Deep phenotyping to predict live birth outcomes in in vitro fertilization

Abstract

Nearly 75% of in vitro fertilization (IVF) treatments do not result in live births and patients are largely guided by a generalized agebased prognostic stratification. We sought to provide personalized and validated prognosis by using available clinical and embryo data from prior, failed treatments to predict live birth probabilities in the subsequent treatment. We generated a boosted tree model, IVF BT , by training it with IVF outcomes data from 1,676 first cycles (C1s) from 2003-2006, followed by external validation with 634 cycles from 2007-2008, respectively. We tested whether this model could predict the probability of having a live birth in the subsequent treatment (C2). By using nondeterministic methods to identify prognostic factors and their relative nonredundant contribution, we generated a prediction model, IVF BT , that was superior to the age-based control by providing over 1,000-fold improvement to fit new data (p < 0.05), and increased discrimination by receiver-operative characteristic analysis (area-under-the-curve, 0.80 vs. 0.68 for C1, 0.68 vs. 0.58 for C2). IVF BT provided predictions that were more accurate for ∼83% of C1 and ∼60% of C2 cycles that were out of the range predicted by age. Over half of those patients were reclassified to have higher live birth probabilities. We showed that data from a prior cycle could be used effectively to provide personalized and validated live birth probabilities in a subsequent cycle. Our approach may be replicated and further validated in other IVF clinics.regression tree model | in vitro fertilization prediction | live birth prediction | in vitro fertilization prognostics | personalized medicine

Cumulative live-birth rates per total number of embryos needed to reach newborn in consecutive in vitro fertilization (IVF) cycles: a new approach to measuring the likelihood of IVF success

“…Thus, in this report, we describe an analysis of our database in which we focused on cumulative live-birth rate (CLBR) per total number of embryos replaced until a live birth was achieved (or EmbR) since 1998, considering each embryo transferred as a chance to achieve a live birth, and taking into consideration the relevance of age, infertility etiology, and day of ET as key parameters affecting IVF success (13)(14)(15).…”

Boosting predictions of treatment success

“…The authors also note that this boosted tree algorithm "allows many variables to be analyzed simultaneously, without need to select variables a priori" (1). In this case, with at most ∼50 potential variables and a reasonable number of patient cases, this is likely true.…”

“…This is a pervasive applied question in medicine. In PNAS, a report by Banerjee et al (1) considers such a problem with respect to personalized prediction of success of in vitro fertilization (IVF) treatments. The majority of IVF procedures do not achieve a live birth; therefore, providing predictions of success for subsequent IVF treatments should assist a patient with decisions, given the financial, physical, and emotional costs of undergoing IVF therapy.…”

“…The majority of IVF procedures do not achieve a live birth; therefore, providing predictions of success for subsequent IVF treatments should assist a patient with decisions, given the financial, physical, and emotional costs of undergoing IVF therapy. The authors note that, at this time, estimates of the success of IVF treatment are primarily based on age-based stratification (1).…”

Assessment of the Field of Assisted Reproductive Technology: Improvements and the Future of Human In Vitro Fertilization

“…While currently there is no universally accepted method for picking such an embryo, some new models show promise, albeit lacking verification and publication of results [18,19]. Certain characteristics, such as duration of infertility, maternal age, number of ART treatment cycles, and patient pathology, are most inversely correlated with embryo quality and, thus, pregnancy success [17,18,20,21]. The hope of these models is that multiple pregnancy rates can be greatly diminished by screening patients to determine good candidates (and embryos) for SET and thereby decreasing use of DET.…”