Atypical squamous cells of undetermined significance (ASC-US) smears comprise approximately half of all abnormal Pap test diagnoses in American women undergoing cervical cytologic screening.1 Although serious disease occurs relatively rarely (estimates of moderate to severe cervical intraepithelial neoplasia (CIN) range from 5% to 17% among women with this cytologic diagnosis2,3), ASC-US remains the most common Pap test abnormality preceding a diagnosis of CIN 2 or worse.4 Because of the lack of evidence surrounding the optimal management of women with mild cytologic abnormalities (including low-grade squamous intraepithelial lesions [LSIL] as well as ASC-US), the National Cancer Institute implemented a multicenter randomized trial: The ASCUS/LSIL Triage Study.5 The ASCUS/LSIL Triage Study trial was designed to compare the sensitivity and specificity of 3 different management strategies (colposcopy, repeat cytology, and oncogenic HPV testing) in detecting severe histologic disease (CIN 3) in women with mild cytologic abnormalities.
Although data initially seemed to support the role of oncogenic HPV testing in women with LSIL,6 the March 2000 publication from the ASCUS/LSIL Triage Study trial demonstrated the opposite. In this analysis, oncogenic HPV was confirmed in 83% of the population with LSIL. Based on the high prevalence of infection, the costs of HPV testing in women with LSIL were determined to outweigh the savings gained from avoiding colposcopy in a minority of cases. For ASC-US, however, the addition of reflex HPV testing resulted in a referral rate of only 56%.1,3 The lower prevalence of infection, combined with the high sensitivity of oncogenic HPV typing, led the authors of the 2001 Consensus Guidelines2 to recommend reflex HPV testing as the preferred method for providers using liquid-based cytologic screening methods.
What remained undetermined was whether the addition of reflex testing to all ASC-US smears is warranted. In the ASCUS/LSIL Triage Study, HPV typing resulted in maintenance of excellent sensitivity and acceptable rates of referral to colposcopy, but only for those aged 29 years and older. For younger women, particularly adolescents, the addition of HPV typing remains unclear. For example, women younger than the age of 18 years were not considered in the ASCUS/LSIL Triage Study, nor was the data stratified for younger women with LSIL (the authors considered women aged 18–28 as a whole). We hypothesize that rates of oncogenic HPV positivity will be inversely related to a woman's age and that, if rates approach 80% in the youngest women, the usefulness of HPV testing in this population becomes questionable. As secondary outcomes, we planned to evaluate other risk factors traditionally associated with cervical neoplasia (such as smoking) to see whether they were associated with the patient's HPV status.
In July of 2002, HPV reflex testing for all women with ASC-US was added as part of the standard of care for women undergoing cervical cancer screening at the Women's Primary Care Center of Women and Infants’ Hospital. Of note, the Women's Primary Care Center serves a large urban, racially and ethnically diverse (25% Latina, 20% African-American, 50% Caucasian) population, more than 70% of whom are eligible for state assistance. We conducted a retrospective cohort study of the 601women diagnosed with ASC-US between July 2002 and February 2004. Patients were identified through a pathology database, and medical chart review was subsequently performed by the study authors. Women and Infant's Institutional Review Board approved the use of the database and medical charts for the purposes of this study.
Of the 601 women identified, 527 (88%) had HPV results available and form the study population. To be included, then, all 527 had had an ASC-US smear in the study period and had oncogenic HPV DNA typing performed. The 71 women without HPV results were analyzed separately. Of note, women with atypical squamous cells, cannot exclude high-grade squamous intraepithelial lesions, were not included in these results. Women without an intact cervix (n = 3) were also excluded from this analysis.
All cervical Pap smears were obtained using fluid-based, thin-layer cytology and were generated from PreservCyt (Cytyc Corporation, Marlborough, MA) samples by use of a semiautomated processor (ThinPrep 3000; Cytyc Corporation). All specimens were evaluated by Women and Infants’ Hospital cytotechnicians and cytopathologists and classified according to Bethesda II terminology. Oncogenic HPV typing was then performed on the residual PreservCyt supernatant from all ASC-US smears. Specimens for Hybrid Capture II (Digene Corporation, Gaithersburg, MD) HPV DNA testing were denatured and hybridized with a mixture of ribonucleic acid (RNA) probes for oncogenic HPV types 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, and 68. The Hybrid Capture II assay is a sandwich capture molecular hybridization assay that uses chemiluminescent detection. The intensity of light measured is proportional to the amount of target DNA in the specimen. A positive test was defined as a relative light unit measurement of 1 pg/mL HPV DNA or greater or approximately 5,000 genomic equivalents of HPV DNA per test. Samples with less than 1 pg/mL oncogenic HPV DNA were considered negative.
Our primary outcome was the age-related prevalence of oncogenic HPV subtypes in women with ASC-US. A sample size calculation was done before study initiation. Assuming an age-dependent difference in the rate of oncogenic HPV infection (75% in women aged less than 18 years compared with 55% in women aged older than 25 years), we calculated a required sample of size of 196 women (98 in each age category) to have 80% power (with α = .05) to detect such a difference. Due to the probability of inconsistent data collection by providers (estimated at 25%), we projected the final sample size required to be 245.
Other information extracted from patients’ medical records included certain patient characteristics historically associated with HPV infection and cervical neoplasia. Because we were limited to chart review, only those characteristics consistently collected were included in the final analysis. In addition to age, then, data collection was possible for the following: race or ethnicity, marital status, insurance status, gravidity and parity, tobacco use, contraception, history and type of sexually transmitted diseases, and history of previous abnormal Pap test or prior treatment for cervical neoplasia. Information on age at first intercourse and number of lifetime sexual partners was missing in 34% and 28% of the study cohort, respectively, and was therefore not included in the analysis.
Descriptive statistics were computed. Differences in characteristics between women with oncogenic HPV were compared with women without oncogenic HPV infection by Student t test for continuous variables and χ2 or Fisher exact test for categorical variables. Multivariate binary regression analyses were used to estimate adjusted relative risks of oncogenic HPV infection by risk factors traditionally associated with cervical neoplasia, including age and current smoking. Stratified analyses were performed to evaluate differences in associations of oral contraceptive pill use and HPV among younger compared with older women. All analyses were performed using Stata 8 (StataCorp LP, College Station, TX).
Between July of 2002 and February of 2004, 601 women seen at the Women's Primary Care Center for cervical cancer screening were found to have ASC-US. Three women were subsequently found to have undergone vaginal screening and were excluded from the analysis. Of the remaining 598, 71 (12%) did not have oncogenic HPV results available for 1 or more of the following reasons: 1) provider failure to order HPV testing in the presence of ASC-US on the initial request form, 2) patient's inability to pay, and 3) residual PreservCyt supernatant inadequate to perform testing. In general, women who were not tested for HPV did not differ from those who were tested save for 2 factors: they were more likely to be sexually inactive and to have had a history of an abnormal Pap test (data not presented).
Of the 527 eligible women, 357 (68%, 95% confidence interval [CI] 64–72%) tested positive for oncogenic HPV infection. Women with evidence of HPV were significantly more likely to be younger (mean age 25.0 years compared with 27.8, P < .01) and never married (82% compared with 73%, P = .02). When stratified by age, adolescent women (aged 19 and younger), compared with women older than the age of 25 years, were more likely to harbor HPV infection (77% compared with 58%, P < .01). On the basis of gravidity and parity, racial or ethnic distribution, insurance status, current smoking status, and prior history of abnormal smears or treatment for neoplasia, the populations were similar (Table 1). Of the women who tested positive for HPV, 46% were not currently using any form of contraception, compared with 38% of those who tested negative (P = .07). Use of tubal ligation (12% among the HPV-negative group compared with 7% among those positive, P = .05) or oral contraceptives (19% compared with 14%, P = .15) was more common among the HPV-negative women, whereas condom use did not differ significantly between the 2 groups. A history of sexually transmitted infections, particularly Chlamydia trachomatis (36% of HPV-positives compared with 26%, P = .02), was seen more often among the HPV-positive patients (Table 2).
In the final adjusted analysis presented in Table 3, age younger than 20 years (relative risk [RR] 1.30, 95% CI 1.14–1.49), current smoking (RR 1.14, 95% CI 1.03–1.25, and a history of C. trachomatis (RR 1.20, 95% CI 1.09–1.32) were all associated with the presence of oncogenic HPV infection in women with ASC-US. On the other hand, a history of treatment for cervical neoplasia (RR 0.85, 95% CI 0.67 –1.08) and current oral contraceptive use (RR 0.86, 95% CI 0.72 –1.03) exhibited a protective effect. With respect to oral contraceptive use, the observed effect was greatest among adolescents (RR 0.57, 95% CI 0.31–1.03), data not presented). Of note, we were concerned that a history of an abnormal Pap test or a history of previous treatment for neoplasia may have exerted an age-dependent effect on our results. In a separate analysis (data not presented), a history of an abnormal Pap test was not found to be associated with the patient's current HPV status when stratified by age. A history of treatment was associated with an increased risk of having HPV infection at the time of a subsequent ASC-US smear only among women aged 20–25 years (Adjusted RR 1.28, 95% CI 1.12–1.46). The small number of adolescent women who had previously undergone treatment precluded assessment of the effect of prior treatment on this population.
In an analysis of data from the ASCUS/LSIL Triage Study, Sherman et al3 proposed that by considering age or viral load, colposcopic triage might be improved. In fact, rates of oncogenic HPV were dramatically lower in women aged older than 29 years compared with women aged 18–22 years (71% positive compared with 31%). Similar findings emerged in the LSIL data from the ASCUS/LSIL Triage Study. Compared with younger women, women aged 30 years or older were significantly less likely to be oncogenic HPV positive (73% compared with 85%, adjusted OR 0.5, 95% CI 0.3–0.9). Based on this lower rate of oncogenic HPV infection, the ASCUS/LSIL Triage Study authors suggested the need for further studies to clarify the role of HPV testing in the management of women aged older than 30 years.6
Current recommendations for the management of women with ASC-US call for the addition of reflex HPV DNA testing when liquid-based cytology is used.2 HPV testing, in this context, demonstrates excellent sensitivity,1,7 and in women aged older than 30 years, is cost-effective.8 Because approximately 50% of women with ASC-US will test positive for oncogenic HPV infection, rates of referral to colposcopy remain acceptable. Management, however, does differ based on certain patient characteristics, such as age (postmenopausal women with evidence of atrophy may receive a course of intravaginal estrogen) and presence of immunosuppression (immediate colposcopy is advised). The authors of the 2001 Consensus Guidelines do not, however, specifically address adolescents with ASC-US, although less invasive triage strategies, such as repeat cytology or HPV testing in 1 year, are recommended for teenagers with LSIL.
Oncogenic HPV infection clearly peaks in young women.9 Among women with normal cytology, incident HPV infection rates seem to be highest in the youngest women.10,11 In a recent study of more than 2,000 women with ASC-US or LSIL, Sarode et al12 found an inverse relationship between age and HPV status with 95% of adolescents and 77% of women aged 20–29 years demonstrating evidence of oncogenic infection. Our findings clearly corroborate such reports: of more than 500 women diagnosed with ASC-US at our institution, nearly 80% of the adolescents and 72% of the young women aged 20–25 years tested positive for high-risk HPV.
Stratification by factors associated with HPV acquisition or persistence did not for the most part prove to be helpful in discriminating those at greatest risk for infection. For example, the risks associated with cigarette smoking or a history of chlamydial infection were only modestly elevated, while the protective effect of oral contraceptive pill use was seen only among adolescents. Similar findings have been previously demonstrated. For example, a history of a sexually transmitted infection has been demonstrated to be a risk factor for HPV infection among both young11 and middle-aged women.13 With respect to oral contraceptive use, however, the data have been less clear. In the study by Moscicki et al11 of young women, the current use of oral contraceptives had a significantly protective effect on HPV acquisition, which the author found interesting in the context of other data demonstrating an association with prolonged use of oral contraceptives and HPV infection14 as well as the development of squamous intraepithelial lesions15 and cancer.16 Although possible mechanisms by which a protective effect occurs have been proposed (eg, hormonal effects on immune regulation11 or the presence of neutralizing antibodies17), the significance of this association and the reasons for why studies of different designs result in inconsistent results with respect to oral contraceptive use remain unclear.
This study has several limitations. Because of its retrospective nature, inferences regarding causality are limited. Second, the characteristics of the population studied limit the generalizability of our results to all colposcopy populations. In addition, information on several risk factors (eg, number of sexual partners) associated with HPV and CIN were not consistently collected. Such omissions occurred primarily among women not seen in colposcopy, and thus limited our ability to make certain comparisons between HPV-positive and -negative women. Finally, providers may not have recorded information on dual contraception, although Ho et al17 found that lack of condom use did not alter the relationship between use of oral contraceptives and a reduced risk of HPV infection.
The strengths of this study include its large sample size, diverse patient population, and ability to assess multiple secondary outcomes. Among our population of women with ASC-US, the high prevalence of oncogenic HPV infection in young women, particularly adolescents, suggests a need to rethink the management of this population. Because the rates of clinically significant disease (ie, CIN 2 or worse) have been demonstrated to be equivalent in women with HPV-positive ASC-US or LSIL,18 and due to the pervasiveness of this infection in adolescents with ASC-US as seen in numerous reports including the present study, alternative management schemes, similar to those proposed for adolescents with LSIL, should be considered.19
Indeed, in 2002, the American Cancer Society,20 and subsequently the American College of Obstetricians and Gynecologists,21 changed their recommendations for when cervical cancer screening should be initiated by suggesting that screening begin approximately 3 years after initiation of sexual intercourse, but no later than age 21 years. The impact of such guidelines on adolescent women might be expected to result in a diminution of the problem we present, yet in 2003, more than 50% of female 11th graders reported a history of sexual intercourse.22 Although fewer adolescents will be screened according to these new guidelines, a significant number will still have a Pap test before the age of 20 years, and rates of HPV infection will most likely be even greater than what we report. With respect to our findings, then, although colposcopy would remain an option for adolescents with high-risk HPV infection, other far less invasive schemes such as repeat cytology or HPV DNA testing would be far more acceptable to this population and the providers who care for them.
1. Solomon D, Schiffman M, Tarone R. Comparison of three management strategies for patients with atypical squamous cells of undetermined significance: baseline results from a randomized trial. J Natl Cancer Inst 2001;93:293–9.
2. Wright TC, Cox JT, Massad LS, Twiggs LB, Wilkinson EJ. 2001 consensus guidelines for the management of women with cervical cytological abnormalities. JAMA 2002;287:2120–30.
3. Cox JT; Atypical Squamous Cells of Undetermined Significance/Low-Grade Squamous Intraepithelial Lesion Triage Study Group. Sherman ME, Schiffman M, Effects of age and human papilloma viral load on colposcopy triage: data from the randomized Atypical Squamous Cells of Undetermined Significance/Low-Grade Squamous Intraepithelial Lesion Triage Study (ALTS). J Natl Cancer Inst 2002;94:102–7.
4. Kinney WK, Manos MM, Hurley LB, Ransley JE. Where's the high-grade cervical neoplasia? The importance of the minimally abnormal Papanicolaou diagnosis. Obstet Gynecol 1998;91:973–6.
5. Schiffman M, Adrianza ME. ASCUS-LSIL triage study: design, methods and characteristics of trial participants. Acta Cytol 2000;44:726–42.
6. Human papillomavirus testing for triage of women with cytologic evidence of low-grade squamous intraepithelial lesions: baseline data from a randomized trial. The Atypical Squamous Cells of Undetermined Significance/Low-Grade Squamous Intraepithelial Lesions Triage Study (ALTS) Group. J Natl Cancer Inst 2000;92:397–402.
7. Arbyn M, Buntinx F, Van Ranst M, Paraskevaidis E, Martin-Hirsch P, Dillner J. Virologic versus cytologic triage of women with equivocal Pap smears: a meta-analysis of the accuracy to detect high-grade intraepithelial neoplasia. J Natl Cancer Inst 2004;96:280–93.
8. Goldie SJ, Kim JJ, Wright TC. Cost-effectiveness of human papillomavirus DNA testing for cervical cancer screening in women aged 30 years or more. Obstet Gynecol 2004;103:619–31.
9. Kulasingam SL, Hughes JP, Kiviat NB, Mao C, Weiss MS, Kuypers JM, et al. Evaluation of human papillomavirus testing in primary screening for cervical abnormalities: comparison of sensitivity, specificity and frequency of referral. JAMA 2002;288:1749–57.
10. Sellors JW, Karwalajtys TL, Kaczorowski J, Mahony JB, Lytwyn A, Chong S, et al. Incidence, clearance and predictors of human papillomavirus infection in women. CMAJ 2003;168:421–5.
11. Moscicki A-B, Hills N, Shiboski S, Powell K, Jay N, Hanson E, et al. Risks for incident human papillomavirus infection and low-grade squamous intraepithelial lesion development in young females. JAMA 2001;285:2995–3002.
12. Sarode VR, Werner C, Gander R, Foster B, Fulmer A, Sabaoorian MH, et al. Reflex human papillomavirus DNA testing on residual liquid-based (TPPTTM) cervical samples: focus on age-stratified clinical performance. Cancer (Cancer Cytopathol) 2003;99:149–55.
13. Munoz N, Kato I, Bosch FX, Eluf-Neto J, de Sanjose S, Ascunce N, et al. Risk factors for HPV DNA detection in middle-aged women. Sex Transm Dis 1996;23:504–10.
14. Kjaer SK, van den Brule AJC, Bock JE, Poll PA, Engholm G, Sherman ME, et al. Determinants for genital human papillomavirus (HPV) infection in 1000 randomly chosen young Danish women with normal Pap smear: are there different risk profiles for oncogenic and nononcogenic HPV types? Cancer Epidemiol Biomarkers Prev 1997;6:799–805.
15. Irwin KL, Rosero-Bixby L, Oberle MW, Lee NC, Whatley AS, Fortney JA, et al. Oral contraceptives and cervical cancer risk in Costa Rica: detection bias or causal association? JAMA 1988;259:59–64.
16. Smith JS, Green J, Berrington de Gonzalez A, Appleby P, Peto J, Plummer M, et al. Cervical cancer and use of hormonal contraceptives: a systematic review. Lancet 2003;361:1159–67.
17. Ho GYF, Studentsov Y, Hall CB, Bierman R, Beardsley L, Lempa M, et al. Risk factors for subsequent cervicovaginal human papillomavirus (HPV) infection and the protective role of antibodies to HPV-16 virus-like particles. J Infect Dis 2002;186:737–42.
18. Solomon D for the ASCUS-LSIL Triage Study (ALTS) Group. Cox JT, Schiffman M, Prospective follow-up suggests similar risk of subsequent cervical intraepithelial neoplasia grade 2 or 3 among women with cervical intraepithelial neoplasia grade 1 or negative colposcopy and directed biopsy. Am J Obstet Gynecol 2003;188:1406–12.
19. Wright TC, Cox JT, Massad LS, Twiggs LB, Wilkinson EJ. 2001 consensus guidelines for the management of women with cervical cytological abnormalities. JAMA 2002;287:2120–9.
20. Saslow D, Runowicz CD, Solomon D, Moscicki A-B, 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.
21. Clinical management guidelines for obstetrician-gynecologists: cervical cancer screening. ACOG Practice Bulletin No. 45. American College of Obstetricians and Gynecologists. Obstet Gynecol 2003;102:417–27.
22. Grunbaum JA, Kann L, Kinchen S, Ross J, Hawkins J, Lowry R, et al. Youth risk behavior surveillance—United States, 2003. MMWR Surveill Summ 2004;53(SS-2):1–96.
© 2005 The American College of Obstetricians and Gynecologists
This article has been cited