Treatment strategies for oncogenic addicted NSCLC patients others than EGFR and ALK and NGS

Jordi Remon MasipThoracic Oncology Unit

Outline

Introduction1

BRAF V600E (~2%)2

RET fusion (1-2%) and NTRK fusion (~1%)

ROS1 fusion (1-2%)4

MET ex14 (~4%)

5

Next Generation Sequencing6

3

Outline

Introduction1

Treatment paradigm in NSCLC

PDL1 statusMolecular

PD-L1≥ 50%PD-L1< 50%

EGFR ALK ROS1BRAF

V600E HER2

RETNTRK MET

1ST LINECT +/- Pembro*,#

CBDCA+Taxol +/- BVZ*#

+/- Atezo@,; CDDP-PemAnd maintenance;

Platin/Pem/Pembr*

Platinum based-CT+/- necitumumab*,#

CT + Pembro / Atezo@

Crizotinib*,#

Ceritinib@

Entrectinib@

2ND LINE• Nivolumab*#

• Atezolizumab*#

• Pembrolizumab*#

If PD-L1 +• Docetaxel +Ramucirumab*,#

Platinumbased-CT

• Nivolumab*#

• Atezolizumab*#

• Pembrolizumab*#

If PD-L1 +• Docetaxel +Nintedanib# / Ramu*,#

T790M + → Osimertinib*#

(if not received as 1st Line)

T790M – or 1st Line Osimertinib→ Pem / Platinum or CT+Atezo+BVZ*#

Ceritinib*#

Alectinib#*

Brigatinib*

Lorlatinib*

Afatinib*,#

Erlotinib*,#

+/- BVZ#

Gefitinib*,#

Dacomitinib@

Osimertinib@

Crizotinib*,#

Ceritinib *#

Alectinib#,*

Dabrafenib+

Trametinib*#

Entrectinib*,#

LOXO101*,

LOXO 292PoziotinibTDM1@…

Crizotinib@

Platinum based-CT

Squamous Non-Squamous

*FDA approved#EMA approved

@Not yet approved

PD-L1Oncogene addiction

50%

Algorithm by Jordi Remon

Ipi + Nivo TMB ≥ 10@

Selection correlates with outcome

PDL1 statusMolecular

PD-L1Oncogene addiction

50%

Mok – JCO 2018 * Brahmer – WCLC 2017

ARCHER 1050 KEYNOTE 024

PD-L1≥ 50%PD-L1< 50%EGFR, ALK, ROS1, BRAF TMB High

~80%~20%

EGFR ALK ROS1

METHER2 RET NTRK

BRAF

Courtesy of Dr Mazières (modified)

ASCO biomarker guidelines

ROS1 rearrangement

RET rearrangement

HER2 mutation

BRAF mutation

MET alterations

✔

✔

Kalemkerian – JCO 2018 * Lindeman – JTO 2018

(Not for IASLC)

TREATMENT APPROVED 1 MONTH AGO!

Anti-PD(L)-1 in oncogenic addicted NSCLC 0

20

40

60

80

10

0

% o

f tu

mo

r ce

lls P

DL1

EGFR KRAS ALK ROS1 BRAF HER2 RET MET

IMMUNOTARGET COHORT (N=551)

Mazièrés – ASCO 2018

Driver PD SD CR/PR

BRAF 46% 30% 24%

MET 50% 34% 16%

KRAS 51% 23% 26%

HER2 67% 26% 7%

EGFR 67% 21% 12%

ALK 68% 32% 0

RET 75% 19% 6%

ROS1 83% 0 17%

TOTAL 57% 24% 19%

PFS 2.8 mo. OS: 13.3 mo.

Spigel – ASCO 2016 * DuDnik- ASCO 2018

Oncogenic addicted NSCLC and TMBMedianMut/Mb 742 4

Outline

BRAF V600E (~2%)2

BRAF mutationEurope1

All histology (Biomarkers France)

(n = 9,911)

BRAF 1.7%

EGFR (sensitizing)

9.5%

EGFR (resistance)

0.8%

HER2 0.9%

KRAS 27%

UKN/Other

53.8%

PI3K 2.6% ALK 3.7%

Nguyen – JTO 2015 * Barlesi – Lancet Oncol 2016 * PaiK - JCO 2011 (Courtesy of Dr Mazières) * Danker – Oncogene 2018

B-RAF

V600E

60%

Most common BRAFmutations in NSCLC Pathway and drugs

Vemurafenib

BRAF shorter OS compared wt. BRAFV600E “CDDP-resistant”

Vemurafenib and dabrafenib in BRAF mutant NSCLC

Hyman – NEJM 2015

VE-Basket trial

20 BRAFV600 NSCLCORR: 42%. PFS: 7.3 mo.

AcSé trial

Mazières – WCLC 2016

AcSé program Principles

Vemurafenib in patients with Non-Small Cell Lung Cancer (NSCLC) harboring BRAF mutation. Preliminary results of the Acsé trial. J. Mazieres1, C. Cropet2, F. Barlesi3, P.J. Souquet4, V. Avrillon2, B. Coudert5, J. Le Treut6, F. Orsini Piocelle7, G. Quere8, E. Fabre9, J. Tredaniel10, M. Wislez11, O. Huillard12, E. Dansin13, D. Moro-Sibilot14, H. Blons9, G. Ferretti14, E. Lonchamp15, N. Hoog-Labouret15,

V. Pezzella16, C. Mahier - Aït Oukhatar16, J.Y. Blay2. 1-CHU-Hôpital Larrey, Toulouse/France, 2-Centre Léon Bérard, Lyon/France, 3-CHU-Hôpital Nord, Marseille/France, 4-CH-Lyon Sud, Pierre Bénite/France, 5-Centre Georges François Leclerc, Dijon/France, 6-CHI Aix Pertuis, Aix En Provence/France, 7-CHR Annecy, Pringy/France, 8-CHRU-Hôpital Morvan, Brest/France, 9-Assistance Publique Hôpitaux de Paris (AP-HP)-Hôpital Européen Georges Pompidou (HEGP), Paris/France, 10-GH Paris Saint Joseph, Paris/France, 11-APHP-Hôpital Tenon, Paris/France, 12-APHP-Hôpital Cochin, Paris/France, 13-Centre Oscar Lambret, Lille/France, 14-CHU de Grenoble, Grenoble/France, 15-French

National Cancer Institute INCa, Boulogne Billancourt/France, 16-UNICANCER, Paris/France

2016 WCLC congress, Vienna, Austria #p3.02a-034

Background

Risk of a wide off label use of the drug

• Vemurafenib is registered as a monotherapy for the treatment of adult

patients with BRAF V600 mutation unresectable or metastatic melanoma

• Responses were observed in other tumour types

Identification of subsets of patients that may benefit from treatment

• Patients ≥18 years with advanced disease harbouring BRAF genomic

alterations (except colorectal cancer and V600 BRAF mutated melanoma) and not eligible for any other active trial targeting the same alteration.

AcSé Program : secured access program to innovative cancer drugs

Consort diagram NSCLC

Patients characteristics

Patients on treatment

A regular dermatological monitoring has been set up. Indeed, all

patients are referred to a dermatologist prior to first intake, after 28

days of Vemurafenib and then every 3 months. An initial dermatologic

history, including photo type, history of sun / UV exposure, previous

skin cancers and immunosuppression is completed. Suspicious skin

lesions are biopsied or excised. Melanoma and squamous cell carcinoma

lesions are submitted for central dermatopathology review.

Furthermore, specialists have been appointed for the management of

BRAF inhibitors specific skin toxicities.

Reasons for treatment discontinuation

Progression 33

Toxicity 18

Cutaneous 6

Hepatic 4

Other 8

Other 8

Patient decision 7

Physician decision 1

• 59 stopped their treatment

Efficacy - Best response

Waterfall plot of best reduction in tumour measurement from baseline

• Vemurafenib provided interesting response rate and PFS in BRAF V600 NSCLC . Our results are consistent with previous findings in a large population of patients closed from our real life practice (heavily pre-treated, PS2 and brain metastasis allowed).

• Vemurafenib was not found efficient in NSCLC with other BRAF mutations.

• Toxicity profile is manageable. • These results underline the interest of integrating BRAF V600 in

biomarkers routine screening. Our trial will continue up to 100 NSCLC patients to provide more comprehensive data on this population.

Collaborative groups • French thoracic oncology Intergroup (IFCT) • French Cooperative Gynecological Cancer Research

Group (ARCAGY Gineco) • French genito-urinary cancer cooperative groups

(GETUG/AFU) • French Sarcoma Group (GSF-GETO) • French Gastro Institutional Groups (UCGI) • French thyroid cancer network (TUTHYREF) • French Innovative Leukemia Organization (FILO) • French of myeloma intergroup (IFM) • Skin Cancer Group (GCC) from the French Society of

Dermatology (SFD)

Thanks Patients

Study support

• financial support of the INCa, and of the ARC Foundation, the Unicancer’s partner for Personalized Medicine research

• institutional support of Roche

Accrual

BRAF V600 screening activity*

(At August 31, 2016)

Number of positive

cases

Patients included

(from october 1, 2014 to

october 17, 2016)

1304 58 22**

Inclusions per cohorts

* : except NSCLC, colorectal and melanoma, tests performed for diagnosis ** : + 82 NSCLC + 17 pts from molecular pangenomic programs or other circuits.

a : 53 V600E, 1 V600D, 1 V600K, 1 V600M b : 1 G466V, 1 G596R, 2 G469A, 1 G469V, 2 K601N, 1 K601E and 1 N581S c : including 1 V600D and 1 V600M d : including 1 V600K

960mg Vemurafenib given twice daily, continuously

Treatment scheme

• Endpoint : Objective Response Rate (ORR) evaluated in each cohort

• Bayesian sequential design

• 100 patients/cohort maximum

• New data could lead to open additional cohorts

• Up to 200 investigating centres

1/ Promote a secured access for all patients with an advanced refractory malignancy and no therapeutical alternative through an academic phase II clinical trial.

• One trial for each targeted treatment selected

• Withdrawal if high toxicity or no efficacy in a predefined number of patients with the same tumour type

• efficacy signal force to inform pharmaceutical firms for drug development decision making

2/ Ensuring equity of access to innovation

• Provide nationwide molecular tumour diagnosis for all patients through INCa molecular genetic centres

• Whatever the healthcare institution status (public hospitals, private hospitals…)

• Perform high quality tests

• Hemopathies, solid tumours

France organization of the 28 molecular centres for personalized medicine

Partnerships between laboratories located in University Hospitals and Cancer Centres • Regional organization • Cooperation between pathologists and

biologists AcSé program

Validated projects

A/Pathology cohorts “one pathology, BRAF V600 mutation”

diagnosed by INCa molecular genetic centres

1. NSCLC

2. Ovarian

3. Cholangiocarcinoma

4. Thyroid cancer

5. Prostatic cancer

6. Bladder cancer

7. Sarcoma / GIST

8. Multiple Myeloma

9. Chronic Lymphocytic Leukaemia (CLL)

10. Hairy Cell Leukaemia (HCL) (this excludes HCL variant types, marginal zone splenic lymphoma (MZL) and splenic red pulp lymphoma (SRPL))

B/ Miscellaneous malignancies

BRAF genomic alterations tested via emerging biomarkers programs or molecular pangenomic programs :

• Non-predefined pathology, V600 mutation

• Same or other non-predefined pathology harbouring, non V600 activating mutations

• Same or other non-predefined pathology, BRAF amplifications

• Disease response assessed every 8 weeks

• Safety assessed continuously

• Treatment pursed until progression, unacceptable toxicity, undercurrent conditions, or patient refusal

For HCL and CLL, Vemurafenib is prescribed for 2 cycles and possibly for 2 additional cycles if CR is not achieved. Treatment is stopped at day 112 max whatever the response.

Disease characteristic Sex

Male 36 (48%)

Age (Years)

Median (range) 68 (40 ; 85)

Time between histological diagnosis and inclusion (years)

Median (range) 1 (0,1 ; 12,4)

Material for molecular analysis

Primary tumor 46 (61,3%) Metastasis 20 (26,7%) Ganglion 9 (12%)

Prior chemotherapy 70 (93,3%) 1 line 37 (52,8%) 2 lines 24 (34,3%) 3 lines and more 9 (12,9%)

Tobacco No smoker 18 (24%) Smoker 50 (66,7%) PA – Median (Range) 30 (0,6 ; 80)

• 16 still on treatment

Frequency N=59

Frequency (%) N=75

The current median treatment duration : 1,9 (0,2 ; 72) months

*Considering a beta(1,1) as the prior distribution of the ORR

PFS

OS

median PFS NSCLC BRAF V600 : 4,2 months [3,6 – 7,1]

38 progressions or deaths

median PFS NSCLC BRAF non V600 : 1,9 months [1,1; NA]

7 progressions or deaths

median OS NSCLC BRAF V600: 8,7 months [5,6 – NA] ; 25 deaths

median OS NSCLC BRAF non V600: 4,7 months [1,2; NA] ; 6 deaths

To October 17, 2016 :

Focus on Non Small Cell Lung Cancers

Phase 2 clinical trial: Secured Access to Vemurafenib for patients with tumours harbouring BRAF genomic alterations – PI : Jean-Yves BLAY / NCT02304809

Related toxicities NSCLC BRAF (n=75)

Grade Numbers of events

Background

Main objective

Population

Molecular diagnosis process

Cohorts

Statistical design

Dermatological monitoring

Objectives

• When a marketed targeted therapy exists in a molecularly defined subgroup of patients

• When the same alteration is found in other tumour types

Current analysis of tumour response 65 pts BRAF

10 pts recently included received < 16 weeks

Current analysis of safety 75 pts BRAF

7 pts included but dead before treatment (2), treated < 4 weeks (3)

or missing data (2)

16 pts not included due to non inclusion criteria

98 pts BRAF screened between 10/14 & 10/16

82 pts BRAF included in the study

a

b

c d c

(*) RECIST evaluation is not evaluable for 19 patients :

12 progression before first evaluation, 4 treatment stop before first evaluation, 1 missing data, 2 deaths before first evaluation

(a) V600D; (b) V600M

a

b

NSCLC BRAF non V600

Conclusion

55 BRAFV600 NSCLCORR: 38.5%. PFS: 4.2

Vemurafenib Vemurafenib

84 BRAFV600E NSCLCORR: 30%. PFS: 5.5

BRF113928 (Cohort A)

Dabrafenib

Planchard- Lancet Oncol 2016

Clinical benefit with dabrafenib and trametinib

1st Line (BRF113928, cohort C)

36 BRAFV600E NSCLCORR: 64%

2nd Line (BRF113928, cohort B)

PFS: 10.9 mo. OS: 24.6 mo.

Planchard- Lancet Oncol 2016 * Planchard – Lancet Oncol 2017

40 BRAFV600E NSCLCORR: 67%

PFS: 10.2 mo. OS: 18.2 mo.

BRAF and immunotherapy

Dudnik – WCLC 2017 * Dudnik – JTO 2018

42% of BRAFV600E & high PD-L1 expression levels (≥ 50% by 22C3 IHQ) 25% of BRAFV600E associated with high TMB (≥20 Mb)

PFS: V600E vs. Non-V600E: 3.7 mo. vs. 4.1 mo.

N=21 V600E

ORR: V600E vs. Non-V600E: 25% vs. 33%

Outline

MET ex14 (~4%) 3

MET aberrations in NSCLC

MET mutant / amplified NSCLC, poor prognosis

Paik – Cancer Discovery 2015 * Tong - Clin Cancer Res 2016 * Drilon – JTO 2016

MET as a primary driver

MET as a secondary / co-driver

MET exon 14 mutations

19%

933 non-squamous54 non-smokers

Frampton - Cancer Discovery 2015 * Awad - JCO 2016 * Heist - The Oncologist 2016 * Liu – JCO 2016

Sarcomatoide ~ 20% MET+

Mutually exclusive

Personalised Treatment impacts in outcome

MET TKI prolonged OS (HR 0.11,p=0.04) Drilon – JTO 2016 * Awad – ASCO 2017

Crizotinib in MET amplified

Camidge – ASCO 2014 * Camidge – ASCO 2018 Moro – Sibilot – WCLC 2016 * Vassal – ASCO 2018

AcSé Trial

mPFS: 3.2 mo.mOS: 7.7 mo.

ORR: 32%No correlation with MET copy number

ORR: 67%

DoR: 18.4 mo.

PROFILE 1001 Trial

N 3 14 20

ORR (%) 33 14.3 40

DoR (mo.) 12.1 3.7 5.5

PFS (mo.) 1.8 1.9 6.7

MET/CEP7 Ratio ≤ 2

MET/CEP7 Ratio ≥ 5

ASCO2018

N 25

ORR (%) 28

Crizotinib and Tepotinib in MET mutation

Objectiveresponserate(ORR) 11/28(39%,95%CI:22,59)

Bestoverallresponsen(%)

Completeresponse

PartialresponseStabledisease

ProgressivediseaseIndeterminate

2(7)

9(32)10(36)

2(7)5(18)

BestPercentChangeFromBaselineinSizeofTargetLesions(n=22)*

%c

ha

ng

ef

ro

mb

ase

lin

e

Progressivedisease

Stabledisease

Partialresponse

Completeresponse

*Includespatientswithmeasurablediseaseatbaselineand≥1responseassessmentscan;excludes1patientwithearlydeath,4patientswith

indeterminateresponseand1patient(CRresponder)withnomeasurabletargetlesionsatbaseline

Mediandurationofresponse:

9.1months(95%CI:5.9,10.5)

40

20

0

-40

-20

-60

-80

-100

Drilon - WCLC 2016

PROFILE 1001 Trial, crizotinib

ORR: 39%. DoR: 9.1 mo

Tepotinib, phase II

Felip – ASCO 2018

500 mg QD. 0-2 prior linesMET+ tissue or liquid biopsy

ORR: 43%. DoR: 12.4 mo

N=31

Poor outcome to ICI in MET ex14 altered NSCLCs

TMB is lower in patients with MET exon 14 altered NSCLCs compared to other

NSCLCs

P = 0.0006PD-L1, Cell Signaling, Clone E1L3N assay, n=54

0 % 1 – 49 % ≥50 %

n, (%) 19 (35) 10 (19) 25 (46)

MET exon 14-altered cancers can express high levels of PD-L1

PD- L1 St at us

-100

-80

-60

-40

-20

0

20

40

60

80

100

Ch

an

ge

fro

m b

ase

lin

e (

%)

PDL1 ≥ 50

No partial responses among the 6 patients with PD-L1 ≥ 50

NANA9080900100NANANA00NA90 100

Sabari – ASCO 2017

ORR 6.7%

Outline

ROS1 fusion (1-2%)4

ROS1 fusion partnersA total of 14 different ROS1 fusion partner genes. ROS1 fusions do not contain kinase domain

Tyrosine kinase domain Transmembrane domain Coiled-coil domain Lin – JTO 2017 * McCoach – CCR 2018

ROS1 accounts 1-2% of NSCLC

Crizotinib in ROS1 NSCLC

ORR: 66%. PFS: 19.3 mo. ORR: 72%. PFS: 15.9 mo.

Shaw –NEJM 2014 * Shaw – ESMO 2016 Wu – JCO 2018

Crizotinib in ROS1 NSCLC patients

Trial N Region ORR PFS (mo.) 1-year OS

PROFILE 1001, ph I 53 World 66% 19.3 79%

OxOnc, ph II 127 East Asia 72% 15.9 83.1%

EUROS, pooled 32 Europe 80% 9.1 NR

AcSé, basket trial 37 France 54% 9.1 NR

EUCROSS, ph II 34 Spain/Germany 69% Not reached NR

METROS, ph II 26 Italy 62% 17.2 Not reached

Shaw- NEJM 2014 * Shaw – ESMO 2016 (1206PD) * Wu – JCO 2018 * Mazières – JCO 2015 * Moro-Sibilot – WCLC 2015 * Vassal – ASCO 2018 * Michels – WCLC 2016 * Landi – JTO 2017

Crizotinib approved by FDA (11 March 2016) and EMA (21 July 2016)

Efficacy of crizotinib among different types of ROS-1 partners

Li – JTO 2018

12.6 mo

44.5 mo

17.6 mo

24.3 mo

PFS OS

It remains unknown whether the partner will drive different resistance mutations

Ceritinib and entrectinib in ROS1

Ceritinib

ORR: 62%. PFS: 19.3 m*

Lim – JCO 2017 (*crizotinib-naïve. Whole: 9.3 mo.) Ahn – WCLC 2017

ORR: 78%. PFS: 29.6 m

(Intracranial RR: 83.3%)

N=32

Entrectinib

N=32

Treatment sequence? Activity in BM ? Toxicity profile?

How do we manage with resistance?

Lin – JTO 2017 * Gainor – JCO Precision Oncology 2017

~50-60% ~20-25%

Frequency 6% 41% 6%

Efficacy in EXP6 (ROS1+ With Any Prior Treatment)

EXP6

(n=47)

ORR, n/N (%) (95% CI)

17/47 (36)(23, 52)

IC ORR, n/N (%) (95% CI)

14/25 (56) (35, 76)

Median DOR, mo (95% CI)

13.8 (11.1, NR)

Disease control rate at 6 months

45%

DOR ≥6 mo, n⁰/n (%) 12/17 (71)

Median PFS, mo (95% CI)

9.6(4.7, NR)

• 25 patients (53%) had brain

metastases at baseline.CI, confidence interval; DOR, duration of response; mo, months; NR, not reached.

70

60

10

0

30

20

50

40

Intracraniala,b

‒10

‒20

‒30

‒40

‒50

‒60

‒70

‒80

‒90

‒100

#

#

# #

#

#

#

#

70

60

10

0

30

20

50

40

‒10

‒20

‒30

‒40

‒50

‒60

‒70

‒80

‒90

‒100

Overalla,b

Off treatment or PD occurred

Complete response

Partial response

Stable disease

Progressive disease (PD)

Indeterminate

Best

Ch

an

ge F

rom

Baselin

e (

%)

Previously received crizotinib#

#

#

#

#

#

# # # # #

#

## # #

# ## #

#

#

#

#

Solomon – WCLC 2017 * Besse – ESMO 2017

Lorlatinib in ROS1 and BM

ORR: 36%. PFS: 9.6 mo.(28% no previous TKI,

64% 1 previous TKI, 8% 2 previous TKI)

ORR: 56%

ROS-1 as long survivors as ALK

Solomon – JCO 2018 * Park – J Thorac Oncol 2018

ROS1ALK: PROFILE 1014

4y-OS: 56.6% 4y-OS: ~52%

N=103 ROS1

Outcome of ROS-1

Park – J Thorac Oncol 2018

Outline

RET fusion (1-2%) and NTRK fusion (~1%)5

RET-altered cancers

10-20%

IHQ not reliableRT-PCR

NSCLC:0.9%-1.8%as a whole

to 6%-14% in WT ADK

FISH

Ferrara- J Thorac Oncol 2017

RET rearrangements in NSCLC

Khono – Transl Lung Cancer 2015 * Gautschi – JCO 2017 * Ferrara – JTO 2018 Dimerization domain

72%

RET TKI in the clinic

Yoh –Lancet Resp Med 2016 * Lee – Ann Oncol 2017 * Drilon – Lancet Oncol 2016 * Velcheti – Ann Oncol 2016

Lenvatinib

LURET. Ph II. N=17

RR 47%. PFS 4.7 mo

Korean. Ph II. N=18

RR 18% . PFS 4.5 mo. OS 11.6 mo.

Vandetanib

Cabozantinib Ph II trial. N=25

RR 28% . PFS 5.5 mo. OS 9.9 mo.

Ph II trial. N=25

RR 16% . PFS 7.3 mo.

Vandetanib

Better outcome CCDC6 vs. KIF5B

Better outcome in non-KIF5BBetter outcome in non-KIF5B

RET tyrosine kinase inhibitors

Higher toxicity, dose reductions in 70%Different clinical activity

BLU-667: ARROW ph I trial (Part I)

Highly selective ATP-competitive RET inhibitor

Subbiah – Cancer Discovery 2018

N= 53 (29 MTC, 19 NSCLC, 3 others) Maximum Tolerated Dose400 mg QD

ORR 46% (43% PR, 3% CR)

Subbiah – AACR 2018

LOXO-292: LIBRETTO-001 trial

Subbiah – Ann Oncol 2018 * Velcheti – WCLC 2017

Highly selective ATP-competitive RET inhibitor

Significant CNS penetration

ORR in NSCLC (n=38): 77%

Drilon – ASCO 2018

RR regardless of fusion partner /dose

Pemetrexed based CT in RET-fusion NSCLC

Drilon – Ann Oncol 2016

ORR in RET: 45%

PFS

NTRK familyNTRK-1, -2, -3

• Encode TrkA, TrkB and TrkCtransmembrane receptors, respectively.

• Ras/Raf/MAPK pathway PI3K/Akt/mTOR pathway PLCc/PKC pathway.

• Trks are involved in physiological CNS development and maturation

• NTRK fusions in NSCLC 0.1%-~3%

Amatu – ESMO Open 2016 (Courtesy of rof Besse)

Multiple NTRK 1/2/3 fusions across multiple tumours

Vaishnavi – Cancer Dsicovery 2015 * Hyman – ASCO 2017

NTRK fusions in Lung Cancer

FISH

FISH

IHC

NGS (DNA/RNA) tests

Vaishnavi - Nature Medicine 2013 * Hyman- ASCO 2017 (Courtesy Dr Reguart)

Larotrectinib (LOXO101)

Drilon – NEJM 2018 * Nyman – ASCO 2017

Larotrectinib (LOXO101)

*

*

***

ORR by IRR: 75%Median time to response: 1.8 mo.

Drilon - NEJM 2018N=55 patients. LOXO101: 100 mg BID. 15% required dose reductions

Larotrectinib (LOXO101)

Drilon - NEJM 2018 * Hyman –ASCO 2017

Efficacy regardless NTRK gene

Efficacy regardless gene partner

45%2%53%

PFS

12-mo. PFS other oncogenic alterations:

BRAF V660E, D+T ~40%

ROS1, Crizotinib ~66%

ROS1, Entrectinib ~70%

EGFR, osimertinib (1st L) ~65%

Entrectinib (ALK, ROS, NTRK TKI)

Drilon – Cancer Discovery 2017 * Drilon – AACR 2016

Ph I: STARTRK-1 & ALKA-372-001. N=25 eligible patients

DoR (mo.) 2.6-15.1 17.4 7.4PFS (mo.) NR 19 8.3

600 mg QDt1/2: 20-22h 100% 86% 57%

LOXO-195 overcomes resistance

Drilon – Cancer Discovery 2017 Phase 1/2 LOXO-195 (NCT03215511) ongoing

ESMO-MCBS in “high unmet need”

Cherney – Ann Oncol 2017

Grade 3 Grade 2 Grade 1

PFS≥ 6 mo. ≥ 3 to < 6 mo. 2 to < 3 mo.

ORR≥ 60 % ≥ 40 % to < 60% ≥ 20 % to < 40%

ORR and DoR

≥ 20 <60%and

DoR ≥ 9 mo.

≥ 20% to < 40%and

DoR ≥ 6 mo to < 9 mo.

> 10 to < 20%and

DoR ≥ 6 mo.

CRIZOTINIB / DABRAFENIB +TRAMETINIBLAROTRECTINIB / ENTRECTINIB

Outline

Next Generation Sequencing6

Performed, actionable, treated

Zehir- Nature med 2017 * Tedan – ASCO 2017 * Massard, Cancer Dis. 2017 * André – Lancet Oncol 2014* Meric-Bernstam – JCO 2015 * Wheler – Cancer Res 2016 * Tsimberidou –CCR 2012 * Stockley – Genome med 2016 * Schwaederle – Mol Cancer Ther 2015 * Sohal –JNCI 2016 * Johnson – The Oncologist 2014 . Safir Lung unpublished

Serie N Molecular profile Actionable mutation Matched trials

MSK-IMPACT 10,945 91% 37% 11%

PROFILER 2,676 73% 52% 7%

MOSCATO 1,035 81% 40% 19%

SAFIR01* 423 71% 46% 13%

SAFIR Lung 686 67% 43% 16%

MDACC 2,601 77% 39% 11%

MDACC 500 68% 64% 37.6%

MDACC 1,283 89% 40% 15%

IMPACT/COMPACT 1,893 86% 50% 5%

PREDICT 347 NR NR 25%

CLEVELAND 250 89% 49% 11%

VANDERBILT 103 94% 83% 25%

RANGE (mean) 68%-94% (73%) 37%-83% (43%) 5%-37.6% (11%)

NGS triggers agnostic approval and

know new predictive biomarkers

Yarchoan – NEJM 2017

Correlation between TMB and RR (p<0.001)

RR= 10.8 x log (X)-0.7X= number of mutations/Mb

Larotrectinib approval NTRK-fusion tumors

Drilon – NEJM 2018

Personalised treatment in NSCLC

Barlesi – Lancet 2015 * Jordan- Cancer Discovery 2017

N=860 ADC. ≥ 300 cancer-associated genes

87% Potentially actionable alterations. 28 days

N=17,664 (76% ADC). 6 cancer-associated genes

21% Potentially actionable alterations. 11 days

Do we treat more ADC lung cancer patients

with massive sequencing?

Jordan Cancer Discovery 2017

37.1% received matched therapies14.4% excluding alterations associatedwith SOC, with clinical benefit in 52%

Strong likelihood receiving matched therapiesWeak likelihood receiving matched therapies

NGS cost vs. sequential test in NSCLC

Pennell – ASCO 2018

(EGFR, ALK, ROS1, BRAF, MET, HER2, RET, NTRK1)

Should we perform NGS regardless of tumor type? Or NGS with specific genes for specific cancers?How many genes?

B-RAF + MEK inhibitor in NSCLC

(V600E BRAF mutation) Lancet Oncol 2017

ALK inhibitor in NSCLC

(ALK rearrangment) NEJM 2010

EGFR inhibitor in NSCLC (EGFR mutation) Lancet Oncol 2012

ROS1 inhibitor in NSCLC(ROS1 rearrangment) NEJM 2014

*

*

Nonsmoker

Smoker, ≤ 40

pack years

Smoker, > 40

pack years

*

*

* ~4% NSCLC

~11% NSCLC ~1-2% NSCLC

~2% NSCLC~2% NSCLC

RET inhibitor in NSCLC

(RET fusions) Drilon ASCO 2018

~1-2% NSCLC

NTRK inhibitor in NSCLC

(NTRK rearrangment) Drilon NEJM 2018

~0.1-3% NSCLC

Conclusions

NGS6

BRAF V600E 1

RET fusion

ROS1 rearranged 3

MET ex14

4

NTRK fusion5

Dabrafenib + Trametinib. Non-V600E, ERKi?

LOXO292 / BLU667. More selective, less toxic

Crizotinib SoC. Entrectinib next 1st Line?

Crizotinib. Pending other MET inhibitors

Example of tissue-agnostic biomarker approval

NGS improve knowledge, basket trials, new drugs. Platforms

2

PAST TIME IN NSCLC TREATMET PRESENT MOVING TIME IN NSCLC TREATMET