Isoniazid preventive therapy in a time of HIV, TB, and MDR

1
Isoniazid preventive therapy in a time of HIV,
TB, and MDR

2
Why do TB and HIV programs and the HIV community
itself ignore IPT?

• WHO Recommended Policy on Collaborative TB/HIV
  Activities
• HIV/AIDS programmes should provide isoniazid
  preventive therapy as part of the package of care
  for people living with HIV/AIDS when active TB is
  safely excluded.
• Countries have been ignoring WHOs advice for IPT
  for at least a decade. Paper policies do not lead
  to actual program implementation in practice.
• 2. Information about isoniazid preventive therapy
  should be made available to all people living
  with HIV/AIDS.
• TB programs provide inaccurate information about
  IPT and deride its effectiveness while HIV
  programs ignore IPT.
• Community activist groups and PWA groups have not
  done enough to educate their peers about IPT and
  create demand.

3
How strong is the evidence?

• Q How strong is the evidence that IPT works?
• A Overwhelming.

4
Effect of IPT on TB in PPDve PLWHAmeta-analysis
of clinical trials
Relative risk, 95 CI
1.0
Placebo
Overall
TST
TST-
Woldehanna 2004, Cochrane review
5
How much IPT is being done?

• Q How much IPT is being done as part of
  worldwide scale-up of HIV prevention, care, and
  treatment services?
• A Pathetically little.Only 27,056 in 2006
  equivalent to less than 0.1 of the estimated 33
  million people estimated to be infected with HIV
  globally While 84 countries reported the
  existence of an IPT policy, only 25 reported any
  provision Numbers on IPT are dominated by
  Botswana, which accounted for 70 of the total
  number globally in 2006.
• -- WHO TB report 2008

6
Excuses for not implementing IPT

1. IPT worsens drug resistance.
2. Its not needed if youre on ART.
3. Its too toxic.
4. Its too hard to rule out active TB.
5. IPT is too complicated and costly.
6. IPT adherence is poor.

7
Objection 1 Resistance

• Claim IPT promotes drug resistant disease and
  renders first-line therapy less effective when
  active TB occurs.
• Fact IPT does not promote drug resistant
  disease it reduces TB incidence by 40-60 and
  when active TB occurs among those given IPT,
  standard four-drug first-line therapy works.

8
Does IPT promote INH resistant TB?

• IPT effective even if relatively high prevalence
  of INH resistance eg 17 in Haiti in 1990s
  (Chaisson ARRCCM 19961541034)
• If TB is latent, few organisms, dividing slowly,
  thus low risk of selection of DR-TB
• Standard quadruple therapy is effective for INH-r
  TB (Nolan IJTLD 20026952)

9
Does IPT promote isoniazid resistance?
Balcells Emerg Infect Dis 200612744
10
Does IPT promote isoniazid resistance?

• Combined average of previous 13 studies analyzed
  shows risk of increased resistance, if any, is
  small
• summary RR 1.45 (95 CI 0.85, 2.47)
• most resistance arises from suboptimal treatment
  of active disease, so preventing active disease
  will reduce resistance
• need for surveillance for resistance
• need for implementation research

11
Objection 2 IPT not necessary because ART is
good enough

• Claim IPT is not necessary because ART alone
  is good enough in reducing TB incidence.
• Fact IPT and ART are synergistic in reducing TB
  incidence among people with HIV taking both.

12
Risk factors for TB on ART UK
Unadj rate ratio (95 CI) Adj. rate ratio (95 CI)
Ethnic group White Other Black African 1 2.29 (1.50, 3.51) 4.81 (3.48, 6.65) 1 1.96 (1.24, 3.11) 2.49 (1.51, 4.11)
HIV exposure Heterosexual Sex between men Other 1 0.26 (0.19, 0.35) 0.45 (0.26, 0.81) 1 0.54 (0.32, 0.89) 0.58 (0.30, 1.11)
CD4 count 500 350-499 200-349 51-199 50 1 2.52 (1.30, 4.91) 3.33 (1.78, 6.25) 9.51 (5.26, 17.21) 24.58 (13.01, 46.46) 1 2.89 (1.41, 5.93) 3.67 (1.75, 7.71) 10.81 (4.87, 24.00) 33.92 (13.68, 84.12)
Nadir CD4 (per 50 cell increase) 0.89 (0.83, 0.96) 1.18 (1.08, 1.28)
Time since ART start (per yr inc) 0.68 (0.62, 0.75) 0.78 (0.69, 0.89)
Calendar year 2002-2004 1999-2001 1996-1998 1 1.58 (1.15, 2.17) 1.68 (1.08, 2.58) 1 1.09 (0.78, 1.76) 0.55 (0.31, 0.98)

• time-updated Grant CROI 2007 abs 846

13
Effect of ART on TB (South Africa)
Lawn AIDS 2005192109
14
Effect of ART IPT in Brazil

• Retrospective review of TB incidence in 11,026
  HIV at 29 public clinics in RJ between 1 Sep
  03-1 Sep 05.
• TB reduction ()
• No ART, no IPT 4.01 cases/100 person years 0
• ART, no IPT 1.90 cases/100 py 52.5
• IPT, no ART 1.27 cases/100 py 68.3
• ART IPT 0.80 cases/100 py 80.0
• Compared with no ART/no IPT.
• After adjusting for age, previous TB, and
  baseline CD4 count the overall reduction seen
  with ARTIPT was 76 vs. no intervention.
• -- Golub JE, Saraceni V, Cavalcante SC, et al.
  The impact of ART and isoniazid preventive
  therapy on TB incidence in HIV-infected patients
  in Rio de Janeiro, Brazil. AIDS. 2007 Jul
  1121(11)1441-8.

15
Objection 3 Toxicity

• Claim IPT is too toxic for people with HIV (on
  or not on ART) and has additive toxicity with
  ARTgt
• Fact IPT is far less toxic than HRZE and has far
  fewer interactions with ART than R IPT toxicity
  is rare and can be managed.

16
IPT hepatotoxicity rare

• Uganda RCT
• 7/931 ASTgt135u total 3 stopped with any adverse
  event (Whalen NEJM 1997337801)
• South Africa, routine, pre ART
• 1/777 stopped INH with asymptomatic raised AST
  ref?AMA 20052932719)
• South Africa, ART cohort
• IPT not associated with higher risk of
  hepatotoxicity (Hoffmann AIDS 2007211301)

17
Objection 4 Its too hard to rule out active TB

• ClaimIts too hard to rule out active TB among
  HIV persons.
• Fact If countries use intensified case finding
  and implement the new WHO diagnostic algorithm
  for smear-negative and extrapulmonary TB active
  TB case detection will increase.

18
Objection 5 IPT is too costly complicated

• IPT is an effective intervention on an
  individual basis if patients can complete the 6-9
  months regimen and if there is a system to
  support them Screening, sensitizing the HIV
  community and ensuring that people living with
  HIV can complete the entire duration of treatment
  is resource intensive
• -- Saidi Egwaga, NTLP Program Manager, Tanzania

19
IPT is cheaper and easier than treating active TB
disease!

• Claim IPT is too costly and complicated.
• Fact Its a lot easier to take one very
  effective and non-toxic drug for six to nine
  months to prevent potentially fatal TB than it is
  to take six months of TB treatment when youve
  got a high risk of early mortality (up to 33)
  even if youre on ART. In this case prevention
  really is cheaper than treatment.

20
Objection 6 IPT adherence is poor

• Claim IPT adherence is poor.
• Fact Lack of clarity about who is responsible
  for IPT means that no one is responsible for
  providing IPT or for promoting adherence.
• Fact People with HIV have amazingly high
  adherence to ART why dont HIV treatment
  programs use HIV adherence training and support
  methods to increase uptake and adherence to IPT.

21
Retention on IPT Zambia
number of patients
23
months on IPT
Ayles Int Conf on AIDS, Durban 2000 Abstract
ThPeB5212
22
What is needed?

• HIV programs must take responsibility.
• HIV community/PHA group must take responsibility.
• Operational research should be done but not as an
  excuse to delay implementation.
• Failure to provide IPT is a violation of human
  rights and will worsen the DR-TB epidemic among
  people with HIV.

23
Issues to consider

• Duration IPT shorter 2 drug PT regimens?
• Better HIV cohort reporting recording
  integrating IPT and CTX.
• Lack of standard guidelines for follow-up of
  healthy (pre-ART) HIV patients in RLS.
• Need to record active TB incidence in districts
  with without IPT.
• Need to measure impact if any on DR-TB.
• Evaluate different adherence approaches.

24
Acknowledgments

• Aurum
• CREATE
• JHU
• LSHTM
• THRio
• WHO HIV STB Departments
• Bill Melinda Gates Foundation
• TB, HIV, and TB/HIV activist colleagues

25
TAC TB march, 2007