Back
Many anti-cancer medications inhibit tubular creatinine excretion, thereby increasing serum creatinine without causing acute kidney damage (“pseudo-kidney injury”) including cyclin dependent kinase 4/6, poly ADP ribose polymerase (PARP), and specific tyrosine kinase inhibitors. The long-term effect of this blockade is largely unknown and these same drugs also block growth factors that mediate cell proliferation, resulting in damage of renal tubular epithelium.These medications include first-line therapies for lung, breast, and kidney cancer and once initiated, patients remain on these daily oral medications for years during which time clinicians struggle to distinguish “pseudo-kidney injury” from true nephrotoxicity that may progress to chronic kidney disease (CKD). With the aim of determining long term kidney outcomes, we constructed a cohort of participants treated with molecularly targeted therapies suspected to acutely block tubular creatinine excretion.
Retrospective cohort study conducted using data from STAnford medicine Research data Repository (STARR) Observational Medical Outcomes Partnership (OMOP) Common Data Model that captures data (e.g., encounters, insurance status, diagnoses, drugs, measurements and procedures) using standardized methods applicable to different institutions. STARR OMOP includes the EHR data of approximately 1.5 million patients seen at Stanford Hospital and outpatient clinics over the last 17 years. Characteristics of 1,749 patients treated for at least 180 days on the following drug classes were studied: CDK 4/6i (n=453), PARPi (n=237), specific TKIs (n=1050) and MEKi (n=9). We will determine the incidence of progressive CKD defined using a common composite outcome (> 30% decline in eGFR or progression to end-stage kidney disease). Initial changes in creatinine from tubular creatinine secretion blockade (“pseudo- kidney injury”) will abruptly manifest at drug onset; therefore, the initial eGFR used for our composite outcome will be defined as first level within two weeks to three months after the start of therapy, so that further decline in kidney function would capture true nephrotoxicity. Mean initial change in eGFR will also be determined and assessed with other factors as potential contributors to risk of progressive CKD.
Table 1 shows the baseline characteristics of our study cohort prior to drug initiation. Average eGFR (ml/min/1.73m2) at the time of drug initiation for the following drug classes are as follows: CKD 4/6i (84 ± 21), PARPi (83 ± 22), TKI (83 ± 22), and MEKi (78 ± 17). Age varied from a median of 64 years in CKD 4/6i patients to 80 years in MEKi. Less than 1% if patients had an eGFR < 30 at the time of drug onset. Twelve percent of patients on CKD 4/6i had and eGFR < 60 compared with 17% on PARPi and 17% on TKIs. Hypertension was present in 40% on TKIs and PARPi and 33% on CDK 4/6i. Diabetes was present in 16% of patients on TKIs, 13% on CDK 4/6i, and 11% on PARPi. We expect outcomes analyses to be complete for presentation prior to WCN 2024.
Molecularly targeted therapies are offered to an elderly population with comorbidity rates matching the general US population and relatively preserved eGFR at presentation; understanding their long-term kidney outcomes will inform clinical-patient decision making.