Transcript Phase I Trials

Phase I Trials of Chemotherapy and
Targeted Agents
ELIZABETH GARRETT-MAYER
(SOME SLIDES BY PAT LORUSSO
OF KARMANOS CANCER
INSTITUTE
WAYNE STATE UNIVERSITY)
Phase I Trials
PHASE I TRIALS ARE CLINICAL EXPERIMENTS
If you don’t think your experiment through well before embarking on it, if
you don’t identify the variables and account for them, the experiment may
not be a success
TEAM APPROACH
You need a research team, including preclinical scientists,
pharmacologists, medical oncologists, biostatisticians, to plan your study.
Phase I Trials
MUST BE REALISTIC WITH QUESTIONS YOU
WANT TO ANSWER
MANY QUESTIONS IN PHASE I ARE CURRENTLY
EXPLORATORY DUE TO SEVERAL FACTORS
Limited patient numbers
Multiple doses with small patient numbers
Unknown assay
Heterogeneous patient population
Definition(s) of a Phase I Trial
First evaluation of a new cancer therapy in humans
◦ First-in-human, first-in-kind (e.g. the first compound
ever evaluated in humans against a new molecular
target), single-agent
◦ First-in-human, but not first-in-kind (i.e. others
agents of the same class have entered human
testing), single-agent
Definition(s) of a Phase I Trial
Multiple types of Phase I Trials
◦ Investigational agent + investigational agent
◦ Investigational agent + approved agent(s)
◦ Approved agent + approved agent(s)
◦ Approved or investigational agent with pharmacokinetic
focus (adding of CYP inhibitor)
◦ Typically considered drug-drug interaction study
◦ Approved or investigational agent with pharmacodynamic
focus (e.g. evaluation using functional imaging)
◦ Approved or investigational agent with radiotherapy
◦ Food effect study
◦ QTc prolongation study
◦ Bioequivalence study
We are in a revolutionary state
of affairs in phase I in cancer
Nivolumab: Phase I trial ended up with ~300 patients
Pembrolizumab: Phase I trial ended up with ~1300 patients.
Novel immunotherapies are not ‘behaving’ like traditional
anti-cancer therapies.
Studies designs have changed but it’s not clear how to most
efficiently tackle the first two phases of drug development
with these novel agents.
Objectives of a Phase I Trial
Primary objective:
◦ Identify dose-limiting toxicities (DLTs) and the recommended
phase II dose(s) (RP2D)
Newer Primary objective:
◦ Identify the optimal dose—induces a clinical effect and
operates with acceptable toxicity
Secondary objectives:
◦
◦
◦
◦
Define toxicity profile of new therapy(s) in the schedule under evaluation
Pharmacokinetics (PK)
Pharmacodynamic (PD) effects in tumor and/or surrogate tissues
Antitumor activity
1.0
Classic Phase I Assumption:
Efficacy and toxicity both increase with dose
DLT =
doselimiting
toxicity
0.8
0.6
0.4
0.2
0.0
Probability of Outcome
Response
DLT
1
2
3
4
Dose Level
5
6
7
Definitions of Key Concepts in Phase I Trials
Dose-limiting toxicity (DLT):
◦ Toxicity that is considered unacceptable (due to severity and/or
irreversibility) and limits further dose escalation
◦ Defined with standard criteria CTCAE 4.1
◦ Dose-limiting toxicity
◦ defined in advance prior to beginning the trial
◦ is protocol-specific
◦ Typically defined based on toxicity seen in the first cycle
◦ With select agents that have a more delayed toxicity (eg
2nd/3rd cycle), time allowed for DLT definition being reevaluated
Definitions of Key Concepts in Phase I Trials
Examples of DLTs – chronic (daily) dosing:
◦ Threshold for DLTs is lower
◦ Some Grade 2 toxicities may be unacceptable and intolerable due to their
persistence and lack of time period for recovery
◦ Examples:
◦ Grade 2 intolerable or worse non-hematologic toxicity despite supportive measures
◦ Grade 3 or worse hematologic toxicity
◦ Inability to complete a pre-specified percentage of treatment during the cycle due to toxicity (e.g.
missing 10-15% of doses)
Two examples from HCC trials
Karim’s study: DLT will be defined as any one of the following attributed to AZD5363/AZD1208 during course 1: grade 4
neutropenia ≥ 1 week; febrile neutropenia; grade 4 thrombocytopenia; grade 3 thrombocytopenia with bleeding; QTcF
prolongation to > 500 msec, or an increase of > 60 msec from baseline QTcF to a QTcF value > 480 msec, confirmed on repeat EKG;
grade 3 or 4 nausea, vomiting, or diarrhea despite use of adequate medical intervention; any other clinically significant ≥ grade 3
non-hematologic toxicity; and/or persistent, intolerable toxicity which delays scheduled treatment for > 14 days. All subjects will be
evaluated for toxicity by history, physical exam, and clinical labs weekly during course 1, every 4 weeks during subsequent courses,
and at EOS. Subjects will undergo imaging every 2 courses (approximately 8 weeks), and will remain on study as long as they are
deriving clinical benefit, tolerating the combination, and consent to continue. Blood samples will be drawn at frequent intervals
during course 1 to assess the pharmacokinetic behavior of AZD5363 + AZD1208. Disease-specific cancer markers, AKT-regulated
biomarkers, and PIM-regulated biomarkers will be evaluated pre-and post-treatment. One specific toxicity concern is
hyperglycemia, which has been noted with both agents.
Adoptive T-cell transfer study: The DLT in our studies will be identical to DLT defined in NCI Surgery Branch adoptive T cell transfer
protocols, which is 1) ≥Grade 2 or more bronchospasm or generalized uticaria (hypersensitivity), 2) ≥Grade 2 or more allergic
reactions, 3) Grade 3 or greater toxicity occurring within 24 hours post cell infusion (related to cell infusion) and not reversible to a
grade 2 or less within 8 hours with two doses of 650 mg po of acetaminophen or two doses of 50 mg po of dyphenhydramine, 4)
any Grade 4 autoimmunity, 5) Grade 3 autoimmunity (excluding vitiligo, the depigmentation of the skin and/or hair) that cannot be
resolved to less than or equal to a grade 2 autoimmune toxicity within 10 days, or 4) any grade 3 or greater non-hematologic
toxicity, excluding injection site reactions, skin rash, pruritis, and local adenopathy (Severe skin rashes such as Steven-Johnson
Syndrome or TEN will be considered DLT), 6) > Grade 3 toxicity of any kind related to the TIL 1383I TCR transduced T cells
administration with particular attention to the following events: a) ≥Grade 3 injection site reaction due to TIL 1383I TCR transduced
T cells administration, b) > Grade 3 hematological or hepatic toxicity that does not subside within 4 weeks after infusion of TIL
1383I TCR transduced T cells, c) > Grade 3 neurotoxicity: New motor or sensory deficits, encephalopathy, or signs and symptoms
that may indicate either tumor progression or a productive immune response will be evaluated as noted in earlier in this section
with radiological imaging, and, if warranted, biopsy.
Definitions of Key Concepts in Phase I Trials
Maximum Administered Dose (MAD), Maximum Tolerated Dose
(MTD): confusing
◦ Usage of these 2 phrases varies with country
More important term: Recommended phase II dose (RPTD or
RD):
◦ Dose associated with DLT in a pre-specified proportion of patients; dose
that will be used in subsequent phase II trials
◦ Often the dose that is selected before launching to phase II is based on
much more than just DLTs in cycle 1
◦
◦
◦
◦
Cycle 2 and beyond DLTs and other AEs
PK profile
Clinical outcomes
Immune responses
◦ Aside: phase I/II study
◦ What is gained and what is lost by combining two phases in one
protocol?
Phase I Trial Design: Classic Algorithmic
Approach

“Standard” Phase I trials (in oncology) use what is often
called the ‘3+3’ design (aka ‘modified Fibonacci’):
Treat 3 patients at dose K
1. If 0 patients experience dose-limiting toxicity (DLT), escalate to dose K+1
2. If 2 or more patients experience DLT, de-escalate to level K-1
3. If 1 patient experiences DLT, treat 3 more patients at dose level K
A. If 1 of 6 experiences DLT, escalate to dose level K+1
B. If 2 or more of 6 experiences DLT, de-escalate to level K-1



Maximum tolerated dose (MTD) is considered highest dose
at which 1 or 0 out of six patients experiences DLT.
Doses need to be pre-specified
Confidence in MTD is usually poor.
Phase I Trial Design: “Modified Fibonacci “
Dose Escalation (Algorithmic)
Attributed to a merchant from the 13th century
◦ Mathematical Model having nothing specifically to do with drug
development. Sounds fancy, but no real basis.
Doses increase by: 100%, 66%, 50%, 40%, 33%, etc.
Standard “3+3” design: 3 patients per cohort, escalating to 6 if
DLT occurs
Dose escalate until DLT observed and MTD/RP2D defined
Advantages:
◦ relatively safe, straightforward, clinician-friendly
Disadvantages:
◦ lacks statistical foundation and precision, potentially treating a large
proportion of patients at sub-therapeutic doses, time consuming
◦ Low confidence in selected dose as ‘safe’ or even near the desired DLT
rate
Phase I Trial Design: Standard 3 + 3 Design
MAD
Dose
DLT
Recommended dose
(some call this MTD in US)
3 pts
3 pts
3 pts
3 pts
3 pts
+
3 pts
DLT
3 pts
Definitions of Key Concepts in Phase I Trials
Optimal biological dose (OBD):
◦ Dose associated with a pre-specified desired effect on a
biomarker
◦ Examples:
◦ Dose at which > XX% of patients have inhibition of a key target in
tumor/surrogate tissues
◦ Dose at which > XX% of patients achieve a pre-specified immunologic
parameter
◦ Challenge with defining OBD is that the “desired effect on
a biomarker” is generally not known or validated before
initiation of the phase I trial
Definitions of Key Concepts in Phase I Trials
Pharmacokinetics (PK):
◦ “what the body does to the drug”
◦ absorption, distribution, metabolism and excretion
◦ PK parameters: Cmax, AUC (drug exposure), t1/2,
Clearance, etc.
Pharmacodynamics (PD):
◦ “what the drug does to the body”
◦ e.g. nadir counts, non-hematologic toxicity, molecular
correlates, imaging endpoints
Phase I Trials: Fundamental Questions
At what dose do you start?
What type of patients?
How many patients per cohort?
How quickly do you escalate?
What are the endpoints?
Phase I Trials: Fundamental Questions
At what dose do you start?
What type of patients?
How many patients per cohort?
How (quickly) do you escalate?
What are the endpoints?
Preclinical Toxicology
Typically a rodent (mouse or rat) and non-rodent (dog or
non-human primate) species
Reality of animal organ specific toxicities – very few
predict for human toxicity
◦ Myelosuppression and GI toxicity more predictable
◦ Hepatic and renal toxicities – large false positive
Toxicologic parameters:
◦ LD10 – lethal dose in 10% of animals
◦ TDL (toxic dose low) – lowest dose that causes any
toxicity in animals
◦ NOAEL – no observed adverse effect level
Phase I Trials: Starting Dose
1/10 of the LD10 in rodents
or
(depending on sensitivity of the species)
1/6 or 1/3 of the TDL in large animals
Unless preclinical studies suggest a very steep
dose/toxicity curve
Phase I Trials: Fundamental Questions
At what dose do you start?
What type of patients?
How many patients per cohort?
How (quickly) do you escalate?
What are the endpoints?
Phase I Patient Population
“Conventional” eligibility criteria- examples:
◦ Advanced solid tumors (or select hematologic
malignancies) unresponsive to standard therapies or for
which there is no known effective treatment
◦ Performance status (e.g. ECOG 0 or 1)
◦ Adequate organ function (e.g. ANC, platelets, creatinine,
AST/ALT, bilirubin)
◦ Specification about prior therapy allowed
◦ Specification about time interval between prior therapy
and initiation of study treatment
◦ No serious uncontrolled medical disorder or active
infection
Phase I Patient Population
“Agent-specific” eligibility criteria - examples:
◦ Restriction to certain patient populations – must have strong scientific
rationale
◦ Specific organ functions:
◦ Cardiac function (e.g. QTc < 450-470 ms, LVEF > 45%, etc)
◦ if preclinical or prior clinical data of similar agents suggest cardiac risks
◦ No recent (6-12 months) history of acute MI/unstable angina, cerebrovascular events,
venous thromboembolism; no uncontrolled hypertension; no significant proteinuria (eg.
antiangiogenic agents)
◦ Prohibited medications if significant risk of drug interaction
How has this occurred in the era of immunotherapies (e.g. PD-1/PD-L1
checkpoint blockade?)
Phase I Trials: Fundamental Questions
At what dose do you start?
What type of patients?
How many patients per cohort?
How (quickly) do you escalate?
What are the endpoints?
Number of Patients per Cohort
Predicated on several factors
◦ What is driving the definition of ‘optimal dose’? That
is, what is the primary endpoint?
◦ For toxicity-driven designs
◦ Usually 3 per cohort for 3+3 and even model-based.
◦ Accelerated Titration Design attempts to minimize Pt #
◦ Typically 1 patient per cohort until >grade 2 toxicity and then increase to
3 (or 6 if DLT)
◦ For efficacy-driven designs
◦ Usually more than 3 per cohort, but trial-dependent.
Cohort Dose Escalation: Standard 3 + 3 Design
# of pts with DLT
0/3
1/3
1/3 + 0/3
1/3 + 1/3
1/3 + 2/3
1/3 + 3/3
2/3
3/3
Action
Increase to next level
Accrue 3 more pts at same dose level
Increase to next dose level
Stop: recommend previous dose level
Stop: recommend previous dose level
Stop: recommend previous dose level
Stop: recommend previous dose level
Stop: recommend previous dose level
Many phase I trials accrue additional patients at the RP2D to
obtain more safety, PK, PD data (but this expansion cohort does
not equal to a phase II trial)
“EXPANSION COHORT”
Phase I Trials: Fundamental Questions
At what dose do you start?
What type of patients?
How many patients per cohort?
How (quickly) do you escalate?
What are the endpoints?
Phase I Trial Basic Principles
Start with a safe starting dose
◦ Often you start with ‘dose level 2’ and define dose level 1 as a ‘step-down dose’ if the
starting dose is too toxic. Easier than amending the trial to add a lower level.
Minimize the number of pts treated at sub-toxic (and thus maybe subtherapeutic) doses
Escalate dose rapidly in the absence of toxicity
Escalate dose slowly in the presence of toxicity
Different therapies have difference ‘increments’
◦
◦
Standard chemo: you might double the dose for the next dose level
Adoptive T-cell transfer: you might increase dose by factor of 10. (e.g. dose level 1
= 1x107; dose level 2 = 1x108)
“Old School” Phase I Trial Assumption
The higher the dose, the greater the likelihood of efficacy
◦ Dose-related acute toxicity regarded as a surrogate for efficacy
◦ Highest safe dose is dose most likely to be efficacious
◦ This dose-effect assumption is primarily for cytotoxic agents and may not
apply to molecularly targeted agents
Modern situations where this isn’t true:
◦ Nivo and Pembro: Pretty flat dose response curves
◦ Temsirolimus in kidney cancer (Atkins et al., JCO, 2004)
◦ Anastrazole in breast cancer (Jonat et al., Eur J Cancer, 1996)
Phase I Trial Design: Accelerated Titration Design
(Rule-based)
First proposed by Simon et al (J Natl Cancer Inst 1997)
Several variations exist:
◦
◦
◦
◦
Doubling dose in single-patient cohorts till Grade 2 toxicity
then revert to standard 3+3 design using a 40% dose escalation
Modifications are often trial/drug dependent
Intra-patient dose escalation allowed in some variations
Advantage
◦ More rapid dose escalation
◦ Fewer patient numbers
◦ Typically does not expedite trial completion
Phase I Trial Design: Accelerated Titration
Dose
DLT
Recommended dose
Gr 2 toxicity
3 pts
3 pts
1 pt
1 pt
1 pt
Starting dose
+
3 pts
DLT
3 pts
Phase I Trial of Combination of Agents
Combination phase I trials:
◦ New drug A + Standard drug B
◦ Need to provide rationale: why add A to B?
◦ Need to think about overlapping toxicity in your definition of DLT
◦ Ideally keep standard drug dose fixed and escalate the new drug
(e.g. 1/2, 2/3, full dose)
Phase I Trial of Combinations of Agents
Relative dose
Relative dose
A
1
B
2
3
4
5
DL
1
C
1
2
3
4
DL
D
2
3
4
DL
1
2
3
4
5
DL
Pitfalls of Phase I Trials
Chronic and cumulative toxicities usually
cannot be assessed & may be missed
◦ Most patients do not stay on trial beyond 2
cycles
Uncommon toxicities will be missed
◦ Too few patient numbers in a phase I trial
◦ Reason for toxicity evaluation & reporting
through phase IV drug testing
Phase I Trials: Fundamental Questions
At what dose do you start?
What type of patients?
How many patients per cohort?
How quickly do you escalate?
What are the endpoints?
Endpoints in Phase I Trials
DLT and other toxicity – safety and tolerability
Pharmacokinetics (PK)
Pharmacodynamics (PD)
◦ biological correlates, imaging endpoints
Preliminary antitumor activity
Expansion cohort: Have become THE NORM in phase I studies.
◦ Historically: treat a few more patients at MTD to ‘confirm’ safety, etc.
◦ More recently: ‘phase II’-like but without clear objectives and endpoints (NB:
statisticians do not like unclear objectives!).
Response Rates and Deaths from Toxic Events in Phase I Oncology Trials Involving the First Use
of Agents in Humans
(Horstman et al, NEJM 352, 2005)
Trial
No. of
Trials
No. of
Patients
Assessed for
Response
Overall
Response
Rate* %
No. of Patients
Assessed for
Toxic Events
Deaths from
Toxic Events
no. %
Total
First use of an agent in
humans
117
3164
4.8
3498
9 (0.26%)
Cytotoxic chemotherapy
First use of an agent in
humans
43
1298
5.0
1422
7 (0.49%)
Immunomodulator
First use of an agent in
humans
16
404
7.4
431
1 (0.23%)
27
742
3.8
853
1 (0.12%)
Antiangiogenesis
First use of an agent in
humans
8
200
7.0
228
0
Gene transfer
First use of an agent in
humans
0
0
0
0
0
Vaccine
First use of an agent in
humans
23
520
3.1
564
0
Receptor or signal
transduction
First use of an agent in
humans
Clinical Development of Molecularly Targeted
Agents: Known Challenges
Patient Pre-selection
◦ Other than typical Inclusion/Exclusion Criteria
◦ Tumor type
◦ Eg. Gorlin’s syndrome, basal cell carcinoma – HH inhibitors
◦ Molecular Profiling
◦ bRAF melanoma
◦ ALK+/EGFR+ NSCLCa
◦ What about other targets?
Clinical Development of Molecularly Targeted
Agents: Known Challenges
General requirement for long-term administration: pharmacology and
formulation critical
Difficulty in determining the optimal dose in phase I: MTD versus OBD
◦ Do we need MTD with targeted agents?
Absent or low-level tumor regression as single agents: problematic for
making go no-go decisions
Need for large randomized trials to definitively assess clinical benefit:
need to maximize chance of success in phase III
Biological Correlative Studies:
Why Do We Need Them?
Conventional cytotoxic drugs have led to predictable effects
on proliferating tissues (neutropenia, mucositis, diarrhea),
thus enabling dose selection and confirming mechanisms of
action
Targeted biological agents may or may not have predictable
effects on normal tissues and often enter the clinic needing
evidence/proof of mechanisms in patients
◦ On target vs off-target effects
Biomarkers: Functional Definitions
Prognostic Markers
Correlate with disease outcome
regardless of intervention
e.g. Clinical: stage, PS
e.g. ER in breast cancer
e.g. HPV in H&N cancer
Predictive Markers
Predict outcome with specific
therapy - match drugs with
appropriate pts
e.g. KRAS mutations and antiEGFR monoclonal antibodies in
colorectal cancer
e.g. HER2 in breast cancer
Pharmacodynamic Markers
 Confirm biological activity
 e.g.  pERK with a
targeted agent (such as a
MEK inhibitor)
Use of Lab Correlates in Clinical Trials
Should be driven by sound scientific rationale
Phase I:
◦ Proof-of-mechanism
◦ Exploratory
◦ Establish optimal biological dose and/or schedule in some trials
(especially if little or no toxicity expected)
◦ Often more practical to perform at an expanded cohort at the
recommended phase II dose
PAR Levels in Tumors of Predose vs. Postdose
Pre ABT
PAR Levels in Tumor
PAR (pg/10mg Protein)
1000000
Post ABT
100000
10000
1000
13
10
6
14
30
11
20
23
9
28 29 2
Patient Number
16
21
7
12
5
3
4
25
Post ABT
Relative PAR Levels in Tumor (Phase I P7977)
100
Relative PAR (%)
75
50
25
0
13
10
6
14
30
11
20
23
9 28 29 2
Patient Number
NCTVL
16
21
7
12
5
3
4
25
Sources of Phase I Drugs
Pharmaceutical industries / biotechnology companies (big
and small)
◦ Big pharmas: often select “preferred sites” for pipeline
development, often intense “test burden”, secure and well
funded
◦ Small pharmas: 1 or 2 drugs as their “life-line”, more
amenable to data sharing, less secure
Academic agencies (NCI US, EORTC, etc)
In-house development (e.g. Apogee phase I study @ HCC)
Challenges:
◦ Getting support for investigator-initiated trial ideas
◦ Getting different agents from different companies for a single trial
The Successful Phase I Team
Scientists
Fellows
Trial nurses
Data coordinators
Investigators
PATIENT
Pharmacists
Biostatisticians
Lab personnel:
reference, PK, PD
Radiologists
Recent Changes