Human papillomavirus (HPV) is the most common viral sexually transmitted infectious agent, peaking in incidence in young, sexually active females.1 For the majority of women, HPV infection is transient. However, in cases of persistent high-risk HPV infections, women with type-specific persistence are at significantly higher risk for precancerous cervical lesions (cervical intraepithelial neoplasia [CIN] grades 2 or 3).1
It is standard that women with CIN 2 or 3 undergo treatment such as cryotherapy, laser ablation, and loop electrosurgical excision procedure to remove the dysplastic lesions, thus preventing cancer.2 In some cases after treatment, there may be recurrence of the lesion.3–5 Hence, postsurgical follow-up is necessary to assess the success of treatment.6
Several studies have estimated the clearance of HPV infection after treatment, with most showing that infections clear in a majority of women; however, a proportion of HPV infection cases do not clear.3,7–15 Many of these studies were limited because of small numbers,13,15–17 inadequate length of follow-up,16–18 or combined measurement of high-risk HPV only, with no HPV genotyping.3,7,10 Knowledge of HPV genotype-specific clearance will inform whether genotypes detected after treatment are persistent infections. Persistent detection of high-risk HPV types is associated with an increased risk of disease recurrence.11,19 The aim of our study was to estimate the clearance of HPV in a large clinical sample of women presenting for gynecologic treatment at a dysplasia clinic in Melbourne, Australia.
MATERIALS AND METHODS
We invited women attending the Dysplasia Clinics of the Royal Women's Hospital in Melbourne, Australia, with a new diagnosis of high-grade dysplasia or persistent low-grade dysplasia (defined as low-grade dysplasia for 12 months or longer, based on biopsy before treatment) requiring treatment by excision (including loop electrosurgical excision procedure or cone biopsy) or laser ablation to participate in a prospective study. More than 90% of these women had a colposcopy-directed punch biopsy performed for histological confirmation of abnormality before determining whether treatment was required. Women were recruited from May 2001 through December 2007. Women who were previously treated for cervical dysplasia, had immunosuppression, or were human immunodeficiency virus-positive were excluded from this study. The majority women in the cohort (97%) were referred because of an abnormal Pap cytology screening result indicating low-grade to high-grade cervical dysplasia. The remaining women were referred because of the presence of other abnormalities such as abnormal bleeding, an abnormal looking cervix, or ongoing review of previous abnormality. All women had at least one examination before the treatment visit, during which cervical specimens were collected for conventional Pap test analysis and a colposcopy was performed. If the colposcopy was indicative of an abnormality, then a biopsy was obtained for histological analysis. Women were excluded if they had a diagnosis of vaginal intraepithelial neoplasia, vulval intraepithelial neoplasia, or lichen simplex by Pap test or histology.
At treatment, women provided informed consent to participate in the study. Ethical approval was received from the Royal Women's Hospital. When this study began, routine surveillance for treatment failure according to the National Health and Medical Research Council of Australia required colposcopy and cervical cytology between 2 and 6 months after treatment. This was followed-up with cervical cytology and sometimes, but not necessarily, colposcopy 12 months after treatment, and annually thereafter until there was no longer evidence of an abnormality.20 These guidelines were rescinded and updated in 2005 to require a follow-up colposcopy and cervical cytology between 4 and 6 months after treatment. Cervical cytology and HPV testing were then required 12 months after treatment and every 12 months thereafter, until a woman had negative results for both tests on two consecutive visits.6
The details of the specimen collection methods at treatment for this population have been described previously.21 Briefly, at treatment, cervical specimens were collected for conventional Pap test analysis. Residual cells on cervical brushes were rinsed into ThinPrep vials containing 20 mL of PreservCyt solution for HPV DNA genotyping. For loop electrosurgical excision procedure or cone biopsy procedures, lesions on the cervix were excised, providing tissue for histological analysis, whereas treatment by laser ablation did not liberate cervical tissue for histology. The treatment modality for each woman was not selected based on specific criteria, but rather was an independent decision by the treating clinician.
During follow-up visits, cervical cells were collected for Pap test and HPV DNA genotyping. If the colposcopy examination indicated an abnormality, then a biopsy was performed for histological analysis. Pap test specimens were classified using the 2004 Australia Modified Bethesda System, which differentiates between possible low-grade atypical squamous cells and possible high-grade atypical squamous cells, and histological specimens were classified using the Bethesda System.6
DNA was isolated from the PreservCyt samples using the MagNA Pure LC DNA Isolation Kit I on the automated MagNA Pure LC extraction system, according to a modified protocol.22
The linear array assay was used to test DNA samples from time of treatment and follow-up visits for specific HPV genotypes. Linear array detects 37 HPV genotypes, including 13 high-risk HPV genotypes (HPV 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, and 68) and 24 low-risk HPV genotypes (HPV 6, 11, 26, 40, 42, 53, 54, 55, 61, 62, 64, 66, 67, 69, 70, 71, 72, 73, 81, 82, 83, 84, IS39, and 89).23,24 Linear array was performed according to the manufacturer's instructions, with several modifications described previously.25,26 Because of cross-reactivity of the HPV 52 probe hybridizing with other high-risk-HPV genotypes (33, 35, and 58), samples positive for HPV 52 alone were classified as HPV 52-positive. Samples positive for HPV 52 in conjunction with HPV 33, 35, or 58 (or a combination of these) underwent a validation assay using a real-time polymerase chain reaction with a HPV 52-specific hydrolysis probe.27
Study time was defined as calendar days on study, beginning with the treatment visit (baseline) and ending with the last visit before the study ended. Participants who completed the baseline visit and at least one follow-up visit from which a valid HPV test was performed were included in this analysis. In the cross-sectional analysis, linear changes in the prevalence of individual HPV genotypes at subsequent visits were compared using a χ2 test for trend.28 Clearance of a single HPV genotype was defined as the first test in which that HPV genotype was not detected. Median time to clearance of individual HPV genotypes was estimated in those positive for that genotype at baseline and by taking into account all follow-up visits within the study period.
In longitudinal analyses, time to clearance was compared for the eight most prevalent HPV genotypes detected at baseline. Clearance was defined as the first visit after baseline that a specified HPV genotype was not detected. The probability of clearance was evaluated using the Kaplan-Meier method, and groups were compared using the log-rank test. Because of the possibility of multiple infections, groups were not mutually exclusive.
The aforementioned longitudinal analysis was repeated with four mutually exclusive categories based on HPV infection as baseline. We created a composite variable: 1) infection with HPV 16 and possibly other HPV genotypes (not HPV 18); 2) infection with HPV18 and possibly other HPV genotypes (not HPV 16); 3) infection with high-risk HPV genotype 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, or 68 (or a combination of these, but not HPV 16 or HPV 18); and 4) infection with low-risk HPV genotype 6, 11, 26, 40, 42, 53, 54, 55, 61, 62, 64, 66, 67, 69, 70, 71, 72, 73, 81, 82, 83, 84, IS39, or 89 (or a combination of these but no high-risk HPV detected). To create four mutually exclusive categories, we excluded women with both HPV 16 and 18 detected at baseline from this composite variable (n=31). Time of clearance was defined as the first visit after baseline in which HPV 16 was not detected (category 1), HPV 18 was not detected (category 2), high-risk HPV (other than types 16 and 18) were not detected (category 3), and low-risk HPV was not detected (category 4). To determine the effect of multiple compared with single infections from HPV on time to clearance, we divided the third and fourth categories into single and multiple high-risk and low-risk genotypes, respectively. For the latter two analyses based on mutually exclusive groups, potential confounders were accounted for using a Cox proportional hazards model. All analyses were conducted in Stata 11.
At baseline, 1,649 women were enrolled in the study and treated for cervical abnormality by either excision or laser ablation. Of these, 1,207 (73%) presented for at least one follow-up visit and had at least two valid HPV tests; therefore, they were included in our analysis data set. Of the 1,207 women in the study (96%, n=1,159) had three or more posttreatment visits, and 72% (n=863) had four or more. At baseline, 45% of women had laser ablative therapy, and 55% underwent excision therapy. More than 70% of women completed at least four visits, including the baseline visit, and 90% had preoperative histology results from a visit that occurred, on average, 3 months before the treatment date (Table 1). Compared with the complete baseline cohort, our restricted sample was slightly older (mean age 30.2 years compared with 29.7 years, P=.10). The histology results at the most recent preoperative visit were comparable between the two groups (Table 1).
At treatment, 84% of PreservCyt samples were positive for HPV by linear array, with 72% positive for high-risk HPV. The proportion positive for high-risk HPV decreased to 31% by the second visit (on average 6 months later), and to 24% and 27% at the third and fourth visits (on average 16 and 24 months later, respectively, P trend <.001; Table 2). The most common HPV genotypes at baseline were HPV 16, 18, 31, 39, 51, 52, 53, and 89, with all showing a decrease in prevalence at subsequent visits (P trend <.01). The detection of multiple types was common, with half of the women positive for HPV at baseline having two or more HPV genotypes detected, decreasing to 30% at visits 2 and 3, and to 22% at visit 4 (P trend <.001) (Table 2).
Histology was performed in all women undergoing excision therapy at baseline (n=665) and a subset of women at follow-up visits (visit 2=57, visit 3=142, visit 4=104). Nearly 40% of women had histologically confirmed CIN 3 (n=248) or cervical carcinoma (n=14) at baseline; however, recurrence of these conditions was rare. There were a total of eight confirmed cases of CIN 3 diagnosed during follow-up: five within 24 months, two within the next 24 months, and one 70 months after treatment. There were no recurrent cases of carcinoma during follow-up (Table 2).
Within the study period, almost all women who were positive for HPV 16 or 18 at treatment had clearance of these types (93% and 97%, respectively), and half occurred within 6 months (Table 3). In most cases, this occurred at the first visit after treatment, which was, on average, 6.3 months after treatment (standard deviation [SD] 5.1 months). The other two most common high-risk HPV genotypes, 51 and 52, showed similar clearance (93% and 90% respectively), with half clearing within 6 months. Low-risk HPV genotypes 53 and 89 also had high clearance (89% and 76%, respectively); however, the median clearance time was longer (10 and 10.8 months, respectively). Just more than half of women with any HPV genotype detected at baseline had clearance of all detectable types by the end of the study, with half clearing within 25 months. These observations were largely unchanged if other HPV genotypes were also detected at baseline (Table 3). The time to clearance did not differ by treatment group (laser compared with excision).
Of the 1,016 women who had HPV detected at baseline, 808 women had at least one of the eight most prevalent HPV genotypes detected (HPV 16, 18, 31, 39, 51, 52, 53, or 89). Two-thirds had more than one HPV genotype detected. Clearance times varied depending on the type of HPV detected at baseline (P<.001). The two low-risk HPV genotypes 53 and 89 cleared slower than the others (P=.006 and P<.001, respectively), whereas high-risk HPV genotypes 16, 18, and 31 cleared faster (P<.001, P=.05, P=.04, respectively; Fig. 1).
The data were divided into four mutually exclusive groups according to type-specific detection at baseline. There were 387 women with HPV 16 (mean age 28, SD 5.0) and possibly other HPV genotypes (not HPV 18), 68 with HPV 18 (mean age 31, SD 8.9) and possibly other HPV genotypes (not HPV 16), 388 with high-risk HPV genotypes (not HPV 16 or HPV 18; mean age 31, SD 8.2), and 142 with low-risk HPV genotypes (mean age 30, SD 9.9). For the latter two categories, 66% and 87% were multiple infections. For each category, 362 (94%), 66 (97%), 289 (74%), and 94 (66%) women had clearance of their HPV, respectively. Time to clearance was faster for women who had HPV 16 or HPV 18 at baseline, compared with other categories (P<.001; Fig. 2). This relationship remained after adjusting for age, preoperative histology, number of preoperative histology results, and treatment type. The model met the assumptions of the Cox proportional hazards model.
This analysis was repeated with the latter two categories separated into four categories (high-risk: 132 single and 256 multiple; low-risk: 86 single and 56 multiple) such that six categories of HPV infection were compared. Women with HPV 16, HPV 18, and other single high-risk HPV types (other than 16 and 18) had clearance of these infections faster than did women with single low-risk types, multiple low-risk types, and multiple high-risk types (other than 16 and 18; P<.001; data not shown).
This is a large, prospective study to measure time to clearance of HPV after treatment for cervical abnormalities. Overall, we observed a high clearance rate of HPV after treatment. Of the six most common high-risk HPV genotypes detected at baseline, nearly 90% cleared during follow-up and half were no longer detectable 6 months after treatment, irrespective of treatment type, excision, or ablation. For many women, this would be their first posttreatment visit, because the Australian guidelines at the time this study began recommended surveillance for treatment failure by colposcopy and cervical cytology between 2 and 6 months after treatment.20
Two of the eight most common HPV genotypes detected at baseline were low-risk (HPV 53 and HPV 89). Interestingly, the time to the first visit when these genotypes were not detected was longer for low-risk than high-risk genotypes. This association was observed irrespective of the presence of other types. When grouped into four mutually exclusive categories, HPV 16, HPV 18, high-risk HPV, and low-risk HPV, women with HPV 16 or 18 had clearance of these infections faster the than did other groups. This association remained after taking into account potential confounding factors such as treatment type (excision or ablation), age, preoperative histological grade, and number of visits. This could be explained by the definition of clearance, which was defined as clearance of a single type for HPV 16 and 18, but clearance of all HPV types for the remaining categories. These groups often comprised multiple infections. However, when these groups were subdivided into single and multiple infections, HPV 16 and HPV 18 still cleared faster than low-risk HPV. Infections with a single high-risk HPV type also cleared faster.
Many of the studies that have performed HPV genotyping to assess type-specific HPV clearance after treatment for cervical abnormalities have been limited by small sample size.8,9,15,17 Most relevant to this study, Kreimer et al12 analyzed data for 1,130 infections in 481 women who underwent loop electrosurgical excision procedure in the ALTS trial. Similar to our results, the authors reported a higher clearance for high-risk compared with low-risk infections 6 months after the procedure (82% and 72%, respectively, P<.001).12 When broken down into individual types, 74% of HPV 16 infections and 81% of HPV 18 infections cleared 6 months after loop electrosurgical excision procedure, with higher clearance rates than what we found in this timeframe. However, like our study, the clearance of HPV 89 at 6 months was substantially lower (60%).12 What our study adds, however, is more detailed data on time to clearance and comparisons of clearance times between HPV types. We observed a low rate of recurrence or residual disease in our study compared with what has been reported previously in similar populations.29,30
This was a clinic-based study, and the timing of the study visits related to the Australian recommendations for women undergoing treatment for cervical abnormalities and the clinician's discretion. To assess the health benefits of a “test-of-cure” scheme, we did not select the midpoint between two visits as the time of clearance, as others have done,31 but defined clearance as the visit date when the infection was no longer detectable. The infection would have cleared before this date. Also, the types of data that were collected were limited. We did not assess HPV viral load, involvement of the endocervical margin with CIN, or the lesion grade of the biopsy as predictors of clearance. These have been previously identified as predictors of clearance after treatment.3,10 However, we had information on treatment type, number of visits, age, and pretreatment cervical diagnosis to inform clearance. Finally, a substantial proportion of patients were lost to follow-up, which is common in this patient population.
In this group of more than 1,000 women attending dysplasia clinics for treatment of cervical abnormalities, we observed a high clearance rate of HPV infection, particularly high-risk HPV, and a low recurrence rate of disease after treatment. Because of our low rate of disease after treatment and possible differences between populations, treatment, and screening algorithms, the clinical utility of posttreatment HPV testing would require further local studies before implementation of HPV testing as a marker of adequacy of treatment or recurrence of cervical abnormalities.
1. Kjaer SK, van den Brule AJ, Paull G, Svare EI, Sherman ME, Thomsen BL, et al. Type specific persistence of high risk human papillomavirus (HPV) as indicator of high grade cervical squamous intraepithelial lesions in young women: population based prospective follow up study. BMJ 2002;325:572.
2. Soutter WP, de Barros Lopes A, Fletcher A, Monaghan JM, Duncan ID, Paraskevaidis E, et al. Invasive cervical cancer after conservative therapy for cervical intraepithelial neoplasia. Lancet 1997;349:978–80.
3. Costa S, De Simone P, Venturoli S, Cricca M, Zerbini ML, Musiani M, et al. Factors predicting human papillomavirus clearance in cervical intraepithelial neoplasia lesions treated by conization. Gynecol Oncol 2003;90:358–65.
4. Distefano AL, Picconi MA, Alonio LV, Dalbert D, Mural J, Bartt O, et al. Persistence of human papillomavirus DNA in cervical lesions after treatment with diathermic large loop excision. Infect Dis Obstet Gynecol 1998;6:214–9.
5. Mitchell MF, Tortolero-Luna G, Cook E, Whittaker L, Rhodes-Morris H, Silva E. A randomized clinical trial of sryotherapy, laser vaporization, and loop electrosurgical excision for the treatment of squamous intraepithelial lesions of the cervix. Obstet Gynecol 1998;92:747–4.
6. National Cancer Screening Program. Screening to prevent cervical cancer: guidelines for management of asymptomatic women with screen detected abnormalities. Canberra (Australia): National Health and Medical Research Council; 2005.
7. Alonso I, Torné A, Puig-Tintoré LM, Esteve R, Quinto L, Campo E, et al. Pre- and post-conization high-risk HPV testing predicts residual/recurrent disease in patients treated for CIN 2–3. Gynecol Oncol 2006;103:631–6.
8. Bais AG, Eijkemans MJ, Rebolj M, Snijders PJ, Verheijen RH, van Ballegooijen M, et al. Post-treatment CIN: randomised clinical trial using hrHPV testing for prediction of residual/recurrent disease. Int J Cancer 2009;124:889–95.
9. Brismar S, Johansson B, Borjesson M, Arbyn M, Andersson S. Follow-up after treatment of cervical intraepithelial neoplasia by human papillomavirus genotyping. Am J Obstet Gynecol 2009;201:17 e1–8.
10. Kim YT, Lee JM, Hur SY, Cho CH, Kim YT, Kim SC, et al. Clearance of human papillomavirus infection after successful conization in patients with cervical intraepithelial neoplasia. Int J Cancer 2010;126:1903–9.
11. Kreimer AR, Guido RS, Solomon D, Schiffman M, Wacholder S, Jeronimo J, et al. Human papillomavirus testing following loop electrosurgical excision procedure identifies women at risk for posttreatment cervical intraepithelial neoplasia grade 2 or 3 disease. Cancer Epidemiol Biomarkers Prev 2006;15:908–14.
12. Kreimer AR, Katki HA, Schiffman M, Wheeler CM, Castle PE, ASCUS-LSIL Triage Study Group. Viral determinants of human papillomavirus persistence following loop electrical excision procedure treatment for cervical intraepithelial neoplasia grade 2 or 3. Cancer Epidemiol Biomarkers Prev 2007;16:11–6.
13. Nagai Y, Maehama T, Asato T, Kanazawa K. Persistence of human papillomavirus infection after therapeutic conization for CIN 3: is it an alarm for disease recurrence? Gynecol Oncol 2000;79:294–9.
14. Sarian LO, Derchain SF, Pitta Dda R, Morais SS, Rabelo-Santos SH. Factors associated with HPV persistence after treatment for high-grade cervical intra-epithelial neoplasia with large loop excision of the transformation zone (LLETZ). J Clin Virol 2004;31:270–4.
15. Venturoli S, Ambretti S, Cricca M, Leo E, Costa S, Musiani M, et al. Correlation of high-risk human papillomavirus genotypes persistence and risk of residual or recurrent cervical disease after surgical treatment. J Med Virol 2008;80:1434–40.
16. Bar-Am A, Gamzu R, Levin I, Fainaru O, Niv J, Almog B. Follow-up by combined cytology and human papillomavirus testing for patients post-cone biopsy: results of a long-term follow-up. Gynecol Oncol 2003;91:149–53.
17. van Ham MA, van Hamont D, Bekkers RL, Bulten J, Melchers WJ, Massuger LF. High-risk HPV presence in cervical specimens after a large loop excision of the cervical transformation zone: significance of newly detected hr-HPV genotypes. J Med Virol 2007;79:314–9.
18. Houfflin Debarge V, Collinet P, Vinatier D, Ego A, Dewilde A, Boman F, et al. Value of human papillomavirus testing after conization by loop electrosurgical excision for high-grade squamous intraepithelial lesions. Gynecol Oncol 2003;90:587–92.
19. Bae JH, Kim CJ, Park TC, Namkoong SE, Park JS. Persistence of human papillomavirus as a predictor for treatment failure after loop electrosurgical excision procedure. Int J Gynecol Cancer 2007;17:1271–7.
20. Screening to prevent cervical cancer: guidelines for the management of women with screen detected abnormalities [rescinded]. Canberra (Australia): National Health and Medical Research Council; 1994.
21. Stevens MP, Garland SM, Tan JH, Quinn MA, Petersen RW, Tabrizi SN. HPV genotype prevalence in women with abnormal pap smears in Melbourne, Australia. J Med Virol 2009;81:1283–91.
22. Stevens MP, Rudland E, Garland SM, Tabrizi SN. Assessment of MagNA pure LC extraction system for detection of human papillomavirus (HPV) DNA in PreservCyt samples by the Roche AMPLICOR and LINEAR ARRAY HPV tests. J Clin Microbiol 2006;44:2428–33.
23. Coutlée F, Rouleau D, Petignat P, Ghattas G, Kornegay JR, Schlag P, et al. Enhanced detection and typing of human papillomavirus (HPV) DNA in anogenital samples with PGMY primers and the Linear array HPV genotyping test. J Clin Microbiol 2006;44:1998–2006.
24. Gravitt PE, Peyton CL, Alessi TQ, Wheeler CM, Coutlée F, Hildesheim A, et al. Improved amplification of genital human papillomaviruses. J Clin Microbiol 2000;38:357–61.
25. Stevens MP, Garland SM, Tabrizi SN. Validation of an automated detection platform for use with the roche linear array human papillomavirus genotyping test. J Clin Microbiol 2008;46:3813–6.
26. Stevens MP, Tabrizi SN, Quinn MA, Garland SM. Human papillomavirus genotype prevalence in cervical biopsies from women diagnosed with cervical intraepithelial or cervical cancer in Melbourne, Australia. Int J of Gynecol Cancer 2006;16:1017–24.
27. Stevens MP, Garland SM, Tabrizi SN. Development and validation of a real-time PCR assay specifically detecting human papillomavirus 52 using the Roche LightCycler 480 system. J Virol Methods 2008;147:290–6.
28. Royston P. PTREND: Stata module for trend analysis for proportions. London (UK): MRC Clinical Trials Unit; 2002.
29. Chan BK, Melnikow J, Slee CA, Arellanes R, Sawaya GF. Posttreatment human papillomavirus testing for recurrent cervical intraepithelial neoplasia: a systematic review. Am J Obstet Gynecol 2009;200:422e1–9.
30. Melnikow J, McGahan C, Sawaya GF, Ehlen T, Coldman A. Cervical intraepithelial neoplasia outcomes after treatment: long-term follow-up from the British Columbia Cohort Study. J Natl Cancer Inst 2009;101:721–8.
31. Insinga RP, Dasbach EJ, Elbasha EH, Liaw KL, Barr E. Incidence and duration of cervical human papillomavirus 6, 11, 16, and 18 infections in young women: an evaluation from multiple analytic perspectives. Cancer Epidemiol Biomarkers Prev 2007;16:709–15.