A prognostic model for advanced colorectal neoplasia recurrence

Lin Liu, Karen Messer, John A. Baron, David Lieberman, Elizabeth T. Jacobs, Amanda J. Cross, Gwen Murphy, Maria Elena Martinez, Samir Gupta

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

Purpose: Following colonoscopic polypectomy, US Multisociety Task Force (USMSTF) guidelines stratify patients based on risk of subsequent advanced neoplasia (AN) using number, size, and histology of resected polyps, but have only moderate sensitivity and specificity. We hypothesized that a state-of-the-art statistical prediction model might improve identification of patients at high risk of future AN and address these challenges. Methods: Data were pooled from seven prospective studies which had follow-up ascertainment of metachronous AN within 3–5 years of baseline polypectomy (combined n = 8,228). Pooled data were randomly split into training (n = 5,483) and validation (n = 2,745) sets. A prognostic model was developed using best practices. Two risk cut-points were identified in the training data which achieved a 10 percentage point improvement in sensitivity and specificity, respectively, over current USMSTF guidelines. Clinical benefit of USMSTF versus model-based risk stratification was then estimated using validation data. Results: The final model included polyp location, prior polyp history, patient age, and number, size and histology of resected polyps. The first risk cut-point improved sensitivity but with loss of specificity. The second risk cut-point improved specificity without loss of sensitivity (specificity 46.2 % model vs. 42.1 % guidelines, p < 0.001; sensitivity 75.8 % model vs. 74.0 % guidelines, p = 0.64). Estimated AUC was 65 % (95 % CI: 62–69 %). Conclusion: This model-based approach allows flexibility in trading sensitivity and specificity, which can optimize colonoscopy over- versus underuse rates. Only modest improvements in prognostic power are possible using currently available clinical data. Research considering additional factors such as adenoma detection rate for risk prediction appears warranted.

Original languageEnglish (US)
Pages (from-to)1-11
Number of pages11
JournalCancer Causes and Control
DOIs
StateAccepted/In press - Aug 12 2016

Fingerprint

Recurrence
Polyps
Advisory Committees
Neoplasms
Guidelines
Sensitivity and Specificity
Histology
Statistical Models
Colonoscopy
Practice Guidelines
Adenoma
Area Under Curve
Prospective Studies
Research

Keywords

  • Colorectal cancer
  • Colorectal polyps
  • Epidemiology
  • Polyp surveillance
  • Risk stratification

ASJC Scopus subject areas

  • Oncology
  • Cancer Research

Cite this

Liu, L., Messer, K., Baron, J. A., Lieberman, D., Jacobs, E. T., Cross, A. J., ... Gupta, S. (Accepted/In press). A prognostic model for advanced colorectal neoplasia recurrence. Cancer Causes and Control, 1-11. https://doi.org/10.1007/s10552-016-0795-5

A prognostic model for advanced colorectal neoplasia recurrence. / Liu, Lin; Messer, Karen; Baron, John A.; Lieberman, David; Jacobs, Elizabeth T.; Cross, Amanda J.; Murphy, Gwen; Martinez, Maria Elena; Gupta, Samir.

In: Cancer Causes and Control, 12.08.2016, p. 1-11.

Research output: Contribution to journalArticle

Liu, L, Messer, K, Baron, JA, Lieberman, D, Jacobs, ET, Cross, AJ, Murphy, G, Martinez, ME & Gupta, S 2016, 'A prognostic model for advanced colorectal neoplasia recurrence', Cancer Causes and Control, pp. 1-11. https://doi.org/10.1007/s10552-016-0795-5
Liu, Lin ; Messer, Karen ; Baron, John A. ; Lieberman, David ; Jacobs, Elizabeth T. ; Cross, Amanda J. ; Murphy, Gwen ; Martinez, Maria Elena ; Gupta, Samir. / A prognostic model for advanced colorectal neoplasia recurrence. In: Cancer Causes and Control. 2016 ; pp. 1-11.
@article{4f454e81187149c1ad0987d1a6ad088c,
title = "A prognostic model for advanced colorectal neoplasia recurrence",
abstract = "Purpose: Following colonoscopic polypectomy, US Multisociety Task Force (USMSTF) guidelines stratify patients based on risk of subsequent advanced neoplasia (AN) using number, size, and histology of resected polyps, but have only moderate sensitivity and specificity. We hypothesized that a state-of-the-art statistical prediction model might improve identification of patients at high risk of future AN and address these challenges. Methods: Data were pooled from seven prospective studies which had follow-up ascertainment of metachronous AN within 3–5 years of baseline polypectomy (combined n = 8,228). Pooled data were randomly split into training (n = 5,483) and validation (n = 2,745) sets. A prognostic model was developed using best practices. Two risk cut-points were identified in the training data which achieved a 10 percentage point improvement in sensitivity and specificity, respectively, over current USMSTF guidelines. Clinical benefit of USMSTF versus model-based risk stratification was then estimated using validation data. Results: The final model included polyp location, prior polyp history, patient age, and number, size and histology of resected polyps. The first risk cut-point improved sensitivity but with loss of specificity. The second risk cut-point improved specificity without loss of sensitivity (specificity 46.2 {\%} model vs. 42.1 {\%} guidelines, p < 0.001; sensitivity 75.8 {\%} model vs. 74.0 {\%} guidelines, p = 0.64). Estimated AUC was 65 {\%} (95 {\%} CI: 62–69 {\%}). Conclusion: This model-based approach allows flexibility in trading sensitivity and specificity, which can optimize colonoscopy over- versus underuse rates. Only modest improvements in prognostic power are possible using currently available clinical data. Research considering additional factors such as adenoma detection rate for risk prediction appears warranted.",
keywords = "Colorectal cancer, Colorectal polyps, Epidemiology, Polyp surveillance, Risk stratification",
author = "Lin Liu and Karen Messer and Baron, {John A.} and David Lieberman and Jacobs, {Elizabeth T.} and Cross, {Amanda J.} and Gwen Murphy and Martinez, {Maria Elena} and Samir Gupta",
year = "2016",
month = "8",
day = "12",
doi = "10.1007/s10552-016-0795-5",
language = "English (US)",
pages = "1--11",
journal = "Cancer Causes and Control",
issn = "0957-5243",
publisher = "Springer Netherlands",

}

TY - JOUR

T1 - A prognostic model for advanced colorectal neoplasia recurrence

AU - Liu, Lin

AU - Messer, Karen

AU - Baron, John A.

AU - Lieberman, David

AU - Jacobs, Elizabeth T.

AU - Cross, Amanda J.

AU - Murphy, Gwen

AU - Martinez, Maria Elena

AU - Gupta, Samir

PY - 2016/8/12

Y1 - 2016/8/12

N2 - Purpose: Following colonoscopic polypectomy, US Multisociety Task Force (USMSTF) guidelines stratify patients based on risk of subsequent advanced neoplasia (AN) using number, size, and histology of resected polyps, but have only moderate sensitivity and specificity. We hypothesized that a state-of-the-art statistical prediction model might improve identification of patients at high risk of future AN and address these challenges. Methods: Data were pooled from seven prospective studies which had follow-up ascertainment of metachronous AN within 3–5 years of baseline polypectomy (combined n = 8,228). Pooled data were randomly split into training (n = 5,483) and validation (n = 2,745) sets. A prognostic model was developed using best practices. Two risk cut-points were identified in the training data which achieved a 10 percentage point improvement in sensitivity and specificity, respectively, over current USMSTF guidelines. Clinical benefit of USMSTF versus model-based risk stratification was then estimated using validation data. Results: The final model included polyp location, prior polyp history, patient age, and number, size and histology of resected polyps. The first risk cut-point improved sensitivity but with loss of specificity. The second risk cut-point improved specificity without loss of sensitivity (specificity 46.2 % model vs. 42.1 % guidelines, p < 0.001; sensitivity 75.8 % model vs. 74.0 % guidelines, p = 0.64). Estimated AUC was 65 % (95 % CI: 62–69 %). Conclusion: This model-based approach allows flexibility in trading sensitivity and specificity, which can optimize colonoscopy over- versus underuse rates. Only modest improvements in prognostic power are possible using currently available clinical data. Research considering additional factors such as adenoma detection rate for risk prediction appears warranted.

AB - Purpose: Following colonoscopic polypectomy, US Multisociety Task Force (USMSTF) guidelines stratify patients based on risk of subsequent advanced neoplasia (AN) using number, size, and histology of resected polyps, but have only moderate sensitivity and specificity. We hypothesized that a state-of-the-art statistical prediction model might improve identification of patients at high risk of future AN and address these challenges. Methods: Data were pooled from seven prospective studies which had follow-up ascertainment of metachronous AN within 3–5 years of baseline polypectomy (combined n = 8,228). Pooled data were randomly split into training (n = 5,483) and validation (n = 2,745) sets. A prognostic model was developed using best practices. Two risk cut-points were identified in the training data which achieved a 10 percentage point improvement in sensitivity and specificity, respectively, over current USMSTF guidelines. Clinical benefit of USMSTF versus model-based risk stratification was then estimated using validation data. Results: The final model included polyp location, prior polyp history, patient age, and number, size and histology of resected polyps. The first risk cut-point improved sensitivity but with loss of specificity. The second risk cut-point improved specificity without loss of sensitivity (specificity 46.2 % model vs. 42.1 % guidelines, p < 0.001; sensitivity 75.8 % model vs. 74.0 % guidelines, p = 0.64). Estimated AUC was 65 % (95 % CI: 62–69 %). Conclusion: This model-based approach allows flexibility in trading sensitivity and specificity, which can optimize colonoscopy over- versus underuse rates. Only modest improvements in prognostic power are possible using currently available clinical data. Research considering additional factors such as adenoma detection rate for risk prediction appears warranted.

KW - Colorectal cancer

KW - Colorectal polyps

KW - Epidemiology

KW - Polyp surveillance

KW - Risk stratification

UR - http://www.scopus.com/inward/record.url?scp=84982132588&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84982132588&partnerID=8YFLogxK

U2 - 10.1007/s10552-016-0795-5

DO - 10.1007/s10552-016-0795-5

M3 - Article

C2 - 27517467

AN - SCOPUS:84982132588

SP - 1

EP - 11

JO - Cancer Causes and Control

JF - Cancer Causes and Control

SN - 0957-5243

ER -