| |
TDF Did Not Reduce Renal Function in RTV-Boosted PI Regimens After 1 Year Preliminary Analysis
|
| |
| |
"Renal Function in Patients Receiving Tenofovir With Ritonavir/Lopinavir or Ritonavir/Atazanavir in the HIV Outpatient Study (HOPS) Cohort"
[Letters to the Editor]
JAIDS Journal of Acquired Immune Deficiency Syndromes: Volume 43(5) 15 December 2006 pp 626-628
Buchacz, Kate PhD*; Young, Benjamin MD, PhD; Baker, Rose K MS; Moorman, Anne BSN, MPH*; Chmiel, Joan S PhD; Wood, Kathy C BSN; Brooks, John T MD*; the HIV Outpatient Study (HOPS) Investigators
*Division of HIV/AIDS Prevention, Centers for Disease Control and Prevention, Atlanta, GA, Rose Medical Center, Denver, CO, Cerner Corporation, Vienna, VA, Feinberg School of Medicine, Northwestern University, Chicago, IL
This work was funded by Contract 200-2001-00133-Centers for Disease Control and Prevention.
"....In conclusion, we found that, despite more advanced HIV disease at baseline that could predispose to incident renal insufficiency, and despite the concerns for increased exposure to TDF, patients on TDF-containing HAART with either rtv/LPV or rtv/ATZ did not experience greater decrements in renal function during the first year of observation than patients on TDF-containing HAART regimens without these agents or other protease inhibitors...."
To the Editor:
Use of tenofovir (TDF)-containing HAART regimens has been associated with modest but statistically significant decrements in creatinine clearance in observational cohort studies1,2 and with rare severe renal dysfunction in multiple case series,3 but has not been associated with incident renal insufficiency in large randomized clinical trials.4-7 Pharmacokinetic studies have shown that TDF exposure is increased when it is co-administered with ritonavir/lopinavir (rtv/LPV) or atazanavir (ATZ)8 raising the concern for increased nephrotoxicity.9 We sought to assess the impact of concurrent use of TDF with rtv/LPV or rtv/ATZ on renal function in HIV-infected patients with normal baseline renal parameters.
We studied patients enrolled in the HIV Outpatient Study (HOPS), an ongoing prospective cohort study that has been continuously enrolling patients in 8 cities in the United States since 1993. We compared renal function in patients receiving TDF with concurrent rtv/LPV or rtv/ATZ (exposed) with patients receiving TDF-containing HAART without these agents or other protease inhibitors (PIs) (unexposed). Analyses included exposed and unexposed patients, both antiretroviral-naive and experienced, who initiated HAART regimens between November 1, 2001 and November 1, 2004, and who had serum creatinine measured at baseline (ie, closest within 12 months before starting HAART regimen of interest; median = 17 days, interquartile range; 0-49 days), and again after 12 months (±3 months) of follow-up while remaining in the same exposure category. We excluded patients who had baseline serum creatinine >1.5 mg/dL or creatinine clearance (CrCl) <50 mL/min, any history of renal disease, any prior or concurrent exposure to cidofovir or adefovir, or concurrent pregnancy. We thus assessed within-patient changes in CrCl and glomerular filtration rate (GFR) among 99 exposed (80 receiving rtv/LPV, 19 receiving rtv/ATZ) and 210 unexposed patients who had both baseline and 12-month follow-up data.
Creatinine clearance was estimated by Cockcroft-Gault equation (CG) using age and weight data closest in time to serum creatinine measurement (±3 months),10 and GFR was estimated by simplified Modification of Diet in Renal Disease Study (MDRD) equation.11 Baseline CD4+ cell count and HIV viral load were those obtained closest to start date of HAART regimen of interest (-12 months to +1 months). Nephrotoxic drug use history included any prior use of aminoglycoside antibiotics (except for ophthalmological use), cisplatin, foscarnet, or intravenous pentamidine or amphotecerin B. We used Wilcoxon rank sum test for comparing distributions of continuous variables, and χ2 or Fisher exact test for categorical variables.
The exposed and unexposed groups were comparable by gender (88% vs. 85% male) and age (median = 43 years for both), but the exposed group had fewer whites (65% vs. 78%), fewer antiretroviral-naive persons (1% vs. 13%), lower baseline CD4+ cell counts (median = 352 vs. 427 cells/mm3), higher baseline HIV viral loads (median = 3.6 vs. 2.5 copies/mL), and had been diagnosed with HIV for a longer time (median = 10.4 vs. 8.6 years), all P < 0.05. The exposed and unexposed patients were comparable by history of hypertension (23% vs. 15%), history of diabetes (6% vs. 4%), and history of nephrotoxic drug use (3% vs. 1%), all P > 0.10. The two groups also had similar baseline serum creatinine (median = 0.9 vs. 0.9 mg/dL, P = 0.21) and baseline CrCl by CG (median = 106 vs. 106 mL/min, P = 0.56) but the exposed group had higher baseline GFR by MDRD (median = 97 vs. 90 mL/min/1.73m2, P = 0.02).
The median 12-month change in CrCl was -5.1 mL/min among the exposed and -2.8 mL/min among the unexposed patients (P = 0.51); the median changes in GFR were -0.6 and -0.5 mL/min/1.73m2 (P = 0.55), respectively (Fig. 1). In generalized linear models adjusted for baseline CrCl or GFR, time from baseline, and other known predictors of renal insufficiency (race, baseline CD4+ cell count, baseline HIV viral load, ARV-naive or -experienced status, time since HIV diagnosis, and history of diabetes and hypertension), the 12-month mean changes among the exposed and unexposed patients were, respectively: -2.8 mL/min (P = 0.41) and -4.3 mL/min (P = 0.12) for CrCl (P = 0.59, null hypothesis of no difference in means), and -1.6 mL/min/1.73m2 (P = 0.62) and -4.5 mL/min/1.73m2 (P = 0.09) for GFR (P = 0.29, null hypothesis of no difference in means). Exposed persons receiving rtv/LPV and rtv/ATZ had comparable changes in CrCl and GFR. There were no diagnoses of incident clinical renal disease in either exposure group during 12 months of observation. Only 1 patient (unexposed) developed Grade 2 serum creatinine elevation. No Grade 3 or 4 toxicities were observed. Two exposed and 3 unexposed patients developed CrCl <50 mL/min (which was a transient occurrence in 4 of 5 cases), with corresponding incidence rates of 2.1 per 100 person-years among the exposed and 1.5 per 100 person-years among the unexposed (P = 0.92).
In this preliminary analysis, patients on TDF-containing HAART regimens with and without concurrent rtv/LPV or rtv/ATZ had comparable profiles of creatinine clearance and GFR during the first year of observation, with modest average declines that did not achieve statistical significance. We recognize that our analyses of about 300 patients may have been underpowered to detect subtle differences between the exposure groups. Also, at the time of our analysis, few patients on other ritonavir-containing HAART regimens had 12-month data, and thus these patients were not analyzed: 21 patients took ritonavir-boosted regimen with saquinavir, nelfinavir, amprenavir, or indinavir; only 3 patients took HAART with full-dose ritonavir.
In conclusion, we found that, despite more advanced HIV disease at baseline that could predispose to incident renal insufficiency, and despite the concerns for increased exposure to TDF, patients on TDF-containing HAART with either rtv/LPV or rtv/ATZ did not experience greater decrements in renal function during the first year of observation than patients on TDF-containing HAART regimens without these agents or other protease inhibitors.
FIGURE 1. Twelve-month changes in renal function among exposed and unexposed patients in the HIV Outpatient Study, November 1, 2001-September 30, 2005. In part (A) and (B), Unexposed are patients on TDF-containing regimens without ritonavir, atazanavir, lopinavir, or other protease inhibitors; Exposed are patients on TDF-containing regimens with ritonavir/atazanavir or ritonavir/lopinavir. Most extreme outliers represented persons who had very low Creatinine measurement (0.4 mg/dL) at baseline or follow-up, and thus large changes in serum creatinine (eg, 0.4 to 0.9 mg/dL; or 0.9 to 0.4 mg/dL). Exclusion of these observations did not affect the overall inferences.
|
|
| |
| |
|
|
|
