A cut-off of 2150 cytokeratin 19 mRNA copy number in sentinel lymph node may be a powerful predictor of non-sentinel lymph node status in breast cancer patients

Irene Terrenato, Valerio D'Alicandro, Beatrice Casini, Letizia Perracchio, Francesca Rollo, Laura De Salvo, Simona Di Filippo, Franco Di Filippo, Edoardo Pescarmona, Marcello Maugeri-Saccà, Marcella Mottolese, Simonetta Buglioni

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

Since 2007, one-step nucleic acid amplification (OSNA) has been used as a diagnostic system for sentinel lymph node (SLN) examination in patients with breast cancer. This study aimed to define a new clinical cut-off of CK19 mRNA copy number based on the calculation of the risk that an axillary lymph node dissection (ALND) will be positive. We analyzed 1529 SLNs from 1140 patients with the OSNA assay and 318 patients with positive SLNs for micrometastasis (250 copies) and macrometastasis (5000 copies) underwent ALND. Axillary non-SLNs were routinely examined. ROC curves and Youden's index were performed in order to identify a new cut-off value. Logistic regression models were performed in order to compare OSNA categorical variables created on the basis of our and traditional cut-off to better identify patients who really need an axillary dissection. 69% and 31% of OSNA positive patients had a negative and positive ALND, respectively. ROC analysis identified a cut-off of 2150 CK19 mRNA copies with 95% sensitivity and 51% specificity. Positive and negative predictive values of this new cut-off were 47% and 96%, respectively. Logistic regression models indicated that the cut-off of 2150 copies better discriminates patients with node negative or positive in comparison with the conventional OSNA cut-off (p<0.0001). This cut-off identifies false positive and false negative cases and true-positive and true negative cases very efficiently, and therefore better identifies which patients really need an ALND and which patients can avoid one. This is why we suggest that the negative cut-off should be raised from 250 to 2150. Furthermore, we propose that for patients with a copy number that ranges between 2150 and 5000, there should be a multidisciplinary discussion concerning the clinical and bio-morphological features of primary breast cancer before any decision is taken on whether to perform an ALND or not.

Original languageEnglish
Pages (from-to)e0171517
JournalPLoS One
Volume12
Issue number2
DOIs
Publication statusPublished - 2017

Fingerprint

Keratin-19
Dissection
breast neoplasms
lymph nodes
Nucleic Acids
Lymph Nodes
Amplification
Breast Neoplasms
Messenger RNA
Lymph Node Excision
nucleic acids
Logistic Models
Logistics
ROC Curve
Sentinel Lymph Node
cyhalothrin
Assays
Neoplasm Micrometastasis
Sensitivity and Specificity

Keywords

  • Biomarkers, Tumor
  • Breast Neoplasms
  • Female
  • Humans
  • Keratin-19
  • Lymphatic Metastasis
  • Molecular Diagnostic Techniques
  • Predictive Value of Tests
  • RNA, Messenger
  • Sentinel Lymph Node
  • Journal Article

Cite this

A cut-off of 2150 cytokeratin 19 mRNA copy number in sentinel lymph node may be a powerful predictor of non-sentinel lymph node status in breast cancer patients. / Terrenato, Irene; D'Alicandro, Valerio; Casini, Beatrice; Perracchio, Letizia; Rollo, Francesca; De Salvo, Laura; Di Filippo, Simona; Di Filippo, Franco; Pescarmona, Edoardo; Maugeri-Saccà, Marcello; Mottolese, Marcella; Buglioni, Simonetta.

In: PLoS One, Vol. 12, No. 2, 2017, p. e0171517.

Research output: Contribution to journalArticle

Terrenato, Irene ; D'Alicandro, Valerio ; Casini, Beatrice ; Perracchio, Letizia ; Rollo, Francesca ; De Salvo, Laura ; Di Filippo, Simona ; Di Filippo, Franco ; Pescarmona, Edoardo ; Maugeri-Saccà, Marcello ; Mottolese, Marcella ; Buglioni, Simonetta. / A cut-off of 2150 cytokeratin 19 mRNA copy number in sentinel lymph node may be a powerful predictor of non-sentinel lymph node status in breast cancer patients. In: PLoS One. 2017 ; Vol. 12, No. 2. pp. e0171517.
@article{9895697595b045c9bdb13112338f65f5,
title = "A cut-off of 2150 cytokeratin 19 mRNA copy number in sentinel lymph node may be a powerful predictor of non-sentinel lymph node status in breast cancer patients",
abstract = "Since 2007, one-step nucleic acid amplification (OSNA) has been used as a diagnostic system for sentinel lymph node (SLN) examination in patients with breast cancer. This study aimed to define a new clinical cut-off of CK19 mRNA copy number based on the calculation of the risk that an axillary lymph node dissection (ALND) will be positive. We analyzed 1529 SLNs from 1140 patients with the OSNA assay and 318 patients with positive SLNs for micrometastasis (250 copies) and macrometastasis (5000 copies) underwent ALND. Axillary non-SLNs were routinely examined. ROC curves and Youden's index were performed in order to identify a new cut-off value. Logistic regression models were performed in order to compare OSNA categorical variables created on the basis of our and traditional cut-off to better identify patients who really need an axillary dissection. 69{\%} and 31{\%} of OSNA positive patients had a negative and positive ALND, respectively. ROC analysis identified a cut-off of 2150 CK19 mRNA copies with 95{\%} sensitivity and 51{\%} specificity. Positive and negative predictive values of this new cut-off were 47{\%} and 96{\%}, respectively. Logistic regression models indicated that the cut-off of 2150 copies better discriminates patients with node negative or positive in comparison with the conventional OSNA cut-off (p<0.0001). This cut-off identifies false positive and false negative cases and true-positive and true negative cases very efficiently, and therefore better identifies which patients really need an ALND and which patients can avoid one. This is why we suggest that the negative cut-off should be raised from 250 to 2150. Furthermore, we propose that for patients with a copy number that ranges between 2150 and 5000, there should be a multidisciplinary discussion concerning the clinical and bio-morphological features of primary breast cancer before any decision is taken on whether to perform an ALND or not.",
keywords = "Biomarkers, Tumor, Breast Neoplasms, Female, Humans, Keratin-19, Lymphatic Metastasis, Molecular Diagnostic Techniques, Predictive Value of Tests, RNA, Messenger, Sentinel Lymph Node, Journal Article",
author = "Irene Terrenato and Valerio D'Alicandro and Beatrice Casini and Letizia Perracchio and Francesca Rollo and {De Salvo}, Laura and {Di Filippo}, Simona and {Di Filippo}, Franco and Edoardo Pescarmona and Marcello Maugeri-Sacc{\`a} and Marcella Mottolese and Simonetta Buglioni",
year = "2017",
doi = "10.1371/journal.pone.0171517",
language = "English",
volume = "12",
pages = "e0171517",
journal = "PLoS One",
issn = "1932-6203",
publisher = "Public Library of Science",
number = "2",

}

TY - JOUR

T1 - A cut-off of 2150 cytokeratin 19 mRNA copy number in sentinel lymph node may be a powerful predictor of non-sentinel lymph node status in breast cancer patients

AU - Terrenato, Irene

AU - D'Alicandro, Valerio

AU - Casini, Beatrice

AU - Perracchio, Letizia

AU - Rollo, Francesca

AU - De Salvo, Laura

AU - Di Filippo, Simona

AU - Di Filippo, Franco

AU - Pescarmona, Edoardo

AU - Maugeri-Saccà, Marcello

AU - Mottolese, Marcella

AU - Buglioni, Simonetta

PY - 2017

Y1 - 2017

N2 - Since 2007, one-step nucleic acid amplification (OSNA) has been used as a diagnostic system for sentinel lymph node (SLN) examination in patients with breast cancer. This study aimed to define a new clinical cut-off of CK19 mRNA copy number based on the calculation of the risk that an axillary lymph node dissection (ALND) will be positive. We analyzed 1529 SLNs from 1140 patients with the OSNA assay and 318 patients with positive SLNs for micrometastasis (250 copies) and macrometastasis (5000 copies) underwent ALND. Axillary non-SLNs were routinely examined. ROC curves and Youden's index were performed in order to identify a new cut-off value. Logistic regression models were performed in order to compare OSNA categorical variables created on the basis of our and traditional cut-off to better identify patients who really need an axillary dissection. 69% and 31% of OSNA positive patients had a negative and positive ALND, respectively. ROC analysis identified a cut-off of 2150 CK19 mRNA copies with 95% sensitivity and 51% specificity. Positive and negative predictive values of this new cut-off were 47% and 96%, respectively. Logistic regression models indicated that the cut-off of 2150 copies better discriminates patients with node negative or positive in comparison with the conventional OSNA cut-off (p<0.0001). This cut-off identifies false positive and false negative cases and true-positive and true negative cases very efficiently, and therefore better identifies which patients really need an ALND and which patients can avoid one. This is why we suggest that the negative cut-off should be raised from 250 to 2150. Furthermore, we propose that for patients with a copy number that ranges between 2150 and 5000, there should be a multidisciplinary discussion concerning the clinical and bio-morphological features of primary breast cancer before any decision is taken on whether to perform an ALND or not.

AB - Since 2007, one-step nucleic acid amplification (OSNA) has been used as a diagnostic system for sentinel lymph node (SLN) examination in patients with breast cancer. This study aimed to define a new clinical cut-off of CK19 mRNA copy number based on the calculation of the risk that an axillary lymph node dissection (ALND) will be positive. We analyzed 1529 SLNs from 1140 patients with the OSNA assay and 318 patients with positive SLNs for micrometastasis (250 copies) and macrometastasis (5000 copies) underwent ALND. Axillary non-SLNs were routinely examined. ROC curves and Youden's index were performed in order to identify a new cut-off value. Logistic regression models were performed in order to compare OSNA categorical variables created on the basis of our and traditional cut-off to better identify patients who really need an axillary dissection. 69% and 31% of OSNA positive patients had a negative and positive ALND, respectively. ROC analysis identified a cut-off of 2150 CK19 mRNA copies with 95% sensitivity and 51% specificity. Positive and negative predictive values of this new cut-off were 47% and 96%, respectively. Logistic regression models indicated that the cut-off of 2150 copies better discriminates patients with node negative or positive in comparison with the conventional OSNA cut-off (p<0.0001). This cut-off identifies false positive and false negative cases and true-positive and true negative cases very efficiently, and therefore better identifies which patients really need an ALND and which patients can avoid one. This is why we suggest that the negative cut-off should be raised from 250 to 2150. Furthermore, we propose that for patients with a copy number that ranges between 2150 and 5000, there should be a multidisciplinary discussion concerning the clinical and bio-morphological features of primary breast cancer before any decision is taken on whether to perform an ALND or not.

KW - Biomarkers, Tumor

KW - Breast Neoplasms

KW - Female

KW - Humans

KW - Keratin-19

KW - Lymphatic Metastasis

KW - Molecular Diagnostic Techniques

KW - Predictive Value of Tests

KW - RNA, Messenger

KW - Sentinel Lymph Node

KW - Journal Article

U2 - 10.1371/journal.pone.0171517

DO - 10.1371/journal.pone.0171517

M3 - Article

VL - 12

SP - e0171517

JO - PLoS One

JF - PLoS One

SN - 1932-6203

IS - 2

ER -