Here, we present novel data on the influence of TDF combination with PI/r cART regimens on renal function in Chinese adult HIV-infected patients over a 48 week period. We found that patients exposed to TDF + LPV/r + 3TC exhibited higher levels of serum creatinine, a lower eGFR rate, and greater renal function decline than control. Importantly, our data demonstrated that serum creatinine rapidly increased, while eGFR remarkably declined during the first 4 weeks.
Only a few patients can afford the cost of TDF if TDF was used as first-line therapy drug. Patients who failed first-line treatment could receive cART containing TDF because TDF was a free second-line therapy provided by the Chinese government to HIV-infected individuals who failed first-line treatment since mid-2009. Patients in the TDF group were infected with HIV for longer than control, and were previously exposed to ART agents. Although patients in the TDF group were HIV seropositive longer, they received first-line cART (e.g., AZT + 3TC +NVP/EFV, D4T + 3TC + NVP/EFV, AZT + DDI + NVP/EFV or D4T + DDI + NVP/EFV) for over a year. To our knowledge, first-line therapies mentioned previously were not known to be associated with renal damage. However, using cART and suppressing HIV RNA may improve kidney function , reduce the rate of eGFR decrease , and/or reduce the risk of renal events . Patients who received initial cART regimes containing TDF and/or PIs were excluded from our study, and baseline serum creatinine, eGFR and HIV-1 RNA viral load were not significantly different between TDF+PI/r and control groups.
Nancy Crum-Cianflone et al.  evaluated the impact of tenofovir and a boosted PI on renal dysfunction among 150 HIV-infected Caucasian patients Initiating both tenofovir and boosted PI therapy greatly impacted kidney function: 78 (52%) of subjects showed a reduction in eGFR, the median change being −12.1 ml/min/1.73m2 [95% confidence interval (CI), -9.1 to −14.1 ml/min/1.73m2]. Moreover, 30% lost >10 mL/min/1.73m2 during the 2-year follow-up period. Goicoechea et al.  compared the estimated decline in renal function among Western HIV-1 infected patients receiving either TDF+PI/r- (n=51) or non-TDF- containing (n=66) regimens. They demonstrated that TDF+PI/r patients showed a greater rate of decline in creatinine clearance (CrCl) than the non-TDF-containing group (for MDRD, -14.7 vs. -4.78 mL/min/1.73m2/year). Consistent with observations in the Western population, median changes of eGFR were −8.8 ml/min/1.73m2 in TDF+PI/r and 6.4 ml/min/1.73m2 in control. Patients treated with TDF and PI/r showed greater decline in renal function over 48 weeks compared to non-TDF-containing regimens. Gallant JE et al. also illustrated that patients taking TDF and a PI/r exhibited a greater median decline in eGFR than those taking TDF and a non-NRTI at 6 months, with trends at 12 and 24 months . Data from the HIV Outpatient Study (HOPS) cohort, however, showed no differences in rate of creatinine clearance at 12 months between patients who were treated with either tenofovir, non-NRTIs, or ritonavir-boosted protease inhibitors (atazanavir or lopinavir) . A smaller renal function change in PI/r-treated patients may result from differences in patient characteristics. Compared with the HOPS cohort, only antiretroviral-experienced patients age 18–65 with eGFR more than 50 ml/min/1.73m2 and less advanced HIV disease were included in our study. Moreover, the comparator group contained both TDF- and other PI+TDF-treated patients. This may have influenced the decreased renal function observed in their study, reducing the window to discern an effect between treatment groups.
For tenofovir-induced nephrotoxic effects, approximately 70% are observed with concomitant use of low-dose ritonavir . Due to a PI/r and tenofovir interaction, renal clearance of tenofovir is retarded. Kiser JJ et al. observed that patients taking tenofovir + LPV/r showed a 17.5% lower tenofovir renal clearance than those taking TDF alone, even after adjusting for GFR differences . The plasma concentration of tenofovir can be increased by approximately 20-37% in PI/r-containing regimens [24, 25]. Studies have also highlighted that ritonavir is a potent inhibitor of multidrug resistance-associated protein 2 (MRP-2) . Postulated mechanisms include a potential unidentified cofactor normally excreted by MRP2 potentiating TDF toxicity or competing for TDF excretion at MRP4 . Whether this mechanism is due to PI/r competitively inhibiting renal tubular transporter function, decreasing TDF renal excretion needs further elucidation.
For our analyses of kidney function over time among patients using TDF+LPV/r+3TC, the impact of TDF+ PI/r was apparent within the first 4 weeks and stabilized to 48 weeks. Consistent with an observation from a Development of Antiretroviral Therapy (DART) trial in a large cohort of HIV-infected African adults from Uganda and Zimbabwe, changes in eGFR were also observed predominantly during early exposure to TDF cART regimens, with kidney function stabilization after ~4 weeks . Contrastingly, in the EuroSIDA (AIDS across Europe) study, data from larger European cohorts have demonstrated an association between TDF cumulative exposure and renal toxicity increases . Andrew N. Phillips et al. performed a prospective study comprised of 85% white patients. 89.8% of enrolled participants were cART-experienced patients, and 99.7% of patients were already receiving a TDF-containing cART regimen. Many enrolled participants exhibited preexisting risk factors for CKD. In the TDF group, 21.7% patients were using nephrotoxic drugs, such as pentamidine, cidofovir, acyclovir, foscarnet, or amphotericin B at or before the trial’s start.
In our study, patients were Chinese, and 90% were of the Han race. All patients in TDF+PI/r group had never taken TDF and/or PIs-containing cART regimens, nor did they use nephrotoxic drugs prior to the trial. Thus, a plausible explanation for this discrepancy resides in the longer follow-up of the EuroSIDA cohort, genetics, difference in treatment history, and inclusion patients who may have had pre-existing risk factors for CKD, subsequently increasing the risk of TDF-related CKD.
The impact of a TDF+PI/r-based cART regimen on renal function was analyzed in patients stratified by baseline eGFR. Baseline eGFR was divided into two categories (eGFR≥90 and 90>eGFR≥50 ml/min/1.73m2). We expect an observable significant decline in patients’ renal function whose baseline is 90>eGFR≥50 ml/min/1.73m2. We were thus surprised to detect no significant differences in median eGFR at 0, 4, 8, 12, 24, 36 and 48 weeks for patients with baseline 90>eGFR≥50 ml/min/1.73m2. Perhaps patients with this basal eGFR remain insensitive to nephrotoxic side effects, but may be hypersensitive to renal dysfuntion. Our cutoff for group sorting may be too broad to detect eGFR changes, since TDF impact on GFR is normally small. Studies confirming our hypotheses require further elucidation.
Early detection and diagnosis of renal dysfunction are essential for preventing or at least slowing kidney function decline to ultimately improve outcome in HIV-infected patients. Individuals with underlying impaired kidney function and patients whose kidney damage was drug-induced would likely benefit from vigilant renal follow-up. Particularly, patients receiving tenofovir may benefit from frequent assessments of kidney function, serum phosphate levels, and urinalyses to monitor for early signs of nephrotoxicity [6, 30]. The HIV Medicine Association of the Infectious Diseases Society of America (IDSA) recommends that patients receiving TDF meet any of three criteria: eGFR<90 mL/min/1.73 m2, use of medications eliminated through renal secretion, comorbid diseases following a ritonavir-boosted protease inhibitor regimen, should have kidney function (via eGFR) and serum phosphate measured no less frequently than every 6 months, and also should be analyzed for proteinuria and glycosuria . Evidence is lacking in instructing patients about frequency of assessment. In agreement with the IDSA, our results suggest that eGFR should be monitored over the first 4 weeks of follow-up when TDF is combined with PI/r. Patients who exhibited decrease eGFR with TDF+PI/r probably are candidates for more frequent assessment of kidney function, serum phosphate levels, and urinalyses, and additionally have their kidney function evaluated for a longer term. TDF+PI/r-based cART regimens should be carefully considered in patients with decreased kidney function.
Our study reports that incidences of proteinuria were not significant for both groups at weeks 0 or 48. Tenofovir is thought to primarily affect proximal renal tubular function , yet glomerular function may be lightly affected, so proteinuria incidences were not noticeable in urine.
Our study does have limitations. First, patients in the TDF+ PI/r group were receiving first-line cart prior to the study, while control patients were naïve. Second, only 3 patients older than 60 were enrolled. This study was thus not designed to decipher an effect of TDF on renal function of older patients, and the safety of TDF on older patients should be assessed in the future. Third, patients whose serum creatinine > 1.5 times the upper limit of laboratory normal and/or eGFR< 50ml/min/1.73m2 were excluded from our study. The influence of TDF based cART regimens on renal function in HIV-1 with compromised renal function remains poorly understood. Patients with renal dysfunction treated with TDF cART regimens should be frequently monitored by nephrologists. Finally, our last time point was 48 weeks. Further follow-up in a larger cohort would be required to more accurately determine long-term TDF-related nephrotoxicity. In view of these limitations, a prospective long-term multicenter clinical trial is being considered to investigate the potential impact of TDF on renal function in Chinese HIV-infected individuals. Future trials may include older patients and patients with moderate to severe renal dysfunction.
In conclusion, TDF+PI/r based cART regimens were associated with a higher level of serum creatinine and a greater decline in renal function over 48 weeks compared to control. For patients with baseline eGFR ≥90 ml/min/1.73m2, TDF+PI/r based cART regimens could influence non-progressive eGFR decreases, but the impact of TDF+PI/r based cART regimens on eGFR in patients with abnormal renal function seemed to be relatively minor. Additionally, renal function rapidly decreased during the first four weeks. Our results suggest that renal function should be monitored over the first four weeks of follow-up, especially when TDF is combined with PI/r.