Computational approaches to support comparative analysis of multiparametric tests: Modelling versus Training

Research output: Contribution to journalArticlepeer-review

Standard

Computational approaches to support comparative analysis of multiparametric tests : Modelling versus Training. / OPTIMA Trial Management Group.

In: PLoS ONE, Vol. 15, No. 9 September, e0238593, 03.09.2020.

Research output: Contribution to journalArticlepeer-review

Harvard

APA

Vancouver

Author

Bibtex

@article{9d7c95451db042558fe5c77d5bdd7695,
title = "Computational approaches to support comparative analysis of multiparametric tests: Modelling versus Training",
abstract = "Multiparametric assays for risk stratification are widely used in the management of breast cancer, with applications being developed for a number of other cancer settings. Recent data from multiple sources suggests that different tests may provide different risk estimates at the individual patient level. There is an increasing need for robust methods to support cost effective comparisons of test performance in multiple settings. The derivation of similar risk classifications using genes comprising the following multi-parametric tests Oncotype DX{\textregistered} (Genomic Health.), Prosigna{\texttrademark} (NanoString Technologies, Inc.), MammaPrint{\textregistered} (Agendia Inc.) was performed using different computational approaches. Results were compared to the actual test results. Two widely used approaches were applied, firstly computational “modelling” of test results using published algorithms and secondly a “training” approach which used reference results from the commercially supplied tests. We demonstrate the potential for errors to arise when using a “modelling” approach without reference to real world test results. Simultaneously we show that a “training” approach can provide a highly cost-effective solution to the development of real-world comparisons between different multigene signatures. Comparisons between existing multiparametric tests is challenging, and evidence on discordance between tests in risk stratification presents further dilemmas. We present an approach, modelled in breast cancer, which can provide health care providers and researchers with the potential to perform robust and meaningful comparisons between multigene tests in a cost-effective manner. We demonstrate that whilst viable estimates of gene signatures can be derived from modelling approaches, in our study using a training approach allowed a close approximation to true signature results.",
author = "{OPTIMA Trial Management Group} and Bartlett, {John M.S.} and Jane Bayani and Kornaga, {Elizabeth N.} and Patrick Danaher and Cheryl Crozier and Tammy Piper and Yao, {Cindy Q.} and Dunn, {Janet A.} and Boutros, {Paul C.} and Stein, {Robert C.} and John Bartlett and Cameron, {David A.} and Amy Campbell and Peter Canney and Jenny Donovan and Janet Dunn and Helena Earl and Mary Falzon and Adele Francis and Peter Hall and Victoria Harmer and Helen Higgins and Luke Hughes-Davies and Claire Hulme and Iain Macpherson and Andrea Marshall and Andreas Makris and Chris McCabe and Adrienne Morgan and Sarah Pinder and Chris Poole and Daniel Rea and Leila Rooshenas and Nigel Stallard and Stein, {Robert C.}",
year = "2020",
month = sep,
day = "3",
doi = "10.1371/journal.pone.0238593",
language = "English",
volume = "15",
journal = "PLoSONE",
issn = "1932-6203",
publisher = "Public Library of Science (PLOS)",
number = "9 September",

}

RIS

TY - JOUR

T1 - Computational approaches to support comparative analysis of multiparametric tests

T2 - Modelling versus Training

AU - OPTIMA Trial Management Group

AU - Bartlett, John M.S.

AU - Bayani, Jane

AU - Kornaga, Elizabeth N.

AU - Danaher, Patrick

AU - Crozier, Cheryl

AU - Piper, Tammy

AU - Yao, Cindy Q.

AU - Dunn, Janet A.

AU - Boutros, Paul C.

AU - Stein, Robert C.

AU - Bartlett, John

AU - Cameron, David A.

AU - Campbell, Amy

AU - Canney, Peter

AU - Donovan, Jenny

AU - Dunn, Janet

AU - Earl, Helena

AU - Falzon, Mary

AU - Francis, Adele

AU - Hall, Peter

AU - Harmer, Victoria

AU - Higgins, Helen

AU - Hughes-Davies, Luke

AU - Hulme, Claire

AU - Macpherson, Iain

AU - Marshall, Andrea

AU - Makris, Andreas

AU - McCabe, Chris

AU - Morgan, Adrienne

AU - Pinder, Sarah

AU - Poole, Chris

AU - Rea, Daniel

AU - Rooshenas, Leila

AU - Stallard, Nigel

AU - Stein, Robert C.

PY - 2020/9/3

Y1 - 2020/9/3

N2 - Multiparametric assays for risk stratification are widely used in the management of breast cancer, with applications being developed for a number of other cancer settings. Recent data from multiple sources suggests that different tests may provide different risk estimates at the individual patient level. There is an increasing need for robust methods to support cost effective comparisons of test performance in multiple settings. The derivation of similar risk classifications using genes comprising the following multi-parametric tests Oncotype DX® (Genomic Health.), Prosigna™ (NanoString Technologies, Inc.), MammaPrint® (Agendia Inc.) was performed using different computational approaches. Results were compared to the actual test results. Two widely used approaches were applied, firstly computational “modelling” of test results using published algorithms and secondly a “training” approach which used reference results from the commercially supplied tests. We demonstrate the potential for errors to arise when using a “modelling” approach without reference to real world test results. Simultaneously we show that a “training” approach can provide a highly cost-effective solution to the development of real-world comparisons between different multigene signatures. Comparisons between existing multiparametric tests is challenging, and evidence on discordance between tests in risk stratification presents further dilemmas. We present an approach, modelled in breast cancer, which can provide health care providers and researchers with the potential to perform robust and meaningful comparisons between multigene tests in a cost-effective manner. We demonstrate that whilst viable estimates of gene signatures can be derived from modelling approaches, in our study using a training approach allowed a close approximation to true signature results.

AB - Multiparametric assays for risk stratification are widely used in the management of breast cancer, with applications being developed for a number of other cancer settings. Recent data from multiple sources suggests that different tests may provide different risk estimates at the individual patient level. There is an increasing need for robust methods to support cost effective comparisons of test performance in multiple settings. The derivation of similar risk classifications using genes comprising the following multi-parametric tests Oncotype DX® (Genomic Health.), Prosigna™ (NanoString Technologies, Inc.), MammaPrint® (Agendia Inc.) was performed using different computational approaches. Results were compared to the actual test results. Two widely used approaches were applied, firstly computational “modelling” of test results using published algorithms and secondly a “training” approach which used reference results from the commercially supplied tests. We demonstrate the potential for errors to arise when using a “modelling” approach without reference to real world test results. Simultaneously we show that a “training” approach can provide a highly cost-effective solution to the development of real-world comparisons between different multigene signatures. Comparisons between existing multiparametric tests is challenging, and evidence on discordance between tests in risk stratification presents further dilemmas. We present an approach, modelled in breast cancer, which can provide health care providers and researchers with the potential to perform robust and meaningful comparisons between multigene tests in a cost-effective manner. We demonstrate that whilst viable estimates of gene signatures can be derived from modelling approaches, in our study using a training approach allowed a close approximation to true signature results.

UR - http://www.scopus.com/inward/record.url?scp=85090320659&partnerID=8YFLogxK

U2 - 10.1371/journal.pone.0238593

DO - 10.1371/journal.pone.0238593

M3 - Article

C2 - 32881987

AN - SCOPUS:85090320659

VL - 15

JO - PLoSONE

JF - PLoSONE

SN - 1932-6203

IS - 9 September

M1 - e0238593

ER -