Histological Differentiated/Undifferentiated Mixed Type Should Not Be Considered as a Non-Curative Factor of Endoscopic Resection for Patients With Early Gastric Cancer

Background Histological differentiated/undifferentiated mixed-type adenocarcinomas are frequently found in patients with early gastric cancer (EGC). Yet it is unclear whether these mixed-type adenocarcinomas can be treated by endoscopic resection (ER) in EGC patients. Aims To evaluate the lymph node metastasis (LNM) rate and long-term outcomes in mixed-type EGC patients and assess the feasibility of ER in these patients. Methods Clinicopathological features, risk factors of LNM, and overall survival (OS) and progression-free survival (PFS) rates of EGC patients were analyzed according to different histological types. Results Patients with mixed-type EGC had higher LNM rates than patients with non-mixed-type EGC (11.4 vs. 6.2%, P = 0.044). In the multivariate analysis, larger tumor diameter, presence of an ulcer, submucosal invasion, histological undifferentiated type, histological mixed type, and lymphovascular invasion resulted as independent risk factors for LNM in EGC patients (all P < 0.05). The LNM rate in mixed-type patients who met the Japanese ER criteria was 3.3%, including fulfilling the absolute criteria 0%. The 5-year OS and PFS rates in mixed-type patients were 94.59 and 91.47%, respectively. There was no statistical significance in the OS (P = 0.870) and PFS (P = 0.705) between mixed-type and non-mixed-type EGC patients fulfilling the Japanese ER criteria. Conclusion Histological differentiated/undifferentiated mixed type in EGC patients meeting the Japanese absolute criteria for ER are associated with low risk of LNM and favorable prognosis, and thus, it should not be considered as a non-curative factor for ER.


INTRODUCTION
Early gastric cancer (EGC) can be histologically classified into differentiated and undifferentiated types (1); the former contains well or moderately differentiated papillary/tubular adenocarcinoma. In contrast, the latter contains poorly differentiated/signet-ring cell/mucinous adenocarcinoma. In addition to these histological types, the differentiated/undifferentiated mixed-type adenocarcinomas, which refer to the intermingled histopathology of differentiated and undifferentiated adenocarcinoma components, are frequently observed in EGC patients (2). The histological types are considered when determining different treatment strategies for EGC patients. For example, for undifferentiatedtype EGC, radical gastrectomy with lymph node dissection is generally selected rather than endoscopic submucosal dissection (ER) because this type is believed to be associated with lymph node metastasis (LNM) (3). However, the LNM rates in EGC patients with differentiated/undifferentiated mixed type still remain controversial. Huh et al. (4) reported that, in EGC patients, adenocarcinoma mixed with signet-ring cell histological components showed more aggressive behavior than other histology and was an independent risk factor for LNM (P < 0.001). Takizawa et al. (5) retrospectively analyzed the LNM rates of different histological types in EGC patients, identifying no statistical significance in LNM rates among mixed type and purely undifferentiated/differentiated type (5.9 vs. 5.1/2.4%, P = 0.124).
Although previous studies focused on the LNM in EGC patients, the LNM rates in patients with mixed-type EGC meeting the ER criteria still remain unclear (6). Moreover, data on the long-term outcomes of mixed-type EGC are scarce, and it is unclear whether mixed type is a non-curative factor for EGC patients fulfilling the ER criteria (7). Therefore, the aim of this study was to analyze the LNM rates and long-term outcomes in patients with mixed-type EGC meeting the criteria for ER and evaluate the feasibility of ER in these patients.

Patients
All 1174 EGC cases registered at Qilu Hospital at Shandong University between February 2003 and December 2018 were prospectively reviewed. The inclusion criteria were as follows: (1) patients underwent radical gastrectomy with D2 lymph node dissection, (2) patients with complete clinical and pathological data, and (3) patients had a postoperative histological diagnosis that confirmed adenocarcinoma. The exclusion criteria were as follows: (1) synchronous multiple gastric tumors, (2) patients with a history of gastrectomy before surgery, (3) patients who died of postoperative complications, (4) patients with less than 16 retrieved lymph nodes, and (5) patients lost to follow-up. Preoperative examinations, including gastroscopy, endoscopic ultrasound, and abdominal computed tomography, were performed in all patients (8). Postoperative follow-up examinations were scheduled at 6-month intervals during the first year and once a year thereafter. Follow-up data were prospectively collected and updated until February 2020. The informed consent was waived due to the retrospective nature of the study. This study was approved by the Ethics Committee Board of Qilu Hospital of Shandong University, China.

Data Collection
The collected data included the patient's age, gender, tumor diameter, tumor location of the stomach, surgical type (subtotal or total gastrectomy), macroscopic appearance (elevated, flat, depressed), presence of an ulcer, depth of invasion (mucosal or submucosal), number of dissected lymph nodes, LNM, pathological tumor stage (pTNM), lymphovascular invasion, perineural invasion, and histological differentiation (purely differentiated, purely undifferentiated, differentiated/undifferentiated mixed type). The intervals between sections of the primary lesion were 2-3 mm. Continuous multiple sectioning was used at 2-mm intervals in the preparation of lymph node sections. The D2-40 was used in lymphatic vessel stains, and CD31/CD34 was used in venous stains. Immunostaining (CK/CK-19) was used for the evaluation of lymph node metastasis. An isolated tumor cell or micrometastasis in the lymph node was defined as metastasis. Lymphovascular invasion (LVI) was defined as carcinoma cells present within a definite, endothelial-lined space (the lymphatic and/or blood vessels). Lymphatic vessels (ly) were not differentiated from the blood vessels (v) in LVI-positive patients. Pathologists reviewed mixed-type EGC pathological slides to evaluate the predominantly differentiated/undifferentiated components needed to meet the ER criteria (9). Pathological tumor stages were classified according to the American Joint Committee on Cancer staging system (7th edition).

Statistical Analysis
Continuous variables were presented as medians (ranges). Differences in distributions of baseline characteristics among participants with mixed-type and non-mixed-type cancers were analyzed with the use of Pearson's chi-square test for categorical variables and the Mann-Whitney test for ordinal or continuous variables with a non-normal distribution. The Kaplan-Meier method was used to calculate OS curves based on the length of time between primary surgical treatment and final follow-up or death and PFS curves based on the length of time between primary surgical treatment and final follow-up, death, recurrence, or metastasis. The log-rank test was used to assess statistical differences between curves. The logistic regression model was used to identify the independent factors associated with LNM   in EGC patients (12). P < 0.05 was considered statistically significant. The statistical analysis was performed using SAS 9.4 (SAS Institute Inc., Kerry, United States).  Table 1. Mixed-type EGC tumors were associated with younger age (P = 0.000), female gender (P = 0.000), larger tumor diameter (P = 0.020), more located at the body of the stomach (P = 0.000), more flat macroscopic appearance (P = 0.000), more positive LNM (P = 0.001), higher pN stage (P = 0.004), and higher pTNM stage (P = 0.004) compared to non-mixed-type (differentiated-type or undifferentiated-type) EGC tumors. There was no significant difference among these three groups in relation to surgical type, presence of an ulcer, depth of invasion (the pT stage), number of dissected lymph nodes, lymphovascular invasion, perineural invasion or follow-up periods (all P > 0.05).

Long-Term Outcomes of Mixed-Type ECG Patients Fulfilling the ER Criteria
We further analyzed the prognosis of EGC patients who met the criteria for ER (absolute or expanded) according to different histological types (13). The median follow-up period for all EGC  The Kaplan-Meier analysis shows that there was no significant difference in both OS (P = 0.870) and PFS (P = 0.705) between the two groups. The survival curves of OS and PFS are shown in Figure 1. A flowchart with long-term outcomes of mixedtype and non-mixed-type EGC patients fulfilling the ER criteria is shown in Figure 2.

DISCUSSION
Following the fast development of endoscopic technology, ER has become an important option for patients with EGC. According to the latest 2018 Japanese gastric cancer treatment guidelines (5th edition), ER criteria are mainly based on histological types, depth of invasion (pT stage), ulcerative findings, and tumor diameter, which make the histological types very important for EGC patients (10). The Japanese guidelines have specified the criteria for ER in patients with histologically differentiated or undifferentiated adenocarcinomas (14). Well or moderately differentiated EGC usually implies low risk of LNM and curative ER (11,15). Interestingly, many studies have reported that patients with pure undifferentiated histological types also have a favorable prognosis (3,(16)(17)(18). According to these results, the foundation for ER in patients with histologically differentiated or undifferentiated types EGC was established (19). However, the feasibility of ER in patients with histologically differentiated/undifferentiated mixed-type adenocarcinomas still remained unclear.
In this study, we compared the clinicopathological features of EGC patients with mixed type and non-mixed type. The mixed-type tumors showed more aggressive behaviors, such as younger age, larger tumor diameter, more positive LNM, higher pN stage, and higher pTNM stage (20). Moreover, in the univariate and multivariate logistic analysis, histological mixed type resulted as an independent risk factor for LNM in patients with EGC. This was consistent with previous FIGURE 2 | Flowchart for long-term outcomes in EGC patients who met the Japanese ER criteria.
Frontiers in Oncology | www.frontiersin.org studies, which also found that the mixed type was more associated with younger age, positive LNM, lymphovascular invasion, and depressed type than other histological types (21). The locally aggressive characteristics in mixed-type EGC may be explained by the increased expression of proteins, such as Ki-67, EMMPRIN, and VEGF, which are involved in the angiogenetic process and cell proliferation in mixed-type EGC (22). Another study suggests that mixed-type gastric carcinomas frequently show CpG island hypermethylation (23).
As is well known, treatment for EGC mainly depends on the risk of LNM and long-term outcomes. Previous studies have shown that the LNM rate in EGC patients ranges from 5% to 18% according to the proportion of submucosa invasion (24). However, the mixed-type EGC shows more aggressive behaviors with a higher LNM rate than other histological types. Huh et al. reviewed 156 mixed-type patients and reported an LNM of 19.2% in mixed-type EGC (4). In our study, the LNM of mixed-type EGC was 11.4%, which was significantly higher than in patients with non-mixed type (6.2%). These findings suggest that the overall LNM rates in mixed-type EGC may not be suitable for ER (25). To further clarify this doubt, we calculated the LNM rate in mixed-type EGC patients meeting the Japanese criteria for ER. The results show that, under the absolute criteria, the LNM in mixed-type EGC was 0% regardless of the predominant histological proportion, which means that EGC patients with mixed-type histology, intramucosal invasion, no ulcerative findings, and diameter ≤2 cm have a very low risk of LNM (26).
As for mixed-type EGC patients meeting the expanded criteria, the LNM rates were 8.3% (1/12, criteria #1), 12.5% (1/8, criteria #2), 0% (0/8, criteria #3), and 0% (0/4, criteria #4), respectively. The LNM rate in patients meeting expanded criteria #1 and #2 appears high, which may be caused by the small number of patients in each group. Moreover, the low LNM rates (0%) among patients meeting expanded criteria #3 (8 cases) and #4 (only 4 cases) may also be caused by the same reason. The 2018 Japanese gastric cancer guidelines newly defined the ER curability (eCura) criteria (27). According to these criteria, EGC patients who met the absolute or expanded criteria for ER received en bloc ER with negative horizontal/vertical margins and had no lymphovascular infiltration should be regarded as suitable candidates for endoscopic treatment (28). Observation and follow-up with necessary examinations should be enough for these patients (29). During our follow-up period, mixed-type EGC patients fulfilling the ER criteria had a 5-year OS rate of 94.59% and a PFS rate of 91.47%. Among 60 patients, only 2 (3.3%) patients died after surgery, and 2 (3.3%) experienced tumor recurrence in the remnant stomach. There was no gastric cancer-related death during the follow-up period. Moreover, the 5-year OS and PFS in mixed-type patients were similar to that of non-mixed-type patients. Although mixed type is a clear risk factor for LNM in EGC patients, the rate of LNM seems to be negligible under certain conditions (30). This implies the mixed-type EGC patients meeting the absolute ER criteria should not be considered as non-curative for ER in EGC patients. For these patients, additional surgery after the ER may not be necessary (31).

CONCLUSION
For EGC patients meeting the Japanese criteria for ER, the mixed type indicates a high risk of LNM; however, in patients fulfilling the absolute or expanded ER criteria, mixed type may indicate a low risk of LNM and favorable prognosis. Mixed type patients who met the absolute ER criteria did not have LNM, and there was no significant difference in the OS and PFS between mixed-type and non-mixed-type EGC patients. Mixed-type histology in EGC patients meeting the Japanese absolute criteria for ER should not be considered as noncurative for endoscopic treatment. For patients meeting the expanded ER criteria, this case series is still limited, and further data are needed.

DATA AVAILABILITY STATEMENT
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

ETHICS STATEMENT
The studies involving human participants were reviewed and approved by the Ethics Committee of Qilu Hospital of Shandong University. The patients/participants provided their written informed consent to participate in this study. Written informed consent was obtained from the individual(s) for the publication of any potentially identifiable images or data included in this manuscript.

AUTHOR CONTRIBUTIONS
HQ, XT, and PG designed this study. XT and MZ contributed in drafting the manuscript. HQ, QH, and GS critically revised the manuscript for important intellectual content. XT, MZ, and CW collected and analyzed the data. All the authors approved the final manuscript submitted. Each author participated sufficiently in the work to take public responsibility for appropriate portions of the content and agreed to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.