| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ORIGINAL RESEARCH |
From the Department of Gynecology, Biostatistics and Cytopathology Laboratories, Ioannina University Hospital, Ioannina, Greece; and the Academic Unit of Obstetrics and Gynecology, St Mary’s Hospital, Manchester, United Kingdom.
Address reprint requests to: Evangelos Paraskevaidis, MD, Visiting Professor, Academic Unit of Obstetrics and Gynaecology, St. Mary’s Hospital, Whitworth Park, Manchester M13 0JH, United Kingdom; E-mail: vangelispar{at}hotmail.com.
 |
ABSTRACT |
|---|
TOP
 ABSTRACT MATERIAL AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
METHODS: We performed a retrospective study of 41 women who developed subsequent CIN after LLETZ (group A) and 82 women without CIN for a minimum of 5 years after LLETZ (group B). The first post-treatment cervical smear was retrieved and examined for high-risk HPV deoxyribonucleic acid. The sensitivity, specificity, positive and negative likelihood ratio of HPV testing, first post-treatment Papanicolaou test, and excision margins for the detection of treatment failure were calculated. Multiple logistic regression analysis was also done.
RESULTS: The HPV test was positive in 38 of 41 women in group A and 13 of 82 in group B (P < .001). An abnormal cytologic result in the first post-treatment smear was found in 20 of 41 in group A and 11 of 82 in group B (P < .001). Sixteen women in group A and 18 in group B had involved margins (P = .046). Values for the sensitivity, specificity, and positive and negative likelihood ratios of the HPV test were 93%, 84%, 5.8, 0.08; for the Papanicolaou test they were 49%, 87%, 3.9, 0.586; and for margin status they were 39%, 78%, 1.8, 0.782, respectively. Positive HPV test presents significantly high odds ratio for treatment failure (P < .001), independent of cytology and margin status.
CONCLUSION: Women who postoperatively have positive HPV testing are at higher risk of treatment failure. This could be performed at the first post-treatment visit and further follow-up could be adjusted accordingly.
Conservative treatment of women with cervical intraepithelial neoplasia (CIN) reduces the risk of appearance of cervical cancer. However, these women are still more likely to develop invasive cancer than the general population.1 They should therefore be followed up for the detection of residual or recurrent lesions. Cytology and colposcopy are both used in the follow-up, but widely accepted protocols do not exist.
Conservative excision methods such as large loop excision of the transformation zone (LLETZ) are increasingly used in the treatment of CIN. One advantage of these methods is the pathologic assessment of excision margins. It is obvious that women with extension of CIN to the margins should be followed up closely, because they have much greater risk of reappearance of CIN after LLETZ2 depending on whether CIN was extending to the endocervical or ectocervical margin. However, even clear margins do not always indicate eradication of the disease. In fact, the risk of treatment failure in such cases ranges from 1.9% to 6%, independent of the excision method used.3–5 In a recent study, satellite lesions of human papillomavirus (HPV) infection detected colposcopically during treatment and located outside the transformation zone were found to be a major independent risk factor for treatment failure in women with clear margins.6
Because post-treatment HPV testing could possibly detect the presence of such satellite lesions, we conducted the present study to investigate whether HPV testing could predict treatment failure after LLETZ and, consequently, its role in the follow-up of women who underwent excisional treatment for CIN.
|
MATERIAL AND METHODS |
|---|
|
TOP
ABSTRACTMATERIAL AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
These 41 women with treatment failure were classified as group A. The women in group A were compared with 82 women classified as group B–women who were free of disease for at least 5 years of follow-up after LLETZ treatment for CIN and were randomly selected from the chart list. Follow-up was by cytology and colposcopy.
For each woman the first post-treatment cervical smear (which had to be taken in the first post-treatment year) was retrieved from the archive and was tested for the presence of high-risk HPV types DNA. The detection of high-risk HPV types DNA (types 16, 18, 31, 33) was done by polymerase chain reaction (PCR).
The glass slides were soaked in xylene and left for 4–7 days until the coverslips could be easily removed. The cells were collected with a sterile razor blade and transferred into an Eppendorf tube containing 1 mL of xylene. Samples were vortexed in xylene and centrifuged at 12,000g for 5 minutes. Then xylene was carefully removed with a clean pipette. The xylene wash was repeated and was followed by two rounds of wash in 1 mL of 100% ethanol. The sample was vortexed vigorously and centrifuged. The pellets were air dried at room temperature. To prepare DNA, we used spin columns according to the QIAamp DNA mini kits (Qiagen GmbH, Hilden, Germany) protocol with minor modifications. Performing PCR for ß-globulin tested the intactness of the DNA for each sample. Specimens testing positive for ß-globulin were then examined for HPV DNA with the L1 consensus primers GP5+/GP6+ that detect a broad spectrum of sequenced and still-unsequenced HPV types at subpicogram level as described previously.7 Specimens containing HPV were then tested for HPV types 16, 18, 31, and 33 by using PCR type-specific primers.
Primary statistical analysis was carried out with the 
2 test. The sensitivity and specificity were calculated for the Papanicolaou test, the HPV test, and the excision margin status. Likelihood ratios were also calculated for normal and abnormal test outcome. Likelihood ratios are formulated for a normal test outcome by 1-sensitivity/specificity and for an abnormal test outcome by sensitivity/1-specificity. The calculation of likelihood ratios enables comparison of diagnostic values of tests in a prevalence-independent way. The likelihood ratios can be multiplied with the odds of the prevalence (pretest probability) to determine the odds of post-test probabilities (predictive values).8 Multiple logistic regression analysis was done using treatment failure as the dependent factor and margin status, Papanicolaou test, and HPV test as independent factors.
|
RESULTS |
|---|
|
TOP
ABSTRACTMATERIAL AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. The HPV test was positive in the first post-treatment cervical smear in 38 of 41 women in group A (92.7%) and 13 of 82 in group B (15.9%) (P < .001). An abnormal cytologic result in the first post-treatment smear was found in 20 of 41 (48.8%) in group A and in 11 of 82 in group B (13.4%) (P < .001). Sixteen of group A (39%) and 18 of group B (21.9%) had positive margins of excision on pathologic examination of the cervical specimen (P = .046). Cytologic and HPV test results on the first post-treatment smear in both groups are shown on Table 2; the overlap of findings is also shown in Table 2. Of the 31 women in group A who had either normal cytology results or clear margins, 28 had a positive HPV test. And of the 26 women in group B who had either abnormal cytology results or involved margins, 19 had a negative HPV test.
|
|
).
|
. A positive HPV test presents a significantly higher odds ratio (OR) for treatment failure (P < .001), independent of cytology and margin status.
|
|
DISCUSSION |
|---|
|
TOP
ABSTRACTMATERIAL AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Two factors could cause potential biases in this study. First, women who had involved margins of excision or abnormal follow-up cytology might have had better compliance; however the follow-up protocol and compliance with the protocol were the same in both groups. Second, 14 cases with treatment failure were excluded from the study for reasons already described, and this could also be related to a selection bias, although it is improbable that this could place limitations in the interpretation of our data. In our study we did not calculate the predictive values for normal and abnormal test outcome, because these values are related to the pretest probability (frequency of treatment failure) that is arbitrary in our study (41 cases for 82 controls). The real pretest probability can only be estimated on the basis of consecutive patients.
Recent data from our group showed that satellite lesions resembling HPV infection detected colposcopically during treatment and located outside the transformation zone are a risk factor for treatment failure after LLETZ in patients with clear margins.6 HPV testing could reveal these residual nests of HPV infection that cause a repopulation of the transformation zone and eventually recurrence. In cases with involved margins, the reason for treatment failure is more likely to be due to an incompletely excised lesion containing HPV DNA, which is detected by the test.
In this study, HPV testing predicted treatment failure more accurately than either the first post-treatment Papanicolaou test or the pathologic evaluation of excision margins. Based on this finding, HPV testing could be used in the early follow-up period of these patients, irrespective of the excision margin status for triage follow-up. In only three cases of treatment failure was the HPV test negative. There is a minor risk that a positive HPV test post-treatment could represent reinfection and not treatment failure, especially if the subsequent lesion is CIN-1. It is, however, very unlikely, because the HPV testing was done on the first follow-up smear, which was taken in the first post-treatment year in all women, and the mean time from treatment to the first follow-up smear was only 4.2 months.
Further prospective studies are required to verify our results and define the role of HPV testing in follow-up. In a future prospective study, the management plan could be based on HPV testing. It could be that if the follow-up smear at 6 months is negative and the HPV test is negative, then the women could return to routine screening. But if the follow-up smear at 6 months is negative and the HPV test is positive, then colposcopy would be the next step. In the event of negative colposcopy, smears would be repeated at 12 months and annually for 5–10 years as is currently recommended. Additional advantages of prospective studies would be the use of a fresh sample for HPV testing and the use of a more suitable HPV DNA detecting method, such as Hybrid Capture II, which can also detect more high-risk HPV types than the method we used.
|
Footnotes |
|---|
Received March 12, 2001. Received in revised form June 18, 2001. Accepted June 28, 2001.
|
REFERENCES |
|---|
|
TOP
ABSTRACTMATERIAL AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
2. Murdoch JB, Morgan PR, Lopes A, Monaghan JM. Histologic incomplete excision of CIN after large loop excision of the transformation zone (LLETZ) merits careful follow up, not retreatment. Br J Obstet Gynaecol 1992;99: 990–3.[Medline]
3. Gardeil F, Barry-Walsh C, Prendiville W, Clinch J, Turner M. Persistent intraepithelial neoplasia after excision for cervical intraepithelial neoplasia grade III. Obstet Gynecol 1997;89:419–22.[Abstract]
4. Flannelly G, Langhan H, Jandial L, Mann E, Campbell M, Kitchener H. A study of treatment failures following large loop excision of the transformation zone for the treatment of cervical intraepithelial neoplasia. Br J Obstet Gynaecol 1997;104:718–22.[Medline]
5. Mohamed-Noor K, Quinn MA, Tan J. Outcomes after surgical cold knife conization with complete and incomplete excision of abnormal epithelium: A review of 699 cases. Gynecol Oncol 1997;67:34–8.[Medline]
6. Paraskevaidis E, Lolis ED, Koliopoulos G, Alamanos Y, Fotiou S, Kitchener HC. Cervical intraepithelial neoplasia outcomes after large loop excision with clear margins. Obstet Gynecol 2000;95:828–31.
7. De Roda Husman A, Walboomers JM, Van de Brule AJ, Meijer CJ, Snijders PJ. The use of general primers GP5 and GP6 elongated at their 3' ends with adjacent highly conserved sequences improves human papillomavirus detection by PCR. J Gen Virol 1995;76:1057–62.
8. Sacket DL, Haynes RB, Guyatt GH, Tugwell P. Clinical epidemiology. A basic science for clinical medicine. 2nd Edition. Toronto: Little, Brown and Co., 1991.
9. Chua KL, Hjerpe A. Human papillomavirus analysis as a prognostic marker following conization of the cervix uteri. Gynecol Oncol 1997;66:108–13.[Medline]
10. Bollen LJM, Tjong-A-Hung SP, van der Velden J, Mol BW, ten Kate FWJ, ter Schegget J, et al. Prediction of recurrent and residual dysplasia by human papillomavirus detection among patients with abnormal cytology. Gynecol Oncol 1999;72:199–201.[Medline]
This article has been cited by other articles:
|
|
 |
|
  D. Solomon Chapter 14: Role of Triage Testing in Cervical Cancer Screening J Natl Cancer Inst Monographs, June 1, 2003; 2003(31): 97 - 101. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. H. Kaufman Is There a Role for Human Papillomavirus Testing in Clinical Practice? Obstet. Gynecol., November 1, 2001; 98(5): 724 - 725. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
