Univfy, Stanford scientists develop first validated prognostic test for predicting IVF live birth outcomes

Jul 20 2010

Univfy, a pioneer in the development of personalized in vitro fertilization (IVF) prognostic tools, announced today the publication of new research findings in the journal Proceedings of the National Academy of Sciences (PNAS) by its founding scientific team and clinical collaborators from Stanford University.  The peer-reviewed paper details their development of the first rigorously validated prognostic test for predicting IVF live birth outcomes for patients who have failed their first IVF attempt. In their validation test, this novel prognostic algorithm showed a more than 1000-fold improved live birth outcome prediction when compared to the age-based prediction model that is commonly used today (p<0.05).  Using this prediction model, the validation data showed that more than 60 percent of patients in the study received a significantly different, yet more accurate, prediction of live birth outcome in a subsequent IVF cycle as compared to conventional age-based prediction.

Stanford scientists, including Univfy's co-founders, developed the proprietary algorithm by examining 52 prognostic factors, such as body mass index, embryo development, and response to hormonal medications, along with analyzing the impact of those factors on IVF live birth success. The researchers extracted IVF outcomes data from more than 5000 IVF cycles performed at Stanford Hospital and Clinics from 2003-2008. 1676 treatments comprised the training data set that was used to generate the prediction model, which was subsequently validated in a separate set of 634 treatments that were performed in 2007-2008 in an unrelated group of patients.  

Univfy has obtained the exclusive license from Stanford University to develop and market this technology (patent pending), and plans to offer test services to patients and physicians in the fall of 2010, subject to clearance for use by the U.S. Food and Drug Administration (FDA).  These new prognostic tests would allow IVF physicians to provide live birth probabilities that are specific to each patient – and thus, are more meaningful – to best inform them as they navigate the difficult decisions that often come with IVF treatment.

"We believe that providing high quality, customized prognostics will empower patients with the information they need to make IVF decisions with greater confidence. We are excited about this merging of personalized medicine with reproductive medicine," stated Mylene W. M. Yao, M.D., president and co-founder of Univfy.  "Since this model uses clinical data from a previously failed IVF attempt, a first IVF treatment can be viewed both as an infertility treatment and as a potential prognostic tool for future cycles. Most remarkably, Univfy's proprietary algorithms do not require any additional blood tests or medical procedures, but rather, utilize a patient's existing clinical data to provide a personalized and highly accurate probability of live birth."

SOURCE Univfy