Molecular Mechanisms And Biomarkers Associated With Chemotherapy-Induced AKI Ⅱ

5. Biomarker-Guided Diagnosis 

The Acute Disease Quality Initiative (ADQI) group proposed an extended definition of AKI, which includes AKI biomarkers classified as functional and damage biomarkers according to the AKI aspects, which they recapitulate. SCr level and urine output are two functional biomarkers widely employed in clinical practice, but they have several limitations [140]. Indeed, in healthy patients, the sCr levels increase only if at least 50% of the functional nephrons are lost, whereas during critical illness (i.e., in ICU patients), many confounding factors likely play a role in creatinine decrease (i.e., cirrhosis, hyperbilirubinemia, fluid overload, elderly patients, muscle wasting), making eGFR based on creatinine unreliable to correctly estimate kidney function [141,142]. These caveats limit the ability of sCr measurements to diagnose early AKI. In addition, creatinine assessment does not clarify to what extent subclinical AKI episodes contribute to shortening the kidney lifespan and CKD. Likewise, urine output can be influenced by hypovolaemia and the use of diuretics, resulting in a relatively low specificity of this parameter [143]. Cystatin C is a low molecular weight molecule produced by all epithelial cells. It is freely filtered by the glomerular filtration barrier and completely reabsorbed by proximal TEC in healthy individuals [144]. Therefore, it is detected in the urine only following tubular epithelial injury. Unlike sCr, its measurement is not confounded by acute and chronic illness, changes in diet, and decreased muscle mass, rendering it a better predictor of mortality compared to the sCr-based eGFR calculations [142,145–148].

NEW HERBAL FORMULATION FOR AKI

In contrast to functional biomarkers, damage-associated biomarkers are specific to tubular injury and can potentially identify patients at higher risk of developing AKI. This is particularly relevant in clinically silent cases or in subclinical AKI, where creatinine level and urine output measurements are unreliable. Early kidney damage does not often cause a relevant change in urine output or sCr, missing the diagnostic criteria of AKI. If the damage is severe or prolonged over time (i.e., progression to AKD), overcoming the renal function reserve, a GFR reduction, and a subsequent alteration of sCr and urine output will be observed. However, kidney damage without any function loss also affects outcomes [149,150]. Importantly, both functional impairment (sCr level elevation and/or urine output decline) and the presence of damage biomarkers indicating structural damage is associated with a marked mortality increase in specific clinical contexts, such as those associated with critical illness [151–154]. Therefore, identifying patients at high risk of chemotherapy-associated AKI development is a challenge, and damage biomarkers offer a potential solution to guide clinicians in their therapeutic decisions to prevent AKI outcomes (Table 2) [155].

Damage biomarkers. Nephrotoxins contribute to approximately 30% of AKI cases in critically ill patients, and mismanagement from excessive nephrotoxic treatment coupled with unnecessary exposure is often a contributing factor [156,157]. As the use of nephrotoxic agents represents one of the few modifiable risk factors for AKI, clinicians must be able to rapidly identify patients at high risk for drug-induced kidney injury.

Several biomarkers with different anatomical origins, kinetics, function, and timing after the insult have been identified and used for clinical and/or research purposes. These molecules are usually produced after a parenchymal lesion and released in the urine due to tubular reabsorption failure. Several neutrophil gelatinase-associated lipocalin (NGAL) isoforms are released by the kidney (thick ascending limb and collecting ducts) and by immune cells [158]. In healthy individuals, the concentration of NGAL in the urine is very low, but it increases considerably after an insult, showing high sensitivity and specificity for predicting AKI in patients with a previously normal kidney function [159,160], as well as in patients with CKD [161,162]. Several in vivo studies evaluating the NGAL response to known nephrotoxins, including aminoglycosides, amphotericin B, cisplatin, paraquat poisoning, methotrexate, nonsteroidal anti-inflammatory drugs, and vancomycin, are described [130–132,163,164].

NGAL measurement in cisplatin and amphotericin-associated AKI was effective in the early detection of AKI, performing better than sCr, but it was not so evident in chronic cisplatin-associated AKI [131,138].

Kidney Injury Molecule-1 (KIM-1) is a type 1 transmembrane glycoprotein that is markedly upregulated in the injured proximal tubular epithelium after ischemic injury or nephrotoxic exposure and shed into the tubular lumen [165]. KIM-1 is suggested to be a more sensitive/specific biomarker for detecting amphotericin and cisplatin-induced AKI [166]. Urinary KIM-1 and NGAL could efficiently discriminate patients with or without vancomycin-associated AKI earlier than sCr, and their combination showed fair discrimination compared with the individual biomarkers [137]. Further studies in patients undergoing platinum chemotherapeutics, urinary levels of KIM-1, NGAL, and cystatin C showed a statistically significant early increase after treatment initiation, preceding sCr rise, in AKI patients [133,167]. Accordingly, a Canadian study showed the ability of KIM-1 and NGAL to provide early AKI detection and their utility in identifying patients at risk of long-term AKI complications in a cohort of pediatric oncologic patients [130]. Following this, the US Food and Drug Administration (FDA) approved KIM-1 as a nephrotoxic biomarker for different drugs in use, resulting in several quantitative KIM-1 measurements having been developed [168–170]. Liver-type Fatty Acid-Binding Protein (L-FABP) is mainly produced by the liver but also by other organs, such as the kidney. L-FABP can be detected in the urine predicting AKI in patients after cardiac surgery or in critically ill patients, apparently better than NGAL [171,172]. Further studies demonstrated an additional benefit of using biomarkers (NGAL, KIM-1, L-FABP) in conjunction with the functional criteria of sCr and urine output, as their combination improves the prediction of worse outcomes [155]. Other biomarkers are represented by the lysosomal enzyme N-acetyl-b-D-glucosaminidase (NAG) and the cytosolic protein lactate dehydrogenase (LDH) [173,174]. The relationship between NAG and drug-induced kidney disease has been evaluated in several studies [135,163,175], focusing mainly on aminoglycoside and cisplatin use, demonstrating that higher NAG levels exhibited a relationship with nephrotoxicity during therapy with aminoglycosides and with a methotrexate and cisplatin combination [134–136]. In a proof-of-concept study, damage urinary biomarkers (KIM-1, NGAL, and NAG) provided an early identification of aminoglycoside-related proximal tubule renal toxicity, enabling treatment adjustment and the identification of infants at risk of long-term kidney impairment [163].

Other biomarkers of nephrotoxicity include gamma-glutamyl transpeptidase (GGT), Glutathione S-transferase (GST), and alanine aminopeptidase (AAP). GGT and NAG predicted AKI in critical care patients, especially in the ICU setting [176], and urinary concentrations of NAG increased in mice exposed to gentamicin or lithium [136,177]. Recently, urinary dickkopf-3 (DKK3), a stress-induced tubular epithelial-derived profibrotic glycoprotein, has been shown to predict postoperative AKI and provide information about ongoing tubulointerstitial fibrosis and short-term eGFR loss [177–179]. The RUBY study demonstrated that elevated urinary CCL14 predicts persistent AKI in a large heterogeneous cohort of critically ill patients with severe AKI [180]. However, there is still no evidence of the potential application of DKK3 and CCL14 in the context of chemotherapy-induced AKI.

Cell cycle arrest biomarkers. Unbiased screening for urinary biomarkers revealed that cell cycle arrest markers were among the top candidates capable of predicting subsequent AKI [181]. Cell cycle arrest of kidney TEC is involved in the pathogenesis of AKI [182]. As G1 cell cycle arrest due to cell stress is one of the first events during AKI, metalloproteinase inhibitor 2 (TIMP2) and insulin-like growth factor-binding protein 7 (IGFBP7) are detectable in the urine very early during AKI development [181,183]. In the Sapphire study [184], combined TIMP2 and IGFBP7 measurements demonstrated an excellent ability to predict moderate to severe AKI, and it was superior to all the other existing AKI markers, considerably improving patient risk stratification [185,186]. The FDA subsequently approved a test incorporating this marker combination (termed Nephro-Check) for clinical use. Several trials have shown that urinary TIMP2 and IGFBP7 levels predict AKI development, kidney recovery, and patient mortality [181,183,187,188]. The PrevAKI trial was the first study to investigate TIMP2 and IGFBP7 in diagnosing AKI associated with cardiac surgery [183,189]. Biomarker level ([TIMP-2]·[IGFBP7] (0.3 ng/mL)2/1000) and time point of measurement (4 h after cardiopulmonary bypass) resulted in a successful predictive performance of those patients at high risk of AKI development. A similar biomarker-guided intervention was applied to prevent AKI after major surgery in the BigPAK trial [190]. The development of moderate as well as severe AKI, the incidence of sCr increase, ICU, and hospitalization length were all significantly reduced in patients whose biomarker levels were within the range of 0.3–2.0 (ng/mL)2/1000. This suggests that patients with higher biomarker levels may have suffered an extended period of kidney stress, resulting in a progression to AKI and AKD. Early biomarker-based prediction of AKI followed by the implementation of the KDIGO (Kidney Disease: Improving Global Outcomes) care bundle reduced AKI severity [190].

Based on the Nephro-Check test results, in both PrevAKI and BigPAK studies, patients with a higher risk of AKI had benefited from the decision to avoid nephrotoxic treatment [189,190]. Thus, it is conceivable that implementing a biomarker-based approach with the detection of the high-risk population might be beneficial for preventing AKI. These biomarkers can also be used to predict adverse long-term outcomes because their early measurement in the setting of critical illness may identify patients with AKI at increased risk of death or KRT in the following months [187]. Moreover, the best results can be achieved by combining different biomarkers. High KIM-1, NGAL, and [TIMP-2]·[IGFBP7] values identified patients with vancomycin-associated AKI earlier than sCr [137,139]. A drug combination that has gained recent attention for an additive risk of nephrotoxicity is vancomycin plus piperacillin-tazobactam. To establish whether kidney injury associated with this combination is a valid clinical concern, [TIMP-2]·[IGFBP7] have been employed. Patients treated with the combination therapy showed higher levels of [TIMP- 2]·[IGFBP7] in comparison to those treated with vancomycin monotherapy, associated with increased long-term adverse outcomes [191]. Collectively, this evidence suggests the benefit of damage biomarker measurement in identifying nephrotoxic AKI early (Figure 3). Practical considerations for the implementation of these biomarkers for predicting and detecting chemotherapy-induced kidney injury need to be evaluated. In particular, a better understanding of the appropriate concentration for each biomarker for each nephrotoxic drug or drug class that increases the risk for drug-induced kidney injury needs to be developed. 

6. Management 

General measures. Given that no specific evidence is available to suggest that AKI in cancer patients should be managed differently from other causes of AKI, strategies based on KDIGO are appropriate for risk- and stage-based prevention and management of AKI [140]. Cancer patients are particularly at risk for infection and sepsis [18,192]. Thus, early detection and management of sepsis, including source control of the infection (e.g., removal of tunneled central venous catheter systems) and optimized antibiotic use based on the pharmacokinetics and pharmacodynamics changes observed in AKI are essential, particularly in patients with neutropenia [18,193]. A review of patients' charts to ascertain the cumulative exposure to chemo- and immunotherapeutic agents and other medications is important to assess the risk of nephrotoxicity and other less common therapy-associated injuries (e.g., thrombotic microangiopathy, tubulointerstitial nephritis, glomerular diseases, and intratubular obstruction from medications) must also be considered [44]. Notably, the risk of nephrotoxicity increases from cumulative exposure to chemotherapeutic agents and other medications [194,195]. The risk of AKI increases with the number of nephrotoxic drugs used, and all potentially nephrotoxic agents that can be stopped should be discontinued [146,196]. Indispensable agents should only be used as long as needed and only at required doses. Careful monitoring of drug concentrations is also mandatory (for example, vancomycin) [44]. 

Patients with chemotherapy-induced AKI may present with symptoms and signs resulting directly from diminished kidney function. These typically include edema, hypertension, decreased urine output, or anuria in severe AKI [1]. However, many patients do not show any symptoms, and the only sign of diminished kidney function may be an increase in creatinine detected by laboratory tests without an overt AKI [1]; otherwise, signs and symptoms are indistinguishable from AKI from other etiologies. SCr remains the only laboratory value used in operative definitions for AKI and the biomarker most used in clinical practice. All subsequent evaluations are directed at determining the underlying cause of AKI to achieve prompt and adequate management. For all patients, the timing of onset often suggests the underlying etiology, albeit sCr concentration should be measured frequently, a goal hardly achievable unless the patient is admitted to the hospital. Careful attention should be given to volume status to avoid hypovolemia, as patients may initially present with relative volume depletion due to fever and gastrointestinal losses as volume resuscitation is rarely performed [197]. Volume management and hemodynamic monitoring are also required at all stages of AKI. Avoiding hyperglycemia is also essential because the filtered glucose increases tubular reabsorption workload and oxidative stress, a process that sensitizes the kidney tubule to injury [198]. Implementation of the 'KDIGO bundle'-consisting of optimizing volume status and hemodynamics, avoiding nephro-toxic drugs, and preventing hyperglycemia in patients at high risk of AKI as identified by biomarkers-can prevent AKI [189].

Kidney replacement therapies. When the severity of AKI necessitates KRT, the jugular veins should be considered as the preferred insertion sites for dialysis catheters. The catheter exit site and anchoring remain visible, and these sites confer a lower risk of infection and thrombosis [199]. Initiation and continuation of dialysis in the cancer patient with AKI should be based on the general clinical condition and overall life expectancy and the personal patient expectations on quality of life after eventual recovery [200]. Hypophosphatemia is common in malnourished cancer patients and those on prolonged continuous KRT and may need to be corrected with supplements to prevent hypophosphatemia-associated complications [201]. Intradialytic seizures may occur in cancer patients on maintenance anticonvulsant therapy due to dialytic removal of the drug, and higher post-dialysis doses may be required to maintain therapeutic serum concentration [202]. Cancer patients are at risk of malnutrition due to various factors, such as prolonged immobilization, catabolic changes, and reduced food intake. Therefore, the current consensus recommendations for the nutritional management of critically ill patients with cancer should be followed [203]. Finally, kidney transplantation is not a valid KRT from the critically ill patient's perspective

AKI to CKD transition. Limiting progression from AKI and AKD to CKD is a crucial issue in chemotherapy-exposed patients [204]. Cancer is strictly linked to AKI and CKD, and the presence of CKD markedly reduces cancer patients' survival [205–207]. Patients with risk factors for CKD (i.e., diabetes, hypertension, obesity, low nephron endowment, and many others) on the verge of receiving chemotherapeutics should be trained adequately about possible CKD onset and progression [2]. All the risk factors for CKD mentioned above should be tightly controlled whenever possible: patients should implement a healthy diet and physical activity, and anemia, high blood pressure, dyslipidemia, and diabetes should be pharmacologically controlled when conservative measures prove to be insufficient [2]. It would be advisable to stop medications that may increase the risk for nephrotoxicity, namely non-steroidal anti-inflammatory drugs, whenever clinically feasible [189]. Expo is sure that iodinated contrast should be limited, too [189]. Concerning cancer patients, all lifestyle modifications and new drug implementation should always be collectively discussed with patients and oncologists. In this setting, nephron overload, the structural adaptations that promote accelerated loss of kidney epithelia in nephrons challenged by hemodynamic and metabolic overload, represents a typical driver of CKD progression and a therapeutic target [208]. Currently, the renin-angiotensin-aldosterone system and SGLT2 (Sodium-glucose Cotransporter-2) inhibitors represent the most effective drugs to slow CKD progression [209–211]. Importantly, there is no evidence to date linking SGLT2 inhibitors and an augmented risk of cancer [212]. 

Future directions. Improvement in AKI diagnosis and treatment remains a significant unmet medical need. Given that AKI is a global health problem, there is an urgent need to train health workers to identify patients at significant risk of kidney disease development and subsequent progression to AKD or CKD. An active and effective proposal should span from health-system surveillance methods to clinical interventions. This should be done by: (1) promoting a stronger collaboration between nephrologists, intensive care specialists, and oncologists; (2) preventing or at least limiting drug-associated AKI through nephrotoxin stewardship, and (3) implementing novel biomarkers aimed at a proper patient classification [213] (Figure 4). The ideal biomarker "for" AKI should be (1) sensitive, it should work as an early predictor of AKI and then be altered following injury in minutes or hours; (2) AKI specific by providing clues regarding the underlying etiology; (3) serve as a prognostic factor; (4) predict the need for KRT; (5) be cost-effective and highly reproducible.

Hence, it is essential to use the best available and novel biomarkers to recognize initial AKI phases and apply protective measures and risk mitigation to avoid worsening of the condition. Finally, even when AKI has fully developed, identifying patients who might progress to AKD or even CKD is important [214]. In these patients, specific biomarkers may help plan the allocation of resources.