Liquid-based pap (L-Pap) test samples are used to detect human papillomavirus,1,2 Chlamydia trachomatis (CT), and Neisseria gonorrhoeae (GC).3,4 Currently, there are 3 L-Pap tests cleared by the US Food and Drug Administration for cytopathology: SurePath (Becton Dickinson, Franklin Lakes, NJ); ThinPrep (Hologic, Bedford, MA); and Cytotek MonoPrep (Monogen, Vernon Hills, IL). We validated the APTIMA Combo 2 (AC2) assay on SurePath L-Pap test samples using a testing protocol from a multicenter trial3 and calculated sensitivity, specificity, and predictive values for L-Pap, cervical swabs (CS), and first-catch urine (FCU). SurePath and ThinPrep collection devices were also compared for adequacy of sampling and cytologic scoring.
As part of the validation, we performed end point dilution and stability experiments for CT and GC as mocked CS and L-Pap samples. CT (serovar: L2–434) was propagated in McCoy cells and GC (ATCC: 43,069) grown in modified New York city agar plates. Ten-fold dilutions were made in APTIMA specimen transport media (STM) and SurePath L-Pap media and samples of both dilution series were transferred into STM for testing: replicates of 3 were tested for 16 weeks. The mocked samples were positive for CT to a dilution of 10−8 in STM and 10−9 in L-Pap media. The GC-mocked specimens were positive in STM to a dilution of 10−7 and in L-Pap media to 10−8. These differences cannot be considered significant because finer dilutions with a large number of replicates were not done. Each showed a gradual decline in relative light unit (RLU) values with dilution and over time, ranging between 10−9 and 10−8 at 1 week to 10−8 and 10−7 at 16 weeks. We also determined the stability of CT in clinical SurePath L-Pap test samples by pooling L-Pap residua within 48 hours of submission for Pap testing. Each of the 5 positive pools was held at room temperature and 3 replicates of each pool were tested on 5 time points from 0 to 10 days. All of the patient pools remained positive for 10 days at RT (Fig. 1). Three of the pools retained >90% and 2 >60% of their original assay RLU values. In a previously published study, human papillomavirus DNA and RNA from ThinPrep and SurePath media stored at room temperature over a 3-week period showed severe degradation of RNA in SurePath media using TRIzol or Qiagen extraction.5 Our ability to preserve CT rRNA in SurePath L-Pap media may have been due to transfer of the specimen into STM before extraction, providing stability to the amplification target.
In the clinical study, a total of 520 women signed consent for collection of an FCU, a CS (collected with an APTIMA unisex swab) and 2 SurePath L-Pap randomized samples (one collected with the SurePath Cervex Brush [cervical broom] and another collected with a Pap Perfect plastic spatula and Cytobrush). Each sampling was placed into separate SurePath tubes. These cytology collection devices are quite different but can be used in the SurePath or ThinPrep systems. The Cervex brush looks like a broom with longer central bristles and was inserted into the cervical os then rotated clockwise 5 times. The other device uses a plastic spatula by scraping the ectocervix before inserting the cytobrush and turning: both the spatula end and cytobrush head were put into the media. The study was performed on the assumption that one might be better than the other. The L-Pap samples were processed for cytopathology using a PrepStain slide processor. The remaining L-Pap material left in the vial was used for CT and GC testing in the AC2 assay within 10 days. This was done by transferring 1 mL to a Gen-Probe APTIMA specimen transfer tube containing 2.9 mL of STM, which was then tested, after the established swab testing procedure in the package insert of the AC2 assay. The CS and FCU samples were also tested using the AC2 assay. A patient was considered infected if more than 1 specimen was positive in an AC2 test or if a single specimen positive by AC2 was confirmed positive by an ACT or AGC assay.6 Statistical analyses were performed using SPSS software for Windows version 11.5. Sensitivities, specificities, positive predictive value (PPV), and negative predictive values (NPV), and their associated 95% confidence intervals were calculated using confidence interval analysis software (version 2.1.2, 2004; T. Bryant, University of Southampton, UK). A P <0.05 was considered statistically significant.
The prevalence of infection was 4.8% (25/520) for CT; 1.0% (5/520) for GC; and 3 (0.6%) of the women had dual infections: 73.1% (19/25) of the CT-positive patients were AC2-positive in all specimens (Table 1). An additional 2 patients had 3 of the samples positive and 3 others had single samples positive, which were confirmed by a positive reaction in an ACT test. Testing the CS diagnosed 96.0% of the CT-positives and recorded 2 positives, which did not confirm in an ACT test (specificity 99.4%) (Table 2). FCU samples yielded positive diagnoses for 88.0% of the CT-infected women with no false positives. The L-Pap samples collected with the SurePath broom yielded 20 (80.0%) positives whereas the L-Pap samples collected with the ThinPrep brush identified 21 of the 25 positives (84%). Both L-Pap samples showed 100% specificity. The PPV and NPV for all of the sample types ranged from 88.8% to 100%. Only 5 women were infected with GC and 3 of them were also infected with CT (data not shown). The CS and L-Pap samples yielded equal numbers of GC infections, but the FCU sample missed 2 of the positive patients. Using the infected patient reference standard of 25 CT-positives, the % sensitivity, specificity, PPV, and NPV for a SurePath L-Pap sample collected with a SurePath broom tested by AC2 were 80.0, 100, 100, and 99.0. The percentages for the SurePath L-Pap samples collected with the ThinPrep brush were 84.0, 99.8, 95.5, and 99.2, respectively. These percentages, in order, are very similar to 85.2, 99.5, 93.2, and 98.7 reported from the SurePath multicentered study,3 which used a reference standard positivity status of a CS measured positive by AC2 and confirmed by ACT or AGC. The values in the current study would approximate even closer the values in the multicentered study if we used a similar less expanded gold standard (without FCU) for comparison. Confirmation of the high predictive values in both studies supports the use of AC2 with SurePath L-Pap specimens processed with the APTIMA specimen transfer kit.
Comparing the 2 collection devices, there were no significant differences for diagnosing CT- or GC-infected women. The ThinPrep brush yielded 1 CT-positive, which was negative in the sample collected with the SurePath broom. Day et al7 examined the impact of additional collection devices compared with the SurePath L-Pap broom on Pap diagnostic utility. They showed little or no enhancement of numbers of patients with abnormal or normal cells. In the current study, all samples showed adequate cells for reading and there were 50 (9.6%) abnormal Pap readings. A total of 17 of the readings differed almost equally between the 2 collection devices. There were no significant differences seen for each cytology category comparing the SurePath broom to the ThinPrep brush (data not shown). There was no correlation of abnormal Pap readings and CT and/or GC infection. Our study corroborates the use of brush or broom L-Pap-samplers for collection into SurePath L-Pap media, which, when processed in AC2 for CT and GC within 10 days of Pap testing, allowed confirmation of diagnostic observations from a previously published multicentered trial.
1. Lörincz AT, Richart RM. Human papillomavirus DNA testing as an adjunct to cytology in cervical screening programs. Arch Pathol Lab Med 2003; 127:959–968.
2. Molijn A, Kleter B, Quint W, et al. Molecular diagnosis of human papillomavirus [HPV] infections. J Clin Virol 2004; 32(suppl 1):S43–S51.
3. Chernesky M, Freund GG, Hook E III, et al. Detection of Chlamydia trachomatis
and Neisseria gonorrhoeae
infections in North Am women by testing SurePath liquid-based Pap specimens in APTIMA assays. J Clin Microbiol 2007; 45:2434–2438.
4. Chernesky M, Jang D, Portillo E, et al. Abilities of APTIMA, AMPLICOR, and Probe-Tec assays to detect Chlamydia trachomatis
and Neisseria gonorrhoeae
in PreservCyt ThinPrep liquid-based Pap samples J Clin Microbiol 2007; 45:2355–2358.
5. Powell N, Smith K, Fiander A. Recovery of human papillomavirus nucleic acids from liquid-based cytology media. J Virol Methods 2006; 137:58–62.
6. Boyadzhyan B, Yashina T, Yatable JH, et al. Comparison of the APTIMA CT and GC assays with the APTIMA Combo 2 assay, the Abbott LCx
assay, and direct fluorescent-antibody and culture assays for detection of Chlamydia trachomatis
and Neisseria gonorrhoeae
. J Clin Microbiol 2004; 42:3089–3093.
© Copyright 2009 American Sexually Transmitted Diseases Association
7. Day SJ, O’Shaughnessy DL, O’Connor JC, et al. Additional collection devices used in conjunction with the SurePath liquid-based Pap test broom device do not enhance diagnostic utility. BMC Women’s Health 2004; 4:6.