Umbrella and Basket Trials

Young Kwang Chae, MD, MPH, MBA

Assistant Professor of Medicine

Co-Director, Developmental Therapeutics

Lurie Cancer Center

Contents

• Background

• Concept of Basket Trial and Examples

• Concept of Umbrella Trial and Examples

• Current and Future Trial Designs

• Lung cancer treatments

(NTRK targeted therapy)

Background

• Next Generation Sequencing

• More Targeted Therapies

• Innovative Clinical Trial Design

• Precision Medicine vs. Treatment designed for average patient

• Personalized Cancer Medicine

Precision Medicine Initiative

Traditional Trial Design

• Phase 1 (dose escalation)

• Phase IB (dose expansion)

• Phase II (non-randomized, randomized)

• Phase III (randomized)

• Histology Based

• Phase 1 (3+3 design)

Drug Development Pathway

Drug Approval

Phase II/III in one

histology

Phase I trial in one histology

Phase I in multiple

histologies

6

Cancer Trials

8

• HER2

• CD30

• B-RAF

• Signature

• NCI-MATCH

Basket Trial

• BATTLE

• LUNG-MAP

• Stand Up to Cancer Melanoma

Umbrella Trial

Traditional Histology Based Trial Histology Agnostic Aberration based Trial

Erika Hurwitz’s cancer of white blood

cells is now undetectable, after she

entered a trial that uses drugs that

block mutations

• BRAF-mutant disease basket trial

with vemurafenib

• Erdheim Chester disease

(polyostotic sclerosing

histiocytosis)

Basket Trials

Across different histologies

What is rare may not be rare any more.

Great for Rare Tumors

Can be marker driven or may not be marker driven

14

Nat Genetics 2013

Rare Tumors

• Definition: annual incidence ≤6 of 100,000, prevalence of ≤200K (Orphan Disease Act)

• Molecular subtypes (Crizotinib in 5% of NSCLC)

• Two third of FDA approval trials non-randomized single arm trials

• 70% of FDA approvals relied on objective response rate as the primary efficacy endpoint.

Gadipatti, CCR, 2012

Basket Trial for Rare Tumors

Navigation Trial

FDA Approval

Trial

Hypothesis Testing

Trial

Imatinib Basket Trial

Hypereosinophilic syndrome/PDGFR (14)

Mastocytosis PDGFR/Kit (5)

MDS/MPD PDGFR (7)

Dematofibrosarcoma protuberans/PDGFR (12)

Heinrich, CCR 2008

186 patients enrolled in 40 diseases Approval in the following:

Vemurafenib basket trial

Colorectal cancer 3%

Ovarian cancer 1%

Breast cancer 1%

Lung cancer 1%

Erdheim-Chester Disease 51%

Others unknown

MEDI4736 (12-179) multiple dose-expansion cohort

Microsatellite unstable tumors

Ovarian HPV+ tumors GBM

SCLC Gastroesophageal HCC Bladder cancer

HNSCC NSCLC TNBC

Immune Checkpoint Inhibitor Basket Trial for Rare Tumors

Navigation

Adaptive Design

Phase II

Approval

No biomarker of consensus Learn as we go model All rare tumors

Biomarker-based DNA repair defect High mutation burden

Basket Trial Molecularly Defined Orphan Diseases

Biomarker-based

Genetic marker positive

Prostate Endometrial Gastroesophageal Ovarian CRC Others

Biology will dictate arm selection.

Baskets in a Basket

The Program

The SIGNATURE program rapidly matches patients to treatments that target their tumor’s molecular abnormality and brings the trial to the patient rather than the patient traveling to a trial

Genetic profiling of tissue sample

Local CLIA certified laboratory

Relapsed/refractory cancer patient

Actionable mutation?

?

1http://www.signaturetrial.com. Accessed May 16, 2014.

Protocol package

• Fixed contract • Central IRB (Quorum) • Standard budget • Standard informed consent

Study open!

• 5.7 weeks vs 34-

week* average

Call center prequalifies the patient: (1) Protocol package sent to site (2) Expedited site visit

When a patient is identified as having an actionable mutation, their oncologist contacts Novartis

1-855-744-6727

How Does the Program Work?

NCI-MATCH scheme

Molecular Analysis for Therapy Choice

NCI MATCH

NCI-MATCH Patient Accrual Thru 04/23/2017

Screened % of Total

Less Common Cancers 2885 61.5% Common Cancers* 1809 38.5%

Total 4694

The trial has far exceeded the goal that 25% of the 6000 patients screened have rare or less common types of cancer

Common cancers screened: Colorectal 15.5%, Breast 12.8%, NSCLC 7.6%, Prostate 2.6%

NCI-MATCH Distribution of ~1100 Participating Sites

• Trial is open and enrolling in every state, DC, and Puerto Rico

NCI-MATCH First 15 of 30 Current Treatment Arms, By Prevalence Rate of Gene Abnormality

Arm Variant Prevalence Rate % Drug Opened Accrual Goal

I PIK3CA 3.47 Taselisib Feb ‘16 70

W FGFR 2.86 AZD4547 May ’16 70

Z1I BRCA1 or BRCA2 2.79 AZD1775 Mar ‘17 35

P PTEN loss 1.93 GSK2636771 Feb ‘16 35

Z1A NRAS 1.90 Binimetinib May ’16 70

S1 NF1 1.77 Mekinist™ Feb ‘16 70

N PTEN 1.75 GSK2636771 Feb ‘16

Z1D dMMR status 1.51 Opdivo® May ‘16 70

Q HER2 amplif. 1.49 Kadcyla® Aug ‘15 70

J HER2 amplif. 1.49 Herceptin® Perjeta® Mar ‘17 35

Z1C CDK4 or CDK6 1.36 Ibrance® Mar ‘17 35

M TSC1 or TSC2 1.11 TAK-228 Mar ‘17 35

B HER2 activating 1.04 Gilotrif® Aug ‘15 70

Z1B CCND1/2/3 0.84 Ibrance® May ‘16 70

R BRAF fusions 0.80 Mekinist™ Aug ‘15 35

NCI-MATCH Remaining 15 of 30 Current Treatment Arms, By Prevalence Rate of Gene Abnormality

Arm Variant Prevalence Rate % Drug Opened Accrual Goal

Y AKT 0.77 AZD5363 May ‘16 35

H BRAF V600 E/K 0.69 Taflinar® Mekinist™ Aug ‘15 35

U NF2 loss 0.69 Defactinib (VS-6063) Aug ‘15 35

C2 MET exon 14 0.61 Xalkori® May ‘16 35

C1 MET amplif. 0.51 Xalkori® May ‘16 35

T SMO/PTCH1 0.42 Erivedge® Feb ‘16 35

L mTOR 0.31 TAK-228 Mar ‘17 35

S2 GNAQ/GNA11 0.16 Mekinist™ Feb ‘16 35

E EGFR T790M 0.11 AZD9291 Aug ‘15 35

V cKIT 0.11 Sutent® Aug ‘15 35

Z1E NTRK 0.10 Larotrectinib Mar ‘17 35

G ROS1 0.05 Xalkori® Aug ‘15 35

A EGFR activating 0.05 Gilotrif® Aug ‘15 35

F ALK 0.03 Xalkori® Aug ‘15 35

X DDR2 0.00 Sprycel® Feb ‘16 35

Levels of Evidence for Target Selection in NCI-MATCH

• Level 1: Gene variant credentialed for selection of an approved drug

• Level 2: Variant is eligibility criterion for an ongoing clinical trial for

that drug OR variant identified in an N of one response

• Level 3: Preclinical inferential data

– Models with variant respond; without variant do not

– Gain of function mutation demonstrated in preclinical model

– Loss of function (tumor suppressor genes or pathway inhibitor e.g.

NF1); stop codon or demonstrated loss of function in pre-clinical

model

Levels of Evidence for Drugs in NCI-MATCH

• Level 1: FDA-approved for any indication for that target

• Level 2: Agent met a clinical endpoint (objective response, PFS, or

OS) with evidence of target inhibition

• Level 3: Agent demonstrated evidence of clinical activity with

evidence of target inhibition at some level

FGFR pathway disruption

FGFR Implicated cancers

FGFR fusions in Lung and Other

Cancers

Capelletti 2014, CCR

Kim 2014, JCO

Anti-FGFR agents in development

Dieci et al. Cancer Discovery 2013

A Phase II Study of AZD4547 in Patients with Tumors with Aberrations in the FGFR Pathway Molecular Analysis for Therapy Choice (MATCH) Arm WYoung Kwang Chae, Christos Vaklavas, Heather H. Cheng, Fangxin Hong, Lyndsay Harris, Edith Mitchell, James Zwiebel, Larry Rubinstein, Lisa McShane, Robert Gray, Shuli Li, Percy Ivy, Sherry Ansher, Stanley Hamilton, Mickey Williams, David Patton, James Tricoli, Carlos Arteaga, Barbara Conley, Peter O'Dwyer, Alice Chen, Keith Flaherty

1Young Kwang Chae

Waterfall plot of best responses

16

No. of patients

(% among

evaluable)

PR 4 (9.5)

SD 20 (47.6)

PD 18 (42.9)

Not

evaluable

8

Total 50

Young Kwang Chae

38 cases on the plot (8, unevaluable; 5, PD due to new lesions without target lesion measurements

Molecular characteristics of PR cases

Disease site Variant Response PFS

Intrahepatic

cholangiocarcinoma

FGFR2

Y375C

69% 334

days

Transitional

cell/urothelial

carcinoma of renal

pelvis

FGFR3-

TACC3.F17T8

70% 304

days

Squamous cell

carcinoma of cervix;

p16+, HPV ISH+, HPV-associated

FGFR3-

TACC3.F17T11

33% 156

days

Transitional cell

carcinoma of bladder

FGFR3

A391

35% 166

days

17

Amp SNV Fusion Total

PR 0 2 2 4

SD 10 5 5 20

PD 7 9 2 18

Not

Eval 4 4 0 8

Total 21 20 9 50

Young Kwang Chae

Swimmer plot of duration of treatment

• Treatment duration for 24 patients who achieved PR (n=4) and SD (n=20).

• 5 cases with PFS times > 168 days (5.5 months; 168=28*6): 2 PRs and 3 SDs

• 2 PRs: Intrahepatic cholangiocarcinoma (FGFR2 Y375C), Transitional cell/urothelial carcinoma of renal pelvis (FGFR3-TACC3.F17T8)

• 3 SDs: Mucinous adenocarcinoma of colon (FGFR1 Amp), Epithelial-myoepithelial carcinoma of parotid/submandibular gland (BAG4-FGFR1.B2F6), Transitional cell carcinoma of renal pelvis (FGFR3 S249)

18Young Kwang Chae

The colored triangle marked when PR started (blue) or PD reported (yellow).

How can we do better?

How to Select Best Patients?

(Optimal Biomarker)

How to Reduce Screen Failure? (Basket Trial & Umbrella Trial)

How to Overcome Resistance?

(Rational Combiation)

NCI MATCH & SWOG DART

DART: Dual Anti-CTLA-4 & Anti-PD-1 blockade in Rare Tumors

Young Chae, MD, MPH, MBA Assistant Professor

Vice Chair, SWOG Early Therapeutics and Rare Cancers Committee

Co-Director Developmental Therapeutics Lurie Cancer Center

Northwestern University

Razelle Kurzrock, MD Professor

Chief, Division of Hematology, Medical Oncology

Chair, SWOG Early Therapeutics and Rare Cancers Committee UCSD Moores Cancer Center

Sandip Patel, MD Assistant Professor

Deputy Director, San Diego Center for Precision Immunotherapy

UCSD Moores Cancer Center

The Cancer Letter 2017

Rare Tumors Basket Study

SWOG DART Treatment/ Schema

• Basket study in rare tumors • Concurrent Combination Immunotherapy:

Ipilimumab 1 mg/kg IV every 6 weeks and nivolumab 240mg IV (fixed dose) every 2 weeks – Nivolumab monotherapy permitted for patients who experience severe immune-related toxicity on combination

ipilimumab/nivolumab

• Treatment cycle length: 6 weeks

• Imaging assessments: q8 wks until Wk 25, then q12 weeks until 3 years

Ipilimumab 1mg/kg IV q6wks

Nivolumab 240mg IV q2wks

PD, unacceptable toxicity, or withdrawal of consent

• Sample size = Minimum 50 pts. / 31 histologic cohorts + unknown primary + NOC cohort (Coordinated referral/accrual strategy from MATCH – MATCH screening/treatment failures that meet accrual criteria will be offered DART therapy at time of screen failure/discontinuation of last eligible MATCH therapy). o Accrual to cohorts will be monitored by SWOG/NCI monthly o Approximately 20% of NCI-MATCH patients met histologic criteria

• Two Stage Design: 87% power with a one-sided alpha of 13% in

each subtype o First stage: 6 eligible patients per histologic subtype • If no response is observed, accrual to that histologic subtype will be

permanently closed. • If ≥ 1 response is observed, an additional 10 patients will be accrued

in the second stage. o Second stage: 2 or more responses out of 16 will be considered

evidence that the combination regimens warrants further study in the histologic subtype • With 16 eligible patients in a histologic subtype, any toxicity with at least

a 10% chance of occurring has an 81% chance of being observed at least once.

Statistical Considerations

DART Activated: 1/13/17; First Patient Treated: 1/30/17

As of 3/16/2019:

• 875 sites approved to enroll through CTSU

• Total enrollment: 560 patients

• 37 Cohorts originally

• 53 cohorts in upcoming amendment 5 including PD-L1 amplified cohort

DART To Date

Potential outcomes

• New treatment options

• New FDA approval

• New NCCN guideline treatment options

• Neuroendocrine carcinoma of any histology origin

• Adrenal Cortical Carcinoma, etc.

Precision Immuno-oncology PD-L1 IHC Immune biomarkers Germline DNA

sequencing

Proteomic immune

signature

cDNA sequencing Tumor DNA/RNA

Performing Lab CIMACs CIMACs Counsyl Biodesix Circulogene MatchBox and CIMACs

WES/RNASeq

Sample source Tumor tissue (FFPE) or

unstained slide

Blood in collected in Tempus

tubes

(one 2cc vial for RNA, another

2cc vial for DNA)

Blood collected in the

EDTA tube

Blood collected in

the EDTA tube

Blood collected in the

EDTA tube

Tumor tissue (FFPE)

collected as part of NCI-

MATCH

Biomarker Target PD-L1 protein expression

by 28-8 IHC analysis

DNA, RNA sequencing 0of

tumor tissue and blood

Leukocyte DNA

sequencing (Illumina)

Serum proteins Cell free DNA sequencing

(Illumina)

Tumor next-generation

sequencing

(Ion Torrent)

Specimen Estimate 150 (baseline tissue) 240 (baseline blood) 240 (baseline blood) 240 (baseline

blood)

240 (baseline blood) 300 (baseline tissue)

Biomarker output PD-L1 strata will be

grouped <1%, 1-5%, 6-

25%, 26-49%, >50%

Immune and Cancer pathway

Nanostring (gene expression of

770 genes assaying 24 immune

cell types and 500 immune

response genes)

Genetic alteration Predictive

signature (good,

intermediate, poor

group)

Genetic alteration and

mutational load

Genetic alteration and

mutational load

Statistical Considerations Binary endpoint by strata Log-expression Categorical variable Categorical variable Percentile rank of

mutational load

Percentile rank of

mutational load

Sample time points Tissue: Baseline Tissue: baseline

Blood: DNA and RNA at three

time points

Blood at baseline Blood: at three

time points

Blood: at three time

points

Tissue: baseline

Other Innovative Trial Designs

• Continuous Reassessment Method (CRM)

• Efficacy-Toxicity Method (EffTox)

• Seamless Phase 1/2/3 Design with interim analyses

• Octopus Design (multiple histology/dose/regimen arms in phase ½ trial)

• Platform Based (BATTLE, I-SPY, MATCH, Lung-MAP)

NTRK as a target

NTRK fusion across tumor types

NRTK fusion assays

NRTK and lung cancer

NTRK 1 & 3 0.23% of all lung cancers

Current NTRK inhibitors

• Larotrectinib (US FDA approval for all tumors including lung cancer with NTRK 1/2/3 fusions. Nov. 2018): ORR = 75% (22% CR and 53% PR), DOR >12 months = 39%

• Entrectinib

• LOXO-195

• TPX-0005

Take Home Message

• Histology agnostic basket trials or histology & pathway based umbrella trials are new innovative ways to develop treatments for many “molecular orphan diseases”.

• Upfront or timely NGS testing is recommended to explore actionable genomic alterations.

• NRTK fusions are found to be a new target in many cancers including lung cancer.