College of American Pathologists
Printable Version

  Feature Story


cap today

Two roads for HPV testing, but only one is heavily traveled

November 2003
Karen Southwick

The use of HPV DNA testing as a followup to an abnormal Pap test result is gaining acceptance among clinicians as the standard of care and generally is reimbursed by insurers, according to the vendor and pathologists who are performing the tests.

However, even though the American College of Obstetricians and Gynecologists and the American Cancer Society have acknowledged the use of HPV as a screening test for cervical abnormalities in women 30 and older, that use of HPV is likely to be accepted more slowly.

Several groups have recognized the clinical utility of HPV as a reflex test for women whose Pap test is defined as atypical squamous cells of undetermined significance, or ASC-US, including ACOG, ACS, and American Society for Colposcopy and Cervical Pathology, or ASCCP.

The consensus guidelines for reflex testing were published in 2002 after the ASCUS-LSILTriage Study concluded that triage with HPV after an ASC-US Pap test result was a viable option. Later studies, published in the Journal of the American Medical Association, suggested that HPV might have value as an additional screening test in older women and permit those with double negatives (on the Pap and HPV tests) to be tested at longer intervals of two or three years.

Doing an HPV reflex test has become "the standard of care for managing ASC-US," says David Bolick, MD, medical director for AmeriPath-Utah, Salt Lake City. "It’s also the most cost-effective way of managing ASC-US." Women not tested for HPV as a followup to ASC-US are usually referred for repeat Pap tests every three to six months or for colposcopy, while if their HPV result is negative, they can return to annual Pap test screening.

Digene Corp.’s sales figures back up the notion that HPV reflex testing has become the norm. The company’s Hybrid Capture 2, the only FDA-approved HPVDNA test, had 38 percent of the ASC-US marketplace in the United States and Canada as of June 2002, reports Jack Butler, director of North American sales. By the end of June 2003, that share had grown to an estimated 62 percent. Once the share passes 50 percent, "many consider that standard of care," he says.

"Most physicians around the country have adopted HPV for reflex testing for ASC-US, although certainly there’s still room for growth," says Susan Garfield, director of women’s health and reimbursement for Digene, based in Gaithersburg, Md. "Reimbursement for this indication is widespread."

In September, Digene launched its DNAwithPap test, approved by the FDA in March, for primary screening in women 30 and older, in accordance with ACOG and ACS guidelines. Garfield says one major insurer, Aetna, already has come out with a coverage policy for HPV in this setting. "We expect other insurers to follow in the next several months," she says.

Butler reports that market share for screening is still low, since Digene’s combined test was launched only recently. "Together, a negative Pap test and a negative Hybrid Capture 2 HPV test have a negative predictive value of 99.2 percent," he says. "The ACOG and ACS guidelines suggest that this high negative predictive value can allow women age 30 and older to go to longer screening intervals." (Younger women have a higher incidence of benign HPV infection, which is why HPV screening is not recommended for them.)

Garfield says more than 50 million people are already covered by payers for the new screening indication. "The test provides incredible value for payers," she says. "The majority of women are going to be double-negative, so scarce resources can be used more effectively, especially in the public sector, where disease burden is highest. Decreasing unnecessary followup leads to cost savings."

Digene’s HPV test can be run either from a liquid-based cytology sample or from Digene’s own transport media, called Standard Transport Media. "Our STM sample makes it possible for users of conventional Pap smears to take advantage of the new combined HPV and Pap screening indication on women age 30 and older," Garfield says.

When HPV DNA testing is used to manage women with ASC-US, a liquid-based cytology sample is considered by many to be the best collection medium. But when HPV DNA testing is used as an adjunct to cytology for primary screening, "a separate specimen placed in its own transport tube is preferable," says Thomas C. Wright Jr., MD, associate professor of pathology at Columbia-Presbyterian Medical Center, New York. Three to 10 percent of liquid-based cytology specimens have insufficient residual fluid after processing for HPV DNA testing, he says. "Telling three to 10 percent of all women that they need to return to their clinician’s office and undergo another pelvic exam is unlikely to be considered acceptable," Dr. Wright says. And liquid-based cytology samples could become cross-contaminated during processing in the lab. "Although that’s undesirable in all circumstances," he says, "the clinical significance of a one percent to two percent contamination rate is quite different when HPV DNA testing is used for managing women with ASC-US as opposed to when it is used for primary screening."

Pathologists are finding that HPV as a reflex test is broadly accepted by clinicians, but HPV screening is in its early stages and may never achieve the same level of acceptance. Several pathologists point out that abandoning a yearly Pap test would require a sea change in the way most gynecologists practice and in the annual exam to which most women are accustomed. Consequently, it will be more difficult to gain acceptance as compared with merely adding HPV as a new test in ASC-US cases.

Garfield concedes that’s a barrier. "We feel there are many reasons why women should see their doctor annually, besides the Pap," she says. "We’re going to be working with patients and physicians to educate them about the need for an annual visit," even as Digene seeks to move them toward HPV screening and three-year Pap/HPV for women with double negatives.

Dr. Bolick, whose AmeriPath-Utah laboratory serves seven hospitals in three states, says he and others in the laboratory had to educate clinicians to move toward HPV reflex testing and will have to do the same for primary screening.

"We talk to our clinicians about the utility of HPV testing and ask them to sign standing orders for reflex HPV testing on different types of Pap abnormalities," he says. Most have complied because reflex HPV testing is clearly "faster, better, and cheaper" than the alternatives, which include expensive, invasive colposcopy or inconvenient followup Pap tests every three to six months.

For the reflex test, "very few clinicians are skeptical at this point," Dr. Bolick says. "Nowadays well over 90 percent of our clients are using standard reflex testing. And we don’t have any reimbursement problems with the majority of insurance companies."

Dr. Bolick is an advocate of polymerase chain reaction testing for HPV because his data indicate, he says, that it has fewer false-negative results and greater specificity than the Hybrid Capture 2. It can also help distinguish between intermediate- and high-risk types of HPV, which Digene’s DNA test cannot, he says. There are no FDA-approved PCR tests for HPV DNA, however.

He also recommends using HPV in a broader way for gynecological testing, including primary screening, but he concedes that clinicians "are going to use what makes them comfortable."

Mark Stoler, MD, professor of pathology and gynecology and associate director of surgical pathology and cytopathology at the University of Virginia Health System, Charlottesville, agrees with Dr. Bolick that educating clinicians about HPV screening will take considerable effort.

Moving to HPV as a primary screen "would have a much bigger impact on clinical practice," he says. Going to a three-year Pap/HPV screen "would decrease clinicians’ business," he says. "If they still want to see women annually they’re going to have to think up another reason."

Dr. Stoler says, however, that a combined test has "almost absolute negative predictive value and reasonable positive predictive value." Adding HPV to Pap improves sensitivity from the 50 to 80 percent range to the 98 to 99 percent range, he says. For those reasons, he believes an HPV screening indication will start to "take off" within the next few years.

Meanwhile, HPV reflex testing is catching on, he says. At the University of Virginia Health System, which does about 30,000 Pap tests annually, more than half of clinicians are checking a box on their requisitions to do HPV for ASC-US results. Some are still resistant, he says.

"We don’t have 100 percent compliance yet, although we would like to get it," Dr. Stoler says. "We can’t force them to do it. If we have a group that’s particularly reticent, we will call them up and offer to do a presentation." Most are convinced when the laboratory presents data from the ASCUS-LSIL Triage Study. "That offers good, statistically valid data that can persuade clinicians and insurance companies," he says.

Since the HPV testing began about two years ago, the Pap test volume has held steady or decreased slightly, Dr. Stoler reports. Previously, the volume had been increasing five to seven percent a year. The change indicates that HPV reflex testing curbs the need for repeat Pap tests and lowers the volume, he says.

For laboratories themselves, HPV testing can be an important quality control indicator, Dr. Stoler notes. "HPV allows you to independently validate whether you’re overcalling or undercalling ASC-US." With ASC-US patients, 40 to 60 percent should be HPV high-risk positive, he says. "If your rate is higher, then you’re being too specific and you’re undercalling. If it’s lower, you’re being too sensitive and you’re overcalling."

On the question of PCR versus Hybrid Capture, Dr. Stoler says it can be hard to validate the results of a "homebrew PCR," since there is no standardized version yet on the market. "The increased sensitivity is a two-edged sword if you can’t validate it," he says.

Attila Lorincz, PhD, chief scientist at Digene, says there is little or no clinical value to distinguishing between intermediate- and high-risk types of HPV and hence these types are all grouped into the high-risk probe set in the HC2 test. It was his research team that first discovered and detailed the classification of HPVs into low-, intermediate-, and high-risk types in a 1992 article in the American Journal of Obstetrics and Gynecology. "There is nothing different that a clinician should do for a woman infected by an intermediate-risk HPV type versus a high-risk HPV type," Dr. Lorincz says. Using "overly sensitive" PCRtests to detect additional HPV infections, he says, will lead to an increase in the rate of clinical false-positives, "which is not in the best interests of the patient or the clinician."

At Johns Hopkins Medical Institutions, Baltimore, reflex HPV on ASC-US has become nearly universal, reports Dorothy L. Rosenthal, MD, professor of pathology, oncology, gynecology, and obstetrics. Johns Hopkins, which serves an urban population, finds that about 6.5 percent of its Pap tests are high-risk ASC-US. "We’re doing HPV on virtually every single one of them," she says. "The only time we wouldn’t do it is if the woman had a previous positive [HPV] test within a year."

If the HPV is negative, "she goes back into the annual repeat Pap cue," says Dr. Rosenthal. If positive, "she goes to colposcopy," which is what the guidelines that arose from the ASCUS-LSILTriage Study, or ALTS, recommend. Overall, since Johns Hopkins began HPV reflex testing two years ago, it has eliminated about 50 percent of cases that previously went to colposcopy or frequent Pap tests. Reimbursement for HPV reflex testing "has not been a problem," Dr. Rosenthal says.

Over the two years of offering the test, physician acceptance has been dramatic, she says. "At first we had to twist some arms, but now clinicians will go ballistic if we don’t do the test [for ASC-US]," she says. Johns Hopkins physicians were educated through memos and grand rounds in which pathologists presented the ALTS data.

In addition, it’s a convenience for patients. "This can triage women immediately. There’s no delay" in running the test because it can be done on the Pap test sample, Dr. Rosenthal says.

At present, Johns Hopkins is using HPV as a primary screen at its HIV clinic. "Each [female] patient there gets an HPV test with her Pap because this population is at terrifically high risk for cervical cancer," Dr. Rosenthal says.

In a retrospective study that looked at 192 of these HIV patients over a six-month period, all had HPV testing and 25 percent had biopsies. Twelve percent of these patients were under age 30. The cytologic breakdown for the group was as follows: negative = 56 percent, ASC-US = 16 percent, LSIL = 20 percent, and HSIL = six percent, for a total abnormal rate of 42 percent, much higher than the usual rate of 12 percent. Of all 192 women, 55 percent were HPV-positive.

"Recognizing that the [HPV] rate in the cytologically negative population is probably around 20 percent, our HIV population is really high risk, even without looking at the cytologic diagnoses," Dr. Rosenthal says. Thus, HPV screening is a useful tool for this group. "Our experience with this high-risk population is enough to make a believer out of even the most skeptical critic of HPV testing," she says.

The system is also doing HPV screening for women with high-risk ASC-US, called ASC-H. "In the ALTS trial, that was not considered a target population, but we’re finding that [HPV screening] helps eliminate unnecessary LEEPs," says Dr. Rosenthal. At Johns Hopkins, if a woman with ASC-H has an indeterminate biopsy result and colposcopy, "we do HPV," says Dr. Rosenthal. If that is negative, "we continue to watch her." If positive, "we go to LEEP."

At Evanston (Ill.) Northwestern Healthcare, where the laboratory serves three hospitals in the northern Chicago suburbs, about 80 percent of clinicians are ordering a reflex HPV test after an ASC-US Pap result, says Charles D. Sturgis, MD, director of cytopathology.

The physicians can check a box that orders the reflex as a matter of course or order it after they receive the report. "We don’t contact them regarding the test," says Dr. Sturgis. He has done two grand rounds with gynecologists and family practitioners on HPV testing.

The 20 percent of clinicians that aren’t using HPV testing include a group of physicians who are resistant to switching to liquid-based cytology or HPV, Dr. Sturgis says. And there’s another group that has switched to liquid Pap tests but still prefers to go to colposcopy or repeat Pap rather than use HPV testing.

Evanston Northwestern serves a largely suburban population and experiences a lower squamous atypia rate than that reported by Dr. Rosenthal at Johns Hopkins. For the first six months of 2003, the system performed 19,314 Pap tests, of which 816, or four percent, were ASC-US. About 22 percent of those end up positive for high-risk HPV, Dr. Sturgis says, and they undergo colposcopy. The other 78 percent return to the annual screening population.

"Most of our patients have private medical insurance and we have not had any problems getting reimbursement," he says. Evanston Northwestern Healthcare is not using HPV as a primary screen. "HPV testing may one day be possible as a primary screening test, but many other factors, such as the future use of vaccines and cost-effectiveness studies on the topic, would need to be considered," says Dr. Sturgis.

Pathology and Cytology Laboratories Inc., Lexington, Ky., has had a similar experience. About 90 percent of clinicians have moved to liquid-based cytology and about three-fourths of those order an HPV test up front for ASC-US, says pathologist Richard Lozano, MD. In cases where HPV is not ordered as a reflex, "we put a comment in our report, including the ASCCP consensus guidelines, for them to call back and order HPV testing," he says. "Nearly all of them do."

Insurers are willing to pay for the test, he says. "It helps that Medicare has listed it as an accepted test, and insurers have fallen in line with that."

However, the HPV screening indication is still so new that "I haven’t really gotten the pulse of clinicians on using it," he says. While his laboratory does plan to start offering HPV as a screening option, "we don’t know what the feedback will be." Getting them to accept reflex testing required a lot of education. "We will probably offer the screen, and let them decide to use it or not," he says.

Labs had a role, he says, in driving use of HPV reflex testing because it offered "improved patient care." Using HPV as a screen is more an issue of convenience than improved patient care, so labs aren’t as likely to push it as hard, Dr. Lozano believes. "There’s the possibility that HPV could be overused because patients come in asking for it when they don’t need it. It’s incumbent upon us as pathologists to make sure it’s used correctly."

Making sure it’s used correctly is not an easy task. Most clinicians at the University of Kentucky Chandler Medical Center, Lexington, order HPV reflex testing for ASC-US. Despite regular informational memos, there is still confusion about appropriate management of the patient after HPV results are reported, says Diane D. Davey, MD, director of cytology.

"Some women with negative results come back for repeat Pap in four to six instead of 12 months, and there are also a few requests for HPV testing right after a SIL diagnosis or repeat testing within a short time interval, which is probably not useful. I think there may be a role for using HPV testing in some older women with LSIL, but not for the general SIL case," she says.

Unlike the others, Coastal Pathology Laboratories, Charleston, SC, has had reimbursement problems with HPV as a reflex test. R. Marshall Austin, MD, PhD, medical director of the laboratory, says the state’s primary insurer, BlueCross BlueShield, has been resistant to paying for the test.

"We started doing the HPV reflex test in 2002," Dr. Austin says, after educating physicians about its use. About 90 percent of clinicians have moved to liquid-based Pap tests and most of them are also ordering reflex HPVs for ASC-US. "The data [in favor of HPV reflex] are overwhelming," he says. "There’s no scientific debate, just a reimbursement debate."

While there has been no problem getting payment for liquid Pap tests, BlueCross BlueShield has not been paying for reflex HPV testing, Dr. Austin says. Other insurers, including Cigna and Aetna, have paid, he adds. "Our position has been to continue to offer the test and to lobby through the media and local politicians to get BlueCross to change its position."

Elizabeth Hammond, a spokesperson for BlueCross BlueShield of South Carolina, says the insurer does cover HPV testing if it is performed after an ASC-US result. "We’re not paying for it as a primary screen," she says. "In fact, a review shows that we have paid and are paying South Carolina providers, including Dr. Austin’s group, per our medical policy."

Karen Southwick is a writer in San Francisco.