- Email: [email protected]
Ovarian tissue banking and fertility preservation in cancer patients: histological and immunohistochemical evaluation
Available online at www.sciencedirect.com R
Gynecologic Oncology 89 (2003) 259 –266
www.elsevier.com/locate/ygyno
Ovarian tissue banking and fertility preservation in cancer patients: histological and immunohistochemical evaluation R. Fabbri, BSc, Ph.D.,a,* S. Venturoli, M.D., Ph.D.,a A. D’Errico, M.D., Ph.D.,b C. Iannascoli, BSc,a E. Gabusi, BSc,b B. Valeri, M.D.,b R. Seracchioli, M.D.,a and W.F. Grigioni, M.D., Ph.D.b a
Human Reproductive Medicine Unit, University of Bologna, S. Orsola Hospital, 40138 Bologna, Italy b Patology of the “F. Addarii” Institute of Oncology, University of Bologna, Italy Received 2 July 2002
Abstract Objective. A combination of chemotherapy and radiotherapy in young females with cancer has greatly enhanced the life expectancy of these patients, even if these treatments have a highly deleterious effect on the ovary and cause a severe depletion of the follicular store. Cryopreservation of ovarian tissue before chemotherapy and/or radiotherapy, followed by autograft after remission or in in vitro maturation, could restore gonadal function and fertility. The aim of this study is to verify the efficiency of the ovarian tissue cryopreservation procedure by histological and immunohistochemical analyses. Methods. Ovarian tissue was obtained by laparoscopy from 22 patients affected with different malignant diseases. Tissue specimens were frozen using a combination of PROH (1,2-propanediol) and sucrose as cryoprotectants, and the cryopreservation protocol used consisted of a slow freezing–rapid thawing program. Both fresh and frozen/thawed tissues were embedded in paraffin blocks for histological and immunohistochemical analyses. Results. Good stromal and follicular morphology was found in fresh and frozen/thawed tissue. No significant differences were found in follicular density, distribution, and diameters in fresh and frozen/thawed tissue. Follicle immunohistochemical analysis showed a high percentage of negative staining for both estrogen receptor (ER) (100% both in fresh and frozen/thawed specimens) and progesterone receptor (PR) (97% versus 91%, respectively). Regarding the Ki67 protein, positive staining was found in both the granulosa cells and/or the oocytes (36% in fresh and 56% in frozen/thawed). For the Bcl2 protein, positive staining was observed in the follicle granulosa cells but not in the oocytes in 74% of the fresh and in 79% of the frozen/thawed specimens. For the stromal cells, ER showed a negative staining distribution in 97% of the fresh and 100% of the frozen/thawed specimens. The stroma staining distribution was diffuse/focal in fresh versus frozen/thawed specimens (50% versus 74% respectively) for PR, patch/focal (70% versus 80%, respectively) for Ki67 protein, and diffuse (55% versus 54%, respectively) for Bcl2. Conclusions. These results suggest that human ovarian tissue morphology, antigenicity, cellular proliferation, and anti-apoptotic index were well preserved by cryopreservation in PROH and sucrose. © 2003 Elsevier Science (USA). All rights reserved. Keywords: Human ovarian tissue cryopreservation; Malignant diseases; Cryoprotectants; Histological analysis; Immunohistochemical analysis
Introduction In the past 25 years, new chemotherapy and/or radiotherapy treatments have significantly improved the survival rate * Corresponding author. Human Reproductive Medicine Unit, University of Bologna, S. Orsola Hospital, Via Massarenti 13, 40138 Bologna, Italy. Fax: ⫹39-051301926. E-mail address: [email protected] (R. Fabbri).
of many young cancer patients, even if these treatments are gonadotoxic. Ovarian damage from ionizing radiation depends on the dosage while that from chemotherapy depends on the type of drug used (alkylating agents, antimetabolites, and vinca alkaloids), on the dosage, on the length of treatment, and on the age of the patient [1– 4]. The cryopreservation of ovarian tissue is an attractive strategy for conserving both steroidogenic and gametogenic
0090-8258/03/$ – see front matter © 2003 Elsevier Science (USA). All rights reserved. doi:10.1016/S0090-8258(02)00098-7
260
R. Fabbri et al. / Gynecologic Oncology 89 (2003) 259 –266
functions in patients who are at risk of losing their ovarian function from chemotherapy and/or radiotherapy treatments. Ovarian tissue collection and storage has several significant advantages over egg or embryo storage. Mature oocytes can only be collected following gonadotropin stimulation and they are never available in large numbers even if recent results have shown a high survival rate in frozen/thawed oocytes [5]. Embryo cryopreservation is only relatively efficient and raises many ethical and legal problems. On the contrary, primordial follicles, available in large numbers in ovarian tissue, are better suited to cryopreservation because they are small, they lack the zona pellucida, and they are metabolically quiescent and undifferentiated [6]. Furthermore, ovarian tissue cryopreservation is the only possibility for maintaining fertility in prepubertal girls undergoing chemotherapy or abdominal radiation as they are too young for ovarian stimulation therapy [7]. The cryopreservation of ovarian tissue, before chemo- or radiotherapy, could be indicated in fertile patients affected with inflammatory diseases and malignant diseases, systemic diseases (Hodgkin’s diseases and non-Hodgkin’s lymphomas), extrapelvic diseases (breast cancer, follicular thyroid carcinoma, hepatocellular carcinoma in a cirrhotic liver, renal carcinoma), and pelvic diseases (colon carcinoma, invasive cervix carcinoma, vaginal lymphoma, ovarian adenocarcinoma). In borderline ovarian exophytic tumors, because zones of ovarian cortex that appear normal at a macroscopic level may exist and the cortex is lost to the patient in any case, the freezing may be useful. The choice to cryopreserve cortical tissue must be made in consultation with a pathologist and, since the risk of reintroducing the cancer by autografting exists, the only alternative to be proposed to these patients is in vitro fertilization [8]. An additional application of ovarian tissue cryopreservation could be the pregnancy management of women having Turner’s syndrome in which the reimplantation or in vitro maturation of the ovarian follicles might be an option for their infertility treatment. The steroidogenic function of ovarian tissue can be restored by grafting so that the patients avoid hormone replacement therapy (HRT) [9]. Finally, ovarian tissue banking could be an opportunity for patients suffering from repeated ovarian cysts who risk losing their ovarian function following surgical treatments. After thawing, the tissue could be grafted into its normal site (orthotopic), which would allow the possibility of pregnancies without further medical assistance [10,11], or into a site other than its normal position (heterotopic), necessitating recourse to IVF to obtain pregnancies [12]. In addition, after thawing, the ovarian follicles could be grown and matured in vitro in order to recover metaphase II oocytes for an IVF program [8]. The aim of this study was to investigate follicular and stromal preservation by histological analysis, and to evaluate the maintenance of ovarian antigenicity using anti-ER
and anti-PR antibodies, the cell proliferation by anti-Ki67, and the cellular anti-apoptotic activity by anti-Bcl2, using immunohistochemical analysis in order to assess the efficiency of the cryopreservation procedure on the human ovarian tissue obtained from patients affected with different malignant diseases.
Materials and methods Ovarian tissue Ovarian tissue was obtained by laparoscopy, from 22 counseled patients between 21 and 39 years old (mean 28 ⫾ 4.4 years), affected by different malignant diseases: breast cancer (4 patients), colon cancer (1 patient), kidney cancer (1 patient), Hodgkin’s disease (9 patients), non-Hodgkin’s lymphoma (4 patients), vaginal lymphoma (1 patient), ovarian adenocarcinoma (1 patient), and ovarian exophytic stage I borderline tumor (1 patient). The study was approved by a local Human Investigations Committee of S.Orsola-Malpighi Hospital (Italy). The patients had a regular menstrual cycle (28 –30 days), except for the two affected with amenorrhea. At the time of biopsy, 15 patients were in the follicular phase and 5 in the luteal phase. After the biopsy, the ovarian tissue was placed in Dulbecco’s phosphatebuffered solution (PBS) (Gibco, Life Technologies LTD, Paisley, Scotland) added with 10% human serum (HS was provided by the Transfusion Centre of S.Orsola-Malpighi Hospital) and cut into pieces (1–2 mm in thickness and about 1–2 cm in length) using a scalpel. One or two pieces per patient were immediately fixed in formalin for subsequent histological and immunohistochemical analyses (control fresh tissue). The other specimens were maintained in PBS ⫹ 10% HS solution before cryopreservation. Freezing/thawing The cryopreservation protocol consisted of a slow freezing/rapid thawing method. The pieces were placed in plastic cryovials (Intermed Nunc Cryotubes, Denmark) containing 1.8 ml of “freezing solution” consisting of 1.5 M PROH (Fluka Chemica, Sigma Aldrich SrL; Milan, Italy) ⫹ 0.2 M sucrose ⫹ 30% HS in PBS and maintained at 0°C in an ice bath. The cryovials were transferred to a rolling system for 30 min at 4°C to allow the cryoprotectant to enter the tissue; they were then cooled in a programmable freezer (Planer Kryo 10/1,7 Series III, SAPIO Life). The starting temperature was 0°C; then it was slowly reduced to ⫺9°C at a rate of ⫺2°C/min. Ice nucleation was induced manually at ⫺9°C (seeding). After a holding time of 10 min at ⫺9°C, the cryovials were cooled slowly to ⫺40°C at a rate of ⫺0.3°C/min and then rapidly to ⫺140°C at a rate of ⫺10°C/min. After 10 min of stabilization temperature, the cryovials were transferred into liquid nitrogen tanks and stored until thawing.
R. Fabbri et al. / Gynecologic Oncology 89 (2003) 259 –266
To thaw, the vials were air-warmed for 30 s and then immersed in a 37°C water bath for 2 min. The cryoprotectants were removed at room temperature by stepwise dilution of PROH in the thawing solutions. The contents of the melted vials were expelled in 1.0 M PROH ⫹ 0.2 M sucrose ⫹ 30% HS and the sample was equilibrated for 5 min. Then the sample was transferred to 0.5 M PROH ⫹ 0.2 M sucrose ⫹ 30% HS for an additional 5 min and then in 0.2 M sucrose ⫹ 30% HS for 10 min before final dilution in PBS ⫹ 30% HS for 20 min (10 min at room temperature and 10 min at 37°C). Finally, the tissue pieces were incubated for about 30 min in a culture medium (Medicult Universal IVF medium) at 37°C in an atmosphere of 5% CO2 in air. A thawed tissue specimen from each patient was fixed in formalin for histological and immunohistochemical analyses. Histological and immunohistochemical analyses Both the control fresh and the frozen thawed ovarian tissue slices were embedded in paraffin blocks and random sections of 4 m thickness for each patient were obtained. From four to six sections were utilized for staining with hematoxylin and eosin (EE) while the others were utilized for the immunohistochemical procedure with anti-estrogen and anti-progesterone receptors and anti-Ki67 and anti-Bcl2 protein antibodies (four sections per procedure). Following this, another four to six sections were utilized for EE staining as a control. The follicles observed in the high-power field (HPF) using a Leitz microscope (⫻100) were classified according Gougeon [13] as primordial (the oocyte is surrounded by flattened GCs), intermediary (the oocyte is surrounded by a mixture of flattened and cuboidal GCs), primary (the oocyte is surrounded by a single layer of cuboidal GCs), secondary (the oocyte is surrounded by more than one layer of cuboidal GCs but the epithelioid cells have not differentiated in the theca layer), pre-antral (the oocyte is surrounded by multiple GC layers without an antrum and the theca is evident), antral (the oocyte is enlarged and is surrounded by multiple GC layers with antrum formation), and atretic (the follicle and/or the oocyte shows an irregular shape and nuclear pyknosis in the granulosa layers is evident). The follicles were measured using Cytometrica (Chieco C&V Technology—Bologna) based on the method of measurement plane and count (see “Image Cytometry,” Chieco P, Jonker A, Van Noorder CJF, BIOS, Oxford, UK, 2001) and the follicular density was expressed as the mean ⫾ SD of the follicle number per 5 mm2 of ovarian tissue. Student’s t test was used for statistical analysis. Four classes (very good, good, fairly bad, and bad) were used to evaluate follicular and stromal morphology. Stromal morphology was considered to be very good when there were no signs of interstitial edema or necrosis in any of the sections and the stromal cells showed no sign of cellular leakage; the follicular morphology was very good when the
261
GCs were regular in shape and the oocyte had no signs of degeneration. When the stromal tissue showed only a slight and limited presence of edema and sporadic GCs were detached from the basement membrane, the stromal and follicular morphology was judged to be good. The morphology was considered fairly bad when the interstitial edema was clearly observed in the sections; the follicles showed a focal exfoliation of the granulosa layers and the oocyte began to show nuclear changes. Finally, the morphology was considered to be bad when the stromal cells had a fibrous aspect, the follicles lost their normal appearance, and the oocytes showed cytoplasmic vacuolations. For immunohistochemical analysis (IHC), the sections were deparaffined, rehydrated, and treated with 3% H2O2 in methanol to inactivate endogenous peroxidase. After being washed in Tris-buffered solution (TBS), the slices were processed using a microwave antigen retrieval method in citrate buffer (10 mM, pH 6) for four cycles of 5 min each (high power 750 W). The sections were then incubated overnight at 4°C with the following primary monoclonal antibodies: anti-estrogen receptor clone 1D5 (Bio Genex, San Ramon, CA) and anti-progesterone receptor clone 1A6 (Bio Genex) diluted 1:20 and anti-proliferating antigen Ki67 (Bio Genex) and anti-protein Bcl2 (Dako, Carpinteria, CA) diluted 1:80. An En Vision monoclonal immunoenzymatic system was used for IHC detection (Dako). The reaction was developed in 3,3-diaminobenzidine (DAB, Sigma, St. Louis, MO, USA). Finally, the sections were counterstained with Mayer’s hematoxylin for 10 s, dehydrated, and mounted in Eukitt. Control procedures were undertaken simultaneously to ensure the specificity of immunoreaction. Slices without primary antibodies were used as a negative control and sections of human breast cancer were used as a positive control. Follicle and stromal positivity for the primary antibodies (anti-ER, anti-PR, anti-Ki67, and anti-Bcl2) were evaluated at 200X magnification under a Leitz microscope. Follicles with at least one stained granulosa cell were considered positive. All sections were examined on a blind basis by the same observer and confirmed by a second observer. In evaluating granulosa cells and oocytes, four classes of follicles were identified: (⫺;⫺) granulosa cells and oocytes negatively stained; (⫺;⫹) granulosa cells negatively and oocytes positively stained; (⫹;⫺) granulosa cells positively and oocytes negatively stained; and (⫹;⫹) granulosa cells and oocytes positively stained. The staining distribution score (0 –100%) in the stromal cells was evaluated according to an optical density scale identifying four classes: negative distribution (no cells positively stained), patch distribution (⬍10% positive), focal distribution (10 –50% positive), and diffuse distribution (⬎50% positive). Considering the frequency of fresh and frozen/thawed samples observed in the stromal and follicular classes, the 2 test was used for statistical analysis.
262 R. Fabbri et al. / Gynecologic Oncology 89 (2003) 259 –266
Fig. 1. Morphological appearance of human ovarian tissue before and after cryopreservation. (A) Primordial (a), intermediary (b), and pre-antral (c) follicles in fresh tissue. Original magnification ⫻10. (B) Primary (a) and early secondary (b) follicles in frozen/thawed tissue. ⫻10. Fig. 2. Immunohistochemical staining for ER (A,B), PR (C,D), Ki67 (E,F), and Bcl2 (G,H) in stromal cells and follicles of human ovarian tissues before and after cryopreservation. (A) Patch positive staining distribution for ER in fresh stromal cells (an enlargement is shown in the square). No staining in the pre-antral follicle. Original magnification ⫻4. (B) No staining for ER in frozen/thawed stromal cells and follicles. ⫻4. (C,D) Positive staining distribution in stromal cells and no staining in primary follicles for PR in fresh (C) and frozen/thawed (D) tissue. Original magnification ⫻10. (E) Granulosa cells and oocyte positive staining for Ki67 in pre-antral fresh follicles. Original magnification ⫻20. (F) Oocyte and stromal cell positive staining for Ki67 in primary frozen follicles. ⫻20. (G) Oocyte granulosa and stromal cell positive staining for Bcl2 in fresh tissue. Original magnification ⫻10. (H) Granulosa and stromal cell positive staining for Bcl2 in frozen/thawed tissue. ⫻10. (Œ) Negative staining; (n) Positive staining.
R. Fabbri et al. / Gynecologic Oncology 89 (2003) 259 –266 Table 1 Follicular distribution and diameters in fresh and frozen/thawed ovarian tissue Follicular stage
Distribution
Diameter
Fresh (%)
Fresh (m)
Frozen/ thawed (%)
Frozen/ thawed (m)
Primordial and intermediary 77 ⫾ 25 79 ⫾ 19 45.8 ⫾ 7.5 43.4 ⫾ 7.2 Primary 14 ⫾ 24 14 ⫾ 18 57.1 ⫾ 9.0 54.5 ⫾ 9.8 Secondary, pre-antral, and cavitary 5⫾9 2 ⫾ 4 116.8 ⫾ 59.8 116.9 ⫾ 95.3 Atretic 3⫾6 5 ⫾ 10 45.3 ⫾ 10.3 45.5 ⫾ 7.1 Note. Statistical evaluation was performed using Student’s t test. Results are presented as means ⫾ SD. P ⫽ NS (not significant).
Results Histological analysis in fresh and frozen/thawed tissue (Fig. 1) Good stromal and follicle morphology was found in both fresh and frozen/thawed tissue (100 and 100% in fresh; 93.8 and 90.3% in frozen/thawed, respectively), and the remaining 6.2 and 9.7% of the stroma and follicle thawed showed a fairly bad morphology (Fig. 1A and B). The follicular density was higher, but not significantly different, in fresh with respect to frozen/thawed tissue (4.8 ⫾ 5.3 and 3.6 ⫾ 4.5, P ⫽ NS). A similar follicular distribution was observed in fresh and frozen/thawed tissue (P ⫽ NS) (Table 1). A high percentage of the follicles was primordial and intermediary while the primary, secondary, pre-antral, and antral follicles were observed only occasionally. Only relatively few follicles were atretic (Table 1). The mean follicle diameter increased according to the follicular development stage (the atretic follicles showed diameters similar to those of primordial follicles) and they were similar in fresh and frozen/ thawed tissue (P ⫽ NS) (Table 1). Immunohistochemical analysis in fresh and frozen/thawed follicles Estrogen receptors (ER) showed negative staining in all fresh and frozen/thawed follicles (Fig. 2A and B; Fig. 3). Progesterone receptors (PR) revealed high negative staining (97% fresh vs 91% frozen/thawed, respectively) and only 3% of the fresh and 9% of the frozen/thawed samples showed positive staining in the granulosa cells and/or the oocytes (Fig. 2C and D; Fig. 3). On the contrary, regarding the cellular proliferation marker Ki67 (Ki67), positive staining was found in both the granulosa cells and/or the oocytes in 36% of the fresh and 56% of the frozen/thawed follicles (Fig. 2E and F; Fig. 3). Concerning the Bcl2 protein (Bcl2), positive staining
263
was observed in the granulosa cells but not in the oocytes in 74% of the fresh and in 79% of the frozen/thawed follicles. In 3% of the fresh follicles, positive staining was found in both the granulosa cells and the oocytes (Fig. 2G and H; Fig. 3). Immunohistochemical analysis in fresh and frozen/thawed stroma The stromal tissue showed a negative staining distribution for ER in 97% of fresh and 100% of the frozen/thawed tissue. Only one fresh specimen (3%) showed patch positive distribution for ER (Fig. 2A and B; Fig. 4). For PR, the stroma was diffusely/focally stained (50% fresh vs 74% frozen/thawed tissue) and patch stained (17% fresh vs 13% frozen/thawed tissue) (Fig. 2C and D; Fig. 4). For the Ki67 protein, the stromal was patch/focal positive stained (70% of fresh and 80% of the frozen/thawed tissue) (Fig. 2E and F; Fig. 4). Concerning Bcl2, the stromal positive distribution was diffuse in 55% vs 54%, and patch/focal in 40% vs 38% in fresh and frozen/thawed samples, respectively (Fig. 2G and H; Fig. 4).
Discussion The improvement of the prognosis for many young cancer patients has increased the interest in ovarian tissue cryopreservation as an alternative to oocyte cryopreservation to reverse infertility induced by chemo- and/or radiotherapy. In this study, the follicular and stromal ovarian cell morphology before and after cryopreservation was evaluated by histological analysis. Furthermore, we investigated the expression of estrogen and progesterone receptors in order to check the maintenance of ovarian antigenicity, and the presence of Ki67 and Bcl2 to verify cell proliferation and anti-apoptotic activity, respectively, after the freezing/ thawing procedure. Our data point out the success of freezing/thawing procedures as evidenced by the high percentage of a good stromal and follicular morphology in the specimens. Hovatta et al. reported similar results in a comparative study on fresh and frozen/thawed ovarian tissue using DMSO or PROH as cryoprotectants, indicating that there was no essential difference in the morphological appearance before and after freezing/thawing [14]. Gook et al., by using electron microscope analysis, showed that reasonable survival of both pregranulosa cells and oocytes can be achieved with a slow freezing procedure in conjunction with one-step dehydration in PROH solution [15]. We did not observe any statistically significant differences in the follicular density of fresh and frozen/thawed cortical slices. Follicular distribution, among individual patients and also among different sections obtained from the
264
R. Fabbri et al. / Gynecologic Oncology 89 (2003) 259 –266
same patient, showed great variability probably because the follicles, within the ovarian stroma, have a particular cluster distribution. These results agree with those obtained by Nisolle et al. [16]. The follicular stage distribution and diameters at various levels of development were similar in the fresh and frozen/thawed tissue as also reported by Nisolle et al. [16] and Lass et al. [17]. Therefore, the morphological analysis did not evidence any significant differences between fresh and frozen/thawed samples, indicating that ovarian tissue showed a good preservation rate after the cryopreservation procedure. During the menstrual cycle of a healthy human ovary, the primordial, intermediary, primary, and pre-antral follicles do not express ER or PR [18]. The expression of ER is evidenced only in antral and pre-ovulatory follicles before and during the LH surge. On the other hand, PR is expressed in pre-ovulatory follicles during the LH surge and after ovulation in luteinized granulosa cells. Our immunohistochemical analysis showed that, in most of the follicles (100% for ER and about 90% for PR), there was no specific positive staining for ER and PR expression, and no difference was found between fresh and frozen/thawed samples. The high negative staining percentage for ER and PR, in our study, might be due to the fact that a large number of the biopsies were retrieved in the follicular phase and also to the massive presence of primordial, intermediary, and primary follicles in the cortical slices. ER is negative while PR is diffusely expressed in stromal cells during the whole menstrual cycle [18]. In accordance with these authors, in our study, the stromal cells were negative for ER in 97% of fresh and 100% of frozen/thawed specimens. Only one fresh specimen (3%) showed ER “patch” positive distribution probably induced by the presence of one antral follicle. Our results regarding the PR stromal expression showed a diffuse positivity that was greater for the thawed tissue (49%) with respect to the fresh tissue (35%). At present, this difference is unexplained even if we can exclude the possibility that technical artifacts might have affected our results. The monoclonal Ki67 antibody recognizes a nuclear antigen which is expressed in all stages of the cell cycle G1, S, G2, M except G0. This antibody can be widely used to assess the proliferative status of the cell population in a variety of neoplastic conditions and nonneoplastic conditions and in normal tissue [18]. Gerdes et al. showed that there is a positive correlation between cell proliferation and their staining by the monoclonal antibody Ki67 [19]. Furthermore the Ki67 protein is a rather fleeting protein which
is sensitive to carelessly performed freezing procedures. Our immumohistochemical analysis for the proliferation index in the follicles showed a positivity in both granulosa cells and/or oocytes. A high percentage of stromal cells showed a “patch/focal” positive distribution. These results indicated that follicular and stromal cells could resume the mitotic cycle and then grow. In the frozen/thawed follicles, a greater positivity to the Ki67 antibody was found as compared to fresh tissue, even if this difference was not statistically significant. This result might be explained by considering that the tissue, after thawing, was maintained in a culture medium at 37°C and in 5% CO2 in air for about 30 min before the fixation in formalin. This time might be sufficient to enable the cells to begin a new mitotic cycle. The Bcl2 protooncogene encodes for an inner mitochondrial membrane protein which blocks programmed cell death and it is often topographically restricted to long-lived or proliferating cell zones [20]. Our analysis of the Bcl2 protein on the fresh and frozen/thawed specimens showed a high percentage of positivity in both granulosa and stromal cells. These results could indicate an inhibition of the cellular apoptotic process, indicating good tissue preservation after the freezing and thawing procedures. The lack of significant differences between fresh and frozen/thawed specimens for ER and PR expression, in both stromal and granulosa cells, showed that ovarian tissue antigenicity was maintained after the freezing/ thawing procedures. Furthermore, the presence of Ki67 and Bcl2 proteins shows that all the cellular types in the tissue might be able to resume a normal cellular cycle and subsequently grow. Nevertheless, it remains to be seen whether function is indeed completely maintained in the cryopreserved tissue. Maintenance of function has been reported after orthotopic and heterotopic transplantation of ovarian cortical strips [10 –12,21] and after long-term culture of thawed ovarian tissue [22–24]. However, unresolved questions do remain regarding restoration of natural fertility with production of normal progeny and the feasibility of prolonged restoration of ovarian function in young patients. Furthermore, the risk of neoplastic transmission cannot be ruled out, especially for certain tumor types [8,21,25]. Finally, optimization is required of the surgical and cryopreservation techniques. In conclusion, although much research still remains to be done, our results provide further encouragement regarding the eventual feasibility of using frozen tissue for transplants or maturation.
Fig. 3. Fresh and frozen/thawed follicle staining for estrogen and progesteron receptors, Ki67 and Bcl2 proteins. Value are expressed as a percentage (%). Four classes of follicles were identified: (G-;O-) granulosa cells and oocytes negatively stained; (G-;O⫹) granulosa cells negatively and oocytes positively stained; (G⫹;O⫺) granulosa cells positively and oocytes negatively stained; (G⫹;O⫹) granulosa cells and oocytes positively stained. Fig. 4. Staining distribution of ER, PR, Ki67, and Bcl2 in fresh and frozen/thawed stromal cells. Value are expressed as a percentage (%). The staining distribution score (0 –100%) in the stromal cells was evaluated according to an optical density scale identifying four classes: negative distribution (no cells positively stained), patch distribution (⬍10% positive), focal distribution (10 –50% positive), and diffuse distribution (⬎50% positive).
R. Fabbri et al. / Gynecologic Oncology 89 (2003) 259 –266
265
266
R. Fabbri et al. / Gynecologic Oncology 89 (2003) 259 –266
Acknowledgments The authors particularly thank Graziella Bracone for her full participation throughout the later phases of this study.
References [1] Posada MN, Kolp L, Garcia JE. Fertility options for female cancer patients: facts and fiction. Fertil Steril 2001;75(4):647–53. [2] Meirow D, Nugent D. The effects of radiotherapy and chemotherapy on female reproduction. Hum Reprod Update 2001;7(6):535– 43. [3] Grundy R, Gosden RG, Hewitt M, et al. Fertility preservation for children treated for cancer: scientific advances and research dilemmas. Arch Dis Child 2001;84:355–9. [4] Arnon J, Meirow D, Lewis-Roness H, et al. Genetic and teratogenic effects of cancer treatments on gametes and embryos. Hum Reprod Update 2001;7(4):394 – 403. [5] Fabbri R, Porcu E, Marsella T, et al. Human oocyte cryopreservation: new perspectives regarding oocyte survival. Hum. Reprod 2001; 16(3):411– 6. [6] Newton H, Aubard Y, Rutherford A, et al. Low temperature storage and grafting of human ovarian tissue. Hum Reprod 1996;7:1487–91. [7] Picton H, Gosden R. Cryopreservation of ovarian tissue. In: Female infertility therapy. London: Martin Dunizt Ltd.; 1997, Vol. 25, p. 275–99. [8] Aubard Y, Poirot C, Piver P, et al. Are there indications for ovarian tissue cryopreservation? Fertil Steril 2001;76(2):414 –5. [9] Hovatta O. Pregnancies in women with Turner’s syndrome. Ann Med 1999;31:106 –10. [10] Oktay K, Alp Aydin B, Karlikaya G, et al. A technique for laparoscopic transplantation of frozen-banked ovarian tissue. Fertil Steril 2001;75(6):1212–16. [11] Radford JA, Lieberman BA, Brison DR, et al. Orthotopic reimplantation of cryopreserved ovarian cortical strips after high-dose chemotherapy for Hodgkin’s lymphoma. Lancet 2001;357:1172–5. [12] Oktay K, Economos K, Kan M, et al. Endocrine function and oocyte retrieval after autologous transplantation of ovarian cortical strips to the forearm. JAMA 2001;286(12):1490 –3.
[13] Gougeon A. Dynamics of follicular growth in the human: a model from preliminary results. Hum Reprod 1986;1(2):81–7. [14] Hovatta O, Silye R, Krausz T, et al. Cryopreservation of human ovarian tissue using dimethylsulphoxide and propanediol-sucrose as cryoprotectant. Hum Reprod 1996;11(6):1268 –72. [15] Gook A, Edgar DH, Stern C. The effects of cryopreservation regimens on the morphology of human ovarian tissue. Mol Cell Endocrinol 2000;169:99 –103. [16] Nisolle M, Casanas-Roux F, Qu J, et al. Histologic and ultrastructural evaluation of fresh and frozen–thawed human ovarian xenografts in nude mice. Fertil Steril 2000;74(1):122–9. [17] Lass A, Silye R, Abrams DC, et al. Follicular density in ovarian biopsy of infertile women: a novel method to assess ovarian reserve. Hum Reprod 1997;12(5):1028 –31. [18] Iwai T, Nanbu Y, Iwai M, et al. Immunohistochemical localization of oestrogen receptors and progesterone receptors in the human ovary throughout the menstrual cycle. Virchows Archiv A Pathol Anat 1990; 417:369–75. [19] Gerdes J, Lemke H, Baisch J, et al. Cell cycle analysis of a cell proliferation associated human nuclear antigen defined by the monoclonal antibody Ki67. J Immunol 1984;133:1710 –5. [20] Hockenbery DM, Zutter M, Hickey W, et al. Bcl2 protein is topographically restricted in tissues characterized by apoptotic cell death. Proc Natl Acad Sci USA 1991;88:6961–5. [21] Oktay K, Yih M. Preliminary experience with orthotopic and heterotopic transplantation of ovarian cortical strips. The cryobiology of assisted reproduction: gametes and gonads. Seminars Reprod Med 2002;20(1):63–74. [22] Hovatta O, Silye R, Abir R, et al. Extracellular matrix improves survival of both stored and fresh human primordial and primary ovarian follicles in long-term culture. Hum Reprod 1997;12(5):1032– 6. [23] Hovatta O, Wright C, Krausz T, et al. Human primordial, primary and secondary ovarian follicles in long-term culture: effect of partial isolation. Hum Reprod 1999;14(10):2519 –24. [24] Wright CS, Hovatta O, Margara R, et al. Effects of follicle-stimulating hormone and serum substitution on the in-vitro growth of human ovarian follicles. Hum Reprod 1999;14(6):1555– 62. [25] Seymour JF. Ovarian tissue cryopreservation for cancer patients: who is appropriate? Reprod Fertil Dev 2001;13:81–9.
Gynecologic Oncology 89 (2003) 259 –266
www.elsevier.com/locate/ygyno
Ovarian tissue banking and fertility preservation in cancer patients: histological and immunohistochemical evaluation R. Fabbri, BSc, Ph.D.,a,* S. Venturoli, M.D., Ph.D.,a A. D’Errico, M.D., Ph.D.,b C. Iannascoli, BSc,a E. Gabusi, BSc,b B. Valeri, M.D.,b R. Seracchioli, M.D.,a and W.F. Grigioni, M.D., Ph.D.b a
Human Reproductive Medicine Unit, University of Bologna, S. Orsola Hospital, 40138 Bologna, Italy b Patology of the “F. Addarii” Institute of Oncology, University of Bologna, Italy Received 2 July 2002
Abstract Objective. A combination of chemotherapy and radiotherapy in young females with cancer has greatly enhanced the life expectancy of these patients, even if these treatments have a highly deleterious effect on the ovary and cause a severe depletion of the follicular store. Cryopreservation of ovarian tissue before chemotherapy and/or radiotherapy, followed by autograft after remission or in in vitro maturation, could restore gonadal function and fertility. The aim of this study is to verify the efficiency of the ovarian tissue cryopreservation procedure by histological and immunohistochemical analyses. Methods. Ovarian tissue was obtained by laparoscopy from 22 patients affected with different malignant diseases. Tissue specimens were frozen using a combination of PROH (1,2-propanediol) and sucrose as cryoprotectants, and the cryopreservation protocol used consisted of a slow freezing–rapid thawing program. Both fresh and frozen/thawed tissues were embedded in paraffin blocks for histological and immunohistochemical analyses. Results. Good stromal and follicular morphology was found in fresh and frozen/thawed tissue. No significant differences were found in follicular density, distribution, and diameters in fresh and frozen/thawed tissue. Follicle immunohistochemical analysis showed a high percentage of negative staining for both estrogen receptor (ER) (100% both in fresh and frozen/thawed specimens) and progesterone receptor (PR) (97% versus 91%, respectively). Regarding the Ki67 protein, positive staining was found in both the granulosa cells and/or the oocytes (36% in fresh and 56% in frozen/thawed). For the Bcl2 protein, positive staining was observed in the follicle granulosa cells but not in the oocytes in 74% of the fresh and in 79% of the frozen/thawed specimens. For the stromal cells, ER showed a negative staining distribution in 97% of the fresh and 100% of the frozen/thawed specimens. The stroma staining distribution was diffuse/focal in fresh versus frozen/thawed specimens (50% versus 74% respectively) for PR, patch/focal (70% versus 80%, respectively) for Ki67 protein, and diffuse (55% versus 54%, respectively) for Bcl2. Conclusions. These results suggest that human ovarian tissue morphology, antigenicity, cellular proliferation, and anti-apoptotic index were well preserved by cryopreservation in PROH and sucrose. © 2003 Elsevier Science (USA). All rights reserved. Keywords: Human ovarian tissue cryopreservation; Malignant diseases; Cryoprotectants; Histological analysis; Immunohistochemical analysis
Introduction In the past 25 years, new chemotherapy and/or radiotherapy treatments have significantly improved the survival rate * Corresponding author. Human Reproductive Medicine Unit, University of Bologna, S. Orsola Hospital, Via Massarenti 13, 40138 Bologna, Italy. Fax: ⫹39-051301926. E-mail address: [email protected] (R. Fabbri).
of many young cancer patients, even if these treatments are gonadotoxic. Ovarian damage from ionizing radiation depends on the dosage while that from chemotherapy depends on the type of drug used (alkylating agents, antimetabolites, and vinca alkaloids), on the dosage, on the length of treatment, and on the age of the patient [1– 4]. The cryopreservation of ovarian tissue is an attractive strategy for conserving both steroidogenic and gametogenic
0090-8258/03/$ – see front matter © 2003 Elsevier Science (USA). All rights reserved. doi:10.1016/S0090-8258(02)00098-7
260
R. Fabbri et al. / Gynecologic Oncology 89 (2003) 259 –266
functions in patients who are at risk of losing their ovarian function from chemotherapy and/or radiotherapy treatments. Ovarian tissue collection and storage has several significant advantages over egg or embryo storage. Mature oocytes can only be collected following gonadotropin stimulation and they are never available in large numbers even if recent results have shown a high survival rate in frozen/thawed oocytes [5]. Embryo cryopreservation is only relatively efficient and raises many ethical and legal problems. On the contrary, primordial follicles, available in large numbers in ovarian tissue, are better suited to cryopreservation because they are small, they lack the zona pellucida, and they are metabolically quiescent and undifferentiated [6]. Furthermore, ovarian tissue cryopreservation is the only possibility for maintaining fertility in prepubertal girls undergoing chemotherapy or abdominal radiation as they are too young for ovarian stimulation therapy [7]. The cryopreservation of ovarian tissue, before chemo- or radiotherapy, could be indicated in fertile patients affected with inflammatory diseases and malignant diseases, systemic diseases (Hodgkin’s diseases and non-Hodgkin’s lymphomas), extrapelvic diseases (breast cancer, follicular thyroid carcinoma, hepatocellular carcinoma in a cirrhotic liver, renal carcinoma), and pelvic diseases (colon carcinoma, invasive cervix carcinoma, vaginal lymphoma, ovarian adenocarcinoma). In borderline ovarian exophytic tumors, because zones of ovarian cortex that appear normal at a macroscopic level may exist and the cortex is lost to the patient in any case, the freezing may be useful. The choice to cryopreserve cortical tissue must be made in consultation with a pathologist and, since the risk of reintroducing the cancer by autografting exists, the only alternative to be proposed to these patients is in vitro fertilization [8]. An additional application of ovarian tissue cryopreservation could be the pregnancy management of women having Turner’s syndrome in which the reimplantation or in vitro maturation of the ovarian follicles might be an option for their infertility treatment. The steroidogenic function of ovarian tissue can be restored by grafting so that the patients avoid hormone replacement therapy (HRT) [9]. Finally, ovarian tissue banking could be an opportunity for patients suffering from repeated ovarian cysts who risk losing their ovarian function following surgical treatments. After thawing, the tissue could be grafted into its normal site (orthotopic), which would allow the possibility of pregnancies without further medical assistance [10,11], or into a site other than its normal position (heterotopic), necessitating recourse to IVF to obtain pregnancies [12]. In addition, after thawing, the ovarian follicles could be grown and matured in vitro in order to recover metaphase II oocytes for an IVF program [8]. The aim of this study was to investigate follicular and stromal preservation by histological analysis, and to evaluate the maintenance of ovarian antigenicity using anti-ER
and anti-PR antibodies, the cell proliferation by anti-Ki67, and the cellular anti-apoptotic activity by anti-Bcl2, using immunohistochemical analysis in order to assess the efficiency of the cryopreservation procedure on the human ovarian tissue obtained from patients affected with different malignant diseases.
Materials and methods Ovarian tissue Ovarian tissue was obtained by laparoscopy, from 22 counseled patients between 21 and 39 years old (mean 28 ⫾ 4.4 years), affected by different malignant diseases: breast cancer (4 patients), colon cancer (1 patient), kidney cancer (1 patient), Hodgkin’s disease (9 patients), non-Hodgkin’s lymphoma (4 patients), vaginal lymphoma (1 patient), ovarian adenocarcinoma (1 patient), and ovarian exophytic stage I borderline tumor (1 patient). The study was approved by a local Human Investigations Committee of S.Orsola-Malpighi Hospital (Italy). The patients had a regular menstrual cycle (28 –30 days), except for the two affected with amenorrhea. At the time of biopsy, 15 patients were in the follicular phase and 5 in the luteal phase. After the biopsy, the ovarian tissue was placed in Dulbecco’s phosphatebuffered solution (PBS) (Gibco, Life Technologies LTD, Paisley, Scotland) added with 10% human serum (HS was provided by the Transfusion Centre of S.Orsola-Malpighi Hospital) and cut into pieces (1–2 mm in thickness and about 1–2 cm in length) using a scalpel. One or two pieces per patient were immediately fixed in formalin for subsequent histological and immunohistochemical analyses (control fresh tissue). The other specimens were maintained in PBS ⫹ 10% HS solution before cryopreservation. Freezing/thawing The cryopreservation protocol consisted of a slow freezing/rapid thawing method. The pieces were placed in plastic cryovials (Intermed Nunc Cryotubes, Denmark) containing 1.8 ml of “freezing solution” consisting of 1.5 M PROH (Fluka Chemica, Sigma Aldrich SrL; Milan, Italy) ⫹ 0.2 M sucrose ⫹ 30% HS in PBS and maintained at 0°C in an ice bath. The cryovials were transferred to a rolling system for 30 min at 4°C to allow the cryoprotectant to enter the tissue; they were then cooled in a programmable freezer (Planer Kryo 10/1,7 Series III, SAPIO Life). The starting temperature was 0°C; then it was slowly reduced to ⫺9°C at a rate of ⫺2°C/min. Ice nucleation was induced manually at ⫺9°C (seeding). After a holding time of 10 min at ⫺9°C, the cryovials were cooled slowly to ⫺40°C at a rate of ⫺0.3°C/min and then rapidly to ⫺140°C at a rate of ⫺10°C/min. After 10 min of stabilization temperature, the cryovials were transferred into liquid nitrogen tanks and stored until thawing.
R. Fabbri et al. / Gynecologic Oncology 89 (2003) 259 –266
To thaw, the vials were air-warmed for 30 s and then immersed in a 37°C water bath for 2 min. The cryoprotectants were removed at room temperature by stepwise dilution of PROH in the thawing solutions. The contents of the melted vials were expelled in 1.0 M PROH ⫹ 0.2 M sucrose ⫹ 30% HS and the sample was equilibrated for 5 min. Then the sample was transferred to 0.5 M PROH ⫹ 0.2 M sucrose ⫹ 30% HS for an additional 5 min and then in 0.2 M sucrose ⫹ 30% HS for 10 min before final dilution in PBS ⫹ 30% HS for 20 min (10 min at room temperature and 10 min at 37°C). Finally, the tissue pieces were incubated for about 30 min in a culture medium (Medicult Universal IVF medium) at 37°C in an atmosphere of 5% CO2 in air. A thawed tissue specimen from each patient was fixed in formalin for histological and immunohistochemical analyses. Histological and immunohistochemical analyses Both the control fresh and the frozen thawed ovarian tissue slices were embedded in paraffin blocks and random sections of 4 m thickness for each patient were obtained. From four to six sections were utilized for staining with hematoxylin and eosin (EE) while the others were utilized for the immunohistochemical procedure with anti-estrogen and anti-progesterone receptors and anti-Ki67 and anti-Bcl2 protein antibodies (four sections per procedure). Following this, another four to six sections were utilized for EE staining as a control. The follicles observed in the high-power field (HPF) using a Leitz microscope (⫻100) were classified according Gougeon [13] as primordial (the oocyte is surrounded by flattened GCs), intermediary (the oocyte is surrounded by a mixture of flattened and cuboidal GCs), primary (the oocyte is surrounded by a single layer of cuboidal GCs), secondary (the oocyte is surrounded by more than one layer of cuboidal GCs but the epithelioid cells have not differentiated in the theca layer), pre-antral (the oocyte is surrounded by multiple GC layers without an antrum and the theca is evident), antral (the oocyte is enlarged and is surrounded by multiple GC layers with antrum formation), and atretic (the follicle and/or the oocyte shows an irregular shape and nuclear pyknosis in the granulosa layers is evident). The follicles were measured using Cytometrica (Chieco C&V Technology—Bologna) based on the method of measurement plane and count (see “Image Cytometry,” Chieco P, Jonker A, Van Noorder CJF, BIOS, Oxford, UK, 2001) and the follicular density was expressed as the mean ⫾ SD of the follicle number per 5 mm2 of ovarian tissue. Student’s t test was used for statistical analysis. Four classes (very good, good, fairly bad, and bad) were used to evaluate follicular and stromal morphology. Stromal morphology was considered to be very good when there were no signs of interstitial edema or necrosis in any of the sections and the stromal cells showed no sign of cellular leakage; the follicular morphology was very good when the
261
GCs were regular in shape and the oocyte had no signs of degeneration. When the stromal tissue showed only a slight and limited presence of edema and sporadic GCs were detached from the basement membrane, the stromal and follicular morphology was judged to be good. The morphology was considered fairly bad when the interstitial edema was clearly observed in the sections; the follicles showed a focal exfoliation of the granulosa layers and the oocyte began to show nuclear changes. Finally, the morphology was considered to be bad when the stromal cells had a fibrous aspect, the follicles lost their normal appearance, and the oocytes showed cytoplasmic vacuolations. For immunohistochemical analysis (IHC), the sections were deparaffined, rehydrated, and treated with 3% H2O2 in methanol to inactivate endogenous peroxidase. After being washed in Tris-buffered solution (TBS), the slices were processed using a microwave antigen retrieval method in citrate buffer (10 mM, pH 6) for four cycles of 5 min each (high power 750 W). The sections were then incubated overnight at 4°C with the following primary monoclonal antibodies: anti-estrogen receptor clone 1D5 (Bio Genex, San Ramon, CA) and anti-progesterone receptor clone 1A6 (Bio Genex) diluted 1:20 and anti-proliferating antigen Ki67 (Bio Genex) and anti-protein Bcl2 (Dako, Carpinteria, CA) diluted 1:80. An En Vision monoclonal immunoenzymatic system was used for IHC detection (Dako). The reaction was developed in 3,3-diaminobenzidine (DAB, Sigma, St. Louis, MO, USA). Finally, the sections were counterstained with Mayer’s hematoxylin for 10 s, dehydrated, and mounted in Eukitt. Control procedures were undertaken simultaneously to ensure the specificity of immunoreaction. Slices without primary antibodies were used as a negative control and sections of human breast cancer were used as a positive control. Follicle and stromal positivity for the primary antibodies (anti-ER, anti-PR, anti-Ki67, and anti-Bcl2) were evaluated at 200X magnification under a Leitz microscope. Follicles with at least one stained granulosa cell were considered positive. All sections were examined on a blind basis by the same observer and confirmed by a second observer. In evaluating granulosa cells and oocytes, four classes of follicles were identified: (⫺;⫺) granulosa cells and oocytes negatively stained; (⫺;⫹) granulosa cells negatively and oocytes positively stained; (⫹;⫺) granulosa cells positively and oocytes negatively stained; and (⫹;⫹) granulosa cells and oocytes positively stained. The staining distribution score (0 –100%) in the stromal cells was evaluated according to an optical density scale identifying four classes: negative distribution (no cells positively stained), patch distribution (⬍10% positive), focal distribution (10 –50% positive), and diffuse distribution (⬎50% positive). Considering the frequency of fresh and frozen/thawed samples observed in the stromal and follicular classes, the 2 test was used for statistical analysis.
262 R. Fabbri et al. / Gynecologic Oncology 89 (2003) 259 –266
Fig. 1. Morphological appearance of human ovarian tissue before and after cryopreservation. (A) Primordial (a), intermediary (b), and pre-antral (c) follicles in fresh tissue. Original magnification ⫻10. (B) Primary (a) and early secondary (b) follicles in frozen/thawed tissue. ⫻10. Fig. 2. Immunohistochemical staining for ER (A,B), PR (C,D), Ki67 (E,F), and Bcl2 (G,H) in stromal cells and follicles of human ovarian tissues before and after cryopreservation. (A) Patch positive staining distribution for ER in fresh stromal cells (an enlargement is shown in the square). No staining in the pre-antral follicle. Original magnification ⫻4. (B) No staining for ER in frozen/thawed stromal cells and follicles. ⫻4. (C,D) Positive staining distribution in stromal cells and no staining in primary follicles for PR in fresh (C) and frozen/thawed (D) tissue. Original magnification ⫻10. (E) Granulosa cells and oocyte positive staining for Ki67 in pre-antral fresh follicles. Original magnification ⫻20. (F) Oocyte and stromal cell positive staining for Ki67 in primary frozen follicles. ⫻20. (G) Oocyte granulosa and stromal cell positive staining for Bcl2 in fresh tissue. Original magnification ⫻10. (H) Granulosa and stromal cell positive staining for Bcl2 in frozen/thawed tissue. ⫻10. (Œ) Negative staining; (n) Positive staining.
R. Fabbri et al. / Gynecologic Oncology 89 (2003) 259 –266 Table 1 Follicular distribution and diameters in fresh and frozen/thawed ovarian tissue Follicular stage
Distribution
Diameter
Fresh (%)
Fresh (m)
Frozen/ thawed (%)
Frozen/ thawed (m)
Primordial and intermediary 77 ⫾ 25 79 ⫾ 19 45.8 ⫾ 7.5 43.4 ⫾ 7.2 Primary 14 ⫾ 24 14 ⫾ 18 57.1 ⫾ 9.0 54.5 ⫾ 9.8 Secondary, pre-antral, and cavitary 5⫾9 2 ⫾ 4 116.8 ⫾ 59.8 116.9 ⫾ 95.3 Atretic 3⫾6 5 ⫾ 10 45.3 ⫾ 10.3 45.5 ⫾ 7.1 Note. Statistical evaluation was performed using Student’s t test. Results are presented as means ⫾ SD. P ⫽ NS (not significant).
Results Histological analysis in fresh and frozen/thawed tissue (Fig. 1) Good stromal and follicle morphology was found in both fresh and frozen/thawed tissue (100 and 100% in fresh; 93.8 and 90.3% in frozen/thawed, respectively), and the remaining 6.2 and 9.7% of the stroma and follicle thawed showed a fairly bad morphology (Fig. 1A and B). The follicular density was higher, but not significantly different, in fresh with respect to frozen/thawed tissue (4.8 ⫾ 5.3 and 3.6 ⫾ 4.5, P ⫽ NS). A similar follicular distribution was observed in fresh and frozen/thawed tissue (P ⫽ NS) (Table 1). A high percentage of the follicles was primordial and intermediary while the primary, secondary, pre-antral, and antral follicles were observed only occasionally. Only relatively few follicles were atretic (Table 1). The mean follicle diameter increased according to the follicular development stage (the atretic follicles showed diameters similar to those of primordial follicles) and they were similar in fresh and frozen/ thawed tissue (P ⫽ NS) (Table 1). Immunohistochemical analysis in fresh and frozen/thawed follicles Estrogen receptors (ER) showed negative staining in all fresh and frozen/thawed follicles (Fig. 2A and B; Fig. 3). Progesterone receptors (PR) revealed high negative staining (97% fresh vs 91% frozen/thawed, respectively) and only 3% of the fresh and 9% of the frozen/thawed samples showed positive staining in the granulosa cells and/or the oocytes (Fig. 2C and D; Fig. 3). On the contrary, regarding the cellular proliferation marker Ki67 (Ki67), positive staining was found in both the granulosa cells and/or the oocytes in 36% of the fresh and 56% of the frozen/thawed follicles (Fig. 2E and F; Fig. 3). Concerning the Bcl2 protein (Bcl2), positive staining
263
was observed in the granulosa cells but not in the oocytes in 74% of the fresh and in 79% of the frozen/thawed follicles. In 3% of the fresh follicles, positive staining was found in both the granulosa cells and the oocytes (Fig. 2G and H; Fig. 3). Immunohistochemical analysis in fresh and frozen/thawed stroma The stromal tissue showed a negative staining distribution for ER in 97% of fresh and 100% of the frozen/thawed tissue. Only one fresh specimen (3%) showed patch positive distribution for ER (Fig. 2A and B; Fig. 4). For PR, the stroma was diffusely/focally stained (50% fresh vs 74% frozen/thawed tissue) and patch stained (17% fresh vs 13% frozen/thawed tissue) (Fig. 2C and D; Fig. 4). For the Ki67 protein, the stromal was patch/focal positive stained (70% of fresh and 80% of the frozen/thawed tissue) (Fig. 2E and F; Fig. 4). Concerning Bcl2, the stromal positive distribution was diffuse in 55% vs 54%, and patch/focal in 40% vs 38% in fresh and frozen/thawed samples, respectively (Fig. 2G and H; Fig. 4).
Discussion The improvement of the prognosis for many young cancer patients has increased the interest in ovarian tissue cryopreservation as an alternative to oocyte cryopreservation to reverse infertility induced by chemo- and/or radiotherapy. In this study, the follicular and stromal ovarian cell morphology before and after cryopreservation was evaluated by histological analysis. Furthermore, we investigated the expression of estrogen and progesterone receptors in order to check the maintenance of ovarian antigenicity, and the presence of Ki67 and Bcl2 to verify cell proliferation and anti-apoptotic activity, respectively, after the freezing/ thawing procedure. Our data point out the success of freezing/thawing procedures as evidenced by the high percentage of a good stromal and follicular morphology in the specimens. Hovatta et al. reported similar results in a comparative study on fresh and frozen/thawed ovarian tissue using DMSO or PROH as cryoprotectants, indicating that there was no essential difference in the morphological appearance before and after freezing/thawing [14]. Gook et al., by using electron microscope analysis, showed that reasonable survival of both pregranulosa cells and oocytes can be achieved with a slow freezing procedure in conjunction with one-step dehydration in PROH solution [15]. We did not observe any statistically significant differences in the follicular density of fresh and frozen/thawed cortical slices. Follicular distribution, among individual patients and also among different sections obtained from the
264
R. Fabbri et al. / Gynecologic Oncology 89 (2003) 259 –266
same patient, showed great variability probably because the follicles, within the ovarian stroma, have a particular cluster distribution. These results agree with those obtained by Nisolle et al. [16]. The follicular stage distribution and diameters at various levels of development were similar in the fresh and frozen/thawed tissue as also reported by Nisolle et al. [16] and Lass et al. [17]. Therefore, the morphological analysis did not evidence any significant differences between fresh and frozen/thawed samples, indicating that ovarian tissue showed a good preservation rate after the cryopreservation procedure. During the menstrual cycle of a healthy human ovary, the primordial, intermediary, primary, and pre-antral follicles do not express ER or PR [18]. The expression of ER is evidenced only in antral and pre-ovulatory follicles before and during the LH surge. On the other hand, PR is expressed in pre-ovulatory follicles during the LH surge and after ovulation in luteinized granulosa cells. Our immunohistochemical analysis showed that, in most of the follicles (100% for ER and about 90% for PR), there was no specific positive staining for ER and PR expression, and no difference was found between fresh and frozen/thawed samples. The high negative staining percentage for ER and PR, in our study, might be due to the fact that a large number of the biopsies were retrieved in the follicular phase and also to the massive presence of primordial, intermediary, and primary follicles in the cortical slices. ER is negative while PR is diffusely expressed in stromal cells during the whole menstrual cycle [18]. In accordance with these authors, in our study, the stromal cells were negative for ER in 97% of fresh and 100% of frozen/thawed specimens. Only one fresh specimen (3%) showed ER “patch” positive distribution probably induced by the presence of one antral follicle. Our results regarding the PR stromal expression showed a diffuse positivity that was greater for the thawed tissue (49%) with respect to the fresh tissue (35%). At present, this difference is unexplained even if we can exclude the possibility that technical artifacts might have affected our results. The monoclonal Ki67 antibody recognizes a nuclear antigen which is expressed in all stages of the cell cycle G1, S, G2, M except G0. This antibody can be widely used to assess the proliferative status of the cell population in a variety of neoplastic conditions and nonneoplastic conditions and in normal tissue [18]. Gerdes et al. showed that there is a positive correlation between cell proliferation and their staining by the monoclonal antibody Ki67 [19]. Furthermore the Ki67 protein is a rather fleeting protein which
is sensitive to carelessly performed freezing procedures. Our immumohistochemical analysis for the proliferation index in the follicles showed a positivity in both granulosa cells and/or oocytes. A high percentage of stromal cells showed a “patch/focal” positive distribution. These results indicated that follicular and stromal cells could resume the mitotic cycle and then grow. In the frozen/thawed follicles, a greater positivity to the Ki67 antibody was found as compared to fresh tissue, even if this difference was not statistically significant. This result might be explained by considering that the tissue, after thawing, was maintained in a culture medium at 37°C and in 5% CO2 in air for about 30 min before the fixation in formalin. This time might be sufficient to enable the cells to begin a new mitotic cycle. The Bcl2 protooncogene encodes for an inner mitochondrial membrane protein which blocks programmed cell death and it is often topographically restricted to long-lived or proliferating cell zones [20]. Our analysis of the Bcl2 protein on the fresh and frozen/thawed specimens showed a high percentage of positivity in both granulosa and stromal cells. These results could indicate an inhibition of the cellular apoptotic process, indicating good tissue preservation after the freezing and thawing procedures. The lack of significant differences between fresh and frozen/thawed specimens for ER and PR expression, in both stromal and granulosa cells, showed that ovarian tissue antigenicity was maintained after the freezing/ thawing procedures. Furthermore, the presence of Ki67 and Bcl2 proteins shows that all the cellular types in the tissue might be able to resume a normal cellular cycle and subsequently grow. Nevertheless, it remains to be seen whether function is indeed completely maintained in the cryopreserved tissue. Maintenance of function has been reported after orthotopic and heterotopic transplantation of ovarian cortical strips [10 –12,21] and after long-term culture of thawed ovarian tissue [22–24]. However, unresolved questions do remain regarding restoration of natural fertility with production of normal progeny and the feasibility of prolonged restoration of ovarian function in young patients. Furthermore, the risk of neoplastic transmission cannot be ruled out, especially for certain tumor types [8,21,25]. Finally, optimization is required of the surgical and cryopreservation techniques. In conclusion, although much research still remains to be done, our results provide further encouragement regarding the eventual feasibility of using frozen tissue for transplants or maturation.
Fig. 3. Fresh and frozen/thawed follicle staining for estrogen and progesteron receptors, Ki67 and Bcl2 proteins. Value are expressed as a percentage (%). Four classes of follicles were identified: (G-;O-) granulosa cells and oocytes negatively stained; (G-;O⫹) granulosa cells negatively and oocytes positively stained; (G⫹;O⫺) granulosa cells positively and oocytes negatively stained; (G⫹;O⫹) granulosa cells and oocytes positively stained. Fig. 4. Staining distribution of ER, PR, Ki67, and Bcl2 in fresh and frozen/thawed stromal cells. Value are expressed as a percentage (%). The staining distribution score (0 –100%) in the stromal cells was evaluated according to an optical density scale identifying four classes: negative distribution (no cells positively stained), patch distribution (⬍10% positive), focal distribution (10 –50% positive), and diffuse distribution (⬎50% positive).
R. Fabbri et al. / Gynecologic Oncology 89 (2003) 259 –266
265
266
R. Fabbri et al. / Gynecologic Oncology 89 (2003) 259 –266
Acknowledgments The authors particularly thank Graziella Bracone for her full participation throughout the later phases of this study.
References [1] Posada MN, Kolp L, Garcia JE. Fertility options for female cancer patients: facts and fiction. Fertil Steril 2001;75(4):647–53. [2] Meirow D, Nugent D. The effects of radiotherapy and chemotherapy on female reproduction. Hum Reprod Update 2001;7(6):535– 43. [3] Grundy R, Gosden RG, Hewitt M, et al. Fertility preservation for children treated for cancer: scientific advances and research dilemmas. Arch Dis Child 2001;84:355–9. [4] Arnon J, Meirow D, Lewis-Roness H, et al. Genetic and teratogenic effects of cancer treatments on gametes and embryos. Hum Reprod Update 2001;7(4):394 – 403. [5] Fabbri R, Porcu E, Marsella T, et al. Human oocyte cryopreservation: new perspectives regarding oocyte survival. Hum. Reprod 2001; 16(3):411– 6. [6] Newton H, Aubard Y, Rutherford A, et al. Low temperature storage and grafting of human ovarian tissue. Hum Reprod 1996;7:1487–91. [7] Picton H, Gosden R. Cryopreservation of ovarian tissue. In: Female infertility therapy. London: Martin Dunizt Ltd.; 1997, Vol. 25, p. 275–99. [8] Aubard Y, Poirot C, Piver P, et al. Are there indications for ovarian tissue cryopreservation? Fertil Steril 2001;76(2):414 –5. [9] Hovatta O. Pregnancies in women with Turner’s syndrome. Ann Med 1999;31:106 –10. [10] Oktay K, Alp Aydin B, Karlikaya G, et al. A technique for laparoscopic transplantation of frozen-banked ovarian tissue. Fertil Steril 2001;75(6):1212–16. [11] Radford JA, Lieberman BA, Brison DR, et al. Orthotopic reimplantation of cryopreserved ovarian cortical strips after high-dose chemotherapy for Hodgkin’s lymphoma. Lancet 2001;357:1172–5. [12] Oktay K, Economos K, Kan M, et al. Endocrine function and oocyte retrieval after autologous transplantation of ovarian cortical strips to the forearm. JAMA 2001;286(12):1490 –3.
[13] Gougeon A. Dynamics of follicular growth in the human: a model from preliminary results. Hum Reprod 1986;1(2):81–7. [14] Hovatta O, Silye R, Krausz T, et al. Cryopreservation of human ovarian tissue using dimethylsulphoxide and propanediol-sucrose as cryoprotectant. Hum Reprod 1996;11(6):1268 –72. [15] Gook A, Edgar DH, Stern C. The effects of cryopreservation regimens on the morphology of human ovarian tissue. Mol Cell Endocrinol 2000;169:99 –103. [16] Nisolle M, Casanas-Roux F, Qu J, et al. Histologic and ultrastructural evaluation of fresh and frozen–thawed human ovarian xenografts in nude mice. Fertil Steril 2000;74(1):122–9. [17] Lass A, Silye R, Abrams DC, et al. Follicular density in ovarian biopsy of infertile women: a novel method to assess ovarian reserve. Hum Reprod 1997;12(5):1028 –31. [18] Iwai T, Nanbu Y, Iwai M, et al. Immunohistochemical localization of oestrogen receptors and progesterone receptors in the human ovary throughout the menstrual cycle. Virchows Archiv A Pathol Anat 1990; 417:369–75. [19] Gerdes J, Lemke H, Baisch J, et al. Cell cycle analysis of a cell proliferation associated human nuclear antigen defined by the monoclonal antibody Ki67. J Immunol 1984;133:1710 –5. [20] Hockenbery DM, Zutter M, Hickey W, et al. Bcl2 protein is topographically restricted in tissues characterized by apoptotic cell death. Proc Natl Acad Sci USA 1991;88:6961–5. [21] Oktay K, Yih M. Preliminary experience with orthotopic and heterotopic transplantation of ovarian cortical strips. The cryobiology of assisted reproduction: gametes and gonads. Seminars Reprod Med 2002;20(1):63–74. [22] Hovatta O, Silye R, Abir R, et al. Extracellular matrix improves survival of both stored and fresh human primordial and primary ovarian follicles in long-term culture. Hum Reprod 1997;12(5):1032– 6. [23] Hovatta O, Wright C, Krausz T, et al. Human primordial, primary and secondary ovarian follicles in long-term culture: effect of partial isolation. Hum Reprod 1999;14(10):2519 –24. [24] Wright CS, Hovatta O, Margara R, et al. Effects of follicle-stimulating hormone and serum substitution on the in-vitro growth of human ovarian follicles. Hum Reprod 1999;14(6):1555– 62. [25] Seymour JF. Ovarian tissue cryopreservation for cancer patients: who is appropriate? Reprod Fertil Dev 2001;13:81–9.