Search

The acquisition of mesenchymal traits is considered a hallmark of breast cancer progression. However, the functional relevance of epithelial-to-mesenchymal transition (EMT) remains controversial and context dependent. Here, we isolate epithelial and mesenchymal populations from human breast cancer metastatic biopsies and assess their functional potential in vivo. Strikingly, progressively decreasing epithelial cell adhesion molecule (EPCAM) levels correlate with declining disease propagation. Mechanistically, we find that persistent EPCAM expression marks epithelial clones that resist EMT induction and propagate competitively. In contrast, loss of EPCAM defines clones arrested in a mesenchymal state, with concomitant suppression of tumorigenicity and metastatic potential. This dichotomy results from distinct clonal trajectories impacting global epigenetic programs that are determined by the interplay between human ZEB1 and its target GRHL2. Collectively, our results indicate that susceptibility to irreversible EMT restrains clonal propagation, whereas resistance to mesenchymal reprogramming sustains disease spread in multiple models of human metastatic breast cancer, including patient-derived cells in vivo.

STUDY QUESTION What is the success rate in terms of ovarian activity (menstrual cycles) as well as pregnancy and delivery rates 1 year after orthotopic ovarian transplantations conducted in a three-country network? SUMMARY ANSWER In 49 women with a follow-up >1 year after transplantation, the ovaries were active in 67% of cases and the pregnancy and delivery rates were 33 and 25%, respectively. WHAT IS KNOWN ALREADY Cryopreservation of ovarian tissue in advance of cytotoxic therapies and later transplantation of the tissue is being performed increasingly often, and the total success rates in terms of pregnancy and delivery have been described in case series. However, published case series have not allowed either a more detailed analysis of patients with premature ovarian insufficiency (POI) or calculation of success rates based on the parameter ‘tissue activity'. STUDY DESIGN, SIZE, DURATION Retrospective analysis of 95 orthotopic transplantations in 74 patients who had been treated for cancer, performed in the FertiPROTEKT network from 2008 to June 2015. Of those 95 transplantations, a first subgroup (Subgroup 1) was defined for further analysis, including 49 women with a follow-up period >1 year after transplantation. Of those 49 women, a second subgroup (Subgroup 5) was further analysed, including 40 women who were transplanted for the first time and who were diagnosed with POI before transplantation. PARTICIPANTS/MATERIALS, SETTING, METHODS Transplantation was performed in 16 centres and data were transferred to the FertiPROTEKT registry. The transplantations were carried out after oncological treatment had been completed and after a remission period of at least 2 years. Tissue was transplanted orthotopically, either into or onto the residual ovaries or into a pelvic peritoneal pocket. The success rates were defined as tissue activity (menstrual cycles) after 1 year (primary outcome) and as pregnancies and deliveries achieved. MAIN RESULTS AND THE ROLE OF CHANCE The average age of all transplanted 74 women was 31 ± 5.9 years at the time of cryopreservation and 35 ± 5.2 at the time of transplantation. Twenty-one pregnancies and 17 deliveries were recorded. In Subgroup 1, tissue was cryopreserved at the age of 30 ± 5.6 and transplanted at 34 ± 4.9 years. Ovaries remained active 1 year after transplantation in 67% of cases (n = 33/49), the pregnancy rate was 33% (n = 16/49) and the delivery rate was 25% (n = 12/49). In Subgroup 5, tissue was cryopreserved at the age 30 ± 5.9 years and transplanted at 34 ± 5.2 years. Ovaries remained active 1 year after transplantation in 63% of cases (n = 25/40), the pregnancy rate was 28% (n = 11/40) and the delivery rate was 23% (n = 9/40). The success rates were age dependant with higher success in women who cryopreserved at a younger age. In Subgroup 5, tissue was exclusively transplanted into the ovary in 10% (n = 4/40) of women and into a peritoneal pocket in 75% (n = 30/40), resulting in spontaneous conceptions in 91% of patients (n = 10/11). LIMITATIONS, REASONS FOR CAUTION The data were drawn from a retrospective analysis. The cryopreservation and transplantation techniques used have changed during the study period. The tissue was stored in many tissue banks and many surgeons were involved, leading to heterogeneity of the procedures. However, this does reflect the realistic situation in many countries. Although patients with POI were evaluated before transplantation to allow specific analysis of the transplanted tissue itself, the possibility cannot be excluded that residual ovarian tissue was also reactivated. WIDER IMPLICATIONS OF THE FINDINGS This is the largest case series worldwide to date and it confirms that cryopreservation and transplantation of ovarian tissue can be a successful option for preserving fertility. Persistent tissue activity 12 months after transplantation suggests that the pregnancy and delivery rates may increase further in the future. As transplantation into the peritoneum results in a high success rate, this approach may be an alternative to transplantation into the ovary. However, in order to establish the best transplantation site, a randomized study is required. STUDY FUNDING/COMPETING INTEREST This study was in part funded from the Deutsche Forschungsgemeinschaft (# DI 1525) and the Wilhelm Sander Foundation (2012.127.1) and did not receive any funding from a commercial company. No competing interests. TRIAL REGISTRATION NUMBER None.