Abstract
Purpose
Lymphovascular infiltration (LVI) may play a critical role in radicality and prognostic assessment of early gastric cancer (EGC). However, risk factors for LVI in endoscopically resected EGC remain unknown. This study evaluated the clinicopathological characteristics and prognoses of patients who underwent endoscopic resection of EGC to identify potential risk factors of LVI.
Methods
A cross-sectional study of patients who received gastric endoscopic submucosal dissection between February 1, 2012, and December 31, 2019, at two institutions was conducted. Among 1,462 lesions, 943 met the criteria for radical treatment considering features other than LVI and were included. The lesions were classified based on the presence of LVI. The clinicopathological characteristics of the two groups were compared.
Results
LVI was detected in 17 lesions (1.8%). The positivity rates of LVI were 0.7% (7/903) for intramucosal cancer and 25% (10/40) for submucosally invasive cancer. The LVI positivity rate was significantly higher for mixed-type cancer (lesions containing differentiated and undifferentiated-type carcinoma) than for non-mixed-type cancer (35.3 vs. 2.8%; P < 0.001) and for submucosally invasive cancer than for intramucosal cancer (58.8 vs. 3.2%; P < 0.001). In the multivariate logistic regression analysis, independent risk factors for LVI were mixed-type cancer (odds ratio; 95% confidence interval: 23.9; 5.0–115; P < 0.001) and submucosal invasion (58.7; 16.0–215; P < 0.001).
Conclusions
Mixed-type cancer and submucosal invasion were risk factors for LVI in endoscopically resected EGC. These factors may play a critical role in the radicality and prognostic assessment of EGC.
Keywords: Early gastric cancer, Lymphovascular infiltration, Endoscopic submucosal dissection, Mixed-type cancer, Submucosal invasion
Introduction
In 2022, due to advances in diagnostic and therapeutic endoscopy in Japan, early-stage gastrointestinal cancers at all locations other than the jejunum and ileum can be diagnosed and treated endoscopically. Furthermore, this treatment is covered by the national health insurance system. In the 1990s, endoscopic submucosal dissection (ESD), an early gastric cancer treatment that allows en bloc resection of lesions without division and detailed pathological evaluation, was developed. This treatment is not only less invasive than surgery but also has higher post-treatment quality of life. Therefore, ESD has been widely used in East Asia, where the prevalence of gastric cancer is particularly high, and Europe (Gotoda et al. 2006; Park et al. 2011; Lian et al. 2012; Pimentel-Nunes et al. 2015). Early gastric cancer is defined in the Japanese classification of gastric carcinoma (3rd English edition) (Japanese Gastric Cancer Association 2011) as cancer that is restricted to the mucosa or submucosal layer, irrespective of lesion size and the presence of lymph node metastasis. Additionally, lymph node metastasis is the most significant prognostic factor for early gastric cancer (Gotoda et al. 2000; Hatta et al. 2017). Radical surgical resection in combination with lymph node dissection has been accepted as the standard treatment for early gastric cancer. Histopathological examination of numerous surgically resected lesions allowed the identification of lesions with a very low probability of having lymph node metastasis, which were suitable for radical endoscopic therapies. As a result, indications for endoscopic treatment have increased (Ono et al. 2021).
Indications for endoscopic treatment depend on preoperative diagnostic findings, including histological type, tumor diameter, depth of invasion, and ulceration and related scarring. In contrast, the radicality is evaluated by the risk of lymph node metastasis, predicted from the local resectability and pathological signs in the resected samples (Ono et al. 2021; Japanese Gastric Cancer Association 2021). Reported risk factors for lymph node metastasis include the predominance of undifferentiated-type carcinoma, large tumor diameter, submucosal invasion, ulceration, presence of lateral and vertical margins, and lymphovascular (lymphatic and/or venous) infiltration (Gotoda et al. 2000; Hatta et al. 2017). However, the specific contribution of these factors to lymph node metastasis remains unknown. In a study with 1,101 early gastric cancer patients with additional surgery performed on a primary non-curative endoscopic resection, Hatta et al. showed that lymphatic infiltration is the most significant risk factor for lymph node metastasis (Hatta et al. 2017). Therefore, lymphovascular infiltration is a crucial feature in the radicality and prognosis assessment of early gastric cancer.
In clinical practice, most lesions with lymphovascular infiltration involve SM2 deep submucosal invasion, e.g., invasion extending > 0.5 mm from the lamina muscularis mucosae. In contrast, most intramucosal cancers (M) that are indicated for endoscopic treatments and subsequently endoscopically resected, are rarely positive for lymphovascular infiltration. Similarly, several cases of submucosally invasive cancer with only SM1 invasion, e.g., invasion extending < 0.5 mm from the lower edge of the lamina muscularis mucosae, are negative for lymphovascular infiltration. Although lymphovascular infiltration is absent in most endoscopically resected cases, patients with M or SM1 lesions are sometimes positive for this pathological feature. Such lesions do not meet the criteria for curative resection and additional surgical resections are recommended (Ono et al. 2021). Additionally, the rate of lymph node metastasis is not high for lesions that meet curative resection criteria other than lymphovascular infiltration. Therefore, these lesions may not need additional resections.
To date, risk factors for endoscopically resected M or SM1 cancer with lymphovascular infiltration have not been investigated. The present study aimed to evaluate the clinicopathological characteristics and prognoses of such lesions to obtain clinical insights to determine the suitability and efficacy of endoscopic treatment in these cases.
Materials and methods
Study subjects
This cross-sectional study included all patients who underwent gastric ESD between February 1, 2012, and December 31, 2019, at the authors’ hospitals. These patients had 1,462 lesions, of which 922 and 540 were treated at Yokohama City University Medical Center and Omori Red Cross Hospital, respectively. Clinical and demographic data were obtained by multi-site collaboration. Since written informed consent from the subjects was difficult to obtain, information about the study was made available to all study subjects to ensure that each had the opportunity to refuse participation.
Analysis of lesions and other pathological features
Lesions with final pathological diagnoses other than gastric cancer were excluded from the study. Moreover, lesions with SM2 or deeper invasion, which are classified as non-curative resection in radicality assessment and not suitable for endoscopic treatment, were excluded. In addition, among the classes of gastric cancer shown below, lesions were excluded if they were not classified as curative resection per the radicality evaluation criteria other than lymphovascular infiltration in the ESD/EMR Guidelines (second edition) (Ono et al. 2021). Thus, the lesions included in the study met the criteria of curative resection considering features other than lymphovascular infiltration.
Endoscopic radicality A (eCuraA): For curative resection, the prognosis must be at least as favorable as that of surgical resection, which has been verified by long-term outcome data. If no ulceration is found during en bloc resection of the tumor, one of the two following criteria must be met: (i) regardless of the tumor diameter, the cancer is intramucosal (M), negative for horizontal and vertical margins and lymphovascular infiltration, and predominantly differentiated type (cancer with a predominance of papillary adenocarcinoma, well-differentiated tubular adenocarcinoma, or moderately differentiated adenocarcinoma) and (ii) the tumor diameter is not > 2 cm, and the cancer is intramucosal (M), negative for horizontal and vertical margins and lymphovascular infiltration, and predominantly differentiated type. Conversely, if ulceration is found, the tumor diameter must not be > 3 cm, and the cancer must be intramucosal (M), negative for horizontal and vertical margins and lymphovascular infiltration, and predominantly differentiated type.
Endoscopic radicality B (eCuraB): Radical treatment can be expected despite insufficient long-term outcome data. En bloc resection is performed. The tumor diameter is not > 3 cm, the cancer is submucosally invasive (SM1), negative for horizontal and vertical margins and lymphovascular infiltration, and predominantly differentiated type.
Endoscopic radicality C (eCuraC): The cancer does not meet eCuraA or eCuraB criteria and was determined to be non-curative resection. Lesions are defined as C-1 if they did not meet one of the following criteria: (i) positive for lateral margin with en bloc resection of differentiated type and (ii) with divided resection. Other cases are considered non-curative resection (C-2). Both progression monitoring and additional local resection are acceptable for C-1 lesions, whereas additional surgical resection is indicated for C-2 due to the risks of metastasis and recurrence.
The lesions above were classified into two groups according to the presence or absence of lymphovascular infiltration. The clinicopathological characteristics of the two groups were compared. A total of nine evaluation items were defined, consisting of (i) four criteria for endoscopic treatment indication and radicality assessment, namely, histological type, tumor diameter, ulcerative (or scarring) signs, and depth of invasion, and (ii) five criteria which have been previously evaluated as risk factors for lymph node metastasis, namely, age, sex, tumor location, macroscopic type, and the presence or absence of mixed histological components. In addition, treatment progression and prognosis were investigated in the lymphovascular-infiltration-positive group.
Pathological signs were classified using post-resection pathological tissues in accordance with the Japanese classification of gastric carcinoma (3rd English edition and 15th Japanese edition) (Japanese Gastric Cancer Association 2011; Japanese Gastric Cancer Association 2017). The histological type was classified using quantitatively predominant tissue patterns. In this regard, well-differentiated tubular adenocarcinoma (tub1), moderately differentiated tubular adenocarcinoma (tub2), and papillary adenocarcinoma (pap) were defined as differentiated-type carcinoma. Poorly differentiated adenocarcinoma (por), signet-ring cell carcinoma (sig), and mucinous carcinoma (muc) were defined as undifferentiated-type carcinoma. When a single lesion contained different histological types, i.e., differentiated and undifferentiated types (mixed-type cancer in a narrow sense), it was defined as mixed-type in this study. If the lesion contained differentiated-type carcinoma or undifferentiated-type carcinoma, as classified above, with different degrees of differentiation (mixed-type cancer in a broad sense), it was defined as non-mixed. If the depth of submucosal invasion was < 500 µm from the lamina muscularis mucosae, the cancer was defined as submucosally invasive cancer (SM1). With a deeper invasion, the cancer was defined as SM2 and excluded from the present study. Therefore, the comparison in this study was between M and SM1. Tumor locations were classified endoscopically. In this regard, the greater and lesser curvatures of the stomach were divided into three equal parts. The corresponding points in each curvature were joined to define the upper, middle, and lower regions. Using resected histopathological samples, the macroscopic type was classified as follows: elevated type (0-I, 0-I + IIa, 0-IIa, etc.), flat type (0-IIb), and depressed type (0-IIc, 0-III, etc.).
Samples were prepared in accordance with the Japanese classification of gastric carcinoma, as detailed below. Resected lesions were extension-fixed in 10% neutral-buffered formalin to the same degree as the endoscopic findings and cut through completely at 2–3 mm intervals. Subsequently, tissue slices were embedded in paraffin. The paraffin block was sliced at a thickness of 2 µm, and the thin slices were stained and sealed to prepare specimens. Lesions with suspected lymphovascular infiltration on the basis of hematoxylin–eosin staining were immunostained with the D2-40 antibody, which recognizes the lymph duct endothelium. Furthermore, these lesions were examined for the presence of elastic fibers using the Elastica van Gieson (EVG) stain to confirm lymphatic and venous infiltration. The procedures were verified by pathologists at Yokohama City University Medical Center and Omori Red Cross Hospital.
Statistical and analysis methods
For comparison between the two groups, the Shapiro–Wilk test was first performed to determine the normality of distributions. The t test was performed for samples with normal distribution. If one of the two groups exhibited non-normal distribution, the Wilcoxon rank-sum test (Mann–Whitney U test) was performed. The proportions in the two groups were tested with the χ2 test or Fisher’s exact probability test. Multiple logistic regression analysis was performed to adjust for confounding factors. The significance level was defined as P < 0.05. Data were analyzed using JMP Pro version 15 for Windows software (SAS Institute Japan).
Results
Among 1462 lesions, 354 were excluded from the analysis because the final pathological diagnosis was not gastric cancer, 102 because the classification was SM2 or deeper submucosal invasion, and 63 because assessment of histological type, tumor diameter, or ulceration (or scarring) was non-curative resection. The final analysis was performed with 943 lesions that met the criteria for radical treatment considering features other than lymphovascular infiltration (Fig. 1).
Fig. 1.
Flowchart of lesion enrollments. Among the 1462 lesions, the final analysis was performed with 943 lesions that met the criteria for radical treatment considering features other than lymphovascular infiltration
Demographic and clinical details of 943 lesions included in the final analysis are shown in Table 1. The median age of all patients was 75 years. A total of 648 (68.7%) lesions were derived from male patients. The tumor location was the middle region in 412 lesions (43.7%). Regarding the macroscopic type, 466 lesions (49.4%) were depressed. The median tumor diameter was 12 mm. Regarding other pathological features, 910 lesions were of differentiated type (96.5%), 910 lesions (96.5%) were non-mixed cancer, 884 lesions (93.7%) showed no ulceration, and 903 lesions (95.8%) were intramucosal cancer. Lymphovascular infiltration was detected in 17 lesions (1.8%), among which the presence of lymphatic infiltration, venous infiltration, and both types of infiltrations were found in 11 (1.2%), eight (0.8%), and two (0.2%) lesions, respectively. The positivity rates of lymphovascular infiltration were 0.7% (7/903) for intramucosal cancer and 25% (10/40) for submucosally invasive cancer.
Table 1.
Characteristics of 943 consecutive lesions
| n = 943 | |
|---|---|
| Age (years) | |
| Median (IQR) | 75 (69–80) |
| Sex, n (%) | |
| Male | 648 (68.7) |
| Female | 295 (31.3) |
| Tumor location, n (%) | |
| Upper | 124 (13.2) |
| Middle | 412 (43.7) |
| Lower | 407 (43.2) |
| Macroscopic type, n (%) | |
| Elevated (0-I,0-IIa) | 397 (42.1) |
| Flat (0-IIb) | 80 (8.5) |
| Depressed (0-IIc,0-III) | 466 (49.4) |
| Tumor size (mm) | |
| Median (IQR) | 12 (8–20) |
| Histological type, n (%) | |
| Differentiated type | 910 (96.5) |
| Undifferentiated type | 33 (3.5) |
| Histological mixture type, n (%) | |
| Non-mixed type | 910 (96.5) |
| Mixed type | 33 (3.5) |
| Ulceration (scar), n (%) | |
| Absent | 884 (93.7) |
| Present | 59 (6.3) |
| Depth of invasion, n (%) | |
| pT1a (M) | 903 (95.8) |
| pT1b1 (SM1) | 40 (4.2) |
| Lymphatic infiltration, n (%) | |
| Negative | 932 (98.8) |
| Positive | 11 (1.2) |
| Venous infiltration, n (%) | |
| Negative | 935 (99.2) |
| Positive | 8 (0.8) |
Differentiated-type includes well-differentiated and moderately differentiated tubular adenocarcinoma, and papillary adenocarcinoma
Undifferentiated-type includes poorly differentiated adenocarcinoma, signet-ring cell carcinoma, and mucinous adenocarcinoma
Non-mixed-type consists of purely differentiated or undifferentiated-type carcinoma
Mixed-type consists of differentiated and undifferentiated-type carcinoma
IQR, interquartile range; pT1a (M), intramucosal cancer (pathological diagnosis); pT1b1 (SM1), submucosally invasive cancer with depth of invasion from the muscularis mucosae < 500 μm
A comparison of the clinical pathological characteristics of lymphovascular-infiltration-positive and -negative groups is shown in Table 2. The lymphovascular infiltration positivity rate was significantly higher in mixed than in non-mixed-type cancer (35.3 vs. 2.8%; P < 0.001) and in submucosally invasive cancer than in intramucosal cancer (58.8 vs. 3.2%; P < 0.001). No significant differences in age, sex, cancer location (upper, middle, or lower region), macroscopic type (elevated, flat, or depressed), tumor diameter, histological type (differentiated or undifferentiated), and ulceration (or scarring) signs were found between lymphovascular-infiltration-positive and -negative groups.
Table 2.
Comparison of clinicopathological features between lymphovascular infiltration (LVI) positive and negative lesions
| Variables | LVI positive n = 17 |
LVI negative n = 926 |
P value |
|---|---|---|---|
| Age (years) | |||
| Median (IQR) | 74 (66–81) | 75 (69–80) | 0.75a |
| Sex, n (%) | |||
| Male | 11 (64.7) | 637 (68.8) | 0.72b |
| Female | 6 (35.3) | 289 (31.2) | |
| Tumor location, n (%) | |||
| Upper third | 2 (11.8) | 122 (13.2) | 0.71b |
| Middle third | 6 (35.3) | 406 (43.8) | |
| Lower third | 9 (52.9) | 398 (43.0) | |
| Macroscopic type, n (%) | |||
| Elevated (0-I, 0-IIa) | 7 (41.2) | 390 (42.1) | 0.41b |
| Flat (0-IIb) | 0 (0) | 80 (8.6) | |
| Depressed (0-IIc, 0-III) | 10 (58.8) | 456 (49.2) | |
| Tumor size (mm) | |||
| Median (IQR) | 15 (12–20.5) | 12 (8–20) | 0.16a |
| Histological type, n (%) | |||
| Differentiated-type | 16 (94.1) | 894 (96.5) | 0.46c |
| Undifferentiated-type | 1 (5.9) | 32 (3.5) | |
| Histological mixture type, n (%) | |||
| Non-mixed-type | 11 (64.7) | 900 (97.2) | < 0.001c |
| Mixed-type | 6 (35.3) | 26 (2.8) | |
| Ulceration (scar), n (%) | |||
| Absent | 17 (100) | 867 (93.6) | 0.62c |
| Present | 0 (0) | 59 (6.3) | |
| Depth of invasion, n (%) | |||
| pT1a (M) | 7 (41.2) | 896 (96.8) | < 0.001c |
| pT1b1 (SM1) | 10 (58.8) | 30 (3.2) |
Differentiated-type includes well-differentiated and moderately differentiated tubular adenocarcinoma, and papillary adenocarcinoma
Undifferentiated-type includes poorly differentiated adenocarcinoma, signet-ring cell carcinoma, and mucinous adenocarcinoma
Non-mixed-type consists of purely differentiated or undifferentiated-type carcinoma
Mixed-type consists of differentiated and undifferentiated-type carcinoma
pT1a (M), intramucosal cancer (pathological diagnosis); pT1b1 (SM1), submucosally invasive cancer with the depth of invasion from the muscularis mucosae < 500 μm
IQR, interquartile range
aA univariate analysis was performed using the Wilcoxon rank-sum test
bA univariate analysis was performed using the chi-square test
cA univariate analysis was performed using the Fisher's exact probability test
The results of the multivariate analysis of lymphovascular infiltration risk are shown in Table 3. Along with the two items with statistical significance in the univariate analysis, seven items from previous research were suspected as risk factors for lymphovascular infiltration. Therefore, a total of nine items were included as variables in the multivariate analysis. In the multiple logistic regression analysis, independent risk factors for lymphovascular infiltration (P < 0.005) were mixed-type cancer (odds ratio; 95% confidence interval: 23.9, 5.0–115; P < 0.001) and submucosal tissue invasion (58.7; 16.0–215; P < 0.001).
Table 3.
Multivariate logistic regression analysis of factors associated with lymphovascular infiltration
| Variables | Univariate | Multivariate | ||||
|---|---|---|---|---|---|---|
| OR | 95% CI | P value | OR | 95% CI | P value | |
| Age (years) | 0.75a | 1.03 | 0.95–1.11 | 0.47 | ||
| Male | 0.83 | 0.30–2.27 | 0.72b | 1.39 | 0.38–5.07 | 0.62 |
| Tumor location | 0.71b | 0.47 | ||||
| Macroscopic type | 0.41b | 0.36 | ||||
| Tumor size (mm) | 0.16a | 0.99 | 0.93–1.05 | 0.75 | ||
| Differentiated-type | 0.57 | 0.07–4.45 | 0.46c | 0.34 | 0.03–4.25 | 0.43 |
| Mixed-type | 18.9 | 6.49–55.0 | < 0.001c | 23.9 | 5.0–115 | < 0.001 |
| Ulceration (scar) | 0.62c | 4.55e-8 | 0.42 | |||
| Submucosal invasion (SM1) | 42.7 | 15.2–120 | < 0.001c | 58.7 | 16.0–215 | < 0.001 |
Differentiated-type includes well-differentiated and moderately differentiated tubular adenocarcinoma, and papillary adenocarcinoma
OR, Odds ratio; 95% CI, 95% confidence interval; SM1, submucosally invasive cancer with the depth of invasion from the muscularis mucosae < 500 μm
aA univariate analysis was performed using the Wilcoxon rank-sum test
bA univariate analysis was performed using the chi-square test
cA univariate analysis was performed using the chi-square test and Fisher's exact probability test
A multivariate logistic regression analysis was used to identify risk factor for lymphovascular infiltration
Clinicopathological signs and time-course of 17 lesions in 17 patients of the lymphovascular-infiltration-positive group are shown in Table 4. Additional surgical resection was performed for 12 lesions (70.6%). Lymph node metastasis was detected in one lesion (5.9%). However, no residual cancer was found in any surgical samples. Surgery was not performed in five lesions (29.4%) due to the patients’ decisions and/or increased severity of a different disease. For three lesions with confirmed time courses, no recurrence was found during the monitoring of progression for 15–56 months (median, 15 months) by periodic endoscopy, and thoracic and abdominal computed tomography. Among the six mixed-type lesions, five were predominantly differentiated-type carcinoma, and one was predominantly undifferentiated-type carcinoma. All lesions were composed of tub2 and por.
Table 4.
Clinicopathological signs and time-course of with 17 lesions in 17 patients in the lymphovascular-infiltration-positive group
| No. | Depth of invasion | Sex | Age | Location | Size (mm) | Macroscopic type | Ulceration | Histopathological classification | Histological mixture type | Lymphatic infiltration | Venous infiltration | Additional resection | LNM | Follow-up time |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | M | Female | 60 | Lower | 12 | Depressed | − | tub2 > por | Mixed | + | − | DG | − | |
| 2 | M | Male | 77 | Middle | 12 | Elevated | − | tub1 > tub2 | Non-mixed | + | + | LECS | + | |
| 3 | M | Male | 71 | Lower | 21 | Depressed | − | tub2 | Non-mixed | − | + | LECS | − | |
| 4 | M | Female | 83 | Middle | 20 | Elevated | − | tub2 > por | Mixed | + | − | Not done | − |
Refusal to operate 15-month relapse free |
| 5 | M | Male | 64 | Lower | 12 | Depressed | − | por > tub2 | Mixed | + | − | LADG | − | |
| 6 | M | Male | 68 | Lower | 19 | Elevated | − | tub2 | Non-mixed | + | − | LADG | − | |
| 7 | M | Male | 86 | Lower | 15 | Elevated | − | tub1 | Non-mixed | + | − | Not done | − | No surgery due to age/15-month relapse free |
| 8 | SM1 | Female | 74 | Lower | 12 | Depressed | − | tub2 > por | Mixed | + | + | DG | − | |
| 9 | SM1 | Male | 78 | Lower | 7 | Elevated | − | tub2 > por | Mixed | − | + | DG | − | |
| 10 | SM1 | Female | 79 | Upper | 17 | Depressed | − | tub2 > tub1 | Non-mixed | + | − | Local resection | – | |
| 11 | SM1 | Male | 61 | Middle | 10 | Elevated | − | tub1 | Non-mixed | − | + | DG | − | |
| 12 | SM1 | Male | 70 | Lower | 23 | Depressed | − | tub2 > tub1 | Non-mixed | + | − | DG | − | |
| 13 | SM1 | Male | 71 | Middle | 13 | Depressed | − | tub1 > tub2 | Non-mixed | + | − | Not done | − |
Refusal to operate 56-month relapse free |
| 14 | SM1 | Male | 75 | Upper | 8 | Depressed | − | tub1 > pap > tub2 | Non-mixed | − | + | PG | − | |
| 15 | SM1 | Male | 64 | Middle | 17 | Depressed | − | tub2 > por | Mixed | − | + | LADG | − | |
| 16 | SM1 | Female | 92 | Lower | 29 | Depressed | − | tub2 | Non-mixed | − | + | Not done | Unknown | No surgery due to age/No follow-up |
| 17 | SM1 | Female | 88 | Middle | 21 | Elevated | − | tub1 | Non-mixed | + | − | Not done | Unknown | Died of heart failure |
LMN lymph node metastasis, M intramucosal cancer, SM1 submucosally invasive cancer with the depth of muscularis mucosae < 500 μm, DG Distal gastrectomy, LECS Laparoscopic and endoscopic cooperative surgery, LADG Laparoscopic assisted distal gastrectomy, PG Proximal gastrectomy
Non-mixed-type consists of purely differentiated or undifferentiated-type carcinoma
Mixed-type consists of differentiated and undifferentiated-type carcinoma
Among the 926 lesions in the lymphovascular-infiltration-negative group, there were 26 mixed-type lesions, 30 submucosally invasive lesions, and one mixed-type and submucosally invasive lesion. For the 26 mixed-type lesions, immunostaining with D2-40 antibody and elastic fiber staining with EVG staining were performed only in eight lesions and no lymphovascular infiltration was confirmed. Among the 30 submucosally invasive lesions, no lymphovascular infiltration was confirmed by immunohistochemistry in 17 lesions.
After obtaining these results, additional immunohistochemical staining was performed with lesions that had not been immunostained to reassess the presence or absence of lymphovascular infiltration. Cross-sectional slide with poorly differentiated tissue or maximum invasion depth were selected. However, no additional lymphovascular infiltration was detected.
Discussion
In this study, mixed-type cancer and submucosal invasion (SM1) were risk factors for lymphovascular infiltration. Mixed-type early gastric cancer has been reported to have higher malignancy than non-mixed (purely differentiated or purely undifferentiated) cancer (Hanaoka et al. 2009; Takizawa et al. 2013; Miyamae et al. 2016; Seo et al. 2019). These findings were consistent with previous single-institution reports of only surgically resected patients. In a study of patients who underwent endoscopic resection, Horiuchi et al. found that mixed-type cancer was a risk factor for non-curative resection (Horiuchi et al. 2018, 2020). Other than this report, endoscopically resected samples have not been extensively studied. Therefore, the present study is the first multi-institution study with endoscopically resected samples.
The fact that mixed-type cancer represents a risk factor for lymphovascular infiltration can be explained by several reasons. Mixed-type cancer includes undifferentiated-type carcinoma, which is a risk factor for lymph node metastasis. Alternatively, several proteins, more strongly expressed in mixed-type cancer than in non-mixed-type cancer, may increase tumor malignancy. These proteins include Ki-67, which is involved in tumor cell proliferation; VEGF, which promotes malignant tumor angiogenesis and lymphangiogenesis; and EMMPRIN, which promotes tumor growth, invasion, and metastasis (Zheng et al. 2008).
Submucosal invasion represents another risk factor for lymphovascular infiltration. In particular, the submucosal layer has more blood and lymph vessels than the mucosa. Therefore, gastric cancer invades blood and lymph vessels more readily at greater depth. As the depth of tumor invasion by gastric cancer increases, the rates of lymphovascular infiltration and lymph node metastasis increase. With submucosally invasive cancer, when the depth of tumor invasion exceeds 500 µm from the lamina muscularis mucosae (SM2), the lymph node metastasis rate increases (Gotoda et al. 2000, 2001). The results of the present study were consistent with the previous findings.
Importantly, all six mixed-type lesions with lymphovascular infiltration in the present study were combinations of tub2 and por. Tubular adenocarcinoma, pap, por, sig, and muc are classified as general-type, malignant epithelial tumors. Furthermore, tubular adenocarcinoma could be subclassified as tub1 or tub2 per the presence of gland formation (Japanese Gastric Cancer Association 2017). Among these tumor types, tub2 consists mainly of glands formed from cuboidal epithelia. Gland formation often does not involve full differentiation and is unclear in parts. Additionally, there have been rare cases of tub2 with crawling-type branching and fusion and a cribriform structure. With respect to the latter, Okamoto et al. have reported that poorly differentiated components were detected in approximately 60% of cases of crawling-type gastric adenocarcinoma, with the formation of poorly differentiated adenocarcinoma in connection with invasion. Such lesions are considered to be precursors of undifferentiated-type adenocarcinoma (Okamoto et al. 2013). Therefore, the risk of containing undifferentiated-type adenocarcinoma components should be considered for lesions with moderately differentiated adenocarcinomatous components. In the present study, when the moderately differentiated tubular adenocarcinoma (tub2) components of six mixed-type lesions with lymphovascular infiltration were re-examined microscopically, one exhibited the diagnostic features of crawling-type cancer. However, no lymph node metastasis was found in this case.
For endoscopically resected lesions, lymphovascular infiltration is a risk factor for lymph node metastasis recurrence and thereby a determining factor for additional surgical resection (Ono et al. 2021). The lymph node metastasis rates in early gastric cancer with lymphovascular infiltration have been reported to be 26.3% for intramucosal cancer in patients who have undergone surgical resection and 36.6% for submucosally invasive cancer (Gotoda et al. 2000). In addition, the lymph node metastasis rates in early gastric cancer with a primary non-curative endoscopic resection and additional surgical resection have been reported to be 15.6% for lymphatic-infiltration-positive cases and 14.1% for venous-infiltration-positive cases (Hatta et al. 2017). Therefore, lymphovascular infiltration does not necessarily indicate lymph node metastasis. In a study of 1,243 early gastric cancer patients with lymphovascular infiltration who underwent surgical resection, Pyo et al. reported that when tumors were ≤ 2 cm in diameter, predominantly differentiated, intramucosal, and without ulcerative signs, no lymph node metastasis was observed even when positive for lymphovascular infiltration (Pyo et al. 2019). Therefore, some patients, who deviate from the curative resection criteria based on lymphovascular infiltration alone, may have lesions that do not require additional surgical resection. Further verification is required in the selection of treatment for lesions that deviate from the curative resection criteria based on lymphovascular infiltration alone.
In the present study, the lymphovascular infiltration positivity rates for intramucosal cancer and submucosally invasive cancer were 0.6% and 25%, respectively, in comparison to the rates of 0.7% and 34.2–35.6%, respectively, in previous reports (Gotoda et al. 2000, 2001). The reason why the lymphovascular infiltration positivity rate for submucosally invasive cancer was lower than those in previous reports is likely due to the restrictive inclusion criteria of the current study. In this regard, the current study only examined cases with submucosally invasive cancer (SM1) that met the criteria for radical treatment with endoscopic resection. Additionally, regarding the lymph node metastasis rate among the 15 lesions that were surveyed for lymphovascular infiltration in the present study, lymph node metastasis was detected in only one (6.7%) lesion. Although this lymph node metastasis rate was lower than the previously reported rate in surgically resected patients, it was higher than the estimated risk of 1% for lymph node metastasis in intramucosal cancer with endoscopic treatment indication. Since the median target age for ESD is 75 years, the requirement for additional surgical resection must be carefully decided for each patient.
To detect lymphovascular infiltration, immunohistochemical staining with D2-40 and EGV staining is useful. However, since these methods are costly and time-consuming, they were not performed for all patients. Therefore, these additional procedures were performed at the discretion of the pathologists and the criteria of the relevant medical institutions.
This study had some limitations. First, all lesions were at very early stages of gastric cancer for which endoscopic treatment is indicated. Therefore, in comparison with studies with surgically resected patients, there may have been a selection bias for lesions with small tumor diameters. Additionally, this study had fewer cases with undifferentiated cancer, cancer with ulceration, or mixed-type cancer. The frequency of mixed-type cancer in this study was 3.5%, which was lower than the previously reported frequencies of 6.8–23.4% (Takizawa et al. 2013; Seo et al. 2019). Second, patients with intramucosal or submucosally invasive cancer in this analysis accounted for only 17 lesions out of the total 943 gastric cancer lesions. Therefore, more patients of this type should be examined in future studies to confirm these findings.
In conclusion, this multi-institution study with endoscopically resected samples revealed that mixed-type cancer and submucosal invasion were risk factors for lymphovascular infiltration in endoscopically resected early gastric cancer. For lesions with these features, an immunohistochemical examination to confirm additional pathological characteristics of lymphovascular infiltration should be considered in the treatment assessment plan.
Acknowledgements
We would like to thank assistant pathologist H. Abe for his assistance with immunostaining. We are grateful to Dr. S. Tsujimoto and Dr. M. Iida for their helpful discussions regarding pathological diagnoses.
Abbreviations
- ESD
Endoscopic submucosal dissection
- pT1a (M)
Intramucosal cancer (histopathological diagnosis)
- pT1b1 (SM1)
Submucosally invasive cancer with depth of invasion from the muscularis mucosae < 500 μm
- pT1b2 (SM2)
Submucosally invasive cancer with depth of invasion from the muscularis mucosae ≧500 μm
- eCuraA (endoscopic curability A)
Curative resection
- eCuraB (endoscopic curability B)
Curative resection is expected under the following conditions: en bloc resection, long diameter ≦3cm, predominantly differentiated type, pT1b1 (SM1), negative for horizontal and vertical margins, and lymphovascular infiltration
- eCuraC (endoscopic curability C)
Non-curative resection
Author contributions
KA: conceptualization; NM: methodology; KA, ST, NM, TT, TY, TN, JA, HC: formal analysis and investigation; TH, AS, SF: writing—review and editing; AN: supervision.
Funding
The authors declare that no funds, grants, or other support were received during the preparation of this manuscript.
Data availability
The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.
Declarations
Conflict of interest
The authors declare that they have no conflict of interest.
Ethical approval
This study was performed in line with the principles of the Declaration of Helsinki. The Institutional Review Boards of Yokohama City University Hospital (approval number, B210100040) Omori Red Cross Hospital (approval number, 19-18) approved this study.
Informed consent
Informed consent was obtained from the participants via an opt-out method. Information about the study was made available to all study subjects to ensure that each had the opportunity to refuse participation.
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- Gotoda T, Sasako M, Ono H, Katai H, Sano T, Shimoda T (2001) Evaluation of the necessity for gastrectomy with lymph node dissection for patients with submucosal invasive gastric cancer. Br J Surg 88:444–449. 10.1046/j.1365-2168.2001.01725.x [DOI] [PubMed] [Google Scholar]
- Gotoda T, Yamamoto H, Soetikno RM (2006) Endoscopic submucosal dissection of early gastric cancer. J Gastroenterol 41:929–942. 10.1007/s00535-006-1954-3 [DOI] [PubMed] [Google Scholar]
- Gotoda T, Yanagisawa A, Sasako M, Ono H, Nakanishi Y, Shimoda T, Kato Y (2000) Incidence of lymph node metastasis from early gastric cancer: estimation with a large number of cases at two large centers. Gastric Cancer 3:219–225. 10.1007/pl00011720 [DOI] [PubMed] [Google Scholar]
- Hanaoka N, Tanabe S, Mikami T, Okayasu I, Saigenji K (2009) Mixed-histologic-type submucosal invasive gastric cancer as a risk factor for lymph node metastasis: feasibility of endoscopic submucosal dissection. Endoscopy 41:427–432. 10.1055/s-0029-1214495 [DOI] [PubMed] [Google Scholar]
- Hatta W, Gotoda T, Oyama T, Kawata N, Takahashi A, Yoshifuku Y et al (2017) A scoring system to stratify curability after endoscopic submucosal dissection for early gastric cancer: “eCura system.” Am J Gastroenterol 112:874–881. 10.1038/ajg.2017.95 [DOI] [PubMed] [Google Scholar]
- Horiuchi Y, Fujisaki J, Yamamoto N, Ishizuka N, Ishiyama A, Yoshio T et al (2020) Undifferentiated-type predominant mixed-type early gastric cancer is a significant risk factor for requiring additional surgeries after endoscopic submucosal dissection. Sci Rep 10:6748. 10.1038/s41598-020-63781-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Horiuchi Y, Fujisaki J, Yamamoto N, Ishizuka N, Omae M, Ishiyama A et al (2018) Undifferentiated-type component mixed with differentiated-type early gastric cancer is a significant risk factor for endoscopic non-curative resection. Dig Endosc 30:624–632. 10.1111/den.13059 [DOI] [PubMed] [Google Scholar]
- Japanese Gastric Cancer Association. Japanese classification of gastric carcinoma: 3rd (2011) English edn. Gastric Cancer 14:101–112 [DOI] [PubMed]
- Japanese Gastric Cancer Association (2017) Japanese classification of gastric carcinoma, 15th edn. Kanehara Shuppan, Tokyo [DOI] [PubMed]
- Japanese Gastric Cancer Association. Japanese gastric cancer treatment guidelines 2018 (2021) 5th edn. Gastric Cancer 24:1–21 [DOI] [PMC free article] [PubMed]
- Lian J, Chen S, Zhang Y, Qiu F (2012) A meta-analysis of endoscopic submucosal dissection and EMR for early gastric cancer. Gastrointest Endosc 76:763–770. 10.1016/j.gie.2012.06.014 [DOI] [PubMed] [Google Scholar]
- Miyamae M, Komatsu S, Ichikawa D, Kosuga T, Kubota T, Okamoto K et al (2016) Histological mixed-type as an independent risk factor for nodal metastasis in submucosal gastric cancer. Tumour Biol 37:709–714. 10.1007/s13277-015-3864-6 [DOI] [PubMed] [Google Scholar]
- Okamoto N, Kawachi H, Yoshida T, Kitagaki K, Sekine M, Kojima K et al (2013) ‘Crawling-type’ adenocarcinoma of the stomach: a distinct entity preceding poorly differentiated adenocarcinoma. Gastric Cancer 16:220–232. 10.1007/s10120-012-0173-2 [DOI] [PubMed] [Google Scholar]
- Ono H, Yao K, Fujishiro M, Oda I, Uedo N, Nimura S, et al (2021) Guidelines for endoscopic submucosal dissection and endoscopic mucosal resection for early gastric cancer (second edition). Dig Endosc, 2nd edn. 33:4–20. 10.1111/den.13883 [DOI] [PubMed]
- Park YM, Cho E, Kang HY, Kim JM (2011) The effectiveness and safety of endoscopic submucosal dissection compared with endoscopic mucosal resection for early gastric cancer: a systematic review and metaanalysis. Surg Endosc 25:2666–2677. 10.1007/s00464-011-1627-z [DOI] [PubMed] [Google Scholar]
- Pimentel-Nunes P, Dinis-Ribeiro M, Ponchon T, Repici A, Vieth M, De Ceglie A et al (2015) Endoscopic submucosal dissection: European Society of Gastrointestinal Endoscopy (ESGE) guideline. Endoscopy 47:829–854. 10.1055/s-0034-1392882 [DOI] [PubMed] [Google Scholar]
- Pyo JH, Lee H, Min YW, Min BH, Lee JH, Kim KM et al (2019) Feasibility of endoscopic resection in early gastric cancer with lymphovascular invasion. Ann Surg Oncol 26:449–455. 10.1245/s10434-018-07119-4 [DOI] [PubMed] [Google Scholar]
- Seo HS, Lee GE, Kang MG, Han KH, Jung ES, Song KY (2019) Mixed histology is a risk factor for lymph node metastasis in early gastric cancer. J Surg Res 236:271–277. 10.1016/j.jss.2018.11.055 [DOI] [PubMed] [Google Scholar]
- Takizawa K, Ono H, Kakushima N, Tanaka M, Hasuike N, Matsubayashi H et al (2013) Risk of lymph node metastases from intramucosal gastric cancer in relation to histological types: how to manage the mixed histological type for endoscopic submucosal dissection. Gastric Cancer 16:531–536. 10.1007/s10120-012-0220-z [DOI] [PubMed] [Google Scholar]
- Zheng HC, Li XH, Hara T, Masuda S, Yang XH, Guan YF, Takano Y (2008) Mixed-type gastric carcinomas exhibit more aggressive features and indicate the histogenesis of carcinomas. Virchows Arch 452:525–534. 10.1007/s00428-007-0572-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.