Sparse feature selection for classification and prediction of metastasis in endometrial cancer

Mehmet Eren Ahsen, Todd P. Boren, Nitin K. Singh, Burook Misganaw, David G. Mutch, Kathleen N. Moore, Floor J. Backes, Carolyn K. McCourt, Jayanthi S. Lea, David S. Miller, Michael A. White, Mathukumalli Vidyasagar

Research output: Contribution to journalArticle

12 Citations (Scopus)

Abstract

Background: Metastasis via pelvic and/or para-aortic lymph nodes is a major risk factor for endometrial cancer. Lymph-node resection ameliorates risk but is associated with significant co-morbidities. Incidence in patients with stage I disease is 4-22% but no mechanism exists to accurately predict it. Therefore, national guidelines for primary staging surgery include pelvic and para-aortic lymph node dissection for all patients whose tumor exceeds 2cm in diameter. We sought to identify a robust molecular signature that can accurately classify risk of lymph node metastasis in endometrial cancer patients. 86 tumors matched for age and race, and evenly distributed between lymph node-positive and lymph node-negative cases, were selected as a training cohort. Genomic micro-RNA expression was profiled for each sample to serve as the predictive feature matrix. An independent set of 28 tumor samples was collected and similarly characterized to serve as a test cohort. Results: A feature selection algorithm was designed for applications where the number of samples is far smaller than the number of measured features per sample. A predictive miRNA expression signature was developed using this algorithm, which was then used to predict the metastatic status of the independent test cohort. A weighted classifier, using 18 micro-RNAs, achieved 100% accuracy on the training cohort. When applied to the testing cohort, the classifier correctly predicted 90% of node-positive cases, and 80% of node-negative cases (FDR = 6.25%). Conclusion: Results indicate that the evaluation of the quantitative sparse-feature classifier proposed here in clinical trials may lead to significant improvement in the prediction of lymphatic metastases in endometrial cancer patients.

Original languageEnglish (US)
Article number233
JournalBMC Genomics
Volume18
DOIs
StatePublished - Mar 27 2017

Fingerprint

Endometrial Neoplasms
Lymph Nodes
Neoplasm Metastasis
MicroRNAs
Lymphatic Metastasis
Neoplasms
Lymph Node Excision
Clinical Trials
Guidelines
Morbidity
Incidence

Keywords

  • Endometrial cancer
  • Lymph node metastasis
  • Machine learning
  • Sparse classification

ASJC Scopus subject areas

  • Biotechnology
  • Genetics

Cite this

Ahsen, M. E., Boren, T. P., Singh, N. K., Misganaw, B., Mutch, D. G., Moore, K. N., ... Vidyasagar, M. (2017). Sparse feature selection for classification and prediction of metastasis in endometrial cancer. BMC Genomics, 18, [233]. https://doi.org/10.1186/s12864-017-3604-y

Sparse feature selection for classification and prediction of metastasis in endometrial cancer. / Ahsen, Mehmet Eren; Boren, Todd P.; Singh, Nitin K.; Misganaw, Burook; Mutch, David G.; Moore, Kathleen N.; Backes, Floor J.; McCourt, Carolyn K.; Lea, Jayanthi S.; Miller, David S.; White, Michael A.; Vidyasagar, Mathukumalli.

In: BMC Genomics, Vol. 18, 233, 27.03.2017.

Research output: Contribution to journalArticle

Ahsen, ME, Boren, TP, Singh, NK, Misganaw, B, Mutch, DG, Moore, KN, Backes, FJ, McCourt, CK, Lea, JS, Miller, DS, White, MA & Vidyasagar, M 2017, 'Sparse feature selection for classification and prediction of metastasis in endometrial cancer', BMC Genomics, vol. 18, 233. https://doi.org/10.1186/s12864-017-3604-y
Ahsen, Mehmet Eren ; Boren, Todd P. ; Singh, Nitin K. ; Misganaw, Burook ; Mutch, David G. ; Moore, Kathleen N. ; Backes, Floor J. ; McCourt, Carolyn K. ; Lea, Jayanthi S. ; Miller, David S. ; White, Michael A. ; Vidyasagar, Mathukumalli. / Sparse feature selection for classification and prediction of metastasis in endometrial cancer. In: BMC Genomics. 2017 ; Vol. 18.
@article{390584a5bbdf4f39adc1e9fc2dfde7aa,
title = "Sparse feature selection for classification and prediction of metastasis in endometrial cancer",
abstract = "Background: Metastasis via pelvic and/or para-aortic lymph nodes is a major risk factor for endometrial cancer. Lymph-node resection ameliorates risk but is associated with significant co-morbidities. Incidence in patients with stage I disease is 4-22{\%} but no mechanism exists to accurately predict it. Therefore, national guidelines for primary staging surgery include pelvic and para-aortic lymph node dissection for all patients whose tumor exceeds 2cm in diameter. We sought to identify a robust molecular signature that can accurately classify risk of lymph node metastasis in endometrial cancer patients. 86 tumors matched for age and race, and evenly distributed between lymph node-positive and lymph node-negative cases, were selected as a training cohort. Genomic micro-RNA expression was profiled for each sample to serve as the predictive feature matrix. An independent set of 28 tumor samples was collected and similarly characterized to serve as a test cohort. Results: A feature selection algorithm was designed for applications where the number of samples is far smaller than the number of measured features per sample. A predictive miRNA expression signature was developed using this algorithm, which was then used to predict the metastatic status of the independent test cohort. A weighted classifier, using 18 micro-RNAs, achieved 100{\%} accuracy on the training cohort. When applied to the testing cohort, the classifier correctly predicted 90{\%} of node-positive cases, and 80{\%} of node-negative cases (FDR = 6.25{\%}). Conclusion: Results indicate that the evaluation of the quantitative sparse-feature classifier proposed here in clinical trials may lead to significant improvement in the prediction of lymphatic metastases in endometrial cancer patients.",
keywords = "Endometrial cancer, Lymph node metastasis, Machine learning, Sparse classification",
author = "Ahsen, {Mehmet Eren} and Boren, {Todd P.} and Singh, {Nitin K.} and Burook Misganaw and Mutch, {David G.} and Moore, {Kathleen N.} and Backes, {Floor J.} and McCourt, {Carolyn K.} and Lea, {Jayanthi S.} and Miller, {David S.} and White, {Michael A.} and Mathukumalli Vidyasagar",
year = "2017",
month = "3",
day = "27",
doi = "10.1186/s12864-017-3604-y",
language = "English (US)",
volume = "18",
journal = "BMC Genomics",
issn = "1471-2164",
publisher = "BioMed Central",

}

TY - JOUR

T1 - Sparse feature selection for classification and prediction of metastasis in endometrial cancer

AU - Ahsen, Mehmet Eren

AU - Boren, Todd P.

AU - Singh, Nitin K.

AU - Misganaw, Burook

AU - Mutch, David G.

AU - Moore, Kathleen N.

AU - Backes, Floor J.

AU - McCourt, Carolyn K.

AU - Lea, Jayanthi S.

AU - Miller, David S.

AU - White, Michael A.

AU - Vidyasagar, Mathukumalli

PY - 2017/3/27

Y1 - 2017/3/27

N2 - Background: Metastasis via pelvic and/or para-aortic lymph nodes is a major risk factor for endometrial cancer. Lymph-node resection ameliorates risk but is associated with significant co-morbidities. Incidence in patients with stage I disease is 4-22% but no mechanism exists to accurately predict it. Therefore, national guidelines for primary staging surgery include pelvic and para-aortic lymph node dissection for all patients whose tumor exceeds 2cm in diameter. We sought to identify a robust molecular signature that can accurately classify risk of lymph node metastasis in endometrial cancer patients. 86 tumors matched for age and race, and evenly distributed between lymph node-positive and lymph node-negative cases, were selected as a training cohort. Genomic micro-RNA expression was profiled for each sample to serve as the predictive feature matrix. An independent set of 28 tumor samples was collected and similarly characterized to serve as a test cohort. Results: A feature selection algorithm was designed for applications where the number of samples is far smaller than the number of measured features per sample. A predictive miRNA expression signature was developed using this algorithm, which was then used to predict the metastatic status of the independent test cohort. A weighted classifier, using 18 micro-RNAs, achieved 100% accuracy on the training cohort. When applied to the testing cohort, the classifier correctly predicted 90% of node-positive cases, and 80% of node-negative cases (FDR = 6.25%). Conclusion: Results indicate that the evaluation of the quantitative sparse-feature classifier proposed here in clinical trials may lead to significant improvement in the prediction of lymphatic metastases in endometrial cancer patients.

AB - Background: Metastasis via pelvic and/or para-aortic lymph nodes is a major risk factor for endometrial cancer. Lymph-node resection ameliorates risk but is associated with significant co-morbidities. Incidence in patients with stage I disease is 4-22% but no mechanism exists to accurately predict it. Therefore, national guidelines for primary staging surgery include pelvic and para-aortic lymph node dissection for all patients whose tumor exceeds 2cm in diameter. We sought to identify a robust molecular signature that can accurately classify risk of lymph node metastasis in endometrial cancer patients. 86 tumors matched for age and race, and evenly distributed between lymph node-positive and lymph node-negative cases, were selected as a training cohort. Genomic micro-RNA expression was profiled for each sample to serve as the predictive feature matrix. An independent set of 28 tumor samples was collected and similarly characterized to serve as a test cohort. Results: A feature selection algorithm was designed for applications where the number of samples is far smaller than the number of measured features per sample. A predictive miRNA expression signature was developed using this algorithm, which was then used to predict the metastatic status of the independent test cohort. A weighted classifier, using 18 micro-RNAs, achieved 100% accuracy on the training cohort. When applied to the testing cohort, the classifier correctly predicted 90% of node-positive cases, and 80% of node-negative cases (FDR = 6.25%). Conclusion: Results indicate that the evaluation of the quantitative sparse-feature classifier proposed here in clinical trials may lead to significant improvement in the prediction of lymphatic metastases in endometrial cancer patients.

KW - Endometrial cancer

KW - Lymph node metastasis

KW - Machine learning

KW - Sparse classification

UR - http://www.scopus.com/inward/record.url?scp=85016148200&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85016148200&partnerID=8YFLogxK

U2 - 10.1186/s12864-017-3604-y

DO - 10.1186/s12864-017-3604-y

M3 - Article

C2 - 28361685

AN - SCOPUS:85016148200

VL - 18

JO - BMC Genomics

JF - BMC Genomics

SN - 1471-2164

M1 - 233

ER -