Objective: To estimate the risk of and risk factors for progression among human immunodeficiency virus (HIV)-seropositive women with abnormal cervical cytology but negative colposcopy.
Methods: In a prospective cohort study, 391 HIV-seropositive and 103 seronegative women with cervical cytology read as atypical squamous cells (ASC) or low-grade squamous intraepithelial lesion (LSIL) but negative colposcopy were followed up for a mean of 4.0 years with cytology at 6-month intervals. Colposcopy was prescribed for any epithelial abnormality.
Results: Progression to CIN2, CIN3, high-grade SIL/severe dysplasia, or cancer occurred in 47 (12%) HIV-seropositive women and 4 (4%) HIV-seronegative women (P = .02). Progression to CIN1 was seen in an additional 12 HIV-seropositive women and 1 seronegative woman. In multivariate analysis, high-risk but not low-risk HPV detection (hazard ratio [HR] 2.46–95% confidence interval [CI] 1.18–5.12, P = .02 for high risk, HR 1.41, 95% CI 0.62–3.21, P = .42 for low risk), satisfactory colposcopy (HR 2.01, 95% CI 1.11–3.65, P = .02), and non-Hispanic African-American ethnicity (HR 5.08, 95% CI 1.72–14.98, P = .003) were the only factors associated with progression, while HIV serostatus was marginally significant (HR 2.53, 95% CI 0.85–7.50, P = .09).
Conclusion: Human immunodeficiency virus–seropositive women with negative colposcopy after borderline cytology face a higher risk of progression than seronegative women, but the absolute risk is low and becomes nonsignificant after controlling for HPV risk type, ethnicity, and colposcopic findings. Observation is appropriate.
Level of Evidence: II-2
Despite negative colposcopy, human immunodeficiency virus-infected women with borderline cervical cytology remain at moderate risk for progression to high-grade cervical disease.
From the 1Southern Illinois University School of Medicine, Springfield, Illinois; 2University of Illinois, Chicago, Illinois; 3Albert Einstein College of Medicine, Bronx, New York; 4National Institute of Child Health and Human Development, Bethesda, Maryland; 5Montefiore Medical Center, Bronx, New York; 6University of California, San Francisco, California; 7Johns Hopkins School of Public Health, Baltimore, Maryland; 8State University of New York, Brooklyn, New York; 9Georgetown University School of Medicine, Washington, DC; and 10Keck School of Medicine, University of Southern California, Los Angeles, California.
The Women's Interagency HIV Study is funded by the National Institute of Allergy and Infectious Diseases, with supplemental funding from the National Cancer Institute, the National Institute of Child Health and Human Development, the National Institute on Drug Abuse, the Agency for Health Care Policy and Research, the National Center for Research Resources, and the Centers for Disease Control and Prevention, grants U01-AI-35004, U01-AI-31834, U01-AI-34994, U01-AI-34989, U01-HD-32632 (NICHD), U01-AI-34993, U01-AI-42590, M01-RR00079, and M01-RR00083. Human papillomavirus DNA testing was supported by Public Health Service Grant CA85178–01 (National Cancer Institute).
Corresponding author: L. Stewart Massad, MD, Department of Obstetrics & Gynecology, Southern Illinois University, PO Box 19640, Springfield, IL 62794-9640; e-mail: LSMASSAD@ameritech.net.
Women infected with the human immunodeficiency virus (HIV)-1 are at increased risk for the development of cervical intraepithelial neoplasia (CIN).1–5 As a reflection of this increased risk for CIN, cervical cancer screening recommendations are more intensive for HIV-infected women than for uninfected women: women with HIV should have 2 Papanicolaou tests after HIV diagnosis, followed by annual screening,6 rather than the annual, biennial, or triennial screening recommended for women in the general population.7 However, most cervical abnormalities in HIV-seropositive women are low grade and may represent only opportunistic infection with human papillomavirus (HPV) rather than premalignant lesions. The long-term cancer risk among women with HIV is unclear but seems to be low among women participating in screening.8 Data supporting an increased risk for progressive cervical disease among women with HIV would in turn support the hypothesis that borderline abnormalities in women with HIV have premalignant potential and require intensive surveillance.
Colposcopy is the first step in the management of HIV-infected women with abnormal cytologic screening results.9 Women with high-grade cervical disease identified at colposcopy receive destructive treatment, whereas women with low-grade disease may be either observed or treated.10 However, data accumulated during the past decade suggest that HIV-uninfected women who have no disease identified at colposcopy nevertheless remain at substantial risk for later development of CIN.11,12 Recently, results from the national randomized trial of management options for HIV-uninfected women with atypical squamous cells or low-grade squamous lesions indicated that women with negative colposcopy but detectable high-risk HPV, like women with CIN1, face an approximately 10% risk of developing CIN3 during 2 years of follow-up,13 and observation with cytology at 6-month intervals is sufficient to identify progression. Risk may be higher in women with HIV, but the level of risk has not been defined. We set out to determine whether high-grade CIN is more frequent among HIV-infected than uninfected women with negative colposcopy after abnormal cytology.
PATIENTS AND METHODS
This investigation was part of the Women's Interagency HIV Study (WIHS), an ongoing multicenter prospective cohort study of the natural history of HIV infection and related health conditions among HIV-seropositive women and at-risk HIV-uninfected comparison women. The protocols, recruitment processes, procedures, and baseline results of the WIHS have been previously described; seropositive WIHS participants are representative of U.S. women with HIV.14 WIHS enrollment began in 1994 at 6 study consortia (Brooklyn, Bronx, Chicago, Los Angeles, San Francisco, and Washington, DC), enrolling 2,628 women. Written informed consent was obtained after approval by local human subjects committees. Follow-up continues, but this analysis includes follow-up information obtained before October 1, 2003. At baseline and every 6 months, participants had medical, social, and sexual histories taken by trained interviewers. They then underwent physical examination and conventional cervical cytology. Smears were interpreted at Dianon Systems (New York, NY) using the 1991 criteria of the Bethesda System for cervicovaginal cytologic diagnosis.15 At baseline, cervicovaginal saline lavage was obtained for polymerase chain reaction (PCR) typing for HPV by MY09/MY11 consensus primer PCR amplification, followed by hybridization with consensus and HPV type-specific probes.16 Beta-globin gene amplification during PCR was used to assess specimen adequacy; β-globin–negative specimens were excluded. Results were not used in management. HPV types 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 68, and 73 were considered to confer high oncogenic risk; all others were considered to confer low oncogenic risk. Colposcopy was prescribed for women with all grades of cytologic epithelial abnormality, including atypia, although some sites initially also performed colposcopy for research purposes on women with normal cytology; only the former were studied. Endocervical curettage was required by protocol for women with unsatisfactory colposcopy and LSIL cytology but was discretionary for other participants. Women with negative colposcopy were identified. Negative colposcopy was defined as an examination without evidence of low- or high-grade neoplasia by colposcopic impression or biopsy. Women with negative colposcopy but with CIN or cancer on endocervical curettage were excluded. Cytologic, colposcopic, and histologic findings were entered into a central database. Biopsy slides were not centrally reviewed.
After identifying women with negative colposcopy, we excluded those who had prior hysterectomy, those with vulvar or vaginal colposcopy, those without prior cytologic abnormality, and those without follow-up. We also excluded those with antecedent Pap tests read as high-grade squamous intraepithelial lesion (SIL) or severe dysplasia or 2 prior smears read as high-grade SIL or moderate dysplasia, because diagnostic excision rather than observation is considered standard for such women.9 Relatively few HIV-seronegative women were included as controls, because abnormal cytology was a requirement for analysis and is markedly less common among these women than among HIV-seropositive women3,17
Demographic and medical characteristics at index colposcopy were tabulated. Women were followed up cytologically at 6-month intervals, with colposcopy repeated for epithelial abnormality of any grade. Cervical disease treatment decisions were not specified by study protocols. Women were followed up until the identification of high-grade cervical disease (CIN2, CIN3, or cancer documented by biopsy, or cytology showing high-grade SIL or severe dysplasia once or high-grade SIL or moderate dysplasia twice), until cervical disease treatment, or until the end of the study period.
HIV-seropositive and seronegative women were compared using the χ2 test for categorical variables and the Wilcoxon 2-sample test for continuous variables. Incidence was calculated as the number of cervical neoplasms diagnosed during the study divided by the person-years at risk. Kaplan-Meier curves illustrating cervical disease-free survival by HIV status were compared using the log-rank test. Cox proportional-hazards models were used to adjust for confounders in multivariate analyses; potential confounders included were HIV status, colposcopy adequacy, Pap test result, HPV risk type, age, ethnicity, smoking status, number of sex partners, and parity, with CD4 lymphocyte count and HIV RNA level included for HIV-seropositive women. Except where noted, time varying factors were assessed at the visit of the index colposcopy.
We identified 1,421 women whom clinicians deemed to have negative colposcopy. Of these, 315 nevertheless had CIN on biopsy and were excluded, as were 2 with cancer and 179 with histologic changes consistent with HPV infection, including condyloma and koilocytosis. Women also were excluded if they had prior hysterectomy (n = 72), noncervical colposcopy (n = 13), no prior cytologic abnormality (n = 235), dysplasia at or before negative colposcopy (n= 52), prior high-grade SIL cytology (n = 5), or no follow-up (n = 47). One woman was excluded because she had high-grade SIL at the same visit as her negative colposcopy. Finally, 5 women were excluded because their HPV results were not available. One woman who seroconverted after HIV infection was also excluded.
After these exclusions, 494 women with negative cervical colposcopy after prior abnormal cytology were available for analysis (391 [79%] HIV-seropositive and 103 [21%] seronegative). HIV-seropositive women were followed up for 1,591 person-years, with a mean follow-up of 4.0 years, and seronegative women were followed up for 378 person-years, with a mean follow-up of 3.7 years. Maximum follow-up was 8.6 years.
The demographic and medical characteristics of HIV-seropositive and -seronegative women are shown in Table 1. Median age was 37.1 years among HIV-seropositive women and 33.8 years among seronegative women (P < .001). HIV-seronegative women had more sexual partners in the 6 months before colposcopy (median among seronegative women 1.0, mean 2.14, range 0–50, compared with 1.0 among seropositive women, mean 1.31, range 0–100, P < .001 by Wilcoxon 2-sample test). Satisfactory colposcopy was more common in HIV-seronegative women than seropositive women (76% compared with 56%, P < .001). HIV-seropositive women had fewer partners in the 6 months before index negative colposcopy (P < .001), yet were more likely to have HPV than seronegative women (P < .001). Among seropositive women, median CD4 count was 342 cells/μL, and median HIV viral load was 12,000 copies/μL.
Among women with negative colposcopy, progression to CIN2, CIN3, high-grade SIL or severe dysplasia, or cancer occurred in 51 (10%). Progression to high-grade disease occurred in 47 (12%) HIV-seropositive women and 4 (4%) HIV-seronegative women (P = .02). Mean time to progression among seropositive women was 2.7 years (median 2.4, range 0.4–7.5 years); mean time to progression among seronegative women was 1.1 year (median 1.0, range 0.4–2.0 years). This is shown graphically in Figure 1. The overall incidence of high-grade disease in the first 2 years of follow-up was 3.1% (25/802 person-years) and in the latter years was 2.2% (26/1167 person-years) (relative risk 1.4, 95% confidence interval [CI] 0.8–2.5, P = .23). In addition, all incident high-grade disease in seronegative women occurred in the first 2 years of follow-up (4/160 years, 2.5%). In contrast, 21 cases in seropositive women occurred in the first 2 years of follow-up (21/642 person-years, 3.3%), with 26 cases in the later years (14/949 person-years, 1.5%). Rates of progression were 5.8% in Brooklyn, 2.2% in Washington, DC, 1.1% in Los Angeles, 2.3% in San Francisco, and 3.0% in Chicago; with the rate in Brooklyn being significantly higher than in other sites combined (2.0%, P < .001). Progression to CIN1 was seen in an additional 12 HIV-seropositive women and 1 seronegative woman.
Table 2 highlights risk factors for the development of high-grade cervical disease after negative colposcopy. HIV-seropositive women had almost 3 times the risk of incident high-grade neoplasia as compared with seronegative women. Detection of high-risk but not low-risk HPV and a prior Pap test read as LSIL rather than atypical squamous cells of undetermined significance were significant predictors of incident high-grade disease. Non-Hispanic African-American women also were more likely to have incident high-grade disease. Among HIV-seropositive women, higher CD4 lymphocyte counts were associated with lower risk of incident high-grade disease.
HIV seroprevalence was not associated with progression in multivariate analysis, which included factors significant in univariate analysis. Detection of high-risk but not low-risk HPV, satisfactory colposcopy, and non-Hispanic African-American ethnicity were the only factors associated with progression to high-grade disease after negative colposcopy (Table 3). Table 3 also shows similar findings from an analysis limited to HIV-seropositive women: non-Hispanic African-American women and women with high-risk but not low-risk HPV were more likely to progress, while those with CD4 counts in the mid but not upper range were less likely to progress. In a model incorporating only CD4 count, HIV viral load, HPV risk type, and cytology grade before colposcopy, women with CD4 counts of 200–500/μL were less likely to progress (HR 0.40 compared with women with CD4 counts < 200/μL, 95% CI 0.21–0.78, P = .008), whereas the lower progression risk among women with CD4 counts over 500/μL was not significant (HR= 0.70, 95% CI = 0.31–1.62, P = .41). Detection of an oncogenic HPV also was associated with higher progression risk (HR 2.65, 95% CI 1.25–5.60, P = .01). Other factors in this model were not significantly correlated with progression risk.
Among women with HIV and borderline cervical cytology results, the risk of progression to high-grade cervical disease is significantly higher than that among HIV-seronegative women in WIHS, although this risk reflects differences in HPV type, ethnicity, and colposcopic findings. This supports the concept that in HIV-infected women borderline cervical changes do not merely reflect very common opportunistic HPV infections but rather have true premalignant potential. Continuing surveillance for progression is warranted even when colposcopy is initially reassuring.
However, the absolute risk of progression over a median of 4 years of follow-up was only 12%. This low risk of progression is similar to that reported in a national prospective trial of women in the general population with HPV-positive atypical smears, low-grade SIL, and CIN1.13 However, no cancers were missed in our study when HIV-seropositive women with negative colposcopy were observed. Thus, HIV-seropositive women with cervical cytology read as atypical squamous cells of undetermined significance or low-grade SIL can be managed according to guidelines established for seronegative women. According to these guidelines, diagnostic excision is not required to identify high-grade disease, and monitoring with cytology 6 and 12 months after colposcopy should be sufficient, provided colposcopy is repeated should Pap tests show any grade of persisting epithelial abnormality.8
Although HIV-seropositive women had a higher risk of progression to high-grade cervical disease than seronegative comparison women in univariate analysis, HIV serostatus was not a significant correlate of progression after controlling for ethnicity, HPV risk type, and ability to visualize the squamocolumnar junction at colposcopy. Inability to control the expression of oncogenic HPV is critical to defining progression risk in HIV-seropositive women. HIV-seropositive women with persistent high-risk HPV types may benefit from cytologic assessment at 6-month intervals for longer than 1 year or from surveillance colposcopy. Nonvisualized portions of the transformation zone within the cervix may harbor occult high-grade disease that becomes evident over time. Unsatisfactory colposcopy may be more common among HIV-seropositive women because they were older, but we cannot exclude the possibility that a higher frequency of unsatisfactory colposcopy reflects scarring after prior treatment that we did not ascertain. Non-Hispanic African-American ethnicity may be a surrogate for biologic factors we did not measure, but was not associated with highly active antiretroviral therapy or with CD4 lymphocyte count.
In this study, HIV-seronegative women who progressed after initial negative colposcopy usually did so within the first 2 years, suggesting that most of these had small lesions that were initially unrecognized. HIV-seronegative women who did not progress during initial follow-up did not progress later. HIV-seropositive women initially progressed to high-grade disease at a rate almost identical to that of seronegative women, suggesting that rates of occult high-grade disease missed by initial colposcopy were similar in the 2 groups. However, in contrast to seronegative women, seropositive women continued to progress to high-grade disease over time. This further supports the concept that high-risk HPV infections persisting as a result of immunodeficiency may lead to oncogenic changes, and continuing cytologic monitoring is important even after initial negative results.
We did not include women with cytology read as high-grade SIL or cancer in our analysis. We have previously shown that these results are uncommon among women HIV-seropositive women.3 When they are reported, management should be identical to that in HIV-seronegative women: diagnostic excision should follow even when colposcopy fails to identify a lesion.9
Our findings have several limitations. We had limited power to detect a truly higher rate of progression after negative colposcopy among women with HIV, and our power to detect significant associations was limited by numbers. However, WIHS is the largest cohort study of women with HIV in the United States, and other studies are unlikely to have greater power. Associations found to be significant, however, should be reliable. We also were limited by our inability to impose centralized training or certification in colposcopy; in consequence, we cannot exclude the possibility that more expert colposcopists might have found high-grade disease in some patients categorized in our study as having negative colposcopy. Nevertheless, our results are similar to those previously reported in HIV-seronegative women, and more expert colposcopy would have resulted in lower progression rates. This would have reinforced our conclusion that despite a significantly increased risk for progression, observation is appropriate for HIV-seropositive women with borderline cytology but negative colposcopy.
1. Wright TC, Ellerbrock TV, Chiasson MA, Van Devanter N, Sun XW. Cervical intraepithelial neoplasia in women infected with human immunodeficiency virus: Prevalence, risk factors, and validity of Papanicolaou smears. New York Cervical Disease Study. Obstet Gynecol 1994;84:591–7.
2. Maiman M, Fruchter RG, Sedlis A, Feldman J, Chen P, Burk RD, et al. Prevalence, risk factors, and accuracy of cytologic screening for cervical intraepithelial neoplasia in women with the human immunodeficiency virus. Gynecol Oncol 1998;68:233–9.
3. Massad LS, Ahdieh L, Benning L, Minkoff H, Greenblatt RM, Watts H, et al. Evolution of cervical abnormalities among women with HIV-1: Evidence from surveillance cytology in the Women's Interagency HIV Study. J Acquir Immune Defic Syndr 2001;27:432–42.
4. Conti M, Agarossi A, Parazzini F, Muggiasca ML, Boschini A, Negri E, et al. HPV, HIV infection, and risk of cervical intraepithelial neoplasia in former intravenous drug abusers. Gynecol Oncol 1993;49:344–8.
5. 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. European Study Group on Natural History of HIV Infection in Women. AIDS 2000;14:1775–84.
6. 1999 USPHS/IDSA Guidelines for the prevention of opportunistic infections in persons infected with the human immunodeficiency virus. MMWR Recomm Rep 1999;48(RR-10):1–82.
7. Saslow D, Runowicz CD, Solomon D, Moscicki AB, Smith RA, Eyre HJ, et al. American Cancer Society guideline for the early detection of cervical neoplasia and cancer. CA Cancer J Clin 2002;52:342–62.
8. Massad LS, Seaberg EC, Watts DH, Hessol NA, Melnick S, Bitterman P, et al. Low incidence of invasive cervical cancer among HIV-infected US women in a prevention program. AIDS 2004;18:109–13.
9. Wright TC, Cox JT, Massad LS, Twiggs LB, Wilkinson EJ; ASCCP-Sponsored Consensus Conference. 2001 consensus guidelines for the management of women with cervical cytological abnormalities. JAMA 2002; 2001;287:2120–9.
10. Wright TC, Cox JT, Massad LS, Carlson J, Twiggs LB, Wilkinson EJ; American Society for Colposcopy and Cervical Pathology. 2001 consensus guidelines for the management of women with cervical intraepithelial neoplasia. Am J Obstet Gynecol 2003;189:295–304.
11. Hellberg D, Nilsson S, Valentin J. Positive cervical smear with subsequent normal colposcopy and histology: Frequency of CIN in a long-term follow-up. Gynecol Oncol 1994;53:148–51.
12. Milne DS, Wadehra V, Mennim D, Wagstaff TI. A prospective follow up study of women with colposcopically unconfirmed positive cervical smears. Br J Obstet Gynaecol 1999;106:38–41.
13. Guido R, Schiffman M, Solomon D, Burke L; ASCUS LSIL Triage Study (ALTS) Group. Postcolposcopy management strategies for women referred with low-grade squamous intraepithelial lesions or human papillomavirus DNA-positive atypical squamous cells of undetermined significance: a two-year prospective study. Am J Obstet Gynecol 2003;188:1401–5.
14. Barkan SE, Melnick SL, Preston-Martin S, Weber K, Kalish LA, Miotti P, et al. The Women's Interagency HIV Study. WIHS Collaborative Study Group. Epidemiol 1998;9:117–25.
15. Kurman RJ, Solomon D, editors. The Bethesda system for reporting cervical/vaginal cytologic diagnoses: definitions, criteria, and explanatory notes for terminology and specimen adequacy. NY: Springer-Verlag; 1994.
16. Palefsky JM, Minkoff H, Kalish LA, Levine A, Sacks HS, Garcia P, et al. Cervicovaginal human papillomavirus infection in Human Immunodeficiency Virus-1 (HIV)-positive and high-risk HIV-negative women. J Natl Cancer Inst 1999;91:226–36.
17. Massad LS, Riester KA, Anastos K. M, Fruchter RG, Palefsky JM, Burk RD, et al. Prevalence and predictors of squamous cell abnormalities in Papanicolaou smears from women infected with HIV-1. Women's Interagency HIV Study Group. J Acquir Immune Defic Syndr 1999;21:33–41.