The OraQuick HCV point‑of‑care test for rapid detection of hepatitis C virus antibodies

Sensitivity (%)

Specificity (%)

Oral fluid

90.8–99.2

92.1–100.0

Whole blood

94.4–100.0

98.8–100.0

Fingerstick blood

95.9–100.0

99.9–100.0

Serum

99.9–100.0

99.8–100.0

Plasma

99.4–100.0

99.7–99.9

Study component

Description

Objectives/hypotheses

To assess the diagnostic performance of the OraQuick HCV compared to standard laboratory methods.

Study design

Diagnostic test accuracy.

Intervention

The OraQuick HCV using either serum or oral fluid.

Setting

2 Korean hospitals: Samsung Medical Centre and Seoul National University Bundang Hospital.

Inclusion/exclusion criteria

People who were previously diagnosed with HCV on the basis of clinical and laboratory tests. The publication did not report whether this diagnosis was from antibody testing or RT–PCR.

Primary outcomes

Clinical sensitivity and specificity, analytical sensitivity, interference and cross‑reactivity.

Methods

Comparator laboratory tests: Architect (Abbott) AxSYM (Abbott), E170 (Roche), ADVIA Centaur (Siemens) and Elecsys (Roche).

A total of 3 other rapid tests were also used as comparators: Asan Easy Test HCV (Asan Pharmaceutical), SD BIOLINE HCV (SD), Genedia HCV Rapid LF (Green Cross Medical Science).

For the OraQuick HCV, oral fluid and serum samples were tested.

After oral fluid testing, venous blood samples were drawn. Sera could only be obtained in 114 patients enrolled in the oral fluid test. If oral test result in these HCV‑positive patients was unreactive, the OraQuick HCV with oral fluid was repeated, and the serum was tested using Architect EIA (Abbott). An additional 200 HCV–RNA‑positive serum samples from a blood bank were also tested.

In healthy blood donors, all oral fluids were tested at the site of blood donation. If the oral test was unreactive, it was immediately repeated using the OraQuick HCV. All results were compared to serum Architect EIA (Abbott). If there was a discrepancy in the results, the Architect EIA serum test was repeated and confirmed by Western blot (HCV Blot 3.0, MP Biomedicals) and HCV RT–PCR (COBAS). AmpliPrep/COBAS TaqMan HCV Test 2.0, Roche Diagnostics). An additional 200 HCV‑negative serum samples from a blood bank were also tested.

Participants

There were 137 previously diagnosed HCV‑positive oral fluid samples. There were 114 serum samples from the same patients.

There were 300 healthy‑patient samples for specificity.

Experience of person undertaking test

Not reported.

Results

Oral fluid sensitivity 95% CI, 97.8 (93.2 to 99.4) [134/137].

Oral fluid specificity 95% CI, 100.0 (98.4 to 100.0) [300/300].

Serum sensitivity 95% CI, 100 (97.7 to 100.0) [200/200].

There were 3 false negatives: all patients who previously had HCV but were treated with antivirals. RNA was not detected in 2 of these patients (no active infection).

Dilution tests:

Dilution tests showed that the OraQuick HCV was 'quite comparable' to laboratory assays such as Architect, Centaur, AxSYM.

HCV genotype sensitivity:

HCV genotypes 1a, 2b and 3a tested.

The OraQuick HCV was more sensitive with genotype 3a panel – detected anti‑HCV antibodies 10 days earlier than Architect and E170 did.

Interference and cross‑reactivity:

Neither bilirubin (up to 171 µmol/l), haemoglobin (5 g/l) nor triglycerides (up to 3.39 mmol/l) showed any interference. No cross reactivity with sera positive for rheumatoid factor, multipara, other viral infections such as HIV, hepatitis A or B.

Adverse events

None

Conclusions

Clinical performance of the OraQuick HCV is comparable to laboratory‑based assays with both serum and oral fluid. This supports the supplementary use of rapid HCV testing using oral fluid in various medical and non‑medical settings.

Abbreviations: CI, confidence interval; EIA, enzyme immunoassay; HCV, hepatitis C virus; RNA, ribonucleic acid; RT–PCR, reverse transcriptase polymerase chain reaction.

Study component

Description

Objectives/hypotheses

To evaluate the accuracy of the OraQuick HCV and assess its feasibility for use by community‑based organizations in the USA for high HCV‑risk populations.

Study design

Diagnostic test accuracy.

Intervention

The OraQuick HCV using oral fluid.

Setting

Six community‑based organizations in the USA.

Inclusion/exclusion criteria

People aged 18 years or older who consented in English, Spanish or French, have a self‑reported history or at least 1 of the following risk factors: injecting drug use, incarceration, HIV infection, sexually transmitted infection, liver disease, receipt of a blood transfusion or clotting factor before 1992, organ transplantation, haemodialysis, non‑licensed tattoo or piercing, and sexual partners who were injecting drug users or HCV‑positive.

Primary outcomes

OraQuick/EIA results concordance, clinical staff impressions of the technology.

Methods

Oral fluid samples were taken as per the manufacturer's instruction and processed at the point of care. Blood samples were sent away for reference testing using Abbott HCV EIA 2.0 (Abbott) testing. If the signal‑to‑cut‑off ratio from EIA was between 1 and 3.8 (low signal), the laboratory performed a recombinant immunoblot assay (Chiron RIBA HCV 3.0 Strip Immunoblot Assay, Chiron).

Participants

There were 503 patients from 6 community‑based organisations recruited between April and September 2009.

Experience of person having test

Researcher questionnaire:

Results

Specimens from 486 patients (96.6%) provided valid results.

From the OraQuick HCV and EIA comparator. The OraQuick HCV and EIA yielded the same result in 474 (97.5%) patients.

EAC calculated diagnostic performance (from data in the publication).

Sensitivity, 93.88%

Specificity, 99.48%

Discordant OraQuick/EIA tests resolved using PCR:

Study results were discordant for 12 patients (2.5%). PCR testing could not be performed on 2 of these patients because sample volume was not sufficient. In 10 discordant results, PCR testing confirmed the OraQuick result in 6 cases (indicating active infection); PCR confirmed the EIA result in 1 case (indicating active infection) and the remaining 3 results were either invalid or indeterminate for OraQuick or EIA.

Adverse events

None

Conclusions

OraQuick accuracy is comparable to EIA comparator. The oral swab rapid test could help HCV screening programs reach individuals unaware of their status and expand.

Abbreviations: HCV, hepatitis C virus; EIA, enzyme immunoassay; PCR, polymerase chain reaction.

Study component

Description

Objectives/hypotheses

To assess the use of the OraQuick HCV as a screening tool in the case of HCV outbreak.

Study design

Diagnostic test accuracy (screening).

Intervention

The OraQuick HCV using either serum or whole blood.

Setting

USA. Outbreak of HCV in a cardiac catheterization laboratory, from an infected healthcare worker.

Inclusion/exclusion criteria

Patients treated during the period of the infected healthcare worker's employment were asked to present to 1 of 8 emergency public health clinics for HCV testing.

Primary outcomes

Sensitivity, specificity, and positive and negative predictive value.

Methods

Patients were informed by phone call and letter to present to 1 of 8 public clinics. Two serum samples (for EIA reference and additional tests if needed), and 1 whole blood sample (for the OraQuick HCV) were collected for each patient. OraQuick results were given on‑site; serum samples were sent away for laboratory testing. Any specimen that was positive or invalid was retested on‑site by a blinded second tester, who was unaware that it was a repeat test. Once confirmed, results were relayed to the patient.

Comparator: Ortho HCV v3.0 EIA (Ortho Clinical Diagnostics).

Discordant results were frozen at −70C and sent for chemilluminescent assay testing at the US Centers for Disease Control and Prevention.

Positive results (by the OraQuick HCV or EIA) also tested for HCV RNA by Cobas Amplicor HCV test v2.0 (Roche).

Participants

1174

Results

There were 22 positives (1.9%), 1134 negatives (98.0%) and 1 invalid result (0.1%).

OraQuick sensitivity, 94.4%

Specificity, 99.4%

PPV, 72.7%

NPV, 99.9%

Six OraQuick positives were negative by EIA and CIA, 1 negative OraQuick was positive by EIA and CIA. All were negative by RT–PCR for HCV RNA (indicating no active infection present).

Adverse events

None reported

Conclusions

Demonstrates that the OraQuick HCV can be used in an outbreak setting to allow rapid screening of a large number of patients. This can identify HCV‑infected patients who may be counselled and prevent further spread of the disease. The OraQuick HCV could be integrated into future HCV outbreak testing algorithms.

Abbreviations: CIA, chemilluminescent; EIA, enzyme immunoassay; HCV, hepatitis C; SCR, signal to cut off ratio; RNA, ribonucleic acid; RT–PCR, reverse transcriptase polymerase chain reaction.

Study component

Description

Objectives/ hypotheses

To assess the preference and acceptability of the OraQuick HCV against standard anti‑HCV antibodies testing in injecting drug users, at high risk of HCV infection.

Study design

Questionnaire study.

Intervention

The OraQuick HCV using fingerstick blood.

Setting

San Francisco, USA. Questionnaire following up patients tested with both OraQuick (fingerstick blood only) and phlebotomy‑based testing.

Inclusion/exclusion criteria

People who had injected drugs in the last 30 days, aged over 30 years, who previously completed a baseline visit. Must self‑report negative or unknown HCV RNA status, or prospective patients with unknown HCV status, or anti‑HCV antibodies positive but HCV RNA negative.

Primary outcomes

Testing preference (rapid or standard), participant demographics/characteristics, comparison with phlebotomy‑based testing from patient perspective.

Methods

Eligible patients asked to complete HCV rapid test acceptability survey. Frequencies and measures of central tendency were used to describe sample characteristics and perception of HCV rapid testing (preferences, perceptions and reasons for testing choice).

All patients participating had phlebotomy and OraQuick test (fingerstick blood only), regardless of testing method selection.

Participants

129

Results

98.4% (127/129) completed the survey.

82.9% (n=107) chose the rapid test over standard HCV testing using phlebotomy.

60.2% (50/83*) chose it because of rapid results. 84.4% (81/96) preferred getting their results on the same day, and 97.5% (94/96) understood their results.

60.5% (78/129) felt the rapid test was at least as accurate as standard testing. 53.5% (53/99) thought it was less painful than tests involving venepuncture.

93.9% (92/98) would recommend this test to a friend.

3.1% (3/97) thought it would be better to wait a week to get lab samples back instead.

Of the patients who opted for the standard lab test, 38.1% (8/21) felt that the older test was more established and therefore reliable. 14.3% (3/21) did not want their test results on that day, and felt the standard HCV test was more convenient. A total of 9.5% (2/21) were afraid of having a finger pricked and 4.8% (1/21) felt that the standard test was less painful. Overall, 13.2% (17/129) felt that the OraQuick HCV was to some degree less accurate.

*Note: Reasons for missing responses include interviewer error (n = 13) and a faulty skip pattern (corrected early in data collection) in which participants who preferred standard anti‑HCV testing were not asked a follow‑up question regarding testing preference reason (n = 7).

Adverse events

None

Conclusions

OraQuick test by finger‑prick is more widely accepted by people who inject drugs than phlebotomy‑based testing.

Abbreviations: HCV, hepatitis C virus; RNA, ribonucleic acid.

Study component

Description

Objectives/hypotheses

Evaluation of a commercially available EIA and OraQuick with fingerstick blood and oral mucosal transudate (OraQuick results only reported in this table).

Study design

Diagnostic test accuracy.

Intervention

The OraQuick HCV with fingerstick blood and oral mucosal transudate.

Setting

Hepatology or infectology units of Grenoble University Hospital, France.

Inclusion/exclusion criteria

Not reported, HCV‑positive and HCV‑negative patients recruited from the hospital units (see above).

Primary outcomes

Sensitivity, specificity, storage conditions of dried blood spots and oral mucosal transudate on test performance.

Methods

For each patient, 2 oral samples were taken (1 for the OraQuick HCV, 1 for EIA). The order of each sample being taken was randomized, with 10 minutes between each.

Fingerstick blood was placed alternatively on the OraQuick Sample loop or onto Protein Saver 903 card for dried blood spot samples (dried for 24 hours, stored at −20C).

Comparator EIA: Monolisa HCV–Ag–Ab–ULTRA (BioRad).

EIA results from serum samples taken at recruitment used as baseline.

Participants

113 HCV‑positive people consecutively recruited between May and September 2011; 88 HCV‑negative people also sampled during this time.

Experience of person having test

Not reported.

Results

Diagnostic performance:

Fingerstick blood

Sensitivity 95% CI, 97.4 (92.5 to 99.1)

Specificity 95% CI, 100 (95.8 to 100)

PPV, 100

NPV, 96.2

LR+ve, Infinity

LR–ve, 0.03

Oral fluid

Sensitivity 95% CI, 94.6 (88.6 to 97.5)

Specificity 95% CI, 100 (95.7 to 100)

PPV, 100

NPV, 92.5

LR+ve, Infinity

LR–ve, 0.05

Of the fingerstick blood samples, 1 false‑negative correlated with spontaneously‑resolved HCV infection with low cEIA index value (1.70) in serum.

There were 2 others that were HCV RNA‑negative, HIV‑positive.

Adverse events

None reported.

Conclusions

Commercial EIA can be used on stored samples as a reliable alternative to OraQuick point‑of‑care testing.

Abbreviations: cEIA, commercial enzyme immunoassay; CI, confidence interval; EIA, enzyme immunoassay; HCV, hepatitis C virus; LR, likelihood ratio; NPV, negative predictive value; PPV, positive predictive value; RNA, ribonucleic acid.

Study component

Description

Objectives/hypotheses

To evaluate the performance of the OraQuick HCV with venous blood, fingerstick blood, serum, plasma, and oral fluid, compared with FDA‑approved laboratory methods.

Study design

Diagnostic test accuracy.

Intervention

The OraQuick HCV using venous blood, fingerstick blood, serum, plasma, and oral fluid.

Setting

Not reported.

Inclusion/exclusion criteria

Not reported.

Primary outcomes

Sensitivity, specificity, factors predictive of false‑positive and false‑negative test results.

Methods

People with HCV infection (n=122) and at low‑risk of HCV infection (n=450) samples (oral fluid, venous whole blood, fingerstick blood, plasma and serum) tested with the OraQuick HCV and comparator tests.

Comparators: EIA (make/manufacturer not named), RIBA strip immunoblot assay.

A positive HCV diagnosis was made when there was a positive EIA and RIBA, or a positive EIA and PCR.

Sensitivity to anti‑HCV antibodies seroconversion was tested in 19 commercially‑available panels of plasma specimens sequentially collected from individuals having HCV seroconversion following recent infection.

Participants

122 HCV‑positive people and 450 low HCV‑risk people, of unknown HCV status.

Results

121/122 (99.2%) detected by the OraQuick HCV.

There was 1 case that tested negative in oral fluid but positive in all other samples. The serum from this individual was positive on EIA and RIBA but PCR‑negative.

There were 449/450 (99.8%) correctly identified as HCV negative by the OraQuick HCV.

There was 1 patient whose sample gave a false positive (RIBA and PCR negative) and 1 low‑risk patient correctly identified as HCV positive by the OraQuick HCV, confirmed by control tests.

Oral fluid sensitivity 95% CI, 99.2% (95.5 to 100)

Oral fluid specificity 95% CI, 100% (99.2 to 100)

Venous whole blood sensitivity 95% CI, 100% (97.0 to 100)

Venous whole blood specificity 95% CI, 100% (99.2 to 100)

Fingerstick blood sensitivity 95% CI, 100% (97.0 to 100)

Fingerstick blood specificity 95% CI, 100% (99.2 to 100)

Plasma sensitivity 95% CI, 100% (97.0 to 100)

Plasma specificity 95% CI, 99.8% (98.8 to 100)

Serum sensitivity 95% CI, 100% (97.0 to 100)

Serum specificity 95% CI, 99.8% (98.8 to 100)

Mean time to detection of seroconversion was 61.3 days in EIA and 56.4 days in the OraQuick HCV.

Of 19 seroconversion panels tested, HCV antibody was detected at the same time by the OraQuick HCV and EIA in 9 cases and earlier by the OraQuick HCV in 10. In no cases did EIA detect before the OraQuick HCV.

EAC diagnostic performance calculations:

HCV‑positive people

Sensitivity (%), 99.18

PPV (%),100.00, NPV (%), 0.00

Low‑risk people

Sensitivity (%), 100.00; Specificity (%), 99.78

PPV (%), 50.00; NPV (%), 100.00; LR (+ve), 449.0; LR (−ve), 0.00

This study had separate 2×2 tables within the publication, as they sampled HCV‑positive people in 1 group and low‑risk HCV‑status‑unknown people in another. These were kept separated, as they do not reflect population prevalence in a real‑world scenario.

Adverse events

None reported.

Conclusions

The OraQuick HCV appears to have a sensitivity and specificity equivalent to laboratory‑based tests, even when antibody levels are low, and is suitable as an aid in the diagnosis of HCV infection.

Abbreviations: CI, confidence interval; EIA, enzyme immunoassay; HCV, hepatitis C virus; PCR, polymerase chain reaction; RIBA, recombinant immunoblot assay.

Study component

Description

Objectives/ hypotheses

To evaluate the prospective diagnostic performance of the OraQuick HCV with venous blood, fingerstick blood, serum, plasma, and oral fluid, compared to FDA‑approved laboratory methods, in HCV‑at‑risk people.

Study design

Diagnostic test accuracy.

Intervention

The OraQuick HCV using venous blood, fingerstick blood, serum, plasma, and oral fluid.

Setting

8 separate clinical testing sites, otherwise not reported.

Inclusion/exclusion criteria

People at risk of HCV infection, or who had signs and/or symptoms of HCV infection.

Risk factors included: had injected intravenous drugs; born to HCV‑positive mother; had sex with known HCV‑positive partner; had sex with more than 2 different sexual partners in the last 6 months; had sex with an intravenous drug user; currently have or ever had a sexually transmitted disease; have been on long‑term haemodialysis; HIV‑positive; had a blood transfusion, blood product or organ transplant prior to 1992; have been incarcerated.

Primary outcomes

Sensitivity, specificity, interference factors for oral testing.

Methods

There were 2206‑HCV positive and low‑risk HCV samples tested with the OraQuick HCV and comparator tests (venous blood, fingerstick blood, serum, plasma, and oral fluid). Each sample type was tested for each patient in an unblinded fashion.

Comparators: EIA (Abbott AxSym); RIBA (Chiron); PCR (COBAS); AMPLICOR Hepatitis C Virus Test v2.0 (Roche).

People were diagnosed HCV‑positive with both positive EIA and positive RIBA, or positive EIA and positive PCR.

Oral interference study:

There were 50 HCV‑negative samples collected and tested under the following conditions: gingivitis present, use of dentures, consumption of food or beverage, use of tobacco products, use of oral care products (tooth brushing, mouthwash, and tooth whitening). Specimens were tested 5, 10, 15 and 30 minutes post‑exposure to the interfering condition.

Seroconversion panel study:

Sensitivity to anti‑HCV antibodies seroconversion was tested in 27 commercially‑available panels of plasma specimens sequentially collected from individuals having HCV seroconversion following recent infection.

Participants

There were 2206 people at risk of HCV and/or with signs or symptoms of hepatitis who took part. Of these, 123 (5.6%) had symptoms of hepatitis and 1930 (87.5%) were asymptomatic.

Results

Sensitivity and specificity

There were 2183 people classed as HCV‑positive (757, 34.3%) or HCV negative (1426, 64.6%). In total 23 were excluded due to indeterminate RIBA test and negative PCR.

Sensitivities (99.7 to 99.9%) for venous blood, fingerstick blood, serum and plasma. Lower (98.1%) in oral fluid.

Specificities were equivalent (99.6‑99.9%) for all sample types.

Oral fluid

Sensitivity 95% CI, 98.1% (96.9 to 99.0) [739/753]

Specificity 95% CI, 99.6% (99.2 to 99.9) [1418/1423]

PPV, 99.33%; NPV, 99.02%; LR+ve, 279.31; LR–ve,0.02

Venous whole blood

Sensitivity 95% CI, 99.7% (99.9 to 100.0) [753/755]

Specificity 95% CI, 99.9% (99.5 to 100.0) [1421/1423]

PPV, 99.74%, NPV, 99.86 %, LR+ve, 709.62; LR–ve=0.00

Fingerstick blood

Sensitivity 95% CI, 99.7% (99.0 to 100.0) [752/754]

Specificity 95% CI, 99.9% (99.6 to 100.0) [1421/1422]

PPV, 99.87%; NPV, 99.86 %; LR+ve, 1418.23; LR–ve=0.00

Plasma sensitivity 95% CI, 99.9% (99.3 to 100.0) [755/756]

Plasma specificity 95% CI 99.9% (99.5 to 100.0) [1420/1422]

PPV, 99.74%; NPV, 99.93%; LR+ve, 711.06; LR–ve, 0.00

Serum sensitivity 95% CI, 99.9% (99.3 to 100.0) [756/757]

Serum specificity 95% CI, 99.9% (99.6 to 100.0) [1422/1423]

PPV, 99.87%; NPV, 99.93%; LR+ve, 1421.12; LR–ve, 0.00

(PPV, NPV and LRs calculated by EAC from raw data using 2×2 tables)

Oral interference study

Consumption of tobacco and most types of food and drink did not affect results, even at the shortest wait time of 5 minutes. A low rate of false positives at the shortest wait times (5 minutes) after use of mouthwash, whitening products, tooth brushing or acidic drinks (such as cola), but this was not observed with 15 minutes (cola) or 30 minutes (oral care products).

Seroconversion panel study:

HCV antibody detected at the same time by the OraQuick HCV and EIA in 19/27 seroconversion series.

The OraQuick HCV detected the anti‑HCV antibodies earlier than EIA in 6 cases and the EIA detected earlier in 2 cases.

Overall, the OraQuick HCV detected the antibodies 0.6 days (CI 0.1 to 1.4) days before EIA. In no series was there a large difference in seroconversion sensitivity between the OraQuick HCV and EIA (maximum 7 days).

Adverse events

None reported.

Conclusions

The OraQuick HCV demonstrated clinical performance equivalent to laboratory‑based tests, and is suitable as an aid in the diagnosis of HCV infection.

Abbreviations: CI, confidence interval; EIA, enzyme immunoassay; HCV, hepatitis C virus; PCR, polymerase chain reaction; RIBA, recombinant immunoblot assay.

Study component

Description

Objectives/hypotheses

To evaluate the use of the OraQuick HCV to screen vulnerable individuals; and linkage to care in a mobile medical clinic.

Study design

Diagnostic screening/demographic study.

Intervention

The OraQuick HCV using fingerstick blood.

Setting

Mobile medical clinic in New Haven, Connecticut, USA (March 2012 –March 2013).

Inclusion/exclusion criteria

Not reported – clients reporting to mobile medical clinic during the time period.

Primary outcomes

Examine difference between clients who preferred point‑of‑care testing compared with phlebotomy HCV testing.

Multivariate logistic models for 2 outcomes: acceptance of HCV testing and preferring point‑of‑care testing over phlebotomy.

Linkage of specialty HCV care among those testing positive using the 2 diagnostic strategies.

Methods

People were offered either:

Everyone who had a positive HCV antibody tests had blood drawn for confirmatory HCV RNA testing (RT–PCR).

Participants

Clients accepting HCV screening at the mobile medical clinic. patients from 'vulnerable neighbourhoods'.

Results

In total 438/1345 (32.6%) people accepted HCV screening.

Of these, 209/438 (47.7%) people chose point‑of‑care testing (the OraQuick HCV) and 27/438 (6.2%) people had a positive test result for anti‑HCV antibodies. No difference in HCV prevalence was detected between point‑of‑care and standard phlebotomy groups (7.7 compared with 4.8%;p=0.219).

Of the patients with a positive anti‑HCV antibody test (detected using gold‑standard screening or the OraQuick HCV), people who chose the OraQuick HCV were significantly more likely to be linked to HCV specialty care (93.8 compared with 18.2%; p<0.0001).

Adverse events

None reported.

Conclusions

Rapid point‑of‑care testing is acceptable in a mobile clinic setting, and was preferred by almost half of the people who had HCV screening. patients who selected the OraQuick HCV were statistically significantly more likely to be linked to HCV specialty care.

Abbreviations: EIA, enzyme immunoassay; HCV, hepatitis C virus; RT–PCR, reverse transcriptase polymerase chain reaction; RNA, ribonucleic acid.

Study component

Description

Objectives/hypotheses

To evaluate the diagnostic performance of the OraQuick HCV and 4 other rapid point‑of care tests for emergency testing of blood transfusions in a military setting (OraQuick results only shown here).

Study design

Diagnostic test accuracy.

Intervention

The OraQuick HCV using donor plasma samples.

Setting

Military research laboratory.

Inclusion/exclusion criteria

HCV‑positive and negative plasma samples, also whole blood to which with HCV‑positive and HCV‑negative plasma had been added.

Primary outcomes

Sensitivity, specificity, performance under stress conditions (for example extreme temperatures).

Methods

Samples of HCV‑positive plasma provided by American Red Cross. Samples of HCV Negative plasma provided by Robertson Blood Centre.

Pathogen‑free whole blood sourced from Biological Speciality Corporation.

Comparator: samples were deemed HCV‑positive if the EIA signal‑to‑cut‑off ratio was at least 1 and the RIBA result was positive.

Whole blood testing

Stress conditions:

Normal; 18–30C

Seroconversion panel study

User survey:

Ease of use was evaluated by technicians and observers who were using the device. There were 6 questions devised, each on a 5‑point Likert scale (1=very difficult; 2=difficult; 3=neither; 4=easy; 5=very easy).

Participants

There were 335 HCV‑positive and 339 HCV‑negative blood donor plasma specimens.

Pathogen‑free whole blood also spiked with HCV‑positive (n=84) and HCV‑negative (n=84) plasma.

Experience of person having test

Not reported.

Results

Sensitivity and specificity:

Sensitivity 95% CI, 99.4% (98.0 to 99.9)

Specificity 95% CI, 99.7% (98.6 to 100.0)

Positive likelihood ratio 95% CI, 336.9 (47.6 to 2385.4)

Negative likelihood ratio95% CI, 0.01 (0.00 to 0.02)

Whole blood testing

Normal conditions

Sensitivity 95% CI, 98.8 (94.3 to 99.9)

Specificity 95% CI, 98.8 (94.3 to 99.9)

LR +ve 95% CI, 83.0 (11.8 to 582.5)

LR –ve 95% CI, 0.01 (0.00 to 0.08)

Hot storage

Sensitivity 95% CI, 97.6 (92.4 to 99.6)

Specificity 95% CI, 100 (96.5 to 100)

LR –ve 95% CI, 0.02 (0.01 to 0.09)

Hot testing

Sensitivity 95% CI, 97.6 (92.4 to 99.6)

Specificity 95% CI, 100 (96.5 to 100)

LR –ve 95% CI, 0.02 (0.01 to 0.09)

Cold storage

Sensitivity 95% CI, 97.6 (92.4 to 99.6)

Specificity 95% CI, 100 (94.5 to 100)

LR –ve 95% CI, 0.02 (0.01 to 0.09)

Seroconversion panel study

The OraQuick HCV was 1 of 2 rapid tests that demonstrated the best performance and most consistency with the reference standard.

User survey:

The OraQuick HCV was rated a maximum score of 5 for all survey questions.

Adverse events

None reported.

Conclusions

The data support OraQuick HCV superiority over the other rapid tests.

Abbreviations: CI, confidence interval; EIA, enzyme immunoassay; HCV, hepatitis C virus; NLR, negative likelihood ratio; PCR, polymerase chain reaction; PLR, positive likelihood ratio; RIBA, recombinant immunoblot assay.

Study component

Description

Objectives/hypotheses

Evaluate performance of rapid tests for anti‑HCV antibodies detection in sera, whole blood and oral fluid samples from individuals with different endemicity profiles and risk behaviours.

Study design

Diagnostic test accuracy.

Intervention

The OraQuick HCV in serum, whole blood and oral fluid samples.

Setting

Brazil, field samples:

Group 1: from a viral hepatitis ambulatory clinic in Rio de Janeiro

Group 2: remote areas in northern and mid‑west regions of Brazil (Tocantins and Mato Grosso do Sul states)

Group 3: people from southeast and northeast regions of Brazil who use crack cocaine and beauty professionals living in Rio de Janeiro state.

Inclusion/exclusion criteria

Group 1: people aged 18 years or older, who read and signed the consent form, and who were either symptomatic or asymptomatic for hepatitis but had at least 1 risk factor (history of intravenous drug use; sexual intercourse with a known carrier of HCV; sexually transmitted disease; long‑term haemodialysis; surgery or blood transfusion before 1994; or piercing). Also the inclusion of 120 individuals for blood/serum samples in OraQuick testing.

Group 2: people who read and signed the consent form, or whose parents or legal guardians read and signed consent form.

Group 3: people who had used crack cocaine within the last 12 months. Beauty professionals (manicurists, pedicurists or hairdressers) who had worked in the last 30 days.

Primary outcomes

Sensitivity, specificity, reproducibility, cross‑reactivity with other infectious agents.

Methods

Field samples: comparator EIA: HCV Ab (Radim) used to test serum samples, confirmed by COBAS AMPLICOR HCV Test 2.0 (Roche Diagnostics). HCV RNA‑reactive samples genotyped using Versant HCV Genotype Assay 2.0 (Siemens).

The OraQuick HCV was tested with 120 paired serum, whole blood and oral fluid samples, evaluated under laboratory conditions. Whole blood and serum samples collected by venepuncture. Oral fluid collected by the OraQuick HCV swab. OraQuick result read after 40 minutes.

Participants

Group 1: 172 suspected cases sampled

Group 2: 459 low‑risk individuals

Group 3: 43 high‑risk individuals (people who use crack cocaine)

Results

Oral fluid testing:

Sensitivity, specificity and other diagnostic performance:

Anti‑HCV antibody positive

Sensitivity 95% CI, 88.52 (81.5 to 93.58)

Specificity 95% CI, 100 (92.89 to 100)

PPV 95% CI, 100 (96.64 to 100)

NPV 95% CI, 78.13 (66.03 to 87.49)

Kappa statistic, 81.77 (72.62 to 90.29)

HCV RNA positive

Sensitivity 95% CI, 95.35 (88.52 to 98.72)

Specificity 95% CI, 100 (92.89 to 100)

PPV 95% CI, 100 (95.6 to 100)

NPV 95% CI, 92.59 (82.11 to 97.94)

Kappa statistic, 93.78 (87.77 to 99.79)

Diagnostic performance in high risk (Group 3, n=43):

TP, 0

TN, 43

FP, 0

FN, 0

Diagnostic performance in low prevalence (Group 2, n=459):

TP, 0

TN, 455

FP, 0

FN, 4

Blood testing

Serum

Sensitivity 95% CI, 100 (95.55 to 100)

Specificity 95% CI, 100 (90.97 to 100

PPV 95% CI, 100 (95.55 to 100)

NPV 95% CI, 100 (90.97 to 100)

Kappa statistic,100

Whole blood

Sensitivity 95% CI, 98.77 (93.31 to 99.97)

Specificity 95% CI, 100 (90.97 to 100)

PPV 95% CI, 100 (95.49 to 100)

NPV 95% CI, 97.5 (86.84 to 99.94)

Kappa statistic, 98.1

Whole blood

Sensitivity 95% CI, 97.53 (91.36 to 99.70)

Specificity 95% CI, 100 (90.97 to 100)

PPV 95% CI, 100 (95.44 to 100)

NPV 95% CI, 95.12 (83.47 to 99.40)

Kappa statistic, 96.25

Adverse events

None reported.

Conclusions

The OraQuick performed the best (of all rapid tests) using oral fluid samples from low‑prevalence individuals. It did not perform as well as BioEasy (a different rapid test) in high‑prevalence individuals, using whole blood. All rapid tests could be used to identify active HCV infection among individuals from different backgrounds and risk profiles.

Abbreviations: CI, confidence interval; EIA, enzyme immunoassay; FN, false negative; FP, false positive; TN, true negative; TP, true positive; PPV, positive predictive value; NPV, negative predictive value; HCV, hepatitis C virus; PCR, polymerase chain reaction; RNA, ribonucleic acid.

Study component

Description

Objectives/hypotheses

To assess the diagnostic performance of 3 rapid, POC anti‑HCV antibodies assays compared to standard laboratory methods (data presented here for OraQuick only).

Study design

Diagnostic test accuracy.

Intervention

OraQuick using either fingerstick blood or oral fluid.

Setting

Community‑based HIV and HCV testing. Two of 21 US cities participating in the National HIV Behavioural Surveillance System (NHBS) studied OraQuick (New York City and Seattle).

Inclusion/exclusion criteria

People who inject drugs (in the last 12 months), aged 18 years or over, living in the participating metropolitan statistical area, who consented to participate and who could complete a survey in English or Spanish.

Primary outcomes

Sensitivity, specificity, factors predictive of false‑positive and false‑negative test results.

Methods

The reference standard was FDA‑approved, laboratory immunoassays according to local study protocols. Positive anti‑HCV antibodies specimens were then confirmed with a third‑generation recombinant immunoblot assay. Two reference standards were used:

Screening assay used as per manufacturer's recommendation: CDC‑recommended algorithm.

Participants

The entire cohort size (3 index technologies) was 1861 patients. Results for the OraQuick HCV are available for 816 tests in 550 patients: of either blood (266 tests in 266 patients in Seattle only) or saliva (550 tests in 550 patients in New York and Seattle).

Results

False‑positive or false‑negative results were not associated with HIV status, gender or birth year.

Use of the CDC reference standard criteria resulted in slightly higher sensitivity across the test, but not statistically significantly so.

New York City (oral)

SA (n=285) Sensitivity 95% CI, 94.4 (90.4 to 96.8)

Specificity 95% CI, 95.8 (88.5 to 98.6)

PPV, 98.6; NPV, 85.7

LR+ve, 23.7; LR−ve, 0.06

CDC (n=285) Sensitivity 95% CI, 94.7 (90.8 to 97.0)

Specificity 95% CI, 92.1 (83.8 to 96.3)

PPV, 97.2; NPV, 86.2

LR+ve, 12.8; LR−ve, 0.06

Seattle (oral)

SA (n=265) Sensitivity 95% CI, 90.8 (85.4 to 93.7)

Specificity 95% CI, 98.6 (92.4 to 99.8)

PPV, 100; NPV, 78.8

LR+ve, N/A; LR−ve, 0.09

CDC (n=264) Sensitivity 95% CI, 92.2 (87.5 to 95.2)

Specificity 95% CI, 97.2 (90.9 to 99.3)

PPV, 98.5; NPV, 81.8

LR+ve, 25.7 LR−ve, 0.09

Seattle (fingerstick blood)

SA (n=266) Sensitivity 95% CI, 95.9 (91.6 to 97.6)

Specificity 95% CI, 100 (94.9 to 100.0)

PPV=100; NPV=89.7

LR+ve, N/A; LR‑ve, 0.04

CDC (n=265) Sensitivity 95% CI, 97.4 (94.1 to 98.9)

Specificity 95% CI, 98.6 (92.9 to 99.8)

PPV, 100; NPV, 93.1

LR+ve, N/A; LR−ve, 0.03

Adverse events

None reported.

Conclusions

The authors reported considerable variability in diagnostic performance across sites.

Abbreviations: CI, confidence interval; CDC, US Centers for Disease Control and Prevention; FDA, US Food and Drug Administration; HCV, hepatitis C virus; LR(−), negative likelihood ratio; LR(+), positive likelihood ratio; NPV, negative predictive value; POC, point of care; . PPV, positive predictive value; SA, (manufacturer's) screening assay.