Surg Endosc (2011) 25:271–277 DOI 10.1007/s00464-010-1173-0
Laparoscopic resection of small gastric submucosal tumors Ki-Jin Ryu • Sae-Rom Jung • Jung-Soon Choi • You-Jin Jang • Jong-Han Kim • Sung-Soo Park • Beom Jin Park • Seong-Heum Park Seung-Joo Kim • Young-Jae Mok • Chong-Suk Kim
•
Received: 11 August 2009 / Accepted: 19 March 2010 / Published online: 18 June 2010 Ó Springer Science+Business Media, LLC 2010
Abstract Background No consensus exists regarding the necessity of operative resection for patients with small, asymptomatic gastric submucosal tumors (SMTs). The purpose of this study is to evaluate clinical outcomes of resection by minimally invasive surgery. Methods The medical records of 20 consecutive patients who had undergone laparoscopic or robotic wedge resection for small (\5 cm) gastric SMTs between March 2008 and February 2009 were reviewed. Operative indications included all SMTs unquestionably visible by endoscopy, irrespective of symptoms. The operative procedures, clinicopathologic features, and operative results were assessed. Results Out of a total of 20 patients, 17 were asymptomatic, and 3 presented with vague abdominal discomfort. One patient had two tumors, therefore 21 total lesions were resected and evaluated (19 by laparoscopy and 2 by robotic procedures). There were 12 exogastric and 9 transgastric K.-J. Ryu S.-R. Jung Y.-J. Jang J.-H. Kim S.-S. Park S.-H. Park S.-J. Kim Y.-J. Mok C.-S. Kim Department of Surgery, Korea University College of Medicine, Seoul, Korea J.-S. Choi Department of Biostatistics, Korea University College of Medicine, Seoul, Korea B. J. Park Department of Radiology, Korea University College of Medicine, Seoul, Korea S.-S. Park (&) Division of Upper Gastrointestinal Surgery, Department of Surgery, Korea University Anam Hospital, Korea University College of Medicine, 126-1, Anam-dong 5-ga, Seongbuk-gu, Seoul 136-705, Korea e-mail:
[email protected]
wedge resections. Mean operative time was 84 ± 28 min, and mean length of hospitalization was 4.7 ± 1.6 days. There were no major peri- or postoperative complications or mortalities. Mean tumor size was 2.4 ± 1.2 cm (range 0.6–4.8 cm). All lesions had microscopically negative resection margins. There were 16 gastrointestinal tumors (GISTs) and 5 other benign lesions. Fifteen of the GISTs had mitotic count (MC) \5 per 50 high-power fields (HPFs), and one lesion measuring 2.5 cm in size had MC of 38 per 50 HPFs. Conclusions Small size cannot guarantee a specific malignant risk for gastric SMTs. Laparoscopic/robotic wedge resection is safe and effective in treating small, asymptomatic lesions. Therefore, an active surgical approach should be considered for management of patients with small gastric SMTs.
Keywords Gastrointestinal stromal tumor Laparoscopy Small submucosal tumor
The management of gastric submucosal tumors (SMTs) has attracted considerable attention, with the widespread application of laparoscopic surgery. In East Asia, where there is a high prevalence of gastric cancer, routine gastrointestinal screening is well established, and the detection of small, asymptomatic gastric SMTs has increased [1]. Due to the difficulties and limitations of a definitive preoperative diagnosis of gastric SMTs, surgical resection plays a decisive role in both diagnosis and treatment [2]. Considering that approximately 80% of SMTs are benign [3], coupled with the possible morbidities associated with surgical intervention, decisions regarding operative indications are prudent.
123
272
Surg Endosc (2011) 25:271–277
The operative indications for SMTs are determined by tumor size, which is used as a predictive marker for the malignant potential of gastrointestinal stromal tumors (GISTs) [4]. Although some investigators have suggested that tumors [2 cm in size are candidates for operative resection [5], the cutoff level for tumor size remains controversial, as every GIST should be considered potentially malignant, regardless of size [6, 7]. Moreover, there is currently limited data pertaining to small SMTs. The current study evaluated small gastric SMTs, with the purpose of reporting our experiences with laparoscopic/ robotic resection, with special emphasis on operative procedures, pathologic diagnoses, and clinical outcomes.
Materials and methods Between March 2008 and February 2009, 20 consecutive patients undergoing laparoscopic/robotic surgical resection of small (\5 cm) gastric SMTs were identified in a prospectively collected database at Korea University Medical Center. Preoperative diagnosis and evaluation of tumor size, location, and presence of metastatic lesions were established in all patients using a combination of gastrofiberoscopy, upper gastrointestinal radiography, and computed tomography (CT). Endoscopic ultrasonography (EUS) and core-needle biopsy were performed only when clinically indicated. Lesions in which the postoperative immunohistochemical staining was positive for CD117 (KIT) were diagnosed as GISTs, and mutation analysis for the c-KIT mutation was also performed. Classification of resected GISTs with aggressive risk was defined according to the size and mitotic rate of the tumors, as proposed by Fletcher et al. [8] and as shown in Table 1. The operative indications included a SMT [10 mm in size and definitely visible by endoscopy, irrespective of symptoms. Tumors \20 mm in size measured on EUS or CT were preoperatively localized by endoscopic clipping at
Table 1 Risk of aggressiveness of GISTs Grouping of tumors (NIH GIST workshop)
Tumor size (cm)
Mitotic count (counts/50 HPFs)
Very low risk
\2
\5
Low risk
2–5
\5
Intermediate risk
\5
6–10
5–10
\5
[5
[5
[10
Any count
Any size
[10
High risk
HPF high-power field, NIH National Institutes of Health
123
the mucosa covering the SMTs (Fig. 1). CT gastrography, including virtual gastroscopy and surface-shaded volumerendering techniques, intuitively provided global information regarding location and shape of small SMTs (Fig. 2). Exogastric wedge resection was preferentially performed for those tumors with an exophytic growth pattern, and transgastric wedge resection was preferentially performed for tumors with an endophytic growth pattern. Lesions located predominantly on the outer wall of the stomach were defined as having an exophytic growth pattern, and lesions located predominantly on the inner wall of the stomach were defined as having an endophytic growth pattern. A second-generation cephalosporin was given to patients as a prophylactic antibiotic once before and once immediately after surgery, as per the policy of our institution, to reduce antibiotic abuse and microbe resistance. The SPSS 12.0 software program for Windows (SPSS Inc., Chicago, IL, USA) was used for all statistical analyses. Quantitative results are expressed as mean ± standard deviation. The Student t-test was used to compare the two groups. The Pearson test and simple linear regression test were used to test the significance of correlation between tumor size and mitotic count. Values of p \ 0.05 were considered statistically significant.
Results A total of 15 women and 5 men were enrolled in the present study, with mean age of 54.4 years (range 30– 75 years). Seventeen (85%) patients were asymptomatic, and among the three (15%) symptomatic patients, vague abdominal discomfort was evident in two patients and dyspepsia in one patient. The tumors were located in the upper third of the stomach in 5 patients (three near the esophagogastric junction), the middle third in 14 patients (70%), and the lower third in 1 patient. As one patient had two tumors, a total of 21 lesions were resected and evaluated, 19 by laparoscopy and 2 by robotic procedures. Patient and tumor characteristics and perioperative outcomes are shown in Table 2. The operative approaches used were exogastric (n = 11) and transgastric wedge resections (n = 10). The patient with two SMTs had an exophytic tumor resected via exogastric wedge resection as well as an endophytic tumor resected via transgastric wedge resection. There were no conversions to open surgery. Mean operative time was 84 ± 28 min (range 40– 140 min). There were no episodes of tumor rupture or spillage and no other major intraoperative complications. Estimated intraoperative bleeding was negligible, and no blood transfusions were required for any case in the perioperative period. None of the patients had any evidence of staple-line leakage. Average length of hospitalization
Surg Endosc (2011) 25:271–277
Fig. 1 Preoperative endoscopic clipping of a 1.2-cm SMT. A Endoscopy showed the tumor to be located in the rugae of the greater curvature of the stomach. B Clipping was successfully performed on the normal mucosa above the tumor
was 4.7 ± 1.6 days (range 3–8 days). All patients were permitted to drink water on postoperative day 2. Five patients undergoing transgastric wedge resection presented with postoperative peritoneal irritation signs on physical examination, all of which subsided within 24 h following empirical antibiotic therapy. There were no other major postoperative complications or mortalities. All lesions had microscopically negative resection margins. The average size of the 21 tumors was 2.4 ± 1.2 cm (range 0.6–4.8 cm). Nine of 21 tumors (43%) were classified as small (T \ 2 cm), and the average size of these tumors was 1.4 ± 0.4 cm. Twelve tumors (57%) were defined as
273
Fig. 2 Two gastric GISTs in a 52-year-old female. A The surfaceshaded image of CT gastrography shows small, closeted polypoid lesions (arrows) on the anterior wall of the gastric high body. B Virtual gastroscopy shows two polypoid lesions (arrows) which measure approximately 1.7 cm and 0.6 cm, respectively, and an upper larger lesion with a lobulated contour in the antero-inferior aspect of the esophagogastric junction (arrowhead) shows a slightly obtuse angle with the adjacent wall, indicating a submucosal tumor
relatively small (2 cm \ T \ 5 cm), and the average size of these tumors was 3.1 ± 1.0 cm. There were no significant differences between the perioperative outcomes of patients with ‘‘small’’ tumors and those with ‘‘relatively small’’ tumors. Pathological and immunohistochemical studies confirmed GIST in 16 lesions. Analysis of c-KIT mutations was performed in 13 lesions, 8 (61%) of which were positive. Although 15 GISTs had mitotic count (MC) \5
123
274
Surg Endosc (2011) 25:271–277
Table 2 Patient and tumor characteristics, and perioperative outcomes Type of Operative Postoperative Complications Mitotic c-KIT countb mutation surgery time (min) hospital stay (days)
Case
Age Gender Tumor Tumor Growth Histologic (years) size (cm) location pattern diagnosis
1
49
F
4.8
B
Exo
GIST
3
P
E-WR
95
4
–
2
61
M
1.2
B
Exo
GIST
0
N
E-WR
80
4
–
3 4
59 63
F F
4.0 4.5
B B
Exo Exo
EP GIST
– 1
– N
E-WR E-WR
85 100
6 4
– –
5
65
F
2.2
Uc
Exo
Leiomyoma
–
–
E-WR
105
4
–
6
44
M
1.2
L
Endo
EP
–
–
E-WR
115
5
–
7
67
F
2.1
U
Endo
GIST
1
P
T-WR
140
6
Abd. pain
8
40
F
2.9
B
Exo
Benign cyst
–
–
E-WR
54
5
–
9
66
M
1.7
U
Endo
GIST
2
P
RT-WR
50
8
Abd. pain
10
62
F
1.6
B
Endo
GIST
0
–
T-WR
63
4
–
11
49
F
2.5
B
Endo
GIST
38
P
T-WR
70
7
Abd. pain
12
35
F
2.4
Uc
Endo
GIST
0
N
RT-WR
90
3
–
13
39
F
4.2
B
Endo
GIST
3
P
E-WR
54
3
–
14
30
F
2.4
B
Endo
EP
–
–
T-WR
92
3
–
15
75
F
1.9
B
Exo
GIST
0
P
E-WR
75
3
–
16
57
F
3.1
B
Exo
GIST
3
P
E-WR
45
3
–
17
60
F
1.2
B
Endo
GIST
0
–
T-WR
88
7
Abd. pain
18-1a 52 18-2a
F
0.6 1.7
B B
Endo Exo
GIST GIST
0 0
N P
T-WR T-WR
110
5
–
19
61
M
2.5
Uc
Endo
GIST
1
N
T-WR
133
7
Abd. pain
20
55
M
1.5
B
Exo
GIST
0
–
E-WR
40
3
–
M male, F female, U upper third of the stomach, B middle third of the stomach, L lower third of the stomach, Endo endophytic, Exo exophytic, GIST gastrointestinal stromal tumor, EP ectopic pancreas, P positive, N negative, E-WR exogastric wedge resection, T-WR transgastric wedge resection, RT-WR robotic transgastric wedge resection, Abd. pain abdominal pain a
Two gastric submucosal tumors in one patient
b
Mitotic count per 50 high-power fields
c
At the gastroesophageal junction (\3 cm)
per 50 HPFs, one lesion (2.5 cm size) had MC of 38 per 50 HPFs. The Pearson correlation coefficient between tumor size and MC in all GISTs was not statistically significant (p = 0.639). However, the Pearson correlation coefficient between tumor size and MC in the 15 GISTs with MC \5 per 50 HPFs was statistically significant (p = 0.002). Figure 3 shows a scatter graph and the simple linear regression equation between tumor size and MC in these 15 GISTs (MC = -0.723 ? 0.710 9 tumor size; p = 0.002). The risk categories and the numbers of patients in each category were as follows: very low, eight (50%); low, seven (43%); intermediate, zero; and high, one (6%). Other histopathologic diagnoses were made in five patients with SMTs and included ectopic pancreas in three patients, leiomyoma in one patient, and a duplication cyst in one patient. Comparison of clinical outcomes between the exogastric and transgastric wedge resections is shown in Table 3. Both operative time and postoperative hospital stay were longer in the transgastric group than in the exogastric
123
group, although the difference in operative time between the two groups was not statistically significant.
Discussion Gastric SMTs are neoplastic or nonneoplastic lesions of various etiologies that develop in the stomach wall between the mucosa and the muscularis propria [3]. Because the tumor is covered with normal mucosa, preoperative histologic diagnosis remains difficult. Even with the assistance of EUS, predictions regarding whether the tumor possesses malignant potential are also difficult [9]. Preoperative core-needle biopsy might be helpful, but is not universally recommended because intra-abdominal tumor spillage is possible [7]. Thus, in order to make a pathologic diagnosis and ensure removal of the lesion, local resection of the stomach is considered the first-line treatment for gastric SMTs. SMTs without clear evidence of benign features should be resected as presumed GISTs, which account for
Surg Endosc (2011) 25:271–277
275
Fig. 3 Relationship between tumor size and mitotic count. R2 = 0.547; p-value = 0.002. One tumor, indicated by a triangle, was not included in the calculation of the simple linear regression equation and fitted line
[70% of all gastric SMTs [1, 2]. With the increasing detection and incidence of small gastric SMTs, the issue of whether surgery is beneficial for these tumors, especially those tumors \2 cm in size, has become increasingly relevant. Twenty patients with small gastric SMTs that were detected incidentally underwent laparoscopic/robotic surgery over the course of 1 year at a single institution. Since there are generally few specific symptoms associated with small SMTs, their detection is primarily incidental and dependent upon a routine gastrointestinal screening program. It has been previously reported that approximately two-thirds of relatively small (\5 cm) SMTs are asymptomatic [1, 10]. Seventeen of 20 patients (85%) in the current study were asymptomatic, and the presenting symptoms of the remaining three patients (abdominal discomfort and dyspepsia) had nonspecific symptoms possibly attributable to other causes. Treatment of choice for gastric SMTs is complete resection with gross negative margins with maintenance of gastrointestinal tract continuity through avoidance of routine lymphadenectomy. As such, laparoscopic wedge
resection has been regarded as the most appropriate procedure in recent years [7]. Laparoscopic wedge resections are mainly divided into exogastric and transgastric resections [11]. In this study, there were 12 exogastric and 9 transgastric wedge resections, determined by the growth pattern, size, and location of the lesions. Transgastric wedge resection allows for direct visualization of the lesion and better control of the surgical margin, thus causing less deformity. However, transgastric wedge resection is also known to have some limitations in that gastrostomy is associated with risk of bleeding and intraperitoneal contamination by gastric juices [12]. However, because we used a Harmonic Scalpel (Johnson and Johnson), bleeding at the gastrostomy site in the current study was negligible, and by placing gauze adjacent to the gastrostomy site during surgery, leakage of gastric juices was prevented. The additional time expenditure was compensated for by using the endo-GIA for gastrostomy site closure. Moreover, there were no intraoperative complications such as tumor rupture or spillage. The perioperative outcomes of the transgastric group were not significantly different from those of the exogastric group. Two patients underwent transgastric wedge resection using the da Vinci robotic surgical system (Intuitive Surgical, Inc., Sunnyvale, CA, USA). Due to the unique characteristics of the robot (wide visual field and steady traction), robotic surgery has the advantage of being an easy and safe operative procedure for SMTs located in difficult positions, especially near the gastroesophageal junction. Nevertheless, use of the da Vinci robotic surgical system carries high costs due to the expense of the instrument. Localization of small SMTs is difficult and is thought to inevitably require intraoperative gastroscopy or laparoscopic ultrasound, both of which are complicated and time consuming. However, as the preoperative endoscopic clipping and CT gastrography facilitated localization in this study, quick and easy resection was performed accurately without use of intraoperative devices. Length of postoperative hospital stay is generally accepted as a reliable parameter of invasiveness of surgery [13]. Mean postoperative hospital stay was approximately 5 days in this study, which is acceptable in comparison with other reports [14]. Peritoneal irritation signs in five
Table 3 Comparison of clinical outcomes between exo- and transgastric wedge resection Clinical outcomes
Exogastric wedge resection (n = 11)a
Transgastric wedge resection (n = 8)a
p-Value
Operation time (min), mean ± SD
77.1 ± 25.7
90.8 ± 31.8
0.890b
4.0 ± 1.0
5.62 ± 2.0
0.008b
Postoperative hospital stay (days), mean ± SD SD standard deviation a
One patient with two SMTs was not included in the data for accurate analysis
b
p-Values were calculated with the Student’s t-test
123
276
patients who had undergone transgastric wedge resection led to an increase in the mean hospital stay. The basis for those complications was thought to be small efflux of gastric juice and that prophylactic antibiotics were administered only once before and once immediately after surgery. Soon after administration of empirical antibiotics, the peritoneal signs abated. There were no other postoperative complications or mortalities. The operative results of tumors \2 cm in size were not significantly different from those of larger tumors. In view of the results achieved so far, we assert that transgastric and exogastric wedge resections for both small and relatively small SMTs are minimally invasive and produce acceptable clinical outcomes. MC has been the only independent prognostic factor identified in some studies and is regarded as a more powerful prognostic factor for GISTs than tumor size [10, 15]. Tumor size was nearly correlated with MC in this study; however, one 2.5-cm tumor was shown to have very high MC (38/50 HPFs), corresponding to the high risk category. Thus, these two factors showed a significant correlation only when the tumor with MC of 38/50 HPFs was excluded from the calculation. Similar to this exceptional case of a small tumor carrying a high risk, two cases of only 1.2-cm, high-risk GISTs have been reported in previous investigations [16]. Moreover, it has been reported that even those tumors with low MC and small size may exhibit unexpected malignant potential [6, 17, 18]. It is noteworthy that some tumors go against the generally accepted rules of clinicopathologic behavior. Consequently, all SMTs, even those small in size, should be considered to have some low malignancy potential. As such, there is credible evidence of a need for early surgical treatment, presupposing a safe operation. Nevertheless, it is still uncertain whether surgical resection would be desirable for small SMTs, less than 2 cm in size, because all eight GISTs of that size belonged to very low risk group in this study. More studies with a larger sample and longer follow-up may be needed.
Conclusions Current size criteria for classifying lesions cannot determine a specific malignant risk for GISTs. Only surgical resection of a lesion provides a definitive pathologic diagnosis, assessment, and removal of malignant potential. In addition, laparoscopic/robotic exogastric and transgastric wedge resections are safe and effective for treating small, asymptomatic gastric SMTs. Given these findings, as well as the advantages afforded by minimally invasive surgery, an active surgical approach should be an option for management of patients with small gastric SMTs.
123
Surg Endosc (2011) 25:271–277 Disclosures Authors Ki-Jin Ryu, Sae-Rom Jung, Jung-Soon Choi, You-Jin Jang, Jong-Han Kim, Sung-Soo Park, Beom Jin Park, SeongHeum Park, Seung-Joo Kim, Young-Jae Mok, and Chong-Suk Kim have no conflicts of interest or financial ties to disclose.
References 1. Iwahashi M, Takifuji K, Ojima T, Nakamura M, Nakamori M, Nakatani Y, Ueda K, Ishida K, Naka T, Ono K, Yamaue H (2006) Surgical management of small gastrointestinal stromal tumors of the stomach. World J Surg 30:28–35 2. Berindoague R, Targarona EM, Feliu X, Artigas V, Balague C, Aldeano A, Lahoud A, Navines J, Fernandez-Sallent E, Trias M (2006) Laparoscopic resection of clinically suspected gastric stromal tumors. Surg Innov 13:231–237 3. Wiech T, Walch A, Werner M (2005) Histopathological classification of nonneoplastic and neoplastic gastrointestinal submucosal lesions. Endoscopy 37:630–634 4. Otani Y, Ohgami M, Igarashi N, Kimata M, Kubota T, Kumai K, Kitajima M, Mukai M (2000) Laparoscopic wedge resection of gastric submucosal tumors. Surg Laparosc Endosc Percutan Tech 10:19–23 5. Otani Y, Furukawa T, Yoshida M, Saikawa Y, Wada N, Ueda M, Kubota T, Mukai M, Kameyama K, Sugino Y, Kumai K, Kitajima M (2006) Operative indications for relatively small (2–5 cm) gastrointestinal stromal tumor of the stomach based on analysis of 60 operated cases. Surgery 139:484–492 6. DeMatteo RP, Lewis JJ, Leung D, Mudan SS, Woodruff JM, Brennan MF (2000) Two hundred gastrointestinal stromal tumors: recurrence patterns and prognostic factors for survival. Ann Surg 231:51–58 7. Blay JY, Bonvalot S, Casali P, Choi H, Debiec-Richter M, Dei Tos AP, Emile JF, Gronchi A, Hogendoorn PCW, Joensuu H, Le Cesne A, Mac Clure J, Maurel J, Nupponen N, Ray-Coquard I, Reichardt P, Sciot R, Stroobants S, van Glabbeke M, van Oosterom A, GIST Demetri GD Consensus Meeting Panelists (2005) Consensus meeting for the management of gastrointestinal stromal tumors–– report of the GIST consensus conference of 20–21 March 2004, under the auspices of ESMO. Ann Oncol 16:566–578 8. Fletcher CD, Berman JJ, Corless C, Gorstein F, Lasota J, Longley BJ, Miettinen M, O’Leary TJ, Remotti H, Rubin BP, Shmookler B, Sobin LH, Weiss SW (2002) Diagnosis of gastrointestinal stromal tumors: a consensus approach. Hum Pathol 33:459–465 9. Kwon JG, Kim EY, Kim YS, Chun JW, Chung JT, You SS, Ha HK, Lee CH, Kim HG, Cho CH (2005) Accuracy of endoscopic ultrasonographic impression compared with pathologic diagnosis in gastrointestinal submucosal tumors. Korean J Gastroenterol 45:88–96 [Korean] 10. Park SS, Ryu JS, Oh SY, Kim WB, Lee JH, Chae YS, Kim SJ, Kim CS, Mok YJ (2007) Surgical outcomes and immunohistochemical features for gastrointestinal stromal tumors (GISTS) of the stomach: with special reference to prognostic factors. Hepatogastroenterology 54:1454–1457 11. Tagaya N, Mikami H, Kogure H, Kubota K, Hosoya Y, Nagai H (2002) Laparoscopic intragastric stapled resection of gastric submucosal tumors located near the esophagogastric junction. Surg Endosc 16:177–179 12. Motson RW, Fisher PW, Dawson JW (1995) Laparoscopic resection of a benign intragastric stromal tumor. Br J Surg 82:1670 13. Kitano S, Shiraishi N, Fujii K, Yasuda K, Inomata M, Adachi Y (2002) A randomized controlled trial comparing open vs laparoscopy-assisted distal gastrectomy for the treatment of early gastric cancer: an interim report. Surgery 131:S306–S311
Surg Endosc (2011) 25:271–277 14. Catena F, Di Battista M, Fusaroli P, Ansaloni L, Di Scioscio V, Santini D, Pantaleo M, Biasco G, Caletti G, Pinna A (2008) Laparoscopic treatment of gastric GIST: report of 21 cases and literature’s review. J Gastrointest Surg 12:561–568 15. Miettinen M, Sobin LH, Lasota J (2005) Gastrointestinal stromal tumors of the stomach: a clinicopathologic, immunohistochemical, and molecular genetic study of 1765 cases with long-term follow-up. Am J Surg Pathol 29:52–68 16. Lee EJ, Kim TD, Oh HA, Lee HC, Kim JH, Jang BI, Kim TN, Chung MK, Bae YK (2005) Is the invasive approach for all the
277 upper gastrointestinal mesenchymal tumors necessary? Korean J Gastroenterol 45:387-–393 [Korean] 17. Hasegawa T, Matsuno Y, Shimoda T, Hirohashi S (2002) Gastrointestinal stromal tumor: consistent CD117 immunostaining for diagnosis, and prognostic classification based on tumor size and MIB-1 grade. Hum Pathol 33:669–676 18. Joensuu H, Fletcher C, Dimitrijevic S, Silberman S, Roberts P, Demetri G (2002) Management of malignant gastrointestinal stromal tumours. Lancet Oncol 3:655–664
123
