Drug Delivery to the CNS: Barriers that May Influence Efficacy in Treating Tuberculosis in the Brain

William F. ElmquistPharmaceutics

SystemicPharmacokinetics

TB-Meningitis

astrocyte endfeet

endothelial cell layerendothelial cell layer

basementmembranecapillary lumen

tightjunctions luminal

abluminal

Blood Brain Barrier

1

TB-MeningitisRockville MD

22-23 May 2017

efflux influxdiffusion

Why Does a Drug Work ??

Connect - Disconnect of the PK-PD Relationship

Pharmacodynamics(events at the target)

Pharmacokinetics(conc-time in blood)

Information

Balance

Why Doesn’t a Drug Work ??

Translational research in CNS Drug Delivery- Must keep in mind the big questions !

Information

Why does this one work, and that one doesn’t ??

Flow

Why

CNS Drug Delivery in the Era of Systems Biology

Dose and dosing regimen Mechanisms of delivery and action

“Traditional PK/PD”“Molecular pharmacokinetics”

Black box modelsSystem – structure function

Nature Biotechnology 23, 191 - 194 (2005)

Systems model“the Big Picture”

Overview

plasma

choroid plexusarachnoid membrane

BBBECF

ICF

ependymapia mater

neuronaland

glial cellmem-branes

cerebrospinal fluid

Lung

HeartKidney

Brain

Liver

Examine drug pharmacokinetics / delivery to CNS sites across several scales

target

Drug

X

Veno

us B

lood

Arterial Blood

Quantitative – Qualitative Oscillations in the Thought Cycle

macro

micro

meso

Genetic Factors- drug targets- drug transporters- drug metabolizing enzymes

Environmental Factors- induction- inhibition

Physiological Factors- age, disease, etc.

Understanding Sources of Variability in Drug ResponseVariability Cycle

Connect - Disconnect of the PK-PD Relationship

“Locations” of Variability in Drug Efficacy In CNS Tuberculosis

Presystemic bioavailability questions (“traditional” bioavailability)

Site-specific bioavailability questions (drug targeting)

Oraldosageform

Systemiccirculation Drug

action

Targeted Bioavailability

Intestinalabsorption

Liver metabolism

Cellular delivery

Intestinalmetabolism

Tissue distribution Target site

BBB

B-TB

TB-Target

6

“Locations” of Variability in Drug Response

Presystemic bioavailability questions (“traditional” bioavailability)

Site-specific bioavailability questions (drug targeting)

Oraldosageform

Systemiccirculation Drug

action

Targeted Bioavailability

Intestinalabsorption

Liver metabolism

Cellular delivery

Intestinalmetabolism

Tissue distribution Target site

Systemicclearance

Mechanisms that influence the fraction of the drug in the systemic circulation that is available for distribution to target tissue and the exposure of the tissue to the drug

- distribution of blood flow- ratio of total clearance to a distributional clearanceDistributional clearance - membrane permeability, competing carrier-mediated transport (influx or efflux), protein-binding, intracellular metabolism, tissue transit time, capillary structureTotal clearance – will affect the availability of the drug in the blood to distribute to the tissue

Physicochemical PropertiesPh

ysio

logy

/ Pa

thol

ogy Drug

MetabolismReceptorAffinity

ProteinBinding

MembranePermeability

Drug Transport

GeneRegulation

Targeted

Bioavailability

Dosage R

egimen

ProteinExpression

Pharmacological / Toxicological Response8

Importance of Transporters in the CNS Disposition of Drugs

From Lee and Gottesmann. Journal of Clinical Investigation, 1998illustration by Naba Bora, Medical College of Georgia.

One “Location” : Blood-brain Barrier

9

Modifed from:

Loscher, Aug. 2005Nature Rev. Neurosci.

GLUT1

P-gp(p-glycoprotein)

Co-localizationGLUT1 - P-gp

brain

brainbrain

blood

Clout

Clin

BCRPP-gp

10

Tight Junction

Compartmental model for solute exchange in the brain

Therapeutic decisions limited by available data at specific sites

plasma

choroid plexusarachnoid membrane

braincapillaries

extra-cellular

fluidintracellular

fluid

ependymapia mater

neuronaland

glial cellmem-branes

cerebrospinal fluid

PO , IV

ICV,IT

bolus IC, CEDinputoutputexchange

To know,is to measure.

Modeling limited by available mechanistic data at specific sites

plasma

choroid plexusarachnoid

braincapillaries

extra-cellular

fluidintracellular

fluid

ependymapia mater

neuronaland

glial cell

cerebrospinal fluid

CSF flowSurface areaPermeabilityTransportersMetabolismRegional variability

ECF productionSurface areaPermeabilityTransportersMetabolismRegional variability

ConvectionDiffusionTransportersPermeabilityMetabolismReceptorsRegional variability

Convection, DiffusionPermeability, Regional variability

Simplified Quantitative Analysis of Drug Transfer In CNS

Drug “Binding” – Plasma and Brain

Plasma

Cu,plasma Cplasma

Brain CSF

BBB BCSFB

PSCLeff CLin

CcsfCu,brainCbrain CLbulk

CLmeta

[CLin, CLeff, PS]

Simplified Quantitative Analysis of Drug Transfer In CNS

Kinetics of distribution - Rate and Extent

Extent (exposure) - described by area under the curve (AUC)

Ratio of areas gives tissue partition coefficient

plasma

brain

AUCAUCKp =

unboundplasma

unboundbrain

AUCAUCuuKp

−

−=,

Total concs Unbound concs

Rate (onset) - described by maximum concentration (Cmax)

Simplified Quantitative Analysis of Drug Transfer In CNS

Extent - partitioning of free concentration

out

infreep CL

CLK =,

bulkmetabolismefflux

uptakefreep CLCLCLPS

CLPSK

++++

=,

Sum of clearances in each direction

Simplified Quantitative Analysis of Drug Transfer In CNS

Extent - partitioning into a specific brain region

bulkmetabolismefflux

uptakefreep CLCLCLPS

CLPSK

++++

=,

Tight-junctionopening

Tight-junctionopening

Substrate forEfflux Transporter

Substrate forInflux Transporter

Active transport CL dependson both capacity and affinity CKm

TCLact += max

BrainMetabolism

Fluid flow

Representative Case Study:The Treatment of Glioblastoma with

Inhibition of P53 Degradation – MDM2 Inhibitor

Minjee Kim, Jann Sarkaria

Choice of PDX Glioma Model

Neu

rosp

here

Cou

nt (%

)

0

25

50

75

100

GBM10 GBM12 GBM102 GBM108 GBM143

0 nM100 nM

0

10

20

30

40

50

GBM10 GBM102 GBM108 GBM143

0 nM100 nM

Apop

tosi

s (%

)

Efficacy of SAR405838 depends on tumor location

Heterotopic Xenograft Orthotopic Xenograft

0

500

1,000

1,500

2,000

2,500

30 35 40 45 50 55 60

PlaceboSAR405838

p = 0.005

n = 7

n = 8

Volu

me

(mm

3 )

GBM108

0.0

0.3

0.5

0.8

1.0

0 20 40 60 80Study Day

PlaceboSAR405838

n = 11

n = 12Su

rviv

al

GBM108

P=0.59

Study Day

50mg/kg qd 50mg/kg qd

SAR405838 Concentration vs. TimeProfiles in Plasma and Brain Plasma Brain

Influence of Efflux Transporters at the BBBOn Brain Penetration of SAR405838

WT, Bcrp-/-

Pgp-/-

TKO-/-

SAR405838 - Plasma and Brain Distribution Kinetics

Wild-type Mdr1a/1b-/- Bcrp1-/- Mdr1a/1b-/- Bcrp1-/-

T1/2 Hr 3.26 4.18 3.02 5.08Tmax Hr 8 8 4 2Cmax ng/ml 4651 3582 3976 6164

AUCinf_pred hr*ng/ml 61195 41382 47804 65867Vz/F ml/kg 1922 3642 2281 2779CL/F ml/hr/kg 409 604 523 380

AUCbrain hr*ng/ml 1335 63234 1956 65442

0.022 1.53 0.041 0.994

Distribution Advantage 1.00 70.0 1.88 45.5

22

Blood brain barrier integrity in orthotopic tumors. Near-moribund mice with orthotopic GBM108 tumors were injected with TexasRed-3 kDa dextran conjugate 10 min before euthanasia and processed for cresyl violet and fluorescent microscopy on serial histology sections. Accumulation of TR-dextran within the tumor reflects disruption of the BBB. Results presented are representative of five mice analyzed. Scale bar = 500 µm.

TexasRed Cresyl Violet

orthotopic GBM108 parental line

23

Heterogeneous Breakdown of Tumor BBB

orthotopic GBM108 parental line

G108-VEGFA Cell Line Generation

Texas-Red Cresyl violet

GB

M10

8-ve

ctor

GB

M10

8-VE

GFA

Texas Red Cresyl violetVe

ctor

VEG

FAA T1 + Gad T2/FLAIRB

Vect

orVE

GFA

Vect

orVE

GFA

C H&E Gadavist SAR405838

Spatial Distribution of SAR405838 (MALDI-MSI)

GBM108-Empty vector GBM108-VEGFA

Orthotopic Survival

Days0 20 40 60 80

Surv

ival

0.0

0.2

0.4

0.6

0.8

1.0

Placebo SAR405838

Vector

Days0 20 40 60 80

Su

rviv

al0.0

0.2

0.4

0.6

0.8

1.0

Placebo SAR405838

VEGFA

45 day

Conclusions for SAR405838 Study

• SAR405838, a potent MDM2 inhibitor, is subject to BBB efflux

• This preclinical study indicates enhanced delivery of SAR405838 will improve its efficacy

• Strategies to overcome limited delivery of drug across BBB will result in better treatment for brain tumors

Translation in the Clinic -Delivery and the BBB

Deb Brinkmann, Jann Sarkaria(Mayo Clinic)

30

Use of Uptake Transporters in BBB

Discordance in tumor delineation by 18F-FDOPA PET and MRI.

Volumes defined for :

A) FDOPA positivity ( ) by PET B) T1 contrast enhancement ( ) on T1 contrast enhanced imagesC) FLAIR positive (blue)

outlined for a single patient

Deb Brinkmann, Jann Sarkaria - Mayo Clinic , Rochester

31

Structure Structure Volume (cc) from Eclipse

PET Volume outside of T1-GAD (cc) from Eclipse

PET Volume outside of FLAIR (cc) from Eclipse

MR Volume outside of PET (cc) fromEclipse

T1-GAD 10.4 N/A N/A 4.9

FLAIR 32.0 N/A N/A 21.7

PET 19.5 13.8 9.1 N/A

RT_FDOPA02Grade IVTotal Multi-focalFLAIR contour in blueT1-GAD contour in redPET contour in yellow Orthogonal views with

crosshairs turned on, for reference

Regions of tumor with intact BBBprotected from treatment

by efflux transporters and TJ

Orthogonal views with crosshairs turned on,

for reference

RT_FDOPA05Grade IV Total Single

FLAIR contour in blueT1-Gad contour in redPET contour in yellow

Structure Structure Volume (cc) from Eclipse

PET Volume outside of T1-GAD (cc) from Eclipse

PET Volume outside of FLAIR (cc) from Eclipse

MR Volume outside of PET (cc) from Eclipse

T1-GAD 47.7 N/A N/A 33.2**

FLAIR 54.3 N/A N/A 36.4

PET 21.7 7.1 3.8 N/A

**T1-GAD contour includes post-op cavity (not just enhancement)

Regions of tumor with intact BBBprotected from treatment

by efflux transporters and TJ

A. B.Jann Sarkaria – Mayo Clinic Rochester

Screen capture of biopsy planning using the registered 18F-DOPA PET and T1-CE MRI in the Stealth™ Neuronavigation System for blue needle locations at:A) a T1 contrast enhancing, PET positive (M+P+) location B) a non-contrast enhancing but PET positive (M-P+) location.

34

CT scan T1 – CE - MRI T2 - MRIRegion specific disease, requires region specific consideration of drug delivery

CT scan T1 – CE - MRI T2 - MRI

Questions:

1) In the tuberculomas, is there a change in drug penetration as they encapsulate?

2) Is there active disease in the peripheral regions that show edema?

3) What is the integrity of the BBB at different sites within an infected brain?

4) Are drug concentrations in each region adequate to treat disease?

5) Are there significant differences in delivery limitations amongst drugsused in the necessary combinations in different regions of disease?

PlasmaCu,plasma Cplasma

Brain CSF

BBB BCSFBPS CLeff CLin

CcsfCu,brainCbrain CLbulk

CLmeta

[CLin, CLeff, PS]

Spatial and Temporal Changes in Delivery of Drugs as Disease Progresses

serum

normal brain

tumor

peritumorextracellularfluid

10.9 µg/ml

0.29 µg/ml

0.32 µg/ml 0.6 µg/ml

1.4 µg/ml

Rifampicin Spatial Differences in Distribution within the Human Brain

microdialysis

MALDI-MSI for Rifampicin Distribution in the Lung

Liver

Lung

Blood

BrainAUCinfected = 0.0012 ng*hr/ml

AUCcontrol = 0.0008 ng*hr/ml

AUCinfected = 0.0034 ng*hr/ml

AUCcontrol = 0.0054 ng*hr/ml

AUCinfected = 0.061 ng*hr/ml

AUCcontrol = 0.067 ng*hr/ml

AUCinfected = 0.0056 ng*hr/ml

AUCcontrol = 0.0061 ng*hr/ml

PET scan for distribution of 11C-Rifampicinin M. tuberculosis Infected Mouse Model

DeMarco et al. 2015

RIF

11C PET

PlasmaCu,plasma Cplasma

Brain CSF

BBB BCSFBPS

CLeff CLin

CcsfCu,brainCbrain CLbulk

CLmeta

[CLin, CLeff, PS]

Parp1 inhibitor - Talazoparib

Talazoparib – Pgp substrate - poor penetration into the brain

toxicity

delivery

potency

toxicity

potencydelivery

The sweet spot

Make things as simple as possible,but not simpler.

Albert Einstein

Delivery across the BBB: Beware of the simplest explanation.

Systems Pharmacology

Mechanistic “Molecular” Pharmacokinetics

Empiric “Black-Box” Pharmacokinetics

“IntelligentDrug Design”

“Sweet Spot” for CNS Rx