Evaluation of a Rapid Assay as an Alternative to Conventional Enzyme Immunoassays for Detection of Hepatitis C Virus-Specific Antibodies
http://www.100md.com
微生物临床杂志 2005年第4期
Department of Clinical Virology, Christian Medical College, Vellore, India
ABSTRACT
A rapid membrane flow-through immunoassay to detect antibodies to hepatitis C virus was compared with a commercial enzyme immunoassay (EIA) and microparticle enzyme immunoassay (MEIA) using 2,590 serum samples. Sensitivity and specificity of the "rapid assay" in comparison to the EIA/MEIA were 99.3 and 99.0%; the correlation coefficient being 0.91. This assay is suitable where infrastructure and laboratory expertise are limited.
TEXT
India has an estimated 15 million hepatitis C virus (HCV) carriers today (1, 4). Despite HCV screening being made mandatory in blood banks in India since June 2001, it is not scrupulously followed. HCV antibody (HCV-Ab) screening by conventional enzyme immunoassay (EIA) continues to be expensive and is not used by most blood banks and clinical laboratories in India for want of adequate infrastructure and trained personnel. To curtail expenses, smaller laboratories and blood banks perform HCV-Ab screening by batch testing which delays the reporting process. Additionally, there is a need for rapid screening for blood-borne viruses prior to invasive procedures to enable health care workers to take adequate precautions without denial of services for patients. Against this background, a rapid, relatively inexpensive commercial assay (U.S. dollars = $3.0/test) was evaluated at the Department of Clinical Virology, Christian Medical College, Vellore, India.
Blood samples were received in our laboratory from preoperative patients prior to high-risk procedures or from the delivery room of the Department of Obstetrics of this tertiary care center. HCV-Ab testing was done with the sole purpose of ensuring appropriate patient handling, the required medical or surgical treatment not being withheld from any patient. General consent is obtained in this center for screening for all blood-borne agents prior to investigations.
In this study, a total of 2,590 serum samples were received from 1,571 (61%) low-risk individuals and 1,019 (39%) high-risk individuals for the purpose of screening for HCV-Ab. Low-risk individuals were preoperative patients (n = 1,421), antenatal women (n = 50), and blood donors (n = 100), while high-risk individuals were patients from the Departments of Gastroenterology, Hepatology, and Nephrology.
All sera in the panel were tested with the rapid assay HCV TRI DOT (J. MITRA &Co. Ltd., New Delhi, India), which is a visual, qualitative, fourth generation HCV-Ab screening assay, based on flowthrough technology, utilizing a unique combination of modified antigens from the putative core, NS3, NS4, and NS5 regions of HCV. These antigens are immobilized on a porous immunofiltration membrane, which includes two test dots, T1 and T2, and an additional dot serving as a quality control or serum control dot. As sample passes through the membrane, HCV-Ab in the serum/plasma, binds to the immobilized antigen on the absorbent pad. Unbound serum or plasma proteins are removed by subsequent washing. Protein A conjugate is then added, which binds to the Fc portion of HCV-specific immunoglobulin G to give a distinct pinkish-purple dot in the test region. The device is user friendly with a built-in control. The control dot should develop color after addition of the patient's serum irrespective of color development in the two test dots (T1 and T2), thereby confirming proper functioning of the device, reagents, and correct procedural application. This control dot thus serves as a built-in quality control. Turnaround time of the test is 5 min. The rapid assay results were compared with a third generation EIA (UBI, HCV EIA 4.0). Positives by EIA and/or microparticle enzyme immunoassay (MEIA) (Axsym; Abbott Laboratories, Ill.) were considered as positives in the panel, and those that were negative by EIA were considered negative. Samples positive by EIA/MEIA were retested in duplicate. All assays were performed as per the manufacturers' instructions. Samples that gave a discrepant result in the rapid assay as compared to EIA/MEIA were retested to rule out any defect in the device. A recombinant immunoblot assay (RIBA) (CHIRON RIBA HCV 3.0 SIA; Ortho-Clinical Diagnostics, Inc., Raritan, N.J.) was done on all samples that yielded discrepant results between the rapid assay and EIA and/or MEIA. Further, HCV RNA testing was done on such discrepant samples by an in-house reverse transcription-PCR (RT-PCR) standardized in this laboratory (2).
In addition, 60 samples from HCV-infected individuals with known genotypes were used to evaluate the capacity of the rapid assay to detect different HCV genotypes prevalent in this region. Genotyping was done with an earlier published type-specific RT-PCR performed in this center (3).
Of the 2,590 study samples, 2,451 samples were negative and 139 were positive by the EIA/MEIA method. Of the 139 positives, 25 (18%) were from low-risk individuals and 114 (82%) were from high-risk individuals. Of these 139 EIA/MEIA positives, 138 samples were positive by the rapid assay while one sample (from a high-risk individual) was negative. This sample was of low reactivity in both EIA and MEIA (sample optical density [S/CO] = 4.0) and was RNA positive. Of the 114 high-risk patient samples, one sample was negative by EIA but positive in MEIA and the rapid assay. RIBA performed on this sample showed positivity for c33c and c100, 5-1-1 antigens: c33c is a recombinant antigen derived from the putative nonstructural (NS3) region, while c100 and 5-1-1 are synthetic peptides derived from the nonstructural (NS4) region of the virus. Of the 2,451 EIA/MEIA negatives, 2,427 were negative by the rapid assay. The high sensitivity (99.3%) of this assay (Table 1) shows that a high proportion of specimens known to contain HCV-Ab (reference test positive) were positive in the rapid assay. The high specificity (99.0%) on the other hand shows that a high proportion of specimens known not to contain HCV-Ab (reference test negative) were negative in this test assay. The likelihood that a person testing negative is truly uninfected (i.e., the negative predictive value [NPV]) is high for this assay, demonstrating its usefulness as a screening test. The likelihood that a person testing positive is truly infected (i.e., the positive predictive value [PPV]) is high but lower than the NPV, suggesting that rapid assay positives may need confirmation by a second assay. The overall degree of agreement between the rapid assay and the EIA/MEIA was excellent (kappa coefficient = 0.91) (Table 1).
Of the 60 additional genotyped samples tested, 4 were missed in the rapid assay. Among these, 3 were genotype 1 and 1 was genotype 3b (Table 2). RIBA was done to confirm the antibody status of these discrepant samples. Of the three genotype 1 samples missed by the rapid assay, one was negative by EIA (S/CO = 0.5) but positive by MEIA (S/CO = 46.3), giving an indeterminate result in RIBA. While the second sample was positive in both systems with S/CO values of 5.8 in EIA and 46.6 in MEIA and gave a positive result in RIBA. The sample volume of the third genotype 1 sample was insufficient for RIBA testing. The genotype 3 sample, which was missed by the rapid assay, was indeterminate by RIBA, yielding a band for the c33c antigen only.
Additionally, 30 samples which included 10 each of the HCV-Ab positives and negatives, 8 weak-positive samples and 2 indeterminate samples (by RIBA) were included to evaluate the interassay variability of the rapid assay. All HCV-Ab-positive, -negative, and -indeterminate samples gave reproducible results on repeat testing. One of the eight weak-positive samples tested negative on retesting, while the remaining seven yielded consistent results.
Screening for HCV is difficult in the community setting. It is important to have a specific, sensitive, simple, and convenient assay for the screening and diagnosis of HCV infection. In this study, we found a good correlation between the rapid assay and the EIA/MEIA HCV-Ab assays. In the second panel of sera obtained from HCV-infected genotyped individuals, the rapid assay detected all HCV genotypes except genotypes 1 and 3, where only 81.3 and 96.8%, respectively, were detected by the rapid assay. These findings suggest that diagnostic kits for HCV-Ab detection particularly for genotype 1 have to be developed with regional strains. Twenty-four samples from low-risk individuals, which were negative by the EIA, gave a positive result in the rapid assay. This again could be due to the differences in configurations of antigens between assays (5). Given the high sensitivity and specificity of over 99.0% and a correlation coefficient of >0.9, this rapid assay can be used as an alternative to EIA or MEIA in health care settings where infrastructure and trained personnel are unavailable. Due to a built-in control, this rapid assay can be performed even for a single sample, avoiding batch testing, thus making HCV-Ab screening quick and cost-effective. This device is easy to interpret and is eminently applicable to a field setting.
ACKNOWLEDGMENTS
This study was partially funded by Fluid Research Grant 4847 from Christian Medical College, Vellore, India.
REFERENCES
Nanu, A., S. P. Sharma, K. Chatterjee, and P. Jyoti. 1997. Markers for transfusion-transmissible infections in north Indian voluntary and replacement blood donors: prevalence and trends 1989-1996. Vox Sang. 73:70-73.
Radhakrishnan, S., P. Abraham, S. Raghuraman, G. T. John, P. P. Thomas, C. K. Jacob, and G. Sridharan. 2000. Role of molecular techniques in the detection of HBV DNA & HCV RNA among renal transplant recipients in India. Indian J. Med. Res. 111:204-211.
Raghuraman, S., R. V. Shaji, G. Sridharan, S. Radhakrishnan, G. Chandy, B. S. Ramakrishna, and P. Abraham. 2003. Distribution of the different genotypes of HCV among patients attending a tertiary care hospital in south India. J. Clin. Virol. 26:61-69.
World Health Organization. 1997. Hepatitis C: global prevalence. Wkly. Epidemiol. Rec. 46:341-348.
Zhang, Z.-X., M. Chen, A. Snnerborg, and M. Sllberg. 1994. Antigenic structure of the complete nonstructural (NS) 2 and 5 proteins of the hepatitis C virus (HCV): anti-HCV NS2 and NS5 antibody reactivities in relation to HCV serotype, presence of HCV RNA, and acute HCV infection. Clin. Diagn. Lab. Immunol. 1:290-294.(Hubert Darius J. Daniel, )
ABSTRACT
A rapid membrane flow-through immunoassay to detect antibodies to hepatitis C virus was compared with a commercial enzyme immunoassay (EIA) and microparticle enzyme immunoassay (MEIA) using 2,590 serum samples. Sensitivity and specificity of the "rapid assay" in comparison to the EIA/MEIA were 99.3 and 99.0%; the correlation coefficient being 0.91. This assay is suitable where infrastructure and laboratory expertise are limited.
TEXT
India has an estimated 15 million hepatitis C virus (HCV) carriers today (1, 4). Despite HCV screening being made mandatory in blood banks in India since June 2001, it is not scrupulously followed. HCV antibody (HCV-Ab) screening by conventional enzyme immunoassay (EIA) continues to be expensive and is not used by most blood banks and clinical laboratories in India for want of adequate infrastructure and trained personnel. To curtail expenses, smaller laboratories and blood banks perform HCV-Ab screening by batch testing which delays the reporting process. Additionally, there is a need for rapid screening for blood-borne viruses prior to invasive procedures to enable health care workers to take adequate precautions without denial of services for patients. Against this background, a rapid, relatively inexpensive commercial assay (U.S. dollars = $3.0/test) was evaluated at the Department of Clinical Virology, Christian Medical College, Vellore, India.
Blood samples were received in our laboratory from preoperative patients prior to high-risk procedures or from the delivery room of the Department of Obstetrics of this tertiary care center. HCV-Ab testing was done with the sole purpose of ensuring appropriate patient handling, the required medical or surgical treatment not being withheld from any patient. General consent is obtained in this center for screening for all blood-borne agents prior to investigations.
In this study, a total of 2,590 serum samples were received from 1,571 (61%) low-risk individuals and 1,019 (39%) high-risk individuals for the purpose of screening for HCV-Ab. Low-risk individuals were preoperative patients (n = 1,421), antenatal women (n = 50), and blood donors (n = 100), while high-risk individuals were patients from the Departments of Gastroenterology, Hepatology, and Nephrology.
All sera in the panel were tested with the rapid assay HCV TRI DOT (J. MITRA &Co. Ltd., New Delhi, India), which is a visual, qualitative, fourth generation HCV-Ab screening assay, based on flowthrough technology, utilizing a unique combination of modified antigens from the putative core, NS3, NS4, and NS5 regions of HCV. These antigens are immobilized on a porous immunofiltration membrane, which includes two test dots, T1 and T2, and an additional dot serving as a quality control or serum control dot. As sample passes through the membrane, HCV-Ab in the serum/plasma, binds to the immobilized antigen on the absorbent pad. Unbound serum or plasma proteins are removed by subsequent washing. Protein A conjugate is then added, which binds to the Fc portion of HCV-specific immunoglobulin G to give a distinct pinkish-purple dot in the test region. The device is user friendly with a built-in control. The control dot should develop color after addition of the patient's serum irrespective of color development in the two test dots (T1 and T2), thereby confirming proper functioning of the device, reagents, and correct procedural application. This control dot thus serves as a built-in quality control. Turnaround time of the test is 5 min. The rapid assay results were compared with a third generation EIA (UBI, HCV EIA 4.0). Positives by EIA and/or microparticle enzyme immunoassay (MEIA) (Axsym; Abbott Laboratories, Ill.) were considered as positives in the panel, and those that were negative by EIA were considered negative. Samples positive by EIA/MEIA were retested in duplicate. All assays were performed as per the manufacturers' instructions. Samples that gave a discrepant result in the rapid assay as compared to EIA/MEIA were retested to rule out any defect in the device. A recombinant immunoblot assay (RIBA) (CHIRON RIBA HCV 3.0 SIA; Ortho-Clinical Diagnostics, Inc., Raritan, N.J.) was done on all samples that yielded discrepant results between the rapid assay and EIA and/or MEIA. Further, HCV RNA testing was done on such discrepant samples by an in-house reverse transcription-PCR (RT-PCR) standardized in this laboratory (2).
In addition, 60 samples from HCV-infected individuals with known genotypes were used to evaluate the capacity of the rapid assay to detect different HCV genotypes prevalent in this region. Genotyping was done with an earlier published type-specific RT-PCR performed in this center (3).
Of the 2,590 study samples, 2,451 samples were negative and 139 were positive by the EIA/MEIA method. Of the 139 positives, 25 (18%) were from low-risk individuals and 114 (82%) were from high-risk individuals. Of these 139 EIA/MEIA positives, 138 samples were positive by the rapid assay while one sample (from a high-risk individual) was negative. This sample was of low reactivity in both EIA and MEIA (sample optical density [S/CO] = 4.0) and was RNA positive. Of the 114 high-risk patient samples, one sample was negative by EIA but positive in MEIA and the rapid assay. RIBA performed on this sample showed positivity for c33c and c100, 5-1-1 antigens: c33c is a recombinant antigen derived from the putative nonstructural (NS3) region, while c100 and 5-1-1 are synthetic peptides derived from the nonstructural (NS4) region of the virus. Of the 2,451 EIA/MEIA negatives, 2,427 were negative by the rapid assay. The high sensitivity (99.3%) of this assay (Table 1) shows that a high proportion of specimens known to contain HCV-Ab (reference test positive) were positive in the rapid assay. The high specificity (99.0%) on the other hand shows that a high proportion of specimens known not to contain HCV-Ab (reference test negative) were negative in this test assay. The likelihood that a person testing negative is truly uninfected (i.e., the negative predictive value [NPV]) is high for this assay, demonstrating its usefulness as a screening test. The likelihood that a person testing positive is truly infected (i.e., the positive predictive value [PPV]) is high but lower than the NPV, suggesting that rapid assay positives may need confirmation by a second assay. The overall degree of agreement between the rapid assay and the EIA/MEIA was excellent (kappa coefficient = 0.91) (Table 1).
Of the 60 additional genotyped samples tested, 4 were missed in the rapid assay. Among these, 3 were genotype 1 and 1 was genotype 3b (Table 2). RIBA was done to confirm the antibody status of these discrepant samples. Of the three genotype 1 samples missed by the rapid assay, one was negative by EIA (S/CO = 0.5) but positive by MEIA (S/CO = 46.3), giving an indeterminate result in RIBA. While the second sample was positive in both systems with S/CO values of 5.8 in EIA and 46.6 in MEIA and gave a positive result in RIBA. The sample volume of the third genotype 1 sample was insufficient for RIBA testing. The genotype 3 sample, which was missed by the rapid assay, was indeterminate by RIBA, yielding a band for the c33c antigen only.
Additionally, 30 samples which included 10 each of the HCV-Ab positives and negatives, 8 weak-positive samples and 2 indeterminate samples (by RIBA) were included to evaluate the interassay variability of the rapid assay. All HCV-Ab-positive, -negative, and -indeterminate samples gave reproducible results on repeat testing. One of the eight weak-positive samples tested negative on retesting, while the remaining seven yielded consistent results.
Screening for HCV is difficult in the community setting. It is important to have a specific, sensitive, simple, and convenient assay for the screening and diagnosis of HCV infection. In this study, we found a good correlation between the rapid assay and the EIA/MEIA HCV-Ab assays. In the second panel of sera obtained from HCV-infected genotyped individuals, the rapid assay detected all HCV genotypes except genotypes 1 and 3, where only 81.3 and 96.8%, respectively, were detected by the rapid assay. These findings suggest that diagnostic kits for HCV-Ab detection particularly for genotype 1 have to be developed with regional strains. Twenty-four samples from low-risk individuals, which were negative by the EIA, gave a positive result in the rapid assay. This again could be due to the differences in configurations of antigens between assays (5). Given the high sensitivity and specificity of over 99.0% and a correlation coefficient of >0.9, this rapid assay can be used as an alternative to EIA or MEIA in health care settings where infrastructure and trained personnel are unavailable. Due to a built-in control, this rapid assay can be performed even for a single sample, avoiding batch testing, thus making HCV-Ab screening quick and cost-effective. This device is easy to interpret and is eminently applicable to a field setting.
ACKNOWLEDGMENTS
This study was partially funded by Fluid Research Grant 4847 from Christian Medical College, Vellore, India.
REFERENCES
Nanu, A., S. P. Sharma, K. Chatterjee, and P. Jyoti. 1997. Markers for transfusion-transmissible infections in north Indian voluntary and replacement blood donors: prevalence and trends 1989-1996. Vox Sang. 73:70-73.
Radhakrishnan, S., P. Abraham, S. Raghuraman, G. T. John, P. P. Thomas, C. K. Jacob, and G. Sridharan. 2000. Role of molecular techniques in the detection of HBV DNA & HCV RNA among renal transplant recipients in India. Indian J. Med. Res. 111:204-211.
Raghuraman, S., R. V. Shaji, G. Sridharan, S. Radhakrishnan, G. Chandy, B. S. Ramakrishna, and P. Abraham. 2003. Distribution of the different genotypes of HCV among patients attending a tertiary care hospital in south India. J. Clin. Virol. 26:61-69.
World Health Organization. 1997. Hepatitis C: global prevalence. Wkly. Epidemiol. Rec. 46:341-348.
Zhang, Z.-X., M. Chen, A. Snnerborg, and M. Sllberg. 1994. Antigenic structure of the complete nonstructural (NS) 2 and 5 proteins of the hepatitis C virus (HCV): anti-HCV NS2 and NS5 antibody reactivities in relation to HCV serotype, presence of HCV RNA, and acute HCV infection. Clin. Diagn. Lab. Immunol. 1:290-294.(Hubert Darius J. Daniel, )