TY - JOUR
T1 - Change-point of multiple biomarkers in women with ovarian cancer
AU - Mariño, Inés P
AU - Blyuss, Oleg
AU - Ryan, Andy
AU - Gentry-Maharaj, Aleksandra
AU - Timms, John F
AU - Dawnay, Anne
AU - Kalsi, Jatinderpal
AU - Jacobs, Ian J
AU - Menon, Usha
AU - Zaikin, Alexey
PY - 2017/3
Y1 - 2017/3
N2 - To date several algorithms for longitudinal analysis of ovarian cancer biomarkers have been proposed in the literature. An issue of specific interest is to determine whether the baseline level of a biomarker changes significantly at some time instant (change-point) prior to the clinical diagnosis of cancer. Such change-points in the serum biomarker Cancer Antigen 125 (CA125) time series data have been used in ovarian cancer screening, resulting in earlier detection with a sensitivity of 85% in the most recent trial, the UK Collaborative Trial of Ovarian Cancer Screening (UKCTOCS, number ISRCTN22488978; NCT00058032). Here we propose to apply a hierarchical Bayesian change-point model to jointly study the features of time series from multiple biomarkers. For this model we have analytically derived the conditional probability distribution of every unknown parameter, thus enabling the design of efficient Markov Chain Monte Carlo methods for their estimation. We have applied these methods to the estimation of change-points in time series data of multiple biomarkers, including CA125 and others, using data from a nested case-control study of women diagnosed with ovarian cancer in UKCTOCS. In this way we assess whether any of these additional biomarkers can play a role in change-point detection and, therefore, aid in the diagnosis of the disease in patients for whom the CA125 time series does not display a change-point. We have also investigated whether the change-points for different biomarkers occur at similar times for the same patient. The main conclusion of our study is that the combined analysis of a group of specific biomarkers may possibly improve the detection of change-points in time series data (compared to the analysis of CA125 alone) which, in turn, are relevant for the early diagnosis of ovarian cancer.
AB - To date several algorithms for longitudinal analysis of ovarian cancer biomarkers have been proposed in the literature. An issue of specific interest is to determine whether the baseline level of a biomarker changes significantly at some time instant (change-point) prior to the clinical diagnosis of cancer. Such change-points in the serum biomarker Cancer Antigen 125 (CA125) time series data have been used in ovarian cancer screening, resulting in earlier detection with a sensitivity of 85% in the most recent trial, the UK Collaborative Trial of Ovarian Cancer Screening (UKCTOCS, number ISRCTN22488978; NCT00058032). Here we propose to apply a hierarchical Bayesian change-point model to jointly study the features of time series from multiple biomarkers. For this model we have analytically derived the conditional probability distribution of every unknown parameter, thus enabling the design of efficient Markov Chain Monte Carlo methods for their estimation. We have applied these methods to the estimation of change-points in time series data of multiple biomarkers, including CA125 and others, using data from a nested case-control study of women diagnosed with ovarian cancer in UKCTOCS. In this way we assess whether any of these additional biomarkers can play a role in change-point detection and, therefore, aid in the diagnosis of the disease in patients for whom the CA125 time series does not display a change-point. We have also investigated whether the change-points for different biomarkers occur at similar times for the same patient. The main conclusion of our study is that the combined analysis of a group of specific biomarkers may possibly improve the detection of change-points in time series data (compared to the analysis of CA125 alone) which, in turn, are relevant for the early diagnosis of ovarian cancer.
U2 - 10.1016/j.bspc.2016.11.015
DO - 10.1016/j.bspc.2016.11.015
M3 - Article
SN - 1746-8094
VL - 33
SP - 169
EP - 177
JO - Biomedical signal processing and control
JF - Biomedical signal processing and control
ER -