PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of jcmPermissionsJournals.ASM.orgJournalJCM ArticleJournal InfoAuthorsReviewers
 
J Clin Microbiol. 2010 September; 48(9): 3068–3072.
Published online 2010 July 7. doi:  10.1128/JCM.00736-10
PMCID: PMC2937663

Evaluation of Transported Dry and Wet Cervical Exfoliated Samples for Detection of Human Papillomavirus Infection[down-pointing small open triangle]

Abstract

We determined the feasibility of human papillomavirus (HPV) detection in cervical exfoliated cells collected as dry swab samples. Both dry cervical swab and specimen transport medium (STM) cervical swab samples were collected from 135 patients attending either colposcopy or women's clinics in Guayaquil, Ecuador, who had a cytology diagnosis within 6 months. HPV was detected by dot blot hybridization and genotyped by the liquid bead microarray assay (LBMA). Overall, 23.1% of dry samples were positive for any high-risk HPV types, and 24.6% of STM samples were positive for any high-risk HPV types. Of 125 paired samples, the type-specific high-risk HPV proportion positive agreement was 60.7% (kappa, 0.69; 95% confidence interval [CI], 0.53 to 0.82). Of six women with cytological evidence of invasive cervical cancer, high-risk HPV DNA was detected in three of their STM samples and in five of their dry samples. Dry samples were more likely to be insufficient for HPV testing than STM samples. Consistent with this observation, the amount of genomic DNA quantitated with the β-actin gene was almost 20 times lower in dry samples than in STM samples when detected by the real-time TaqMan assay; however, HPV DNA viral loads in dry samples were only 1.6 times lower than those in matched STM samples. We concluded that exfoliated cervical cells could be collected as dry swab samples for HPV detection.

Human papillomavirus (HPV) infection causes warts and various cancerous and precancerous lesions in men and women. It has been established that high-risk HPV infection is the etiological agent of cervical cancer, which affects almost half a million women worldwide and has a 50% mortality rate (12, 18). In developed countries, cervical cancer control relies on routine cytology screening to detect and treat cervical cancer precursor lesions, and HPV detection has been used for the management of women with equivocal cytology results (1, 4, 14). In resource-poor settings where cytology-based screening is difficult to implement, HPV detection has been proposed as the alternative primary screening test for cervical cancer (4, 17).

Currently, HPV detection is performed on cervical samples collected in liquid medium by a trained clinician, which is often impractical or unavailable in remote areas and developing countries. Several approaches have been proposed to simplify the sample collection process for HPV detection. For example, HPV detection in self-collected vaginal swab samples has been proposed as a method that would eliminate the need for clinical visits, and we have shown that HPV detected in self-collected vaginal swab samples stored in specimen transport medium (STM) had sensitivity and specificity similar to those of clinician-collected cervical swab samples stored in STM for detecting cervical neoplasia (16). Currently, cervical exfoliated cells are collected either in phosphate-buffered saline (PBS), which is inexpensive but requires constant refrigeration, or in various liquid media, such as STM or Preservcyt, which preserve HPV DNA at room temperature but are expensive. Additionally, Preservcyt is flammable. Thus, the elimination of the requirement of refrigeration or special media would facilitate the collection, storage, and transport of cervical exfoliated cells in remote areas or resource-poor settings and ultimately reduce cost.

A few studies have investigated the feasibility of dry sample collection for HPV detection. For example, dry swab samples could be collected and stored at 4°C before being processed for HPV detection (9, 13). Cervical smear samples could also be collected on filter paper for HPV detection (2, 7, 8). However, no studies have investigated the feasibility of collecting, storing, and shipping dry swab samples at room temperature for HPV detection or the effect of dry sample collection on DNA degradation.

In this study, we first determined whether cells of cervical cancer cell lines could be stored at room temperature without significant DNA degradation. We then collected both dry cervical swab and STM cervical swab samples from 135 patients from Guayaquil, Ecuador. HPV detection and genotyping were determined by dot blot hybridization and the liquid bead microarray assay (LBMA), respectively (6). In addition, HPV DNA and the housekeeping gene β-actin were quantitated by using real-time TaqMan assays and 42 paired samples, 32 of which were HPV positive.

MATERIALS AND METHODS

Cervical cancer cell lines.

Both Siha and Caski cell lines were obtained from the ATCC (Manassas, VA) and grown in RPMI 1640 medium supplemented with 10% fetal bovine serum. Cells were detached from culture plates by using trypsin digestion and were counted with a hemocytometer. About 5 × 106 cells were centrifuged, and the resulting cell pellet was collected with a sterile Dacron swab and put into either a dry sterile 4.5-ml Nunc cryotube or a 4.5-ml Nunc cryotube containing 1 ml STM (Qiagen). Both samples were kept at room temperature for 1 to 4 weeks before storing at −20°C and processed later.

Clinical sample collection.

Patients were enrolled from either the colposcopy clinic or the women's clinic at the hospital Teodoro Madonado Carbo, Guayaquil, Ecuador, between June 2008 and April 2009. To collect exfoliated cervical samples, Dacron swabs were inserted into the endocervix by a clinician and rotated in both directions five times. Two samples were collected from each patient in a random order: one collected into a 4.5-ml Nunc cryotube containing 1 ml STM and the other sample collected into a dry sterile 4.5-ml Nunc cryotube. Samples were stored at 4°C for up to 6 months before they were shipped at room temperature to the HPV research laboratory at the University of Washington for HPV detection and genotyping.

Genomic DNA isolation from cell lines and cervical swab samples.

The STM samples, including the swabs, were digested with 20 μg/ml proteinase K at 37°C for 1 h. About 800 to 900 μl of the digested sample was collected, and genomic DNA was isolated from 200 μl of the digested samples by using a QIAamp DNA blood minicolumn according to the manufacturer's protocol (Qiagen, Valencia, CA) and resuspended in 100 μl Tris-EDTA (TE). The cell pellet and dry swab samples were first rehydrated in 1 ml STM at room temperature and then immediately digested with proteinase K as STM samples.

HPV detection and genotyping by LBMA.

The presence of HPV DNA was determined by PCR amplification followed by dot blot hybridization (15). Briefly, HPV and β-globin were amplified from 4 μl of isolated genomic DNA using L1 consensus primers MY09/MY11/HMB01 and β-globin primers PC04/GH20 (10). Ten microliters of PCR products was then dotted onto nylon filters and probed with biotin-labeled β-globin and generic HPV probes (3). Samples were considered insufficient for HPV detection if they were negative for β-globin and HPV, and HPV-positive samples were genotyped by using the LBMA (6). Briefly, HPV L1 fragments were amplified by using primers MY09/MY11/HMB01, purified with a QIAquick PCR purification column (Qiagen, Valencia, CA). Biotin-labeled single-stranded HPV PCR products were generated from the PCR-amplified HPV L1 fragment using biotin-labeled primer MY11. Subsequently, biotin-labeled PCR products were genotyped by using the Luminex assay, which distinguishes 37 HPV types. A sample was considered to be positive or weakly positive for a specific HPV type if the relative light units (RLU) were greater than 10 or 7 times the background RLU, respectively, for that specific HPV type.

HPV detection by real-time PCR.

Type-specific TaqMan assays were designed for HPV-6, -16, -18, -39, -52, -53, and -66 based on the HPV E7 gene. Primers and probe sequences are listed in Table Table1.1. Quantification of β-actin was performed by using TaqMan β-actin detection reagents (ABI, Foster City, CA). The real-time TaqMan PCR assays were performed with the ABI Prism 7900 sequence detection system. Relative quantifications (RQs) of both HPV and β-actin were determined by using the 2−ΔCT method. For both HPV and β-actin, the RQ of the STM sample was set at 1.

TABLE 1.
Primers and probes for HPV detection by real-time TaqMan PCR assays

Statistical analysis.

The RQs of both HPV and β-actin were determined by real-time PCR assays and calculated by using the 2−ΔCT method. In a real-time PCR assay, the CT (cycle threshold) is defined as the number of cycles required for the fluorescent signal to cross the background threshold. Therefore, CT levels are inversely proportional to the amount of template DNA in the sample. Because the CT is measured during the PCR exponential amplification phase, one CT cycle difference translates into a 2-fold difference in the original DNA concentration. In the current study, for both β-actin and HPV, the RQ of the STM sample was set at 1. For example, in analyzing the β-actin quantity, if the CT for the dry sample is 21 while the CT for the paired STM sample is 20, the RQ for β-actin in the dry sample would be 0.5.

In this study, high-risk HPV types are defined as HPV types 16, 18, 26, 31, 33, 35, 39, 45, 51, 52, 53, 56, 58, 59, 66, 68, 73, 82, and is39 (11). The proportion positive agreement (PPA) between paired STM and dry swab samples was calculated by dividing the number of samples testing positive for HPV for both the STM and dry swab samples by the number of samples testing positive for either STM or dry swab samples. Conversely, the proportion negative agreement (PNA) was calculated by dividing the number of samples testing negative for HPV for both the STM and dry swab samples by the number of samples testing negative for either sample. To determine the PPA (or PNA) beyond that expected by chance, a modified unweighted kappa statistic (referred to as “kappa+” or “kappa”) was calculated by dividing the difference between the observed positive (or negative) agreement and the expected positive (or negative) agreement by 1 minus the expected positive (or negative) agreement. Expected positive (or negative) agreement was computed under the assumption of independence of the assay results. To account for correlation within subjects due to the assessment of multiple potential HPV types per individual, 95% confidence intervals (CIs) were computed by using percentile bootstrap methods with 1,000 repetitions.

Logistic regression was used to evaluate whether the cytological diagnosis (categorized as negative or “ASC-US+” [including ASC-US {atypical squamous cells of undetermined significance}, LSIL {low-grade squamous intraepithelial lesion}, HSIL {high-grade squamous intraepithelial lesion}, or cancer]) was associated with the likelihood of high-risk type-specific positive agreement (defined as testing positive for a specific high-risk HPV type in both samples) between STM and dry swab samples. Robust variance estimates were used to account for correlation within subjects. Paired t tests were used to compare viral loads of β-actin and HPV in STM versus dry samples.

RESULTS

Stability of genomic and HPV DNA in dry cervical cancer cell pellets.

We quantitated β-actin and HPV DNA in dry cervical cancer cell pellets stored at room temperature for different amounts of time. Both Siha and Caski cancer cell lines were used since Siha cells contain low copy numbers (1 to 2 copies) of integrated HPV-16 genomes per cell, and Caski cells contain high copy numbers (~600 copies) of integrated HPV-16 genomes per cell. The cell pellets were kept at room temperature either as dry cell pellets or as STM samples for 1 to 4 weeks before being stored at −20°C. Since STM is routinely used to preserve HPV DNA at room temperature, we calculated the RQ for the dry cell pellet sample by setting the matched STM sample at 1. The RQ of β-actin for dry cell pellet samples varied from 0.75 to 3.25, while the RQ of HPV-16 for dry cell pellet samples varied from 0.69 to 5.31 (Table (Table22).

TABLE 2.
HPV-16 and β-actin DNA quantitation in cervical cancer cell lines

Patients and clinical data.

A total of 135 pairs of samples were collected from 73 patients enrolled from the colposcopy clinic and 62 patients enrolled from the women's clinic in Guayaquil, Ecuador. Of 73 patients who attended the colposcopy clinic, 2 had insufficient cytology, 2 had normal cytology, 17 had ASC-US, 43 had LSIL, 3 had HSIL, and 6 had cancer. All 62 patients who attended the regular clinic had normal cytology. The mean age of patients was 38.2 years (standard deviation [SD], 11.5 years), with patients from the colposcopy clinic (mean age, 44.4 years [SD, 11.4 years]) being older than patients from the regular clinic (mean age, 31.2 years [SD, 6.7 years]) (P < 0.001).

Overall HPV concordance between dry swab samples and STM swab samples.

HPV detection and genotyping were determined for 135 paired (dry versus STM) cervical swab samples. Only one (0.7%) STM sample was insufficient for HPV detection, while nine (6.7%) dry swab samples were insufficient for HPV detection (P < 0.01). Of 134 sufficient STM samples, 31 (23.1%) samples were positive for any high-risk HPV, and 48 type-specific high-risk HPV infections were detected. Of 126 sufficient dry samples, 31 (24.6%) were positive for any high-risk HPV, and 43 type-specific high-risk HPV infections were detected. Among the 125 dually sufficient paired samples, 25 were high-risk HPV positive for both STM and dry samples, 6 were high-risk HPV positive for the STM sample but not the dry sample, 5 were high-risk HPV positive for the dry but not the STM sample, and 89 were negative for both samples (Table (Table3).3). Hence, the proportion positive agreement for any high-risk HPV between dry and STM samples was 25 out of 36 samples (69.4%) (kappa+ = 0.61; 95% CI, 0.44 to 0.79). Pooling across the 19 high-risk HPV types (2,375 total type-specific pairs), there were 34 occurrences in which both the STM and dry samples were positive for a given type, 14 occurrences in which the STM sample was positive for a given type but the dry sample was negative for that type, 8 occurrences in which the dry sample was positive for a given type but the STM sample was negative for that specific type, and 2,319 occurrences when both the STM and dry samples were negative for a given type (Table (Table3).3). Hence, the type-specific proportion positive agreement for all high-risk HPV types between dry and STM samples was 34 out of 56, or 60.7% (kappa+ = 0.69; 95% CI, 0.53 to 0.82). The proportion negative agreement for any high-risk HPV between dry and STM samples was 89% (kappa = 0.62; 95% CI, 0.42 to 0.79), and the type-specific proportion negative agreement for high-risk HPV types between dry and STM samples was 99.1% (kappa = 0.60; 95% CI, 0.47 to 0.72).

TABLE 3.
HPV detection agreement of STM and dry samples stratified by cytology diagnosis

HPV concordance in dry and STM samples by cytology.

Next, we determined HPV concordance between dry and STM samples stratified by cytology diagnosis. Of subjects with normal cytology, the type-specific proportion positive agreement for high-risk HPV between dry and STM samples was 45.0% (kappa+ = 0.45; 95% CI, 0.23 to 0.69). Of subjects with an abnormal cytology diagnosis (ASC-US+) (including ASC-US, LSIL, HSIL, and cancer), the type-specific proportion positive agreement for high-risk HPV was 69.4% (kappa+ = 0.69; 95% CI, 0.54 to 0.83). Therefore, a cytological diagnosis of ASC-US+ was associated with a borderline statistically significant increased likelihood of type-specific positive agreement for high-risk HPV (odds ratio [OR] = 2.78; 95% CI, 0.93 to 8.29). Similarly, of subjects with normal cytology, the type-specific proportion negative agreement for high-risk HPV between dry and STM samples was 99% (kappa = 0.44; 95% CI, 0.22 to 0.70). Of subjects with an abnormal cytology diagnosis (ASC-US+) (including ASC-US, LSIL, HSIL, and cancer), the type-specific proportion negative agreement for high-risk HPV was 99.1% (kappa = 0.69; 95% CI, 0.55 to 0.82).

Of nine patients with a cytological diagnosis of HSIL or invasive cervical cancer, three patients had an STM sample that was negative for high-risk HPV and one patient had an STM sample that was insufficient for HPV DNA testing, while only two of these nine patients had a dry sample that was negative for high-risk HPV DNA, and none were insufficient (Table (Table44).

TABLE 4.
HPV detected in nine patients with HSIL or cancer diagnosis

Quantitation of β-actin and HPV DNA in dry swab and STM swab samples.

Because dry swab samples were more likely to be insufficient for HPV DNA detection than STM swab samples, we wanted to determine whether DNA was preferentially degraded in dry swab samples. Genomic DNA was quantitated by a β-actin real-time TaqMan assay, and HPV DNA was quantitated by type-specific real-time TaqMan assays targeting the HPV E7 region on pairs of samples that were either positive for at least one of the following HPV types (type 6, 16, 18, 52, 53, or 66) or insufficient for HPV detection in one sample. β-Actin quantification was performed on 42 paired samples, and HPV quantification was performed on 32 paired HPV type-specific infections.

Four samples (two dry and two STM samples) were insufficient for β-actin quantitation by real-time PCR. For the remaining 38 samples, the average amount of β-actin quantitated by RQ in dry samples was 19.5 times lower than that in matched STM samples (P < 0.001). For five pairs, HPV DNA was undetected in the dry sample (two pairs), the STM sample (one pair), or both samples (two pairs). For the remaining 27 samples, the average HPV viral load in dry samples was 1.6 times lower than that in matched STM samples (P = 0.013) (Fig. (Fig.1).1). The difference in viral loads measured in dry samples compared to those in STM samples was greater for β-actin than for HPV (P < 0.001).

FIG. 1.
Quantitation of β-actin and HPV in STM and dry samples. Quantitation of β-actin and HPV viral load were determined with paired STM and dry samples by quantitative TaqMan PCR assays. The relative quantification (RQ) was determined by using ...

DISCUSSION

In this study, we determined whether HPV could be detected in dry swab samples with sensitivities similar to those for STM samples. Using cell line cell pellets, we showed that there was no significant DNA degradation (both β-actin and HPV) in dry cell pellet samples stored at room temperature for up to 1 month regardless of HPV copy numbers. Our data on clinical samples showed that dry swab samples were more likely to be insufficient for HPV detection than the STM swab samples. The presence and number of HPV types detected in STM samples were somewhat greater than in paired dry swab samples, although these differences did not reach statistical significance. The type-specific high-risk HPV positive agreement was 60.7% between dry swab samples and STM samples (kappa = 0.69; 95% CI, 0.53 to 0.82). Importantly, positive agreement for the detection of any high-risk HPV type was better (borderline statistical significance) for samples from women with ASC-US+ (kappa+ = 0.74; 95% CI, 0.52 to 0.94) than for samples from those with negative cytological findings (kappa+ = 0.46; 95% CI, 0.16 to 0.74). Using real-time TaqMan assays, we reported that the degree of genomic DNA degradation in dry versus STM samples was greater than the degree of HPV DNA degradation in dry versus STM samples. Consistent with these observations, we showed that the sensitivities of high-risk HPV DNA detection appeared to be similar for dry and wet swab samples when the patient had a clinically significant cervical lesion (HSIL and cancer) (Table (Table44).

While a few studies have reported using dry sample collection for HPV detection (including one large national survey of women in the United States) (5, 9, 13), only Shah et al. compared dry swab samples directly to wet swab samples. In that study, three types of clinician-collected samples were compared for HPV detection: vaginal dry swabs, vaginal swabs collected in STM, and cervical swabs collected in STM (13). HPV was detected and genotyped by dot blot hybridization. The positive agreement of detection of any HPV in vaginal dry swab and vaginal STM swab samples was 68.9% (62/90). The frequency and types of HPV detected in vaginal dry swabs were similar to those found in vaginal STM swab samples.

Several studies investigated the collection of dry cervical exfoliated cell samples using filter papers. With this method, swab samples can be smeared directly onto sterile filter paper, which is then easily air dried and stored at room temperature for up to 1 year. A small piece of the paper smear can then be punched out, boiled in water, and used directly for PCR amplification. Using this method, Kailash et al. showed previously that HPV-16 was correctly identified in 50 cervical paper scrape samples (8). Similarly, using chemically treated Flinders Technology Associates (FTA) elute microcards, Gustavsson et al. showed previously that the agreement in HPV positivity for 50 cervical samples detected by a real-time PCR assay between cytobrush and FTA samples was 94% (kappa = 0.88; 95% CI, 0.75 to 1) (7). On the other hand, Banura et al. compared the detection of 25 HPV types in paper and PBS samples from 111 young women. HPV was genotyped by using the SPF10 line probe assay (LiPA). The prevalence of any HPV types was 82.9% for PBS samples compared to 32.4% for paper samples. In addition, fewer HPV types and fewer multiple HPV infections were detected in paper samples (2). In summary, these studies further support the feasibility of dry sample collection.

To estimate the extent of DNA degradation in dry swab samples, we quantitated β-actin as well as HPV DNA in paired samples using real-time TaqMan assays. Unlike what we observed for the cancer cell line experiment, where no DNA degradation was observed for both β-actin and HPV DNA in dry cell pellet samples, in clinical samples, both β-actin and HPV were significantly degraded in dry swab samples, although genomic DNA degradation was greater than HPV DNA degradation. We hypothesize that the degradation of genomic DNA is likely due to the presence of additional nuclease activity in dry clinical samples and that HPV DNA present predominantly as episomal copies in these samples was less affected. It would be interesting to determine whether integrated HPV DNA, such as that present in cervical cancer samples, would be degraded to an extent similar to that of genomic DNA in dry swab samples.

In our study, since wet swab samples were stored in STM, we used STM to rehydrate the dry swab samples to simplify the comparison. Because dry swab samples are immediately digested during rehydration, it is unlikely to introduce significant degradation by rehydrating dry swab samples in PBS instead of STM. Although our dry swab samples were stored refrigerated before shipment, we do not see any problems storing them at room temperature, since they were transported from clinics (Guayaquil, Ecuador) to the laboratory (Seattle, WA) at ambient temperature. Future studies should investigate the effect of rehydration in PBS and storage at room temperature to reduce the cost of sample collection.

In summary, our study provides evidence that HPV detection can be performed by using dry swab samples stored at room temperature although with an increase in the number of insufficient samples and with marginally less sensitivity for HPV detection. Despite the finding that genomic DNA was significantly degraded in dry swab samples versus wet samples, very little effect was observed on HPV DNA if it was present as episomal copies. Future studies should focus on identifying methods to inactivate nuclease activities in clinical samples and preserve genomic DNA in dry swab samples.

Acknowledgments

Informed consent was obtained according to procedures approved by the Human Subjects Committee of the Hospital Dr. Teodoro Maldonado Carbo.

We have no commercial or other associations that might pose a conflict of interest.

Footnotes

[down-pointing small open triangle]Published ahead of print on 7 July 2010.

REFERENCES

1. ASCUS-LSIL Triage Study Group. 2003. Results of a randomized trial on the management of cytology interpretations of atypical squamous cells of undetermined significance. Am. J. Obstet. Gynecol. 188:1383-1392. [PubMed]
2. Banura, C., S. Franceschi, L. J. van Doorn, F. Wabwire-Mangen, E. K. Mbidde, and E. Weiderpass. 2008. Detection of cervical human papillomavirus infection in filter paper samples: a comparative study. J. Med. Microbiol. 57:253-255. [PubMed]
3. Bauer, H. M., Y. Ting, C. E. Greer, J. C. Chambers, C. J. Tashiro, J. Chimera, A. Reingold, and M. M. Manos. 1991. Genital human papillomavirus infection in female university students as determined by a PCR-based method. JAMA 265:472-477. [PubMed]
4. Cox, J. T. 2009. History of the use of HPV testing in cervical screening and in the management of abnormal cervical screening results. J. Clin. Virol. 45(Suppl. 1):S3-S12. [PubMed]
5. Dunne, E. F., E. R. Unger, M. Sternberg, G. McQuillan, D. C. Swan, S. S. Patel, and L. E. Markowitz. 2007. Prevalence of HPV infection among females in the United States. JAMA 297:813-819. [PubMed]
6. Feng, Q., S. Cherne, R. L. Winer, A. Balasubramanian, S. K. Lee, S. E. Hawes, N. B. Kiviat, and L. A. Koutsky. 2009. Development and evaluation of a liquid bead microarray assay for genotyping genital human papillomaviruses. J. Clin. Microbiol. 47:547-553. [PMC free article] [PubMed]
7. Gustavsson, I., M. Lindell, E. Wilander, A. Strand, and U. Gyllensten. 2009. Use of FTA card for dry collection, transportation and storage of cervical cell specimen to detect high-risk HPV. J. Clin. Virol. 46:112-116. [PubMed]
8. Kailash, U., S. Hedau, V. Gopalkrishna, S. Katiyar, and B. C. Das. 2002. A simple ‘paper smear’ method for dry collection, transport and storage of cervical cytological specimens for rapid screening of HPV infection by PCR. J. Med. Microbiol. 51:606-610. [PubMed]
9. Krech, T., S. Castriciano, D. Jang, M. Smieja, G. Enders, and M. Chernesky. 2009. Detection of high risk HPV and Chlamydia trachomatis in vaginal and cervical samples collected with flocked nylon and wrapped rayon dual swabs transported in dry tubes. J. Virol. Methods 162:291-293. [PubMed]
10. Manos, M. M., Y. Ting, D. K. Wright, A. J. Lewis, T. R. Broker, and S. M. Wolinsky. 1989. The use of polymerase chain reaction amplification for the detection of genital human papillomavirus, p. 209-214. In M. Furth and M. Greaves (ed.), Molecular diagnostics of human cancer, vol. 7. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
11. Munoz, N., X. Castellsague, A. B. de Gonzalez, and L. Gissmann. 2006. Chapter 1: HPV in the etiology of human cancer. Vaccine 24(S3):S1-S10. [PubMed]
12. Parkin, D. M. 2006. The global health burden of infection-associated cancers in the year 2002. Int. J. Cancer 118:3030-3044. [PubMed]
13. Shah, K. V., R. W. Daniel, M. K. Tennant, N. Shah, K. T. McKee, Jr., C. A. Gaydos, J. C. Gaydos, and A. Rompalo. 2001. Diagnosis of human papillomavirus infection by dry vaginal swabs in military women. Sex. Transm. Infect. 77:260-264. [PMC free article] [PubMed]
14. Solomon, D., M. Schiffman, and R. Tarone. 2001. Comparison of three management strategies for patients with atypical squamous cells of undetermined significance: baseline results from a randomized trial. J. Natl. Cancer Inst. 93:293-299. [PubMed]
15. Weaver, B. A., Q. Feng, K. K. Holmes, N. Kiviat, S. K. Lee, C. Meyer, M. Stern, and L. A. Koutsky. 2004. Evaluation of genital sites and sampling techniques for detection of human papillomavirus DNA in men. J. Infect. Dis. 189:677-685. [PubMed]
16. Winer, R. L., Q. Feng, J. P. Hughes, M. Yu, N. B. Kiviat, S. O'Reilly, and L. A. Koutsky. 2007. Concordance of self-collected and clinician-collected swab samples for detecting human papillomavirus DNA in women 18 to 32 years of age. Sex. Transm. Dis. 34:371-377. [PubMed]
17. Wright, T. C., Jr., M. Schiffman, D. Solomon, J. T. Cox, F. Garcia, S. Goldie, K. Hatch, K. L. Noller, N. Roach, C. Runowicz, and D. Saslow. 2004. Interim guidance for the use of human papillomavirus DNA testing as an adjunct to cervical cytology for screening. Obstet. Gynecol. 103:304-309. [PubMed]
18. zur Hausen, H. 2009. Papillomaviruses in the causation of human cancers—a brief historical account. Virology 384:260-265. [PubMed]

Articles from Journal of Clinical Microbiology are provided here courtesy of American Society for Microbiology (ASM)