Excluding race factor from kidney function estimates may affect clinical outcomes for Black cancer patients

Removing race as a factor in a kidney function-estimating equation used to guide cancer treatment dosing and eligibility may decrease the proportion of Black patients recommended to receive full doses of anticancer drugs, according to an analysis of 15 years of phase 1 clinical trial data published in The Lancet Oncology journal. This may ultimately affect clinical outcomes for Black cancer patients.

The Chronic Kidney Disease-Epidemiology Collaboration (CKD-EPI) equation estimates a key indicator of kidney function called glomerular filtration rate (GFR). The CKD-EPI equation calculates a higher GFR for Black patients and recently, the use of race in kidney function-estimating equations has been questioned, prompting discussion on the potential consequences of removing race from equations that calculate GFR.

Since the kidney is a major organ for eliminating substances from the body, the CKD-EPI equation can be used by clinicians to help determine cancer patients' eligibility for and dosage of anticancer medications. To date, no analysis has looked at the potential impact of removing race from the CKD-EPI equation on cancer treatment.

Findings from the new study suggest that since removing race from the CKD-EPI equation calculates a lower GFR for Black cancer patients, it could decrease the proportion of patients eligible to receive anticancer medications.

Dr Thomas D. Nolin, of the University of Pittsburgh School of Pharmacy, USA, who co-authored the study, said: "Our findings suggest that removing race as a variable in the CKD-EPI equation – a widely used equation for estimating kidney function – could impact care for Black cancer patients. Specifically, removing race may exclude more Black patients, who are already disproportionately affected by cancer, from receiving full doses of potentially life-saving cancer therapy, which could affect their outcomes.

We hope our research can enhance wider discussions around race-based adjustments that will explore the complexities of including or removing race factors from treatment considerations. Ultimately, our goal is to ensure that clinicians are using the best available evidence to provide carefully considered, precise, and personalised treatment for every patient."

Dr Thomas D. Nolin, Study Co-Author, University of Pittsburgh School of Pharmacy

Study co-author Dr Morgan A Casal, of the University of Pittsburgh School of Pharmacy, USA, adds, "A race-agnostic equation for evaluating kidney function that is at least as accurate as CKD-EPI would be ideal. Until such an equation is ready for widespread clinical implementation, physicians must take an individualised, patient-centered approach to evaluating kidney function when prescribing cancer pharmacotherapy.That includes fully recognizing the limitations and implications of the various GFR-estimating equations, such as whether or not race is included, especially if the resulting choice or dosage of drug can have a substantial impact on survival."

The authors conducted a retrospective analysis of 340 Black patients enrolled in National Cancer Institute phase 1 clinical trials between 1995 and 2010. The average age of patients was 57 years, and 172 were men and 168 were women.

Patients' kidney function was estimated using the CKD-EPI equation with and without race as a variable. For comparison, the authors also generated estimates using the less precise Cockcroft-Gault (CG) equation, which was developed in a predominantly white cohort, does not take account of race, and is often utilised in oncology practice.

Dosing simulations based on each kidney function estimate were performed for 10 chemotherapeutic medications with kidney function cut offs for dosing or eligibility. The proportions of patients eligible for treatment and recommended to have a decreased dose were calculated. The proportion of patients who received conflicting eligibility or dosing recommendations based on the estimates of the three equations was also calculated.

The median GFR calculated by CKD-EPI with race included was 103 mL/min, compared to 89 mL/min for CKD-EPI without the race factor. Using CG, the median was 90 mL/min. Removing the race term from CKD-EPI nearly doubled the proportion of patients (from 7% to 13%) with a GFR of less than 60 mL/min, a key cut-off point at which many drugs, including anticancer drugs, begin to have recommended changes to dosage and eligibility.

This led to increases of between 61% and 163% in the number of patients ineligible to receive a drug or who were recommended a dosage reduction. Overall, up to 5% and 18% of all patients in the analysis had discrepant eligibility or dosing recommendations when race was excluded from the CKD-EPI equation.

Since the removal of race from the CKD-EPI equation lowered the number of patients who would be recommended to be eligible for or receive full doses of anticancer medications, the authors acknowledge that doing so might reduce anticancer drug toxicity, but further research is required.

The authors also acknowledge some limitations to their study. The cohort included only patients enrolled in phase I trials, and therefore the observed rate of kidney impairment and the proportion of patients deemed ineligible for a drug or requiring a reduced dose may underestimate the true effect in the general cancer population.

The accuracy and precision of the CKD-EPI equation with and without race has not been assessed specifically in Black cancer patients; however, this is currently being investigated as part of an NCI-sponsored trial. The authors were also only able to assess the impact on Black cancer patients of removing race from the CKD-EPI equation, although estimates for other races may also be affected by the removal of a race variable.

Writing in a linked Comment, Dr Andrew S Levey of Tufts University, USA and Dr Neil R Powe of University of California at San Francisco, USA, who were not involved in the study, say: "Although race is a social rather than a biological variable, race and ancestry are correlated, and the interplay of biological and social effects on the non-GFR determinants of serum creatinine is poorly understood… Omission of race from eGFRCr could have substantial clinical implications for cancer chemotherapy; however, adverse consequences for cancer outcomes could be minimised. eGFRCr should be the first step rather than the last step in GFR evaluation… More frequent use of confirmatory tests for GFR evaluation will increase effort and cost, but it will promote appropriate decisions and prevent disparities in cancer treatment."