Abstract: TH-PO055
AKI Is Common in Patients Receiving Immune Checkpoint-Inhibitors (CPIs)
Session Information
- AKI: Biomarkers, Drugs, Onco-Nephrology
October 25, 2018 | Location: Exhibit Hall, San Diego Convention Center
Abstract Time: 10:00 AM - 12:00 PM
Category: Acute Kidney Injury
- 101 AKI: Epidemiology, Risk Factors, and Prevention
Authors
- Seethapathy, Harish Shanthanu, Massachusetts General Hospital, Boston, Massachusetts, United States
- Chute, Donald F., Massachusetts General Hospital, Boston, Massachusetts, United States
- Oppong, Yaa, Massachusetts General Hospital, Boston, Massachusetts, United States
- Cortazar, Frank B., Massachusetts General Hospital, Boston, Massachusetts, United States
- Leaf, David E., Brigham and Women's Hospital, Boston, Massachusetts, United States
- Reynolds, Kerry, Massachusetts General Hospital, Boston, Massachusetts, United States
- Sise, Meghan E., Massachusetts General Hospital, Boston, Massachusetts, United States
Background
Immune CPI use for first-line or salvage cancer immunotherapy is rapidly increasing. Because acute interstitial nephritis may complicate CPI therapy, the American Society of Clinical Oncology suggests holding CPIs and evaluating any patient (pt) whose serum creatinine (SCr) rises at least 1.5-fold above baseline (BL) i.e., ≥ grade 1 AKI. Empiric steroid therapy is recommended for pts with a Grade ≥ 2 AKI without an alternate cause. We sought to determine the frequency, severity, and predictors of AKI in a real-world population receiving CPIs.
Methods
We included all pts on CPIs from 2011-2017 at our institution. We determined CPI start date by oncology infusion records. BL SCr was established by averaging all SCr values during the 6 months prior to CPI start date. We compared all SCr values within 6 months after starting CPIs to the BL SCr to determine the fold-change from BL. We used logistic regression to determine if age, sex, race, hypertension, diabetes, cirrhosis or baseline CKD (GFR < 60) predicted AKI.
Results
1843 pts started CPI therapy. 1039 (56%) had sufficient data to determine a BL SCr and had ≥1 SCr checked within 6 months of starting CPIs. Average age was 63 (SD, 13) yrs, 60% were male and 91% were Caucasian. The mean BL SCr was 0.9 (SD, 0.4) mg/dl, and 175 (17%) had CKD at BL. Pts had a median of 7 (IQR, 2-14) SCr values measured during the BL period, and 12 (IQR, 8-18) SCr values measured in the 6 months after CPI. 140 pts (13.5%) had an AKI event within 6 months of starting CPI therapy; 79 (56%) were grade 1 (>1.5-2-fold rise in SCr), 46 (33%) were grade 2 (>2-3-fold rise in SCr) and 15 (11%) were grade 3 (>3-fold rise in SCr or needed dialysis). Of the 140 pts with AKI, 42 (30%) had multiple AKI events within the first 6 months of starting CPIs. The first AKI event occurred on average 67 (SD, 52) days after starting CPIs. 69 pts (6.6% of the total cohort) had a sustained AKI event lasting > 48 hours. No baseline pt characteristics, including BL SCr, predicted AKI in a multivariable models.
Conclusion
AKI events are common in pts receiving CPIs. BL CKD does not appear to be a risk factor for AKI after CPIs. Nephrologists are likely to be increasingly called upon to evaluate the cause of AKI and assist oncologists in determining the best course of therapy in pts receiving CPIs.