Underrepresentation of black individuals in pivotal trials for novel anticancer drugs: Potential consequence of using estimated creatinine clearance to assess kidney function?

BACKGROUND: Black individuals are historically underrepresented in oncology clinical trials. One potential reason for this is the prevalence of kidney disease in Black individuals, utilization of estimated creatinine clearance as a surrogate for glomerular filtration rate (GFR) in oncology, and GFR-based trial eligibility criteria. We characterized the representation of racial minorities in anticancer agent pivotal trials and examined if GFR-based trial eligibility criteria impact the proportion of Black individuals in trial populations. METHODS: We constructed a data repository for anticancer drugs FDA-approved 2015-2019 and associated pivotal trials, from which we extracted trial population racial compositions and GFR-based trial eligibility criteria. We calculated the participation-to-incidence ratio (PIR) and participation-to-mortality ratio (PMR) for a variety of cancer sites, where PIR or PMR >1.2 and < 0.8 indicate overrepresentation and underrepresentation, respectively. We evaluated the relationship between GFR eligibility cutoffs and the proportion of Black enrollees with Spearman rank correlation coefficient. RESULTS: We assessed 24,698 patients in 74 trials. Black individuals were underrepresented in all trials (PIR ≤0.48, PMR ≤0.50). For trials with GFR-based eligibility criteria (n = 49), a lower GFR cutoff was modestly associated with a higher proportion of Black enrollees (r = -0.29, p = 0.039). This relationship was strengthened for trials that only used estimated creatinine clearance to estimate GFR (r = -0.43, p = 0.004). CONCLUSIONS: GFR-related eligibility, and specifically the use of estimated creatinine clearance, may contribute to Black individuals being disproportionately excluded from cancer clinical trials. This highlights the need for implementation of contemporary GFR equations and other interventions to boost racial minority trial enrollment.

Impact of the 2021 CKD-EPI equation on anticancer pharmacotherapy in black and non-black cancer patients.

Cancer and kidney disease disproportionately impact Black patients. The CKD-EPI2021 equation was developed to estimate glomerular filtration rate (eGFR) without using race. We assessed the impact of using CKD-EPI2021 instead of CKD-EPI2009 or Cockcroft-Gault (CG) on dosing and eligibility of anticancer drugs in Black and non-Black patients. Utilizing the National Cancer Institute Theradex database, deindexed eGFR (mL/min) was calculated for 3931 patients (8.6 % Black) using CKD-EPI2021, CKD-EPI2009, and CG. Dosing simulations based on each eGFR were performed for ten anticancer drugs with kidney function-based eligibility or dosing cutoffs. eGFR differences using CKD-EPI2021 versus CKD-EPI2009 varied between Black and non-Black patients (p < 0.001); on average, Black patients had 10.3 mL/min lower eGFR and non-Black patients had 4.2 mL/min higher eGFR using CKD-EPI2021. This corresponded to a difference in relative odds of cisplatin ineligibility using CKD-EPI2021 versus CKD-EPI2009; Black patients had 48 % higher odds of ineligibility and non-Black patients had 27 % lower odds of ineligibility using CKD-EPI2021 (p < 0.001). When using CKD-EPI2021 versus CG, eGFR differences were similar between Black and non-Black patients (p = 0.679) and relative difference in odds of cisplatin ineligibility did not vary. Using CKD-EPI2021 versus CKD-EPI2009 differentially impacts Black versus non-Black cancer patients; Black patients have lower calculated eGFR and are less likely to receive full doses of drug using CKD-EPI2021. From the historical default of CG, adopting CKD-EPI2021 would not disparately impact patients based on race, but would result in Black patients being less likely to receive full doses of drug than if CKD-EPI2009 were used.

Inclusion of Participants with CKD and Other Kidney-Related Considerations during Clinical Drug Development: Landscape Analysis of Anticancer Agents Approved from 2015 to 2019.

BACKGROUND: The US Food and Drug Administration has prioritized efforts to expand availability of therapies, including anticancer agents, for patients with CKD. US Food and Drug Administration Guidance recommends inclusion of study participants with CKD in clinical trials, improving pharmacokinetic characterization in people with decreased GFR, and using contemporary GFR assessment methods during drug development. We performed a landscape analysis of anticancer agents approved from 2015 to 2019 to evaluate inclusion of study participants with CKD and GFR assessment methods used during drug development and subsequent translation to kidney-related safety and dosing data in product labeling. METHODS: Oncology drugs approved from 2015 to 2019 and associated pivotal trials were identified. We evaluated inclusion of study participants with CKD in pivotal trials and pharmacokinetic analyses, investigated GFR assessment methods used for pivotal trial eligibility and renal pharmacokinetic analyses, and identified kidney-related adverse drug event and dosing information. RESULTS: A total of 55 drugs and 74 pivotal trials were included. Of the pivotal trials, 95% contained kidney-related eligibility criteria, including 68% with GFR-based eligibility. The median lower limit of GFR required for inclusion was 45 ml/min or ml/min per 1.73 m 2 . Pharmacokinetic analyses were performed in CKD stages 4-5 and hemodialysis for only 29% and 6% of drugs, respectively. Estimated creatinine clearance was used in over 60% and 80% of pivotal trials and pharmacokinetic analyses, respectively. Reporting of kidney-related adverse drug events was highly variable. Product labeling for 49% of drugs contained no kidney dosing information. CONCLUSIONS: Study participants with CKD continue to be excluded from anticancer drug development, and GFR estimation in pivotal trials and renal pharmacokinetic analyses remains imprecise and heterogeneous. Furthermore, kidney-related safety and dosing information is scarcely and inconsistently presented.