Use of surrogate markers for HIV disease monitoring and disease progression in the Australian HIV Observational Database (AHOD)

Background: In the current era of combination therapy (cART) patients are achieving long-term viral suppression and living significantly longer. In the context of lifelong patient management understanding the ongoing utility of well-known surrogate markers of disease progression and exploring potential new ones is increasingly pertinent. Objectives: The aims of this thesis are: 1) investigating the effect of reducing viral-load test frequency among patients with long-term virological control; 2) assessing the predictive value of CD4/CD8 ratio for disease progression and factors associated with a normalised CD4/CD8 ratio among patients in the Australian HIV Observational Database (AHOD). Methods: 1) The effect of reducing viral-load frequency was estimated by determining the rate of accumulated drug resistance by applying the rate of virological failure calculated in AHOD and the estimated additional time a patient remained on a failing regimen if HIV viral-load testing occurred annually. 2) The predictive value of CD4/CD8 ratio was assessed by investigating the predictors of AIDS and non-AIDS death using competing risk regression methods. Factors associated with achieving a normalised CD4/CD8 ratio (>1) was assessed using Cox regression methods.Results: 1) The rate of virological failure even among stable and suppressed patients was relatively high in AHOD (4/100 person years). If viral load monitoring was reduced to annually in this population, the mean extra time patients remained on a failing regimen was estimated at 6 months. Overall, this would equate to a potential increase in treatment resistance from 10% to 32%. 2) CD4/CD8 ratio was not an independent predictor of non-AIDS and AIDS deaths once adjusted for factors including CD4 cell count and suppressed viral load. Over a median of 6 years of follow-up, only 27% of patients achieved a normalised CD4/CD8 ratio. Key predictors of normalisation include viral-load, CD4 and CD8 cell count. Conclusions: Despite effective and well-tolerated regimens, stable and suppressed patients continue to fail therapy. Annual viral load monitoring could lead to a substantial increase in the numbers of failing patients who develop resistance. CD4/CD8 was not independent predictor of non-AIDS and AIDS death, and CD4 cell count still remains the key factor predicting HIV disease progression in AHOD