Journal Pre-proof

Borderline Ovarian Tumors: French Guidelines from the CNGOF. Part 1.

Epidemiology, Biopathology, Imaging and Biomarkers

C. Huchon, N. Bourdel, Wahab C. Abdel, H. Azaı̈s, S. Bendifallah,

P.A. Bolze, J.L. Brun, G. Canlorbe, P. Chauvet, E. Chereau, B.
Courbiere, T. De La Motte Rouge, M. Devouassoux-Shisheboran, C.
Eymerit-Morin, R. Fauvet, E. Gauroy, T. Gauthier, M. Grynberg, M.
Koskas, E. Larouzee, L. Lecointre, J. Levêque, F. Margueritte, E.
Mathieu D’argent, K. Nyangoh-Timoh, L. Ouldamer, J. Raad, E.
Raimond, R. Ramanah, L. Rolland, P. Rousset, C.
Rousset-Jablonski, I. Thomassin-Naggara, C. Uzan, M. Zilliox, E.
Daraı̈

PII: S2468-7847(20)30335-4
DOI: https://doi.org/10.1016/j.jogoh.2020.101965
Reference: JOGOH 101965

To appear in: Journal of Gynecology Obstetrics and Human Reproduction

Please cite this article as: Huchon C, Bourdel N, Abdel WC, Azaı̈s H, Bendifallah S, Bolze PA,
Brun JL, Canlorbe G, Chauvet P, Chereau E, Courbiere B, De La Motte Rouge T,
Devouassoux-Shisheboran M, Eymerit-Morin C, Fauvet R, Gauroy E, Gauthier T, Grynberg M,
Koskas M, Larouzee E, Lecointre L, Levêque J, Margueritte F, Mathieu D’argent E,
Nyangoh-Timoh K, Ouldamer L, Raad J, Raimond E, Ramanah R, Rolland L, Rousset P,
Rousset-Jablonski C, Thomassin-Naggara I, Uzan C, Zilliox M, Daraı̈ E, Borderline Ovarian
Tumors: French Guidelines from the CNGOF. Part 1. Epidemiology, Biopathology, Imaging
and Biomarkers, Journal of Gynecology Obstetrics and Human Reproduction (2020),
doi: https://doi.org/10.1016/j.jogoh.2020.101965
This is a PDF file of an article that has undergone enhancements after acceptance, such as
the addition of a cover page and metadata, and formatting for readability, but it is not yet the
definitive version of record. This version will undergo additional copyediting, typesetting and
review before it is published in its final form, but we are providing this version to give early
visibility of the article. Please note that, during the production process, errors may be
discovered which could affect the content, and all legal disclaimers that apply to the journal
pertain.

© 2020 Published by Elsevier.

Borderline Ovarian Tumors: French Guidelines from the CNGOF. Part 1. Epidemiology,

Biopathology, Imaging and Biomarkers.

Huchon C* (1), Bourdel N (2), Abdel Wahab C (3) , Azaïs H (4) , Bendifallah S (5), BOLZE PA (6), Brun JL
(7), Canlorbe G (4), Chauvet P (1), Chereau E (8), Courbiere B (9), De La Motte Rouge T (10),
Devouassoux-Shisheboran M (11), Eymerit-Morin C (12), Fauvet R (13), Gauroy E (14), Gauthier T (15),
Grynberg M (16), Koskas M (14), Larouzee E (14), Lecointre L (17), Levêque J (18), Margueritte F (15),
Mathieu D’argent E (5), Nyangoh-Timoh K (18), Ouldamer L (19), Raad J (16), Raimond E (20), Ramanah
R (21), Rolland L (9), Rousset P (22), Rousset-Jablonski C (23), Thomassin-Naggara I (3), Uzan C (4),
Zilliox M (17) , E. Daraï (5).

1. Service de gynécologie & obstétrique, GH Saint-Louis Lariboisière-Fernand Widal,

of
Hôpital Lariboisière, Université de Paris, 2, rue Ambroise Paré, 75010 Paris, France.
2. Service de Chirurgie Gynécologique, CHU de Clermont Ferrand, 1 Place Lucie Aubrac,
63 003 Clermont Ferrand.

ro
3. APHP.6 Service de Radiologie, Hôpital Tenon, 4 rue de la Chine, Faculté de Médecine
UPMC, Sorbonne Université, 75020, Paris.
4. AP-HP, Hôpital Pitié-Salpêtrière, service de chirurgie et oncologie gynécologique et

5.
-p
mammaire,. Faculté de Médecine UPMC, Sorbonne Université, 75013 Paris France.

Service de Gynécologie-Obstétrique et Médecine de la Reproduction, Hôpital Tenon,

Assistance Publique des Hôpitaux de Paris (AP-HP), Institut Universitaire de Cancérologie
re
(IUC), Centre CALG (Cancer Associé à La Grossesse), UMRS-938, Faculté de Médecine UPMC,
Sorbonne Université, 75013 Paris France.
6. Service de chirurgie gynécologique et oncologique, obstétrique,165 Chemin du Grand Revoyet,
lP

69310, Lyon Sud, Pierre Bénite, France. Université Lyon 1, 43 Boulevard du 11 Novembre 1918, 69100,
Villeurbanne, France
7. Service de Chirurgie Gynécologique, Centre Aliénor d’Aquitaine, Hôpital Pellegrin, 33076
Bordeaux. Société Française de Gynéco Pathologie, 81 rue verte, 76000 Rouen.
na

8. Service de gynécologie obstétrique, Hopital Saint Joseph, 13005 Marseille, France

9. Centre Clinico-Biologique d’AMP, Pôle Femmes – Parents- Enfants, AP-HM, Hôpital de La
Conception, 147 Bd Baille, 13005 Marseille.
10. Département d’oncologie médicale, Centre Eugène Marquis, 35000 Rennes France
11. Institut de Pathologie multi-sites des HOSPICES CIVILS DE LYON, Centre Hospitalier Lyon Sud,
ur

Centre de biologie et pathologie Sud, 165 Chemin du Grand revoyet, 69495 Pierre Bénite. Société
Française de Gynéco Pathologie, 81 rue verte, 76000 Rouen.
12. Service d'Anatomie et Cytologie Pathologiques, Hôpital Tenon, HUEP, 4 rue de la Chine, 75020
Jo

Paris, UPMC Paris VI, Sorbonne Universities, France. ; Institut de Pathologie de Paris,35 boulevard
Stalingrad, 92240 Malakoff
13. Service de Gynécologie Obstétrique, Centre Hospitalier Universitaire de Caen, 14000 Caen,
France

14. Service de Gynécologie-Obstétrique, Hôpital Bichat, 46 Rue Henri Huchard, Université

de Paris, 75018 Paris, France.
15. Service de Gynécologie-Obstétrique, Hôpital Mère-Enfant, CHU Limoges, 8 av
Dominique Larrey 87042 Limoges, France

1
16. Service de Médecine de la Reproduction, Hôpital Antoine Béclère, 157 rue de la Porte
de Trivaux, 92140 Clamart.
17. Centre Hospitalier Universitaire Hautepierre, Hôpital de Hautepierre, CHRU
Strasbourg, 1 avenue Molière, 67000 Strasbourg, France.
18. Département de Gynécologie Obstétrique et Reproduction Humaine, 16, boulevard de
Bulgarie, 35000 Rennes CHU Anne de Bretagne, UFR Médecine Université de Rennes 1, 35000
Rennes, Bretagne, France.
19. Département de Gynécologie, Centre hospitalier universitaire de Tours, Hôpital
Bretonneau, 2 Boulevard Tonnellé, 37000, Tours, France.
20. Département de Gynécologie Obstétrique, Institut Alix de Champagne, CHU Reims, 51000
Reims, France.

21. Pôle Mère-Femme, CHU Besançon, 3 boulevard Fleming, 25000 Besançon, France
22. Service de Radiologie, Centre Hospitalier Lyon Sud, HCL, EMR 3738, 165 Chemin du Grand
Revoyet, 69310, Lyon Sud, Pierre-Bénite, France. Université Lyon 1, 43 Boulevard du 11 Novembre

of
1918, 69100, Villeurbanne, France
23. Centre Léon Bérard, 28 Rue Laënnec, 69008, Lyon, France. Centre Hospitalier Lyon Sud, Pierre-
Bénite, France. Université Claude Bernard Lyon 1, EA 7425 Hesper, Health Service and Performance
Research, Domaine Rockefeller, 8 Avenue Rockefeller, 69373, Lyon Cedex 8, France

ro
-p
*Corresponding author : Cyrille Huchon. Service de gynécologie & obstétrique, GH Saint-Louis

Lariboisière-Fernand Widal, Hôpital Lariboisière, Université de Paris, 2, rue Ambroise Paré,

re
75010 Paris, France. E-mail address: cyrillehuchon@yahoo.fr
lP
na
ur
Jo

2
ABSTRACT

The incidence (rate per 100 000) of borderline ovarian tumors (BOTs) increases

progressively with age, starting at 15-19 years and peaking at around 4.5 cases per 100 000 at

an age of 55-59 years (LE3) with a median age of 46 years. The five year survival for FIGO

stages I, II, III and IV is 99.7% (95% CI: 96.2-100%), 99.6% (95% CI: 92.6-100%), 95.3% (95% CI:

91.8-97.4%) and 77.1% (95% CI: 58.0-88.3%), respectively (LE3).

An epidemiological association exists between the individual risk of BOT and family

of
history of BOT and certain other cancers (pancreatic, lung, bone, leukemia) (LE3), a personal

ro
history of benign ovarian cyst (LE2), a personal history of tubo-ovarian infection (LE3), the use

of a levonorgestrel intrauterine device (LE3), oral contraceptive use (LE3), multiparity (LE3),
-p
Hormonal replacement therapy (LE3), high consumption of Coumestrol (LE4), medical
re
treatment for infertility with progesterone (LE3) and non-steroidal anti-inflammatory drug use

(LE3).
lP

Screening for BOTs is not recommended for patients (Grade C).

The overall risk of recurrence of BOTs varies between 2% and 24%, with an overall
na

survival greater than 94% at 10 years, and the risk of an invasive recurrence of a BOT ranges

from 0.5% to 3.8%. The use of scores and nomograms can be useful in assessing the risk of
ur

recurrence, and providing patients with information (Grade C).

Jo

The WHO classification is recommended for classifying BOTs. It is recommended that

the presence of a microinvasive focus (<5mm) and microinvasive carcinoma (<5mm with an

atypical nuclei and a desmoplastic stroma reaction) within a BOT be reported. In cases of

serous BOT, it is recommended to specify the classic histological subtype or micropapillary /

cribriform type (Grade C).

3
When confronted with a BOT, it is recommended that the invasive or non-invasive nature of

peritoneal implants can be investigated based solely on the invasion and destruction of

underlying adipose or peritoneal tissue which has a desmoplastic stromal reaction where in

contact with the invasive clusters (Grade B).

For bilateral mucinous BOTs and / or in cases with peritoneal implants or peritoneal

pseudomyxoma, it is recommended to also look for a primitive digestive or pancreato-biliary

cancer (Grade C).

It is recommended to sample ovarian tumors suspected of being BOTs by focusing samples on

of
vegetations and solid components, with at least 1 sample per cm in tumors with a size less

ro
than 10 cm and 2 samples per cm in tumors with a size greater than 10 cm (Grade C). In cases

of BOTs and in the absence of macroscopic omental involvement after careful macroscopic
-p
examination, it is recommended to perform at least 4 to 6 systematic sampling blocks and to
re
include all peritoneal implants (Grade C).

It is recommended to consult an expert pathologist in gynecology when a BOT suspicion

lP

requires intraoperative extemporaneous histology (grade C).

Endo-vaginal and suprapubic ultrasonography are recommended for the analysis of an ovarian
na

mass (Grade A). In case of an undetermined ovarian lesion on ultrasonography, it is

recommended that a pelvic MRI be performed (Grade A). To analyze an adnexal mass with
ur

MRI, it is recommended to use an MRI protocol with T2, T1, T1 Fat Sat, dynamic and diffusion
Jo

sequences as well as gadolinium injection (Grade B). To characterize an adnexal mass with

MRI, it is recommended to include a score system for malignancy (ADNEX MR/O-RADS) (Grade

C) in the report and to formulate a histological hypothesis (Grade C). Pelvic MRI is

recommended to characterize a tumor suspected of being a BOT (Grade C). Macroscopic MRI

features should be analyzed to differentiate BOT subtypes (Grade C).

4
Pelvic ultrasound is the first-line examination for the detection and characterization of

adnexal masses during pregnancy (Grade C). Pelvic MRI is recommended from 12 weeks of

gestation in case of an indeterminate adnexal mass and should provide a diagnostic score

(Grade C). Gadolinium injection must be minimized as fetal impairment has been proven

(Grade C).

It is recommended that serum levels of HE4 and CA125 be evaluated and that the ROMA score

for the diagnosis of an indeterminate ovarian mass on imaging be used (grade A). In case of

suspicion of a mucinous BOT on imaging, dosage of serum levels of CA 19-9 can be considered

of
(Grade C).

ro
If the determination of tumor markers is normal preoperatively, routine dosage of tumor

markers in BOT follow-up is not recommended (Grade C).

-p
In case of preoperative elevation in tumor markers, the determination of serum CA 125 levels
re
is recommended in the follow-up of BOT (Grade B). When conservative treatment of a BOT

has been adopted, the use of endovaginal and transabdominal ultrasonography is

lP

recommended during follow-up (Grade B).

na

Key words: Borderline ovarian tumor, Guidelines

ur
Jo

5
Introduction

Borderline Ovarian Tumors (BOTs) are believed to account for 10 to 20% of all epithelial

tumors of the ovary. Their particularity is that they occur on average 10 years earlier and have

a much better prognosis than ovarian carcinomas. The two main histological forms of BOTs

are serous and mucinous. Survival, all stages combined, is 95% at 5 years and 90% at 10 years.

The preoperative diagnosis of BOTs is difficult and differentiating between benign and

malignant tumors requires precise preoperative examinations. Knowledge of the

ultrasonographic and MRI characteristics of BOTs allows for a better determination of the

of
surgical procedures that can be performed. Given the good prognosis of BOTs, the young age

ro
of patients at onset and the societal evolution towards later pregnancies, surgical treatment

of patients with a desire for pregnancy is evolving to be increasingly conservative. Currently

-p
the management of BOTs is disparate across France. The surgical possibilities of conservative
re
treatment, including for patients presenting advanced stages of BOTs, must be known to all

surgeons and in particular the absence of benefit of systematic hysterectomy or systematic

lP

appendectomy and the possibility of laparoscopy to reduce postoperative morbidity. The

treatment of BOTs is surgical with a goal of complete cytoreduction without adjuvant therapy.
na

Staging or even restaging procedures must be integrated into the treatment strategy of BOTs

(cystectomy, adnexectomy) and must provide the FIGO stage and the detection of implants.
ur

In patients wishing to preserve their fertility, a consultation with a specialist in Assisted

Jo

Reproduction Techniques (ART) must be offered. However, patients should be informed of

the higher risk of recurrence after conservative treatment compared to radical treatment.

Follow-up should be long-lasting as recurrences may occur more than 10 years after initial

treatment. Finally, in patients undergoing radical treatment, the possibility of hormone

replacement therapy should be discussed.

6
The “Collège National des Gynécologues et Obstétriciens Français” (CNGOF) thus decided to

issue guidelines for clinical best practice and management of BOTs with the objective of

improving the standard of healthcare. The methodology used in the elaboration of these

guidelines followed the standards of the French National Authority for Health (HAS) (1). This

text summarizes the epidemiology, the biopathology and the contribution of biomarkers and

imaging to the diagnosis of BOTs.

These guidelines are intended for healthcare professionals involved in the diagnosis, initial

treatment and follow-up of patients with BOTs: surgeons, obstetricians and gynecologists,

of
medical oncologists, midwives, anatomo-pathologists, general practitioners and radiologists.

ro
-p
re
lP
na
ur
Jo

7
Epidemiology (2 ,3)

1/ Descriptive epidemiology

In France, no specific epidemiological data exists with respect to the incidence or prevalence

of BOTs.

BOTs are believed to account for 10% to 20% of all epithelial tumors of the ovary. Their

particularity is their age at onset, on average 10 years earlier, and their prognosis which is

much better than that of ovarian carcinomas. At diagnosis, the median age of patients is 46

years (LE3).

of
Survival, all stages combined, is 95% at 5 years and 90% at 10 years. The five-year survival for

ro
stages I, II, III, IV is 99.7% (95% CI: 96.2-100%), 99.6% (95% CI: 92.6-100%), 95.3% (95% CI:

91.8-97.4%) and 77.1% (95% CI: 58.0-88.3%) (LE3), respectively. The five-years survival for
-p
serous and mucinous tumors is 99.7% (95% CI: 99.2-99.9%) and 98.5% (95% CI: 96.9-99.3%),
re
respectively (LE3).

The incidence (rate per 100 000) of BOTs increases progressively from the age of 15 to 19
lP

years and peaks at about 4.5 cases per 100 000 at an age of 55-59 years (LE3).

In the presence of an ovarian mass believed to be benign, the standardized risk ratio of
na

diagnosing a serous or mucinous BOT is 1.69, (95% CI: 1.39-2.03) and 1.75, (95% CI: 1.45-2.10),

respectively (LE2). The two main factors associated with an excess risk of diagnosing a BOT in
ur

the presence of an ovarian mass are an age of under 40 years and a solid mass (versus cystic)
Jo

(LE2).

Unilateral forms (79.7% of cases) are predominant compared to cancers (45.3%) (<0.001) and

FIGO stage I tumors account for approximately 63.7% of BOT cases (LE3).

2/ Risk factors of BOT

8
There is an epidemiological association between individual risk of BOT and family history of

BOTs or other cancers (pancreatic, lung, bone and leukemia) (LE3).

The risk of BOTs increases with a personal history of a benign ovarian cyst (LE2) or pelvic

infectious disease (LE3).

There is an epidemiological association between smoking and the risk of a mucinous BOT

(LE2).

An epidemiological association exists between being overweight/obesity and the risk of

of
serous BOT (LE2). However, no association exists between physical activity and the risk of

ro
BOTs (LE4).

No epidemiological association has been found between a personal history of tubal ligation
-p
and the risk of BOTs (LE4).
re
An epidemiological association exists between exposure to hormones and the risk of BOTs

(LE3). The relative risk of developing a BOT is 0.76 (95% CI: 0.57-0.99) with a Levonorgestrel
lP

intrauterine device (LE3). Data concerning the risk of BOTs with oral contraception varies

between no epidemiological association and reduced risk of BOT. The relative risk of BOTs
na

decreases as parity increases, with a relative risk of BOT at 0.44 (0.26 to 0.75) among

multiparous women (LE3). There is an epidemiological association between BOTs and

ur

hormonal replacement therapy (LE2) as well as treatment of infertility with progesterone

Jo

(LE3).

There is an epidemiological association between high Coumestrol consumption and the risk

of developing a BOT (LE4). High vitamin D consumption is associated with a decreased risk of

serous BOTs (LE4). There is conflicting evidence in literature regarding the epidemiological

associations between coffee, tea and caffeine consumption and the risk of developing a BOT.

9
There is an epidemiological association between the use of non-steroidal anti-inflammatory

drugs (NSAIDs) and the risk of BOTs (LE3) and an association exists between the use of

paracetamol and the risk of mucinous BOTs (LE3).

3/ Screening and risk of BOT

In the absence of sufficient publications on clinical, imaging and biological markers for BOTs,

including risk factors, no guidelines can be made on BOT screening strategies.

4/ Recurrences

of
The overall risk of BOT recurrence is estimated at between 2% and 24% (LE2). The risk of BOT

ro
recurrence with an invasive form ranges from 0.5% to 3.8% (LE2). The delay until recurrence

can be long, sometimes over 10 years after initial treatment (LE2).

-p
An age of under 40 years is a risk factor for recurrence (LE3). An age greater than 50 years is a
re
pejorative prognostic factor associated with lower survival and a higher risk of progression to

invasive carcinoma (LE3). The risk of recurrence increases with the initial FIGO stage (LE3).
lP

Conservative BOT treatment is associated with a higher risk of recurrence compared to radical

treatment of bilateral adnexectomy (LE2), although survival is not impacted (LE2).

na

In cases of serous BOTs, complete initial surgical staging significantly reduces the risk of

recurrence, but with no impact on overall survival (LE2). BOT residue decreases recurrence-
ur

free survival (LE4).

Jo

Studies are inconsistent as to the impact of micropapillary architecture and microinvasion

with respect to non-invasive recurrences. In serous BOTs, micropapillary architecture and

microinvasion do not appear to be risk factors of recurrence (LE3).

10
The presence of micropapillary architecture increases the risk of lethal recurrences (invasive

recurrences or death) of serous BOTs, (LE2). In cases of serous BOTs with implants (invasive

or not), the risk of recurrence (invasive or not) is increased compared to a serous BOT without

implants (LE2). For mucinous BOTs, the presence of intraepithelial lesions or microinvasion is

not associated with an excess risk of lethal recurrence (LE3).

Laparoscopy is not associated with a higher risk of recurrence compared to laparotomy (LE2).

The presence of lymph node involvement in cases of BOT is not associated with a higher risk

of recurrence (LE3).

of
An elevation at diagnosis of CA 125 above the norm (≥ 35 IU/ml) for serous BOTs is an

ro
independent risk factor of recurrence (LE4).

The score of Ouldamer et al. and the nomogram of Bendifallah et al. are efficient tools for
-p
assessing the risk of BOT recurrence after surgical treatment (LE3).
re
The use of these scores and nomograms after surgical treatment of a BOT can therefore be

useful to assess the risk of recurrence and to provide information to patients (Grade C).
lP
na
ur
Jo

11
Biopathology (4 )

1/ Histological diagnostic criteria and classification of borderline ovarian tumors:

BOTs include 6 histological subtypes (serous, mucinous, seromucinous, endometrioid, clear

cell, Brenner's), of which serous and mucinous are the most frequent (LE2). It is recommended

to classify BOTs according to the WHO classification.

Micropapillary/cribriform serous BOTs are more often bilateral, exophytic, FIGO stage >1 and

with invasive peritoneal implants compared to conventional serous BOTs (LE2). Serous BOTs

of the micropapillary/cribriform type are considered to be the precursor to low-grade invasive

of
serous carcinoma (LE3). In case of serous BOTs, it is recommended to specify the classic

ro
histological or micropapillary/cribriform subtype (Grade C). For serous BOTs with a

micropapillary patterns, it is recommended to investigate peritoneal implants for invasive

-p
potential (Grade B).
re
lP

The presence of invasive implants characterized by the invasion of underlying adipose tissue

is associated with a greater risk of recurrence, and death, than non-invasive implants (LE2). It
na

is recommended to verify whether or not implants are invasive (Grade B). An invasive implant

is defined by the destruction of the underlying adipose or peritoneal tissue associated with a
ur

desmoplastic stromal reaction on contact with the invasive clumps. In the absence of
Jo

underlying adipose tissue and in case of uncertainty about the invasiveness of the implant, it

is recommended to use the term implant of undetermined type.

For mucinous BOTs, the presence of peritoneal implants and/or bilaterality strongly raise the

possibility of ovarian metastasis of a mucinous adenocarcinoma of non-ovarian origin (LE3).

In case of bilateral mucinous BOTs and/or peritoneal implants or peritoneal pseudomyxoma,

12
it is recommended to look for a primitive digestive or pancreato-biliary cancer with the

appropriate examinations (Grade C).

Clear cell BOTs are exceptional and are often associated with an invasive contingent of clear

cell carcinoma (LE3). In the case of a clear cell BOT, it is recommended that the tumor be

sampled extensively to rule out the presence of an associated clear cell adenocarcinoma

(Grade C).

If there is any doubt about the histological subtype, an immunohistochemical study can help

of
in classifying the BOT histological subtype (LE3). No immunohistochemical marker is able to

ro
distinguish a borderline tumor from an invasive carcinoma. Invasiveness is based solely on

morphological findings. Therefore, systematic immunohistochemical analysis is not

recommended when searching for invasion (Grade C).

-p
re
It is recommended that the presence of microinvasive focus (< 5mm) or microinvasive

carcinoma (< 5 mm with atypical nuclei and a desmoplastic stroma reaction) within a BOT be
lP

reported.
na

In view of the variable reproducibility of both tumor and implant diagnosis and the overall risk

of over-diagnosis, it is recommended that a review or double reading by an expert pathologist

ur

in gynecology be done in the following situations (INCa guidelines- TMRG Network) (Grade C)

(5):
Jo

1- In case of doubt of: the borderline nature of the tumor, the histological subtype of BOT, or

the invasive nature of an implant.

2- Systematically for non-classical serous BOT and in the presence of peritoneal implants.

3- Systematically for mucinous and clear-cell BOT.

13
2. Histological Methodology

It is recommended that ovarian tumors suspected of being borderline should be sampled

extensively in order to rule out an invasion. Samples should be directed especially towards

solid component and vegetations, to the tumor capsule and to areas with different

macroscopic appearances, with at least 1 sample per cm for tumors with a size < 10 cm and 2

samples per cm for tumors with a size > 10 cm as well as on all papillary and solid areas (Grade

of
C).

ro
Serous BOTs with micropapillary patterns require a more extensive sampling (2 blocks per cm)

of the tumor to rule out microinvasive or frankly invasive areas (LE3). In case of a serous BOT
-p
with a micropapillary pattern (Grade C), sampling with 2 blocks per cm of tumor is
re
recommended (Grade C).

Mucinous BOTs are more heterogeneous than other histological subtypes and are associated
lP

with a non-negligible risk of microinvasion or invasive carcinoma, justifying more extensive

sampling than for serous BOTs (LE3). In light of the tumor heterogeneity of mucinous BOTs,
na

sampling of 2 blocks per cm of tumor is recommended even if the tumor size is < 10 cm (grade

C).
ur

In case of microinvasion, or intraepithelial carcinoma, it is recommended that the tumor be

Jo

further sampled for more extensive invasion (Grade C).

Peritoneal implants must be included in their entirety to formally exclude an invasive

component.

14
Sampling of an omentectomy specimen without macroscopic lesions, with an average of 6

blocks (depending on the size of the specimen) allows for the detection of the majority of

microscopic involvement of the omentum (LE3). Thus, in the absence of macroscopic

involvement of the omentum, after careful macroscopic examination, it is recommended that

at least 4 to 6 blocks of systematic sampling be performed on an omentectomy specimen

(depending on the size of the epiploic resection) to allow for the detection of the majority of

microscopic involvement (Grade C).

of
It is recommended that all lymph nodes and implants be included (Grade C).

ro
3/ Diagnostic value of intraoperative histology

-p
Extemporaneous histology is less efficient in BOTs than for benign or malignant ovarian

tumors. The concordance rate between extemporaneous and definitive histology is 69-73%
re
for BOTs (LE2). The rate of underdiagnosis (borderline at extemporaneous, malignant at
lP

definitive) is 20% to 21%. The rate of overdiagnosis (borderline in extemporaneous, benign to

definitive) is 6% to 10% (LE2). Recent studies tend to support the improvement of diagnostic
na

performances of extemporaneous histology with less overdiagnosis (LE4).

Factors decreasing diagnostic performance of extemporaneous histology in BOTs are

ur

mucinous subtype and unilaterality (LE2). Large tumor size increases the risk of discrepancy

between the extemporaneous and the definitive histology (LE4). The Impact that the level of
Jo

expertise that the pathology department and that the gynecological pathologist has on

performance is controversial. The diagnostic performance of extemporaneous histology in

BOTs is not significantly modified by the type of structure (general pathology department vs.

gynecopathology department, university hospital centre and cancer centre vs. non-

universitary hospital) (LE4). However, in a general anatomopathology department, the

15
performance of extemporaneous histology is increased when the diagnosis is made by a

pathologist expert in gynecology compared to a pathologist non-expert in gynecology (LE4). It

is therefore recommended that a pathologist with an expertise in gynecology should be

consulted when extemporaneous histology is required for a suspected BOT (Grade C).

4/ Performance of laparoscopic surgical specimen analysis

The diagnostic performance of cytology for the diagnosis of BOTs is poor (LE3). In addition,

unprotected puncture of a suspected ovarian cyst exposes the patient to the risk of peritoneal

of
dissemination. It is recommended that a suspicious ovarian cyst not be punctured (grade C).

ro
At the end of surgery, there is no reason to address cystic fluid to cytology when the cyst or

ovary is available for histology (LE3). -p

Peritoneal swabs from implants accompanied by a crown of apparently healthy tissue
re
facilitate the diagnosis of invasion (LE4). For BOTs, it is recommended that the implant sample

be large enough to contain macroscopically normal adjacent tissue (grade C).

lP

5/ Methods of fixation and tissue transport and preservation

na

Pre-analytical factors such as cold ischemia time, type of fixative and fixation time modify

morphology, protein and nucleic acid (LE4) preservation. It is recommended that tissue
ur

samples be fixed in a neutral buffered formalin (with 4 % formaldehyde) no later than 1 hour
Jo

after removal (grade C). Vacuum packing and storage at +4°C can be an alternative, only for

large operative specimens, that allows for this time to be extended to a maximum of 48 hours

(grade C). It is recommended that tissue sample fixation be done at least 6 h (for biopsies)

before pathological examination (grade C).

16
6/ The use of biological tools

No biomolecular tool has clearly demonstrated a diagnostic role in the management of a

patient with BOTs (LE4). The evaluation methods for the most interesting tools (TILs, BRAF

and KRAS mutations) are not standardized. The presence of BRAF and KRAS mutations in the

tumor or implants have no diagnostic utility but could have an influence on prognosis (LE4).

Testing for gene mutations (BRCA, MMR, BRAF, KRAS) in women treated for BOTs is not

recommended (grade C).

of
7/ Quality Criteria of the Pathologic Examination and Elements of the Report

ro
The presence of an ovarian capsule rupture has an impact on the FIGO classification (LE2). It

is recommended that the pathology report include the gross examination of tissue samples
-p
including a description of the specimens received and their integrity (intact or ruptured
re
ovarian capsule), tumor sites and a description of the omentum (Grade B).

The application of BOT microscopic diagnostic criteria, possibly associated with an antibody
lP

panel, allows a better definition of the histological subtype and improves intra- and inter-

observer coherence (LE2). It is recommended to specify the histological subtype of the BOT,
na

and for serous BOTs, the presence or absence of micropapillary patterns, the presence of

implants (invasive or non-invasive), the presence of microinvasions, other tumor locations,

ur

the results of the peritoneal washing and the FIGO classification (Grade B).
Jo

In case of doubt concerning the BOT histological subtype, it is recommended that an

immunohistochemical study with an optimal panel of antibodies be performed (Grade C).

In case of any doubt between a benign, BOT and malignant tumor, further lesion samples

should be referred for review to expert pathologist in gynecology (Grade C) (5).

17
Imaging (6)

1/ Diagnosis of adnexal masses (Figure 1)

Ultrasonography is the first-line imaging technique for the diagnosis of adnexal masses and

magnetic resonance imaging (MRI) is the most accurate non-invasive technique for the

preoperative diagnosis of epithelial tumors of the ovary. The guidelines published jointly by

INCa, CNGOF and the FRANCOGYN group, ARCAGY-GINECO for the initial management of

epithelial tumors of the ovary are presented below (7):

1. Transvaginal and suprapubic ultrasonography are recommended for the analysis of an

of
ovarian mass (Grade A).

ro
2. When ultrasonography is performed by an expert, a subjective analysis is recommended

(Grade A). -p
3. In case of ultrasonography performed by a non-expert, the use of simple rules ("Simple
re
Rules") is recommended (Grade A). This approach must be combined with a subjective analysis

in order to match the examination quality of an expert sonographer (Grade A).

lP

4. In cases of undetermined ovarian lesions on transvaginal and suprapubic ultrasonography,

a pelvic MRI is recommended (Grade A).

na

5. For adnexal mass analysis, when an MRI is performed, MRI protocol with T2, T1, T1

sequences with fat saturation, diffusion, injected dynamics, and after gadolinium injection is
ur

recommended (Grade B).

Jo

6. To characterize an adnexal mass on MRI, it is recommended to include a malignancy risk

score (ADNEX MR/O-RADS) (Grade C) in the report and to formulate a histological hypothesis

(Grade C).

2/ Differential diagnosis between BOT and benign ovarian lesions

18
For ovarian cysts with endocystic vegetations, the predictive signs of benignity are a low

number of vegetations, a small size, a presence of calcifications, an absence of Doppler flow

when size is greater than 10 mm with ultrasonography (LE4) and a type 1 curve in MRI (LE4).

MRI is recommended for indeterminate lesions (Grade A) or for lesions suspected of being

BOTs with ultrasonography (Grade B).

There is no data to support the use of CT or PET-CT in this indication.

3/ Differential diagnosis between BOT and ovarian cancers

of
Morphological criteria with ultrasonography and MRI can help to differentiate a BOT from an

ro
invasive tumors of any grade (LE2): solid portions, a crescent sign, a round or oval shape, well-

defined contours, a purely cystic character, exophytic or branched vegetations and the
-p
apparent diffusion coefficient (ADC).
re
Pelvic MRI is recommended to characterize a tumor suspect of being a BOT or an invasive

tumor with ultrasonography (Grade C).

lP

PET-CT is not efficient in differentiating BOTs from invasive tumors (LE2). CT may be useful in
na

excluding peritoneal carcinomatosis in cases where there is doubt between a BOT and an

invasive tumor (Grade C).

ur
Jo

4/ Combining biological markers and imaging criteria

The combination of ultrasonography and biological parameters with menopausal status to

characterize an adnexal mass as a BOT has been studied in a limited number of studies, and

with a low level of evidence due to their predominantly retrospective nature and their

heterogeneity in terms of FIGO stages and histological types.

19
No recommendation can therefore be made on the use of scores combining ultrasonographic

and biological parameters with menopausal status for the diagnosis of BOTs.

5/ Imaging of implants

The diagnostic performance of imaging to detect peritoneal metastases of BOTs is not known.

The use of imaging to analyze the invasiveness of peritoneal metastases of BOTs has not been

evaluated.

of
6/ Morphological diagnosis of histological subtypes

ro
The association of macroscopic signs on MRI allows for the differentiation between the

serous, seromucinous and mucinous (intestinal type) BOT subtypes, despite similarities
-p
between certain presentations (LE3). The analysis of macroscopic MRI signs should therefore
re
be performed in order to differentiate between BOT subtypes (Grade C).
lP

7/ Prediction of the feasibility of surgical conservative treatment

Prediction of the possibility of conservative treatment using imaging has been studied in a
na

limited number of studies with conflicting results. No recommendations can be made

regarding imaging as a tool for the prediction of the feasibility of conservative treatment of
ur

BOTs.
Jo

8/ BOT and pregnancy.

During pregnancy, the additional difficulty will be to differentiate BOTs from functional

pathologies specific to pregnancy (hyperreactio luteinalis in particular). Recent data has

confirmed the maternal-fetal safety of MRI if performed after the first trimester of pregnancy.

20
However, pelvic ultrasonography is the first-line examination for the detection and

characterization of adnexal mass associated with pregnancy (Grade C).

Pelvic MRI is recommended from the 12th week of gestation in case of an indeterminate

adnexal mass and should be concluded with a diagnostic score (ADNEX MR/O-RADS) (Grade

C).

The injection of gadolinium should be restricted due to the proven risks to the fetus and

should be discussed on a case-by-case basis after informing the patient (Grade C).

of
9/ Imaging of recurrences

ro
Serous BOT recurrences usually appear as a thin-walled liquid cyst with vegetations,

corresponding, in the IOTA classification, to a solid unilocular cyst (LE2). In this case, a cyst size
-p
of less than 20 mm is not sufficient to eliminate the diagnosis of a BOT recurrence (LE2).
re
Recurrences of mucinous BOTs mostly appear as multilocular or as solid multilocular cysts

(LE4).
lP

The ovarian crescent sign is a criterion in favor of a non-invasive recurrence in the case of a

complex mass (NP4).

na

After initial conservative treatment, and even among patients with a normal clinical

examination, transvaginal and suprapubic ultrasonography are recommended during follow-

ur

up to detect recurrences (Grade B).

Jo

Tumor markers ( 3, 8 )

1/ Relevance of tumor markers in the initial diagnosis for the preoperative differential

diagnosis between benign, borderline and malignant ovarian tumors.

21
A/ CA125

CA125 (cancer antigen) also known as mucin 16 or MUC16 is a glycoprotein encoded by the

MUC16 gene, expressed by the epithelial cells of several normal tissue types and potentially

over-expressed in the case of epithelial cancer. CA125 can also be elevated in various benign

pathologies (uterine fibroids, endometriosis, benign inflammatory pathologies), during

pregnancy and sometimes during ovulation and menstruation. This explains its low specificity

especially in non-menopaused patients.

of
Preoperative serum CA125 levels are higher among patients with a serous BOT (LE4) and

ro
increase with the size and FIGO stage of BOTs (LE4), especially for serous BOTs. A normal

CA125 level does not rule out a BOT (LE4). -p

There is little evidence in literature regarding the discriminating value of serum CA125 levels
re
to differentiate between benign, BOT and malignant ovarian tumors.

No recommendation can be proposed for the use of the serum CA125 level for preoperative
lP

differential diagnosis between a presumed benign ovarian tumor, a BOT and a malignant

ovarian tumor.
na

B/ CA19 9
ur

CA19 9 is a glucolipid (monosialoganglioside) contained in cell membranes and a marker for

Jo

mucinous tumors: it can therefore be elevated in gastrointestinal tumors (such as colorectal

cancer), esophageal and pancreatic cancer.

For BOTs, the preoperative elevation of CA19 9 is less frequent than that of CA125.

Preoperative CA19 9 levels increase with the size and FIGO stage of BOTs and are higher in

mucinous BOTs (LE4).

22
There is little evidence in literature regarding the discriminating value of serum CA19 9 levels

to differentiate between benign ovarian tumors, BOTs and malignant ovarian tumors.

No recommendation can be proposed regarding the use of a CA19 9 assay for the preoperative

differential diagnosis between benign ovarian tumors, BOTs and malignant ovarian tumors.

When a mucinous BOT is suspected on imaging, a CA19 9 assay may be proposed (Grade C).

C/ CEA

of
CEA's positivity rates are lower than those of CA125 and CA19 9 in BOTs (LE4). There is little

ro
evidence in literature regarding the discriminating value of serum CEA to differentiate

between benign ovarian tumors, BOTs, and malignant ovarian tumors.

-p
re
No recommendation can therefore be proposed regarding the use of a CEA assay for the

preoperative differential diagnosis between benign ovarian tumors, BOTs, malignant ovarian
lP

tumors.
na

D/ HE4

HE4 (human epididymis protein 4) is encoded by the WFDC2 (WAP four-disulfide core domain)
ur

gene which is located on chromosome 20q12-13.

Jo

Serum levels of HE4 do not differ according to the histological subtypes of BOTs (LE4).

Faced with an indeterminate ovarian mass on imaging, according to INCa 2018 "Guidelines for

Patients with Epithelial Ovarian Cancer" [7], it is recommended that serum HE4 (Grade A) and

serum CA125 (Grade A) be measured and that the ROMA score be used (Grade A).

23
2/ Role of tumor markers and scores (clinical and biological) for the diagnosis of BOT in cases

of an indeterminate ovarian mass on imaging

There is little evidence in literature regarding the use of tumor markers and scores for the

diagnosis of BOTs.

No recommendation can therefore be proposed regarding the use of specific markers and

scores for the specific diagnosis of BOTs.

of
3/ Value of tumor markers in the evaluation of pejorative histo-pronostic factors and

ro
recurrence

High preoperative tumor marker levels are a predictive factor for the existence of implants
-p
(CA 125) (LE4), and an independent factor for BOT recurrence (CA 125) (LE4). An elevation of
re
CA 125 ≥ 35 IU/ml at diagnosis of serous BOT is an independent risk factor for recurrence

(LE4). Tumor markers detect recurrence with a sensitivity of 33% to 66.6% (LE2).
lP

When tumor markers are normal preoperatively, their evaluation is not recommended as part

of a BOT follow-up strategy (Grade C).

na

In case of a preoperative elevation of tumor markers, it is recommended to use the CA 125

assay during BOT follow-up (Grade B).

ur
Jo

24
Acknowledgments to reviewers:

J. Alexandre (oncologist, Paris), Y. Aubard (gynecologist surgeon, CHU, Limoges), C. Akladios

(gynecologist surgeon, Strasbourg), FX. Aubriot (gynecologist surgeon, Paris), AS. Bats
(gynecologist surgeon, Paris), L. Benoit (gynecologist surgeon, Paris), JL. Benifla (gynecologist
surgeon, Paris), P. Berveiller (obstetrician, Poissy), M. Canis (gynecologist surgeon, Clermont-
Ferrand), X. Carcopino (gynecologist surgeon, Marseille), N. Chabbert-Buffet (endocrinologist,
Paris), J. Dubuisson (gynecologist surgeon, Genève), L. Fournier (radiologist, Paris), C. Genestié
(anatomo-pathologist, Villejuif), F. Golfier (gynecologist surgeon, Lyon), A. Guivarch
(gynecologist, Rennes), T. Hebert (gynecologist surgeon, Tours), M. Mezzadri (gynecologist
surgeon, Paris), C. Mimoun (gynecologist surgeon, Paris), A. Jalaguier (radiologist, Marseille),
V. Lavoué (gynecologist surgeon, Rennes), MA. Le Frère-Belda (anatomo-pathologists, Paris),
G. Legendre (gynecologist surgeon, Angers), D. Toth (gynecologist surgeon, Brive-la-Gaillarde),
LP. Tran (gynecologist surgeon, La Réunion).

of
Promotor of the guidelines:
CNGOF (Collège National des Gynécologues et Obstétriciens Français) 91 boulevard de
Sébastopol – 75002 Paris

ro
Organisation Commitee : Emile Daraï (Président, Paris), Nicolas Bourdel (Coordonnator,
Clermont-Ferrand), Cyrille Huchon (Méthodologist, Paris).
-p
re
lP
na
ur
Jo

25
References

1. Huchon C, Bourdel N, Daraï E. [Borderline Ovarian Tumours: CNGOF guidelines for

Clinical Practice - Methods & Organization]. Gynecol Obstet Fertil Senol. 2020
Mar;48(3):236-238. doi: 10.1016/j.gofs.2020.01.023. Epub 2020 Jan 28.
2. Ouldamer L, Body G, Daraï E, Bendifallah S. [Borderline Ovarian Tumours: CNGOF
Guidelines for Clinical Practice - Epidemiological Aspects and Risk Factors]. Gynecol
Obstet Fertil Senol. 2020 Mar;48(3):239-247. doi: 10.1016/j.gofs.2020.01.012. Epub
2020 Jan 28.
3. Margueritte F, Sallee C, Lacorre A, Gauroy E, Larouzee E, Chereau E, De La Motte Rouge
T, Koskas M, Gauthier T. [Borderline Ovarian Tumours: CNGOF Guidelines for Clinical

of
Practice - Epidemiology and Risk Factors of Relapse, Follow-up and Interest of a
Completion Surgery]. Gynecol Obstet Fertil Senol. 2020 Mar;48(3):248-259. doi:

ro
10.1016/j.gofs.2020.01.013. Epub 2020 Jan 28.
4. Eymerit-Morin C, Brun JL, Vabret O, Devouassoux-Shisheboran M. [Borderline ovarian
-p
tumours: CNGOF Guidelines for clinical practice - Biopathology of ovarian borderline
tumors]. Gynecol Obstet Fertil Senol. 2020 May 16:S2468-7189(20)30187-2. doi:
re
10.1016/j.gofs.2020.05.007.
5. « Référentiel de l’observatoire des tumeurs rares gynécologiques : les tumeurs
lP

ovariennes borderline, Mai 2019 ; https://www.ovaire-rare.org.

6. Abdel Wahab C, Rousset P, Bolze PA, Thomassin-Naggara I. [Borderline Ovarian
na

Tumours: CNGOF Guidelines for Clinical Practice - Imaging]. Gynecol Obstet Fertil
Senol. 2020 Mar;48(3):260-276. doi: 10.1016/j.gofs.2020.01.014. Epub 2020 Jan 28.
7. Lavoue V, Huchon C, Akladios C, Alfonsi P, Bakrin N, Ballester M, Bendifallah S, Bolze
ur

PA, Bonnet F, Bourgin C, Chabbert-Buffet N, Collinet P, Courbiere B, De la Motte

Rouge T, Devouassoux-Shisheboran M, Falandry C, Ferron G, Fournier L, Gladieff L,
Jo

Golfier F, Gouy S, Guyon F, Lambaudie E, Leary A, Lecuru F, Lefrere-Belda MA,

Leblanc E, Lemoine A, Narducci F, Ouldamer L, Pautier P, Planchamp F, Pouget N,
Ray-Coquard I, Rousset-Jablonski C, Senechal-Davin C, Touboul C, Thomassin-Naggara
I, Uzan C, You B, Daraï E. Management of epithelial cancer of the ovary, fallopian
tube, and primary peritoneum. Short text of the French Clinical Practice Guidelines

26
 issued by FRANCOGYN, CNGOF, SFOG, and GINECO-ARCAGY, and endorsed by INCa.
Eur J Obstet Gynecol Reprod Biol. 2019 May;236:214-223.

8. Nyangoh-Timoh K, Bendifallah S, Dion L, Ouldamer L, Levêque J. [Borderline Ovarian

Tumours: CNGOF Guidelines for Clinical Practice - Value of Tumor Markers]. Gynecol
Obstet Fertil Senol. 2020 Mar;48(3):277-286. doi: 10.1016/j.gofs.2020.01.015. Epub
2020 Jan 28.

of
ro
-p
re
lP
na
ur
Jo

27
Figure list.

Figure 1. Imaging of Borderline Ovarian Tumors. 2020 Guidelines of the French National
College of Gynecologists and Obstetricians (CNGOF).

of
ro
-p
re
lP
na
ur
Jo

28