NATAP - http://www.natap.org

9% of Patients with Multi-Drug Resistant HIV Died Within 2 Years

Reported by Jules Levin

At the 14th HIV Drug Resistance Workshop (June 7-11, 2005, Quebec City, Canada) Deenan Pillay (University College London/Health Protection Agency, UK) reported on behalf of the UK Collaborative Group on HIV Drug resistance & UK Collaborative HIV Cohort Study.

The estimated probability of death after two years with multi-drug resistance to the 3 main classes of drugs was 9%. Pillay found that lower CD4, higher viral load, greater number of previous drugs used, and earlier time of diagnosis were associated with increased risk of death. Of all treatment strategies used, which may have excluded T20 & tipranavir, including treatment interruption, changing therapy alone, or changing with increased GSS (Genotypic Sensitivity Score) was associated with reduced risk of death compared to continuing the same therapy.

Changing therapy to a regimen where GSS increased had the best benefit in terms of increasing CD4 count (+51) compared to other strategies including stopping therapy. The point was made from the audience that the use of T20 plus tipranavir provides a different treatment strategy. And of course new classes of drugs are in development; entry (CCR5) inhibitors are in phase II & III studies in patients & the Merck integrase inhibitor is in phase II studies in patients, with both therapies appearing to be promising for patients with HIV drug resistance. As well, several new protease inhibitors and NNRTIs specifically designed for patients with resistance to current drugs are also in development & in phase II & III studies in patients including TMC114, TMC125, and TMC278.

As a side note, in a poster here presented by the CDC, the rates of sexually transmitted HIV drug resistance was 15% for men & 13% for women. The CDC survey took place in 9 US states. Concern has been expressed about the high level of sexually transmitted HIV reported from this survey. The rates were similar regardless of HIV transmission category (MSM, heterosexual, IDU), and race/ethnicity (Black, Hispanic, White, Asian/Pacific Islanders).

“Predictors of Death and Response to Therapy in Patients with MDR HIV-1 in the UK”

Background provided by Pillay. The prolonged survival of HIV infected individuals has led to an increasing proportion with exposure to all 3 main drug classes of antiretrovirals. 25% of patients have had viral load failure while on all 3 classes. An increasing number of patients in the UK are developing multi-drug (3 classes) resistance. Pillay reported that the prevalence of MDR to 3 classes has been steady for several years at 4% but the absolute number of patients with MDR is increased since more patients are receiving treatment.

The aim of this study is to evaluate predictors of survival and virologic/immunologic outcome in patients with MDR (3 classes) HIV in the UK Drug Resistance Database.

METHODS & ANALYSIS
All patients with MDR HIV were identified from the UK HIV Drug Resistance database (1197-2003). MDR=major resistance mutations to all 3 main classes of antiretroviral drugs based on IAS-USA (2003), on cumulative tests. Level of resistance was defined by Stanford HIV database. Each drug was given a score (eg (1-sensitive). The Genotypic Sensitivity Score (GSS) is the sum of scores for each drug in the regimen. Clinical data obtained from UK CHIC cohort and other local databases.

Factors considered for survival:
treatment strategy compared to continuing current regimen.

--change in regimen with increased GSS
--change in regimen with same GSS
--change in regimen with decreased GSS
--off treatment with same GSS
--off treatment with reduced GSS

FACTORS CONSIDERED FOR SURVIVAL
CD4 count
VL
Years of MDR diagnosis
Number of previous drugs
Years on ART
GSS of existing regimen
Number of inactive drugs in existing regimen
Change in number of inactive drugs
Presence of boosted PI

STATISTICAL METHODS
Kaplan Meier analysis for survival estimates. Poison regression was used to determine factors associated with survival. Linear regression was used to determine factors associated with virologic/immunologic response at 24-48 weeks.

NUMBER IN STUDY
628 with MDR HIV-1
54 deaths over follow-up
321 with change of therapy within 6 months of diagnosis, and >3 months on new strategy.

DEMOGRAPHICS
85% male.
Median age: 43 yrs.
Calendar year of MDR diagnosis:
1998: 7%
1999: 26%
2000: 29%
2001: 20%
2002: 14%
2003: 4%
median CD4 count: 238
median viral load: 4.2 log copies/ml
median number of previous ART drugs: 8
median time on ART: 4.5 yrs
median no. of inactive drugs: 12 (9-16)
number of previous resistance tests:
1, 60%
2, 30%
3+, 10%

ESTIMATED PROBABILITY OF SURVIVAL AFTER MDR
Estimated probability of death was 3%, 8%, & 13% by 12, 24, and 36 months, respectively. Median followup was 24 months (IDR: 11-37).

RELATIONSHIP BETWEEN CHARACTERISTICS AT TIME OF MDR DIAGNOSIS & DEATH

CD4, VL, calendar year (the earlier the year) MDR was diagnosed, and the number of drugs previously used were predictive.

Category                                               Multivariate
                                               IRR (95% CI)     p
CD4 (per 100)                         0.58                0.013
VL (log copies/ml)                   2.09                0.006
Calendar Year MDR               0.56                0.024
(earlier the year, the worse)
no. of drugs prev exp             1.33                0.002
no. of resistance tests             1.0                 0.997
boosted PI in regimen            1.33                0.507
yrs of ART                               0.94               0.481


RELATIONSHIP BETWEEN TREATMENT STRATEGY & DEATH

The two key categories come out as significant: changing therapy when GSS is the same & changing therapy when GSS increases, both of which had beneficial effects. The other categories had too few numbers of patients to analyze.

Treatment strategy                         Multivariate
                                N             IRR (95% CI)     p
                         Total 321       
On therapy           150              1.00               0.143
Continued

Change therapy    96               0.35
GSS same

Change therapy   10               1.49
GSS decreases                   

Change therapy    82              0.31
GSS increases

Stop therapy        20               0.24
GSS same

Stop therapy        10              0.60
GSS decreases


RELATIONSHIP BETWEEN TREATMENT STRATEGY & VIRAL LOAD CHANGE: 24-48 weeks:

Change in therapy was the only category that showed benefit compared to the category of continuing therapy:
IRR: -0.32
--The other categories of strategy did not show significant viral load reduction.

RELATIONSHIP BETWEEN TREATMENT STRATEGY & CD4 COUNT CHANGE
Changing therapy when GSS increases was associated with significant CD4 increase (+51) compared to continuing therapy.