Optimized prediction of extreme treatment outcomes in ovarian cancer

Burook Misganaw, Eren Ahsen, Nitin Singh, Keith A. Baggerly, Anna Unruh, Michael A. White, M. Vidyasagar

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

Ovarian cancer is the fifth leading cause of death among female cancers. Front-line therapy for ovarian cancer is platinum-based chemotherapy. However, the response of patients is highly nonuniform. The TCGA database of serous ovarian carcinomas shows that ∼10% of patients respond poorly to platinum-based chemotherapy, with tumors relapsing in seven months or less. Another 10% or so enjoy disease-free survival of three years or more. The objective of the present research is to identify a small number of highly predictive biomarkers that can distinguish between the two extreme responders and then extrapolate to all patients. This is achieved using the lone star algorithm that is specifically developed for biological applications. Using this algorithm, we are able to identify biomarker panels of 25 genes (of 12,000 genes) that can be used to classify patients into one of the three groups: super respond-ers, medium responders, and nonresponders. We are also able to determine a discriminant function that can divide the entire patient population into two classes, such that one group has a clear survival advantage over the other. These biomarkers are developed using the TCGA Agilent platform data and cross-validated on the TCGA Affymetrix platform data, as well as entirely independent data from Tothill et al. The P-values on the training data are extremely small, sometimes below machine zero, while the P-values on cross-validation are well below the widely accepted threshold of 0.05.

Original languageEnglish (US)
Pages (from-to)45-55
Number of pages11
JournalCancer Informatics
Volume15
DOIs
StatePublished - Mar 21 2016

Fingerprint

Ovarian Neoplasms
Biomarkers
Platinum
Drug Therapy
Genes
Disease-Free Survival
Cause of Death
Neoplasms
Databases
Carcinoma
Survival
Research
Population
Therapeutics

Keywords

  • Ovarian cancer
  • Platinum chemotherapy
  • Prediction of patient response

ASJC Scopus subject areas

  • Cancer Research
  • Oncology

Cite this

Misganaw, B., Ahsen, E., Singh, N., Baggerly, K. A., Unruh, A., White, M. A., & Vidyasagar, M. (2016). Optimized prediction of extreme treatment outcomes in ovarian cancer. Cancer Informatics, 15, 45-55. https://doi.org/10.4137/CIN.S30803

Optimized prediction of extreme treatment outcomes in ovarian cancer. / Misganaw, Burook; Ahsen, Eren; Singh, Nitin; Baggerly, Keith A.; Unruh, Anna; White, Michael A.; Vidyasagar, M.

In: Cancer Informatics, Vol. 15, 21.03.2016, p. 45-55.

Research output: Contribution to journalArticle

Misganaw, B, Ahsen, E, Singh, N, Baggerly, KA, Unruh, A, White, MA & Vidyasagar, M 2016, 'Optimized prediction of extreme treatment outcomes in ovarian cancer', Cancer Informatics, vol. 15, pp. 45-55. https://doi.org/10.4137/CIN.S30803
Misganaw, Burook ; Ahsen, Eren ; Singh, Nitin ; Baggerly, Keith A. ; Unruh, Anna ; White, Michael A. ; Vidyasagar, M. / Optimized prediction of extreme treatment outcomes in ovarian cancer. In: Cancer Informatics. 2016 ; Vol. 15. pp. 45-55.
@article{fe48345f235a4dcf8b526315507b1f32,
title = "Optimized prediction of extreme treatment outcomes in ovarian cancer",
abstract = "Ovarian cancer is the fifth leading cause of death among female cancers. Front-line therapy for ovarian cancer is platinum-based chemotherapy. However, the response of patients is highly nonuniform. The TCGA database of serous ovarian carcinomas shows that ∼10{\%} of patients respond poorly to platinum-based chemotherapy, with tumors relapsing in seven months or less. Another 10{\%} or so enjoy disease-free survival of three years or more. The objective of the present research is to identify a small number of highly predictive biomarkers that can distinguish between the two extreme responders and then extrapolate to all patients. This is achieved using the lone star algorithm that is specifically developed for biological applications. Using this algorithm, we are able to identify biomarker panels of 25 genes (of 12,000 genes) that can be used to classify patients into one of the three groups: super respond-ers, medium responders, and nonresponders. We are also able to determine a discriminant function that can divide the entire patient population into two classes, such that one group has a clear survival advantage over the other. These biomarkers are developed using the TCGA Agilent platform data and cross-validated on the TCGA Affymetrix platform data, as well as entirely independent data from Tothill et al. The P-values on the training data are extremely small, sometimes below machine zero, while the P-values on cross-validation are well below the widely accepted threshold of 0.05.",
keywords = "Ovarian cancer, Platinum chemotherapy, Prediction of patient response",
author = "Burook Misganaw and Eren Ahsen and Nitin Singh and Baggerly, {Keith A.} and Anna Unruh and White, {Michael A.} and M. Vidyasagar",
year = "2016",
month = "3",
day = "21",
doi = "10.4137/CIN.S30803",
language = "English (US)",
volume = "15",
pages = "45--55",
journal = "Cancer Informatics",
issn = "1176-9351",
publisher = "Libertas Academica Ltd.",

}

TY - JOUR

T1 - Optimized prediction of extreme treatment outcomes in ovarian cancer

AU - Misganaw, Burook

AU - Ahsen, Eren

AU - Singh, Nitin

AU - Baggerly, Keith A.

AU - Unruh, Anna

AU - White, Michael A.

AU - Vidyasagar, M.

PY - 2016/3/21

Y1 - 2016/3/21

N2 - Ovarian cancer is the fifth leading cause of death among female cancers. Front-line therapy for ovarian cancer is platinum-based chemotherapy. However, the response of patients is highly nonuniform. The TCGA database of serous ovarian carcinomas shows that ∼10% of patients respond poorly to platinum-based chemotherapy, with tumors relapsing in seven months or less. Another 10% or so enjoy disease-free survival of three years or more. The objective of the present research is to identify a small number of highly predictive biomarkers that can distinguish between the two extreme responders and then extrapolate to all patients. This is achieved using the lone star algorithm that is specifically developed for biological applications. Using this algorithm, we are able to identify biomarker panels of 25 genes (of 12,000 genes) that can be used to classify patients into one of the three groups: super respond-ers, medium responders, and nonresponders. We are also able to determine a discriminant function that can divide the entire patient population into two classes, such that one group has a clear survival advantage over the other. These biomarkers are developed using the TCGA Agilent platform data and cross-validated on the TCGA Affymetrix platform data, as well as entirely independent data from Tothill et al. The P-values on the training data are extremely small, sometimes below machine zero, while the P-values on cross-validation are well below the widely accepted threshold of 0.05.

AB - Ovarian cancer is the fifth leading cause of death among female cancers. Front-line therapy for ovarian cancer is platinum-based chemotherapy. However, the response of patients is highly nonuniform. The TCGA database of serous ovarian carcinomas shows that ∼10% of patients respond poorly to platinum-based chemotherapy, with tumors relapsing in seven months or less. Another 10% or so enjoy disease-free survival of three years or more. The objective of the present research is to identify a small number of highly predictive biomarkers that can distinguish between the two extreme responders and then extrapolate to all patients. This is achieved using the lone star algorithm that is specifically developed for biological applications. Using this algorithm, we are able to identify biomarker panels of 25 genes (of 12,000 genes) that can be used to classify patients into one of the three groups: super respond-ers, medium responders, and nonresponders. We are also able to determine a discriminant function that can divide the entire patient population into two classes, such that one group has a clear survival advantage over the other. These biomarkers are developed using the TCGA Agilent platform data and cross-validated on the TCGA Affymetrix platform data, as well as entirely independent data from Tothill et al. The P-values on the training data are extremely small, sometimes below machine zero, while the P-values on cross-validation are well below the widely accepted threshold of 0.05.

KW - Ovarian cancer

KW - Platinum chemotherapy

KW - Prediction of patient response

UR - http://www.scopus.com/inward/record.url?scp=84963525792&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84963525792&partnerID=8YFLogxK

U2 - 10.4137/CIN.S30803

DO - 10.4137/CIN.S30803

M3 - Article

C2 - 27034613

AN - SCOPUS:84963525792

VL - 15

SP - 45

EP - 55

JO - Cancer Informatics

JF - Cancer Informatics

SN - 1176-9351

ER -