A Gynecologic Oncology Group study, headed by Cornelia
Trimble of the Johns Hopkins
Kimmel
Cancer Center, has revealed that pathologists who
evaluate uterine biopsies disagree 60 percent of the time
on whether the specimens contain cancerous cells. The
authors, who presented their findings at the 40th annual
American Society of Clinical Oncology, say new standards
for collecting and classifying biopsies are needed to
improve the accuracy of diagnoses.
Cancer of the uterus is the most common cancer
affecting the female reproductive tract in the United
States. It is diagnosed from biopsied uterine or
endometrial cells that indicate the presence of cancer or
precancerous lesions called atypical endometrial
hyperplasia, or AEH.
"This study brought into sharp focus the fact that it
is very difficult to make an accurate diagnosis from
uterine biopsies," says Trimble, associate professor at the
Johns Hopkins Kimmel Cancer Center. "Yet women receiving a
diagnosis of AEH face complete removal of their wombs
through hysterectomy. So we suggest these patients get a
second opinion from a pathologist who specializes in
gynecology."
Trimble and colleagues in the national cooperative
Gynecologic Oncology Group set out to get a baseline
estimate of the percentage of actual cancers found in
hysterectomy samples of women diagnosed with AEH biopsies.
The baseline, reported to be anywhere from 17 percent to 52
percent, would provide information needed to design studies
to find nonsurgical treatments for AEH that preserve
fertility in young patients or eliminate the need for
surgery in women with diabetes, hypertension or other
complicating disorders. But after reviewing biopsies from
289 patients classified as AEH, they were surprised to find
a high degree of disagreement with the initial diagnosis of
most of the biopsies.
In their review, the GOG investigators downgraded the
biopsy diagnoses to "less than AEH," which denotes a range
of benign disorders, in 25.6 percent of cases. They
upgraded the AEH diagnosis to cancer in 29.1 percent of
cases and were unable to agree on a diagnosis in 5.5
percent of the biopsies. The investigators agreed with the
diagnosis of AEH in only 39.8 percent of cases. The overall
baseline estimate of women who also had cancer present in
their corresponding hysterectomy specimen was 42
percent.
Underscoring the level of disagreement and difficulty
in making an accurate diagnosis, even with the GOG review,
was the fact that cancer was found in corresponding
hysterectomy specimens of 14 of 74 downgraded AEH biopsies,
45 of 115 AEH-diagnosed biopsies, 54 of 84 AEH specimens
upgraded to cancer and 10 of 16 cases where there was no
agreement.
"We will be looking at these data more closely to
determine why there was such disagreement in the diagnosis
and find a better system for classifying and grading these
biopsies, including using improvements in molecular markers
to help identify better ways to diagnose uterine cancer,"
Trimble says.
Although guidelines for distinguishing benign
endometrial biopsies from AEH and cancer have been accepted
by the World Health Organization and Society of Gynecologic
Pathologists, Trimble and her colleagues noted that the
classification system has not undergone the same rigorous
evaluation as other lesions, such as cervical cancer.
Trimble and the GOG investigators are developing more
precise classification guidelines.
Uterine cancer, also called endometrial cancer, will
strike more than 40,000 U.S. women annually and will kill
7,090. The most common identifiable symptom is irregular
bleeding.
This research was funded by the National Cancer
Institute.
— Vanessa Wasta