Available online at www.sciencedirect.com

EJSO 39 (2013) 150e155 www.ejso.com

Surgical treatment of locally advanced, non-metastatic, gastrointestinal stromaltumours after treatment with imatinib

R. Tielen a,*, C. Verhoef c, F. van Coevorden e, H. Gelderblom g, S. Sleijfer d, H.H. Hartgrink f,J.J. Bonenkamp a, W.T.A. van der Graaf b, J.H.W. de Wilt a

aDepartment of Surgical Oncology, Radboud University Nijmegen Medical Centre, Geert Grooteplein 10, 6525 GA Nijmegen, The NetherlandsbDepartment of Medical Oncology, Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands

cDepartment of Surgical Oncology, Erasmus Medical Centre, Daniel Den Hoed Cancer Centre, Rotterdam, The NetherlandsdDepartment of Medical Oncology, Erasmus Medical Centre, Daniel Den Hoed Cancer Centre, Rotterdam, The Netherlands

eDepartment of Surgical Oncology, The Netherlands Cancer Institute, Amsterdam, The NetherlandsfDepartment of Surgical Oncology, Leiden University Medical Centre, The NetherlandsgDepartment of Medical Oncology, Leiden University Medical Centre, The Netherlands

Accepted 26 September 2012

Available online 16 October 2012

Abstract

Aims: Patients with locally advanced gastrointestinal stromal tumours (GISTs) have a high risk of tumour perforation, incomplete tumourresections and often require multivisceral resections. Long-term disease-free and overall survival is usually impaired in this group of pa-tients. Induction therapy with imatinib followed by surgery seems to be beneficial in terms of improved surgical results and long-term out-come. We report on a large cohort of locally advanced GIST patients who have been treated in four centres in the Netherlands specialized inthe treatment of sarcomas.Methods: Between August 2001 and June 2011, 57 patients underwent surgery for locally advanced GISTs after imatinib treatment. Data ofall patients were retrospectively collected. Endpoints were progression-free and overall survival.Results: The patients underwent surgery after a median of 8 (range 1e55) months of imatinib treatment. Median tumour size before treat-ment was 12.2 (range 5.2e30) cm and reduced to 6.2 (range 1e20) cm before surgery. No tumour perforation occurred and a surgicalcomplete (R0) resection was achieved in 48 (84%) patients. Five-year PFS and OS were 77% and 88%. Eight patients had recurrent/met-astatic disease.Conclusions: Imatinib in locally advanced GIST is feasible and enables a high complete resection rate without tumour rupture. The com-bination of imatinib and surgery in patients with locally advanced GIST seems to improve PFS and OS.� 2012 Elsevier Ltd. All rights reserved.

Keywords: Gastrointestinal stromal tumour; Surgery; Imatinib

Introduction

Gastrointestinal stromal tumours (GISTs) are the mostcommon soft tissue tumours of the gastrointestinal tract,which arise from the interstitial cells of Cajal.1 The esti-mated prevalence is 1e2 per 100 000 persons.2,3 MostGISTs express KIT, a tyrosine kinase receptor, which canbe detected by immunohistochemistry using the CD117

* Corresponding author. Tel.: þ31 24 361 3808; fax: þ31 24 354 0501.

E-mail address: r.tielen@chir.umcn.nl (R. Tielen).

0748-7983/$ - see front matter � 2012 Elsevier Ltd. All rights reserved.

http://dx.doi.org/10.1016/j.ejso.2012.09.004

antibody.1 Over 80% of GISTs have an activating mutationin the KIT or, less often, the PDGFRA gene.4e6

Historically, treatment for locally advanced GISTs hasrelied on surgery as the first-line intervention, as responserates to conventional chemotherapy were less than10%.7,8 Approximately 70e85% of GISTs are primary re-sectable at first presentation depending on anatomic siteand/or tumour size. A high mitotic activity, large tumoursize, incomplete surgical resection and tumour perforationhave been identified as negative prognostic factors for re-lapse and survival.1,3,9e11

151R. Tielen et al. / EJSO 39 (2013) 150e155

In 2001, imatinib mesylate, a small molecular receptortyrosine kinase inhibitor, was found to inhibit mutatedKIT or PDGFA receptor tyrosine kinases.12,13 Imatinibhas been demonstrated to achieve partial response or stabledisease in nearly 80% of patients with advanced GIST.12e14

Treatment with imatinib is generally well-tolerated withmild side effects and is considered first-line treatment inmetastasized GIST patients.14 Nevertheless, progressionof disease occurs at a median time of 2 years from startof treatment through acquisition of additional activatingc-KIT or PDGFA mutations in tumour clones renderingthem imatinib refractory.15,16

Before the era of imatinib, surgical resection of locallyadvanced GISTs larger than 5 cm resulted in a median over-all survival (OS) of approximately 30 months and a recur-rence rate of up to 60% within 2 years.7e9 Because imatinibinduces downsizing of large tumours, it could potentiallyreduce the risk of tumour rupture during surgery and pro-vide an opportunity for a surgical complete and less morbid(i.e. organ-sparing) resection.17 This could lead to an im-proved disease-free and overall survival in patients with lo-cally advanced GISTs. Reports on surgical resectionfollowing imatinib treatment in patients with locally ad-vanced GIST are limited and usually consist of small retro-spective patient series.17e22 Most of these studies alsoincluded patients with both locally advanced and metastaticdisease. The present study is the first to retrospectivelyevaluate the long-term outcome in a large group of patientswho underwent surgery for locally advanced, non-metastatic, GIST after treatment with imatinib.

Methods

Patients and preoperative treatment

We reviewed all patients with a locally advanced GISTwho received imatinib before surgery was undertaken atfour Dutch institutions (The Netherlands Cancer Institute,Amsterdam; Leiden University Medical Centre, Leiden;Radboud University Nijmegen Medical Centre, Nijmegen;Erasmus Medical Centre, Daniel Den Hoed Cancer Centre,Rotterdam). These patients were evaluated in a multidisci-plinary sarcoma board at each centre before start of treat-ment. All tumours were considered too large (>5 cm)and/or ill-located for surgery by the sarcoma board. There-fore, imatinib was started in an attempt to downsize the tu-mour and prevent peroperative tumour rupture witha possibly less mutilating resection. Before start of imatiniba baseline CT was performed, and all patients were clini-cally and radiographically re-evaluated until surgery. Pa-tients were classified as having a complete response (CR),partial response (PR), stable disease (SD) or progressivedisease (PD) on the use of imatinib, based on serial imagingand scored according to Response Evaluation Criteria inSolid Tumours (RECIST).23 The decision when to performsurgery was tailor-made for each patient at the time the

multidisciplinary sarcoma board thought of a maximumtherapeutic response.

We collected patient- and treatment-specific data fromprospectively kept sarcoma databases, medical records da-tabases and patient charts at every institution. Data includedinitial symptoms, date of diagnosis, histopathological anal-yses, duration and dose of imatinib, complications on ima-tinib, best response to imatinib, date of surgery, type ofsurgical resection and (postoperative) complications, adju-vant imatinib treatment, date of recurrent/metastatic diseaseafter surgery, last follow-up and disease status at lastfollow-up, and if applicable, date of death.

Surgery and postoperative treatment

All resections were classified as R0 (macroscopicallycomplete resection with negative microscopic margins),R1 (macroscopically complete resection with positive mi-croscopic margins) or R2 (macroscopically incomplete re-section). Recurrent disease appearing after surgery in theregion of the previously located tumour is called ‘recur-rence’, and disease that had spread to distant sites, suchas the liver, is called ‘metastasis’. Imatinib treatment wasrestarted depending on completeness of resection and pref-erence of the treating physicians. Status of disease at lastfollow-up was determined using the most recent physicaland radiographical evaluation. If a patient had deceased,date of death and status of disease at death were recorded.The data of each patient was updated until July 2011.

Endpoints and statistics

Progression-free survival (PFS) was defined as the timefrom date of surgery to the date of clinical evidence of re-current or metastatic disease, date of last follow-up or deathfrom any cause, whatever occurred first. OS was defined asthe time from surgery to date of last follow-up or patientdeath. PFS and OS were estimated using the KaplaneMeiermethod. Statistical analysis was performed using SPSS sta-tistical software, version 16.0.

Results

Patients and preoperative imatinib treatment

A total of 57 patients (35 men and 22 women) were el-igible for evaluation. The median age was 61 (range29e82) years at the time of surgery after treatment with im-atinib. Details on tumour location and imatinib treatmentare summarized in Table 1. All GISTs were confirmed byexperienced sarcoma pathologists at each centre and werecharacterized by a positive c-KIT expression. Other tumourmarkers were not commonly assessed. Mutation status wasavailable in 30 patients: KIT exon 11 (n ¼ 18), KIT exon 9(n ¼ 1), KIT exon 12 mutation (n ¼ 1), KIT exon 18 mu-tation (n ¼ 1), KIT exon 9 and 17 mutation (n ¼ 1),

Table 1

Tumour location and initial therapy details.

n ¼ 57

Primary location

Stomach 37

Duodenum 2

Small intestine 6

Rectum 12

Preoperative imatinib therapya

Imatinib 400 50

Imatinib 400, switch 300b 2

Imatinib 400, switch 800c 5

a In mg daily dose.b Decreased dose due to complications.c Increased dose due to progressive disease.

Table 2

Operative procedures.

Procedures No.

Gastrectomy 22

Gastrectomy þ splenectomy þ pancreatectomy � bowel resection 14

Small bowel resection � bowel resection 6

Abdominoperineal resection 5

Duodenal resection 2

Rectosigmoid resection 1

Intersphincteric rectum amputation 1

Partial rectum resection 3

Exploratory laparotomya 1

Total/posterior exenteration 2

a Resection was not possible.

Table 3

Postoperative complications.

152 R. Tielen et al. / EJSO 39 (2013) 150e155

wildtype (n ¼ 4), no KIT exon 9 or 11 mutation (n ¼ 3). Inthe remaining 27 patients, the mutation status was not de-termined because it was not routinely performed in thepast or analysis was not possible due to technical difficul-ties. Median tumour size was 12.2 (range 5.2e30) cm be-fore start of imatinib. Treatment with imatinib 400 mgdaily was the first choice of treatment in all patients. Theprimary tumour size, possible invasion of surrounding or-gans on CT-imaging and technical difficult surgical proce-dures (i.e. ill-location), surgery was not the first choice intreatment. Two patients experienced gastrointestinal com-plications and imatinib was lowered to 300 mg daily beforesurgery. Two patients with a PR had to stop using imatinibafter 1 and 4 months because of severe toxidermic compli-cations and progressive (pre-existent) renal failure. Surgeryfollowed within two weeks after stopping imatinib. Two pa-tients shortly interrupted imatinib because of gastrointesti-nal complications and oedema, after a short stop imatinibwas continued at 800 mg daily dose because of disease pro-gression. Three patients experienced disease progressionfrom start of imatinib and switched to 800 mg daily. Twopatients experienced a PR and one patient PD after startingthe higher dose and surgery followed after 2, 2, and 4months, respectively. No patient was switched to second-line therapy. The tumour size after a median of 8 (range1e55) months of treatment with imatinib was 6.2 (range1e20) cm. One patient had a CR, 46 patients had a PR, 7patients had SD and 3 patients had PD at the time of sur-gery. One patient experienced an ongoing (partial) responsebefore disease stabilization at 51 months. Surgery followedat 55 months.

Complications No.

Intra-abdominal bleeding 4

Surgical outcome and postoperative treatment Anastomotic leakagea 2

Bowel perforation 1

Urinary tract infection 3

Enterocutaneous fistula 1

Gastric perforation 1

Abscess 2

Wound infection 5

Fascial dehiscence 1

a Involved anastomotic leakage of large bowel segments.

All patients underwent elective surgery and the proce-dures are listed in Table 2. In 6 patients no viable tumourcould be demonstrated at final pathology. An R0 resectionwas achieved in 48 patients and an R1 resection in 8 pa-tients. In 1 patient, resection of the tumour was not consid-ered feasible during surgery because of extensive tumourinvasion in liver, spleen, pancreas and duodenum. Despite

tumour shrinkage 19 patients were surgically treated withan en-bloc multivisceral resection. In the other patients,a less mutilating procedure was performed. One patient un-derwent a partial resection of the anterior wall of the rec-tum because the tumour was located between the prostateand rectum. The tumour was removed without performinga low anterior resection. Thirteen patients experienced atleast one surgical complication, with a total of 20 compli-cations (Table 3). Reoperations for complications were re-quired in four patients; postoperative bleeding (one), bowelperforation (one), and anastomotic leakage of large bowel(two). No postoperative mortality was observed within 30days of surgery. One patient with a bowel perforationdied 44 days after surgery. In 33 patients, imatinib was con-tinued following surgery for 1, 2 years or lifelong afterevaluation in the sarcoma board.

Progression-free and overall survival

Complete follow-up data were available for 55 patients.Two 2 patients were lost to follow-up. Median PFS mea-sured from time of surgery has not been reached, andone-, three- and five-year PFS have been estimated at96%, 87% and 77%. Eight patients experienced recurrent/metastatic disease; 3 patients during adjuvant imatinibtreatment and 5 patients without adjuvant imatinib treat-ment. These five patients were treated with imatinib atthe time of diagnosis of recurrent/metastatic disease. PFSbased on adjuvant imatinib treatment is shown in Fig. 1.

0 20 40 60 80 10050

60

70

80

90

100

110

Group BGroup A

Time (months)

Surv

ival

Figure 1. Progression-free survival (PFS) for patients treated adjuvant with

imatinib (Group A) and patients not treated adjuvant with imatinib (Group

B), calculated from date of surgery (months).

0 20 40 60 80 1000

20

40

60

80

100

Time (months)

Surv

ival

Figure 2. Overall survival (OS) of operated patients, calculated from start

of imatinib (months).

Table 4

Outcome of patients treated with imatinib followed by surgery for locally

advanced GIST.

Author No. of

patients

Follow-up

(median/months)

Survival and disease

status

Andtbacka

et al., 20062011 19.5 All alive at last

follow-up; 10 NED,

1 RD

Raut et al., 200621 9 14.6a 95% 1-year survival

SD, 86% 1-year

survival LP, 0% 1-year

survival GPa

Mearadji

et al., 2008199 40 All alive at last FU;

7 NED, 2 AWD

Eisenberg

et al., 20091730 36 83% 2-year PFS, 93%

2-year OS

Fiore et al., 200918 15 34 14 alive at last FU:

12 NED, 2 RD, 1 DD

Blesius

et al., 2011229 53.5 67% 3-year PFS,

89% 3-year OS

Present study 57 43 47 alive at last FU,

4 AWD, 3 DD

NED, no evidence of disease; RD, recurrent disease; AWD, alive with dis-

ease; DD, died of disease.a Patients were divided in three categories: stable disease (SD), limited

progression (LP), and generalized progression (GP).

153R. Tielen et al. / EJSO 39 (2013) 150e155

Although a trend to a higher PFS in the adjuvant imatinibgroup was shown, it was not significantly different( p ¼ 0.11). Seven patients developed distant metastasis:peritoneal (n ¼ 4), liver (n ¼ 1), peritoneal and liver(n ¼ 1) and abdominal wall (n ¼ 1). One patient experi-enced a local recurrence after an intersphincteric rectumamputation. Surgical procedures were performed in threepatients because of metastatic lesions. One patient under-went resection of several metastatic peritoneal lesions anda splenectomy 11 months after initial surgery, and has noevidence of disease 23 months after initial surgery. One pa-tient underwent a partial liver resection 26 months after re-section of the primary tumour. Because of re-recurrentmetastases, radiofrequency ablation of liver metastasesand resection of 2 peritoneal metastatic lesions combinedwith a partial small bowel resection was performed after51 months. This patient died of disease 59 months after ini-tial surgical treatment. Finally, one patient underwent re-section of abdominal wall metastases and resection of 2peritoneal lesions 9 months after initial surgery. This pa-tient is alive without evidence of disease.

At recent follow-up, 44 patients had no evidence of dis-ease, 4 patients were alive with disease, 4 patients had diedof disease and 3 patients died because of other reasons. Me-dian OS has not been reached, one-, three-, and five-yearOS have been estimated at 100%, 96% and 88% (Fig. 2).Four patients died of GIST. No correlation between clini-cal, pathological, and treatment variables with prognosiscould be demonstrated due to small sample size.

Discussion

This study is the largest study to date to report long-termoutcome of patients who underwent surgical resection oflocally advanced, non-metastatic GIST after imatinib ther-apy. Larger studies are available reporting the results ofboth locally advanced and metastatic GIST together asone group with or without surgery after imatinib therapy,which makes it difficult to compare with the results ofthis study.15,20 Several reports differentiate between locallyadvanced and metastatic GIST, but are usually comprised

of small patient series with a limited follow-up (Table 4).The present series demonstrates a high 5-year PFS andOS of 77% and 88% in a multicentre collected group of pa-tients with locally advanced GISTs.

Surgery remains the only possible curative treatment forGIST. Approximately 70e85% of patients with localizedGISTs can undergo a complete resection at first presenta-tion.7,24 A surgical complete resection is the most importantprognostic factor for patients with locally advanced, non-metastatic GIST. Furthermore, large tumours carry an in-creased risk of tumour rupture, which has a detrimental ef-fect on disease-free and overall survival. Tumour rupturereduced the median survival to approximately 17 months,

154 R. Tielen et al. / EJSO 39 (2013) 150e155

comparable to the median survival of 21 months after an in-complete resection in the pre-imatinib era as reported by Nget al.9 Recently, Hohenberger et al. reported that nearly allpatients develop abdominal metastases after rupture ofGIST.11 Therefore, if a GIST is large and the risk of tumourrupture is considered high, imatinib treatment should bestarted to increase the chance of an R0 resection and decreasethe potential risk of tumour perforation during surgery.

The discovery of gain-of-function KIT mutations inGIST by Hirota et al. and the introduction of imatinib,the small-molecular targeted therapy, revolutionized themanagement of GIST.4,12 Currently, imatinib is approvedworldwide as the first-line treatment of metastatic GIST.Toxicity and primary resistance to imatinib are the mainlimitations of this drug.13,25 Secondary resistance, definedas progressive disease at least 3 months after initiation ofimatinib, usually occurs at a median time of 2 years afterstart of treatment.13 Timing of resection is important if im-atinib is used as induction therapy in locally advanced tu-mours. If surgery is performed beyond the window oftherapeutic response, resistance or metastases might de-velop.15 In the present study, the median interval betweenstart of imatinib and surgery was 8 months, and no patientswitched to second-line therapy because of disease progres-sion. This might indicate that all tumours reached a plateauin their response to imatinib. A favourable outcome for re-sponding patients undergoing surgery following imatinibhas already been suggested in metastasized patients.19e21

Although this has not been confirmed in a randomized trial.In this study, tumour response on imatinib has been as-

sessed by a combination of physical and radiological exam-ination. A partial response on CT scan according toRECIST requires at least 6 months of imatinib therapy.15

Using size-based response criteria such as RECIST mightunderestimate the response to imatinib as suggested byclinical experience in the last few years.26,27 Changes in tu-mour nodules and vascularization should be combined withtumour density and smaller changes in size to evaluate po-tential responses by CT faster. A positron emission tomog-raphy (PET) using 18F-fluorodeoxyglucose (18FDG) maypredict responses to therapy better on short-term follow-up. It is instrumental when CT findings are inconsistentwith clinical findings.26,28,29 Response measurement usingthese 18FDG-PET scan criteria was not always possible be-cause of the retrospective nature of this study. Most patientswere diagnosed with GIST in other hospitals and an 18FDG-PET scan was not always available.

Given the anti-tumour activity of imatinib in the meta-static setting, with response rates of over 50%, imatinib is in-creasingly used. In patients with locally advanced GISTs, i.e.when resection is judged impossible or at the cost of consid-erable morbidity, it is now commonly used as induction ther-apy. Randomized controlled studies are hard to conduct formultiple reasons and the only prospective multicentric neo-adjuvant imatinib trial was the RTOG study.17 A better in-sight needs to be obtained by retrospective series. In the

present study, adequate clinical downsizing of the tumourwith preoperative imatinib therapy was demonstrated in 52patients who underwent an R0 resection without tumour per-foration. This reflects the main advantage of imatinib as in-duction therapy in patients with locally advanced GIST. Aless extensive resection rate was not clearly demonstratedin this study as 19 patients underwent multivisceral resec-tions. However, most multivisceral resections (n ¼ 14)were performed in the early treated patients and after 2007less extensive resections have become more common. Thisbias is probably caused by increased knowledge of the poten-tial benefits of imatinib and the tendency of locally advancedGIST not to invade surrounding organs.

At a median postoperative follow-up time of 40 months,the 5-year estimated PFS was 77% measured from surgery.This seems an improvement compared to historical data,since patients with tumours larger than 10 cm experienceda disease-specific 5-year survival of only 20% after resec-tion in recent literature.7,8 However, this comparison is bi-ased because 33 patients received adjuvant treatment withimatinib. Median PFS for patients treated with adjuvant im-atinib has not been reached and for patients who receivedno adjuvant treatment it was 49 (range 9e56) months. Ina phase III trial, patients with GISTs larger than 3 cm un-derwent surgical resection and received adjuvant imatinibor placebo for 1 year.25 Significantly fewer recurrenceswere noted in patients who received imatinib for 1 yearafter complete resection compared to patients receivingplacebo. Median recurrence-free survival was not reachedfor tumours between 6 and 10 cm, and medianrecurrence-free survival for tumours 10 cm or greater wasapproximately 35 months in the imatinib group. Recently,Joensuu et al. reported that administration of imatinib for36 months after surgery improves recurrence-free and over-all survival compared to 12 months in patients with a highestimated risk of recurrence.30 Although these results haveto be published, adjuvant treatment with imatinib is nowconsidered standard treatment in the referral hospitals ofthis study group.

Median OS after surgical resection of GISTs larger than5 cm is reported to be approximately 27e32 months.7-9 TheOS in the present study is substantially higher with 83% ofpatients alive at 5 years, and median OS not reached after49 months. This might suggest that imatinib therapy fol-lowed by surgical resection enables adequate surgical re-sections with a low chance on tumour perforation andmight prolong OS. Once again, firm conclusions are hardto draw as the current median OS of patients with meta-static disease treated with imatinib and/or sunitinib hasmeanwhile dramatically improved as well.

Conclusion

Evaluation of patients with a locally advanced, non-metastatic GIST in a multidisciplinary tumour board inhigh-volume GIST centres has proven to be successful. It

155R. Tielen et al. / EJSO 39 (2013) 150e155

supplies the best strategy for treatment and prevention ofdisease progression. Imatinib as induction therapy is con-sidered a useful tool in patients with locally advancedGIST. This results in a decrease of tumour size in most pa-tients, and thereby increases the chances of a surgical com-plete resection without tumour rupture. Combining imatiniband surgery in patients with locally advanced GIST seemsto improve PFS and OS compared to available historical re-ported series.

Acknowledgements

The author(s) indicated no potential conflict of interest.No support of funding was obtained for this report.

This abstract was accepted as a poster presentation dur-ing the CTOS 16th Annual Meeting 2010 (abstract ID892339).

Conflict of interest

Hereby, all authors declare to have no conflicts of inter-est as stated and signed by all authors in the “Author Form”.

References

1. Fletcher CD, Berman JJ, Corless C, et al. Diagnosis of gastrointestinal

stromal tumors: a consensus approach. Hum Pathol 2002 May;33(5):

459–65.

2. Joensuu H, Kindblom LG. Gastrointestinal stromal tumors e a review.

Acta Orthop Scand Suppl 2004 Apr;75(311):62–71.

3. Miettinen M, Lasota J. Gastrointestinal stromal tumors e definition,

clinical, histological, immunohistochemical, and molecular genetic

features and differential diagnosis. Virchows Arch 2001 Jan;438(1):

1–12.

4. Hirota S, Isozaki K, Moriyama Y, et al. Gain-of-function mutations of

c-kit in human gastrointestinal stromal tumors. Science 1998 Jan 23;

279(5350):577–80.

5. Hirota S, Ohashi A, Nishida T, et al. Gain-of-function mutations of

platelet-derived growth factor receptor alpha gene in gastrointestinal

stromal tumors. Gastroenterology 2003 Sep;125(3):660–7.

6. Heinrich MC, Corless CL, Duensing A, et al. PDGFRA activating mu-

tations in gastrointestinal stromal tumors. Science 2003 Jan 31;

299(5607):708–10.

7. 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 Jan;231(1):

51–8.

8. Dematteo RP, Heinrich MC, El-Rifai WM, Demetri G. Clinical man-

agement of gastrointestinal stromal tumors: before and after STI-571.

Hum Pathol 2002 May;33(5):466–77.

9. Ng EH, Pollock RE, Munsell MF, Atkinson EN, Romsdahl MM. Prog-

nostic factors influencing survival in gastrointestinal leiomyosarco-

mas. Implications for surgical management and staging. Ann Surg

1992 Jan;215(1):68–77.

10. Langer C, Gunawan B, Schuler P, Huber W, Fuzesi L, Becker H. Prog-

nostic factors influencing surgical management and outcome of gastro-

intestinal stromal tumours. Br J Surg 2003 Mar;90(3):332–9.

11. Hohenberger P, Ronellenfitsch U, Oladeji O, et al. Pattern of recur-

rence in patients with ruptured primary gastrointestinal stromal

tumour. Br J Surg 2010 Dec;97(12):1854–9.

12. Joensuu H, Roberts PJ, Sarlomo-Rikala M, et al. Effect of the tyrosine

kinase inhibitor STI571 in a patient with a metastatic gastrointestinal

stromal tumor. N Engl J Med 2001 Apr 5;344(14):1052–6.

13. Demetri GD, von Mehren M, Blanke CD, et al. Efficacy and safety of

imatinib mesylate in advanced gastrointestinal stromal tumors. N Engl

J Med 2002 Aug 15;347(7):472–80.

14. van Oosterom AT, Judson I, Verweij J, et al. Safety and efficacy of im-

atinib (STI571) in metastatic gastrointestinal stromal tumours: a phase

I study. Lancet 2001 Oct 27;358(9291):1421–3.

15. Verweij J, Casali PG, Zalcberg J, et al. Progression-free survival in

gastrointestinal stromal tumours with high-dose imatinib: randomised

trial. Lancet 2004 Sep 25eOct 1;364(9440):1127–34.

16. Antonescu CR, Besmer P, Guo T, et al. Acquired resistance to imatinib

in gastrointestinal stromal tumor occurs through secondary gene mu-

tation. Clin Cancer Res 2005 Jun 1;11(11):4182–90.

17. Eisenberg BL, Harris J, Blanke CD, et al. Phase II trial of neoadjuvant/

adjuvant imatinib mesylate (IM) for advanced primary and metastatic/

recurrent operable gastrointestinal stromal tumor (GIST): early results

of RTOG 0132/ACRIN 6665. J Surg Oncol 2009 Jan 1;99(1):42–7.

18. Fiore M, Palassini E, Fumagalli E, et al. Preoperative imatinib mesy-

late for unresectable or locally advanced primary gastrointestinal stro-

mal tumors (GIST). Eur J Surg Oncol 2009 Jul;35(7):739–45.

19. Mearadji A, den Bakker MA, van Geel AN, et al. Decrease of CD117

expression as possible prognostic marker for recurrence in the resected

specimen after imatinib treatment in patients with initially unresect-

able gastrointestinal stromal tumors: a clinicopathological analysis.

Anticancer Drugs 2008 Jul;19(6):607–12.

20. Andtbacka RH, Ng CS, Scaife CL, et al. Surgical resection of gastro-

intestinal stromal tumors after treatment with imatinib. Ann Surg

Oncol 2007 Jan;14(1):14–24.

21. Raut CP, Posner M, Desai J, et al. Surgical management of advanced

gastrointestinal stromal tumors after treatment with targeted systemic

therapy using kinase inhibitors. J Clin Oncol 2006 May 20;24(15):

2325–31.

22. Blesius A, Cassier PA, Bertucci F, et al. Neoadjuvant imatinib in pa-

tients with locally advanced non metastatic GIST in the prospective

BFR14 trial. BMC Cancer 2011 Feb 15;11(1):72.

23. Duffaud F, Therasse P. New guidelines to evaluate the response to

treatment in solid tumors. Bull Cancer 2000 Dec;87(12):881–6.

24. Roberts PJ, Eisenberg B. Clinical presentation of gastrointestinal stro-

mal tumors and treatment of operable disease. Eur J Cancer 2002 Sep;

38(Suppl. 5):S37–8.

25. Dematteo RP, Ballman KV, Antonescu CR, et al. Adjuvant imatinib

mesylate after resection of localised, primary gastrointestinal stromal

tumour: a randomised, double-blind, placebo-controlled trial. Lancet

2009 Mar 28;373(9669):1097–104.

26. Choi H, Charnsangavej C, de Castro Faria S, et al. CT evaluation of

the response of gastrointestinal stromal tumors after imatinib mesylate

treatment: a quantitative analysis correlated with FDG PET findings.

AJR Am J Roentgenol 2004 Dec;183(6):1619–28.

27. Benjamin RS, Choi H, Macapinlac HA, et al. We should desist using

RECIST, at least in GIST. J Clin Oncol 2007 May 1;25(13):1760–4.

28. Choi H. Response evaluation of gastrointestinal stromal tumors.

Oncologist 2008;13(Suppl. 2):4–7.

29. Van den Abbeele AD. The lessons of GISTePET and PET/CT: a new

paradigm for imaging. Oncologist 2008;13(Suppl. 2):8–13.

30. Joensuu H, Eriksson M, Hatrmann J, et al. Twelve versus 36 months of

adjuvant imatinib (IM) as treatment of operable GIST with a high risk

of recurrence: final results of a randomized trial (SSGXVIII/AIO).

ASCO meeting abstracts. 2011 June 23, vol. 29(Suppl. 18): LBA1;

2011.