Introduction to Human Challenge Models (HCMs) for ... · immunological status can be tightly...
Transcript of Introduction to Human Challenge Models (HCMs) for ... · immunological status can be tightly...
1
Introduction to Human Challenge Models (HCMs) for Respiratory Viruses and the Application of Quantitative Pharmacology to Enhance Drug Development
Craig Rayner PharmD MBA
22nd August 2016
Disclosures and Acknowledgements
The information is derived from publically available sources and key
material for further reading is provided. Any representation/interpretation
does not necessarily reflect perspectives of others.
Data presented represents the efforts of many colleagues, over many
years across many organizations
- Roche / Genentech
- d3 Medicine LLC
- ICPD, University at Buffalo, Monash University
- Alios Biopharma, hVIVO, University of Tennessee
- 360 Biolabs
Primary investigators and clinical trials staff
Clinical trials participants
2
Related Work
• Publications
1. Wollacott AM et al. Safety and Upper Respiratory Pharmacokinetics of the Hemagglutinin Stalk-
Binding Antibody VIS410 Support Treatment and Prophylaxis Based on Population Modeling of
Seasonal Influenza A Outbreaks. EBioMedicine. 2016 Feb 26;5:147-55
2. Kamal M, Rayner C et al. A Drug-Disease Model Describing the Effect of Oseltamivir Neuraminidase
Inhibition on Influenza Virus Progression. Antimicrob Agents Chemother (2016).
3. Rayner C et al. Pharmacokinetic-pharmacodynamic determinants of oseltamivir efficacy using data
from phase 2 inoculation studies. Antimicrob Agents Chemother (2013).
4. Dobrovolny H, Rayner C et al. Assessing mathematical models of influenza infections using features
of the immune response. PLoS One (2013).
5. Reddy MB, · Yang KH, · Rao G, · Rayner CR, · Nie J, · Pamulapati C, · Marathe BM, · Forrest A, ·
Govorkova EA. Oseltamivir Population Pharmacokinetics in the Ferret: Model Application for
Pharmacokinetic/Pharmacodynamic Study Design. PLoS ONE 10/2015; 10(10)
3
Related Work• Abstracts
1. Lutz J, Patel K et al. A mechanistic model describing
the effect of respiratory syncytial virus (RSV) kinetics
on clinical symptom score by presatovir (GS-5806).
ACOP7 (2016 accepted).
2. Fidler M, Patel K et al. A Symptom Driven Multiscale
Model of Influenza. ISIRV (2016 accepted).
3. Patel K et al. Population PK/PD Modeling of Human
Respiratory Syncytial Virus Infection and the Antiviral
Effect of AL-8176. IDWeek (2015).
4. Patel K et al. Population
Pharmacokinetic/Pharmacodynamic (PK/PD) Modeling
of MHAA4549A, an Anti-influenza Monoclonal
Antibody, In Subjects Challenged With Influenza A
Virus. ICAAC (2015).
5. Patel K et al. Population modelling of influenza viral
kinetics, immune response, symptom dynamics and
the effect of oseltamivir. PAGANZ (2015).
6. Rayner C, Kirkpatrick et al. A Novel Interdisciplinary
Pharmacometric Approach: A Systems Approach to
Support Pharmacology to the Payer. ACOP (2015).
7. Patel K at al. Modelling the kinetics of human
respiratory syncytial virus (RSV) and clinical disease
symptoms. ACOP (2014).
4
8. Wu D, Chaiyakunapruk N, Pratoomsoot C, Lee K, Chong HY,
Nelson NE, Smith PF, Kirkpatrick CM, Kamal M, Nieforth K,
Dall G, · Toovey S, ·Kong DC, · Kamauu A, Rayner CR. Cost-
Utility Analysis of Optimal Dosing of Oseltamivir Under Pandemic
Influenza Using a Novel Approach: Linking Health Economics
and Transmission Dynamic Models Value in Health 11/2014;
17(7):A807.
9. Wu D, Chaiyakunapruk N, Pratoomsoot C, Lee K, Chong HY,
Nelson NE, Smith PF, Kirkpatrick CM, Kamal M, Nieforth K,
Dall G, · Toovey S, ·Kong DC, · Kamauu A, Rayner CR.
Oseltamivir use in an Influenza Outbreak: Linking Pharmacology
to Pharmacoeconomics. IDWeek 2014 Meeting of the Infectious
Diseases Society of America; 10/2014
10. Smith PF, Kirkpatrick CM, Rayner CR Wu D, Chaiyakunapruk N,
Pratoomsoot C, Lee K, Chong HY, Nelson NE, Nieforth K, Dall
G, · Toovey S, ·Kong DC, Kamauu A, Kamal M. Estimating
Health Outcomes of Antiviral Use in Influenza Outbreaks by
Linking PK/PD and Epidemiology via a Transmission Dynamic
Model: A Novel Approach. IDWeek 2014 Meeting of the
Infectious Diseases Society of America; 10/2014
11. Rayner CR, Bulik CC, Kamal MA, Reynolds DK, Toovey S,
Hammel JP, Smith PF, Bhavnani SM, Ambrose PG, Forrest A.
Pharmacokinetic-Pharmacodynamic (PK-PD) Determinants of
Oseltamivir Efficacy Using Data from Two Phase 2 Inoculation
Studies. XIV International Symposium on Respiratory Viral
Infections (ISRVI), Istanbul, Turkey, 23–26 March 2012
12. Yang KH, Reddy M, Pamulapati C, Rayner CR,
Forrest A. Development of a Pharmacokinetic
Model for Oseltamivir in Ferrets Using Iterative
Two-Stage Analysis. American College of Clinical
Pharmacy Virtual Poster Symposium, May 22–
24, 2012
13. E. Hershberger et al. Pharmacokinetics of the
Hemagglutinin (HA) Stalk-Binding Antibody,
VIS410, in a Human Challenge Model of Infection
with a p2009 H1N1 Virus. ISIRV 2016
14. GG Rao et al. Population Modelling of Influenza A
Kinetics, Innate Immune Response & Symptom
Dynamics. 54th Interscience Conference on
Antimicrobial Agents and Chemotherapy (ICAAC).
Washington, DC Sept 2014.
15. M Fidler et al. Modeling the Spread of Pandemic
Influenza in the United States: Impact of Antiviral
Interventions, Pharmacology, and Resistance.
American Conference on Pharmacometrics;
October 2014, Las Vegas, NV.
16. E Lakota et al. Development and optimization of
an adaptive study design for respiratory virus
human challenge models. ACCP, 2014.
Background
• Human challenge models (HCM) are employed as part of many
development programs for ARI such as influenza, rhinovirus and respiratory
syncytial virus (RSV)
• Infections are deliberately induced under carefully controlled and monitored
conditions, involving virus inocula of known virulence and provenance
• Key variables, such as baseline infection load, timing of Rx, and
immunological status can be tightly controlled to isolate drug effect
• PK, PD (Virologic, clinical, biomarker) and safety can be diligently evaluated
• Quantitative pharmacology can optimize HCM study design and analysis to
enhance drug development for respiratory viruses.
5
HCMs as Applied to Respiratory Viruses
• Highly controlled nature of experimental
pharmacology HCMs enables “rich and clean”
interrogation of drug effect
• Intensive safety surveillance
• Specialist units creates opportunity for novel
designs (eg. adaptive design to capture ER
surface, expansive dose ranges, frequent
invasive sampling) to mitigate expense and
accelerate
• Not dependent on seasonal disease, means
study durations can be short
• Rich data enables quantitative pharmacological
approaches (MBM) to support development
6
Pro’s Con’s
• Regulatory challenges in obtaining an
acceptable inocula
• Ethical considerations (infecting volunteers,
transmission risk)
• Transferability to the field is debated as
important variability ignored including virus
(strains, virulence, replication, IC50, tropism,
baseline VL), time to Rx, host immune status
etc), abrogated/different clinical course of
disease
• Artefact of study including inoculation
techniques, impost of frequent invasive
sampling (dilution, loss of virus)
• Specialist clinical trial units, strain availability
and need to pre-screen to obtain sero-negative
subjects can create availability bottlenecks and
high costs
Considerations for Quantitative Pharmacology Application to HCM of Respiratory Viruses
- Case 1: HCM for Oseltamivir for influenza
- Case 2: HCM for ALS-008176 for RSV
7
CASE 1 - HCM FOR OSELTAMIVIR FOR INFLUENZA
9
Methods – Pooled Available Ph2 Studies
• Healthy volunteers in Studies PV15616 and NP15717 were experimentally infected with
influenza (TCID50 of virus 106)and treatment was initiated with oseltamivir or placebo after 24h
• Symptoms (feeling feverish, headache, muscle ache, sore throat, cough, overall discomfort,
nasal symptoms) individually ranked b.i.d. for symptom severity (0, 1, 2, or 3) for 9 days
• Nasal lavage b.i.d. for viral culture Days 1–3 and o.d. on Days 4–8
Study
Number of
subjects
(infected) Virus
IC50
(nM) Dosing regimens PK Sampling
PV15616 80 (69)Influenza
A/Texas0.18
20, 100 or 200mg b.i.d.
or
200mg q.d.
or placebo for 5 days
Sparse PK samples
pre-dose and on Days
3, 4 and 7
NP15717 60 (46)Influenza
B/Yamagata16.76
75 or 150mg b.i.d.
or placebo b.i.d. for
5 days
Intensive PK on Days
1 and 5
10
Methods - PK
• A robust population PK model based on nine clinical studies was developed to provide individual
OC exposures
• Individual predicted steady-state exposures for OC were determined: AUC0–24h, Cmin, Cmax
CLoc = clearance of OC; CLop = clearance of OP; F = fractional bioavailability
of OP; Fm = fractional bioavailability of oseltamivir that is metabolised;
Ka = the first order rate of absorption ; Q = inter-compartmental clearance
term; Voc = volume of OC; Vop = central volume of OP; Vp = peripheral volume
of OP
(Fm)
Oral
Depot
Ka
(F)
Central (OP)
Vop
Q/Vop
Q/Vp
Peripheral
(OP) Vp
CLop/Vop
Central (OC)
Voc
CLoc/Voc
11
Independent Variables
Exposure, PK-PD
or potency measureDemographic / other
AUC0–24h Age
AUC0–24h:IC50 ratio Race
Cmax Sex
Cmax:IC50 ratio Height
Cmin Weight
Cmin:IC50 ratio Body mass index
Total daily dose Creatinine clearance
Treatment regimen Antipyretic anilides
IC50
Note: co-linearity reduced focus to AUC related metrics
consistent with PKPD index from preclinical models
Methods:
12JAMA. 1999;282(13):1240-1246. doi:10.1001/jama.282.13.1240
The viral titer area under the curve value was lower in the combined oseltamivir group (n=56) compared with placebo (n=13); P=.02.
“Typical” time-course of viral load (PV15616)
Time-course is a blend of “fact”
and experimental “fiction”
- Inoculation
- Serial nasal washings
- Measurement
Detection of PK/PD effect influenced by VT time-course and form of dependent variable (eg. Peak VT<rate of decline< AUC<< TSVR =Tshed )
13JAMA. 1999;282(13):1240-1246. doi:10.1001/jama.282.13.1240
“Typical” symptom and cytokine time-course (PV15616)
The total symptom score area under the curve value was lower in the combined oseltamivir groups (n=56) compared with placebo
(n=13); P=.05; Cytokine levels for days 1, 3, 5, and 9 were determined using commercially available enzyme-linked immunosorbent
assay kits. Asterisk indicates P≤.01; dagger, P≤.001; and double dagger, P<.05.
Patel et al. ICAAC 2015
MHAA4549A
Oseltamivir
HCM viruses can provide biophase PK, VK, cytokines and symptoms to inform MBMs
Understand the inputs!
14
Sampling procedures
- Procedure (disruptive, dilutive)
- Replicate (same/different site)
- Pooled/Combined(wash + swab)
- Standardisation (within study,
across studies)
VT / biomarker measurement
- Sampling error
- Method (culture/ PCR)
- Performance (BLQ)
It is essential to have a detailed understanding of the sampling and measurement methodologies
CROs and PROs
- Validated instruments
- Relevance to HCM
- Fever
- Composite vs
individual symptom
scores
- Frequency / learning
15
Methods:
Time-to-event Continuous
• Time to alleviation of composite symptom
score
• Time at which any of the seven individual
symptoms scores were >1 to the time at
which all applicable individual symptom
scores were ≤1
• Time to cessation of viral shedding
• Time from the first positive viral culture to the
time of the first negative viral culture
• Composite symptom score AUC
– Symptom scores added together and AUC of
composite score calculated from the time at
which any of the seven individual symptoms
scores were >1 to the time at which all
applicable individual symptom scores were ≤1
• Viral titre AUC
– The AUC of the viral titre values spanning the
time from the first positive viral culture to the
time of the first negative viral culture
• Peak viral titre– The maximum viral titre value achieved for
each subject during study Days 1–9
A diverse set of univariate and multivariable PK-PD analyses were performed
against the following dependent variables:
Results
• Higher OC Exposures are
Associated with a Faster
Cessation of Viral Shedding - Peak VT and AUCVT ND
• Higher OC Exposures are
Associated with a Lower
Severity of Illness
• A Faster Time to Alleviation of
Composite Symptoms is Seen
with Higher OC Exposures
16
Findings consistent with observations of exposure-response nested within Phase III trials1,2
1Evaluation of Pharmacokinetic-Pharmacodynamic (PK-PD) Relationships for Influenza
Symptom and Quality of Life (QOL) Endpoints Among Oseltamivir- Treated Patients
(ICAAC 2014); 2Pharmacokinetic-Pharmacodynamic (PK-PD) Evaluation of the Impact
of Oseltamivir on Influenza Viral Endpoints (ICAAC 2014)
CASE 2 - ALS008176 INOCULATION STUDY FOR RSV
18
- PCR guided Rx start- Adaptive design (safety, PK, PD) to establish ER/DR
-008176 is an oral nucleoside prodrug converted intracellularly to its active
triphosphate ALS-008112.
Methods
19
Results ALS-8176 Dose-Response in a Human RSV Challenge Model
500 mg LD/500 mg q12h
500 mg LD/ 150 mg q12h
375 mg q12hLonger VT time-course than influenza HCM
ALS-8176 had significant improvements on VT AUC, Time to undetectable PCR, Peak VT and AUC symptom score
N Engl J Med 373;21 (2015)
20
K Patel, et al. ID Week 2015
Even Abbreviated Designs Enable MBM Approaches
Rich PK/PD data provided in RSV HCM enables quantitative pharmacology approaches to address complex nucleoside pharmacology
Considerations for Quantitative Pharmacology for Respiratory Virus HCMs
• Critical to understand sampling and assay methods as they can impact
integrity of observations and assumptions
• Consider the time-course of PD markers (eg. VK) to guide the most
appropriate form of a PD variable
• Don’t forget about other endpoints, as HCM virus can cause symptoms
• Consider potential for MBM to connect PK, VK, immune-system
components, and symptoms to enable better extrapolation to clinic
(especially for complex examples such as nucleoside analogues)
• Numerous opportunities to optimise HCMs through quantitative
pharmacology methods (adaptive design, optimal design based on ER)
21