EXTH-64. COMPARISON OF PANOBINOSTAT CSF PENETRATION WITH CNS PENETRATION FOLLOWING SYSTEMIC ADMINISTRATION IN A PRE-CLINICAL NON-HUMAN PRIMATE MODEL

Targeted therapies developed for diffuse midline gliomas (DMG) expressing H3K27M have focused on histone deacetylase inhibitors (HDACi). High-throughput drug screening with patient derived DMG cell lines identified the HDACi panobinostat as a prominent clinical agent as well as pre-clinical studies with orthotopic mouse tumors models proving efficacious. Diametrically there is a pronounced lack of measurable panobinostat CSF concentrations in a non-human primate (NHP) non-tumor bearing pre-clinical model and in pediatric brain tumor patients. Notwithstanding, adult and pediatric glioma clinical trials and clinical observation with panobinostat alone or in combination have demonstrated minor responses. Pharmacokinetic models utilize the premise that CSF drug penetration is a surrogate of CNS drug penetration. However, the direct correlation between CSF and CNS drug levels is undefined especially in lieu of geographic CNS extracellular fluid drug variability previously demonstrated in the same NHP pre-clinical model. Utilizing the same NHP model, this study sought to compare panobinostat CSF penetration to CNS penetration via analysis of homogenized normal cerebrum, cerebellum, and brainstem tissue utilizing LC-MS/MS. METHODS: Panobinostat was administered orally as a single dose to three non-human primates. Pre panobinostat plasma and CSF were collected. Following panobinostat administration (1-hr Tmax) CSF, cerebrum, cerebellum, and brain stem tissue were collected as well as plasma to confirm the presence of panobinostat. Tissue slices were individually homogenized and panobinostat extracted via protein precipitation. Plasma, CSF, and tissue panobinostat concentrations were quantified using a LC-MS/MS assay. The lower limit of quantitation (LLOQ) for plasma-0.1 ng/ml, CSF-0.5 ng/ml, and tissue-10.0 pg/mg. RESULTS: Panobinostat was quantifiable in plasma (n=2) at the 1 hour (20.033 ng/mL and 0.153 ng/mL). CSF and CNS tissue samples were below the LLOQ for panobinostat in all samples. CONCLUSIONS: Panobinostat was not measurable from CSF and homogenized brain tissue in a non-tumor bearing NHP model at 1-hour post-administration using LC-MS/MS.

Population Pharmacokinetic Properties of Antituberculosis Drugs in Vietnamese Children with Tuberculous Meningitis

ABSTRACTOptimal dosing of children with tuberculous meningitis (TBM) remains uncertain and is currently based on the treatment of pulmonary tuberculosis in adults. This study aimed to investigate the population pharmacokinetics of isoniazid, rifampin, pyrazinamide, and ethambutol in Vietnamese children with TBM, to propose optimal dosing in these patients, and to determine the relationship between drug exposure and treatment outcome. A total of 100 Vietnamese children with TBM were treated with an 8-month antituberculosis regimen. Nonlinear mixed-effects modeling was used to evaluate the pharmacokinetic properties of the four drugs and to simulate different dosing strategies. The pharmacokinetic properties of rifampin and pyrazinamide in plasma were described successfully by one-compartment disposition models, while those of isoniazid and ethambutol in plasma were described by two-compartment disposition models. All drug models included allometric scaling of body weight and enzyme maturation during the first years of life. Cerebrospinal fluid (CSF) penetration of rifampin was relatively poor and increased with increasing protein levels in CSF, a marker of CSF inflammation. Isoniazid and pyrazinamide showed good CSF penetration. Currently recommended doses of isoniazid and pyrazinamide, but not ethambutol and rifampin, were sufficient to achieve target exposures. The ethambutol dose cannot be increased because of ocular toxicity. Simulation results suggested that rifampin dosing at 50 mg/kg of body weight/day would be required to achieve the target exposure. Moreover, low rifampin plasma exposure was associated with an increased risk of neurological disability. Therefore, higher doses of rifampin could be considered, but further studies are needed to establish the safety and efficacy of increased dosing.

EXTH-25. THE CSF PENETRATION OF THE PROCASPASE-ACTIVATING COMPOUND (PAC-1) IN COMBINATION WITH TEMOZOLOMIDE (TMZ) IN A NONHUMAN PRIMATE CSF ACCESS MODEL

BACKGROUND The synergistic activity of temozolomide (TMZ) administered in combination with procaspase-activating compound (PAC-1) has been reported in pre-clinical mouse models and canine patients, leading to clinical trials in adults with glioblastoma. To optimize pediatric clinical trial design, a translational pharmacokinetic CSF penetration study was conducted using a pre-clinical nonhuman primate non-tumor bearing CSF access model with TMZ and PAC-1 administered alone and in combination. METHODS Four male rhesus macaques with CSF lateral ventricular reservoirs received PAC-1, 15 mg/kg orally [Human Equivalent Dose (HED) 558 mg/m2/day] or TMZ, 1-hr IV infusion, 7.5 mg/kg (HED 150 mg/m2) as single and combination agent administration. Paired plasma and CSF samples were collected for 0–96 hours. PAC-1 and TMZ were quantified by LC-MS/MS. Pharmacokinetic parameters were calculated using noncompartmental methods. Statistics were determined via Mann-Whitney test. RESULTS (Mean ± Standard Deviation): For TMZ: Plasma AUC0-24 (hr*ng/ml) single agent (n=4): 28590 ± 4888 and in combination (n=4): 32736 ± 10147. CSF AUC0-24 (hr*ng/ml) single agent (n=4):14406 ± 1279 and in combination (n=4):15614 ± 1767. CSF penetration (% AUCCSF: AUCPLASMA) single agent: 50.9% ± 2.2 and in combination:49.6% ± 8.4. For PAC-1: Plasma PAC-1 AUC0-24 (hr*ng/ml) single agent (n=4): 2556 ± 2157 and in combination (n=4): 1947 ± 1311. CSF PAC-1 AUC0-24 (hr*ng/ml) single agent (n=2): 8.7 ± 1.1 and in combination (n=3):12.9 ± 10.2. CSF penetration single agent: 0.2% ± 0.03 and in combination: 0.4% ± 0.38. CSF PAC-1 tlag and Tmax pharmacokinetic parameters decreased with concurrent TMZ administration. CONCLUSIONS In this non-tumor bearing pre-clinical nonhuman primate model the CSF penetration of PAC-1 was low and not notably affected by the concurrent administration of TMZ. TMZ CSF penetration for single administration was within previously reported ranges and also appeared unaffected when administered in combination with PAC-1.

Ceftaroline Cerebrospinal Fluid Penetration in the Treatment of a Ventriculopleural Shunt Infection: A Case Report

Pharmacokinetic data regarding ceftaroline fosamil (CPT) penetration into cerebrospinal fluid (CSF) are limited to a rabbit model (15% inflamed) and adult case reports. We describe serum and CSF CPT concentrations in a 21-year-old, 34.8 kg female, medically complex patient presented with a 4-day history of fevers (Tmax 39.2°C), tachypnea, tachycardia, fatigue, and a 1-week history of pus and blood draining from the ventriculopleural (VPL) shunt. A head CT and an ultrasound of the neck revealed septated complex fluid collection surrounding the shunt. Therapy was initiated with vancomycin and ceftriaxone. Blood and CSF cultures from hospital day (HD) 1 were positive for methicillin-resistant Staphylococcus aureus with a CPT MIC of 0.5 mg/L and a vancomycin MIC range of 0.5 to 1 mg/L. On HD 3, CPT was added. On HD 7, simultaneous serum (69.4, 44, and 30.2 mg/L) and CSF (1.7, 2.3, and 2.3 mg/L) concentrations were obtained at 0.25, 1.5, and 4.75 hours from the end of an infusion. Based on these concentrations, CPT CSF penetration ratio ranged from 2.4% to 7.6%. After addition of CPT, the blood and CSF cultures remained negative on a regimen of vancomycin plus CPT. On HD 14, a new left-sided VPL shunt was placed. The patient continued on CPT for a period of 7 days after the new VPL shunt placement. This case demonstrated CPT CSF penetration in a range of 2.4% to 7.6%, approximately half of the rabbit model. This allowed for CSF concentrations at least 50% free time > 4 to 6× MIC of the dosing interval with a dosing regimen of 600 mg IV every 8 hours in a 34.8 kg chronic patient and resulted in a successful clinical outcome with no identified adverse outcomes.

EXTH-65. PLASMA AND CEREBROSPINAL FLUID PHARMACOKINETICS COMPARISON OF BRAF AND MEK INHIBITORS FOLLOWING SINGLE AND EXTENDED ADMINISTRATION IN A PRE-CLINICAL NON-HUMAN PRIMATE MODEL

BACKGROUND Aberrations in the MAPK signaling pathway are present in adult and pediatric CNS tumors. The BRAF and MEK inhibitors Dabrafenib, Vemurafenib, Selumetinib and Trametinib, alone or in combination, have demonstrated clinical efficacy. However, minimal CSF penetration has been demonstrated in pre-clinical models with single dose administration. CSF penetration of these agents was compared following single and extended oral administration in a nonhuman CSF Lateral Reservoir model to determine if CSF penetration increases at steady state. METHODS Agents were administered orally, to each subject (n=9), in single or multiple doses, as serial single agent studies, with washout periods between studies. The following dosages (Human equivalent dose HED) were utilized in the studies: Single Dosage-Selumetinib 2.5 mg/kg (50.0 mg/m2), Dabrafenib 8.0 mg/kg (160 mg/m2), Vemurafenib 26 mg/kg (520 mg/m2), Trametinib 0.06 and 0.12 mg/kg (1.2 and 2.4 mg/m2). Extended dosages-Selumetinib 2.5 mg/kg BID x 3 days (50.0 mg/m2), Dabrafenib 4.0 mg/kg BID x 2 days (160 mg/m2/day), Vemurafenib 26 mg/kg BID x 3 days (520 mg/m2), Trametinib 0.05 mg/kg x 10 days BID (1.0mg/m2). Paired serial plasma and CSF samples were collected. Agents were quantified by uHPLC-MS/MS. Pharmacokinetic parameters were calculated via noncompartmental methods. Plasma exposure reported as Area Under the Curve (AUC) and CSF penetration as ratio of CSF:plasma AUC. Statistical significance determined by t-test. RESULTS Single administration AUCinf/D (hr*ng/ml/mg) vs. extended administration AUCtau/D (hr*ng/ml/mg) for each agent: Vemurafenib-Plasma 30.9–351.4 vs.29.2–379.7, CSF-non-quantifiable; Dabrafenib-Plasma 29.3–102.6 vs.3.4–58.2, CSF-0.12–0.54 vs 0.04–0.19, % CSF:Plasma-0.3–0.84 vs.0.32–1.3; Trametinib-Plasma AUC indeterminable, CSF-non-quantifiable; Selumetinib-Plasma 69.8–133.0 vs.26.5–81.2; CSF-AUC indeterminable vs. 0.37–0.86, % CSF:Plasma-Indeterminable vs. 0.78–1.4. T-test: A downward trend in Dabrafenib and Selumetinib plasma AUC and a marked upward trend in Selumetinib CSF AUC. CONCLUSIONS With the exception of Selumetinib, the CSF exposure of BRAF and MEK inhibitors studied did not markedly increase following extended dosing.

Plasma and cerebrospinal fluid pharmacokinetics of doxil after intravenous administration in adults with primary CNS lymphoma.

e14067 Background: Doxil, a pegylated liposomal doxorubicin (PLD) has excellent CNS penetration in rodent brain tumor models and is active in patients with lymphoma. The CNS penetration of doxil in humans is unknown. We evaluated the plasma and cerebrospinal fluid (CSF) pharmacokinetics (PK) of doxil in adults primary CNS lymphoma (PCNSL) as a surrogate for CNS penetration. Methods: Adults with PCNSL enrolled on a NCI phase I study of ibrutinib and multi-agent immune-chemotherapy (NCT02203526) received doxil 50 mg/m2 IV over 1 hour on day 2 of 21-day cycles of chemotherapy. In 4 patients with indwelling Ommaya reservoirs serial blood and CSF samples were obtained prior to infusion and until a median of 345 hours, (range 72-477 hours) after doxil administration. Total doxorubicin concentration (liposome bound + protein bound + free) was quantified with a validated liquid chromatography/tandem mass spectrometry assay (lower limit of quantification plasma=0.29 ng/mL, and CSF=0.06 ng/mL). PK parameters were estimated using non-compartmental methods. CSF penetration was calculated from the AUCCSF:AUCplasma. Results:Total doxorubicin plasma concentration time curves were characterized by sustained drug exposure and a median terminal half-life of 64.5 hours. (range, 61.9-65.1 hours). Doxorubicin was measurable in CSF in all patients, but CSF penetration was low. The terminal half-life of doxorubicin in CSF could not be calculated due to persistently measurable concentrations throughout the sampling times. Conclusions: In patients with PCNSL doxorubicin was measurable in CSF for prolonged time periods after doxil administration, but CSF penetration is low. Clinical trial information: NCT02203526. [Table: see text]

Plasma and cerebrospinal fluid pharmacokinetics of selumetinib in non-human primates (NHP).

e14070 Background: Selumetinib (AZD6244 hyd sulfate) is an orally bioavailable agent targeting mitogen-activated protein kinase (MEK) 1/2. Selumetinib has shown activity in children with NF1 plexiform neurofibromas and refractory low-grade gliomas. The central nervous system (CNS) penetration of selumetinib in humans is unknown. Methods: We analyzed selumetinib in the cerebrospinal fluid (CSF) of NHP as a surrogate for CNS penetration. Four adult male rhesus macaques were each given a single dose of selumetinib capsules 2.5 mg/kg (human equivalent dose 50 mg/m2) by mouth on an empty stomach. Selumetinib was quantified in serial paired plasma and CSF samples from 0 to 48 hours after administration by Covance Bioanalytical Services (lower limit of quantification 0.5 ng/mL). Free selumetinib was estimated at 2.3% of total based on reported plasma protein binding of 97.7% in NHP. Pharmacokinetic parameters including area under the concentration x time curve (AUC) of selumetinib were calculated using non-compartmental methods and CSF penetration was calculated from the ratio of AUC (CSF): AUC (plasma). Results: Selumetinib was detected in plasma in all four NHPs, but in the CSF of only one NHP (ZH60). The mean (± SD) AUC0-∞ and half-life (t1/2) in plasma were 3350 ± 489 ng•h/mL and 10.8 ± 2.5 hours, respectively. In ZH60, the CSF AUC0-48 hr was 10.6 ng•h/mL. The CSF penetration of total and free selumetinib in this animal was 0.4%, and 14%, respectively. The mean plasma AUC and t1/2 were comparable to that of children (2842 ng•h/mL and 6.9 hours, respectively) at a similar dose in a phase I trial. However, the average peak concentration in children occurred at approximately 1 hour with a Cmax of 841 ng/mL, while the Cmax in NHP plasma ranged from 1 – 4 hours and averaged 198.8 ng/mL. Conclusions: The CSF penetration of oral selumetinib is limited; in one NHP it was 0.4% and 14% when corrected for plasma protein binding of selumetinib. The fact that plasma AUC0-∞ and half-lives in NHP were similar to previously reported human plasma PK indicates that this model may accurately reflect human CSF penetration as well.