Elsevier

Gynecologic Oncology

Volume 157, Issue 3, June 2020, Pages 793-798
Gynecologic Oncology

Carboplatin dosing in the era of IDMS-creatinine; the Cockroft-Gault formula no longer provides a sufficiently accurate estimate of glomerular filtration rate for routine use in clinical care

https://doi.org/10.1016/j.ygyno.2020.03.017Get rights and content

Highlights

  • 51Cr-GFR is considered the gold standard for renal function assessment.

  • The Cockroft-Gault formula can generate spuriously high GFR estimates.

  • AUC5 carboplatin based on 51Cr-GFR and AUC6 carboplatin based on CG-GFR are not equivalent.

  • Imprecise GFR estimates results in incorrect carboplatin dosing.

  • When 51Cr-GFR is unavailable, capping CG-GFR at 125 mL/min is recommended.

Abstract

Background

Glomerular filtration rate (GFR) measured by Chromium-51-EDTA excretion (51Cr-GFR) is considered the gold standard of renal function assessment, but serum creatinine in the Cockcroft-Gault (CG) formula is routinely used to estimate GFR for carboplatin dosing. Serum creatinine measured by isotope-dilution-mass-spectrometry (IDMS) can generate spuriously high GFR estimates when used in the CG formula. We hypothesized that GFR calculated using IDMS-creatinine in the CG formula (CG-GFR) exposes patients to inaccurate carboplatin dosing.

Methods

This is a multicenter retrospective study of patients who had a 51Cr-GFR assessment for malignant or non-malignant indications, with a matched CG-GFR. Carboplatin dose based on 51Cr-GFR at AUC5 was used as the reference.

Results

550 patients were analyzed, median age 62 (19–90), 64% female. Indication for GFR evaluation: malignancy (85%), assessment for live kidney donation (12%), other (3%). Median ratio of CG-GFR: 51Cr-GFR 1.04 (0.43–3.38); <0.8 in 72 patients (13%), >1.2 in 180 patients (33%). Despite capping of CG-GFR at 125 mL/min, dosing according to AUC6 would have resulted in 18% of patients being underdosed and 23% overdosed by >100 mg compared to 51Cr-GFR. Subgroup analysis identified BMI (>35, MPE 39%), gender (female MPE 15%), GFR indication (malignancy MPE 11%) as risk factors for overestimate of CG-GFR, and BMI < 20 for underestimate (MPE −3.5%).

Conclusions

The convention of considering AUC5 carboplatin based on 51Cr-GFR, and AUC6 carboplatin based on CG-GFR as equivalent is invalid and should be abandoned. When 51Cr-GFR is unavailable, capping CG-GFR at 125 mL/min is recommended.

Introduction

Platinum-based chemotherapy has been the standard of care for women with gynecological cancers since the late 1980s [1]. Although the majority of chemotherapeutic agents used in oncology are dosed proportional to a patient's body surface area (BSA), as carboplatin is predominantly renally excreted, it is dosed according to glomerular filtration rate (GFR), especially since dosing according to BSA results in variable exposure and toxicity [2]. In 1989, Calvert et altera (et al.) proposed a formula for carboplatin dosing, that accounted for both renal and non-renal excretion, based on a measured GFR, using Chromium-51 EDTA (51Cr-GFR) [3].Total Carboplatin Dosemg=targetAUC×GFR+25,whereAUCis area under curveCalvert Formula

The clinical use of 51Cr-GFR is limited by cost and availability of the study, which is technically demanding and requires involvement of a nuclear medicine department. As such, common clinical practice substitutes a calculated GFR using serum creatinine, as an estimate of renal function. A variety of equations were developed and adopted according to regional preferences, but GFR based on the Cockcroft-Gault (CG) formula (CG-GFR) [4] is currently the most commonly used, in both clinical trials and in clinical practice.Creatinine clearanceGFRmL/min=140age×weightkg×0.85if femaleSerum creatininemg/mL×72

In the mid-1990s, Calvert et al. demonstrated significant disparity between CG-GFR and 51Cr-GFR, such that the CG formula underestimates GFR, leading to underdosing in the order of an AUC of 1 [5]. Accordingly, most international gynecology trials since have required investigators to dose carboplatin at AUC5 if using 51Cr-GFR, or AUC6 if using CG-GFR [6,7]. European ovarian cancer clinical trials [8] have predominantly favored 51Cr-GFR with carboplatin dosing at AUC5, whereas North American trials [9] have predominantly used formula-based GFR (including CG-GFR), with doses up to AUC7.5. Similarly, preferences for the formula used to estimate GFR have varied geographically, with the Gynecologic Oncology Group (GOG) favoring the Jelliffe formula [10], creating a lack of consistency in clinical practice [11].

In 2006, the National Kidney Disease Education Program (NKDEP) endorsed the isotope-dilution-mass-spectrometry (IDMS) traceable assay as a standard assay for measurement of serum creatinine [12]. This was implemented as the new international standard between 2005 and 2010. Although providing a more reliable and consistent result than the preceding enzymatic assays, IDMS produces values 5–20% lower than previously. When incorporated into the CG equation, IDMS creatinine increases the CG-GFR, leading to an increase in carboplatin dose of approximately 10% [13]. This observation led the National Cancer Institute [14] to issue an Information Letter in 2010 recommending that all clinical trial protocols be amended so that estimates of GFR, based on IDMS-creatinine, would be capped at 125 mg/mL to prevent overdosing such patients [14].

The combined impact of Calvert's publication and the NCI's Information Letter is that the most common current practice is that carboplatin is dosed at AUC5 when 51Cr-GFR is used, and at AUC6, but with GFR capped at 125 mL/min, when CG-GFR (or any other IDMS-creatinine-based estimate of GFR) is used. CG-GFR, however, has never been validated using IDMS-creatinine values. This study aims to assess the validity of the current dosing conventions.

Section snippets

Study design

This study was a multicenter, retrospective audit performed across three Melbourne institutions. The primary objective was to examine whether 51Cr-GFR is accurately predicted by CG-GFR. Secondary objectives included an analysis of whether the convention of using AUC6 in the CG-GFR is a valid surrogate for AUC5 in the 51Cr- GFR, or whether an alternate formula better predicts 51Cr- GFR. Other formulae assessed include the Wright formula (using enzymatic serum creatinine and without CK) [15], the

Patients

Between 2011 and 2014, a total of 550 patients were available for analysis. The median age was 62 years and 64% of patients were female, with 85% of patients undergoing GFR assessment for a malignant indication (Table 1). The high proportion of female patients is in keeping with gynecological cancers representing the largest population (Table 1). Baseline GFR results and patient demographics are detailed in Table 1 with stratification of primary tumor types. Of the 15% undergoing assessment for

Discussion

Carboplatin is dosed according to renal function but controversy surrounding appropriate method for assessment of renal function as well as optimum dosing continues. Although convenient and readily available, measures that estimate GFR using serum creatinine are unreliable. Moreover, the change to IDMS-creatinine assessment since 2005 has resulted in lower creatinine values and, thus, spuriously high GFR results, which further increases inaccuracies in carboplatin dosing. We therefore compared

Research/Financial support

This was a non-funded study

Author contributions

All authors have all contributed to the work describe sufficiently to be named as authors.

Where else presented

This work was presented in part as an abstract at 2015 American Society of Clinical Oncology Annual Meeting.

Disclaimers

Nil.

CRediT authorship contribution statement

Luke McLean:Data curation, Writing - original draft, Writing - review & editing.James R. Whittle:Conceptualization, Methodology, Data curation, Writing - original draft, Writing - review & editing.Jonathan Graham:Conceptualization, Methodology, Data curation, Writing - review & editing.Huda Ismail:Data curation, Writing - review & editing.Meir Lichtenstein:Conceptualization, Methodology, Data curation, Writing - review & editing.Rodney J. Hicks:Conceptualization, Methodology, Writing - review &

Declaration of competing interest

The authors declare no relationships to disclose.

References (32)

  • D.W. Cockcroft et al.

    Prediction of creatinine clearance from serum creatinine

    Nephron

    (1976)
  • Calvert AH, Boddy A, Bailey NP, et al: Carboplatin in combination with paclitaxel in advanced ovarian cancer: dose...
  • M.A. Bookman et al.

    Evaluation of new platinum-based treatment regimens in advanced-stage ovarian cancer: a phase III trial of the gynecologic Cancer intergroup

    J. Clin. Oncol.

    (2009)
  • T.J. Perren et al.

    A phase 3 trial of bevacizumab in ovarian cancer

    N. Engl. J. Med.

    (2011)
  • P.A. Vasey et al.

    Phase III randomized trial of docetaxel-carboplatin versus paclitaxel-carboplatin as first-line chemotherapy for ovarian carcinoma

    J. Natl. Cancer Inst.

    (2004)
  • R.F. Ozols et al.

    Phase III trial of carboplatin and paclitaxel compared with cisplatin and paclitaxel in patients with optimally resected stage III ovarian cancer: a gynecologic oncology group study

    J. Clin. Oncol.

    (2003)
  • Cited by (8)

    • Improving Cancer Care for Patients With CKD: The Need for Changes in Clinical Trials

      2022, Kidney International Reports
      Citation Excerpt :

      We would recommend that the CKD-EPI equation is adopted as the de facto standard regression equation to determine GFR in patients with cancer. In addition, several studies in cancer patients have recently shown that estimation of creatinine clearance using the Cockcroft and Gault formula is suboptimal and should be replaced by eGFR estimation by the CKD-EPI equation, because this is associated with more precise estimation of kidney function, more precise determination of cancer drug-eligibility, more accurate dose calculation, and prevention of chemotherapy-related adverse events.67–73 The discussion about whether body surface area (BSA)-indexing should be used or not is multifaceted and complex.

    • Glomerular filtration rate estimation for carboplatin dosing in patients with gynaecological cancers

      2022, ESMO Open
      Citation Excerpt :

      We also found that the Wright formulae generally performed very poorly in this cohort of patients and finally, that estimation with AUC6 dosing resulted in overdosing by at least 10% in the majority of cases, again representing a risk of toxicity to patients. This is consistent with recent data suggesting that that the switch to IDMS creatinine a few years ago risks overdosing of carboplatin when using AUC6 dosing.20 Finally, when using CG to estimate GFR, increasing BMI was associated with increasing bias when using actual or ideal body weight but not when using adjusted weight, which appears agnostic to the influence of BMI by quartiles.

    View all citing articles on Scopus
    1

    These authors contributed equally.

    View full text