| Introduction
In the last decade endometriosis
has been associated with a definite increase in
risk of various malignancies (1). In order to assess
the entity and the nature of this association, we
have critically evaluated observational, cohort
and case-control studies performed on this topic
with a specific focus on their problems and limitations
(1).
Evidence
from clinical series
Based on Sampson’s (2) and Scott’s (3)
criteria to identify malignant tumors raised from
endometriosis, few groups have evaluated the prevalence
of ovarian malignancy in large series of patients
operated for endometriosis. A prevalence of about
0.9% has been observed. Unfortunately, all currently
available series are retrospective, and thus probably
unable to properly address this point. In support
to this idea, the same prevalence differed significantly
according to the pathologists who performed the
analysis.
A prospective,
sufficiently large and unbiased series evaluating
the frequency of ovarian cancer concomitant to endometriosis
is currently not available.
Whether or not
endometriosis should be considered a preneoplastic
disease represents a major and controversial issue.
Studies on the epithelial lining of cystic ovarian
endometriosis have documented the presence of metaplastic,
hyperplastic or atypical changes whose prevalence
in endometriosis is not defined (revised in Somigliana
et al, 2006) (1). Data on metaplasia are
scanty and controversial. Differences among the
studies may be due, at least in part, to a different
study population and/or to selection biases. Moreover,
the presence of neoplasm per se might induce metaplasia
in the adjacent endometriosis.
Overall, there
is insufficient evidence supporting metaplasia of
endometriotic lesions as a preneoplastic condition.
Data on atypia are more univocal and document a
highly significant increased presence of severe
atypia in endometriotic lesions in patients with
endometriosis-associated ovarian cancer when compared
to patients with endometriosis alone. Finally, data
on endometriotic hyperplasia, although very limited,
also support a possible association with the malignant
transformation. These data suggest that carcinoma
may arise from endometriosis through a multi-step
phenomenon where typical endometriosis may change
into severe atypia with or without hyperplasia and
then into carcinoma. Nevertheless, conclusive evidence
supporting this model is lacking.
Studies prospectively
evaluating the risk of cancer in patients with endometriotic
atypia and/or complex hyperplasia are extremely
scarce and inconclusive. A major limit in investigating
the potential malignant transformation of these
lesions is that during a surgical intervention for
endometriosis, excision of endometriomas is generally
complete and, consequently, it may be argued that
atypical lesions are practically absent after surgery.
A 4-29% frequency of endometriosis was found in
cases operated for ovarian tumors (Table I). These
percentages do not appear to be very different from
the 10% supposed prevalence of the disease in the
reproductive age. On the other hand, a consistent
body of evidence has documented a clear association
between endometriosis and endometrioid and/or clear
cell ovarian carcinomas (Table I). This observation
represents one of the most important issues supporting
a possible causal relationship between endometriosis
and ovarian cancer.
TABLE
I
In regard to the
clinical behaviour and prognostic factors in ovarian
cancer patients with or without concomitant endometriosis,
patients affected tended to be younger and to be
diagnosed in earlier stages and with lower grade
lesions (4-8). A better prognosis could be demonstrated
in these patients. It remains to be clarified whether
the less frequent dissemination outside the ovaries
in cancers arising from endometriosis may be due
to a different pathological behaviour of the malignancy
per se or whether it may be related to the destruction
of endometriotic lesions in more advanced cancers.
Furthermore, a diagnostic bias could also explain,
at least in part, the increased diagnosis in initial
stages. The typical symptoms of endometriosis might
facilitate earlier diagnosis.
TABLE
II
Endometriosis
and ovarian cancer: evidence from population-based
studies
Data from the majority of the available cohort and
case-control studies tend to suggest an association
between endometriosis and ovarian cancer, although
it is difficult to precisely estimate the effect
size as the observed increase in risk ranges from
30 to 90% (Table II) (revised in Somigliana et
al, 2006) (1). In epidemiology, a relative
risk of less than two is considered to indicate
a weak association and weak associations are more
likely to be explained by unrelated biases. Some
limitations of available studies have to be considered:
1. confounders
have not always been controlled adequately. It is
well known that parity and oral contraceptive use
represent strong preventive factors and measures
of association should at least be controlled for
these two factors. It cannot be ruled out that some
medical treatment options of endometriosis may also
influence the hazardof ovarian cancer;
2. studies assume that the identification of endometriotic
lesions during a surgical intervention corresponds
to the presence of the disease in that particular
patient. However, surgery is aimed to eradicate
the disease. Since recurrence of endometriosis is
not a systematic occurrence, the assumption that
all operated patients remain affected may lead to
an underestimation of the risk;
3. the generalisation of the results to all women
with endometriosis might be incorrect, as most of
the observations refer to women affected by advanced
forms necessitatinghospitalisation and surgery;
4. insights from clinical series indicate that endometriosis
is linked only to endometrioid and clear cell ovarian
carcinomas and not to other malignant histotypes
(Table I). These histologic diagnoses represent
about one third of all epithelial ovarian cancers.
Unfortunately, most epidemiologic studies focus
on ovarian cancer in general. This may have attenuated
the entity of the observed associations. Studies
specifically considering these two histotypes suggest
an increased association but number of observed
cases are extremely low (9, 10).
Hence, some limitations
may have biased results towards the null hypothesis
whereas others may have led to overestimate the
association.
TABLE
III
Endometriosis
and other cancers
The potential association between endometriosis
and breast cancer remains unclear should be interpreted
with caution, because of the small number of subjects
reporting the condition, the trend towards a protective
effect in postmenopausal women observed in some
studies and the lack of consistency among studies
(Table III) (revised in Somigliana et al, 2006)
(1). The documentation of a reduced risk of cervical
cancer in patients with endometriosis using the
National Swedish Cancer Register is unexpected (Table
III) and could be interpreted in terms of increased
number of referrals to a gynaecologist from patients
with endometriosis. An association between melanoma
and endometriosis has been repeatedly reported by
a single research group (Table III). In general,
these studies collected a limited number of subjects,
concerned dysplastic nevi, a well-known precursor
of melanoma, rather than the cancer itself and were
characterized by too many sub-analyses in relation
to the casistics evaluated. Results from large independent
observational studies are controversial in this
regard (Table III). A statistically significant
increased risk of melanoma in the order of two-fold
was documented in infertile women with endometriosis
when compared to patients with other causes of infertility
in a recent study by Brinton et al (11).
The two largest
population-based cohort studies have independently
documented the association between non-Hodgkin’s
lymphoma and endometriosis (Table III).
However, the statistically significant results from
these studies are based on a small number of observed
cases and need further confirmation.
Discussion
Current evidence is robust enough to sustain a link
between endometriosis and ovarian cancer. However,
the demonstration of an association between two
conditions cannot be used to infer causality. Two
possible scenarios may be envisioned to explain
this link:
1. Endometriotic cells might undergo somatic mutational
events able to confer to the cells the malignant
potential characteristics of cancer. Endometriosis
would be the precursor of some, if not all, ovarian
cancers of endometrioid and clear cell histologic
types. Support to the idea of the ectopic tissue
as the benign precursor of some ovarian cancers
comes from those cases of evident histologically-proven
transition from the benign disease to the malignant
entity.
2. Alternatively, endometriosis and ovarian carcinoma
might represent two distinct biological entities
characterised by a different set of causative molecular
events and their relative frequent coexistence may
derive from the sharing of some risk factors or
antecedent mechanisms. Nulliparity and menstrual
characteristics are well known determinants of the
risk for both the conditions.
At present, we
are unable to disentangle this issue; the idea that,
in rare cases, the ectopic tissue undergoes malignant
transformation cannot be refuted but we are unable
to quantify the entity of this process and consequently
to infer that causality should be advocated to explain
the increased risk of specific histotypes of ovarian
cancer in women with endometriosis. Consequently,
from a clinical point of view, it is questionable
whether a systematic and serial surgery may be justifiable
in women with endometriosis based on the assumption
that eradication of visible lesions would abolish
this increase in risk. On the other hand, it is
well known that oral contraceptives have a protective
effect of on ovarian cancer risk in general. Oral
contraceptive use is associated with a substantial
and duration-dependent reduction in risk, with an
observed 80% lower occurrence of ovarian cancer
in women with endometriosis who use the drug for
>10 years (12). This approach seems preferable
to the uncertain results of a repetitive surgery
that, in these patients, is sometimes associated
with major morbidity due to severely distorted anatomical
conditions.
For other types
of tumours, no modifications in the standard, age-related
diagnostic evaluations for the early detection of
cancer are suggested. Women with endometriosis should
undergo the usual dermatologic surveillance suggested
for the general population. General practitioners
and gynaecologists should be aware of the possibility
that women with endometriosis are at increased risk
of non-Hodgkin’s lymphomas, and consequently,
should seek prompt evaluation in case of recurrent
infections, unexplained fever, persistent cough,
or weight loss but due to the small number of patients
that develop non-Hodgkin’s lymphoma in the
general population, no major invasive diagnostic
investigations seem warranted in women with endometriosis
undergoing routine screening programs.
REFERENCES
1. Somigliana E, Vigano P,
Parazzini F, Stoppelli S, Giambattista E, Vercellini
P. Association between endometriosis and cancer:
A comprehensive review and a critical analysis of
clinical and epidemiological evidence. Gynecol Oncol
2006 Feb 10; [Epub ahead of print].
2. Sampson JA. Endometrial carcinoma of the ovary
arising in endometrial tissue in that organ. Arch
Surg 1925; 10: 1-72.
3. Scott RB. Malignant changes in endometriosis.
Obstet Gynecol 1953; 2: 283-9.
4. McMeekin DS, Burger RA, Manetta A, DiSaia P,
Berman ML. Endometrioid adenocarcinoma of the ovary
and its relationship to endometriosis. Gynecol Oncol
1995; 59: 81-6.
5. Komiyama S, Aoki D, Tominaga E, Susumu N, Udagawa
Y, Nozawa S. Prognosis of Japanese patients with
ovarian clear cell carcinoma associated with pelvic
endometriosis: clinicopathologic evaluation. Gynecol
Oncol 1999; 72: 342-6.
6. Erzen M, Rakar S, Klancnik B, Syrjanen K, Klancar
B. Endometriosis-associated ovarian carcinoma (EAOC):
an entity distinct from other ovarian carcinomas
as suggested by a nested case-control study. Gynecol
Oncol 2001; 83: 100-8.
7. DePriest PD, Banks ER, Powell DE, Van Nagell
JR jr, Gallion HH, Puls LE, Hunter JE, Kryscio RJ,
Royalty MB. Endometrioid carcinoma of the ovary
and endometriosis: the association in postmenopausal
women. Gynecol Oncol 1992; 47(1): 71-5.
8. Erzen M, Kovacic J. Relationship between endometriosis
and ovarian cancer. Eur J Gynaecol Oncol 1998; 19(6):
553-5.
9. Brinton LA, Sakoda LC, Sherman ME, Frederiksen
K, Kruger Kjaer S, Graubard BI, Olsen JH, Mellemkjaer
L. Relationship of benign gynecologic diseases to
subsequent risk of ovarian and uterine tumors. Cancer
Epidemiol Biomarkers Prev 2005; 14: 2929-2935.
10. Ness RB, Cramer DW, Goodman MT, et al. Infertility,
fertility drugs, and ovarian cancer: a pooled analysis
of case-control studies. Am J Epidemiol 2002; 155:
217-224.
11. Brinton LA, Westhoff CL, Scoccia B, Lamb EJ,
Althuis MD, Mabie JE, Moghissi KS. Causes of infertility
as predictors of subsequent cancer risk. Epidemiology
2005; 16(4): 500-7.
12. Modugno F, Ness RB, Allen GO, Schildkraut JM,
Davis FG, Goodman MT. Oral contraceptive use, reproductive
history, and risk of epithelial ovarian cancer in
women with and without endometriosis. Am J Obstet
Gynecol 2004; 191: 733-40.
Next... |
