Background: In virologically suppressed HIV-infected adults, CD4/CD8 ratio has been inversely associated with risk of non-communicable diseases (NCDs). Very low CD4/CD8 ratio (<0.4) despite antiretroviral therapy (ART) also correlates with higher measures of immunosenescence. The interaction of age and CD4/CD8 and its association with future NCD events has not been described.
Methods: We assessed the association of CD4/CD8 ratio and serious NCDs (cardiovascular [coronary artery, cerebrovascular, and peripheral vascular disease], malignancy, liver, and renal diseases) in a cohort of HIV-infected adults after their first year of suppressed HIV RNA (defined as baseline). We examined patient characteristics by baseline CD4/CD8 ratio and CD4/CD8 ratio change from baseline by linear regression. We used Cox proportional hazard models to assess baseline CD4/CD8 ratio and risk of future NCDs.
Results: Between 1998-2010, 1700 patients achieved virologic suppression for one year and were included in this study (median follow-up time=3 years). Compared to those with CD4/CD8 ratio >0.7, patients with low CD4/CD8 ratio (<0.4) were older (median 44 vs. 40 years, p<0.01), more likely to be male (86 vs. 70%, p<0.01), and had lower CD4+ lymphocyte counts (median 279 vs. 640 cells/mL, p<0.01). There was no difference in prior ART or follow-up duration. Among those with consistent virologic suppression after three years (n=454), older patients (>50 years) had lower CD4/CD8 gain compared to younger patients (<40 years), after adjusting for baseline ratio (beta = -0.07, p=0.03). There were 123 serious NCDs, including 48 cardiovascular disease (CVD) and 30 cancer events (median time to first CVD or cancer event=2.3 years). Compared to patients with no NCDs during follow-up, only those with CVD and cancer outcomes had statistically lower baseline CD4/CD8 ratios (see Figure). In a model adjusting for age, sex, and CD4+ lymphocyte count, a higher CD4/CD8 ratio remained associated with a lower risk of CVD and cancer events (composite aHR per 0.1 increase = 0.90 [95% CI: 0.81-1.00]). An interaction term for age and CD4/CD8 ratio was not statistically significant (p=0.46).
Conclusions: Low CD4/CD8 ratio after one year of suppressed HIV RNA was independently predictive of serious CVD and cancer events. Older adults had lower CD4/CD8 ratio and had less improvement in CD4/CD8 ratio over time. Further study of CD4/CD8 ratio as a biomarker for immunosenescence and risk factor for CVD and cancer in aging HIV-infected adults is needed.
Distribution of CD4/CD8 ratio by non-communicable disease (NCD) outcomes in a cohort of 1,700 virologically suppressed HIV-infected adults, 1998-2010