Author manuscript, published in "European Journal of Surgical Oncology (EJSO) 37, 9 (2011) 747" DOI : 10.1016/j.ejso.2011.06.018
Accepted Manuscript Title: Lymph node metastases and prognosis in oesophageal carcinoma- a systematic review
peer-00723456, version 1 - 10 Aug 2012
Authors: B. Kayani, MRCS, MBBS, BSc E. Zacharakis, MD, PhD K. Ahmed, MRCS, MBBS G.B. Hanna, PhD, FRCS PII:
S0748-7983(11)00355-6
DOI:
10.1016/j.ejso.2011.06.018
Reference:
YEJSO 3192
To appear in:
European Journal of Surgical Oncology
Accepted Date: 28 June 2011
Please cite this article as: Kayani B, Zacharakis E, Ahmed K, Hanna GB. Lymph node metastases and prognosis in oesophageal carcinoma- a systematic review, European Journal of Surgical Oncology (2011), doi: 10.1016/j.ejso.2011.06.018 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Lymph node metastases and prognosis in oesophageal carcinoma- a systematic review
RI PT
BKayani (MRCS MBBS BSc), E Zacharakis (MD PhD), K Ahmed (MRCS MBBS), GB Hanna (PhD FRCS)
SC
Department of Surgery and Cancer, Imperial College London, St Mary’s Hospital, Praed
Mr Emmanouil Zacharakis MD PhD Clinical Senior Lecturer in Surgery
M AN U
Address for correspondence:
TE D
Consultant General & Upper Gastrointestinal Surgeon
Department of Surgery and Cancer, Imperial College, St Mary’s Hospital, Praed Street, London W2 1NY, UK
EP
Tel: +44 (0)20 331 21012
Fax: +44 (0)20 331 26950
Email:
[email protected]
AC C
peer-00723456, version 1 - 10 Aug 2012
Street, London W2 1NY, UK
ACCEPTED MANUSCRIPT
Abstract Oesophageal cancer is the seventh most common cause of cancer-related death in the developed world and the incidence of oesophageal adenocarcinoma is now the fastest
RI PT
growing of any gastrointestinal cancer. Lymph node involvement is the single most important prognostic factor in oesophageal cancer. Imaging to determine the extent of
lymph node involvement and plan treatment often requires a combination of modalities to
SC
avoid under-staging. The seventh edition of the staging system released by the
International Union Against Cancer (IUCC) has stratified lymph node involvement
M AN U
of metastatic lymph node involvement. This review discusses the prognostic and treatment implications of these modifications and explores micrometastatic lymph node involvement, capsular infiltration and lymph node ratio as possible additions to the staging
EP
TE D
system.
AC C
peer-00723456, version 1 - 10 Aug 2012
according to the number of lymph nodes involved and redefined its groupings for location
ACCEPTED MANUSCRIPT
Introduction
Identification of lymph node metastases on histopathological examination is the single
RI PT
most important prognostic factor in oesophageal cancer [1-10]. The overall 5-year survival rate after surgical resection is between 70% and 92% for patients without nodal involvement, but only 18% to 47% for patients with lymph node metastasis [8,10]. It is an
SC
accurate predictor of disease-free survival and can be used to identify patients who may require adjuvant chemotherapy or chemoradiotherapy to treat systemic metastases
M AN U
The International Union Against Cancer (UICC) staging system for oesophageal carcinoma stratifies patients into prognostic groups based on their TNM score. However, several authors have criticised the TNM oesophageal staging system, claiming that it
TE D
poorly stratifies what is a heterogeneous group of patients who have variable prognostic outcomes and undergo different treatment regimens [12-23]. The UICC has responded to these recommendations in its seventh edition of TNM staging, released in 2010, by
EP
modifying all three aspects of the scoring system. Nevertheless, the reliability and accuracy of the nodal portion of the TNM staging system still remains the epicentre of much debate as it omits some potentially critical prognostic information relating to lymph
AC C
peer-00723456, version 1 - 10 Aug 2012
developing after primary resection [10,11].
node status. This review discusses the changes to the nodal staging system and their treatment implications and also explores micrometastases, capsular infiltration and lymph node ratio as possible additions to the staging system.
Methods:
Electronic searches were performed using PubMed, Embase and the Cochrane library for
ACCEPTED MANUSCRIPT
nodal factors affecting prognosis in Oesophageal Cancer. The following Medical Subject Headings (MeSH) were used to carry out a systematic search of the literature: “Oesophageal”, “Cancer”, “Staging”, “Imaging”, “Lymph node”, “Micrometastases”,
RI PT
“Capsule”, “Location” and “Prognosis.” Articles were selected based on pre-defined inand exclusion criteria based on literature published between January 1990 and January 2011 (Figure 1). The reference lists from the retrieved articles were also reviewed for
SC M AN U
Literature Search: 212Articles
27 Articles excluded after not fulfilling inclusion criteria: 1. Study Population: Humans 2. Language: English 3. Study subject: Lymph node disease 4. Patient Population: Patients with Oesophageal Cancer
TE D
41 Articles excluded after reviewing title and abstract
144 Articles remaining after selection process
EP
67 Articles selected from references
111 Articles excluded after review of full paper
AC C
peer-00723456, version 1 - 10 Aug 2012
additional articles. In total, one hundred articles were included.
100 Articles included in Reviewarticle
Fig1: Flow chart showing selection process of articles for review of nodal staging in oesophageal cancer
Results:
ACCEPTED MANUSCRIPT
Of 100 included articles, 67 were retrospective studies, 23 were prospective cohort studies, and six were randomised controlled trials. Furthermore, seven studies were review articles,
RI PT
two studies were meta-analyses and one article contained the seventh edition of TNM staging by UICC. Studies for both squamous cell carcinoma and adenocarcinoma of the
oesophagus were included. The review is up to date with articles published up until May
SC
2011.
M AN U
Preoperative staging of lymph nodes
Estimate of the extent of lymph node metastases before surgery is crucial for planning the
TE D
treatment strategy and for predicting the prognosis. Preoperatively, the presence of malignant regional lymph nodes can be determined with endoscopic ultrasound (EUS), computed tomography (CT), and 18F-Fluoro-2-deoxy-D-glucose Positron Emission
EP
Tomography (FDG-PET).
Endoscopic Ultrasound (EUS)
AC C
peer-00723456, version 1 - 10 Aug 2012
Discussion:
Endoscopic ultrasound provides useful information about lymph node metastases and the depth of tumour invasion before surgery. It utilises an echo-endoscope which is able to visualise the individual histological layers of the oesophagus. The number of lymph nodes detected by this method correlates significantly with 5-year survival rates [24] and EUS detects more locoregional node involvement than CT or PET [25-27]. Preoperative
ACCEPTED MANUSCRIPT
endoscopic diagnosis of regional lymph node metastases has a specificity of 70% (95% CI: 65-65%) and a sensitivity of 80% (95% CI: 75-84%) [25]. Combining endoscopic Ultrasound-guided Fine Needle Aspiration Cytology (FNAC) with CT further improves
RI PT
the diagnostic accuracy of local lymph node involvement [28].
Knowledge of celiac and other abdominal lymph node metastases is important for tumour
SC
staging and management. In the case of the former, EUS is the preferred diagnostic
technique. However, not all metastatic abdominal lymph nodes can be detected in the same
M AN U
may cause luminal obstruction, thus limiting the passage of the endoscope. Furthermore, skip metastases may occur, in which distant lymph nodes are involved, without regional lymph node disease [29, 30].In view of these facts, other imaging methods such as CT and
TE D
PET scans are also used for accurate staging of the disease.
CT and FDG-PET scans:
EP
Both CT and FDG-PET scans are used in the initial staging of nodal disease, evaluation of response to neoadjuvant therapy, and detection of recurrence. Computed tomography focuses on three areas: local invasion, lymph node status, and distant metastases [31-33].
AC C
peer-00723456, version 1 - 10 Aug 2012
way, as the probe has a limited depth penetration of about 5 cm and increased tumour size
With respect to regional lymph node status, the value of CT scanning is limited, as it has a sensitivity of 50% (95% CI: 41-60%) and specificity of 83% (95% CI 0.77–0.89) [25]. Furthermore, FDG-PET has a sensitivity of 57% (95% CI: 43-70%) and specificity of 85% (95% CI 76–95%) in detecting regional lymph node metastases [25]. In one particular study, patients with no evidence of metastases on CT scan, FDG-PET identified distant metastatic disease in 4.8% (95% CI: 2.2-8.9%) of patients [34]. However, these FDG-PET-
ACCEPTED MANUSCRIPT
detected metastases should be confirmed before excluding a patient from surgery, as at least 3.7% (95% CI: 1.5-7.5%) are false positives [34]. Some 5% of patients will have
RI PT
metastatic disease that evades detection by CT or FDG-PET [29, 34].
Multimodality imaging
SC
As there is no definitive test providing information for the staging of nodal status,
clinicians often use a combination of EUS, CT and PET to minimise the risk of under-
M AN U
regional lymph node metastases. It is typically used to exclude the presence of regional lymph node metastases, whereas positive disease can be confirmed with FNAC, CT or PET [25].
TE D
Micrometastases in lymph nodes
Lymph node micrometastases are small metastatic lesions, defined as clusters of five or
EP
fewer tumour cells in a lymph node, which are not detected by routine histological examination [35,
36]. These micrometastatic deposits can be detected with
immunohistochemical analysis using monoclonal antibodies to stain tumour-specific
AC C
peer-00723456, version 1 - 10 Aug 2012
staging. EUS demonstrates the highest sensitivity but also the lowest specificity for
antigens or using Polymerase Chain reaction (PCR) at the molecular level [37-44]. It is unclear whether these nodal micrometastases represent tumour cells with metastatic potential or just shredded cells, which will be cleared by the immune system or remain dormant in a non-proliferating phase [39, 45].
Prognostic implications of micrometastases:
ACCEPTED MANUSCRIPT
In oesophageal cancer, it has been shown that between 11% and 50% of pN0 patients have nodal micrometastases [46-54]. There is a strong body of evidence suggesting that
RI PT
immunohistochemically detected micrometastatic deposits in oesophageal cancer are associated with reduced survival and increased risk of disease recurrence [55-65]. Several studies have identified micrometastatic spread to be an independent prognostic indicator in
SC
oesophageal adenocarcinoma [35, 39, 41, 43, 66] and it has been postulated that these
micrometastatic deposits may help to explain why up to 50% of histologically node-
M AN U
the primary tumour [67, 68].
On the other hand, other studies have shown prognosis in immunohistochemically positive lymph nodes to be equal to that detected by conventional histological testing [39, 53, 63,
TE D
69]. Others have found no prognostic significance [50,55,56]. The conflicting evidence arises because of variability in the numbers of retrieved lymph nodes, differences in the mean follow-up times and pooling of data for squamous cell carcinoma and
EP
adenocarcinoma despite the two malignancies having different biological and epidemiological profiles.
AC C
peer-00723456, version 1 - 10 Aug 2012
negative oesophageal cancer patients develop recurrence, despite a curative resection of
Predicting nodal micrometastatic involvement
Nodal micrometastases have been shown to correlate with lymphatic infiltration [70], tumour size [71],vascular invasion [67] and risk of lymph node recurrence [56]. However, there is no evidence as yet that correlates these occult deposits to the location or timing of nodal recurrence. As hypothesised in Paget’s seed-and-soil hypothesis, the growth of
ACCEPTED MANUSCRIPT
disseminated tumour cells into overt metastases is determined by the presence of growth factors in the environment of the cell and the ability of the cell to respond to them. Since there is likely to be significant interpersonal and internodal variation in these factors, it
RI PT
may prove much more difficult to predict the timing and location of recurrences. At present, there are no accepted guidelines for upstaging nodal status even if
M AN U
Extracapsular lymph node involvement is the term used to describe extension of cancer cells though the nodal capsule into the perinodal fatty tissue [54]. This is present in 57% (95% CI: 53-61%) of node positive patients with oesophageal carcinoma [68, 72-75].
TE D
Prognostic implications of extracapsular disease
Extracapsular lymph node involvement is associated with increased disease recurrence in
EP
patients with local infiltration in the neck/mediastinum. The incidence of recurrence is 21% with intracapsular disease compared to 60% with extracapsular disease [76,77]. Additionally, the median survival in patients with intracapsular disease is 41 months
AC C
peer-00723456, version 1 - 10 Aug 2012
Extracapsular lymph node involvement
SC
micrometastases are detected by immunohistochemical analysis.
(reported range 19-64 months) but only 15 months (reported range 12-18 months) with extracapsular lymph node involvement [76]. The presence of extracapsular nodal invasion is an independent prognostic factor [68, 74, 75, 78-80] and can be used to stratify patients according to the number of lymph nodes in which it is present. For example, in patients with extracapsular lymph node infiltration in only one lymph node, the median survival is
ACCEPTED MANUSCRIPT
21 months (95% CI: 13-28 months) but if present in two or more lymph nodes, the median survival drops to 12 months (95% CI: 8-15 months) [76].
RI PT
Extracapsular lymph node involvement has been associated with a number of clinicopathological features [68, 72-73] including T stage [76], N stage [14, 81], and number of
positive lymph nodes [14]. The increased risk of recurrence and reduced survival in
SC
patients with extracapsular disease may be a reflection of the invasiveness and aggressiveness of the primary tumour [77, 82]. Studies have indeed demonstrated that a
M AN U
of nodes with extracapsular infiltration [14, 77]. However, it is unknown whether this occurs early or late in cancer progression. The “metastases to metastases theory” indicates that extracapsular metastatic deposits give rise to newer deposits, thus increasing the speed
TE D
of dissemination and reducing survival [76].
Number of positive lymph nodes
EP
The changes from the sixth to the seventh edition of the UICC staging system for regional node staging in oesophageal cancer are shown in figures two and three [83]. These modifications reflect findings which show improved prognostic stratification with
AC C
peer-00723456, version 1 - 10 Aug 2012
significant correlation exists between the number of positive lymph nodes and the number
consideration of the number ofdiseased lymph nodes [13, 84-86]. Several studies have shown the number of positive lymph nodes to be an TnQTable1Nodal
independent prognostic indicator [13,14, 87-90].
Figure
3:
Sixth
edition
Figure 4: Seventh edition (2010)
(2002):
Lymph node involvement
stage Nx
Regional nodes cannot be assessed
N0
No regional lymph node metastases
N1
1 - 2 regional lymph node metastases
ACCEPTED MANUSCRIPT
N2
3 – 6 regional lymph node metastases
N3
Greater than 6 regional lymph node
RI PT
metastases
Solitary lymph node involvement
SC
It has been demonstrated that the presence of a solitary metastatic node - though an indicator of a poorer prognostic outcome compared with pN0 disease, is associated with
M AN U
example, when comparing patients with solitary lymph node involvement, two or three metastatic lymph nodes and greater than three lymph nodes the median survival was 17 months [Hazard Ratio-1, P