Andrew M. Stein
ACoP - November 10, 2021
Executive Summary
2
CAR-T therapy overview
3
In vivo
Ex vivo
Lentiviral vector
Patient’s T cell
Target
Native TCR
Tumor cell
Cancer cells
Dead tumor cell
Anti-target
CAR construct
Cytokine release
CAR-T cell expansion
Antigen-binding domain
Hinge�Transmembrane
Costimulatory domain
�Signaling domain
CART Cellular kinetic model
4
CAR-T Expander
(CE)
CAR-T Persister
(CP)
Expand until tmax
Application: Describes CART kinetics
ρ
~α
β
Stein AM, Grupp SA, Levine JE, et al. Tisagenlecleucel model based cellular kinetic analysis of chimeric antigen receptor–T Cells. CPT Pharmacometrics Syst Pharmacol. 2019;8(5):285-295.
Comed
CART Cellular Kinetic – Pharmacodynamic model
5
CAR-T Expander
(CE)
CAR-T Persister
(CP)
Tumor
(U)
ρ
Application
Question/Challenge
g
Example: Lotka-Volterra Predator-Prey Model1-2 is frequently used, but has cycles
6
Tumor
(U)
g
ρ
d
CART
(C)
Predator
(CART)
Time
Population
Prey
(tumor)
Cyclic behavior is not observed
in the clinic.
List of criteria a CellK-PD model should have
7
EC50 expansion + EC50 killing model had all desired properties
8
Criteria | Mayer et al. + Gompertz | Sahoo et al. | Kuznetsov et al. | Lotka-Volterra | Newly proposed model structure |
1. Exponential Expansion | yes | yes | yes | yes | yes |
2. Fast initial decline | yes | yes | yes | yes | yes |
3. Slow long term slow decline | no | no | no | no | yes |
4. Antigen drives expansion | yes | no | yes | yes | yes |
5. CART expansion not infinite | no | no | no | no | yes |
6. No cyclic behavior | possibly | possibly | possibly | no | yes |
9
Newly proposed CART CellK-PD model structure that meets all criteria
CAR-T Expander
(CE)
CAR-T Persister
(CP)
Tumor
(U)
ρ
g
kEP
kPE
d
EC50 formula is derived from a traditional binding model.
Mayer, Andreas, et al. "Regulation of T cell expansion by antigen presentation dynamics." Proceedings of the National Academy of Sciences 116.13 (2019): 5914-5919.
Confounding factors that make E-R analyses challenging1-3
10
CAR-T Expander
(CE)
CAR-T Persister
(CP)
Tumor
(U)
Challenge
Path Forward
Introduction of binding kinetics to account for antigen burden
11
CE
CP
CE
U
+
U
+
U
CP
Application
Challenge: potential umbrella shaped affinity-response curve [1]
Immunophenotype modeling framework
12
CART
Effector
CART
Effector
Memory
Tumor
CART
Naive
CART
Central
Memory
CART
Stem Cell
Memory
Application
Challenge
Importance of genotype heterogeneity
13
CE,1
CP,1
CE,2
CP,2
CE,n
CP,n
…
…
U
Challenge
Path Forward
Summary
14
Acknowledgements
15