Umbrella, Basket and Platform trials

BASKET,
UMBRELLA AND
PLATFORM
TRIALS
Dr. Govind Mishra, MD
DM Resident (Clinical Pharmacology)
AIIMS Bhubaneswar

Background
• With increasing advancements in genomics, there have been increasing
interests in biomarkers and how they can be used to improve biomedical
interventions
• Biomarker-guided trials are becoming more popular to identify therapies
that can specifically affect disease targets based on their genetic make-up
– known as ‘targeted therapies’
Adapted from Baxter C. After the double helix. Nature Reviews Genetics. 2003 Apr 1;4(4):246-7.
1953 1958 1980 1983 1990 1999 2003 2017 2018
European
consensus
statement
released on CIDs
2019 2020
C
2
DNA
helix
structur
e
First DNA
replicatio
n
First
genome
sequenced
(FX174)
PCR
develope
d
Huma
n
genom
e
project
1st
chromoso
me
decoded
Human
genome
project
completed
First tumor-
agnostic therapy
approved by the
FDA
(Pembrolizumab)
FDA draft
guidance on
master
protocols
FDA draft
guidance
on
Complex
Innovative
Designs (CIDs)

Master
Protocol
A master protocol is defined as “a
protocol designed with multiple sub
studies, which may have different
objectives and involves coordinated
efforts to evaluate one or more
investigational drugs in one or more
disease subtypes within the overall trial
structure.”
• Often categorized as basket
trials, umbrella trials, and
platform trials
Master Protocols. Efficient Clinical Trial Design Strategies to Expedite Development of Oncology
Drugs and Biologics Guidance for Industry (Draft Guidance)

Types of Master protocols
Type of Trial Objective
Umbrella To study multiple targeted therapies in the context of
a single disease.
Basket To study a single targeted therapy in the context of
multiple diseases or disease subtypes.
Platform To study multiple targeted therapies in the context of
a single disease in a perpetual manner, with therapies
allowed to enter or leave the platform on the basis of
a decision algorithm.
Woodcock J, LaVange LM. Master Protocols to Study Multiple Therapies, Multiple Diseases, or
Both. N Engl J Med. 2017 Jul 6;377(1):62-70. doi: 10.1056/NEJMra1510062. PMID: 28679092.

Rising popularity of master protocols
60
90
80
70
50
40
30
20
10
0
2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019
Basket trials Umbrella trials
Platform trials
Number of Master Protocols over Time:
Basket Trials, Umbrella Trials, and Platform Trials
Cytel Inc. 6
Park JJ et al. BMC Trials. 2019 Dec;20(1):1-0.
Systematic review of basket trials, umbrella trials, and platform trials: a landscape analysis of master protocols.
Imatinib Target
Exploration
Consortium
Study B2225

What is a basket trial?
7
• A clinical trial that tests one or more targeted
interventions across multiple types of diseases
that share common molecular alterations and/or
other risk factors
• There are unifying eligibility criteria usually
based on predictive risk factors that combine
patients with different diseases into a single
“basket”
• Predictive risk factors are usually based on the
intervention’s mechanism of action since it can
help predict whether the patient will respond to a
specific intervention
A basket trial in cancer
• Multiple histological types of cancer
• Common targets as unifying
eligibility criteria
• Interventions agnostic to tumor and
histology

An illustrate example of a single-arm basket trial
Park JJH et al. 2020 Feb 7 CA: A Cancer Journal for Clinicians.
An overview of precision oncology basket and umbrella trials for clinicians

An illustrate example of a randomized basket trial

What is an umbrella trial?
10
• A clinical trial that tests multiple targeted interventions
for a single disease based on predictive biomarkers
and/or other risk factors
• In an umbrella trial, a single disease (e.g. breast
cancer) is stratified into multiple groups
• For example, the eligibility for each group can be
defined by the intervention’s mechanism of action
An umbrella trial in cancer
• A single histological cancer type
• Multiple targets used to stratify
patients
• Multiple interventions

An illustrate example of a single-arm
umbrella trial
Cytel Inc.
10

An illustrate example of a randomized
umbrella trial
Cytel Inc.
11

Illustrations of basket and umbrella trials
Cytel Inc.
12

Basket and umbrella trials
14
A basket trial in cancer An umbrella trial in cancer
• Multiple histological types of
cancer
• A single histological cancer type
• Common targets as unifying
eligibility criteria
• Multiple targets used to stratify
patients
• Interventions agnostic to tumor
and histology
• Multiple sub-cancer
interventions

Are basket
and umbrella
trials similar
in any way?
• In both trials, a common
molecular screening protocol is
used to determine eligibility
• Standardized biomarker assays are
used within the trial ecosystem –
features of master protocols
• Intervention assignment may
or may not be determined
using randomization in these
trials
15

Basket trials
• The inherent nature of basket trials may
be
described as “unification of diseases”
• Patients in a basket trial will represent
multiple diseases that share a common
unifying predictive risk factor
• Given that the disease sub-type is often a
prognostic factor, patient subgroups may be
defined based on the disease sub-types, but
they are usually not powered to detect
subgroups
Lung
cancer
Salivary gland
cancer
Ovarian
cancer
Biliary tract
cancer
Colorectal
cancer
Bladder
cancer
HER2 amplification
or mutation
Unification of diseases
Adapted from NCT02675829: “Trial of Ado-Trastuzumab
Emtansine for Patients With HER2 Amplified or Mutant Cancers"
Ado-trastuzumab
emtasine

Umbrella trials
• Umbrella trials have an inherent feature
of using multiple predictive risk factors to
stratify single-disease patients into
multiple groups (patient stratification)
• Each group is statistically powered as
each sub-study of the master protocol
Adapted from NICE’s advanced non-squamousnon-small
cell lung cancer systematic anti-cancer therapyguideline
17

Basket and umbrella trials: Trial design
characteristics
18
Trial design
characteristics*
Basket trials
(N = 49)
Umbrella trials
(N = 18)
Exploratory
(phase I or II)
96%
(n = 47)
89%
(n = 16)
Use of
randomization
10%
(n = 5)
44%
(n = 8)
Number of interventions
investigated
Median: 1
(IQR: 3-1 = 2)
Median: 5
(IQR: 6-4 = 2)
*Park JJ et al. BMC Trials. 2019 Dec;20(1):1-0.
Systematic review of basket trials, umbrella trials, and platform trials: a landscape analysis of master protocols.
Acronym: Interquartile range - IQR (Q3 –
Q1)

Basket and umbrella trials vs. other biomarker-guided trials
• Similar to other biomarker-guided trials, basket and
umbrella trials aim to use “omic” technologies to
define disease and eligibility criteria for improved
characterization and identification of predictive
biomarkers and targeted therapies
• Basket trials aim to identify histology-agnostic
therapies
• Traditionally, it is not uncommon for phase I trials to
recruit multiple tumors to test for the existence of
signal, but basket trials’ histology-agnostic
approaches are now being considered for phase 2
and 3 evaluations
Transcriptomics
mRNA
Proteomics
Protein
Metabolomics
Metabolite
Genomics
DNA
19

Basket and
umbrella
trials vs.
other
biomarker-
guided trials
• The use of a single master protocol
with standardized operating
procedures is a key difference!
• In umbrella trials, multiple histology-
dependent targeted therapies are
evaluated as different sub-studies that are
molecularly differentiated
• Use of a master protocol in an
umbrella trial allows for screening
efficiency
20

Screening for eligible biomarker-positive
population
Screening population
Cytel Inc.
Eligible biomarker
population
Screening

Basket and
umbrella trials
vs. other
biomarker-
guided trials
• For example, if we assume that 10% of
breast cancer patients will have the
biomarker of interest, an expected 1000
cancer patients will need to be screening
to reach the recruitment target of 100
patients
• In principle, each sub-study of an umbrella
trial can be conducted separately as a non-
master protocol, but conducting them
independently would require a much
larger number of patients that would
need to be screened.

Key design considerations for basket and umbrellatrials
23
Biological
plausibility
Accuracy of
biomarker
assays
Biospecimen
collection
Biomarker
prevalence
Sample size
calculations
Use of
randomization

Biological plausibility
24
Most important to consider the biological
plausibility of the targeted intervention strategies
being evaluated
It is common for cancers to have multiple genetic
mutations, but most mutations are passenger
mutations that do not affect the underlying
carcinogenic process.
Intervention strategies should be targeting driver
mutations, but it can be difficult to separate
driver mutations from passenger mutations

Accuracy of biomarker
assays
• Conceptually, a targeted intervention should be
more efficacious against the disease with the
biomarker target versus the disease without the
target
• Given that all medical tests have some degree of
diagnostic inaccuracy, a proportion of false positive
patients are expected in basket and umbrella trials
• In exploratory biomarker-guided trials, it has
been shown that increasing false positive rates of
biomarker tests will reduce statistical power

Biospecimen collection
• Ensuring adequate biospecimen collection
procedures and standards across multiple
trial institutions are important
• Centralized biospecimen processing and
molecular analyses established through a
master protocol can help
• Ease of biospecimen collection,
biospecimen quality, and biospecimen yield
should be similar between different tumors
26

Biomarker
prevalence
• Patient recruitment is a key determinant for
any clinical trials
• Biomarker prevalence will affect the size of
eligible patient pool. If the target
biomarker has a low prevalence,
recruitment challenges can be amplified
• Planning for comprehensive recruitment
strategies to reach the target sample size
within the desired trial duration will be
essential

Sample size calculations
• For basket trials, sample size calculations
may be done for the overall cohort.
• It can be difficult to differentiate
’responders’ and ‘non-responders’
between different disease sub-types
• For umbrella trials, sample size calculations
may be done for each of the sub-studies
given that multiple targeted interventions
are being evaluated

Use of randomization
• Randomization is generally preferable, as it can help determine
whether the risk factors being used as part of the targeted
intervention strategies are indeed predictive
• In single-arm basket and umbrella trials, it can be difficult to
differentiate between predictive and prognostic factors
• Statistical adjustments may be made on disease sub-types and/or
other prognostic risk factors, but adjustments are difficult in
smaller data sets

Use of
randomization
• Choice in control arm:
• If there are multiple standard-of-
care across different tumors, it
might be difficult to pick a single
control in a basket trial
• In an umbrella trial, each sub-study
can have its own control and
powered accordingly

Vemurafenib in Nonmelanoma Cancers Harboring BRAF
V600 Mutations
• *ECD = Erdheim-Chester
disease; LCH = Langerhans cell
histiocytosis.
Hyman DM, Puzanov I, Subbiah V, Faris JE, Chau I, Blay JY, Wolf J, Raje NS, Diamond EL, Hollebecque A, Gervais R, Elez-Fernandez ME, Italiano A, Hofheinz RD, Hidalgo M, Chan E,
Schuler M, Lasserre SF, Makrutzki M, Sirzen F, Veronese ML, Tabernero J, Baselga J. Vemurafenib in Multiple Nonmelanoma Cancers with BRAF V600 Mutations. N Engl J Med.
2015 Aug 20;373(8):726-36. doi: 10.1056/NEJMoa1502309. Erratum in: N Engl J Med. 2018 Oct 18;379(16)

Vemurafenib in Nonmelanoma Cancers Harboring BRAF V600
Mutations

Signature program: a platform of basket trials
• A series of 8 phase 2, agent-specific basket protocols.
• Each protocol evaluated 1 agent (buparlisib, dovitinib, binimetinib, encorafenib,
sonidegib, BGJ398, ceritinib, or ribociclib) in patients with solid or hematologic
malignancies and an actionable mutation.
• The primary endpoint - clinical benefit rate (ie, complete or partial response, or
stable disease) at 16 weeks.
• The most common tumor types among the 595 treated patients were colorectal
(9.2%), non-small cell lung adenocarcinoma (9.1%), and ovarian (8.4%).
• Frequent genetic alterations were in PIK3CA, RAS, p16, and PTEN. Overall, 30 partial
or complete responses were observed with 6 compounds in 16 tumor types.

Agents Target Mutations Required Tumor Types Excluded
Buparlisib (BKM120) Pan-Pl3K PIK3CA mutation/amplification, PTEN
mutation/loss, or PIK3R1 mutation
Endometrial, glioblastoma, NSCLC,
prostate, breast
Dovitinib (TKI258) Various RTKs FGFR1-3, FLT3, or c-
KIT mutation/amplification
or PDGFRα/β, VEGFR1-
2, RET, TrkA (NTRK1), or CSF-
1R mutation
Multiple myeloma, urothelial, AML
(FLT3+), hepatocellular, endometrial,
renal cell, breast (metastatic),
squamous NSCLC
Binimetinib
(MEK162)
MEK (RAS pathway) RAS, RAF, MEK1/MEK2, or NF1 Pancreatic, biliary, colorectal, ovarian
(low-grade serous), melanoma
Encorafenib
(LGX818)
BRAF BRAF V600E Melanoma, colorectal, primary CNS
Ceritinib (LDK378) ALK/ROS1 ALK/ROS1 mutation/amplification/tra
nslocation/rearrangement
ALK+ NSCLC
Signature program: a platform of basket trials

BATTLE trial
Design-
• BATTLE was a randomized phase II, single-center, open-label study
in patients with advanced NSCLC refractory to prior chemotherapy.
Study groups-
• Following molecular tumor biomarker assessments, patients were
randomly assigned to oral treatment with erlotinib (150 mg once
daily), vandetanib (300 mg once daily), erlotinib (150 mg once daily)
plus bexarotene (400 mg/m2 once daily), or sorafenib (400 mg
twice daily)

BATTLE trial
Primary end point- disease control rate (DCR) at eight weeks.
Secondary end points-
• Included response rate
• Progression-free survival (PFS)
• Overall survival (OS)
• Toxicity.
Planned exploratory objectives were each treatment's efficacy in
relation to patient biomarker profiles.

Results
• Sorafenib was active against tumors with
mutated or wild-type KRAS, but had a
worse DCR (compared with other study
agents) in patients with EGFR mutations.
• Erlotinib was beneficial in patients with
mutated-EGFR tumors. Erlotinib plus
bexarotene improved DC in patients with a
higher expression of cyclin D1 and also
improved DC in the KRAS-mutant patient
population.

• FOCUS4-D has shown no evidence of efficacy of single-agent EGFR,
HER2, and HER3 inhibition with AZD8931 (Sapitinib) in patients with
advanced colorectal cancer whose tumours are wild-type for BRAF,
PIK3CA, KRAS, and NRAS after first-line induction therapy.
FOCUS 4 Conclusion

Platform Trials
• A randomized design with a common control
arm and many different experimental arms
that enter and exit the trial as futility or
efficacy are demonstrated, often according to
Bayesian decision rules.

Platform
Trials
• The trial itself then comprises a platform or
standing infrastructure to which novel
therapies may be added or from which they
may be dropped.
• Adaptive randomization, i.e. mid-trial shifts
in the randomization ratios for patients with
a given biomarker signature to favor the
treatment showing the most promise in that
signature, may also be present.

I-SPY 2 TRIAL (Investigation of Serial studies to Predict The Therapeutic
Response with Imaging and Molecular Analysis 2, NCT01042379)
• Multicenter, open-label, adaptive phase 2 clinical trial design for women with high-risk clinical
stage II or III breast cancer.
• Goal- I-SPY 2 trial is to target rapid, focused, and individualized clinical development of promising
agents or agent combinations based upon breast cancer biomarker subtypes.
• The trial classifies breast cancer into 10 subtypes based on ER, HER2, and MammaPrint scores.
• Study population – Women with invasive tumor. ≥ 2.5 cm in Diameter.

I-SPY 2 TRIAL
• The control group receives standard neoadjuvant therapy consisting
of weekly intravenous paclitaxel (T) at a dose of 80 mg per m2 of
body surface area for 12 cycles, followed by 4 cycles of doxorubicin
(A) at a dose of 60 mg per m2 and cyclophosphamide (C) at a dose
of 600 mg per m2 , administered intravenously every 2 weeks to 3
weeks.
• The primary endpoint - pCR, defined as elimination of invasive
cancer in the breast and lymph nodes at the time of surgery.
• Drugs that increase pCR rates within a specific molecular subtype
will be preferentially assigned to new patients entering the trial with
that particular subtype allowing for a more rapid determination of
success. On the other hand, drugs doing poorly within a particular
subtype will be less likely assigned to that subtype.

I-SPY 2 TRIAL
• Criteria for Stopping enrolment in an experimental group-
Bayesian predictive probability of success reaches a
prespecified threshold (usually 85%) for any biomarker
signature- Agent/combination is deemed as “graduated”
from the trial.
Futility - if the probability falls to below 10% for all
biomarker signatures.

The steps involved in the enrollment, randomization, and treatment
process.

Veliparib/Carboplatin graduated with triple
negative signatures
Estimated qCR rates : 26% Estimated qCR : 52%
95% interval : 11% - 40% 95% interval : 35%-69%

Neratinib
graduated in
HR-/HER2+

Summary
• There is great promise that use of the Master Protocol design to
discover, explore, and test multiple biomarkers and therapeutic
regimens under one study mechanism will result in getting new and
effective therapies to patients more quickly.
• Methodological advancements in basket and umbrella trials will help
catalyze the adoption of precision medicine and oncology into clinical
practice.
• There are some important considerations that are need to be made
for these trials.

Umbrella, Basket and Platform trials

Recommended

Recommended

More Related Content

What's hot

What's hot (20)

Similar to Umbrella, Basket and Platform trials

Similar to Umbrella, Basket and Platform trials (20)

Recently uploaded

Recently uploaded (20)

Umbrella, Basket and Platform trials