College of American Pathologists
Printable Version

  Feature Story


cap today

Leaning toward HPV triage to clear up ASCUS

HPV triage data

May 2001
Karen Southwick

Clear-cut results from a recently completed study involving women with abnormal Pap tests should accelerate the push to develop practice guidelines on what to do in such cases.

About four percent of the 50 million Pap tests performed annually in the United States are classified as ASCUS, or atypical squamous cells of undetermined significance. Currently, there is no definitive treatment. The options are colposcopy and biopsy; repeat Pap tests at four-to-six month intervals, with colposcopy if the results continue to be abnormal; or triage using DNA testing for cancer-associated human papillomavirus types.

Sponsored by the National Cancer Institute, the ASCUS/LSIL Triage Study (ALTS) compared these three approaches for timely detection of cervical precancerous lesions. The 3,488 women with an ASCUS diagnosis that enrolled in the study were randomly assigned to undergo one of the three options, with immediate colposcopy considered the reference standard.

In the HPV arm, sensitivity to detect cervical intraepithelial neoplasia grade 3 (CIN3) or worse was 96.3 percent, while referring 56.1 percent of women for colposcopy. Sensitivity of a single repeat Pap specimen at a triage threshold of HSIL was 44.1 percent, with 6.9 percent referred for colposcopy. Sensitivity using a lower threshold of atypical cells or worse on the Pap test was 85.3 percent, with 58.6 percent referred. The study concluded that HPV testing is "a viable option for the management of ASCUS."

"For the majority of practitioners, this is news because there’s a valid information base for both people in the lab and the clinician," says R. Marshall Austin, MD, PhD, director of cytopathology and gynecologic pathology services at Coastal Pathology Laboratories, Charleston, SC. While "everyone agrees what needs to be done with high-grade lesions and cancer, there’s much more variation with low-grade abnormal results."

Diane Solomon, MD, ALTS project director, says these early findings from the study only scratch the surface. "We just reported the enrollment results," she says, in an article published in the Feb. 21 issue of the Journal of the National Cancer Institute. "There will probably be 20 more papers coming out of this study."

Nonetheless, she adds, the results so far indicate that HPV triage should be considered by women and their physicians. "If you get an ASCUS result and the HPV test is negative, that provides the woman 99.5 percent reassurance that she doesn’t have a high-grade lesion," Dr. Solomon says. In that case, going back to an annual Pap test would probably be indicated.

If the HPV test is positive, "it doesn’t necessarily mean the woman has a high-grade lesion, but it does indicate that she requires additional followup," usually a colposcopy. "There’s a 10 to 20 percent chance that the woman has a lesion that needs to be treated," she says.

HPV triage is especially useful for older women—those over 35—"because the specificity of the test improves in older women," Dr. Solomon adds. (HPV is far more common among women in their late teens and early 20s who may not have cervical abnormalities.) In fact, other studies are considering whether to use HPV as a primary screening tool in older women, replacing the Pap test, which can yield ambiguous results on cervical changes caused by menopause.

Dr. Solomon doesn’t expect the ALTS cost-effectiveness analysis to be completed until 2002. Currently, HPV testing is two to three times as expensive as a single Pap test but cheaper than a colposcopy. However, the HPV cost could decline if volume increases.

In addition, HPV testing, rather than colposcopy or repeat Pap tests, could save the woman another visit to her physician. If liquid cytology is used, the HPV test can be performed on the same sample. With a conventional Pap smear, the clinician must obtain two separate samples if an HPV test is anticipated, or ask the patient to return for a second sample.

Consensus conference

Getting clinicians to change their practice methods is never easy, but the study should encourage use of HPV as an ASCUS triage, particularly after a consensus conference in September that’s expected to put the stamp of approval on HPV testing.

Representatives of medical groups involved with cervical cancer, including the CAP and American College of Obstetricians and Gynecologists, are expected to attend the consensus conference of the American Society of Colposcopy and Cervical Pathology, or ASCCP. "The aim is to develop consensus guidelines for the management of cytological abnormalities and cervical cancer precursors. This will include dealing with management of ASCUS," says Edward J. Wilkinson, MD, ASCCP president and director of anatomic pathology and cytology at the University of Florida College of Medicine, Gainesville.

ASCCP’s current guidelines for ASCUS followup list the three treatment options and the pros and cons for each. The repeat Pap test, for example, may delay the diagnosis of cervical cancer. Immediate colposcopy is more costly and invasive than the other alternatives. HPV triage also has its negatives, such as an additional office visit if a liquid sample was not obtained initially.

The ASCCP conference is scheduled for Sept. 6-9 in Washington, by invitation only. The experts who attend will develop guidelines, and they’ll be presented to the ASCCP annual meeting in March 2002 and distributed to other medical groups for their comments.

The ALTS results will be included in the meeting agenda, Dr. Wilkinson says, praising the NCI trial as "one of the best controlled and largest studies ever done in this area." Because it included "approximately 5,000 patients with two years of followup data and was focused on management of these [cervical] abnormalities, the results are very, very important."

He adds, "The conclusion at this point would be using HPV after ASCUS." Although HPV "is not a diagnostic test, it does tell you who’s at risk [for a high-grade lesion]," he says. "The negative predictive value of HPV is very strong. The positive predictive value is more difficult to judge."

One expert, though, cautions that using HPV in this context could present difficulties. HPV is a sexually transmitted infection, and "STDs still have an associated stigma in our society," notes Elizabeth Unger, MD, PhD, acting chief of the HPV section at the Centers for Disease Control and Prevention. If Pap tests become associated with testing for an STD, that might scare off some women from undergoing the screening.

Then, too, HPV testing "is not likely the answer to reducing cervical cancer," Dr. Unger adds. For that, increased screening of hard-to-reach at-risk women is needed.

She believes that U.S. clinicians "probably overtreat women with cervical lesions that have very little chance of progressing to cancer." Dr. Unger suggests that HPV triage could be a useful "interim step" in moving toward molecular markers that are more specific for precancerous lesions that are likely to invade. "We shouldn’t bill this [HPV triage] as a perfect solution," she says.

Practice patterns

At the University of Florida, current practice is to schedule a woman with an abnormal Pap for colposcopy, Dr. Wilkinson says. "Our gynecology department will have to make some decisions now that we have these results. Where you have a high-risk population [ASCUS rates above five percent], it makes sense to look at HPV triage."

The ALTS trial could have an impact also in encouraging more use of liquid cytology rather than a conventional Pap, since the same sample of the former can be stored to do HPV testing. Or clinicians can be trained to obtain two cervical samples during an exam. "Both methods would work," Dr. Wilkinson says.

Dr. Austin is an enthusiastic proponent of broadening the use of HPV triage and of moving to liquid Pap testing. "The savings appear in your system costs," and in making it easier for the patient who has an ASCUS result. Though liquid cytology is more expensive than a conventional Pap test, "you have to take the cost of a repeat visit into account, and the time the patient must spend away from work or home."

"This will catch on as health systems and third-party payers realize it is cost-effective," Dr. Austin maintains. In addition, he says, liquid cytology will become the standard because it is more specific and more sensitive. "Last year our detection of abnormal high-grade lesions was 85 percent higher than with a conventional smear."

Coastal Pathology charges $30 for liquid cytology and $15 for a conventional Pap test. The HPV test is about $75. But Dr. Austin says the additional testing is worth it because, "if you take a combination of ThinPrep [a liquid cytology test] and HPV, you can detect almost 100 percent of cancers and high-grade lesions. No other method has ever been able to accomplish that level of sensitivity."

For women over 35, using HPV testing "can virtually eliminate false negatives." And, with the incidence of endocervical cancer increasing, the HPV test is a better indicator than the Pap test, which is only about 50 percent sensitive to that type of cancer, Dr. Austin says.

Reimbursement is uneven, both for liquid cytology and for HPV testing, he acknowledges. "We’ve had to convince payers that it was worth covering liquid cytology because it’s better at detecting a problem. Now they pay for that here." He sees the same educational process occurring with HPV triage.

Cunningham Pathology, which serves five major hospitals and many outpatient clinics in the Birmingham, Ala., area, was an early adopter of HPV testing. "The ALTS results definitely strengthen our case," says Rick Lozano, MD, director of cytopathology at Cunningham. In fact, he was familiar with ALTS even before the results were published because he trained at the local University of Alabama at Birmingham campus, where one of the clinical centers of the trial was located.

"I knew the study design and the presumed outcome," he says. Consequently, about two years ago, Cunningham went to 100 percent liquid cytology, with HPV triage for ASCUS. Now, a cytopathologist reviews every ASCUS diagnosis. "If the cytopathologist agrees it’s ASCUS, it goes for HPV typing," he says.

Cunningham’s reference lab, Molecular Pathology in Maryville, Tenn., which handles the HPV testing, helped to educate clinicians about the test’s importance. "Our main point was that an ASCUS diagnosis continues to be a treatment dilemma," Dr. Lozano says. By funneling ASCUS results through a cytopathologist and then HPV triage, "we are working hard to eliminate the ASCUS diagnosis."

Since the new process went into effect, in 49 cases where the preliminary diagnosis was ASCUS, the HPV typing was used to change the diagnosis to something more definitive, such as low-grade or high-grade lesion. (see "HPV triage data," side story.) "We now regard ASCUS as a temporary diagnosis until we can refine it to something more definitive," Dr. Lozano says. That means clinicians don’t have to deal with "confusing data" or "clouds of verbiage" about an unclear result.

Insurers have reimbursed for HPV testing because Cunningham’s ASCUS rate is less than one percent, thanks to the initial triage by a cytopathologist. "If we had an ASCUS rate of 40 percent, I’m sure they’d start protesting," he says.

Once the ALTS results were published, Diane D. Davey, MD, director of cytopathology at the University of Kentucky Chandler Medical Center, Lexington, sent out a memo to clinicians in her system suggesting broader use of HPV testing.

She wrote: "Preliminary results of the National Cancer Institute ALTS trial confirm that HPV testing is a useful triage for patients with ASCUS on cytology. Our laboratory has been offering adjunctive HPV testing for patients with ASCUS or a history of abnormality for several months using the remainder of the liquid-based ThinPrep vial.... Our recommendation is that HPV testing be done the first time the patient has a cytology result of unqualified ASCUS."

Dr. Davey says adopting guidelines at the ASCCP consensus meeting should also drive use of HPV testing. "Clinicians will get these guidelines at their meetings, and they’ll be more accepted," she adds. The ALTS results and the guidelines should also help in obtaining "appropriate reimbursement" for liquid cytology and HPV testing. In her experience, payers often pay for an HPV test but at a level insufficient to cover the typical reference laboratory’s charges.

Karen Southwick is a freelance writer in San Francisco.