1. Home>
2. Specialties>
3. HIV/AIDS Clinical Care>
4. Feature

Anal Cancer in HIV Infection: To Screen or Not to Screen?

Anal cancer screening for HIV-infected patients remains a controversial topic . To discuss the potential risks and benefits of such screening, ACC convened a discussion with Dr. Sue Goldie, who has done research into the cost-effectiveness of anal cancer screening in HIV-infected patients; Dr. Joel M. Palefsky, who has extensive clinical and research experience in this area; and Dr. Kimberly Workowski, who is Chief of the Guidelines Unit of the epidemiology and surveillance branch in the CDC's division of STD prevention.

ACC: Let's start with some background: What are the incidence and prevalence of anal cancer in HIV-infected patients, and how do these numbers compare with those in the general population?

Dr. Palefsky: We don't really have a handle on the true incidence of anal cancer. Before the AIDS epidemic, the incidence was estimated at 12 to 35 per 100,000 among men with a history of anal intercourse, compared with about 0.7 to 0.9 per 100,000 in the general population. So clearly, a history of anal intercourse puts people at substantially increased risk.

Dr. Goldie: But the risk was and is probably even higher.

Dr. Palefsky: That's a key point because we've never routinely screened for anal disease. As far as the HIV-positive population is concerned, a paper by Bob Biggar's group shows at least a 2-fold increase in HIV-positive men who have sex with men (MSM) compared with HIV-negative MSM, and Mortin Frisch has shown an overall 37-fold increase in HIV-positive MSM compared with the general population. So, we estimate an incidence of roughly 70 to 75 per 100,000 HIV-positive MSM -- a substantially higher rate than that of cervical cancer in the general population of women.

Dr. Workowski: Anogenital sex is rarely addressed as a risk factor for human papillomavirus (HPV) infection. What is its role in HPV transmission?

Dr. Palefsky: HPV transmission is probably most efficient through anal intercourse, but it probably also is possible through anogenital intercourse. If you look at the Journal of the National Cancer Institute paper on anal abnormalities in women, one of the risk factors was abnormal cervical cytology, and an abundant literature shows that women at risk for anal cancer are at increased risk for cervical cancer and vice versa. It's safe to say that in many of these people, HPV is a field infection. It's hard to track down in which direction the infection got transferred.

Dr. Workowski: And what about the incidence among women?

Dr. Palefsky: The data we have on women are more limited than in men, but they point in the same direction. Frisch and colleagues showed a 6.8-fold increase in cancer among HIV-positive women compared with women in the general population. The logic you use to support screening in HIV-infected men probably applies to HIV-infected women.

ACC: How frequently is anal cancer a cause of hospitalization or death among HIV-infected people?

Dr. Goldie: Like cervical cancer, it's still an uncommon cause relative to other causes of mortality. However, the current incidence of anal cancer in MSM, even in those without HIV infection, is well above that of cervical cancer. Most data, albeit indirectly, point to a higher risk for anal cancer in MSM than for cervical cancer in women, independent of screening and treatment, but we don't know the magnitude of that risk.

Dr. Palefsky: I agree. Also, data from the Scandinavian registries base suggest that the incidence of anal cancer is steadily increasing in the general population in both men and women, by something like 2% per year for the last 10 years. I imagine several factors are influencing this trend: Some of the increase may be accounted for by the HIV-infected population, and some probably reflects a cohort effect -- a hangover from the days of sexual liberation beginning in the '60s and '70s -- which probably explains most of the increase in women.

Dr. Workowski: Clinicians rarely ask their patients, particularly women, about receptive anal intercourse, and they should, given the increased frequency of intercourse in college-age women.

Dr. Palefsky: We need to know more about the risk factors for HPV infection, and we need to study cytologic and histologic changes in HIV-negative women who have been risk-matched to HIV-positive women. We also need to better understand the status of average women who come into a cervical dysplasia clinic because they're at risk for cervical disease and of even lower-risk women who just come in for screening.

ACC: Who should receive anal Pap smears and how often? For instance, should HIV-infected women receive anal Pap smears and, if so, at the same rate as HIV-infected men?

Dr. Palefsky: These are incredibly loaded questions. We have a pretty good handle on who's at risk, but making the jump to deciding who should receive anal Pap smears is another matter. At this stage, it's a philosophical issue. Given the evidence that certain populations aren't at high enough risk of disease to merit screening, widespread implementation of Pap-smear screening at this point would not meet the most stringent evidence-based guidelines. Thus, the question becomes how much evidence do we need, and what is involved in awaiting such evidence? Are we putting patients at risk by insisting on evidence, or are we putting them at risk by going forward with the Pap-smear program now?

Dr. Goldie: Keep in mind that the level of evidence that we have required for establishing other cancer-screening programs has been variable. For example, there has never been a randomized controlled trial of cervical-cancer screening that assigned patients to two arms -- screening or no screening -- and followed them prospectively to ascertain mortality differences due to invasive cancer. Rather, modeling efforts in conjunction with data have in large part shaped our current cervical-cancer screening program. We need to think carefully about what we are going to require in terms of evidence for the establishment of an anal-cancer screening program.

Dr. Workowski: The other issue is a lack of consistent or cogent natural history data. We're trying to extrapolate from what we know about cervical cancer, but the anal mucosa is very different from the cervical mucosa. What if a Pap smear comes back abnormal? What treatment do we suggest? What do we know about the progression from low-grade to high-grade lesions? We may get some answers from the ongoing SUN study, a CDC-funded, prospective, observational study with collection of anal Pap smears in HIV-infected men and women.

Dr. Palefsky: Unfortunately, I don't believe that anal Pap smears in this study are being followed up routinely with anoscopies and biopsies. Certainly, clinicians should not do anal Pap smears unless they are prepared to deal with the results.

Dr. Workowski: They've left it up to each site to do it; there's no standardization between sites; but I'm not aware of any other ongoing, prospective, cohort study collecting these data.

Dr. Goldie: Just to be clear about the definition of natural history: We have the published data that we've used in the past to infer the incidence of low-grade lesions, the progression rate to high-grade lesions, and the incidence of high-grade lesions, but we never will be able to observe or calculate actual incidence.

Dr. Palefsky: But I think that we can all agree that the focus needs to be on high-grade disease, which is of most concern to us from a mortality standpoint.

Dr. Goldie: As Dr. Palefsky alluded to, it is unclear if we will be able to collect reliable data on high-grade lesions before making decisions about screening and treatment programs.

Dr. Workowski: Unfortunately, in the meantime, few colorectal surgeons are buying into the results of such screenings, and so the clinician who sees a low-grade or a high-grade lesion doesn't know what to do. Once we have the Pap-smear results, we need guidance on what to do.

Dr. Palefsky: We published guidelines in 2002, in which we said that before embarking on a screening program, a clinic should set up an infrastructure that includes people with different kinds of expertise. This will also be discussed in the new treatment guidelines for HIV-related opportunistic infections.

Dr. Workowski: I have questions about the text of the draft guidelines on HIV opportunistic infections that mention that there are insufficient data to recommend a specific treatment approach. The guidelines for anal intraepithelial neoplasia (AIN) describe treatment options based on lesion size, which may be difficult for inexperienced clinicians to discern. Are there published data that validate such recommendations? Some of the treatments listed are quite caustic, and I am concerned about some of the intra-anal treatment options.

Dr. Palefsky: Well, given the paucity of available evidence-based medical data, the guidelines reflect specialists' practice. We have made it quite clear that there isn't a lot of evidence to back up the efficacy of these treatments.

Dr. Workowski: Right. However, it would be useful if the tables indicated that the guidelines are based on clinical experience and not published literature [see Tables 1 and 2 with the suggested change incorporated].

Dr. Palefsky: Fair point. But although these treatments can be caustic, they are extraordinarily well tolerated in the anal canal. If applied properly with the small end of the stick, not the cotton end, such treatments have a very, very low morbidity and are extremely cheap.

Dr. Goldie: What we need right now is a large-scale study that looks at alternative management strategies for high-grade lesions -- a study with large enough numbers and with enough support. Do we have anything at all?

Dr. Palefsky: No. We're doing a retrospective chart review of our various treatment approaches to high-grade disease to collect data on how successful we were in making those high-grade lesions go away and for how long. I don't think anybody is doing a randomized, prospective study. I do, however, think that listing treatment options and clarifying the supporting evidence demonstrates that treatment options exist. Sometimes I hear from doctors, "Well, why should I look, because nothing can be done?" That's not true. We need to know more about how good these approaches are, but I think people are extremely nihilistic to use the absence of information as a reason not to look.

ACC: What are the data regarding the effects of antiretroviral therapy on the development of neoplasia and anal cancer?

Dr. Palefsky: Data we are now preparing for publication are very clear: potent combination antiretroviral therapy offers little benefit. We have evidence from San Francisco County that the incidence of anal cancer has continued to rise since the advent of potent antiretroviral therapy.

ACC: How sensitive and specific are anal Pap smears and anal cytology for detection and grading? When are biopsies necessary and who should be performing them? And is the question of who should be performing screening and treatment in any way controversial?

Dr. Palefsky: Everything is controversial. The anal Pap smears are equivalent to cervical Pap smears, warts and all. Their sensitivity is better in HIV-positive men than in HIV-negative men, probably because the lesions are bigger in HIV-positive men. Pap smears have a sensitivity of 50% to 80% in HIV-positive men. And just as in cervical Pap smears, they shouldn't be used for grading a lesion. The positive predictive value of a high-grade lesion is high, but the positive predictive value of low-grade smears is quite a bit lower. In fact, when we looked at the Pap-smear results in men who had high-grade lesions, roughly half of them were low-grade on cytology or less. We tend to use the anal Pap smear for yes/no screening, more or less the way the cervical Pap smear is used. Individuals who have an abnormal Pap smear are referred for visualization of the lesion and biopsy.

ACC: And does the biopsy involve referral to a specialist?

Dr. Palefsky: Not necessarily. In our clinic, we have an active interest in the disease, and so we often do the biopsies ourselves, but we do get a lot of referrals from the community. So, there are a lot of people, primary care providers, who are doing Pap smears on their patients and referring them to us if the Pap smears come back abnormal.

Dr. Workowski: I guess you've developed a nice infrastructure, but where the infrastructure doesn't exist, finding someone capable of performing a screening with a high-resolution anoscopy is somewhat challenging.

Dr. Palefsky: I agree, but infrastructures exist because an individual or a group has made a commitment. Resources are out there. For instance, we are starting up a study as part of the CPCRA in which we will offer training to individuals who want to participate in the study.

Dr. Workowski: And getting the word out that this training is available?

Dr. Palefsky: Well, we can't offer it to everybody, but we are offering it to people who are participating in the multisite studies. We almost always have trainees in our clinic -- they mostly observe, but we hope they'll go home and get practice and start to train others. It's important to point out that the field is new: Only a handful of people have been involved in clinical care of anal disease -- and only over the past 10 years.

Dr. Goldie: Look at the pattern of colorectal cancer screening: The workshops for family practitioners on how to perform outpatient sigmoidoscopy didn't gain momentum until there were some screening recommendations in place. Right now, the clinical guidelines for anoscopy aren't supported with hard evidence of benefit, infrastructure is inadequate, and indirect evidence of efficacy won't be sufficient to promote infrastructure development. I can see a circular process in which nothing happens unless research is put into the infrastructure.

Dr. Workowski: Who's interested? Is the buy-in mostly from colorectal surgeons, or from gastroenterologists, or is it a mix?

Dr. Palefsky: Actually, it doesn't need to be either. Screeners can be primary care providers, and anoscopists can be infectious disease people, nurse practitioners, physician assistants, or others. It depends on who takes an interest. A parallel is cervical colposcopy, which is available from a wide range of people.

Dr. Workowski: But I'm not aware of many HIV clinics that perform on-site colposcopies unless they are involved in research. Gynecologists do most colposcopies. You're talking about on-site, high-resolution anoscopy being performed in the clinics, which would be a financial burden for a lot of sites.

Dr. Palefsky: I guess it's site-specific. At UCSF, a high proportion of our patients are on a MediCal plan, and we get by.

ACC: Dr. Palefsky, you mentioned that clinicians shouldn't be engaging in screening unless they're prepared to deal with the results. What does "being prepared to deal with the results" look like? How do you advise patients regarding moderate-grade dysplasia or high-grade dysplasia? And how do you work through those options with patients?

Dr. Palefsky: I tell patients that mild dysplasia is not likely to lead directly to cancer and that treatment is optional. Most patients prefer treatment, and here there's a cost/benefit ratio. I tell them that if it's easy to treat then it might be worth considering. Some of these lesions are going to progress to high-grade over time, and if a patient changes his or her mind later, treatment options will be more difficult. If the disease is widespread and treatment would be difficult and painful my recommendation changes.

At the opposite end of the spectrum is severe dysplasia, in which the strength of my recommendation for treatment goes up as the potential for progression to cancer increases. Here also, the ease or difficulty of treatment is important. For some patients with diffuse disease, treatment is going to be extremely challenging and both physically and financially costly. I tell those individuals, that although we'd prefer to treat, the evidence suggests that severe dysplasia progresses to cancer only in a minority of people. If such an individual agrees to defer treatment, he or she is followed closely, with the goal of catching a cancer very early. We have diagnosed a number of individuals with such early cancer that we think that we might be able to treat them with wide local excision, which is sphincter-sparing, and spares patients the morbidity associated with chemo and radiation therapy.

We don't have any natural history data that compare the likelihood of developing cancer with moderate versus severe dysplasia because we don't have a lot of progression-to-cancer data. By and large, my recommendations for moderate dysplasia are similar to those for severe, although I tell patients that the risk of progression to cancer is likely to be lower with moderate than with severe dysplasia. My hope is that if we do implement more proactive anal-screening programs, we'll be catching lesions earlier, and when lesions are small, we think they can be treated quite easily and inexpensively.

Dr. Workowski: In the HIV opportunistic infection guidelines, would you recommend that treatment of AIN 2 and 3 include observation as an option?

Dr. Palefsky: It's always an option, and it is something we do.

Dr. Goldie: In cervical guidelines, the option for such levels of dysplasia is usually listed as "management." In following a high-grade lesion, maybe guidelines should specify a screening frequency.

Dr. Workowski: Indeed, this would be helpful in the upcoming guidelines. There's great variability in current clinical practice.

ACC: Where are we with the guidelines? How close are we to having the same sort of dependable recommendations that support cervical cancer screening and treatment?

Dr. Palefsky: We have put out guidelines, and the evidence supporting them is mostly based on natural history data, our personal experience with treatment, and, of course, our experience treating cervical cancer, so there's a certain degree of cogency. It will be a long time before every step of the screening and treatment is strongly supported by good evidence-based medicine, but guidelines, such as they are, are out there.

Dr. Workowski: We're probably planning to have another consensus meeting for the CDC STD Treatment Guidelines a year from now, and then publish our guidelines a year later -- probably in 2006 -- so we hope we'll have some more information.

FURTHER READING

— Sue Goldie, MD, MPH, Joel M. Palefsky, MD, and Kimberly Workowski, MD

Dr. Goldie is Associate Professor Health Decision Science in the Department of Health Policy and Management at the Harvard School of Public Health in Boston. Dr. Palefsky is Professor, Laboratory Medicine and Medicine, at the University of California, San Francisco, and Program Member, UCSF Comprehensive Cancer Center. Dr. Workowski is Chief of the Guidelines Unit, Epidemiology and Surveillance Branch in the CDC's Division of STD Prevention, and Associate Professor, Division of Infectious Diseases, at Emory University in Atlanta.

Published in Journal Watch HIV/AIDS Clinical Care July 1, 2004

Citation(s):

Goldie SJ et al. The clinical effectiveness and cost-effectiveness of screening for anal squamous intraepithelial lesions in homosexual and bisexual HIV-positive men. JAMA 1999 May 19; 281:1822-9.

• Original article (Subscription may be required)
• Medline abstract (Free)

Frisch M et al. Cancer in a population-based cohort of men and women in registered homosexual partnerships. Am J Epidemiol 2003 Jun 1; 157:966-72.

• Original article (Subscription may be required)
• Medline abstract (Free)

Stanley M. Chapter 17: Genital human papillomavirus infections--current and prospective therapies. J Natl Cancer Inst Monogr 2003; 31:117-24.

Chin-Hong PV and Palefsky JM. Natural history and clinical management of anal human papillomavirus disease in men and w omen infected with human immunodeficiency virus. Clin Infect Dis 2002 Nov 1; 35:1127-34.

• Medline abstract (Free)

CDC, NIH, IDSA Guidelines for the treatment of opportunistic infections in adults and adolescents infected with the human immunodeficiency virus. MMWR Recomm Rep 2004 (in press).