- Email: [email protected]
Comparison of treatment outcomes between laparoscopic and endoscopic surgeries for relatively small gastric gastrointestinal stromal tumors
Surgical Oncology 27 (2018) 737–742
Contents lists available at ScienceDirect
Surgical Oncology journal homepage: www.elsevier.com/locate/suronc
Comparison of treatment outcomes between laparoscopic and endoscopic surgeries for relatively small gastric gastrointestinal stromal tumors
T
Liang Chena,b,1, Qiang Zhanga,1, Feng-Yuan Lia,1, Li Yangc, Dian-Cai Zhangc, Lin-Jun Wangc, Wei-Zhi Wangc, Zheng Lic, Jiang-Hao Xuc, Zhong-Yuan Hec, Kang-Jing Xua, Ming Chena, Hao Xua,∗, Ze-Kuan Xua,∗∗ a
Department of General Surgery, The First Affiliated Hospital of Nanjing Medical University, No.300, Guangzhou Road, Nanjing, Jiangsu province, 210029, China Department of General Surgery, Jiangsu Cancer Hospital & Jiangsu Institute of Cancer Research, The Affiliated Cancer Hospital of Nanjing Medical University,Nanjing, 210009, China c Jiangsu Key Lab of Cancer Biomarkers, Prevention and Treatment, Jiangsu Collaborative Innovation Center for Cancer Personalized Medicine, School of Public Health, Nanjing Medical University, Nanjing 211166, China b
ARTICLE INFO
ABSTRACT
Keywords: Gastric gastrointestinal stromal tumors Laparoscopy Endoscopy Prognosis
Background: Complete surgical resection remains the predominant treatment modality for primary gastrointestinal stromal tumors (GISTs). No therapeutic consensus exists for 2–5 cm gastric GISTs. We compared the efficacy, safety, and prognosis of laparoscopic and endoscopic surgeries in the treatment of relatively small (2–5 cm) intraluminal gastric GISTs. Methods: We collected 101 patients with relatively small intraluminal gastric GISTs who had integrated clinicopathological data and underwent laparoscopic or endoscopic resection (laparoscopic group n = 66; endoscopic group n = 35). Clinicopathological characteristics, perioperative data, and long-term oncological outcomes were retrospectively analyzed. Comparative analysis of clinicopathological data in the two groups was performed by using a chi-square test, Fisher's exact test, and Student's t-test. Recurrence-free survival (RFS) was analyzed by the log-rank test. Results: All clinicopathological characteristics had no significant difference between the two groups. Patients in the endoscopic group had shorter operation time (P < 0.001), postoperative hospital stay (P < 0.001), time to a liquid diet (P < 0.01), and time to a semi-liquid diet (P < 0.01), and lower hospital charges (P < 0.001), compared to those in the laparoscopic group. Four patients (6.1%) in the laparoscopic group and one patient (2.9%) in the endoscopic group had perioperative complications, but with no significant difference. Recurrence occurred in 6 patients (9.1%) and 2 patients (5.7%) in the laparoscopic and endoscopic groups, respectively. There was no significant difference in RFS between the two groups. Conclusion: Endoscopic resection is a feasible and safe treatment modality for patients with relatively small (2–5 cm) intraluminal gastric GISTs. Due to faster recovery and lower cost, endoscopic resection is more suitable for elderly and weak patients, or patients with a poor financial situation.
1. Introduction Gastrointestinal stromal tumors (GISTs) are the most common mesenchymal tumors of the gastrointestinal tract, and occur most commonly in the stomach (50–60%) [1,2]. Incidence of GISTs is about 10–22 per million [2]. It remains a threat to human health because of potential malignancy and reoccurrence.
Mutations of c-kit and platelet-derived growth factor receptor alpha (PDGFRA) are the leading causes of GISTs [3,4]. Complete surgical resection remains the main and preferred treatment method for primary GISTs [5]. Due to the low rate of lymphatic metastasis and infiltrating growth, GISTs can be treated by local surgery resection rather than extensive surgery with lymph node dissection [6]. As the NCCN Clinical Practice Guidelines [7] and NCCN Task Force Report [8] indicated,
Corresponding author. Corresponding author. Department of General Surgery, the First Affiliated Hospital of Nanjing Medical University, No.300, Guangzhou Road, Nanjing, 210029, Jiangsu province, China. E-mail addresses: [email protected] (H. Xu), [email protected] (Z.-K. Xu). 1 These authors contributed equally to this work. ∗
∗∗
https://doi.org/10.1016/j.suronc.2018.10.002 Received 15 April 2018; Received in revised form 19 August 2018; Accepted 1 October 2018 0960-7404/ © 2018 Elsevier Ltd. All rights reserved.
Surgical Oncology 27 (2018) 737–742
L. Chen et al.
different surgical procedures should be recommended according to tumor size, location and adjacent tissue structure. For patients with gastric GISTs < 2 cm, most of which are almost universally benign, surgical treatment is controversial and follow-up observation by endoscopy is much more recommended because of its general very low risk of relapse. However, all GISTs ≥ 2 cm should be resected because of the risk of reoccurrence. Especially for gastric GISTs > 5 cm, which generally have an intermediate or high risk of recurrence, laparoscopic or open surgeries are both applied depending on the tumor size and location. However, for 2–5 cm intraluminal gastric GISTs, no standardized treatment criteria have been recognized. Due to their respective merits and demerits, both laparoscopic and endoscopic surgeries are possible therapeutic modalities. In recent years, as a minimally invasive procedure, laparoscopic resection has been increasingly recognized as a standard surgical procedure for the treatment of GISTs [9]. Generally, laparoscopic wedge resection with negative margins is the ideal treatment choice for relatively small gastric GISTs. However, gastric GISTs > 5 cm and close to cardia or pylorus are benefit more from laparoscopy-assisted and open operations [10]. The feasibility and safety of laparoscopic resection in the treatment of gastric GISTs had been confirmed in several previous studies [9–11]. Compared to endoscopic resection, laparoscopic surgery is more invasive and has a longer operation time, longer postoperative hospitalization, more complications, and higher treatment costs for GIST patients [12,13]. Endoscopic resection, with its advantages of being less invasive and preserving the integrity and organ function of the stomach, has been reported to treat gastric GISTs [14,15]. Endoscopic band ligation with follow-up by endoscopic ultrasonography has been effectively and safely used in the treatment of small gastric stromal tumors [14]. With the development of endoscopic technology, endoscopic dissection was confirmed as an effective, safe, and feasible treatment modality for relatively small (2–5 cm) GISTs in the esophagus and stomach [16]. Different to laparoscopic resection, with endoscopic resection it is more difficult to achieve R0 resection because of limited operating spaces. Furthermore, iatrogenic perforation and bleeding are common complications of endoscopic resection for gastric GIST patients, which may result in asking for surgical assistance [16–18]. Few studies have reported which method, laparoscopic resection or endoscopic resection, is the optimal therapeutic modality for patients with relatively small (2–5 cm) intraluminal gastric GISTs. Due to controversial treatment choice, the aim of this study is to compare the efficacy, safety, and prognosis of laparoscopic and endoscopic resections for patients with relatively small (2–5 cm) intraluminal gastric GISTs.
Board of the First Affiliated Hospital of Nanjing Medical University. All enrolled patients were informed of this study and signed informed consents. 2.2. Data collection Clinicopathological characteristics (including age, gender, tumor location, tumor size, mitotic count, risk level of recurrence, cell subtype, CD117, CD34 and Ki-67), perioperative characteristics (including operation time, postoperative hospital stay, hospital charges, time to a liquid diet, time to a semi-liquid diet, and complications) and long-term oncological outcomes (including recurrence, recurrence-related death and follow-up time) of patients in two groups were collected and analyzed respectively. Time to a liquid/semi-liquid diet was calculated from the operation date. Time to first anal exhaust was similar to liquid diet time. First anal exhaust indicated the beginning of liquid diet. No intraoperative tumor rupture occurred, R0 resection was achieved and resection margins were negative in all patients. Risk levels of reoccurrence were evaluated according to Armed Forces Institute of Pathology criteria [19]. 2.3. Operative procedure 2.3.1. Laparoscopic procedure Patients were placed in a supine position with separated legs. After general anesthesia with trachea cannula, pneumoperitoneum pressure was established at 12 mmHg. As Fig. 1 shows, one trocar (Johnson & Johnson, Shanghai, China) 10 mm in diameter was inserted into the superior margin of the umbilicus as a camera port. Four additional trocars (two trocars 12 mm in diameter, and two trocars 5 mm in diameter) were inserted into the upper left, lower left, upper right, and lower right quadrants of the abdomen above the umbilicus to serve as operative ports. The tumor site was confirmed in a laparoscopic field of
2. Methods 2.1. Patients We enrolled 101 relatively small intraluminal gastric GIST patients who had entire clinicopathological data and underwent laparoscopic or endoscopic resection in the First Affiliated Hospital of Nanjing Medical University from April 2009 to April 2016. All patients were divided into two groups, laparoscopic resection group (n = 66) and endoscopic resection group (n = 35). General clinical information, perioperative data and long-term prognosis of all patients were retrospectively analyzed. Inclusion criteria: patients with pathologically confirmed GISTs which were 2–5 cm in diameter, and in good physical condition were eligible. Patients suffering from other tumors simultaneously, with history of abdominal surgery, preoperative hemoglobin < 90 g/L, dysfunction of important organs, and combined resection of other organs would be excluded. If the subjects had taken imatinib postoperatively or simultaneously participated in other clinical trials were not included. Besides, female patients who were pregnant or in lactation would be excluded as well. This study was approved by The Institutional Ethical
Fig. 1. The location of ports used in laparoscopic procedure. A (10 mm), the camera port in superior margin of umbilicus. B (12 mm), C (5 mm), D (5 mm) and E (12 mm), the operative ports in left upper, left lower, right upper and right lower quadrants of abdomen of upper level of umbilicus. 738
Surgical Oncology 27 (2018) 737–742
L. Chen et al.
vision (Olympus Corporation, Japan), and gastroscopy (Olympus Corporation, Japan) was occasionally used to assist with identifying the site and extent of the tumor. Then, blood vessels and tissues around the tumor were separated using an ultrasound knife (Johnson & Johnson, Shanghai, China). Laparoscopic wedge resection of the stomach was performed employing an EndoGIA stapler (Johnson & Johnson, Shanghai, China) when ensuring negative resection margins. All surgeries were performed by surgeons with similar high levels of experience.
Table 1 Clinicopathological characteristics of all patients. Variables
Age, median years (range) Gender Male Female Tumor site Cardia Fundus Body Antrum Gastric body Lesser curve Greater curve Tumor size, median (range, cm) Mitotic figure (per 50 HPF) ≤5 > 5, ≤10 > 10 Risk level Low Intermediate High Cell subtype Epitheloid Spindle Mixed CD117 Positive Negative CD34 Positive Negative Ki-67 ≤ 2% > 2%
2.3.2. Endoscopic procedure In the gastroscopic field of vision (Olympus Corporation, Japan), endoscopic resection was performed to resect gastric GISTs. Patients were placed in a left-lateral position, and general anesthesia was adopted without a trachea cannula. The procedure of the endoscopic resection was as follows. First, electrocoagulation marking was performed using an acicular cutting knife (Johnson & Johnson, Shanghai, China) in the edge mucosa around the protruded lesion. Then, 5 mL indicarmine solution (0.1%–0.4% in concentration; Cuicr Corporation, Chengdu, China), 1 mL adrenaline hydrochloride injection (0.1% in concentration; Dandong Yichuang, Liaoning, China) and 100 mL physiological saline (Dandong Yichuang, Liaoning, China) were mixed as the marking solution, which was submucosally injected into the outside mucosa at multiple points marked around the protruding lesion. An acicular cutting knife was employed to cut the outside mucosa along the marked points. Then, submucosa was cut using the acicular cutting knife with anfractuous terminal. After exposing the lesion in muscularis propria, peeling off of the tumor was performed around the edge of the lesion. Near the end of the peeling process, a snare (Johnson & Johnson, Shanghai, China) was occasionally used to perform integral electroexcision of the tumor in the root of the lesion. After cutting the tumor, argon plasm coagulation was adopted to treat the active bleeding points. Then, the wound was closed using titanium clips (Olympus Corporation, Japan). All endoscopic procedures were handled by endoscopic doctors with similar high levels of experience.
Surgery type
P value
L (n = 66)
E (n = 35)
61 (34–80)
57 (31–78)
27 39
12 23
11 22 17 16
4 18 6 7
8 9 3.0 (2.0–4.8)
2 4 2.5 (2.0–4.6)
59 6 1
32 3 0
59 6 1
32 3 0
8 55 3
3 31 1
64 2
34 1
61 5
32 3
30 36
18 17
0.354 0.515 0.356
0.660# 0.0945 1.000#
1.000#
0.899#
1.000# 1.000# 0.567
Table 1. No significant difference existed in all clinicopathological characteristics of patients between laparoscopic and endoscopic groups. L, laparoscopic group; E, endoscopic group. # Fisher's exact test.
2.4. Postoperative follow-up Follow-up was regularly performed through telephone or outpatient service. Tumor recurrence, metastasis and death of patients were recorded. During the follow-up, all patients were required for abdominal CT scanning and gastroscopy according to risk level of recurrence. For patients with low risk level, abdominal CT scanning was performed every 6 months for the first two years and every 12 months thereafter, and gastroscopy was performed every 12 months. Abdominal CT scanning was conducted every 3 months (every 6 months for gastroscopy) for the first two years and every 6 months (every 12 months for gastroscopy) thereafter in patients with intermediate or high risk level.
The median tumor size was 3.0 cm (2.0–4.8 cm) and 2.5 cm (2.0–4.6 cm) in the laparoscopic and endoscopic groups, respectively. Thirty-nine of the patients were male (38.6%) while 62 were female (61.4%). The primary tumor site was cardia in 15 patients (14.8%), gastric fundus in 40 patients (39.6%), gastric body in 23 patients (22.8%), and gastric antrum in 23 patients (22.8%). The 23 patients with tumors in the gastric body included 10 patients (9.9%) in the lesser curve and 13 patients (12.9%) in the greater curve. The mitotic figures per 50 high-power fields were ≤5 in ninety-one patients (90.1%), > 5 and ≤ 10 in nine patients (8.9%), and > 10 in one patient (1.0%). The risk of recurrence was low in ninety-one patients (90.1%), intermediate in nine patients (8.9%), and high in one patient (1.0%). The cell subtype of the tumor was epithelioid in 11 patients (10.9%), spindle in 86 patients (85.1%), and mixed in 4 patients (4.0%). CD117 expression was positive in 98 patients (97.0%), and negative in 3 patients (3.0%). Expression of CD34 was positive in 93 patients (92.1%) and negative in 8 patients (7.9%). Expression of Ki-67 was ≤2% in 48 patients (47.5%) and > 2% in 53 patients (52.5%). As Table 1 shows, all clinicopathological characteristics indicated no significant difference between the laparoscopic group and endoscopic group.
2.5. Statistical analysis Chi-square test, Fisher's exact test and Student's t-test were performed by IBM SPSS 22 software in comparative analysis of clinicopathological data from patients in two groups. Log-rank test was employed in analysis of recurrence-free survival (RFS) of patients. R0 resection was defined as the microscopically resection with negative margins. RFS was calculated from the operation date to date of disease recurrence or last follow-up. P < 0.05 was considered statistically significant. 3. Results
3.2. Perioperative outcomes
3.1. Clinicopathological characteristics of all patients
The median operation time was 119 min (75–180 min) in the laparoscopic group, which was obviously longer than the 77 min (50–113 min) in the endoscopic group (P < 0.001). In the laparoscopic group, the median postoperative hospital stay was 9 days (3–18 days),
The median age of patients was 61 years (34–80 years) in the laparoscopic group and 57 years (31–78 years) in the endoscopic group. 739
Surgical Oncology 27 (2018) 737–742
L. Chen et al.
Table 2 Perioperative outcomes of all patients. Variables
Surgery type
Operation time, median min (range) Postoperative hospital stay, median days (range) Hospital charges, median yuan (range) Time to liquid diet, median days (range) Time to semi-liquid diet, median days (range) Complications
P value
L (n = 66)
E (n = 35)
119 (75–180) 9 (3–18) 39210 (20009–59026) 3 (1–6) 4 (3–7) 4 (6.1%)
77 (50–113) 6 (3–8) 24577 (18832–39671) 2 (1–3) 3 (2–4) 1 (2.9%)
< 0.001* < 0.001* < 0.001* < 0.01* < 0.01* 0.656#
Table 2. Patients in endoscopic group had shorter operation time, postoperative hospital stay, time to liquid diet, time to semi-liquid diet and lower hospital charges. L, laparoscopic group; E, endoscopic group. * Statistically significant; # Fisher's exact test.
which was remarkably longer than the 6 days (3–8 days) in the endoscopic group (P < 0.001). The median hospital charge of the laparoscopic group (39210 yuan; 20009 to 59026 yuan) was significantly higher than that of the endoscopic group (24577 yuan; 18832 to 39671 yuan) (P < 0.001). Furthermore, the median time to a liquid diet in the endoscopic group (2 days; 1–3 days) was markedly shorter than that in the laparoscopic group (3 days; 1–6 days) (P < 0.01). Similarly, the median time to a semi-liquid diet was 3 days (2–4 days) in the endoscopic group, which was notably shorter than the 4 days (3–7 days) in the laparoscopic group (P < 0.01). Perioperative complications occurred in four patients (including incisional infection, suture leakage, and peritonitis; 6.1%) and one patient (bleeding; 2.9%) in laparoscopic and endoscopic groups, respectively, with no significant difference. All complications were controlled safely. The bleeding was handled immediately in endoscopy. These results are summarized in Table 2.
Fig. 2. The RFS curves of laparoscopic group and endoscopic group. Recurrence of GIST occurred in 6 patients (9.1%) and 2 patients (5.7%) in laparoscopic and endoscopic groups, respectively. Log-rank test indicated no significant difference (P = 0.6006).
3.3. Long-term oncological outcomes Prior to the last follow-up, recurrence of GIST occurred in six patients (9.1%; including two patients who died because of the recurrence) and two patients (5.7%) in the laparoscopic and endoscopic groups, respectively, but with no significant difference. For the laparoscopic group, 3 patients had recurrent liver metastases and 3 patients experienced relapse in situ. 2 patients were recorded recurrence in situ in endoscopy group and no statistical significance was found about the recurrent location between the two groups. No patient in endoscopic group died. The median follow-up time was 60 months (15–60 months) in the laparoscopic group, which was significantly longer than the 45 months (17–60 months) in the endoscopic group (P < 0.001) (see Table 3). Furthermore, recurrence-free survival (RFS) was calculated and analyzed by the log-rank test. No significant difference in RFS existed between the laparoscopic and endoscopic groups (90.9% versus 94.3%, respectively; P = 0.6006). This is illustrated in Fig. 2.
GISTs are still a severe threat to human health owing to their risk of recurrence [1]. Despite recent revolutionary breakthroughs in their treatment due to the introduction of therapies targeting tyrosine kinase inhibitors, such as imatinib mesylate, radical surgery and postoperative follow-up remain the primary choice for patients with primary resectable GISTs [1,2]. However, with the development of minimally invasive surgery (MIS) technology, MIS (including laparoscopic and endoscopic approaches) has been increasingly applied in the treatment of gastric GISTs. According to presentations by the NCCN [7,8], and ESMO [20] clinical practice guidelines, annual follow-up with endoscopic ultrasound is the standard approach for patients with gastric GISTs < 2 cm, reserving surgical resection for patients whose tumor increases in size or becomes symptomatic. For patients with gastric GISTs > 5 cm, integral surgical excision by laparoscopic or open approaches is essential because of the higher risk of recurrence. However, for the treatment of patients with intraluminal gastric GISTs 2–5 cm in size, no therapeutic consensus has been presented in any treatment guidelines. Both laparoscopic and endoscopic approaches have been employed in treating relatively small gastric GISTs in several previous studies [15,16,21]. However, no studies have compared laparoscopic and endoscopic resections for patients with relatively small (2–5 cm) intraluminal gastric GISTs. In this study, we comparatively analyzed the short-term and long-term outcomes between laparoscopic and endoscopic approaches in patients with intraluminal gastric GISTs 2–5 cm in size. Our results showed that there was no significant difference in all clinicopathological characteristics of patients between the laparoscopic group and endoscopic group. Due to the limitation of retrospective studies, there may be some uncertain bias affecting the results that require randomized prospective experiments to validate. Laparoscopic resection was implicated in treating gastric GISTs in several studies.
4. Discussion As the most common mesenchymal neoplasms of the digestive tract, Table 3 Long-term oncological outcomes of all patients. Variables
Recurrence Recurrence-related death Follow-up, median months (range)
Surgery type
P value
L (n = 66)
E (n = 35)
6 (9.1%) 2 60 (15–60)
2 (5.7%) 0 45 (17–60)
0.711# 0.543# < 0.001*
Table 3. Differences of recurrence and recurrence-related death were not significant between two groups. The follow-up time of endoscopic group was significantly shorter. L, laparoscopic group; E, endoscopic group. * Statistically significant; # Fisher's exact test. 740
Surgical Oncology 27 (2018) 737–742
L. Chen et al.
Compared with open surgery, laparoscopic resection of gastric GISTs was confirmed to be a more effective and safe treatment method, with a shorter operation time, shorter hospital stay, lower rate of recurrence, and lesser complications [22,23]. Honda M et al. [21] indicated that the incidence of postoperative complications of laparoscopic surgery for patients with gastric GISTs < 5 cm was 2.6% (two patients with anastomotic leakage), and the reoccurrence rate was 1.3%. In contrast to these results, in our study, there were four patients (6.1%) in the laparoscopic resection group with postoperative complications, including reflux esophagitis, incisional infection, anastomotic leakage, and peritonitis; furthermore, six patients (9.1%) experienced disease relapse. Endoscopic resection has been increasingly widely used in treating gastric GISTs because of its merits of being less invasive and reserving gastric integrity and function; however, its common complications include perforation and bleeding [14,16]. A study of 31 patients with 2–5 cm gastric GISTs who underwent ESD indicated that six patients (19.35%) experienced iatrogenic perforation and three patients (9.68%) suffered from intraoperative bleeding [16]. Li L et al. [17] showed that iatrogenic perforation occurred in three of ten patients (30%) who underwent ESD, and one patient (10%) experienced postESD bleeding during treatment for gastric fundus subepithelial tumors originating from muscularis propria. However, intraoperative bleeding occurred in just one patient (2.9%) in the endoscopic resection group in the present study, and no patient suffered from perforation. Taken together, the endoscopic resection still has some limitations in the treatment of small intraluminal gastric GISTs. Firstly, the endoscopic operation is difficult for the patient with GIST located at the cardia and fundus or large tumor size. Secondly, the resection range may be insufficient because of the difficulty of fully judging the depth of invasion of GIST in the endoscopic procedure. Lastly, the endoscopic approach may induce the high incidence of complications after treatment, such as bleeding and perforation. Previous studies have compared laparoscopic and endoscopic resections in treating GIST patients. For gastric GIST patients with a tumor size ≤3.5 cm, the endoscopic approach resulted in shorter operation time, re-alimentation days, and postoperative hospital stay, and lower costs compared to the laparoscopic approach [13]. Moreover, Jeong IH et al. [15] revealed that relapse of GIST occurred in two patients (3.5%) who underwent laparoscopic resection, and the incidence of complication was lower in the endoscopic group compared to the laparoscopic group (7.4% versus 10.5%). Furthermore, a comparative analysis of laparoscopic resection, laparoscopic and endoscopic cooperative surgery, and endoscopic resection for patients with gastric GISTs ≤5 cm indicated that the operative time and intraoperative blood loss were significantly less in the endoscopic group compared to the other groups [24]. Similar to previous studies, we found that endoscopic resection resulted in a shorter operation time, postoperative hospital stay, time to a liquid diet, and time to a semi-liquid diet, as well as lower hospital charges; additionally, no significant difference was found in perioperative complication and tumor recurrence between the two groups. In conclusion, we confirmed the feasibility and safety of endoscopic resection in treating relatively small (2–5 cm) intraluminal gastric GISTs; this treatment is recommended because of the faster recovery and lower cost compared to laparoscopic resection. We suggest that endoscopic resection is adopted in elderly and weak patients or patients with a poor financial situation. However, laparoscopic resection should be considered as the treatment modality for other patients with relatively small (2–5 cm) intraluminal gastric GISTs. In addition, considering the underlying bias, the complexity of endoscopic procedure and lack of long term follow-up data support in the current study, the reproducibility and validity of the related results need to be verified by future prospective studies.
Acknowledgements Our work was sponsored by funds from Natural Science Foundation of Jiangsu Province (BK20141493, to H. Xu), Jiangsu Key Medical Discipline (General Surgery) (ZDXKA2016005, to Z.K. Xu), the Priority Academic Program Development of Jiangsu Higher Education Institutions (PAPD, JX10231801, to Z.K. Xu) and Jiangsu Key Lab of Cancer Biomarkers, Prevention and Treatment, Collaborative Innovation Center for Cancer Personalized Medicine, Nanjing Medical University (to Z.K. Xu). Abbreviations APC EUS GISTs HPF MIS PDGFRA RFS TKIs NCCN ESMO
argon plasma coagulation endoscopic ultrasonography gastrointestinal stromal tumors high-power fields minimal invasive surgery platelet-derived growth factor receptor alpha recurrence-free survival tyrosine kinase inhibitors National Comprehensive Cancer Network European Society for Medical Oncology
Authors’ contributions Conception and design: H. Xu, ZK. Xu, L. Chen; Acquisition of data: L. Chen, FY. Li, Q. Zhang; Analysis and interpretation of data: L. Chen; Drafting, review, and revision of manuscript: L. Chen, FY. Li, Q. Zhang, KJ. Xu, M. Chen; Study supervision: H. Xu, ZK. Xu; Final approval of the article: All authors. Conflicts of interest All authors have no conflicts of interest or financial ties to disclose. References [1] P. Rutkowski, A. Gronchi, P. Hohenberger, et al., Neoadjuvant imatinib in locally advanced gastrointestinal stromal tumors (GIST): the EORTC STBSG experience, Ann. Surg. Oncol. 20 (9) (2013) 2937–2943. [2] H. Joensuu, P. Hohenberger, C.L. Corless, Gastrointestinal stromal tumour, Lancet 382 (9896) (2013) 973–983. [3] J. Li, J.F. Gong, J. Li, et al., Efficacy of imatinib dose escalation in Chinese gastrointestinal stromal tumor patients, World J. Gastroenterol. 18 (7) (2012) 698–703. [4] B.K. Bednarski, D.M. Araujo, M. Yi, et al., Analysis of prognostic factors impacting oncologic outcomes after neoadjuvant tyrosine kinase inhibitor therapy for gastrointestinal stromal tumors, Ann. Surg. Oncol. 21 (8) (2014) 2499–2505. [5] B.K. Goh, P.K. Chow, W.M. Yap, et al., Which is the optimal risk stratification system for surgically treated localized primary GIST? Comparison of three contemporary prognostic criteria in 171 tumors and a proposal for a modified Armed Forces Institute of Pathology risk criteria, Ann. Surg. Oncol. 15 (8) (2008) 2153–2163. [6] M.I. Gluzman, V.A. Kashchenko, A.M. Karachun, et al., Technical success and shortterm results of surgical treatment of gastrointestinal stromal tumors: an experience of three centers, Transl. Gastroenterol. Hepatol. 2 (2017) 56. [7] M. von Mehren, R.L. Randall, R.S. Benjamin, et al., Soft tissue sarcoma, version 2.2016, NCCN clinical practice guidelines in Oncology, J. Natl. Compr. Canc. Netw. 14 (6) (2016) 758–786. [8] G.D. Demetri, R.S. Benjamin, C.D. Blanke, et al., NCCN Task Force report: management of patients with gastrointestinal stromal tumor (GIST)–update of the NCCN clinical practice guidelines, J. Natl. Compr. Canc. Netw. 5 (Suppl 2) (2007) S1–S29. [9] P. Tabrizian, S.Q. Nguyen, C.M. Divino, Laparoscopic management and longterm outcomes of gastrointestinal stromal tumors, J. Am. Coll. Surg. 208 (1) (2009) 80–86. [10] Y. Otani, T. Furukawa, M. Yoshida, et al., Operative indications for relatively small (2-5 cm) gastrointestinal stromal tumor of the stomach based on analysis of 60 operated cases, Surgery 139 (4) (2006) 484–492. [11] B.K. Goh, P.K. Chow, A.Y. Chok, et al., Impact of the introduction of laparoscopic wedge resection as a surgical option for suspected small/medium-sized gastrointestinal stromal tumors of the stomach on perioperative and oncologic outcomes, World J. Surg. 34 (8) (2010) 1847–1852. [12] H. Wang, X. Feng, S. Ye, et al., A comparison of the efficacy and safety of endoscopic
741
Surgical Oncology 27 (2018) 737–742
L. Chen et al.
[13] [14] [15] [16] [17] [18]
full-thickness resection and laparoscopic-assisted surgery for small gastrointestinal stromal tumors, Surg. Endosc. 30 (8) (2016) 3357–3361. W.J. Dai, G. Liu, M. Wang, et al., Endoscopic versus laparoscopic resection of gastric gastrointestinal stromal tumors: a multicenter study, Oncotarget 8 (7) (2017) 11259–11267. S. Sun, N. Ge, C. Wang, et al., Endoscopic band ligation of small gastric stromal tumors and follow-up by endoscopic ultrasonography, Surg. Endosc. 21 (4) (2007) 574–578. I.H. Jeong, J.H. Kim, S.R. Lee, et al., Minimally invasive treatment of gastric gastrointestinal stromal tumors: laparoscopic and endoscopic approach, Surg. Laparosc. Endosc. Percutaneous Tech. 22 (3) (2012) 244–250. Z. He, C. Sun, Z. Zheng, et al., Endoscopic submucosal dissection of large gastrointestinal stromal tumors in the esophagus and stomach, J. Gastroenterol. Hepatol. 28 (2) (2013) 262–267. L. Li, F. Wang, B. Wu, et al., Endoscopic submucosal dissection of gastric fundus subepithelial tumors originating from the muscularis propria, Exp. Ther. Med. 6 (2) (2013) 391–395. B. Mangiavillano, P. Viaggi, E. Masci, Endoscopic closure of acute iatrogenic perforations during diagnostic and therapeutic endoscopy in the gastrointestinal tract
using metallic clips: a literature review, J. Digest. Dis. 11 (1) (2010) 12–18. [19] M. Miettinen, J. Lasota, Gastrointestinal stromal tumors: Pathology and prognosis at different sites, Semin. Diagn. Pathol. 23 (2) (2006) 70–83. [20] Group ESESNW, Gastrointestinal stromal tumours: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up, Ann. Oncol. 25 (Suppl 3) (2014) iii21–26. [21] M. Honda, N. Hiki, S. Nunobe, et al., Long-term and surgical outcomes of laparoscopic surgery for gastric gastrointestinal stromal tumors, Surg. Endosc. 28 (8) (2014) 2317–2322. [22] K. De Vogelaere, A. Hoorens, P. Haentjens, et al., Laparoscopic versus open resection of gastrointestinal stromal tumors of the stomach, Surg. Endosc. 27 (5) (2013) 1546–1554. [23] J.-Q. Cai, K. Chen, Y.-P. Mou, et al., Laparoscopic versus open wedge resection for gastrointestinal stromal tumors of the stomach: a single-center 8-year retrospective cohort study of 156 patients with long-term follow-up, BMC Surg. 15 (1) (2015) 58. [24] X. Yin, Y. Yin, H. Chen, et al., Comparison analysis of three different types of minimally invasive procedures for gastrointestinal stromal tumors < /=5 cm, J. Laparoendosc. Adv. Surg. Tech. 28 (1) (2018) 58–64.
742
Contents lists available at ScienceDirect
Surgical Oncology journal homepage: www.elsevier.com/locate/suronc
Comparison of treatment outcomes between laparoscopic and endoscopic surgeries for relatively small gastric gastrointestinal stromal tumors
T
Liang Chena,b,1, Qiang Zhanga,1, Feng-Yuan Lia,1, Li Yangc, Dian-Cai Zhangc, Lin-Jun Wangc, Wei-Zhi Wangc, Zheng Lic, Jiang-Hao Xuc, Zhong-Yuan Hec, Kang-Jing Xua, Ming Chena, Hao Xua,∗, Ze-Kuan Xua,∗∗ a
Department of General Surgery, The First Affiliated Hospital of Nanjing Medical University, No.300, Guangzhou Road, Nanjing, Jiangsu province, 210029, China Department of General Surgery, Jiangsu Cancer Hospital & Jiangsu Institute of Cancer Research, The Affiliated Cancer Hospital of Nanjing Medical University,Nanjing, 210009, China c Jiangsu Key Lab of Cancer Biomarkers, Prevention and Treatment, Jiangsu Collaborative Innovation Center for Cancer Personalized Medicine, School of Public Health, Nanjing Medical University, Nanjing 211166, China b
ARTICLE INFO
ABSTRACT
Keywords: Gastric gastrointestinal stromal tumors Laparoscopy Endoscopy Prognosis
Background: Complete surgical resection remains the predominant treatment modality for primary gastrointestinal stromal tumors (GISTs). No therapeutic consensus exists for 2–5 cm gastric GISTs. We compared the efficacy, safety, and prognosis of laparoscopic and endoscopic surgeries in the treatment of relatively small (2–5 cm) intraluminal gastric GISTs. Methods: We collected 101 patients with relatively small intraluminal gastric GISTs who had integrated clinicopathological data and underwent laparoscopic or endoscopic resection (laparoscopic group n = 66; endoscopic group n = 35). Clinicopathological characteristics, perioperative data, and long-term oncological outcomes were retrospectively analyzed. Comparative analysis of clinicopathological data in the two groups was performed by using a chi-square test, Fisher's exact test, and Student's t-test. Recurrence-free survival (RFS) was analyzed by the log-rank test. Results: All clinicopathological characteristics had no significant difference between the two groups. Patients in the endoscopic group had shorter operation time (P < 0.001), postoperative hospital stay (P < 0.001), time to a liquid diet (P < 0.01), and time to a semi-liquid diet (P < 0.01), and lower hospital charges (P < 0.001), compared to those in the laparoscopic group. Four patients (6.1%) in the laparoscopic group and one patient (2.9%) in the endoscopic group had perioperative complications, but with no significant difference. Recurrence occurred in 6 patients (9.1%) and 2 patients (5.7%) in the laparoscopic and endoscopic groups, respectively. There was no significant difference in RFS between the two groups. Conclusion: Endoscopic resection is a feasible and safe treatment modality for patients with relatively small (2–5 cm) intraluminal gastric GISTs. Due to faster recovery and lower cost, endoscopic resection is more suitable for elderly and weak patients, or patients with a poor financial situation.
1. Introduction Gastrointestinal stromal tumors (GISTs) are the most common mesenchymal tumors of the gastrointestinal tract, and occur most commonly in the stomach (50–60%) [1,2]. Incidence of GISTs is about 10–22 per million [2]. It remains a threat to human health because of potential malignancy and reoccurrence.
Mutations of c-kit and platelet-derived growth factor receptor alpha (PDGFRA) are the leading causes of GISTs [3,4]. Complete surgical resection remains the main and preferred treatment method for primary GISTs [5]. Due to the low rate of lymphatic metastasis and infiltrating growth, GISTs can be treated by local surgery resection rather than extensive surgery with lymph node dissection [6]. As the NCCN Clinical Practice Guidelines [7] and NCCN Task Force Report [8] indicated,
Corresponding author. Corresponding author. Department of General Surgery, the First Affiliated Hospital of Nanjing Medical University, No.300, Guangzhou Road, Nanjing, 210029, Jiangsu province, China. E-mail addresses: [email protected] (H. Xu), [email protected] (Z.-K. Xu). 1 These authors contributed equally to this work. ∗
∗∗
https://doi.org/10.1016/j.suronc.2018.10.002 Received 15 April 2018; Received in revised form 19 August 2018; Accepted 1 October 2018 0960-7404/ © 2018 Elsevier Ltd. All rights reserved.
Surgical Oncology 27 (2018) 737–742
L. Chen et al.
different surgical procedures should be recommended according to tumor size, location and adjacent tissue structure. For patients with gastric GISTs < 2 cm, most of which are almost universally benign, surgical treatment is controversial and follow-up observation by endoscopy is much more recommended because of its general very low risk of relapse. However, all GISTs ≥ 2 cm should be resected because of the risk of reoccurrence. Especially for gastric GISTs > 5 cm, which generally have an intermediate or high risk of recurrence, laparoscopic or open surgeries are both applied depending on the tumor size and location. However, for 2–5 cm intraluminal gastric GISTs, no standardized treatment criteria have been recognized. Due to their respective merits and demerits, both laparoscopic and endoscopic surgeries are possible therapeutic modalities. In recent years, as a minimally invasive procedure, laparoscopic resection has been increasingly recognized as a standard surgical procedure for the treatment of GISTs [9]. Generally, laparoscopic wedge resection with negative margins is the ideal treatment choice for relatively small gastric GISTs. However, gastric GISTs > 5 cm and close to cardia or pylorus are benefit more from laparoscopy-assisted and open operations [10]. The feasibility and safety of laparoscopic resection in the treatment of gastric GISTs had been confirmed in several previous studies [9–11]. Compared to endoscopic resection, laparoscopic surgery is more invasive and has a longer operation time, longer postoperative hospitalization, more complications, and higher treatment costs for GIST patients [12,13]. Endoscopic resection, with its advantages of being less invasive and preserving the integrity and organ function of the stomach, has been reported to treat gastric GISTs [14,15]. Endoscopic band ligation with follow-up by endoscopic ultrasonography has been effectively and safely used in the treatment of small gastric stromal tumors [14]. With the development of endoscopic technology, endoscopic dissection was confirmed as an effective, safe, and feasible treatment modality for relatively small (2–5 cm) GISTs in the esophagus and stomach [16]. Different to laparoscopic resection, with endoscopic resection it is more difficult to achieve R0 resection because of limited operating spaces. Furthermore, iatrogenic perforation and bleeding are common complications of endoscopic resection for gastric GIST patients, which may result in asking for surgical assistance [16–18]. Few studies have reported which method, laparoscopic resection or endoscopic resection, is the optimal therapeutic modality for patients with relatively small (2–5 cm) intraluminal gastric GISTs. Due to controversial treatment choice, the aim of this study is to compare the efficacy, safety, and prognosis of laparoscopic and endoscopic resections for patients with relatively small (2–5 cm) intraluminal gastric GISTs.
Board of the First Affiliated Hospital of Nanjing Medical University. All enrolled patients were informed of this study and signed informed consents. 2.2. Data collection Clinicopathological characteristics (including age, gender, tumor location, tumor size, mitotic count, risk level of recurrence, cell subtype, CD117, CD34 and Ki-67), perioperative characteristics (including operation time, postoperative hospital stay, hospital charges, time to a liquid diet, time to a semi-liquid diet, and complications) and long-term oncological outcomes (including recurrence, recurrence-related death and follow-up time) of patients in two groups were collected and analyzed respectively. Time to a liquid/semi-liquid diet was calculated from the operation date. Time to first anal exhaust was similar to liquid diet time. First anal exhaust indicated the beginning of liquid diet. No intraoperative tumor rupture occurred, R0 resection was achieved and resection margins were negative in all patients. Risk levels of reoccurrence were evaluated according to Armed Forces Institute of Pathology criteria [19]. 2.3. Operative procedure 2.3.1. Laparoscopic procedure Patients were placed in a supine position with separated legs. After general anesthesia with trachea cannula, pneumoperitoneum pressure was established at 12 mmHg. As Fig. 1 shows, one trocar (Johnson & Johnson, Shanghai, China) 10 mm in diameter was inserted into the superior margin of the umbilicus as a camera port. Four additional trocars (two trocars 12 mm in diameter, and two trocars 5 mm in diameter) were inserted into the upper left, lower left, upper right, and lower right quadrants of the abdomen above the umbilicus to serve as operative ports. The tumor site was confirmed in a laparoscopic field of
2. Methods 2.1. Patients We enrolled 101 relatively small intraluminal gastric GIST patients who had entire clinicopathological data and underwent laparoscopic or endoscopic resection in the First Affiliated Hospital of Nanjing Medical University from April 2009 to April 2016. All patients were divided into two groups, laparoscopic resection group (n = 66) and endoscopic resection group (n = 35). General clinical information, perioperative data and long-term prognosis of all patients were retrospectively analyzed. Inclusion criteria: patients with pathologically confirmed GISTs which were 2–5 cm in diameter, and in good physical condition were eligible. Patients suffering from other tumors simultaneously, with history of abdominal surgery, preoperative hemoglobin < 90 g/L, dysfunction of important organs, and combined resection of other organs would be excluded. If the subjects had taken imatinib postoperatively or simultaneously participated in other clinical trials were not included. Besides, female patients who were pregnant or in lactation would be excluded as well. This study was approved by The Institutional Ethical
Fig. 1. The location of ports used in laparoscopic procedure. A (10 mm), the camera port in superior margin of umbilicus. B (12 mm), C (5 mm), D (5 mm) and E (12 mm), the operative ports in left upper, left lower, right upper and right lower quadrants of abdomen of upper level of umbilicus. 738
Surgical Oncology 27 (2018) 737–742
L. Chen et al.
vision (Olympus Corporation, Japan), and gastroscopy (Olympus Corporation, Japan) was occasionally used to assist with identifying the site and extent of the tumor. Then, blood vessels and tissues around the tumor were separated using an ultrasound knife (Johnson & Johnson, Shanghai, China). Laparoscopic wedge resection of the stomach was performed employing an EndoGIA stapler (Johnson & Johnson, Shanghai, China) when ensuring negative resection margins. All surgeries were performed by surgeons with similar high levels of experience.
Table 1 Clinicopathological characteristics of all patients. Variables
Age, median years (range) Gender Male Female Tumor site Cardia Fundus Body Antrum Gastric body Lesser curve Greater curve Tumor size, median (range, cm) Mitotic figure (per 50 HPF) ≤5 > 5, ≤10 > 10 Risk level Low Intermediate High Cell subtype Epitheloid Spindle Mixed CD117 Positive Negative CD34 Positive Negative Ki-67 ≤ 2% > 2%
2.3.2. Endoscopic procedure In the gastroscopic field of vision (Olympus Corporation, Japan), endoscopic resection was performed to resect gastric GISTs. Patients were placed in a left-lateral position, and general anesthesia was adopted without a trachea cannula. The procedure of the endoscopic resection was as follows. First, electrocoagulation marking was performed using an acicular cutting knife (Johnson & Johnson, Shanghai, China) in the edge mucosa around the protruded lesion. Then, 5 mL indicarmine solution (0.1%–0.4% in concentration; Cuicr Corporation, Chengdu, China), 1 mL adrenaline hydrochloride injection (0.1% in concentration; Dandong Yichuang, Liaoning, China) and 100 mL physiological saline (Dandong Yichuang, Liaoning, China) were mixed as the marking solution, which was submucosally injected into the outside mucosa at multiple points marked around the protruding lesion. An acicular cutting knife was employed to cut the outside mucosa along the marked points. Then, submucosa was cut using the acicular cutting knife with anfractuous terminal. After exposing the lesion in muscularis propria, peeling off of the tumor was performed around the edge of the lesion. Near the end of the peeling process, a snare (Johnson & Johnson, Shanghai, China) was occasionally used to perform integral electroexcision of the tumor in the root of the lesion. After cutting the tumor, argon plasm coagulation was adopted to treat the active bleeding points. Then, the wound was closed using titanium clips (Olympus Corporation, Japan). All endoscopic procedures were handled by endoscopic doctors with similar high levels of experience.
Surgery type
P value
L (n = 66)
E (n = 35)
61 (34–80)
57 (31–78)
27 39
12 23
11 22 17 16
4 18 6 7
8 9 3.0 (2.0–4.8)
2 4 2.5 (2.0–4.6)
59 6 1
32 3 0
59 6 1
32 3 0
8 55 3
3 31 1
64 2
34 1
61 5
32 3
30 36
18 17
0.354 0.515 0.356
0.660# 0.0945 1.000#
1.000#
0.899#
1.000# 1.000# 0.567
Table 1. No significant difference existed in all clinicopathological characteristics of patients between laparoscopic and endoscopic groups. L, laparoscopic group; E, endoscopic group. # Fisher's exact test.
2.4. Postoperative follow-up Follow-up was regularly performed through telephone or outpatient service. Tumor recurrence, metastasis and death of patients were recorded. During the follow-up, all patients were required for abdominal CT scanning and gastroscopy according to risk level of recurrence. For patients with low risk level, abdominal CT scanning was performed every 6 months for the first two years and every 12 months thereafter, and gastroscopy was performed every 12 months. Abdominal CT scanning was conducted every 3 months (every 6 months for gastroscopy) for the first two years and every 6 months (every 12 months for gastroscopy) thereafter in patients with intermediate or high risk level.
The median tumor size was 3.0 cm (2.0–4.8 cm) and 2.5 cm (2.0–4.6 cm) in the laparoscopic and endoscopic groups, respectively. Thirty-nine of the patients were male (38.6%) while 62 were female (61.4%). The primary tumor site was cardia in 15 patients (14.8%), gastric fundus in 40 patients (39.6%), gastric body in 23 patients (22.8%), and gastric antrum in 23 patients (22.8%). The 23 patients with tumors in the gastric body included 10 patients (9.9%) in the lesser curve and 13 patients (12.9%) in the greater curve. The mitotic figures per 50 high-power fields were ≤5 in ninety-one patients (90.1%), > 5 and ≤ 10 in nine patients (8.9%), and > 10 in one patient (1.0%). The risk of recurrence was low in ninety-one patients (90.1%), intermediate in nine patients (8.9%), and high in one patient (1.0%). The cell subtype of the tumor was epithelioid in 11 patients (10.9%), spindle in 86 patients (85.1%), and mixed in 4 patients (4.0%). CD117 expression was positive in 98 patients (97.0%), and negative in 3 patients (3.0%). Expression of CD34 was positive in 93 patients (92.1%) and negative in 8 patients (7.9%). Expression of Ki-67 was ≤2% in 48 patients (47.5%) and > 2% in 53 patients (52.5%). As Table 1 shows, all clinicopathological characteristics indicated no significant difference between the laparoscopic group and endoscopic group.
2.5. Statistical analysis Chi-square test, Fisher's exact test and Student's t-test were performed by IBM SPSS 22 software in comparative analysis of clinicopathological data from patients in two groups. Log-rank test was employed in analysis of recurrence-free survival (RFS) of patients. R0 resection was defined as the microscopically resection with negative margins. RFS was calculated from the operation date to date of disease recurrence or last follow-up. P < 0.05 was considered statistically significant. 3. Results
3.2. Perioperative outcomes
3.1. Clinicopathological characteristics of all patients
The median operation time was 119 min (75–180 min) in the laparoscopic group, which was obviously longer than the 77 min (50–113 min) in the endoscopic group (P < 0.001). In the laparoscopic group, the median postoperative hospital stay was 9 days (3–18 days),
The median age of patients was 61 years (34–80 years) in the laparoscopic group and 57 years (31–78 years) in the endoscopic group. 739
Surgical Oncology 27 (2018) 737–742
L. Chen et al.
Table 2 Perioperative outcomes of all patients. Variables
Surgery type
Operation time, median min (range) Postoperative hospital stay, median days (range) Hospital charges, median yuan (range) Time to liquid diet, median days (range) Time to semi-liquid diet, median days (range) Complications
P value
L (n = 66)
E (n = 35)
119 (75–180) 9 (3–18) 39210 (20009–59026) 3 (1–6) 4 (3–7) 4 (6.1%)
77 (50–113) 6 (3–8) 24577 (18832–39671) 2 (1–3) 3 (2–4) 1 (2.9%)
< 0.001* < 0.001* < 0.001* < 0.01* < 0.01* 0.656#
Table 2. Patients in endoscopic group had shorter operation time, postoperative hospital stay, time to liquid diet, time to semi-liquid diet and lower hospital charges. L, laparoscopic group; E, endoscopic group. * Statistically significant; # Fisher's exact test.
which was remarkably longer than the 6 days (3–8 days) in the endoscopic group (P < 0.001). The median hospital charge of the laparoscopic group (39210 yuan; 20009 to 59026 yuan) was significantly higher than that of the endoscopic group (24577 yuan; 18832 to 39671 yuan) (P < 0.001). Furthermore, the median time to a liquid diet in the endoscopic group (2 days; 1–3 days) was markedly shorter than that in the laparoscopic group (3 days; 1–6 days) (P < 0.01). Similarly, the median time to a semi-liquid diet was 3 days (2–4 days) in the endoscopic group, which was notably shorter than the 4 days (3–7 days) in the laparoscopic group (P < 0.01). Perioperative complications occurred in four patients (including incisional infection, suture leakage, and peritonitis; 6.1%) and one patient (bleeding; 2.9%) in laparoscopic and endoscopic groups, respectively, with no significant difference. All complications were controlled safely. The bleeding was handled immediately in endoscopy. These results are summarized in Table 2.
Fig. 2. The RFS curves of laparoscopic group and endoscopic group. Recurrence of GIST occurred in 6 patients (9.1%) and 2 patients (5.7%) in laparoscopic and endoscopic groups, respectively. Log-rank test indicated no significant difference (P = 0.6006).
3.3. Long-term oncological outcomes Prior to the last follow-up, recurrence of GIST occurred in six patients (9.1%; including two patients who died because of the recurrence) and two patients (5.7%) in the laparoscopic and endoscopic groups, respectively, but with no significant difference. For the laparoscopic group, 3 patients had recurrent liver metastases and 3 patients experienced relapse in situ. 2 patients were recorded recurrence in situ in endoscopy group and no statistical significance was found about the recurrent location between the two groups. No patient in endoscopic group died. The median follow-up time was 60 months (15–60 months) in the laparoscopic group, which was significantly longer than the 45 months (17–60 months) in the endoscopic group (P < 0.001) (see Table 3). Furthermore, recurrence-free survival (RFS) was calculated and analyzed by the log-rank test. No significant difference in RFS existed between the laparoscopic and endoscopic groups (90.9% versus 94.3%, respectively; P = 0.6006). This is illustrated in Fig. 2.
GISTs are still a severe threat to human health owing to their risk of recurrence [1]. Despite recent revolutionary breakthroughs in their treatment due to the introduction of therapies targeting tyrosine kinase inhibitors, such as imatinib mesylate, radical surgery and postoperative follow-up remain the primary choice for patients with primary resectable GISTs [1,2]. However, with the development of minimally invasive surgery (MIS) technology, MIS (including laparoscopic and endoscopic approaches) has been increasingly applied in the treatment of gastric GISTs. According to presentations by the NCCN [7,8], and ESMO [20] clinical practice guidelines, annual follow-up with endoscopic ultrasound is the standard approach for patients with gastric GISTs < 2 cm, reserving surgical resection for patients whose tumor increases in size or becomes symptomatic. For patients with gastric GISTs > 5 cm, integral surgical excision by laparoscopic or open approaches is essential because of the higher risk of recurrence. However, for the treatment of patients with intraluminal gastric GISTs 2–5 cm in size, no therapeutic consensus has been presented in any treatment guidelines. Both laparoscopic and endoscopic approaches have been employed in treating relatively small gastric GISTs in several previous studies [15,16,21]. However, no studies have compared laparoscopic and endoscopic resections for patients with relatively small (2–5 cm) intraluminal gastric GISTs. In this study, we comparatively analyzed the short-term and long-term outcomes between laparoscopic and endoscopic approaches in patients with intraluminal gastric GISTs 2–5 cm in size. Our results showed that there was no significant difference in all clinicopathological characteristics of patients between the laparoscopic group and endoscopic group. Due to the limitation of retrospective studies, there may be some uncertain bias affecting the results that require randomized prospective experiments to validate. Laparoscopic resection was implicated in treating gastric GISTs in several studies.
4. Discussion As the most common mesenchymal neoplasms of the digestive tract, Table 3 Long-term oncological outcomes of all patients. Variables
Recurrence Recurrence-related death Follow-up, median months (range)
Surgery type
P value
L (n = 66)
E (n = 35)
6 (9.1%) 2 60 (15–60)
2 (5.7%) 0 45 (17–60)
0.711# 0.543# < 0.001*
Table 3. Differences of recurrence and recurrence-related death were not significant between two groups. The follow-up time of endoscopic group was significantly shorter. L, laparoscopic group; E, endoscopic group. * Statistically significant; # Fisher's exact test. 740
Surgical Oncology 27 (2018) 737–742
L. Chen et al.
Compared with open surgery, laparoscopic resection of gastric GISTs was confirmed to be a more effective and safe treatment method, with a shorter operation time, shorter hospital stay, lower rate of recurrence, and lesser complications [22,23]. Honda M et al. [21] indicated that the incidence of postoperative complications of laparoscopic surgery for patients with gastric GISTs < 5 cm was 2.6% (two patients with anastomotic leakage), and the reoccurrence rate was 1.3%. In contrast to these results, in our study, there were four patients (6.1%) in the laparoscopic resection group with postoperative complications, including reflux esophagitis, incisional infection, anastomotic leakage, and peritonitis; furthermore, six patients (9.1%) experienced disease relapse. Endoscopic resection has been increasingly widely used in treating gastric GISTs because of its merits of being less invasive and reserving gastric integrity and function; however, its common complications include perforation and bleeding [14,16]. A study of 31 patients with 2–5 cm gastric GISTs who underwent ESD indicated that six patients (19.35%) experienced iatrogenic perforation and three patients (9.68%) suffered from intraoperative bleeding [16]. Li L et al. [17] showed that iatrogenic perforation occurred in three of ten patients (30%) who underwent ESD, and one patient (10%) experienced postESD bleeding during treatment for gastric fundus subepithelial tumors originating from muscularis propria. However, intraoperative bleeding occurred in just one patient (2.9%) in the endoscopic resection group in the present study, and no patient suffered from perforation. Taken together, the endoscopic resection still has some limitations in the treatment of small intraluminal gastric GISTs. Firstly, the endoscopic operation is difficult for the patient with GIST located at the cardia and fundus or large tumor size. Secondly, the resection range may be insufficient because of the difficulty of fully judging the depth of invasion of GIST in the endoscopic procedure. Lastly, the endoscopic approach may induce the high incidence of complications after treatment, such as bleeding and perforation. Previous studies have compared laparoscopic and endoscopic resections in treating GIST patients. For gastric GIST patients with a tumor size ≤3.5 cm, the endoscopic approach resulted in shorter operation time, re-alimentation days, and postoperative hospital stay, and lower costs compared to the laparoscopic approach [13]. Moreover, Jeong IH et al. [15] revealed that relapse of GIST occurred in two patients (3.5%) who underwent laparoscopic resection, and the incidence of complication was lower in the endoscopic group compared to the laparoscopic group (7.4% versus 10.5%). Furthermore, a comparative analysis of laparoscopic resection, laparoscopic and endoscopic cooperative surgery, and endoscopic resection for patients with gastric GISTs ≤5 cm indicated that the operative time and intraoperative blood loss were significantly less in the endoscopic group compared to the other groups [24]. Similar to previous studies, we found that endoscopic resection resulted in a shorter operation time, postoperative hospital stay, time to a liquid diet, and time to a semi-liquid diet, as well as lower hospital charges; additionally, no significant difference was found in perioperative complication and tumor recurrence between the two groups. In conclusion, we confirmed the feasibility and safety of endoscopic resection in treating relatively small (2–5 cm) intraluminal gastric GISTs; this treatment is recommended because of the faster recovery and lower cost compared to laparoscopic resection. We suggest that endoscopic resection is adopted in elderly and weak patients or patients with a poor financial situation. However, laparoscopic resection should be considered as the treatment modality for other patients with relatively small (2–5 cm) intraluminal gastric GISTs. In addition, considering the underlying bias, the complexity of endoscopic procedure and lack of long term follow-up data support in the current study, the reproducibility and validity of the related results need to be verified by future prospective studies.
Acknowledgements Our work was sponsored by funds from Natural Science Foundation of Jiangsu Province (BK20141493, to H. Xu), Jiangsu Key Medical Discipline (General Surgery) (ZDXKA2016005, to Z.K. Xu), the Priority Academic Program Development of Jiangsu Higher Education Institutions (PAPD, JX10231801, to Z.K. Xu) and Jiangsu Key Lab of Cancer Biomarkers, Prevention and Treatment, Collaborative Innovation Center for Cancer Personalized Medicine, Nanjing Medical University (to Z.K. Xu). Abbreviations APC EUS GISTs HPF MIS PDGFRA RFS TKIs NCCN ESMO
argon plasma coagulation endoscopic ultrasonography gastrointestinal stromal tumors high-power fields minimal invasive surgery platelet-derived growth factor receptor alpha recurrence-free survival tyrosine kinase inhibitors National Comprehensive Cancer Network European Society for Medical Oncology
Authors’ contributions Conception and design: H. Xu, ZK. Xu, L. Chen; Acquisition of data: L. Chen, FY. Li, Q. Zhang; Analysis and interpretation of data: L. Chen; Drafting, review, and revision of manuscript: L. Chen, FY. Li, Q. Zhang, KJ. Xu, M. Chen; Study supervision: H. Xu, ZK. Xu; Final approval of the article: All authors. Conflicts of interest All authors have no conflicts of interest or financial ties to disclose. References [1] P. Rutkowski, A. Gronchi, P. Hohenberger, et al., Neoadjuvant imatinib in locally advanced gastrointestinal stromal tumors (GIST): the EORTC STBSG experience, Ann. Surg. Oncol. 20 (9) (2013) 2937–2943. [2] H. Joensuu, P. Hohenberger, C.L. Corless, Gastrointestinal stromal tumour, Lancet 382 (9896) (2013) 973–983. [3] J. Li, J.F. Gong, J. Li, et al., Efficacy of imatinib dose escalation in Chinese gastrointestinal stromal tumor patients, World J. Gastroenterol. 18 (7) (2012) 698–703. [4] B.K. Bednarski, D.M. Araujo, M. Yi, et al., Analysis of prognostic factors impacting oncologic outcomes after neoadjuvant tyrosine kinase inhibitor therapy for gastrointestinal stromal tumors, Ann. Surg. Oncol. 21 (8) (2014) 2499–2505. [5] B.K. Goh, P.K. Chow, W.M. Yap, et al., Which is the optimal risk stratification system for surgically treated localized primary GIST? Comparison of three contemporary prognostic criteria in 171 tumors and a proposal for a modified Armed Forces Institute of Pathology risk criteria, Ann. Surg. Oncol. 15 (8) (2008) 2153–2163. [6] M.I. Gluzman, V.A. Kashchenko, A.M. Karachun, et al., Technical success and shortterm results of surgical treatment of gastrointestinal stromal tumors: an experience of three centers, Transl. Gastroenterol. Hepatol. 2 (2017) 56. [7] M. von Mehren, R.L. Randall, R.S. Benjamin, et al., Soft tissue sarcoma, version 2.2016, NCCN clinical practice guidelines in Oncology, J. Natl. Compr. Canc. Netw. 14 (6) (2016) 758–786. [8] G.D. Demetri, R.S. Benjamin, C.D. Blanke, et al., NCCN Task Force report: management of patients with gastrointestinal stromal tumor (GIST)–update of the NCCN clinical practice guidelines, J. Natl. Compr. Canc. Netw. 5 (Suppl 2) (2007) S1–S29. [9] P. Tabrizian, S.Q. Nguyen, C.M. Divino, Laparoscopic management and longterm outcomes of gastrointestinal stromal tumors, J. Am. Coll. Surg. 208 (1) (2009) 80–86. [10] Y. Otani, T. Furukawa, M. Yoshida, et al., Operative indications for relatively small (2-5 cm) gastrointestinal stromal tumor of the stomach based on analysis of 60 operated cases, Surgery 139 (4) (2006) 484–492. [11] B.K. Goh, P.K. Chow, A.Y. Chok, et al., Impact of the introduction of laparoscopic wedge resection as a surgical option for suspected small/medium-sized gastrointestinal stromal tumors of the stomach on perioperative and oncologic outcomes, World J. Surg. 34 (8) (2010) 1847–1852. [12] H. Wang, X. Feng, S. Ye, et al., A comparison of the efficacy and safety of endoscopic
741
Surgical Oncology 27 (2018) 737–742
L. Chen et al.
[13] [14] [15] [16] [17] [18]
full-thickness resection and laparoscopic-assisted surgery for small gastrointestinal stromal tumors, Surg. Endosc. 30 (8) (2016) 3357–3361. W.J. Dai, G. Liu, M. Wang, et al., Endoscopic versus laparoscopic resection of gastric gastrointestinal stromal tumors: a multicenter study, Oncotarget 8 (7) (2017) 11259–11267. S. Sun, N. Ge, C. Wang, et al., Endoscopic band ligation of small gastric stromal tumors and follow-up by endoscopic ultrasonography, Surg. Endosc. 21 (4) (2007) 574–578. I.H. Jeong, J.H. Kim, S.R. Lee, et al., Minimally invasive treatment of gastric gastrointestinal stromal tumors: laparoscopic and endoscopic approach, Surg. Laparosc. Endosc. Percutaneous Tech. 22 (3) (2012) 244–250. Z. He, C. Sun, Z. Zheng, et al., Endoscopic submucosal dissection of large gastrointestinal stromal tumors in the esophagus and stomach, J. Gastroenterol. Hepatol. 28 (2) (2013) 262–267. L. Li, F. Wang, B. Wu, et al., Endoscopic submucosal dissection of gastric fundus subepithelial tumors originating from the muscularis propria, Exp. Ther. Med. 6 (2) (2013) 391–395. B. Mangiavillano, P. Viaggi, E. Masci, Endoscopic closure of acute iatrogenic perforations during diagnostic and therapeutic endoscopy in the gastrointestinal tract
using metallic clips: a literature review, J. Digest. Dis. 11 (1) (2010) 12–18. [19] M. Miettinen, J. Lasota, Gastrointestinal stromal tumors: Pathology and prognosis at different sites, Semin. Diagn. Pathol. 23 (2) (2006) 70–83. [20] Group ESESNW, Gastrointestinal stromal tumours: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up, Ann. Oncol. 25 (Suppl 3) (2014) iii21–26. [21] M. Honda, N. Hiki, S. Nunobe, et al., Long-term and surgical outcomes of laparoscopic surgery for gastric gastrointestinal stromal tumors, Surg. Endosc. 28 (8) (2014) 2317–2322. [22] K. De Vogelaere, A. Hoorens, P. Haentjens, et al., Laparoscopic versus open resection of gastrointestinal stromal tumors of the stomach, Surg. Endosc. 27 (5) (2013) 1546–1554. [23] J.-Q. Cai, K. Chen, Y.-P. Mou, et al., Laparoscopic versus open wedge resection for gastrointestinal stromal tumors of the stomach: a single-center 8-year retrospective cohort study of 156 patients with long-term follow-up, BMC Surg. 15 (1) (2015) 58. [24] X. Yin, Y. Yin, H. Chen, et al., Comparison analysis of three different types of minimally invasive procedures for gastrointestinal stromal tumors < /=5 cm, J. Laparoendosc. Adv. Surg. Tech. 28 (1) (2018) 58–64.
742