advances in understanding the biology of talimogene ... · t-vec (herpes virus) modifications...

Advances in Understanding the Biology of Talimogene Laherparepvec (T-VEC)

Howard L. Kaufman

Disclosures

• Advisory Boards

– Amgen, Celldex, Compass Therapeutics, EMD Serono, Merck, Prometheus, Turnstone Biologics

• Speaker’s Bureau

– Merck (funds given to Rutgers University)

• Research Funding

– Amgen, BMS, Merck, Viralytics

– NCI, The Gateway Foundation

Why Immunotherapy?

• Specificity

• Memory

• Adaptability

Credited Jim Allison, PhD

Why Oncolytic Viruses?

• Cytotoxicity

• Immunity

• Safety

T-VEC (Herpes Virus)Modifications Rationale

JS-1 strain Improved Cancer Cell Lysis

Deletion of ICP34.5 Cancer Cell Specific Replication

Early Expression of US11

Cancer Cell Specific Replication

Deletion of ICP47 Permits Antigen Presentation

Insertion of GM-CSF Augments anti-tumor Immune response

Kaufman, Kohlhapp, and Zloza Nat Rev Drug Discov. 2015 Sep;14(9):642-62

Talimogene laherparepvec (T-VEC; Imlygic™): Engineered HSV-1 Oncolytic Virus

T-VEC has profound lytic effect against SK-MEL-28 melanoma cells

7

Anti-tumor activity of T-VEC in other tumor cell lines in vitro

Cell lines Tissue Cell survival (%) (MOI=1)

24 hrs 48 hrs 3 days 6 days

A549 Lung cancer 82.5 76.0 55.8 43.1

H460 Lung cancer 65.2 64.0 44.0 27.6

CALU-1 Lung cancer 71.1 60.0 41.9 40.4

PANC-1 Pancreatic cancer 74.6 57.6 24.1 9.4

MIA PACA-2 Pancreatic cancer 66.5 38.5 18.6 1.4

CAPAN-1 Pancreatic cancer 81.0 42.2 56.6 20.3

BxPC-1 Pancreatic cancer 57.6 15.1 16.1 8

HCT116 Colorectal cancer 65.7 27.4 14 1.1

HT29 Colorectal cancer 51.6 22.0 24.3 3.9

SW620 Colorectal cancer 80.4 66.8 45.0 3.9

COLO205 Colorectal cancer 49.8 20.0 9.7 3.1

MTT assays(Higher MOI’s result in complete lysis of all cells)

8

T-VEC improves PFS in patients with melanoma

•Modified PFS was defined as time from the first dose of study treatment until death or development of

the first clinically significant progression for which no objective response was subsequently achieved

Perc

en

t P

rog

ressio

n F

ree

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

GM-CSF

T-VEC

Unadjusted Log Rank: P < 0.0001*

Hazard Ratio: 0.42 (0.32, 0.54)

Study Month0 5 10 15 20

141295

29175

1696

257

00

2.9 (2.8, 4.0) months

8.2 (6.5, 9.9) months

Median (95% CI)

GM-CSF (N = 141)

T-VEC (N = 295)

*P-value is descriptive

only

Risk set, n

163 (55.3%)

Events n (%)

84 (59.6%)

Antdbacka, Kaufman, Collicio et al. JCO 2015 17

9

T-VEC improved overall survival in patients with melanoma

Patients at risk:

T-VEC

GM-CSF 295 269 230 187 159 145 125 95 66 36 16 2

0141 124 100 83 63 52 46 36 27 15 5 0

Events / N (%)Median (95% CI)

in Months

T-VEC 189 / 295 (64) 23.3 (19.5 - 29.6)

GM-CSF 101 / 141 (72) 18.9 (16.0 - 23.7)

HR = 0.787 (95% CI: 0.62, 1.00)

Unadjusted Log-rank P = 0.051

0

0 5 10 15 20 25 30 35 40 45 50 55 60

0%

20%

40%

60%

80%

100%

Ka

pla

n-M

eie

r P

erc

en

t

Survival T-VEC GM-CSFDifference

% (95% CI)

12-month 73.7% 69.4% 4.3 (-4.9, 13.5)

24-month 49.6% 41.3% 8.3 (-1.9, 18.5)

36-month 40.6% 27.8% 12.8 (1.0, 24.6)

Antdbacka, Kaufman, Collicio et al. JCO 2015

10

T-VEC has more profound effect in patients

with stage III and IV M1a tumors

Study Month

0%

20%

40%

60%

80%

100%

Log Rank: p = 0.7094 (descriptive)Hazard Ratio: 1.07 (95% CI: 0.75, 1.52)

T-VECRisk set, n

0 5 10 15 20 25 30 35 40 45 50 55 60

13111284 58 46 41 32 22 15 13 6 1 0GM-CSF 55 46 35 28 20 17 16 14 10 5 3 0

Stage IVM1b/cStage IIIB/C, IVM1a

16315714612911310493 73 51 23 10 1 086 78 65 55 43 35 30 22 17 10 2 0

0 5 10 15 20 25 30 35 40 45 50 55 60

Study Month

0%

20%

40%

60%

80%

100%

Log Rank: p = 0.0009 (descriptive)Hazard Ratio: 0.57 (95% CI: 0.40, 0.80)

0

T-VECGM-CSF

Risk set, n

Ka

pla

n-M

eie

r P

erc

en

t

T-VEC 13.4 (11.4-16.2)

GM-CSF 15.9 (10.2-19.7)

T-VEC 41.1 (30.6, NE)

GM-CSF 21.5 (17.4, 29.6)

0 0

Events / N (%)

60 / 183 (49)

57 / 86 (66)

109 / 131 (83)

44 / 55 (80)

Median (95% CI), mos Events / N (%) Median (95% CI), mos

Antdbacka, Kaufman, Collicio et al. JCO 2015

Tumor regression in injected lesions is greater than in non-injected lesions

Part-1

How does T-VEC kill tumor cells?

HSV-1 Life Cycle

Melanoma cell lines exhibit a range of permissiveness to T-VEC infection

0 1 2 3 4 5 6 7 8 9 100

50

100

MOI

Perc

en

t su

rviv

al

10_24

_16 T-VEC _5 days

M14_TVEC_5day

SK_2_T-VEC_5day

SK_5_T-VEC_5day

SK_28_T-VEC_72hr

LOX_28_T-VEC_72hr

Kaufman, Kohlhapp, and Zloza Nat Rev Drug Discov. 2015 Sep;14(9):642-62

HSV-1 enters cells through HVEM, Nectin-1 and Nectin-2

SKMEL5 and SKMEL28: BRAF V600E; N-RAS WT

PE Isotype average MFI: 629

HVEM triplicate

average MFI: 2711

PE Isotype average MFI: 629

NECTIN1 triplicate

average MFI: 1349

APC Isotype average MFI: 2061

NECTIN2 triplicate

average MFI: 8329

HSV-1 utilizes HVEM, Nectin-1 and Nectin-2 to enter tumor cells

SKMEL5

SKMEL28

T-VEC induces calreticulin expression on the surface of SK-MEL28 cells in a dose-dependent manner*

0.1 MOI (10X) 1.0 MOI (10X)

10.0 MOI (10X) 1.0 MOI (40X)

*Shown at 18 hours post-infection

SK28

-CTR

L-soup-3

6hr

SK28

-T-V

EC-1

MOI-s

oup-36h

r0

10

20

30

AT

P n

M

SK28-ATP-assay

p=0.02

SM

1-CTR

L-soup-3

6hr

SM

1-TV

EC-1

MOI-s

oup-36h

r0

5

10

AT

P n

M

SM1-ATP assay

p=0.001

T-VEC induces extracellular ATP release following infection of melanoma cells

T-VEC induces minimal apoptosis in melanoma cells, which can be enhanced by MEK inhibition

SKMEL 28

-T-V

EC 0

.5 m

oi2

4h

r+M

eki 1

uM

3

6 h

rs

-MEK

i1U

M 3

6 h

rs

-T-V

EC 0

.5 M

OI 2

4 h

r

-No

tre

atm

ent

Vinculin

Cleaved casp3

-T-V

EC 0

.5 m

oi2

4h

r+M

eki 1

uM

3

6 h

rs

-MEK

i1U

M 3

6 h

rs

-T-V

EC 0

.5 M

OI 2

4 h

r

-No

tre

atm

ent

-20-10

-15

T-VEC induces caspase 3 cleavage in the presence of MEK inhibition

TVEC

TVEC+4

0nM

Tra

mei

nib

TVEC+2

50uM

ZVA

D

contr

ol0

20000

40000

60000

80000

100000

MOI 0.12 3days

Lu

min

iscen

ce

TVEC

TVEC+40nM Trameinib

control

TVEC+250uM ZVAD

p=0.15

p=0.01

ns

Caspase 3/7 assay


T-VEC does not appear to induce autophagy

PKR and IRF7 protein expression is variable across melanoma cell lines:

Correlation with T-VEC-mediated lysis

IRF3 protein expression is variable across melanoma cell lines: Correlation with T-VEC lysis

0 1 2 3 4 5 6 7 8 9 100

50

100

MOI

Perc

en

t su

rviv

al

10_24

_16 T-VEC _5 days

M14_TVEC_5day

SK_2_T-VEC_5day

SK_5_T-VEC_5day

SK_28_T-VEC_72hr

LOX_28_T-VEC_72hr

STING expression correlates with tumor cell resistance to T-VEC infection

Conclusions: T-VEC tumor cell killing

• Melanoma cell express variable levels of HSV entry receptors (HVEM, Nectin-1 and Nectin-2)– All cell susceptible to infection but receptor expression is

variable across cell lines– Nectin may be important for viral entry

• T-VEC induces immunogenic cell death/releases DAMPs– Calreticulin, ATP and HMGB-1

• T-VEC induces apoptosis– Effect can be enhanced by MEK inhibition

• T-VEC does not appear to induce autophagy• T-VEC-mediated lysis may depend on PKR and IRF3/7

Part-2

Can combinations treatment enhance T-VEC-mediated tumor cell lysis?

Herpes virus replicates more efficiently in RAS-mutated tumor cells

29 Cuddington and Mossman J Virol 2014

MEK inhibition sensitizes melanoma cell line to T-VEC lysis

0.0001 0.001 0.01 0.1 1 100

20

40

60

80

100

MOI

sk28 T

VE

C a

lon

e

TVEC+MEKi

sk28 TVEC alone

5nM Trametinib + TVEC

10nM Trametinib + TVEC

T-VEC contributes to MEK inhibitor cell killing in a dose-dependent manner

SK MEL 28

Increased cell lysis by combination T-VEC/MEKimay be due to increased viral replication

-SK

28

_Cel

ls o

nly

-SK

28

_Cel

ls+

1 M

OI-

T-V

EC

-SK

28

_Cel

ls+

1 M

OI-

T-V

EC+

DM

SO

-SK

28

_Cel

ls+

MEK

Inh

ibit

or

10

nM

-SK

28

_cel

ls +

DM

SO

-SK

28

_Cel

ls+

10

nm

MEK

Inh

ibit

or+

1 M

OI-

T-V

EC

HSV1 gD ~43kD

Vinculin

Sk28 cells pretreated with 10nM Trametinib for 12 hours and then infected with T-VEC 1 MOI for 16 hours. Protein lysate collected and 40ug is loaded.

Indicated cells were pretreated with 10nM MEKi for 12 hours and then infected with 0.1 MOI

T-VEC; 36 hrs post-infection viral titers were measured from supernatants by plaque assay.

MEK Inhibition increases viral titers in T-VEC treated Melanoma cell lines

Lumacyte uses radiance via velocity histogram to identify T-VEC infected cells

Example of typical histogram data

Radiance infection metric measures viral infection

Principal Component Analysis differentiates infected vs. non-infected, and MEK-inhibited vs. uninhibited cells

10/26 11/1 11/15 11/290

200

400

Tu

mo

r vo

lum

e

T-VEC10^6 pfu_MEK_1mg/kg NSG Data

Contol

T-VEC

MEK Inhibitor

Combo

Initial studies suggest Combination T-VEC and MEKiinduce anti-tumor immunity in murine melanoma

Conclusions

• MEK inhibition enhances T-VEC-mediated cell killing in vitro in a dose-dependent manner

• Combination treatment changed the morphology of tumor cells and infection can be detected by changes in cell radiance and velocity

• MEK inhibition may allow an increase in viral replication

• MEK inhibition may enhance T-VEC-mediated melanoma killing in vivo

Part-3

Immunological effects of T-VEC

Nectin1 expression is required for T-VEC-mediated lysis in murine B16 melanoma cells

5 10 15 20 25

-100

0

100

200

300

400

mm

2

PBS

T-VEC

5 10 15 20 25

-100

0

100

200

300

400

mm

2

PBS

T-VEC

Left FlankRight Flank

T-VECInfection

Contralateral B16-Nectin Treated with T-VEC

B16 B16-nectin-1

CD3

CD

11

c

16.2% 31.4%

PBS T-VEC

4 Days Post Injection

PBS T-VEC0

10

20

30

40

%C

D1

1c+

**

PBS T-VEC

CD3

CD

11

c

31.5% 27.8%

PBS T-VEC


PBS T-VEC0

10

20

30

40

%C

D1

1c+

ns

PBS T-VEC

T-VEC Increases Monocyte Infiltration

PBS T-VEC0

5

10

15

20

25

PBS T-VEC0

2

4

6

8

CD3

6.7% 19.5%

PBS T-VEC


CD3

NK

1.1

4.6% 3.0%

PBS T-VEC


%N

K1

.1+

ns

PBS T-VEC

%N

K1

.1+

**

PBS T-VEC

NK

1.1

T-VEC Increases NK Cell Infiltration

CD3

CD

8

16.7% 19.3%

PBS T-VEC


CD3

CD

8

9.6% 48.8%

PBS T-VEC


PBS T-VEC0

20

40

60

%C

D8

+

p < 0.056

PBS T-VECPBS T-VEC0

5

10

15

20

25

%C

D8

+

ns

PBS T-VEC

T-VEC Increases CD8 T Cell Infiltration

PBS T-VEC0

20

40

60

CD3

0.8% 15.2%

PBS T-VEC


CD3

HSV

-1 T

et

0.0% 42.9%

PBS T-VEC


%H

SV-1

Tet

+

**

PBS T-VECPBS T-VEC0

10

20

30

40

%H

SV-1

Tet

+

ns

PBS T-VEC

HSV

-1 T

et

T-VEC Increases HSV-1 gD-Specific CD8 T Cell Infiltration

HSV gD-specific CD8+ T cells accumulate in both injected and uninjected tumors

Conclusions from Immunologic Effects

• Tumor regression is more pronounced in injected tumors but does occur in non-injected tumors

• T-VEC appears to induce local innate and adaptive immune responses

• HSV-specific CD8+ T cells are detected in both injected and un-injected tumors (hypothesis: chemokine gradient?)

• Cross presentation of tumor antigens is being evaluated

The “inflamed” vs. “non-inflamed” tumor microenvironment

“Inflamed”

“Non-inflamed”

Patient #1

Patient #2

a Dosing of T-VEC was 4 mL × 106 PFU/mL once, then after 3 weeks, 4 mL × 108 PFU/mL Q2W.

Ipilimumab 3mg/kg IV Q3W x 4

Primary Endpoint: Incidence of dose-limiting toxicities / ORR per mirRC

Key Secondary Endpoints: ORRirRC, Safety / PFS, OS, BORR, time to response, duration of response

Study schema of phase 1b/II trial of combination T-VEC and ipilimumab

Unresectable Stage IIIB-IV

Melanoma:

•Injectable

•Treatment naive

•ECOG PS 0 or 1

•No CNS mets

T-VEC Intralesional

106 PFU/mL, after 3 weeks 108 PFU/mL Q2Wa

Week 6

Phase Ib N = 19

Phase II N=198

T-VEC dosing until CR, all injectable tumors disappeared, PD per irRC, or intolerance whichever comes first.

Week 1

Puzanov et al. JCO 2016Chesney et al. Submitted, 2017

Ph II stratification:Disease stageBRAF status

Spider plot of individual subject responses in Phase Ib T-VEC and Ipilimumab trial

51Puzanov et al. JCO 2016

52

Phase 1b: Treatment-Emergent Adverse Events*

Preferred TermTotal

N (%)

Grade 3

N (%)

Any event 19 (100) 5 (26)

Any attributed to T-VEC 17 (90) 3 (16)†

Any attributed to ipilimumab 15 (80) 3 (16)†

Chills 11 (58) -

Fatigue 11 (58) 1 (5)

Pyrexia 11 (58) 1 (5)

Nausea 9 (47) 2 (11)

Rash 9 (47) -

Diarrhea 8 (42) 1 (5)

Headache 8 (42) -

Pruritis 7 (37) -

Decreased appetite 4 (21) -

Hyperglycemia 4 (21) -

Vomiting 4 (21) 1 (5)

ALT increased 3 (16) -

Back pain 3 (16) 1 (5)

Influenza-like illness 3 (16) 1 (5)

Pain 3 (16) -

Vision blurred 3 (16) -

*All events of any grade occurring in > 15% of patients during treatment or up to 30 days after last T-VEC or 60 days after last

ipilimumab, whichever is later; †Grade 3 events in these patients: pyrexia attributed to T-VEC; hypophysitis and abdominal

distention attributed to ipilimumab; and nausea, diarrhea, fatigue, influenza-like illness, vomiting, adrenal insufficiency, and

dehydration attributed to both; ALT: alanine aminotransferase

• The only grade 3 event

occurring in > 1 patient was

nausea

• The only two grade 4 events

were in a patient with elevated

amylase and lipase (attributed

to ipilimumab)

• There was one grade 5 event of

metastases to central nervous

system (preferred term)

Puzanov et al. JCO 2016

Changes in activated CD4+ICOS+ and CD8+DR+T cells after T-VEC and Ipilimumab

53Puzanov et al. JCO, 2016

CD4+ICOS+ T cells

CD8+DR+ T cells

Phase II baseline patient/tumor characteristics

Chesney et al. Submitted, 2017

Incidence of Adverse Events

Waterfall plots of response rates

Output: f14-04-001-001-eff-waterfall-bestchg-tum-burd-itt-p2-l.rtf (Date Generated: 06FEB17 13:57) Source Data: adtminv, adsl

Program: /userdata/stat/amg678/onc/20110264/analysis/pa_p2_201610_ub_f/figures/f-eff-waterfall-bestchg-tum-burd-p2.sas

Evaluable subjects are defined as subjects who had baseline and at least one post baseline tumor burden.

Perc

enta

ge C

hange f

rom

Baselin

e

-100

-75

-50

-25

0

25

50

75

100

2:Stage IV M1c (N=26)2:Stage IV M1b (N=10)2:Stage IV M1a (N=15)2:Stage IIIC (N=27)2:Stage IIIB (N=8)


-100

-75

-50

-25

0

25

50

75

100



100% 17 (19.8%)

≥50% 30 (34.9%)

Any 46 (53.5%)

Max. decrease

200

400

600

2:Ipilimumab (N=86)

-100

-75

-50

-25

0

25

50

75

100

-100

-75

-50

-25

0

25

50

75

100

100% 22 (24.7%)

≥50% 45 (50.6%)

Any 53 (59.6%)

Max. decrease

200

400

600

1:Talimogene Laherparepvec + Ipilimumab (N=89)

Chesney et al. Submitted, 2017

T-VEC and Ipilimumab vs. Ipilimumabresponse rates

There is no difference in PFS between combination T-VEC/Ipi and ipilimumab alone

||||||||||||

||

||||||||||||||

||| |

|

|||||| |

||

|| |||| | | |

|||

||

|||

||

|| || | |

||| |||

||

| |

|

|||

|

||| |||

| | || | |

|

100 50 31 23 18 16 6 3 3 3 3 1 0

98 59 44 36 30 21 16 7 5 2 1 1 0

0 3 6 9 12 15 18 21 24 27 30 33 36

Study Month

0%

25%

50%

75%

100%

Kapla

n-M

eie

r P

erc

ent

Ipil imumab

Talimogene Laherparepvec+Ipil imumab

Talimogene Laherparepvec + Ipilimumab (N = 98) Median (95% CI) 8.2 (4.2, 21.5)

Ipilimumab (N = 100) Median (95% CI) 6.4 (3.2, 16.5)

Censor indicated by vertical |.

Subjects that have not had an event of death or disease progression are censored at their last evaluable tumor assessment date.

Progression-free survival (PFS) is defined as time from randomization to the date of f irst of confirmed disease progression per modif ied irRC criteria, or death, w hicheveroccurs earlier.

N = Number of subjects in the analysis set. CI = Confidence Interval. One month = 365.25/12 days.

Program: /userdata/stat/amg678/onc/20110264/analysis/pa_p2_201610_ub_f/figures/f-eff-km-pfs-mirrc-p2.sas

Output: f14-04-004-001-eff-km-pfs-mirrc-itt-p2-l.rtf (Date Generated: 01NOV16 14:49) Source Data: adtte, adsl

||||||||||||

||

||||||||||||||

||| |

|

|||||| |

||

|| |||| | | |

|||

||

|||

||

|| || | |

||| |||

||

| |

|

|||

|

||| |||

| | || | |

|

100 50 31 23 18 16 6 3 3 3 3 1 0

98 59 44 36 30 21 16 7 5 2 1 1 0

0 3 6 9 12 15 18 21 24 27 30 33 36

Study Month

0%

25%

50%

75%

100%

Kapla

n-M

eie

r P

erc

ent

Ipil imumab

Talimogene Laherparepvec+Ipil imumab

Log Rank: p = 0.348

Hazard Ratio: 0.83 (0.56, 1.23)

Number of Subjects at Risk:

Talimogene Laherparepvec + Ipilimumab (N = 98) Median (95% CI) 8.2 (4.2, 21.5)

Ipilimumab (N = 100) Median (95% CI) 6.4 (3.2, 16.5)

Censor indicated by vertical |.

Subjects that have not had an event of death or disease progression are censored at their last evaluable tumor assessment date.

Progression-free survival (PFS) is defined as time from randomization to the date of f irst of confirmed disease progression per modif ied irRC criteria, or death, w hicheveroccurs earlier.

N = Number of subjects in the analysis set. CI = Confidence Interval. One month = 365.25/12 days.

Program: /userdata/stat/amg678/onc/20110264/analysis/pa_p2_201610_ub_f/figures/f-eff-km-pfs-mirrc-p2.sas

Output: f14-04-004-001-eff-km-pfs-mirrc-itt-p2-l.rtf (Date Generated: 01NOV16 14:49) Source Data: adtte, adsl

Disease subset analysis

Response rate Disease control rate

Progression-free survival

60Chesney et al. Submitted, 2017Hodi et al. NEJM 2010

Overall survival for T-VEC + Ipilimumab

61

Chesney et al. Submitted, 2017Hodi et al. NEJM 2010

Pembrolizumab and T-VEC in melanoma

Long et al. SMR Abstract 2015

Conclusions

• T-VEC is the first oncolytic virus to achieve regulatory approval for the treatment of melanoma

• Patients with stage III/IV M1a and in first-line setting has the best outcomes

• Combination T-VEC and T cell checkpoint inhibitors appears promising

• T-VEC kills melanoma cells through immunogenic and apoptotic mechanisms, releases, and lysis may be enhanced by MEK inhibition

• T-VEC releases DAMPs and appears to generate an HSV-specific T cells responses (and tumor-specific?)

Acknowledgments

Clinical Trial Investigators• Robert Andtbacka, MD• Jason Chesney, MD• Frances Collichio, MD• Philip Friedlander, MD• Claus Garbe, MD• John Glaspy, MD• Omid Hamid, MD• Axel Hauschild, MD• Celeste Lebbe, MD• Ted Logan, MD• Mohammed Milhem, MD• Igor Puzanov, MD• Merrick Ross, MD• Parminder Singh, MD

Study Participants & Supporters

• Patients and Families

• Jennifer Gansert, MD, PhD

• Lisa Chen

• Jenny J. Kim

Acknowledgments

Rutgers Cancer Institute of NJ

• Ann Silk, MD

• James Goydos, MD

• Vadim Koshenkov, MD

• Janice Mehnert, MD

• Tracie Saunders, RN

• Jennifer Bryant, RN

• Andrew Zloza, MD, PhD

• Dan Medina, PhD

• Megha Shettigar, PhD

• Praveen Bommareddy

• Ricardo Estupinian

Mass General Hospital• Sam Rabkin, PhD• Robert Martuza, PhD

• Don Lawrence, MD• Keith Flaherty, MD, PhD• Ryan Sullivan, MD• Justine Cohen, DO• Ken Tanabe, MD

FundingNCI UM1 CA 186716-01NCI NO1 CA 10-034NCI RO1 CA 093696 The Gateway Foundation

advances in understanding the biology of talimogene ... · t-vec (herpes virus) modifications...

Documents