CROI 2016 Abstract eBook

Abstract Listing

Poster Abstracts

Results: 62 paradoxical cryptococcal-IRIS events did not consistently present with substantial reconstitution of CD4+ T cell counts (61% increased by >25 cells/mL), nor did these CD4 values distinguish IRIS from 13 culture-positive relapse events (54%with increases of >25 cells/µL). All IRIS cases had excellent, appropriate virologic responses, yet 50% (6/12) with culture-positive relapse did not have HIV viral suppression (>40,000 copies/mL min). The median CSF WBC counts did not differ at time of cryptococcal-IRIS event (median 25; IQR, 5 to 85, max 240 cells/mcL) versus relapse (median 15; IQR, <5 to 40, max 1150 cells/mcL) ( P =.53). Immune parameters which distinguished IRIS at event included 3-4 fold greater CSF levels of: Th1 cytokine IFN-g ( P =.006), Th2 cytokine IL-4 ( P =.011); pro-inflammatory IL-17( P =.038). Persons with culture-positive relapse did have markedly higher CSF IL-13 levels (relapse: median 253 (IQR, 63-471) pg/mL vs. IRIS: 6.9 (IQR, 2.4-17) pg/mL, P =.006) and non-statistically higher GM-CSF levels ( P =.17). Conclusions: CD4 counts, plasma HIV RNA, nor CSF WBCs at event onset did not consistently discriminated IRIS from relapse. The distinct immunologic signatures that did distinguish the two clinical scenarios, included 35-fold higher CSF IL-13 levels in relapse, a Th2 cytokine which is non-protective and associated with uncontrolled cryptococcal infection in murine models. In IRIS, multiple T-cell cytokines were increased (IFN-g, IL-4, IL-17) at event. In resource-limited settings, those with CSF culture-positive relapse should be targeted for viral load testing. 763 Acute Kidney Injury and Urine Biomarkers in HIV-Associated Cryptococcal Meningitis Charlotte Schutz 1 ; David R. Boulware 2 ; Katherine Huppler Hullsiek 2 ; Joshua R. Rhein 2 ; KabandaTaseera 3 ; Conrad Muzoora 3 ; David B. Meya 4 ; FriedrichThienemann 1 ; Maximilian von Hohenberg 2 ; Graeme Meintjes 1 ; for the COATTrialTeam 1 Univ of Cape Town, Cape Town, South Africa; 2 Univ of Minnesota, Minneapolis, MN, USA; 3 Mbarara Univ of Sci and Tech, Mbarara, Uganda; 4 Infectious Diseases Inst, Kampala, Uganda Background: Cryptococcal meningitis is a major cause of mortality in HIV-infected persons. Amphotericin B deoxycholate remains key to effective treatment despite its toxicities, including acute kidney injury (AKI). Current AKI definitions are based on serum creatinine or urine output, both late indicators of kidney injury. Biomarkers that could predict AKI earlier could be useful in directing amphotericin management. Neutrophil gelatinase-associated lipocalin (NGAL), is an early marker of ischemic AKI. Urinary Cystatin C (CysC) is a marker of tubular dysfunction. Tissue inhibitor of metallaproteinases-2 (TIMP-2) can predict AKI in critically ill patients. We assessed these urine biomarkers in patients with cryptococcosis treated with amphotericin. Methods: Participants were prospectively enrolled into the Cryptococcal Optimal ART Timing (COAT) trial from 2010 to 2012 and were treated with daily amphotericin 0.7-1.0 mg/kg and fluconazole for 2 weeks. Baseline and follow up serum creatinine concentrations were measured. Estimated glomerular filtration rate (GFR) was calculated using the Modification of Diet in Renal Disease (MDRD) study equation, and AKI defined as decrease in GFR to <60 mL/min/1.73m 2 within 4 weeks of diagnosis. We measured NGAL, CysC and TIMP-2 among 130 participants with stored urine samples collected a median of 4 days (IQR: 1- 4) from diagnosis. We explored AKI incidence, risk factors for incident AKI, and associations with mortality using univariate and multivariate Cox proportional hazards models. Results: Among 130 participants, median age was 35 years (IQR: 30-40), 52% (n=68) were men, median CD4 T cell count was 21 cells/mcL (IQR: 9-74), and 44% (n=57) died within 12 months. Four patients (3%) had GFR<60 at baseline. Incident AKI occurred in 42% (53/126) with GFR<60. Incident AKI was independently associated with mortality (adjusted hazard ratio [aHR]=3.5, 95%CI: 1.95-6.1). Persons developing AKI had higher baseline CD4 T cell counts (49 vs. 14 cells/mcL, p<0.01). Higher urinary TIMP-2 was associated with development of AKI in univariate (p=0.02) and multivariate analysis (aHR=1.4, 95%CI: 1.02-1.94 per each doubling of TIMP-2) but TIMP-2 was not associated with mortality. Conclusions: AKI occurred in 42% of HIV-infected patients treated with amphotericin B deoxycholate for cryptococcal meningitis, and AKI was associated with mortality. TIMP-2 shows promise as a urine marker for antecedent prediction of amphotericin-associated AKI.

Poster Abstracts

764 Pulmonary Aspergillosis May Be Common in AIDS With Smear Negative Tuberculosis Iain D. Page 1 ;WilliamWorodria 2 ; Alfred Andama 2 ; Irene Ayakaka 2 ; Lucian Davis 3 ; Laurence Huang 3 ; Malcolm Richardson 1 ; DavidW. Denning 1 1 Univ of Manchester, Manchester, UK; 2 Mulago Hosp, Kampala, Uganda; 3 Univ of California San Francisco, San Francisco, CA, USA

Background: Smear-negative tuberculosis (TB) is common in AIDS in the developing world. WHO diagnostic guidelines require persistent productive cough plus abnormal chest x-ray and failure to respond to antibiotics. Microbiological proof of TB infection is not required. Subacute invasive aspergillosis (SAIA) also occurs in AIDS, with a clinical and radiological presentation that is very similar to tuberculosis. Autopsy studies of AIDS patients show aspergillosis occurs in 3% of all AIDS deaths, but ante-mortem diagnosis occurs in less than 10% of these cases. Measurement of Aspergillus -specific IgG is key to the diagnosis of SAIA, but rarely available in areas of high tuberculosis prevalence. The frequency of SAIA in patients presenting with AIDS and apparent smear-negative tuberculosis is not known. The Siemens Aspergillus -specific IgG test has a sensitivity of 96% and specificity of 98% for the diagnosis of chronic pulmonary aspergillosis using a diagnostic cut-off of 10 mg/L (1). We have used this assay to investigate levels of Aspergillus -specific IgG in patients with AIDS and smear negative tuberculosis. Methods: We tested sera from 100 healthy Ugandan controls and 39 HIV infected persons admitted to Mulago Hospital, Kampala, Uganda with apparent smear negative tuberculosis, with chronic cough and an abnormal chest X-ray, but no evidence of tuberculosis or other diagnosis after thorough investigation including sputum culture, GeneXpert PCR testing and bronchoscopy. Results: The mean patient age was 35 years and 59% of patients were female. Mean CD4 count was 109 cells/mL and 44% of patients had CD4 count <50 cells/mL. Raised Aspergillus -specific IgG was present in 2% of healthy controls, but 26% of patients (95% CI 14 – 41%, p 0.000). 40% of those with a positive test died within 2 months of sampling. Conclusions: The presence of raised Aspergillus -specific IgG in the context of chronic cough, abnormal chest X-ray and exclusion of other conditions is highly suggestive of either subacute or chronic pulmonary aspergillosis. SAIA is probably occurring frequently in Ugandan in patients with AIDS and being misdiagnosed as smear-negative tuberculosis. Further prospective studies with CT scanning, plus effective fungal culture and serology should be performed to investigate this possibility. Reference 1 - Page ID, Richardson M, Denning DW. Comparison of six Aspergillus -specific IgG assays for the diagnosis of chronic pulmonary aspergillosis (CPA). 2015. Journal of Infection. In press.

318

CROI 2016