College of American Pathologists
Printable Version

  Feature Story


cap today

New Pap guidelines are here—and now the real work begins

June 2002
Karen Southwick

Now that the long-awaited consensus guidelines for managing women with abnormal Pap smears have been published in the April 24 issue of JAMA, you’d think the controversy would be over.

Not quite.

First, clinicians, insurers, patients, and laboratory professionals will have to be educated about the new guidelines, which do, in some cases, require significant changes from what had been accepted practice.

Second, laboratories will have to decide whether it’s cost-effective to add the liquid-based cytology and human papillomavirus testing recommended by the guidelines.

Third, the role of HPV testing in diagnosing and managing cervical cytological abnormalities continues to evolve. The May 8 issue of JAMA contains two articles on HPV testing, one suggesting it could be used as a primary screen for cervical cancer.

Consequently, there should be a lot of fodder for discussion at the CAP’s upcoming conference on HPV testing, to be held Sept. 21 and 22 in Chicago. "A lot of times when guidelines are issued, clinicians read them but don’t really know what to do," says Rick Lozano, MD, co-organizer of the conference. The meeting will focus on questions such as, When should HPV testing be used? How should it be reported? How should clinicians act on it? How should it be brought into the lab?

Dr. Lozano, who is director of cytopathology at Cunningham Pathology, Birmingham, Ala., says the consensus guidelines and the new Bethesda terminology for reporting results of cervical cytology will be featured prominently at the conference, the timing of which is fortuitous. "We’ve been planning this for about 12 months," he says, while publication of the guidelines had been expected much earlier than April.

Diane D. Davey, MD, the other co-organizer of the conference, says it will be "a high-intensity meeting focused on both the science and the practical aspects of incorporating HPV testing into your lab." The conference "will help you make a decision on whether you want to set up this test in your own lab and tell you how to educate clinicians, how to do quality control, and what the regulatory requirements are," says Dr. Davey, director of cytopathology at the University of Kentucky Chandler Medical Center, Lexington.

Reaching agreement

The consensus guidelines say that HPV should be used as a triage test for the 2 million women whose Pap test results are classified as ASC-US (atypical squamous cells, undetermined significance), rather than repeat the Pap test or perform an immediate colposcopy. The guidelines strongly suggest that liquid-based cytology be used for all Pap tests, since a "reflex" HPV test can be done on the same sample. "Requiring women to return for HPV DNA testing or repeat cervical cytological testing is inconvenient and would be expected to increase costs," according to the April 24 JAMA article "2001 consensus guidelines for the management of women with cervical cytological abnormalities."

Dr. Lozano notes that there are "a lot of logistical issues" in taking two separate samples—that is, a Pap and a repeat sample if HPV testing is needed. "If you do it as one specimen and the lab can reflex it, you ensure that you maintain the specimen integrity. That makes it easier and more functional for the lab to do the additional test." Besides, he adds, the sensitivity of liquid-based cytology is much greater than that of a Pap test.

Women with ASC-US whose HPV test is negative should be followed up with a repeat Pap test at 12 months, according to the guidelines. Women who are positive for high-risk HPV should be referred for colposcopy. If cervical intraepithelial neoplasia, or CIN, is ruled out, those women should return to repeat the Pap test at six and 12 months or get an HPV test at 12 months. If CIN is found, the women should be treated accordingly. "Because of the potential for overtreatment," the April 24 JAMA article says, procedures such as LEEP (loop electrosurgical excision procedure) "should not routinely be used to treat women with ASC in the absence of biopsy-confirmed CIN."

Because women with ASC-H (cannot exclude high-grade squamous intraepithelial lesion, or HSIL) have a higher risk for CIN3+ than ASC-US, women with this Pap interpretation should be referred for colposcopy. Women with a finding of atypical glandular cells (AGC, formerly called AGUS) should undergo colposcopy with endocervical sampling, except women with atypical endometrial cells, who should be evaluated with endometrial sampling.

For the first time, "the guidelines make clear that when liquid cytology is used, the preferred method for triaging ASC cases is reflex HPV testing," says Dr. Lozano. Although those results had been expected (as reported in the October 2001 CAP TODAY article "Closer to consensus on abnormal Pap tests"), "we needed the auspices of a consensus conference and publication in a peer-review journal," he adds. "This should guarantee reimbursement for HPV tests that are ordered and used appropriately." Ironically, however, a higher volume of HPV testing might result in lower payments per test. "If a lot of claims start coming in, insurance companies may start ratcheting down reimbursement," Dr. Lozano says.

"Approximately 60 percent of the total Pap market is now liquid-based, and, by numbers, liquid-based cytology is now the standard," says J. Thomas Cox, MD, a gynecologist who directs the gynecology clinic in the Health Service at the University of California, Santa Barbara. "The conventional Pap will disappear in the next few years," he predicts. The new standard will be liquid-based cytology every two years, Dr. Cox adds, with reflex testing for ASC-US. "Even hold-out payers will be convinced by these guidelines because this approach is more cost-effective than the present, common approach of annual conventional Paps and repeat cytology for ASC-US," he says.

The guidelines are backed by a number of prestigious medical organizations, including the National Cancer Institute and the American Society for Colposcopy and Cervical Pathology. "We really got a lot of people together behind HPV testing," says Dr. Davey, although "there’s still the option of doing colposcopy or repeat Pap."

Says Dr. Cox: "I totally agree with the guidelines. HPV is 96 percent sensitive for the highest grade pre-cancer." A negative HPV finding "gives you enough reassurance to go back to annual Pap testing." For women whose Pap result is ASC-US and who receive an HPV-positive finding, "their risk of high-grade precancer is almost identical to those with LSIL," Dr. Cox adds. For both groups, "you go to colposcopy."

David R. Bolick, MD, medical director of Reference Pathology Services, Sandy, Utah, proclaims himself a longtime proponent of HPV testing and liquid Pap, so he’s pleased with the new guidelines. "In my opinion, the traditional Pap smear is antiquated," he says. Even if you don’t wind up running an HPV test, "you get a better specimen with liquid Pap."

The Food and Drug Administration has approved only one liquid Pap sample for Digene HPV testing, Cytyc’s ThinPrep, but Dr. Bolick says his lab uses TriPath’s SurePath, which is awaiting FDA approval. (Laboratories can perform HPV testing on SurePath samples if they perform a validation study.)

Dr. Bolick believes more reflex HPV testing will increase the 61 percent market share for liquid Pap tests. A combination of more volume and competition should drive prices down, he says.

The experts agree that HPV reflex testing should be done on all women, regardless of age. Although HPV is a sexually transmitted disease and much more common in younger than older women, a positive HPV finding can be more significant in an older woman, Dr. Lozano says. "Although fewer [older women] are positive, when they are, there’s a higher risk of high-grade lesion and cancer," he says.

A study of the cost-effectiveness of different triage strategies for ASC-US, published in the May 8 issue of JAMA ("Cost-effectiveness of alternative triage strategies for atypical squamous cells of undetermined significance"), concluded that reflex HPV testing provides the same or greater life-expectancy benefits as the alternatives, including colposcopy and repeat Pap, and is more cost-effective than other strategies. Biennial liquid cytology with reflex HPV testing had a cost of $174,200 per year of life gained, compared with more than $200,000 for colposcopy.

The study also found that annual screening "provided minimal incremental life-expectancy gains," especially when compared to biennial liquid Pap screening. It concluded that shifting those women who are being screened annually with conventional cytology coupled with repeat Pap for ASC-US to biennial liquid cytology and reflex HPV testing "would save more than $15 billion over the lifetime of a typical cohort of 18-year-old to 24-year-old women."

Educating clinicians

Laboratories play an important role in implementing the new guidelines. While an HPV test can’t be performed without a clinician’s order, laboratorians can help speed the process by educating clinicians, informing them that the test is available, and adopting forms that make it easy to order the reflex test.

Dr. Bolick, whose presentation at the CAP conference will cover laboratory considerations in HPV testing, says reflex tests can be ordered by calling the clinician if the Pap test is abnormal, using a check-off box, or using a standing order signed by the clinician. The latter two are preferable for the laboratory.

"What you can’t do is restrict the ability of a clinician to make a choice," he says. "You have to reflex test based on a clinician’s instructions."

Dr. Bolick adds that laboratories should move to liquid-based cytology and reflex HPV testing. "If you’re a lab doing Pap smears, you should offer the best technology available to your clinicians," he says. "If you can’t do it yourself, then make arrangements with another lab."

Laboratories will want to analyze their volume to determine if HPV testing is cost-effective. "We’ve found the threshold to be roughly 10 to 15 HPV tests per run," says Dr. Bolick. That is, the lab has enough volume to perform that many tests during the six to eight hours it takes to run the assay. He estimates that if a lab’s ASC-US rate is around four percent, which is typical, "you would have to be running 400 liquid-based Pap smears a week before this [HPV] would be cost-effective in your lab."

That calculation, however, is based on the assumption that all clinicians will order reflex HPV testing. "My experience is that only one-half of them do it now," Dr. Bolick says. "With the new guidelines, and by making it convenient for them to order, we should be able to get that up to 75 or 80 percent."

HPV is a labor-intensive assay that takes one technologist a full day to complete, although multiple samples can be run at once. "We do 80 or 90 on a run all the time," says Dr. Bolick. His lab, which performs HPV for about 40 other labs nationwide, does 1,500 Pap tests a week. For most labs, "if you’re doing 30 to 40 HPV tests a day, it becomes very cost-effective for you to offer that test."

This applies to reference and hospital laboratories, he says. Even if the HPV test is not quite break-even, "you may choose to do it to offer full service to your clients. In my opinion, if you don’t offer liquid cytology and HPV testing in a surgical cytology practice, you’re not offering the standard of care."

Pathologists should also become knowledgeable about HPV itself, says George Birdsong, MD, director of anatomic pathology at Grady Hospital, Atlanta, which is affiliated with Emory University. "We can help clinicians explain to patients some of the natural history of HPV," he says. (In most women, HPV is cleared from the system in 12 to 18 months.) "As HPV testing becomes more prominent, we will be counseling women whose Pap is negative but whose HPV is positive," he adds.

Laboratories can also inform clients that HPV testing is available and make clear when it should be used and when it’s not beneficial. For example, "there has not been a demonstrated benefit for HPV testing with HSIL," Dr. Birdsong says. (The guidelines recommend colposcopy with endocervical assessment of women with HSIL.)

As a gynecologist, Dr. Cox agrees that it’s appropriate for pathologists to get involved in educating others about HPV testing. "It would be very helpful if labs were to send out to all their clients a short description of what an HPV test is and a copy of the JAMA guidelines," he says.

Additional uses for HPV testing

Disagreement continues over the use of HPV as a primary screen and its role for managing abnormal Pap test results other than ASC-US, such as atypical glandular cells. The May 8 JAMA article "Benefits and costs of using HPV testing to screen for cervical cancer" recommends biennial screening with HPV plus Pap. Screening every two years using a combined HPV-Pap test costs $76,183 per year of life saved. (Discontinuing biennial screening with HPV and Pap at age 75 captures 97.8 percent of the benefits of lifetime screening at a slightly lower cost of $70,347 per year of life saved.)

The study concluded: "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. ... Pap results have been noted to have low sensitivity, poor reproducibility, and high potential for misclassification. Thus, parallel screening with cytology and HPV testing improves outcomes by increasing sensitivity ... where the additional savings in life-years are achieved at a reasonable incremental cost."

So far, though, the FDAhas not endorsed Pap plus HPV screening. Digene, which makes the Hybrid Capture 2 HPV DNA test (and plans to merge with ThinPrep maker Cytyc), received a conditional recommendation for approval of the test, known as the DNA Pap, from an FDA advisory panel in March. Until the conditions are met, the FDA won’t rule on the recommendation.

Dr. Birdsong, a member of the advisory panel, says members were divided on the combined test. Discussion was contentious, he says, and resulted in a split vote. The final decision was to recommend approval, with conditions, of the HPV-Pap for women 30 and over. (HPV is so common in younger women that a positive finding yields little information.) Conditions for Digene to obtain final approval are as follows:

  • Develop an algorithm for using the test in clinical management—that is, how often screening should take place under various combinations of test results, such as negative on Pap but positive on HPV or negative on both. "Digene suggested that if women are negative on both tests, then put them in the lowest risk group, regardless of the management protocol being followed," says Dr. Birdsong. "But the majority of the panel felt that a specific protocol needs to be specified. A particular problem," he adds, "was that the company didn’t define what should be done with women who are negative on the Pap but high-risk positive on the HPV test." (Digene says its proposed labeling recommends that such women be followed up with a repeat DNA Pap test in six to 12 months.)
  • Demonstrate that the combination test has an impact on clinical outcome by showing that women who are negative on Pap and HPV don’t develop cancer. "Digene had not done any specific study for this presentation," says Dr. Birdsong. "The majority of the panel wanted more evidence from either existing data or a new study that this [double test] was beneficial."
  • Include educational material with the double test to educate laboratorians and clinicians as well as patients.
  • Conduct postmarket surveillance to assess the impact of the DNA Pap on clinical outcomes.

Digene "will have to come back to the FDA with the additional information," says Dr. Birdsong. The agency will then determine whether to approve, reject, or convene another advisory panel. Meanwhile, clinicians "can continue to use it off-label."

At Grady Hospital, "we’re not yet doing that [the double test]," says Dr. Birdsong. In fact, Grady isn’t offering reflex HPV testing or routine liquid Paps. Dr. Birdsong hopes that, with the new guidelines, the hospital will quickly adopt both, as he has been urging them to do. As far as using HPV as a primary screen, "we would certainly make it available to our clinicians," he says, "but I don’t know if I would push for it yet."

Other pathologists already advocate using HPV with Pap as a primary screen, most notably Dr. Bolick.

"There are many uses for HPV as a screen," he says. "We’ve found that a woman with a normal Pap and the presence of oncogenic HPV has a 20-fold greater risk over the baseline population of developing a high-grade lesion in the next two years. If a clinician wants to know that information, it’s relevant to do an HPV test with every Pap."

The HPV test provides information that a Pap test doesn’t, he says, including whether the woman is positive for oncogenic (high-risk) or non-oncogenic HPV. It appears, says Dr. Bolick, that non-oncogenic HPV may inhibit oncogenic HPV and help prevent cancer. "We’ve found that a woman who has both types of HPV and a normal Pap has one-tenth the risk of developing a high-grade lesion, compared to a woman with oncogenic HPV only," he says.

Dr. Bolick gives this example of how HPV testing can guide a diagnosis. "If you believe you have HSIL in a patient who’s HPV negative, you better be very concerned about that diagnosis," he says. "Our data show that only eight percent of women with HSIL do not have oncogenic HPV."

For combined HPV-Pap testing to be broadly adopted, however, Dr. Bolick believes the cost must come down and the test must be improved by providing needed information. This includes obtaining answers to the questions: What is the viral burden of oncogenic and non-oncogenic HPV? If it’s oncogenic HPV, is there cell transformation?

Dr. Cox, who was a presenter to the FDA advisory panel, is another advocate of using HPV more broadly. He suggests liquid-based cytology every two years until a woman is 30 or 35 and then using an HPV test every three years with the Pap test. "That gives you nearly 100 percent negative predictive value," he says.

Dr. Cox agrees with Dr. Bolick that the HPV test has predictive value that the Pap test lacks. "With the HPV test, you can say whether you’re more likely or less likely to develop disease," he says. In fact, some experts believe "you could use HPV without the Pap as a primary screen, then do the Pap as a secondary test," Dr. Cox says. "But that would take a considerable amount of retraining of clinicians and the public to accept that."

Dr. Cox also thinks that atypical glandular cells in some women could eventually be managed by HPV testing, "but we don’t have enough data to support that yet." He ran the AGC section of the consensus guidelines, which did not recommend an HPV reflex test for that finding. If HPV testing is used with AGC, it would have to be in women under 35, Dr. Cox says. That’s because in older women when an abnormality is detected in the evaluation of AGC, it is more likely to be endometrial in origin, which HPV cannot detect.

But in women under 35, "you’re worried about endocervical adenocarcinoma in situ," he says, which has a similar rate of HPV detection as high-grade squamous intraepithelial neoplasia. "At this point," he adds, "until more data are available, I can see only using HPV testing as a followup for women with AGC-NOS [atypical glandular cells, not otherwise specified] who are negative on colposcopy and endocervical sampling to provide some measure of increased reassurance when negative and increased vigilance if HPV-positive." In contrast, women with AGC"favor neoplasia" or AISand negative on colposcopy should have a cone or LEEP, he says.

Dr. Bolick is of the same opinion. "There’s a lot of promise in using HPV testing for AGC, but we need about 10 times the amount of data," he says. While ASC-US makes up three to five percent of all Pap test results, AGC is only 0.3 to 0.5 percent. At his lab, Dr. Bolick says, "my data show that 100 percent of patients with endocervical adenocarcinoma in situ have oncogenic HPV."

What is clear is that use of HPV tests will increase dramatically in managing cervical cytological abnormalities. This means that clinicians and laboratorians will have to explain to women why they’re undergoing sexually transmitted disease testing and the trade-offs in not doing it. Says Dr. Bolick: "It’s a woman’s right to know if she has HPV. It’s an appropriate test for clinicians to offer all women."

Karen Southwick is a writer in San Francisco.