Surg Endosc DOI 10.1007/s00464-015-4095-z
and Other Interventional Techniques
Clinicopathological factors of multiple lateral margin involvement after endoscopic submucosal dissection for early gastric cancer Jun Hee Lee • Jun Haeng Lee • Kyoung-Mee Kim Ki Joo Kang • Byung-Hoon Min • Jae J. Kim
•
Received: 5 October 2014 / Accepted: 19 January 2015 Ó Springer Science+Business Media New York 2015
Abstract Background and study aims In some ESD specimens of EGC, tumors involve multiple lateral margins. However, the factors related to the number of lateral margins involved are unclear. We evaluated the factors related to the multiplicity of lateral margin involvement in specimens of ESD for EGC. Patients and methods The study included 1,358 patients treated with ESD for EGC between March 2004 and September 2011 at a single tertiary hospital. Of those, 71 patients (5.2 %) were found to have lateral margin-positive specimens. The demographic, endoscopic, and pathological features between the single lateral margin-positive lesions (SLM? group) and the multiple lateral margin-positive lesions (MLM? group) were compared retrospectively. Results Single lateral margin involvement was noted in 43 lesions (60.6 %), and multiple lateral margin involvement was seen in 28 lesions (39.4 %). Extremely welldifferentiated adenocarcinoma (EWDA) and histological heterogeneity were more common in the MLM? group (p = 0.043 and p = 0.070, respectively). In multivariate analysis, EWDA was the only significant risk factor for multiple lateral margin involvement (OR 4.453 [1.011–19.624, 95 % CI], p = 0.048). Surgery was performed in 65 % (46/71) of the patients as an additional treatment for positive lateral margin, while 20 % (14/71) of J. H. Lee J. H. Lee (&) K. J. Kang B.-H. Min J. J. Kim Department of Medicine, Samsung Medical Center, Sungkyunkwan University School of Medicine, 50 Irwon-dong, Gangnam-gu, Seoul 135-710, Korea e-mail: [email protected]; [email protected] K.-M. Kim Department of Pathology, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Korea
the patients underwent an additional ESD. After additional treatment, residual tumors were observed in 65 % (39/60) of the specimens. There was no local recurrence among the patients treated with either type of additional treatment. Conclusions In ESD for EGC, multiple lateral margin involvement was related to the histological characteristics of the tumor, such as extremely well-differentiated adenocarcinoma and histological heterogeneity. Keywords Stomach neoplasms Endoscopic submucosal dissection Lateral margin positivity Gastric cancer is the second leading cause of global cancer mortality and the most common malignancy in Korea [1, 2]. Early gastric cancer (EGC) is defined as a gastric cancer that invades only as far as the submucosa, irrespective of lymph node metastasis. The proportion of EGC within gastric cancer as a whole is increasing due to technical advancements and the introduction of a mass screening program [2]. Endoscopic submucosal dissection (ESD) is one of the most advanced endoscopic techniques, and it has become the standard of treatment for selected cases of EGC [3, 4]. The absence of lateral margin involvement is an essential condition for complete resection during endoscopic treatment. It was demonstrated that lateral margin involvement is associated with local recurrence [5]. Despite technical improvements, lateral margin involvement remains a problem and usually requires additional treatment [6]. Surgical resection is the usual treatment option after an incomplete resection of EGC. In patients with very low risk of lymph node metastasis, however, the non-surgical treatment such as repeated ESD, endoscopic ablation, or close observation could be considered as alternatives [4, 7]. The number of lateral margin involvements is an important factor in deciding the method of additional treatment. To the best of
123
Surg Endosc
our knowledge, the factors related to multiple lateral margin involvement in the ESD specimen have not been carefully studied. The aim of the present study is to clarify the factors related to the multiplicity of lateral margin involvement in specimens of ESD for EGC.
Methods Patients Between March 2004 to September 2011, 1,358 EGCs were treated by ESD at Samsung Medical Center in Korea. The indication for ESD was based on a diagnosis of EGC with no risk of regional lymph node metastasis, in accordance with published criteria [8]. Four experienced endoscopists (JJK, JHL, BHM, and KJK) performed all procedures using the standardized ESD techniques with same instruments. In the pathological examination of the ESD specimen, 74 lesions (5.5 %) were positive for lateral resection margin. After three patients were excluded due to diagnostic ESD (n = 2) and unevaluable specimen (n = 1), 71 lesions (5.2 %) with positive lateral resection margin were included. When the tumor involvement of the lateral margin was evaluated in four directions, 43 lesions (60.6 %) were involved by just one lateral margin, and 28 lesions (39.4 %) were involved by two or more lateral margins. The lesions were divided into two groups, the single lateral margin-positive group (SLM? group) and the multiple (two or more) lateral margin-positive group (MLM? group). The demographic, endoscopic, and pathological features between the SLM? and MLM? groups were compared to evaluate the risk factors for multiple lateral margin involvement in ESD specimens for EGC. Clinicopathological data were collected from medical records and pathology reports. The Institutional Review Board of the Samsung Medical Center approved this study (IRB No. 2012-10-083-001). Endoscopic evaluation and ESD technique Preoperative endoscopy was performed at least once before the ESD. The tumor location was classified into the upper, middle, or lower third of the stomach. The tumor’s macroscopic type was classified as either an elevated type or a flat or depressed type. To assess the extent of the tumor’s border, chromoendoscopy with 0.2 % indigo carmine was used in all cases, and the endoscopically suspected tumor size was measured. ESD was usually performed under sedation with midazolam or propofol. Cardiorespiratory function was continuously monitored during the procedure. After identifying the target lesion, marking dots were made
123
circumferentially at about 5 mm lateral to the margin of the lesion using a needle knife. After marking, a submucosal injection of various solutions, such as normal saline and an epinephrine or glycerol mixture, was performed around the lesion to make a submucosal cushion. Circumferential mucosal cutting was performed outside the marking dots to separate the lesion from the surrounding non-neoplastic mucosa using a dual knife (KD-650L, Olympus Optical Co., Tokyo, Japan). After the circumferential cutting, an additional submucosal injection was carried out. Finally, direct dissection of the submucosal layer was performed using an insulation-tipped knife (KD-610L, Olympus Optical Co., Tokyo, Japan) or dual knife. An electrocautery snare was used at the final step of ESD as needed. During the ESD procedure, endoscopic hemostasis was performed with hemostatic forceps such as Coagrasper (FD-410LR, Olympus Optical Co., Tokyo, Japan). Evaluation of technical factors We established three kinds of technical variables to evaluate the procedural factors of the multiple lateral margin involvement: (1) the rate of en bloc resection, (2) the presence of technical difficulty, and (3) the procedure time. En bloc resection was defined as resection in a single piece as opposed to piecemeal resection. The presence of technical difficulty was decided when the intended procedure failed, for example, when there was no gross safety margin outside the circumferential marking. Figure 1 shows an example of an ESD procedure with technical difficulty. Procedure time was the time length from the initial endoscopic evaluation of the lesion to the final hemostasis after the tumor resection. Pathological evaluation After resection of the lesion, the specimens were placed on Styrofoam and pinned to identify the horizontal margin. The proximal, distal, anterior, and posterior of the specimen were positioned at the 12 o’clock, 6 o’clock, 9 o’clock, and 3 o’clock positions, respectively. The ESD specimen was serially sectioned parallel to the line of 12–6 o’clock at 2-mm intervals and completely embedded. A gross photograph with cutting lines was taken to compare macroscopic and microscopic findings. All ESD specimens were reviewed histopathologically by one gastrointestinal pathologist (KMK). The margins of the tumor, size, and a variety of pathological variables were assessed. The presence of cancer involvement and the level of safety margin in each of the four lateral directions (anterior, posterior, proximal, and distal) and the vertical margin of the ESD specimen were reported.
Surg Endosc
Fig. 1 An example of technical difficulty. ESD specimen showed that there was no safety margin outside the circumferential marking in lateral side (arrow)
The pathological tumor size of the ESD specimen was measured, and the discrepancy between the endoscopically suspected tumor size and pathologic tumor size was calculated. The histological type of tumor was classified according to the revised Vienna classification of epithelial neoplasia of the gastrointestinal tract [9]. A tumor consisting of components of both differentiated- and undifferentiated-type carcinoma was classified according to the quantitative predominance [10]. The presence of histological heterogeneity, extremely well-differentiated adenocarcinoma, submucosal invasion, lymphovascular invasion, and microulceration was assessed. Histological heterogeneity was defined as differentiated cancer mixed with 5 % or more poorly differentiated components [11]. Extremely well-differentiated gastric adenocarcinoma (EWDA) was defined as intestinal-type carcinoma that mimics the complete type of intestinal metaplasia while showing only lowgrade cytologic atypia. EWDA also featured considerable structural atypia of neoplastic tubules in the mucosa showing branching, tortuous, anastomosing, and plexiform structures [12]. Additional treatment and follow-up The majority of the patients with lateral margin tumor involvement were treated with additional method such as surgery, additional ESD, or argon plasma coagulation. The selection of modality of additional treatment was based on the histological findings of the ESD specimen and clinical condition of each patient. However, surgery was always recommended for patients at a high risk of regional LN metastasis, such as with lymphatic invasion, vertical margin involvement, or deep submucosal cancer. After the
additional treatment, the patients underwent an upper endoscopy and abdominal computed tomography (CT) to find local or metachronous recurrence and extragastric metastasis. Upper endoscopy with a forceps biopsy was performed two months and six months after the ESD and then annually. CT scans were performed every six months for the first year and then annually. Statistical analysis To evaluate the risk factors for multiple lateral margin involvement, we used the Chi-square test, Fisher’s exact test, unpaired t test, and Mann–Whitney U test. Multivariate analysis was performed, by logistic regression, to determine the predictive factors. A p value of less than 0.05 was considered significant.
Results Baseline characteristics of patients Table 1 presents the baseline characteristics of the 71 patients with lateral margin-positive ESD specimens. The median follow-up period was 38 months (range 6–93 months). About 80 % of the lesions were located at the middle (40.8 %) and lower (38.0 %) third of the stomach. Half of the lesions were of a macroscopically flat or depressed type (50.7 %). The mean endoscopically suspected tumor size was 1.76 cm. The differentiated-type cancer, with well-differentiated or moderately differentiated tubular adenocarcinoma, was 81.7 %, and the histological undifferentiated type, with poorly differentiated
123
Surg Endosc Table 1 Baseline characteristics of patients Variables
Number of patient (%)
Age (years)
62.8 ± 1.3
Sex (male:female)
50:21
Median follow-up (months)
38.1 ± 2.5
Tumor location Upper third
15 (21.2 %)
Middle third
29 (40.8 %)
Lower third Macroscopic shape Elevated Flat or depressed Endoscopically suspected tumor size, cm
27 (38.0 %) 35 (49.3 %) 36 (50.7 %) 1.76 ± 0.12
Histology by forcep biopsy Differentiated (W/D, M/D)
59 (83.1 %)
Undifferentiated (P/D, SRC)
12 (16.9 %)
No. of positive lateral margin after ESD (anterior/posterior/proximal/distal) 1
43 (60.5 %)
2
17 (24 %)
3 4
5 (7 %) 6 (8.5 %)
W/D well differentiated, M/D moderately differentiated, P/D poorly differentiated, SRC signet ring cell, ESD endoscopic submucosal dissection
tubular adenocarcinoma or signet ring cell carcinoma, was 16.9 %. The ratio of submucosal invasion was 38 %. Patients with lateral margin involvement in 1, 2, 3, and all 4 directions of ESD specimens were 43 (60.5 %), 17 (24 %), 5 (7 %), and 6 (8.5 %), respectively. Comparison of endoscopic and procedural features between the SLM? group and the MLM? group Endoscopic and procedural features between single lateral margin-positive cases (SLM? group) and multiple lateral margin-positive cases (MLM? group) were compared to clarify the risk factors for multiple lateral margin involvement (Table 2). Age, sex, tumor location, and macroscopic type were not significantly different between the two groups. The mean endoscopically suspected tumor size was 1.74 cm in the SLM? group and 1.78 cm in the MLM? group. The rate of en bloc resection and the presence of technical difficulty were not different between the two groups. The procedure time was slightly shorter to a significant degree in the SLM? group than in the MLM? group (55 vs. 84 min, respectively, p = 0.09). Cases taken more than 100 min of procedure time were 25.6 % (11/43) of single lateral margin-positive lesions and 7.1 % (2/28) of multiple lateral margin-positive lesions (p = 0.10). The
123
distribution of single and multiple lateral margin involvement was not significantly different among four endoscopists (SLM? cases vs. MLM? cases, 25 vs. 14, 11 vs. 12, 4 vs. 1, 3 vs. 0, and 0 vs. 1, respectively, in each endoscopist; p = 0.196). Comparison of pathological features between the SLM? group and the MLM? group Table 3 demonstrates the difference of pathological features between the SLM? group and the MLM? group. The difference between the pathological size and the endoscopic size was almost the same between the two groups. The occurrence of poorly differentiated tubular adenocarcinoma or signet ring cell carcinoma histological findings was 14.3 and 21.4 % in the SLM? group and the MLM? group, respectively. Histological heterogeneity had a tendency to be more frequent in the MLM? group than in the SLM? group (46.4 vs. 25.6 %, respectively, p = 0.07). EWDA was significantly more frequent in the MLM? group than in the SLM? group (25 vs. 7 %, p = 0.043). In multivariate analysis, the odds ratio of histological heterogeneity and EWDA was 2.526 (range 0.889–7.171, p = 0.082) and 4.453 (range 1.011–19.624, p = 0.048), respectively. Figure 2 and 3 show the cases of histological heterogeneity and EWDA with multiple lateral margin positivity, respectively. The rates of submucosal invasion, vertical margin involvement, lymphovascular invasion, and microscopic ulceration were not significantly different between the two groups (p = 0.410, 0.675, 0.461, and 1.000, respectively). In the subgroup analysis of the number of lateral margins involved, the higher number of lateral margin-positive directions correlated with the higher proportion of histological heterogeneity or EWDA, compared to other histological types (W/D, M/D, P/D or SRC) (Fig. 4). These were 25 % (11/43), 58 % (10/17), 60 % (3/5), and 83 % (5/ 6), respectively. The difference was statistically significant (p = 0.002 by Cochran–Armitage test). Clinical outcome of additional treatment after lateral margin-positive ESD As an additional treatment for the 71 patients with lateral margin involvement, surgery was performed in 46 patients (65 %). Among them, residual cancer was found in 27 patients (59 %), and there was no patient with lymph node involvement (Fig. 5). Additional ESD was done in 14 patients (20 %), and residual cancer was observed in 12 patients (17 %). Ablation therapy using argon plasma gas was carried out in five patients (7 %). Two patients were observed without undergoing additional treatment due to poor general condition and the clinician’s discretion. Two
Surg Endosc Table 2 Comparison of endoscopic and procedural features between the SLM? group and the MLM? group
Factors
SLM? group (n = 43)
MLM? group (n = 28)
P value
Age
62.0 ± 1.8
64.1 ± 2.1
0.455
Sex (male:female)
31:12
19:9
Tumor location
SLM? single lateral marginpositive, MLM? multiple lateral margin-positive, W/D well differentiated, M/D moderately differentiated, P/D poorly differentiated, SRC signet ring cell, EWDA extremely welldifferentiated adenocarcinoma * Fisher’s exact test
Mann–Whitney test
0.702 0.747
Upper
9 (20.9 %)
6 (21.4 %)
Middle third
19 (44.2 %)
10 (35.7 %)
Lower third
15 (34.9 %)
12 (42.9 %)
Elevated
20 (46.5 %)
15 (53.6 %)
Flat or depressed
23 (53.5 %)
13 (46.4 %)
1.74 ± 0.14
1.78 ± 0.22
0.758
En bloc resection
38 (88.4 %)
25 (89.3 %)
1.000*
Technical difficulty
7 (16.3 %)
5 (21.4 %)
0.583
Procedure time (min)
84.35 ± 7.46
55.85 ± 5.31
0.009
Macroscopic type
0.561
Endoscopically suspected tumor size (cm) Technical factor
Table 3 Comparison of pathological features between the SLM? group and the MLM? group Factors
SLM? group (n = 43)
MLM? group (n = 28)
P value
Endoscopically suspected
1.74 ± 0.14
1.78 ± 0.22
0.758
Pathologically measured
3.14 ± 0.20
3.15 ± 0.23
0.971
Size difference (Patho.–Endo.)à
1.43 ± 0.16
1.49 ± 0.22
0.829
37 (86.0 %)
22 (78.6 %)
Tumor size (cm)
Histology by forcep biopsy Differentiated (W/D, M/D) Undifferentiated (P/D, SRC)
0.437 6 (14.0 %)
6 (21.4 %)
Histological heterogeneity
11 (25.6 %)
13 (46.4 %)
0.070
EWDA
3 (7.0 %)
7 (25.0 %)
0.043*
18 (41.9 %)
9 (32.1 %)
0.410
Vertical margin involvement Lymphovascular invasion
3 (7.3 %) 5 (11.6 %)
3 (11.1 %) 5 (17.9 %)
0.675* 0.461
Microscopic ulceration
6 (14.0 %)
4 (14.3 %)
1.000*
Submucosal invasion
SLM? single lateral margin-positive, MLM? multiple lateral margin-positive, W/D well differentiated, M/D moderately differentiated, P/D poorly differentiated, SRC signet ring cell carcinoma, EWDA extremely well-differentiated adenocarcinoma * Fisher’s exact test
Mann–Whitney test
à
Size difference between pathologic size and endoscopic size
patients were lost to follow-up, and two patients refused surgery. During the median follow-up period of 38 months, there was neither local recurrence nor metachronous recurrence in the patients who underwent any kind of additional treatment. Only one patient, who had refused surgery, developed advanced gastric cancer. Comparing the additional treatment pathological factors between the surgery group and the additional ESD group, the rate of histologically undifferentiated cancer (24 vs. 7 %, respectively), lateral margin involvement in two or more directions (42 vs. 29 %, respectively), and
submucosal invasion (46 vs. 21 %, respectively) was higher in the surgery group than in the additional ESD group. All patients (11 cases) with vertical margin involvement or lymphovascular invasion were treated with surgery.
Discussion Surgery is considered the standard treatment for noncurative endoscopic resection for EGC. However,
123
Surg Endosc
Fig. 2 A case of histological heterogeneity with all four lateral margin positivity. (A) The tumor was located in the antrum. Endoscopically, the size of tumor was estimated to be 0.6 cm. (B) The marking on the 5 mm safety margin from the reddish color change. (C) The precutting by needle knife after the submucosal injection. (D) The fixed ESD specimen after en bloc resection. The
pathologic report showed four directions of lateral margin involvement, and the tumor size was 4.0 9 3.0 cm. (E) The pathological findings of the initial ESD specimen showed histological heterogeneity. The well-differentiated tubular adenocarcinoma (arrow) was confined to the lamina propria and mixed with poorly differentiated tubular adenocarcinoma (empty arrow) (H&E 9100)
retrospective studies showed that the risk of LN metastasis is negligible when the reason for the non-curative resection was lateral margin involvement only [13, 14]. Therefore, less invasive treatments such as repeated ESD, ablation, or close observation could be applied in selected cases [4, 13]. There is no definite treatment guideline for lateral marginpositive cases. Usually, additional endoscopic treatment, such as ablation or early additional ESD, is chosen for single lateral margin-positive cases. However, surgical resection is commonly chosen for multiple lateral marginpositive cases. There are only a few studies about the factors relating to the lateral margin involvement of an ESD specimen. Kakushima et al. [15] reported that the diameter of the tumor, recurrent-type cancer, submucosal cancer, and undifferentiated histology were related to lateral margin-positive resection. However, the study was limited by the small size of its population (16 cases). Lateral resection margin involvement by cancer is a common cause of incomplete resection and needs to be further treated [14, 16]. Following an incomplete ESD due to positive lateral margin involvement, residual tumors have been reported in 47.0–63.0 % in surgical specimens [13, 14]. Our data showed that the 65.0 % (39 of 60 cases) of patients with positive lateral margin had residual tumors
in the specimen obtained by repeated ESD or surgery. This finding suggests that additional treatment after lateral margin-positive ESD is necessary in most cases. We recently reported that additional ESD could achieve curative resection in 94 % (15 of 16 cases) of the patients with positive lateral margin ESD and that the adverse event rate was not high [7]. This study also showed that there was no recurrence in the 19 patients treated with repeated ESD or ablation. In the present study, technical factors such as the presence of technical difficulty and the rate of en bloc resection were not associated with multiple lateral margin positivity. To our surprise, the procedure time was shorter in the multiple lateral margin-positive groups. Additionally, the time was not different from the usual procedure time of ESD at our institution. This suggests that the initially intended procedure was performed in the MLM? group. There were no differences in the factors known to be related to the procedure time of ESD, such as tumor location, size, and presence of microulceration, between the SLM? group and the MLM? group [17, 18]. The reason for the longer procedure time in the SLM? group in the present study is uncertain. Cases taken more than 100 min of procedure time were slightly more common in
123
Surg Endosc
Fig. 3 A case of extremely well-differentiated adenocarcinoma with all four lateral margin positivity. (A) The tumor was located on the posterior side of angle. Endoscopically, the size of tumor was estimated to be 1.5 cm. (B) The marking on the 5 mm safety margin from the nodular mucosal change. (C) The precutting by needle knife after the submucosal injection. (D) The fixed ESD specimen after en bloc resection. The pathologic report showed four directions lateral
margin involvement, and the tumor size was 4.2 9 3.5 cm. (E) The pathological findings of the surgical specimen after an incomplete ESD show tumor cell involvement of the ESD resection margin (arrow) (H&E 940). (F) Magnification photograph showed the EWDA with branching, tortuous, or anastomosing tubular structure (H&E 9200). EWDA extremely well-differentiated adenocarcinoma
Fig. 4 Proportion of histological heterogeneity or EWDA compared to other histological types according to the number of positive lateral margin directions. The proportion of histological heterogeneity or EWDA increased according to the number of lateral margins in the
positive direction, and the difference was statistically significant (p = 0.02). HH histological heterogeneity, EWDA extremely welldifferentiated adenocarcinoma
single lateral margin-positive lesions, and most of prolonged cases had significant submucosal fibrosis. From these findings, we infer that single lateral margin involvement of ESD is mostly related to the technical difficulty. The main reason of multiple lateral margin
involvement was not the technical difficulty, but other factors like unclear margin due to pathological features. Pathologically, gastric carcinoma can be classified into two types: differentiated and undifferentiated [19]. In keeping with the Japanese classification of gastric
123
Surg Endosc Fig. 5 Clinical outcome of additional treatment of lateral margin-positive ESD
carcinoma [4], the histological diagnostic criteria of gastric cancer with histological heterogeneity are supposed to follow the predominant histological type. However, gastric cancer shows greater histological diversity than other types of cancer. Several studies [11, 20] have demonstrated that the risk of lymph node metastasis was higher in gastric cancers with histological heterogeneity than in more homogenous cases. Mita et al. [11] reported that the rate of lymph node metastasis was higher in combined differentiated-type gastric cancer (23 of 84 cases, 27 %) than that of the pure differentiated type (12 of 112 cases, 7 %). Hanaoka et al. [20] reported that the rate of lymph node metastasis in differentiated-type-predominant mixed type (20 of 104 cases, 19 %) and undifferentiated-type-predominant mixed type (23 of 63 cases, 36.5 %) was higher than in those of pure differentiated type (7 of 129 cases, 5.4 %) and pure undifferentiated type (12 of 80 cases, 15 %). Recent study in our institution showed that mixed differentiated-type EGC mixed with an undifferentiated component was significantly associated with more frequent lateral margin involvement compared to pure differentiated-type EGC (10.7 vs. 2.5 %, p \ 0.001) [21]. In the present study, multiple lateral margin involvement was slightly more frequent in mixed type cancer than in its counterpart. In our opinion, the difference is due to the difficulty in the determination of cancer margins in cases with histological heterogeneity. The EWDA is very rare subgroup of intestinal-type gastric cancer and is suggested to have a close link with complete-type intestinal metaplasia [12, 22]. It has distinguishing features, such as neoplastic tubules in the mucosa showing branching, tortuous, anastomosing, and plexiform structures, which are more pathognomonic than the
123
cytological features. Although EWDA pathologically demonstrated a low-grade cytological atypia and considerable structural atypia, it is often difficult to make a pathological diagnosis because it closely mimics completetype intestinal metaplasia. Endoh et al. reported that in cases of EWDA, intramucosal elements were well preserved, and macroscopic investigation of the lesion’s surface showed only minor changes, making the diagnosis of EWDA difficult [12, 23]. It is important for pathologists and endoscopists to be aware of this variant of gastric cancer. Recent study reported that the histological diagnosis is best established using a combination of at least three of the six architectural features, including anastomosing glands, spiky glands, distended glands, discohesive cells, abortive glands, and glandular outgrowth [24]. Kang et al. reported that involvement of lateral resection margin (29.4 vs. 2.5 %, p \ 0.005) and incomplete resection (52.8 vs. 19.6 %, p = 0.01) of ESD was more common in the EWDAs compared to the non-EWDAs, respectively [23]. Our data also showed a higher rate of multiple lateral margin positivity in cases with EWDA. The present study was limited by the retrospective design and a relatively short follow-up period. This study did not include negative lateral margin cases as a control group. Also, we did not evaluate whether new methods, such as magnification endoscopy, narrow band imaging, or acetic acid-indigo carmine chromoendoscopy, could help to delineate the exact extent of the tumor during the ESD procedure. However, this is the first study to clarify the factors related to the multiplicity of lateral margin involvement in specimens of ESD for EGC. In conclusion, multiple lateral margin involvement after ESD for EGC was related to certain histological
Surg Endosc
characteristics, such as EWDA and histological heterogeneity. Pathologists and endoscopists should take such pathological variants into account in making a pathologic diagnosis of gastric cancer and performing an ESD.
12.
13. Disclosures Drs. J. H. Lee, J. H. Lee, K. M. Kim, K. J. Kang, B. H. Min, and Jae J. Kim have no conflicts of interest or financial ties to disclose. 14.
References 1. Jemal A, Bray F, Center MM, Ferlay J, Ward E, Forman D (2011) Global cancer statistics. CA Cancer J Clin 61:69–90 2. Kang KJ, Kim KM, Min BH, Lee JH, Kim JJ (2011) Endoscopic submucosal dissection of early gastric cancer. Gut Liver 5:418–426 3. Kim JJ, Lee JH, Jung HY, Lee GH, Cho JY, Ryu CB, Chun HJ, Park JJ, Lee WS, Kim HS, Chung MG, Moon JS, Choi SR, Song GA, Jeong HY, Jee SR, Seol SY, Yoon YB (2007) EMR for early gastric cancer in Korea: a multicenter retrospective study. Gastrointest Endosc 66:693–700 4. Japanese Gastric Cancer Association (2011) Japanese gastric cancer treatment guidelines 2010 (ver. 3). Gastric Cancer 14:113–123 5. Lauwers GY, Ban S, Mino M, Ota S, Matsumoto T, Arai S, Chan HH, Brugge WR, Shimizu M (2004) Endoscopic mucosal resection for gastric epithelial neoplasms: a study of 39 cases with emphasis on the evaluation of specimens and recommendations for optimal pathologic analysis. Mod Pathol 17:2–8 6. Ono H, Kondo H, Gotoda T, Shirao K, Yamaguchi H, Saito D, Hosokawa K, Shimoda T, Yoshida S (2001) Endoscopic mucosal resection for treatment of early gastric cancer. Gut 48:225–229 7. Bae SY, Jang TH, Min BH, Lee JH, Rhee PL, Rhee JC, Kim JJ (2012) Early additional endoscopic submucosal dissection in patients with positive lateral resection margins after initial endoscopic submucosal dissection for early gastric cancer. Gastrointest Endosc 75:432–436 8. Soetikno R, Kaltenbach T, Yeh R, Gotoda T (2005) Endoscopic mucosal resection for early cancers of the upper gastrointestinal tract. J Clin Oncol 23:4490–4498 9. Schlemper RJ, Riddell RH, Kato Y, Borchard F, Cooper HS, Dawsey SM, Dixon MF, Fenoglio-Preiser CM, Flejou JF, Geboes K, Hattori T, Hirota T, Itabashi M, Iwafuchi M, Iwashita A, Kim YI, Kirchner T, Klimpfinger M, Koike M, Lauwers GY, Lewin KJ, Oberhuber G, Offner F, Price AB, Rubio CA, Shimizu M, Shimoda T, Sipponen P, Solcia E, Stolte M, Watanabe H, Yamabe H (2000) The Vienna classification of gastrointestinal epithelial neoplasia. Gut 47:251–255 10. Kim WH, Park CK, Kim YB, Kim YW, Kim HG, Bae HI, Song KS, Chang HK, Chang HJ, Chae YS (2005) A standardized pathology report for gastric cancer. Korean J Pathol 39:106–113 11. Mita T, Shimoda T (2001) Risk factors for lymph node metastasis of submucosal invasive differentiated type gastric carcinoma:
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
clinical significance of histological heterogeneity. J Gastroenterol 36:661–668 Endoh Y, Tamura G, Motoyama T, Ajioka Y, Watanabe H (1999) Well-differentiated adenocarcinoma mimicking complete-type intestinal metaplasia in the stomach. Hum Pathol 30:826–832 Oda I, Gotoda T, Sasako M, Sano T, Katai H, Fukagawa T, Shimoda T, Emura F, Saito D (2008) Treatment strategy after non-curative endoscopic resection of early gastric cancer. Br J Surg 95:1495–1500 Jung H, Bae JM, Choi MG, Noh JH, Sohn TS, Kim S (2011) Surgical outcome after incomplete endoscopic submucosal dissection of gastric cancer. Br J Surg 98:73–78 Kakushima N, Ono H, Tanaka M, Takizawa K, Yamaguchi Y, Matsubayashi H (2011) Factors related to lateral margin positivity for cancer in gastric specimens of endoscopic submucosal dissection. Dig Endosc 23:227–232 Noh H, Park JJ, Yun JW, Kwon M, Yoon DW, Chang WJ, Oh HY, Joo MK, Lee BJ, Kim JH, Yeon JE, Kim JS, Byun KS, Bak YT (2012) Clinicopathologic characteristics of patients who underwent curative additional gastrectomy after endoscopic submucosal dissection for early gastric cancer or adenoma. Korean J Gastroenterol 59:289–295 Chung IK, Lee JH, Lee SH, Kim SJ, Cho JY, Cho WY, Hwangbo Y, Keum BR, Park JJ, Chun HJ, Kim HJ, Kim JJ, Ji SR, Seol SY (2009) Therapeutic outcomes in 1000 cases of endoscopic submucosal dissection for early gastric neoplasms: Korean ESD Study Group multicenter study. Gastrointest Endosc 69:1228–1235 Goto O, Fujishiro M, Kodashima S, Ono S, Omata M (2009) Is it possible to predict the procedural time of endoscopic submucosal dissection for early gastric cancer? J Gastroenterol Hepatol 24:379–383 Nakamura K, Sugano H, Takagi K (1968) Carcinoma of the stomach in incipient phase: its histogenesis and histological appearances. Gann 59:251–258 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 Min BH, Kim KM, Park CK, Lee JH, Rhee PL, Rhee JC, Kim JJ (2014) Outcomes of endoscopic submucosal dissection for differentiated-type early gastric cancer with histological heterogeneity. Gastric Cancer. doi:10.1007/s10120-014-0378-7 Endoh Y, Watanabe H, Hitomi J (1994) Intestinal-type adenocarcinoma in the fundic gland area of the stomach. Stomach Intest 28:1009–1023 Kang KJ, Kim KM, Kim JJ, Rhee PL, Lee JH, Min BH, Rhee JC, Kushima R, Lauwers GY (2012) Gastric extremely well-differentiated intestinal-type adenocarcinoma: a challenging lesion to achieve complete endoscopic resection. Endoscopy 44:949–952 Ushiku T, Arnason T, Ban S, Hishima T, Shimizu M, Fukayama M, Lauwers GY (2013) Very well-differentiated gastric carcinoma of intestinal type: analysis of diagnostic criteria. Mod Pathol 26:1620–1631
123
and Other Interventional Techniques
Clinicopathological factors of multiple lateral margin involvement after endoscopic submucosal dissection for early gastric cancer Jun Hee Lee • Jun Haeng Lee • Kyoung-Mee Kim Ki Joo Kang • Byung-Hoon Min • Jae J. Kim
•
Received: 5 October 2014 / Accepted: 19 January 2015 Ó Springer Science+Business Media New York 2015
Abstract Background and study aims In some ESD specimens of EGC, tumors involve multiple lateral margins. However, the factors related to the number of lateral margins involved are unclear. We evaluated the factors related to the multiplicity of lateral margin involvement in specimens of ESD for EGC. Patients and methods The study included 1,358 patients treated with ESD for EGC between March 2004 and September 2011 at a single tertiary hospital. Of those, 71 patients (5.2 %) were found to have lateral margin-positive specimens. The demographic, endoscopic, and pathological features between the single lateral margin-positive lesions (SLM? group) and the multiple lateral margin-positive lesions (MLM? group) were compared retrospectively. Results Single lateral margin involvement was noted in 43 lesions (60.6 %), and multiple lateral margin involvement was seen in 28 lesions (39.4 %). Extremely welldifferentiated adenocarcinoma (EWDA) and histological heterogeneity were more common in the MLM? group (p = 0.043 and p = 0.070, respectively). In multivariate analysis, EWDA was the only significant risk factor for multiple lateral margin involvement (OR 4.453 [1.011–19.624, 95 % CI], p = 0.048). Surgery was performed in 65 % (46/71) of the patients as an additional treatment for positive lateral margin, while 20 % (14/71) of J. H. Lee J. H. Lee (&) K. J. Kang B.-H. Min J. J. Kim Department of Medicine, Samsung Medical Center, Sungkyunkwan University School of Medicine, 50 Irwon-dong, Gangnam-gu, Seoul 135-710, Korea e-mail: [email protected]; [email protected] K.-M. Kim Department of Pathology, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Korea
the patients underwent an additional ESD. After additional treatment, residual tumors were observed in 65 % (39/60) of the specimens. There was no local recurrence among the patients treated with either type of additional treatment. Conclusions In ESD for EGC, multiple lateral margin involvement was related to the histological characteristics of the tumor, such as extremely well-differentiated adenocarcinoma and histological heterogeneity. Keywords Stomach neoplasms Endoscopic submucosal dissection Lateral margin positivity Gastric cancer is the second leading cause of global cancer mortality and the most common malignancy in Korea [1, 2]. Early gastric cancer (EGC) is defined as a gastric cancer that invades only as far as the submucosa, irrespective of lymph node metastasis. The proportion of EGC within gastric cancer as a whole is increasing due to technical advancements and the introduction of a mass screening program [2]. Endoscopic submucosal dissection (ESD) is one of the most advanced endoscopic techniques, and it has become the standard of treatment for selected cases of EGC [3, 4]. The absence of lateral margin involvement is an essential condition for complete resection during endoscopic treatment. It was demonstrated that lateral margin involvement is associated with local recurrence [5]. Despite technical improvements, lateral margin involvement remains a problem and usually requires additional treatment [6]. Surgical resection is the usual treatment option after an incomplete resection of EGC. In patients with very low risk of lymph node metastasis, however, the non-surgical treatment such as repeated ESD, endoscopic ablation, or close observation could be considered as alternatives [4, 7]. The number of lateral margin involvements is an important factor in deciding the method of additional treatment. To the best of
123
Surg Endosc
our knowledge, the factors related to multiple lateral margin involvement in the ESD specimen have not been carefully studied. The aim of the present study is to clarify the factors related to the multiplicity of lateral margin involvement in specimens of ESD for EGC.
Methods Patients Between March 2004 to September 2011, 1,358 EGCs were treated by ESD at Samsung Medical Center in Korea. The indication for ESD was based on a diagnosis of EGC with no risk of regional lymph node metastasis, in accordance with published criteria [8]. Four experienced endoscopists (JJK, JHL, BHM, and KJK) performed all procedures using the standardized ESD techniques with same instruments. In the pathological examination of the ESD specimen, 74 lesions (5.5 %) were positive for lateral resection margin. After three patients were excluded due to diagnostic ESD (n = 2) and unevaluable specimen (n = 1), 71 lesions (5.2 %) with positive lateral resection margin were included. When the tumor involvement of the lateral margin was evaluated in four directions, 43 lesions (60.6 %) were involved by just one lateral margin, and 28 lesions (39.4 %) were involved by two or more lateral margins. The lesions were divided into two groups, the single lateral margin-positive group (SLM? group) and the multiple (two or more) lateral margin-positive group (MLM? group). The demographic, endoscopic, and pathological features between the SLM? and MLM? groups were compared to evaluate the risk factors for multiple lateral margin involvement in ESD specimens for EGC. Clinicopathological data were collected from medical records and pathology reports. The Institutional Review Board of the Samsung Medical Center approved this study (IRB No. 2012-10-083-001). Endoscopic evaluation and ESD technique Preoperative endoscopy was performed at least once before the ESD. The tumor location was classified into the upper, middle, or lower third of the stomach. The tumor’s macroscopic type was classified as either an elevated type or a flat or depressed type. To assess the extent of the tumor’s border, chromoendoscopy with 0.2 % indigo carmine was used in all cases, and the endoscopically suspected tumor size was measured. ESD was usually performed under sedation with midazolam or propofol. Cardiorespiratory function was continuously monitored during the procedure. After identifying the target lesion, marking dots were made
123
circumferentially at about 5 mm lateral to the margin of the lesion using a needle knife. After marking, a submucosal injection of various solutions, such as normal saline and an epinephrine or glycerol mixture, was performed around the lesion to make a submucosal cushion. Circumferential mucosal cutting was performed outside the marking dots to separate the lesion from the surrounding non-neoplastic mucosa using a dual knife (KD-650L, Olympus Optical Co., Tokyo, Japan). After the circumferential cutting, an additional submucosal injection was carried out. Finally, direct dissection of the submucosal layer was performed using an insulation-tipped knife (KD-610L, Olympus Optical Co., Tokyo, Japan) or dual knife. An electrocautery snare was used at the final step of ESD as needed. During the ESD procedure, endoscopic hemostasis was performed with hemostatic forceps such as Coagrasper (FD-410LR, Olympus Optical Co., Tokyo, Japan). Evaluation of technical factors We established three kinds of technical variables to evaluate the procedural factors of the multiple lateral margin involvement: (1) the rate of en bloc resection, (2) the presence of technical difficulty, and (3) the procedure time. En bloc resection was defined as resection in a single piece as opposed to piecemeal resection. The presence of technical difficulty was decided when the intended procedure failed, for example, when there was no gross safety margin outside the circumferential marking. Figure 1 shows an example of an ESD procedure with technical difficulty. Procedure time was the time length from the initial endoscopic evaluation of the lesion to the final hemostasis after the tumor resection. Pathological evaluation After resection of the lesion, the specimens were placed on Styrofoam and pinned to identify the horizontal margin. The proximal, distal, anterior, and posterior of the specimen were positioned at the 12 o’clock, 6 o’clock, 9 o’clock, and 3 o’clock positions, respectively. The ESD specimen was serially sectioned parallel to the line of 12–6 o’clock at 2-mm intervals and completely embedded. A gross photograph with cutting lines was taken to compare macroscopic and microscopic findings. All ESD specimens were reviewed histopathologically by one gastrointestinal pathologist (KMK). The margins of the tumor, size, and a variety of pathological variables were assessed. The presence of cancer involvement and the level of safety margin in each of the four lateral directions (anterior, posterior, proximal, and distal) and the vertical margin of the ESD specimen were reported.
Surg Endosc
Fig. 1 An example of technical difficulty. ESD specimen showed that there was no safety margin outside the circumferential marking in lateral side (arrow)
The pathological tumor size of the ESD specimen was measured, and the discrepancy between the endoscopically suspected tumor size and pathologic tumor size was calculated. The histological type of tumor was classified according to the revised Vienna classification of epithelial neoplasia of the gastrointestinal tract [9]. A tumor consisting of components of both differentiated- and undifferentiated-type carcinoma was classified according to the quantitative predominance [10]. The presence of histological heterogeneity, extremely well-differentiated adenocarcinoma, submucosal invasion, lymphovascular invasion, and microulceration was assessed. Histological heterogeneity was defined as differentiated cancer mixed with 5 % or more poorly differentiated components [11]. Extremely well-differentiated gastric adenocarcinoma (EWDA) was defined as intestinal-type carcinoma that mimics the complete type of intestinal metaplasia while showing only lowgrade cytologic atypia. EWDA also featured considerable structural atypia of neoplastic tubules in the mucosa showing branching, tortuous, anastomosing, and plexiform structures [12]. Additional treatment and follow-up The majority of the patients with lateral margin tumor involvement were treated with additional method such as surgery, additional ESD, or argon plasma coagulation. The selection of modality of additional treatment was based on the histological findings of the ESD specimen and clinical condition of each patient. However, surgery was always recommended for patients at a high risk of regional LN metastasis, such as with lymphatic invasion, vertical margin involvement, or deep submucosal cancer. After the
additional treatment, the patients underwent an upper endoscopy and abdominal computed tomography (CT) to find local or metachronous recurrence and extragastric metastasis. Upper endoscopy with a forceps biopsy was performed two months and six months after the ESD and then annually. CT scans were performed every six months for the first year and then annually. Statistical analysis To evaluate the risk factors for multiple lateral margin involvement, we used the Chi-square test, Fisher’s exact test, unpaired t test, and Mann–Whitney U test. Multivariate analysis was performed, by logistic regression, to determine the predictive factors. A p value of less than 0.05 was considered significant.
Results Baseline characteristics of patients Table 1 presents the baseline characteristics of the 71 patients with lateral margin-positive ESD specimens. The median follow-up period was 38 months (range 6–93 months). About 80 % of the lesions were located at the middle (40.8 %) and lower (38.0 %) third of the stomach. Half of the lesions were of a macroscopically flat or depressed type (50.7 %). The mean endoscopically suspected tumor size was 1.76 cm. The differentiated-type cancer, with well-differentiated or moderately differentiated tubular adenocarcinoma, was 81.7 %, and the histological undifferentiated type, with poorly differentiated
123
Surg Endosc Table 1 Baseline characteristics of patients Variables
Number of patient (%)
Age (years)
62.8 ± 1.3
Sex (male:female)
50:21
Median follow-up (months)
38.1 ± 2.5
Tumor location Upper third
15 (21.2 %)
Middle third
29 (40.8 %)
Lower third Macroscopic shape Elevated Flat or depressed Endoscopically suspected tumor size, cm
27 (38.0 %) 35 (49.3 %) 36 (50.7 %) 1.76 ± 0.12
Histology by forcep biopsy Differentiated (W/D, M/D)
59 (83.1 %)
Undifferentiated (P/D, SRC)
12 (16.9 %)
No. of positive lateral margin after ESD (anterior/posterior/proximal/distal) 1
43 (60.5 %)
2
17 (24 %)
3 4
5 (7 %) 6 (8.5 %)
W/D well differentiated, M/D moderately differentiated, P/D poorly differentiated, SRC signet ring cell, ESD endoscopic submucosal dissection
tubular adenocarcinoma or signet ring cell carcinoma, was 16.9 %. The ratio of submucosal invasion was 38 %. Patients with lateral margin involvement in 1, 2, 3, and all 4 directions of ESD specimens were 43 (60.5 %), 17 (24 %), 5 (7 %), and 6 (8.5 %), respectively. Comparison of endoscopic and procedural features between the SLM? group and the MLM? group Endoscopic and procedural features between single lateral margin-positive cases (SLM? group) and multiple lateral margin-positive cases (MLM? group) were compared to clarify the risk factors for multiple lateral margin involvement (Table 2). Age, sex, tumor location, and macroscopic type were not significantly different between the two groups. The mean endoscopically suspected tumor size was 1.74 cm in the SLM? group and 1.78 cm in the MLM? group. The rate of en bloc resection and the presence of technical difficulty were not different between the two groups. The procedure time was slightly shorter to a significant degree in the SLM? group than in the MLM? group (55 vs. 84 min, respectively, p = 0.09). Cases taken more than 100 min of procedure time were 25.6 % (11/43) of single lateral margin-positive lesions and 7.1 % (2/28) of multiple lateral margin-positive lesions (p = 0.10). The
123
distribution of single and multiple lateral margin involvement was not significantly different among four endoscopists (SLM? cases vs. MLM? cases, 25 vs. 14, 11 vs. 12, 4 vs. 1, 3 vs. 0, and 0 vs. 1, respectively, in each endoscopist; p = 0.196). Comparison of pathological features between the SLM? group and the MLM? group Table 3 demonstrates the difference of pathological features between the SLM? group and the MLM? group. The difference between the pathological size and the endoscopic size was almost the same between the two groups. The occurrence of poorly differentiated tubular adenocarcinoma or signet ring cell carcinoma histological findings was 14.3 and 21.4 % in the SLM? group and the MLM? group, respectively. Histological heterogeneity had a tendency to be more frequent in the MLM? group than in the SLM? group (46.4 vs. 25.6 %, respectively, p = 0.07). EWDA was significantly more frequent in the MLM? group than in the SLM? group (25 vs. 7 %, p = 0.043). In multivariate analysis, the odds ratio of histological heterogeneity and EWDA was 2.526 (range 0.889–7.171, p = 0.082) and 4.453 (range 1.011–19.624, p = 0.048), respectively. Figure 2 and 3 show the cases of histological heterogeneity and EWDA with multiple lateral margin positivity, respectively. The rates of submucosal invasion, vertical margin involvement, lymphovascular invasion, and microscopic ulceration were not significantly different between the two groups (p = 0.410, 0.675, 0.461, and 1.000, respectively). In the subgroup analysis of the number of lateral margins involved, the higher number of lateral margin-positive directions correlated with the higher proportion of histological heterogeneity or EWDA, compared to other histological types (W/D, M/D, P/D or SRC) (Fig. 4). These were 25 % (11/43), 58 % (10/17), 60 % (3/5), and 83 % (5/ 6), respectively. The difference was statistically significant (p = 0.002 by Cochran–Armitage test). Clinical outcome of additional treatment after lateral margin-positive ESD As an additional treatment for the 71 patients with lateral margin involvement, surgery was performed in 46 patients (65 %). Among them, residual cancer was found in 27 patients (59 %), and there was no patient with lymph node involvement (Fig. 5). Additional ESD was done in 14 patients (20 %), and residual cancer was observed in 12 patients (17 %). Ablation therapy using argon plasma gas was carried out in five patients (7 %). Two patients were observed without undergoing additional treatment due to poor general condition and the clinician’s discretion. Two
Surg Endosc Table 2 Comparison of endoscopic and procedural features between the SLM? group and the MLM? group
Factors
SLM? group (n = 43)
MLM? group (n = 28)
P value
Age
62.0 ± 1.8
64.1 ± 2.1
0.455
Sex (male:female)
31:12
19:9
Tumor location
SLM? single lateral marginpositive, MLM? multiple lateral margin-positive, W/D well differentiated, M/D moderately differentiated, P/D poorly differentiated, SRC signet ring cell, EWDA extremely welldifferentiated adenocarcinoma * Fisher’s exact test
Mann–Whitney test
0.702 0.747
Upper
9 (20.9 %)
6 (21.4 %)
Middle third
19 (44.2 %)
10 (35.7 %)
Lower third
15 (34.9 %)
12 (42.9 %)
Elevated
20 (46.5 %)
15 (53.6 %)
Flat or depressed
23 (53.5 %)
13 (46.4 %)
1.74 ± 0.14
1.78 ± 0.22
0.758
En bloc resection
38 (88.4 %)
25 (89.3 %)
1.000*
Technical difficulty
7 (16.3 %)
5 (21.4 %)
0.583
Procedure time (min)
84.35 ± 7.46
55.85 ± 5.31
0.009
Macroscopic type
0.561
Endoscopically suspected tumor size (cm) Technical factor
Table 3 Comparison of pathological features between the SLM? group and the MLM? group Factors
SLM? group (n = 43)
MLM? group (n = 28)
P value
Endoscopically suspected
1.74 ± 0.14
1.78 ± 0.22
0.758
Pathologically measured
3.14 ± 0.20
3.15 ± 0.23
0.971
Size difference (Patho.–Endo.)à
1.43 ± 0.16
1.49 ± 0.22
0.829
37 (86.0 %)
22 (78.6 %)
Tumor size (cm)
Histology by forcep biopsy Differentiated (W/D, M/D) Undifferentiated (P/D, SRC)
0.437 6 (14.0 %)
6 (21.4 %)
Histological heterogeneity
11 (25.6 %)
13 (46.4 %)
0.070
EWDA
3 (7.0 %)
7 (25.0 %)
0.043*
18 (41.9 %)
9 (32.1 %)
0.410
Vertical margin involvement Lymphovascular invasion
3 (7.3 %) 5 (11.6 %)
3 (11.1 %) 5 (17.9 %)
0.675* 0.461
Microscopic ulceration
6 (14.0 %)
4 (14.3 %)
1.000*
Submucosal invasion
SLM? single lateral margin-positive, MLM? multiple lateral margin-positive, W/D well differentiated, M/D moderately differentiated, P/D poorly differentiated, SRC signet ring cell carcinoma, EWDA extremely well-differentiated adenocarcinoma * Fisher’s exact test
Mann–Whitney test
à
Size difference between pathologic size and endoscopic size
patients were lost to follow-up, and two patients refused surgery. During the median follow-up period of 38 months, there was neither local recurrence nor metachronous recurrence in the patients who underwent any kind of additional treatment. Only one patient, who had refused surgery, developed advanced gastric cancer. Comparing the additional treatment pathological factors between the surgery group and the additional ESD group, the rate of histologically undifferentiated cancer (24 vs. 7 %, respectively), lateral margin involvement in two or more directions (42 vs. 29 %, respectively), and
submucosal invasion (46 vs. 21 %, respectively) was higher in the surgery group than in the additional ESD group. All patients (11 cases) with vertical margin involvement or lymphovascular invasion were treated with surgery.
Discussion Surgery is considered the standard treatment for noncurative endoscopic resection for EGC. However,
123
Surg Endosc
Fig. 2 A case of histological heterogeneity with all four lateral margin positivity. (A) The tumor was located in the antrum. Endoscopically, the size of tumor was estimated to be 0.6 cm. (B) The marking on the 5 mm safety margin from the reddish color change. (C) The precutting by needle knife after the submucosal injection. (D) The fixed ESD specimen after en bloc resection. The
pathologic report showed four directions of lateral margin involvement, and the tumor size was 4.0 9 3.0 cm. (E) The pathological findings of the initial ESD specimen showed histological heterogeneity. The well-differentiated tubular adenocarcinoma (arrow) was confined to the lamina propria and mixed with poorly differentiated tubular adenocarcinoma (empty arrow) (H&E 9100)
retrospective studies showed that the risk of LN metastasis is negligible when the reason for the non-curative resection was lateral margin involvement only [13, 14]. Therefore, less invasive treatments such as repeated ESD, ablation, or close observation could be applied in selected cases [4, 13]. There is no definite treatment guideline for lateral marginpositive cases. Usually, additional endoscopic treatment, such as ablation or early additional ESD, is chosen for single lateral margin-positive cases. However, surgical resection is commonly chosen for multiple lateral marginpositive cases. There are only a few studies about the factors relating to the lateral margin involvement of an ESD specimen. Kakushima et al. [15] reported that the diameter of the tumor, recurrent-type cancer, submucosal cancer, and undifferentiated histology were related to lateral margin-positive resection. However, the study was limited by the small size of its population (16 cases). Lateral resection margin involvement by cancer is a common cause of incomplete resection and needs to be further treated [14, 16]. Following an incomplete ESD due to positive lateral margin involvement, residual tumors have been reported in 47.0–63.0 % in surgical specimens [13, 14]. Our data showed that the 65.0 % (39 of 60 cases) of patients with positive lateral margin had residual tumors
in the specimen obtained by repeated ESD or surgery. This finding suggests that additional treatment after lateral margin-positive ESD is necessary in most cases. We recently reported that additional ESD could achieve curative resection in 94 % (15 of 16 cases) of the patients with positive lateral margin ESD and that the adverse event rate was not high [7]. This study also showed that there was no recurrence in the 19 patients treated with repeated ESD or ablation. In the present study, technical factors such as the presence of technical difficulty and the rate of en bloc resection were not associated with multiple lateral margin positivity. To our surprise, the procedure time was shorter in the multiple lateral margin-positive groups. Additionally, the time was not different from the usual procedure time of ESD at our institution. This suggests that the initially intended procedure was performed in the MLM? group. There were no differences in the factors known to be related to the procedure time of ESD, such as tumor location, size, and presence of microulceration, between the SLM? group and the MLM? group [17, 18]. The reason for the longer procedure time in the SLM? group in the present study is uncertain. Cases taken more than 100 min of procedure time were slightly more common in
123
Surg Endosc
Fig. 3 A case of extremely well-differentiated adenocarcinoma with all four lateral margin positivity. (A) The tumor was located on the posterior side of angle. Endoscopically, the size of tumor was estimated to be 1.5 cm. (B) The marking on the 5 mm safety margin from the nodular mucosal change. (C) The precutting by needle knife after the submucosal injection. (D) The fixed ESD specimen after en bloc resection. The pathologic report showed four directions lateral
margin involvement, and the tumor size was 4.2 9 3.5 cm. (E) The pathological findings of the surgical specimen after an incomplete ESD show tumor cell involvement of the ESD resection margin (arrow) (H&E 940). (F) Magnification photograph showed the EWDA with branching, tortuous, or anastomosing tubular structure (H&E 9200). EWDA extremely well-differentiated adenocarcinoma
Fig. 4 Proportion of histological heterogeneity or EWDA compared to other histological types according to the number of positive lateral margin directions. The proportion of histological heterogeneity or EWDA increased according to the number of lateral margins in the
positive direction, and the difference was statistically significant (p = 0.02). HH histological heterogeneity, EWDA extremely welldifferentiated adenocarcinoma
single lateral margin-positive lesions, and most of prolonged cases had significant submucosal fibrosis. From these findings, we infer that single lateral margin involvement of ESD is mostly related to the technical difficulty. The main reason of multiple lateral margin
involvement was not the technical difficulty, but other factors like unclear margin due to pathological features. Pathologically, gastric carcinoma can be classified into two types: differentiated and undifferentiated [19]. In keeping with the Japanese classification of gastric
123
Surg Endosc Fig. 5 Clinical outcome of additional treatment of lateral margin-positive ESD
carcinoma [4], the histological diagnostic criteria of gastric cancer with histological heterogeneity are supposed to follow the predominant histological type. However, gastric cancer shows greater histological diversity than other types of cancer. Several studies [11, 20] have demonstrated that the risk of lymph node metastasis was higher in gastric cancers with histological heterogeneity than in more homogenous cases. Mita et al. [11] reported that the rate of lymph node metastasis was higher in combined differentiated-type gastric cancer (23 of 84 cases, 27 %) than that of the pure differentiated type (12 of 112 cases, 7 %). Hanaoka et al. [20] reported that the rate of lymph node metastasis in differentiated-type-predominant mixed type (20 of 104 cases, 19 %) and undifferentiated-type-predominant mixed type (23 of 63 cases, 36.5 %) was higher than in those of pure differentiated type (7 of 129 cases, 5.4 %) and pure undifferentiated type (12 of 80 cases, 15 %). Recent study in our institution showed that mixed differentiated-type EGC mixed with an undifferentiated component was significantly associated with more frequent lateral margin involvement compared to pure differentiated-type EGC (10.7 vs. 2.5 %, p \ 0.001) [21]. In the present study, multiple lateral margin involvement was slightly more frequent in mixed type cancer than in its counterpart. In our opinion, the difference is due to the difficulty in the determination of cancer margins in cases with histological heterogeneity. The EWDA is very rare subgroup of intestinal-type gastric cancer and is suggested to have a close link with complete-type intestinal metaplasia [12, 22]. It has distinguishing features, such as neoplastic tubules in the mucosa showing branching, tortuous, anastomosing, and plexiform structures, which are more pathognomonic than the
123
cytological features. Although EWDA pathologically demonstrated a low-grade cytological atypia and considerable structural atypia, it is often difficult to make a pathological diagnosis because it closely mimics completetype intestinal metaplasia. Endoh et al. reported that in cases of EWDA, intramucosal elements were well preserved, and macroscopic investigation of the lesion’s surface showed only minor changes, making the diagnosis of EWDA difficult [12, 23]. It is important for pathologists and endoscopists to be aware of this variant of gastric cancer. Recent study reported that the histological diagnosis is best established using a combination of at least three of the six architectural features, including anastomosing glands, spiky glands, distended glands, discohesive cells, abortive glands, and glandular outgrowth [24]. Kang et al. reported that involvement of lateral resection margin (29.4 vs. 2.5 %, p \ 0.005) and incomplete resection (52.8 vs. 19.6 %, p = 0.01) of ESD was more common in the EWDAs compared to the non-EWDAs, respectively [23]. Our data also showed a higher rate of multiple lateral margin positivity in cases with EWDA. The present study was limited by the retrospective design and a relatively short follow-up period. This study did not include negative lateral margin cases as a control group. Also, we did not evaluate whether new methods, such as magnification endoscopy, narrow band imaging, or acetic acid-indigo carmine chromoendoscopy, could help to delineate the exact extent of the tumor during the ESD procedure. However, this is the first study to clarify the factors related to the multiplicity of lateral margin involvement in specimens of ESD for EGC. In conclusion, multiple lateral margin involvement after ESD for EGC was related to certain histological
Surg Endosc
characteristics, such as EWDA and histological heterogeneity. Pathologists and endoscopists should take such pathological variants into account in making a pathologic diagnosis of gastric cancer and performing an ESD.
12.
13. Disclosures Drs. J. H. Lee, J. H. Lee, K. M. Kim, K. J. Kang, B. H. Min, and Jae J. Kim have no conflicts of interest or financial ties to disclose. 14.
References 1. Jemal A, Bray F, Center MM, Ferlay J, Ward E, Forman D (2011) Global cancer statistics. CA Cancer J Clin 61:69–90 2. Kang KJ, Kim KM, Min BH, Lee JH, Kim JJ (2011) Endoscopic submucosal dissection of early gastric cancer. Gut Liver 5:418–426 3. Kim JJ, Lee JH, Jung HY, Lee GH, Cho JY, Ryu CB, Chun HJ, Park JJ, Lee WS, Kim HS, Chung MG, Moon JS, Choi SR, Song GA, Jeong HY, Jee SR, Seol SY, Yoon YB (2007) EMR for early gastric cancer in Korea: a multicenter retrospective study. Gastrointest Endosc 66:693–700 4. Japanese Gastric Cancer Association (2011) Japanese gastric cancer treatment guidelines 2010 (ver. 3). Gastric Cancer 14:113–123 5. Lauwers GY, Ban S, Mino M, Ota S, Matsumoto T, Arai S, Chan HH, Brugge WR, Shimizu M (2004) Endoscopic mucosal resection for gastric epithelial neoplasms: a study of 39 cases with emphasis on the evaluation of specimens and recommendations for optimal pathologic analysis. Mod Pathol 17:2–8 6. Ono H, Kondo H, Gotoda T, Shirao K, Yamaguchi H, Saito D, Hosokawa K, Shimoda T, Yoshida S (2001) Endoscopic mucosal resection for treatment of early gastric cancer. Gut 48:225–229 7. Bae SY, Jang TH, Min BH, Lee JH, Rhee PL, Rhee JC, Kim JJ (2012) Early additional endoscopic submucosal dissection in patients with positive lateral resection margins after initial endoscopic submucosal dissection for early gastric cancer. Gastrointest Endosc 75:432–436 8. Soetikno R, Kaltenbach T, Yeh R, Gotoda T (2005) Endoscopic mucosal resection for early cancers of the upper gastrointestinal tract. J Clin Oncol 23:4490–4498 9. Schlemper RJ, Riddell RH, Kato Y, Borchard F, Cooper HS, Dawsey SM, Dixon MF, Fenoglio-Preiser CM, Flejou JF, Geboes K, Hattori T, Hirota T, Itabashi M, Iwafuchi M, Iwashita A, Kim YI, Kirchner T, Klimpfinger M, Koike M, Lauwers GY, Lewin KJ, Oberhuber G, Offner F, Price AB, Rubio CA, Shimizu M, Shimoda T, Sipponen P, Solcia E, Stolte M, Watanabe H, Yamabe H (2000) The Vienna classification of gastrointestinal epithelial neoplasia. Gut 47:251–255 10. Kim WH, Park CK, Kim YB, Kim YW, Kim HG, Bae HI, Song KS, Chang HK, Chang HJ, Chae YS (2005) A standardized pathology report for gastric cancer. Korean J Pathol 39:106–113 11. Mita T, Shimoda T (2001) Risk factors for lymph node metastasis of submucosal invasive differentiated type gastric carcinoma:
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
clinical significance of histological heterogeneity. J Gastroenterol 36:661–668 Endoh Y, Tamura G, Motoyama T, Ajioka Y, Watanabe H (1999) Well-differentiated adenocarcinoma mimicking complete-type intestinal metaplasia in the stomach. Hum Pathol 30:826–832 Oda I, Gotoda T, Sasako M, Sano T, Katai H, Fukagawa T, Shimoda T, Emura F, Saito D (2008) Treatment strategy after non-curative endoscopic resection of early gastric cancer. Br J Surg 95:1495–1500 Jung H, Bae JM, Choi MG, Noh JH, Sohn TS, Kim S (2011) Surgical outcome after incomplete endoscopic submucosal dissection of gastric cancer. Br J Surg 98:73–78 Kakushima N, Ono H, Tanaka M, Takizawa K, Yamaguchi Y, Matsubayashi H (2011) Factors related to lateral margin positivity for cancer in gastric specimens of endoscopic submucosal dissection. Dig Endosc 23:227–232 Noh H, Park JJ, Yun JW, Kwon M, Yoon DW, Chang WJ, Oh HY, Joo MK, Lee BJ, Kim JH, Yeon JE, Kim JS, Byun KS, Bak YT (2012) Clinicopathologic characteristics of patients who underwent curative additional gastrectomy after endoscopic submucosal dissection for early gastric cancer or adenoma. Korean J Gastroenterol 59:289–295 Chung IK, Lee JH, Lee SH, Kim SJ, Cho JY, Cho WY, Hwangbo Y, Keum BR, Park JJ, Chun HJ, Kim HJ, Kim JJ, Ji SR, Seol SY (2009) Therapeutic outcomes in 1000 cases of endoscopic submucosal dissection for early gastric neoplasms: Korean ESD Study Group multicenter study. Gastrointest Endosc 69:1228–1235 Goto O, Fujishiro M, Kodashima S, Ono S, Omata M (2009) Is it possible to predict the procedural time of endoscopic submucosal dissection for early gastric cancer? J Gastroenterol Hepatol 24:379–383 Nakamura K, Sugano H, Takagi K (1968) Carcinoma of the stomach in incipient phase: its histogenesis and histological appearances. Gann 59:251–258 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 Min BH, Kim KM, Park CK, Lee JH, Rhee PL, Rhee JC, Kim JJ (2014) Outcomes of endoscopic submucosal dissection for differentiated-type early gastric cancer with histological heterogeneity. Gastric Cancer. doi:10.1007/s10120-014-0378-7 Endoh Y, Watanabe H, Hitomi J (1994) Intestinal-type adenocarcinoma in the fundic gland area of the stomach. Stomach Intest 28:1009–1023 Kang KJ, Kim KM, Kim JJ, Rhee PL, Lee JH, Min BH, Rhee JC, Kushima R, Lauwers GY (2012) Gastric extremely well-differentiated intestinal-type adenocarcinoma: a challenging lesion to achieve complete endoscopic resection. Endoscopy 44:949–952 Ushiku T, Arnason T, Ban S, Hishima T, Shimizu M, Fukayama M, Lauwers GY (2013) Very well-differentiated gastric carcinoma of intestinal type: analysis of diagnostic criteria. Mod Pathol 26:1620–1631
123
