(Ackerman's Surgical Pathology, 8th Ed.)
General and clinical features
Carcinoma of the endometrium is the most common gynecologic
malignancy in the United States, and its incidence is rising.
It typically occurs in elderly patients; approximately 80%
are postmenopausal at the time of diagnosis. It has been suggested
that endometrial carcinoma can be divided in two distinct
types on the basis of their pathogenesis: oneby far the more
commonoccurring as a result of excess estrogenic stimulation
and developing against a background of endometrial hyperplasia
and the other developing de novo. Patients at high risk for
the first category include the obese, diabetic, hypertensive,
infertile; those with failure of ovulation (including the
Stein-Leventhal syndrome) and dysfunctional bleeding; long-standing
estrogen users; those with severe degrees of endometrial hyperplasia,
andto a much lesser degree those with functioning granulosa
cell tumors and thecomas.
In the majority of patients with
Stein-Leventhal syndrome, the endometrial pathology is that
of hyperplasia and, as such, it will regress with medical
therapy. However, a few well-documented cases of carcinoma
have been reported; these have almost always been of a welldifferentiated
nature, and myometrial invasion, if present at all, has been
minimal. Fechner and Kaufman pointed out that the lesion may
be reversible when treated by curettage followed by therapy
directed toward reestablishment of ovulation and have urged
a conservative approach to these patients. In support of this
policy, they emphasized the fact that not a single case of
well-differentiated adenocarcinoma in a patient with Stein-Leventhal
syndrome has been proved to metastasize, recur locally, or
cause death. The situation is quite similar regarding the
relationship between endometrial pathology and functioning
Gonadal dysgenesis (Turner's
syndrome) can also be associated with endometrial adenocarcinoma,
usually of the well-differentiated type. McCarty et al. found
thirteen reported cases; eleven patients had received replacement
estrogen therapy, usually in high doses and for prolonged
periods. It is not clear whether this association represents
a complication of long-term estrogen exposure or a rare expression
of the Turner phenotype. Interestingly, almost two thirds
of the carcinomas exhibited squamous differentiation.
Some cases of endometrial carcinoma
have been seen years after pelvic irradiation for some other
condition, but whether these are spontaneous or radiation-induced
is not clear.
Recently several reports have
appeared suggesting that patients who receive tamoxifen as
long-term treatment for breast carcinoma may be at an increased
risk for the development of endometrial adenocarcinoma; of
particular concern is the fact that in two series a significant
number of these cases have been high-grade tumors associated
with a poor prognosis.
Grossly, carcinoma of the endometrium may form broad-based
polypoid masses or infiltrate diffusely into the myometrium.
In general, extensive myometrial invasion is accompanied by
clinically detectable uterine enlargement. However, notable
exceptions occur; sometimes deep myometrial extension is accompanied
by a normal-sized uterus. At times, the tumor begins in a
cornu and is missed by D&C.
Microscopically, about 80% of
endometrial malignant epithelial tumors are conventional adenocarcinomas,
which are usually divided into well (grade I, 50%), moderately
(grade II, 35%), and poorly differentiated (grade III, 15%)
tumors (Figs. 19-91 and 19-92). The FIGO grading system is
primarily based on the growth pattern (relative proportion
of glandular and solid areas) but it also makes provisions
for nuclear atypia.
The better differentiated tumors
closely recapitulate the light and electron microscopic features
of the non-neoplastic endometrium, hence the term "endometrioid"
that is used for them. Over a quarter of the carcinomas have
papillary (villoglandular) foci, either on the surface or
in the invasive areas. These tumors should be sharply separated
from the much more aggressive papillary serous carcinomas.
The stroma of endometrial adenocarcinoma
usually has a desmoplastic quality. It may contain collections
of foamy cells, probably the result of tumor necrosis and
a good marker for the presence of carcinoma. However, these
cells can also be seen in hyperplasia and, exceptionally,
in otherwise normal endometria. They are said to form from
endometrial stroma rather than histiocytes. They are fat positive
and mucin negative, in contrast to the mucin-positive macrophages
sometimes seen in the stroma of benign endometrial polyps.
The non-neoplastic endometrium
of a uterus harboring an adenocarcinoma is often hyperplastic,
is sometimes atrophic, and only exceptionally exhibits a normal
proliferative or secretory pattern; when it does, the assumption
has been made that the carcinoma has arisen in a "progesterone
refractory" mucosal area.
The frequency and extent of myometrial
invasion by carcinoma are directly related to the microscopic
grade of the tumor. Care should be exercised to distinguish
true myometrial extension by carcinoma from expansion of the
endometrial-myometrial junction and from atypical or malignant
changes involving pre-existent foci of adenomyosis; the latter
condition is recognized by the presence of endometrial stroma
around the intramyometrial proliferating glandular foci. Extension
of the endometrial carcinoma into the cervix occurs in over
10% of the cases, usually by direct invasion, but sometimes
by implantation following D&C. This may be grossly evident
or become apparent only on microscopic examination; it may
involve the surface only, the fibrous stroma, or both. The
presence and type of cervical extensionwhich influences the
staging of the tumoris best detected by fractional curettage;
care should be exercised in distinguishing bona fide cervical
extension from isolated tumor fragments, or else a high false
positive rate will occur.