Multiplexed Serum Biomarkers to Discriminate Nonviable and Ectopic Pregnancy

Objective The use of multiplexed biomarkers may improve the diagnosis of normal and abnormal early pregnancies. In this study we assessed 24 markers with multiple machine learning-based methodologies to evaluate combinations of top candidates to develop a multiplexed prediction model for identification of 1) viability and 2) location of an early pregnancy. Design A nested case-control design evaluating the predictive ability and discrimination of biomarkers in patients at risk of early pregnancy failure in the first trimester to classify viability and location Subjects 218 individuals with a symptomatic (pain and/or bleeding) early pregnancy: 75 with an ongoing intrauterine gestation, 68 ectopic pregnancies, and 75 miscarriages. Interventions Serum values of 24 biomarkers were assessed in the same patients. Multiple machine learning-based methodologies to evaluate combinations of these top candidates to develop a multiplexed prediction model for identification of 1) a nonviable pregnancy (ongoing intrauterine pregnancy vs miscarriage or ectopic pregnancy) and 2) an ectopic pregnancy (ectopic pregnancy vs ongoing intrauterine pregnancy or miscarriage). Main Outcome Measures The predicted classification by each model was compared to actual diagnosis and sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), conclusive classification, and accuracy were calculated. Results Models using classification regression tree analysis using three markers (PSG3, CG-Alpha and PAPPA) were able to predict a maximum sensitivity 93.3%, a maximum specificity 98.6%. The model with the highest accuracy was 97.4% (with 70.2% receiving classification). Models using an overlapping group of three markers (sFLT, PSG3 and TFP12) achieved a maximum sensitivity of 98.5%. and a maximum specificity of 95.3%. The model with the highest accuracy was 94.4% (with 65.6% receiving classification). When the models were used simultaneously the conclusive classification increased to 72.7% with an accuracy 95.9%. The predictive ability of the biomarkers random forest produced similar test characteristics when using 11 predictive markers. Conclusion We have demonstrated a pool of biomarkers from divergent biological pathways that can be used to classify individuals with potential early pregnancy loss. The biomarkers CG-Alpha, PAPPA and PSG3 can be used to predict viability and sFLT, TPFI2 and PSG3 can be used to predict pregnancy location.