CROI 2011: Immediate ART Improves Survival for HIV+ People with TB


Tuberculosis patients who immediately start antiretroviral therapy are less likely to progress to AIDS or die, but among people with higher CD4 counts, waiting reduces the risk of IRIS, researchers reported at CROI 2011.

People with tuberculosis (TB) who start antiretroviral therapy (ART) immediately upon HIV diagnosis were less likely to progress to AIDS or death in the multinational ACTG 5221 STRIDE study, according to a report at 18the Conference on Retroviruses and Opportunistic Infections (CROI 2011) this month in Boston.

For people with less advanced HIV disease and higher CD4 T-cell counts, however, a short delay of 8-12 weeks reduced the likelihood of developing immune reconstitution inflammatory syndrome (IRIS).

TB is a leading cause of death for people with HIV/AIDS worldwide and is considered an opportunistic infection. In general, people diagnosed with late-stage HIV disease or AIDS are advised to start combination ART right away. TB medications can interact with HIV drugs, however, and increase the frequency of adverse side effects, so the optimal time to start ART in patients with HIV/TB coinfection has been the subject of debate.

Diane Havlir and fellow investigators aimed to shed further light on the benefits of starting antiretroviral treatment early in HIV/TB coinfected patients with varying levels of immune dysfunction. ACTG 5221 was an open-label strategy trial comparing immediate ART (starting 2 weeks after TB treatment initiation) versus early ART (starting at 8-12 weeks).

The researchers enrolled more than 800 patients with confirmed or suspected TB at more than 20 sites on 4 continents. At entry, all participants had fewer than 250 cells/mm3. This level -- indicating moderate immune deficiency -- was above the World Health Organization's ART threshold of 200 cells/mm3 when the study began, but WHO raised this to 350 cells/mm3 in 2009. They were stratified by CD4 count below or above 50 cells/mm3. The median baseline CD4 count was 77 cells/mm3.

All participants received standard rifampicin-containing TB treatment and trimethoprim/sulfamethoxazole (TMP/SMX) prophylaxis. In addition, they were randomly assigned (1:1) to start HIV therapy consisting of efavirenz (Sustiva) plus tenofovir/emtricitabine (Truvada), either within 2 weeks (median 10 days) or within 8-12 weeks (median 10 weeks) after starting TB treatment.


  • During 48 weeks of follow-up, 13.0% of participants in the immediate ART arm and 16.1% in the early ART arm progressed to AIDS or died, not a statistically significant difference (P = 0.45).
  • Among people with < 50 CD4 cells, however, the difference was more apparent: 15.5% on immediate ART and 26.6% on early ART progressed to AIDS or death, which was significant (P = 0.02).
  • Among people with 50 cells/mm3 or higher, 11.5% on immediate ART and 10.3% on early ART progressed to AIDS or death, which again was not significant (P = 0.67).
  • The risk of TB IRIS was significantly higher in the immediate ART group than in the early ART group (11.0% vs 5.0%, respectively; P = 0.009).
  • There were no deaths in either group attributed to IRIS.
  • Overall, serious or severe (grade 3 or 4) toxicities did not differ between the groups.
  • Viral load suppression and CD4 cell increases at 48 weeks were also similar (about 75% with HIV RNA < 400 copies/mL and 60% achieving CD4 gains of at least 100 cells/mm3).

Based on these findings, the researchers concluded, "Overall, immediate ART did not reduce AIDS and death compared to early ART."

But they continued, "For persons with CD4 [counts] < 50 cells/mm3 immediate ART resulted in lower rates of AIDS and death compared to early ART."

"This was really a dramatic reduction, and it's very urgent that antiretroviral therapy be started in this population," Havlir said at a CROI press conference. "We absolutely need to act on this data. This study shows that minutes matter if you have low CD4 counts."

These findings support current (2009) WHO guidelines, which recommend that HIV/TB coinfected people should start ART as soon as possible after beginning TB treatment.


In a second presentation, however, Salim Abdool Karim and fellow investigators with the CAPRISA 003 SAPiT trial sounded a note of caution regarding HIV/TB patients with better-preserved immune function.

This open-label study included more than 600 HIV positive participants starting TB treatment in Durban, South Africa. The SAPIT enrollees had less advanced HIV disease as a group, with a median baseline CD4 count of 150 cells/mm3 -- about double that of the STRIDE participants.

Patients were originally evenly allocated to 3 arms: starting ART within the first 4 weeks of TB treatment (the intensive phase); starting ART later, within 4 weeks after completing the intensive phase; waiting to start ART until after TB treatment was completed. In this study ART consisted of efavirenz plus didanosine (ddI; Videx) plus lamivudine (3TC; Epivir).

The SAPiT investigators previously reported that starting ART during TB treatment significantly reduced the risk of death compared with waiting until TB treatment was finished, and that arm of the trial was halted. In this analysis, they compared outcomes among people in the early ART arm (median 9 days into TB treatment) and later ART arm (median 95 days into TB treatment).


  • The overall incidence of AIDS or death was 6.9 per 100 person-years in the early ART group compared with 7.8 per 100 person-years in the late ART group, which was not a significant difference (incidence rate ratio [IRR] 0.89; P = 0.73).
  • Among the 72 participants with baseline CD4 counts < 50 cells/mm3, the difference was much greater and did reach statistical significance, with AIDS or death rates of 8.5 in the early group vs 26.3 per 100 person-years in the late group (IRR 0.32 -- a 68% reduction; P = 0.06).
  • IRIS incidence was nearly 5 times higher in the early compared with the late ART group, 46.8 vs 9.9 per 100 person/years, respectively (IRR 4.7; P = 0.01).
  • 3 people -- all in the early ART group -- required antiretroviral drug switches due to adverse events.
  • Among the 357 participants with baseline CD4 counts of at least 50 cells/mm3, the incidence rates of AIDS or death were 6.6 per 100 person/years in the early ART group vs 4.4 per 100 person/years in the late ART group, not a significant difference (P = 0.34).
  • The difference in IRIS incidence was much smaller among these patients, but still statistically significant: 15.8 vs 7.2 per 100 person-years in the early and late ART groups, respectively (P = 0.02).
  • Again, the early ART group had more antiretroviral drug switches due to adverse events than the late group (7 vs 1, respectively; P = 0.04).

"In patients with pulmonary TB/HIV coinfection with CD4 counts < 50 cells/mm3, early ART initiation within 4 weeks of TB treatment initiation was associated with better AIDS-free survival, albeit with increased risk of IRIS," the investigators concluded.

However, they added, "in patients with CD4 [counts] > 50 cells/mm3, delaying initiation of ART to the first 4 weeks of continuation phase of TB reduced the risk of IRIS and drug switches without compromising AIDS-free survival."

"Those with severe immune deficiency should start antiretroviral therapy ASAP," Abdool Karim said at the press conference. But if CD4 count is greater than 50 cells/mm3, "physicians can make a judgment" about whether the benefits of early ART outweigh the risks.

Investigator affiliations:

Abstract 38: Univ of California, San Francisco, CA; Univ of the Witwatersrand, Johannesburg, South Africa; Harvard School of Publix Health, Boston, MA; Univ of Nebraska, Omaha, NE; Johns Hopkins Univ Research Project, Blantyre, Malawi; National Institutes of Health, Bethesda, MD; Social & Sci Systems, Inc, Silver Spring, MD.

Abstract 39LB: Ctr for the AIDS Program of Research in South Africa, Durban, South Africa; Univ of KwaZulu-Natal, Durban, South Africa; Columbia Univ, Mailman School of Public Health, New York, NY; International Ctr for AIDS Care and Treatment Programs, Columbia Univ, Mailman School of Public Health, New York, NY; Yale Univ, New Haven, CT.



D Havlir, P Ive, M Kendall, and others. International Randomized Trial of Immediate vs Early ART in HIV+ Patients Treated for TB: ACTG 5221 STRIDE Study. 18th Conference on Retroviruses and Opportunistic Infections. Boston. February 27-March 2, 2011. Abstract 38.

S Abdool Karim, K Naidoo, N Padayatchi, and others. Optimal Timing of ART during TB Therapy: Findings of the SAPiT Trial. 18th Conference on Retroviruses and Opportunistic Infections. Boston. February 27-March 2, 2011. Abstract 39LB.