Abstract Body

Background: In the Cryptococcal Optimal ART Timing (COAT) trial, early ART initiation (7-13 days) after cryptococcal meningitis (CM) was associated with higher mortality compared with deferred (5 weeks) ART initiation. A significant proportion of deaths (70%, 49/70) occurred after completion of 14 days of amphotericin induction therapy. We hypothesized that serum biomarkers may enable risk stratification of short and long term mortality. Methodology: 177 COAT subjects were randomized in Kampala and Mbarara, Uganda and Cape Town, South Africa. 156 participants survived to the end of amphotericin therapy of which 153 had 22 serum biomarkers profiled using ELISA or Luminex multiplex assay. We compared by logistic regression log2- transformed biomarkers versus 5 week survival (i.e. time of outpatient clinic registration) and 46 week survival, adjusting for randomized arm (early ART vs. deferred ART). Results: Of 153 persons who survived 14 days after cryptococcal meningitis diagnosis, 121 (81%) survived for 5 weeks until time of outpatient clinic registration and 104 (68%) survived >1 year. Twelve serum biomarkers were associated with increased risk of mortality. Odds Ratio (OR) for mortality for each two-fold increase in serum biomarker is presented in the Table. Serum CRP was a consistent risk factor for short and long term mortality. Additionally, when CRP was measured at 21 days after diagnosis (2 weeks after randomization), CRP was associated with mortality between 3-5 weeks (OR 2.22 per 2-fold increase (95%CI: 1.35, 3.66, P=.0002) as well as mortality through one year (OR 1.45 per 2-fold increase, 95% CI: 1.08, 1.95; P=.01). Conclusions: Serum biomarkers at the end of amphotericin induction therapy may enable risk stratification of short and long term mortality, most notably including CRP, a commonly available routine lab test. Through risk stratification, close clinical follow up may be wise for individuals who have survived their initial cryptococcal meningitis but remain at high risk of death.