Elfgren, Kristina MD; Jacobs, Marcel MD; Walboomers, Jan M. M. PhD†; Meijer, Chris J. L. M. MD, PhD; Dillner, Joakim MD, PhD
Cervical cancer is the third most common cancer among women worldwide, with approximately 370,000 cases a year and 190,000 deaths.1,2 Organized cytologic screening protects against cervical cancer, but there has been a recent increase of cervical cancer also in a thoroughly screened population.3 Screening programs today identify women with abnormal cytology for further examination by colposcopy and cervical biopsy and eventually surgical removal of a histologically verified cervical intraepithelial neoplasia (CIN), the precursor to cervical cancer. Follow‐up after treatment has so far consisted of repeat cytology and, possibly, colposcopy.
Infection with human papillomavirus (HPV) is established as a prerequisite for the development and maintenance of the vast majority of cervical cancers and CINs.4–7 We previously reported that HPV DNA is usually no longer present 2 years after effective treatment for CIN,8 suggesting that strategies for follow‐up after treatment of CIN based on monitoring the clearance of the major risk factor for CIN (ie, HPV) might be feasible. To expand on these findings, we observed a larger group of patients with more frequent HPV DNA testing after treatment and determined the time course of HPV clearance and the determinants of HPV persistence after treatment, including examination of the current sexual partners.
MATERIALS AND METHODS
The institutional review board of Huddinge Hospital approved the study (decision no. 126/95).
For maximum generalizability, all women admitted to the Department of Gynecology and Obstetrics, Huddinge University Hospital, Stockholm, Sweden, for treatment of CIN I–III (66 women with CIN III, 21 with CIN II, and 22 with CIN I) from September 1995 to September 1997 and giving informed consent were eligible to be enrolled. There was no HIV screening. The CIN diagnosis was based on diagnostic biopsy from the cervix and also, in most cases, endocervical curettage. The lag between the diagnostic biopsy taken at the gynecology outpatient clinics in the catchment area (southern Stockholm County) and treatment at the hospital was on average about 2.5 months. The diagnostic biopsies had been taken because of an abnormal Papanicolaou smear, taken in either organized or opportunistic screening.
At the time of the study, the main features of the established regional guidelines for choice of treatment of CIN were that CIN I and II should be treated by cryotherapy and CIN III by conization.
Seventy‐one women (mean age 35 years, range 21–71) were treated with conization (66 had CIN III and five had CIN II), and 38 women (mean age 30 years, range 20–55) were treated with cryotherapy (22 had CIN I and 16 had CIN II).
Of the 109 women, 100 were white, three from the Middle East, three Latin American, and three Asian. One woman had immunosuppressive treatment because of a renal transplant. Two women became pregnant at the beginning of the study, but continued to participate. Three women were pregnant at the last visit.
All women were examined by the same gynecologist (KE) immediately before surgical treatment. A cervical brush sample (Cytobrush; Medscand, Malmö, Sweden) was collected from the endo‐ and ectocervices. The brush was put in a plastic tube with 1 mL of phosphate‐buffered saline solution containing a 5‐mmol/L ethylenediaminetetra‐acetic acid buffer, immediately frozen at −20C, and later transferred to −70C storage for future analysis side by side with the cervical samples taken at the follow‐up visits.
The conization was performed as an electrosurgical excision with microneedle diathermy on all 71 women. Cryotherapy was performed using nitrous oxide as the refrigerant. The cryosurgical probe was applied to the cervix and a procedure of 2 × 3–minute freeze with thawing in between was carried out.
Follow‐up visits were scheduled at 3, 6, 9, 12, and 24 months after treatment. The actual mean times for the follow‐up visits were 3 months and 8 days, 6 months and 8 days, 9 months and 9 days, 13 months and 1 day, and 24 months. Of the initial 109 enrolled patients, 91% attended the 3‐month visit. Three women could not be sampled because they had had a hysterectomy, because of the CIN lesion. The attendance rates of the originally enrolled 109 women were 84.5% at 6 months, 84% at 9 months, and 85% at 12 months. On March 1, 1998, when the study was closed, only 46 patients had been observed for 24 months after treatment and were eligible for invitation. Seventy‐three percent of these invited women attended. The mean length of follow‐up for the eligible women was 15.7 months.
At all follow‐up visits, the women were examined by the same gynecologist (blinded to the HPV DNA status) who examined them at baseline. Cervical brush sampling was conducted in the same way as at the baseline visit, before a Papanicolaou smear, colposcopy, and, if necessary (ie, in case of colposcopic or cytologic signs of CIN), a punch biopsy was performed. Altogether 407 follow‐up visits were conducted, and at 34 of these visits (8%) a punch biopsy was taken. Four of these resulted in a new treatment before the study closed, two conizations (CIN II, atypical squamous cells of undetermined significance [ASCUS] plus immunosuppression), one cryotherapy (CIN I), and one reconization (CIN II).
At each follow‐up visit, all women were asked about new sexual partners since the previous consultation. Among the 72 women subjected to conization, there were 18 changes of male sexual partners (25%). Among the 38 women in the cryotherapy group, 12 new male sexual partners (32%) were reported during the follow‐up period.
All sexual partners of the women at inclusion and during follow‐up were invited for a single examination including sampling of HPV DNA by rotating a cyto‐brush over the glans penis and sulcus coronarius, after which the same brush was inserted into the urethral orifice. The brush was thereafter handled in the same manner as the cervical brushes. Forty‐eight (51%) of the 95 reported male partners of the 109 women attended. Thirty were partners of women treated with conization, and 18 were partners of women undergoing cryosurgery. Samples from 39 men were β‐globin positive. The men were examined after treatment of their female partner. None of the CIN case women reported female sexual partners in the study.
After the study was closed, all cervical samples from baseline and follow‐up visits were thawed and the tube was vortexed to dispense mucus and cells from the brush. The brush was then removed from the tube in a sterile manner while the remaining mucus and cells on the brush were pressed towards the edge of the tube. After vortexing, 350 μL of the sample was transferred to a new tube, frozen, and analyzed for HPV DNA. The analyzing laboratory was blinded to the identity of the samples, but an analysis order list ensuring that samples from the same women were analyzed side by side was provided.
The brush containing HPV DNA from the male partner was treated in the same manner and analyzed side by side with the samples from the female patients.
Testing for HPV was performed using an HPV general primer GP5+/6+‐mediated polymerase chain reaction (PCR)–enzyme immunoassay method. The PCR amplification products were individually probed with oligonucleotide probes for HPV typing of 14 high‐risk‐(HPV 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 66, 68) and six low‐risk (HPV 6, 11, 40, 42, 43, 44) HPV types. Three times the mean optical density value of the PCR‐negative controls was used as a cutoff point. This value revealed a 100% agreement between enzyme immunoassay results and results from the radioactive method. Testing for β‐globin ensured that adequate samples containing intact human DNA had been taken. The method used is described in detail elsewhere.9 There is also high interlaboratory agreement of the method (κ = 0.88–1.0).10
Survival analysis with the Cox‐Mantel test analyzed the clearance rate after treatment.11 Losses to follow‐up and hysterectomies were treated as censored observations.
We defined women as having cleared their HPV infection if the type of HPV infection at the baseline visit could not be detected at follow‐up. For women positive for multiple HPV types at baseline, clearance from all HPV types was required.
One hundred nine women (mean age 32.5 years, range 20–71) were enrolled in the study. These women were sampled before treatment of CIN, and 104 women had an adequate β‐globin–positive sample. Twenty of 104 women (19%) were negative for HPV DNA. Six of these women became HPV positive in at least one visit during follow‐up.
Among the 84 women with positive HPV DNA tests, 18 double infections were detected, giving a total of 102 cervical infections. Human papillomaviruses 16, 18, and 31 were by far the most common types (Table 1). Four of the initially HPV DNA–positive women could not be observed, three because of hysterectomy and one because of a β‐globin–negative sample at follow‐up. Of the 80 HPV‐positive women who could be followed up, 49 had had conization and 31 had had cryotherapy.
Among the women treated with conization, only six remained positive with an HPV DNA type present before therapy at 3 months after treatment. Two women remained continuously positive during the first year of follow‐up visits. Both of these women had HPV 16 (Figure 1 and Table 2) and involved margins in the cone specimen.
Fifteen of the 31 women who were treated with cryosurgery had an HPV DNA type at the 3‐month follow‐up visit that had also been present before treatment. The rates of persistence declined gradually during follow‐up, but at the follow‐up visit at 12 months five patients were still persistently HPV positive (Figure 1), four of them having CIN I in cytology during follow‐up (Figure 1 and Table 2).
By survival analysis the clearance rate was significantly slower in the cryotherapy group (P < .002) (Figure 1).
Presence of multiple infections was common among cryotherapy‐treated women, both before treatment and during the follow‐up (Tables 1 and 2).
Only four couples had HPV DNA of the same HPV type detected (three couples with HPV 16 and one couple with HPV 6). Two women in these couples were persistently HPV DNA positive with the same type (6 or 16) at all follow‐up visits, whereas the other two women did show clearance at follow‐up (odds ratio for HPV persistence in case of detection of the same HPV DNA type in the partner = 10.8 [95% confidence interval 0.55, 164.3]).
During the 2 years of follow‐up, all together 17 women (16% of 104) had abnormal cytology at least once. Seven women treated with conization had cytological CIN I or II in a total of 15 smears. Six of these women had a positive HPV test and a positive cytology concurrently at least once. Two of them had the same HPV type as before treatment, and three women had positive margins in the cone specimen. Three women had CIN II (two with a new HPV type and one with persistence).
Ten women from the cryotherapy group had cytologic CIN I, CIN II, or ASCUS in at least one smear of a total of 21 smears. Eight of these ten women were HPV positive concurrently with the abnormal smear, four of them with the same HPV type as before treatment.
The objective of the present study was not to investigate whether HPV infection causes CIN III, as this is already well known, but to investigate whether HPV can be cleared after treatment and at what rate. Clearance of HPV DNA was rapid and usually occurred within 3 months after treatment. Our mean length of follow‐up was only 16 months, but there were few additional clearances occurring after 6 months. HPV DNA clearance was primarily found in the conization group, suggesting that this type of treatment in this group of women had indeed resulted in clearance of HPV infection. However, it should be stressed that the purpose of the present study was not to compare treatments per se, but to describe the rates of HPV clearance in different treatment groups, which is essential for planning of when to take follow‐up samples. Studies comparing different treatments should be based on randomized allocation of treatment.
In recent years, several studies have confirmed our finding that after successful CIN III treatment HPV DNA is no longer detectable even by highly sensitive PCR methodology. In our original study of 23 patients with HPV‐positive CIN III,8 only one woman was still positive for the same HPV type 2 years after treatment. Similar results were published by Kjellberg et al,12 who found only three women HPV DNA positive 3 years after laser conization of 82 initially HPV DNA–positive women with CIN I–III. All three women had a new HPV type. In two studies by Bollen et al,13,14 88–90% of women treated for CIN had cleared their initial HPV infection at follow‐up 1 year after treatment. Strand et al15 also found that 27 of 30 women treated for CIN I–III were negative for HPV DNA at follow‐up 6–12 months after treatment. Kanamori et al16 found that two of 27 HPV DNA–positive women treated with conization were HPV DNA positive after conization. Nagai et al17 reported an HPV DNA persistence after conization for HPV DNA–positive CIN III of 19.6%, and the recurrence of CIN in this group of patients to 46%. None of the patients with negative HPV DNA tests after treatment had recurrence of CIN. Chua and Hjerpe18 analyzed the presence of HPV DNA in normal Papanicolaou smears taken after treatment of CIN and found HPV DNA in 25 of 26 women who later developed recurrent disease, whereas none of the 22 treated women who remained healthy had detectable DNA. Nobbenhuis et al19 found that HPV DNA negativity 6 months after treatment had a 99% negative predictive value that CIN II–III would not develop. Tate et al,20 who performed PCR on microdissected specimens, found that HPV DNA is very uncommon in normal tissue adjacent to CIN, suggesting that the mechanism of clearance might simply be removal of the infected cells. Our results, with a more frequent HPV clearance in the conization group, could be interpreted as a more effective treatment of CIN and the underlying HPV infection. However, the conization group had a higher CIN grade, were older, and had a lower rate of sexual partner change than the cryotherapy group. Several of these explanations may have contributed to the differences in HPV clearance. It is possible, for example, that CIN I may reflect an early infection with substantial production of virus, whereas CIN III lesions produce less virus and may be more prone to heal after treatment. The clearance rates of different types could not be studied because 39% of the infections were HPV 16 and the other 12 HPV types were represented only by one to four cases per type. The fact that two women with a partner positive for the same HPV DNA type were persistently positive at all visits suggests that type‐specific reinfection may also be a determinant of persistence.
Several different treatment strategies exist for the treatment of CIN. At electrosurgical conization there is a better control for removal of the CIN lesion, and thus the HPV infection, relative to the blind surgery of cryotherapy. However, Mitchell et al21 observed no significant difference in success rate for curing the CIN I–III lesions when comparing cryosurgery, loop electrosurgical excision, and laser vaporization. Studies comparing different treatments are hampered by the long follow‐up times required until CIN recurrence is observed and by limited statistical power because of few recurrent cases. We suggest that the use of HPV testing after treatment could be useful as an intermediate end point to achieve more rapid results and more power when comparing different CIN treatments.
The majority of pathologic Papanicolaou smears during follow‐up in our study were CIN I, which might reflect true neoplasia but can also be a nonspecific finding or reflect a recent new HPV infection. The results in our study show that the CIN treatment strategies used in a vast majority of cases will result in negative cytology and eradication of the disease. The fact that some women become disease free, at least temporarily, but do not clear the HPV infection after treatment raises questions as to what the mechanisms of clearance of the virus infection might be. Of the seven patients persistently HPV positive at 12 months in our study, the two patients from the conization group had unclear margins in the cone specimen. In the cryotherapy group one had multifocal genital warts and one had immunosuppressive therapy because of an organ transplant. Immunosuppression is a well‐known risk factor for the development, persistence, and recurrence of HPV infection and disease.22,23 For the other three patients with persistence, no known risk factor could be identified. All patients presenting with CIN II (the true risk group for developing CIN III) during follow‐up had positive HPV tests concurrent with the CIN II smear. However, there were also patients with persistently positive HPV DNA tests who did not show cytologic abnormalities in the Papanicolaou smear during follow‐up. It is well established that the persistent HPV infection is a prerequisite for the development of CIN III.7
Our present study extends earlier findings with smaller patient groups that effective treatment of CIN will indeed rapidly clear the HPV infection. Our results imply that removal of the main risk factor for CIN, the HPV infection, is an attainable goal. They also suggest that HPV testing could be used as an important intermediate end point in the follow‐up after treatment of CIN for evaluation of efficacy of treatments.
1. Parkin M, Pisani P, Ferlay J. Estimates of the worldwide incidence of 25 major cancers in 1990. Int J Cancer 1999; 80:827–41.
2. Pisani P, Parkin M, Bray F, Ferlay J. Estimates of the world-wide mortality from 25 cancers in 1990. Int J Cancer 1999;83:18–29.
3. Anttila A, Pukkala E, Söderman B, Kallio M, Nieminen P, Hakama M. Effect of organised screening on cervical cancer incidence and mortality in Finland 1963–1995: Recent increase in cervical cancer incidence. Int J Cancer 1999;83:59–65.
4. zur Hausen H. Papillomavirus infections—a major cause of human cancers. Biochim Biophys Acta 1996;1288:F55–78.
5. Schiffman MH, Bauer HM, Hoover RN, Glass AG, Cadell DM, Rush B, et al. Epidemiological evidence that human papillomavirus infection causes most cervical intraepithelial neoplasia. J Natl Cancer Inst 1993;85:958–63.
6. Koutsky L, Holmes KK, Critchlow CW, Stevens CE, Paavonen J, Beckman AM, et al. A cohort study of the risk of cervical intraepithelial neoplasia grade 2 or 3 in relation to papillomavirus infection. N Engl J Med 1992;327:1272–8.
7. Nobbenhuis MAE, Walboomers JMM, Helmerhorst TJM, Rozendaal L, Remmink AJ, Risse EK, et al. Relation of human papillomavirus status to cervical lesions and consequences for cervical-cancer screening: A prospective study. Lancet 1999;354:20–5.
8. Elfgren K, Bistoletti P, Dillner L, Walboomers JMM, Meijer CJ, Dillner J. Conization for cervical intraepithelial neoplasia is followed by disappearance of human papillomavirus deoxyribonucleic acid and a decline in serum and cervical mucus antibodies against human papillomavirus antigens. Am J Obstet Gynecol 1996;174:937–42.
9. Jacobs MV, van den Brule AJC, Snijders PJF, Meijer CJLM, Helmerhorst ThJM, Walboomers JMM. A general primer GP5+/6+ mediated PCR EIA for rapid detection of 14 high-risk and 6 low-risk human papillomavirus genotypes in cervical scrapings. J Clin Microbiol 1997;35:791–5.
10. Jacobs MV, Snijders PJF, Voorhorst FJ, Dillner J, Forslund O, Johansson B, et al. Reliable high risk HPV DNA testing by polymerase chain reaction: An intermethod and intra-method comparison. J Clin Pathol 1999;52:498–503.
11. Statsoft Inc. STATISTICA for Windows [manual]. Tulsa, Oklahoma: StatSoft Inc., 1995.
12. Kjellberg L, Wadell G, Bergman F, Isaksson M, Angström T, Dillner J. Regular disappearance of the human papillomavirus genome after conization of cervical epithelial dysplasia by carbon dioxide laser. Am J Obstet Gynecol 2000;183:1238–42.
13. Bollen LJM, Tjong-A-Hung S, van der Velden J, Mol BWJ, Lammes FB, ten Kate FWJ, et al. Human papillomavirus DNA after treatment of cervical dysplasia. Low prevalence in normal cytologic smears. Cancer 1996;77:2538–43.
14. Bollen LJM, Tjong-A-Hung S, van der Velden J, Mol BW, Boer K, ten Kate FWJ, et al. Clearance of human papillomavirus infection by treatment for cervical dysplasia. Sex Transm Dis 1997;24:456–60.
15. Strand A, Wilander E, Zehbe I, Rylander E. High risk HPV persists after treatment of genital papillomavirus infection but not after treatment of cervical intraepithelial neoplasia. Acta Obstet Gynecol Scand 1997;76:140–4.
16. Kanamori Y, Kigawa J, Minagawa Y, Irie T, Oishi T, Itamochi H, et al. Residual disease and presence of human papillomavirus after conization. Oncology 1998;55:517–20.
17. Nagai Y, Maehama T, Asato T, Kanawaza 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.
18. Chua K-L, Hjerpe A. Human papillomavirus analysis as a prognostic marker following conization of the cervix uteri. Gynecol Oncol 1997;66:108–13.
19. Nobbenhuis MA, Meijer CJ, van Brule AJ, Rozendaal L, Voorhorst FJ, Risse EK, et al. Addition of high-risk HPV testing improves the current quidelines on follow up after treatment for cervical intraepithelial neoplasia. Br J Cancer 2000;84:796–801.
20. Tate JE, Murray R, Sheets EE, Crum CP. Absence of papillomavirus DNA in normal tissue adjacent to most cervical intraepithelial neoplasms. Obstet Gynecol 1996; 88:257–60.
21. Mitchell MF, Tortolero-Luna G, Cook E, Whittaker L, Rhodes-Morris H, Silva E. A randomized clinical trial of cryotherapy, laser vaporization, and loop electrosurgical excision for treatment of squamous intraepithelial lesions of the cervix. Obstet Gynecol 1998;92:737–44.
22. Wright TC, Ellerbrook TV, Chiasson MA, Van Devanter N, Sun XW, New York Cervical Disease Study. Cervical intraepithelial neoplasia in women infected with human immunodeficency virus: Prevalence, risk factors and validity of Papanicolau smears. Obstet Gynecol 1994;84:591–7.
23. Delmas MC, Larsen C, van Benthem B, Hamers FF, Bergeron C, Poveda JD, et al. Cervical squamous intraepithelial lesions in HIV-infected women: Prevalence, incidence and regression. AIDS 2000;14:1775–84.