Reasons
for Adverse SMART Outcomes Remain Poorly Understood By
Liz Highleyman The
SMART treatment interruption study was prematurely cancelled in January 2006 after
it became apparent that patients in the interruption arm had higher rates of opportunistic
illnesses (OIs) and death compared with those receiving continuous therapy. Unexpectedly,
subjects in the treatment interruption group also had higher rates of non-HIV-related
conditions such as cardiovascular and kidney problems, which are often assumed
to be associated with antiretroviral therapy. A
set of studies presented at the XVI International AIDS Conference, taking place
this week in Toronto, offered information from further analyses of the SMART data. SMART
included 5,472 participants worldwide, 95% of whom had prior treatment experience.
The subjects, who started with CD4 cell counts above 350 cells/mm3, were randomly
assigned to one of two treatment strategy arms:
 Drug conservation
(DC): subjects stopped antiretroviral therapy when their CD4 cell counts reached
350 cells/mm3, and resumed when counts fell back down to 250 cells/mm3 or less;
Viral suppression
(VS): subjects remained on continuous therapy.
As
reported this past February at the Conference on Retroviruses and Opportunistic
Illnesses, participants in the DC arm were 2.6 times more likely to experience
OIs or death. CD4,
Viral Load, and Adverse Outcomes Jens
Lundgren presented data from an analysis looking at the relationship between adverse
outcomes and "proximal" CD4 cell counts and viral load levels (the measurement
closest to the onset of illness or the time of death). Results
A total of 95
fatal or non-fatal OIs and 72 deaths occurred during follow-up.
There were 120
OIs or deaths in the DC arm, compared with 47 in the VS arm (3.3 vs 1.3 per 100
person years, respectively).
The median proximal
CD4 counts were 343 cells/mm3 in the DC arm compared and 540 cells/mm3 in the
VS arm.
Participants in the DC arm spent a total of 32% of follow-up time with CD4 counts
below 350 cells/mm3, compared with 7% in the VS arm.
In both arms,
OIs and death occurred more often among patients with lower CD4 counts and higher
viral loads.
The risk of adverse
outcomes was similar in the two arms among patients with CD4 counts below 350
cells/mm3, but significantly higher in the DC arm among patients with higher CD4
counts (P < 0.05).
The lower CD4
cell counts in the DC arm only partially explained the difference in OI and death
rates in the two strategy arms.
"Residual
excess risk" was noted in the DC arm among patients with proximal CD4 counts
of 350 cells/mm3 or greater.
Among patients
with viral loads above 350 cells/mm3, the overall median viral load was 4.0 logs
in the DC arm, compared with less than 2.6 logs in the VS arm.
Conclusion Lundgren
said there must be a "missing link" that would explain the unexpected
high risk of adverse outcomes in patients undergoing treatment interruption. The
researchers concluded that there may be some "impairment of immune function
not reflected in peripheral blood CD4 count." Risk
Factors A
second study, presented by Wafa El-Sadr, examined risk factors that might help
explain the elevated OI and death rates in the DC arm. Results
The overall hazard
ratio was 2.6, meaning participants in the DC were more than twice as likely to
experience OIs or death.
Hazard ratios
were similar across a wide range of subgroups.
There was no
significant effect related to:
- Age;
- Gender;
- History of injection
drug use;
- Prior AIDS diagnosis;
- Hepatitis B or C co-infection;
-
Type or duration of antiretroviral therapy.
The hazard ratio
was greater for black patients compared with whites and Latino patients, driven
by a higher rate of non-OI deaths in the former group.
In contrast with
some past studies, baseline and nadir (lowest ever) CD4 cell count and CD4 percentage
were not significant predictors of adverse outcomes.
Among patients
with viral load levels of 400 copies/mL or less, the rate of OIs or death was
3.2 in the DC arm, compared with 0.8 in the VS arm (P < 0.001).
However, among
patients with HIV RNA levels higher than 400 copies/mL, there was no significant
difference in OI or death rates in the DC and VS arms.
Patients who
were doing best on HAART before treatment interruption (i.e., those with higher
CD4 cell counts and lower viral loads) had comparatively worse outcomes in the
DC arm.
Conclusion The
researchers concluded that rates of adverse outcomes were higher in the DC arm
across all analyzed subgroup, and none appeared to benefit from treatment interruption.
However, some groups did experience "particularly inferior outcomes."
Types and
Severity of Clinical Events A
third substudy analyzed the types and severity of non-fatal clinical events seen
in the SMART study and their association with proximal CD4 cell count. In
a poster presentation, D. Cohn and colleagues analyzed the type and severity of
non-fatal clinical events and their association with proximal CD4 cell count in
the SMART study. Events were classified as serious (life-threatening) or non-serious;
the serious events were:
Mycobacterium
avium complex (MAC);
Cytomegalovirus (CMV);
Toxoplasmosis;
Histoplasmosis;
Cryptococcosis;
Progressive multifocal leukoencephalopathy (PML);
AIDS-related dementia;
AIDS wasting;
Lymphoma;
Visceral Kaposi's sarcoma (KS).
Results
Over a mean follow-up
period of 14 months, 90 patients experienced a clinical event of interest.
70 clinical events
occurred in the DC arm (2.2 per 100 PY), compared with 20 (0.6 per 100 PY) in
the VS arm (P < 0.0001).
57% of patients
in the DC arm and 30% in the VS arm were off antiretroviral therapy when the clinical
events occured.
9 serious events
occurred in patients with CD4 cell counts below 350 cells/mm3, all of them in
the DC arm.
34 non-serious
events occurred in the DC arm and 7 in the VS arm in patients with CD4 cell counts
below 350 cells/mm3.
6 serious events
were reported among patients with CD4 cells counts of 350 cells/mm3, 4 of them
in the DC arm.
Among patients
with CD4 counts of 350 cells/mm3, there were 26 non-serious events in the DC arm
and 11 in the VS arm.
Among patients
with CD4 counts of 350 cells/mm3, the rate of serious events was similar in the
DC and VS arms (13% vs 15%, respectively).
Conclusion The
researchers concluded that, overall, the incidence of both serious and non-serious
events was greater in the DC arm than in the VS arm. However, among patients with
proximal CD4 cell counts of 350 cells/mm3, the proportion of serious events was
similar in both arms. Quality
of Life Finally,
W. Burman and colleagues conducted a substudy to examine quality of life among
1225 SMART participants, using the multidimensional SF-12 health questionnaire. Results
At study entry,
50% rated their health as very good or excellent.
After a mean
follow-up of 2.4 years, current health status and general perception of health
declined in the DC arm, but increased in the VS arm (P = 0.09 and 0.02, respectively).
Patients in the
DC arm also had lower physical health (P = 0.005) and energy (P = 0.05) scores.
Differences in
other SF-12 rating dimensions were small and not statistically significant.
Conclusion The
researchers concluded that episodic use of antiretroviral therapy did not improve
quality of life in the SMART study. "Physical functioning, general health
perception, and energy scores worsened among patients in the DC group compared
to the VS group," they wrote The
Big Picture Taken
together, these studies suggest that the type of treatment interruption strategy
used in the SMART study provides little if any benefit and considerable risk of
both minor and life-threatening adverse events. Treatment
interruption has been explored in an effort to improve the quality of life of
people with HIV (e.g., reduced toxicity, inconvenience, and cost), even if it
meant an increased risk of HIV disease progression. In SMART, however, quality
of life not only failed to improve, but actually decreased. Lundgren
and El-Sadr agreed that until more is known about the causes of the high rate
of adverse outcomes in SMART, CD4-guided strategic treatment interruption should
only be attempted in the context of randomized clinical trials. 8/18/06 References Lundgren
JD et al. Progression of HIV-related disease or death (POD) in the randomized
SMART study: why was the risk of POD greater in the CD4-guided ((re)-initiate
ART at CD4 < 250 cells/mcL) drug conservation (DC) vs the virological suppression
(VS) arm? XVI International AIDS Conference. Toronto, August 13-18. Abstract WEAB0203. El-Sadr
W et al. Inferior clinical outcomes with episodic CD4-guided antiretroviral therapy
aimed at drug conservation in SMART study: consistency of finding in all patient
groups. XVI International AIDS Conference. Toronto, August 13-18. Abstract WEAB0204. D
Cohn, for the SMART Study Group. Severity and types of clinical events by proximal
CD4 cell counts in the SMART study. XVI International AIDS Conference. Toronto,
August 13-18. Abstract 10520. W
Burman for the SMART Study Group. The effect of episodic CD4-guided antiretroviral
therapy on quality of life: results of the quality of life substudy of SMART.
XVI International AIDS Conference. Toronto, August 13-18. Abstract 18588
| 
