Risk factors for lymph node metastasis in non-sentinel node basins in early gastric cancer: sentinel node concept

Masashi Takeuchi, Hiroya Takeuchi, Hirofumi Kawakubo, Ayako Shimada, Tadaki Nakahara, Shuhei Mayanagi, Masahiro Niihara, Kazumasa Fukuda, Rieko Nakamura, Koichi Suda, Norihito Wada, Yuukou Kitagawa

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

Background: Sentinel node (SN) concept is being applied to early gastric cancer. However, when SNs are positive for metastasis, it is unclear how often LNs in other LN basins show metastasis. We aimed to investigate LN metastasis possibility in LN basins without SNs (non-SN basins). We determined risk factors for metastasis in non-SN basins and identified a prediction model for non-SN basin metastasis using classification and regression tree (CART) analysis. Methods: We enrolled 550 patients who were diagnosed with cT1N0M0 or cT2N0M0 gastric cancer with a single lesion and underwent SN mapping. We adopted a dual-tracer method using a radioactive colloid and blue dye to detect SNs. Results: Of all, 45 (8.2%) patients had SN metastasis; we divided them into two groups: LN metastasis positive and LN metastasis negative in non-SN basins. Univariate analysis showed that the groups differed significantly regarding lymphatic invasion (p = 0.007), number of identified SNs (p = 0.032), and macrometastasis in SN basins (p = 0.005). The CART decision tree for predicting LN metastasis in non-SN basins had area under the curve value of 0.86. Moreover, there were significantly differences in cancer-specific survival (CSS) between the two groups (p = 0.028). Conclusions: Macrometastasis in SN basins, lymphatic invasion, and number of identified SNs ≥ 5 are risk factors for LN metastasis in non-SN basins among gastric cancer patients. We identified a prediction model with CART analysis; patients with macrometastasis in SN basins and lymphatic invasion were considered to be at the highest risk for LN metastasis.

Original languageEnglish
Pages (from-to)1-8
Number of pages8
JournalGastric Cancer
DOIs
Publication statusAccepted/In press - 2018 May 24

Fingerprint

Stomach Neoplasms
Lymph Nodes
Neoplasm Metastasis
cyhalothrin
Regression Analysis
Decision Trees
Colloids
Area Under Curve
Coloring Agents
Survival

Keywords

  • Gastrectomy
  • Sentinel lymph node biopsy
  • Stomach neoplasms

ASJC Scopus subject areas

  • Oncology
  • Gastroenterology
  • Cancer Research

Cite this

Risk factors for lymph node metastasis in non-sentinel node basins in early gastric cancer : sentinel node concept. / Takeuchi, Masashi; Takeuchi, Hiroya; Kawakubo, Hirofumi; Shimada, Ayako; Nakahara, Tadaki; Mayanagi, Shuhei; Niihara, Masahiro; Fukuda, Kazumasa; Nakamura, Rieko; Suda, Koichi; Wada, Norihito; Kitagawa, Yuukou.

In: Gastric Cancer, 24.05.2018, p. 1-8.

Research output: Contribution to journalArticle

@article{102d149d0afb44548f7c1b52f1582b25,
title = "Risk factors for lymph node metastasis in non-sentinel node basins in early gastric cancer: sentinel node concept",
abstract = "Background: Sentinel node (SN) concept is being applied to early gastric cancer. However, when SNs are positive for metastasis, it is unclear how often LNs in other LN basins show metastasis. We aimed to investigate LN metastasis possibility in LN basins without SNs (non-SN basins). We determined risk factors for metastasis in non-SN basins and identified a prediction model for non-SN basin metastasis using classification and regression tree (CART) analysis. Methods: We enrolled 550 patients who were diagnosed with cT1N0M0 or cT2N0M0 gastric cancer with a single lesion and underwent SN mapping. We adopted a dual-tracer method using a radioactive colloid and blue dye to detect SNs. Results: Of all, 45 (8.2{\%}) patients had SN metastasis; we divided them into two groups: LN metastasis positive and LN metastasis negative in non-SN basins. Univariate analysis showed that the groups differed significantly regarding lymphatic invasion (p = 0.007), number of identified SNs (p = 0.032), and macrometastasis in SN basins (p = 0.005). The CART decision tree for predicting LN metastasis in non-SN basins had area under the curve value of 0.86. Moreover, there were significantly differences in cancer-specific survival (CSS) between the two groups (p = 0.028). Conclusions: Macrometastasis in SN basins, lymphatic invasion, and number of identified SNs ≥ 5 are risk factors for LN metastasis in non-SN basins among gastric cancer patients. We identified a prediction model with CART analysis; patients with macrometastasis in SN basins and lymphatic invasion were considered to be at the highest risk for LN metastasis.",
keywords = "Gastrectomy, Sentinel lymph node biopsy, Stomach neoplasms",
author = "Masashi Takeuchi and Hiroya Takeuchi and Hirofumi Kawakubo and Ayako Shimada and Tadaki Nakahara and Shuhei Mayanagi and Masahiro Niihara and Kazumasa Fukuda and Rieko Nakamura and Koichi Suda and Norihito Wada and Yuukou Kitagawa",
year = "2018",
month = "5",
day = "24",
doi = "10.1007/s10120-018-0840-z",
language = "English",
pages = "1--8",
journal = "Gastric Cancer",
issn = "1436-3291",
publisher = "Springer Japan",

}

TY - JOUR

T1 - Risk factors for lymph node metastasis in non-sentinel node basins in early gastric cancer

T2 - sentinel node concept

AU - Takeuchi, Masashi

AU - Takeuchi, Hiroya

AU - Kawakubo, Hirofumi

AU - Shimada, Ayako

AU - Nakahara, Tadaki

AU - Mayanagi, Shuhei

AU - Niihara, Masahiro

AU - Fukuda, Kazumasa

AU - Nakamura, Rieko

AU - Suda, Koichi

AU - Wada, Norihito

AU - Kitagawa, Yuukou

PY - 2018/5/24

Y1 - 2018/5/24

N2 - Background: Sentinel node (SN) concept is being applied to early gastric cancer. However, when SNs are positive for metastasis, it is unclear how often LNs in other LN basins show metastasis. We aimed to investigate LN metastasis possibility in LN basins without SNs (non-SN basins). We determined risk factors for metastasis in non-SN basins and identified a prediction model for non-SN basin metastasis using classification and regression tree (CART) analysis. Methods: We enrolled 550 patients who were diagnosed with cT1N0M0 or cT2N0M0 gastric cancer with a single lesion and underwent SN mapping. We adopted a dual-tracer method using a radioactive colloid and blue dye to detect SNs. Results: Of all, 45 (8.2%) patients had SN metastasis; we divided them into two groups: LN metastasis positive and LN metastasis negative in non-SN basins. Univariate analysis showed that the groups differed significantly regarding lymphatic invasion (p = 0.007), number of identified SNs (p = 0.032), and macrometastasis in SN basins (p = 0.005). The CART decision tree for predicting LN metastasis in non-SN basins had area under the curve value of 0.86. Moreover, there were significantly differences in cancer-specific survival (CSS) between the two groups (p = 0.028). Conclusions: Macrometastasis in SN basins, lymphatic invasion, and number of identified SNs ≥ 5 are risk factors for LN metastasis in non-SN basins among gastric cancer patients. We identified a prediction model with CART analysis; patients with macrometastasis in SN basins and lymphatic invasion were considered to be at the highest risk for LN metastasis.

AB - Background: Sentinel node (SN) concept is being applied to early gastric cancer. However, when SNs are positive for metastasis, it is unclear how often LNs in other LN basins show metastasis. We aimed to investigate LN metastasis possibility in LN basins without SNs (non-SN basins). We determined risk factors for metastasis in non-SN basins and identified a prediction model for non-SN basin metastasis using classification and regression tree (CART) analysis. Methods: We enrolled 550 patients who were diagnosed with cT1N0M0 or cT2N0M0 gastric cancer with a single lesion and underwent SN mapping. We adopted a dual-tracer method using a radioactive colloid and blue dye to detect SNs. Results: Of all, 45 (8.2%) patients had SN metastasis; we divided them into two groups: LN metastasis positive and LN metastasis negative in non-SN basins. Univariate analysis showed that the groups differed significantly regarding lymphatic invasion (p = 0.007), number of identified SNs (p = 0.032), and macrometastasis in SN basins (p = 0.005). The CART decision tree for predicting LN metastasis in non-SN basins had area under the curve value of 0.86. Moreover, there were significantly differences in cancer-specific survival (CSS) between the two groups (p = 0.028). Conclusions: Macrometastasis in SN basins, lymphatic invasion, and number of identified SNs ≥ 5 are risk factors for LN metastasis in non-SN basins among gastric cancer patients. We identified a prediction model with CART analysis; patients with macrometastasis in SN basins and lymphatic invasion were considered to be at the highest risk for LN metastasis.

KW - Gastrectomy

KW - Sentinel lymph node biopsy

KW - Stomach neoplasms

UR - http://www.scopus.com/inward/record.url?scp=85047330204&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85047330204&partnerID=8YFLogxK

U2 - 10.1007/s10120-018-0840-z

DO - 10.1007/s10120-018-0840-z

M3 - Article

C2 - 29799060

AN - SCOPUS:85047330204

SP - 1

EP - 8

JO - Gastric Cancer

JF - Gastric Cancer

SN - 1436-3291

ER -