Artificial neural network analysis of circulating tumor cells in metastatic breast cancer patients

Antonio Giordano, Mario Giuliano, Michelino De Laurentiis, Antonio Eleuteri, Francesco Iorio, Roberto Tagliaferri, Gabriel N. Hortobagyi, Lajos Pusztai, Sabino De Placido, Kenneth Hess, Massimo Cristofanilli, James M. Reuben

Research output: Contribution to journalArticle

21 Citations (Scopus)

Abstract

A cut-off of 5 circulating tumor cells (CTCs) per 7.5 ml of blood in metastatic breast cancer (MBC) patients is highly predictive of outcome. We analyzed the relationship between CTCs as a continuous variable and overall survival in immunohistochemically defined primary tumor molecular subtypes using an artificial neural network (ANN) prognostic tool to determine the shape of the relationship between risk of death and CTC count and to predict individual survival. We analyzed a training dataset of 311 of 517 (60%) consecutive MBC patients who had been treated at MD Anderson Cancer Center from September 2004 to 2009 and who had undergone pre-therapy CTC counts (CellSearch ®). Age; estrogen, progesterone receptor, and HER2 status; visceral metastasis; metastatic disease sites; therapy type and line; and CTCs as a continuous value were evaluated using ANN. A model with parameter estimates obtained from the training data was tested in a validation set of the remaining 206 (40%) patients. The model estimates were accurate, with good discrimination and calibration. Risk of death, as estimated by ANN, linearly increased with increasing CTC count in all molecular tumor subtypes but was higher in ER+ and triple-negative MBC than in HER2+. The probabilities of survival for the four subtypes with 0 CTC were as follows: ER+/HER2- 0.947, ER+/HER2+ 0.959, ER-/HER2+ 0.902, and ER-/HER2- 0.875. For patients with 200 CTCs, they were ER+/HER2- 0.439, ER+/HER2+ 0.621, ER-/HER2+ 0.307, ER-/HER2- 0.130. In this large study, ANN revealed a linear increase of risk of death in MBC patients with increasing CTC counts in all tumor subtypes. CTCs' prognostic effect was less evident in HER2+ MBC patients treated with targeted therapy. This study may support the concept that the number of CTCs, along with the biologic characteristics, needs to be carefully taken into account in future analysis.

Original languageEnglish
Pages (from-to)451-458
Number of pages8
JournalBreast Cancer Research and Treatment
Volume129
Issue number2
DOIs
Publication statusPublished - Sep 2011

Fingerprint

Circulating Neoplastic Cells
Breast Neoplasms
Cell Count
Survival
Neoplasms
Triple Negative Breast Neoplasms
Progesterone Receptors
Estrogen Receptors
Calibration
Therapeutics

Keywords

  • Artificial neural network
  • Circulating tumor cells
  • HER2
  • Metastatic breast cancer
  • Prognosis

ASJC Scopus subject areas

  • Oncology
  • Cancer Research

Cite this

Artificial neural network analysis of circulating tumor cells in metastatic breast cancer patients. / Giordano, Antonio; Giuliano, Mario; De Laurentiis, Michelino; Eleuteri, Antonio; Iorio, Francesco; Tagliaferri, Roberto; Hortobagyi, Gabriel N.; Pusztai, Lajos; De Placido, Sabino; Hess, Kenneth; Cristofanilli, Massimo; Reuben, James M.

In: Breast Cancer Research and Treatment, Vol. 129, No. 2, 09.2011, p. 451-458.

Research output: Contribution to journalArticle

Giordano, A, Giuliano, M, De Laurentiis, M, Eleuteri, A, Iorio, F, Tagliaferri, R, Hortobagyi, GN, Pusztai, L, De Placido, S, Hess, K, Cristofanilli, M & Reuben, JM 2011, 'Artificial neural network analysis of circulating tumor cells in metastatic breast cancer patients', Breast Cancer Research and Treatment, vol. 129, no. 2, pp. 451-458. https://doi.org/10.1007/s10549-011-1645-5
Giordano, Antonio ; Giuliano, Mario ; De Laurentiis, Michelino ; Eleuteri, Antonio ; Iorio, Francesco ; Tagliaferri, Roberto ; Hortobagyi, Gabriel N. ; Pusztai, Lajos ; De Placido, Sabino ; Hess, Kenneth ; Cristofanilli, Massimo ; Reuben, James M. / Artificial neural network analysis of circulating tumor cells in metastatic breast cancer patients. In: Breast Cancer Research and Treatment. 2011 ; Vol. 129, No. 2. pp. 451-458.
@article{9894a7664eba4ce9aaa2bb259aba22d2,
title = "Artificial neural network analysis of circulating tumor cells in metastatic breast cancer patients",
abstract = "A cut-off of 5 circulating tumor cells (CTCs) per 7.5 ml of blood in metastatic breast cancer (MBC) patients is highly predictive of outcome. We analyzed the relationship between CTCs as a continuous variable and overall survival in immunohistochemically defined primary tumor molecular subtypes using an artificial neural network (ANN) prognostic tool to determine the shape of the relationship between risk of death and CTC count and to predict individual survival. We analyzed a training dataset of 311 of 517 (60{\%}) consecutive MBC patients who had been treated at MD Anderson Cancer Center from September 2004 to 2009 and who had undergone pre-therapy CTC counts (CellSearch {\circledR}). Age; estrogen, progesterone receptor, and HER2 status; visceral metastasis; metastatic disease sites; therapy type and line; and CTCs as a continuous value were evaluated using ANN. A model with parameter estimates obtained from the training data was tested in a validation set of the remaining 206 (40{\%}) patients. The model estimates were accurate, with good discrimination and calibration. Risk of death, as estimated by ANN, linearly increased with increasing CTC count in all molecular tumor subtypes but was higher in ER+ and triple-negative MBC than in HER2+. The probabilities of survival for the four subtypes with 0 CTC were as follows: ER+/HER2- 0.947, ER+/HER2+ 0.959, ER-/HER2+ 0.902, and ER-/HER2- 0.875. For patients with 200 CTCs, they were ER+/HER2- 0.439, ER+/HER2+ 0.621, ER-/HER2+ 0.307, ER-/HER2- 0.130. In this large study, ANN revealed a linear increase of risk of death in MBC patients with increasing CTC counts in all tumor subtypes. CTCs' prognostic effect was less evident in HER2+ MBC patients treated with targeted therapy. This study may support the concept that the number of CTCs, along with the biologic characteristics, needs to be carefully taken into account in future analysis.",
keywords = "Artificial neural network, Circulating tumor cells, HER2, Metastatic breast cancer, Prognosis",
author = "Antonio Giordano and Mario Giuliano and {De Laurentiis}, Michelino and Antonio Eleuteri and Francesco Iorio and Roberto Tagliaferri and Hortobagyi, {Gabriel N.} and Lajos Pusztai and {De Placido}, Sabino and Kenneth Hess and Massimo Cristofanilli and Reuben, {James M.}",
year = "2011",
month = "9",
doi = "10.1007/s10549-011-1645-5",
language = "English",
volume = "129",
pages = "451--458",
journal = "Breast Cancer Research and Treatment",
issn = "0167-6806",
publisher = "Springer New York LLC",
number = "2",

}

TY - JOUR

T1 - Artificial neural network analysis of circulating tumor cells in metastatic breast cancer patients

AU - Giordano, Antonio

AU - Giuliano, Mario

AU - De Laurentiis, Michelino

AU - Eleuteri, Antonio

AU - Iorio, Francesco

AU - Tagliaferri, Roberto

AU - Hortobagyi, Gabriel N.

AU - Pusztai, Lajos

AU - De Placido, Sabino

AU - Hess, Kenneth

AU - Cristofanilli, Massimo

AU - Reuben, James M.

PY - 2011/9

Y1 - 2011/9

N2 - A cut-off of 5 circulating tumor cells (CTCs) per 7.5 ml of blood in metastatic breast cancer (MBC) patients is highly predictive of outcome. We analyzed the relationship between CTCs as a continuous variable and overall survival in immunohistochemically defined primary tumor molecular subtypes using an artificial neural network (ANN) prognostic tool to determine the shape of the relationship between risk of death and CTC count and to predict individual survival. We analyzed a training dataset of 311 of 517 (60%) consecutive MBC patients who had been treated at MD Anderson Cancer Center from September 2004 to 2009 and who had undergone pre-therapy CTC counts (CellSearch ®). Age; estrogen, progesterone receptor, and HER2 status; visceral metastasis; metastatic disease sites; therapy type and line; and CTCs as a continuous value were evaluated using ANN. A model with parameter estimates obtained from the training data was tested in a validation set of the remaining 206 (40%) patients. The model estimates were accurate, with good discrimination and calibration. Risk of death, as estimated by ANN, linearly increased with increasing CTC count in all molecular tumor subtypes but was higher in ER+ and triple-negative MBC than in HER2+. The probabilities of survival for the four subtypes with 0 CTC were as follows: ER+/HER2- 0.947, ER+/HER2+ 0.959, ER-/HER2+ 0.902, and ER-/HER2- 0.875. For patients with 200 CTCs, they were ER+/HER2- 0.439, ER+/HER2+ 0.621, ER-/HER2+ 0.307, ER-/HER2- 0.130. In this large study, ANN revealed a linear increase of risk of death in MBC patients with increasing CTC counts in all tumor subtypes. CTCs' prognostic effect was less evident in HER2+ MBC patients treated with targeted therapy. This study may support the concept that the number of CTCs, along with the biologic characteristics, needs to be carefully taken into account in future analysis.

AB - A cut-off of 5 circulating tumor cells (CTCs) per 7.5 ml of blood in metastatic breast cancer (MBC) patients is highly predictive of outcome. We analyzed the relationship between CTCs as a continuous variable and overall survival in immunohistochemically defined primary tumor molecular subtypes using an artificial neural network (ANN) prognostic tool to determine the shape of the relationship between risk of death and CTC count and to predict individual survival. We analyzed a training dataset of 311 of 517 (60%) consecutive MBC patients who had been treated at MD Anderson Cancer Center from September 2004 to 2009 and who had undergone pre-therapy CTC counts (CellSearch ®). Age; estrogen, progesterone receptor, and HER2 status; visceral metastasis; metastatic disease sites; therapy type and line; and CTCs as a continuous value were evaluated using ANN. A model with parameter estimates obtained from the training data was tested in a validation set of the remaining 206 (40%) patients. The model estimates were accurate, with good discrimination and calibration. Risk of death, as estimated by ANN, linearly increased with increasing CTC count in all molecular tumor subtypes but was higher in ER+ and triple-negative MBC than in HER2+. The probabilities of survival for the four subtypes with 0 CTC were as follows: ER+/HER2- 0.947, ER+/HER2+ 0.959, ER-/HER2+ 0.902, and ER-/HER2- 0.875. For patients with 200 CTCs, they were ER+/HER2- 0.439, ER+/HER2+ 0.621, ER-/HER2+ 0.307, ER-/HER2- 0.130. In this large study, ANN revealed a linear increase of risk of death in MBC patients with increasing CTC counts in all tumor subtypes. CTCs' prognostic effect was less evident in HER2+ MBC patients treated with targeted therapy. This study may support the concept that the number of CTCs, along with the biologic characteristics, needs to be carefully taken into account in future analysis.

KW - Artificial neural network

KW - Circulating tumor cells

KW - HER2

KW - Metastatic breast cancer

KW - Prognosis

UR - http://www.scopus.com/inward/record.url?scp=80052632043&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=80052632043&partnerID=8YFLogxK

U2 - 10.1007/s10549-011-1645-5

DO - 10.1007/s10549-011-1645-5

M3 - Article

VL - 129

SP - 451

EP - 458

JO - Breast Cancer Research and Treatment

JF - Breast Cancer Research and Treatment

SN - 0167-6806

IS - 2

ER -