Confirming Positive Results of Nucleic Acid Amplification Tests (NAATs) for Chlamydia trachomatis: All NAATs Are Not Created Equal
http://www.100md.com
微生物临床杂志 2005年第3期
University of California, San Francisco, California
University of Alabama, Birmingham, Alabama
Louisiana State University Medical Center, New Orleans, Louisiana
Florida State Department of Health, Jacksonville, Florida
Planned Parenthood Foundation of Houston and Southeast Texas, Houston, Texas
Wishard Memorial Hospital, Indianapolis, Indiana
Magee-Womens Research Institute, Pittsburgh, Pennsylvania
University of Illinois, Chicago, Illinois
St. Joseph's Health Care Regional Virology and Chlamydiology Laboratory, Hamilton, Ontario, Canada
ABSTRACT
The Centers for Disease Control and Prevention recommended confirming positive screening tests for Chlamydia trachomatis when positive predictive values are <90%. It is accepted that less sensitive tests (i.e., culture and immunoassays) should not be used to confirm the results of more sensitive nucleic acid amplification tests (NAATs). We show that the same principle applies when NAATs are used for confirmation.
TEXT
Chlamydia trachomatis is the most common sexually transmitted bacterial pathogen (4). It causes adverse reproductive consequences in women and conjunctivitis and pneumonia in infants exposed to the organism at birth. Screening women in family planning or in obstetrics and gynecology clinics and treatment of those infected with C. trachomatis prevents pelvic inflammatory disease and neonatal infections (7, 8). Because most genital chlamydial infections are asymptomatic, Centers for Disease Control and Prevention (CDC) guidelines call for routine screening of sexually active women between ages 15 and 24.
Nucleic acid amplification tests (NAATs) are the most sensitive tests for the screening and diagnosis of genital chlamydial infections (6). NAATs are more sensitive than previously available diagnostic tests (culture, antigen detection, or nucleic acid hybridization) by at least 20 to 30%. That NAATs can be used with noninvasively collected specimens, such as first-catch urine samples (FCU) from men or women and self- or clinician-collected vaginal swabs, improves our ability to screen for chlamydia.
As more screening for genital infections has been done by using NAATs, concerns have been raised about their specificity, particularly in screening low-prevalence populations. In some studies, positive NAAT results could not be reproduced (1-3, 5, 9). There are obvious concerns about the social ramifications of incorrectly informing individuals that they have a sexually transmitted disease based on false-positive screening test results. These concerns led the CDC to recommend confirmatory testing of positive test results when the positive predictive value is <90% (4). Several strategies for confirmation were suggested. One was to perform a second NAAT, targeting a different nucleic acid sequence, with either the original specimen or a specimen collected in duplicate.
Gen-Probe, Inc., has introduced three assays based on its transcription-mediated amplification technology: the APTIMA CT assay (ACT), which detects C. trachomatis, the APTIMA GC assay (AGC), which detects Neisseria gonorrhoeae, and the APTIMA COMBO 2 assay (AC2), which detects both pathogens. The ACT and AGC oligonucleotide probes target rRNA sequences different from those for the AC2. These tests are thus each well suited for confirming results obtained with the other. During a clinical assessment of the performance of these assays, we had an opportunity to evaluate the confirmation of NAAT-positive results by using other NAATs.
The study population consisted of 1,465 females and 1,322 males who attended sexually transmitted disease, obstetrics and gynecology, teen, or family planning clinics in seven centers in North America. FCU and endocervical swabs (CS) from women and urethral swabs and FCU from males were collected and tested by ACT and AC2. The FCU and a duplicate swab from each subject were also tested with BDProbeTec ET System C. trachomatis assay (BD; Becton Dickinson and Co. Diagnostics, Sparks, Md.). With females, FCU were collected first, and then two randomized CS were collected. In males, two randomized urethral swabs were collected first, followed by the FCU. Thus, three NAATs, each having a different target, were used to test specimens from each subject. For this evaluation, the results from the BD, ACT, and AC2 were each considered as primary screening tests, and positive results for each test were then confirmed by the two other tests. The APTIMA assays do not have a control for inhibitors; the BD test does have such a control. All results presented here were valid (there were no indeterminate results due to inhibition).
Of all of the male and female swab specimens and FCU tested, there were 850 positive results with the AC2, 927 positive results with the ACT, and 720 positive results with the BD. Both the ACT and AC2 confirmed 96.9% of the positive results with the BD. Of the positive results with the AC2, 98.1% were positive with the ACT, but only 82% were positive with the BD. Of the positive results with the ACT, the AC2 confirmed 89.8%, but the BD confirmed only 75.1%. The positive results and confirmatory tests by sex and specimen type are shown in Table 1. There were no major differences in the observed patterns by sex or specimen type.
In the CDC guidelines, it was pointed out that less sensitive diagnostic tests, such as culture and enzyme immunoassays, should not be used to confirm positive results of the more sensitive NAATs for C. trachomatis. This guideline exists because 30% or more of specimens positive by NAATs will be negative by culture or enzyme immunoassay. Our study shows that the same principle applies when only NAATs are being used. Both APTIMA assays yielded more total and more confirmed-positive results than the BD assay. If reporting a positive result is based upon confirmation of the initial positive result, then using BD to confirm positive results by ACT or AC2 would result in the incorrect reporting of approximately 15% of confirmable positive results as negative (i.e., not confirmed).
In this evaluation, the ACT had the greatest number of positive results (927), and the BD had the least number of positive results (720). While a rule of thumb has been that the less sensitive test is more specific, that was not the case here (assuming that confirmation is a proxy for specificity). Approximately 97% of the 720 positive BD results were confirmed by the more sensitive AC2 and ACT. However, the AC2 had 850 positives, 98% of which were confirmed by ACT. The AC2 confirmed only about 90% of the positive results of the ACT. There are two obvious interpretations for the latter result. It may be that the ACT is less specific and that the AC2 failed to confirm positive results because they were false positives. The alternate explanation is that ACT is actually more sensitive than the other tests and that the inability to confirm its greater number of positive results reflects the lower sensitivity of the confirmatory assays. Further studies are needed to answer this question.
It is clear from our results that some NAATs cannot be used to confirm positive results from other NAATs. Even among NAATs, the assays employed should have equivalent sensitivities. The BD should not be used to confirm the APTIMA assays. If, however, the BD were used for the primary screening, the APTIMA assays could be used for confirmation (96.9% confirmed). Based on our results, we would recommend the AC2 for initial screening and the use of the ACT for confirmation, as 98% of AC2 positive results were confirmed by ACT. This strategy would identify more infected individuals.
ACKNOWLEDGMENTS
The clinical trial was funded by Gen-Probe, Inc.
REFERENCES
Castriciano, S., K. Luinstra, D. Jang, J. Patel, J. Mahony, J. Kapala, and M. Chernesky. 2002. Accuracy of results obtained by performing a second ligase chain reaction assay and PCR analysis on urine samples with positive or near-cutoff results in the LCx test for Chlamydia trachomatis. J. Clin. Microbiol. 40:2632-2634.
Culler, E. E., A. M. Caliendo, and F. S. Nolte. 2003. Reproducibility of positive test results in the BDProbeTec ET system for detection of Chlamydia trachomatis and Neisseria gonorrhoeae. J. Clin. Microbiol. 41:3911-3914.
Gronowski, A. M., S. Copper, D. Baorto, and P. R. Murray. 2000. Reproducibility problems with the Abbott laboratories LCx assay for Chlamydia trachomatis and Neisseria gonorrhoeae. J. Clin. Microbiol. 38:2416-2418.
Johnson, R. E., W. J. Newhall, J. R. Papp, J. S. Knapp, C. M. Black, T. L. Gift, R. Steece, L. E. Markowitz, O. J. Devine, C. M. Walsh, S. Wang, D. C. Gunter, K. L. Irwin, S. DeLisle, and S. M. Berman. 2002. Screening tests to detect Chlamydia trachomatis and Neisseria gonorrhoeae infections—2002. Morb. Mortal. Wkly. Rep. Recomm. Rep. 51(RR-15):1-38;CE1-CE4.
Peterson, E. M., V. Darrow, J. Blanding, S. Aarnaes, and L. M. de la Maza. 1997. Reproducibility problems with the AMPLICOR PCR Chlamydia trachomatis test. J. Clin. Microbiol. 35:957-995.
Schachter, J. 2001. NAATs to diagnose Chlamydia trachomatis genital infection: a promise still unfulfilled. Expert Rev. Mol. Diagn. 1:137-144.
Schachter, J., R. L. Sweet, M. Grossman, D. Landers, M. Robbie, and E. Bishop. 1986. Experience with the routine use of erythromycin for chlamydial infections in pregnancy. N. Engl. J. Med. 314:276-279.
Scholes, D., A. Stergachis, F. E. Heidrich, H. Andrilla, K. K. Holmes, and W. E. Stamm. 1996. Prevention of pelvic inflammatory disease by screening for cervical chlamydial infection. N. Engl. J. Med. 334:1362-1366.
Verkooyen, R. P., G. T. Noordhoek, P. E. Klapper, J. Reid, J. Schirm, G. M. Cleator, M. Ieven, and G. Hoddevik. 2003. Reliability of nucleic acid amplification methods for detection of Chlamydia trachomatis in urine: results of the first international collaborative quality control study among 96 laboratories. J. Clin. Microbiol. 41:3013-3016.(J. Schachter, E. W. Hook,)
University of Alabama, Birmingham, Alabama
Louisiana State University Medical Center, New Orleans, Louisiana
Florida State Department of Health, Jacksonville, Florida
Planned Parenthood Foundation of Houston and Southeast Texas, Houston, Texas
Wishard Memorial Hospital, Indianapolis, Indiana
Magee-Womens Research Institute, Pittsburgh, Pennsylvania
University of Illinois, Chicago, Illinois
St. Joseph's Health Care Regional Virology and Chlamydiology Laboratory, Hamilton, Ontario, Canada
ABSTRACT
The Centers for Disease Control and Prevention recommended confirming positive screening tests for Chlamydia trachomatis when positive predictive values are <90%. It is accepted that less sensitive tests (i.e., culture and immunoassays) should not be used to confirm the results of more sensitive nucleic acid amplification tests (NAATs). We show that the same principle applies when NAATs are used for confirmation.
TEXT
Chlamydia trachomatis is the most common sexually transmitted bacterial pathogen (4). It causes adverse reproductive consequences in women and conjunctivitis and pneumonia in infants exposed to the organism at birth. Screening women in family planning or in obstetrics and gynecology clinics and treatment of those infected with C. trachomatis prevents pelvic inflammatory disease and neonatal infections (7, 8). Because most genital chlamydial infections are asymptomatic, Centers for Disease Control and Prevention (CDC) guidelines call for routine screening of sexually active women between ages 15 and 24.
Nucleic acid amplification tests (NAATs) are the most sensitive tests for the screening and diagnosis of genital chlamydial infections (6). NAATs are more sensitive than previously available diagnostic tests (culture, antigen detection, or nucleic acid hybridization) by at least 20 to 30%. That NAATs can be used with noninvasively collected specimens, such as first-catch urine samples (FCU) from men or women and self- or clinician-collected vaginal swabs, improves our ability to screen for chlamydia.
As more screening for genital infections has been done by using NAATs, concerns have been raised about their specificity, particularly in screening low-prevalence populations. In some studies, positive NAAT results could not be reproduced (1-3, 5, 9). There are obvious concerns about the social ramifications of incorrectly informing individuals that they have a sexually transmitted disease based on false-positive screening test results. These concerns led the CDC to recommend confirmatory testing of positive test results when the positive predictive value is <90% (4). Several strategies for confirmation were suggested. One was to perform a second NAAT, targeting a different nucleic acid sequence, with either the original specimen or a specimen collected in duplicate.
Gen-Probe, Inc., has introduced three assays based on its transcription-mediated amplification technology: the APTIMA CT assay (ACT), which detects C. trachomatis, the APTIMA GC assay (AGC), which detects Neisseria gonorrhoeae, and the APTIMA COMBO 2 assay (AC2), which detects both pathogens. The ACT and AGC oligonucleotide probes target rRNA sequences different from those for the AC2. These tests are thus each well suited for confirming results obtained with the other. During a clinical assessment of the performance of these assays, we had an opportunity to evaluate the confirmation of NAAT-positive results by using other NAATs.
The study population consisted of 1,465 females and 1,322 males who attended sexually transmitted disease, obstetrics and gynecology, teen, or family planning clinics in seven centers in North America. FCU and endocervical swabs (CS) from women and urethral swabs and FCU from males were collected and tested by ACT and AC2. The FCU and a duplicate swab from each subject were also tested with BDProbeTec ET System C. trachomatis assay (BD; Becton Dickinson and Co. Diagnostics, Sparks, Md.). With females, FCU were collected first, and then two randomized CS were collected. In males, two randomized urethral swabs were collected first, followed by the FCU. Thus, three NAATs, each having a different target, were used to test specimens from each subject. For this evaluation, the results from the BD, ACT, and AC2 were each considered as primary screening tests, and positive results for each test were then confirmed by the two other tests. The APTIMA assays do not have a control for inhibitors; the BD test does have such a control. All results presented here were valid (there were no indeterminate results due to inhibition).
Of all of the male and female swab specimens and FCU tested, there were 850 positive results with the AC2, 927 positive results with the ACT, and 720 positive results with the BD. Both the ACT and AC2 confirmed 96.9% of the positive results with the BD. Of the positive results with the AC2, 98.1% were positive with the ACT, but only 82% were positive with the BD. Of the positive results with the ACT, the AC2 confirmed 89.8%, but the BD confirmed only 75.1%. The positive results and confirmatory tests by sex and specimen type are shown in Table 1. There were no major differences in the observed patterns by sex or specimen type.
In the CDC guidelines, it was pointed out that less sensitive diagnostic tests, such as culture and enzyme immunoassays, should not be used to confirm positive results of the more sensitive NAATs for C. trachomatis. This guideline exists because 30% or more of specimens positive by NAATs will be negative by culture or enzyme immunoassay. Our study shows that the same principle applies when only NAATs are being used. Both APTIMA assays yielded more total and more confirmed-positive results than the BD assay. If reporting a positive result is based upon confirmation of the initial positive result, then using BD to confirm positive results by ACT or AC2 would result in the incorrect reporting of approximately 15% of confirmable positive results as negative (i.e., not confirmed).
In this evaluation, the ACT had the greatest number of positive results (927), and the BD had the least number of positive results (720). While a rule of thumb has been that the less sensitive test is more specific, that was not the case here (assuming that confirmation is a proxy for specificity). Approximately 97% of the 720 positive BD results were confirmed by the more sensitive AC2 and ACT. However, the AC2 had 850 positives, 98% of which were confirmed by ACT. The AC2 confirmed only about 90% of the positive results of the ACT. There are two obvious interpretations for the latter result. It may be that the ACT is less specific and that the AC2 failed to confirm positive results because they were false positives. The alternate explanation is that ACT is actually more sensitive than the other tests and that the inability to confirm its greater number of positive results reflects the lower sensitivity of the confirmatory assays. Further studies are needed to answer this question.
It is clear from our results that some NAATs cannot be used to confirm positive results from other NAATs. Even among NAATs, the assays employed should have equivalent sensitivities. The BD should not be used to confirm the APTIMA assays. If, however, the BD were used for the primary screening, the APTIMA assays could be used for confirmation (96.9% confirmed). Based on our results, we would recommend the AC2 for initial screening and the use of the ACT for confirmation, as 98% of AC2 positive results were confirmed by ACT. This strategy would identify more infected individuals.
ACKNOWLEDGMENTS
The clinical trial was funded by Gen-Probe, Inc.
REFERENCES
Castriciano, S., K. Luinstra, D. Jang, J. Patel, J. Mahony, J. Kapala, and M. Chernesky. 2002. Accuracy of results obtained by performing a second ligase chain reaction assay and PCR analysis on urine samples with positive or near-cutoff results in the LCx test for Chlamydia trachomatis. J. Clin. Microbiol. 40:2632-2634.
Culler, E. E., A. M. Caliendo, and F. S. Nolte. 2003. Reproducibility of positive test results in the BDProbeTec ET system for detection of Chlamydia trachomatis and Neisseria gonorrhoeae. J. Clin. Microbiol. 41:3911-3914.
Gronowski, A. M., S. Copper, D. Baorto, and P. R. Murray. 2000. Reproducibility problems with the Abbott laboratories LCx assay for Chlamydia trachomatis and Neisseria gonorrhoeae. J. Clin. Microbiol. 38:2416-2418.
Johnson, R. E., W. J. Newhall, J. R. Papp, J. S. Knapp, C. M. Black, T. L. Gift, R. Steece, L. E. Markowitz, O. J. Devine, C. M. Walsh, S. Wang, D. C. Gunter, K. L. Irwin, S. DeLisle, and S. M. Berman. 2002. Screening tests to detect Chlamydia trachomatis and Neisseria gonorrhoeae infections—2002. Morb. Mortal. Wkly. Rep. Recomm. Rep. 51(RR-15):1-38;CE1-CE4.
Peterson, E. M., V. Darrow, J. Blanding, S. Aarnaes, and L. M. de la Maza. 1997. Reproducibility problems with the AMPLICOR PCR Chlamydia trachomatis test. J. Clin. Microbiol. 35:957-995.
Schachter, J. 2001. NAATs to diagnose Chlamydia trachomatis genital infection: a promise still unfulfilled. Expert Rev. Mol. Diagn. 1:137-144.
Schachter, J., R. L. Sweet, M. Grossman, D. Landers, M. Robbie, and E. Bishop. 1986. Experience with the routine use of erythromycin for chlamydial infections in pregnancy. N. Engl. J. Med. 314:276-279.
Scholes, D., A. Stergachis, F. E. Heidrich, H. Andrilla, K. K. Holmes, and W. E. Stamm. 1996. Prevention of pelvic inflammatory disease by screening for cervical chlamydial infection. N. Engl. J. Med. 334:1362-1366.
Verkooyen, R. P., G. T. Noordhoek, P. E. Klapper, J. Reid, J. Schirm, G. M. Cleator, M. Ieven, and G. Hoddevik. 2003. Reliability of nucleic acid amplification methods for detection of Chlamydia trachomatis in urine: results of the first international collaborative quality control study among 96 laboratories. J. Clin. Microbiol. 41:3013-3016.(J. Schachter, E. W. Hook,)