Key words borderline tumour - fertility preservation - future wish for child - luteal phase
stimulation - ovarian cryopreservation
Schlüsselwörter Borderline-Tumor - Fertilitätserhalt - späterer Kinderwunsch - Lutealphasenstimulation
- Kryokonservierung von Ovargewebe und Eizellen
Introduction
Around 10 to 20 % of all epithelial ovarian tumours are serous or seromucinous borderline
ovarian tumours. Their incidence is around 24 per 1 000 000 women [1 ]. The World Health Organization (WHO) defines borderline tumours as having both malignant
characteristics (usually cellular atypicality) and benign characteristics such as
the absence of invasive-destructive growth or stromal invasion. Borderline tumours
exhibit a complex papillary architecture with at least 2 of the following characteristics:
epithelial cell proliferation in the form of small papillae without a connection to
the internal tissue)
multilayered epithelium with atypicality (layers of atypical epithelial cells on the
papillae)
mitotic activity (increased mitoses or changes in the distribution of the mitoses)
nuclear atypicality (increase in nucleus size and chromate density) [2 ]
A further peculiarity of this type of tumour is the fact that it is primarily diagnosed
in women of reproductive age [3 ]. While 80 % of borderline ovarian tumours involve only 1 ovary, around 20 % occur
bilaterally. The prognosis for bilateral involvement is less favourable [4 ]. While many authors propose an ablative surgical procedure with bilateral adnexectomy
for treating early-stage borderline ovarian tumours, there are also indications that
organ-sparing procedures that preserve fertility also have very good oncological outcomes.
Fertility-preserving procedures include tumour enucleation, cystectomy and unilateral
adnexectomy [5 ]. The disadvantages of this type of procedure are the higher risk of recurrence and
need for follow-up surgery. Other factors promoting a risk of recurrence are age < 30,
histology (micropapillary tumours have the least favourable prognosis) and tumour
stage [6 ], [7 ]. [Table 1 ] presents an overview of the tumour stages for borderline ovarian tumour, while [Table 2 ] presents an overview of the histological subtypes and their characteristics. The
decision for or against organ-sparing surgery depends on a variety of factors, including
the patientʼs desire for fertility, age, gynaecological findings and the volume of
ovarian tissue remaining [8 ]. If an organ-sparing surgical procedure is not an option despite the patientʼs current
or future desire to become pregnant, for example, in the case of bilateral ovarian
involvement or due to the disease progression, cryopreservation of oocytes after hormonal
stimulation therapy and/or of ovarian tissue may be considered [9 ], [10 ].
Table 1 Tumour staging for borderline ovarian tumour.
Tumour stage
Characteristics
Stage I
Borderline tumour limited to the ovary
Stage II
Tumour extension into the lesser pelvis
Stage III
Tumour extension below the lesser pelvis and/or in the area of the retroperitoneum
Stage IV
Distant metastases
Table 2 Histologic subtypes of borderline ovarian tumour [32 ].
Histologic subtype
Characteristics
Serous
Papillary protuberances in the area of the internal cyst wall
Mucinous
Contains around 50 % solid components
Endometrioid
Primarily solid
Brenner tumour
Coarse stromal tumour up to several centimetres in size with low mitotic activity
Clear cell-micropapillary
Irregular tubuli with atypical clear cells
Case Report
We report on a 25-year-old woman (0 gravida) who first presented to our department
on 12 October 2015 for consultation on fertility preservation after external diagnosis
of bilateral seromucinous borderline ovarian tumour.
Patient history
Externally determined findings
The patient initially presented for urological examination in July 2015 due to lower
abdominal pain of unknown origin. Based on the imaging results showing enlarged cystically
changed ovaries bilaterally, gynaecological diagnostics were recommended. Subsequently,
laparoscopic extirpation of an ovarian cyst was performed. Intraoperatively, extensive
adhesions in the lower pelvis were observed, along with an approx. 3 cm pseudoperitoneal
cyst in the area of the left ovary that ruptured spontaneously. Moreover, on the left
ovary a lawn of roe-like deposits were observed that appeared brittle upon contact.
The right ovary made a cystically changed impression, was 3 × 4 cm in size and was
otherwise unremarkable. In addition to adhesiolysis, the pseudoperitoneal cyst with
a cystoid follicle on the left was removed, the tumour on the left was resected and
the cyst removed. During histological examination, an atypically proliferated seromucinous
borderline ovarian tumour was diagnosed bilaterally. The proliferation was low to
moderate and focally pronounced, so that the risk of recurrence was deemed to be low.
The 25-year-old woman was advised to undergo stage-appropriate surgery with a bilateral
adnexectomy and resection of the greater omentum, the appendix and all macroscopically
visible tumour foci. She was also advised to present to a second centre to obtain
a second opinion on the histological findings. Since the young woman was planning
to become pregnant in the future, she presented to our department for advice about
fertility-preserving treatment options.
Clinical findings
At initial presentation, the patient reported that her overall state of health was
good. The patient reported a regular menstrual cycle (28/3 days) and dysmenorrhoea.
Apart from light smoking (2 cigarettes/day), there were no other remarkable aspects
in the patient history. During the gynaecological examination, the only palpable abnormalities
were a coarse resistance in the left adnexal area and in the ultrasonography, a prominent
right ovary and little free fluid in the recto-uterine pouch ([Fig. 1 ]). We explained to the patient that the non-organ-conserving surgery proposed at
the external facility was considered the standard treatment in the specific situation
with respect to achieving maximum oncological certainty. At the same time, in view
of the patientʼs explicit desire to preserve fertility, we discussed possible fertility-preserving
options, including oocyte and ovarian tissue cryopreservation, as well as the attempt
to spare an ovary and the uterus. We informed the patient about an increased risk
of recurrence with this procedure, which, however, would not guarantee a future pregnancy.
Our patient requested oocyte and ovarian tissue cryopreservation. We discussed hormonal
stimulation therapy with oocyte cryopreservation as well as the subsequent surgical
laparoscopy with bilateral tumour removal and unilateral ovarian tissue cryopreservation.
The patient consented to a two-stage procedure comprising primary oocyte retrieval
and preservation of biopsies and secondary stage-appropriate surgery in the event
that the need for bilateral adnexectomy and, in the event of uterine involvement,
hysterectomy, was determined intraoperatively. The patient presented to our endocrinological
reproductive medicine out-patient department on 19 October 2015 to plan oocyte cryopreservation.
Since the patient was on day 11 of her menstrual cycle, in consensus with the patient,
we agreed to initiate experimental luteal phase stimulation starting on day 18 of
the cycle (26 October 2015) to prevent a 14-day wait and thus counteract further dissemination
of the tumour.
Fig. 1 Preoperative sonographic image of the right ovary.
Treatment
We initiated hormonal stimulation therapy in the luteal phase on 26 October 2015 in
the short antagonist protocol and our patient self-administered 200 (day 18 to 20
of the cycle)/150 IU FSH as a stimulation drug and the GnRH antagonist cetrorelix
once a day. The stimulation lasted 16 days. During this period, we performed 3 sonographic
check-ups with laboratory work-ups. The treatment resulted in the development of 3
follicles > 10 mm on the right and 1 follicle on the left. Two days prior to oocyte
retrieval, the endometrium was well-established with 12.7 mm. The estradiol value
was 2126 pg/nl. Our patient inducted ovulation with 0.2 mg triptorelin. Oocyte retrieval
was performed on 11 November 2015 and was uncomplicated. Cryopreservation of 5 unfertilised
oocytes was subsequently undertaken. The 25-year-old woman then presented to our department
for the planned surgical laparoscopy on 27 November 2015. Intraoperatively, the organs
of the upper abdomen presented as unremarkable. In the lesser pelvis there were veil-like
peritoneal adhesions in the left adnexa area. Both ovaries presented with cysts. We
performed bilateral tumour extirpation and removed approximal one third of the right
ovary for cryopreservation ([Fig. 2 ]). Since the other adnexa was unremarkable, we removed only the macroscopically visible
tumour on the ovary. We also took several peritoneal biopsies in the area of the abdomen
as a whole and performed an omentectomy. As there were no signs of peritoneal tumour
dissemination, no other interventions were performed. The tumour dissemination corresponded
with FIGO stage IC2. There were no perioperative or postoperative complications. The
histological testing of the biopsies by our pathology department yielded a seromucinous
borderline tumour in the area of the left ovary not related to the ovarian surface
that had already been diagnosed by the referring facility. The right ovary showed
tumour-free tissue with focal low-grade chronic inflammation. The peritoneal biopsy
was also unremarkable. We discussed the results with our patient in detail and advised
her to present for follow-up with rectovaginal palpation and ultrasonography every
3 months in the first 2 years post-surgery. The patient was advised to become pregnant
soon after this period, no later than age 35, in order to reduce the risk of recurrence.
After this period, the patient was advised to present for follow-up on a 6-monthly
basis up to 5 years post-surgery and on a yearly basis afterward. We also advised
her to undergo bilateral adnexectomy and subsequent hormone replacement therapy after
completion of childbearing.
Fig. 2 Surgical site: Removal of a part of the ovary for cryopreservation.
Discussion
The case report presented above describes a deviation from the standard treatment
for bilateral ovarian seromucinous borderline tumour due to the individual situation
of our 25-year-old patientʼs pronounced desire to become pregnant. In consensus with
our patient, we elected not to perform an adnexectomy and instead to employ an experimental
treatment method with stage-appropriate surgery and luteal phase stimulation. In the
specific situation, the decision to perform the fertility-preserving surgery was particularly
difficult in light of the fact that bilateral ovarian involvement and the performance
of a cystectomy rather than an adnexectomy are definitively associated with increased
risk of recurrence [19 ]. This also applies to a cystic rupture, which had occurred during the previous surgery
performed at the referring facility. While the risk of borderline tumour recurrence
is generally reported to be approx. 5 % after stage-appropriate surgery, this percentage
can increase four-fold after fertility-preserving cystectomy such as that performed
on our patient [11 ]. However, the literature also contains reports on the possibility of bilateral cystectomy
in the case of bilateral ovarian involvement and explicit fertility-preserving desire.
Vasconcelos et al. concluded that in this situation, there are no indications of increased
recurrence rates compared to unilateral adnexectomy with contralateral cystectomy
[12 ]. It should also be noted that the surgical procedure in this case complied with
guidelines. The S3 guidelines for ovarian cancer state that if the patient wishes
to have children and the tumour is limited to the ovary, a fertility-preserving procedure
can be carried out [13 ]. This brings to light the fact that among young patients, oncological certainty
and the wish to preserve fertility may be at odds with each other. For each situation,
the attending physician needs to weigh the increased risk of recurrence against the
gain in quality of life for the patient yielded by the ability to become pregnant
in the future. This may cause an ethical dilemma for the physician that may be best
resolved by addressing the topic of fertility directly with the patient. It is known
that up to 1 in 7 patients would be prepared to take compromises in oncological certainty
into account if it meant being able to have children later on [14 ]. At the same time, it is also known that as the patientʼs awareness of the significance
of and options for preserving fertility increases, so does the frequency of educating
the patients about this and in turn, the use of fertility-preserving measures [15 ].
Nevertheless, the procedure continues to be a case-by-case decision that needs to
be carefully considered by the young cancer patient. Zanetta et al. reported a recurrence
rate of 18.5 % for early disseminated cancer after fertility-sparing surgery as compared
to a rate of only 4.7 % after radical surgery. However, at the 70-month follow-up,
the higher recurrence rate was not associated with a higher mortality. The rate of
progression into invasive ovarian cancer has been reported at 2 % [12 ], [16 ]. However, it must be critically noted that in up to 30 % of all cases, the recurrence
of a borderline tumour represents a classic invasive epithelial ovarian cancer with
a much worse prognosis [17 ]. Thus, prior to taking the decision for or against a fertility-preserving procedure,
it is indispensable to perform sufficient surgical staging including assessment of
the internal genitalia, peritoneum and greater omentum, as was performed in our patient
at the referring facility. It also includes the removal of multiple peritoneal biopsies
in both the pelvis and the upper abdomen, as well as an omentectomy [17 ], [18 ]. In 2013, Du Bois et al. published results similar to those reported by Zanetta
et al. While the organ-sparing surgery was associated with a three-fold increase in
the recurrence rate, the overall survival of the patients was not decreased [19 ]. The fact that the risk of mortality was not increased for patients experiencing
recurrence of borderline tumours encouraged us to use the organ-sparing procedure.
Thus, sparing the adnexa is a legitimate approach in the primary situation and a cystectomy
is only absolutely necessary in the case of recurrence. We also informed our 25-year-old
patient with the stage I borderline tumour about the recurrence rate of approx. 1 : 5
with the use of an organ-sparing procedure. The safety of hormonal stimulation therapy
in patients with known borderline ovarian tumours is hotly debated. For example, in
the late 1990s, Parazzini et al. reported a higher incidence of borderline tumours
in women who had undergone hormonal stimulation therapy as part of infertility treatment
[8 ]. All in all, very few studies focus on stimulation therapy in women with borderline
tumours. In a study conducted by Denschlag et al. with 62 women and 152 treatment
cycles, the recurrence rate of 19.4 % after 52 months was comparable to the rate reported
by Zanetta et al. [20 ]. Some other authors recommend that women with borderline tumours who are basically
capable of spontaneous conception should wait 1 or 2 years before becoming pregnant,
since most recurrences of borderline ovarian tumours occur within 24 months of treatment
[7 ]. According to a study performed by Beiner et al., only 1 out of 7 women who underwent
infertility treatment experienced a recurrence of a borderline tumour [21 ]. In a retrospective multicentric study from 2007, the risk of recurrence was reported
as 1 : 9, with recurrence occurring only in women undergoing IVF treatment [22 ]. In vitro, no stimulatory effect of FSH or estradiol on the growth of borderline
tumour cell cultures could be detected [23 ]. This may indicate that hormonal stimulation may be recommended in good faith for
patients with borderline tumour who wish to preserve fertility. However, due to the
sparse data available, the treatment should be used prudently after weighing the risks
and benefits and only after the patient has been adequately informed.
Since our patient explicitly requested hormonal ovarian stimulation but was not at
the beginning of her menstrual cycle when she presented to our endocrinological reproductive
medicine out-patient department, after an in-depth consultation, we decided to initiate
hormonal ovarian stimulation in the luteal phase starting on day 18 of the cycle in
accordance with the stimulation protocol published by Wolff et al. [24 ]. In so doing, we weighed a potential compromise in oocyte quantity and quality against
the risk of delaying the procedure, i.e. tumour cell dissemination. The low number
of oocytes ultimately harvested, at only 5, is however most likely due to the patientʼs
low ovarian reserve after partial ovarian resection and tumour involvement of the
ovarian tissue rather than to the stimulation protocol. One positive outcome is the
fact that intraoperatively no involvement of the peritoneum was detected. Whether
the more timely surgery was responsible for this cannot be conclusively assessed in
retrospect. The literature contains several publications that describe luteal phase
stimulation for preserving fertility in patients with cancer as an option with good
results [25 ].
If our 25-year-old patient is unable to become pregnant, either spontaneously or by
means of fertilisation of the cryopreserved oocytes with subsequent transfer, she
will still have the option of transplantation of the resected ovarian tissue. The
tissue can be implanted into the peritoneal pouch by means of laparoscopy and in 70 %
of cases it regains its function [9 ]. Since hormone production usually only occurs for a limited period, the patient
should try to become pregnant very soon after the transplantation. In most cases,
short-term ovarian stimulation, oocyte retrieval, in vitro fertilisation and subsequent
embryo transfer are required. Worldwide, over 75 babies have been born thanks to this
procedure. At present, the likelihood of women becoming pregnant who have undergone
this procedure is around 25 % [26 ]. The method is therefore now the standard procedure for preserving fertility in
women with cancer. Reasons for the seemingly low pregnancy rate include the challenge
of creating a favourable ovarian milieu during transplantation, low ovarian reserve,
often already prior to the procedure, and the as yet still limited experience with
the relatively new procedure. If our patient eventually decides to undergo ovarian
tissue transplantation, in addition to informing her about the general risks associated
with the laparoscopic procedure, we will inform her about a potential risk of transplanting
tumour cells. In the literature, this risk is reported to be around 1 % across all
tumour types and is highest for haematological neoplasms [27 ]. Due to the sparse number of cases similar to the case we have described here, there
are no reports on statistics on the risk of transplanting malignant cells in patients
with borderline ovarian tumour in the literature. However, for all neoplasms primarily
involving the ovary an elevated risk must be assumed, that approximately coincides
with the risk of recurrence in the case of the fertility-preserving procedure. In
the case of recurrence, stage-appropriate surgery must be performed. Nevertheless,
Fain-Kahn et al. reported that cryopreservation of ovarian tissue is possible in over
half of patients with borderline ovarian tumour [28 ]. Other research groups also recommend removal and later transplantation of ovarian
tissue for preserving fertility in women with a borderline tumour [9 ]. While the removal and later transplantation of ovarian tissue in general is no
longer considered to be experimental, and is now the standard procedure for preserving
fertility, in the specific situation of our patient, it must be considered to be an
individual deviation from the generally recommended approach due to the elevated risk
of ovarian tumour implants owing to the bilateral ovarian involvement [29 ]. In 2011, Lotz et al. analysed ovarian biopsies from 23 premenopausal women with
epithelial and non-epithelial ovarian malignomas that were transplanted into SCID
mice. Twenty-four weeks after transplantation of these biopsies, no malignant cells
could be detected in the mice either histologically or under a light microscope [30 ]. Histological examination of the tissue can help lower the risk of transplanting
tumour cells.
Whether or not the patient wishes to become pregnant, regular check-ups are necessary
during follow-up. Contrary to most other tumour types, a borderline ovarian tumour
can still recur even 15 years after initial diagnosis [19 ]. We advised our patient to present for follow-up on a 3-monthly basis in the first
3 years post-surgery and then on a 6-monthly basis in the 2 years after that. Afterward,
the patient should undergo life-long annual check-ups. While current debate focuses
on the determination of tumour markers such as CA 125 and CA 19-9 in particular during
follow-up, we do not perform these tests [31 ]. Since no adjuvant chemotherapy is required after borderline tumour surgery, we
have advised the patient to become pregnant soon after surgery. After completion of
childbearing, stage-appropriate surgery is planned.
Conclusions
The attending physicianʼs desire for maximum oncological certainty and the patientʼs
need to preserve fertility in order to plan a pregnancy may be at odds during the
treatment of malignant or semimalignant tumours. A special feature of semimalignant
borderline ovarian tumours is that they often occur in young women who have not yet
completed childbearing. For this reason, before conducting any cancer treatment, and
especially in this particular situation, the patient should receive in-depth information
from a physician with experience in oncofertility. If a patient is currently planning
to become pregnant, she should be informed about an elevated postoperative risk of
infertility due to the low ovarian reserve and possible tubal functional disorders,
as well as about the elevated risk of recurrence, especially in the first 24 months
post-surgery. If the ovarian reserve is normal and the spermiogram unremarkable, the
patient can attempt to become pregnant 3 months post-surgery if she absolutely does
not want to delay childbearing. If the woman wishes to become pregnant later, is without
a partner or has diagnosed infertility, oocyte cryopreservation after prior hormonal
stimulation therapy and/or ovarian tissue cryopreservation are conceivable options.
Our case report demonstrates the practical relevance of the issues associated with
planning a pregnancy in women with cancer and presents possible options for action
in clinical routine.
