Transcript David Lewinsohn, MD, PhD

The Royal Society
SATELLITE MEETING ON
Human evolution – plagues,
pathogens and selection
Recognition of the Mtb-infected Cell:
What IGRAS Have Allowed Us To Understand About TB
David Lewinsohn, MD, PhD
Pulmonary & Critical Care Medicine
Portland VA Medical Center
Oregon Health & Sciences University
TB Diagnostics: Opportunities for Impact
• TB Disease
– Case Finding
• Improved detection of bacterial products
– LED / NAAT / Urine Dipstick / Novel Plasma or Urine Markers
• Rapid identification of drug resistance
– Line Probe / Cepheid
• Improved understanding of immune markers of antigenic load
• Identification of markers of active infection
– Treatment
• Tools to identify those at risk for treatment failure
• “Latent” TB Infection
– Identify those at risk for progression
• Improved understanding of bacterial persistence
• Improved understanding of immune markers of antigenic load
Incidence
Mortality
Abu-Raddad L J et al. PNAS 2009;106:13980-13985
©2009 by National Academy of Sciences
The Testing Strategy Should be Matched to Benefit
TST ≥ 10 mm
TST ≥ 15 mm
TST ≥
5 mm
The TST Immune Phenotype
Rieder, Epidemiologic Basis of TB Control, IUTLD (1999)
Acute TB Infection
Persistent
TB Infection
 Risk highest
shortly after
infection
 Late disease might
reflect re-exposures
 TST poor at
identifying those at
risk
Outcome of contact evaluations
ARPES,
2001; S+
Reichler,
1996; all
Marks, 1998;
S+ close
Contacts
Evaluated
55-81% of contacts complete TST
LTBI dx’d
22-36% found newly infected
Start TLTBI
63-74% begin treatment
Complete Rx
51-62% finish treatment
Outcome
~17-37% of target population actually
complete TLTBI
Courtesy Andy Vernon, CDC
Lesson 1: David vs Goliath
TST
–
–
–
Associated with
Exposure to Mtb
Conversion
associated with risk
of disease
Treatment of
benefit
IGRAs
Peter
Andersen, SSI
Winner 2011
Novo Nordisk
Prize
Steve Reed, IDRI
The Immune Phenotype: Beyond TST
Lesson 1: We Can do At Least as Well as the TST
IFN-TNF
CD4
Lymphocyte
CD4
TCR
“Professional”
Antigen Presenting
Cell
Protein
Class II
MHC
Peptide
12-18 AA
Quantiferon
T-SPOT
IFN-γ ELISA
CD4
IFN-γ ELISPOT
CD4 / CD8
ESAT-6
CFP-10
IGRAs and The Diagnosis of LTBI
• Test Accuracy
– What is the Sensitivity of the test with regard to TB?
– What is the Specificity in those unlikely to have been exposed?
– What is the Likelihood of infection in those being tested?
• Association with Conditions Linked to Treatment Success
– Is the test associated with exposure to Mtb?
– Is the test associated with progression to TB?
• Are there other risks / benefits that might affect the decision as to which test
to use?
– Test logistics
– Client and provider perceptions
– Cost
• Is treatment based on the test associated with improved outcomes?
– More accurate identification of those at risk?
– Do more patients accept or complete treatment based on the test results?
Test Accuracy
– Sensitivity Active TB*
• QFT-IT
• TSPOT
• TST
84% (81,86)
93% (90,95)
77% (71,82)
– Association with Exposure
• Both QFT and TSPOT are more closely associated with
exposure than TST.
* Courtesy Madhu Pai, High prevalence countries
Test Accuracy
– Specificity*
• QFT-IT
• TSPOT
• TST
99% (98,100)
85-98%
– BCG
49% (46,73)
– No BCG 97% (95,99)
* Courtesy Madhu Pai, High prevalence countries
Progression to TB
–
Diel R et al. Predictive value of a whole blood IFN-gamma assay for the development of active
tuberculosis disease after recent infection with Mycobacterium tuberculosis. Am J Respir Crit Care Med.
2008.
•
•
•
•
•
–
601 household contacts
40.4% (243/601) TST +
11% (66/601( (11%) QFT+
Six contacts progressed to TB disease within the 2-year follow-up.
All were QFT positive and had declined preventive treatment, equating to a progression rate of 14.6% (6/41) among
those who were QFT positive.
Kik et al. Predictive value for progression to tuberculosis by IGRA and TST in immigrant contacts. Eur
Respir J. 2009 .
•
•
•
•
•
339 TST+ household contacts (5mm); prophylaxis not given
9/288=3.1% (95% CI; 1.3-5.0) for TST ≥10
5/178=2.8% (95% CI; 1.0-4.6) for QFTGIT (54% QFT+; 5/8 cases detected)
6/181=3.3% (95% CI; 1.3-5.3) for T-SPOT.TB (63% TSPOT +; 6/8 cases detected)
Review of TB case reporting did not find cases in those who were TST-
High Risk Populations
•
•
•
•
Heterogeneous Population
Very Limited Data in High Income Settings
TSPOT May be Less Affected by Low CD4 Counts
Immunosupression is Associated with Impaired Test Performance for Cellular Assays
•
Sensitivity Active TB in HIV*
– QFT-IT / TSPOT
– TST
* Courtesy John Metcalfe
•
84% (65,100)
43% (25,85)
Association with Progression to TB
–
–
–
–
–
–
–
Aichelburg et al., Clin Infect Dis. 2009 Apr 1;48(7):954-62.
830 HIV+ subjects
44 QFT+
8 found to have TB at time of testing
7QFT+
1 QFT –
3/37 QFT+ developed TB during follow up
Low Risk Populations
– Not candidates for Rx
– Specificity of IGRAs may be superior to TST,
particularly in those with a history of BCG
vaccination.
Khoury et al., JOEM, 2011
Lesson 2: There is nothing Latent about LTBI
IGRAs Reflect a Dynamic Relationship of the Host,
the Microbe, and the Environment
• The Host
– Vulnerability to infection
– Infection / disease status
• The Microbe
– Strain
– Metabolic activity
• Environment
– Likelihood of exposure / re-exposure
– Other mycobacteria
– Other infections
TST Variability
• Menzies, AJRCCM 1999
– 6mm represents 2SD in repeated TST
measurements
– TST reversions
• Following INH therapy among those recently exposed on a
submarine
– Houk et al., 1968. Arch. Environ. Health 16:46–50
• Estimated 8% in exposed children
• Estimated 8% annually in older adults
Do IGRAs Reflect Bacterial Burden?
• Metcalfe et al., AJRCCM 2009
– 660 TB suspects in San Francisco
– Increased IFN-γ associated with increased likelihood of TB
Kobashi et al., Lung 2010
Hinks et al., Infection and Immunity
2009
Are IGRAs a Reflection of
the Environment and/or Host?
Detjen et al.,
Clinical Vaccine Immunology
2009
Ringshausen et al,
Clinical and Vaccine Immunology
2011
Transient Exposure?
• Friedman et al., Infect Control Hosp Epidemiol
2006
– 45 % HCW previously positive by TST were negative on
repeat testing
– Best correlation with persistent TST positivity
• Foreign birth
• BCG
• TST > 15mm
– Best correlation with TST reversion
• Working in study clinic
Transient Exposure?
Ewer al al., Am J Respir Crit Care Med Vol 174.
pp 831–839, 2006
Transient Exposure
• Natural exposure of guinea pigs to MDR/XDR
– 75% of the 362 exposed guinea pigs had positive skin
test reactions [6 mm].
– 12% had histopathologic evidence of active disease.
– Reversions (6mm change) in skin test reactivity were
seen in 22% of animals, exclusively among those with
reactions of 6-13 mm.
– 2of 86 guinea pigs with reversion had histological
evidence of disease vs 47% (31/66) of guinea pigs with
large, non-reverting reactions.
– Immunosuppression had no effect
Can IGRAs Discern Recent from Remote Exposures?
• Millington et al., JID 2010
– 27 subjects with selfhealed, untreated TB
– >50 years
– 17 + Ex-vivo ELISPOT (16
IFN-γ )
– 6/10 + long term assay
• Recent vs Remote
– Hinks et al., I&I 2009
– Kik et al., IJTLD 2009
Cellular Immunology: CD4 Cells
IFN-TNF
CD4
Lymphocyte
CD4
TCR
“Professional”
Antigen Presenting
Cell
Protein
Class II
MHC
Peptide
12-18 AA
Cytokines
Cytotoxicity
How Do We Explain the Variability Seen in
Longitudinal Testing?
• Assumptions
– ELISA/ELISPOT have inherent variability (10-15%)
– There is a learning curve when IGRAs are introduced
van Zyl-Smit et al., AJRCCM, 2009
Ringshausen et al., BMC Infect Dis
2010
Lesson 3: We Need Better Surrogates of
Bacterial Burden
Novel T Cell Phenotypes
Adapted from Seder et al., Nat Rev Immunol, 2008
The Host: Sampling the Intracellular Environment
IFN-TNF
All Cells
CD8
Lymphocyte
CD8
TCR
2m
Class I
MHC
Cytoplasmic
Protein
Peptide
9-11AA
Cytokines
Cytotoxicity
ESAT-6 / CFP-10 Responses During TB Treatment
CD4 Response
CD8 Response
BMI Has a Profound Effect on the CD4 Response
CD4 Response
CD8 Response
* p = 0.02
BMI ≤ 17
BMI > 17
BMI ≤ 17
BMI > 17
Conclusions
• IGRAs perform at least as well as TST in those
who might benefit from therapy.
• Quantitatively, IGRAs likely a reflection of
persistent and variable Mtb-antigen.
• Improved surrogates of bacterial / disease
burden are likely needed.
Acknowledgements
•
•
•
Portland VA Medical Center
Melanie Harriff
Lynne Swarbrick
Marielle Gold
OHSU
Deborah Lewinsohn
Megan Null
Amanda Duncan
Veena Rajaraman
Tomi Mori
Todd Vogt
Ervina Winata
Sue Smyk-Pearson
Joel Weekley
Liz Canfield
Melissa Nyendak
Katelynne Garner-Toren
Meghan Cansler
Shannon McWeeney
Byung Park
Guanming Wu
Special thanks to:
Aeras
Collaborators
Henry Boom
Denise Johnson
Keith Chervenak
Dave Sherman
Karen Dobos
John Belisle
Lisa Wolf
Henry Boom
Roy Mugerwa
Harriett Myanja
Sarah Kiguli
Dennis Dobbs
Sarah Zalwango
Mary Nsereko
Payam Nahid
Alesandro Sette
John Altman
CWRU
NIH
CWRU
CWRU
University of Washington
Colorado State University
Colorado State University
Colorado State University
OHSU
Case Western Reserve University
Makerere University
Makerere University
Makerere University
TBRU
Makerere University
TBRU
UCSF
The La Jolla Institute for Allergy and Immunology
Emory University
Sean Bennett
Bruce McClain
Jerry Sadoff
Donata Sizemore
John Fulkerson
David Hokey
Tom Evans
Christine Sizemore
Alison Deckhut
Timothy Gondre-Lewis
Lynne Swarbrick
Amanda Duncan
Melissa Nyendak
Melissa Kumagi
Phyllis Carello
Lynne Swarbrick
Ervina Winata