A Novel Classifier Based on Urinary Proteomics for Distinguishing Between Benign and Malignant Ovarian Tumors.
machine learning
mass spectrometry
non-invasive biomarkers
ovarian cancer
urinary proteomics
Journal
Frontiers in cell and developmental biology
ISSN: 2296-634X
Titre abrégé: Front Cell Dev Biol
Pays: Switzerland
ID NLM: 101630250
Informations de publication
Date de publication:
2021
2021
Historique:
received:
20
05
2021
accepted:
09
08
2021
entrez:
16
9
2021
pubmed:
17
9
2021
medline:
17
9
2021
Statut:
epublish
Résumé
Preoperative differentiation of benign and malignant tumor types is critical for providing individualized treatment interventions to improve prognosis of patients with ovarian cancer. High-throughput proteomics analysis of urine samples was performed to identify reliable and non-invasive biomarkers that could effectively discriminate between the two ovarian tumor types. In total, 132 urine samples from 73 malignant and 59 benign cases of ovarian carcinoma were divided into C1 (training and test datasets) and C2 (validation dataset) cohorts. Mass spectrometry (MS) data of all samples were acquired in data-independent acquisition (DIA) mode with an Orbitrap mass spectrometer and analyzed using DIA-NN software. The generated classifier was trained with Random Forest algorithm from the training dataset and validated in the test and validation datasets. Serum CA125 and HE4 levels were additionally determined in all patients. Finally, classification accuracy of the classifier, serum CA125 and serum HE4 in all samples were evaluated and plotted via receiver operating characteristic (ROC) analysis. In total, 2,199 proteins were quantified and 69 identified with differential expression in benign and malignant groups of the C1 cohort. A classifier incorporating five proteins (WFDC2, PTMA, PVRL4, FIBA, and PVRL2) was trained and validated in this study. Evaluation of the performance of the classifier revealed AUC values of 0.970 and 0.952 in the test and validation datasets, respectively. In all 132 patients, AUCs of 0.966, 0.947, and 0.979 were achieved with the classifier, serum CA125, and serum HE4, respectively. Among eight patients with early stage malignancy, 7, 6, and 4 were accurately diagnosed based on classifier, serum CA125, and serum HE4, respectively. The novel classifier incorporating a urinary protein panel presents a promising non-invasive diagnostic biomarker for classifying benign and malignant ovarian tumors.
Sections du résumé
BACKGROUND
BACKGROUND
Preoperative differentiation of benign and malignant tumor types is critical for providing individualized treatment interventions to improve prognosis of patients with ovarian cancer. High-throughput proteomics analysis of urine samples was performed to identify reliable and non-invasive biomarkers that could effectively discriminate between the two ovarian tumor types.
METHODS
METHODS
In total, 132 urine samples from 73 malignant and 59 benign cases of ovarian carcinoma were divided into C1 (training and test datasets) and C2 (validation dataset) cohorts. Mass spectrometry (MS) data of all samples were acquired in data-independent acquisition (DIA) mode with an Orbitrap mass spectrometer and analyzed using DIA-NN software. The generated classifier was trained with Random Forest algorithm from the training dataset and validated in the test and validation datasets. Serum CA125 and HE4 levels were additionally determined in all patients. Finally, classification accuracy of the classifier, serum CA125 and serum HE4 in all samples were evaluated and plotted via receiver operating characteristic (ROC) analysis.
RESULTS
RESULTS
In total, 2,199 proteins were quantified and 69 identified with differential expression in benign and malignant groups of the C1 cohort. A classifier incorporating five proteins (WFDC2, PTMA, PVRL4, FIBA, and PVRL2) was trained and validated in this study. Evaluation of the performance of the classifier revealed AUC values of 0.970 and 0.952 in the test and validation datasets, respectively. In all 132 patients, AUCs of 0.966, 0.947, and 0.979 were achieved with the classifier, serum CA125, and serum HE4, respectively. Among eight patients with early stage malignancy, 7, 6, and 4 were accurately diagnosed based on classifier, serum CA125, and serum HE4, respectively.
CONCLUSION
CONCLUSIONS
The novel classifier incorporating a urinary protein panel presents a promising non-invasive diagnostic biomarker for classifying benign and malignant ovarian tumors.
Identifiants
pubmed: 34527671
doi: 10.3389/fcell.2021.712196
pmc: PMC8437375
doi:
Types de publication
Journal Article
Langues
eng
Pagination
712196Informations de copyright
Copyright © 2021 Ni, Zhou, Zhu, Yuan, Gong, Zhu, Zheng and Zhao.
Déclaration de conflit d'intérêts
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Références
Clin Proteomics. 2019 Apr 5;16:12
pubmed: 30988666
Nan Fang Yi Ke Da Xue Xue Bao. 2019 Dec 30;39(12):1393-1401
pubmed: 31907150
Curr Med Chem. 2017;24(17):1747-1760
pubmed: 28521686
OMICS. 2013 Dec;17(12):595-610
pubmed: 24116388
JAMA. 2004 Jun 9;291(22):2705-12
pubmed: 15187051
Cancer Immunol Res. 2019 Feb;7(2):257-268
pubmed: 30659054
CA Cancer J Clin. 2021 Jan;71(1):7-33
pubmed: 33433946
J Ovarian Res. 2019 Mar 27;12(1):28
pubmed: 30917847
CA Cancer J Clin. 2019 Jul;69(4):280-304
pubmed: 31099893
Mol Cell Proteomics. 2020 Jan;19(1):114-127
pubmed: 31243064
Clin Proteomics. 2020 Mar 12;17:10
pubmed: 32190014
Asian Pac J Cancer Prev. 2012;13(9):4695-8
pubmed: 23167404
J Ovarian Res. 2019 Sep 16;12(1):87
pubmed: 31526390
Mol Cell Proteomics. 2019 Dec;18(12):2492-2505
pubmed: 31585987
Proteomics Clin Appl. 2015 Apr;9(3-4):307-21
pubmed: 25504613
Int J Gynecol Cancer. 2016 Jan;26(1):43-51
pubmed: 26588231
Oncotarget. 2017 Feb 7;8(6):9660-9671
pubmed: 28039447
Discov Med. 2018 Feb;25(136):57-65
pubmed: 29579412
J Proteome Res. 2015 Nov 6;14(11):4752-62
pubmed: 26423119
Cancer Res. 2003 Jul 1;63(13):3695-700
pubmed: 12839961
J Vis Exp. 2021 May 8;(171):
pubmed: 34028441
Proteomics Clin Appl. 2016 Apr;10(4):358-70
pubmed: 26703953
Int J Clin Exp Pathol. 2014 Jul 15;7(8):4867-76
pubmed: 25197357
Clin Proteomics. 2018 Apr 30;15:18
pubmed: 29719494
Arch Gynecol Obstet. 2014 Mar;289(3):491-8
pubmed: 24253338
Front Public Health. 2021 Jul 02;9:678856
pubmed: 34277544
Gynecol Oncol. 2015 Jun;137(3):430-5
pubmed: 25866324
Mol Med Rep. 2018 Oct;18(4):3727-3736
pubmed: 30106150
Biomark Cancer. 2019 Feb 26;11:1179299X19830977
pubmed: 30833816
Nat Methods. 2020 Jan;17(1):41-44
pubmed: 31768060
Clin Dev Immunol. 2010;2010:791603
pubmed: 21318181
Rapid Commun Mass Spectrom. 2017 Jan 30;31(2):171-178
pubmed: 27794190
Am J Obstet Gynecol. 2017 Dec;217(6):652-660
pubmed: 28735703
Chin Med J (Engl). 2020 Jun 5;133(11):1285-1291
pubmed: 32404690
Nat Methods. 2009 May;6(5):359-62
pubmed: 19377485
Expert Rev Proteomics. 2015;12(6):623-36
pubmed: 26472227
Front Oncol. 2020 Nov 03;10:559016
pubmed: 33224875
Nat Commun. 2019 Apr 3;10(1):1523
pubmed: 30944313
Mol Cell Proteomics. 2021 Feb 24;20:100021
pubmed: 33288685
Gynecol Obstet Invest. 2019;84(6):591-598
pubmed: 31311023
Oncotarget. 2017 Aug 24;8(43):74806-74819
pubmed: 29088825
Cancers (Basel). 2019 May 20;11(5):
pubmed: 31137558
Chin Neurosurg J. 2020 Apr 14;6:11
pubmed: 32922940
J Proteome Res. 2014 Apr 4;13(4):1885-95
pubmed: 24552128
Oncotarget. 2017 Feb 7;8(6):9717-9738
pubmed: 28038455
Cell. 2020 Jul 9;182(1):59-72.e15
pubmed: 32492406
Obstet Gynecol. 2016 May;127(5):848-858
pubmed: 27054927
Biochem Cell Biol. 2013 Aug;91(4):265-9
pubmed: 23859021
Hepatobiliary Pancreat Dis Int. 2015 Apr;14(2):171-7
pubmed: 25865690
J Proteomics. 2020 Aug 15;225:103780
pubmed: 32298775