Towards personalized p immunotherapy of cancer Hans-Georg - - PowerPoint PPT Presentation

NCT Conference 2013 New Cancer Targets

Towards personalized p immunotherapy of cancer

Hans-Georg Rammensee

Interfakultäres Institut für Zellbiologie Abteilung Immunologie

• Prof. Dr. Hans‐Georg Rammensee

Towards patient-specific tumor antigen selection for i ti vaccination.

Rammensee HG, Weinschenk T, Gouttefangeas C, Stevanović S. I l R 188 164 76 2002 Immunol Rev 188:164-76, 2002

• 1. MHC, peptides, and cancer
• 2. Our strategy to identify cancer associated peptides

3 Cli i l t di

• 3. Clinical studies
• 4. Why mutated antigens should be the better targets

y g g

• 5. Problems

protein fragment

this can be recognized by T cells

This is true for all

Antigen presentation

p g = peptid This is true for all proteins inside a cell, also for tumor

Antigen presentation by HLA molecule

associated changes, including mutations

Antigen processing Antigen processing

Cancer cells differ from normal cells from normal cells

eg., in gene expression

This can be sensed by T cells - no matter which cellular compartment is affected by the change a ected by t e c a ge

3 classes of tumor antigens

Limited to few cancer entities We know plenty of such antigens; immune responses tend to be weak but stronger immune responses may lead This is our vision but stronger immune responses may lead to autoimmunity

I so la tio n o f Na tura lly Pre se nte d HL A I so la tio n o f Na tura lly Pre se nte d HL A-

• L

ig a nds L ig a nds I so la tio n o f Na tura lly Pre se nte d HL A I so la tio n o f Na tura lly Pre se nte d HL A L ig a nds L ig a nds

07.07.2011 L a b Me e ting Da nie l K

• wa le wski

Peptide preparation

General survey

Andy Weinzierl

y

• Protein A affinity chromatography

i W6/32 tib d using W6/32-antibody

• elution with citrate buffer, pH 3

Technological progress

HLA ligandome analysis, next generation

Daniel Kowalewski

D

Daniel Kowalewski

D

Technological advance: lt HPLC ultraHPLC

Comparative HLA ligandome analysis

Off‐the‐shelf vaccines: Allele‐specific target identification (A*02) p g ( )

6000+ peptides exclusively identified in A*02+ tumor ligandomes 362 peptides identified in ≥ 50 % of A*02+ tumor ligandomes tumor ligandomes

Department of Immunology Oncology Hematology Immunology Rheumatology

Ligandomics in leukemia

Juliane Stickel

HLA ligand isolation

ells

g

ignant ce

Purification Affinity Chromatography

s Mali

Leukemia Patient Cell lysate HLA antibody

id l i

PBMCs

Acid elution Ultrafiltration 10 kDa

Isolated peptides LC-MS/MS Healthy donor

• d_70521_28 1 (0.019) Sm (Mn, 2x3.00)

• d_70521_28 1 (0.019) Sm (Mn, 2x3.00)

Healthy donor Peptide sequences & Source protein annotation

Overall comparison HLA Ligandome

HLA ligandome comparison CLL HLA ligandome comparison CLL

30

CLL Patient PBMC (n = 20) Healthy Volunteer PBMC (n = 31)

10 20

e ind ividu als

• 30
• 20
• 10

m b er of p ositive

• 40

30

N u m

als

10 12

ive individua

6 8 10

mber of posit

2 4

1732 CLL exclusive proteins

Num

HLA ligandome comparison AML HLA ligandome comparison AML

5 10 15 s

AML patient PBMC (n=10) healthy volunteer PBMC (n=21)

• 10
• 5

f positive individuals

• 25
• 20
• 15

number of

• 30

HLA ligand source proteins

als sitive individua Number of po

1545 AML exclusive proteins

AML20

Peptide-specific CD8-T-cells after “recall” stimulation in AML patients

AML20

P1

• APLP2
• DKGZ

HIV A*03 HIV A*03 P1 P1 P2 P2

DKGZ

• FAF1
• MITCH 2

P2 KLF2

• KLF2
• MET7A
• VCIP 1
• WIPI 1

P4 P4 PBMCs PHA

WIPI 1

HIV A*03 P2 P1 FΥ+

2,4%

FΥ+ FΥ+

1,5% 0%

INF INF INF CD8+ CD8+ CD8+

Peptide-based Immunotherapy for Leukemia

Leukemia patient

n“

Exome sequencing Immunoprecipitation

rd therapy

• n induction

Mutations + Idiotype sequences

Epitope prediction

LC/MS analysis

Peptide sequences, Source protein annotation

Standar „ remissio

Mutated HLA-Ligands + Idiotype HLA Ligands

SYFPEITHI Source protein annotation, Comparison to healthy tissue

Leukemia associated HLA ligands

Check for immunogenicity Check for immunogenicity

P ti t ifi i ti tid Patient specific vaccination peptides Peptide vaccination = maintenance therapy

Acknowledgments Acknowledgments g

• Prof. L. Kanz
• Prof H Salih
• Prof. H.-G. Rammensee
• Prof S Stevanovic
• Prof. H. Salih
• Prof. S. Stevanovic
• Juliane Stickel
• Daniel Kowalewski
• Heiko Schuster
• Heiko Schuster
• Claudia Berlin

fortüne Programm

Clinical studies

immatics Nature Medicine 2012 Nature Medicine, 2012

Multipeptidvakzinierung beim Nierenzellkarzinom

28

RCC phase 2 trial (IMA901-202) European multicenter study (10 countries, 50 centers)

Cyclophosphamide (300 mg/m2 as single infusion) Advanced RCC (N=68)

• HLA-A*02-positive
• 1 previous line of therapy
• Measurable lesion(s)

Documented progression IMA901 plus GM-CSF (i.d.)

R

IMA901 plus GM CSF (i d )

• Documented progression

Follow-up for OS 17 vaccinations over 9 months IMA901 plus GM-CSF (i.d.) Stratification: 1) First-line therapy: cytokine(s) vs TKI 2) Risk score: low vs intermediate [MSKCC]

Patients: mRCC patients following failure of first-line treatment (TKI and/ or cytokines) Treatm ent: Single infusion of cyclophosphamide as immunomodulator (300 mg/ m 2) prior to the 1st vaccination with IMA901 (413 g i d ) and GM CSF (75 g i d ) to the 1st vaccination with IMA901 (413 g i.d.) and GM-CSF (75g i.d.) Endpoints: Primary EP was DCR after 6 months. Secondary EP included PFS, OS, and

• safety. Immunomonitoring for peptide-specific T-cells (ELISpot and MHC

multimer) and Tregs Conduct: 68 patients were randomized between 2007 and 2009 in 23 centers across 10 European countries

29

IMA901 Renal Cell Cancer Phase II Trial (N=68) Low-dose CY shows benefit only in immune responders espo de s

Single-dose CY pre-treatment associated with survival benefit

• Only in patients with vaccine-induced immune responses (HR=0.38, p=0.040)
• Not in patients without immune responses, arguing against single agent effect of CY (HR=0.92,

Not in patients without immune responses, arguing against single agent effect of CY (HR 0.92, p=0.870)

Number of immune responses associated with survival

Walter, Weinschenk et al. (2012), Nature Medicine

p

• (p=0.023)

IMA910 composition (colorectal carcinoma) 13 tumor-associated peptides

30

TUMAP Name Function / Comments HLA-A*02 TUMAPs C20-001 Chromosome 20 open reading frame 42 Poorly characterized, strong overexpression CCN-001 Cyclin D1 Cell cycle regulation, frequently upregulated in many cancer types CEA-004 Carcinoembryonic antigen-related cell adhesion molecule 5 (CEA) Well-established TAA in CRC, cell adhesion, metastasis; MET-001 Met proto-oncogene Proliferation, motility, adhesion, invasion MUC-001 Mucin 1 Well-established TAA in CRC, unmasking of epitope d t lt d l l ti i t due to altered glycosylation in tumors NOX-001 NADPH oxidase 1 Strong overexpression, inhibition of apoptosis ODC-001 Ornithine decarboxylase 1 Transformation, pro-angiogenic PCN-001 Proliferating cell nuclear antigen Proliferation (DNA replication) g g ( p ) TGFBI-001 Transforming growth factor beta- induced Tissue remodelling, angiogenesis TOP-001 Topoisomerase (DNA) II Proliferation (DNA replication) HLA-DR TUMAPs HLA-DR TUMAPs CEA-006 Carcinoembryonic antigen-related cell adhesion molecule 5 (CEA) Well-established TAA in CRC, cell adhesion, metastasis MMP-001 Matrix metallopeptidase 7 (matrilysin, uterine) Tissue remodelling, inhibition of apoptosis (matrilysin, uterine) TGFBI-004 Transforming growth factor beta- induced Tissue remodelling, angiogenesis

IMA910 Colorectal Cancer Phase I/II Study Overall survival of multi-TUMAP responders

31

100 100

HR=0.59 0 083 HR=0.60 p=0 12

OS vs. Class II response OS vs. Class I response

50 75

survival %

50 75

l survival %

p=0.083 p=0.12

25

0-1 >=2 Overall

25

0-1 >=2 Overall

N=46 N=35 N=25 N=46

100

100 200 300 400 500 600 700 800 900 1000 1100 100 200 300 400 500 600 700 800 900 1000 1100 1200

OS vs. Class I and Class II resp.

Overall survival relative to Visit VC (Follow up 1). Per Protocol population p-values from Log Rank statistics; HR from Cox prop. hazards model

days days

75 100

vival %

Trend for increased OS in multi-TUMAP responders

• Trend for increased OS observed in Class I

HR=0.53 p=0.088 25 50

O thers Overall surv

• r
• r Class II Multi-TUMAP responders
• Consistent finding: effect most pronounced

in patients responding to multiple Class I C

N=47

100 200 300 400 500 600 700 800 900 10001100

>=2 Days

and Class II TUMAPs

N=24

Active immunotherapy can work, even with self antigens

Lessons:

1 One should aim to have more multipeptide responders

• 1. One should aim to have more multipeptide responders.

2 Single low dose of Cyclophosphamide appears to be

• 2. Single low dose of Cyclophosphamide appears to be

good.

• 3. Immunotherapy of this kind may have no impact on

progression free survival but on overall survival progression free survival but on overall survival.

multipeptide vaccination, prostate carcinoma

Clinic for Urology, Dept. Immunology Tübingen

P tid b d i ti Peptide-based vaccination

• f patients with prostate carcinoma

Department of Urology

• design and schedule -
• Phase I/II randomized study, start 2004
• 40 HLA-A2+ patients with biochemical relapse after radical prostatectomy

40 HLA A2 patients with biochemical relapse after radical prostatectomy

• 14 peptides, 300 µg / peptide in 500 µl Montanide ISA51

for PSA measurement for immunomonitoring ( d l h t ) Blood samples

Peptide cocktail s.c. in Montanide Peptide cocktail s.c. in Montanide GM-CSF Imiquimod Hyperthermia

(serum and lymphocytes)

in Montanide in Montanide Stabilized RNA w/o

vaccination timepoints vaccination timepoints

monthly up to 15 vacc

• 1. 2. 3. 4. 5. 6.
• 7. 8. 9.

1week

• 1. 2. 3. 4. 5. 6.
• 7. 8. 9.

1week

up to 15 vacc.

blood samples blood samples

Peptides in the vaccination cocktail

Protein Position Sequenz HLA-Restriction PSA 141-150 FLTPKKLQCV HLA-A*0201 146-154 KLQCVDLHV HLA-A*0201 154-163 VISNDVCAQV HLA-A*0201 PSCA 14-22 ALQPGTALL HLA-A*0201 105-113 AILALLPAL HLA-A*0201 PSMA 4-12 LLHETDSAV HLA-A*0201 711-719 ALFDIESKV HLA-A*0201 Survivin 95-104 ELTLGEFLKL HLA-A*0201 5-14 TLPPAWQPFL HLA-A*0201 5 14 TLPPAWQPFL HLA A 0201 TRP-P8 187-195 GLMKYIGEV HLA-A*0201 Prostein 31-39 CLAAGITYV HLA-A*0201 PSMA 459-473 NYTLRVDCTPLMYSL DRB1*0301,0401,0701,1101 Survivin 97-111 TLGEFLKLDRERAKN DR (Vorhersage)

53

Influenza MP 58-66 GILGFVFTL HLA-A*0201

clinical observations PSA time to next therapy PSA, time to next therapy

2 mo No response 0,0,0,0 time to next therapy

(days after 1. vaccination)

390 Biometric analysis by Prof. K. Dietz

3 imi Response 1,1,0,0 time to next therapy

(days after 1. vaccination)

>2908

5 mo transient response 1,0,0,0 time to next therapy

(days after 1. vaccination)

1117

100 120 60 80 100 NR% R trans% R f ll%

PSA values

20 40 R full%

values %

imi RNA mo GM Hyp

time to next therapy or to end of observation next therapy no other therapy

Clinical parameters TTNT vs PSA Clinical parameters TTNT vs PSA

3000 2000 3000

erapy or

• n

1000 2000

• next the

bservatio

1000

days to

• b

NR tR R

Anti HLA class II T cells Anti‐HLA class II T cells

2 I peptides 1 ed HLA class I b of recognize Nb

V i 11 J 2013 Version 11 June 2013 Pro22 excluded, Pro3 and 4 n=1, Pro38 n.i., Pro34 n=1

Anti HLA class I T cells Anti‐HLA class I T cells

5 6 7 gen-derived es 3 4 5 ed tumor antig class I peptide 1 2 b of recognize HLA c Nb

V i 11 J 2013 Version 11 June 2013 Pro22 excluded, Pro7,24 n.i.

Immunogenicity of vaccine peptides

Version 11 June 2013

Immunogenicity of vaccine peptides

120 80 100

(ELISPOT)

40 60

ing patients

20

% reacti

Survivin (II) PSMA (II) Prostein PSMA 711 TRP‐P8 Survivin 95 PSA 146 PSCA 105 PSA 141 Survivin 5 PSCA 14 PSA 154 PSMA 4 32 25 23 22 14 9 9 7 4 34 35 34 34 34 34 34 34 34 34 34 34 34 91,4 74,3 68 65 41 26 26 20 12

Conclusions

Multipeptide immunotherapy in prostate carcinoma patients

• is well tolerated
• induces T cell responses in most patients
• shows clinical benefit in a fraction of patients
• gives best results with TLR7 ligands as adjuvants

some peptides are not immunogenic

• some peptides are not immunogenic

Widenmeyer et al., Promiscuous survivin peptide induces robust CD4(+) T-cell responses in the majority of vaccinated cancer patients. Int J Cancer. 2011 Feyerabend et al., Novel multi-peptide vaccination in Hla-A2+ hormone sensitive patients with biochemical relapse of prostate cancer. Prostate. 2009

Why mutated antigens should be the better targets targets

Limited to few cancer entities We know plenty of such antigens; immune responses tend to be weak but stronger immune responses may lead This is our vision but stronger immune responses may lead to autoimmunity

• 1. T cells having left the thymus before tumor
• ccurence should not be tolerant of tumor
• ccurence should not be tolerant of tumor

specific mutations

N C i i d i d ll i h li l ll (B Note: Cancer testis antigens are expressed in medullary epithelial cells (Bruno Kyewski et al.)

indeed, patients develop efficient T cell responses against mutations (Thomas Wölfel)

• 2. T cells specific for tumor specific mutations

should not react to self peptides should not react to self peptides, since those with crossreactivity to non- mutated peptides should have undergone negative thymic selection. negative thymic selection.

i i d > no toxicity expected

RCC-Patient

vaccination

Kidney Blood

Tumor Normal

T-cells

Our new strategy for the

mRNA, DNA Peptides

Immuno A

gy identification of tumor

Genome and t i t

MS Assays

specific peptides

transcriptome sequence

HLA-Ligands

Specific Recognition Recognition candidate peptides for immunotherapy

CCC IND‐01, Genome sequencing and prediction of mutated HLA‐ligands by SYFPEITHI

KCNJ12 239E/K QLIKPRVTK, HLA-A*03, score 33 HGC6.3 128M/V VVTPTASSF, HLA-A*03, score 25

(Bold: anchor; underlined: auxillary anchor AA)

CCC IND‐01, Genome sequencing and prediction of mutated HLA‐ligands by SYFPEITHI

88 unique mutated peptide sequences in tumor tissue fitting to the patient's HLA Type: tissue fitting to the patient s HLA‐Type:

• 40 fit to HLA*A03

40 fit to HLA A03

• 26 fit to HLA*B14
• 22 fit to HLA*B44

67 stem from SNV and 21 from InDel Mutations

Search for mutated peptides by Search for mutated peptides by mass spectrometry

Bsp 1:

• Bsp. 1:
• In Gen: sphingosine‐1‐phosphate phosphatase

(SGPP1) ( )

• WT:

MVGLSITF M i MVGFSITF

• Mutiert: MVGFSITF
• Position: 425

Nico Trautwein

MVGFSITF Tumor MVGFSITF Tumor

MVGFSITF synpep MVGFSITF synpep

Bsp 2:

• Bsp. 2:
• Gen: Glutamate receptor, ionotropic (GRIN1)
• WT:

FYRIPVLGL WT: FYRIPVLGL

• Mutiert: FYCIPVLGL
• Position: 115

FYCIPVLGL Tumor FYCIPVLGL Tumor

FYCIPVLGL synpep FYCIPVLGL synpep

RCC792 Tumor W6/32 Rep#1/ GVPIMLVY (B*15, A*03) RNF19B%NM_001127361@A429V Daniel Kowalewski

RCC792 Tumor W6/32 Rep#1/ GVPIMLVY (B*15, A*03) RNF19B%NM_001127361@A429V

726.5191

b 250 Extracted from: E:\RCC792 Exhaustion & Spectral Counting\Tumor W\120608_ARDK_RCC792_Tumor_W_10%_1sDynExl_Rep#1_msms1.RAW #3522 RT: 46.38 ITMS, CID, z=+2, Mono m/z=454.25336 Da, MH+=907.49944 Da, Match Tol.=0.8 Da 281.2502

y 566.5985 314 2938

b 418 3171 254 2081

b 354.8043

b 363.5275

b 182.0476

y 627.1843

b 50 100 150 200 Intensity [counts] 543.6866 314.2938 418.3171 254.2081 200 300 400 500 600 700 800 900 m/z 50

Sequence: GVPIMLVY M5‐Oxidation (15.99492 Da) Charge: +2, Monoisotopic m/z: 454.25336 Da (+1.82 mmu/+4.01 ppm), MH+: 907.49944 Da, RT: 46.38 min, Identified with: Mascot (v1.15); IonScore:26, Exp Value:2.1E‐002, Matched Ions: 4/74 Fragment Match Tolerance: 0.8 Da Protein References (2): gnl|58|do rcc792 cegat long|RNF19B%NM 153341@A430V| ‐ gnl|58|do:rcc792_cegat_long|RNF19B%NM_153341@A430V| ‐ gnl|59|do:rcc792_cegat_long|RNF19B%NM_001127361@A429V|

RCC792 Tumor W6/32 Rep#1/ GVPIMLVY (B*15, A*03)

726.5191

b 250 Extracted from: E:\RCC792 Exhaustion & Spectral Counting\Tumor W\120608_ARDK_RCC792_Tumor_W_10%_1sDynExl_Rep#1_msms1.RAW #3522 RT: 46.38 ITMS, CID, z=+2, Mono m/z=454.25336 Da, MH+=907.49944 Da, Match Tol.=0.8 Da

Tumor Sample

281.2502

y 354.8043

b

b

y 627.1843

b 100 150 200 Intensity [counts] 566.5985 543.6866 314.2938

b 418.3171 254.2081

b 363.5275

b 182.0476

y 200 300 400 500 600 700 800 900 m/z 50 727.5488

b 12 14 16 Extracted from: F:\120711_DK_SynPep_120224_IVAC792B15_labeled_5pmol_msms2.RAW #5702 RT: 139.79 ITMS, CID, z=+2, Mono m/z=454.74988 Da, MH+=908.49248 Da, Match Tol.=0.8 Da

SynPep (Val7‐ 15N labeled, 5 pmol

b 182 0665

y 216.3578 627.3995

b 269.2565 322.2579 420.0762 396.0079

y 355.3776

b 4 6 8 10 12 Intensity [counts] 513.9355

b 663.6210 498.9489 182.0665 200 300 400 500 600 700 800 900 m/z 2

Zoom‐In

354.8043

b

b 182 0476

y 60 70 Extracted from: E:\RCC792 Exhaustion & Spectral Counting\Tumor W\120608_ARDK_RCC792_Tumor_W_10%_1sDynExl_Rep#1_msms1.RAW #3522 RT: 46.38 ITMS, CID, z=+2, Mono m/z=454.25336 Da, MH+=907.49944 Da, Match Tol.=0.8 Da

Tumor Sample

246 3437 382.5796 408.2029 427.0581 281.2502

y 336.3707 165.1427

y 314.2938

b 418.3171 254.2081

b 363.5275

b 182.0476 20 30 40 50 Intensity [counts] 246.3437 382.5796 200 300 400 500 m/z 10 Extracted from: F:\120713 DK SynPep 120224 IVAC792B15 labeled 10pmol msms1 RAW #6699 RT: 139 61 182.1375

y 80 100 ts] Extracted from: F:\120713_DK_SynPep_120224_IVAC792B15_labeled_10pmol_msms1.RAW #6699 RT: 139.61 ITMS, CID, z=+2, Mono m/z=454.74976 Da, MH+=908.49224 Da, Match Tol.=0.8 Da 227.0332 410.9032 136.0168 272.2562

y 313.9926

b 286.2734 350.4114

b 376.9111

• 1+

y 395.2849

y 165.0696

y 364.3462

b 20 40 60 Intensity [coun

SynPep (Val7‐ 15N labeled, 5 pmol

200 300 400 500 m/z

New:

Two step strategy for individualized p gy immunotherapy:

1st step, as fast as possible (eg., right after surgery): vaccination with off the shelf peptides according to HLA-expression 2nd step, after tumor mutation analysis: vaccination with mutated peptides mutated peptides

Our first example: RCC 792 (HLA-A2,A3,B15,B27) p ( )

A2 Nr. Sequenz Quelle HLA Note 51005 STAPPVHNV MUC-1 A*02 RCC ligand 51004 SVASTITGV ADFP 1 A*02 RCC li d

RCC 792, Two step individualized vaccination

51004 SVASTITGV ADFP-1 A*02 RCC ligand 51007 YVDPVITSI MET-1 A*02 RCC ligand 51001 LAALPHSCL RGS-2 A*02 RCC ligand 50060 FLGENISNFL APOL 1 A*02 RCC ligand

HLA-A2 cocktail for RCC

50060 FLGENISNFL APOL-1 A 02 RCC ligand 51003 ALADGVQKV APOL-2 A*02 RCC ligand 51006 ALFDGDPHL KIAA0367-1 A*02 RCC ligand 51002 SVFAGVVGV GUCY1A31 A*02 RCC ligand

RCC

g 50061 LLYPTEITV ITGA3-1 A*02 RCC ligand 51009 LLGATCMFV CCND1-1 A*02 RCC ligand 51008 FLPSDFFPSV HBV-1 A*02 pos

Vaccination started April 2012, Med. Klinik II Vaccination started April 2012, Med. Klinik II

50076 GILGFVFTL IMP-1 A*02 pos 50089 NPPSMVAAGSVVAAV 2CCND1-1 DR helper 50091 HSKIIIIKKGHAKDSQ 2IBP3-1 DR helper

Klinik II Klinik II

51018 SQDDIKGIQKLYGKRS 2MMP7-1 DR helper 50097 EIHVVHLSTAFARVDEALGR 2G250-1 DR helper

HLA-A3 cocktail for RCC

Mutation analysis of RCC792

#gene transcript allele ref_pep ref_score mut_pep mut_score binder mutation ZNF799 NM_001080821 B_2705_9 CGKAFIDFY 11 CRKAFIDFY 21 True G454R C2orf28 NM_080592 A_0301_9 TQCPGSVQN 12 TLCPGSVQN 22 True Q92L C2orf28 NM 001170795 A 0301 9 TQCPGSVQN 12 TLCPGSVQN 22 True Q37L _ _ _ SNX27 NM_030918 A_0201_9 SAVLPGGAA 12 SAVLPGGAV 18 True A90V FAM55B NM_182495 A_0301_9 KNINDCLER 14 KIINDCLER 22 True N351I SAMM50 NM_015380 A_0201_9 IVLRLGNIA 14 IILRLGNIA 18 True V431I ZNF799 NM_001080821 B_2705_9 KCGKAFIDF 15 KRGKAFIDF 25 True C453R C2orf28 NM 080592 B 1501 9 ALPEICTQC 15 ALPEICTLC 14 True Q92L C2orf28 NM_080592 B_1501_9 ALPEICTQC 15 ALPEICTLC 14 True Q92L C2orf28 NM_001170795 B_1501_9 ALPEICTQC 15 ALPEICTLC 14 True Q37L C2orf28 NM_080592 A_0201_9 LALPEICTQ 16 LALPEICTL 26 True Q92L C2orf28 NM_001170795 A_0201_9 LALPEICTQ 16 LALPEICTL 26 True Q37L C2orf28 NM_080592 A_0201_9 QCPGSVQNL 16 LCPGSVQNL 18 True Q92L C2orf28 NM_001170795 A_0201_9 QCPGSVQNL 16 LCPGSVQNL 18 True Q37L JPH2 NM_020433 A_0201_9 SVGSQRSRV 16 SVGGQRSRV 18 True S241G ZNF799 NM_001080821 B_2705_9 KPYKCKCGK 17 KPYKCKCRK 19 True G454R SAMM50 NM_015380 B_2705_9 YGAGIVLRL 17 YGAGIILRL 19 True V431I PLA2G3 NM 015715 B 2705 9 VALGGSPAL 18 VALGGSPSL 19 True A23S PLA2G3 NM_015715 B_2705_9 VALGGSPAL 18 VALGGSPSL 19 True A23S CALHM2 NM_015916 A_0201_9 QLFGWLLIG 18 QLFEWLLIG 18 True G188E AP2A2 NM_012305 A_0201_9 NLVKVGGYI 18 NLVKVGDYI 18 True G500D AP2A2 NM_012305 A_0201_9 LVKVGGYIL 18 LVKVGDYIL 18 True G500D FLNA NM_001456 A_0201_9 ARRLTVSSL 18 ARCLTVSSL 18 True R2326C FLNA NM 001110556 A 0201 9 ARRLTVSSL 18 ARCLTVSSL 18 T R2334C FLNA NM_001110556 A_0201_9 ARRLTVSSL 18 ARCLTVSSL 18 True R2334C TLR8 NM_138636 A_0201_9 YTLTDKYNL 19 YTLTDKYKL 21 True N604K SHH NM_000193 B_2705_9 RLLLTAAHL 19 RMLLTAAHL 19 True L264M OR4K15 NM_001005486 A_0201_9 TLRNKEVKA 19 MLRNKEVKA 19 True T312M KIAA1211 NM 020722 A 0201 9 SAAKHKLAV 20 SAANHKLAV 20 True K158N _ _ _ SETD2 NM_014159 A_0201_9 KELDSLSKV 20 NELDSLSKV 19 True K650N SAMM50 NM_015380 A_0201_9 GIVLRLGNI 21 GIILRLGNI 23 True V431I SAMM50 NM_015380 A_0301_9 VLRLGNIAR 21 ILRLGNIAR 23 True V431I SHH NM_000193 B_2705_9 TREPRERLL 21 TREPRERML 21 True L264M SAMM50 NM 015380 A 0201 9 YGAGIVLRL 21 YGAGIILRL 21 True V431I

IND Nr Sequenz Quelle Position HLA Note Nr. Sequenz Quelle Position HLA Note 121148 AQNSRIQQL ANGL4 111-119 A*02:01/B*15 ligand of 792 121172 GVPIMLVY RN19B 424-431 B*15 mutation 121164 RLDNSAANH KIA 1211 151-159 A*03:01 mutation 121150 ARCLTVSSL FLNA 2332-2340 B*27 mutation

Vaccination started

• Sept. 2012, Med.

Klinik II (off the shelf peptides Vaccination started

• Sept. 2012, Med.

Klinik II (off the shelf peptides

( p p continued) ( p p continued)

RCC792: T-Zellmonitoring after 12day stimulation (16.01.13) with IND-peptide cocktail

IFN- Elispot 250.000 cells/well

10% DMSO 121150 ARCLTVSSL 121172 GVPIMLVY 121197 RLDNSAANH 121148 AQNSRIQQL PHA 02.07.12 11.10.12 28.12.12 120470 ARRLTVSSL 121290 GVPIMLAY 120471 RLDNSAAKH 02.07.12 11.10.12

02 07 2012 8 weeks before 1 V with IND‐Cocktail 02.07.2012 8 weeks before 1.V with IND‐Cocktail 11.10.2012 8 weeks after 1.V mit IND‐Cocktail (5V IND Cocktail) 28.12.2012 4 months after 1.V mit IND‐Cocktail (8V IND‐Cocktail)

Karoline Laske

RCC 792 Two step RCC 792, Two step individualized vaccination

• 1. Mutated peptides could not be identified by mass

spectrometry

• 2. Patient RCC 792 appeared to be nonresponder against all

class I peptides tried

But we still try, and we will try harder y, y

Interfakultäres Institut für Zellbiologie Abteilung Immunologie

• Prof. Dr. Hans‐Georg Rammensee

Collaboration B b S h ib l b d R /St i l b Bob Schreiber lab and Rammensee/Stevanovic lab Heiko Schuster

Courtesy Bob Schreiber

Expansion and IFNγ treatment of d42mt1 progressor cell line T3

1) Expansion to 500 Mio cells 1) Expansion to 500 Mio cells 2) IFNγ treatment to increase MHC expression MCA Sarcoma cell line d42mt1 ‐T3 (R913L‐Spectrin‐β2 negative (R913L Spectrin β2 negative = “progressor clone”)

Isolation of H2Kb presented peptides and subsequent LC‐MS analysis

H2Kb Affinity chromatography (specific antibody Y3) Lysis with detergent Acidic elution and

• In total 224 different

Lysis with detergent Acidic elution and ultrafiltration LC‐MS

• a

d e e H2Kb presented peptides identified (FDR < 0 05) (FDR < 0.05)

• Identification of

mutated tumor rejection antigen Lama4 VGFNFRTL

First time identification and successful validation of a mutated MHC ligand validation of a mutated MHC ligand

VGFNFRTL mLama4(G1254V) V lid i i h Validation with an isotope labeled synthetic peptide VGFNFRTL VGFNFRTL (13C6, 15N1)

Problem: Patient individualized peptides have to be produced according to GMP conditions, like all other drugs.

GMP center for individualized substances

Bau: Finanziert durch MWK und UKT Betrieb: Anfinanziert durch MWK und Med. Fak. Betrieb: Anfinanziert durch MWK und Med. Fak.

Erteilung des Zertifikats für Wi k t ff tidh t ll 7 Mä 2012 Wirkstoffpeptidherstellung 7. März 2012 Antrag auf Arzneimittelherstellung aus diesen Antrag auf Arzneimittelherstellung aus diesen Wirkstoffen Mai 2012 Dritter Mängelbeseitigungsbericht geht am 19.7. 2013 ans Regierungspräsidium 2013 ans Regierungspräsidium Inspektion der (neuen) GMP-Räume am 23. und p ( )

• 24. 9. 2013.

Problem: What will the regulatory authorities say to an individualized study design?

Levels of personalization

Concept by CIMT

81

(A) (B) (C)

G).

Patient

search Group (RRG

? ? ? ?

Tumor

MT) Regulatory Res

passive personalization active personalization Stratification

Thera- nostic

mmunotherapy (CIM

(„AP“)

nostic

ciation for Cancer Im

Drug Product(s)

Slide by Assoc

Confidential Confidential

Invariant DP Variant DPs Variant DPs

THE REGULATORY LANDSCAPE FOR ACTIVELY PERSONALIZED CANCER IMMUNOTHERAPIES IMMUNOTHERAPIES Cedrik M. Britten*1,2, Harpreet Singh-Jasuja*3, Bruno Flamion4, Axel Hoos5, Christoph Huber6, Karl- Josef Kallen7, Samir N. Khleif8, Sebastian Kreiter1, Michaela Nielsen9, Hans-Georg Rammensee10, U S hi

1 11 Th

Hi

#12

d Ul i h K li k

#13

Ugur Sahin1,11, Thomas Hinz#12, and Ulrich Kalinke#13

• n behalf of the Association of Cancer Immunotherapy (CIMT) Regulatory Research Group (RRG)

ABSTRACT Tumors carry multiple somatic mutations, the majority of which are unique to individual patients Recent data imply that immunogenic tumor mutations can be individual patients. Recent data imply that immunogenic tumor mutations can be exploited for the treatment of cancer patients. Here we propose a development strategy for actively personalized vaccines (APVACs) targeting multiple tumor

• mutations. This strategy is based on the existing regulatory framework thus

facilitating the way towards first clinical testing.

N t Bi t h l 2013 i Nature Biotechnology 2013, in press

Glioma Actively Personalized Vaccine Consortium

83

GAPVAC

• Establish an actively personalized vaccinaton (APVAC) approach for

treatment of glioblastoma patients C ti ith

• Consortium with

14 partners funded by EU with 6 mn EUR

• Led by Immatics

(Coordinator) and BioNTech (Vice ( Coordinator)

• Clinical study planned

t t t i 2014 to start in 2014

• Up to 30 glioblastoma

patients will receive APVACs patients will receive APVACs composed of warehouse-selected and mutanome-derived peptides.

84 84

Tumor-associated peptides – shared vs. individual

Tumor 1 Tumor 2 SHAILEALA TQMPDPKTF HVNDLFLQY TUMAPs potentially suitable for personalized therapy ALRDVRQQY HQITVLHVY GLATDVQTV KLHGVNINV LEEDSAREI SHAILEALA FAEGFVRAL HVIDVKFLY GQFPGHNEF GLNDETYGY LEEDSAREI LLAERDLYL IAMATVTAL LLAERDLYL QEQSFVIRA RLASYLDKV MEDIKILIA MQKEITAL Off-the-shelf Multi-TUMAP Vaccine Tumor 3 MQKEITAL Tumor 3

Glioma Actively Personalized Vaccine Consortium

85

GAPVAC design

To do for each patient: To do for each patient:

• 1. start vaccination with off-the-shelf peptides
• 2. select 5 - 15 peptides representing mutated AND

patient-specific wild type HLA class I and class II ligands synthesize and formulate to a cocktail start ligands, synthesize and formulate to a cocktail, start vaccination

Interfakultäres Institut für Zellbiologie Abteilung Immunologie

• Prof. Dr. Hans‐Georg Rammensee

87

Co-workers and Collaborators

IMA901 Chief Investigators US: Brian Rini, Cleveland EU: Tim Eisen, London GER: Arnulf Stenzl, Tübingen Pfizer Inc. COIN Study Group University of Tübingen Hans-Georg Rammensee Stefan Stevanovic Discovery Toni Weinschenk Oliver Schoor IMA910 Clinical Team Andrea Mayer-Mokler Jörg Ludwig Cécile Gouttefangeas Arnulf Stenzl Susan Feyerabend Jörg Hennenlotter Jens Bedke Norbert Hilf Jens Fritzsche Andrea Mahr Nina Pawlowski IMA910 Chief Investigator Frank Mayer, Tübingen IMA950 Project Leaders Oliver Schoor (CR-UK study) BioNTech Ugur Sahin Cedrik Britten John Castle Sandra Heesch Jens Bedke Graham Pawelec Evelyna Derhovanessian University of Verona Vincenzo Bronte Immunology Steffen Walter Dominik Maurer Sabrina Kuttruff R i M d k Oliver Schoor (CR-UK study) Norbert Hilf (NCI study) IMA950 Chief Investigators Roy Rampling, Glasgow (CR-UK study) H d Fi d J h S l Sandra Heesch Sebastian Kreiter Vincenzo Bronte University of Padova Susanna Mandruzzato Regina Mendrzyk Verona Vass CMC Peter Lewandrowski Howard Fine and Joohee Sul, Bethesda (NCI study) SAB & DSMB Members Hans-Georg Rammensee Cancer Research UK James Ritchie Lesley McGuigan University of Geneva Pierre-Yves Dietrich Valérie Dutoit University of Heidelberg Christian Flohr Werner Stüber IMA901 Clinical Team Alexandra Kirner g Cornelius Melief Christoph Huber Pedro Romero Craig Slingluff Jr. Christian Ottensmeier y g Sarah Halford y g Christel Herold-Mende Philipp Beckhove Wolfgang Wick Alexandra Kirner Maud LePriellec Annette Schmid Christian Ottensmeier Hakan Melsted Chief Medical Officer Carsten Reinhardt Chief Scientific Officer Harpreet Singh Immunomonitoring Proficiency Panel & Regulatory Research Group

Acknowledgements

• Dept. Immunology

University Hospital

Molecular immunology Stefan Stevanovic Urology Arnulf Stenzl Immunomonitoring Cécile Gouttefangeas Surgery Alfred Königsrainer Mathias Walzer Armin Rabsteyn Daniel Kowalewski Jens Bedke Susan Feyerabend Jörg Hennenlotter Karoline Laske Annemarie Dröge Melanie Widenmayer Stefan Löb Markus Löffler Philipp Horvath Daniel Kowalewski Heiko Schuster Janet Peper Jörg Hennenlotter Pathology Melanie Widenmayer Heinrich Griesemann

Bioinformatics

Philipp Horvath Internal Medicine II Nico Trautwein Christina Kyzirakos Falko Fend H G ti

Bioinformatics

Oliver Kohlbacher Mathias Walzer Lothar Kanz Susanne Rittig Christoph Grabenbauer Thomas Feger Christian Hotz Human Genetics Olaf Riess Peter Bauer

immatics

Harpreet Singh Helmut Salih Sebastian Haen Julia Stickel Christian Hotz Lea Prokop Stefanie Souczek Michael Bonin Christopher Schröder Harpreet Singh Toni Weinschenk Steffen Walter Julia Stickel Claudia Berlin Rita Pfeifer

Support: SFB 685, GK 794, BMBF, EU, Krebshilfe, fortüne, AKF