Tumor Biol. DOI 10.1007/s13277-013-1572-7
RESEARCH ARTICLE
Clinicopathologic factors predicting outcomes in patients with gastrointestinal stromal tumors of the rectum and colon Zhongguo Zhou & Zhanpeng Chen & Minshan Chen & Ruojing Wang & Ying Yin & Yuan Yao
Received: 7 November 2013 / Accepted: 17 December 2013 # International Society of Oncology and BioMarkers (ISOBM) 2013
Abstract Limited information is available concerning the clinicopathologic profile of colorectal gastrointestinal stromal tumors (GISTs), which are relatively rare, as well as survival rates following surgical resection. The present study was designed to describe the clinicopathologic characteristics of patients with colorectal GISTs and identify potential factors that may predict postoperative survival outcomes. We reviewed the medical records of 67 patients with colorectal GISTs who underwent surgical resection between January 2000 and December 2012. Clinicopathologic factors affecting overall survival were assessed using the Kaplan-Meier method and multivariate Cox proportional hazards models. The median age at diagnosis was 57 years (range, 32–79 years), with a male-to-female ratio of 1.68. Tumor size varied from 0.2 to 11 cm, with a median size of 5.7 cm. Sixty-two tumors (93 %) were positive for CD117, 53 (79 %) for CD34, 6 (9 %)
for PDGFRA, 15 (22 %) for SMA, 5 (8 %) for S100, and 39 (58 %) for vimentin. The overall median survival time was 54 months (95 % confidence interval, 32–59 months), and the 1-, 3-, and 5-year overall survival rates were 89.6, 63.9, and 34.4 %, respectively. High-risk tumors (determined by histologic grade, size, and other histologic variables) were associated with poor prognosis (hazard ratio, 1.83; 95 % confidence interval, 1.21–2.78), and patients who received adjuvant treatment with imatinib had significantly longer median overall survival times than patients who did not (hazard ratio, 0.43; 95 % confidence interval, 0.24–0.80). In patients with colorectal GISTs, high-risk tumor histologic grade can predict poor prognosis, and patients may benefit from adjuvant treatment with imatinib. Findings from the present study may provide information to establish evidence-based management strategies for colorectal GISTs.
Zhongguo Zhou and Zhanpeng Chen contributed to this work equally.
Keywords Gastrointestinal stromal tumors . Imatinib . Overall survival . CD117
Z. Zhou : M. Chen Department of Hepatobiliary Surgery, Sun Yat-sen University Cancer Center; State Key Laboratory of Oncology in South China, Collaborative Innovation Center for Cancer Medicine, 651# DongFeng East Road, GuangZhou 510060, People’s Republic of China Z. Chen Department of General Surgery, The 5th People’s Hospital of Foshan City, 63# Jiangpu East Road, Nanhai District, FoShan 528211, China R. Wang Department of Neurosurgery, The First Affiliated Hospital of Sun Yat-sen University, 74# Zhongshan Er Lu, Guangzhou 510089, People’s Republic of China Y. Yin : Y. Yao (*) Department of General Surgery, Guangdong General Hospital, Guangdong Academy of Medical Sciences, 96 Dong Chuan Lu, Guangzhou 510080, People’s Republic of China e-mail: [email protected]
Introduction Gastrointestinal stromal tumors (GISTs), which arise from the interstitial cell of Cajal or its precursor, are the most common mesenchymal neoplasms of the gastrointestinal tract [1]. GISTs are morphologically, immunohistochemically, and molecularly distinct from other gastrointestinal tract spindle cell neoplasms of mesenchymal origin [2]. Although GISTs occur throughout the gastrointestinal tract, those involving the rectum and colon are relatively rare, accounting for only 5–10 % of all diagnosed cases and 0.1 % of all malignant colorectal tumors [3]. Clinically, colorectal GISTs may present in various ways, ranging from small, asymptomatic intramural nodules to large, outward-bulging masses that cause occult bleeding, abdominal pain, and a palpable mass [4].
Tumor Biol.
Complete surgical resection with negative tumor margins (R0) is generally agreed to be the principal curative procedure for primary and nonmetastatic GISTs because they are resistant to conventional chemotherapy, and the role of radiotherapy has not yet been decisively established [5, 6]. However, R0 resections are difficult to achieve for rectal GISTs because of involvement of surrounding structures, such as the vagina, bladder, prostate, or pelvis [7]. In addition, colonic GISTs are often large at the time of presentation and large tumors predict a poor prognosis [8]. Therefore, adjuvant treatment, such as treatment with imatinib, a tyrosine kinase inhibitor, may be necessary in patients with colorectal GISTs more so than in patients with GISTs at other sites. Some studies specifically indicate that most patients with colorectal GISTs are at increased risk for local or distant recurrence compared with patients with GISTs of other origins [9, 10]. Unfortunately, studies involving large groups of patients under long-term follow-up observation are lacking, mainly owing to the extremely low incidence of colorectal GISTs, making it difficult to establish effective treatment strategies. Moreover, the clinicopathologic profile of colorectal GISTs has not yet been accurately characterized, and therefore, the prognostic factors for colorectal GISTs are often assumed to be the same as for GISTs at other sites [11, 12]. Therefore, the purpose of our study is to describe the clinicopathologic profile of colorectal GISTs and to identify potential factors that may predict postoperative outcomes.
Methods Patients with GISTs arising from the colon or rectum who underwent surgical resection in our hospital were included in our analysis. We reviewed medical records for data on patients’ age, sex, clinical symptoms, tumor site, maximal tumor diameter, surgical procedure, extent of surgical resection, and clinical outcome until the last follow-up. Survival data were obtained for all patients by reviewing the clinical records and through direct communication with the attending physicians. Patients with incomplete medical records were excluded from analysis. All pathologic slides were reviewed by two experienced pathologists from our institute. Histopathologic examination was carried out using standard hematoxylin and eosin staining as well as specific immunohistochemical techniques allowing identification of tumor grade (low, intermediate, or high), size, stromal background, stage, and number of mitoses per highpower field. Tumors were classified according to the National Institutes of Health (NIH) workshop [13] into 4 groups: very low-risk tumors, low-risk tumors, intermediate-risk tumors, and high-risk tumors. For our analysis, we combined the very low-risk group with the low-risk group, owing to the small
number of tumors in these groups. Surgical procedures were independently determined by the corresponding physician, mainly on the basis of tumor size and site. Additionally, the need for preoperative or postoperative treatment with tyrosine kinase inhibitors was assessed by the treatment team. Statistical analyses were performed using SPSS software (version 10.0; SPSS, Inc., Chicago, IL). The Kaplan-Meier method was used to calculate the overall survival rate. Variables among the groups were compared using the nonparametric log-rank test. Multivariate Cox proportional hazard models were used to estimate hazard ratios for overall survival and to determine independent risk factors. P50 years Sex Male Female Anatomic location Colon Rectum Tumor size
Median overall survival time, months (95 % confidence interval)
20 (30) 47 (70)
56 (29–74) 52 (30–59)
42 (63) 25 (37)
54 (30–68) 56 (18–59)
19 (28) 48 (72)
51 (30–58) 60 (26–69)
28 (42) 31 (46) 8 (12)
60 (30–71) 54 (30–68) 34 (15–57)
P value
0.41
0.22
0.92
0.11
≤5 cm 5.1–10 cm >10 cm National Institutes of Health risk category Low Intermediate High Type of resection Local resection Local resection + lymphadenectomy Multivisceral resection Adjuvant therapy Yes No
received adjuvant treatment with imatinib at a daily dose of 400 mg for a median of 8 months (range, 4–24 months). Among these patients, nine received imatinib before surgery and 20 after surgery.
N (%)
0.034 30 (45) 23 (34) 14 (21)
56 (30–77) 56 (32–69) 18 (8–57)
25 (37) 28 (42) 14 (21)
58 (29–68) 51 (30–64) 30 (13–70)
29 (43) 38 (57)
56 (43–74) 39 (19–59)
0.20
0.01
No differences in median survival times were found between patients whose tumors stained positive for CD34, VIM, PDGFRA, SMA, or S100 and those whose tumors did not (Table 2). Survival analysis was not performed according to
Survival analysis The overall median survival time was 54 months (95 % confidence interval, 32–59 months), and the 1-, 3-, and 5-year overall survival rates were 89.6, 63.9, and 34.4 %, respectively (Fig. 1). Univariate analysis of age, gender, anatomic location, tumor size, and surgery procedures showed a tendency towards differences in mean survival time for patients older than 50 years, male patients, patients with colonic tumors, patients with tumors larger than 10 cm, and patients who underwent multivisceral resection had shorter median overall survival times than their counterparts, but the differences were not statistically significant. However, a significantly longer median overall survival time was observed in patients with low-risk tumors compared with those with intermediate or high-risk tumors (Fig. 2). In addition, patients who received adjuvant therapy had significantly longer median overall survival times than those who did not (56 vs 39 months; Fig. 3).
Fig. 1 Kaplan-Meier curve for overall survival among patients with gastrointestinal stromal tumors of rectal or colonic origin (n=67)
Tumor Biol. Table 2 Median overall survival time by histopathologic features of patients with gastrointestinal stromal tumors of rectal or colonic origin (n=67) Feature
N (%)
Median overall survival time, months (95 % confidence interval)
5 (7) 62 (93)
– –
28 (42) 39 (58)
50 (19–59) 56 (32–70)
14 (21) 53 (79)
50 (14–74) 54 (30–60)
61 (91) 6 (9)
56 (34–60) 28 (12–120)
–
CD117
Fig. 2 Kaplan-Meier curves for overall survival among patients with gastrointestinal stromal tumors of rectal or colonic origin by tumor risk classification
CD117 status because of the small number of tumors staining negative for CD117. The multivariate Cox regression model confirmed that NIH risk categories and adjuvant therapy were independent prognostic factors (Table 3). High-risk tumors were associated with poor prognosis (hazard ratio, 1.83; 95 % confidence interval, 1.21–2.78), and patients who received adjuvant treatment with imatinib had significantly longer median overall survival times than patients who did not (hazard ratio, 0.43; 95 % confidence interval, 0.24–0.80).
− + VIM − + CD34 − + PDGFRA − + SMA − + S100 − +
P value
0.27
0.79
0.55
0.64 52 (78) 15 (22)
57 (39–68) 28 (12–51)
62 (93) 5 (7)
54 (34–60) 30 (20–71)
0.53
Discussion Although a growing number of studies concerning GISTs have been published, little information is available concerning the management of colorectal GISTs, mainly owing to the rarity of these tumors. Clinical outcomes after surgical treatment remain unclear, and the efficacy of tyrosine kinase inhibitors, such as imatinib mesylate, is also unclear. Findings from the present study may provide information to establish evidence-based management strategies for colorectal GISTs. Our findings demonstrated that high-risk tumors were associated with poor prognosis for patients with colorectal GISTs and that adjuvant treatment with imatinib led to improved postoperative overall survival time. Among patients in our study with rectal GISTs, the median age at onset was 59 years, which was in line with previous Table 3 Independent prognostic factors for overall survival in patients with gastrointestinal stromal tumors of rectal or colonic origin (n=67) according to multivariate analysis
Fig. 3 Kaplan-Meier curves for overall survival among patients with gastrointestinal stromal tumors of rectal or colonic origin by adjuvant therapy
Prognostic factor
β
Hazard ratio (95 % confidence interval)
P value
National Institutes of Health risk category Adjuvant therapy
0.61
1.83 (1.21–2.78)
0.034
−0.84
0.43 (0.24–0.80)
0.027
Tumor Biol.
studies. Similarly, our cohort of patients with rectal GISTs showed a substantial male predominance, with a male-tofemale ratio of 1.71 [14, 15]. However, in patients with colonic tumors, the median age at onset was 55 years in our study, which was slightly lower than that reported in another study (62 years) [4]. Moreover, although Miettinen et al. reported an even sex distribution among patients with colonic GISTs [4], a substantial sex difference was observed in our patients, with a male-to-female ratio of 1.67. However, the median tumor size in our cohort (including both those with rectal tumors and those with colonic tumors) was 5.7 cm, which was in agreement with a previous report (5.68 ± 2.64 cm) [14]. CD117, the antigen corresponding to the KIT protein, is considered the most specific defining marker for GISTs. Consistent with previous studies [4, 14, 15], we found that 93 % of patients in our cohort had CD117-positive tumors, confirming the key role of CD117 in the diagnosis of GIST. CD34, a hematopoietic progenitor cell antigen, is also commonly seen in colorectal GISTs. In this study, 79 % of patients had tumors showing positive immunostaining for CD34, whereas the CD34-positive rate of colorectal GISTs across other studies varies greatly, ranging from 50 to 100 %. However, the rates of positive staining for SMA, PDGFRA, and S100 in our study (22, 9, and 8 %, respectively) were similar to those reported elsewhere. Radical resection is universally recognized as a curative treatment for GISTs, regardless of where the tumor originates [5, 6]. The tumor should be removed en bloc with its pseudocapsule to an adequate resection margin. In patients in our cohort, local resection was performed for patients without metastasis (37 %). Although previous studies have indicated that GISTs rarely gave rise to lymph node metastasis [3, 16], we preferred to perform regional lymph node resection in patients with an enlarged or suspicious lymph node (41.8 %) to decrease the potential risk of metastasis as much as possible. Multivisceral resections (involving contiguous organ metastasis) were performed in 14 patients (21 %), and none of these patients experienced late complications, confirming the feasibility of en bloc resection. The 5-year overall survival rate for all 67 patients in our study was 34.4 %, which is consistent with the findings of DeMatteo et al [17],. who reported a 5-year overall survival rate of 35 % in a group of 200 patients encountered over 16 years, and higher than the rate reported by Ng et al [18],. who found a 5-year overall survival rate of 28 % in 191 patients over 30 years. In both of these studies, as in ours, patients with metastasis or local recurrence were included in the analysis. However, our 5-year overall survival rate was substantially lower than the 40–65 % rates reported for GISTs of various origins in a number of other studies. Although admission rate bias may be a possible explanation for the difference because outcomes depend on the status and extent
of disease when treated, the difference might also be a reflection of the fact that colorectal GISTs involve more risk than GISTs originating in other sites. Consistent with previous studies [14, 17], we found longer overall survival times in patients with low-risk tumors than in patients with intermediate and high-risk tumors (hazard ratio, 1.83; 95 % confidence interval, 1.21–2.78), confirming the strong impact of NIH risk classification on prognosis of patients with GISTs. However, tumor size did not significantly predict overall survival time, although obvious differences were observed between groups according to tumor size. This seemed inconsistent with results presented in a consensus report [18–20]. Similar findings (i.e., that tumor size was not associated with overall survival) were also reported by Wu et al [21],. who suggested that patient selection bias was a possible reason for the discrepancy, and by El-Zohairy et al [22],. who ascribed the difference to small sample size and lower overall frequency of deaths in their study group. Small sample size might also be the cogent explanation for the lack of association between tumor size and overall survival in our study; however, further study is needed for clarification. Imatinib is now more or less accepted as a valid treatment for advanced GISTs; however, critical questions remain about the appropriate duration of imatinib therapy in different contexts, and this depends mainly on the efficacy and tolerability of the drug in patients with GISTs [23, 24]. Information on the long-term survival of patients with GISTs is very limited because imatinib does not have a long history of use for the treatment of GISTs. In our study, patients who received adjuvant treatment with imatinib for a median duration of 8 months had significantly longer median overall survival times than patients who did not receive the treatment (hazard ratio, 0.43; 95 % confidence interval, 0.24–0.80), confirming the efficacy of imatinib as a treatment for GISTs. In conclusion, GISTs are a new and growing clinical problem, and our retrospective study provided useful information on the clinicopathologic and histopathologic features of patients with colorectal GISTs.
Conflicts of interest None
References 1. Miettinen M, Sarlomo-Rikala M, Lasota J. Gastrointestinal stromal tumours. Ann Chir Gynaecol. 1998;87(4):278–81. 2. Miettinen M, Sarlomo-Rikala M, Lasota J. Gastrointestinal stromal tumors: recent advances in understanding of their biology. Hum Pathol. 1999;30(10):1213–20. 3. Kisluk J, Gryko M, Guzinska-Ustymowicz K, Kemona A, Kedra B. Immunohistochemical diagnosis of gastrointestinal stromal tumors— an analysis of 80 cases from 2004 to 2010. Adv Clin Exp Med. 2013;22(1):33–9.
Tumor Biol. 4. Miettinen M, Sarlomo-Rikala M, Sobin LH, Lasota J. Gastrointestinal stromal tumors and leiomyosarcomas in the colon: a clinicopathologic, immunohistochemical, and molecular genetic study of 44 cases. Am J Surg Pathol. 2000;24(10):1339–52. 5. Bamboat ZM, Dematteo RP. Updates on the management of gastrointestinal stromal tumors. Surg Oncol Clin N Am. 2012;21(2):301– 16. doi:10.1016/j.soc.2011.12.004. 6. Rammohan A, Sathyanesan J, Rajendran K, Pitchaimuthu A, Perumal SK, Srinivasan U, et al. A gist of gastrointestinal stromal tumors: a review. World J Gastrointest Oncol. 2013;5(6):102–12. doi:10.4251/wjgo.v5.i6.102. 7. Xiao CC, Zhang S, Wang MH, Huang LY, Wu P, Xu Y, et al. Clinicopathological features and prognostic factors of rectal gastrointestinal stromal tumors. J Gastrointest Surg. 2013;17(4):793–8. doi: 10.1007/s11605-012-2086-0. 8. Macias-Garcia F, Parada P, Martinez-Lesquereux L, Pintos E, Fraga M, Dominguez-Munoz JE. Gastrointestinal stromal tumors (GISTs) of the colon. Rev Esp Enferm Dig. 2010;102(6):388–90. 9. Hassan I, You YN, Dozois EJ, Shayyan R, Smyrk TC, Okuno SH, et al. Clinical, pathologic, and immunohistochemical characteristics of gastrointestinal stromal tumors of the colon and rectum: implications for surgical management and adjuvant therapies. Dis Colon Rectum. 2006;49(5):609–15. doi:10.1007/s10350-006-0503-8. 10. Hassan I, You YN, Shyyan R, Dozois EJ, Smyrk TC, Okuno SH, et al. Surgically managed gastrointestinal stromal tumors: a comparative and prognostic analysis. Ann Surg Oncol. 2008;15(1):52–9. doi:10.1245/s10434-007-9633-z. 11. Dematteo RP, Gold JS, Saran L, Gonen M, Liau KH, Maki RG, et al. Tumor mitotic rate, size, and location independently predict recurrence after resection of primary gastrointestinal stromal tumor (GIST). Cancer. 2008;112(3):608–15. doi:10.1002/cncr.23199. 12. Martin J, Poveda A, Llombart-Bosch A, Ramos R, Lopez-Guerrero JA, Garcia del Muro J, et al. Deletions affecting codons 557–558 of the cKIT gene indicate a poor prognosis in patients with completely resected gastrointestinal stromal tumors: a study by the Spanish Group for Sarcoma Research (GEIS). J Clin Oncol. 2005;23(25):6190–8. 13. Miettinen M, Lasota J. Gastrointestinal stromal tumors—definition, clinical, histological, immunohistochemical, and molecular genetic features and differential diagnosis. Virchows Arch. 2001;438(1):1–12.
14. Shi Y, Hou YY, Lu SH, Zhou Y, Xu JF, Ji Y, et al. Clinical and pathological studies of borderline gastrointestinal stromal tumors. Chin Med J (Engl). 2010;123(18):2514–20. 15. Bulbul DG. Gastrointestinal stromal tumors: a multicenter study of 1160 Turkish cases. Turk J Gastroenterol. 2012;23(3):203–11. 16. Boni L, Benevento A, Dionigi G, Rovera F, Dionigi R. Surgical resection for gastrointestinal stromal tumors (GIST): experience on 25 patients. World J Surg Oncol. 2005;3:78. doi:10.1186/1477-78193-78. 17. DeMatteo RP, Lewis JJ, Leung D, Mudan SS, Woodruff JM, Brennan MF. Two hundred gastrointestinal stromal tumors: recurrence patterns and prognostic factors for survival. Ann Surg. 2000;231(1):51– 8. 18. Ng EH, Pollock RE, Munsell MF, Atkinson EN, Romsdahl MM. Prognostic factors influencing survival in gastrointestinal leiomyosarcomas. Implications for surgical management and staging. Ann Surg. 1992;215(1):68–77. 19. Fletcher CD, Berman JJ, Corless C, Gorstein F, Lasota J, Longley BJ, et al. Diagnosis of gastrointestinal stromal tumors: a consensus approach. Int J Surg Pathol. 2002;10(2):81–9. 20. Theodoropoulos DG. Gastrointestinal tumors of the colon and rectum. Clin Colon Rectal Surg. 2011;24(3):161–70. doi:10.1055/s0031-1286000. 21. Wu TJ, Lee LY, Yeh CN, Wu PY, Chao TC, Hwang TL, et al. Surgical treatment and prognostic analysis for gastrointestinal stromal tumors (GISTs) of the small intestine: before the era of imatinib mesylate. BMC Gastroenterol. 2006;6:29. doi:10.1186/1471-230X-6-29. 22. El-Zohairy M, el Khalil SA, Fakhr I, El-Shahawy M, Gouda I. Gastrointestinal stromal tumor (GIST)’s surgical treatment, NCI experience. J Egypt Natl Canc Inst. 2005;17(2):56–66. 23. Linch M, Claus J, Benson C. Update on imatinib for gastrointestinal stromal tumors: duration of treatment. Onco Targets Ther. 2013;6: 1011–23. doi:10.2147/OTT.S31260. 24. Tielen R, Verhoef C, van Coevorden F, Gelderblom H, Sleijfer S, Hartgrink HH, et al. Surgery after treatment with imatinib and/or sunitinib in patients with metastasized gastrointestinal stromal tumors: is it worthwhile? World J Surg Oncol. 2012;10:111. doi:10. 1186/1477-7819-10-111.
RESEARCH ARTICLE
Clinicopathologic factors predicting outcomes in patients with gastrointestinal stromal tumors of the rectum and colon Zhongguo Zhou & Zhanpeng Chen & Minshan Chen & Ruojing Wang & Ying Yin & Yuan Yao
Received: 7 November 2013 / Accepted: 17 December 2013 # International Society of Oncology and BioMarkers (ISOBM) 2013
Abstract Limited information is available concerning the clinicopathologic profile of colorectal gastrointestinal stromal tumors (GISTs), which are relatively rare, as well as survival rates following surgical resection. The present study was designed to describe the clinicopathologic characteristics of patients with colorectal GISTs and identify potential factors that may predict postoperative survival outcomes. We reviewed the medical records of 67 patients with colorectal GISTs who underwent surgical resection between January 2000 and December 2012. Clinicopathologic factors affecting overall survival were assessed using the Kaplan-Meier method and multivariate Cox proportional hazards models. The median age at diagnosis was 57 years (range, 32–79 years), with a male-to-female ratio of 1.68. Tumor size varied from 0.2 to 11 cm, with a median size of 5.7 cm. Sixty-two tumors (93 %) were positive for CD117, 53 (79 %) for CD34, 6 (9 %)
for PDGFRA, 15 (22 %) for SMA, 5 (8 %) for S100, and 39 (58 %) for vimentin. The overall median survival time was 54 months (95 % confidence interval, 32–59 months), and the 1-, 3-, and 5-year overall survival rates were 89.6, 63.9, and 34.4 %, respectively. High-risk tumors (determined by histologic grade, size, and other histologic variables) were associated with poor prognosis (hazard ratio, 1.83; 95 % confidence interval, 1.21–2.78), and patients who received adjuvant treatment with imatinib had significantly longer median overall survival times than patients who did not (hazard ratio, 0.43; 95 % confidence interval, 0.24–0.80). In patients with colorectal GISTs, high-risk tumor histologic grade can predict poor prognosis, and patients may benefit from adjuvant treatment with imatinib. Findings from the present study may provide information to establish evidence-based management strategies for colorectal GISTs.
Zhongguo Zhou and Zhanpeng Chen contributed to this work equally.
Keywords Gastrointestinal stromal tumors . Imatinib . Overall survival . CD117
Z. Zhou : M. Chen Department of Hepatobiliary Surgery, Sun Yat-sen University Cancer Center; State Key Laboratory of Oncology in South China, Collaborative Innovation Center for Cancer Medicine, 651# DongFeng East Road, GuangZhou 510060, People’s Republic of China Z. Chen Department of General Surgery, The 5th People’s Hospital of Foshan City, 63# Jiangpu East Road, Nanhai District, FoShan 528211, China R. Wang Department of Neurosurgery, The First Affiliated Hospital of Sun Yat-sen University, 74# Zhongshan Er Lu, Guangzhou 510089, People’s Republic of China Y. Yin : Y. Yao (*) Department of General Surgery, Guangdong General Hospital, Guangdong Academy of Medical Sciences, 96 Dong Chuan Lu, Guangzhou 510080, People’s Republic of China e-mail: [email protected]
Introduction Gastrointestinal stromal tumors (GISTs), which arise from the interstitial cell of Cajal or its precursor, are the most common mesenchymal neoplasms of the gastrointestinal tract [1]. GISTs are morphologically, immunohistochemically, and molecularly distinct from other gastrointestinal tract spindle cell neoplasms of mesenchymal origin [2]. Although GISTs occur throughout the gastrointestinal tract, those involving the rectum and colon are relatively rare, accounting for only 5–10 % of all diagnosed cases and 0.1 % of all malignant colorectal tumors [3]. Clinically, colorectal GISTs may present in various ways, ranging from small, asymptomatic intramural nodules to large, outward-bulging masses that cause occult bleeding, abdominal pain, and a palpable mass [4].
Tumor Biol.
Complete surgical resection with negative tumor margins (R0) is generally agreed to be the principal curative procedure for primary and nonmetastatic GISTs because they are resistant to conventional chemotherapy, and the role of radiotherapy has not yet been decisively established [5, 6]. However, R0 resections are difficult to achieve for rectal GISTs because of involvement of surrounding structures, such as the vagina, bladder, prostate, or pelvis [7]. In addition, colonic GISTs are often large at the time of presentation and large tumors predict a poor prognosis [8]. Therefore, adjuvant treatment, such as treatment with imatinib, a tyrosine kinase inhibitor, may be necessary in patients with colorectal GISTs more so than in patients with GISTs at other sites. Some studies specifically indicate that most patients with colorectal GISTs are at increased risk for local or distant recurrence compared with patients with GISTs of other origins [9, 10]. Unfortunately, studies involving large groups of patients under long-term follow-up observation are lacking, mainly owing to the extremely low incidence of colorectal GISTs, making it difficult to establish effective treatment strategies. Moreover, the clinicopathologic profile of colorectal GISTs has not yet been accurately characterized, and therefore, the prognostic factors for colorectal GISTs are often assumed to be the same as for GISTs at other sites [11, 12]. Therefore, the purpose of our study is to describe the clinicopathologic profile of colorectal GISTs and to identify potential factors that may predict postoperative outcomes.
Methods Patients with GISTs arising from the colon or rectum who underwent surgical resection in our hospital were included in our analysis. We reviewed medical records for data on patients’ age, sex, clinical symptoms, tumor site, maximal tumor diameter, surgical procedure, extent of surgical resection, and clinical outcome until the last follow-up. Survival data were obtained for all patients by reviewing the clinical records and through direct communication with the attending physicians. Patients with incomplete medical records were excluded from analysis. All pathologic slides were reviewed by two experienced pathologists from our institute. Histopathologic examination was carried out using standard hematoxylin and eosin staining as well as specific immunohistochemical techniques allowing identification of tumor grade (low, intermediate, or high), size, stromal background, stage, and number of mitoses per highpower field. Tumors were classified according to the National Institutes of Health (NIH) workshop [13] into 4 groups: very low-risk tumors, low-risk tumors, intermediate-risk tumors, and high-risk tumors. For our analysis, we combined the very low-risk group with the low-risk group, owing to the small
number of tumors in these groups. Surgical procedures were independently determined by the corresponding physician, mainly on the basis of tumor size and site. Additionally, the need for preoperative or postoperative treatment with tyrosine kinase inhibitors was assessed by the treatment team. Statistical analyses were performed using SPSS software (version 10.0; SPSS, Inc., Chicago, IL). The Kaplan-Meier method was used to calculate the overall survival rate. Variables among the groups were compared using the nonparametric log-rank test. Multivariate Cox proportional hazard models were used to estimate hazard ratios for overall survival and to determine independent risk factors. P50 years Sex Male Female Anatomic location Colon Rectum Tumor size
Median overall survival time, months (95 % confidence interval)
20 (30) 47 (70)
56 (29–74) 52 (30–59)
42 (63) 25 (37)
54 (30–68) 56 (18–59)
19 (28) 48 (72)
51 (30–58) 60 (26–69)
28 (42) 31 (46) 8 (12)
60 (30–71) 54 (30–68) 34 (15–57)
P value
0.41
0.22
0.92
0.11
≤5 cm 5.1–10 cm >10 cm National Institutes of Health risk category Low Intermediate High Type of resection Local resection Local resection + lymphadenectomy Multivisceral resection Adjuvant therapy Yes No
received adjuvant treatment with imatinib at a daily dose of 400 mg for a median of 8 months (range, 4–24 months). Among these patients, nine received imatinib before surgery and 20 after surgery.
N (%)
0.034 30 (45) 23 (34) 14 (21)
56 (30–77) 56 (32–69) 18 (8–57)
25 (37) 28 (42) 14 (21)
58 (29–68) 51 (30–64) 30 (13–70)
29 (43) 38 (57)
56 (43–74) 39 (19–59)
0.20
0.01
No differences in median survival times were found between patients whose tumors stained positive for CD34, VIM, PDGFRA, SMA, or S100 and those whose tumors did not (Table 2). Survival analysis was not performed according to
Survival analysis The overall median survival time was 54 months (95 % confidence interval, 32–59 months), and the 1-, 3-, and 5-year overall survival rates were 89.6, 63.9, and 34.4 %, respectively (Fig. 1). Univariate analysis of age, gender, anatomic location, tumor size, and surgery procedures showed a tendency towards differences in mean survival time for patients older than 50 years, male patients, patients with colonic tumors, patients with tumors larger than 10 cm, and patients who underwent multivisceral resection had shorter median overall survival times than their counterparts, but the differences were not statistically significant. However, a significantly longer median overall survival time was observed in patients with low-risk tumors compared with those with intermediate or high-risk tumors (Fig. 2). In addition, patients who received adjuvant therapy had significantly longer median overall survival times than those who did not (56 vs 39 months; Fig. 3).
Fig. 1 Kaplan-Meier curve for overall survival among patients with gastrointestinal stromal tumors of rectal or colonic origin (n=67)
Tumor Biol. Table 2 Median overall survival time by histopathologic features of patients with gastrointestinal stromal tumors of rectal or colonic origin (n=67) Feature
N (%)
Median overall survival time, months (95 % confidence interval)
5 (7) 62 (93)
– –
28 (42) 39 (58)
50 (19–59) 56 (32–70)
14 (21) 53 (79)
50 (14–74) 54 (30–60)
61 (91) 6 (9)
56 (34–60) 28 (12–120)
–
CD117
Fig. 2 Kaplan-Meier curves for overall survival among patients with gastrointestinal stromal tumors of rectal or colonic origin by tumor risk classification
CD117 status because of the small number of tumors staining negative for CD117. The multivariate Cox regression model confirmed that NIH risk categories and adjuvant therapy were independent prognostic factors (Table 3). High-risk tumors were associated with poor prognosis (hazard ratio, 1.83; 95 % confidence interval, 1.21–2.78), and patients who received adjuvant treatment with imatinib had significantly longer median overall survival times than patients who did not (hazard ratio, 0.43; 95 % confidence interval, 0.24–0.80).
− + VIM − + CD34 − + PDGFRA − + SMA − + S100 − +
P value
0.27
0.79
0.55
0.64 52 (78) 15 (22)
57 (39–68) 28 (12–51)
62 (93) 5 (7)
54 (34–60) 30 (20–71)
0.53
Discussion Although a growing number of studies concerning GISTs have been published, little information is available concerning the management of colorectal GISTs, mainly owing to the rarity of these tumors. Clinical outcomes after surgical treatment remain unclear, and the efficacy of tyrosine kinase inhibitors, such as imatinib mesylate, is also unclear. Findings from the present study may provide information to establish evidence-based management strategies for colorectal GISTs. Our findings demonstrated that high-risk tumors were associated with poor prognosis for patients with colorectal GISTs and that adjuvant treatment with imatinib led to improved postoperative overall survival time. Among patients in our study with rectal GISTs, the median age at onset was 59 years, which was in line with previous Table 3 Independent prognostic factors for overall survival in patients with gastrointestinal stromal tumors of rectal or colonic origin (n=67) according to multivariate analysis
Fig. 3 Kaplan-Meier curves for overall survival among patients with gastrointestinal stromal tumors of rectal or colonic origin by adjuvant therapy
Prognostic factor
β
Hazard ratio (95 % confidence interval)
P value
National Institutes of Health risk category Adjuvant therapy
0.61
1.83 (1.21–2.78)
0.034
−0.84
0.43 (0.24–0.80)
0.027
Tumor Biol.
studies. Similarly, our cohort of patients with rectal GISTs showed a substantial male predominance, with a male-tofemale ratio of 1.71 [14, 15]. However, in patients with colonic tumors, the median age at onset was 55 years in our study, which was slightly lower than that reported in another study (62 years) [4]. Moreover, although Miettinen et al. reported an even sex distribution among patients with colonic GISTs [4], a substantial sex difference was observed in our patients, with a male-to-female ratio of 1.67. However, the median tumor size in our cohort (including both those with rectal tumors and those with colonic tumors) was 5.7 cm, which was in agreement with a previous report (5.68 ± 2.64 cm) [14]. CD117, the antigen corresponding to the KIT protein, is considered the most specific defining marker for GISTs. Consistent with previous studies [4, 14, 15], we found that 93 % of patients in our cohort had CD117-positive tumors, confirming the key role of CD117 in the diagnosis of GIST. CD34, a hematopoietic progenitor cell antigen, is also commonly seen in colorectal GISTs. In this study, 79 % of patients had tumors showing positive immunostaining for CD34, whereas the CD34-positive rate of colorectal GISTs across other studies varies greatly, ranging from 50 to 100 %. However, the rates of positive staining for SMA, PDGFRA, and S100 in our study (22, 9, and 8 %, respectively) were similar to those reported elsewhere. Radical resection is universally recognized as a curative treatment for GISTs, regardless of where the tumor originates [5, 6]. The tumor should be removed en bloc with its pseudocapsule to an adequate resection margin. In patients in our cohort, local resection was performed for patients without metastasis (37 %). Although previous studies have indicated that GISTs rarely gave rise to lymph node metastasis [3, 16], we preferred to perform regional lymph node resection in patients with an enlarged or suspicious lymph node (41.8 %) to decrease the potential risk of metastasis as much as possible. Multivisceral resections (involving contiguous organ metastasis) were performed in 14 patients (21 %), and none of these patients experienced late complications, confirming the feasibility of en bloc resection. The 5-year overall survival rate for all 67 patients in our study was 34.4 %, which is consistent with the findings of DeMatteo et al [17],. who reported a 5-year overall survival rate of 35 % in a group of 200 patients encountered over 16 years, and higher than the rate reported by Ng et al [18],. who found a 5-year overall survival rate of 28 % in 191 patients over 30 years. In both of these studies, as in ours, patients with metastasis or local recurrence were included in the analysis. However, our 5-year overall survival rate was substantially lower than the 40–65 % rates reported for GISTs of various origins in a number of other studies. Although admission rate bias may be a possible explanation for the difference because outcomes depend on the status and extent
of disease when treated, the difference might also be a reflection of the fact that colorectal GISTs involve more risk than GISTs originating in other sites. Consistent with previous studies [14, 17], we found longer overall survival times in patients with low-risk tumors than in patients with intermediate and high-risk tumors (hazard ratio, 1.83; 95 % confidence interval, 1.21–2.78), confirming the strong impact of NIH risk classification on prognosis of patients with GISTs. However, tumor size did not significantly predict overall survival time, although obvious differences were observed between groups according to tumor size. This seemed inconsistent with results presented in a consensus report [18–20]. Similar findings (i.e., that tumor size was not associated with overall survival) were also reported by Wu et al [21],. who suggested that patient selection bias was a possible reason for the discrepancy, and by El-Zohairy et al [22],. who ascribed the difference to small sample size and lower overall frequency of deaths in their study group. Small sample size might also be the cogent explanation for the lack of association between tumor size and overall survival in our study; however, further study is needed for clarification. Imatinib is now more or less accepted as a valid treatment for advanced GISTs; however, critical questions remain about the appropriate duration of imatinib therapy in different contexts, and this depends mainly on the efficacy and tolerability of the drug in patients with GISTs [23, 24]. Information on the long-term survival of patients with GISTs is very limited because imatinib does not have a long history of use for the treatment of GISTs. In our study, patients who received adjuvant treatment with imatinib for a median duration of 8 months had significantly longer median overall survival times than patients who did not receive the treatment (hazard ratio, 0.43; 95 % confidence interval, 0.24–0.80), confirming the efficacy of imatinib as a treatment for GISTs. In conclusion, GISTs are a new and growing clinical problem, and our retrospective study provided useful information on the clinicopathologic and histopathologic features of patients with colorectal GISTs.
Conflicts of interest None
References 1. Miettinen M, Sarlomo-Rikala M, Lasota J. Gastrointestinal stromal tumours. Ann Chir Gynaecol. 1998;87(4):278–81. 2. Miettinen M, Sarlomo-Rikala M, Lasota J. Gastrointestinal stromal tumors: recent advances in understanding of their biology. Hum Pathol. 1999;30(10):1213–20. 3. Kisluk J, Gryko M, Guzinska-Ustymowicz K, Kemona A, Kedra B. Immunohistochemical diagnosis of gastrointestinal stromal tumors— an analysis of 80 cases from 2004 to 2010. Adv Clin Exp Med. 2013;22(1):33–9.
Tumor Biol. 4. Miettinen M, Sarlomo-Rikala M, Sobin LH, Lasota J. Gastrointestinal stromal tumors and leiomyosarcomas in the colon: a clinicopathologic, immunohistochemical, and molecular genetic study of 44 cases. Am J Surg Pathol. 2000;24(10):1339–52. 5. Bamboat ZM, Dematteo RP. Updates on the management of gastrointestinal stromal tumors. Surg Oncol Clin N Am. 2012;21(2):301– 16. doi:10.1016/j.soc.2011.12.004. 6. Rammohan A, Sathyanesan J, Rajendran K, Pitchaimuthu A, Perumal SK, Srinivasan U, et al. A gist of gastrointestinal stromal tumors: a review. World J Gastrointest Oncol. 2013;5(6):102–12. doi:10.4251/wjgo.v5.i6.102. 7. Xiao CC, Zhang S, Wang MH, Huang LY, Wu P, Xu Y, et al. Clinicopathological features and prognostic factors of rectal gastrointestinal stromal tumors. J Gastrointest Surg. 2013;17(4):793–8. doi: 10.1007/s11605-012-2086-0. 8. Macias-Garcia F, Parada P, Martinez-Lesquereux L, Pintos E, Fraga M, Dominguez-Munoz JE. Gastrointestinal stromal tumors (GISTs) of the colon. Rev Esp Enferm Dig. 2010;102(6):388–90. 9. Hassan I, You YN, Dozois EJ, Shayyan R, Smyrk TC, Okuno SH, et al. Clinical, pathologic, and immunohistochemical characteristics of gastrointestinal stromal tumors of the colon and rectum: implications for surgical management and adjuvant therapies. Dis Colon Rectum. 2006;49(5):609–15. doi:10.1007/s10350-006-0503-8. 10. Hassan I, You YN, Shyyan R, Dozois EJ, Smyrk TC, Okuno SH, et al. Surgically managed gastrointestinal stromal tumors: a comparative and prognostic analysis. Ann Surg Oncol. 2008;15(1):52–9. doi:10.1245/s10434-007-9633-z. 11. Dematteo RP, Gold JS, Saran L, Gonen M, Liau KH, Maki RG, et al. Tumor mitotic rate, size, and location independently predict recurrence after resection of primary gastrointestinal stromal tumor (GIST). Cancer. 2008;112(3):608–15. doi:10.1002/cncr.23199. 12. Martin J, Poveda A, Llombart-Bosch A, Ramos R, Lopez-Guerrero JA, Garcia del Muro J, et al. Deletions affecting codons 557–558 of the cKIT gene indicate a poor prognosis in patients with completely resected gastrointestinal stromal tumors: a study by the Spanish Group for Sarcoma Research (GEIS). J Clin Oncol. 2005;23(25):6190–8. 13. Miettinen M, Lasota J. Gastrointestinal stromal tumors—definition, clinical, histological, immunohistochemical, and molecular genetic features and differential diagnosis. Virchows Arch. 2001;438(1):1–12.
14. Shi Y, Hou YY, Lu SH, Zhou Y, Xu JF, Ji Y, et al. Clinical and pathological studies of borderline gastrointestinal stromal tumors. Chin Med J (Engl). 2010;123(18):2514–20. 15. Bulbul DG. Gastrointestinal stromal tumors: a multicenter study of 1160 Turkish cases. Turk J Gastroenterol. 2012;23(3):203–11. 16. Boni L, Benevento A, Dionigi G, Rovera F, Dionigi R. Surgical resection for gastrointestinal stromal tumors (GIST): experience on 25 patients. World J Surg Oncol. 2005;3:78. doi:10.1186/1477-78193-78. 17. DeMatteo RP, Lewis JJ, Leung D, Mudan SS, Woodruff JM, Brennan MF. Two hundred gastrointestinal stromal tumors: recurrence patterns and prognostic factors for survival. Ann Surg. 2000;231(1):51– 8. 18. Ng EH, Pollock RE, Munsell MF, Atkinson EN, Romsdahl MM. Prognostic factors influencing survival in gastrointestinal leiomyosarcomas. Implications for surgical management and staging. Ann Surg. 1992;215(1):68–77. 19. Fletcher CD, Berman JJ, Corless C, Gorstein F, Lasota J, Longley BJ, et al. Diagnosis of gastrointestinal stromal tumors: a consensus approach. Int J Surg Pathol. 2002;10(2):81–9. 20. Theodoropoulos DG. Gastrointestinal tumors of the colon and rectum. Clin Colon Rectal Surg. 2011;24(3):161–70. doi:10.1055/s0031-1286000. 21. Wu TJ, Lee LY, Yeh CN, Wu PY, Chao TC, Hwang TL, et al. Surgical treatment and prognostic analysis for gastrointestinal stromal tumors (GISTs) of the small intestine: before the era of imatinib mesylate. BMC Gastroenterol. 2006;6:29. doi:10.1186/1471-230X-6-29. 22. El-Zohairy M, el Khalil SA, Fakhr I, El-Shahawy M, Gouda I. Gastrointestinal stromal tumor (GIST)’s surgical treatment, NCI experience. J Egypt Natl Canc Inst. 2005;17(2):56–66. 23. Linch M, Claus J, Benson C. Update on imatinib for gastrointestinal stromal tumors: duration of treatment. Onco Targets Ther. 2013;6: 1011–23. doi:10.2147/OTT.S31260. 24. Tielen R, Verhoef C, van Coevorden F, Gelderblom H, Sleijfer S, Hartgrink HH, et al. Surgery after treatment with imatinib and/or sunitinib in patients with metastasized gastrointestinal stromal tumors: is it worthwhile? World J Surg Oncol. 2012;10:111. doi:10. 1186/1477-7819-10-111.
