College of American Pathologists
Printable Version

  Feature Story


cap today

Closer to consensus on abnormal Pap tests

October 2001
Karen Southwick

Experts at two recent meetings have clarified the terminology used to report abnormal results on Pap tests and developed new consensus guidelines for how clinicians should manage patients with those results.

Taken together, the meetings represent significant progress in achieving agreement among pathologists, clinicians, and other experts in dealing with about 2 million abnormal Pap test results reported annually in the
United States.

The first meeting, held in May, was the Bethesda System 2001 Workshop sponsored by the National Cancer Institute, which focused on terminology and specimen adequacy. The second meeting, in September, was the ASCCP Consensus Conference for the Management of Cytological Abnormalities and Cervical Cancer Precursors. An invitation-only gathering, it was sponsored by the American Society for Colposcopy and Cervical Pathology (ASCCP). CAP representatives attended both meetings.

"It’s the first time there has been such close, complementary development of lab terminology and management guidelines for abnormal results," says Diane Solomon, MD, a senior investigator for NCI who organized the Bethesda workshop attended by about 400 people. "It represents landmark collaboration between pathologists and clinicians."

Though laboratorians and clinicians were involved in both conferences, laboratorians were in the majority at the Bethesda workshop and clinicians were in the majority at the ASCCP conference. "There was very vigorous debate on some issues," Dr. Solomon says, "but consensus was reached in the end."

Bethesda changes

Among the changes adopted at the Bethesda workshop was replacing the previous dichotomy of ASCUS as favor reactive or favor SIL with a single major category: "atypical squamous cells." Two subcategories are "atypical squamous cells, undetermined significance" and "atypical squamous cells, cannot exclude HSIL." The commentary "favor reactive" on ASCUS will be eliminated.

That notation "sent mixed messages to clinicians," says Diane Davey, MD, professor of pathology and laboratory medicine at the University of Kentucky Chandler Medical Center in Lexington and chair of the CAP’s Cytopathology Committee. She attended both meetings. "Atypical, favor reactive" meant the sample could not be called "normal," but at the same time, "we didn’t think there was anything seriously wrong," Dr. Davey says. Consequently, "some women weren’t followed aggressively enough."

Other Bethesda System language that was eliminated are the terms "within normal limits" and "benign cellular changes." These will be replaced with a single term, "negative for intraepithelial lesion or malignancy." Here, too, says Dr. Davey, the aim is to eliminate confusion and mixed messages. Report comments should include discussion of whether organisms are present, such as Trichomonas vaginalis or fungi, or other non-neoplastic findings.

There are also changes in the reporting of endometrial cells. In women under 40, endometrial cells need not be reported, but exfoliated endometrial cells should be reported in all women from age 40 on.

Previously, benign-appearing endometrial cells were reported as epithelial cell abnormalities in postmenopausal women not receiving hormone therapy. The literature and historical precedent behind this custom were reviewed during the Bethesda workshop, says Ann T. Moriarty, MD, a member of the CAPCytopathology Committee and moderator of the forum group that reviewed endometrial cells. With extensive input from the Bethesda Web site and the workshop itself, her group established the following:

  • Endometrial cells that may be significant are exfoliated glandular cells (not histiocytes, abraded endometrial cells, or stromal cells).
  • Significant endometrial abnormalities (atypical hyperplasia or carcinoma) rarely occur in women under 40.
  • The scant scientific data that are available are usually based on the age of women in the study, not their menstrual status.
  • There are no data that support that women on hormones are not at risk for endometrial pathology when they have benign-appearing endometrial cells on their Pap tests.
  • The most common histologic finding in women with endometrial cells on their Pap tests is normal endometrium; therefore, finding endometrial cells "out of cycle" or in the menopausal woman does not carry strong predictive value of endometrial pathology.
  • Menstrual data, menopausal status, risk factors for endometrial pathology, and hormone therapy are not regularly or reliably communicated to the laboratory.

"The Pap test is a screening test for squamous lesions, not glandular lesions," notes Dr. Moriarty, a cytopathologist with AmeriPath Indiana in Indianapolis. "We know that in a vanishingly rare percentage of women with endometrial pathology, the only sign may be benign-appearing endometrial cells on Pap smears. Therefore, to be the most sensitive to that small population, we recommended that endometrial cells be reported in all women at the age of 40." This report also specifies that the Pap test is "negative for squamous intraepithelial lesion." Also recommended is an educational comment about the potential significance of endometrial cells.

"Most importantly, these cells are not reported under the general categorization of ’epithelial cell abnormality.’ Endometrial cells are reported under a general categorization of ’other,’ leaving the laboratory more flexibility in hierarchical review, as well as calling attention to the fact that these cells are not abnormal-appearing epithelial cells," Dr. Moriarty says.

With regard to specimen adequacy, there previously were three categories: "satisfactory," "unsatisfactory," and "satisfactory but limited," which could include a lack of cells from the transformation zone or partially obscuring blood or inflammation. Now the reports will be labeled either "satisfactory for evaluation" or "unsatisfactory for evaluation," with, for example, the presence or absence of endocervical/transformation zone components indicated after the satisfactory term.

Finally, the Bethesda workshop participants agreed on "more specific criteria as to how many squamous cells are needed" for a good sample, Dr. Davey says. For liquid cytology, a minimum of 5,000 squamous cells is recommended. "We’re still working on methods for estimating cellularity of a conventional Pap," she adds, but the plan is that reference images should be used to estimate cellularity and that at least 8,000 to 12,000 cells should be present.

ASCCP changes

Detailed guidelines from the ASCCP meeting are not yet available. Conference organizer and ASCCP president Edward J. Wilkinson, MD, asked participants not to divulge them until they are published in a peer-reviewed journal, which is expected in about three months. "The major publications won’t publish our results if they’re pre-released," says Dr. Wilkinson, who is director of anatomic pathology and cytology at the University of Florida College of Medicine, Gainesville.

However, participants did discuss the key recommendations in a general way and affirm the importance of the consensus conference. Twenty-nine organizations were represented, Dr. Wilkinson says, including the CAP and the NCI, and they agreed on a series of guidelines for managing abnormal Pap test results.

Topics at the meeting included the Bethesda conference report, results of the ASCUS/LSIL Triage Study (ALTS) sponsored by the NCI, and liquid versus conventional Pap smear. "There will be some significant management recommendations for abnormal cervical cytology coming out of this meeting," Dr. Wilkinson predicts. He hopes to start presenting the findings at medical meetings, including an American Cancer Society conference that was scheduled for late September. Following publication in a peer-reviewed journal, "we would release our findings to all medical journals that would be interested."

He convened the "unprecedented" ASCCP meeting in an effort to achieve a national consensus. "Part of our goal was to arrive at consistency between pathologists and clinicians," Dr. Wilkinson says. "The pathologists often offer recommendations in the cervical Pap report, so this [ASCCP consensus] could help guide them in their findings."

The timing for the meeting was right, he says, because of the "huge amount of new information coming out in regard to cervical neoplasia," including the ALTS trial and NCI studies in Latin America. In addition, meta analyses were published recently, and there are new findings on the use of HPV triage (DNA testing for cancer-associated human papillomavirus).

"What came out of the conference will continue to accelerate the trends for liquid cytology and HPV testing as an important part of the screening process," predicts R. Marshall Austin, MD, PhD, director of cytopathology and gynecologic pathology services at Coastal Pathology Laboratories, Charleston, SC. He was a CAP representative to the ASCCP meeting.

(The ALTS trial results, reported in May in CAP TODAY, showed that HPV triage can be valuable for women who receive an ASCUS result on their Pap test. If the HPV test is negative, it is unlikely that a high-grade lesion is present. If positive, colposcopy or other immediate followup is recommended. HPV testing is especially useful for women over 35 because its specificity is higher in that age group.)

Dr. Austin was enthusiastic about the consensus findings. "You’re going to see a very detailed set of management recommendations that will have tremendous impact in terms of driving followup," he says. "Almost immediately they will become the standard of practice."

The recommendations are important to pathologists, especially those accustomed to advising clinicians on how to manage abnormalities. "We will have to be intimately familiar with these new guidelines," he says.

The need to improve the sample and the sensitivity of testing were made clear. "Anybody seeing these results with their eyes open would understand that liquid cytology and adjunctive HPV testing are the way to go," he says.

Dr. Davey agrees with Dr. Austin that HPV testing will increase both for ASCUStriage and as followup for women with a previously identified abnormality." She adds, though, that while liquid cytology lends itself to HPV triage, "I don’t think we’ll see a recommendation favoring liquid or conventional. Both will continue to be used."

According to Dr. Davey, another topic at the ASCCP conference was specimen adequacy, though that was added so late that discussion is continuing. "We hope to have a guideline coming out later that will define the followup on a woman who has no transformation zone component or obscuring factors," she says.

Finally, "there was agreement that labs can include references to these guidelines in their reports," Dr. Davey says, directing clinicians to a Web site or a printed version, once published. "Eventually, we’re hoping to include algorithms on the Internet [related to abnormal results] so that everyone is on the same wavelength on management of the patient," she says.

Karen Southwick is a writer in San Francisco.