Estimating the frequency of indolent breast cancer in screening trials

Yu Shen, Wenli Dong, Roman Gulati, Marc D. Ryser, Ruth Etzioni

Research output: Contribution to journalArticle

Abstract

Cancer screening can detect cancer that would not have been detected in a patient’s lifetime without screening. Standard methods for analyzing screening data do not explicitly account for the possibility that a fraction of tumors may remain latent indefinitely. We extend these methods by representing cancers as a mixture of those that progress to symptoms (progressive) and those that remain latent (indolent). Given sensitivity of the screening test, we derive likelihood expressions to simultaneously estimate (1) the rate of onset of preclinical cancer, (2) the average preclinical duration of progressive cancers, and (3) the fraction of preclinical cancers that are indolent. Simulations demonstrate satisfactory performance of the estimation approach to identify model parameters subject to precise specifications of input parameters and adequate numbers of interval cancers. In application to four breast cancer screening trials, the estimated indolent fraction among preclinical cancers varies between 2% and 35% when assuming 80% test sensitivity and varying specifications for the earliest time that participants could plausibly have developed cancer. We conclude that standard methods for analyzing screening data can be extended to allow some indolent cancers, but accurate estimation depends on correctly specifying key inputs that may be difficult to determine precisely in practice.

Original languageEnglish (US)
JournalStatistical Methods in Medical Research
DOIs
StateAccepted/In press - Jan 1 2018
Externally publishedYes

Fingerprint

Breast Cancer
Early Detection of Cancer
Screening
Cancer
Breast Neoplasms
Neoplasms
Specification
Tumor
Lifetime
Likelihood
Vary
Interval

Keywords

  • Breast cancer
  • indolent cancer
  • mammography screening
  • maximum likelihood estimation
  • overdiagnosis
  • randomized controlled trials

ASJC Scopus subject areas

  • Epidemiology
  • Statistics and Probability
  • Health Information Management

Cite this

Estimating the frequency of indolent breast cancer in screening trials. / Shen, Yu; Dong, Wenli; Gulati, Roman; Ryser, Marc D.; Etzioni, Ruth.

In: Statistical Methods in Medical Research, 01.01.2018.

Research output: Contribution to journalArticle

Shen, Yu ; Dong, Wenli ; Gulati, Roman ; Ryser, Marc D. ; Etzioni, Ruth. / Estimating the frequency of indolent breast cancer in screening trials. In: Statistical Methods in Medical Research. 2018.
@article{233127b068a848d4abcc2ee446c37e84,
title = "Estimating the frequency of indolent breast cancer in screening trials",
abstract = "Cancer screening can detect cancer that would not have been detected in a patient’s lifetime without screening. Standard methods for analyzing screening data do not explicitly account for the possibility that a fraction of tumors may remain latent indefinitely. We extend these methods by representing cancers as a mixture of those that progress to symptoms (progressive) and those that remain latent (indolent). Given sensitivity of the screening test, we derive likelihood expressions to simultaneously estimate (1) the rate of onset of preclinical cancer, (2) the average preclinical duration of progressive cancers, and (3) the fraction of preclinical cancers that are indolent. Simulations demonstrate satisfactory performance of the estimation approach to identify model parameters subject to precise specifications of input parameters and adequate numbers of interval cancers. In application to four breast cancer screening trials, the estimated indolent fraction among preclinical cancers varies between 2{\%} and 35{\%} when assuming 80{\%} test sensitivity and varying specifications for the earliest time that participants could plausibly have developed cancer. We conclude that standard methods for analyzing screening data can be extended to allow some indolent cancers, but accurate estimation depends on correctly specifying key inputs that may be difficult to determine precisely in practice.",
keywords = "Breast cancer, indolent cancer, mammography screening, maximum likelihood estimation, overdiagnosis, randomized controlled trials",
author = "Yu Shen and Wenli Dong and Roman Gulati and Ryser, {Marc D.} and Ruth Etzioni",
year = "2018",
month = "1",
day = "1",
doi = "10.1177/0962280217754232",
language = "English (US)",
journal = "Statistical Methods in Medical Research",
issn = "0962-2802",
publisher = "SAGE Publications Ltd",

}

TY - JOUR

T1 - Estimating the frequency of indolent breast cancer in screening trials

AU - Shen, Yu

AU - Dong, Wenli

AU - Gulati, Roman

AU - Ryser, Marc D.

AU - Etzioni, Ruth

PY - 2018/1/1

Y1 - 2018/1/1

N2 - Cancer screening can detect cancer that would not have been detected in a patient’s lifetime without screening. Standard methods for analyzing screening data do not explicitly account for the possibility that a fraction of tumors may remain latent indefinitely. We extend these methods by representing cancers as a mixture of those that progress to symptoms (progressive) and those that remain latent (indolent). Given sensitivity of the screening test, we derive likelihood expressions to simultaneously estimate (1) the rate of onset of preclinical cancer, (2) the average preclinical duration of progressive cancers, and (3) the fraction of preclinical cancers that are indolent. Simulations demonstrate satisfactory performance of the estimation approach to identify model parameters subject to precise specifications of input parameters and adequate numbers of interval cancers. In application to four breast cancer screening trials, the estimated indolent fraction among preclinical cancers varies between 2% and 35% when assuming 80% test sensitivity and varying specifications for the earliest time that participants could plausibly have developed cancer. We conclude that standard methods for analyzing screening data can be extended to allow some indolent cancers, but accurate estimation depends on correctly specifying key inputs that may be difficult to determine precisely in practice.

AB - Cancer screening can detect cancer that would not have been detected in a patient’s lifetime without screening. Standard methods for analyzing screening data do not explicitly account for the possibility that a fraction of tumors may remain latent indefinitely. We extend these methods by representing cancers as a mixture of those that progress to symptoms (progressive) and those that remain latent (indolent). Given sensitivity of the screening test, we derive likelihood expressions to simultaneously estimate (1) the rate of onset of preclinical cancer, (2) the average preclinical duration of progressive cancers, and (3) the fraction of preclinical cancers that are indolent. Simulations demonstrate satisfactory performance of the estimation approach to identify model parameters subject to precise specifications of input parameters and adequate numbers of interval cancers. In application to four breast cancer screening trials, the estimated indolent fraction among preclinical cancers varies between 2% and 35% when assuming 80% test sensitivity and varying specifications for the earliest time that participants could plausibly have developed cancer. We conclude that standard methods for analyzing screening data can be extended to allow some indolent cancers, but accurate estimation depends on correctly specifying key inputs that may be difficult to determine precisely in practice.

KW - Breast cancer

KW - indolent cancer

KW - mammography screening

KW - maximum likelihood estimation

KW - overdiagnosis

KW - randomized controlled trials

UR - http://www.scopus.com/inward/record.url?scp=85043459693&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85043459693&partnerID=8YFLogxK

U2 - 10.1177/0962280217754232

DO - 10.1177/0962280217754232

M3 - Article

C2 - 29402176

AN - SCOPUS:85043459693

JO - Statistical Methods in Medical Research

JF - Statistical Methods in Medical Research

SN - 0962-2802

ER -