WHO/HIV/2013.39
© World Health Organization 2013
GRADE tables: What ARV regimen to switch to in adults, pregnant women, adolescents and children living with HIV (once-daily PI regimens)? ATZ/r versus LPV/r Author(s): Larry W Chang, Alicen B Spaulding, George W Rutherford Date: 2012-09-19 Question: Should ATZ/r vs LPV/r be used for HIV treatment failure? Settings: Low and Middle-Income Countries Bibliography: BMS 045 (2005, 2006)
Quality assessment
No. of patients
Effect Quality Importance
No. of studies
Design
Risk of bias
Inconsistency
Indirectness
Imprecision
Other considerations
ATZ/r LPV/r
Relative (95% CI)
Absolute
Mortality (96 Weeks) 1 randomized trials no serious risk of bias no serious inconsistency very serious1,2,3 very serious imprecision4 none 0/120 2/123 RR 0.2 (0.01 to (0%) (1.6%) 4.23) 13 fewer per 1000 (from 16 fewer to 53 more) ⊕OOO VERY LOW CRITICAL
Viral response (intent to treat, time to loss of viral response, <50 copies/ml, noncompleter = failure, 96 weeks) 1 randomized trials no serious risk of bias no serious inconsistency very serious1,2,3 very serious imprecision4 none 39/120 44/123 RR 0.91 (0.64 to 32 fewer per 1000 (from 129 fewer to (32.5%) (35.8%) 104 more) 1.29) ⊕OOO VERY LOW CRITICAL
Severe adverse events (96 weeks) 1 randomized trials no serious risk of bias no serious inconsistency very serious1,2,3 very serious imprecision4 none 16/120 13/123 RR 1.26 (0.63 to 27 more per 1000 (from 39 fewer to (13.3%) (10.6%) 2.51) 160 more) ⊕OOO VERY LOW CRITICAL
Immune response (96 weeks) 1 randomized trials no serious risk of bias no serious inconsistency very serious1,2,3 very serious imprecision4 none 120 123 Median increase, 160 cells/mm3 versus 142 cells/mm3 ⊕OOO VERY LOW IMPORTANT
Retention on treatment (96 weeks) 1 randomized trials no serious risk of bias no serious inconsistency very serious1,2,3 very serious imprecision4 none 67/120 65/123 RR 1.06 (0.84 to 32 more per 1000 (from 85 fewer to (55.8%) (52.8%) 1.33) 174 more) ⊕OOO VERY LOW IMPORTANT
This work was commissioned by the World Health Organization and carried out by The University of California, San Francisco (UCSF), Cochrane Review Group on HIV/AIDS 1
WHO/HIV/2013.39 1 2
© World Health Organization 2013
The study was conducted primarily in high-income countries. The regimens previously failing participants differed significantly from WHO first-line regimens, both in quantity and in content. 3 The NRTI backbones used in new regimen included tenofovir and another NRTI selected with guidance from phenotypic sensitivity testing, an option not available in most low- and middle-income settings. 4 Very few events.
DRV/r versus LPV/r Author(s): Larry W. Chang, Alicen B. Spaulding, George W. Rutherford Date: 2012-12-10 Question: DRV/r versus LPV/r or boosted PI for second-line ART Settings: Low- and middle-income countries Bibliography: POWER (2007, 2009); TITAN (2007, 2012)
Quality assessment
No. of patients
Effect Quality Importance
No. of studies
Design
Risk of bias
Inconsistency Indirectness Imprecision
Other LPV/r or DRV/r considerations boosted PI
Relative (95% CI)
Absolute
Mortality (48 Weeks) 2 randomized no serious no serious trials risk of bias inconsistency serious1–5 very serious6 none 7/429 (1.6%) 4/421 (0.95%) RR 1.7 (0.5 to 5.81) 7 more per 1000 (from 5 fewer to 46 more) CRITICAL ⊕OOO VERY LOW
Virologic Response (intent to treat, time to loss of viral response, <50 copies/ml, 96 weeks) 2 randomized no serious no serious trials risk of bias inconsistency serious1–5 no serious imprecision none 231/429 (53.8%) 175/421 (41.6%) RR 1.31 (1.14 to 1.5) 129 more per 1000 (from 58 more to 208 more) CRITICAL ⊕⊕⊕O MODERATE
Severe adverse events (48 weeks) 2 randomized no serious no serious trials risk of bias inconsistency serious1–5 very serious6 none 54/429 (12.6%) 48/421 (11.4%) RR 1.1 (0.76 to 1.58) 11 more per 1000 (from 27 fewer to 66 more) CRITICAL ⊕OOO VERY LOW
Immune response (96 weeks) (better indicated by higher values) 2 randomized no serious no serious trials risk of bias inconsistency serious1,2,3,4,5 no serious imprecision none 429 421 – First trial = median increase from baseline, 81 cells/mm3 versus 93 cells/mm3; second trial = mean absolute increase, 133 cells/mm3 versus 15 cells/mm3 ⊕⊕OO LOW IMPORTANT
On Treatment Retention (48 Weeks)
This work was commissioned by the World Health Organization and carried out by The University of California, San Francisco (UCSF), Cochrane Review Group on HIV/AIDS 2
WHO/HIV/2013.39 2 randomized no serious no serious trials risk of bias inconsistency serious1–5 no serious imprecision none 339/429 (79%) 235/421 (55.8%) RR 1.42 (1.29 to 1.57)
© World Health Organization 2013 CRITICAL ⊕⊕⊕O MODERATE
234 more per 1000 (from 162 more to 318 more)
1 2
The study was conducted primarily in high-income countries. The regimens previously failing participants differed significantly from WHO first-line regimens, both in quantity and in content. 3 DRV/r was twice-daily 600 mg/100 mg dosing rather than daily dosing. 4 Comparison group in POWER 1, 2 studies included a mix of boosted PIs; only 36% were on a LPV/r-based regimen. 5 NRTI backbones were used in a new regimen selected using resistance testing data and may have included antiretroviral drugs not typically available in most low- and middle-income settings. 6 Very few events.
Use of ATZ/r or DRV/r in first-line regimens Author(s): George W. Rutherford, Tara Horvath Date: 2012-11-09 Question: ATV/r versus LPV/r for second-line antiretroviral therapy among people living with HIV (first-line therapy trials) Settings: Argentina, Australia, Austria, Belgium, Brazil, Canada, Chile, Columbia, Costa Rica, France, Germany, Hong Kong SAR, Indonesia, Italy, Mexico, Netherlands, Norway, Panama, Peru, Portugal, Puerto Rico, Singapore, Spain, South Africa, Sweden, Thailand, United Kingdom, United States Bibliography: Molina 2008, Molina 2010, Josephson 2010
Quality assessment
No. of patients
Effect Quality Importance
No. of studies
Design
Risk of bias
Inconsistency
Indirectness
Imprecision
Other considerations
ATV/r LPV/r
Relative (95% CI)
Absolute
Mortality (96 weeks) 1 randomized trials no serious risk of bias no serious inconsistency serious1 very serious2 none 6/440 5/443 RR 1.21 (0.37 to 2 more per 1000 (from 7 fewer to 33 CRITICAL ⊕OOO (1.4%) (1.1%) 3.93) more) VERY LOW
Plasma VL <50 copies/ml (96 weeks) 1 randomized trials no serious risk of bias no serious inconsistency serious1 no serious imprecision none 326/440 295/443 (74.1%) (66.6%) RR 1.09 (1 to 1.19) 60 more per 1000 (from 0 more to 127 CRITICAL ⊕⊕⊕O more) MODERATE
Plasma VL <50 copies/ml (48 weeks) 2 randomized trials no serious risk of bias no serious inconsistency serious1 no serious imprecision none 405/508 400/508 RR 1.01 (0.95 to 8 more per 1000 (from 39 fewer to 63 CRITICAL ⊕⊕⊕O (79.7%) (78.7%) 1.08) more) MODERATE
Increase in CD4 cells/mm3 (96 weeks) (better indicated by higher values)
This work was commissioned by the World Health Organization and carried out by The University of California, San Francisco (UCSF), Cochrane Review Group on HIV/AIDS 3
WHO/HIV/2013.39 1 randomized trials no serious risk of bias no serious inconsistency serious1 no serious imprecision none 440 443 –
© World Health Organization 2013 CRITICAL ⊕⊕⊕O MODERATE
Mean difference 21.2 lower (43.3 lower to 0.9 higher)
Severe adverse events (96 weeks) 1 randomized trials no serious risk of bias no serious inconsistency serious1 serious3 none 62/441 48/437 RR 1.28 (0.9 to 31 more per 1000 (from 11 fewer to 90 (14.1%) (11%) 1.82) more) ⊕⊕OO LOW CRITICAL
Retained in care (96 weeks) 1 randomized trials no serious risk of bias no serious inconsistency serious1 serious3 none 72/438 95/440 (16.4%) (21.6%) RR 1.07 (1 to 1.14) 15 more per 1000 (from 0 more to 30 more) ⊕⊕OO LOW CRITICAL
<100% adherence (self-report) (96 weeks) 1 randomized trials no serious risk of bias no serious inconsistency very serious1,4 no serious imprecision none 361/440 372/443 RR 0.98 (0.92 to (82%) (84%) 1.04) 17 fewer per 1000 (from 67 fewer to 34 more) ⊕⊕OO LOW CRITICAL
1 2
First-line therapy. Very few events. 3 Few events. 4 Self-report. Author(s): George W. Rutherford, Tara Horvath Date: 2012-11-09 Question: DRV/r versus LPV/r for second-line antiretroviral therapy among people living with HIV (first-line therapy trials) Settings: Argentina, Australia, Austria, Belgium, Canada, Chile, Costa Rica, Denmark, France, Germany, Greece, Guatemala, Italy, Malaysia, Mexico, Panama, Puerto Rico, Russian Federation, Singapore, South Africa, Spain, Switzerland, Taiwan, China, Thailand, United Kingdom, United States Bibliography: Orkin 2012
Quality assessment
No. of patients
Effect Quality Importance
No. of studies
Design
Risk of bias
Inconsistency
Indirectness
Imprecision
Other considerations
DRV/r LPV/r
Relative (95% CI)
Absolute
Mortality (192 weeks) 1 randomized trials no serious risk of bias no serious inconsistency serious1 very serious2 none 4/343 7/346 RR 0.58 (0.17 to 8 fewer per 1000 (from 17 fewer to 19 CRITICAL ⊕OOO (1.2%) (2%) 1.95) more) VERY LOW
Plasma VL <50 copies/ml (192 weeks)
This work was commissioned by the World Health Organization and carried out by The University of California, San Francisco (UCSF), Cochrane Review Group on HIV/AIDS 4
WHO/HIV/2013.39 1 randomized trials no serious risk of bias no serious inconsistency serious1 no serious imprecision none 236/343 198/346 RR 1.2 (1.07 to (68.8%) (57.2%) 1.35)
© World Health Organization 2013
114 more per 1000 (from 40 more to CRITICAL ⊕⊕⊕O 200 more) MODERATE
Increase in CD4 cells/mm3 (192 weeks) (better indicated by higher values) 1 randomized trials no serious risk of bias no serious inconsistency serious1 no serious imprecision none 343 346 – Mean difference 5 lower (19.93 lower CRITICAL ⊕⊕⊕O to 9.93 higher) MODERATE
≥1 grade 3 or 4 adverse events (192 weeks) 1 randomized trials no serious risk of bias no serious inconsistency serious1 serious3 none 103/343 110/346 RR 0.94 (0.76 to (30%) (31.8%) 1.18) 19 fewer per 1000 (from 76 fewer to 57 more) ⊕⊕OO LOW CRITICAL
>95% adherence (Modified Medication Adherence Self-Report Inventory) (192 weeks) 1 randomized trials no serious risk of bias no serious inconsistency serious1 no serious imprecision none 286/343 271/346 RR 1.06 (0.99 to 47 more per 1000 (from 8 fewer to 117 CRITICAL ⊕⊕⊕O (83.4%) (78.3%) 1.15) more) MODERATE
1 2
First-line therapy. Very few events. 3 Few events.
This work was commissioned by the World Health Organization and carried out by The University of California, San Francisco (UCSF), Cochrane Review Group on HIV/AIDS 5