/min/1.73 m ² [1−P30]) of each equation. The 2012 CKD-EPI equation, using both serum creatinine and cystatin C, performed the best among all equations. Among the GFR estimating equations that used serum creatinine alone, Cockcroft-Gault and 2009 CKD-EPI had the largest bias and Cockcroft-Gault had the least accuracy. In subgroup analyses, the authors showed that several patient-specific factors strongly influenced the accuracy of GFR estimation. Creatinine-based equations were much more likely to overestimate GFR in women and in those with low body mass index (BMI [<25 kg/m ² ]). In these populations, the CKD-EPI 2012 equation that uses cystatin C alone was most accurate. This suggests that GFR estimation could be personalized on the basis of patient-specific factors such as BMI and sex. The authors also demonstrate that in patients with measured GFR <60 mL/min/1.73 m ² , all equations overestimated GFR. In another study, data from 2011 to 2023 on adult patients with cancer who underwent measured GFR (mGFR) at Mayo Clinic were analyzed. GFR estimating equations were compared using performance metrics. Among 1145 patients with both serum creatinine and cystatin C concentrations available, the CKD-EPI creatinine–cystatin C equation showed the best performance, with low bias and highest precision across both solid and hematologic cancers. The European Kidney Function Consortium equation performed similarly, while the Cockcroft–Gault equation had the poorest precision. Findings support the use of combined creatinine–cystatin C equations over creatinine-only or cystatin C–only equations in patients with cancer.

It is noteworthy that the new 2021 CKD-EPI race-free equation that uses both serum creatinine and cystatin C concentrations was not used in this study.

Currently, we recommend use of the CKD-EPI creatinine–cystatin C formula for determination of eGFR in most patients with cancer. Note that the eGFR is not reliable in the setting of AKI, and the KDIGO guidelines have therefore recommended that timed clearances of urea and creatinine may be of great value for persons with AKI. See Chapter 23 for a detailed discussion of methods for measuring and estimating GFR.

Anticancer drugs are cleared from the body predominantly by the kidneys, extrarenal pathways, or a combination of the two. One of the most important drug-related problems in patients with cancer and impaired kidney function is inappropriate medication use and dosing errors. For example, many cytotoxic drugs and their active and potentially toxic metabolites are eliminated by the kidneys via some combination of glomerular filtration, tubular secretion, and/or tubular reabsorption. Furthermore, patients with CKD receiving chemotherapeutic agents often have alterations in one or more of their pharmacokinetic parameters. Alterations in medication absorption, volume of distribution and drug distribution in the body, protein binding, biotransformation, and kidney excretion may result in the accumulation of potentially toxic components and overdosage. ^, It is therefore critical that clinicians adjust doses of drugs that are excreted primarily by the kidneys appropriately. Drug dosing will use either the calculated or measured creatinine clearance or GFR estimating equations (as discussed earlier), which should allow for safer and more effective use of chemotherapy in patients with cancer and underlying CKD. Drug dosing and appropriate adjustments for the most commonly used chemotherapeutic agents, which may require dose modification in the setting of underlying kidney disease, are noted in Table 57.5 . ^{, ,}

It is important to recognize that CKD also alters the pharmacokinetics of drugs that are cleared by nonrenal pathways. It has been demonstrated that uremia (in the setting of both AKI and CKD) affects hepatic drug metabolism and coupled transport. Selected uremic toxins interfere with transcriptional activation, cause downregulation of gene expression mediated by proinflammatory cytokines, and directly inhibit the activity of the cytochrome P450 enzymes and drug transporters. These additional nuances in pharmacokinetics will make prescribing of some drugs to patients with CKD that much more complicated. As a result, clinicians will need to watch closely for symptoms and signs of drug-related overdose and toxicity even when the patient has received what appears to be appropriate renal dosing of an anticancer drug regimen. Thus clinicians should work closely with clinical pharmacists when these drugs are prescribed to CKD patients, and special considerations should be taken to optimize exposure to cytotoxic drugs and reduce the risk of adverse effects. Precautions should include consideration of current medication pharmacokinetic and pharmacodynamic information, use of less nephrotoxic agents, and documented preventive measures, which are well integrated into oncology practice and vigilant surveillance. ^,

The management of cancer in the ESKD population is particularly challenging. Little is known about anticancer drug management in ESKD patients and even less about the optimal timing and necessary dosage adjustments, depending on all forms of dialysis. In most patients receiving maintenance hemodialysis, kidney function no longer contributes a significant proportion of drug clearance. As a result, these patients may require a reduction in drug dose to avoid excessive drug exposure and associated toxicity. This is clearly needed with drugs that are normally eliminated by the kidneys and also when extrarenal (hepatic primarily) metabolic and excretory pathways are impaired due to uremia. Because most cytotoxic drugs used are excreted predominantly in the urine as unchanged drug or active or toxic metabolite(s), any reduction in kidney clearance (as in CKD and ESKD) may result in the accumulation of potentially toxic components and overdosage. Careful dosage adjustment is mandatory to optimize exposure to cytotoxic drugs and reduce the risk of adverse effects.

The efficiency (or lack thereof) of drug clearance by the modality of dialysis must also be considered to allow for appropriate timing of chemotherapy administration. Therefore it is necessary to determine what fraction of drug(s) is removed by hemodialysis to allow chemotherapy dosing after hemodialysis sessions (for medications with excellent dialytic clearance) to avoid drug removal, which may result in a loss of anticancer efficacy. Drugs that are not removed efficiently by hemodialysis can be administered at any time, either before or after the hemodialysis session. Hemodialysis influences drug pharmacokinetics primarily by the following three properties—drug clearance by dialysis, extraction coefficient, and dialysis extraction factor. Hemodialysis clearance is the removal rate relative to blood concentration when entering the dialyzer. It is calculated by the amount removed (in mg/min) relative to the rate of presentation (mg/mL). Extraction coefficient, also called the “extraction ratio,” is the percentage of drug removed from blood across the dialyzer. It is calculated by the rate of removal (mL/min) relative to the rate of presentation (mL/min). ^, Hemodialysis clearance and the extraction coefficient measure the ability of a hemodialysis system to remove a drug from the blood but do not indicate how readily the drug is removed from the body. Because hemodialysis clearance and extraction ratio cannot be extrapolated to the clinical setting, the hemodialysis extraction factor of the drug (when available) should be used to determine the clinical impact of hemodialysis session on drug removal. This extraction factor, which is derived from total body clearance and the dialysis clearance of a drug, represents the actual influence of dialysis on drug pharmacokinetics. Hemodialysis is considered clinically relevant if the dialysis extraction factor exceeds 25% of the overall drug elimination. ^,

Ultimately, it is important to measure the pharmacokinetic parameters in patients receiving maintenance hemodialysis and to compare these data with those of patients with normal kidney function. The current lack of pharmacokinetic information of anticancer drug therapy in patients receiving maintenance hemodialysis underlines the need for prospective studies in these patients. Data on dosage adjustments in peritoneal dialysis, hemodiafiltration, and continued forms of renal replacement therapy are limited in most anticancer agents.

Although cancer screening recommendations for the general population focus on early detection and treatment to improve outcomes and survival, it is not clear that these same principles apply to patients with ESKD. Reasons to screen or not screen for malignancy in ESKD include a foreshortened lifespan, occurrence rates (common vs. rare) of specific cancers, variable accuracy of screening tools, and safety and efficacy of interventions when a cancer is identified.

Although maintenance dialysis prevents death from uremia, mortality among patients with ESKD remains high. Generally, patients with ESKD younger than 80 years of age live less than one-third as long as their non-ESKD counterparts, whereas those aged older than 80 years are projected to live less than half as long as those without ESKD. The most common cause for mortality, accounting for approximately 50% of deaths, is cardiovascular disease; death due to malignancy is relatively uncommon. Individual characteristics should be taken into account because groups include those with higher and lower mortality risks. For example, the 50-year-old man with uncontrolled type 2 diabetes mellitus, heart failure, and peripheral arterial disease has a much higher mortality than the 50-year-old man with polycystic kidney disease and no other significant comorbid conditions. Thus a patient-specific understanding of life expectancy is required. Due to a likely shortened life expectancy in ESKD patients primarily due to cardiovascular disease and infection, cancer screening strategies developed in the general population may not apply to them.

The goal of cancer screening is early cancer detection to enhance patient longevity. This appears to be an effective strategy for many cancers (e.g., using flexible sigmoidoscopy for colon cancer screening) in the general population when a patient’s expected lifespan exceeds 10 years. Thus colon cancer screening is unlikely to be effective in many ESKD patients with multiple comorbidities whose lifespan is 3 to 5 years. In addition, screening requires patients to undergo additional procedures and tests, which have the potential for harm. Thus in the absence of a survival benefit or if the survival benefit of screening is only realized in those with an expected lifespan of more than 10 years, cancer screening in most patients with ESKD is unlikely to improve outcomes or be cost-effective. It is noteworthy, however, that certain ESKD patients enjoy life expectancies more than 10 years and may benefit from individualized cancer screening. While life expectancy plays a key role in determining potential benefits of screening, underlying disease prevalence also plays a role. Both CKD and ESKD have been associated with an increased incidence of certain cancers, some of which may be amenable to screening strategies. The degree to which incidence estimates reflect a true increase versus a detection bias remains unclear.

Membranous nephropathy is the most commonly reported glomerular disease in patients with cancer. The evidence base for this assertion is modest and, in the largest study of 240 patients with membranous nephropathy, 10% of patients were found to have cancer that was more commonly found in the lung, prostate, and stomach. In this case series, 21 cases of cancer were identified at the time of kidney biopsy, and the remaining 3 cases were identified within 1 year post biopsy. Overall, the standardized incidence ratio of cancer in patients with membranous nephropathy has been estimated to be 2.25, with 80% of cancers detected before or at the time of kidney biopsy. From case-series data, the cancers that have been most often associated with membranous nephropathy include lung, gastric, renal cell, prostate cancers, and thymomas. ^, Other cancers that have been reported with membranous nephropathy include breast, colorectal, pancreatic, esophageal, and hepatic cancers. ^, In patients older than 65 years or with a significant smoking history, the incidence of malignancy with membranous nephropathy may be higher.

Malignancy-associated membranous nephropathy may be due to one of several mechanisms that lead to subepithelial deposits forming in the glomerular basement membrane. Tumor cells may shed antigens that can do the following: 1. form circulating immune complexes that become trapped in the capillary wall and elicit an immune reaction in a subepithelial location; 2. form immune complexes that initially deposit in a subendothelial location but dissociate and reform in a subepithelial location; and 3. deposit in a subepithelial location (depending on the size and charge of the antigen) where they can react with a circulating antibody. A fourth mechanism may be related to an infection with an oncogenic virus or altered immune function, which can simultaneously lead to both malignancy and membranous nephropathy.

Distinguishing primary (idiopathic) versus secondary (malignancy-related) membranous nephropathy is challenging but perhaps made easier by the recognition that a high percentage of patients with primary membranous nephropathy have an elevated circulating level of autoantibodies to the podocyte transmembrane glycoprotein M-type phospholipase A2 receptor (anti-PLA2R) or increased expression of anti-PLA2R on kidney biopsy. Fig. 57.2 lists features of malignancy-associated membranous nephropathy compared with primary membranous nephropathy that can be helpful in diagnosis. There should be a high index of suspicion for an underlying malignancy in patients with a new diagnosis of membranous nephropathy, especially if there are other risk factors for cancer. However, the most efficient and cost-effective method for screening these patients has not been identified. It is recommended that age- and sex-appropriate cancer screening be performed on patients with a new diagnosis of membranous nephropathy. In those patients in whom no malignancy is found on initial screening, follow-up vigilance for cancer should be maintained because there are case reports of cancers being detected up to 5 years after the initial diagnosis of membranous nephropathy. Whether these cancers are directly related to the occurrence of glomerular disease is unknown.

Malignancy-associated membranous nephropathy may resolve with surgical removal of the tumor or with effective chemotherapy. In the absence of curative therapy, standard therapy for membranous nephropathy can be attempted but may have a lower likelihood of inducing remission when compared with patients with primary membranous nephropathy.

There are reported associations among solid-organ malignancies and glomerular diseases for minimal change disease (MCD; e.g., lung, colorectal, thymoma, and renal cell cancers); focal segmental glomerulosclerosis (e.g., renal cell carcinoma, thymoma); membranoproliferative glomerulonephritis (e.g., lung, renal cell, and gastric carcinomas); IgA nephropathy (e.g., renal cell carcinoma, upper respiratory tract tumors, nasopharyngeal cancers); Henoch-Schönlein purpura (e.g., lung, upper respiratory tract tumors, digestive tract tumors); crescentic glomerulonephritis (e.g., renal cell, gastric, and lung carcinomas); and AA-type amyloidosis (e.g., renal cell carcinoma). These potential associations highlight the fact that in patients presenting with glomerular diseases, clinicians should remain vigilant for the possibilities of an underlying associated malignancy. Age- and sex-appropriate cancer screening should be undertaken in these patients. The association is made stronger if effective therapy for the cancer leads to concomitant remission of the glomerular disease.

Lymphomas and leukemias have been associated with an increased risk of various forms of glomerular diseases. The associations are strongest between Hodgkin disease and MCD and chronic lymphocytic leukemia (CLL) and membranoproliferative glomerulonephritis (MPGN).

Plasma Cell Dyscrasia

Plasma cell dyscrasias are associated with a host of glomerular diseases and nephrotic syndrome, the most common of which is AL amyloidosis; AL amyloidosis has been found in 5% to 11% of patients with myeloma at autopsy. Other glomerular diseases include light-chain deposition disease (up to 50% of these patients are found to have myeloma) and heavy-chain deposition disease (up to 25% of these patients have myeloma). In AL amyloidosis, the predominant light-chain isotope is lambda (λ), whereas in light-chain deposition disease, it is kappa (κ). The diagnosis of glomerular diseases in a patient with suspected myeloma is facilitated by finding urine albumin excretion rates >2 g/day, and a kidney biopsy is indicated in these patients (see Fig. 57.3 ). Several rarer forms of glomerular disease can be found in the presence of a plasma cell dyscrasia and paraproteinemia. In all cases, effective chemotherapy or HSCT can lead to remission of the glomerular disease.

An approach to the differential diagnosis of kidney disease associated with paraproteinemias.

AH, Amyloid heavy chain; AHL, amyloid light and heavy chain; AIN, acute interstitial nephritis; AL, amyloid light chain; ATI, acute tubular injury; BRAFi, v-raf murine sarcoma viral oncogene homolog B1 inhibitor; C3, complement 3; FSGS, focal segmental glomerulosclerosis; HCDD, heavy chain deposition disease; ICPi, immune checkpoint inhibitor; Ig, immunoglobulin; IgG 1&4, immunoglobulin G types 1 and 4; IgM, immunoglobulin M; LCDD, light chain deposition disease; LHCDD, light-heavy chain deposition disease; MIDD, monoclonal immunoglobulin deposition disease; mTORi, mammalian target of rapamycin inhibitor; PGMIDD, proliferative N with monoclonal IgG deposits; TLS, tumor lysis syndrome; TMA, thrombotic microangiopathy; TNF, tumor necrosis factor.

Disorders of water homeostasis leading to hyponatremia are among the most common electrolyte disorders encountered in patients with cancer, with a wide range of prevalence reported in the literature (4%–47%). ^, In patients presenting to the hospital with severe hyponatremia, approximately 14% had an underlying malignancy. Perhaps most importantly, hyponatremia in patients with cancer is associated with poor outcomes; the hospital length of stay is doubled, and mortality (within 90 days) is increased 3- to 5-fold. ^, Whether hyponatremia has a direct link to these poor outcomes is debatable; it is more likely that hyponatremia is a marker of severity of illness, disease progression, and overall debility.

The symptoms attributable to hyponatremia are generally nonspecific (e.g., lethargy, nausea) or neurologic (e.g., headache, confusion, seizures) in origin, and the diagnosis relies on correlating laboratory studies with these symptoms. Elucidating the cause of hyponatremia requires careful history taking, physical examination, and judicious laboratory evaluation. An approach to the patient with hyponatremia and cancer is shown in Fig. 57.4 . Clearly understanding the cause of a patient’s hyponatremia is critical in tailoring appropriate therapy.

Diagnostic approach to hyponatremia in the patient with cancer.

The syndrome of inappropriate antidiuretic hormone secretion (SIADH) is a common cause of hyponatremia in the patient with cancer and has numerous causes, ranging from paraneoplastic to medication induced. Paraneoplastic causes include small cell lung cancer (most common and seen in 10%–15% of patients) and head and neck tumors. Case reports have described SIADH with numerous other solid-organ malignancies, from brain to sarcoma. Hyponatremia in these patients typically develops slowly and insidiously, and patients rarely have symptoms unless their serum sodium level is <125 mmol/L. Serial measurements of ADH, at least for small cell lung cancer, reflect the state of the cancer, and levels fall with remission and rise with recurrence. Many chemotherapeutic agents can also lead to SIADH. These include cyclophosphamide, cisplatin, vinblastine, and vincristine. Many of the associations between chemotherapeutic agents and hyponatremia are derived from case reports, and the causal linkage between the drug and SIADH is not conclusive. Of note, in many cases, chemotherapeutic agents (including cyclophosphamide) are administered with substantial volumes of IV hydration, which, depending on the levels of antidiuretic hormone(ADH) and urine osmolality, may lower serum sodium concentation further.

Therapies for hyponatremia in the patient with cancer are similar to those for other causes of hyponatremia and should be guided by three factors: 1. the presence of hyponatremia-related symptoms; 2. the duration of hyponatremia; and 3. the volume status of the patient. When possible, removal or correction of the underlying cause should always be pursued. In the vast majority of cases, hyponatremia associated with cancer is chronic in nature; thus, correction rates should be slow (no more than 6–8 mmol/L/day). In the presence of severe neurologic symptoms (e.g., seizures, obtundation, coma), 3% hypertonic saline can be administered as small boluses or continuous infusion to raise the serum sodium level by 4 to 6 mmol/L, which usually improves these symptoms dramatically. Fluid restriction (typically defined as oral intake 500 mL less than urine output) is a reasonable treatment for patients with mild and transient hyponatremia due to SIADH. However, given the need for hydration protocols with several chemotherapies, as well as the need to maintain adequate nutrition, fluid restriction can be extremely challenging. In these cases, as well as those cases of SIADH in which the urine osmolality is high (>600 mOsm/kg), the use of vasopressin type 2 receptor antagonists (e.g., conivaptan [IV only] or tolvaptan [oral only]) can be considered. These drugs block the action of antidiuretic hormone (ADH) in the distal tubule and allow for the excretion of more dilute urine with increased free water excretion. Tolvaptan is contraindicated in patients with hypovolemic hyponatremia, and volume depletion and/or in patients who cannot perceive or respond to thirst. In addition, given concerns for hepatoxicity with tolvaptan, the current recommendation is to limit its use to 1 month or less. Finally, oral urea administration can also induce a free water diuresis and can be considered as a therapy for SIADH refractory to fluid restriction. Relatively poor palatability of commercially available urea formulations can limit their utility in patients with cancer, many of whom have impaired taste and a general propensity toward nausea and vomiting.

Hypokalemia is commonly encountered in patients with cancer. The causes of hypokalemia are often overlapping and multifactorial in a given patient ( Table 57.7 ). Unique causes in the patient with cancer include the following: 1. tubular damage associated with chemotherapeutic agents and antimicrobials (e.g., cisplatin, ifosfamide, amphotericin B, aminoglycosides); 2. tubular damage associated with a light-chain−induced proximal tubulopathy, such as in multiple myeloma; 3. ectopic adrenocorticotropic hormone (ACTH) syndrome; 4. lysozymuria associated with hematologic malignancies; and 5. chemotherapy-induced gastrointestinal losses (vomiting and diarrhea).

Transcellular shifts that occur post phlebotomy, while the blood is in a sample tube, can lead to spurious hypokalemia, especially when the patient has marked leukocytosis (>100,000/μL) or when the blood is kept at room temperature for prolonged periods. When spurious hypokalemia is suspected, rapid separation of the plasma and storage of blood at 4°C will usually confirm a normal serum potassium level.

The syndrome of paraneoplastic ACTH production can be encountered with several tumors, with the most common causes being bronchial carcinoid tumors, small cell lung carcinoma, lung adenocarcinoma, thymic cancers, and pancreatic tumors. Patients typically present with signs of severe hypercortisolism—diabetes, bone loss, hyperlipidemia, HTN, and, depending on the length of symptoms, Cushingoid habitus—along with severe hypokalemia. The hypokalemia results from excessive cortisol production, which overwhelms the distal tubular mechanism that normally limits mineralocorticoid receptor activation by cortisol (11β-hydroxysteroid dehydrogenase). Excessive production of cortisol essentially creates a syndrome similar to that of excessive aldosterone secretion, with resulting HTN and hypokalemia. Paraneoplastic ACTH production can be diagnosed through measurement of excessive ACTH, hypercortisolism, and localization of the primary tumor with radiographic studies. Optimal management requires control of the primary tumor, which may or may not be possible, so drugs that antagonize glucocorticoid synthesis (metyrapone and ketoconazole) are often required. Rarely, adrenalectomy may be required.

A rare form of hypokalemia associated with acute myelogenous leukemia (subtypes M4 and M5) has been reported. It is usually associated with other electrolyte and acid-base disorders including hyponatremia, hypocalcemia, hypophosphatemia, hypomagnesemia, and a non–anion gap acidosis. ^, The magnitude and breadth of these electrolyte issues suggest a global tubular defect and inappropriate renal potassium losses. This defect has been postulated to be secondary to increased tumor-derived lysozyme secretion and lysozymuria-induced tubular damage.

Hyperkalemia in patients with cancer is often seen in association with AKI, rhabdomyolysis, or tumor lysis syndrome. Less common causes include adrenal insufficiency (secondary to adrenal metastases) and drugs (e.g., heparin, ketoconazole, trimethoprim, calcineurin inhibitors, nonsteroidal anti-inflammatory drugs). An important consideration in patients with marked leukocytosis or thrombocytosis is pseudohyperkalemia, which results from a shift of potassium out of platelets or white blood cells post phlebotomy and after clotting of the sample. If pseudohyperkalemia is suspected, plasma rather than serum samples should be used for laboratory measurement.

Hypercalcemia associated with malignancy occurs in up to 30% of patients with cancer. Hypercalcemia portends a poor prognosis and limited lifespan. ^, Symptoms attributable to hypercalcemia are generally nonspecific (e.g., lethargy, nausea, vomiting, confusion) and are related to both the absolute level of serum calcium and rapidity at which the disorder develops. Patients presenting with serum calcium concentrations above 13 mg/dL are generally symptomatic.

There are three mechanisms that describe the pathophysiology in most cases of malignancy-associated hypercalcemia ( Fig. 57.5 ). The most common cause is paraneoplastic production of parathyroid hormone related peptide (PTHrP). PTHrP is similar to parathyroid hormone (PTH) and acts to increase bone resorption (increases bone calcium release) via activation of osteoclasts and increase reabsorption of calcium in the renal tubules. However, PTHrP is less likely than PTH to stimulate 1,25-dihydroxyvitamin D production. Therefore PTHrP does not increase intestinal calcium absorption. Hypercalcemia is typically accompanied by hypophosphatemia, reduced tubular reabsorption of phosphorus, enhanced tubular reabsorption of calcium, and increased excretion of nephrogenous cyclic adenosine monophosphate (cAMP), reflecting the PTH-like actions of PTHrP. The second most common mechanism of malignancy-associated hypercalcemia is local osteolytic effects of the cancer and is seen in patients with myeloma, breast cancer, lymphomas, and other cancers. For example, myeloma has a predilection to grow in bone environments, facilitated by the secretion of bone reabsorbing factors such as interleukin-1 (IL-1), IL-6, and tumor necrosis factor, leading to extensive bone destruction, lytic bone lesions, and the release of calcium. Typically, these patients will have advanced cancers, with skeletal surveys demonstrating lytic bone lesions. The third mechanism of malignancy-associated hypercalcemia is due to activation of vitamin D by the tumor itself, which can be seen in lymphomas and, rarely, with myeloma. These patients will have elevated serum 1,25 dihydroxyvitamin D and suppressed PTH concentrations. Elevated activated vitamin D concentrations facilitate increased gastrointestinal calcium absorption and increased bone calcium release, both leading to hypercalcemia.

Causes of malignancy-associated hypercalcemia.

PTHrP, Parathyroid hormone related peptide.

The diagnostic approach to the patient with hypercalcemia is presented in Fig. 57.6 . Once hypercalcemia is confirmed with measurement of an elevated serum ionized calcium, the diagnostic algorithm centers on measurement of PTH. Serum PTH concentrations are suppressed in patients with malignancy-associated hypercalcemia, with the rare exceptions of tumors such as parathyroid adenocarcinomas, which produce excessive PTH. In patients with suppressed PTH, measurement of PTHrP and 25-hydroxy vitamin D is the next diagnostic step and allows for accurate diagnosis in most cases.

Diagnostic algorithm for the patient presenting with hypercalcemia.

Measurement of PTH is the first step in the diagnostic approach. Patients with malignancy-associated hypercalcemia typically have low levels of PTH due to feedback inhibition of the high blood calcium levels on the parathyroid gland. In patients with low levels of PTH, measurement of PTHrP and 1,25-dihydroxyvitamin D (1,25(OH) ₂ D) levels will allow classification of patients into several diagnostic groupings. In patients with elevated PTH levels, a 24-hour urine collection for calcium and creatinine levels should be obtained. CaSR, Calcium-sensing receptor; FHH, familial hypocalciuric hypercalcemia; iPTH, intact parathyroid hormone; PTH, parathyroid hormone; PTHrP, parathyroid-related peptide; 25(OH)D, 25-hydroxy vitamin D.

From Reagan P, Pani A, Rosner MH. Approach to diagnosis and treatment of hypercalcemia in a patient with malignancy. Am J Kidney Dis. 2013;63(1):141–147.

Patients with hypercalcemia are usually severely volume depleted, as hypercalcemia frequently leads to nephrogenic diabetes insipidus, as well as nausea and vomiting. In addition, hypercalcemia, via activation of the calcium-sensing receptor in the thick ascending limb of the loop of Henle, leads to natriuresis and worsening of the volume depletion. Thus the cornerstone of the treatment of hypercalcemia is aggressive restoration of the extracellular volume, as well as induction of natriuresis with IV fluids (typically, 0.9% saline), facilitating excretion of excess calcium. Loop diuretics should only be added in the presence of volume overload and are not a routine part of management. Data support the use of bisphosphonates as highly effective agents in the management of patients with malignancy-associated hypercalcemia, largely due to their effect to inhibit osteoclast activity. Both pamidronate and zoledronic acid are U.S. Food and Drug Administration (FDA) approved for this indication; dosing guidelines are shown in Table 57.8 . These agents are generally well tolerated, with the most common side effect being infusion-related fever, but serious nephrotoxicity of these agents has been described including glomerular lesions and tubular toxicity. Thus the serum creatinine should be monitored before each dose of bisphosphonate and the drug withheld in patients with declining kidney function. Urine albumin excretion should be monitored at 3- to 6-month intervals, and the drug should be held if albuminuria develops. Patients with PTHrP-mediated hypercalcemia may remain mildly hypercalcemic, even after bisphosphonate administration, due to the effects of PTHrP on renal calcium excretion. An alternative antiresorptive agent is denosumab, a monoclonal antibody directed against the receptor activator of nuclear factor kappa B ligand (RANKL). Denosumab does not require dose adjustment for impaired kidney function and has demonstrated benefit in reducing skeletal-related events in patients with metastatic cancer. For patients with hypercalcemia secondary to tumors producing 1,25-dihydroxyvitamin D, corticosteroid therapy may be of benefit. Finally, in patients with kidney failure and hypercalcemia, hemodialysis with a low calcium dialysate can be beneficial.

Cancer can lead to disorders of phosphate homeostasis via a number of mechanisms. Hypophosphatemia can result in the setting of malnutrition and cachexia and with vitamin D deficiency; these patients typically have low normal serum calcium, low vitamin D, and elevated PTH concentrations. In addition, patients with hypercalcemia due to elevated PTH or PTHrP levels may have low-normal serum phosphate concentrations due to the effects of these hormones on renal phosphorus handling. In some cases, proximal tubular injury (either through chemotherapy such as cisplatin or ifosfamide or due to light-chain–induced tubular injury from myeloma) can lead to phosphaturia and hypophosphatemia.

A rare and unique form of hypophosphatemia is also encountered in patients with cancer, the syndrome of tumor-induced osteomalacia (TIO), or oncogenic osteomalacia. In this syndrome, tumor production of phosphaturic factors such as fibroblast growth factor-23 (FGF-23) results in renal phosphate wasting, hypophosphatemia, and osteomalacia. Tumor types producing these factors include chondrosarcoma, osteoblastoma, and hemangiopericytomas. These tumors may be small and indolent and may require extensive and specialized imaging for localization. Patients with TIO typically have normal serum calcium concentrations, and the diagnosis is facilitated by measuring the percentage tubular reabsorption of phosphate or the tubular maximum for phosphate, corrected for the GFR. Once renal phosphate wasting is confirmed, measurement of elevated serum concentrations of FGF-23 can be diagnostic of TIO. Importantly, 1,25-dihydroxyvitamin D levels are low because FGF-23 inhibits the activity of the 1α-hydroxylase. Therapy rests on surgical removal of the tumor, which is usually curative.

Hypomagnesemia is a relatively common electrolyte disorder in patients with cancer; the unique causes in this patient population relate to chemotherapeutic agents such as cisplatin and cetuximab. Otherwise, patients usually develop hypomagnesemia due to gastrointestinal losses, especially with prolonged diarrhea, or renal losses due to diuretic therapy. Other drugs associated with hypomagnesemia in the patient with cancer may include proton pump inhibitors, aminoglycoside antibiotics, calcineurin inhibitors (in patients with stem cell transplants), and amphotericin B.

Cisplatin causes magnesium wasting in more than 50% of treated patients, and the incidence increases with the cumulative dose. Renal magnesium wasting continues after cessation of the drug for several months but can persist for years. The occurrence of magnesium wasting does not correlate with cisplatin-induced AKI. Cetuximab, a monoclonal antibody against the epithelial growth factor receptor (EGFR), has been associated with hypomagnesemia due to renal losses in up to 30% of patients receiving this drug. The mechanism of magnesium wasting during treatment with cetuximab is associated with the blockage of the EGFR-dependent transient receptor potential channel 6 (TRPM6) in the nephron, which results in insufficient activation of the TRPM6 epithelial ion channel and magnesium wasting in the distal convoluted tubule. Cessation of therapy with cetuximab leads to reversal of the magnesium wasting and return of the serum magnesium concentration to normal.

The incidence of AKI with HSCT is quite varied (ranging 15%–73%) due to the AKI definition used, the type of HSCT performed (allogenic vs. autologous), and the chemotherapeutic conditioning regimen used (high dose vs. reduced intensity). The incidence and clinical course of AKI following allogeneic HSCT are well documented. In one study, AKI developed in 53% of patients, of whom half required RRT. Subsequent studies have noted that the incidence of moderate to severe AKI (defined as a doubling of the baseline serum creatinine concentration) ranges from 36% to 78% with allogenic HSCT. ^{, ,} Dialysis was required in 21% to 33% of these patients, who suffered a mortality of 78% to 90%. The incidence of AKI with autologous HSCT is less common. In patients with breast cancer, autologous HSCT was associated with the development of moderate to severe AKI in 21% of patients and a mortality rate of 18%. The lower AKI incidence in this group may be due to calcineurin inhibitor avoidance.

Nonmyeloablative HSCT is also associated with a lower incidence of AKI, which is attributed to the milder conditioning regimen used. One study in nonmyeloablative HSCT patients demonstrated a cumulative AKI incidence of 40.4% at 4 months, and dialysis was required in 4.4% of patients. In contrast to myeloablative HSCT, AKI in these patients was usually associated with calcineurin inhibitor exposure. In addition, the development of AKI during nonmyeloablative HSCT occurs over the first 3 months, which is in contrast to myeloablative HSCT, in which AKI typically develops in the first 3 weeks.

As with AKI incidence, the risk for requiring KRT after HSCT varies widely. Overall, approximately 5% of HSCT recipients with AKI require dialysis, which is associated with increased mortality risk. However, the need for dialysis ranges from 0% to 30%, and the risk is higher with myeloablative therapy. Outcomes in patients requiring acute KRT are difficult to quantify accurately. Most studies have reported that acute dialysis in this group is associated with an extremely high mortality rate, ranging from 80% to 100%. ^{, , , ,}

Common risk factors for and causes of AKI after HSCT include volume depletion, sepsis, nephrotoxic medication exposure, SOS, and GVHD ( Table 57.9 ). ^{, , ,} Because of a propensity for increased gastrointestinal fluid losses and poor oral intake, HSCT patients are highly susceptible to volume depletion and prerenal AKI. GVHD is unique to HSCT and causes tissue and endothelial damage via T cell and cytokine-mediated injury. The gastrointestinal mucosa is a common site of injury by GVHD, contributing to inadequate fluid intake and increased fluid losses. SOS is also an independent risk factor for AKI and results in a picture similar to hepatorenal syndrome, with sodium avidity and edema formation. Medications commonly prescribed that cause AKI include vancomycin, aminoglycosides, acyclovir, and amphotericin. These agents cause AKI by direct tubular toxicity, acute interstitial nephritis, and crystal-related tubular injury (acyclovir). Calcineurin inhibitors can lead to renal arteriolar vasoconstriction and have been associated with the development of TMA.

Hepatic Sinusoidal Obstruction Syndrome

AKI following HSCT can also be due to acute liver injury. Hepatic SOS, formerly known as venoocclusive disease, is a commonly encountered complication after HSCT and has been implicated as an independent risk factor for the development of AKI. It typically occurs within 30 days of HSCT, and the incidence varies widely, ranging from 0% to 62.3% (mean, 13.7%). Although the pathophysiology of AKI in the SOS setting is poorly defined, it appears to be a variant of hepatorenal syndrome (hemodynamic AKI). Hepatic sinusoidal obstruction occurs due to iatrogenic damage to sinusoidal endothelial cells and hepatocytes, which results in fibrous narrowing of small hepatic venules and sinusoids. Injury appears to be triggered by the pretransplantation cytoreductive regimen and is more common after allogeneic than autologous HSCT. Furthermore, cytokine release and glutathione depletion in this setting cause hepatocellular necrosis and fibrosis. ^, The development of SOS is usually associated with pretreatment with busulfan, cyclophosphamide, and/or total body irradiation.

Criteria have been established for the diagnosis of hepatic SOS. The Seattle criteria require the presence of at least two of three manifestations—jaundice, painful hepatomegaly, and fluid retention or weight gain within 20 days of HSCT. ^, The Baltimore criteria require a bilirubin level of >2 mg/dL, plus two or more of the following: painful hepatomegaly, weight gain >5%, or ascites, within 21 days of HSCT. Hepatic SOS severity is graded as mild (the illness fulfills SOS diagnostic criteria but is self-limiting), moderate (SOS subsides with diuretic or analgesic treatment), and severe (SOS persists for more than 100 days, or death).

Hepatic SOS is characterized by painful hepatomegaly, jaundice, fluid retention, and ascites. Patients have hyperdynamic vital signs along with hyponatremia, oliguria, and low fractional excretion of sodium. Urinalysis reveals minimal proteinuria and muddy brown granular casts as a result of the toxic tubular effects of bile acids and bilirubin in the urine. Hypervolemia is usually resistant to diuretics, and spontaneous recovery is rare.

Risk factors that predispose to the development of AKI include a baseline serum creatinine >0.7 mg/dL, weight gain, hyperbilirubinemia, and exposure to amphotericin B, vancomycin, or acyclovir. AKI in this setting is associated with mortality. In patients who require KRT, the mortality rate approaches 80%.

The management of hepatic SOS has included a number of agents. Small trials using infusions of prostaglandin E, pentoxifylline, and low-dose heparin have shown promise in the prevention and treatment of SOS. Smaller trials with defibrotide, an antithrombotic and fibrinolytic agent, have also shown benefit in patients with SOS, especially when used within the first 2 days of HSCT. ^, Beneficial effects are thought to be due to the prevention of platelet aggregation and antiinflammatory properties. However, bleeding risk is a concern and must be closely monitored.

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related posts:

Diabetes Initial Hospitalization Care Clinical Syndromes of Metabolic Acidosis Glomerulogenesis and De Novo Nephrogenesis in Medaka Fish: An Evolutionary Approach Obstructive uropathy in Chronic Kidney Disease

Stay updated, free articles. Join our Telegram channel

Join