Limits...
Peptidomic Identification of Serum Peptides Diagnosing Preeclampsia.

Wen Q, Liu LY, Yang T, Alev C, Wu S, Stevenson DK, Sheng G, Butte AJ, Ling XB - PLoS ONE (2013)

Bottom Line: We concluded that serum peptides can accurately discriminate active PE.Measurement of a 19-peptide panel could be performed quickly and in a quantitative mass spectrometric platform available in clinical laboratories.This serum peptide panel quantification could provide clinical utility in predicting PE or differential diagnosis of PE from confounding chronic hypertension.

View Article: PubMed Central - PubMed

Affiliation: Department of Surgery, Stanford University, Stanford, California, United States of America.

ABSTRACT
We sought to identify serological markers capable of diagnosing preeclampsia (PE). We performed serum peptide analysis (liquid chromatography mass spectrometry) of 62 unique samples from 31 PE patients and 31 healthy pregnant controls, with two-thirds used as a training set and the other third as a testing set. Differential serum peptide profiling identified 52 significant serum peptides, and a 19-peptide panel collectively discriminating PE in training sets (n = 21 PE, n = 21 control; specificity = 85.7% and sensitivity = 100%) and testing sets (n = 10 PE, n = 10 control; specificity = 80% and sensitivity = 100%). The panel peptides were derived from 6 different protein precursors: 13 from fibrinogen alpha (FGA), 1 from alpha-1-antitrypsin (A1AT), 1 from apolipoprotein L1 (APO-L1), 1 from inter-alpha-trypsin inhibitor heavy chain H4 (ITIH4), 2 from kininogen-1 (KNG1), and 1 from thymosin beta-4 (TMSB4). We concluded that serum peptides can accurately discriminate active PE. Measurement of a 19-peptide panel could be performed quickly and in a quantitative mass spectrometric platform available in clinical laboratories. This serum peptide panel quantification could provide clinical utility in predicting PE or differential diagnosis of PE from confounding chronic hypertension.

No MeSH data available.


Related in: MedlinePlus

Diagnosis of PE from control with serum biomarkers.Left panel: estimated PE scores were computed from the PE serum peptide panel PAM model as a function of the gestational weeks; right panel: the log sFlt-1/PIGF serum concentration ratio was plotted as a function of the gestational weeks. Red indicates known PE cases; green indicates known healthy pregnancy controls. For either PE or control sample category, a loess curve was fitted to represent the overall trend of biomarker scoring as a function of gestational age.
© Copyright Policy
Related In: Results  -  Collection


getmorefigures.php?uid=PMC3686758&req=5

pone-0065571-g004: Diagnosis of PE from control with serum biomarkers.Left panel: estimated PE scores were computed from the PE serum peptide panel PAM model as a function of the gestational weeks; right panel: the log sFlt-1/PIGF serum concentration ratio was plotted as a function of the gestational weeks. Red indicates known PE cases; green indicates known healthy pregnancy controls. For either PE or control sample category, a loess curve was fitted to represent the overall trend of biomarker scoring as a function of gestational age.

Mentions: With the selected biomarker panel and trained PAM prediction model, the PE prediction performance was analyzed as in Figure 3. The left panel of Figure 3 shows the prediction performance on the training set (n = 42), while the right panel of Figure 3 shows the prediction performance on the blind testing set (n = 20). On the training set, all PE samples (n = 21) were predicted correctly, while 3 of the 21 (14.3%) control samples were false positive. Thus, the sensitivity on the training set was 85.7% and the specificity was 100%, resulting in the overall prediction accuracy of 92.9%. Similarly, on the testing set, the overall prediction accuracy is 90%, with sensitivity 80% and specificity 100%. The scatter plot of the PAM predicted scores along with gestational ages is shown as in Figure 4. The predicted score represents the probability of being PE according to the PAM prediction model. Both the prediction accuracy and the scatter plot show that the selected biomarker panel with 19 peptides can be used to effectively predict the occurrence of PE. The early and late gestational age discriminative analyses demonstrated a comparable performance, indicating the potential usefulness of our serum peptide panel in the early diagnosis of PE. The sFlt-1/PIGF ratio's PE assessment utility, previously through the multicenter trial validation [23], was confirmed in this study and used as a benchmark for our newly derived biomarker panels. As shown in Figure 4, the PE diagnostic performance of our peptide panel was comparable to the sFlt-1/PIGF ratio. If we use 0.66, rather than 0.5, as the cutoff of our PE classification panel, as the dotted line in Figure 4, there is only 1 misclassified sample. In contrast with it, the sFlt-1/PIGF ratio results to at least 4 misclassified samples.


Peptidomic Identification of Serum Peptides Diagnosing Preeclampsia.

Wen Q, Liu LY, Yang T, Alev C, Wu S, Stevenson DK, Sheng G, Butte AJ, Ling XB - PLoS ONE (2013)

Diagnosis of PE from control with serum biomarkers.Left panel: estimated PE scores were computed from the PE serum peptide panel PAM model as a function of the gestational weeks; right panel: the log sFlt-1/PIGF serum concentration ratio was plotted as a function of the gestational weeks. Red indicates known PE cases; green indicates known healthy pregnancy controls. For either PE or control sample category, a loess curve was fitted to represent the overall trend of biomarker scoring as a function of gestational age.
© Copyright Policy
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC3686758&req=5

pone-0065571-g004: Diagnosis of PE from control with serum biomarkers.Left panel: estimated PE scores were computed from the PE serum peptide panel PAM model as a function of the gestational weeks; right panel: the log sFlt-1/PIGF serum concentration ratio was plotted as a function of the gestational weeks. Red indicates known PE cases; green indicates known healthy pregnancy controls. For either PE or control sample category, a loess curve was fitted to represent the overall trend of biomarker scoring as a function of gestational age.
Mentions: With the selected biomarker panel and trained PAM prediction model, the PE prediction performance was analyzed as in Figure 3. The left panel of Figure 3 shows the prediction performance on the training set (n = 42), while the right panel of Figure 3 shows the prediction performance on the blind testing set (n = 20). On the training set, all PE samples (n = 21) were predicted correctly, while 3 of the 21 (14.3%) control samples were false positive. Thus, the sensitivity on the training set was 85.7% and the specificity was 100%, resulting in the overall prediction accuracy of 92.9%. Similarly, on the testing set, the overall prediction accuracy is 90%, with sensitivity 80% and specificity 100%. The scatter plot of the PAM predicted scores along with gestational ages is shown as in Figure 4. The predicted score represents the probability of being PE according to the PAM prediction model. Both the prediction accuracy and the scatter plot show that the selected biomarker panel with 19 peptides can be used to effectively predict the occurrence of PE. The early and late gestational age discriminative analyses demonstrated a comparable performance, indicating the potential usefulness of our serum peptide panel in the early diagnosis of PE. The sFlt-1/PIGF ratio's PE assessment utility, previously through the multicenter trial validation [23], was confirmed in this study and used as a benchmark for our newly derived biomarker panels. As shown in Figure 4, the PE diagnostic performance of our peptide panel was comparable to the sFlt-1/PIGF ratio. If we use 0.66, rather than 0.5, as the cutoff of our PE classification panel, as the dotted line in Figure 4, there is only 1 misclassified sample. In contrast with it, the sFlt-1/PIGF ratio results to at least 4 misclassified samples.

Bottom Line: We concluded that serum peptides can accurately discriminate active PE.Measurement of a 19-peptide panel could be performed quickly and in a quantitative mass spectrometric platform available in clinical laboratories.This serum peptide panel quantification could provide clinical utility in predicting PE or differential diagnosis of PE from confounding chronic hypertension.

View Article: PubMed Central - PubMed

Affiliation: Department of Surgery, Stanford University, Stanford, California, United States of America.

ABSTRACT
We sought to identify serological markers capable of diagnosing preeclampsia (PE). We performed serum peptide analysis (liquid chromatography mass spectrometry) of 62 unique samples from 31 PE patients and 31 healthy pregnant controls, with two-thirds used as a training set and the other third as a testing set. Differential serum peptide profiling identified 52 significant serum peptides, and a 19-peptide panel collectively discriminating PE in training sets (n = 21 PE, n = 21 control; specificity = 85.7% and sensitivity = 100%) and testing sets (n = 10 PE, n = 10 control; specificity = 80% and sensitivity = 100%). The panel peptides were derived from 6 different protein precursors: 13 from fibrinogen alpha (FGA), 1 from alpha-1-antitrypsin (A1AT), 1 from apolipoprotein L1 (APO-L1), 1 from inter-alpha-trypsin inhibitor heavy chain H4 (ITIH4), 2 from kininogen-1 (KNG1), and 1 from thymosin beta-4 (TMSB4). We concluded that serum peptides can accurately discriminate active PE. Measurement of a 19-peptide panel could be performed quickly and in a quantitative mass spectrometric platform available in clinical laboratories. This serum peptide panel quantification could provide clinical utility in predicting PE or differential diagnosis of PE from confounding chronic hypertension.

No MeSH data available.


Related in: MedlinePlus