Zhou K, Li ZW, Wu Y, Wang ZJ, Wang LQ, Zhou LX, Jia L, Ji K, Yang XS, Zhang J, Wu XJ, Wang AQ, Bu ZD. Lymph node metastatic patterns of gastric carcinoma with a combination of adenocarcinoma and neuroendocrine carcinoma components. World J Gastroenterol 2025; 31(8): 102347 [DOI: 10.3748/wjg.v31.i8.102347]
Corresponding Author of This Article
Zhao-De Bu, PhD, Professor, Department of Gastrointestinal Surgery, Peking University Cancer Hospital and Institute, No. 52 Fucheng Road, Haidian District, Beijing 100142, China. buzhaode@cjcrcn.org
Research Domain of This Article
Gastroenterology & Hepatology
Article-Type of This Article
Retrospective Cohort Study
Open-Access Policy of This Article
This article is an open-access article which was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/
Kai Zhou, Zhi-Jie Wang, Ling-Qian Wang, Ke Ji, Xue-Song Yang, Ji Zhang, Xiao-Jiang Wu, An-Qiang Wang, Zhao-De Bu, Department of Gastrointestinal Surgery, Peking University Cancer Hospital and Institute, Beijing 100142, China
Zhong-Wu Li, Yan Wu, Li-Xin Zhou, Ling Jia, Department of Pathology, Key Laboratory of Carcinogenesis and Translational Research (Ministry of Education), Peking University Cancer Hospital and Institute, Beijing 100142, China
Co-corresponding authors: An-Qiang Wang and Zhao-De Bu.
Author contributions: Wang AQ and Bu ZD contributed to conceptualization, funding acquisition, project administration; Zhou K, Wu Y, Li ZW, Wang ZJ, Wang LQ, Zhou LX, Jia L, Ji K, Yang XS contributed to data curation; Zhou K contributed to formal analysis, software and writing original draft; Zhou K, Wu Y, Li ZW contributed to validation, visualization and investigation; Zhou K, Wang AQ contributed to methodology; Bu ZD, Li ZW , Zhang J, Wu XJ contributed to resources; Wang AQ, Bu ZD, Li ZW contributed to supervision; Wang AQ, Wu Y, Bu ZD contributed to writing, review and editing.
Supported by the National Key Research and Development Program of China, No. 2023YFF1204702; the National Natural Science Foundation of China, No. 82173151; Capital’s Funds for Health Improvement and Research, No. CFH 2022-4-1025; Beijing Hospitals Authority Clinical Medicine Development of Special Funding, No. XMLX202119; and Science Foundation of Peking University Cancer Hospital, No. PY202329.
Institutional review board statement: The study was reviewed and approved by the Medical Ethics Committee of Beijing Cancer Hospital, Institutional Review Board (Approval No. 2022KT04).
Informed consent statement: All study participants, or their legal guardian, provided informed written consent prior to study enrollment.
Conflict-of-interest statement: The authors declare that they have no conflict of interest.
STROBE statement: The authors have read the STROBE Statement—a checklist of items, and the manuscript was prepared and revised according to the STROBE Statement-a checklist of items.
Data sharing statement: Technical appendix, statistical code, and dataset available from the corresponding author at buzhaode@cjcrcn.org.
Open Access: This article is an open-access article that was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution NonCommercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: https://creativecommons.org/Licenses/by-nc/4.0/
Corresponding author: Zhao-De Bu, PhD, Professor, Department of Gastrointestinal Surgery, Peking University Cancer Hospital and Institute, No. 52 Fucheng Road, Haidian District, Beijing 100142, China. buzhaode@cjcrcn.org
Received: October 16, 2024 Revised: December 14, 2024 Accepted: January 7, 2025 Published online: February 28, 2025 Processing time: 98 Days and 20.7 Hours
Abstract
BACKGROUND
Gastric mixed-adenoneuroendocrine carcinoma (G-MANEC) is a subtype of gastric cancer. Building upon prior research findings, we propose that tumours containing both neuroendocrine carcinoma (NEC) and adenocarcinoma (AC) components, with each component ranging from 1% to 99% of the tumour, be classified as a distinct entity. We hereby term this adenoneuroendocrine mixed gastric cancer (G-ANEC). Research on lymph node (LN) involvement in G-MANEC has focused mainly on metastasis status, with limited studies on metastatic composition.
AIM
To investigate the LN metastasis patterns of G-ANEC, the clinicopathological features associated with these metastasis patterns, and to explore adjuvant chemotherapy regimens for G-ANEC.
METHODS
We analyzed 68 G-ANEC cases treated with radical surgery and confirmed LN metastasis at Peking University Cancer Hospital between August 2012 and June 2022. Utilizing χ2 tests in IBM statistical product and service solutions statistics and R software.
RESULTS
We identified three distinct LN metastasis patterns in G-ANEC that were significantly associated with the NEC proportion, tumour invasion depth, Lauren classification, and tumour location (P values: 0.008, 0.015, 0.01, and 0.004, respectively). When the SOX/XELOX regimen was applied for adjuvant chemotherapy, patients with LN metastasis comprising only AC exhibited better overall survival (OS) (94.25 ± 11.07 months vs 54.36 ± 11.36 months) than did those with NEC. When LN metastasis components contained NEC, there was a trend towards improved OS (64 ± 10.77 months vs 54.35 ± 11.36 months) and disease-free survival (71.28 ± 9.92 months vs 66.28 ± 11.93 months) in patients treated with the etoposide and cisplatin compared to those receiving the SOX/XELOX regimen.
CONCLUSION
We found a significant correlation between the NEC percentage, tumour invasion depth, Lauren classification, and tumour location and LN metastasis patterns in G-ANEC. For G-ANEC, a lower proportion of NEC or AC in the primary lesion does not preclude the possibility of these components metastasizing to the LNs. Different adjuvant chemotherapy regimens should be administered on the basis of the varying components of LN metastasis in patients with G-ANEC.
Core Tip: Rare individual studies on the lymph node metastasis components and treatment issues of adenoneuroendocrine mixed gastric cancer (G-ANEC). For G-ANEC, even if the neuroendocrine carcinoma (NEC) or adenocarcinoma (AC) component is relatively small in the primary lesion, it may still metastasize to the lymph nodes (LNs). If the metastasis in the LNs consists only of AC components, treatment of G-ANEC should primarily be based on fluoropyrimidine drugs. If the metastasis in the LNs includes NEC components, treatment should primarily be based on the etoposide and cisplatin/irinotecan and cisplatin regimen.
Citation: Zhou K, Li ZW, Wu Y, Wang ZJ, Wang LQ, Zhou LX, Jia L, Ji K, Yang XS, Zhang J, Wu XJ, Wang AQ, Bu ZD. Lymph node metastatic patterns of gastric carcinoma with a combination of adenocarcinoma and neuroendocrine carcinoma components. World J Gastroenterol 2025; 31(8): 102347
Gastric mixed-adenoneuroendocrine carcinoma (G-MANEC) is a particular type of gastric mixed neuroendocrine-non-neuroendocrine neoplasms (G-MiNEN), characterized by the presence of both neuroendocrine carcinoma (NEC) and adenocarcinoma (AC) components[1], with each component comprising more than 30% of total tumour volume[2]. The G-MANEC diagnostic criterion designation of a 30% threshold is deemed arbitrary. In 1987, this threshold was suggested to aid pathologists in identifying and establishing a novel tumour diagnostic entity[3]. Nevertheless, the absence of evidence-based medical justification undermines the validity of this threshold[4]. In 2006, Volante reported that the 30% restriction between NEC and AC lacks a plausible etiological and histogenetic explanation[5]. Previous studies have demonstrated that when the proportion of NEC in the primary lesion is less than 30%, 10%, or even 5%, the overall survival (OS) time is inferior to that of gastric AC. Nevertheless, there is no statistically significant disparity in survival prognosis among groups with primary lesion NEC exceeding 30%, 10%, or 5%[6-8]. Consequently, it is postulated that even a lesser proportion of NEC in the primary lesion may have a detrimental impact on OS. Additionally, the prevailing consensus in the field suggests that the NEC and AC components in G-MANEC originate from a single multi-potent epithelial stem cell[9]. Therefore, mixed gastric cancer comprising AC and NEC constituents, irrespective of the NEC proportion, should be categorized as a singular tumour type. As distinction from G-MANEC with rigorous definition, we denoted tumours with AC and NEC components as adenoneuroendocrine mixed gastric cancer (G-ANEC).
In the cases of G-ANEC, both the NEC and AC components all have the potential to metastasize either concurrently or independently. However, research pertaining to metastatic components remains insufficient. Previous studies on G-MANEC have focused mostly on whether there is lymph node (LN) metastasis[10], often conflating G-NEC, gastric neuroendocrine tumour, and G-MANEC together, or mixing them with other types of gastroenteropancreatic neuroendocrine tumours. Our study represents the largest-scale retrospective analysis of the LN metastasis pattern in patients undergoing curative treatment for G-ANEC. Understanding the metastatic characteristics and patterns of LN metastasis in G-ANEC contributes to a better comprehension of the biological behaviour of this type of disease. This study lays a theoretical foundation for further research on molecular mechanisms, provides a reference for further therapeutic options, and aids in optimizing individualized treatment plans.
MATERIALS AND METHODS
Patient inclusion
This study included patients who underwent curative surgery for G-ANEC at Peking University Cancer Hospital between August 2012 and June 2022. Their postoperative pathology of these patients distinctly identified only two components, namely, NEC and AC, with respective proportions ranging from 1% to 99%. Moreover, these patients all had a definitive presence of local LN metastasis. The exclusion criteria were patients with significant missing clinicopathological information, and patients whose hematoxylin-eosin (HE)-stained histological sections were reassessed by two pathologists and ascertained as having gastric AC with neuroendocrine differentiation or pure NEC. Our study features one of the largest cohorts in single-center retrospective research of G-ANEC.
Data collection
Patient information was obtained from the electronic medical record system, encompassing clinical data such as sex, age, tumour location, neoadjuvant chemotherapy (NAC) status, was obtained from the electronic medical record system. For each case, all pathological data were independently reviewed and confirmed by two pathologists. Slides of the primary tumour and all metastatic LNs were reexamined, and the following parameters were collected: The proportion of AC and NEC components in the primary lesion, the components in the metastatic LNs, postoperative pathological tumour invasion depth (pT stage), lymphatic invasion status, perineural invasion status, Lauren types, LN periosteal violation, synaptophysin (Syn), chromogranin A (CgA), and neural cell adhesion molecule 1, cluster of differentiation (CD) 56, the histological subtypes (large cell, small cell, or both), the level of tissue differentiation and grading, mitotic rate, Ki-67 index. The diagnosis of NEC components is based on typical NEC histological morphology and positive immunohistochemical staining for Syn, CgA, or CD56. The determination of pT stage was conducted in accordance with the American Joint Committee on Cancer (AJCC) 8th Edition cancer staging criteria.
Follow-up assessments were primarily carried out through telephone conversations or outpatient clinic appointments. The term OS refers to the length of time between the initial neoadjuvant or curative surgery and the occurrence of death from any cause or the most recent follow-up. Similarly, disease-free survival time (DFS) is defined as the duration from the first neoadjuvant or curative surgery to the point of gastric cancer recurrence, death from any cause, or the last follow-up.
Statistical analysis
The analysis of the associations between different clinicopathological features and LN metastatic components was conducted using the χ2 test in IBM statistical product and service solutions statistics (version 29). The significance of the relationship between each clinicopathological feature and LN metastatic component was determined by examining the χ2 and P values. Post hoc testing was employed to address the intricate nature of pairwise comparisons among clinicopathological feature groups. Differences between groups were assessed using adjusted standardized residuals, employing an absolute value threshold of 3 to ensure a more cautious estimation. Clinicopathological features with a P value of less than 0.05 were incorporated into the multinomial logistic regression analysis to enhance the reliability of the findings. Heatmaps were generated using the R software (version 2023) to visualize the associations between different clinicopathological features and OS and DFS. Statistical significance was determined by considering a P value of less than 0.05. Kaplan-Meier survival analysis curves were constructed using GraphPad Prism, while Origin 2022 was used to create violin plots.
RESULTS
Recruitment and follow-up results
We retrospectively collected clinical, pathological, and survival data from patients diagnosed with G-MANEC who underwent curative surgery at our hospital between August 2012 and June 2022, using a case retrieval system. The pathology database was further examined to identify 73 patients with a pathological diagnosis of G-MANEC, defined as tumours in which both adenocarcinoma and neuroendocrine carcinoma components each account for at least 30, according to the 2019 World Health Organization classification of tumours of the digestive system. Additionally, 92 patients with tumours containing AC and NEC components, where one of the components accounted for less than 30% of the tumour, were also included in the study. In total, 165 patients (73 + 92) with G-ANEC were enrolled. Tumour stages were assigned based on the AJCC 8th Edition cancer staging system. The exclusion criteria for patients were as follows. Two pathologists reviewed the primary lesion HE-stained slides, and 35 patients were diagnosed with gastric AC with neuroendocrine differentiation (immunohistochemical Syn, CgA, and CD56 positivity, without typical NEC histological morphology), and 6 patients were rediagnosed with pure G-NEC (NEC component accounted for 100%) were thus excluded. Information on the clinicopathological characteristics was missing for 25 patients. Further screening excluded 31 patients who did not develop LN metastasis. Ultimately, 68 patients with LN metastasis who underwent curative surgery were included in this study. A selection flow diagram of patients is shown in Supplementary Figure 1. All patients were followed up, with a follow-up duration ranging from 1 to 120 months. The average follow-up period was 38.8 ± 31.73 months.
LN metastatic patterns of G-AECs
A total of 68 patients with G-ANEC who had LN metastasis and who underwent curative surgery were ultimately included. The total number of LNs was 2335, of which 426 were positive (426/2335). Among the 27 patients who presented exhibited LNs solely with NEC metastasis (157/809), 27 patients presented AC metastasis (143/1016), 7 patients presented both components within the same LN (66/271), and 7 patients displayed distinct components in separate LNs (some LN metastases consisted of pure NEC or pure AC, and/or some LN metastases contained both components within a single LN) (60/236). The LN metastasis patterns of G-ANEC can be described as follows: Firstly, LN metastasis may occur solely with one of the components (Figure 1A and B). Secondly, both components may exhibit LN metastasis. The latter scenario, it can be further classified into LN containing only one component (Figure 1C) and a single LN encompassing both components (Figure 1D).
Figure 1 Lymph node metastasis pattern of adenoneuroendocrine mixed gastric cancer.
A: The components of lymph node metastasis are purely neuroendocrine carcinoma; B: The components of lymph node metastasis are purely adenocarcinoma; C: Different lymph nodes metastasize different components; D: The same lymph node contains both neuroendocrine carcinoma and adenocarcinoma metastatic components.
Clinicopathological characteristics and their associations with components of LN metastasis
The associations between the patients’ clinicopathological characteristics and the components of LN metastasis are shown in Table 1. The research findings revealed a notable disparity between the prevalence of NEC (categorical variable) at the primary tumour site and its presence in LN metastasis, which was statistically significant (χ2 = 12.604, P = 0.008). Specifically, when the proportion of NEC in the primary tumour ranged from 71% to 99%, there was a higher likelihood of NEC metastasis in the LN (adjusted standardized residual of 3.0), whereas the likelihood of AC component metastasis in the LN decreased (adjusted standardized residual of -3.0). When the proportion of NEC in the primary tumour ranged from 30% to 70%, the likelihood of metastatic LNs involving both NEC and AC components was significantly low (adjusted standardized residual = -3.0). Nevertheless, the χ2 analysis results and the violin plot (Figure 2) suggest that in the primary lesion, a higher percentage of NEC (continuous variable) or AC correlated with an increased probability of LN metastasis involving NEC or AC, respectively. However, even when the percentage of NEC or AC in the primary tumour was relatively low, there was still a possibility for LN metastasis components to be either NEC or AC.
Figure 2 Violin plot analysis of neuroendocrine carcinoma proportions and lymph node metastasis components probabilities in adenoneuroendocrine mixed gastric cancer.
The violin plot illustrates the fluctuating proportions of neuroendocrine carcinoma in the primary lesion of adenoneuroendocrine mixed gastric cancer, as well as the probabilities associated with different lymph node metastatic components. NEC: Neuroendocrine carcinoma; AC: Adenocarcinoma; LN: Lymph node.
Table 1 Clinicopathological features among lymph node metastasis patterns, n (%).
Parameters
NEC
AC
Both
Total
χ2 value
P value
pT stage
8.36
0.015
pT1-2 stage
3 (25)
9 (75)
0 (0)
12 (100)
pT3-4 stage
24 (42.9)
18 (32.1)
14 (25)
56 (100)
LVI
0.917
0.68
-
4 (57.1)
2 (28.6)
1 (14.3)
7 (100)
+
23 (37.7)
25 (41)
13 (21.3)
61 (100)
PNI
1.267
0.582
-
10 (50)
7 (35)
3 (15)
20 (100)
+
17 (35.4)
20 (41.7)
11 (22.9)
48 (100)
NEC percentage (%)
12.604
0.008
1-29
1 (12.5)
5 (62.5)
2 (25)
8 (100)
30-70
13 (31.7)
20 (48.8)
8 (19.5)
41 (100)
71-99
13 (68.4)
2 (10.5)
4 (21.1)
19 (100)
Histology
3.350
0.484
Small cell
9 (40.9)
8 (36.4)
5 (22.7)
22 (100)
Large cell
14 (34.1)
18 (43.9)
9 (22)
41 (100)
Both
4 (80)
1 (20)
0 (0)
5 (100)
Lauren type
12.46
0.01
Intestinal type
23 (56.1)
12 (29.3)
6 (14.6)
41 (100)
Diffuse type
1 (9.1)
6 (54.5)
4 (36.4)
11 (100)
Mixed type
3 (18.8)
9 (56.3)
4 (25)
16 (100)
LN periosteal violation
0.269
0.94
No
8 (36.4)
10 (45.5)
4 (18.2)
22 (100)
Yes
15 (37.5)
16 (40)
9 (22.5)
40 (100)
Syn
1.629
1.00
-
0 (0)
1 (100)
0 (0)
1 (100)
+
27 (40.3)
26 (38.8)
14 (20.9)
67 (100)
CgA
1.994
0.44
-
11 (52.4)
7 (33.3)
3 (14.3)
21 (100)
+
16 (34)
20 (42.6)
11 (23.4)
47 (100)
CD56
2.837
0.22
-
4 (25)
9 (56.3)
3 (18.8)
16 (100)
+
22 (45.8)
16 (33.3)
10 (20.8)
48 (100)
Sex
0.345
1.00
Female
3 (37.5)
3 (37.5)
2 (25)
8 (100)
Male
24 (40)
24 (40)
12 (20)
60 (100)
Age (year)
4.546
0.29
≤ 45
0 (0)
2 (100)
0 (0)
2 (100)
45-70
23 (45.1)
19 (37.3)
9 (17.6)
51 (100)
≥ 70
4 (26.7)
6 (40)
5 (33.3)
15 (100)
With NAC
0.339
0.94
No
19 (38)
20 (40)
11 (22)
50 (100)
Yes
8 (44.4)
7 (38.9)
3 (16.7)
18 (100)
Location
11.041
0.004
EGJ
20 (58.8)
8 (23.5)
6 (17.6)
34 (100)
Non-EGJ
7 (20.6)
19 (55.9)
8 (23.5)
34 (100)
There was a statistically significant disparity in the distribution of LN metastasis components between the pT1-2 and pT3-4 groups (χ2 = 8.36, P = 0.015). Furthermore, a significant difference was found in the distribution of LN metastasis components across various Lauren types (χ2 = 12.46, P = 0.01). Specifically, when the Lauren type was classified as intestinal, there was a higher likelihood of NEC metastasis in LN (adjusted standardized residual = 3.4). A statistically significant difference in the distribution of LN metastasis components was also observed between primary tumours located at the esophagogastric junction (EGJ) and those located at non-EGJ (χ2 = 11.041, P = 0.004). Tumours in the EGJ displayed a greater tendency for LN with NEC component metastasis (with an adjusted standardized residual of 3.2). There was no statistically significance difference in the distribution of LN metastasis components across the other groups (P > 0.05) (Table 1).
Clinicopathological features with statistical significance in the univariate analysis (P < 0.05), such as pT, NEC percentage (categorical variable), tumour location, and Lauren type, were incorporated into the unordered multinomial logistic regression. Using LN metastasis with “pure” AC as a reference, the results indicated that the NEC percentage (categorical variable) in the primary tumour [P = 0.006, 95% confidence interval (CI): 1.752-27.891] and tumour location (P = 0.008, 95%CI: 0.030-0.591) were independent risk factors affecting the components of LN metastasis (Table 2).
Table 2 Multivariate logistic regression analysis of independent risk factors.
Indicators
β
SE
Wald
P values
OR
95%CI
pT stage
1.060
0.955
1.230
0.267
2.885
0.44-18.762
NEC percentage (%)
1.945
0.706
7.588
0.006
6.991
1.752-27.891
Tumour location
-2.015
0.759
7.035
0.008
0.133
0.030-0.591
Lauren type
-0.896
0.479
3.507
0.061
0.408
0.160-1.043
Constant
-1.780
1.296
1.887
0.169
Considering the potential influence of NAC on LN metastasis, we excluded 18 patients who underwent NAC from the study. The analysis was subsequently conducted on the remaining 50 G-ANEC patients who did not receive NAC. Differences in the LN metastasis component among the groups of G-ANEC patients who did not receive neoadjuvant therapy was presented in Table 3. Our research findings revealed a significant difference in the prevalence of NEC (categorical variable) at the primary tumour site with its occurrence in LN metastases, which was statistically significant (χ2 = 9.38, P = 0.038). The Cramer’s V value was 0.314, indicating a moderate association. Specifically, NEC proportions ranging from 71% to 99% at the primary site were associated with a higher probability of LN metastasis, with an adjusted standardized residual of 3.0. There was also a statistically significant difference in the distribution of LN metastasis components between the pT1-2 and pT3-4 groups (χ2 = 5.4, P = 0.045). Additionally, significant variations were observed in LN metastasis distributions across different Lauren types (χ2 = 14.92, P = 0.002), particularly a greater likelihood of NEC metastasis in LNs when the Lauren type was classified as intestinal (adjusted standardized residual = 3.7). However, no statistically significant difference was found in LN metastasis components between the EGJ and non-EGJ groups (χ2 = 4.31, P = 0.116).
Table 3 Clinicopathological features among lymph node metastasis patterns in adenoneuroendocrine mixed gastric cancer patients who did not undergo neoadjuvant therapy, n (%).
Parameters
NEC
AC
Both
Total
χ2 value
P value
pT stage
5.4
0.045
pT1-2 stage
3 (30)
7 (70)
0 (0)
10 (100)
pT3-4 stage
16 (40)
13 (32.5)
11 (27.5)
40 (100)
LVI
1.6
0.344
-
3 (60)
2 (40)
0 (0)
5 (100)
+
16 (35.6)
18 (40)
11 (24.4)
45 (100)
PNI
1.89
0.396
-
8 (50)
6 (37.5)
2 (12.5)
16 (100)
+
11 (32.4)
14 (41.2)
9 (26.5)
34 (100)
NEC percentage (%)
9.38
0.038
1-29
1 (14.3)
4 (57.1)
2 (28.6)
7 (100)
30-70
8 (27.6)
14 (48.3)
7 (24.1)
29 (100)
71-99
10 (71.4)
2 (14.3)
2 (14.3)
14 (100)
Histology
5.80
0.185
Small cell
9 (52.9)
5 (29.4)
3 (17.6)
17 (100)
Large cell
7 (24.1)
14 (48.3)
8 (27.6)
29 (100)
Both
3 (75)
1 (25)
0 (0)
4 (100)
Lauren type
14.92
0.002
Intestinal type
17 (60.7)
7 (25)
4 (14.3)
28 (100)
Diffuse type
0 (0)
5 (62.5)
3 (37.5)
8 (100)
Mixed type
2 (14.3)
8 (57.1)
4 (28.6)
14 (100)
LN periosteal violation
0.13
1.00
No
6 (40)
6 (40)
3 (20)
15 (100)
Yes
12 (37.5)
13 (40.6)
7 (21.9)
32 (100)
Syn
1.59
1.00
-
0 (0)
1 (100)
0 (0)
1 (100)
+
19 (38.8)
19 (38.8)
11 (22.4)
49 (100)
CgA
2.25
0.298
-
8 (53.3)
4 (26.7)
3 (20)
15 (100)
+
11 (31.4)
16 (45.7)
8 (22.9)
35 (100)
CD56
4.0
0.143
-
2 (18.2)
7 (63.6)
2 (18.2)
11 (100)
+
17 (47.2)
11 (30.6)
8 (22.2)
36 (100)
Sex
0.51
0.865
Female
3 (50)
2 (33.3)
1 (16.7)
6 (100)
Male
16 (36.4)
18 (40.9)
10 (22.7)
44 (100)
Age (year)
3.62
0.494
≤ 45
0 (0)
1 (100)
0 (0)
1 (100)
45-70
16 (44.4)
13 (36.1)
7 (19.5)
36 (100)
≥ 70
3 (23.4)
6 (46.2)
4 (30.7)
13 (100)
Location
4.31
0.116
EGJ
12 (52.2)
6 (26.1)
5 (21.7)
23 (100)
Non-EGJ
7 (25.9)
14 (51.9)
6 (22.2)
27 (100)
Clinicopathological features identified as statistically significant in the univariate analysis (P < 0.05), such as the pT stage, NEC percentage (categorical variable), and Lauren classification, were incorporated into the unordered multinomial logistic regression model. Additionally, we considered the location of the tumour to have clinical significance in influencing LN metastasis components, and thus it was also included in the model. Using LN metastasis with “pure” AC as a reference, the results indicated that the NEC percentage in the primary tumour (P = 0.027, 95%CI: 1.225-26.984) and Lauren type (P = 0.026, 95%CI: 0.085-0.859) were independent risk factors affecting the components of LN metastasis (Table 4).
Table 4 Multivariate logistic regression analysis of independent risk factors in adenoneuroendocrine mixed gastric cancer patients who did not undergo neoadjuvant therapy.
Indicators
β
SE
Wald
P values
OR
95%CI
pT
0.983
1.033
0.906
0.341
2.674
0.353-20.256
NEC percentage (%)
1.749
0.789
4.917
0.027
5.750
1.225-26.984
Lauren type
-1.306
0.588
4.924
0.026
0.271
0.085-0.859
Tumour location
-1.714
0.882
3.777
0.052
0.180
0.032-1.015
Constant
-1.336
1.414
0.892
0.345
The relationship between clinicopathological characteristics and prognosis survival
We conducted an analysis to examine the correlations between clinicopathological features and DFS (Supplementary Figure 2). A statistically significant correlation was solely observed between the NEC% (categorical variable) in the primary tumour and DFS (P = 0.01, Supplementary Figure 2). Statistical analysis revealed no significant difference in DFS between the AC and NEC metastasis in LN (P = 0.519, χ2 = 1.313). Furthermore, the survival curves did not indicate any trend suggesting a DFS advantage between the two groups. We also conducted an analysis to examine the correlations between clinicopathological features and OS (Supplementary Figure 3). Our findings revealed that only the presence of LN periosteal violation was correlated with OS, and a statistically significant difference was observed between the groups (P= 0.033, Figure 3A and B). Furthermore, the Kaplan-Meier survival analysis curves revealed no statistically significant differences in OS across groups categorized by the NEC percentage (categorical variable) according to primary tumour type (P = 0.591), histological type (P = 0.381), or Lauren type (P = 0.768) (Figure 3C-E). Although there was no statistically significant difference in OS between different LN metastatic components (P = 0.147 for pure NEC vs AC, χ2 = 2.107, 95%CI: 38.33-73.91, Figure 3F and G), there was a trend suggesting a better prognosis for patients with AC metastasis in LN (average survival time: 85.83 ± 9.81 months) compared with those with NEC metastasis (average survival time: 56.12 ± 9.08 months). The 1-year, 3-year, and 5-year survival rates for “pure” NEC metastasis in LNs were found to be 77.9%, 57.2%, and 45.8%, respectively. Similarly, for AC metastasis in LNs, the survival rates were 88.7%, 74.9%, and 69.6% at 1-year, 3-year, and 5-year intervals, respectively.
Figure 3 Kaplan-Meier curves.
A: Kaplan-Meier curves for lymph node periosteal violation and overall survival; B: Kaplan-Meier curves for lymph node periosteal violation and disease-free survival; Kaplan-Meier cumulative survival curves according to overall tumour survival of neuroendocrine carcinoma%: C: Primary tumour; D: Histological type; E: Lauren type; F: Comparison of overall survival in three patterns of lymph node metastasis; G: Comparison of overall survival in lymph node metastasis comprising “pure” adenocarcinoma and “pure” neuroendocrine carcinoma. OS: Overall survival; DFS: Disease-free survival; NEC: Neuroendocrine carcinoma; AC: Adenocarcinoma; LN: Lymph node.
To further exclude bias caused by inconsistent adjuvant chemotherapy regimens, we selected 31 patients who received SOX/XELOX as their adjuvant chemotherapy protocol. For clinical treatment, G-ANEC patients with LN metastasis components consisting solely of AC are treated with 5-fluorouracil (5-FU)-based regimens, which are commonly used for gastric AC. In contrast, for patients with LN metastasis components containing pure NEC or a mixture of NEC and AC, platinum-based therapeutic regimens, which are standard for gastric NEC, are utilized. Therefore, based on the corresponding treatment strategy, G-ANEC patients are classified into two groups: The “LN with NEC-containing group” (Supplementary Figure 4A-C) and “LN with AC-only group” (Supplementary Figure 4D). The results revealed no significant difference in OS between the two groups (P = 0.061 for NEC and both vs AC, χ2 = 3.499, 54.36 ± 11.36 months vs 94.25 ± 11.07 months, Supplementary Figure 5A). However, patients whose LN metastases consisted solely of AC had a significantly better OS than those with NEC. The results also revealed no significant difference in DFS between the two groups (P = 0.84 for NEC and both vs AC, Supplementary Figure 5B).
We further screened patients with G-ANEC showing LN metastasis composed of NEC elements. Patients were divided into the etoposide and cisplatin (EP) group (10 patients) and the SOX/XELOX group (14 patients) on the basis of the postoperative adjuvant chemotherapy protocols. Except for a statistically significant difference in the proportion of NEC components (categorical variable) in the primary lesion (P = 0.014), no other clinical or pathological differences were observed between the two groups (Supplementary Table 1). There were no significant differences in OS (P = 0.27) or DFS (P = 0.296) between the groups. However, there was a trend toward better OS (64 ± 10.77 months vs 54.35 ± 11.36 months, Supplementary Figure 5C) and DFS (71.28 ± 9.92 months vs 66.28 ± 11.93 months, Supplementary Figure 5D) in the EP group than in the SOX/XELOX group.
DISCUSSION
It is commonly believed that gastric cancer has four metastatic pathways: Lymphatic spread, direct infiltration, peritoneal dissemination, and haematogenous spread[11]. Among these pathways, LN metastasis is the primary pathway for gastric cancer[12], especially in G-MANEC, where the metastasis and composition of the LNs are particularly important. Since cancer cells can spread from both the LN and the primary tumour itself, LN metastasis can also be considered a collateral circuit of haematogenous spread[13]. Therefore, the greater the number of metastatic LNs is, the greater likelihood of circulating tumour cells and micro-metastases in organs. LN metastasis is a key factor in disease staging and progression, making lymphadenectomy a crucial aspect of treatment for most tumours[14]. The LN metastatic patterns of G-ANEC in our study, are consistent with the results of Zhang et al[15] regarding the LN metastasis pattern in gastroenteropancreatic MiNENs. Regarding the different metastatic patterns of LN in G-ANEC, there are many gaps waiting to be filled. First, the timing of NEC and AC is important for the primary lesion to acquire invasive and metastatic abilities. Second, for LNs with single-component metastasis, whether there is a single metastatic clone in the primary lesion that subsequently spreads to the LNs. For LNs with two or more components metastasizing, whether multiple subclones in the primary tumour independently produce lymphatic and distant metastases. Finally, whether we can use bioinformatics methods to finely map the LN and distant metastasis landscape of G-ANEC patients on a temporal and spatial scale, identify the pathogenic pathways involved in the turning point that leads to the metastasis of a certain component, and provide answers to important questions related to G-ANEC metastasis.
Currently, there is controversy regarding the factors related to the biological behaviour and prognosis of MiNENs. Some investigators consider that the invasiveness and prognosis of mixed tumours are determined by the component with the highest degree of malignancy[16]. Thus, certain scholars posit that the clinical behaviour and outcomes of G-MANEC may be influenced by the presence of an NEC component in the primary lesion[15]. Consequently, perioperative chemotherapy for G-MANEC predominantly refers to NEC and is primarily based on platinum-based therapeutic regimens[17]. Some researchers believe that the prognosis of MiNEN is related to the main percentage component of the primary lesion, and is not affected by components less than 30%[3,18]. However, certain studies have indicated that the proportion of tumour components present in the primary tumour may not comprehensively indicate the degree of malignancy in G-MANEC[19]. Some scholars have suggested that the therapeutic approach for G-MiNEN should ideally focus on addressing the metastasized tumour component rather than solely relying on the main component of the primary lesion[20]. Although in our study, when the percentage of NEC was between 71% and 99%, LN metastases were more likely to consist of NEC, we also found that the presence of a low percentage (10%) of NEC or AC in the primary tumour does not preclude the possibility of LN metastasis comprising NEC or AC. In addition, certain scholars posit that G-MANEC with distinct metastatic components may exhibit varying responses to chemotherapy, with the component that infiltrates profoundly and metastasizes often being the most influential[16,17]. In the research conducted by Xie et al[19], their findings imply that patients in the AC group may experience advantageous effects from the administration of 5-FU-based adjuvant chemotherapy. Similarly, our study revealed that although there was no statistically significant difference in OS between patients with different LN metastatic components, there was a trend that the survival of patients with LN metastasis with an AC component was better than that with an NEC component. Notably, most of our chemotherapy regimens were also based on 5-FU. Our study revealed that the SOX/XELOX adjuvant chemotherapy regimen achieves better OS when LN metastasis consists of AC compared to NEC. Our research also revealed that for LN metastases associated with NEC, the EP adjuvant chemotherapy regimen tends to result in better OS and DFS than the SOX/XELOX regimen does. Our study suggested that LN metastasis components may play an important role in guiding the selection of adjuvant chemotherapy regimens. Specifically, when the LN metastasis component is AC, we recommend the use of SOX/XELOX, whereas for patients with NEC as the LN metastasis component, EP regimens should be considered. This strategy has the potential to optimize treatment outcomes and improve survival of patients. However, this hypothesis has not yet been validated through prospective randomized controlled trials (RCTs), so it should be approached with caution in clinical practice.
Certain scholars have observed that in certain mice and patients, the correlation between distant metastasis and LN metastasis is more prominent than that between distant metastasis the primary tumour, as evidenced by investigations of colon cancer[21,22]. However, in terms of treatment for G-ANEC, owing to its rarity and substantial heterogeneity, there currently exists no unified standard for its perioperative chemotherapy[6]. In this context, our study’s findings suggested that routine pathological diagnosis of LN metastasis components not only helps to formulate more rational adjuvant chemotherapy regimens but also provides clinicians with more precise treatment references. This approach can reduce ineffective treatments, improve treatment efficacy, and minimize unnecessary side effects for patients. Moreover, this strategy offers an important direction for future clinical trials, particularly for large-scale prospective studies on chemotherapeutic responses of patients with different LN metastasis components, to validate the hypotheses and treatment regimens we have proposed. Moreover, although our findings strongly support the selection of a chemotherapy regimen on the basis of LN metastasis components, this hypothesis still needs further validation through prospective RCTs. Therefore, future research should focus on exploring the impact of LN metastasis components on chemotherapy regimens and conduct large-scale, multicenter clinical trials to validate the true impact of this individualized treatment strategy on patient outcomes. This would represent a significant shift from traditional methods of treatment based solely on the primary tumour component, potentially improving treatment efficacy. From the perspective of treatment strategies for G-ANEC, we strongly believe that exploring whether the metastatic components in LNs can more accurately reflect the biological behaviour and prognosis of the tumour than the most malignant component in the primary lesion is of great clinical value. This unique pattern of LN metastasis undoubtedly provides valuable clues for a deeper understanding of this tumour type. However, its underlying mechanism and specific significance remain the focus and challenge of our future research.
According to recent research findings, carcinomas of the EGJ demonstrate specific clinicopathological features and a distinctive genomic profile[23]. Due to the small size and superficial positioning of early G-MANEC lesions, as well as their lack of symptoms and the complex anatomical complexities of the EGJ, there is a greater likelihood of overlooked diagnoses during endoscopic assessments. The NEC component located in the EGJ demonstrates a significant level of aggressiveness, which may result in the progression of diseases within a limited timeframe[24]. Consequently, NEC of the EGJ frequently presents at advanced stages, and is characterized by poor differentiation and an increased likelihood of LN metastasis[25]. This results in a comparatively poorer OS rate for NEC located at the EGJ than for those with NEC located in non-EGJ regions[26]. Scholars have reported that NEC of the EGJ have a greater incidence of pathological pT3-T4 tumour than their non-EGJ counterparts do[27], a finding that aligns with our research outcomes. According to research findings, the extent of infiltration of gastric cancer has a substantial influence on the incidence of LN metastasis, which is closely tied to the anatomical arrangement of lymphatic networks within the gastric wall[28]. It should be noted that NEC components are frequently found in the submucosal layer or deeper, where lymphatic vessels are abundant, explaining the link between pT in EGJ and increased chances of LN metastasis with NEC in G-ANEC. In addition, considering the potential impact of NAC on the components of LN metastasis, we analyzed patients who did not receive such treatment. The results indicated that there was no significant association between tumour location and LN metastasis components, which is inconsistent with the findings from studies that included patients who underwent NAC. These findings suggest that NAC may have played a significant role in altering the metastatic profile of G-ANEC. Although our study revealed no statistically significant differences in LN metastasis components between groups with and without NAC, this may be due to the small sample size and the presence of confounding factors such as variations in neoadjuvant treatment protocols. Importantly, NAC, which primarily targets AC, may alter the pathological characteristics and behaviour of the tumour, especially affecting the metastasis of the NEC component in LNs, by reducing tumour burden and possibly affecting the lymphatic drainage or immune response[29]. Furthermore, the small sample size of 50 patients in the non-NAC group could have increased variability, reducing the statistical power to identify significant differences.
LN periosteal violation has historically been recognized as an indicator of an unfavorable prognosis in diverse malignant tumours, encompassing gastrointestinal tract malignancies[30]. This characteristic has already been integrated into the staging system for vulvar squamous cell carcinoma, head and neck cancers[31]. In gastric cancer, LN periosteal violation is associated with a reduced 5-year OS and DFS[32]. In our study, we revealed that the occurrence of LN periosteal violation plays a crucial role in determining OS among patients diagnosed with G-ANEC. However, its impact on DFS was not statistically significant, which could be attributed to the limited sample size of our cohort. Previous research has established a correlation between LN periosteal violation by gastric cancer and the occurrence of metastases in the peritoneum and liver[11]. The presence of LN periosteal violation may serve as a representation of the tumour’s aggressive invasiveness and deteriorating biological course[33], thereby explaining the significant decrease in OS observed in these individuals. In conclusion, our study provides novel perspectives on the prognosis of G-ANEC, emphasizing the crucial importance of LN periosteal violation. These discoveries lay the foundation for future investigations and therapeutic approaches, particularly for patients with a heightened susceptibility to LN periosteal violation.
As the pioneering investigation into the relationship between the clinicopathological characteristics of G-ANEC and LN metastatic components, our research boasts one of the largest sample sizes among single-center retrospective studies. Nevertheless, our study is subject to several limitations. Firstly, 26.5% of our patients underwent NAC, predominantly following the 5-FU regimen. The potential impact of NAC on LN metastatic components remains ambiguous. We advocate for future multicenter studies with expanded sample sizes or prospective research endeavors to elucidate the effects of NAC on LN metastatic components. Secondly, we lack pathological results for distant metastatic lesions and have not further validated the relationships between NEC and AC in the primary G-ANEC lesion and the components of distant metastatic lesions, as well as between the components of LN metastases and distant metastatic lesions. In addition, one of the main challenges our study faces were the limitation of sample size. Owing to the low incidence of G-ANEC and specific inclusion criteria (including only patients with LN metastasis G-ANEC and specified LN metastasis components), the number of samples we were able to collect was relatively small. This issue is particularly pronounced in subgroup analyses. The insufficiency of our sample size may limit the generalizability and statistical power of our study results, necessitating more cautious interpretation. Additionally, the small sample size makes it difficult to control for potential biases and confounding factors. To address this, we employed multivariable analysis to mitigate these issues.
CONCLUSION
In conclusion, our study focused primarily on the components of metastatic LNs in G-ANEC, aiming to systematically depict their unique metastatic patterns. Surprisingly, we discovered that even if NEC or AC components are present in a lower proportion in the primary lesion, they can still appear as the components in the LNs. Our study suggested a need for routine pathological diagnosis of metastatic LN components in G-ANEC patients, which could guide adjuvant chemotherapy strategies.
Footnotes
Provenance and peer review: Unsolicited article; Externally peer reviewed.
Peer-review model: Single blind
Specialty type: Gastroenterology and hepatology
Country of origin: China
Peer-review report’s classification
Scientific Quality: Grade A, Grade B, Grade B, Grade D
Novelty: Grade B, Grade B, Grade B, Grade C
Creativity or Innovation: Grade B, Grade B, Grade B, Grade C
Scientific Significance: Grade B, Grade B, Grade B, Grade C
P-Reviewer: Bhowmick M; Luo ZN; Raonic J S-Editor: Fan M L-Editor: A P-Editor: Zheng XM
Kawasaki K, Toshimitsu K, Matano M, Fujita M, Fujii M, Togasaki K, Ebisudani T, Shimokawa M, Takano A, Takahashi S, Ohta Y, Nanki K, Igarashi R, Ishimaru K, Ishida H, Sukawa Y, Sugimoto S, Saito Y, Maejima K, Sasagawa S, Lee H, Kim HG, Ha K, Hamamoto J, Fukunaga K, Maekawa A, Tanabe M, Ishihara S, Hamamoto Y, Yasuda H, Sekine S, Kudo A, Kitagawa Y, Kanai T, Nakagawa H, Sato T. An Organoid Biobank of Neuroendocrine Neoplasms Enables Genotype-Phenotype Mapping.Cell. 2020;183:1420-1435.e21.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 49][Cited by in F6Publishing: 118][Article Influence: 23.6][Reference Citation Analysis (0)]
Sasako M, Sano T, Yamamoto S, Kurokawa Y, Nashimoto A, Kurita A, Hiratsuka M, Tsujinaka T, Kinoshita T, Arai K, Yamamura Y, Okajima K; Japan Clinical Oncology Group. D2 lymphadenectomy alone or with para-aortic nodal dissection for gastric cancer.N Engl J Med. 2008;359:453-462.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 698][Cited by in F6Publishing: 729][Article Influence: 42.9][Reference Citation Analysis (0)]
Zhang P, Li Z, Li J, Li J, Zhang X, Lu Z, Sun Y, Li Y, Zhou J, Wang X, Peng Z, Shen L, Lu M. Clinicopathological features and lymph node and distant metastasis patterns in patients with gastroenteropancreatic mixed neuroendocrine-non-neuroendocrine neoplasm.Cancer Med. 2021;10:4855-4863.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 4][Cited by in F6Publishing: 14][Article Influence: 3.5][Reference Citation Analysis (0)]
Xie JW, Lu J, Wang JB, Lin JX, Chen QY, Cao LL, Lin M, Tu RH, Huang ZN, Lin JL, Zheng CH, Li P, Huang CM. Prognostic factors for survival after curative resection of gastric mixed adenoneuroendocrine carcinoma: a series of 80 patients.BMC Cancer. 2018;18:1021.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 21][Cited by in F6Publishing: 29][Article Influence: 4.1][Reference Citation Analysis (0)]
Lim SM, Kim H, Kang B, Kim HS, Rha SY, Noh SH, Hyung WJ, Cheong JH, Kim HI, Chung HC, Yun M, Cho A, Jung M. Prognostic value of (18)F-fluorodeoxyglucose positron emission tomography in patients with gastric neuroendocrine carcinoma and mixed adenoneuroendocrine carcinoma.Ann Nucl Med. 2016;30:279-286.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 8][Cited by in F6Publishing: 13][Article Influence: 1.4][Reference Citation Analysis (0)]
McFadden DG, Papagiannakopoulos T, Taylor-Weiner A, Stewart C, Carter SL, Cibulskis K, Bhutkar A, McKenna A, Dooley A, Vernon A, Sougnez C, Malstrom S, Heimann M, Park J, Chen F, Farago AF, Dayton T, Shefler E, Gabriel S, Getz G, Jacks T. Genetic and clonal dissection of murine small cell lung carcinoma progression by genome sequencing.Cell. 2014;156:1298-1311.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 197][Cited by in F6Publishing: 220][Article Influence: 20.0][Reference Citation Analysis (0)]
Huang Q, Cheng Y, Chen L, Mingzhan D, Wang Y, Xu G, Shi J, Fan X, Zhou X, Zhang Y, Guo L, Liu T, Zhou D, Mashimo H, Gold JS, Zou X; and The Jiangsu Province Early Gastric Carcinoma Multicenter Study Team. Low risk of lymph node metastasis in 495 early gastric cardiac carcinomas: a multicenter clinicopathologic study of 2101 radical gastrectomies for early gastric carcinoma.Mod Pathol. 2018;31:1599-1607.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 14][Cited by in F6Publishing: 17][Article Influence: 2.4][Reference Citation Analysis (0)]
Zhang D, Nan Q. Patterns of the lymph node metastasis and the influencing factors in esophagogastric junction cancers.Asian J Surg. 2023;46:3512-3519.
[PubMed] [DOI][Cited in This Article: ][Reference Citation Analysis (0)]
Tomita Y, Seki H, Matsuzono E, Kobayashi Y, Sogabe S, Sugai N, Fujita J, Suzuki A. Early gastric mixed neuroendocrine-non-neuroendocrine neoplasm with early poor prognosis after endoscopic submucosal dissection: A case report.DEN Open. 2022;2:e10.
[PubMed] [DOI][Cited in This Article: ][Reference Citation Analysis (0)]
Cheng Y, Zhang X, Zhou X, Xu K, Lin M, Huang Q. Differences in clinicopathology and prognosis between gastroesophageal junctional and gastric non-cardiac neuroendocrine carcinomas: a retrospective comparison study of consecutive 56 cases from a single institution in China.Am J Cancer Res. 2022;12:4737-4750.
[PubMed] [DOI][Cited in This Article: ]
Luchini C, Nottegar A, Solmi M, Sergi G, Manzato E, Capelli P, Scarpa A, Veronese N. Prognostic implications of extranodal extension in node-positive squamous cell carcinoma of the vulva: A systematic review and meta-analysis.Surg Oncol. 2016;25:60-65.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 45][Cited by in F6Publishing: 42][Article Influence: 4.2][Reference Citation Analysis (0)]
Di Giorgio A, Botti C, Sammartino P, Mingazzini P, Flammia M, Stipa V. Extracapsular lymphnode metastases in the staging and prognosis of gastric cancer.Int Surg. 1991;76:218-221.
[PubMed] [DOI][Cited in This Article: ]
