The central nervous system (CNS) is believed to be a significant reservoir for pathogens such as HIV and Cryptococcus; however, current understanding of drug penetration into the CNS is limited and largely based on cerebrospinal fluid (CSF) concentrations. However, CSF is not brain tissue. Herein we used tissues collected post-mortem from HIV-infected Ugandan subjects co-infected with Cryptococcus to characterize the relative distribution of antiretrovirals and antifungal agents across plasma and CNS compartments.
Following obtainment of written, informed consent from next of kin, post-mortems were performed on five subjects co-infected with HIV and cryptococcal meningitis. Tissues from cerebellum, pons, and CSF were snap frozen in liquid nitrogen. Whole blood was collected from femoral vein into EDTA tubes and stored on ice for 1 hour before separating plasma from cell pellets. All samples were transferred to -80 ºC for storage. Following tissue homogenization, drug quantification was performed using high performance (efavirenz) or ultrahigh performance (tenofovir, lamivudine, fluconazole) liquid chromatography coupled with triple quadrupole mass spectrometer. Calibrator standards and quality control (QC) samples were prepared in the matrix to match the sample tested; bovine brain homogenate was used for brain tissue, a solution of salt and proteins for CSF, and lithium heparin for plasma. Data are reported as median (range).
Post-mortems were performed 5.2 (2.2-13.7) hours following death. Three individuals receiving daily tenofovir/lamivudine/efavirenz (300/300/600 mg) had detectable drug in all tissue compartments. Likewise, four individuals receiving fluconazole (400-1200 mg daily) had detectable drug in all compartments. CSF: plasma ratios were similar to values reported in the literature. Drug exposure in brain tissues was consistently lower than CSF for tenofovir, lamivudine, and fluconazole and higher than CSF for efavirenz (see Table).
Tenofovir, lamivudine, and fluconazole exposure in CSF over-predicted brain tissue penetration. In contrast, CSF exposure of efavirenz under-predicted brain exposure. These findings highlight the limitations of CSF as a surrogate for overall CNS drug exposure. Validation in larger cohorts and in additional CNS tissue compartments is warranted. These data support the use of post-mortem tissues to assess drug distribution to and within the CNS, relevant for HIV reservoir eradication.