Urinary β2 microglobulin can predict tenofovir disoproxil fumarate-related renal dysfunction in HIV-1-infected patients who initiate tenofovir disoproxil fumarate-containing antiretroviral therapy

Abstract

In nephrotoxicity induced by tenofovir disoproxil fumarate (TDF), tubular dysfunction precedes the decline in GFR, suggesting that tubular markers are more sensitive than estimated glomerular filtration rate (eGFR). The hypothesis that urinary β2 microglobulin (β2 M), a tubular function marker, can predict TDF-renal dysfunction in HIV-1-infected patients was tested. A single-center observational study. The inclusion criteria were: HIV-1-infected patients who started TDF-containing antiretroviral therapy from 2004 to 2013, urinary β2 M after and closest to the day of TDF initiation within 180 days (termed 'β2 M after TDF') was measured. The associations between 'β2 M after TDF' and four renal end points (>10 ml/min per 1.73 m decrement in eGFR relative to baseline, >20 decrement, >25% decrement, and eGFR < 60) were estimated with logistic regression model. The association between 'β2 M after TDF' and longitudinal changes in eGFR after initiation of TDF was estimated with a mixed-model. A total 655 study patients were analyzed (96% men, median age 38, median CD4 238 cells/μl, 63% treatment naïve). The median baseline eGFR was 117 ml/min per 1.73 m (IQR 110-125), and the median duration of TDF use was 3.32 years (IQR 2.02-5.31). 'β2 M after TDF' was significantly associated with more than 20 decrement in eGFR (P = 0.024) and more than 25% decrement (P = 0.014), and was marginally associated with eGFR less than 60 (P = 0.076). It was also significantly associated with the longitudinal eGFR after initiation of TDF (P < 0.0001). 'β2 M after TDF' of 1700 μg/l was identified as the optimal cutoff value for the prediction of longitudinal eGFR. Urinary β2 M measured within 180 days after initiation of TDF predicts renal dysfunction related to long-term TDF use.