European Journal of Obstetrics & Gynecology and Reproductive Biology 181 (2014) 267–274

Circulating endometrial cells in peripheral blood

Vladimir Bobek a,b,c,*, Katarina Kolostova a,d, Eduard Kucera e

a Department of Laboratory Genetics, University Hospital Kralovske Vinohrady, Prague, Czech Republicb Department of Histology and Embryology, Wroclaw Medical University, Wrocław, Polandc 3rd Department of Surgery, First Faculty of Medicine Charles University in Prague and University Hospital Motol, Prague, Czech Republicd Department of Tumor Biology, 3rd Faculty of Medicine, Charles University Prague, Ruska 87, 100 97 Prague, Czech Republice Department of Gynecology and Obstetrics, 3rd Faculty of Medicine Charles University Prague and University Hospital Kralovske Vinohrady, Srobarova 50,

100 97 Prague, Czech Republic

A R T I C L E I N F O

Article history:

Received 26 February 2014

Received in revised form 23 July 2014

Accepted 29 July 2014

Keywords:

Endometriosis

CEC

Circulating endometrial cells

Cultivation

MetaCell

Biomarker

A B S T R A C T

Objectives: Endometriosis is a common disorder amongst women of reproductive age. Despite extensive

research, no reliable blood tests currently exist for the diagnosis of endometriosis

Study design: We report several new approaches enabling study of cell specific characteristic of

endometrial cells, introducing enrichment and culturing of viable circulating endometrial cells (CECs)

isolated from peripheral blood (PB) and peritoneal endometrial cells (PECs) from peritoneal washing

(PW). Size-based enrichment method (MetaCell1, Czech Republic) has been used for the filtration of PB

and PW in patients with diagnosed endometriosis.

Results: The PECs were found in the PW in all of the tested patients (n = 17), but CECs) only in 23.5% (4/

17) cases. Their endometrial origin has been proved by immunohistochemistry. PECs were successfully

cultured in vitro directly on the separating membrane (9/17) exhibiting both endometrial cell

phenotypes: stromal and glandular within the culture. CECs were successfully cultured in the two of the

four positive cases, but in none of them confluence has been reached. The occurrence in CECs in PB is

clear and very specific evidence of an active endometrial disease.

Conclusions: We demonstrated efficient, quick and user friendly endometrial cells capture platform

based on a cell size. Furthermore, we demonstrated an ability to culture the captured cells, a critical

requirement for post-isolation cellular analysis directed to better understanding of endometriosis

pathogenesis.

� 2014 Elsevier Ireland Ltd. All rights reserved.

Contents lists available at ScienceDirect

European Journal of Obstetrics & Gynecology andReproductive Biology

jou r nal h o mep ag e: w ww .e lsev ier . co m / loc ate /e jo g rb

Introduction

Endometriosis is a common disorder amongst women ofreproductive age. Endometriosis, the presence of endometrial-liketissue outside the uterus, is a disease associated with pelvic painand infertility. Endometriosis afflicts about 10% of womenworldwide [1]. In women with pain, infertility or both, thefrequency increases to 35–60% [2]. It is one of the most frequentbenign gynecological diseases.

Relatively little is known about the pathophysiological mecha-nisms of endometriosis and so far endometriosis continues toremain a significantly under diagnosed and under-treated disease.

* Corresponding author at: University Hospital Kralovske Vinohrady, Department

of Laboratory Genetics, Srobarova 50, 100 34 Prague, Czech Republic.

Tel.: +420 267163578.

E-mail address: vbobek@centrum.cz (V. Bobek).

http://dx.doi.org/10.1016/j.ejogrb.2014.07.037

0301-2115/� 2014 Elsevier Ireland Ltd. All rights reserved.

The mainstay of diagnosis is still the direct visualization of thelesions by invasive procedure such as a laparoscopy [3].

One of the major challenges facing gynecologists is the inabilityto diagnose endometriosis without the need of laparoscopy orlaparotomy. At present we are not able to identify reliablebiomarkers of the disease.

Many studies have focused on identifying biomarkers in bloodor urine because of the ease of sampling. Despite extensiveresearch, no reliable blood tests currently exist for the diagnosis ofendometriosis. And a noninvasive test would be particularlywelcome because early diagnosis is important in endometriosis,and its need in both asymptomatic and symptomatic disease [4–7].

Critical information to understand the pathogenesis of endo-metriosis may come from studies of controlled in vitro models.Such cellular systems could be then used for investigation oftherapy targeted to endometriotic lesions [8–12]. However,establishment of human endometriosis derived permanent celllines were exceptionally successful. The two recently reported in

Fig. 1. Scheme of separating membrane culture system. The blue arrows appoints

the flow of the filtrated liquid (PW and PB). The PECs and CECs-cells are captured on

the membrane kept in a plastic ring and cultured directly after separation in vitro

under the standard culturing conditions. The specific diffusion flow is shown by red

arrow. Cells exhibiting high potential of plasticism can grow over the membrane to

the culture flask bottom. Finally, two cell populations can be identified after several

days of in vitro culture. The first one on growing the membrane and the second one

growing under the membrane. (For interpretation of the references to color in this

figure legend, the reader is referred to the web version of this article.)

V. Bobek et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 181 (2014) 267–274268

vitro models consist of immortalized endometrial cells. These cellsusually exhibit undifferentiated phenotype what does not repre-sent situation in vivo [13,14].

We report a new testing method for detection and culturing ofviable circulating endometrial cells. The aim of the presented study

Fig. 2. Viable endometrial gland cells (PECs) isolated from PW, grown in vitro. The presen

form a kind of cell-islets of epitheloid character (A), following they start to scallope the bo

floating clupms (D). A bar represents 10 mm.

was to identify endometrial cells in peritoneal washing (PW) and inperipheral blood (PB) in patients with a diagnosed or pre-diagnosed endometriosis. There are recently no standardlyintroduced methods to identify viable endometrial cells in PWand PB, so the challenge would be to have more of them keepingthen in in vitro culture. The cells could then further be used notonly for diagnosis support, but could also improve the treatmentstrategy in the direction of personalizing the hormonal endome-trial treatment.

Materials and methods

Patients

Seventeen patients with diagnosed endometriosis or pre-diagnosed endometriosis undergoing planned laparoscopy wereenrolled into the study. The final diagnoses were made byhistopathology. Based on the informed consent the clinical datawere collected from all participating patients. PB was collectedprior the laparoscopy. For each patient, approximately 8 mL of PBwas drawn from the antecubital veins and placed into S-Monovettetubes (Sarstedt AG & Co., Numbrecht, Germany) containing 1.6 mgEDTA/mL blood as an anticoagulant. During the laparoscopy PWwas collected. The samples were processed at room temperatureusing an isolation procedure completed within 24 hours after theblood draw. The ethics committee both participated University and

ted PECs were growing under the membrane. There they start to proliferate and to

rders of the cell-islets and to detach from the bottom (B and C) resulting finally in the

V. Bobek et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 181 (2014) 267–274 269

Hospital approved the study protocol according to the Declarationof Helsinki.

Endometriosis-related cells enrichment and in vitro culture

Size-based enrichment method (MetaCell1, Czech Republic) hasbeen used for the filtration of PB and PW through porouspolycarbonate membrane (pores with 8 mm diameter). The mini-mum and maximum volume of the filtered liquid may be adjustedup to 50 mL of fluid. Standardly, 8 mL PB from patients suffering withendometriosis can be transferred into the filtration tube, andapproximately 25 mL of PW is filtered. Based on the consistence ofPW, the PW could be before filtering centrifuged and the cell pelletreleased in PBS and subsequently transferred into the filtration tube.The PB or PW flow is supported by capillary action of the absorbenttouching the membrane filter. Afterwards the membrane filter keptin a plastic ring is transferred into the 6-well cultivation plate, 4 mLRPMI media is added to the filter top and captured cells are culturedon the membrane in vitro under the conditions of standard cellcultures (37 8C, 5% atmosphere of CO2) and observed by invertedmicroscope. The cells grow in the FBS enriched RPMI medium (10%)for the period of minimum 7–14 days on the membrane (see Fig. 1).The grown cells are analyzed by means of histochemistry (May-Grunwald staining, MGG) and immunohistochemistry (IHC). Anti-bodies against pan-cytokeratin, CD10, vimentin (DAKO, Denmark)were used for cell origin identification.

Fig. 3. Endometrial gland cells isolated from PW, grown in vitro (PECs). The presented PE

form cell-bulks (PECs-bulks) (A), sometimes floating as a group of cells (B). Following the

exhibiting also an abundance of the stromal-like PECs in the culture—see arrow (D). A

Alternatively the enriched cell fraction can be transferred fromthe membrane and cultured directly on any plastic surface or amicroscopic slide. Microscopic slide culture is preferred if the IHC/immunofluorescence analysis is planned. If an intermediate cellanalysis is awaited, the cell fraction is transferred in PBS (1.5 mL) tothe cytospin slide. The slide is then dried for 24 h and analyzed byIHC.

Cytological analysis

The fixed and stained cells on the membranes were examinedusing light microscopy in two steps: (1) screening to locate cells,(2) observation at higher magnification for detailed cytomorpho-logical analysis. Isolated cells and/or clusters of cells of interestwere selected, digitized, and examined by an experiencedresearcher and pathologist. We compared the cells enriched fromPB with the cells captured from PW, to identify the endometrialgland cell morphology and stromal cell morphology.

Results

In total, 17 patients have been enrolled into the study in theyears 2012–2013 (age mean 39.3 years). Following the aims of thestudy, we have identified viable endometrial gland and stromalcells in PW (see Figs. 2 and 3) and endometrial stromal cells in PB(see Fig. 4). This is the first time to our knowledge reporting the

Cs are growing on the separating membrane. There they start to proliferate and to

y start to adhere to the membrane creating cell-islets of an epitheloid character (C),

bar represents 10 mm.

Fig. 4. Endometrial cells isolated from PB, grown in vitro (CECs) (A) and (B) CECs in vitro, adherent to the separating membrane, shown after May-Gruenwald stain (see arrow)

(B). A bar represents 20 mm.

V. Bobek et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 181 (2014) 267–274270

endometrial cells have been isolated from PB. We call themcirculating endometrial cells (CECs), the endometrial cells founddisseminated in peritoneal washing (PW) could be assigned asperitoneal endometrial cells (PECs).

Generally, the PECs were found in the PW in all of the testedpatients using the size-based filtration method, but CECs only in23.5% (4/17) cases in PB. CECs were kept viable in the two of thefour positive cases (Fig. 4) without reaching confluence in 14 days,

Fig. 5. Endometrial cells isolated from PW, grown in vitro on the membrane in vitro (PE

staining is shown forb (A) pan-cytokeratin (B) CD10 (C) pancytokeratin (D) vimentin.

but enabling to have enough cell material for introduction ofsubsequent histochemistry testing. The origin of endometrial cells(PECs and CECs) has been proven by IHC using antibodies againstcytokeratin, vimentin and CD10 (see Fig. 5).

In the 9 of the 17 cases we have successfully cultured the PECs in

vitro on the separating membrane (Fig. 6A and B). Some of the PECsgrow through the membrane and set up culture under themembrane on the culture flask- bottom (Fig. 6C and D). Analyzing

Cs). Their endometrial origin has been proved by immunohistochemistry. Positive

A bar represents 20 mm.

V. Bobek et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 181 (2014) 267–274 271

the cytomorphology of the PECs captured on the membrane, wewere able to identify endometrial stromal cells growing on themembrane and under the membrane (Figs. 6 and 7) within the first5 days of culture. Later on much smaller endometrial glandularcells started to be visible (Fig. 8). The presence of the smallendometrial glandular cells (up to 8um) on the membrane withrelatively big pores (8 mM) could be explained most probably bythe fact, that the endometrial glandular cells are usually found inPW in an organized cell clusters (e.g. gland-like structures,honeycomb, polarized-sheets, spheroids). Some of the smallendometrial gland cells can be found later growing under themembrane as well reflecting the dynamics of endometrialglandular cells folding (Fig. 8).

The PECs cultures were introduced for the purpose of the real-time cell proliferation monitoring by means of bioimpedance-based real-time testing, but as observed in the culture, theendometrial glandular cells do not grow adherently. According toour observations, we may conclude following: After the filtration ofPW through the membrane we have identified several cell types onthe membrane, some of them were PECs. We assume that the PECs-cells are of both endometrial cell types: glandular and stromal.

After the 7 days in vitro culture, some of the PECs exhibitedheavy proliferation, which yielded in the generation of numerousPECs glandular-bulks, with a typical morphology-consisting out ofthe cells with nuclei without detectable nucleoli (Fig. 6A and B).

Fig. 6. Endometrial cells isolated from PW, grown in vitro (PECs) after May-Gruenwald s

stromal and glandular (see arrows) (A and B). Some of the cells captured on the membra

they start to proliferate and form cell-islets of epitheloid character (C). Big nuclei of endo

(D). A bar represents 20 mm.

Within the 7-days, some of the PECs, have grown through theporous membrane to the culture-plate bottom. There a new cellpropagation started, yielding into the typical proliferation-islets,which demonstrated an endometrial gland-like cell character(Fig. 2A). After reaching a specific time point (we assume this couldbe a kind of internal islet-confluency point), the cells on theborders of the islets start to scallop (and to detach from the bottom(see Fig. 2B and C).

Next, the cell population starts to behave as a suspensionculture, with floating cell-clumps with a typically scallopedborders (Fig. 2D). This is mainly true for the PECs-cells of theendometrial gland character. If the cell islet does not detach, itstarts to develop into the—PECs-bulk, similar to them seenpreviously on the membrane (Fig. 3A). But still some of therelatively big stromal cells can be found growing on the plate-bottom and their heavily proliferation can be observed (see Fig. 7).We could assume that the big endometrial stromal cells possess aspecific character of plasticity or invasiveness, which enables themto grow over the membrane pores and behave according theconditions in the medium. Typically, the stromal cells exhibit bigrounded nucleus (sometimes bigger than 30 mM) with several verybig nucleoli. The bi- or tri-nucleated stages of stromal cells arealso very typical (Fig. 8). We observe several very interestingcell differentiation points of the in vitro culture, but to studythe differentiation dynamics it will be necessary to separate

taining. The presented PECs growing on the membrane do show both phenotypes:

ne grow through the pores in the direction to the bottom of the culture flask. There

metrial stromal cells exhibit also several big nucleoli, glandular cells in the middle

Fig. 7. Endometrial cells enriched from PW (PECs), cultured in vitro for 7 days. The cells show more stroma-like character, having a nucleus with several prominent nucleoli

(see arrows). The cells building up monolayer culture, starting to form lacunaes. A bar represents 10 mm.

V. Bobek et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 181 (2014) 267–274272

endometrial stromal cells from the endometrial glandular cells infuture experiments.

Comments

Endometriosis is highly unpredictable. Some women may havea few isolated implants that never spread or grow, while in othersthe disease may spread throughout the pelvis or extra-abdomi-nopelvic localization. Establishing a correct diagnosis of endome-triosis is often problematic, because the presenting symptoms canbe non-specific and associated with a number of differentconditions [1]. Imaging methods such as transvaginal ultrasoundand magnetic resonance imaging may help to identify ovarianendometriomas or a rectovaginal endometriotic nodule, but theyhave no value in diagnosing peritoneal or generalized endometri-osis [15,16]. Consequently, it is recommended that pelvicendometriosis should be diagnosed surgically [16].

Considerable effort has been invested in searching for non-invasive methods of diagnosis. A biomarker that is simple tomeasure could help clinicians to diagnose or exclude endometri-osis; it might also allow the effects of treatment to be monitored. Ifeffective, such a marker or panel of markers could preventunnecessary diagnostic procedures and/or recognize treatmentfailure at an early stage [4–7].

A biomarker is a measurable ‘‘biologic marker’’ that correlateswith a specific outcome or state of the disease [17]. The biomarkerin peripheral blood may be not represented only by chemical orbiological substances (indirect biomarker) but also by directpresence of endometrial cells (CECs) in circulation. The occurrencein CECs in peripheral blood is clear and very specific evidence ofendometrial disease. However, CECs are probably extraordinarilyrare, with only a few CECs circulating among billions of blood cells,making their isolation and characterization a tremendous techni-cal challenge. Thus, high-efficiency and high-purity isolation of

CECs from patient blood is urgently needed to obtain accurateinformation of CECs.

The MetaCell separation system was utilized to isolate CECs fromendometriosis patient blood samples and peritoneal washingssamples, with CECs detected in 4 of 17 samples and PECs detected in100% of the diagnosed endometriosis. We demonstrated efficient,quick and user friendly CECs and PECs capture platform based on acell size. Furthermore, we demonstrated the ability to culture thecaptured cells, a critical requirement for post-isolation cellularanalysis. Although it is extremely challenging to culture the isolatedCECs from patient blood and to develop a new cell line, our systemshows the possibility to culture endometrial cells out of PW, afterthe sophisticated capture and release process, while maintainingtheir viability and proliferation capability. Therefore, our CECscapture system shows great potential for efficient CECs enrichment,isolation, and cellular/genetic analysis, leading to possible ‘‘bloodbiopsy’’ of generalized endometriosis.

It could be of interest for the next research design to comparethe cell populations growing on the membrane and growingthrough the membrane.

Most probably, the invasiveness of the endometriosis can beassigned to the endometrial stromal cells, which create a supportfor the endometrial glandular cells growth. Another point ofinterest could be to define the amount of the endometrial glandcells and stromal cells in the connection to the disease progress. So,that it could be of interest to develop more sophisticated strategiesfor in vitro PECs and CECs propagation. The CECs-cells captured onthe membrane did not show as much proliferation potential incomparison to the PECs, but it could be definitely developed infuture with a help of growing medium substitutes.

Various histological and molecular genetic studies have evenindicated that endometriosis may transform into cancer or that itcould be considered a precursor of cancer. Epidemiological studieshave shown that women with endometriosis have an increased

Fig. 8. Endometrial cells enriched from a PW (PECs), grown in vitro. The presented PECs were found to be grown under the membrane (A) cells with big nuclei are abundant,

some in bi-, or tri-nucleated stages (B and C) the cell-islets of endometrial gland cells (see arrows) do usually grow with the support of stromal cells. (B–D) A bar represents

20 mm.

V. Bobek et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 181 (2014) 267–274 273

risk of different types of malignancies, especially ovarian cancerand non-Hodgkin’s lymphoma [18,19] and dysplastic nevi, andmelanoma, and breast cancer [20–22].

Clinical studies have demonstrated that circulating tumor cells(CTCs) are correlated with disease progression for a wide range ofcancers, such as breast, colorectal, and prostate cancer andgynecological cancers too [16]. CTCs are tumor cells disseminatedfrom primary tumors which subsequently travel through the bloodcirculation to distant organs. Therefore, CTCs hold the key to trackmetastasis, and they can be used for cancer diagnosis andmonitoring of cancer status. On the above mentioned principlethe presented MetaCell technology shows great possibility todetect and test not only CTCs [23,24], but CECs as well.

It could be of interest for the next research design to comparethe cell populations growing on the membrane and growingthrough the membrane.

Most probably, the invasiveness of the endometriosis can beassigned to the endometrial stromal cells, which create a supportfor the endometrial gland cells growth. Another point of interestcould be to define the amount of the endometrial gland cells andstromal cells in PW in the connection to the disease progress. So,that it could be of interest to develop more sophisticated strategiesfor in vitro PECs and CECs propagation. The CECs-cells captured onthe membrane did not show as much proliferation potential incomparison to the PECs, but it could be definitely developed infuture with a help of growing medium substitutes.

Peripheral biomarkers show promise as diagnostic aids, butfurther research is necessary before they can be recommended inroutine clinical care. Our future efforts include culturing thecaptured CECs from patient, cellular and genetic study of isolatedCECs.

Condensation

The paper describes an isolation and cultivation of thecirculating endometrial cells from peripheral blood in patientswith endometriosis. It should be new biomarker.

Acknowledgements

This study was supported by the Research project P 27/2012awarded by Charles University in Prague, 3rd Faculty of Medicine,Prague, Czech Republic and grant of the Czech Ministry of Health:IGA NT14441-3/2013.

References

[1] Giudice L, Kao L. Endometriosis. Lancet 2004;364:1789–99.[2] ASRM. Endometriosis and infertility. Fertil Steril 2004;81:1441–6.[3] Valle RF, Sciarra JJ. Endometriosis: treatment strategies. Ann NY Acad Sci

2001;955:281–92.

V. Bobek et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 181 (2014) 267–274274

[4] May KE, Conduit-Hulbert SA, Villar J, Kirtley S, Kennedy SH, Becker CM.Peripheral biomarkers of endometriosis: a systematic review. Hum ReprodUpdate 2010;16:651–74.

[5] May KE, Villar J, Kirtley S, Kennedy SH, Becker CM. Endometrial alterations inendometriosis: a systematic review of putative biomarkers. Hum Reprod Update2011;17:637–53.

[6] Fassbender A, Vodolazkaia A, Saunders P, Lebovic D, Waelkens E, De Moor B,et al. Biomarkers of endometriosis. Fertil Steril 2013;99:1135–45.

[7] Guo SW. Recurrence of endometriosis and its control. Hum Reprod Update2009;15:441–61.

[8] Bergqvist A, Ljundberg O, Skoog L. Immunohistochemical analysis of oestrogenand progesterone receptors in endometriotic tissue and endometrium. HumReprod 1993;8:1915–22. 15.

[9] Prentice A, Randall BJ, Weddell A, McGill A, Henry L, Horne CH, et al. Ovariansteroid receptor expression in endometriosis and in vivo potential parentepithelia: endometrium and peritoneal mesothelium. Hum Reprod 1992;7:1318–25. 16.

[10] Oral E, Arici A. Peritoneal growth factors and endometriosis. Sem ReprodEndocrinol 1996;14:257–67. http://dx.doi.org/10.1055/s-2007-1016335 17.

[11] Haining RE, Schofield JP, Jones DS, Rajput-Williams J, Smith SK. Identificationof mRNA for epidermal growth factor and transforming growth factor-alphapresent in low copy number in human endometrium and deciduausing reversetranscriptase-polymerase chainreaction. J Mol Endocrinol 1991;6:207–14.http://dx.doi.org/10.1677/jme.0.0060207 18.

[12] Giudice LC. Growth factors and growth modulators in human uterine endo-metrium: their potential relevance to reproductive medicine. Fertil Steril1994;61:1–17.

[13] Bouquet de Joliniere J, Validire P, Canis M, Doussau M, Levardon M, Gogusev J.Human endometriosis-derived permanent cell line (FbEM-1): establishmentand characterization. Hum Reprod Update 1997;3:11723.

[14] Akoum A, Lavoie J, Drouin R, Jolicoeur C, Lemay A, Maheux R, et al. Physiologi-cal and cytogenetic characterization of immortalized human endometrioticcells containing episomal simian virus 40 DNA. Am J Pathol 1999;154:1245–57. http://dx.doi.org/10.1016/S0002-9440(10)65376-X.

[15] Moore J, Copley S, Morris J, Lindsell D, Golding S, Kennedy S. A systematicreview of the accuracy of ultrasound in the diagnosis of endometriosis.Ultrasound Obstet Gynecol 2002;20:630–4.

[16] Kennedy S, Bergqvist A, Chapron C, D’Hooghe T, Dunselman G, Greb R, et al.ESHRE guideline for the diagnosis and treatment of endometriosis. HumReprod 2005;20:2698–704.

[17] Kingsmore SF. Multiplexed protein measurement: technologies and applica-tions of protein and antibody arrays. Nat Rev Drug Discovery 2006;5:310–20.

[18] Brinton LA, Gridley G, Persson I, Baron J, Bergqvist A. Cancer risk after a hospitaldischarge diagnosis of endometriosis. Am J Gynecol 1997;176:72–579.

[19] Melin A, Sparen P, Bergqvist A. Endometriosis and the risk of cancer withspecial emphasis on ovarian cancer. Hum Reprod 2006;21:1237–42.

[20] Bertelsen L, Mellemkjer L, Frederiksen K, Kyer SK, Brinton LA, Sakoda LC, et al.Risk for breast cancer among women with endometriosis. Int J Cancer2007;120:1372–5.

[21] Wyshak G, Frisch RE, Albright NL, Albright TE, Schiff I. Reproductive factors andmelanoma of the skin among women. Int J Dermatol 1989;28:527–30.

[22] Hornstein MD, Thomas PP, Sober AJ, Wyshak G, Albright NL, Frisch RE.Association between endometriosis, dysplastic nevi and history of melanomain women of reproductive age. Hum Reprod 1997;12:143–5.

[23] Liberko M, Kolostova K, Bobek V. Essentials of circulating tumor cells forclinical research and practice. Crit Rev Oncol Hematol 2013;88:338–56.

[24] Kolostova K, Broul M, Schraml J, Cegan M, Matkowski R, Fiutowski M, et al.Circulating tumor cells in localized prostate cancer: isolation, cultivation in vitroand relationship to T-stage and Gleason score. Anticancer Res 2014;34(7):3641–6.