Validation of tumor protein marker quantification by two independent automated immunofluorescence image analysis platforms

Amy R. Peck, Melanie A. Girondo, Chengbao Liu, Albert J. Kovatich, Jeffrey A. Hooke, Craig D. Shriver, Hai Hu, Edith P. Mitchell, Boris Freydin, Terry Hyslop, Inna Chervoneva, Hallgeir Rui

Research output: Contribution to journalArticle

Abstract

Protein marker levels in formalin-fixed, paraffin-embedded tissue sections traditionally have been assayed by chromogenic immunohistochemistry and evaluated visually by pathologists. Pathologist scoring of chromogen staining intensity is subjective and generates low-resolution ordinal or nominal data rather than continuous data. Emerging digital pathology platforms now allow quantification of chromogen or fluorescence signals by computer-assisted image analysis, providing continuous immunohistochemistry values. Fluorescence immunohistochemistry offers greater dynamic signal range than chromogen immunohistochemistry, and combined with image analysis holds the promise of enhanced sensitivity and analytic resolution, and consequently more robust quantification. However, commercial fluorescence scanners and image analysis software differ in features and capabilities, and claims of objective quantitative immunohistochemistry are difficult to validate as pathologist scoring is subjective and there is no accepted gold standard. Here we provide the first side-by-side validation of two technologically distinct commercial fluorescence immunohistochemistry analysis platforms. We document highly consistent results by (1) concordance analysis of fluorescence immunohistochemistry values and (2) agreement in outcome predictions both for objective, data-driven cutpoint dichotomization with Kaplan-Meier analyses or employment of continuous marker values to compute receiver-operating curves. The two platforms examined rely on distinct fluorescence immunohistochemistry imaging hardware, microscopy vs line scanning, and functionally distinct image analysis software. Fluorescence immunohistochemistry values for nuclear-localized and tyrosine-phosphorylated Stat5a/b computed by each platform on a cohort of 323 breast cancer cases revealed high concordance after linear calibration, a finding confirmed on an independent 382 case cohort, with concordance correlation coefficients >0.98. Data-driven optimal cutpoints for outcome prediction by either platform were reciprocally applicable to the data derived by the alternate platform, identifying patients with low Nuc-pYStat5 at ∼3.5-fold increased risk of disease progression. Our analyses identified two highly concordant fluorescence immunohistochemistry platforms that may serve as benchmarks for testing of other platforms, and low interoperator variability supports the implementation of objective tumor marker quantification in pathology laboratories.

Original languageEnglish (US)
Pages (from-to)1143-1154
Number of pages12
JournalModern Pathology
Volume29
Issue number10
DOIs
StatePublished - Oct 1 2016
Externally publishedYes

Fingerprint

Tumor Biomarkers
Fluorescent Antibody Technique
Immunohistochemistry
Fluorescence
Proteins
Software
Pathology
Benchmarking
Computer-Assisted Image Processing
Optical Imaging
Kaplan-Meier Estimate
Paraffin
Formaldehyde
Calibration
Tyrosine
Disease Progression
Microscopy
Staining and Labeling
Breast Neoplasms

ASJC Scopus subject areas

  • Pathology and Forensic Medicine

Cite this

Peck, A. R., Girondo, M. A., Liu, C., Kovatich, A. J., Hooke, J. A., Shriver, C. D., ... Rui, H. (2016). Validation of tumor protein marker quantification by two independent automated immunofluorescence image analysis platforms. Modern Pathology, 29(10), 1143-1154. https://doi.org/10.1038/modpathol.2016.112

Validation of tumor protein marker quantification by two independent automated immunofluorescence image analysis platforms. / Peck, Amy R.; Girondo, Melanie A.; Liu, Chengbao; Kovatich, Albert J.; Hooke, Jeffrey A.; Shriver, Craig D.; Hu, Hai; Mitchell, Edith P.; Freydin, Boris; Hyslop, Terry; Chervoneva, Inna; Rui, Hallgeir.

In: Modern Pathology, Vol. 29, No. 10, 01.10.2016, p. 1143-1154.

Research output: Contribution to journalArticle

Peck, AR, Girondo, MA, Liu, C, Kovatich, AJ, Hooke, JA, Shriver, CD, Hu, H, Mitchell, EP, Freydin, B, Hyslop, T, Chervoneva, I & Rui, H 2016, 'Validation of tumor protein marker quantification by two independent automated immunofluorescence image analysis platforms', Modern Pathology, vol. 29, no. 10, pp. 1143-1154. https://doi.org/10.1038/modpathol.2016.112
Peck, Amy R. ; Girondo, Melanie A. ; Liu, Chengbao ; Kovatich, Albert J. ; Hooke, Jeffrey A. ; Shriver, Craig D. ; Hu, Hai ; Mitchell, Edith P. ; Freydin, Boris ; Hyslop, Terry ; Chervoneva, Inna ; Rui, Hallgeir. / Validation of tumor protein marker quantification by two independent automated immunofluorescence image analysis platforms. In: Modern Pathology. 2016 ; Vol. 29, No. 10. pp. 1143-1154.
@article{808561fcdca2487787dce9b5b02d3e3e,
title = "Validation of tumor protein marker quantification by two independent automated immunofluorescence image analysis platforms",
abstract = "Protein marker levels in formalin-fixed, paraffin-embedded tissue sections traditionally have been assayed by chromogenic immunohistochemistry and evaluated visually by pathologists. Pathologist scoring of chromogen staining intensity is subjective and generates low-resolution ordinal or nominal data rather than continuous data. Emerging digital pathology platforms now allow quantification of chromogen or fluorescence signals by computer-assisted image analysis, providing continuous immunohistochemistry values. Fluorescence immunohistochemistry offers greater dynamic signal range than chromogen immunohistochemistry, and combined with image analysis holds the promise of enhanced sensitivity and analytic resolution, and consequently more robust quantification. However, commercial fluorescence scanners and image analysis software differ in features and capabilities, and claims of objective quantitative immunohistochemistry are difficult to validate as pathologist scoring is subjective and there is no accepted gold standard. Here we provide the first side-by-side validation of two technologically distinct commercial fluorescence immunohistochemistry analysis platforms. We document highly consistent results by (1) concordance analysis of fluorescence immunohistochemistry values and (2) agreement in outcome predictions both for objective, data-driven cutpoint dichotomization with Kaplan-Meier analyses or employment of continuous marker values to compute receiver-operating curves. The two platforms examined rely on distinct fluorescence immunohistochemistry imaging hardware, microscopy vs line scanning, and functionally distinct image analysis software. Fluorescence immunohistochemistry values for nuclear-localized and tyrosine-phosphorylated Stat5a/b computed by each platform on a cohort of 323 breast cancer cases revealed high concordance after linear calibration, a finding confirmed on an independent 382 case cohort, with concordance correlation coefficients >0.98. Data-driven optimal cutpoints for outcome prediction by either platform were reciprocally applicable to the data derived by the alternate platform, identifying patients with low Nuc-pYStat5 at ∼3.5-fold increased risk of disease progression. Our analyses identified two highly concordant fluorescence immunohistochemistry platforms that may serve as benchmarks for testing of other platforms, and low interoperator variability supports the implementation of objective tumor marker quantification in pathology laboratories.",
author = "Peck, {Amy R.} and Girondo, {Melanie A.} and Chengbao Liu and Kovatich, {Albert J.} and Hooke, {Jeffrey A.} and Shriver, {Craig D.} and Hai Hu and Mitchell, {Edith P.} and Boris Freydin and Terry Hyslop and Inna Chervoneva and Hallgeir Rui",
year = "2016",
month = "10",
day = "1",
doi = "10.1038/modpathol.2016.112",
language = "English (US)",
volume = "29",
pages = "1143--1154",
journal = "Modern Pathology",
issn = "0893-3952",
publisher = "Nature Publishing Group",
number = "10",

}

TY - JOUR

T1 - Validation of tumor protein marker quantification by two independent automated immunofluorescence image analysis platforms

AU - Peck, Amy R.

AU - Girondo, Melanie A.

AU - Liu, Chengbao

AU - Kovatich, Albert J.

AU - Hooke, Jeffrey A.

AU - Shriver, Craig D.

AU - Hu, Hai

AU - Mitchell, Edith P.

AU - Freydin, Boris

AU - Hyslop, Terry

AU - Chervoneva, Inna

AU - Rui, Hallgeir

PY - 2016/10/1

Y1 - 2016/10/1

N2 - Protein marker levels in formalin-fixed, paraffin-embedded tissue sections traditionally have been assayed by chromogenic immunohistochemistry and evaluated visually by pathologists. Pathologist scoring of chromogen staining intensity is subjective and generates low-resolution ordinal or nominal data rather than continuous data. Emerging digital pathology platforms now allow quantification of chromogen or fluorescence signals by computer-assisted image analysis, providing continuous immunohistochemistry values. Fluorescence immunohistochemistry offers greater dynamic signal range than chromogen immunohistochemistry, and combined with image analysis holds the promise of enhanced sensitivity and analytic resolution, and consequently more robust quantification. However, commercial fluorescence scanners and image analysis software differ in features and capabilities, and claims of objective quantitative immunohistochemistry are difficult to validate as pathologist scoring is subjective and there is no accepted gold standard. Here we provide the first side-by-side validation of two technologically distinct commercial fluorescence immunohistochemistry analysis platforms. We document highly consistent results by (1) concordance analysis of fluorescence immunohistochemistry values and (2) agreement in outcome predictions both for objective, data-driven cutpoint dichotomization with Kaplan-Meier analyses or employment of continuous marker values to compute receiver-operating curves. The two platforms examined rely on distinct fluorescence immunohistochemistry imaging hardware, microscopy vs line scanning, and functionally distinct image analysis software. Fluorescence immunohistochemistry values for nuclear-localized and tyrosine-phosphorylated Stat5a/b computed by each platform on a cohort of 323 breast cancer cases revealed high concordance after linear calibration, a finding confirmed on an independent 382 case cohort, with concordance correlation coefficients >0.98. Data-driven optimal cutpoints for outcome prediction by either platform were reciprocally applicable to the data derived by the alternate platform, identifying patients with low Nuc-pYStat5 at ∼3.5-fold increased risk of disease progression. Our analyses identified two highly concordant fluorescence immunohistochemistry platforms that may serve as benchmarks for testing of other platforms, and low interoperator variability supports the implementation of objective tumor marker quantification in pathology laboratories.

AB - Protein marker levels in formalin-fixed, paraffin-embedded tissue sections traditionally have been assayed by chromogenic immunohistochemistry and evaluated visually by pathologists. Pathologist scoring of chromogen staining intensity is subjective and generates low-resolution ordinal or nominal data rather than continuous data. Emerging digital pathology platforms now allow quantification of chromogen or fluorescence signals by computer-assisted image analysis, providing continuous immunohistochemistry values. Fluorescence immunohistochemistry offers greater dynamic signal range than chromogen immunohistochemistry, and combined with image analysis holds the promise of enhanced sensitivity and analytic resolution, and consequently more robust quantification. However, commercial fluorescence scanners and image analysis software differ in features and capabilities, and claims of objective quantitative immunohistochemistry are difficult to validate as pathologist scoring is subjective and there is no accepted gold standard. Here we provide the first side-by-side validation of two technologically distinct commercial fluorescence immunohistochemistry analysis platforms. We document highly consistent results by (1) concordance analysis of fluorescence immunohistochemistry values and (2) agreement in outcome predictions both for objective, data-driven cutpoint dichotomization with Kaplan-Meier analyses or employment of continuous marker values to compute receiver-operating curves. The two platforms examined rely on distinct fluorescence immunohistochemistry imaging hardware, microscopy vs line scanning, and functionally distinct image analysis software. Fluorescence immunohistochemistry values for nuclear-localized and tyrosine-phosphorylated Stat5a/b computed by each platform on a cohort of 323 breast cancer cases revealed high concordance after linear calibration, a finding confirmed on an independent 382 case cohort, with concordance correlation coefficients >0.98. Data-driven optimal cutpoints for outcome prediction by either platform were reciprocally applicable to the data derived by the alternate platform, identifying patients with low Nuc-pYStat5 at ∼3.5-fold increased risk of disease progression. Our analyses identified two highly concordant fluorescence immunohistochemistry platforms that may serve as benchmarks for testing of other platforms, and low interoperator variability supports the implementation of objective tumor marker quantification in pathology laboratories.

UR - http://www.scopus.com/inward/record.url?scp=84975132610&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84975132610&partnerID=8YFLogxK

U2 - 10.1038/modpathol.2016.112

DO - 10.1038/modpathol.2016.112

M3 - Article

C2 - 27312066

AN - SCOPUS:84975132610

VL - 29

SP - 1143

EP - 1154

JO - Modern Pathology

JF - Modern Pathology

SN - 0893-3952

IS - 10

ER -