Successful live births after transplantation of frozen-thawed human ovarian tissue have been observed since the first report in 2004 . Up to 2017, approximately 130 live births have been achieved . Currently, this procedure is performed for children, adolescents, and young adults with cancer in many countries.
The standard method of ovarian tissue cryopreservation is slow freezing, but rapid freezing (also called “vitrification”) has increasingly been reported as an alternate cryopreservation method in recent years. The present article reviews recent findings with regard to techniques for vitrification of ovarian tissue.
According to the most recent cancer statistics, more than 870,000 new diagnoses of cancer are expected in the US female population in 2018, with the three most common cancers in women being breast, lung, and colorectal cancers .
Several improvements have been made in the early diagnosis and treatment of infant and adult cancer and these advances have resulted in greatly increased life expectancy and chances of survival. Nevertheless, some oncological treatments, although leading to cancer cure rates higher than 90%, have a detrimental effect in the reproductive potential of children and young women, resulting in a population at high-risk of developing premature ovarian insufficiency (POI) and therefore infertility .
In order to prevent the risk of facing this outcome, fertility preservation options are offered to these patients in order to protect their fertility potential prior to gonadotoxic treatment. Among the available options, ovarian tissue cryopreservation and transplantation is the only method suitable for prepubertal girls and adult women who require urgent treatment.
Indications for ovarian tissue cryopreservation are many and range from malignant to benign pathologies, as summarized in Table 26.1 . Chemotherapy regimens based on alkylating agents, as cyclophosphamide and busulfan, and pelvic radiotherapy are likely to cause ovarian damage and ultimately ovarian dysfunction and premature ovarian failure (POF).
Moreover, the probability of ovarian failure development after gonadotoxic treatment is dependent on the ovarian reserve, which is highly variable among women and anyway tends to decrease with age .
|Malignant diseases requiring gonadotoxic chemotherapy, radiotherapy, or bone marrow transplantation|
|Hematologic diseases (leukemia, Hodgkin’s lymphoma, non-Hodgkin’s lymphoma)|
|Some pelvic cancers|
|Systemic diseases requiring chemotherapy, radiotherapy, or bone marrow transplantation|
|Bilateral benign ovarian tumors|
|Severe and recurrent ovarian endometriosis|
|Possible ovarian torsion|
|Risk of premature ovarian insufficiency|
|Childbearing postponed until later in life|
From  [N Engl J Med, Donnez, J, Dolmans MM. Fertility preservation in women. 377(17):1657–1665. Copyright © (2017) Massachusetts Medical Society. Reprinted with permission.
Pelvic surgery has also an impact on future fertility, and it is the case of bilateral ovarian tumors, ovarian endometriosis, and ovarian torsion. Repetitive surgical interventions on the ovaries, as cystectomy for endometriomas or recurrent borderline ovarian tumors (BOT), may damage severely the healthy ovarian tissue surrounding the lesions, reducing the ovarian reserve [5–8]. For this reason, fertility preservation should be offered in case of recurrence after initial surgery [9, 10]. Although malignant diseases as hematological cancers and breast cancer are the most frequent indications for the procedure, ovarian tissue cryopreservation should also be proposed in the counselling for some benign pathologies, such as autoimmune and hematological conditions that may require bone marrow transplantation .
After transplantation of frozen-thawed ovarian tissue in the pelvic cavity when the patient is free of the disease, ovarian function is restored in more than 95% of cases and maintained for 4–5 years, according to the follicle reserve prior to cryopreservation [3, 11].
The first livebirth after ovarian tissue transplantation of frozen-thawed ovarian tissue was obtained in 2004 in a patient with stage IV Hodgkin’s lymphoma. Since then, we record more than 130 live births . In our series of 22 women who underwent ovarian tissue reimplantation, pregnancy and live-birth rates were, respectively, 41% (9 of 22) and 36% (8 of 22), with a total of 15 live births .
Although satisfactory results have been obtained, ovarian tissue cryopreservation is still considered an experimental procedure and the selection criteria to take into consideration are strict: patients should be less than 35 years old, with a chance of survival of 5 years and a risk of POI of 50% [12–14].
The major concern of the transplantation procedure is related to the risk of reintroducing malignant cells that may be within the tissue and that may have survived to the cryopreservation, leading to disease recurrence upon transplantation in the patient. Therefore, prior to performing the procedure, it is of vital importance to take into account such risk and to detect the possible presence of minimal disseminated disease (MDD) in the frozen-thawed ovarian tissue.
There are three methods to evaluate ovarian involvement by malignant cells:
1. Conventional pathological examinations, by hematoxylin and eosin staining (H&E) and immunohistochemistry (IHC).
2. Molecular methods to detect specific chromosomal abnormality: RT–PCR amplification, flow cytometry and fluorescence in situ hybridization are useful to detect tumor-specific chromosomal translocations that result in the fusion genes and their protein products.
3. Xenografting to immunodeficient mice to examine the ability of cancer cells to survive and induce disease recurrence.
Leukemia is a hematological disease and the most frequent cancer in children. About 75% of leukemia diagnoses in children and teenagers are acute lymphoblastic leukemia (ALL) and the majority of the remaining cases are acute myeloid leukemia (AML) .
Thanks to advances in cancer treatments, survival rates have greatly increased, exceeding 80%, but such treatments should be started in emergency and, in case of relapse, require high-dose chemotherapy with alkylating agents and bone marrow transplantation [16, 17]. Leukemia patients are therefore at high risk of POI and infertility, rendering fertility preservation with ovarian tissue transplantation an option to be offered. Being a blood-borne disease, malignant leukemic cells can be found in the bloodstream and the chances of ovarian involvement are high as well. Kyono et al.  detected ovarian involvement in 8–10% of the autopsy files of patients with leukemia. Several efforts have been made to detect MDD in cryopreserved ovarian tissue from leukemia patients. The first who identified leukemic cells in ovarian tissue was Meirow in 2008 . He reported the case of two patients who suffered from chronic myeloid leukemia (CML) and who requested transplantation of their ovarian tissue for fertility restoration. Analysis prior to transplantation was performed: pathological and immunohistochemical results were negative in both patients, while RT–PCR detected the BCR-ABL1 fusion transcripts, hallmark of CML, in the ovarian tissue of one patient, while the other was negative.
Dolmans et al.  evaluated the presence of leukemic cells in the ovarian tissue from six patients with CML and 12 patients with ALL, cryopreserved during the active phase of disease. The analysis was performed by testing the specific cytogenetic abnormality or gene rearrangement found in the blood or bone marrow at diagnosis in the frozen-thawed ovarian tissue by PCR for each patient; the six CML patients were tested for the fusion gene BCR-ABL1, while concerning the ALL patients, two showed specific translocations, in eight cases immunoglobulins and/or T-cell receptor (TCR-γ) rearrangement genes were tested and in the remaining two, no molecular markers of disease were found and were therefore excluded from the analysis. Also in this case, no malignant cells were found by histology in the ovarian tissue of any patients, while RT–PCR detected the BCR-ABL leukemic marker in the ovarian tissue of two of six CML patients (33%) and positive molecular markers in the ovarian tissue of 7 of 10 ALL patients (70%). The frozen-thawed ovarian tissue was also grafted into SCID mice for 6 months to evaluate disease recurrence. No macroscopic (visible masses) nor microscopic (histology) disease was detected in mice grafted with ovarian tissue from CML patients, while 5 of 12 grafts from ALL patients showed to be invaded by lymphoblasts (Figure 26.1). Noteworthy, all CML patients received treatment prior to ovarian tissue cryopreservation, while among the ALL patients, of the seven patients with positive results, four had not undergone any treatment before ovarian tissue cryopreservation and three had received one regimen of chemotherapy prior to ovarian tissue preservation. The three ALL patients with negative results had undergone chemotherapy prior to ovarian tissue preservation, as well.
Figure 26.1 (A) Human follicle encircled by a large number of lymphocytes. Normal ovarian stroma is no longer present. Original magnification ×100. (B) Massive lymphocytic invasion of the ovarian graft. The histologic abnormalities observed in these lymphocytes were identified as malignant in nature and attributed to leukemic invasion
In two similar studies, Rosendahl et al.  and Greve et al.  analyzed the cryopreserved ovarian tissue from leukemia patients in complete remission. Also in this case, histology and immunohistochemistry were not able to detect leukemic cell infiltration in any of the frozen-thawed ovarian tissue under analysis. RT–qPCR, instead, allowed detection of leukemic cells in the frozen-thawed ovarian tissue of two of four patients with known molecular markers. Nevertheless, these results were not confirmed by the subsequent xenografting experiment in nude mice, as RT–PCR performed on ovarian tissue xenografted for 20 weeks did not demonstrate the presence of any leukemic cells.
In 2018, Shapira et al.  reported the first live birth in an AML patient after transplantation of cryopreserved ovarian tissue collected in complete remission. Ovarian tissue fragments were tested prior to reimplantation, under light microscopy after H&E and IHC for CD43 and C-KIT (CD117) to detect leukemic cells. A xenografting experiment was also conducted on SCID mice for 6 months. Molecular analysis techniques were also performed, as fluorescence in situ hybridization for MLL gene rearrangement and a next-generation sequencing panel to assess any leukemogenic molecular aberrations. The entire panel of analysis was unable to show the presence of leukemic cells, therefore transplantation was allowed.
To conclude, leukemic cells are found in >50% of ovarian tissue cryopreserved during the active phase of the disease, while ovarian tissue collected in complete remission, although showing positive RT–PCR results, appear to contain a number of viable cells not high enough to retransmit the disease upon transplantation.
Among adolescents, lymphoma is the most common malignancy, causing >25% of newly diagnosed cancers in young aged from 15 to 19. Hodgkin’s lymphoma (HL) accounts for the majority of cases, while the remaining cases are affected by one of the subtypes of non‐Hodgkin’s lymphoma (NHL) . Kyono et al. reports ovarian involvement in 4.3% (5/115) of patients with Hodgkin’s lymphoma and 9.8% (5/51) of patients with non-Hodgkin’s lymphoma. Both types of lymphomas are frequent indication of ovarian tissue cryopreservation and transplantation .
The first live birth after orthotopic transplantation of cryopreserved ovarian tissue has been reported in a patient affected by stage IV HL in 2004 . Several teams have studied the safety of ovarian tissue cryopreservation in HL patients. Meirow et al.  was the first to test the cryopreserved ovarian tissue from seven patients with advanced-stage HL: no involvement by Reed–Sternberg (RS) cells, characteristic of HL, was detected at H&E. Kim et al.  grafted ovarian tissue from 13 HL patients to SCID mice for 16 weeks and he was unable to detect disease recurrence. Seshadri et al.  tested the cryopreserved OT from 26 patients with HL, nine of whom had already received some chemotherapy, and he performed IHC for CD15 and CD30, characteristic of RS cells. Also in this case, there was no evidence of HL involvement by H&E or IHC in any of the samples examined. Meirow in 2008  performed another study to detect malignant cells on 33 HL patients, 11 of whom had stage IV disease, and no malignant cell was detected in none of the patients with H&E and CD30 IHC. In one case, however, ovarian involvement was reported in a patient affected by HL stage IIIB: at laparoscopy for OTC prior to chemotherapy, one portion of the ovary appeared suspect and was removed for pathological examination: the fragment contained no healthy ovarian tissue and was entirely composed of histiocytes, lymphocytes, and eosinophils with scattered large cells with polylobated nuclei, that were for CD 30 and CD 15 positive, proving to be HL (Figure 26.2) .
Figure 26.2 (A) Reactive lymphocytes and occasional atypical Hodgkin’s cells with large, irregular multilobated nuclei (arrow) and plump cytoplasm (H & E; ×200). (B) Neoplastic cells showing distinctive cell membrane and paranuclear dot staining positive for CD30 with immunoperoxidase (200). (C) Scattered B lymphocytes staining positive for CD15 with immunoperoxidase, unstained small cells are reactive T lymphocytes (200). (D) Neoplastic cells (arrow) showing no staining with CD20 immunoperoxidase (400)
Meirow et al.  and Kim et al.  both failed to detect NHL cells in the cryopreserved OT from 14 and 5 NHL patients, respectively, either at histology or after xenografting for 16 months. Nevertheless, in a series of 32 patients with NHL, Dolmans et al.  report detection of malignant cells in the cryopreserved ovarian tissue of two NHL patients (6%) by H&E and anti-CD20 IHC, one in the medulla and one in the cortex (Figure 26.3). Although the risk is low, it nevertheless exists and therefore warrants investigation prior to transplantation .
Figure 26.3 Ovarian fragments from a non-Hodgkin’s lymphoma patient. (A) Immunohistochemistry anti-CD20 (against NHL cells) is negative in the cortex (blue staining) and positive in the medulla (brown staining). Magnification of the white rectangle is represented in (B) and magnification of the black rectangle in (C). (B) Cells with an enlarged nucleus and patchy chromatin are disseminated in the tissue and stain strongly for anti-CD20. (C) Piece of cancer cell-free cortex. Follicles present in the cortex appear healthy. (D) Another fragment from the same patient shows massive invasion by NHL cells in the cortex.
Breast cancer is the most common cancer in women after skin cancer. The average risk of developing breast cancer is about 12%, and it accounts for about 30% of new cancer diagnoses in women . It is estimated that more than 250,000 cases of invasive breast cancer have been diagnosed in the United States in 2017 . The incidence of ovarian metastases in breast cancer patients varies from 13.2% to 37.8%, with a tendency to be more frequently observed in patients with advanced stage [18, 34–36]. Sanchez-Serrano et al.  analyzed the ovarian cortical biopsy of 63 women affected by stages I to IIIa infiltrating ductal breast carcinoma by H&E and anti-low molecular weight cytokeratin (CAM 5.2), anti-Gross Cystic Disease Fluid Protein-15 (GCDFP1), anti-Mammaglobin 1 (MGB1) and anti-Wilms’ tumor antigen-1 (WT1) IHC. Rosendahl et al.  performed an analogous study on 51 breast cancer patient and performed H&E and IHC anti-cytokeratin 7 (CK7), anti-CK-aecam, CA125, and anti-WT1 IHC. Both studies failed to detect any malignant breast cancer cell in the ovarian tissue under analysis. Several breast cancer patients have been transplanted so far, and in one case, a local relapse occurred at the site of the primary tumor about one year after the transplant, with no sign of metastatic activity in the ovarian graft.
Concerning advanced stage breast cancer patient, Luyckx et al.  conducted a study to detect minimal disseminated disease in cryopreserved ovarian tissue from 13 breast cancer patients with stage 2 A with positive nodes, ≥stage 2B or large tumors. They performed H&E, IHC for epithelial membrane antigen (EMA), Her-2/neu, and GCDFP15, and also performed qPCR for mammaglobin 2 (MGB2) gene. They also grafted the frozen-thawed ovarian tissue to SCID mice for 6 months. Results of histology and IHC failed to detect any malignant cell in the ovarian tissue and no mouse developed cancerous masses, but MGB2 gene expression was detected at qPCR in four frozen-thawed ovarian tissue and in one graft. They concluded that, using sensitive methods and xenografting, there is a potential risk of having malignant cells in cryopreserved ovarian tissue from advanced-stage breast cancer. Nevertheless, Bockstaele et al.  proved that the expression level of specific breast cancer genes in the ovarian tissue is highly variable. MGB1 and GCDFP15 genes proved to have a good positive predictive value to detect breast cancer cells in the ovarian cortex, while MGB2 detection had greater specificity in the ovarian medulla.
Ovarian cancer is the fifth most lethal cancer among women and frequently occurs in postmenopausal women. Nevertheless, 12% occur in patients of child-bearing age, rendering fertility preservation fundamental . Ovarian tissue cryopreservation can be performed from the contralateral ovary, but safety concerns easily arise, as ovarian tumors originate from the same organ that has to be cryopreserved and transplanted, therefore increasing the chances of preserving malignant cells within the ovarian tissue. In 2011, Lotz et al.  conducted a study to evaluate MDD detection in ovarian tissue from ovarian cancer patients. The frozen-thawed ovarian tissue from 10 ovarian cancer patients (three mucinous adenocarcinoma, two dermoid cysts, two borderline ovarian tumors, two dysgerminoma, and one granulosa cell tumor) was xenografted into mice for 24 weeks, and MDD was evaluated by means of H&E staining and IHC with a pan-cytokeratin panel. No malignant cell contamination was evidenced in the tissue in analysis.
Moreover, transplantation of ovarian tissue from ovarian cancer patients is always a risky procedure as there is the chance of redeveloping the primary disease in the grafted tissue after transplantation.
Four cases of OTC and reimplantation in ovarian cancer patients have been reported in the literature (Table 26.2) [43–46]. In three patients, pregnancies and live births of healthy babies have been obtained, while in one case hormonal activity never restarted. In one patient, a relapse occurred: she was affected by granulosa cell tumor that required ovariectomy and prophylactic removal of the contralateral ovary with OTC. After 9 years, the patient asked for reimplantation of her ovarian tissue, and after low-dose hormonal stimulation and in vitro fertilization, two embryos were transferred and gave a twin pregnancy. The patient underwent a cesarean section and delivered two healthy babies, but during the procedure, macroscopic tumor dissemination was observed by the surgeon, at the level of the diaphragm and of the peritoneum, but not in the graft sites. What caused the relapse remains unknown, on one side it could be the consequence of the transplanted ovarian tissue, on the other the response of a microscopic peritoneal disease to the pregnancy-induced hormonal environment.
|Reference||Diagnosis||Age at cryo||Site of transplant||ART||Pregnancy||Relapse|
|Stern et al. [43, 44]||21||Heterotopic||IVF||Twin pregnancy||Yes|
|Dittrich et al. ||Dysgerminoma, T1a||30||Orthotopic||No||No hormone activity||No|
|Dittrich et al. ||31||Orthotopic||No||Live birth||No|
|Kristensen et al. ||Mucinous cystadenocarcinoma 1 C||23||Heterotopic||IVF x2||No|