Substrate Binding by DnaK ���(HSP70 family)

Zhu X et al. Science 1996;272:1606–1614.

Normal, mutated and cancer peptides

HSP

Antigenic Peptides

Antigenic Peptides Bind to HSPs HSPs Bind Peptides: Normal, Mutated and Cancer Peptides HSPs function as peptide chaperones in every nucleated cell

APC

Injection of HSP-peptide complexes Internalization

and re-presentation of peptides

CD91

Cross- presentation

CD4+

T cell

MHC class I/peptide

MHC class II/peptide

TCR

CD8+

T cell

? CD36 CD91 CD40 CD14 TL-R2 TL-R4

NK cells

MCP-1 MIP-1α RANTES NO

IL-12 TNF-α IL-1β GM-CSF

Nucleus Induction of cytokines and chemokines

Srivastava P. Nature Rev 2002;2:185–194.

Mechanism of Action

Oncophage Manufacturing: Purification

Primary recovery

Purification

Finishing

Tissue processing (break open cells)

Remove solids (clarification)

Affinity chromatography

Ion-exchange chromatography

Buffer exchange

Sterile filtration

Vial fill

96-kD Starting Purified Sample Product

MK-0069-01

0

3

6

9

12

Pre-Op Final Vacc

Fold

Ind

uctio

n

r gp96a gp96

%  of  T

 cells  IFNγ  +  

%  of  N

K  cells  IFNγ  +  

Adap%ve   immune   responses   (CD8+)   and   innate   immune   responses   (NK)   are  measured   in   an   ex   vivo   an%gen  presenta%on   assay   controlled   for   non-­‐specific  effects   of   HSP   with   recombinant   HSP   (r   gp96).     For   pa%ents   who   undergo  resec%on   aLer   vaccina%on   at   progression   site,   directed   brain   biopsies   were  performed  to  assess  extent  of  gamma  interferon  posi%ve  CD  8  +  T  cells  and  NK  cells.    All  pa%ents  tested  to  date  have  demonstrated  a  significant  Adap%ve  and  Innate  Immune  response  peripherally,  and  at  the  site  of  tumor  resec%on  in  situ  when  biopsies  were  performed.    A   typical   pa%ent’s   response  profile   is   shown  here;   demonstra%ng   significant   peripheral   and   site   specific   an%-­‐tumor  immunity.     This   pa%ent   received   a   total   of   8   injec%ons   of   HSPPC-­‐96   and  underwent  biopsy  for  suspicion  of  disease  progression.  

*  

*

*P<0.01  

*  

Adap%ve  and  Innate  Immune  Responses  

CD8+  T  Cell  Response   NK  Cell  Response  

PBL  IFNγ  Response  

Phase  2  Mul+-­‐Center  

Parsa et al. J Clin Oncol 29: 2011 (suppl; abstr 2565)

68.6

CD3  15.1 15.7

30.338.8

CD8  

CD8  

Pre-­‐Vaccine   Post-­‐Vaccine  

IFNγ  

1.31 1.17

46.650.9

57.6

Localiza%on  of  CD8+  IFNγ  T  cells  to  Tumor  Site  Phase  2  Mul+-­‐Center  

Parsa et al. J Clin Oncol 29: 2011 (suppl; abstr 2565)

CD3  

CD56

 

CD56  

IFNγ  

26.1 13.3

4416.6

27.5 5.63

28.138.8

0 19.6

80.40

0 2.98

970

Pre-­‐Vaccine   Post-­‐Vaccine  

Localiza%on  of  IFNγ+  Natural  Killer  cells  to  Tumor  Site  

Phase  2  Mul+-­‐Center  

Parsa et al. J Clin Oncol 29: 2011 (suppl; abstr 2565)

Median:    47.6  wks    (ITT:    42.6  wks)  

Median  Follow-­‐‑up  Time:  46  wks

≥  26  wk  survival:    93%  (ITT:    91%)

Surv

ival

Dis

tribu

tion

Func

tion  

0.0  

0.1  

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

Time to death (wks)  

0   13   26   39   52   65   78   91   104   117   130  

30   30   28   20   10   4   3   2   2   1   1  

G-200 (HSPPC-96) (n = 30)   Censored  

Overall  Survival  

Efficacy  popula%on  (≥4  doses)  

Phase  2  Mul+-­‐Center  

Parsa et al. J Clin Oncol 29: 2011 (suppl; abstr 2565)

Survival  Outcomes  Compared  with  Similar  Surgical  Popula%ons  

Data  Set   Median  Survival   6-­‐Month  Survival  Rate  

G-­‐200   47.6  weeks   93%  

Gliadel    (Kumar  et  al.,  2010)  

39.8  weeks   62%  

NABTC  Phase  2  trials  from  Feb.  1998  –  Nov.  2008  (Clarke  et  al.,  2011)  

31.4  weeks*   56%*  

UCSF  Contemporary  Control  Database**    

32.8  weeks   68%  

*From  Table  4,  combined  data  with  surgery  **N=86,  Jan.  2005  –  Aug.  2009,  recurrent  GBM  with  surgery,  median  age  =  53  years,    all  pts  with  primary  GBM,  all  previously  received  Stupp  protocol,  none  received  G-­‐200,  median  KPS  =  90    

Event G-200 (N=33) Number of Patients with as Least One Related Adverse Event 17 (51.5%)

Fatigue 4 (12.1%)

Injection Site Stinging 1 (3.0%)

Injection Site Reaction 14 (42.4.%)

Skin Desquamation 1 (3%)

Adverse Events Related to G-200

No  Related  Grade  3  or  4  Adverse  Events  

Phase  2  Mul+-­‐Center  

Parsa et al. J Clin Oncol 29: 2011 (suppl; abstr 2565)

Specific Concerns That We Addressed 1) Proposed eligible patient population, single agent activity of the HSPPC-96 vaccine in GBM, and uncertain clinical benefit of bevacizumab in the study patients (post-complete surgical resection of the GBM recurrence)—implications for the control and experimental arms. 2) Broadening the eligibility to include patients with residual disease after resection of the GBM recurrence. 3) The study is designed with a very optimistic view of the effect size which is not sufficiently supported bythe available data. At the same time, type I error rate is relaxed to control the sample size. A more conservative estimate of the effect size and tighter error rates should be considered. 4) Study should be monitored for futility and be closed for if the desirable effect is found to be an unlikely outcome. 5) QOL and PRO outcomes should be evaluated. 6) Previous failure of HSPPC-96 in renal cell carcinoma and melanoma

Results:  Modified  CD4  cells  Pre-­‐op,  Post-­‐op  and  at  recurrence  of  GBM  

Pre-­‐surgery  

Post-­‐surgery  

Recurrence  

0  

2  

4  

6  

8  

10  

12  

14  

MPBL  Levels  

CD25

 

FoxP3  

CD25

 

FoxP3  

CD25

 

FoxP3  

XXX   Censored  

HSPPC-­‐96  (N=25)  

XXX   Censored  

PC              (N=33)  

|  |  |   Censored  

HSPPC-­‐96  (N=184)  

|  |  |   Censored  

Observa%on  (N=178)  

HSPPC  in  Pa%ents  With  Less  Tumor  Burden  

Melanoma  M1a+M1b   RCC  Intermediate  Risk  P=0.034  HR=0.45  

P=0.004  HR=0.52  

Wood C et al. Lancet. 2008;372(9633):145-154. Testori et al. J Clin Oncol. 2008;26(6):955-962.

“Alliance  Trial”  G-­‐200+Bevacizumab    

RECURRENT GBM  

S U R V I V A L  

>90% RESECTION;

KPS>70  

Bevacizumab alone  

G-200 + bevacizumab  

R A N D O M I Z E  

PATH CONF I RMED GBM  

MANUFA CTURE ≥ 6 v I a l s  

18 Months Follow up  

1º  Endpoint:   Overall  survival   2º  Endpoints: Safety,  

Progression  Free  Survival

Sample  Size:     HR:  1.5  (estimates  13.5  months  median  survival  in  combination  arms  

vs.  9  months  in  bevacizumab α  (1-­‐‑sided)  =  0.05;  Power  =  80%

                                                      Randomized  1:1:1;  N  =  255  (85  per  arm)

Dosing: G-­‐‑200  given  id  1x  per  wk  for  4  wks;  every  other  wk  thereafter  up  to  12  total  injections  

(up  to  5  months  of  treatment).    Bevacizumab  given  as  standard  dose  (10  mg/kg)    every  2  wks  

Mul%-­‐center,  three  arm,  randomized  trial  

Potential  Exploratory/Correlative  Studies: -­‐‑Immune  response -­‐‑B7H1  expression -­‐‑PI3  Kinase  activation -­‐‑Molecular  profiling   (initial  tumor  tissue  and  on  biopsy  at  presumed  progression)  

G-200 alone followed by

bevacizumab at progression