ESMO 2019 Update Axel Grothey Director, GI Oncology Research West - - PowerPoint PPT Presentation

ESMO 2019 Update

Axel Grothey Director, GI Oncology Research West Cancer Center Research Institute

Encorafenib plus Cetuximab With or Without Binimetinib for BRAF V600E–Mutant Metastatic Colorectal Cancer: Expanded Results from a Randomized, 3-Arm, Phase 3 Study vs. the Choice of Either Irinotecan or FOLFIRI plus Cetuximab (BEACON CRC)

Josep Tabernero, Axel Grothey, Eric Van Cutsem, Rona Yaeger, Harpreet Wasan, Takayuki Yoshino, Jayesh Desai, Fortunato Ciardiello, Fotios Loupakis, Yong Sang Hong, Neeltje Steeghs, Tormod Kyrre Guren, Hendrik-Tobias Arkenau, Pilar Garcia-Alfonso, Ashwin Gollerkeri, Michael Pickard, Kati Maharry, Janna Christy-Bittel, Lisa Anderson, and Scott Kopetz

2

BEACON CRC: Binimetinib, Encorafenib, And Cetuximab COmbiNed to Treat BRAF-mutant ColoRectal Cancer

Triplet therapy ENCORAFENIB + BINIMETINIB + CETUXIMAB n = 205 Doublet therapy ENCORAFENIB + CETUXIMAB n = 205 Control arm FOLFIRI + CETUXIMAB, or irinotecan + CETUXIMAB n = 205

R 1:1:1 Phase 3

Study Design

Primary Endpoints:

OS

(All randomized Pts)

Randomization was stratified by ECOG PS (0 vs. 1), prior use of irinotecan (yes vs. no), and cetuximab source (US-licensed vs. EU-approved) Patients with BRAFV600E mCRC with disease progression after 1 or 2 prior regimens; ECOG PS of 0 or 1; and no prior treatment with any RAF inhibitor, MEK inhibitor, or EGFR inhibitor

Triplet vs Control

Secondary Endpoints: Doublet vs Control and Triplet vs Doublet - OS & ORR, PFS, Safety QOL Assessments: EORTC QOL Questionnaire (QLQ C30), Functional Assessment of Cancer Therapy Colon Cancer, EuroQol 5D5L, and Patient Global Impression of Change).

ORR – Blinded Central Review

(1st 331 randomized Pts)

ENCORAFENIB + BINIMETINIB + CETUXIMAB

N = 30

Encorafenib 300 mg PO daily Binimetinib 45 mg PO bid Cetuximab standard weekly dosing

Safety Lead-in

Overall Survival and Objective Response Rate

HR (95% CI): 0.52 (0.39-0.70) 2-sided P<0.0001

Median OS in months (95% CI)

Triplet 9.0 (8.0-11.4) Control 5.4 (4.8-6.6)

Triplet Control

HR (95% CI): 0.60 (0.45-0.79) 2-sided P=0.0003

Median OS in months (95% CI) Doublet 8.4 (7.5-11.0) Control 5.4 (4.8-6.6) Control Doublet

Objective Response Rate (First 331 Randomized Patients)

Confirmed Response by BICR Triplet N=111 Doublet N=113 Control N=107

Objective Response Rate 26% 20% 2%

95% (CI) (18, 35) (13, 29) (<1, 7) p-value vs. Control <0.0001 <0.0001

Kopetz et al. Ann Oncol 2019;30 (Supplement 4): iv137–iv151 (LBA-006).

Triplet vs Control Doublet vs Control

5

Overall Survival: Triplet vs Doublet (All Randomized Patients)

Study not powered to formally compare the results of the triplet combination to the doublet combination

33% of patients alive with

less than 6 months FU

Median Follow up: 7.8 Months* HR (95% CI): 0.79 (0.59-1.06)

Median OS in months (95% CI) Triplet 9.0 (8.0-11.4) Doublet 8.4 (7.5-11.0)

Doublet Triplet

*all randomized patients.

6

Overall Survival: Triplet vs Doublet (First 331 Randomized Patients*)

Study not powered to formally compare the results of the triplet combination to the doublet combination

*Post-hoc descriptive analysis.

1% of patients alive with

less than 6 months FU

Doublet Triplet HR (95% CI): 0.74 (0.53-1.04)

Median OS in months (95% CI) Triplet 9.5 (8.1-12.0) Doublet 8.3 (6.2-10.7)

Median Follow up: 12.5 Months

7

Waterfall Plots of Best Change in Sum of Diameters

(based on central review)

*Patients whose SoD was contraindicated by assessment of PD. SoD=sum of longest diameter. Includes patients with measurable disease with a baseline and at least one post-baseline scan.

Doublet

N=87 N=98

Triplet Doublet

N=73

Control

Combined Waterfall Plots: Triplet and Doublet

8

*Patients whose Sum of longest diameter was contraindicated by assessment of PD. 2-sided p-value is descriptive only for this analysis. Comparison of the two waterfall plots is suggestive of a shift towards greater tumor reduction from baseline in the Triplet. Doublet

Triplet (N=87) Doublet (N=98) Wilcoxon Rank Sum P-Value: 0.033

Grade ≥3 Adverse Events and Laboratory Abnormalities*

9

*Occurring in at least 2% of patients in either triplet or doublet arms.

Event Triplet N=222 Doublet N=216 Control N=193 Grade ≥3 Grade ≥3 Grade ≥3 Diarrhea 10% 2% 10% Abdominal pain 6% 2% 5% Nausea 5% <1% 1% Vomiting 4% 1% 3% Pulmonary embolism 4% 1% 4% Intestinal obstruction 3% 4% 3% Asthenia 3% 3% 5% Acute kidney injury 3% 2% <1% Fatigue 2% 4% 4% Dermatitis acneiform 2% <1% 3% Ileus 2% 1% 2% Urinary tract infection 1% 2% 1% Cancer pain <1% 2% <1% Laboratory Abnormality Grade ≥3 Grade ≥3 Grade ≥3 Hemoglobin (g/L), hypo 11% 4% 4% Creatinine (umol/L), hyper 5% 2% 1% Creatine Kinase (IU/L), hyper 3% Bilirubin (umol/L), hyper 2% 2% 3%

Other CRC Highlights

• FOxTROT:
• Neoadjuvant FOLFOX (x3) in locally advanced colon cancer decreases R1 resection rate
• Decreased surgical morbidity
• 20% of patients overstaged (unnecessary chemo)
• No tumor regression in MSI-H cancer
• MOUNTAINEER:
• HER-2 pos CRC: Tucatinib + Trastuzumab generate 55% RR (in 22 pts)
• IDEA high-risk stage II:
• Same regimen-effect as seen in stage III (CAPOX vs FOLFOX)
• IDEA France:
• ctDNA in stage III is prognostic (we knew that), but intensified chemo can improve outcome

(that’s new)

ctDNA in IDEA France

• First documentation that

“intensification” of adjuvant chemotherapy (6 vs 3 months) can improve outcome in ctDNA pos stage III colon cancer

Taieb, ESMO 2019

KEYNOTE-062 Study Design (NCT02494583)

aEU/North America/Australia, Asia (South Korea, Hong Kong, Taiwan, Japan), Rest of World (including South America). bAdministration of pembrolizumab monotherapy was not blinded. cChemotherapy: Cisplatin 80 mg/m2 Q3W + 5-FU 800 mg/m2/d for 5 days Q3W or capecitabine BID d1-14 Q3W (Cisplatin may be capped at 6 cycles as per country guidelines).

Stratification Factors

• Regiona
• Locally advanced or

metastatic disease

• 5-FU or Capecitabine

Key Eligibility Criteria

• Locally advanced,

unresectable or metastatic gastric or gastroesophageal adenocarcinoma

• HER2/neu negative,

PD-L1-positive disease (CPS ≥1)

• ECOG PS 0 or 1

Placebo + Chemotherapy Pembrolizumab 200 mg Q3W for up to 35 cyclesb Until unacceptable toxicity, disease progression, or patient/physician withdrawal decision

R (1:1:1)

Pembrolizumab 200 mg Q3W (to 35 cycles) + Chemotherapyc

N = 256 N = 257 N = 250

Primary endpoints: OS and PFS Secondary endpoints: ORR, Safety

N = 763

Tabernero ASCO 2019

Overall Survival: P vs C (CPS ≥1)

aNI, non-inferiority margin; bHR (95% CI) = 0.91 (0.74-1.10), P = 0.162 for superiority of P vs C; Data cutoff: March 26, 2019.

Median (95% CI)b 10.6 mo (7.7-13.8) 11.1 mo (9.2-12.8)

Pembro Events HR (99.2% CI) NIa Chemo 79% 86% 0.91 (0.69-1.18) 1.2

OS, %

3 6 9 12 15 18 21 24 27 30 33 36 39 42 10 20 30 40 50 60 70 80 90 100

256 201 162 139 120 107 94 83 59 38 23 12 4 250 230 192 144 114 94 75 49 38 21 15 6 2 2

12-mo rate 47% 46% 24-mo rate 27% 19%

Time, months

• No. at Risk

Tabernero ASCO 2019

Shitara ESMO 2019

FGFR1-3 fusions, mutations & amplifications (11-45%) IDH1/2 mutation (5-36%) RNF43 mutation (9%) PIK3CA mutations (3-9%) BRAF mutations (3-7%) ERBB3 amplification (7%) MET amplification (2-7%) ERBB3 mutation (7%) MET mutation (5%) EGFR mutation (1-2%) ERBB2/3 amplification (11-17%) IDH1/2 mutation (0-7%) PIK3CA mutation (7%) MET mutation (4%) BRAF mutations (3%) MET amplification (1%) ERBB2/3 amplification (10-19%) PIK3CA mutation (6- 13%) BRAF mutation (1-6%) RNF43 mutation (4%) MAP2K4 mutation (4%) EGFR mutation (4%) FGFR1-3 fusions, mutations & amplifications (3%) IDH1/2 mutation (2%)

Genetic Landscape of Biliary Cancers

ECC GBCA ICC

Frequencies of Actionable Mutations in BTCs

0% 5% 10% 15% 20% 25% 30% ICC ECC GBCA

Goyal et al., ESMO GI 2017

Phase II: BGJ398 in FGFR altered Cholangiocarcinomas

RR: 14.8% SD: 60.7% 2/3 patients with more than 2 lines

• f prior therapy

Javle et al, JCO 2018

Pemigatinib – Study Cohorts

Pemigatinib RR (Cohort A – FGRF Fusions)

Pemigatinib – Overall Survival

1

ClarIDHy: A global, phase 3, randomized, double-blind study of ivosidenib vs placebo in patients with advanced cholangiocarcinoma with an isocitrate dehydrogenase 1 (IDH1) mutation

Ghassan K. Abou-Alfa,1,2 Teresa Maraculla,3 Milind Javle,4 R. Kate Kelley,5 Sam Lubner,6 Jorge Adeva,7 James M. Cleary,8 Daniel V. Catenacci,9 Mitesh J. Borad,10 John Bridgewater,11 William P . Harris,12 Adrian G. Murphy,13 Do-Youn Oh,14 Jonathan Whisenant,15 Bin Wu,16 Liewen Jiang,16 Camelia Gliser,16 Shuchi S. Pandya,16 Juan W. Valle,17 Andrew X. Zhu18

1Memorial Sloan-Kettering Cancer Center, New York, NY, USA; 2Weill Medical College at Cornell University, New York, NY, USA; 3Vall d’Hebron University

Hospital, Vall d’Hebron Institute of Oncology (VHIO), Barcelona, Spain; 4MD Anderson Cancer Center, Houston, TX, USA; 5University of California San Francisco, San Francisco, CA, USA; 6University of Wisconsin Carbone Cancer Center, Madison, WI, USA; 7Hospital Universitario 12 de Octubre, Madrid, Spain;

8Dana-Faber Cancer Institute, Boston, MA, USA; 9University of Chicago Medical Center, Chicago, IL, USA; 10Mayo Clinic Cancer Center, Phoenix, AZ, USA; 11UCL Cancer Institute, London, UK; 12University of Washington, Seattle, WA, USA; 13Johns Hopkins University, Baltimore, MD, USA; 14Seoul National University

Hospital, Cancer Research Institute, Seoul National University College of Medicine, Seoul, Korea; 15Utah Cancer Specialists, Salt Lake City, UT, USA;

16Agios Pharmaceuticals, Inc., Cambridge, MA, USA; 17University of Manchester, The Christie NHS Foundation Trust, Manchester, UK; 18Massachusetts General Hospital Cancer Center, Harvard Medical School, Boston, MA, USA

Presented at the European Society for Medical Oncology Congress, September 27– October 1, 2019, Barcelona, Spain

3

2-HG=D-2-hydroxyglutarate; a-KG=alpha-ketoglutarate; AML=acute myeloid leukemia; FDA=Food and Drug Administration; Me=methyl groups; ND=newly-diagnosed; OS=overall survival; PFS=progression-free survival; R/R=relapsed/refractory.

• 1. Boscoe AN, et al. J Gastrointest Oncol. 2019;10:751-765. 2. Popovici-Muller J, et al. ACS Med Chem Lett. 2018;9:300-305. 3. TIBSOVO highlights of prescribing information.

https://www.accessdata.fda.gov/drugsatfda_docs/label/2019/211192s001lbl.pdf. Accessed August 5, 2019. 4. Lowery MA, et al. Lancet Gastroenterol Hepatol. 2019;4:711-720.

IDH1 mutations in advanced cholangiocarcinoma

▪ Advanced cholangiocarcinoma is an aggressive rare cancer with treatment options limited primarily to chemotherapy1 ▪ IDH1 mutations occur in up to 20% of cholangiocarcinoma and do not confer a favorable prognosis1 ▪ Ivosidenib (AG-120) is a first-in-class, oral, targeted, small-molecule inhibitor of the mutant IDH1 (mIDH1) protein,2 and is FDA-approved for mIDH1 R/R AML and ND AML not eligible for intensive chemotherapy3 ▪ A phase 1 study of ivosidenib included 73 previously treated mIDH1 cholangiocarcinoma patients and was associated with: median PFS, 3.8 months; 6- and 12-month PFS rates, 40.1% and 21.8%, respectively; and median OS 13.8 months4

Isocitrate a- KG

IDH1

NADPH mIDH1

Metabolic dysregulation Cytoplasm

Citrate

a-KG-dependent dioxygenases Nucleus Epigenetic changes Impaired cellular differentiation 2-HG 2-HG 2-HG

Me Me Me

4

ClarIDHy: Study design and endpoints

ECOG PS=Eastern Cooperative Oncology Group Performance Status; EORTC=European Organisation for Research and Treatment of Cancer; EQ-5D-5L=5-level EuroQoL-5 Dimension questionnaire; FU=fluorouracil; NGS=next-generation sequencing; PGI=Patient Global Impression; QD=once daily; QLQ-BIL21=Cholangiocarcinoma and Gallbladder Cancer module; QLQ-C30=Quality of Life Questionnaire Core 30; RECIST=Response Evaluation Criteria in Solid Tumors.

Key eligibility criteria

• ≥18 years of age
• Histologically confirmed diagnosis of cholangiocarcinoma
• Centrally confirmed mIDH1* status by NGS
• ECOG PS score 0 or 1
• 1-2 prior therapies (at least 1 gemcitabine- or 5-FU-

containing regimen)

• Measurable lesion as defined by RECIST v1.1
• Adequate hematologic, hepatic, and renal function

2:1 double-blind randomization (n=185) Ivosidenib 500 mg QD orally in continuous 28-day (±2 days) cycles (n=124) Placebo (n=61)

Crossover permitted at radiographic disease progression

Pre-screening for IDH1 mutation

NCT02989857 An independent data monitoring committee monitored the safety data throughout the study

▪ Primary endpoint: PFS by blinded independent radiology center (IRC) ▪ Secondary endpoints included: safety and tolerability; PFS by local review; OS; objective response rate; quality of life (QoL)†; pharmacokinetics/pharmacodynamics ▪ Sample size of ~186 patients based on hazard ratio (HR)=0.5, 96% power, 1-sided alpha=0.025 ▪ 780 patients were screened for IDH1 mutations across 49 sites and 6 countries

*IDH1 mutation status prospectively confirmed by NGS-based Oncomine™ Focus Assay on formalin-fixed, paraffin-embedded tumor tissue in a Clinical Laboratory Improvement Amendments-certified laboratory.

†Assessed using EQ-5D-5L, EORTC QLQ-C30, EORTC QLQ-BIL21, and PGI questions.

Stratified by number

• f prior therapies

ClarIDHy: Patient disposition

▪ As of the January 31, 2019 data cut, 35 placebo-treated patients (57.4%) crossed over to open-label ivosidenib upon radiographic disease progression and unblinding ▪ 26 placebo-treated patients (42.6%) did not cross over due to the following reasons: death (n=13), still on placebo treatment (n=8), never dosed (n=2), withdrawal of consent (n=2), received another treatment (n=1)

5

Ivosidenib (n=124) Placebo (n=61) Treated, n (%) 121 (97.6) 59 (96.7) On treatment 38 (31.4) 8 (13.6) Discontinued treatment 83 (68.6) 51 (86.4) Progressive disease 65 (53.7) 44 (74.6) Adverse events 6 (5.0) 4 (6.8) Death 4 (3.3) Withdrawal by patient 6 (5.0) 2 (3.4) Withdrawal of consent 1 (0.8) 1 (1.7) Other 1 (0.8) Not treated, n (%) 3 (2.4) 2 (3.3) On study, n (%) 71 (57.3) 27 (44.3)

7

ClarIDHy: PFS by IRC

PFS probability Survival (months)

0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 0.8 0.9 1.0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Censored Ivosidenib

+

Placebo

HR=0.37 (95% CI 0.25, 0.54) P<0.001

Number of patients at risk:

NE=not estimable; PR=partial response; SD=stable disease.

124 105 54 40 36 28 22 16 14 10 9 6 5 4 3 3 2 1 1 Ivoside nib 61 46 11 6 4 1 Placebo

Ivosidenib Placebo PFS Median, months 2.7 1.4 6-month rate 32% NE 12-month rate 22% NE Disease control rate (PR+SD) 53% (2% PR, 51% SD) 28% (0% PR, 28% SD)

8

0.37 0.41 0.219 0.234 0.612 0.730 0.36 0.45 0.220 0.249 0.589 0.811 0.20 0.41 0.035 0.277 1.111 0.601 0.38 0.257 0.567 0.26 0.52 0.124 0.332 0.540 0.803 0.40 0.39 0.42 0.249 0.188 0.110 0.631 0.830 1.597

ClarIDHy: Ivosidenib efficacy consistent across subgroups*

PFS by IRC

Favors ivosidenib Favors placebo Overall Prior lines of therapy 1 ≥2 Gender Female Male Extent of disease at screening Locally advanced Metastatic Cancer type at initial diagnosis Intrahepatic cholangiocarcinoma extrahepatic cholangiocarcinoma unknown ECOG PS score at baseline ≥1 Regions NorthAmerica Europe Asia 66/106 60/79 74/117 52/68 7/14 119/171 114/169 3/6 9/10 41/68 85/117 83/124 34/49 9/12 Lower 95% CI 0.252 Upper 95% CI 0.543

*Subgroups with events number ≤10 were not plotted.

1 2

Events/N Hazard ratio (HR) 126/185 HR 0.37

9

ClarIDHy: OS by intent-to-treat (ITT)

OS probability Survival (months) +

Placebo Censored* Ivosidenib Placebo(RPSFT-adjusted) Number of patients at risk:

▪ Median OS based on 78 events was numerically longer with ivosidenib than placebo (10.8 vs. 9.7 months) − OS rates at 6 and 12 months for ivosidenib: 67% and 48% vs. 59% and 38% for placebo ▪ With the RPSFT method, the median OS with placebo adjusts to 6 months ▪ Rank-preserving structural failure time (RPSFT)1,2 method used to reconstruct the survival curve for the placebo subjects as if they had never crossed over to ivosidenib

HR=0.69 (95% CI 0.44, 1.10); P=0.06 HR=0.46 (95% CI 0.28, 0.75); P<0.001 (RPSFT-adjusted)

124 11 7 101 88 75 64 52 49 39 34 30 23 19 16 15 10 9 7 4 3 1 1 1 Ivosidenib 61 55 45 39 34 25 22 19 17 17 14 12 5 4 4 3 2 2 1 1 1 Placebo 61 55 42 32 22 16 10 4 1 1 Placebo (RPSFT- adjusted) *Patients without documentation of death at the data cutoff date were censored at the date the patient was last known to be alive or the data cutoff date, whichever was earlier.

• 1. Watkins C, et al. Pharm Stat. 2013;12:348-357. 2. Robins JM, Tsiatis AA. Commun Stat Theory Methods. 1991;20:2609-2631.

0.0 0.5 0.4 0.3 0.2 0.1 0.6 0.7 1.0 0.9 0.8 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

10

ClarIDHy: Treatment-emergent adverse events (TEAEs)

Placebo (n=59) Ivosidenib (n=121) Total ivosidenib (n=156)* Any TEAE, n (%) 57 (96.6) 115 (95.0) 146 (93.6) Most common TEAEs, n (%) Nausea 15 (25.4) 43 (35.5) 50 (32.1) Diarrhea 9 (15.3) 37 (30.6) 45 (28.8) Fatigue 10 (16.9) 32 (26.4) 37 (23.7) Cough 5 (8.5) 25 (20.7) 30 (19.2) Abdominal pain 8 (13.6) 26 (21.5) 29 (18.6) Ascites 9 (15.3) 25 (20.7) 29 (18.6) Decreased appetite 11 (18.6) 23 (19.0) 27 (17.3) Anemia 3 (5.1) 18 (14.9) 25 (16.0) Vomiting 10 (16.9) 23 (19.0) 25 (16.0)

*Total ivosidenib includes 35 patients initially assigned to placebo who had crossed over to ivosidenib upon radiographic disease progression and unblinding. >15% TEAEs based on total ivosidenib

▪ Grade >3 TEAE: 35.6% for placebo vs. 46.2% for total

• ivosidenib. Most common (placebo vs. total ivosidenib):

ascites (6.8% vs. 7.7%), bilirubin increase (1.7% vs. 5.8%), anemia (0% vs. 5.1%), AST increase (1.7% vs. 5.1%) ▪ TEAEs leading to discontinuation were more common for placebo (8.5% vs. 5.8%) than total ivosidenib ▪ TEAEs leading to dose reductions (2.6% vs. 0%) and interruptions (26.3% vs. 16.9%) were more common for total ivosidenib relative to placebo