To the Editor:
In a recent article by Schubert et al. (1), human ovarian cortical samples obtained from benign ovarian cysts were xenografted into SCID mice subcutaneously as either fresh or frozen-thawed pieces . Follicle counts in the grafts and their E2 production as a measure of endocrine function were assessed. While recognizing their valuable work and previous contributions to the field, there are certain points in their study that need to be clarified to the readers.
First, authors used a viability assay kit to determine living follicle density in fresh ovarian samples in order to determine follicle density and transplant the samples with higher follicle counts. These kinds of viability/cytotoxicity assays rely on the assessment of live and dead cells, which produce different staining color or signal under fluorescent microscopy, whereas follicle counts can only be determined by serial sectioning and counting follicles in a fixed specimen under light microscopy. Viability of individual cells using that assay can easily be determined. But in the case of a piece of ovarian tissue, how a follicle will be classified as dead or non-viable is questionable unless the whole follicle, granulosa cells or oocyte are stained positive for the dead signal. Furthermore, architecture of follicles and their compartments will not be easily identified in fresh, unfixed tissue, and only a limited portion of the sample, perhaps only a single plane, can be evaluated for viability under fluorescent microscopy. In addition, performing viability tests to frozen-thawed samples as well could have been more useful and informative, since the concern for the viability of frozen-thawed samples are much graver than for the fresh ones.
Second, in the formula that they used to calculate the mean number of primordial follicles in the samples, the thicknesses of ovarian cortical slices analyzed was taken as 0.06 mm (60 micrometers) even though they state in another part of the text as 1-2 mm. Determining follicle count in such an ultra-thin section might lead to underestimation of real follicle numbers, since fresh ovarian cortex has a thickness of at least 1-2 mm.
Third, some of the fresh, ungrafted ovarian samples have follicle densities somewhat lower than grafted tissues in their table. It should be emphasized that one of the most important factors determining the survival of follicles, hence the success of an ovarian transplant, is the ischemia that occurs until sufficient revascularization is established in the graft bed. During this hypoxic period, the graft relies upon simple diffusion for survival, and more than half of the primordial follicles are lost (2,3). An additional 6-10% will be lost during freezing-thawing processes. Therefore at least 50-6 % of primordial follicles will disappear until sufficient revascularization to prevent further follicle loss is established in a transplanted tissue. Furthermore, the follicle loss in the frozen-thawed cortical tissue after engraftment might be even greater than in fresh tissue due to slower and lower revascularization capability in the former. Therefore, the presence of a higher number of follicles in grafted tissue compared to fresh ones is contradictory. Even though the authors claimed that follicle densities increased in patients 3 and 5 postgrafting, this is more likely to be due to atechnical error in calculating follicle counts rather than a real increase or regeneration of follicles in the grafts analyzed. Therefore it is important to differentiate the term “increase” from postnatal regeneration or renewal of germ cells in order to prevent any misconception that may arise in the readers (4).
A recent study quantified the number of primordial follicles in age-matched cancer patients who underwent ovarian tissue freezing before and after chemotherapy (5). In that study, authors determined the number of primordial follicles (PF/mm2) in 0.5×0.5×0.3 cm ovarian samples serially sectioned 7 microns apart. Although follicle numbers decreased and dwindled progressively with age, follicle counts in age-matched untreated patients were comparable, underscoring the importance of the size of the samples used for histomorphometric analysis in order to determine accurate follicle densities without fluctuating numbers.
In conclusion, a thorough histological assessment of harvested ovarian samples is a prerequisite, not only for any evidence of disease inoculation, but also for an accurate estimation of follicle counts, since the success of the procedure is greatly hampered by follicle loss-induced cryodamage during freezing-thawing process and ischemic insult post-transplantation.
Ozgur Oktem MD
Bulent Urman, MD
Women’s Health and Assisted Reproduction
Fertility Preservation Program and Molecular Reproduction
American Hospital of Istanbul
1. Schubert B, Canis M, Darcha C, Artonne C, Smitz J, Grizard G . Follicular growth and estradiol follow-up after subcutaneous xenografting of fresh and cryopreserved human ovarian tissue. Fertil Steril. 2008 Jun;89(6):1787-94.
2. Oktem O, Oktay K. Ovarian tissue cryopreservation and other fertility preserving strategies. In Gardner DK, Weismann A, Howles CM, Shoham Z eds. Textbook of Assisted Reproductive Techniques, 2nd Edition. Florida Taylor & Francis 2004:315-327.
3. Baird DT, Webb R, Campbell BK, Harkness LM, Gosden RG. Long-term ovarian function in sheep after ovariectomy and transplantation of autografts stored at -196 C. Endocrinology. 1999; 140:462-71.
4. Johnson J, Canning J, Kaneko T, Pru JK, Tilly JL. Germline stem cells and follicular revewal in the postnatal mammalian ovary. Nature 2004; 428:145-50.
5. Oktem O, Oktay K. Quantitative assessment of the impact of chemotherapy on ovarian follicle reserve and stromal function. Cancer 2007; 110:2222-9.
Published online in Fertility and Sterility doi:10.1016/j.fertnstert.2008.12.084
The Authors Respond:
We thank Dr Oktem and Dr. Urman for their interest in our article.
The viability/cytotoxicity assay we used has the simple aim to answer the following question: Does the sample of ovarian cortex harvested from the patient contain a sufficient amount of viable follicles to be used for research? Advantages and inconveniences of this assay have been previously discussed (1).
The 0.06 used in our formula is referred to the height of 2 primordial follicles. When a primordial follicle is observed it can depend from the 30 µm above or beneath the section. Therefore around 0.06 mm of tissue depth is observed through one section. The sections were performed at 120 µm intervals, consequently 10-20 sections were analysed for a sample of 1-2 mm thickness (1).
Neither regeneration nor renewal of follicles was hypothesized in our study after grafting. Our explanation of the higher follicular density after grafting than in fresh tissue likely reflects the huge heterogeneity of follicles in human ovarian cortex. It is related to a different amount of follicles in different pieces of cortex rather than an apparent “increase.” Note the trend of a follicular density decrease after grafting, especially in all cases after grafting of frozen/thawed tissue.
Benoît Schubert, MD, PhD
Department of Biology of Reproduction
1. Schubert B., Canis M., Darcha A., Artonne C., Pouly J-L., Déchelotte P. et al. Human ovarian tissue from cortex surrounding benign cysts : a model to study ovarian tissue cryopreservation. Human Reproduction, 2005; 20:1786-92.
Published online in Fertility and Sterility doi:10.1016/j.fertnstert.2008.12.085