Introduction to Computational Vaccinology and iVAX by EpiVax

Using Computa.onal Vaccinology to Design Genome-‐Derived Vaccines for Infec.ous Diseases,

Cancer, Allergy and Autoimmune Disease

1

Anne S. De Groot, Lenny Moise, Leslie Cousens, Frances Terry, William Mar<n Ins<tute for Immunology and Informa<cs, University of Rhode Island and EpiVax, Inc. www.epvax.com www.immunome.org

22 January 2014

1

Your Speaker – Annie De Groot MD

2

The Company: EpiVax

3 hOp://bit.ly/EpiPubs

EpiVax Collaborates with the Ins*tute for Immunology and Informa*cs @ URI

Collabora<ve Research on Immunome-‐Derived Accelerated Vaccine Design and Development Funded by the NIH CCHI U19, COBRE, and P01 awards. www.immunome.org


Addi.onal Collaborators

Bill Mar<n Lenny Moise Frances Terry Leslie Cousens Ryan Tassone Howie La<mer Mindy Cote Lauren Levitz Chris<ne Boyle

Alan Rothman Carey Medin Andres Gui<errez Danielle Aguirre Joe Desrosiers Thomas Mather Wendy Coy Loren Fast

Don Drake, Brian Schanen

Sharon Frey Mark Buller Jill Schreiwer

Hardy Kornfeld Jinhee Lee Liisa Selin

Connie Schmaljohn Lesley C. Dupuy

Ted Ross

Mark Poznansky Tim Brauns Pierre LeBlanc

AI058326, AI058376, AI078800, AI082642


•  Why Computational Immunology •  Tools to Produce IDVs

– Antigen selection – Vaccine design –  New concepts

•  Case Studies

6

Outline

Predic<ng the future is something that weather experts do with the assistance of informa<cs models. These forecasts enable us to make decisions on a daily basis, and they are accurate enough to mobilize millions if and when severe storms are predicted. Why then, are we so slow to use informa<cs in vaccine and protein therapeu<cs design?

In todays talk, I will discuss the use of immunoinforma<cs tools for vaccine design, mechanism of ac<on studies, and efficacy evalua<ons. I believe that the <me is ripe for vaccine developers to ac<vely apply, evaluate and improve vaccines through the use of computa<onal immunogenicity predic<on tools.

“Old Style” Vaccines

Grow . . . and use whole pathogen

Whole (live/killed) vaccines

Subunit vaccines (Flu, Hepa<<s B, HPV vaccines, for

example)

Genome-‐Derived, Epitope Driven (GD-‐ED)

Vaccines

BeOer understanding of vaccine MOA

Improve vaccine safety and efficacy

Accelerate Vaccine Design

The focus of our work Can we make vaccines beJer/faster


iVAX Vaccine Design Toolkit

•  For Example: – HIV – HCV – Malaria – Universal Influenza Vaccine – Vaccines against Cancer – Vaccines for immunotherapy of AI – Vaccines for diseases affecting food animals

Why? New Vaccines Needed

•  For Example: Pandemic influenza 2009 – Traditional flu vaccine production methods

require large lead time – 20 weeks to first vaccine dose –  “Pandemic” influenza had already peaked by

the time the first shots were being delivered. – Vaccine manufacturing failed the test. –  Is H7N9 the next pandemic? If so, we are

worried. . .

Why? Unacceptable Delays

Emergent H7N9 disease in China


Spread to Beijing on 4/13/13 . . . Spread to Hong Kong on 12/6/ 13

15

Markedly Increased ac.vity in late 2013 and early 2014!


Con.nuing Expansion of H7N9 First confirmed cases occurred in Shanghai (3/30/13) but case ac<vity rapidly increased in Zheijang and Jiangsu provinces shortly aier. Now, we have a problem!

Image credit to VDU and Dr. Ian M Mackay hOp://www.uq.edu.au/vdu/VDUInfluenza_H7N9.htm 17 hOp://bit.ly/EpiPubs

Ci.es that are one stop from H7N9

An es<mated 70% of the world popula<on resides within two hours’ travel <me of des<na<on airports (calculated using gridded popula<on-‐density maps and a data set of global travel <mes, map supplied by A. J. Tatem, Z. Huang and S. I. Hay (2013).

Quick numbers... •  Total confirmed human cases of

influenza A virus H7N9: > 200

•  Total deaths aOributed to infec<on with influenza A virus H7N9: > 50

•  Case Fatality Rate (CFR): 29% (current)

•  Average <me from illness onset to first confirma<on of H7N9 (days): <10

•  Median age of the H7N9-‐confirmed cases (including deaths; years): 63

•  Males: 71% of cases, 74% of deaths

•  Younger pa<ents are recovering . . .

hOp://pandemicinforma<onnews.blogspot.com hOp://www.uq.edu.au/vdu/VDUInfluenza_H7N9.htm

H7N9 Morbidity and Mortality

19

Virus Transmission Mechanism – source is s.ll at large •  Human to human transmission has not been proved (or disproved) many cases show uninfected family members

•  Poultry iden<fied as poten<al natural host and H7N9 samples were found in poultry market environment in Shanghai. However not many poultry vendors infected and many cases have no indica<on of poultry exposure Image credit to VDU and Dr. Ian M Mackay hOp://

www.uq.edu.au/vdu/VDUInfluenza_H7N9.htm 20 hOp://bit.ly/EpiPubs

Distribu.on of Cases

This picture shows the

geographically wide distribu<on of flu cases -‐ sugges<ng widespread

distribu<on of the virus rather than a point outbreak.


Why are immunoinforma.cs tools important in this sedng?

•  Immunoinforma<cs predicted low immunogenicity of ‘cri<cal an<gen’ H7 HA

•  hOp://bit.ly/H7N9_2013

(reminder) Flu Vaccine – HA protein

Ian Mackey hOp://www.uq.edu.au/vduVDUInfluenza_H7N9.htm 23

hOp://bit.ly/EpiPubs

What Can We Learn About H7N9?

HA (hemagglu<nin) is the ‘Cri<cal An<gen’ used for Flu vaccines, especially recombinant vaccines – – which are currently in produc*on.


H7N9 is a unique virus

•  Low conserva<on of HA, NA surface proteins is not surprising

•  Internal proteins are more conserved


gB-2 (EPX Score: -24.56)

- 80 -

- 70 -

- 60 -

- 50 -

- 40 -

- 30 -

- 20 -

- 10 -

- 00 -

- -10 -

- -20 -

- -30 -

- -40 -

- -50 -

- -60 -

- -70 -

- -80 -

Thrombopoietin

Human EPO

Tetanus Toxin

Influenza-HA

Albumin

IgG FC Region

EBV-BKRF3

Fibrinogen-Alpha

Follitropin-Beta

HA A/California/07/2009 (H1N1)

HA A/Victoria/361/2011 (H3N2)

HA A/Texas/50/2012 (H3N2)

HA A/Shanghai/1/2013 (H7N9) . . . . . . . .. . . . . . . . -‐8.11 HA A/mallard/Netherlands/09/2005 (H7N7) . . . . . . -‐8.63

Random Expecta.on

HA A/mallard/Netherlands/12/2000 (H7N3) .. . . . . .-‐9.91

HA A/chicken/Italy/13474/1999 (H7N1) . . . . . . . . . -‐6.23

H7 HA Immunogenic Poten.al

New H7N9 Flu is Predicted to be POORLY IMMUNOGENIC




•  Vaccine was developed but is low immunogenicity as predicted.

hOp://bit.ly/H7N9_NovaVax

Unadjuvanted Influenza Vaccine Effectiveness



•  Vaccine was developed but is low immunogenicity as predicted

•  Sero-‐conversion is delayed, diminished in pa<ents infected with H7N9.

hOp://bit.ly/H7N9_Serology

. . . Low and Slow . . .



•  Vaccine was developed but is low immunogenicity as predicted

•  Sero-‐conversion is delayed, diminished in pa<ents infected with H7N9.

•  New vaccine approaches are needed. •  . . . Now that you are convinced, let’s talk about computa<onal vaccine design



•  Case Studies

31

Outline

Computational Vaccinology: Genomes-to-Vaccines

•  Lots of Genomes now Published! •  On line tools for Pathogen Gene finding

(GLIMMER, ORPHEUS, GeneMark) •  Tools for selecting subsets of protein –

such as subcellular localization of hypothetical proteins (PSORTb, CELLO, Proteome Analyst)

Selection of vaccine antigens is key

Strain 1

Strain 3

Strain 2

core genome dispensable genes

strain-‐specific genes pangenome

Comparative Genomics Impacts Vaccine Immunogen Selection

. . . Need “informa*on” = T cell and B cell epitopes

. . . And the correct “milieu”

= delivery vehicle, adjuvants/TLR ligands

“Fine tune” the immune response?

. . And there is ample evidence that this approach to vaccine design produces protective immunity

Immunome-Derived Vaccines . . .

Payload

Adjuvant

Delivery Vehicle

Vaccine

HLA (Human MHC), are comprised of peptide specific pockets

EpiMatrix predicts how well a peptide sequence will bind to a specific pocket.

Binding is the prerequisite for immunogenicity

8 class II HLA supertypes which taken together incorporate 95% of human

populations (and pockets) worldwide.

Each 9-mer/10-mer is analyzed for binding potential to each of those 8

allele matrices. .

Payload: Predic.ng Epitopes that Drive Immune Response is our Exper.se

Mature APC

Protein MHC II Pocket

Southwood et al. J. Immunology 1998 Sturniolo et al. Nature Biotechnology, 1999

The EpiMatrix Score describes the binding affinity of the pep<de sequence to the HLA complex

Peptide Epitope


epitope

Vaccine an<gen

1 + 1 + 1 = Response

epitope epitope

Immune response to a vaccine an<gen can be predicted by measuring the number of T cell epitopes contained in the an<gen with immunoinforma<cs tools.

How do we measure Immunogenicity?


Non Immunogenic

proteins

Immunogenic proteins

“Immunogenicity Scale”


Easy easy to deliver as pep<des

42

DRB1*0101

DRB1*0301

DRB1*0401

DRB1*0701

DRB1*0801

DRB1*1101

DRB1*1301

DRB1*1501

ClustiMer: Screen for Epitope Clusters

43

Conservatrix: Overcome the Challenge of Variability

HIV HCV Influenza

44

Identifying the most conserved 9-mers allows for protection against more strains with fewer epitopes

Conservatrix Finds Conserved 9-mers

Conserved epitope

CTRPNNTRK

CTRPNNTRK CTRPNNTRK

CTRPNNTRK CTRPNNTRK

CTRPNNTRK

CTRPNNTRK

45

BlastiMer: Epitope Exclusion

Confidential

In all of our vaccines we eliminate cross-‐reac<ve epitopes

Self Foreign

Human

Pathogen

Human

Microbiome

Protec.ve epitopes

Poten.ally detrimental cross-‐reac.ve epitopes

Poten.ally detrimental cross-‐reac.ve epitopes

Epitope Cross-‐Reac<vity Impacts Vaccine Immunogen Selec<on


Each MHC ligand has two faces, The MHC-binding face (aggretope), and the TCR-interacting face (epitope)

JanusMatrix

TCR

MHC

The JanusMatrix algorithm searches for putative MHC ligands which are identical at the contact residues but may vary at the MHC-binding residues.

MHC/HLA

TCR

•  Identical T cell-facing residues •  Same HLA allele and minimally

different MHC-facing residues

Find predicted 9-mer ligands with:

http://bit.ly/JanusMatrix

48

HCV T Effector Epitopes

HCV_G1_1605

HCV_G1_DEXDC_1246

HCV_G1_NS5A_1988

HCV_G1_NS4B_1725

HCV_G1_2898

HCV_G1_2913

HCV_G1_ENV_359

HCV_G1_2941

HCV_G1_NS4B_1910

HCV_G1_ENV_255

HCV_G1_2440

HCV_G1_NS2_732

HCV_G1_NS2_748

HCV_G1_2840

HCV_G1_1941

HCV_G1_NS4B_1769

HCV_G1_NS2_909

HCV_G1_2485

HCV_G1_NS4b_1798 HCV_G1_NS4B_1790

HCV_G1_NS4B_1876

HCV_G1_2879

Treg-‐like-‐Epitope: HCV

HCV_G1_NS2_794


– Antigen selection – Vaccine design

•  Case Studies

51

Outline

Epi-Assembler

Immunogenic consensus

CTRPNNTRK CTRPNNTRK

CTRPNNTRK CTRPNNTRK

CTRPNNTRK

CTRPNNTRK

EpiAssembler Constructs Immunogenic Consensus Sequences

STRAIN 01 Q X S W P K V E Q F W A K H X W N X I S X I Q Y LSTRAIN 02 Q A S W P K V E X F W A K H M W N F I S G I Q Y LSTRAIN 03 Q X S W P K X E Q F W A K H M W N F I S G I Q Y XSTRAIN 04 Q A S W X K V E Q F W A K H M W N F X S X I Q Y LSTRAIN 05 Q X S W P K V E Q F W A K H M W N F I S G I Q Y LSTRAIN 06 Q A S W P K X E Q F W A X H M W N F I S G I Q Y XSTRAIN 07 Q X S W P K V E Q F W A K H M X N F I S G I Q Y LSTRAIN 08 Q A S W X K V E Q F W A K H M W N F I S G I Q Y LSTRAIN 09 Q X S W P K X E Q F W A K H M W N F X S X I X Y XSTRAIN 10 Q A S W P R V E Q F W A K H M W N F I X G I Q Y LSTRAIN 11 Q A S W P K V E Q F W A K H M W N F I S G I Q Y LSTRAIN 12 Q A S W X K V E Q F W A X H M W N F I S G I Q Y XSTRAIN 13 Q A S W P K V E Q F W A K H M W N F I S G I Q Y LSTRAIN 14 Q A S W X K X E Q F W A K H M W N F I S X I Q Y LSTRAIN 15 Q A S W P K V E X F W X K H M W N F I S G I Q Y LSTRAIN 16 Q X S W P K V E Q F W A K H M W N F I X G I Q Y LSTRAIN 17 X A S W X K V E Q F W A K H M W N F I S G I Q Y XSTRAIN 18 Q X S W P K X E Q F W A K H M W N X I S G I Q Y LSTRAIN 19 Q A S W X K V E Q F W A K H M W N F I S X I Q Y LSTRAIN 20 Q A S W P K V E Q F W A X H M W N F I S G I Q Y L

x

F W A K H M W N F

EpiAssembler: Core Epitope


x

F W A K H M W N FW P K V E Q F W A

Q A S W P K V E Q N F I S G I Q Y LM W N F I S G I Q

EpiAssembler: Flanking Epitopes


x

F W A K H M W N FW P K V E Q F W A

Q A S W P K V E Q N F I S G I Q Y LM W N F I S G I Q

Q A S W P K V E Q F W A K H M W N F I S G I Q Y L

EpiAssembler: Final Immunogenic Consensus Sequence

VaxCAD Identifies and Eliminates Junctional Epitopes

VaxCAD will identify junctional epitopes and rearrange chosen epitopes to reduce junctional epitope formation

57

-10

0

10

20

30

40

50

HP

4117

H

P41

79

HP

4007

H

P41

11

HP

4018

H

P40

70

HP

4034

H

P41

93

HP

4065

H

P41

81

HP

4157

H

P40

60

HP

4068

H

P41

64

HP

4160

H

P41

75

HP

4127

H

P41

20

HP

4126

H

P41

54

HP

4168

H

P41

19

HP

4100

H

P40

01

HP

4061

EpiM

atrix

Clu

ster

Sco

re

Peptides in Default order in construct HP_IIB

Epitope Cluster Score Junctional Cluster Score

-10

0

10

20

30

40

50

HP

4117

H

P40

61

HP

4181

H

P41

11

HP

4018

H

P40

70

HP

4060

H

P41

57

HP

4065

H

P40

01

HP

4193

H

P40

34

HP

4068

H

P41

68

HP

4160

H

P41

75

HP

4127

H

P41

26

HP

4007

H

P41

54

HP

4164

H

P41

19

HP

4100

H

P41

20

HP

4179

EpiM

atrix

Clu

ster

Sco

re

Peptides in Optimized order in construct HP_IIB

Epitope Cluster Score Junctional Cluster Score

VaxCAD Example

58

DNA Vector

DNA insert

Intended Protein Product: Many epitopes strung together in a “String-of-Beads”

Protein product (folded)

Multi-Epitope Gene Design

DNA – chain of epitopes, or pep<de in liposomes ICS-‐op<mized proteins in VLP ICS-‐op<mized whole proteins

Immunogenic Consensus Sequence Formulations

HLA A2

HLA DR3

HLA B7

HLA DR2

HLA A2/DR1

HLA DR4

In Vivo Model for Validation: HLA Transgenic Mice

•  Why Computational Immunology •  Tools to Produce IDVs •  Case Studies

– Tularemia – Smallpox – H. pylori – VEEV (multi-pathogen vaccine) –  Influenza

61

Outline

Burk/Tuly/MP

Current Vaccine Design Pipeline

Epitope Discovery

Epitope Validation

Construct Design

Immuno-genicity

HIV/TB Epitope Discovery

Epitope Validation

Construct Design

Immuno-genicity

Tularemia Epitope Discovery

Epitope Validation

Construct Design

Immuno-genicity

Animal Model Validation

Smallpox Epitope Discovery

Epitope Validation

Construct Design


VEEV

Epitope Discovery

Epitope Validation

Construct Design

Animal Model Validation H. pylori

Epitope Discovery

Epitope Validation

Construct Design




Immuno-genicity

Immuno-genicity

Immuno-genicity

62

Epitope Discovery

Epitope Validation

Construct Design


Immuno-genicity Influenza

GDV Approach Applied to F. tularensis

63

McMurry JA, Gregory SH, Moise L, Rivera DS, Buus S, and De Groot AS. Diversity of Francisella tularensis Schu4 an<gens recognized by T lymphocytes aier natural infec<ons in humans: Iden<fica<on of candidate epitopes for inclusion in a ra<onally designed tularemia vaccine. Vaccine 2007 Apr 20;25(16):3179-‐91.

In 24 months:

•  Took one genome

•  Mapped class I + Class II

•  Selected 165 epitopes

•  Confirmed in human

•  Cloned into vaccine

•  Performed Challenge studies. . .

High Responder Frequency to Class II Epitopes in Pa.ents with Prior Exposure

64

Percent of subjects responding by IFN gamma ELISpot Significant Spot Forming Cells averaged across subjects

22/25 pep<des; Average response to the pool was over 1,000 gamma producing cells per million above background.

TulyVax: 6 epitope in LVS Challenge Strain

0

50

100

150

200

250

30030

04

3005

3017

3018

F102

F176

3001

3003

3015

3019

3007

3023

3024

3025

Schu4 peptides with perfect LVS match

Schu4 peptides with partial LVS match

Schu4 peptides without LVS match

IFN

-g S

FC

/10^

6 sp

leno

cyte

s ov

er b

ackg

roun

d

Placebo-immunizedFT_II_v1-immunized

950 -

900 -

TulyVax Immunogenicity in HLA Tg Epitope-‐specific IFNγ Response

Nearly identical immunogenicity profile observed in HLA DR3 mouse immunizations performed in collaboration with Dr. Terry Wu (UNM), illustrating broad reactivity of immunoinformatic predicted epitopes.

57%

0%0%

20%

40%

60%

80%

100%

0 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21Days after lethal bacterial challenge

Perc

ent S

urvi

val

TuliVax Immunized MicePlacebo Recipient Mice

14 epitopes: T cell-‐epitope-‐immunized mice were protected against live challenge with tularemia. Placebo-‐recipient mice died within 10 days.

Rapidity: from genome to candidate vaccine in 24 months . . . Efficacy: 14 epitope vaccine protects against live challenge

TulyVax Efficacy

McMurry et al. Vaccine 2007;25:3179-91 and Gregory et al. Vaccine 2009 27:5299-306

Vaccine

Immunogenic Epitopes

Shared Immunogenic Epitopes

smallpox

vaccinia

Immunome-Derived Smallpox Vaccine: VennVax

88% of predicted T cell epitopes confirmed in vitro using hu PBMC

20

VennVax Class II Epitopes are Antigenic in Dryvax Vaccinees

Moise et al. Vaccine. 2009 27:6471-9

Immunogenicity Day 56

1. epitope DNA vaccine prime (IM) 2. epitope peptide boost (IN)

Immunizations Days 0, 14, 28, 42

Challenge Day 65

VennVax Immunization in HLA DR3 Transgenic Mice

Moise L et al. Vaccine. 2011;29:501-11

Survival of VennVax-‐Vaccinated Mice Aqer Aerosol Challenge

73

0

10

20

30

40

50

60

70

80

90

100

0 5 10 15 20 25

Perc

ent S

urvi

val

Day Post Immunization

Placebo

Vaccinated

DNA DNA boost boost Challenge 17%

0 20 40 60 80 100

100% survival of Vaccinated mice vs. 17% of placebo

Moise et al. Vaccine. 2011; 29:501-11

0

0.5

1

1.5

2

2.5

3

100 200 400 800 1600 3200 6400 12800

OD

490

1/Dilution Factor

Pre-challenge Placebo

Pre-challenge Vaccine

Post-challenge Placebo

Post-challenge Vaccine

Post-challenge

Pre-challenge

Protection Without Vaccine-Induced Antibodies

Therapeutic H. pylori Vaccination

Week 0 Week 6 Week 12-19 Week 51

IFN-gamma and IL-4 ELISpot

Histology

1. epitope DNA vaccine prime IM 2. epitope peptide boost IN

H. pylori SS1

H. pylori SS1

H. pylori SS1

H. pylori SS1 lysate IN

1. epitope DNA vaccine prime IN 2. epitope peptide boost IN

1. control DNA prime IN 2. control peptide boost IN

H. pylori SS1

IFN-gamma Secretion in Response to Splenocyte Restimulation following immunization

0

100

200

300

400

500

600

700

HP

4009

HP

4029

HP

4032

HP

4040

HP

4054

HP

4055

HP

4067

HP

4071

HP

4077

HP

4152

HP

4153

HP

4156

HP

4165

HP

4174

HP

4189

HP

4197

HP

4199

HP

POO

L 1

HP

POO

L 2

HP

POO

L 3

HP

4018

HP

4060

HP

4068

HP

4070

HP

4111

HP

4117

HP

4119

HP

4120

HP

4127

HP

4154

HP

4157

HP

4160

HP

4164

HP

4175

HP

4179

HP

POO

L 4

HP

POO

L 5

HP

POO

L 6

Con

A

SFC/

10̂6

over

bac

kgro

und

Average Helico-Vax

Average SS1

SS1 (whole lysate-immunized mice) recognized few epitopes (white bars); HelicoVax-immunized mice recognized 45 of 50 (dark bars). 45/50 were immunogenic.

HelicoVax: Broad Epitope Recognition

Lysate pVAX DNA IM DNA IN

0

20

40

60

80

100

120

140

160

180

600

800

H. p

ylor

i qPC

R(S

SA/G

APDH

)

*** P<0.001

** P<0.01

*** P<0.001

HelicoVax Eradicates H. pylori Infection

This result accomplished in just over 24 months . . .

Moss et al, Vaccine 2011;29:2085-91

Two Whole Gene Constructs –  Ebola Zaire GP –  VEEV 26S* –  subcloned into pWRG-7077

*Dupuy LC, Richards MJ, Ellefsen B, Chau L, Luxembourg A, Hannaman D, Livingston BD, Schmaljohn CS. A DNA Vaccine for Venezuelan Equine Encephalitis Virus Delivered by Intramuscular Electro-poration Elicits High Levels of Neutralizing Antibodies in Multiple Animal Models and Provides Protective Immunity to Mice and Nonhuman Primates. Clin Vaccine Immunol. 2011 Mar 30.

One Multi-Epitope Construct –  Ebola Zaire/Sudan GP epitopes –  VEEV 26S epitopes –  subcloned into pWRG-7077

VS.

VEEV IDV Development: Comparison with Whole Antigen Vaccine

IFNγ ELISpot responses to VEEV peptide pools

VEEV E1 VEEV E2

USAMRIID DR3 Mouse StudyVEEV Challenge Group ELISA

Day 56 Serum Samples

Neg Con Arm Pos Con Arm Vaccine Arm0

1

2

3

4

5

Log 1

0 Ti

ter

VEEV IDV Elicits Antibody Response

Negative Control

Negative Control

Whole Antigen Vaccine


Epitope-Driven Vaccine


VEEV IDV Protects Against Lethal Challenge

USAMRIID DR3 Mouse StudyVEEV Challenge Survival

0 5 100

102030405060708090

100Neg Con ArmPos Con ArmVaccine Arm

Days postchallenge

Perc

ent s

urvi

val

USAMRIID DR3 Mouse StudyVEEV Challenge Weights

0 1 2 3 4 5 6 7 8 9 10 11 12 1350

60

70

80

90

100 Neg Con ArmPos Con ArmVaccine Arm

Days Postchallenge

% M

ean

Star

ting

Wei

ght

Whole Antigen Negative Control

Epitope-Driven

Vaccine

Negative Control



T helper Epitopes B cell epitopes

Other? CTL? Th2?

Subset of Th epitopes stimulate IFNγ secretion""Combination of immunogenic Th epitopes that overlap B cell epitopes???" "Contribution from other Th epitopes (stimulate other cytokines) that overlap with B-cell epitopes""""Th epitopes that stimulate different subpopulations""""What is clear: that whole Ag is not necessary for protection"

What Drives Protection?

T cells = Immune System Body Armor

T cell response cannot prevent Infec<on but . . .

T cell response can arm against Disease

The "New" Flu (H1N1 2009 California)


2009 Worry: CDC – No Cross-‐reac.ve Ab

•  Preliminary studies of individuals showed that an<bodies induced by seasonal influenza vaccina<on were not cross-‐reac<ve with novel H1N1.

•  What if the T cell epitopes were cross-‐reac<ve? Would that help?

•  (Note that the situa<on is very similar for H7N9 – no cross-‐reac<ve an<body).

Centers for Disease Control and Preven<on. Serum an<body response to a novel influenza A (H1N1) virus aier vaccina<on with seasonal influenza vaccine. MMWR Morb Mortal Wkly Rep 2009;58(19):521–4.


2009 H1N1 contains conserved epitope Sequences – Predicted Cross Protec.on

TIV

2008-‐2009 HA and NA

Novel H1N1

HA and NA

Conserved T-‐Cell Epitopes

Immunogenic T cell

epitopes

TIV

2008-‐2009 HA and NA

Novel H1N1

HA and NA


Immunogenic T cell

epitopes

TIV

2008-‐2009 HA and NA

Novel H1N1

HA and NA


Immunogenic T cell

epitopes

TIV

2008-‐2009 HA and NA

Novel H1N1

HA and NA


Immunogenic T cell

epitopes

Conserved T-Cell

Epitopes

Immunogenic T cell

epitopes

De Groot et al. Vaccine 2009;27:5740-7

Enough Cross-‐protec<ve Epitopes that Seasonal Flu vaccina<on or

exposure may protect


hOp://www.ncbi.nlm.nih.gov/pubmed/19660593

EpiVax Predicted Cross-‐Protec.on


1.00E+06

1.00E+07

1.00E+08

Placebo FluVax 2009

Placebo FluVax 2009

PFU/m

l

2 Days 4 Days

*P= 0.002

Immuniza.on with FluVax cross-‐conserved T cell epitopes decreases lung viral load

108

107

106

Post-‐Infec.on

A handful of conserved epitopes protected against disease

hOp://bit.ly/Moise_Universal_Flu hOp://bit.ly/H1N1_DR3_2013


H1N1 Conclusions

•  This work recapitulates other projects already completed: Complete protection using ONLY T cell epitopes (H. pylori, Tularemia, VennVax)

•  Results of our published studies demonstrate that conserved T cell epitope sequences, important to viral fitness, also may be immunologically significant contributors to protection against newly emerging influenza strains.

•  The conserved epitope approach promises to answer the need for prompt preparedness and delivery of a safe, efficacious vaccine without requiring a new vaccine for every emergent influenza strain.

hOp://bit.ly/Moise_Universal_Flu hOp://bit.ly/H1N1_DR3_2013


What about H7N9?


What Can We Learn About H7N9? Epitopes Novel or Conserved?

H7N9 Circula<ng Flu

As it turns out -‐ -‐ -‐ Very Poor Cross-‐Conserva<on – Only within Internal Proteins


gB-2 (EPX Score: -24.56)

- 80 -

- 70 -

- 60 -

- 50 -

- 40 -

- 30 -

- 20 -

- 10 -

- 00 -

- -10 -

- -20 -

- -30 -

- -40 -

- -50 -

- -60 -

- -70 -

- -80 -

Thrombopoietin

Human EPO

Tetanus Toxin

Influenza-HA

Albumin

IgG FC Region

EBV-BKRF3

Fibrinogen-Alpha

Follitropin-Beta

HA A/California/07/2009 (H1N1)

HA A/Victoria/361/2011 (H3N2)

HA A/Texas/50/2012 (H3N2)

HA A/Shanghai/1/2013 (H7N9) . . . . . . . .. . . . . . . . -‐8.11 HA A/mallard/Netherlands/09/2005 (H7N7) . . . . . . -‐8.63

Random Expecta.on

HA A/mallard/Netherlands/12/2000 (H7N3) .. . . . . .-‐9.91

HA A/chicken/Italy/13474/1999 (H7N1) . . . . . . . . . -‐6.23

H7 HA Immunogenic Poten.al

New H7N9 Flu is Predicted to be POORLY IMMUNOGENIC

hOp://bit.ly/H7N9_HVandI

This is a unique virus

•  Low conserva<on of HA, NA surface proteins is not surprising

•  Internal proteins are more conserved •  And – HA is has unusually low immunogenicity •  Could that explain why infec<on is widespread?

•  Difficult to make an<bodies to the HA


Differen<al Cross-‐reac<vity with the human genome-‐ significance?

H1N1 H7N9


New and unpublished: The “Classic Epitope” Is much more cross-‐conserve with the human genome in the case of H7N9.

This is a unique virus

•  Unusually low immunogenicity •  Cross-‐reac<vity with human genome •  How do we overcome this problem?


•  EpiMatrix – maps T cell epitopes •  ClustiMer - Promiscuous / Supertype Epitopes •  BlastiMer - Avoiding “self” - autoimmunity •  Conservatrix – Identifies Conserved Segments •  EpiAssembler - Immunogenic Consensus Sequences •  Aggregatrix – Optimizing the coverage of vaccines •  VaxCAD - Processing and Assembly

Immunoinforma.cs Toolkit

Seamless Vaccine Design

Integrated toolkit is

unique to iVax


FastVax: Vaccines on demand

•  High throughput computing

•  Immunoinformatics

•  Vaccine design algorithms

•  Vaccine Production

•  Delivery device

•  Animal safety/tox/immunogenicity/validation

•  Deployment by established distribution systems

Prebuilt

Rapid deployment when genome

sequence is in hand

Pilot program Funded by DARPA


20 hours -‐ April 05 – April 06 2013 Extremely Rapid H7N9 Vaccine Design

April 05, 2013: Obtain H7N9 Sequences (4 human-‐sourced; GISAID)

EpiMatrix Analysis: Iden<fica<on of H7N9 Class I and Class II Epitopes

101 H7N9 ICS* Class II Epitopes + 586 Class I Epitopes

April 06, 2013: H7N9 Vaccine: Two Constructs, Class I and Class II

Eliminate Epitopes highly conserved with Human Design vaccine: 12 hours (Logged).

Compare with previous epitopes (IEDB) And other H7N9 strains; create final list 20 hours (Logged).

Obtain all available H7N9 sequences


Regulatory Agency approval

As Currently Proposed with Genome-‐derived Epitope-‐driven Influenza Vaccines (R21 / NIAID / NIH)

Gedng FastVax into the clinic: 4 Steps

1. In silico Design

2. Produc<on and Packaging

3. Clinical Trial

(correlates of immunity)

4. Deployment

Emergency use authoriza<on


H7N9 at EpiVax

•  String-‐of-‐epitopes DNA vaccine (Doug Lowrie) •  String-‐of-‐epitopes Phage vaccine (Ft. Detrick) •  Op<mized HA (fix epitopes) recombinant (TBD?)

•  Op<mized HA + epitope string VLP (Ted Ross) •  Collabora<on with NIID/Japan – in progress

EpiVax Contacts: Anthony Marcello, BDA, [email protected] Anne S. De Groot CEO/CSO [email protected]

105

DNA – chain of epitopes, or pep<de in liposomes ICS-‐op<mized proteins in VLP ICS-‐op<mized whole proteins

H7N9 Delivery vehicles

And . . . Cancer, Allergy and Autoimmune Disease?

107 107

•  Cancer Vax = Epitopes + Adjuvant + ?

•  Tregitope = Novel “adjuvant” that induces tolerance

•  Allergy Vax = Epitopes +Tregitope+Delivery vehicle

•  Autoimmunity Vax= AutoAg+Tregitope+Del. vehicle

• Payload+Adjuvant+ Delivery vehicle = Vaccine



•  Case Studies •  . . . Questions?

108

Outline

EpiVax: Four Core Strengths

Confiden<al

Contact: Anthony Marcello, BDA, [email protected]

Introduction to Computational Vaccinology and iVAX by EpiVax

Technology

Transcript of Introduction to Computational Vaccinology and iVAX by EpiVax