Comparison of Siriraj liquid-based solution and standard transport media for the detection of high-risk human papillomavirus in cervical specimens

1 INTRODUCTION

Cervical cancer is the second most common cancer and the leading cause of cancer-related deaths among Thai women.1 Most of the studies showed that human papillomavirus (HPV) related cancer was approximately 89% (81.4%-93.5%).2-7 However, the study in Thailand found 94.7% prevalence of HPV related cervical cancer.8

According to the recent classification by the International Agency for Research on Cancer, 12 HPV genotypes including type 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, and 59 were classified in group one (carcinogenic to humans).9 HPV types 16 and 18 are responsible for 70% of all cervical cancer cases worldwide.10, 11 Likewise, in Thailand, 74% of cervical cancer derived from these two HPV types.8 Most of the HPV infections can resolve spontaneously and do not cause symptoms. However, persistent infection with high-risk HPV types may lead to precancerous lesions and if left untreated, some will eventually progress to invasive cervical cancer within 10 to 20 years.11, 12

There is a causal relationship between persistent high-risk HPV infection, cervical cancer precursors, and invasive cervical cancer. Current guidelines in the United States recommend HPV DNA testing as an adjunct to cytology for cervical cancer screening in women aged ≥30 years, so-called cotesting. Moreover, HPV DNA testing is a preferable option for triaging women with atypical squamous cells of undetermined significance (ASCUS) and an acceptable option for management of low-grade squamous intraepithelial lesions (LSIL) in young women aged 21 to 24 years and also in postmenopause.13-15 Recently, a pooled analysis of four European trials demonstrated the effectiveness of HPV based screening in preventing cervical carcinoma.16 The Indian trial also reported the efficacy in reducing the numbers of advanced cervical cancers and mortality with a single round HPV DNA testing.17

Because HPV DNA testing is still costly in many developing countries, cervical cancer screening in Thailand mostly depends on cytology alone. We have developed a new preservative solution for liquid-based cytology using modified Saccomanno technique of cell preparation, called Siriraj liquid-based solution.18 Our previous studies have demonstrated that the screening performance of Siriraj liquid-based cytology was superior to that of conventional cytology and cost 2 to 3 times less than commercially available liquid-based cytology. Subsequent study has shown that the detection rate was stable with increased use.19-21

Siriraj liquid-based solution, an ethanol-based preservative, was designed to fix not only cellular but also subcellular components, including DNA, which are targeted in HPV DNA testing.22-24 According to our previous studies, Siriraj liquid-based solution efficiently preserved cells for cytology study.19-21 However, there have been no studies to date investigating the performance of Siriraj liquid-based solution in preserving cervical samples for HPV DNA testing. This study was aimed to compare the detection of high-risk HPV types in cervical samples between using Siriraj liquid-based solution and the commercially available standard transport media.

2 MATERIALS AND METHODS

3 RESULTS

Two hundred and seventeen women were enrolled in this study. One sample was lost owing to inappropriate packaging; therefore, 216 patients were included in the analysis. Participants’ mean age was 46.7 ± 13.1 years. Table 1 summarizes the association between cervical cytology and the percentage positive for high-risk HPV, for each transport medium. Eighty-four women had normal cytology, while 114 women had cytological abnormalities. Eighteen participants did not have cervical cytology results due to the presence of a gross cervical lesion, and instead, cervical biopsies were performed and revealed the cervical carcinoma. The percentage of positive high-risk HPV preserving with standard transport media and Siriraj liquid-based solution were similar in most cytological classifications (normal 3.6% vs 4.8%, ASCUS 93.9% vs 90.4%, atypical squamous cells cannot exclude HSIL 73.7% vs 73.7%, LSIL 80% vs 77.5%, HSIL 95.2% vs 90.5%, and adenocarcinoma 66.7% vs 66.7%) except for atypical glandular cell (50% vs 25%) and squamous cell carcinoma (100% vs 75%) which might be attributed to the small sample size in these subgroups.

Table 2 reveals almost perfect agreement between the use of standard transport media and Siriraj liquid-based solution for HPV DNA testing, with a kappa value of 0.935 (95% CI, 0.893-0.976; P < 0.001). There were nine discorrelated cases out of 216 cases. The details of this group are shown in Table 3. There was no statistical difference between the median storage time of the correlated and discorrelated group, which was 13 (0-78) days and 12 (0-22) days, respectively (P = 0.49).

Cervical histology was considered the gold standard for the diagnosis of CIN2+. Among 216 women, 102 had the final cervical histological diagnoses from the specimens obtained during cervical biopsy, loop electrosurgical excision procedure, or hysterectomy. The performance of HPV DNA testing in detecting CIN2+ lesions on pathology between standard transport media and Siriraj liquid-based solution was similar with the sensitivity, specificity, positive predictive value, negative predictive value and accuracy of 92.9% versus 91.2%, 31.1% versus 33.3%, 63.1% versus 63.4%, 77.8% versus 75%, and 65.7% versus 65.7%, respectively (Table 4). The results of comparing between standard transport media and Siriraj liquid-based solution for sensitivity, specificity, positive predictive value, negative predictive value, and accuracy were 95.4%, 98.1%, 98.1%, 95.5%, and 96.8%.

The relative sensitivity of Siriraj liquid-based solution and standard transport media in patients with CIN2+ is 98% (50/51; Table 5). The relative specificity of Siriraj liquid-based solution and standard transport media in patients with non-CIN2+ is 98.1% (102/104; Table 6).

4 DISCUSSION

Abbott RealTime HR HPV assay was used for detection of high-risk HPV infection in our study. This test is currently used worldwide despite the lack of US FDA approval. Studies from the Netherlands and Slovenia have revealed the sensitivity of this RealTime assay in detection of CIN3 and cervical cancer to be 97% and 98%, respectively, which are comparable with other clinically validated HPV DNA assays.26-28 The analytical specificity studies have shown no cross-reactivity with any organisms, including all relevant low-risk HPV genotypes.25 Furthermore, intralaboratory reproducibility and interlaboratory agreement of this assay has been confirmed by several studies.29-31 According to the Abbott RealTime HR HPV assay manufacturer's instructions, authorized cervical specimens should be collected using ThinPrep PreservCyt Solution, SurePath Preservative Fluid, or the Abbott Cervi-Collect Specimen Collection Kit.

Siriraj liquid-based solution, an ethanol-based preservative, was designed to fix not only cellular but also subcellular components, including DNA, which is the targets for HPV DNA testing.22-24 Our study demonstrated almost perfect agreement between standard transport media (Abbott Cervi-Collect Specimen Collection Kit) and the Siriraj liquid-based solution for use in HPV DNA testing. Characteristics of the nine discordant samples are given in Table 3 and no discorrelation in detecting HPV16, which accounts for 50% of cervical cancer,32 was found. When we considered only positive/negative result of high-risk HPV infection, the positive agreement, negative agreement, and overall agreement were 93.7% (104/111), 93.8% (105/112), and 96.8% (209/216) respectively.

According to the validation guidelines for candidate HPV assays by Meijer et al33, the sensitivity of the candidate test for CIN2+ should be at least 90%. The specificity of the candidate test for non CIN2+ should be at least 98%. Our result met the criteria by having the relative sensitivity of Siriraj liquid-based solution and standard transport media in patients with CIN2+ of 98% (50/51; Table 5). The relative specificity of Siriraj liquid-based solution and standard transport media in patients with non-CIN2+ has also met the criteria at 98.1% (102/104; Table 6). Nevertheless, as for Meijer validation protocol, our sample size for assessing a relative specificity was not large enough to ensure adequate power.

Our study was aimed to compare the detection of high-risk HPV types in cervical samples between using Siriraj liquid-based solution and Abbott RealTime HR HPV assay, therefore, cervical biopsies were not be obtained from all participants. Nevertheless, among 216 women, 102 had final cervical histological diagnoses. Test performance including sensitivity, specificity, positive predictive value, negative predictive value, and accuracy, was similar between the two transport media. The previous study, evaluating the clinical performance of Abbott RealTime High-Risk HPV test for detection of CIN2+ in women with abnormal cytology, revealed clinical sensitivity, specificity, positive predictive value and negative predictive value were 97.8%, 32.8%, 41.3%, and 96.9% respectively.34 However, we admitted that our Siriraj liquid-based solution is not totally perfect which we have one false negative with CIN2+ from cervical histology. Another reason for false negative might be sampling error during the procedure. Zazove, et al35 addressed HPV falsely negative usually due to (a) the level of a cut point for HPV detection or (b) the sampling error.

There are two factors that might affect our results. It was unclear whether HPV DNA detection rates were dependent on the order of cervical specimen collection. de Sanjose et al36 has shown that detection rates were unchanged when the order of sample collection was compared. For ethical reasons, our study assigned standard transport media for the first sample and Siriraj liquid-based solution for the second one. Secondly, with respect to storage time, we compared the median storage time of the correlated (13 days) and discorrelated groups (12 days), which revealed no statistically significant differences. According to the Abbott RealTime HR HPV assay manufacturer's instructions; cervical specimens may be stored up to 90 days. However, there have been no data as yet as for the storage time of cervical specimens preserved using Siriraj liquid-based solution for HPV DNA testing.

This is the first study demonstrating a good performance of Siriraj liquid-based solution in preserving cervical specimens for HPV DNA testing with high sensitivity and specificity compared to standard transport media. In our institute, Siriraj liquid-based solution and Siriraj liquid-based cytology have replaced the conventional cytology since 2006 because of the lower cost and better performance compared with conventional cytology.19-21 Siriraj liquid-based cytology is an accessible cervical cancer screening method for women of low socioeconomic status. Use of Siriraj liquid-based solution has also increased outside our institute for the cervical cytology screening. Nowadays, cervical cancer screening is in the era of HPV DNA testing. Nevertheless, most women in Thailand are not able to afford the costly commercially available tests. Our study has shown that the use of Siriraj liquid-based solution for HPV DNA testing is comparable to that of standard transport media. Siriraj liquid-based solution may thus be considered an alternative medium for HPV DNA testing, with a cost of about 2 to 3 times lower than commercially available standard media. Furthermore, reflex HPV DNA testing can be performed on residual cervical samples stored in Siriraj liquid-based solution.

5 CONCLUSION

Siriraj liquid-based solution showed almost perfect agreement with the standard transport media for HPV DNA testing. This medium can be considered as an alternative option for use with the Abbott RealTime HR HPV assay.

CONFLICTS OF INTEREST

The authors declare no conflicts of interest.