Tumor mutation burden as predictive biomarker …...Correlation Between TMB and Neoantigen Load TMB=tumor mutation burden. 1. Hellmann M. Oral presentation at AACR 2017. 2. Van Allen

Tumor mutation burden as predictivebiomarker –

from discovery to standardization

Nicola Normanno

ISTITUTO NAZIONALE PER LO STUDIO E LA CURA DEI TUMORI FONDAZIONE G. Pascale – NAPOLI

SC Biologia Cellulare e Bioterapie

CENTRO RICERCHE ONCOLOGICHEMERCOGLIANO (AV)

Laboratorio di Farmacogenomica

Somatic mutation frequencies in cancer

Lawrence Nature 2013

Tumor Mutation Burden in NSCLC

Spigel ASCO 2016

Adeno(n=7,925)

SCC(n=1,324)

NSCLC NOS

(n=1,773)

SCLC(n=640)

EGFR mutation

(n=1,775)

ALK or ROS1 fusion(n=489)

METex14 alteration

(n=286)

BRAF mutation

(n=493)

KRAS mutation

(n=3,155)

MeanMutations/Mb

9.1 11.3 11.0 10.3 4.5 3.1 6.2 9.7 10.3

TMB >10 (%)2350 (30)

541 (41)

711 (40)

269 (42)

129 (7)

17 (3)

27 (9)

153 (31)

1,238 (39)

TMB >20 (%)760 (10)

113 (9)

233 (13)

42 (7)

21 (1)

4 (1)

4 (1)

51 (10)

298 (9)

Wilcoxon signed-rank test vs KRAS

p<0.001 p<0.001 p<0.001 p<0.001

Estimate of the neoantigen repertoire in human cancer

Schumacher Science 2015

Classes of human tumour antigens that are recognized by T lymphocytes

Coulie Nat Rev Cancer 2014

Candidate neoantigens and response to pembrolizumab

Rizvi Science 2015

Correlation Between TMB and NeoantigenLoad

TMB=tumor mutation burden.1. Hellmann M. Oral presentation at AACR 2017. 2. Van Allen EM et al. Science. 2015;350(6257):207-211. 3. Rooney MS et al. Cell. 2015;160:48-61.

• Higher levels of TMB have been shown to correlate with higher neoantigen load1-3

‒ Melanoma, lung, bladder, liver, and stomach cancers are predicted to have the highest neoantigen loads3

Pre

dic

ted

Ne

oan

tige

ns

Pan-Tumor1

Total Mutations

0

500

1000

0 1000 2000 3000 4000 5000

R2=0.91Spearman rho 0.92

Correlation between TMB and Response Rate with Anti–PD-1 or Anti–PDL1 Therapy in 27

Tumor Types

Yarchoan NEJM 2018

Whole Exome Sequencing Is a Technical and Complex Process

1. Frampton GM et al. Nat Biotechnol. 2013;31(11):1023-1031. 2. Ulahannan D. Br J Cancer. 2013; 109(4):827-835.

DNA Extraction

Clinical Report

Data Analysis

ACTGTGCGACGA

CGCGTAGATCGC

AACGTCGCGATA

AlignmentDNA

SequencingLibrary

PreparationHybrid Capture

• Sample attrition may occur within the process, and technical complexity may require specialty labs2

• Use of consolidated gene panels may alleviate some of the challenges associated with whole-exome sequencing2

NGS-based genomic profiling workflow1

Assessment of TMB with targetedsequencing

Chalmers Genom Med 2017

FoundationOne®panel 315 genes

TMB assessed by targeted NGS correlates with WES and durable clinical benefit (DCB)

Rizvi JCO 2018

MSK-IMPACT panel 468 genes

CheckMate 227: Nivo + Ipi in 1L NSCLC With High TMB (≥10 mut/Mb)

CheckMate 227 Part 1 Study Designa

Database lock: January 24, 2018; minimum follow-up: 11.2 months

N = 1189

<1% PD-L1expression

N = 550

Nivolumab 3 mg/kg Q2W Ipilimumab 1 mg/kg Q6W

n = 396

Histology-based chemotherapyb

n = 397

Nivolumab 240 mg Q2Wn = 396

Nivolumab 3 mg/kg Q2W Ipilimumab 1 mg/kg Q6W

n = 187

Histology-based chemotherapyb

n = 186

Nivolumab 360 mg Q3W + histology-based chemotherapyb

n = 177

R

1:1:1

Key Eligibility Criteria• Stage IV or recurrent NSCLC

• No prior systemic therapy

• No known sensitizing

EGFR/ALK alterations

• ECOG PS 0–1

Stratified by SQ vs NSQ

R

1:1:1

7

aNCT02477826 bNSQ: pemetrexed + cisplatin or carboplatin, Q3W for ≤4 cycles, with optional pemetrexed maintenance following chemotherapy or nivolumab + pemetrexed maintenance following nivolumab + chemotherapy; SQ: gemcitabine + cisplatin, or gemcitabine + carboplatin, Q3W for ≤4 cycles; cThe TMB co-primary analysis was conducted in the subset of patients randomized to nivolumab + ipilimumab or chemotherapy who had evaluable TMB ≥10 mut/Mb

≥1% PD-L1expression

Nivolumab + ipilimumab n = 396

Chemotherapyb

n = 397

Patients for PD-L1 co-primary analysis

Co-primary endpoints: Nivolumab +

ipilimumab vs chemotherapy

• OS in PD-L1–selected populations

• PFS in TMB-selected populations

Nivolumab + ipilimumab n = 139

Chemotherapyb

n = 160

Patients for TMB co-primary analysisc

CheckMate 568: Identification of Optimal TMB Cutoff (≥10 mut/Mb) for Nivo + Ipi in 1L NSCLC

TMB Analysis With FoundationOne CDx™

Assay

14FoundationOne CDx™ Technical Information. Cambridge, MA: Foundation Medicine Inc; 2018.

FoundationOne CDxTM (F1CDx)

FoundationOne CDxTM TMB result

Somatic mutations

per megabase

FoundationOne

CDxTM

Tumor sample

(FFPE)

• FoundationOne CDx™ uses next-generation

sequencing to detect substitutions, insertions and

deletions, and copy number alterations in 324 genes

and select gene rearrangements

– TMB: total number of synonymous and non-

synonymous variants (≥5% allele frequency)

after filtering germline mutations

CheckMate 568: Identification of Optimal TMB Cutoff (≥10 mut/Mb) for Nivo + Ipi in 1L NSCLC

ORR by TMBa,b

15

0

10

20

30

40

50

OR

R (

%)

n/N

TMB (mut/Mb)

<5c

2/23

9

≥5 to <10d

4/27

15

≥10e

21/48

44

≥15f

11/28

39

aIrrespective of PD-L1 expression; b12% ORR for <10 mut/Mb and 50% ORR for ≥10 to <15 mut/Mb; cCR = 0; dCR = 4%; eCR = 8%; fCR = 7%

CheckMate 568: Identification of Optimal TMB Cutoff (≥10 mut/Mb) for Nivo + Ipi in 1L NSCLC

ROC Curves by Objective Response

16

Tru

e-p

os

itiv

e f

rac

tio

n

False-positive fraction

AUC = 0.70

TMB (n = 98)

AUC = 0.73

9-10 mut/Mb

False-positive fraction

PD-L1 (n = 252)

0.00

0.25

0.50

0.75

0.00 0.25 0.50 0.75 1.00

1.00

0.00

0.25

0.50

0.75

0.00 0.25 0.50 0.75 1.00

1.00

Tru

e-p

os

itiv

e f

rac

tio

n

CheckMate 227: Nivo + Ipi in 1L NSCLC With High TMB (≥10 mut/Mb)

Co-primary Endpoint: PFS With Nivolumab + Ipilimumab vs Chemotherapy in Patients With High TMB (≥10 mut/Mb)a

17

Nivo + ipi 139 85 66 55 36 24 11 3 0

Chemo 160 103 51 17 7 6 4 0 0

Nivo + ipi

(n = 139)

Chemo

(n = 160)

Median PFS,b mo 7.2 5.4

HRc

97.5% CI

0.58

0.41, 0.81

P = 0.0002

Months

0

20

40

60

80

100

0 6 12 183 9 15 21 24

PF

S (

%)

Chemotherapy

Nivolumab +ipilimumab

1-y PFS = 43%

1-y PFS = 13%

aPer blinded independent central review (BICR); median (range) of follow-up in the co-primary analysis population was 13.6 mo (0.4, 25.1) for nivo + ipi and 13.2 mo (0.2, 26.0) for chemo;b95% CI: nivo + ipi (5.5, 13.2 mo), chemo (4.4, 5.8 mo); c95% CI: 0.43, 0.77 mo; dThe P-value for the treatment interaction was 0.0018

No. at risk

• In patients with TMB <10 mut/Mb treated with nivo + ipi vs chemo, the HR was 1.07 (95% CI: 0.84, 1.35)d

Blood-based TMB

Gandara Nat Med 2018

TMB testing: Critical issues

TMB is emerging as a relevant biomarker for response to immunotherapy. Standardization and harmonization of TMB testing are pivotal. Different open questions need to be answered:

1. Current methods for TMB measurement are not standardized and vary from WES to targeted sequencing panels (commercial kits, laboratory-developedassays…)

2. As a consequence, bioinformatic pipelines for TMB calculation differ amongthe different assays

3. No univocal cut-offs to identify low-medium-high TMB tumors exist (the onlyvalidated cut-off is ≥10mut/mb with FoundationOne®CDX, Checkmate-227)

FoundationOne CDx™ 324 genes - Threshold10 mut/Mb

FoundationOne 324 genes value 10

WES value ⁓200Mutations/Mb

Thermo Fisher 409 genes value 12

Qiagen⁓400 genes value 12

Illumina ⁓500 genes value 14

Aims of the study:a) to organize a pilot European EQA scheme for TMB;b) to harmonize the different TMB methods;c) to define recommendations for TMB assessment.

Academia: ESMO, AIOM, Gen&Tiss, ESP, UKNEQAS, EMQN, cIQc, RCPA QualityAssurance Programs

Sponsors: BMS, AstraZeneca, Roche/Foundation Medicine, Thermo Fisher Scientific, Illumina, Qiagen, Genentech

Tumour mutation burden:from recommendations for testing to external quality assessment schemes

Project Kick-off MeetingNaples, 10th May 2018

Participants:

ESMO BMSAIOM Thermo FisherGen&Tiss IlluminaESP QiagenEMQN GenentechcIQc Merck KGaA

Format of pilot EQA

Selection of participants

(survey)

EQA material validation study

EQA distribution

Results submission

- Reports- Data collection

EQA assessment

Release of EQA results and

Scheme Report

EQA sample validation• FFPE cell lines with different mutational load (high, intermediate, low TMB)• Different validating laboratories will be identified • Perform sample validation (DNA extraction + TMB testing)• Different TMB panels to test the samples (Thermo Fisher – Qiagen – Illumina -

FoundationOne)

EQA sample distribution• Identification of 30 laboratories with a survey

• taking into account labs with molecular pathology expertise, already performing TMB testing, with different methods

Analysis of EQA results/Issue of results • Specific scoring system• TMB result and clinical interpretation

Project Planninga) EQA Pilot scheme for TMB testing

a) Creation of an harmonization networka) Identification of a panel of experts to provide FFPE material from NSCLC

patients. b) Identify the appropriate number of centers to be involved

b) Sample selection and analysisa) Collection of samples in order to obtain a significant number of analyses

to allow the creation of a conversion table

c) Comparative analysisa) Performance of comparative analysis of the scores resulting from different

TMB tests to create a reference table that maps the score of one text to another, to enable harmonisation.

d) Publication of the results/Workshops

Project Planningb) Harmonization phase

Project Timeline

TMB project

Timeline

Jun/

Jul

2018

Aug/

Sept

2018

Oct/

Nov

2018

Dec

2018/

Jan

2019

Feb/

Mar

2019

Apr/

May

2019

Jun/

Jul

2019

Aug/

Sept

2019

Oct

2019

Val

idat

ion

ph

ase

(A)

Tender for control material

Identification of the participating

centers (survey)

Place order for EQA material

Material manufacture

Requirements for data collections

EQA sample validation

EQA sample distribution

Analysis of EQA results/Issue of

results

Har

mo

niz

atio

n

ph

ase

(B)

Creation of an harmonization

network

Sample selection and analysis

Comparative analyses

Publication of the

results/Workshops

27

Immuno-oncology Consortium with Leading EU Institutes

For Research Use Only. Not for use in diagnostic procedures.

The immuno-oncology consortia members:

Sabine Merkelbach-Bruse

Institute of Pathology, University Hospital Cologne International

Albrecht Stenzinger

Institute of Pathology, Heidelberg University Hospital

Michael Hummel

Institute of Pathology, Charité - University Medical Center

Berlin

Wilko Weichert and Nicole Pfarr

Institute of Pathology, Technical University of Munich

Bea Bellosillo

Hospital del Mar, Barcelona

Nicola Normanno

IRCCS National Cancer Institute "G. Pascale Foundation"

Jose Carlos Machado

i3S and Institute of Molecular Pathology and Immunology of

the University of Porto (IPATIMUP), Portugal

Maurice Jansen and Winand Dinjens

Erasmus MC Cancer Institute, Rotterdam

Empower clinical trials with

Ion Torrent™ immuno-oncology assays

28

Phase 1 - Controls

• 6 x cell line FFPE controls (AccuRef)• predicted to be high and low TMB plus 2 x Acrometric Oncology Hotspot Control (AOHC)

• check reproducible results across 8 labs

Phase 2 - Shared sample

• IPATIMUP providing 8 CRC samples for all labs which have MSI status,• no exome or clinical response data

• 8 shared samples run as standard TML workflow 8plex chip without normal

comparison

• TUM providing access to tumour-normal exome analysis via DKFZ

• Comparison to exome and interlay reproducibility

Phase 3 - Own samples

• 16 samples from own biobank

• high and low mutational burden status

• Ideally with clinical response data and/or any supporting biomarker/exome data

Tumour Mutational Load assay project - Experimental plan

29

Preliminary Phase I ResultsCell-Ref™ FFPE Slides and Acrometrix™ Control

TMB

(Mu

t/M

b)

A549 AOHC H2228 HCC2998 MCF7 SKMEL2 T47D

Average TMB 7.0 255.5 7.8 198.2 3.9 19.0 2.8

% CV 12% 9% 8% 6% 17% 9% 14%

Range 5.9-8.1

229.2-

270.8 6.8-8.4

176.5-

209.2 3.4-5.1 16.9-21.3 2.5-3.4

30

Preliminary data Phase II results (2 Laboratories)

R² = 0,9697

0

10

20

30

40

50

60

70

0 10 20 30 40 50 60 70

Re

p 2

(C

RO

M)

Rep 1 (IPATIMUP)

Reproducibility of Shared Clinical Samples

TMB testing in CRC cell lines (1)

MSI Status Mutation Count (cBbioPortal)

TML (Oncomine™TML Assay)

RKO MSI-H 382 105,62

LOVO MSI-H 267 80,64

HCT116 MSI-H 227 74,62

COLO320 MSS 23 15,89

HT29 MSS 64 15,89

SW1116 MSS 40 15,87

H1650 MSS 25 6,74

H1975 MSS 54 10,37

TMB testing in CRC cell lines (2)

R² = 0,979

0

50

100

150

200

250

300

350

400

450

0 20 40 60 80 100 120

Mu

tati

on

Co

un

t-

Cb

ioP

ort

al

TMB (Oncomine™TML Assay)

TMB in primary CRC FFPE samples

10

30

50

70

90

110

130

150

MSI MSI-L/MSS

Mu

tati

on

Lo

ad/M

B

(On

com

ine

™TM

LA

ssay

)

Genomics-Driven Oncology

Garraway JCO 2013

Thermo Fisher Scientific and its affiliates are not endorsing, recommending, or promoting any use or

application of Thermo Fisher Scientific products presented by third parties during this seminar.

Information and materials presented or provided by third parties are provided as-is and without

warranty of any kind, including regarding intellectual property rights and reported results. Parties presenting images, text and material represent they have the rights to do so.