Estimation of age-dependent decrease in blastocyst euploidy by next generation sequencing: development of a novel prediction model.

BACKGROUND: We developed a model to estimate the female age-dependent decrease in blastocyst euploidy and the impact of blastocyst cohort size on the likelihood of having at least one euploid blastocyst for transfer.
METHODS: Retrospective analysis of 1296 trophectoderm biopsies by next-generation sequencing analysis from 436 infertile couples undergoing intracytoplasmic sperm injection and preimplantation genetic testing for aneuploidy. A logistic regression model was fit to the data. The dependent and independent variables were embryo genetic status and female age, respectively. The method of fitting was quadratic on age, and the model was validated with cross validation by a data splitting technique.
RESULTS: The decrease in the probability of blastocyst euploidy follows an age-dependent binomial distribution, progressing with every year of female age, from 1.2% to 24.5% in 28-44 years-old women (P<0.0001). The minimum number of blastocysts needed to obtain at least one euploid blastocyst for transfer was computed for different probabilities and female ages. At the age of 28 years, a total of three blastocysts is required to obtain at least one euploid blastocyst with 90% probability, whereas it is 4, 5, 6, 9, 16 and 29 for ages 35, 37, 39, 41, 43, and 45, respectively.
CONCLUSIONS: A novel prediction model estimates the probability of blastocyst euploidy and the number of blastocysts required to obtain at least one euploid embryo for transfer. This new resource based on f emale age and blastocyst cohort size will aid clinicians counsel and plan treatment of infertile couples undergoing IVF/ICSI.