Marginal Structural Cox Model Estimates for the Effect of the Third Active Class in Antiretroviral Therapy (ART) on CD4/CD8 Ratio Among Treatment-Experienced Patients Receiving a Second or Subsequent ART Regimen
ART Class . | Person-years . | Incident Number . | Crude HR (95% CI) . | IPTWa IPCW Weighted Marginal Structural Model, HR (95% CI) . |
---|---|---|---|---|
Ratio CD4/CD8 ≥0.3 | ||||
INSTI + 2 NRTIs | 659.9 | 553 | 1 (reference) | 1 (reference) |
NNRTI + 2 NRTIs | 1016.2 | 686 | .79 (.69–.89) | .78 (.64–.95) |
PI + 2 NRTIs | 1924.6 | 1431 | .86 (.77–.96) | .80 (.67–.95) |
Ratio CD4/CD8 ≥0.5 | ||||
INSTI + 2 NRTIs | 902.7 | 506 | 1 (reference) | 1 (reference) |
NNRTI + 2 NRTIs | 1464.8 | 626 | .75 (.65–.85) | .73 (.61–.87) |
PI + 2 NRTIs | 3000.3 | 1228 | .70 (.63–.79) | .64 (.55–.75) |
Ratio CD/CD8 ≥0.8 | ||||
INSTIs + 2 NRTIs | 1412.5 | 365 | 1 (reference) | 1 (reference) |
NNRTIs + 2 NRTIs | 2498.3 | 485 | .75 (.65–.87) | .77 (.65–.90) |
PIs + 2 NRTIs | 5006.1 | 810 | .62 (.54–.70) | .57 (.49–.67) |
Ratio CD4/CD8 ≥1 | ||||
INSTIs + 2 NRTIs | 1729.2 | 290 | 1 (reference) | 1 (reference) |
NNRTIs + 2 NRTIs | 3039.7 | 387 | .77 (.65–.91) | .78 (.66–.92) |
PIs + 2 NRTIs | 5955.2 | 601 | .59 (.51–.69) | .57 (.49–.67) |
Ratio CD4/CD8 ≥1.2 | ||||
INSTIs + 2 NRTIs | 1978.3 | 211 | 1 (reference) | 1 (reference) |
NNRTIs + 2 NRTIs | 3495.6 | 284 | .76 (.64–.92) | .78 (.65–.95) |
PIs + 2 NRTIs | 6740.7 | 414 | .57 (.48–.67) | .56 (.47–.68) |
ART Class . | Person-years . | Incident Number . | Crude HR (95% CI) . | IPTWa IPCW Weighted Marginal Structural Model, HR (95% CI) . |
---|---|---|---|---|
Ratio CD4/CD8 ≥0.3 | ||||
INSTI + 2 NRTIs | 659.9 | 553 | 1 (reference) | 1 (reference) |
NNRTI + 2 NRTIs | 1016.2 | 686 | .79 (.69–.89) | .78 (.64–.95) |
PI + 2 NRTIs | 1924.6 | 1431 | .86 (.77–.96) | .80 (.67–.95) |
Ratio CD4/CD8 ≥0.5 | ||||
INSTI + 2 NRTIs | 902.7 | 506 | 1 (reference) | 1 (reference) |
NNRTI + 2 NRTIs | 1464.8 | 626 | .75 (.65–.85) | .73 (.61–.87) |
PI + 2 NRTIs | 3000.3 | 1228 | .70 (.63–.79) | .64 (.55–.75) |
Ratio CD/CD8 ≥0.8 | ||||
INSTIs + 2 NRTIs | 1412.5 | 365 | 1 (reference) | 1 (reference) |
NNRTIs + 2 NRTIs | 2498.3 | 485 | .75 (.65–.87) | .77 (.65–.90) |
PIs + 2 NRTIs | 5006.1 | 810 | .62 (.54–.70) | .57 (.49–.67) |
Ratio CD4/CD8 ≥1 | ||||
INSTIs + 2 NRTIs | 1729.2 | 290 | 1 (reference) | 1 (reference) |
NNRTIs + 2 NRTIs | 3039.7 | 387 | .77 (.65–.91) | .78 (.66–.92) |
PIs + 2 NRTIs | 5955.2 | 601 | .59 (.51–.69) | .57 (.49–.67) |
Ratio CD4/CD8 ≥1.2 | ||||
INSTIs + 2 NRTIs | 1978.3 | 211 | 1 (reference) | 1 (reference) |
NNRTIs + 2 NRTIs | 3495.6 | 284 | .76 (.64–.92) | .78 (.65–.95) |
PIs + 2 NRTIs | 6740.7 | 414 | .57 (.48–.67) | .56 (.47–.68) |
N = 2866. Abbreviations: CI, confidence interval; HR, hazard ratio; INSTI, integrase strand transfer inhibitor; IPCW, inverse probability of censoring weights; IPTW, inverse probability of treatment weights; NNRTI, nonnucleoside reverse transcriptase inhibitor; NRTI, nucleoside reverse transcriptase inhibitor; PI, protease inhibitor.
Weighted variables: age (continuous), delay in ART treatment initiation (continuous), nadir CD4 (<200, 200–350 and >350 cells/mm3), and cytomegalovirus serostatus before inclusion (yes or no). CD4, CD8, and viral load variables (continuous) were considered time-dependent during follow-up.
Marginal Structural Cox Model Estimates for the Effect of the Third Active Class in Antiretroviral Therapy (ART) on CD4/CD8 Ratio Among Treatment-Experienced Patients Receiving a Second or Subsequent ART Regimen
ART Class . | Person-years . | Incident Number . | Crude HR (95% CI) . | IPTWa IPCW Weighted Marginal Structural Model, HR (95% CI) . |
---|---|---|---|---|
Ratio CD4/CD8 ≥0.3 | ||||
INSTI + 2 NRTIs | 659.9 | 553 | 1 (reference) | 1 (reference) |
NNRTI + 2 NRTIs | 1016.2 | 686 | .79 (.69–.89) | .78 (.64–.95) |
PI + 2 NRTIs | 1924.6 | 1431 | .86 (.77–.96) | .80 (.67–.95) |
Ratio CD4/CD8 ≥0.5 | ||||
INSTI + 2 NRTIs | 902.7 | 506 | 1 (reference) | 1 (reference) |
NNRTI + 2 NRTIs | 1464.8 | 626 | .75 (.65–.85) | .73 (.61–.87) |
PI + 2 NRTIs | 3000.3 | 1228 | .70 (.63–.79) | .64 (.55–.75) |
Ratio CD/CD8 ≥0.8 | ||||
INSTIs + 2 NRTIs | 1412.5 | 365 | 1 (reference) | 1 (reference) |
NNRTIs + 2 NRTIs | 2498.3 | 485 | .75 (.65–.87) | .77 (.65–.90) |
PIs + 2 NRTIs | 5006.1 | 810 | .62 (.54–.70) | .57 (.49–.67) |
Ratio CD4/CD8 ≥1 | ||||
INSTIs + 2 NRTIs | 1729.2 | 290 | 1 (reference) | 1 (reference) |
NNRTIs + 2 NRTIs | 3039.7 | 387 | .77 (.65–.91) | .78 (.66–.92) |
PIs + 2 NRTIs | 5955.2 | 601 | .59 (.51–.69) | .57 (.49–.67) |
Ratio CD4/CD8 ≥1.2 | ||||
INSTIs + 2 NRTIs | 1978.3 | 211 | 1 (reference) | 1 (reference) |
NNRTIs + 2 NRTIs | 3495.6 | 284 | .76 (.64–.92) | .78 (.65–.95) |
PIs + 2 NRTIs | 6740.7 | 414 | .57 (.48–.67) | .56 (.47–.68) |
ART Class . | Person-years . | Incident Number . | Crude HR (95% CI) . | IPTWa IPCW Weighted Marginal Structural Model, HR (95% CI) . |
---|---|---|---|---|
Ratio CD4/CD8 ≥0.3 | ||||
INSTI + 2 NRTIs | 659.9 | 553 | 1 (reference) | 1 (reference) |
NNRTI + 2 NRTIs | 1016.2 | 686 | .79 (.69–.89) | .78 (.64–.95) |
PI + 2 NRTIs | 1924.6 | 1431 | .86 (.77–.96) | .80 (.67–.95) |
Ratio CD4/CD8 ≥0.5 | ||||
INSTI + 2 NRTIs | 902.7 | 506 | 1 (reference) | 1 (reference) |
NNRTI + 2 NRTIs | 1464.8 | 626 | .75 (.65–.85) | .73 (.61–.87) |
PI + 2 NRTIs | 3000.3 | 1228 | .70 (.63–.79) | .64 (.55–.75) |
Ratio CD/CD8 ≥0.8 | ||||
INSTIs + 2 NRTIs | 1412.5 | 365 | 1 (reference) | 1 (reference) |
NNRTIs + 2 NRTIs | 2498.3 | 485 | .75 (.65–.87) | .77 (.65–.90) |
PIs + 2 NRTIs | 5006.1 | 810 | .62 (.54–.70) | .57 (.49–.67) |
Ratio CD4/CD8 ≥1 | ||||
INSTIs + 2 NRTIs | 1729.2 | 290 | 1 (reference) | 1 (reference) |
NNRTIs + 2 NRTIs | 3039.7 | 387 | .77 (.65–.91) | .78 (.66–.92) |
PIs + 2 NRTIs | 5955.2 | 601 | .59 (.51–.69) | .57 (.49–.67) |
Ratio CD4/CD8 ≥1.2 | ||||
INSTIs + 2 NRTIs | 1978.3 | 211 | 1 (reference) | 1 (reference) |
NNRTIs + 2 NRTIs | 3495.6 | 284 | .76 (.64–.92) | .78 (.65–.95) |
PIs + 2 NRTIs | 6740.7 | 414 | .57 (.48–.67) | .56 (.47–.68) |
N = 2866. Abbreviations: CI, confidence interval; HR, hazard ratio; INSTI, integrase strand transfer inhibitor; IPCW, inverse probability of censoring weights; IPTW, inverse probability of treatment weights; NNRTI, nonnucleoside reverse transcriptase inhibitor; NRTI, nucleoside reverse transcriptase inhibitor; PI, protease inhibitor.
Weighted variables: age (continuous), delay in ART treatment initiation (continuous), nadir CD4 (<200, 200–350 and >350 cells/mm3), and cytomegalovirus serostatus before inclusion (yes or no). CD4, CD8, and viral load variables (continuous) were considered time-dependent during follow-up.
This PDF is available to Subscribers Only
View Article Abstract & Purchase OptionsFor full access to this pdf, sign in to an existing account, or purchase an annual subscription.