Inferior Clinical Outcome of the CD4 Count Guided Antiretroviral Treatment Interruption Strategy in

1
Inferior Clinical Outcome of the CD4 Count
Guided Antiretroviral TreatmentInterruption
Strategy in SMARTRole of CD4 Counts and
HIV-RNA Levels During Follow-up   Jens D.
Lundgren for The Strategies for Management of
Anti-Retroviral Therapy(SMART)Study Group
2
Background

• SMART is open-labelled randomised trial of 5,472
  patients with CD4 counts gt 350 cells/µL comparing
  risk of opportunistic disease or death (OD/death)
  from two strategies of using antiretroviral
  therapy
• Drug conservation (DC) (CD4-guided STI threshold
  for stopping and (re)-initiation 350 and 250
  cells/µL)
• versus
• Virological suppression (VS)
• Enrolment in SMART was prematurely halted due to
  an excess risk of OD/deaths in the DC arm
• hazard ratio (HR)(DC/VS) 2.61

3
Rationale and aim

• Intermittent use of ART results in dramatic
  alterations in CD4 count and HIV-RNA levels
• Question did these alterations explain the
  excess risk of OD/death in the DC group?

4
Method

• Assessment of LATEST CD4 counts and HIV-RNA
  levels and follow-up-time
• LATEST time-updated
• analysis updated when next value in the course of
  follow-up for individual patient is determined
• Follow-up time Time from latest to next value
• Rate ( events/100 person-years)
• Cox PH mode and Poisson regression
• Predictors of clinical outcome
• Hazard ratio (HR) of DC versus VS (HR (DC/VS)

5
Clinical outcomes assessed

• Opportunistic disease (OD) or death (OD/death)
• the primary outcome of the study, and its two
  components
• Opportunistic disease (OD)
• regardless of whether the outcome of the OD was
  fatal or not
• Non-OD deaths
• without a prior non-fatal OD after entry in SMART

6
Distribution of Follow-up Time (FUT) ()by
latest CD4 Count and HIV-RNA

• DC versus VS
• of FUT with latest
• CD4 gt350 c/µL
• 68 vs 93
• HIV-RNA lt400 c/mL
• 22 vs 69

60
50
of total follow-up time in treatment group
40
30
20
gt500
350-499
10
250-350
lt250
0
Latest CD4 Cell count (cells/µL)
gt400
lt400
gt400
lt400
DC Group
VS Group
Treatment group and latest HIV-RNA level
(copies/mL)
7
CD4 count, CD4 and HIV-RNA levels latest
value prior to OD/deathand overall during
follow-up
Overall
Prior to OD/death
8
Comparison of unadjusted and adjusted Hazard
Ratios (DC/VS)
Unadjusted
Adjusted for latest CD4 count
Adjusted for latest HIV-RNA level
Adjusted for latest CD4 count and HIV-RNA level
OD/death
OD
Non-OD death
Endpoints assessed
Use of CD4 rather than CD4 count similar
findings
9
Opportunistic disease or death by latest CD4
cell count
plt0.05 (DC/VS)


of events 31 7 31
6 29 7 29 27
of PYs 305 64 838
201 1225 821 1227 2592
OD and non/OD deaths separately similar findings
10
Conclusion (1)

• Lower CD4 counts and higher HIV-RNA levels
  during follow-up in the DC arm explains a large
  proportion of the observed difference in risk
  (DV/VS) of all three clinical outcomes assessed
• Greater risk of OD/death among patients with
  latest CD4 counts in the range of 200-350
  cells/µL (compared with gt 350)
• Since the of total follow-up time with latest
  CD4 count lt350 cells/µL for DC vs VS
• 32 vs 7
• Hence, this finding provides a partial
  explanation for the higher overall risk of
  OD/death in the DC arm of SMART

11
Conclusion (2)

• However, a residual excess risk in DC group was
  observed
• Confined to patients w/ latest CD4 gt350 cells/µL
• Among patients with latest CD4 count gt350
  cells/µL
• Overall HIV-RNA levels (DC vs VS)
• 4.0 vs. lt2.6 (log10 copies/mL)
• Overall CD4 count (DC vs. VS)
• 350-500 stratum 418 vs 432 (cells/µL)
• 500 stratum 631 vs 721 (cells/µL)
• Residual excess risk in DC group linked to higher
  HIV-RNA level resulting in impairment of immune
  function not reflected in peripheral blood CD4
  count

12
Conclusion (3)

• Interruption of ART is not recommended unless it
  is done in the context of randomized studies
  adequately powered to assess clinical risks and
  benefits of such strategies

13
Acknowledgements

• Colleagues in Writing committee for WHY Paper
• Birgit Grund, Sean Emery, Abdel Babiker, Roberto
  C. Arduino, Jacqueline Neuhaus, Wafaa El-Sadr,
  Gerd Fätkenheuer, Norman Markowitz, Nathan
  Clumeck, Brian Gazzard, James D. Neaton, Andrew
  N. Phillips on behalf of the SMART study group
• SMART Executive committee
• SMART investigators and patients
• NIAID, NIH for sponsorship of SMART

14
Incidence of opportunistic disease or death for
patients with latest CD4 count gt350 cells/µLand
latest HIV-RNA level lt400 or gt400 copies/mL
of events 4 3 25
4 11 14 18 13 of
PYs 329 511 867 317
460 2034 852 535
15
The mechanism of actiondirect or indirect ?
CD4 count HIV-RNA
DC or VS strategies of using Antiretroviral Therap
y (ART)
Clinical outcome
If indirect Adjustment for factors on causal
pathway to explain difference in risk of
clinical outcome between DC and VS would reduce
HR (DC/VS)
16
Bivariate Distribution of Overall CD4 Count and
HIV-RNA Level During Follow-up Univariate Box
Plots (median, IQR)

• DC versus VS
• Exposure to ART
• 33 vs 94

17
Predictors for OD/Death in DC and VS
Adjusted for all factors shown
Worse Prognosis?
?Better Prognosis