Cervical Cancer and Women with HIV
Cervical cancer is a preventable condition, and when diagnosed early, it can be cured. Deaths from cervical cancer have gone down by 70% in the U.S. since 1943, when routine Pap smears were introduced for sexually active women. Pap smears consist of a sample of cells scraped from the cervix and examined under a microscope for precancerous abnormalities. Abnormal cells may be referred to as dysplasia, cervical intraepithelial neoplasia (CIN) or squamous intraepithelial lesions (SIL). In this article these three terms are used interchangeably. Only high-grade dysplasia is considered to be a true cancer precursor. The significance of mild dysplasia is presently uncertain.
Cervical cancer has long been known to be a disease of sexually active women. Though a causal connection has not been completely established, malignancy in the cervix is now firmly associated with infection by specific subtypes of human papillomavirus (or HPV, the accepted cause of genital and anal warts). Malignancy in the cervix is believed to begin to develop in adolescents or young adults and progress slowly over the decades. It is also strongly linked to tobacco use (Winkelstein, 1990; Feldman, 1997).
The Centers for Disease Control and Prevention (CDC) added invasive cervical cancer to its list of AIDS-defining conditions in 1993. Still, much remains unknown about the incidence and natural history of this disease in HIV-positive women. Pap smear screening of HIV-positive women has demonstrated a four- to tenfold elevation in abnormal results compared to HIV-negative women. It is hoped that such screening appreciably decreases the potential for cervical cancer in identified HIV-positive women. The proportion of women diagnosed with dysplasia and cervical cancer who are also HIV-infected is unknown, though, since many women with these conditions may not be aware that they also have HIV. They may have never been tested for the virus, or their cervical pathology may have preceded other signs and symptoms of AIDS. Just as rigorous surveillance for cervical cancer and its precursors must continue in women diagnosed with HIV, aggressive surveillance of women with cervical abnormalities must pursue the connection with HIV infection.
Over the past ten years, a great deal of research has amassed regarding Pap smear findings and their management in HIV-positive women. There continue to be a number of controversies, divergent findings and a range of clinical recommendations for the prevention, diagnosis and treatment of lower genital tract dysplasia in HIV-positive women. (It should be noted that research on HPV-related cancers and their precursors has not been adequately conducted in HIV-positive men, in whom anal cancer is associated with the presence of HPV -- as is also true for women. Prevention and treatment recommendations are lacking for men as a result of this paucity of research.)
Incidence of HPV and Dysplasia in HIVIn an early description of the common occurrence of HPV and dysplasia in HIV-positive women, Byrne (1989) reported that 7 of 19 (37%) HIV-positive, asymptomatic women had abnormal Pap smears. Other findings in these women included vaginal and vulvar abnormalities including both warts and dysplasia. Feingold (1990), compared HIV-negative and HIV-positive women and reported that positive women have higher rates of cervical HPV infection.
Current research is attempting to evaluate the subtypes of HPV present in HIV-positive women, and examine the value of testing for HPV with PCR viral detection technologies. Cervicovaginal lavage samples used for PCR testing confirm that HIV-positive women are more likely to have HPV than are negative women. Investigators also have found that women with HIV have more subtypes of HPV and a greater prevalence of high-risk HPV subtypes (Sun, 1995).
Women on immune suppresive therapies have long been known to experience increased HPV-related events, including warts, dysplasia and cancer (Vermund, 1990). In 1987, Bradbeer published an early inquiry into the relationship between HIV and cervical dysplasia. This was followed by numerous reports including anecdotal reports, descriptive studies, cross-sectional studies, longitudinal studies and case-controlled studies (Provencher, 1988; Feingold, 1990; Maiman, 1990; Vermund, 1991; Mandelblatt, 1992; Maggwa, 1993 and others). All of these reports verified that HIV-positive women are at a greater than normal risk for acquisition, persistence and progression of HPV-related genital tract dysplasia.
Studies have observed, too, that the incidence of HPV-related dysplasia increases as immune function declines in HIV-positive women. Cross-sectional and longitudinal samples of HIV-positive women demonstrate that both rates of HPV infection and disease and of high-grade dysplasia increase as CD4 counts lower (Schafer, 1991; Vermund, 1991; Anastos, 1992; Wright, 1994).
HPV "Field Effect"Early reports on HPV disease in HIV-infected women described that warts, and other, more serious HPV-related growths, frequently showed a "field effect" in the genital tract. They commonly occurred in two, three or even more locations including the cervix, vagina, vulva, periurethra, perineum, and the perianal and rectal areas (Sillman, 1987). In 1990, Maiman compared cases of HIV-positive and -negative women with preinvasive cervical disease. Fifty percent of the positive women had perianal disease, "field effect" and/or extensive cervical disease. None of the negative women had any of these conditions. Further, anal HPV infection is found as commonly as cervical HPV infection in HIV-positive women (Williams, 1994).
A cross-sectional study of 396 HIV-positive and 395 negative women further confirmed these findings (Chiasson, 1997). All of the women were evaluated by Pap smear, colposcopy and cervicovaginal lavage (CVL) for HPV typing. In these women, prevalence of lower genital tract condyloma (warts) was higher in HIV-positive (5.6%) than HIV-negative women (0.8%). High-grade vulvar dysplasia (VIN) was present in two positive women, but none of the negative women, and HPV growths were far more likely to demonstrate the "field effect" in the women with HIV.
Cervical Cancer in HIVDoes all this mean that more HIV-positive women will develop cervical cancer? At present, there is only speculation that these trends will lead to increasing incidence of cervical cancer and related deaths. Some researchers have looked aggressively for answers. In 1990, Maiman and colleagues evaluated 37 women diagnosed with invasive cervical cancer. Nineteen percent of the cancer patients under age 50 were HIV-positive. Positive patients had more advanced invasive disease than negative ones, and disease persisted or recurred in all positive patients compared to 37% of negative patients. Other researchers (Relliman, 1990; Schwartz, 1991; Maiman, 1993) have published case reports of rapidly progressive invasive cervical cancer in women with HIV. There have also been case reports of invasive vulvar cancer in HIV-positive women (Giorda, 1992; Wright, 1996).
Maiman and his colleagues have continued to study the relationship between HIV and cervical cancer in a large cohort of women in Brooklyn, where there is a high female HIV rate. In this cohort, women with HIV infection and cervical cancer have a higher recurrence rate than HIV-negative women diagnosed at comparable stages of cancer (Maiman, 1993). The investigators also reported on their findings from a colposcopy clinic at a Brooklyn hospital, where patients were routinely offered HIV counseling and testing. The positive women had increased rates of severe dysplasia, larger lesions, and greater "field effect" than the women testing negative (Fruchter, 1997). The authors point out that since HIV-positive women are four to ten times more likely than similar HIV-negative women to have an abnormal Pap smear, colposcopy clinics tend to have a concentration of HIV-positive women, and thus should represent an important site for HIV information, counseling and testing.
But do the Brooklyn findings apply to all women with HIV infection? When the CDC in 1993 added invasive cervical cancer to the list of AIDS-defining conditions in women with HIV infection, researchers began to undertake surveillance investigations of women with AIDS and invasive cancer (Klevens, 1995) and to review invasive cancer cases for AIDS surveillance purposes (Reardon, 1994; Kelley, 1995). In a retrospective review of data on cervical cancer and AIDS in women registered through the New York City Department of Health and Institutional Tumor Registries from 1987 to 1995, Maiman (1997) found a strong correlation between cervical cancer and HIV disease throughout New York City.
Between 1980-1992, 377 women living in Contra Costa County, near San Francisco, were diagnosed with invasive cervical cancer. A surveillance review of 79 charts identified only two additional cases of AIDS. However, HIV test results were available for fewer than 15% of cases. The investigators did not recommend aggressive HIV surveillance of cervical cancer cases, but strongly recommended establishment of protocols by which clinicians diagnosing cervical dysplasia provide HIV counseling and testing for clients.
The Gynecologic Oncology Group has recently begun a nationwide study (GOG 154) examining the issue of HIV testing and follow-up for women younger than 50 with invasive cervical cancer. It is hoped that this will provide some insight concerning various clinical, pathologic, epidemiologic and demographic factors. Without an organized HIV screening program for women with invasive cervical cancer, it is not possible to gauge the magnitude of the disease in those with HIV infection.
Pap Smear Screening vs. Colposcopy in HIVAnother important question to consider is, does Pap smear screening adequately identify HIV-positive women who are at high risk for cervical cancer? We have known for quite a while that HIV-positive women have high rates of abnormal Pap smears. What we are not certain about is what these abnormal results reveal about risk for cervical cancer in women with HIV. Conversely, we also do not know if the Pap smear is sensitive enough to identify abnormalities in HIV-positive women in time to prevent cervical cancer. Should all HIV-positive women be screened for dysplasia with more sensitive methods, such as. colposcopy and biopsy? So-called "screening colposcopy," which involves directly examining a woman's cervix with a special microscope, has several critical disadvantages. It requires management by a specialist and the procedure can be painful. Further, the biopsies that frequently accompany the colposcopic exam are invasive and entail a risk of bleeding or infection. All in all, this is not a cost-effective way to identify cervical cancer in the general population. It is difficult to say whether colposcopy screening should be advised for a specific HIV-positive subgroup.
Women's successive Pap smear reports were evaluated by investigators at Jackson Memorial Hospital in Miami (Provencher, 1988). The overall incidence of abnormal Pap smears for the HIV-negative group was 5%, while the overall incidence of abnormal Pap smears for the positive women was 63%. Certainly the number of abnormal Pap smears identified over time in groups of positive women is impressive. But when Maiman and colleagues (1991) compared routine Pap smears to routine colposcopy in HIV-positive women they found the Pap smears to be less than adequate. Thirteen of 32 patients (41%) had CIN by colposcopy and biopsy, yet only one of these women had CIN on Pap smears taken at the same time. The authors felt that cytologic screening may not be adequately predictive of CIN in HIV-positive women. Some clinicians began to recommend that colposcopy be part of the routine management of HIV-positive women. When other researchers (Adachi, 1993; Korn, 1994) looked at Pap smear adequacy, though, Pap smears were found to be as sensitive in HIV-positive women as in other women.
Recurrent Dysplasia with HIVLongitudinal reports also find that dysplasia frequently recurs after treatment in HIV-positive women. Treatment for dysplasia involves various modes of removing or destroying the abnormal tissue (e.g. cryotherapy, electrocautery, laser and surgical excision). In the Brooklyn cohort (Maiman, 1993), 39% of HIV-positive women, compared to 9% of negative women had recurrences of dysplasia after standard treatments. The same authors later reported that 62% of the positive cohort had developed recurrent dysplasia by 36 months. And recurrence rates reached 87% in women with CD4 counts below 200. After a second treatment, dysplasia recurred in 42%, and three of six women had a third recurrence (Fruchter, 1996). The authors note that the risk of recurrence is related to degree of immune compromise and that innovative therapies are needed for controlling CIN in women with HIV.
Interpretation of Mild Atypical Cell PatternsA further controversy among clinicians is appropriate management for "atypical" Pap smear results that are not conclusive for dysplasia and for results that indicate "inflammation" without an apparent cause. Atypical squamous cells are those cells that show minimally abnormal findings including the characteristic findings associated with HPV-infected cells (called koilocytes). Inflammation involves the presence of white blood cells, pathogens, and/or clumping of cells that make the specimen more difficult to evaluate, but do not by themselves indicate dysplasia. HIV-positive women have a high incidence of atypical cells and inflammation, which in the general population are often managed by treating any infection that is present and repeating the Pap smear in three months (treat and repeat). It is known that a small percentage of women with these types of Pap results actually do have high-grade lesions or cancer, thus any woman with repeated atypical cells on Pap smear results should be evaluated by colposcopy and biopsy.
It is unknown to what degree HIV infection in the genital tract contributes to Pap smear findings of "atypia" or inflammation, or if there are other variables that may account for an increased incidence of these findings. A large prospective study of Pap smear and colposcopic findings from 453 HIV-positive and 401 HIV-negative women was begun in 1990, and continued at six-month intervals (Wright, 1996). In this cohort, Pap smears from HIV-positive women were far more likely to be diagnosed with mild cytologic atypia than were smears from negative women. The HIV-positive women with mild atypia on Pap smear findings were 2.7 times more likely to have mild dysplasia on colposcopy/biopsy than the negative women, and 12% of such smears were associated with high-grade dysplasia on colposcopy/biopsy. The authors conclude that mild cytologic atypia, a frequent diagnosis on Pap smears from HIV-positive women, is strongly associated with CIN. They recommend that all HIV-infected women with mild cytologic atypia be referred for colposcopy. For inflammatory changes they recommend the same management as for other women; either to treat and repeat the Pap smear in three to four months or to refer for immediate colposcopy.
Update from the National Conference on Women and HIVSome interesting areas of inquiry into Pap smear management and the natural history of HPV infection in HIV-positive women were reported at the National Conference on Women and HIV. Several researchers discussed the use of HPV DNA assays for detecting high-risk HPV subtypes in HIV-positive women. Clinicians have been eager to have an adjunctive test, besides the Pap smear, to use in screening women to determine whom to refer for colposcopy. It is disappointing that HPV DNA typing tests appear to improve the sensitivity of Pap smear screening only slightly.
Biomedical research into the interaction between HIV and HPV in laboratory cell lines also was described (Cage, 1997). In the lab, HPV-infected cells display an increased growth response when exposed to interleukin-6 (IL-6), tumor necrosis factor (TNF-a), and the HIV-produced tat protein. Since HIV can up-regulate the production of IL-6 and TNF-a, there are several possible pathways by which HIV in the genital tract may enhance the growth potential for HPV in infected cells, conceivably leading to an increased risk of cervical cancer.
Current Guidelines for Women with HIVFor many years now, clinicians who care for large numbers of HIV-positive women have recommended increasing the frequency of their Pap smear screening. Many experts in HIV primary care for women recommend that all HIV-positive women receive gynecologic evaluations including Pap smear screening in their usual primary care setting every six months (Marte, 1989; Allen, 1990; Minkoff, 1991; Denenberg, 1993 and others).
Unfortunately, there is not yet a consensus regarding frequency of Pap smears or the management of abnormal findings. In 1990, the CDC published recommendations for annual Pap smears for HIV-positive women unless the results were abnormal. In 1991, Minkoff and DeHovitz published a counterpoint argument, raising concerns about the conservative nature of the CDC recommendations. They suggested that since HIV-positive clients are seen more than once a year, semi-annual Pap tests would seem a reasonable approach in the primary care setting, with liberal referral for colposcopy. They further recommended that treatment of dysplasia should be aggressive.
In 1994, the federal government's Agency for Health Care Policy and Research (AHCPR) published "Clinical Practice Guideline: Evaluation and Management of Early HIV Infection," which included a Pap smear algorithm for women with early HIV infection (p. 174). Among the document's recommendations were the following:
In 1995, the CDC published "USPHS/IDSA Guidelines for the Prevention of Opportunistic Infections in Persons Infected with HIV: A Summary," which included recommendations for prevention and management of HPV disease. These recommendations (pp. 22-23) included:
In the recently published revision (CDC, 1997), annual Pap smears are recommended for women with two consecutive negative Pap reports, and management of all abnormal Pap smears should follow the management strategies described by the National Cancer Institute Consensus Panel as appropriate for all women. These recommendations are an improvement over the 1995 version, but they are still at variance with the AHCPR algorithm, and they do not provide recommendations for treatment and long-term follow-up.
Management of Abnormal Pap Smear ResultsThe goal of Pap smear screening is to prevent cervical and/or lower genital tract cancer. The Pap smear is a screening test with an average sensitivity of about 70% to 80% (it may miss 20% to 30% of true abnormalities) and an average specificity of about 70% (about 30% of abnormal reports may be from women without a true abnormality). Thus, Pap smears may miss true abnormalities or identify abnormal cells for which no true abnormal growth is present. The best management strategy is one that will not fail to detect a true precursor to cancer, but also does not require that large numbers of women undergo unneeded invasive biopsies.
It is evident that the sensitivity and specificity of the Pap smear does not differ significantly among HIV-positive and HIV-negative women. Yet the proportion of HIV-positive women who receive abnormal Pap smear results is much higher than for HIV-negative women, therefore a greater number of HIV-infected women who have abnormalities will go undetected. Coupled with this lack of detection, there is the risk that an undetected abnormality may progress to cancer more rapidly in immune-compromised women. The need to create a strategy to identify these women early is readily apparent.
Management of Pap smear findings should rightly be a part of primary care for all women. Unfortunately, all too often, primary care providers are not adequately trained in general management strategies for abnormal Pap results in the general population of women, and there is still more confusion about managing Pap smears for women with HIV. The desire to refer all such women to expert colposcopists is tempting, but not reasonable, nor does it represent quality primary care. On the other hand, improper management of abnormal smears can cost lives. The three algorithms proposed are based on a thorough review of the available research and clinical recommendations, and a belief that Pap smear management is not outside the scope of primary care for HIV-positive women. They are provided as a guide for primary care providers and their HIV-positive female patients.
Managing True Cancer PrecursorsTreatment failure and recurrence of dysplasia in positive women is a discouraging finding of the present research. It remains to be seen if highly active antiretroviral therapies will improve local immune function, thus potentially improving treatment success for some women. Hundreds of women, both HIV-positive and -negative have undergone thousands of colposcopies, biopsies and treatments for dysplasia both as research subjects and as women receiving standard care to prevent lower genital tract dysplasia. Several researchers point out that better treatments are needed for positive women with HPV-related disease, but little treatment research is in progress.
Again, the goal is to prevent or treat lower genital tract cancer. Thus, until better treatment regimens are available, some clinicians are offering to provide an observational approach (see algorithms in print version) to management of warts and biopsy-proven mild dysplasia, since the recurrence rate after treatment is so high. Women can be examined every three to six months to search for any indication of true cancer precursors. Treatment for warts can be at the option of the client (see "Treatment for Warts"). When there are signs of higher grades of dysplasia, aggressive treatment should be offered. Recurrent local treatment or more extensive surgical treatment should be determined by discussion between clinician and patient.
ReferencesAdachi A et al. Obstetrics and Gynecology. March 1993; 81(3):372-7.
Anastos K et al. 8th International Conference on AIDS. 1992; abstract Tu.B.0532.
Bradbeer C. Lancet. November 28, 1987; 2(8570):1277-8
Byrne MA et al. AIDS. June 1989; 3(6):379-82.
CDC. Morbidity and Mortality Weekly Report. July 14, 1995; 44(RR-8):22-3.
CDC. Morbidity and Mortality Weekly Report. June 27, 1997; 46(RR-12):25-7.
Chiasson MA et al. Obstetrics & Gynecology. May1997; 89(5):690-4.
Denenberg R (1993). Gynecological Care Manual for HIV-Positive Women.
Durant, OK: Essential Medical Information Services
Feingold AR et al. Journal of Acquired Immune Deficiency Syndromes. September 1990; 3(9):896-903.
Feldman JG et al. Obstetrics & Gynecology. March 1997; 89(3):346-50.
Fink MJ et al. Gynecologic Oncology. October 1994; 55(1):133-7.
Fruchter RG et al. Obstetrics & Gynecology. March 1996; 87(3):338-44.
Giorda G et al. Gynecologic Oncology. February 1992; 44(2):191-4.
Heard I et al. Obstetrics & Gynecology. November 1995; 86(5):749-53.
Kelley KF et al. Journal Acquired Immune Deficiency Syndromes. January 1995; 8(1):102-3.
Klevens MR et al. Obstetrics & Gynecology. August 1996; 88(2):169-73.
Korn AP et al. Obstetrics & Gynecology. March 1994; 83(3):401-4.
Koutsky LA et al.The New England Journal of Medicine. October 1992; 327(18):1272-8.
Kurman RJ et al. Journal of the American Medical Association. June 15, 1994; 271(23):1866-9.
Maggwa BN et al. AIDS. May 1993; 7(5):733-8.
Maiman M et al. Obstetrics & Gynecology. January 1997; 89(1):76-80.
Maiman M et al. Obstetrics & Gynecology. August 1993; 82(2):170-4.
Maiman M et al. Cancer. January 15, 1993; 71(2):402-6.
Maiman M et al. American College Obstetrics and Gynecology. July 1991; 78(1): 84-8.
Maiman M et al. Gynecology Oncology. September 1990; 38(3):377-82.
Maiman M et al. Journal of the American Medical Association. October 21, 1988; 260(15):2214-5.
Mandelblatt JS et al. AIDS. February 1992; 6(2):173-8.
Minkoff HL, Dehovitz JA. Journal of the American Medical Association. October 23-30, 1991; 226(16):2253-8.
Olaitan A & Johnson MA. Journal of the International Association of Physicians in AIDS Care. May 1997; 3(5):15-7.
Provencher D et al. Gyneologic Oncology. September 1988; 31(1):184-90.
Reardon J et al. Journal of Acquired Immune Deficiency Syndromes. June 1994; 7(6):631.
Relliman MA et al. Gynecologic Oncology. March 1990; 36(3):435-8.
Schafer A et al. American Journal of Obstetrics/ Gynecology. February 1991; 164(2):593-9.
Sillman FH, Sedlis A. Obstetrics & Gynecology Clinics North America. June 1987; 14(2):537-58.
Schwartz LB et al. Gynecologic Oncology. June 1991; 41(3):255-8.
Vermund SH et al. American Journal Obstetrics/Gynecology. August 1991; 165(2):392-400.
Williams AB et al. Obstetrics & Gynecology. February 1994; 83(2):205-11.
Winkelstein W Jr. American Journal Epidemiology. June 1990; 131(6):945-57,
Wright TC et al. Gynecologic Oncology. March 1996; 60(3):500-3.
Wright TC et al. Obstetrics and Gynecology. April 1996; 87(4):515-9.
Wright TC et al. Obstetrics & Gynecology. October 1994; 84(4):591-7.
This article was provided by Gay Men's Health Crisis. It is a part of the publication GMHC Treatment Issues. Visit GMHC's website to find out more about their activities, publications and services.