"2001 Consensus Guidelines for the Management of Women With Cervical Cytological Abnormalities" by Thomas C. Wright, MD, and others was published in the April 24, 2002 issue of the Journal of the American Medical Association (JAMA). The article presents new guidelines for managing abnormal changes in the cells of the uterine cervix, as detected by Pap smear testing. The guidelines are a consensus statement developed by a panel of 121 experts in the diagnosis and management of cervical cancer. At a conference in Bethesda, Maryland, this past September, participants met to discuss drafts and issue final guidelines.
Routine Pap smear screening for cervical cell changes and cervical cancer precursors is an important tool for optimizing women's health. Management of abnormal Pap smear test results hinges on whether the results are ASC-US (atypical squamous cells of undetermined significance) or ASC-H (atypical squamous cells, cannot exclude high-grade squamous intraepithelial lesion [HSIL]). Women with the most common abnormal result, ASC-US, are recommended to have two more Pap smears, immediate colposcopy (examination of the tissues of the cervix with a low-powered, lighted microscope to identify abnormal cell growth and, if necessary, removal of a tissue sample for biopsy), or DNA testing for high-risk types of human papillomavirus (HPV), the cause of anogenital warts and cancers. Though all three are valid options, testing for HPV DNA is strongly preferred when liquid-based cytology (e.g., a Pap smear test) is used for screening.
In most cases of ASC-US, the same cell sample that produced the abnormal Pap smear results also can be used for the HPV testing, so that women would not need to come back for another examination or screening. If the HPV test is negative for the strains of virus that are considered high risk, women can more or less be assured that they are not at risk for subsequent cancer and need no further special follow-up.
Women with more definitive Pap smear results such as ASC-H, low- or high-grade squamous intraepithelial lesion(s), or atypical glandular cells should proceed immediately to colposcopic evaluation. Greater convenience, as well as fewer follow-up visits, lower cost, and less anxiety for the patient all should be advantages conferred by the adoption of the new recommendations.
The guidelines also discuss special circumstances, stating that "referral for colposcopy is recommended for all immunosuppressed patients with ASC-US. This includes all women infected with HIV, irrespective of CD4 cell count, HIV viral load, or antiretroviral therapy." Considerations also are given for postmenopausal and pregnant women.
Algorithms describing the guidelines and a glossary of terms are available at the ASCCP Web site; the glossary of terms used in the guidelines is also available at http://jama.ama-assn.org.
Two articles in the May 8, 2002 issue of JAMA support the use of HPV testing in combination with Pap smears as a cost-effective, convenient approach to routine cervical cancer screening.
In "Benefits and Costs of Using HPV Testing to Screen for Cervical Cancer," lead author Jeanne Mandelblatt, MD, MPH, of Georgetown University Medical Center in Washington, DC, states that adding HPV tests to a lifetime of biennial (every two years) Pap smears would detect 225 more cases of cancer per 100,000 women and decrease cervical cancer deaths by 59%. She notes that Pap tests have a number of drawbacks, including frequent false-positive results that necessitate repeat or alternate testing, and explains that HPV tests in essence "hit the mark" (screen for indicators of cancerous cervical cells) much more reliably than do Pap tests alone. Rather than continue with the standard approach that prescribes an annual Pap smear, Dr. Mandelblatt explains that, "maximal savings in life could be achieved by screening every two years from age 20 to death with a combination of HPV and Pap tests. Cessation of screening at age 65 or 75 years is less expensive and captures 86.6 and 97.8% of the benefits of lifetime screening, respectively."
In the second JAMA article, "Cost-Effectiveness of Alternative Triage Strategies," Sue Goldie, MD, MPH, of the Harvard Center for Risk Analysis, and colleagues used a computer-based mathematical model to determine the most efficient and cost-effective management strategy for U.S. women with ASC-US results. Dr. Goldie's team also found that the most cost-effective approach would be to use a biennial schedule in which all women receive Pap smears; the same Pap smear samples that produce ASC-US results could then be used to provide cells for HPV testing. Positive HPV results would indicate follow-up colposcopy only when they reveal a strain linked to cervical cancer. Results of this cost-effectiveness study lend full support to the new guidelines described in the previous section. As with most recommendations, rarely does one size fit all: clinicians must consider each woman's history and individualize decision-making.
One measure that may help individualize the management of abnormal Pap smear results is the number of lifetime sexual partners a woman has had, according to a new study by John Chan, MD, of the University of California at Irvine, and colleagues. For two years, investigators followed 93 women with cervical cell abnormalities known as cervical intraepithelial neoplasia (CIN) grades 2 or 3, with 3 being the most likely to progress to cancer. Dr. Chan presented his findings at the annual meeting of the American College of Obstetricians and Gynecologists, held in early May.
As noted above, infection with certain strains of HPV is the chief risk factor for developing cervical cancer. However, a number of cofactors -- including smoking -- also contribute to disease progression. In some investigations, having multiple sexual partners also has been linked to increased risk for HPV. In this study, the researchers found that having had more than five lifetime sexual partners was an independent risk factor for cancerous changes, even in the absence of HPV infection. The study did not elucidate the reasons for this finding, although the investigators hypothesized that other sexually transmitted infections (STIs) in addition to HPV might contribute to higher risk for cancer-like cell changes, and that an increased number of sexual partners is associated with increased risk of developing STIs.
The investigators found that women with five or fewer lifetime sexual partners had a greater than 60% chance of having cervical cell abnormalities clear up of their own accord, without treatment. The presence or absence of HPV had a significant impact on risk for progression; 80% of cervical lesions that did not contain HPV regressed, compared with 46% of those that did contain HPV. The authors suggested that some women with CIN grade 2 or 3 might delay treatment -- if, for instance, they have had five or fewer lifetime sexual partners and do not have HPV infection. (Physicians typically do not treat CIN grade 1 because these abnormalities usually resolve without treatment; however, CIN 2 or 3 generally is treated with a technique that destroys the affected cervical tissue.)
In addition, women whose HPV infection resolved were more likely to see their cervical lesions regress, compared with women with persistent HPV infection. Other factors, including a woman's race and whether she had given birth, did not affect whether abnormalities regressed, research showed. Overall, 52% of abnormalities resolved on their own. One limitation to this report is that strain of HPV was not considered, and it is well known that HPV strain is of foremost concern when it comes to risk for anogenital cancers.
According to a French study published this year in Sexually Transmitted Infections, women are seven times more likely than men to tell a partner that they have an STI. The study involved three groups of participants, all of whom completed surveys. Over 13,000 adults and adolescents 15 years of age and older participated.
A total of 45 adolescents and 179 adults reported a history of STIs. Nine of ten adults reported that they had told partners about an STI, but one in four adolescents had not. Less than 8% of one group of adults (adults were surveyed in two groups) said they had not informed their main partner, but three-quarters had not told "other" partners. Among the second group of adults, almost all said they had not told any of their previous partners, although most told their current partner. Almost one-third of adolescents had failed to notify current partners.
Gender was a strong determinant of willingness to admit to having an STI: 14% of men, compared with just 2% of women, had not told their main partner. Among adolescents, half of the boys did not tell their partner, but fewer than one in ten girls did not tell. Those 16 years of age and younger were least likely to admit to having an STI. The authors concluded that "these results are all the more worrying in that they probably underestimate the true situation," adding that many of the respondents might have said they had told their main or current partner when, in fact, they had not. Because many STIs are asymptomatic, especially in women, partner notification is very important.
In the U.S., the incidence of HIV/AIDS is increasing most rapidly among African-American women. While the vast majority of cases occur in urban areas and nearby communities, the spread of HIV to and through rural areas is of increasing concern. Results of a recent study that looked at HIV risk in this group of women were published by Richard Crosby, PhD, and others, in the April 2002 issue of American Journal of Public Health. This study compared HIV-associated sexual health history, risk perceptions, and sexual risk behavior of low-income rural and non-rural African-American women who were attending federally funded Special Supplemental Nutrition Program for Women, Infants, and Children (WIC) clinics in Missouri.
Participants were women receiving WIC benefits aged 18 years or older from 27 clinics in 23 counties who agreed to participate in an anonymous survey that focused on condom use. A total of 571 women (25% of women who filled out the survey) reported that they were African American; of these, 72% were from urban counties, 4% from suburban counties, and 24% from rural counties. (More than 90% of Missouri counties are rural.)
Rural women were twice as likely as urban women to report that they had received no HIV counseling during pregnancy. They also were twice as likely to report that a sex partner had been tested for HIV infection but that they "did not worry about becoming infected" by STIs. They were twice as likely to say that they never used condoms, because they believed that their current partner was HIV negative. Rural women also reported 50% less gonorrhea or syphilis.
In their report, the authors cited another recent study indicating that rural minority and low-income women living with HIV/AIDS were likely to have believed, before their HIV diagnosis, that they could not get infected or that their partners were HIV negative. The authors said that the studies provide initial evidence suggesting that low-income rural African-American women believe themselves to be less susceptible to HIV infection than their urban counterparts. The authors noted, "Given the potential diffusion of HIV from high-concentration epicenters to rural areas, these perceptions may be highly problematic in regard to the adoption of protective practices such as condom use."
Long-term data from the Petra (formerly PETRA) study were been published in the April 6, 2002 issue of The Lancet. The data highlight the devastating effects breast-feeding can have on the early benefits of perinatal anti-HIV treatment to prevent transmission of the virus from mothers to infants. In brief, many women participating in the Petra study who took peripartum antiretroviral therapy and delivered HIV negative babies subsequently infected their babies through breast-feeding. The results also depict some of the shortcomings of applying "short-course" treatment alone to mothers in Africa.
The aim of Petra, which ran from June 1996 through January 2000, was to discover the most effective way of administering two common oral antiretrovirals, AZT (zidovudine, Retrovir) and 3TC (lamivudine, Epivir), to prevent mother-to-child transmission (MTCT). [Ed. note: AZT and 3TC were dosed separately, not as Combivir, which is one pill combining both medications.] Researchers enrolled 1,797 pregnant women with HIV in sites in Tanzania, South Africa, and Uganda, and randomized them into four groups following four different regimens. The first group received the two drugs during pregnancy (starting one month before delivery), during labor, and for one week after delivery (i.e., both mothers and infants). The second group received the same two drugs, but only during labor and after delivery (i.e., both mothers and infants). The third group received AZT and 3TC during labor only (i.e., no treatment for infants after birth), and the fourth group was given placebo. (The placebo arm was halted and those participants were randomized to one of the other three groups receiving active therapy in 1998, halfway through the study, when results from a Thai trial of short-course AZT showed 50% efficacy compared with placebo. It would have been unethical to continue the placebo arm at that point, with proof of the efficacy of short-course therapy then available. The analysis of results consider only data from women enrolled through 1998.)
The resulting infection rates among babies at the main time points for evaluation, at six and 18 months after birth, are as follows. Six weeks after birth, 5.7% of newborns in the first group were HIV positive; the rates in the other groups were 8.9%, 14.2%, and 15.3%, respectively. But regardless of group the infection rates thereafter soared because mothers were transmitting the virus through their breast milk, and after 18 months, infection rates were 15%, 18%, 20%, and 22%, respectively.
Unlike the standard of care for women in developed countries, which typically involves use of multiple antiretrovirals for several weeks, the standard available to women in African countries is limited by funds. To date, the only affordable options involve short-course treatment focusing on labor and delivery -- and this is not available to the majority. Joep Lange, MD, of the Academic Medical Centre-International Antiviral Therapy Evaluation Centre at the University of Amsterdam, has said that the treatment regimen used by the first Petra group (therapy before, during, and after delivery) continues to be most beneficial for infants at risk of HIV infection from their mothers. At the Third Conference on Global Strategies for the Prevention of HIV Transmission from Mothers to Infants this past September in Kampala, Uganda, Dr. Lange discussed some disappointing developments related to this study (see "The Dilemma of Breast-Feeding," BETA, Winter 2002, page 36). He pointed out that despite the potency and efficacy of the regimen used by group one, which initially reduced MTCT by 61% (measured at six weeks) benefits were effectively lost by 18 months when mothers also breast-fed. The second regimen also provided initial efficacy, but those benefits also diminished considerably over the infants' first 18 months of life, again largely due to breast-feeding. Dr. Lange stressed that preventing MTCT in developing countries must be addressed in the larger context of care, in which breast-feeding is but one important issue for HIV positive mothers and families. He also suggested adding a third potent antiretroviral to the regimen used by group one in this study, which by today's standards in developed countries is outdated.
Currently, Petra investigators recommend evaluating whether weaning from breast-feeding at six months of age and simultaneously administering prophylactic antiretrovirals to nursing mothers and uninfected infants may provide further protection against HIV. They also acknowledge that in developing settings such as those where the study was conducted, breast-feeding will continue, for "social, hygienic, and financial reasons."
Orphans face enormous challenges -- ranging from inadequate nutrition, health care, and education, to homelessness and abuse -- which in turn impact the societies within which they live. Dr. Beckerman states that while knowledge about effective strategies to reduce MTCT has advanced, maternal health has yet to become a full-fledged concern. She notes that even Petra, a well-designed if now outmoded trial, studied and reported no maternal health endpoints. Since there are clear links between maternal and infant health and survival, continuing negligence by MTCT researchers and care providers is shortsighted at best.
Appealing to scientific interests, Dr. Beckerman referenced evidence from Western studies showing that treating pregnant women for HIV results in "far lower" HIV transmission rates than do mother-to-child interventions alone. She concludes, "Now, in the third decade of AIDS, HIV prevention and AIDS treatment can and must be integral parts of the global response to the AIDS catastrophe. What better place to start than with mothers and children?"
An article published in the March 8, 2002 issue of AIDS by a team of French researchers from INSERM, the French national research institute, reported that a short course of AZT reduced the risk of HIV transmission to babies as long as 24 months after birth, even when mothers breast-fed. (These results contrast with those from other studies such as Petra, reported above; what seems to have made the difference in this study was the mothers' CD4 cell counts at the time of delivery, which researchers say underscores the significant role of maternal health in MTCT.)
In this evaluation researchers pooled and analyzed data from two clinical trials conducted in Côte d'Ivoire and Burkina Faso. In those trials mothers took 300 mg of AZT or placebo twice daily from 36 to 38 weeks gestation until delivery; one group also took the same regimen for seven days after delivery.
Results were available for 641 children, of whom all but 12 were breast-fed and all of whom had at least one HIV-1 test. Overall, at 24 months of age, those children whose mothers had received treatment were 26% less likely to have become infected with HIV. However, when researchers looked at MTCT results based on mothers' CD4 cell counts at the time of delivery, they saw strikingly different results. Among children whose mothers had greater than 500 cells/mm3 and who took AZT, the risk of having been infected with HIV was reduced by 59%. Among children whose mothers had fewer than 500 cells/mm3 and took AZT, the chances at 24 months of having been infected with HIV were roughly equal to those of children whose mothers took placebo; i.e., mothers' use of AZT conferred no risk reduction for these children.
While having fewer than 500 CD4 cells/mm3 is not a sign of advanced HIV disease, it does indicate a significant decline in the immune system's capabilities. (Many of the women likely had far fewer than 500 cells/mm3 at the time of childbirth.) Therefore, the authors said, disease stage at the time of childbirth is a strong factor in the overall odds of MTCT. Lead author Valeriane Leroy, MD, PhD, and colleagues suggested starting AZT therapy earlier in pregnancy or, better yet, using AZT in combination with other anti-HIV drugs.
Another article in the April 6, 2002 issue of The Lancet concerns women and HIV. Audrey French and colleagues reported data obtained through a large longitudinal study of the natural history of HIV in women called the Women's Interagency HIV Study (WIHS). (See "Lessons Learned From Natural History Studies in Women," BETA, Winter 2001, page 22.) In the San Francisco Bay Area, researchers at the University of California at San Francisco (UCSF) are actively recruiting for WIHS; for more information, call (toll-free) 866-476-5109.
An association between vitamin A, or retinol, deficiency and MTCT has been reported in several previous studies. Although results of such studies have been contradictory, retinol is known to play a role in maintaining healthy genital mucosa, and so the theory remains that the association is related to HIV viral load in the female genital tract. That is, a deficiency of retinol damages mucosal integrity, thereby elevating risk of HIV transmission.
In this study, researchers explicitly examined the relationship between retinol and the amount of HIV RNA in genital secretions. To do so, they tested samples of blood and genital secretions from 301 of the 418 WIHS participants from the Bronx, New York site, obtained through cervicovaginal lavage taken at the first study visit. (Samples had been stored for later studies such as this.)
The researchers found that plasma (blood) levels of HIV correlated significantly with genital HIV levels, but that blood levels of retinol did not. Women with detectable levels of genital HIV and women without detectable levels had roughly similar retinol levels. Only six women had retinol deficiencies, which is too small a number to permit useful analyses.
The researchers commented that their findings are similar to those of a 1999 study by R.N. Burns and Richard Semba of retinol deficiency, which also found no association between retinol status and MTCT. Yet their findings differed from African studies, which did find an association between depressed retinol levels and genital HIV levels. This may be due to the fact that the women in the African studies had much more severe retinol deficiencies. Thus, the researchers concluded that in the population of U.S. women they studied, it appears that "a correlation between retinol status and genital HIV-1 burden does not exist." While these data do not mean that vitamin A supplementation for women with profound retinol deficiencies will have no protective anti-HIV effect, neither do they bolster the hope that arose several years ago that vitamin A supplementation might become a cheap MTCT prevention strategy widely available to poor African women.
Leslie Hanna is editor of BETA.
Back to the SFAF BETA Summer/Autumn 2002 contents page.