Back Side Effects - HIV Liver, Kidney & Bone Toxicity ICAAC 2008: Studies Look at Long-term Efficacy Safety of Tenofovir (Viread/Truvada/Atripla), including Kidney Toxicity and Bone Loss

ICAAC 2008: Studies Look at Long-term Efficacy Safety of Tenofovir (Viread/Truvada/Atripla), including Kidney Toxicity and Bone Loss

Tenofovir (Viread) is among the most widely prescribed antiretroviral drugs, and is a component of the Truvada (tenofovir/emtricitabine) and Atripla (tenofovir/emtricitabine/ efavirenz) fixed-dose combination pills. Several presentations last month at the 48th International Conference on Antimicrobial Agents and Chemotherapy (ICAAC 2008) provided new data on tenofovir's long-term efficacy and safety, in particular the occurrence of kidney and bone problems, which have been attributed to tenofovir in some previous studies.

Atripla Efficacy and Side Effects

In study AI26607, participants on stable antiretroviral therapy with HIV suppression < 200 cells/mm3 for at least 3 months were randomly assigned to stay on their current HAART regimen (n = 97) or switch to Atripla (n = 203), which provides a full NNRTI-based antiretroviral regimen in a single once-daily pill.

Most patients (88%) were men, about two-thirds were white, and the mean age was 43 years. The median baseline CD4 count was about 540 cells/mm3 and 96% had HIV RNA < 50 copies/mL.


• At 48 weeks in an intent-to-treat TLOVR analysis, 87% of patients who switched to Atripla and 85% who stayed on their current regimen maintained HIV RNA < 50 copies/mL.
• Participants who switched from a NNRTI-based regimen were more liked to maintain viral suppression than those who switched from a protease inhibitor (PI)-based regimen (92% vs 83%).
• Discontinuation rates were similar for those who switched to Atripla and those who stayed on their baseline regimen.
• Adverse event rates were significantly higher in the Atripla arm (5% vs 1%), largely due to efavirenz-related neuropsychiatric symptoms, most of which were mild and temporary.
• At 48 weeks, estimated GFR (glomerular filtration rate, a measure of kidney function) was unchanged from baseline in both arms.
• Fasting triglyceride levels declined more in the Atripla arm (median -20 vs -3 mg/dL), almost entirely attributable to those who switched from PI-based regimens.
• Excellent adherence (96% or better) was reported in both arms.
• Participants taking Atripla were more likely to report that their regimen was convenient (97% vs 81%).

In conclusion, the investigators wrote, "High and comparable rates of virologic suppression were maintained with [efavirenz/emtricitabine/tenofovir] vs [baseline regimen], regardless of type of prior antiretroviral therapy."

"Mild, transient nervous system symptoms occurred with [efavirenz/emtricitabine/tenofovir], particularly with prior PI-based antiretroviral therapy," they added.

Orlando Immunology Center, Orlando, FL; Denver Infectious Disease Consultants, Denver, CO; Gilead Sciences, Inc., Foster City, CA; Bristol-Myers Squibb, Princeton, NJ.

Long-term Atripla

In Gilead's Study 934, participants starting antiretroviral therapy for the first time were randomly assigned to receive Truvada or the Combivir (zidovudine/lamivudine) coformulation pill, both in combination with efavirenz (Sustiva). After 144 weeks, 286 patients from both arms switched to the Atripla pill and continued follow-up.

Most participants in the extension study (88%) were men, 65% were white, the mean age was 40 years, the median CD4 count was 535 cells/mm3, and about 95% had HIV RNA < 50 copies/mL.


• 48 weeks after switching to Atripla, 94% of patients originally assigned to Truvada and 90% originally assigned to Combivir maintained viral suppression < 50 copies/mL in a missing=failure analysis.
• Mean CD4 counts increased by 1 and 21 cells/mm3, respectively.
• Patients who switched from Combivir were more likely to experience adverse events (those switching from Truvada were already taking the same drugs in a different formulation).
• Serum creatinine, serum phosphorous, and estimated GFR (all measures of kidney function) remained stable in both arms.
• Total cholesterol rose by 6 mg/dL in the Truvada-to-Atripla arm and fell by 9 mg/dL in the Combivir-to-Atripla arm.
• LDL ("bad") cholesterol remained unchanged in the Truvada-to-Atripla arm and fell by 6 mg/dL in the Combivir-to-Atripla arm.
• Fasting triglycerides fell by 3 and 21 mg/dL, respectively.
• Total limb fat did not change significantly in either arm.

"Switching Truvada or Combivir + efavirenz to a single tablet once-daily regimen of [efavirenz/emtricitabine/tenofovir] was well-tolerated and resulted in maintenance of virologic suppression through 48 weeks," the researchers concluded.

However, they added, "In patients on Combivir + efavirenz for 3 years, switching to Atripla did not significantly improve limb fat after 48 weeks."

Orlando Immunology Ctr., Orlando, FL; Chelsea & Westminster Hosp, London, UK; Johns Hopkins Univ, Baltimore, MD; Univ Hosp La Paz, Madrid, Spain; Gilead Sciences, Foster City, CA.

Bone Loss

Another research team looked at changes in bone mineral density among patients taking tenofovir/emtricitabine. Studies have shown that people with HIV have an elevated risk of bone loss (osteopenia and the more serious osteoporosis), but the reasons for this are not fully understood.

The investigators hypothesized that tenofovir might contribute to abnormal calcium metabolism via secondary hyperparathyroidism, or excess production of parathyroid hormone, which regulates calcium and phosphate levels.

In this prospective cross-sectional study, they reviewed medical records and interviewed 51 HIV positive men on HAART with normal kidney function (GFR > 60) and normal serum calcium levels; 34 were taking antiretroviral regimens containing tenofovir plus emtricitabine (Emtriva; also in the Truvada and Atripla pills).

Parathyroid hormone levels were measured using the Immulite 2000 assay with an upper limit of normal (ULN) of 65 pg/mL, and compared in patients taking or not taking tenofovir. Most participants (80%) were white, the mean age was 49 years, the mean duration of HIV infection was 12 years, the median CD4 count was 443 cells/mm3, and 75% had undetectable HIV RNA. Tenofovir recipients and those not taking tenofovir did not differ with respect to age, duration of infection, body mass index, CD4 count, viral load, or baseline GFR.


• 82% of participants had suboptimal vitamin D levels, but the likelihood was similar in both tenofovir recipients and non-recipients.
• Overall, parathyroid hormone levels were higher in tenofovir recipients than in people taking non-tenofovir-containing regimens.
• Among men with low vitamin D levels, the mean parathyroid hormone level was significantly higher in those taking tenofovir (80 vs 55 pg/mL; P = 0.02).
• Within the low vitamin D group, tenofovir recipients were more likely than non-recipients to have an above-normal parathyroid hormone level (39% vs 7%; P = 0.036).
• Among tenofovir recipients, parathyroid hormone levels were far higher in those with low versus sufficient vitamin D (87 vs 43 pg/mL; P = 0.002).

• No patients with an optimal vitamin D level had elevated parathyroid hormone.

Based on these findings, the investigators concluded, "[Tenofovir/emtricitabine] appears to be associated with [secondary hyperparathyroidism] in patients with low vitamin D, but not in patients with sufficient vitamin D, suggesting that adequate doses of vitamin D supplements along with tenofovir may prevent [secondary hyperparathyroidism], a serious condition linked to bone loss and cardiovascular disease."

They suggested that tenofovir in the setting of low vitamin D may decrease whole-body calcium levels, triggering increased parathyroid hormone production and increased calcium resorption, leading to reduced bone mineral density.

They recommended that patients taking tenofovir should have their vitamin D and parathyroid hormone levels checked, and urged further research to investigate whether prophylactic use of vitamin D and calcium supplements might prevent increases in parathyroid hormone and preserve bone in this population.

Mount Sinai School of Medicine, New York, NY.

Kidney Function

Finally, 2 reports looked at kidney (renal) dysfunction in people taking tenofovir. The drug has not been associated with kidney dysfunction in clinical trials, but some observational studies indicate that certain susceptible individuals (for example, those with pre-existing kidney disease) may experience tenofovir-related kidney toxicity. This has mostly been seen in highly treatment-experienced patients.

Researchers at Johns Hopkins assessed whether tenofovir was associated with renal dysfunction in patients starting HAART for the first time. They compared all treatment-naive patients with an estimated GFR > 50 ml/min/1.73m2 (by MDRD) who started either tenofovir (n=201) or alternative nucleoside reverse transcriptase inhibitors (NRTIs) (n=231) after January 2002, analyzing the time to 25% and 50% decline in GFR.

About 60% of participants were men, about 75% were black (a group with a higher rate of HIV-associate kidney disease), the mean age was 40 years, and the mean CD4 count was about 250 cells/mm3. Baseline GFR was similar in both groups (101 vs 110), and similar proportions had pre-existing hypertension (25%) and diabetes (5%).


• A modest decline in GFR was observed among patients starting both tenofovir and alterative NRTIs as part of first-line therapy.

• There were no significant differences in percentages of patients who experienced a 25% decline in renal function after 1 year (23.4% in the tenofovir arm vs 20.2% in the alternative-NRTI arm; P = 0.51) or after 2 years (30.1% vs 28.4%, respectivelyl P = 0.59).

• Likewise, there were no differences in proportions with a 50% decline in renal function after 1 year (4.7% taking tenofovir, 5.5% taking other NRTIs; P = 0.49) or after 2 years (5.1% vs 6.9%, respectively; P = 0.65).

• There remained no differences between the tenofovir and alternative-NRTI groups in 25% or 50% kidney function decline after adjusting for age, race, baseline GFR, baseline CD4 cell count, use of a ritonavir-boosted PI vs a NNRTI, presence of diabetes, or hypertension.

• However, older age, lower CD4 count, and hypertension were each associated with GFR decline independent of tenofovir use (all P < 0.05).

• There was a statistical interaction between tenofovir and boosted PI use, yielding a 2.1-fold greater risk of a 50% decline in GFR compared with use of alternative NRTIs + a boosted PI.

In conclusion, the researchers stated, "Our results emphasize the importance of having a control group who receive an alternative NRTI, since a modest decline in [GFR] was observed among patients taking both tenofovir and [alternative NRTIs]."

"Our results support use of tenofovir within the initial antiretroviral therapy regimen," they concluded. "Our data also suggest that GFR should be monitored more closely in older patients, those with CD4 counts < 200 cells/mm3, when hypertension is present, and when a [ritonavir-boosted PI] is used.'"

Johns Hopkins Univ. School of Medicine, Baltimore, MD.

In a related report, researchers in Philadelphia also compared kidney function in patients taking tenofovir with NNRTIs (n = 110) versus boosted PIs (n = 140). At baseline, study participants had been on HAART for at least 3 months. Here too, a majority were men and about two-thirds were African-American.

Overall, 3 patients changed their regimens due to renal problems. However, the investigators observed no evidence of significant changes in GFR from baseline during the first 2 years of therapy (P > 0.05). Multivariate analysis including age, sex, hypertension, diabetes, and treatment-naive or experienced status did not alter these findings.

"There was no change in creatinine clearance with either a boosted PI or NNRTI," they concluded. "Sub-analysis of the [African-American] population did show a significant difference in first year of therapy; however, these differences were not statistically different over the second year of treatment."

Jefferson Med. College, Philadelphia, PA; Drexel Univ. College of Med., Philadelphia, PA.



E DeJesus, B Young, J Flaherty, and others. Simplification of Antiretroviral Therapy (ART) with Efavirenz (EFV)/Emtricitabine (FTC)/Tenofovir DF (TDF) Single-Tablet-Regimen vs. Continued Unmodified ART in Virologically-Suppressed, HIV-1-Infected Patients. 48th International Conference on Antimicrobial Agents and Chemotherapy (ICAAC 2008). Washington, DC. October 25-28, 2008. Abstract H-1234.

E. DeJesus, A Pozniak, J Gallant, and others. The 48-Week Efficacy and Safety of Switching to Fixed-Dose Efavirenz/Emtricitabine/Tenofovir DF in HIV-1-Infected Patients Receiving HAART. 48th International Conference on Antimicrobial Agents and Chemotherapy (ICAAC 2008). Washington, DC. October 25-28, 2008. Abstract H-1235.

K Childs, S Fishman, K Bateman, and others. Should Vitamin D Be Prescribed with Tenofovir/FTC? 48th International Conference on Antimicrobial Agents and Chemotherapy (ICAAC 2008). Washington, DC. October 25-28, 2008. Abstract H-2300.

R Moore and J Gallant. Renal Function after Use of Tenofovir as Part of the Initial ART Regimen. 48th International Conference on Antimicrobial Agents and Chemotherapy (ICAAC 2008). Washington, DC. October 25-28, 2008. Abstract H-2297.

WR Short, P Solari, and B Kaplan. Comparison of Renal Function on NNRTI vs. Boosted PI based Tenofovir based Regimens. 48th International Conference on Antimicrobial Agents and Chemotherapy (ICAAC 2008). Washington, DC. October 25-28, 2008. Abstract H-2298.