Pseudo contrastive labeling for predicting IVF embryo developmental potential.

In vitro fertilization is typically associated with high failure rates per transfer, leading to an acute need for the identification of embryos with high developmental potential. Current methods are tailored to specific times after fertilization, often require expert inspection, and have low predictive power. Automatic methods are challenged by ambiguous labels, clinical heterogeneity, and the inability to utilize multiple developmental points. In this work, we propose a novel method that trains a classifier conditioned on the time since fertilization. This classifier is then integrated over time and its output is used to assign soft labels to pairs of samples. The classifier obtained by training on these soft labels presents a significant improvement in accuracy, even as early as 30 h post-fertilization. By integrating the classification scores, the predictive power is further improved. Our results are superior to previously reported methods, including the commercial KIDScore-D3 system, and a group of eight senior professionals, in classifying multiple groups of favorable embryos into groups defined as less favorable based on implantation outcomes, expert decisions based on developmental trajectories, and/or genetic tests.

Prediction of sperm extraction in non-obstructive azoospermia patients: a machine-learning perspective.

STUDY QUESTION: Can a machine-learning-based model trained in clinical and biological variables support the prediction of the presence or absence of sperm in testicular biopsy in non-obstructive azoospermia (NOA) patients? SUMMARY ANSWER: Our machine-learning model was able to accurately predict (AUC of 0.8) the presence or absence of spermatozoa in patients with NOA. WHAT IS KNOWN ALREADY: Patients with NOA can conceive with their own biological gametes using ICSI in combination with successful testicular sperm extraction (TESE). Testicular sperm retrieval is successful in up to 50% of men with NOA. However, to the best of our knowledge, there is no existing model that can accurately predict the success of sperm retrieval in TESE. Moreover, machine-learning has never been used for this purpose. STUDY DESIGN, SIZE, DURATION: A retrospective cohort study of 119 patients who underwent TESE in a single IVF unit between 1995 and 2017 was conducted. All patients with NOA who underwent TESE during their fertility treatments were included. The development of gradient-boosted trees (GBTs) aimed to predict the presence or absence of spermatozoa in patients with NOA. The accuracy of these GBTs was then compared to a similar multivariate logistic regression model (MvLRM). PARTICIPANTS/MATERIALS, SETTING, METHODS: We employed univariate and multivariate binary logistic regression models to predict the probability of successful TESE using a dataset from a retrospective cohort. In addition, we examined various ensemble machine-learning models (GBT and random forest) and evaluated their predictive performance using the leave-one-out cross-validation procedure. A cutoff value for successful/unsuccessful TESE was calculated with receiver operating characteristic (ROC) curve analysis. MAIN RESULTS AND THE ROLE OF CHANCE: ROC analysis resulted in an AUC of 0.807 ± 0.032 (95% CI 0.743-0.871) for the proposed GBTs and 0.75 ± 0.052 (95% CI 0.65-0.85) for the MvLRM for the prediction of presence or absence of spermatozoa in patients with NOA. The GBT approach and the MvLRM yielded a sensitivity of 91% vs. 97%, respectively, but the GBT approach has a specificity of 51% compared with 25% for the MvLRM. A total of 78 (65.3%) men with NOA experienced successful TESE. FSH, LH, testosterone, semen volume, age, BMI, ethnicity and testicular size on clinical evaluation were included in these models. LIMITATIONS, REASONS FOR CAUTION: This study is a retrospective cohort study, with all the associated inherent biases of such studies. This model was used only for TESE, since micro-TESE is not performed at our center. WIDER IMPLICATIONS OF THE FINDINGS: Machine-learning models may lay the foundation for a decision support system for clinicians together with their NOA patients concerning TESE. The findings of this study should be confirmed with further larger and prospective studies. STUDY FUNDING/COMPETING INTEREST(S): The study was funded by the Division of Obstetrics and Gynecology, Soroka University Medical Center, there are no potential conflicts of interest for all authors.

Cell-free DNA and telomere length among women undergoing in vitro fertilization treatment.

PURPOSE: The current research is aimed at finding potential non-invasive bio-markers that will help us learn more about the mechanisms at play in failed assisted reproduction treatment. This exploratory pilot study examined the relationship between cell-free DNA (CFD) in plasma and telomere length in lymphocytes among women undergoing in vitro fertilization (IVF) and compared telomere length and CFD levels to a healthy control group. METHODS: Blood of 20 women undergoing IVF was collected at three time points during the IVF cycle. We assessed the relationship between CFD and telomere length as well as controlling for morning cortisol levels. We also collected blood of 10 healthy controls at two time points (luteal and follicular phases of the menstrual cycle) and compared mean telomere length, CFD, and cortisol levels between the IVF patients and healthy controls. RESULTS: The results revealed an inverse relationship between CFD levels and telomere lengths at several time points that remained significant even after controlling for cortisol levels. Women undergoing IVF had statistically significant higher levels of CFD and shorter telomeres compared to healthy controls. CONCLUSIONS: The relationship between telomere length and CFD should be further explored in larger studies in order to uncover potential mechanisms that cause both shortened telomere length and elevated CFD in women undergoing IVF.

Inhibition of topoisomerase I by anti-cancer drug altered the endometrial cyclicity and receptivity.

Topoisomerase I (topo I) is an essential nuclear enzyme involved in virtually all aspects of gene expression, and is the target of the anti-cancer drugs- camptothecin (CPT) and its derivatives. Improvement of the survival rates of young women with cancer has led to the consideration of the effects of long-term chemotherapy on their fertility. The effect of anticancer drugs on ovarian function was previously investigated; however, no reports are available concerning their effect on the endometrium, whose integrity is an important factor in embryo implantation. Here we used a rat animal model to investigate the expression and activity of topo I in the various physiologic phases of the endometrium and the influence of CPT on its integrity and receptivity. The results show, for the first time, that the endometrial topo I level and activity are influenced by the physiologic phases of the endometrium (estrous cycle) and correlate with the estrogen blood concentration. Treatment with the anti-cancer drug CPT caused histological disruption of the endometrium and deleterious effect on its cyclicity. Moreover, CPT treatment significantly reduced the implantation rate of embryos, suggesting alteration in the receptivity of the endometrium. These results suggest that topo I is important for maintaining the normal physiologic cyclicity and functionality of the endometrium in rats. Anti-cancer agents that target topo I severely impair estrous cycle progression and endometrial integrity and receptivity, emphasizing the importance of addressing the effect of chemotherapy on the endometrial functionality.

DNA topoisomerases I and II in human mature sperm cells: characterization and unique properties.

BACKGROUND: The condensed state of the human sperm's chromatin is essential for the compact structure of the spermatozoon head, which is important for the fertilization process. The enzymes DNA topoisomerases (topo I and topo II) are responsible for the topological structure of the chromatin in somatic cells. The activities and the characterization of topoisomerases in mature human sperm cells have not been previously investigated. METHODS: Sperm cells were purified from the semen of healthy donors by standard procedures and assays measuring the activities, protein size and sensitivity to inhibitors of topoisomerases were performed. RESULTS: Topo I and topo II DNA relaxation activities are present in nuclear extracts derived from human sperm. The sperm topo I activity is inhibited by camptothecin, similarly to the somatic enzyme. An 85 kDa sperm protein, compared with the 100 kDa somatic topo IB enzyme, reacted with anti-human topo I antibody. Sperm topo II lacks the DNA decatenation activity of the somatic enzyme and a 97 kDa protein, compared with the 170 kDa somatic topo IIalpha enzyme, was detected with anti-human topo II antibody. Sperm nuclear extracts contained inhibitors of somatic topo II decatenation activity. CONCLUSIONS: Topoisomerase I and II activities as well as topo I and topo II proteins are present in mature human sperm cells. These enzymes possess unique properties compared with their somatic counterparts.

Relationship between sexual abstinence duration and the acrosome index in teratozoospermic semen: analysis of 1800 semen samples.

A high acrosome index (percentage of sperm with normal acrosome morphology--cutoff value > or =10%) is known to be associated with an improved fertilization rate in conventional IVF. A retrospective evaluation of the relationship between duration of sexual abstinence and acrosome index among oligozoospermic and normozoospermic semen samples with teratozoospermia was undertaken. A significant (P = 0.001) decrease in the acrosome index was observed among the normozoospermic samples (n = 1264) between the peak value of 10.2 +/- 3.6% on day 2 and 8.5 +/- 4.0% on day 5 of abstinence, while for the oligozoospermic samples (n = 536) the peak value of 8.7 +/- 3.5% was observed on day 1 and the lowest values of 6.8 +/- 3.7% (P = 0.04) on day 5 of abstinence. The results suggest that an optimal acrosome index will be obtained following a short sexual abstinence.

Effect of cryopreservation on quality and fertilization capacity of human sperm.

OBJECTIVE: To analyze retrospectively the effect of cryopreservation on donor's sperm. STUDY DESIGN: Data were collected on 178 cryopreserved-thawed sperm specimens from 44 donors and 624 oocytes from 62 women, which underwent in vitro fertilization-embryo transfer (IVF-ET) treatments with donor's sperm. Data on fresh sperm, 175 sperm specimens from 76 couples which underwent IVF-ET treatments, were used as a control group. Semen analysis was done by cell concentration, percent of motility, and quality of motility according to the World Health Organization (WHO) recommendation. Sperm quality parameters which had the strongest impact on fertilization capacity were determined using the statistical response surface model and conjoint analysis. RESULTS: Passing sperm through Percoll column decreased sperm concentration, with no improvement in sperm motility but with a slight increase in quality of motility. Quality of motility of donor's sperm had the strongest impact on fertilization capacity. CONCLUSION: Current freezing-thawing protocols of sperm cause a decrease in sperm parameters without affecting fertilization capacity. Furthermore, quality of motility of frozen-thawed sperm seems to be a significant measure of sperm fertilization capacity.

Monozygotic twins following blastocyst transfer: a report of two cases.

Several studies have shown an increased risk for monozygotic twinning after fertilization treatments. We present the clinical and sonographic characteristics of two monozygotic twin cases following blastocyst transfer. It is suggested that delayed transfer of the embryo in the blastocyst stage is a contributing factor to monozygotic twinning.

Differential expression of the H-2Dk MHC class-I antigen and tissue inhibitor of metalloproteinases in metastatic and nonmetastatic T-10 fibrosarcoma cells.

In previous studies, we have demonstrated that the highly metastatic IE-7 cell clone, derived from the T-10 fibrosarcoma, expressed both the H-2Dk and H-2Db genes, whereas a nonmetastatic IC-9 clone, derived from the same tumor, expressed only H-2Db, suggesting that the H-2Dk product might be involved in the metastatic phenotype. To substantiate this notion, IC-9 cells were transfected with an H-2Dk-expressing vector. Although all of the 4 randomly selected transfectant subclones elicited high H-2Dk expression, only one was as metastatic as IE-7 cells. This metastatic transfectant resembled IE-7 cells also in its inability to evoke CTL response in syngeneic mice, whereas the other transfectants were quite competent in this respect. It thus appears that the H-2Dk product may contribute to the metastatic phenotype provided that it is immunogenically abnormal. In addition, the present study provides evidence to suggest that lack of production of the tissue inhibitor of metalloproteinases TIMP-1/TIMP-2 is another important determinant in the metastatic phenotype of these cells.

Resistance to NK and metastatic potential of fibrosarcoma cells is associated with products encoded by the H-2D region.

We have used the murine 3-methylcholanthrene induced T10 fibrosarcoma tumor cell system originating in (C3II/en x C57BL/6)F1 mice (H-2b x H-2k) to elucidate the possible correlation between metastatic potential, expression of individual H-2 antigens and susceptibility to NK cells. Transfection of the non metastatic and NK sensitive IC9 cells (Db+, Kk, Kb, Kk-) with the H-2Dk gene, altered the metastatic phenotype of the parental cells, yet had no effect on the susceptibility of these tumor cells to lysis by NK and did not elicit a specific CTL response in syngeneic hosts. Variants of the metastatic and NK resistant IE7 clone (Db+, Kk-, Kb-, Kk-), lacking H-2Dk, were selected by treatment with monoclonal anti H-2Dk antibodies and complement. These variants were sensitive to NK and poorly or non metastatic. Transfection of Dk negative variants with the H-2Dk gene, resulted in the isolation of several clones which expressed a wide range of metastatic phenotypes but maintained sensitivity to NK. In addition, by cloning the cDNA of the H-2Dk gene of the metastatic T10-IE7 variant cells and analyzing its nucleotide sequence, we found four single nucleotide changes. Two of them are not expected to alter the encoded amino acids, whereas the others should result in two amino acid substitutions in the alpha-2 domain of the class I H-2Kd protein product. These changes might account, at least partially, for the failure of the transfection of H-2Dk to restore resistance to NK.(ABSTRACT TRUNCATED AT 250 WORDS)

Differential transcriptional control of the H-2K and H-2D loci of the major histocompatibility complex in fibrosarcoma cells.

In this study we demonstrate a differential transcription of H-2K and H-2D class-I genes in two different tumor cell clones; one is highly metastatic (IE-7) and the other is not metastatic (IC-9), both derived from the same fibrosarcoma, T-10, induced in an (H-2b x H-2k)F1 mouse. The expression of the two parental H-2K alleles is transcriptionally suppressed in both of these clones. In addition the IC-9 clone does not transcribe also the H-2Dk allele. Our data rule out the possibility that this suppression results from enhanced RNA degradation, impaired polyadenylation, DNA rearrangement, or changes in DNA methylation within these genes. Interferons (IFN) are known to enhance MHC expression by acting on a consensus IFN responsive element present in the promoter region of MHC genes. However, IFN-gamma, which is the most potent IFN in this respect, failed to activate the expression of the silent MHC genes in our cells. This finding may reflect a defect within the promoter region of these genes or changes in their chromatin structure.

Changes in the coding sequence of the H-2Dk gene of metastatic cells that might account for immunogenic abnormality of its encoded antigen.

In previous studies we have demonstrated that metastatic cells, derived from T-10 fibrosarcoma, express an immunogenically abnormal H-2Dk glycoprotein which is involved in manifesting their metastatic phenotype. In the present study we show that these cells contain a remarkably high level of H-2Dk specific mRNA. Moreover, by cloning cDNA of this gene and analyzing its nucleotide sequence, we found 4 single nucleotide changes. Two of them did not change the encoded amino acids, whereas the others resulted in two amino acid substitutions in the alpha-2 domain of the protein product that might account for its immunogenic abnormality.

NK sensitivity, H-2 expression and metastatic potential: analysis of H-2Dk gene transfected fibrosarcoma cells.

We have used the 3-Methylcholanthrene induced T-10 fibrosarcoma tumour cell system (H-2b xH-2k)F1 to elucidate the possible correlation between metastatic potential, expression of individual H-2 antigens and susceptibility to NK cells. Transfection of the non-metastatic and NK sensitive IC9 cells (Db+, Dk-, Kb-, Kk-) with the H-2Dk gene, altered the metastatic phenotype of the parental cells, yet had no effect on the susceptibility of these tumour cells to lysis by NK and did not elicit a specific CTL response in syngeneic hosts. Variants of the metastatic and NK resistant IE7 clone (Db+, Dk+, Kb-, Kk-), lacking H-2Dk, were selected by treatment with monoclonal anti H-2Dk antibodies and complement. These variants were sensitive to NK and poorly or non metastatic. Retransfection of 'Dk' 'loss' variants with the H-2Dk gene, resulted in the isolation of several clones expressing a wide range of metastatic phenotypes but maintained sensitivity to NK. These results indicate that the H-2D region of the MHC and or closely linked genes may be involved in the complex interrelationship between target susceptibility to NK and metastasis.

NK sensitivity, H-2, c-K-ras proto-oncogene expression and metastases: analysis of the metastatic potential of H-2 gene transfected fibrosarcoma cells.

The transformation of a potentially neoplastic cell into an autonomous highly malignant and metastatic tumor cell involves a multifactorial cascade of events. This will eventually lead not only to the emergence of a tumor cell with an unlimited potential of replication, but more important will contribute to its ability to ignore and evade homeostatic immune and non-immune regulatory mechanisms. Specifically, those mechanisms which may restrict and direct its growth, dissemination, patterns of differentiation and interaction with the cellular and humoral factors comprising its environment. However, many different factors may contribute to a highly invasive and malignant phenotype. It is obvious that one should expect that a cardinal role should be assigned to alterations in those factors which contribute to the capacity of the malignant cells with its environment at the cell membrane level, which in turn is dependent on the concerted functional expression of specialized membrane associated components (i.e. receptors, cyto adhesion molecules (CAM's), histocompatibility antigens, GAP junction complexes, extracellular matrix components, etc.). In the present studies, we have investigated the contribution of three major factors, which maybe the cause or result of alterations at the level of the cell membrane: MHC encoded antigen expression, susceptibility to the cytolytic activity of NK cells and enhanced expression of the c-K-ras proto-oncogene, as to their development of metastatic capacity of a malignant cell. To address these questions, we used metastatic (IE7) and non-metastatic (IC9) variants of the murine 3-methylcholanthrene induced T-10 fibrosarcoma. Using this system, the following major conceptually important observations were made: A. The restoration by transfection of the expression of membrane associated H-2K encoded glycoproteins abrogates the metastatic capacity of the highly metastatic tumor cell clone, IE7, irrespective of the degree of susceptibility to NK or c-K-ras oncogene expression. This reduction in metastatic capacity is followed by a significant decrease in its tumorigenicity which is concomitant with its ability to induce in vivo potent H-2K restricted CTL's. These results clearly indicate that H-2K region encoded molecules play no apparent role in determining the susceptibility of tumor cells to NK cells, and yet their loss or aberrant expression is a cardinal event in tumor progression towards metastatic capacity, a fact which is supported by similar observations achieved in other murine models (18).(ABSTRACT TRUNCATED AT 400 WORDS)

The immunobiology of metastatic processes: analysis of NK sensitivity and the metastatic potential of H-2 gene transfected fibrosarcoma cells.

The transformation of a potentially neoplastic cell into an autonomous highly malignant and metastatic tumor cell involves a multifactorial cascade of events. This will eventually lead not only to the emergence of a tumor cell with an unlimited potential of replication, but more important will contribute to its ability to ignore and evade homeostatic immune and nonimmune regulatory mechanisms. Specifically, those mechanisms which may restrict and direct its growth, dissemination, patterns of differentiation and interaction with the cellular and humoral factors comprising its environment. In the present studies we have investigated the contribution of three major factors which may be the cause or result of alterations at the level of the cell membrane: MHC encoded antigen expression, susceptibility to the cytolytic activity of NK cells and enhanced expression of the c-K-ras proto-oncogene, as to their development of the metastatic capacity of a malignant cell. To address these questions we used metastatic (IE7) and nonmetastatic (IC9) variants of the murine 3-methylcholanthrene-induced T-10 fibrosarcoma. Using this system, the following major conceptually important observations were made: (A) The restoration by transfection of the expression of membrane associated H-2K encoded glycoproteins abrogates the metastatic capacity of the highly metastatic tumor cell clone, IE7, irrespective of the degree of susceptibility to NK or c-K-ras oncogene expression.(ABSTRACT TRUNCATED AT 250 WORDS)