CROI 2020 Abstract eBook

Abstract eBook

Poster Abstracts

patients from Arms 1 and 2. PK samples were collected prior to and 1, 2, 5, 8, and 24 hr post-dose on day 14 and a standard meal was provided. Results: A one-compartment model with first-order elimination and transit compartment absorption fitted the data well. Allometric scaling using body weight was applied to clearance (CL) and volume of distribution. Patients taking rifampin had 44 % reduction in AUC compared to rifabutin. The individual median estimates of C max were 2.15 and 3.40 mg/L for rifampin and rifabutin, respectively. For the AUC0-24, the values were 29.9 and 58.9 mg•h/L. CL was 20% higher in men than women. Conclusion: As part of a multidrug regimen, co-administration of pretomanid with rifampin increases the CL substantially compared to rifabutin. However, exposures in the rifampin arm in our study were similar to those seen in patients taking 200mg of pretomanid alone, without food (AUC of 36 mg•h/L). Though pretomanid co-administered with rifabutin is more likely to maintain exposure levels equal to or exceeding regimens that do not contain rifamycins, the reduced exposure with rifampin is less pronounced when given with food and may still permit co-administration.

longer durations. The significant PK/PLT effect was small, consistent with thrombocytopenia reported in only 6% of subjects. Conclusion: QD and BID dosing had comparable toxicity. PN is reversible and week 4 Hb levels can be used to guide early dose adjustments to prevent anemia.

Poster Abstracts

735 HIV VIROLOGICAL OUTCOMES DURING TENOFOVIR ALAFENAMIDE AND RIFABUTIN COADMINISTRATION Shannon Balcombe 1 , Thomas C. Martin 1 , Michael E. Tang 1 , Lucas Hill1 1 University of California San Diego, San Diego, CA, USA Background: Tenofovir alafenamide (TAF) absorption may be decreased during rifabutin (RFB) therapy due to induction of P-glycoprotein. Current USA HIV guidelines therefore recommend against coadministration of TAF and RFB. Based on expert opinion, some centres have administered TAF containing HIV regimens with RFB; however, clinical outcomes have not been assessed. Methods: Retrospective observational study of all individuals at UC San Diego who received a TAF containing HIV regimen coadministered with RFB for ≥1 month between April 2016 and July 2019. The primary outcome was defined as documented HIV VL ≤200 copies/ml after initiating TAF/RFB therapy or maintenance of viral suppression (≤200 copies/ml) for participants already established on antiretroviral therapy (ART) prior to TAF/RFB coadministration. Cases with suspected virological failure were further evaluated for potential alternate causes other than TAF/RFB interaction. Results: 23 patients were included in the analysis. Demographics and HIV related variables are shown in table 1. The majority of patients (78.3%) received RFB 300mg daily and all patients received TAF 25mg daily. Thirteen patients (56.5%) were being treated for mycobacterium tuberculosis, 5 (21.7%) for mycobacterium avium complex, and 5 for other mycobacterium infections. The median duration of TAF and RFB overlap was 33 weeks (IQR 12-44 weeks). Of the 6 patients with baseline viral load ≤200 copies/mL, 4 maintained viral load ≤200 copies/mL through the duration of TAF and RFB overlap. One patient stopped ART during treatment and the other was also receiving chemotherapy for cancer. Of the 17 patients with viral load >200 copies/mL at baseline, 14 (82.4%) achieved viral load ≤200 copies/mL. Of the patients with end of treatment data, 77.8% had a viral load ≤200 copies/mL at the end of TAF and RFB overlap. Mean change in CD4 count from baseline to end of treatment was +105 cells/uL (95% CI -16 -227). Conclusion: Despite a predicted decrease in TAF absorption when coadministered with RFB, the majority of individuals achieved or maintained HIV suppression during TAF/RFB therapy. This data supports further study of TAF and RFB coadministration in HIV and mycobacterial infection.

734 CLINICAL PHARMACOKINETICS AND TOXICODYNAMICS OF LINEZOLID IN THE NIX-TB TRIAL Marjorie Imperial 1 , Jerry Nedelman 2 , Radojka M. Savic 1 1 University of California San Francisco, San Francisco, CA, USA, 2 Global Alliance for TB Drug Development, New York, NY, USA Background: FDA recently approved a high dose linezolid (LZD) containing regimen in combination with pretomanid and bedaquiline (BPaL) for treatment of extremely drug resistant-tuberculosis (XDR-TB). WHO also prioritized LZD for the treatment of DR-TB. Use of LZD is associated with significant toxicities, but limited data is available on optimal dosing and best clinical practices for LZD. We performed population pharmacokinetic (PK)-toxicodynamic modeling and simulation to quantify PK/toxicity relationships of LZD as part of a 6-month BPaL regimen from the NiX-TB study. Methods: Data was available for 88 patients; all initially administered 1200 mg LZD daily (BID or QD schedules). Dose adjustments of LZD were allowed per discretion of the investigator to manage LZD toxicity. LZD PK profiles that accounted for individual dosing histories were predicted from the PK model and linked to safety profiles. Delayed PK/toxicity response models described suppression of platelets (PLT) and hemoglobin (Hb). A proportional odds model described graded peripheral neuropathy (PN) rates over time. Final models were used to simulate and compare PK and safety outcomes following daily doses of 1200 mg LZD as well as alternative dosing regimens. Results: LZD PK was described by a two-compartment model with nonlinear clearance. At 1200 mg QD and 600 mg BID nonlinearity was mild with steady- state average concentrations of 11 and 9.8mg/L. Hb and PLT suppression and PN were each significantly (p<0.001) related to LZD PK. Hb production was largely suppressed for LZD concentrations above 7.7mg/L, consistent with anemia reported for ~40% of subjects. Simulations indicated that the median time to onset of severe anemia was 9 weeks and Hb values at baseline and week4 might predict severe anemia (ROC AUC=0.91) better than LZD PK troughs (0.56). A greater than 10% decrease in Hb levels at week 4 had maximum sensitivity and specificity to predict severe anemia and a dose reduction to 600 mg QD for these patients prevented 65% of severe anemia cases. For PN, simulations showed reversal by 3 months for most patients following dose reductions or termination, with the more aggressive adjustments better minimizing

CROI 2020 270