How To Determine Kidney Function in Cancer Patients?

Contact: emily.li@wecistanche.com

Ben Sprangers, et al

Abstract

A precise and effificient method for estimating kidney function in cancer patients is important to determine their eligibility for clinical trials and surgery and to allow for appropriate dose adjustment of anti-cancer drugs, especially toxic drugs with a narrow therapeutic index. Since direct measurement of glomerular filtration rate (GFR) is cumbersome, severalformulae have been developed to estimate kidney function. Most of these are based on serum creatinine concentration. Though the CKD-EPI formula is recognized as being the most accurate, there is an ongoing debate on which is the optimal formula for cancer patients. In this review, we provide an overview of different GFR estimating equations for kidney function and the advantages and disadvantages of each method and compare their performance in cancer patients. We discuss the importance of body surface area-indexing and propose a framework for evaluating kidney function in cancer patients.

KEYWORDS: Kidney function; Glomerular filtration; GFR formula; BSA indexing

Click here to get detailed information about Cistanche

1. Introduction

1.1. Importance of evaluating kidney function in cancer patients

Precise estimation of kidney function is important in hematology and oncology to determine eligibility for clinical trials and surgery and to facilitate dose adjustments of chemotherapy, antibiotics, opioid analgesics, and other medications, especially for toxic medications with a narrow therapeutic index. Since many cancer drugs are eliminated by the kidney, dose adjustments are necessary for patients with decreased kidney function to avoid both under-dosing and over-dosing. Kidney dysfunction is common among cancer patients and significant losses of kidney function often occur during cancer therapy [1]. A cross-sectional study evaluating cancer patients found a reduction in the estimated glomerular filtration rate (eGFR) of 13 mL/min/1.73 m2 after 2 years, and 17.7% of patients changed from chronic kidney disease (CKD) Stage 2 to CKD stage 3 or 4 at follow-up [2]. Since it is neither practical nor feasible to determine serum drug concentrations or to directly measure GFR repetitively in daily clinical practice, it is important to determine the most precise and feasible method for evaluating kidney function (e.g. estimate GFR).

Several methods are available to directly measure GFR (Table 1). However, all these methods are labor-intensive, complex, and time-consuming making it impossible to perform these assays on all cancer patients on a regular basis. Inulin clearance is the gold standard but it is only rarely used in clinical practice [3], and alternative and simpler methods have been developed such as ethylenediaminetetraacetic acid, iohexol, iothalamate, and diethylenetriaminepentaacetate clearance methods [4,5]. The only method routinely used in clinical practice to measure kidney function is creatinine clearance calculation, which is based on serum creatinine and urine creatinine concentration in a 24- h collection of urine. This method is problematic as creatinine clearance measurement has not been validated in cancer patients [5] and urine collections are known to be cumbersome and subject to error, especially in the outpatient setting. There are currently no randomized trials supporting the need to systematically perform direct measurement of GFR in oncology. However, direct measurement of GFR should be considered to guide drug dosage adjustment for chemotherapeutics with potentially severe nephrotoxicity and with a narrow therapeutic index, such as cis- or carboplatin, or in patients where the available equations exhibit low accuracy [6].

Currently, there is no consensus regarding the optimal means of estimating GFR to allow for adjustments of chemotherapeutics (Table 2) and to define a patient’s eligibility for novel cancer drug trials. Historically, patients with impaired kidney function have been excluded from phase 1 studies of anticancer drugs because of the perceived increased risk for major dose-limiting toxicity. A recent study demonstrated that 85% of clinical drug trials for the five most common malignancies published in high-impact factor journals excluded the vast majority of patients with CKD [7]. A retrospective analysis of over 10,000 patients from 373 single-agent phase 1 clinical trials found no clinically meaningful increase in grade 3 or 4 non-haematologic, grade 4 hematologic, or any clinically relevant toxicities in patients with mild kidney impairment (defined according to the FDA as CrCl 50e79 mL/min) compared with those with normal kidney function [8]. In recent years, some have advocated that clinical trials be more inclusive of patients with mild to moderate kidney impairment [9].

1.2. Importance of the assay used to estimate kidney function

Kidney function is composed of both glomerular and tubular functions. It is important to realize that all commonly used methods to estimate kidney function only evaluate GFR. The essay that is used to estimate GFR is important since there can be important inter-assay variability. This variability is exemplified by applying the different estimating formulae to determine a patient’s eligibility to receive cisplatin. When compared to eGFR using the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation, the CockcrofteGault (CG) formula estimated that creatinine clearance (CrCl) results in 20% more patients being excluded from eligibility for cisplatin. This difference is even more pronounced among Caucasians, elderly and female patients [10e15]. Whether the CG or CKD-EPI eGFR formula is used will change a patient’s eligibility for cisplatin in approximately 15% of cases [11,12,14]. Moreover, in a study by Bennis et al., cisplatin dose adjustments were necessary for 9.7% using the CG formula, but only in 4.8% using the Modification of Diet in Renal Disease (MDRD) study formula [16]. Compared to direct measurement of CrCl based on 24-h urine specimens, more patients are classified as ineligible for cisplatin when CrCl or GFR is estimated [15]. This effect is most pronounced in patients over 65 years old with 24e53% of these patients being denied cisplatin when estimated CrCl or GFR is compared to measured CrCl [10]. These differences are obviously clinically important. Furthermore, there is evidence that measured CrCl correlates with a patient’s ability to complete three full cycles of chemotherapy whereas estimated CrCl and GFR do not [10]. For carboplatin, the assay used to determine kidney function is also an important determinant of dosage. The Calvert formula is used to determine the dose of carboplatin (Calvert: total dose [mg] Z [target area under the curve]
[GFR þ 25]). Short et al. retrospectively studied the dose of carboplatin given to a patient using the CG formula [17]. If MDRD was used instead, a discrepant dose of carboplatin (defined as a difference of more than 20%) would have occurred in 48% of patients. This begs the question of whether the thresholds used for drug selection are appropriate and, even more importantly, what is the most useful method to estimate kidney function in patients with cancer. There are several methods for measuring or estimating GFR in the general population, and each has its attendant limitations. There is no consensus on which of the available methods is ideal for the general population, and even less so in cancer patients [6,18].

1.3. Estimating kidney function using serum creatinine

For multiple reasons, serum creatinine concentration is an imperfect surrogate for kidney function. Nonetheless, it is the most commonly used method to estimate GFR. Creatinine is produced by muscles and removed from the body via glomerular filtration and tubular secretion (Fig. 1). It is important to realize that cancer patients constitute a heterogeneous population and that weight, nutritional status, and muscle mass can vary signifificantly within a single patient over the course of their treatment. Muscle wasting is common among cancer patients and is frequently progressive during cancer therapy, especially among those with advanced disease undergoing chemotherapy [19]. Importantly, the relationship between serum creatinine and GFR is not linear but is rather hyperbolic, meaning that at low serum creatinine concentrations, a small change in serum creatinine concentration corresponds to a large change in GFR. Conversely, at high serum creatinine concentrations, a big change in serum creatinine corresponds to a relatively small change in GFR.

There are also analytical issues associated with serum creatinine measurements. Historically, two techniques are employed to measure serum: the classical Jaffe reaction and the enzymatic method. In the Jaffe method, a reaction between picrate and creatinine in an alkaline milieu produces a red-orange product that can be quantified. Endogenous components (glucose, proteins, ketonic acids, ascorbic acid, acetoacetate, and pyruvate) are also picked up in this assay and this pseudo-chromogens account for 15e20% of the Jaffe reaction if the serum creatinine is in the normal range. Different enzymatic methods have been described, but they all have higher specificity for serum creatinine than the Jaffe assays and are thus considered more accurate and precise than the Jaffe method. Until recently, there was signifificant heterogeneity among the different enzymatic assays [20]. The isotope dilution mass spectrometry (IDMS )-traceable method was introduced to improve standardization [21]. For all the aforementioned limitations, serum creatinine concentrations alone should not be used to monitor kidney function in cancer patients. Of note, the study of Kithclu et al. demonstrated that of clinical drug trials that excluded the vast majority of patients with CKD, serum creatinine threshold values were the exclusion criteria in 62% of patients [7].

1.4. Estimated creatinine clearance and eGFR

The most frequently used formulae to estimate kidney function using serum creatinine are the CG equation, which estimates creatinine clearance, the MDRD formula, and the CKD-EPI equations, which both estimate GFR [4]. Several additional formulae have been developed to estimate GFR. In general, the results of these formulae will be within 30% of the results of measured GFR by a reference method (nuclear medicine studies) in 85e90% of subjects [22]. Since all these formulae use serum creatinine to estimate GFR, based on the prior discussion on the limitations of creatinine measurements and the fact that anorexia, weight loss, and muscle wasting are common fifindings in cancer patients, these formulae may not provide accurate estimates of kidney function in this population [23].

1.5. CG formula

The CG formula uses serum creatinine in combination with age, weight, and gender to estimate creatinine clearance. The formula does not compensate for nonkidney function determinants of serum creatinine such as race, diet, tubular secretion, and extrarenal elimination of creatinine. Furthermore, the formula was developed using measured creatinine clearance from 24-h urine collections as a surrogate for true GFR and at a time when non-standardized non-enzymatic assays for serum creatinine measurement were employed. Consequently, the CG formula is an imprecise estimate of true GFR. Despite these signifificant shortcomings, the CGformula has become the most commonly used assay for kidney function-based drug dosing and for determination of drug eligibility since its incorporation into the the1998 Federal Drug Administration (FDA) guidelines on pharmacokinetics for patients with impaired kidney function.

1.6. MDRD and CKD-EPI

Both the MDRD and CKD-EPI equations were developed using iothalamate GFR measurement, standardized enzymatic serum creatinine assays, and they incorporate readily available non-kidney function determinants of serum creatinine such as age, sex, and race. Compared to the CG formula, the MDRD and CKD EPI formulae result in GFR estimates closer to the true GFR, especially among the elderly and in patients with a large body surface area (BSA) [24]. Although both the Kidney Disease Improving Global Outcomes (KDIGO) and the National Kidney Foundation-Kidney Disease Outcomes Quality Initiative (NKF-KDOQI) guidelines recommend the use of the CKD-EPI formula in clinical practice, this recommendation has not yet been fully adopted by the medical community [25]. Not surprisingly, cancer patients were not well represented in the original studies from which the MDRD and CKD-EPI formulae were developed. There are few studies that have compared the performance of the different kidney function estimating equations among cancer patients.

In a study by Redal-Baigorri et al. [26], the performance of MDRD and CKD-EPI was evaluated in 185 cancer patients with relatively well-preserved kidney function. Only 17% had a measured GFR below 60 mL/ min/1.73 m2. When 51Cr-EDTA was used to measure GFR and IDMS-traceable serum creatinine measurements were obtained, the MDRD and CKD-EPI performed similarly and acceptably, around 89% for both equations [26]. In another study, Lauritsen et al. [27] compared the performance of the CG, MDRD, and CKD-EPI formulae in germ cell cancer patients with preserved kidney function who received treatment with bleomycin, etoposide, and cisplatin. 51Cr-EDTA was used to measure GFR and IDMS-traceable serum creatinine measurements were obtained before chemotherapy and at multiple time points during treatment. The performance of CG, MDRD, and CKD-EPI equations was acceptable. However, among patients with increasing cycles of chemotherapy, the accuracy (defined as within 30% of measured GFR) decreased quickly from 85e90% to 76% for CG, 80% for MDRD, and 50%, for CKD-EPI [27]. Similar fifindings were reported by Funakoshi et al. [28] who reported declining accuracy for all formulae after administration of cisplatin to 60% for CKD-EPI and 56% for CG. Before cisplatin therapy and in patients with measured GFR (mGFR) over 50 mL/min, the performance of the CKD-EPI was superior to the CG formula (accuracy of 92% versus 78%) [28]. After chemotherapy, the accuracy of the CKD-EPI formula decreased. One-quarter of patients with CKD-EPI values over 60 mL/min actually had an mGFR below 50 mL/min [28]. Hingorani et al. [29] compared mGFR (by iohexol plasma clearance) with CG (nonindexed for BSA), MDRD, and CKD-EPI (both indexed for BSA) in 50 patients undergoing hematopoietic cell transplant before and 100 days after transplantation [29]. At baseline, CKD-EPI and MDRD underestimated the GFR and CG overestimated it. The accuracies were low for patients with mean normal GFR values. Indeed, accuracy within 30% at baseline was 79% for CKD-EPI, 70% for MDRD, and 57% for CG. After 100 days, the accuracy observed was similar for CKD-EPI and MDRD and slightly better for CG [29].

1.7. Other formulae

While several formulae specifific to cancer patients have been developed, these equations are not widely used because of the lack of any clear benefit over the more established MDRD and CKD-EPI formulae [30,31]. In an interesting recent report, Janowitz et al. assessed the most accurate and least biased method to estimate GFR in a population of 2,471 Caucasian adult cancer patients receiving carboplatin chemotherapy [32]. The authors compared 51Cr-EDTA GFR measurement with the eGFR determined by seven published formulae with their newly developed formula. They found that the BSA-adjusted CKD-EPI formula was the most accurate published model to estimate GFR in cancer patients. The author’s newly developed model (including serum creatinine, age, gender, and BSA) improved the accuracy of eGFR estimation and carboplatin dosing. The new formula reduced the fraction of patients with a carboplatin dose with an absolute percentage error >20% (14.17% versus 18.62% for the BSA-adjusted CKDEPI and 25.51% for the CG formula). Of note, this study had some important limitations including the use of non-IDMS standardized creatinine measurements, lack of actual carboplatin dose measurements, and an almost exclusive Caucasian population. We suggest that this new model be further examined, along with the BSAadjusted CKD-EPI, in clinical co-nephrology practice [33].

1.8. BSA or not-BSA adjusted

An often neglected but important issue is whether BSAindexed or non-BSA-indexed estimates of kidney function should be used when dosing chemotherapy drugs. This is not a theoretical or trivial discussion as this choice will signifificantly affect drug dosing and possibly clinical outcomes [34,35]. The goal of BSA indexing is to make GFR results comparable between subjects with different body sizes. For example, differences in carboplatin dosing are dependent both on the method used to calculate GFR and whether the BSA-indexed or absolute eGFR are incorporated into the Calvert formula. When eGFR indexed for BSA is calculated by the CKDEPI equation, it is less likely to be associated with drug overdosing but more likely to under-dose a drug in patients as compared with non-BSA indexed eGFR calculated by the same method [35]. BSA indexing will particularly impact GFR in cancer patients with extreme weight and/or height values. It has been observed that cancer patients with a large BSA are frequently under-treated because oncologists will often empirically reduce the dose of chemotherapy based on the belief that using lean body mass is preferable to total body mass for dose calculation [36]. However, in the context of drug dosage adaptation, the goal is to get a precise estimate of the individual’s capacity to excrete a particular drug or drug metabolite.

The FDA and the European Medicines Agency(EMA) recommend drug dosage adaptation be based on non-indexed GFR. Even though many cancer drugs are dosed according to BSA, the most commonly used method to estimate GFR in oncology is the CG formula, which yields an absolute kidney function metric(milliliters per minute) that is not indexed to BSA. Using an absolute kidney function estimate to prescribe anti-cancer drugs that are dosed according to BSA will likely alter the dose compared with dosing decisions on the basis of BSA-indexed kidney function estimates. So, in general, non-indexed GFR estimates should be used to calculate cancer drug dosages. However, when drugs are dosed absolutely or based on non-BSA parameters, estimates of kidney function in milliliters per minute should be used.

1.9. Other methods to evaluate kidney function

Large studies in the general population have established that the measurement of cystatin C in combination with creatinine provides more precise GFR estimates [37]. Recently, Stabuc et al. demonstrated that GFR estimates using cystatin C with 24-h creatinine clearance performed better than eGFR formulae using serum creatinine in patients with solid tumors receiving cisplatin-based chemotherapy [38]. In contrast, Hingorani et al. also evaluated cystatin C-based formulae and demonstrated that the combined equation showed a slightly better accuracy within 30% (at 89%) compared to creatinine-based equations, only at baseline, but not at day 100 after transplantation [29]. These conflicting fifindings suggest that it is too early to recommend cystatin C-based assays to estimate kidney function in cancer patients. There are additional potential limitations to cystatin C-based assays. First, currently, the data on cancer patients are limited, lack a reference method for measuring GFR, and/or include too few patients [39e41]. Moreover, theoretically cancer cells might also produce cystatin C [42,43]. Finally, cystatin C production is also affected by other GFR-independent factors that are not uncommon among cancer patients, such as corticoid exposure, thyroid dysfunction, inflflammation, and obesity [44e46].

1.10. Available guidelines

Several scientific societies, including the InternationalSociety of Geriatric Oncology (SIOG) and the NationalComprehensive Cancer Network (NCCN), recommend an assessment of kidney function to allow for cancer drug dose adjustment to reduce toxicity before chemotherapy, even when serum creatinine concentration is within the normal range. In contrast, there are few guidelines that provide any specifific recommendations regarding the preferred method to estimate kidney function in cancer patients. The SIOG suggests using the MDRD study equation for cancer patients older than 65 years [6,47], while the NCCN suggests using CrCl in elderly patients and “GFR calculations” in adolescents and young adults [48,49]. The current FDA guidelines recommend the CG formula for determining kidney function. However, a draft revision of the guidelines for assessing pharmacokinetics in kidney impairment suggests that the eGFR formula also should be used to estimate kidney function without stating a preference as to which formula to be used.

2. Conclusion

There is an ongoing debate on whether to use the CG formula or CKD-EPI formula to guide drug dose adjustments for cancer drugs in patients with CKD (Fig. 2). Arguments favoring the use of the CKD-EPI equation are as follows. First, in the general population, the CKD-EPI is superior to the CG equation to estimate GFR [50,51]. Second, the CKD-EPI formula estimates GFR, whereas the CG formula estimates CrCl, which is a poor estimation of true GFR. Third, the CG equation was developed using non-calibrated and non-IDMStraceable serum creatinine values [21]. On the other side, historically the CG formula has been widely used to determine drug dosage adjustments for most drugs [52,53] and it has been demonstrated to predict the risk of drug-induced adverse events [54]. For the cancer patient who is being evaluated for inclusion in a clinical trial, the method chosen to estimate kidney function is of particular importance. The current FDA classification of mild kidney impairment is a CrCl of 50e79 mL/ min, and most phase 1 trials disqualify patients from enrolment at CrCl < 60 mL/min. Since the CG formula systemically underestimates kidney function to a higher degree than either CKD-EPI or MDRD, it may unnecessarily exclude patients with mild kidney impairment from clinical trials.

Any definitive recommendations regarding the best method for estimating kidney function in cancer patients would require performing a prospective randomized trial where chemotherapy dosage is calculated using both mGFR and eGFR and then collecting data on subsequent cancer and adverse outcomes among the different groups. For many reasons, there is a low likelihood that such a study would be undertaken. The data that are available generally show differences in chemotherapy dosages when using mGFR and eGFR calculations. Uniformly, these studies have demonstrated differences in dose calculations when these two methods are used. Whether these dosing differences would result in different outcomes is not firmly established. There is only limited data available regarding the performance of GTR estimating formulae in cancer patients. From this, one can conclude that the formulae are at best suboptimal for estimating GFR in cancer patients and their inaccuracy becomes more pronounced during or after cycles of chemotherapy.

One approach includes using different eGFRformulae and calculating the absolute and relative difference between different formulae. If results are concordant (difference <10 mL/min of <10%), drug dosage recommendations available in the literature can be used. However, when signifificant discrepancies are noted, clinicians should consider the patient and their drug profile. For highly effective concentration-dependent drugs with a low risk of nephrotoxicity, the equation that gives the higher eGFR results (and thus higher dosage of the chemotherapeutic) could be considered. Conversely, for a drug with significant nephrotoxicity, a narrow therapeutic range, or invulnerable patient populations, it may be advisable to adjust the dosage based on a formula giving the lowereGFR result. The CG formula is known to give systematically lower eGFR values compared to CKD-EPI, particularly in the elderly. As such, the use of the CGformula will thus result in a more protective behavior in terms of drug dosage.

In our opinion, in addition to deciding on which formula to use, it is important to consider whether BSA-indexed versus non-BSA-indexed estimates of kidney function should be employed to determine dosing and eligibility for anticancer drugs. It is important to emphasize that the assumption that estimates of kidney function are numerically equivalent across congruent units is incorrect. In the future, guidelines should be developed to improve consistency and advocate for the use of the absolute or BSA-indexed measure of kidney function (milliliters per minute) for drugs dosed absolutely or on the basis of any non-BSA parameter versus BSA-indexed measure of kidney function (milliliters per minute per 1.73 m2).

Conflict of interest statementThe authors has no conflicts of interest in relation to the submitted manuscript.

Acknowledgments

BS is a senior clinical investigator of The ResearchFoundation Flanders (F.W.O.) (1842919N).

References
[1] Launay-Vacher V, Oudard S, Janus N, Gligorov J, Pourrat X, Rixe O, et al. Prevalence of Renal Insufficiency in cancer patients and implications for anticancer drug management: the renal insufficiency and anticancer medications (IRMA) study. Cancer 2007;110:1376e84.

[2] Janus N, Oudard S, Beuzeboc P, Gligorov J, Ray-Coquard I, Morere JF, et al. Prevalence of renal insufficiency in cancer patients: data from the IRMA-2 study [Abstract]. J Clin Oncol 2009; 27:9559

[3] Rule AD, Bergstralh EJ, Slezak JM, Bergert J, Larson TS. Glomerular filtration rate estimated by cystatin C among different clinical presentations. Kidney Int 2006;69:399e405.

[4] Levey AS, de Jong PE, Coresh J, El NM, Astor BC, Matsushita K, et al. The definition, classification, and prognosis of chronic kidney disease: a KDIGO Controversies Conference report. Kidney Int 2011;80:17e28.
[5] Soveri I, Berg UB, Bjork J, Elinder CG, Grubb A, Mejare I, et al. Back SE: measuring GFR: a systematic review. Am J Kidney Dis 2014;64:411e24.
[6] Launay-Vacher V, Chatelut E, Lichtman SM, Wildiers H, Steer C, Aapro M. Renal insufficiency in elderly cancer patients: International Society of Geriatric Oncology clinical practice recommendations. Ann Oncol 2007;18:1314e21.
[7] Kitchlu A, Shapiro J, Amir E, Garg AX, Kim SJ, Wald R, et al. Representation of patients with chronic kidney disease in trials of cancer therapy. J Am Med Assoc 2018;319:2437e9.
[8] Beumer JH, Ding F, Tawbi H, Lin Y, Viluh D, Chatterjee I, et al. Effect of renal dysfunction on toxicity in three decades of cancer therapy evaluation program-sponsored single-agent phase I studies. J Clin Oncol 2016;34:110e6.
[9] Sprangers B, Jhaveri KD, Perazella MA. Improving cancer care for patients with chronic kidney disease. J Clin Oncol 2020;38(3): 188e92. epub ahead of print.
[10] Raj GV, Iasonos A, Herr H, Donat SM. Formulas calculating creatinine clearance are inadequate for determining eligibility for Cisplatin-based chemotherapy in bladder cancer. J Clin Oncol 2006;24:3095e100.
[11] Tsao CK, Moshier E, Seng SM, Godbold J, Grossman S, Winston J, et al. Impact of the CKD-EPI equation for estimating renal function on eligibility for cisplatin-based chemotherapy in patients with urothelial cancer. Clin Genitourin Canc 2012;10:15e20.
[12] Horn T, Ladwein B, Maurer T, Redlin J, Seitz AK, Gschwend JE, et al. The method of GFR determination impacts the estimation of cisplatin eligibility in patients with advanced urothelial cancer. World J Urol 2014;32:359e63.
[13] Dash A, Galsky MD, Vickers AJ, Serio AM, Koppie TM, Dalbagni G, et al. Impact of renal impairment on eligibility for adjuvant cisplatin-based chemotherapy in patients with urothelial carcinoma of the bladder. Cancer 2006;107:506e13.
[14] Canter D, Viterbo R, Kutikov A, Wong YN, Plimack E, Zhu F, et al. Baseline renal function status limits patient eligibility to receive perioperative chemotherapy for invasive bladder cancer and is minimally affected by radical cystectomy. Urology 2011;77: 160e5.
[15] Niwa N, Kikuchi E, Masashi M, Tanaka N, Nishiyama T, Miyajima A, et al. Are the formulas used to estimate renal function adequate for patients treated with cisplatin-based chemotherapy after nephroureterectomy for upper tract urothelial carcinoma? Clin Genitourin Canc 2016;14:e501e7.

[16] Bennis Y, Savry A, Rocca M, Gauthier-Villano L, Pisano P, Pourroy B. Cisplatin dose adjustment in patients with renal impairment, which recommendations should we follow? Int J Clin Pharm 2014;36:420e9.

[17] Shord SS, Bressler LR, Radhakrishnan L, Chen N, Villano JL. Evaluation of the modified diet in renal disease equation for calculation of carboplatin dose. Ann Pharmacother 2009;43: 235e41.

[18] Matzke GR, Aronoff GR, Atkinson Jr AJ, Bennett WM, Decker BS, Eckardt KU, et al. Drug dosing consideration in patients with acute and chronic kidney disease-a clinical update from Kidney Disease: Improving Global Outcomes (KDIGO). Kidney Int 2011;80:1122e37.
[19] Stene GB, Helbostad JL, Amundsen T, Sorhaug S, Hjelde H, Kaasa S, et al. Changes in skeletal muscle mass during palliative chemotherapy in patients with advanced lung cancer. Acta Oncol 2015;54:340e8.
[20] Delanaye P, Cavalier E, Pottel H. Serum creatinine: not so simple! Nephron 2017;136:302e8.
[21] Pieroni L, Delanaye P, Boutten A, Bargnoux AS, Rozet E, Delatour V, et al. A multicentric evaluation of IDMS-traceable creatinine enzymatic assays. Clin Chim Acta 2011;412:2070e5.
[22] Delanaye P, Potter H, Botev R, Inker LA, Levey AS. Con: should we abandon the use of the MDRD equation in favor of the CKD-EPI equation? Nephrol Dial Transplant 2013;28:1396e403.
[23] Delanaye P, Mariat C. The applicability of eGFR equations to different populations. Nat Rev Nephrol 2013;9:513e22.
[24] Kilbride HS, Stevens PE, Eaglestone G, Knight S, Carter JL, Delaney MP, et al. Accuracy of the MDRD (modification of diet in renal disease) study and CKD-EPI (CKD Epidemiology Collaboration) equations for estimation of GFR in the elderly. Am J Kidney Dis 2013;61:57e66.
[25] Levey AS, Inker LA, Coresh J. GFR estimation: from physiology to public health. Am J Kidney Dis 2014;63:820e34.
[26] Redal-Baigorri B, Stokholm KH, Rasmussen K, Jeppesen N. Estimation of kidney function in cancer patients. Dan Med Bull 2011;58:A4236.
[27] Lauritsen J, Gundgaard MG, Mortensen MS, Oturai PS, Feldt-Rasmussen B, Daugaard G. Reliability of estimated glomerular filtration rate in patients treated with platinum-containing therapy. Int J Canc 2014;135:1733e9.
[28] Funakoshi Y, Fujiwara Y, Kiyota N, Mukohara T, Shimada T, Toyoda M, et al. Validity of new methods to evaluate renal function in cancer patients treated with cisplatin. Canc Chemother Pharmacol 2016;77:281e8.
[29] Hingorani S, Pao E, Schoch G, Gooley T, Schwartz GJ. Estimating GFR in adult patients with hematopoietic cell transplant: comparison of estimating equations with an iohexol reference standard. Clin J Am Soc Nephrol 2015;10:601e10.
[30] Wright JG, Boddy AV, Highley M, Fenwick J, McGill A, Calvert AH. Estimation of glomerular filtration rate in cancer patients. Br J Canc 2001;84:452e9.
[31] Holweger K, Bokemeyer C, Lipp HP. Accurate measurement of individual glomerular filtration rate in cancer patients: an ongoing challenge. J Canc Res Clin Oncol 2005;131:559e67.
[32] Janowitz T, Williams EH, Marshall A, Ainsworth N, Thomas PB, Sammut SJ, et al. New model for estimating glomerular filtration rate in patients with cancer. J Clin Oncol 2017;35:2798e805.
[33] Beumer JH, Inker LA, Levey AS. Improving carboplatin dosing based on estimated GFR. Am J Kidney Dis 2018;71:163e5.
[34] Takimoto CH, Remick SC, Sharma S, Mani S, Ramanathan RK, Doroshow J, et al. Dose-escalating and pharmacological study of oxaliplatin in adult cancer patients with impaired renal function: a National Cancer Institute Organ Dysfunction Working Group Study. J Clin Oncol 2003;21:2664e72.
[35] Shepherd ST, Gillen G, Morrison P, Forte C, Macpherson IR, White JD, et al. Performance of formulae based estimates of glomerular filtration rate for carboplatin dosing in stage 1 seminoma. Eur J Canc 2014;50:944e52.
[36] Lyman GH, Sparreboom A. Chemotherapy dosing in overweight and obese patients with cancer. Nat Rev Clin Oncol 2013;10: 451e9.
[37] Inker LA, Schmid CH, Tighiouart H, Eckfeldt JH, Feldman HI, Greene T, et al. Estimating glomerular filtration rate from serum creatinine and cystatin C. N Engl J Med 2012;367:20e9.
[38] Stabuc B, Vrhovec L, Stabuc-Silih M, Cizej TE. Improved prediction of decreased creatinine clearance by serum cystatin C: use in cancer patients before and during chemotherapy. Clin Chem 2000;46:193e7.
[39] Bretagne M, Jouinot A, Durand JP, Huillard O, Boudou RP, Tlemsani C, et al. Estimation of glomerular filtration rate in cancer patients with abnormal body composition and relation with carboplatin toxicity. Canc Chemother Pharmacol 2017;80: 45e53.

[40] Cavalcanti E, Barchiesi V, Cerasuolo D, Di PF, Cantile M, Cecere SC, et al. Correlation of serum cystatin C with glomerular filtration rate in patients receiving platinum-based chemotherapy. Anal Cell Pathol 2016:4918325. 2016.

[41] Schmitt A, Gladieff L, Lansiaux A, Bobin-Dubigeon C, EtienneGrimaldi MC, Boisdron-Celle M, et al. A universal formula based on cystatin C to perform individual dosing of carboplatin in normal weight, underweight, and obese patients. Clin Canc Res 2009;15:3633e9.
[42] Bodnar L, Wcislo GB, Smoter M, Gasowska-Bodnar A, Stec R, Synowiec A, et al. Cystatin C as a parameter of glomerular filtration rate in patients with ovarian cancer. Kidney Blood Press Res 2010;33:360e7.
[43] Kos J, Werle B, Lah T, Brunner N. Cysteine proteinases and their inhibitors in extracellular fluids: markers for diagnosis and prognosis in cancer. Int J Biol Markers 2000;15:84e9.
[44] Kimmel M, Braun N, Alscher MD. Influence of thyroid function on different kidney function tests. Kidney Blood Press Res 2012; 35:9e17.
[45] Knight EL, Verhave JC, Spiegelman D, Hillege HL, De ZD, Curhan GC, et al. Factors influencing serum cystatin C levels other than renal function and the impact on renal function measurement. Kidney Int 2004;65:1416e21.
[46] Stevens LA, Schmid CH, Greene T, Li L, Beck GJ, Joffe MM, et al. Factors other than glomerular filtration rate affect serum cystatin C levels. Kidney Int 2009;75:652e60.
[47] Lichtman SM, Wildiers H, Launay-Vacher V, Steer C, Chatelut E, Aapro M. International Society of Geriatric Oncology (SIOG) recommendations for the adjustment of dosing in elderly cancer patients with renal insufficiency. Eur J Canc 2007;43:14e34.
[48] NCCN NCCN. Clinical practice guidelines in oncology: older adult oncology version 1. 2018. [49] NCCN NCCN. Clinical Practice Guidelines in oncology: Adolescent and young adult version 2. 2017.
[50] Froissart M, Rosset J, Jacquot C, Paillard M, Houllier P. Predictive performance of the modification of diet in renal disease and Cockcroft-Gault equations for estimating renal function. J Am Soc Nephrol 2005;16:763e73.
[51] Levey AS, Stevens LA, Schmid CH, Zhang YL, Castro III AF, Feldman HI, et al. A new equation to estimate glomerular filtration rate. Ann Intern Med 2009;150:604e12.
[52] Ainsworth NL, Marshall A, Hatcher H, Whitehead L, Whitfield GA, Earl HM. Evaluation of glomerular filtration rate estimation by Cockcroft-gault, Jelliffe, wright, and modification of diet in renal disease (MDRD) formulae in oncology patients. Ann Oncol 2012;23:1845e53.
[53] Nyman HA, Dowling TC, Hudson JQ, Peter WL, Joy MS, Nolin TD. Comparative evaluation of the Cockcroft-gault equation and the modification of diet in renal disease (MDRD) study equation for drug dosing: an opinion of the nephrology practice and Research Network of the American college of clinical pharmacy. Pharmacotherapy 2011;31:1130e44.
[54] Dufour B, Toussaint-Hacquard M, Kearney-Schwartz A, Manckoundia MD, Laurain MC, Joly L, et al. Glomerular filtration rate estimated by Cockcroft-Gault formula better predicts anti-Xa levels than modification of the diet in renal disease equation in older patients with prophylactic enoxaparin. J Nutr Health Aging 2012;16:647e52.