Clinical Laboratory Testing for Detection of Influenza. - PowerPoint Presentation

Slide1

Clinical Laboratory Testing for Detection of Influenza.

Rangaraj Selvarangan. BVSc, PhD, D(ABMM).Associate Professor, UMKC-SOMDirector, Microbiology LaboratoryDirector, Laboratory Medicine Research AffairsChildren's Mercy Hospitals and ClinicsKansas City, MO 64108

International Conference on Flu, June 8-10, 2015Chicago, USASlide2

ObjectivesDescribe laboratory tests available for detection of influenza infection

Identify strengths and weaknesses of traditional and molecular laboratory tests for InfluenzaDiscuss FDA ruling on Rapid antigen tests Describe the potential clinical impact of rapid molecular testing for influenzaSlide3

IntroductionRespiratory viral illness is one of the most common infection in children and adults

Influenza and RSV cause seasonal outbreaks; early diagnosis improves care.Non-influenza respiratory illness are also common; estimated at 500 million/year, economic impact of $40 billion/year1Several studies show that a substantial percentage of both out patient (>70%) 2 and in-patient antibiotic use (up to 50%) is either unnecessary or inappropriate.In 2008, there were 142,000 visits to emergency departments for adverse events attributed to antibiotics.31Fendrick et al Arch Intern Med 2003,

2 Hersch et al Ped

2011,

3

Shehab et al

Clin

Inf

Dis

2008

Slide4

Influenza Epidemics

Member of the Orthomyxovirus, Influenza type A and B cause seasonal epidemics in humans.Transmission via large-particle droplets.Incubation 1-4 days. Every year in the United States, on average:5% to 20% of the population gets the flumore than 200,000 people are hospitalized from flu-related complications about 36,000 people die from flu-related causes Shedding: Adults 5-10 days, Children >10 days,

immunocompromised weeks or months.Slide5

Influenza in Children

Children aged 0–4 years 100/100,000 children for those without high-risk medical conditions. 500/100,000 children for those with high-risk medical conditions Highest among children aged 0–1 years and are comparable to rates reported among persons aged > 65 years.Signs: Abrupt onset of constitutional and respiratory signs (fever, myalgia, headache, sore throat, cough). Children- otitis media, nausea and vomiting.Hospitalization: 4%-11% ICU, 3% mechanical ventilation.

Complications: Otitis media, Bacterial Pneumonia, Encephalitis, encephalopathy,

Myocarditis

,

Myositis

PPV for clinical definition in children 79% to 88%.Slide6

Influenza Epidemics

ProphylaxisDiagnosisTreatmentSlide7

Influenza Culture

Tube culture- PMK, MDCK (trypsin in medium), MLCo-cultured cells: R-Mix (DHI)- A549+ML, R-Mix Too- A549+MDCK. Rapid results >80% of flu specimens are positive on Day 1.R-Mix Too- may be better in FluB recovery and it does not support SARS-CoV.Dunn et al JCM 2004. Total 3803 respiratory specimens; compared tube culture to R-Mix SV. Flu A 238/241 (99%), Flu B 36/38 (95%) and RSV 52/60 (87%).

Cryopreserved R-Mix Ready cells: Kim et al JCV 2008. Higher and early recovery of viruses in R- Mix ready cells Vs tube culture.

Option for small volume laboratories and surplus inventory during peak winter season.Slide8

Influenza DFA

DFA for influenza is a highly sensitive and specific test.Technical expertise needed, TAT few hoursSensitivity is affected by sample quality- flocked swabs provide good cell recovery for DFA. Screening antibodies for 7 viruses are available from several vendors. Identification is by mAb.Cytospin improves DFA performance on slide preparations.Combination antibodies FluA/respiratory virus help for quick screening during peak influenza season. D3 Duet (DHI)

SimulFluor (Millipore).Slide9

Influenza Rapid Antigen Test

Rapid antigen test for influenza have moderate sensitivity for seasonal Flu A/B (50-80%) and poor sensitivity (10% to 40%) for 2009 H1N1 virus.Specificity usually >90%.Influenza antigen negative results need confirmatory test performed; Culture, DFA, RT-PCR.Benefit of Influenza rapid antigen test in the OP setting patient triage, infection control, reduced antibiotic usage, timely antiviral treatment, reduce unnecessary diagnostic workups, overall reduction in medical care cost.Slide10
Slide11

Rapid Antigen Test-Flu A/BHost Factors:

AgeDuration of illness Specimen typeSpecimen transportSlide12

Host factorsAge: Children excrete high viral loads during influenza infections

Duration of illness: Viral load in respiratory secretions decrease with duration of illness. Testing within 3 days of illness onset improves detection Specimen Type: Nasopharyngeal aspirates, nasopharyngeal washes, nasopharyngeal swabs, mid turbinate swabs are preferred specimens. Nasal and throat swabs results in sub-optimal yieldSpecimen Transport: Collection in viral transport medium and rapid transport on ice improves yield Cheng et al., 2009; Esposito et al., 2011; Loeb et al., 2012; Talbot et al., 2010Slide13

Specimen collection

Nasopharyngeal aspirates and nasal washes: High viral load, diluted, mucous can interfere with EIA/DFA performance. Cell quantity may vary between collection. Suitable for culture, PCR and later flow tests.Recent experience indicates that a combination of NP swab and throat swab in VTM may improve diagnostic yield.Flocked swabs: More cells collected, less mucous, well suited for DFA. Nasopharyngeal and mid-turbinate swabs available.Slide14

Tilt patient head 70 degree angle and against wall

Insert swab straight back (not upward) until resistance is met

Rotate the swab 5-10 times to loosen cells

Remove swab and inoculate VTM Slide15

Swab studies

Abu-Diab et al JCM 2007; NPA Vs NP swab. 455 children. Sensitivity of NP swab 98.5%.Allen et al PASCV 2008 poster # M28; MTS Vs NP swab. 203 children, PCR gold standard. NPS =87.5%, MTS =79%. 8% of both NPS and MTS insufficient for DFA. Nurse prefer MTS.Selvarangan et al PASCV 2009 poster # M51. 200 children. NPA Vs MTS. RSV antigen and SV culture. RSV antigen sensitivity 66% for MTS and 70% for NPA. Nurses prefer MTS over NPA.Slide16

Rapid Antigen Test-Flu A/BViral Factors:

Influenza activityViral subtype Slide17

Test Performance

Sensitivity : proportion of actual positives which are correctly identified as such (i.e. the percentage of sick people who are identified as having the condition) Specificity : proportion of negatives which are correctly identified (i.e. the percentage of well people who are identified as not having the condition). The probability of the presence or absence of disease given the results of a test.Positive predictive value (PPV) : proportion of patients with positive test results who are correctly diagnosed.

Negative predictive value (NPV): proportion of patients with negative test results who are correctly diagnosed.

Wikipedia.orgSlide18

1% Flu

Infected

Not-infected

Positive

TP = 19

FP = 80

PPV

TP / (TP+FP)

19 / 99 = 19%

Negative

FN = 1

TN = 1900

NPV

TN / (TN + FN)

1900 / 1901= 100%

Sensitivity TP / (TP + FN)

19 / 20 = 95%

Specificity TN / (FP +TN)

1900 / 1980 = 96%

20% Flu

Infected

Not-infected

Positive

TP = 380

FP = 64

PPV

TP / (TP + FP)

380 / 444 = 86%

Negative

FN = 20

TN = 1536

NPV

TN / (TN + FN)

1536 / 1556 = 99%

Sensitivity TP / (TP + FN)

380 / 400 = 95%

Specificity TN / (FP + TN)

1536 / 1600 =96%

Beginning and end of the season-

It is advisable to confirm all positives by culture due to low PPV

Test Result

Test ResultSlide19

RADT

RVPRADT,RVP

ProFlu

PCR

LDT-PCR

FluA

/B

Cx

Influneza

pandemic 2009Slide20

Rapid Antigen

PCR

Pre pH1N1

Post pH1N1

Pre pH1N1

Post pH1N1

NPO (n=22)

19 (86.4%)

14 (63.6%)

0

1 (4.5%)

Hosp

Ntwrk

(n=31)

24 (77.4%)

19 (61.3%)

4 (12.9%)

8 (25.8%)

Comm

Hosp (n=43)

40 (93%)

36 (83.7%)

0

0

Academic Inst (n=33)

18 (55%)

8 (24%)*

5 (15%)

14 (42%)*

Survey of Lab Methods

Selvarangan

et al

ASM conference 2010

Use of molecular methods for detection of respiratory viruses

2009 = 18% (Selvarangan,

ClinMicronet

Survey)

2011= 36% (Miller,

ClinMicronet

Survey)

2013 = 63% (Miller,

ClinMicronet

Survey)Slide21

The updated VE estimate against influenza A H3N2 viruses was 18% (95% confidence interval (CI): 6%-29%).

The VE estimate against influenza B viruses this season was 45% (95% CI: 14% – 65%).Slide22

Rapid Antigen Test-Flu A/BAnalytical Factors

Test characteristicsResult interpretationPost Analytical FactorsReporting TimeClinical interpretationSlide23

Next –Generation Influenza RADT

3M Influenza

BD Veritor

Quidel SofiaSlide24

Rapid Antigen Tests- Instrument Reader

Hassan et al JCM 2014Slide25

Rapid Antigen Tests- Instrument Reader

Dunn et al DMID 2014Slide26
Slide27

Manufacturer 1: ref method Culture; 79% sensitivity (75-83%)Slide28

FDA Ruling May 2014-RIDT

Reference Method: Viral culture SensitivityFlu A Point estimate of 90% (95% C.I . >80%)Flu B Point estimate of 80% (95% C.I. >70%)Specificitylower bound of the 95

% CI exceeding 90% for both, Flu A and Flu B.Reference Method:

Molecular

method

Sensitivity

Flu A

Point estimate of 80%

(95% C.I. >70%)

Flu B

Point estimate of 80%

(95% C.I. >70%)

Specificity

lower bound

of the 95%

CI exceeding

90% for both, influenza A

and influenza

B.Slide29

FDA Ruling May 2014-RIDTLaboratory Tests:

Influenza viral antigens or influenza viral gene segments (protein or nucleic acid), either in single unit test formats or multi-test formatsMonitoring Performance:Conduct annual analytical testing of their device with contemporary strainsStandardized panel of viruses selected in coordination with FDA. Dilutions at 10e2 and 10e5 TCID50/mL in triplicate

.Detection of all replicates at least at one dilution. Standardized panels of well characterized viral stocks could possibly be available from CDC or commercial

vendors.

The

testing could

be conducted

in-house or at a contract laboratory.

Absence

of analytical reactivity would

be reflected

in labeling as a limitation

.

Emerging Influenza strains:

Provide analytical reactivity report to FDA within 60 days of emerged virus strain availability.Slide30

Diagnostic Challenges Problem1: Lack of a rapid and highly sensitive method for detection of Influenza at POC setting.

Problem 2: Lack of sensitive method for extensive detection of respiratory viruses in hospitalized patientsImpact: Improper use of antivirals, overuse of antibiotics, additional diagnostic workup, improper infection control measures, over all increase in health care cost. Slide31

Diagnostic Challenge-Solution Problem 1: Lack of a rapid and highly sensitive method for detection of Influenza at POC setting.

Solution- RADT-Instrument read, Rapid multiplex RP-NAAT- Alere™ i Influenza A & B, iQuum Liat Influenza A/B, Focus Simplexa™ FluA/B & RSV Direct , Cepheid Xpert® Flu, and GenXpert and BioFire Filmarray™ RP Considerations: Clinical performance data limited, Cost, clinical experience limited, Lack of complete understanding of the hierarchy among these tests Slide32

Molecular Nucleic Acid Amplification Methods for Respiratory Virus DetectionSlide33

Multiplex RP-NAAT: Rapid

BioFire Filmarray™ RP (~60 min)Cepheid Xpert

® Flu(75 min)Focus Simplexa™ FluA

/B &

RSV Direct (~60 min)

iQuum

Liat

Influenza A/B (20 min )

Alere™ i Influenza A & B

(15 min )Slide34

Alere i Influenza Vs Culture

Specimen DetectionTPFPTN

FN

Total

% Sensitivity (95% CI)

% Specificity (95% CI)

PPV (95% CI)

NPV (95% CI)

Flu A

Children

128

5

320

1

454

99.2

98.4

96.2

99.6

(95.1-99.9)

(96.2-99.4)

(91.0-98.6)

(98.0-99.9)

Adults

17

2

56

0

75

100

96.5

89.4

100

(77.0-100)

(87.0-99.4)

(65.4-98.1)

(92.0-100)

Flu B

Children

70

0

379

2

451

97.2

100

100

99.5

(89.4-99.5)

(98.7-100)

(93.5-100)

(97.9-99.9)

Adults13062075100100100100(71.6-100)(92.7-100)(71.6-100)(92.7-100)Bell et al; Journal of Clinical Virology 2014 61, 81-86Slide35

Alere i Influneza Vs

ProFlu PCR Alere™ i Influenza A&B vs. real-time RT-PCRTPFP

TN

FN

Total

% Sensitivity (95% CI)

% Specificity (95% CI)

PPV (95% CI)

NPV (95% CI)

Influenza A

MTS

76

1

70

9

156

89.4

98.6

98.7

88.6

(80.4-94.7)

(91.3-99.9)

(92.0-99.9)

(79.0-94.3)

NPW/A

27

1

45

4

77

87.1

97.8

96.4

91.8

(69.2-95.8)

(87.0-99.9)

(79.8-99.8)

(79.5-97.4)

Total

103

2

115

13

233

88.8

98.3

98.1

89.8

(81.3-93.7)

(93.3-99.7)

(92.6-99.7)

(82.9-94.3)

Influenza BMTS31−124−155100100100100(86.3-100)(96.3-100)(86.3-100)(96.3-100)NPW/A27−50−

77100100100100(84.5-100)(91.1-100)(84.5-100)(91.1-100)Total58−174−232100100100100(92.3-100)(97.3-100)(92.3-100)(97.3-100)Bell et al; J Clin Microbiol. 2014 Nov;52(11):3992-5. Slide36

Alere i Influenza Vs ProFlu PCRSlide37

Cobas Liat Influenza assay

FDA-cleared, rapid (<20 min) PCR assay (Roche cobas® Liat) to Focus Simplexa™ Flu A/B & RSV Direct using respiratory swabs (n=197). The cobas Liat influenza A and B assays demonstrated sensitivities of 99.2% (123/124) and 100% (23/23), respectively, while showing a specificity of 100% for both targets.Binnicker et al: J Clin Microbiol. 2015 Apr 29. pii: JCM.00791-15.Slide38

Multiplex RP-NAAT: Extended Panel

Luminex xTAG RVP (n=12) and RVP FAST (n=8)BioFire Filmarray RP (n= 20 )

Nanosphere RV Plus test (n=7)GenMark

eSensor

RVP (n=14)Slide39

CAP-ID Resp Panel-FLu 2014

2014 IDR-A summarySlide40

Flu Antiviral Resistance

Influenza viruses

Antiviral

2009 H1N1

Seasonal

H1N1

Seasonal

H3N2

Flu B

Adamantanes

Resistant

Susceptible

Resistant

N/A

Oseltamivir

Susceptible*

Resistant

Susceptible

Susceptible

Zanamivir

Susceptible

Susceptible

Susceptible

Susceptible

* Few oseltamivir-resistant 2009 H1N1 strains reported

CDCSlide41

Cost of the testReimbursement issuesLoss of detection due to mutations*Competitive inhibition of

analytes in a multiplex assayLimited clinical experience for certain infections: coronavirus, rhinovirus and co-infections. *Hawkinson et al DMID 2013Multiplex RP NAAT DisadvantagesSlide42

Influenza Diagnostics- Summary

Rapid antigen tests are useful for OP management. Antigenic drift and shift in Influenza strains influences test performance. Specimen collection method and VTM approved for the antigen test need to be followedMonitor QC controls and review data periodicallyReflex antigen negative test result to confirmatory testing

Rapid molecular tests for Influenza will improve diagnostic yield SV

culture is a simple assay with Flu isolation mostly by day

1

Multiplex PCR assays improve Flu detection and other respiratory infections associated with ILI

Epidemiological data or individual Flu subtype testing may be necessary to determine antiviral resistance.