Objective: To evaluate the accuracy of concentrations of gingival crevicular fluid (GCF) placental alkaline phosphatase (PLAP) during early pregnancy in identifying women at risk of subsequently developing preeclampsia (PE). Design: Prospective cohort study. Setting: Hospital Sotero del Río, Santiago, Chile. Population: Pregnant women recruited at 11-14 weeks of gestation. Methods: Maternal obstetric and periodontal histories were obtained. GCF samples were collected for PLAP determination by ELISA assay. Multiple logistic regression models estimated the association between GCF-PLAP concentrations, maternal variables, and PE development. The accuracy performance of the prediction model was established. Results: 460 women were recruited into the study, and 412 completed their pregnancy follow–up visit. 18 (4.3%) women developed PE. GCF-PLAP concentrations and systolic blood pressure measurements were significantly higher in women who developed PE (p=0.015 and p<0.001, respectively). An association between first-trimester systolic blood pressure, GCF-PLAP, and PE were established. The predictive model had a sensitivity of 83%, specificity of 72%, a positive predictive value (PPV) of 12%, and a negative predictive value (NPV) of 99%. The positive and negative likelihood ratios were 2.9 and 0.3, respectively, thus classifying correctly 72% of women who subsequently developed PE. The area under the receiver operating characteristic curve was 0.77 for PE and 0.85 for preterm PE. Conclusions: An algorithm that includes PLAP concentrations in GCF and blood pressure during early pregnancy may aid in the identification of women at risk of developing PE. Funding: FONDEF IDeA: ID16I10452. NICHD/NIH/DHHS: HHSN275201300006C. Keywords: a cohort study, gestation, hypertension, placental biomarkers, risk prediction model.