2/26/2009 1 BHIVA ‘Best of CROI’ Feedback Meetings Opportunistic infections Hepatitis Co-infections Malignancies February 2009, Manchester, Birmingham, London Opportunistic infection/hepatitis co- infection & malignancies Opportunistic infection: Primary prophylaxis for cryptococcal infection Prednisolone for TB-IRIS Early vs. late HAART in cryptococcal meningitis Early vs. late HAART in TB (SAPIT) Once daily NVP vs. EFV in TB
25
Embed
FINALco-infections-malignancies-hepatitisEW mini ... · 1 BHIVA ‘Best of CROI’ Feedback Meetings Opportunistic infections Hepatitis Co-infections Malignancies February 2009, Manchester,
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
2/26/2009
1
BHIVA ‘Best of CROI’ Feedback MeetingsOpportunistic infections
Rosalind Parkes-Ratanshi #32Median time to starting ART 11 weeks
Follow-up at 4 weeks and then 8 weekly
Results summary
Fluconazole Placebo Adjusted HR P value
Cryptococcal
event
Pre-ARV 1
Post-ARV 0
11
7
18.74 95% CI
(2.50; 140.67)P=0.004
All cause
mortality100 98
0.98 95% CI
(0.74; 1.30)
P=0.89
Oesophageal
candida8 55
7.1, 95% CI
(3.38;14.92)P<0.0001
Oral and
vaginal candida28 163
6.83, 95% CI
(4.54; 10.27), P<0.001
Adverse events
stopping trial
drug
58 581.06 95% CI
(0.74; 1.53)P=0.74
CD4 Cryptococcal
events
(% of CD4 group)
< 50 13 (3.3%)
50-99 4 (1.3%)
100-149 2 (0.6%)
150-199 0
Rosalind Parkes-Ratanshi #32Main AE abnormal LFTs: = between arms/no association with NVP use
2/26/2009
4
Conclusions� Significant reduction in cryptococcal disease, oesophageal
candidiasis, and oral/vaginal candidiasis� Rate of cryptococcal infection was low because of:
� Exclusion of CrAg positive at screening � Early ART� Only 50% had CD4 count less than 100 cells/mm3
� No difference in all-cause mortality � Only 7 deaths attributed to cryptococcal disease
� Implications for clinical practice � In African clinical practice unlikely to have CrAg screening
and limited access to ART and effective cryptococcal treatment
� Fluconazole prophylaxis may have potentially even greater impact in this setting
Rosalind Parkes-Ratanshi #32
Prednisolone for TB-IRIS
� Background: occurs in 8-43%, anecdotal reports steroids beneficial but concerns may worsen KS/CMV
� Hypothesis: 4w prednisolone would reduce need for medical interventions, be safe and not ↑ infections
� Design:� Prednisolone or placebo, randomised double blind
� 1.5mg/kg for 2 weeks then 0.75mg/kg for 2 weeks
� Follow-up assessments: 1, 2, 4, 8, and 12 weeks
� Open-label at physicians’ discretion if clinical deterioration/relapse
� Primary endpoint:� Cumulative number of days and OPD therapeutic procedures
(arbitrarily counted as 1 additional day), ITT analysisMeintges #35
2/26/2009
5
Case definition & enrolment
287 screened (June
2005-Dec 2007)
Alternative diagnosis = 44
Did not fulfil case definition = 65
Exclusion criterion = 55
Unwilling/unable to consent = 13
Enrolled = 110
Placebo = 55
Died (2), defaulted (6),
rifampicin resistance (6),
discontinued study drug (3)
Prednisolone = 55
Died (3), defaulted (0),
rifampicin resistance (4),
discontinued study drug (1)
Case definition:Prior to ART•Evidence of TB•Initial improvement with TB treatment•Still on TB treatment•RIF-sensitive strainWithin 3m of starting ART•New/recurrent TB symptoms•Presence of >1 of: ↑LN, cold abscess, serous effusions, lung infiltrates
� ART initiated ASAP after diagnosis during intensive phase of TB treatment
� ART initiated after intensive phase
� ART initiated after TB treatment completed
� TB treatment standard regimen
� Co-trimoxazole prophylaxis given to all patients
� ART was ddI/3TC/EFV given OD with TB-DOT
Integrated arm
Karim #36a
Status of trial at DSMB review in Sept 2008
Screened n=1331
Enrolled n=642
Integrated arm n=429 Sequential arm n=213
Initiated ARVs n=358 Initiated ARVs n=132
Completed trial n=94
(22%): Rest continuing
follow-up
Completed trial n=38
(18%): Rest continuing
follow-up
DSMB stopped
Sequential arm
2/26/2009
12
Mortality rates per CD4 count
CD4 count
<200 cells/mm3 >200 cells/mm3
Integrated arm•# dead/ py (n)•Mortality rate
23/281 (273)8.2 (5.2-12.3)
2/185 (156)1.1 (0.1-3.9)
Sequential arm•# dead/ py (n)•Mortality rate
21/137 (138)15.3 (9.57-23.5)
6/86 (75)7.0 (2.6-15.3)
Hazard ratio•Cox regression
0.54 (0.34-0.98)P=0.04
0.16 (0.03-0.79)P=0.02
Reduction in mortality rates is present in patients with
CD4 counts above and below 200 cells/mm3
ART adherence and TB and HIV outcomes
Integrated arm Sequential arm
ART adherence >95% pill count
90.4% (311/344) 87.1% (115/132)
VL <1000c/ml at 12 months * (same <400)
91.0% (201/221) 80.0% (72/90)
TB treatment successful
78.4% (258/331) 73.3% (121/165)
Incidence of IRIS * 12.1% (52/249) 3.8% (8/213)
Mortality in MDR-TB patients
20% (3/15) 71% (5/7
63% vs. 62% initiated ART in integrated and sequential arm at time of analysis.
*P<0.05
2/26/2009
13
Conclusions
� Outcome at halt of sequential arm:
� Deaths: integrated arm (25/429), sequential arm (27/213): HR 0.44 (95% CI 0.25-0.79) P=0.003
� Reduced mortality with integrated treatment (56%)
� TB outcomes similar in both arms
� IRIS rate higher in the integrated treatment arm
� VL suppression higher in the integrated treatment arm
� 2 integrated treatment arms of SAPIT continuing (ASAP vs. after initiation phase)
Karim #36a
Once daily NVP vs. EFV in the treatment of HIV/TB� Aim:
� To compare efficacy and safety of two NNRTI regimens given once daily in patients with TB receiving standard TB treatment (EFV preferred but NVP more available)
� Design:
� Randomised non-inferiority open-label CT
� Randomised to EFV or NVP OD after 8w of TB treatment with ddI/3TC
� NVP given with 2w lead in dose of 200mg
� Standard TB treatment given for 6m
� Patients received DOT for TBSwaminathan #35
2/26/2009
14
Once daily NVP vs. EFV in the treatment of HIV/TB� Inclusion criteria:
� Age <18y
� Pulmonary or extrapulmonary TB
� No previous anti-TB or HIV treatment
� CD4 <250 cells/mm3
� No major co-morbidity/pregnancy
� Outcomes:
� Favourable TB outcome: -ve cultures last 2m of TB treatment
� Favourable HIV outcome: <400 c/ml at 24w
� Median BL CD4 84 VL 300,000 c/mlSwaminathan #35
ITT analysis at 24w
Response EFV regimen
N=59
NVP regimen
N=57
Favourable HIV
response (<400c/ml)50 (85%)* 38 (67%)
Unfavourable
responses9 19
• Failure 6 11
• Death 0 5
• Lost to F/U 3 3
Favourable TB
response95% 84%
RR of failure with NVP 1.28 (1.03-1.6)*P=0.038 Swaminathan #35
2/26/2009
15
Conclusions
� 3TC/ddI/EFV good virological outcome (85%) and safety with dosing advantages of ddI over d4T:
� Caveats pregnancy and cost
� OD NVP inferior to EFV when given with RIF containing TB treatment
� 90% of patients had a favourable response to TB treatment
� KS-IRIS more common and more severe in SSA� >HHV-8 antigenic burden with later presentaiton
� Lower CD4 at presentation
� Aim:� Prospectively define incidence/ clinical characteristics and
outcomes of KS-IRIS
� Inclusion criteria:� Bx-confirmed
� No indication for chemotherapy
� No prior HAART/chemotherapy
� All participating in ART comparative trial for initial AIDS-KS treatment
Martin #31
KS-IRISDefinition of KS-IRIS:Any of the following <12w of ART
•Signs and symptoms of inflammation within pre-existing lesions (swelling, tenderness, warmth, paraesthesia, or erythema) •New or worse oedema (facial, genital or limb)•New or worse pulmonary disease (symptom, CXR, or pleural effusion)
N=55Age 34
40% femaleVL Log 5.4
CD4 116 (26-251)No. of skin lesions >50
Baseline At 4 weeks
At 4 weeksBaseline
2/26/2009
17
KS-IRIS
� Prognosis where no oedema or pulmonary involvement good
� 40% mortality where oedema or pulmonary involvement
Mortality
FINDINGS AT 12 WEEKS
22%
61%
Baseline 2 weeks 12 weeks
8 weeks 48 weeks
40%
Oedema/lung
Localised
Pre
-exi
stin
g le
sio
ns
Conclusions
� KS-IRIS more common and severe in SSA
� Determinants such as CD4 not evaluated yet
� May be difficult to distinguish from KS progression
� Wide spectrum of clinical manifestations
� Mucocutaneous lesions get better with HAART alone but may take time
� Lymphnode manifested by oedema or pulmonary disease:
� Often severe and early chemotherapy required
� May contribute to mortalityMartin #31
2/26/2009
18
Non-ADC in HIV: infection related vs. unrelated
� Review of NADC over 3 time periods:� Infection related
� Infection unrelated
� Comparison of trends in incidence ratios between HIV +ve/-ve
� Findings:� 7x rate of IRC in HIV+ve:
significant reduction over time period
� Mild increase in HIV –ve but no change over time
Silverberg #30
80x19.4x7.4x
Conclusions
� 46% of all NADC in HIV +ve were infection related; 13% in HIV –ve
� HIV +ve have 7.4 fold higher risk of infection-related NADC
� This is driven by anal cancer and Hodgkin's disease
� Smaller differences in more recent years
� Only 20% increased risk for infection unrelated NADC
� Higher for lung, melanoma, kidney and possibly colorectal; lower for prostate
� CD4 count and stage/prognosis
2/26/2009
19
Hyaluronic acid as a marker of liver fibrosis progression in SMART� Aims:
� Evaluate impact of treatment interruptions on liver fibrosis progression in both HCV/HBV co-infected and HIV monoinfected using an indirect marker of liver fibrosis – hyaluronic acid (HA)
� Determine if baseline level and change in HA levels were associated with risk of opportunistic disease, non-AIDS death or major liver events
� Control group of HIV monoinfected participants matched 1:1 on randomization date (+/- 6 months), gender, age (+/- 5 years), treatment group (DC vs. VS), history of alcohol abuse
� HA measured in stored plasma samples at BL and at month 6, 12 (co-infected only) and 24
Mocroft #821
Hyaluronic acid as marker of liver fibrosis progression in SMART� Results:
� 110 (16.3%) were HBV+, 553 (81.9%) were HCV+ and 12 (1.8%) were both HBV+ and HCV+
� Median follow-up 33m for co-infected and 35m for HIV monoinfected controls
� Among co-infected participants:
� 52 (31 in DC, 21 in VS) died from non-AIDS causes
� 29 developed an OI
� 21 developed a major liver event (17 cirrhosis, 4 liver-related deaths)
Similar trends, were seen for liver-related outcomes
ART interruption associated with significant
increase in HA levels in DC arm in co-infected
2/26/2009
20
Conclusions� Hepatitis co-infected patients had higher median
plasma levels of HA at BL and during follow-up than HIV monoinfected
� Interruption of ART was associated with a significant increase in HA levels at month 6 among co-infected patients randomized to the DC arm but not sustained at months 12 and 24
� Baseline HA was an independent predictor of time to development of non-AIDS death, but not OI disease
� Co-infected patients randomized to the DC arm with a baseline HA level > 75 ng/mL had a 37.5% risk of non-AIDS death after 48 months, whereas the risk was only 5% for those with a baseline HA ≤ 75 ng/mL
Mocroft #821
Hyaluronic Acid as a Prognostic Marker of Hepatic Encephalopathy and Liver-related Death in the EUROSIDA study
� Aim:
� To investigate plasma hyaluronic acid (HA) as a prognostic marker for LRE in HIV/viral hepatitis coinfected patients within the EuroSIDA study
� Design:
� 1377 HIV-1 infected patients were included. Of them were 1309 positive for HBsAg (denoted HBV) and/or anti-HCV (denoted HCV). Control group of 68 HIV-monoinfected patients
Peters #821
2/26/2009
21
Conclusions
� Baseline HA level was a very strong predictor of later development hepatic encephalopathy or liver-related death in HIV-1 patients coinfected with HBV +/- HCV
� Patients, who during follow-up experienced a liver-related event, had higher annual increase in HA compared to patients without an event
� High CD4 cell count reduced the risk of increasing HA
� Plasma HA may be useful, either alone or in combination with other non-invasive methods to monitor
� Progression of liver disease in patients with chronic viral hepatitis
Peters #821
Background trials� Powderley 1995 NEJM
� Fluconazole vs. clotrimoxazole alone
� 428 participants, hazard ratio placebo vs. fluconazole 8.5 � 17 cryptococcal events; 2 fluconazole group, 15 control group
� All cause mortality equivalent
� Chetchotisakd 2004 HIV Medicine� Fluconazole vs. placebo
� 90 participants
� 10 cryptococcal events; 3 fluconazole group, 7 control group