Anal cancer shares many biologic similarities with cervical cancer, including a strong association with human papillomavirus (HPV) infection (1,2). Although anal cancer is uncommon in the general population, certain groups such as homosexual/bisexual (hereafter referred to as homosexual) men with a history of receptive anal intercourse are at especially high risk (3,4). The estimated incidence of anal cancer in homosexual men before onset of the HIV epidemic was approximately 35/100,000. This incidence is similar to that of cervical cancer in women before introduction of cervical cytology screening and is considerably higher than the current incidence of cervical cancer, which is approximately 8/100,000 (5). It is unknown if the incidence of anal cancer is higher in HIV-positive homosexual men than in HIV-negative homosexual men (6,7).
Invasive cervical cancers arise from squamous intraepithelial lesions (SIL) (8). SIL may also be detected in the anal canal and is likely to represent the precursor to anal cancer. The degree of cervical SIL (CSIL) or anal SIL (ASIL) ranges from low grade to high grade. Highgrade SIL (HSIL) is likely to represent the true invasive cancer-precursor lesion, and low-grade SIL (LSIL) is not believed to progress directly to invasive cancer. However, anal LSIL is clinically important, inasmuch as 62% of HIV-positive and 36% of HIV-negative men with LSIL progressed to HSIL within 2 years (9). In addition, in both the cervix and the anus, "atypical squamous cells of undetermined significance" (ASCUS) may be found on cytologic examination (9,10).
Identification of women with CSIL through routine cervical cytology screening followed by treatment of the lesion has been an important tool to reduce incidence of invasive cervical cancer. No screening program for ASIL exists that is comparable to that in place for CSIL, but it is possible that an ASIL screening program similar to that for CSIL may be of value to prevent anal cancer in high-risk individuals. Such a program requires a thorough understanding of the risk groups that might derive maximum benefit from its implementation. The aim of this study was to describe ASCUS and ASIL in HIV-positive and HIV-negative homosexual men and to characterize risk factors for these lesions.
This study was conducted with the approval of the University of California, San Francisco Committee on Human Research. Participants in this study were recruited between November 1991 and March 1994 from among homosexual men in the San Francisco Men's Health Study (SFMHS) (11) and the San Francisco General Hospital Cohort (SFGH) Study (12). These groups were combined in the late 1980s and some of their prior data were used at baseline in the present study. Another group was recruited at the University of California at San Francisco (UCSF) using newspaper advertisements. After informed consent was obtained, 346 HIV-positive and 262 HIV-negative men were enrolled. Subjects were interviewed in person at baseline regarding their medical history, sexual practices, and substance use. A thorough anal examination was conducted that included insertion of Dacron swabs for anal cytology and HPV testing, and anoscopy using a colposcope as described previously, except that 3%, rather than 5%, acetic acid was used (1,13). Consenting subjects underwent anal biopsy if a lesion was present. If more than one lesion was seen, biopsy specimens were obtained from areas with different colposcopic appearances. Biopsy specimens were fixed in 10% formalin for routine histopathologic examination. During some examinations, lesions were seen but no biopsy specimen was obtained because of subject refusal or medical contraindications such as recent acetylsalicylic acid intake, neutropenia (<1000/mm3), thrombocytopenia (platelet count <65,000/mm3), or concurrent bacterial or viral infection in or around the anus.
Anal cytology and anal histology were evaluated without knowledge of clinical status of the subject, HPV results, or results of other diagnostic procedures. Anal cytologic results were classified as normal, ASCUS, LSIL, or HSIL using current criteria for evaluation of cervical cytology (14). Anal histologic results were classified as normal, atypical, LSIL, or HSIL. If both cytologic and histologic results were available for analysis, a subject's diagnosis was categorized in terms of the most severe result. Subjects were classified as normal if they had normal cytology and either no lesions visible at anoscopy or had a lesion seen that had normal histology. If a lesion was seen at anoscopy but was not biopsied, the cytology result was used for the lesion diagnosis unless the subject had normal cytology. In that case, the diagnosis was considered indeterminate. Cytology was not available for one subject who was classified as normal because he had no visible anal lesions.
HIV status was determined for all subjects using an enzyme-linked immunosorbent assay (ELISA) screening assay. All positive results of ELISA assays were confirmed using Western blot assay. Over the course of this study, different techniques for the measurement of CD4 counts were used. To adjust for differences in these techniques over time, calibration curves that compared the results of the different methods were computed using the same samples, and appropriate adjustments were made. For HIV-positive subjects, CD4 levels were classified as >500/mm3, 200 to 500/mm3, or <200/mm3.
Two different HPV detection methods, polymerase chain reaction (PCR) and hybrid capture (HC) (Digene Diagnostics Inc., Silver Spring, MD, U.S.A.) were used. PCR is an amplification-based test that is highly sensitive and detects low-level HPV infection but does not discriminate between low-level and high-level infection. HC is a nonamplification-based test that detects high-level infection only.
Anal swabs that were used to collect samples for HPV testing were inserted into tubes containing 1 ml of Sample Transport Medium (Digene Diagnostics Inc.), and frozen at −70°C until analysis. For PCR analysis, 200 μl were used according to the MY09/MY11 consensus HPV L1 primers as described elsewhere (15). HC was conducted according to the manufacturer's recommendations. Two separate tests were performed on each specimen to determine the relative light unit (RLU) ratio for the HPV types associated with a low risk of cervical cancer in probe group A (6/11/42/43/44) and for the HPV types associated with a intermediate and high risk of cervical cancer in probe group B (16/18/31/33/35/45/51/52/56). The results of HC are expressed as an RLU ratio, determined by dividing the chemiluminescent signal of the test sample by that obtained with a control sample containing 10 pg/mL of HPV 16 DNA. Magnitude of the RLU ratio increases with increasing quantity of HPV DNA in the specimen. The RLU ratio was computed and classified as negative (<0.65) or positive (≥0.65) and for some analyses, positive specimens were further categorized as 0.65 to <10, 10 to 100, and >100.
Subjects in the cohort were interviewed about past medical history, behavioral factors such as smoking, drinking, recreational drug use, rectal drug use (i.e., insertion of recreational drugs such as cocaine into the rectum), and sexual practices. The presence or absence of a disease, treatment, behavior, or HPV infection was used to evaluate the association between anal lesions and potential risk factors whenever practical. For many of the behavioral variables, almost all subjects had engaged in the particular activity at least occasionally. For potential risk factors, summary variables were developed to characterize lifetime exposure to the activity (15).
The SAS statistical package (SAS Institute, Cary, NC, U.S.A.) was used for data analysis. Analyses were conducted separately for HIV-positive and HIV-negative subjects. ASCUS was included in the abnormal category because it is often associated with HPV infection, is found in the presence of colposcopically or anoscopically detectable SIL, and because it frequently predates detection of SIL (1,9,10). Results classified as indeterminate were included in the abnormal category for analysis because approximately 80% of men with normal cytology and a visible lesion have been shown on biopsy to have a histologic abnormality (10). Relative risk (RR) and 95% confidence intervals (CI) were calculated separately for ASIL and for all grades of anal abnormality combined (16). The Mantel-Haenszel procedure was used to compute adjusted RR.
A total of 608 men participated in this study with 346 (57%) HIV-positive and 262 (43%) HIV-negative subjects. The SFGH, UCSF, and SFMHS subjects had similar demographic characteristics. The ethnic breakdown was 91% white non-Hispanic, 2% black, 7% Hispanic, and 1% other. In terms of education, 9% reported ≤12 years, 54% reported 13 to 16 years, and 37% reported >16 years of education. No differences were found between the HIV-positive and HIV-negative subjects by race, ethnicity, or education. The mean age of the HIV-positive men was 42 years (range, 24-64 years) and that of the HIV-negative men 45 years (range, 26-73 years).
Table 1 shows results of cytologic and histologic testing for the HIV-positive and HIV-negative subjects. Of 346 HIV-positive men, 137 (40%) had normal cytology and no lesion or a normal biopsy specimen. Of 262 HIV-negative men, 206 (79%) had normal cytology and no lesion or a normal biopsy specimen. ASIL, including LSIL and HSIL, was diagnosed cytologically or histologically in 124 of 346 (36%) HIV-positive men and in 19 of 262 (7%) evaluable HIV-negative men. HSIL specifically was found in 17 HIV-positive men (5%) and in only 1 of the HIV-negative men.
Table 2 shows the relation between anal lesions at baseline, HIV status, and CD4 level. The RR was 2.9 for having had any grade of anal lesion in HIV-positive men compared with HIV-negative men and was 5.7 for ASIL specifically. The RR for any grade of anal lesion increased as the CD4 level decreased, and RR associated with ASIL were higher than those for all abnormalities combined.
Table 3 presents the relation between anal disease and HPV infection measured by PCR and HC for HIV-positive and HIV-negative men. When analyzing ASIL alone, HIV-positive men who were PCR-positive for multiple HPV types had more than a threefold increased RR for ASIL compared with HIV-positive men who were PCR-negative. The RR for ASIL in both HIV-positive and HIV-negative men increased with number of HPV types detected using PCR. The RR for ASIL associated with multiple HPV types was higher in HIV-negative men (13) than in HIV-positive men (RR = 3.3), compared with HPV-negative men. Among both HIV-positive and HIV-negative men, RR associated with HPV infection were higher for ASIL than for all grades of anal abnormality combined.
HPV infection measured by HC was associated with higher RR for ASIL in both HIV-positive and HIV-negative men than was infection measured by PCR. Among HIV-positive men, the RR was higher when positive for the B group alone than the A group alone and was highest when both groups were detected. The RR associated with ASIL were consistently higher than those associated with all anal lesions combined (Table 3). Mutual adjustment of the RR for HC positivity (group A or B) and CD4 counts in the HIV-positive men showed little change in RR. Among HIV-negative men, the RR for ASIL was highest for men positive for group A alone or in combination with group B.
Table 4 shows the relation between detection of anal lesions and the level of HPV DNA in the anal specimen as determined by the HC RLU ratio. Among the HIV-positive men, increasing RR for anal lesions were associated with higher levels of HPV DNA in both the A and B groups. Among the HIV-negative men, the highest RR were in those with the highest levels of group A DNA. Except for the HIV-negative men using the HC group B probe mixture, tests for trend confirm a strong relation between amount of HPV DNA and anal lesions for both HIV-positive and HIV-negative men.
Among HIV-positive men, the RR for ASIL at different levels of HPV DNA positivity were stratified by CD4 level to examine the relation between ASIL, the level of HPV positivity measured by HC group B RLU ratios and immunosuppression measured by CD4 levels (Table 5). The RR for ASIL increased with increasing levels of HPV DNA in men with CD4 counts > or <500/mm3. At each level of HC group B DNA positivity, the RR for ASIL higher in those with CD4 counts <500/mm3 than in those with counts >500/mm3.
Lifestyle and medical risk factors for anal lesions were analyzed separately for HIV-positive and HIV-negative men. For HIV-positive men and for all risk factors, little difference appeared between the RR associated with having all grades of lesions combined and those associated with having ASIL only. Risk of anal lesions was slightly increased in HIV-positive men with a history of genital warts (all lesions: RR = 1.3, 95% CI, 1.1-1.5; ASIL: RR = 1.5, 95% CI, 1.1-2.0) and a slightly increased risk associated with rectal discharge (all lesions: RR = 1.2, 95% CI, 1.0-1.5; ASIL: RR = 1.4, 95% CI, 1.0-1.8). When these risk factors were adjusted for presence of HPV infection by HC or for CD4 count, the RR and 95% CIs remained unchanged (data not shown). HIV-positive men who reported moderate or high frequency of receptive anal intercourse had an increased RR for anal lesions when compared with those who reported infrequent or no anal intercourse (RR = 1.6, 95% CI, .95-2.8) but the confidence limits overlapped unity.
Among HIV-negative subjects, risk of anal lesions was elevated in those who reported using phencyclidine piperdine (PCP), lysergic acid diethylamide (LSD), or psilocybin (magic mushrooms) (RR = 1.7; 95% CI, 1.1-2.8) compared with those who did not. Those who engaged in receptive fisting also had an increased RR (RR = 2.0; 95% CI, 1.0-3.8). In multivariate analyses, these RR were slightly diminished when adjusted for HPV infection as detected by HC (RR = 1.5 for drugs and 1.5 for fisting) and confidence limits overlapped unity. History of rectal drug use (RR = 1.7, 95% CI, .96-3.1), gonorrhea (RR = 1.6; 95% CI, .90-2.7) and Chlamydia (RR = 1.5; 95% CI, .87-2.7) were associated with increased RR for anal lesions in HIV-negative men but confidence limits overlapped 1.0. Other factors that were examined and found not to be associated with increased risk of anal lesions in either HIV-positive or HIV-negative men included cigarette smoking, alcohol use, overall illicit drug use, history of herpes, rectal bleeding, history of bladder or bowel radiographs, hemorrhoids, anal itching, anal fissures or fistulas, and use of laxatives, enemas, or suppositories.
A screening program to detect ASIL in high-risk individuals may be of value to prevent anal cancer. Therefore, knowledge of the prevalence and risk factors for anal lesions is important. Our study differs from previous work in several ways. First, few studies of risk factors have been conducted for anal lesions in HIV-positive and HIV-negative homosexual men. Second, we used PCR to detect 29 individual HPV types and a mixture of 10 grouped types, permitting analysis of the role of multiple types in anal disease. Third, we used HC to detect HPV infection, permitting semiquantitative analysis of the amount of HPV DNA in the anal specimen and its relation to detection of anal lesions. Fourth, our study also differs in its use of anal histology to classify the stage of anal lesions. As in the cervix, cytology results do not always correlate with histology (10), and the combination of cytology and histology provides a more accurate assessment of anal lesions than cytology alone.
Because risk factors for anal lesions may differ for HIV-positive and HIV-negative men, these groups were analyzed separately. A high proportion of HIV-positive men had abnormal cytology or histology (60%), similar to that observed in a previous study (17). Most of these were mild, with ASCUS or LSIL in 55% and HSIL in 5%. HIV infection was one of the strongest risk factors for ASIL, with a RR of 5.7 in HIV-positive men compared with HIV-negative men, and high rates of ASIL were found in HIV-positive men at all CD4 levels. Our data showed that the most important risk factors for anal lesions in HIV-positive men were low CD4 counts, infection with HPV types in both the HC A and B groups, and high levels of HPV DNA as measured by RLU ratios. Consistent with the findings of an earlier study (18), HIV-related immunosuppression may contribute to the development of ASIL independent of its effects on HPV replication, because our data showed that the RR for ASIL was higher in those with lower CD4 counts even when stratified by the level of HPV DNA. Other risk factors for ASIL in HIV-positive men included history of genital warts and history of rectal discharge. Although history of rectal discharge may be associated with sexual behaviors that increased exposure to anal HPV infection, the RR for ASIL for rectal discharge remained elevated after adjustment for HPV infection as measured by HC, as did the RR of history of genital warts.
Anal lesions were less common in HIV-negative men than in HIV-positive men. HSIL was found in only one HIV-negative subject. Our data showed that the most important risk factor for anal lesions in HIV-negative men was HPV infection, as measured by PCR or HC. The RR for anal lesions increased at high levels of HPV DNA group as measured by RLU ratios, particularly in the HC A group. The association between anal abnormalities and the HC A group of nononcogenic HPV types rather than the oncogenic B group may reflect the near absence of HSIL in HIV-negative men and the low rate of progression to HSIL within 2 years (9). Other risk factors in these men included use of PCP, LSD, or magic mushroom use, rectal drug use, and receptive fisting. These factors were no longer significant after adjustment for HPV infection and may reflect sexual behaviors that increased the risk of acquisition of HPV.
Based on this study, several groups of men should be considered at high risk for ASIL and may benefit from a screening program. Our data show that a large proportion of HIV-positive men with a history of receptive anal intercourse had anal lesions, regardless of CD4 counts. HIV-positive men with anal HPV detected by HC were at particularly high risk of ASIL. Our study shows that the HIV-negative men most likely to have anal lesions were those who tested positive for HPV. Whereas HPV testing may be of value to identify high-risk individuals, further work is needed to determine its role in a screening algorithm. Women are also at risk for anal lesions and anal cancer, particularly those with high-grade cervical lesions (19), cervical cancer (20), vulvar cancer (21) and those who are HIV-positive (22-24).
After an anal cytologic abnormality has been identified, the patient should undergo anoscopy with biopsy of visible lesions for histologic confirmation. Among patients likely to survive for several years, HSIL should be treated because of its potential to progress to anal cancer. Treatment usually consists of surgical removal (13). LSIL probably does not progress directly to invasive anal cancer but men with LSIL may consider treatment if they have symptoms such as itching or bleeding. Men with LSIL, particularly those who are HIV-positive, require careful follow-up due to the high progression rate of LSIL to HSIL (9). Further studies are needed to define the benefits and costs of a screening strategy to prevent anal cancer in men and women at high-risk for ASIL.
Acknowledgments: The authors are grateful to Jennifer Kristianson, Pam Elmore, and Rachel Remler for their expert technical assistance. Supported by grant R01CA54053 from the National Cancer Institute. These studies were carried out in the General Clinical Research Center, University of California-San Francisco with funds provided by the Division of Research Resources 5 M01-RR-00079, U.S. Public Health Service.
1. Palefsky JM, Gonzales J, Greenblatt RM, Ahn DK, Hollander H. Anal intraepithelial neoplasia
and anal papillomavirus infection among homosexual males with group IV HIV disease. JAMA
2. Zaki SR, Judd R, Coffield LM, Greer P, Rolston F, Evatt BL. Human papillomavirus
infection and anal carcinoma: retrospective analysis by in situ hybridization and the polymerase chain reaction. Am J Pathol
3. Daling JR, Weiss NS, Hislop TG, et al. Sexual practices, sexually transmitted diseases, and the incidence of anal cancer
. N Engl J Med
4. Holly EA, Whittemore AS, Aston DA, Ahn DK, Nickoloff BJ, Kristiansen JJ. Anal cancer
incidence: genital warts, anal fissure or fistula, hemorrhoids, and smoking. J Natl Cancer Inst
5. Qualters JR, Lee NC, Smith RA, Aubert RE. Breast and cervical cancer surveillance, United States, 1973-1987. MMWR Surveill Summ
6. Rabkin CS, Yellin F. Cancer incidence in a population with a high prevalence of infection with human immunodeficiency virus type 1. J Natl Cancer Inst
7. Melbye M, Cote TR, Kessler L, Gail M, Biggar RJ. High incidence of anal cancer
among AIDS patients: the AIDS/Cancer Working Group. Lancet
8. Palefsky JM, Holly EA. Molecular virology and epidemiology of human papillomavirus
and cervical cancer. Cancer Epidemiol Biomarkers Prev
9. Palefsky JM, Holly EA, Hogeboom CJ, et al. Viral, immunologic and clinical parameters in the incidence and progression of anal squamous intraepithelial lesions in HIV-positive and HIV-negative homosexual men. J Acquir Immune Defic Syndr Hum Retrovirol
10. Palefsky JM, Holly EA, Hogeboom CJ, Jay N, Berry JM, Darragh TM. Anal cytology as a screening tool for anal squamous intraepithelial lesions. J Acquir Immune Defic Syndr Hum Retrovirol
11. Winkelstein W, Jr., Wiley JA, Padian NS, et al. The San Francisco Men's Health Study: continued decline in HIV seroconversion rates among homosexual/bisexual men. Am J Public Health
12. Moss AR, Bacchetti P, Osmond D, et al. Seropositivity for HIV and the development of AIDS or AIDS related condition: three-year follow up of the San Francisco-General Hospital cohort. BMJ
13. Palefsky JM. Anal cancer in HIV-positive individuals: an emerging problem. AIDS
14. Kurman RJ, Solomon D. The Bethesda system for reporting cervical/vaginal cytologic diagnoses: definitions, criteria and explanatory notes for terminology and specimen adequacy.
New York: Springer-Verlag, 1994.
15. Palefsky JM, Holly EA, Ralston ML, Jay N. Prevalence and risk factors for human papillomavirus
infection of the anal canal in HIV-positive and HIV-negative homosexual men. J Infect Dis
16. Breslow NE, Day NE. The design and analysis of cohort studies. Statistical methods in cancer research,
vol. 2, no. 82. Lyons, France: International Agency for Research on Cancer Scientific Publications, 1987.
17. Kiviat N, Critchlow C, Holmes K, et al. Association of anal dysplasia
and human papillomavirus
with immunosuppression and HIV infection among homosexual men. AIDS
18. Critchlow CW, Surawicz CM, Holmes KK, et al. Prospective study of high grade anal squamous intraepithelial neoplasia in a cohort of homosexual men: influence of HIV infection, immunosuppression and human papillomavirus
19. Scholefield JH, Sonnex C, Talbot IC, et al. Anal and cervical intraepithelial neoplasia
: possible parallel. Lancet
20. Melbye M, Sprogel P. Aetiological parallel between anal cancer
and cervical cancer. Lancet
21. Ogunbiyi OA, Scholefield JH, Robertson G, et al. Anal human papillomavirus
infection and squamous neoplasia in patients with invasive vulvar cancer. Obstet Gynecol
22. Williams AB, Darragh TM, Vranizan K, Ochia C, Moss AR, Palefsky JM. Anal and cervical human papillomavirus
infection and risk of anal and cervical epithelial abnormalities in human immunodeficiency virus-infected women. Obstet Gynecol
23. Melbye M, Smith E, Wohlfahrt J, et al. Anal and cervical abnormality in women-prediction by human papillomavirus
tests Int J Cancer
24. Hillemans P, Ellerbrock TV, McPhillips S, et al. Prevalence of anal human papillomavirus
infection and anal cytologic abnormalities in HIV-seropositive women. AIDS