| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
EDUCATIONAL REVIEW |
From the Norris Cotton Cancer Center, Lebanon, New Hampshire, (BLE); and the Institute of Cancer Research, Surrey, United Kingdom (IJ).
Correspondence: Address correspondence and reprint requests to: Burton L. Eisenberg, MD, Section of Surgical Oncology, Norris Cotton Cancer Center, Rubin Building, 8th Floor, Dartmouth Hitchcock Medical Center, One Medical Center Drive, Lebanon, NH 03756; Fax: 603-653-9003; E-mail: burton.l.eisenberg{at}dartmouth.edu
ABSTRACT
Gastrointestinal stromal tumor (GIST) is a neoplasm of the gastrointestinal tract, mesentery, or omentum that expresses the protein-tyrosine kinase KIT (CD117) and is the most common mesenchymal tumor arising at these sites. Surgical resection is the first-line intervention for operable GISTs, particularly localized primary tumors, and it was historically the only effective treatment. However, more than half of all GIST patients present with locally advanced, recurrent, or metastatic disease. The 5-year survival rate ranges from 50% to 65% after complete resection of a localized primary GIST and decreases to approximately 35% for patients with advanced disease who undergo complete surgical resection. A total of 40% to 90% of all GIST surgical patients subsequently have postoperative recurrence or metastasis. Imatinib is a potent, specific inhibitor of KIT that has demonstrated significant activity and tolerability in the treatment of malignant unresectable or metastatic GIST, inducing tumor shrinkage of 50% or more or stabilizing disease in most patients. A key strategy for prolonging the survival of patients with GIST is to improve the outcome of surgery. It is possible that the adjuvant and neoadjuvant use of imatinib (e.g., rendering initially inoperable tumors resectable) in the overall management approach to advanced GIST may contribute to surgeons’ success in attaining this objective.
Key Words: Gastrointestinal stromal tumor • Surgery • Imatinib • KIT • Signal • Transduction inhibitor
Gastrointestinal stromal tumor (GIST) is the most common mesenchymal tumor of the gastrointestinal tract. Although GIST has been clinically recognized for almost 20 years, a standard definition has only recently been clarified. In addition to being regarded for many years as a smooth-muscle neoplasm, GIST was proposed to have a neurogenic origin, which included possible involvement of Auerbach’s plexus.1 Kindblom et al.2 and others3 observed that GISTs express CD117 (KIT protein) as well as CD34 and thus share similar markers with the interstitial cells of Cajal. These findings followed a report by Hirota et al.4 indicating that mutations in the KIT oncogene resulted in constitutive activation of the KIT receptor tyrosine kinase in GISTs.4–7 These tumors are now categorized morphologically as spindle-cell or mixed epithelioid neoplasms located within the gastrointestinal tract, presumably sharing a common progenitor cell with the interstitial cells of Cajal. GISTs are characterized by immunohistochemical expression of the CD117 antigen.2,8,9
The insights gained during the past 5 years concerning the molecular pathogenesis of GIST have led to rapid advances in knowledge about this disease and to rapid improvements in its management. GIST seems to be more common than previously thought, which we now know because KIT positivity as a defining characteristic allows for increased accuracy in the identification of cases. It is important to note that KIT is not only the principal marker for diagnostic purposes; it is now also a specific target for systemic therapy. Imatinib (Glivec, Gleevec, formerly STI571; Novartis Pharma AG, Basel, Switzerland) is an orally bioavailable 2-phenylaminopyrimidine derivative that potently and selectively inhibits KIT, Bcr-Abl, and platelet-derived growth factor receptor-
(PDGF-R) and -ß (PDGF-Rß) protein-tyrosine kinases.10–12
Imatinib is currently approved for the treatment of KIT-positive malignant metastatic or unresectable GIST and for first-line use in adults and children with chronic myeloid leukemia in all phases. In the clinical trials of imatinib in patients with advanced GIST, a tumor previously known to be resistant to all conventional chemotherapy, most patients achieved significant tumor regression or stable disease; side effects were generally mild to moderate (grade 1 or 2).13–15
Surgery is the principal initial treatment for patients with resectable GIST, particularly primary, nonmetastatic tumors, but it is seldom curative. Fewer than half of GIST patients present with localized primary disease, and postoperative recurrence or metastasis is seen in 40% to 90% of all cases treated surgically.16–18 A key challenge in the treatment of GIST is to improve the outcome of resection. The availability of a highly effective and well-tolerated systemic therapy may potentially offer benefits for GIST patients who are—or might become—candidates for surgery. This review examines the current roles of surgery and imatinib in the management of GIST. It also discusses findings that may affect decision making about the possible use of both surgery and imatinib in combination, with an update on the ongoing clinical trials that are specifically addressing the issue of adjuvant and neoadjuvant imatinib administration.
CLINICAL OVERVIEW
GIST accounts for approximately .1% to 3% of all gastrointestinal tract neoplasms, 5% to 6% of all sarcomas, and 80% of gastrointestinal mesenchymal tumors.16,19–21 Before the recognition of KIT expression as a ubiquitous feature of GISTs, they were frequently classified as leiomyomas, leiomyosarcomas, leiomyoblastomas, or gastrointestinal autonomic nerve tumors.8,22 Consequently, the true frequency of GIST has been difficult to determine, and past estimates seem to have been too low. A recent population-based study in Sweden of 600 confirmed KIT-positive GISTs indicated an incidence of approximately 15 to 20 cases per million persons per year for symptomatic and clinically detected tumors.23 In addition, asymptomatic GISTs are commonly found incidentally during surgery or endoscopy for other conditions, during digital rectal examination, and at autopsy. The rapid enrollment of GIST patients into current clinical trials has also prompted upward revision of incidence estimates recently, from 300 to 500 cases per year in the United States to 5,000 to 10,000 cases.9,24 These data suggest that an increased index of suspicion for GIST is warranted when neoplasms are detected in the gastrointestinal tract and adjacent sites.
The presence of KIT expression in a gastrointestinal mesenchymal tumor with a histological pattern characteristic of GIST establishes the diagnosis. The KIT tyrosine kinase is detected immunohistochemically by reactivity with the CD117 antigen, generally resulting in strong and diffuse cytoplasmic staining.9,22 Types of tumors that should be routinely assessed by CD117 immunohistochemistry are listed in Table 1.25 A small subset of GISTs (
5%) have very low levels of KIT expression, and confounding factors and atypical findings, such as focal staining, may complicate the interpretation of CD117 assay results.26,27 Therefore, experienced histopathologic evaluation remains an important facet of GIST diagnosis. In unusual circumstances, investigation for KIT mutations is recommended when other findings are insufficient to establish the diagnosis and when the selection of treatment is at issue.26 Because most GISTs develop submucosally and may not allow for adequate endoscopic sampling of tissue, a definitive diagnosis often cannot be made before surgery. In addition, there are concerns that percutaneous fine-needle aspiration may risk tumor rupture, leading to intra-abdominal seeding, and may be nondiagnostic for large tumors containing hemorrhagic or necrotic areas.28,29
|
2 to 5 cm) with low mitotic activity have been found to recur 
5 years after presumed curative surgery.31,32 It is in fact somewhat questionable whether any GIST is truly benign, because long-term follow-up may be necessary to determine the biological activity of any individual tumor.17,33
Overall characteristics of disease at diagnosis are a general prognostic indicator. GISTs initially presenting with clinical signs and symptoms are more likely to have a malignant course than asymptomatic, incidentally detected solitary tumors.16,34,35 Tumor location has also been cited as a prognostic factor: gastric GISTs tend to be at lower risk for recurrence than esophageal or small- or large-bowel GISTs.20,34–36 However, the site of origin is not a consistently reliable predictor of patient survival.16 The two factors most strongly predictive of risk of aggressive tumor behavior and adverse outcome are tumor size and mitotic rate.9 Two criteria—size >5 cm and mitotic rate 5 per 50 high-power fields (HPFs)—increase the risk estimate for disease recurrence from low to intermediate or higher.
SURGERY
Complete surgical resection is the primary treatment modality for GIST. Total excision of the tumor is the most significant factor for outcome and can be accomplished in 40% to 60% of all GIST patients and in 70% of those with primary, nonmetastatic disease.16,17,20,37 Although the disease is metastatic or locally recurrent at presentation in approximately 50% to 60% of patients with GIST,16,37,38 complete resection was achieved in 24% to 46% of such patients in one large series.16
The objective of surgery is removal of all gross tumor, which may require total or subtotal resection, depending on tumor location and size. En-bloc resection of the GIST and its pseudocapsule, if present, should be performed and is the recommended approach whenever feasible in cases of contiguous organ involvement. Wide margins are not generally necessary for disease clearance, but the achievement of clear margins may require consideration of complete or partial organ sacrifice. Lymphadenectomy is not routinely necessary, because regional lymph node involvement is rare in GIST30; however, a local peritonectomy should be performed when practicable because of the frequency of local peritoneal seeding.17 Avoidance of tumor rupture is imperative. In general, the standard guidelines for organ resection, organ preservation, and reanastomosis should govern the surgical resection technique for GIST. En-bloc resection may involve radical excision (e.g., total gastrectomy, pancreaticoduodenectomy, or abdominoperineal resection), which can lead to considerable postoperative morbidity. Potential complications are associated with the complexity of the procedure and can include bleeding, infection, and anastomotic leak.
Successful use of laparoscopic techniques for the resection of primary GISTs and other gastric tumors has been reported in individual cases and small series.39–47 The GISTs were localized and were small (3 cm) or characterized as benign or of low-grade malignancy. Advantages of laparoscopic resection cited by the investigators included minimal manipulation of the tumor as well as effectiveness for diagnosis and treatment in patients presenting with acute gastrointestinal bleeding.41,44,47,48 One group that used laparoscopic wedge resection to treat 34 patients with submucosal tumors of the stomach, including 14 GISTs, reported no disease recurrences over a 5-year follow-up period.45 However, long-term data for patients who have undergone laparoscopic resection for GIST are generally lacking, and the number of GIST patients in published cases or series is small.
Outcomes
Evidence from long-term follow-up of patients who have undergone surgical resection of a high-risk GIST indicates that surgery alone is generally not curative. As many as 85% to 90% have an adverse outcome—recurrence, metastasis, or death.17,49 In general, local recurrences or metastases develop in approximately half of patients who have potentially curative operations for GIST, regardless of the site of the primary tumor, and 5- and 10-year survival rates after potentially curative surgery are 32% to 78% and 19% to 63%, respectively.17,30 The median disease-specific survival for patients with primary GIST is approximately 5 years.50 Outcomes reported in recent studies are consistent with those in earlier series (Table 2).
|
Resection of Recurrent or Metastatic GIST
Outcomes in patients with metastatic GIST and in those with GIST recurrence after primary resection were usually extremely poor in the era before the introduction of imatinib: the median survival of such patients generally ranged from 6 months to approximately 18 months.16,24 In most cases, GIST recurrence and dissemination are intra-abdominal only and become evident by a median of 20 to 25 months after primary resection.16,20,54 Metastases develop most frequently in the liver, and the peritoneum or omentum is the next most common site; extra-abdominal spread to the regional lymph nodes, lungs, bones, or subcutaneous sites is an uncommon finding.8,16,20,38 The liver is the sole site of recurrence or metastasis in approximately 40% to 50% of patients.16,20,55
Results of surgical management of GIST recurrence or spread have been variable, depending on such factors as the stage of disease, tumor risk profile, and length of the disease-free interval after initial resection. In some patients whose primary tumor was a very-low-risk or low-risk (2 to 5 cm; 5 mitoses per 50 HPFs) rectal or anal GIST, locally recurrent disease has been treated successfully with total excision, without further recurrence over follow-up periods ranging from 4 to >10 years.31 In their study of 200 GISTs, DeMatteo et al.16 analyzed outcomes after first recurrence in the patients who underwent complete resection of primary disease. Complete resection of a localized recurrent tumor resulted in a median survival (54 months) comparable to that after complete resection of a localized primary GIST. However, median survival declined to 5 months with incomplete resection of either locally recurrent or concomitant local and metastatic recurrent disease and declined to 10 months with incomplete resection and to 16 months with complete resection of metastatic recurrent disease.
Mudan et al.54 reported a median survival of 15 months after surgery for recurrent GIST; the longest survival was observed in patients whose recurrence consisted of hepatic metastasis alone. In this study, the only significant determinant of survival was the duration of the disease-free period between initial surgery and GIST recurrence, an indicator of the biological aggressiveness of the tumor. In another study of 56 patients (34 with GIST or gastrointestinal leiomyosarcomas) who underwent complete resection for liver metastasis of sarcoma, an interval >2 years between diagnosis of the primary tumor and development of the metastasis was found to be a significant predictor of survival after hepatectomy.56 Complete resection of hepatic metastases was associated with prolonged survival in this study.
When the clinical presentation suggests that a patient with recurrent GIST might be a candidate for surgery, comprehensive diagnostic imaging is required for preoperative staging. In most cases, computed tomography is satisfactory for the demonstration of GIST in the liver, although magnetic resonance imaging affords greater sensitivity for small lesions.38,57 Positron emission tomography is proving to be a very sensitive staging tool. Complete surgical resection should be attempted in selected patients whose recurrent or metastatic disease is localized in a single site (e.g., liver) or consists of low-volume, multiple-site lesions on the peritoneal surfaces. Resection of multiple intra-abdominal organs and surgery for tumor debulking are not warranted and provide no appreciable benefit, except perhaps for palliation of localized bleeding or obstruction in patients whose performance status is otherwise excellent.24 Surgery for recurrent and/or metastatic GIST is contraindicated in patients with poor performance status and significant comorbid disease.
Adjuvant Chemotherapy or Radiotherapy
Before the availability of imatinib, the only treatments for GIST other than surgery were conventional chemotherapy and radiotherapy. However, lack of efficacy has been a consistent finding in studies that evaluated results of radiation or chemotherapy in GIST patients.28,33 A total of 11% and 33% of the patients analyzed by DeMatteo et al.16 received radiotherapy and chemotherapy, respectively, but neither modality had an effect on outcome. In the series of 50 malignant small-intestinal GISTs analyzed by Crosby et al.,20 all of the 10 patients who received adjuvant radiotherapy subsequently relapsed. Recurrence was inside the radiation field in three patients, and no patient received adjuvant chemotherapy. Recently, a report by investigators from the Italian Sarcoma Group on multicenter experience with chemotherapy for GISTs diagnosed between 1979 and 1999 provided further confirmation of previous findings.55 In a retrospective analysis that included 67 patients with advanced GIST given either combination chemotherapy (n = 51) or monochemotherapy (n = 16), no patient had a complete response, 6 had partial responses, and median survival calculated from the start of chemotherapy was 16 months (median follow-up, 11 months).55 Among 15 patients who received adjuvant chemotherapy, 7 subsequently relapsed. The median survival was 38 months (median follow-up, 25 months), and the overall long-term survival (37% at 4 years) did not differ from that of advanced GIST patients generally.
Intraperitoneal chemotherapy, either alone or as an adjuvant to surgery, has been used in an attempt to improve outcomes in patients with GIST.58,59 Eilber et al.58 found that although aggressive resection of recurrent disease followed by intraperitoneal chemotherapy reduced the rate of peritoneal GIST recurrence, this approach was ineffective in preventing hepatic metastasis and therefore had a negligible effect on overall survival. Hepatic arterial chemoembolization is an option for patients with cancer metastatic to the liver and has been associated with a mean survival of 9.5 to 11.4 months.59–62 The addition of radiofrequency ablation to transcatheter arterial chemoembolization increased mean survival to 25 months, which was similar to the survival of 19 to 23 months reported for hepatic arterial infusion of chemotherapy and approached that for hepatic resection.61,62 Obviously, the clinical experience with these modalities in the context of metastatic GIST is quite limited.
IMATINIB IN GIST
The signal transduction inhibitor imatinib exerts its activity in GIST through blockade of the adenosine triphosphate–binding site of KIT, a transmembrane receptor protein-tyrosine kinase. Normal activation of KIT depends on its ligand, stem cell factor (also called Steel factor or mast-cell growth factor), and is essential for maintenance of hematopoiesis, melanogenesis, and gametogenesis and development of mast cells and interstitial cells of Cajal (gastrointestinal tract pacemaker cells).12,63 Gain-of-function mutations in the KIT gene, found in nearly all GISTs, result in ligand-independent activation of the abnormal KIT protein, an early and pivotal event in GIST oncogenesis. In the small subset of GISTs apparently lacking KIT mutations (<5% to 10%), either high levels of KIT activation or alternative oncogenic activating mutations (e.g., in PDGFRA) may be involved.63,64 Imatinib inhibits proliferation and promotes apoptosis in GIST cells by interrupting tyrosine kinase–mediated intracellular signaling. More than 80% of patients with malignant metastatic or inoperable GIST have achieved a decrease in tumor burden of 50% or more (partial response) or had no disease progression in clinical studies of imatinib, for which at least 9 months of follow-up has been reported (Table 3).13,15,65 Rapid and dramatic results have been documented, including a 52% reduction in tumor size and a complete metabolic response (i.e., absence of abnormal tumor uptake of [18F]fluorodeoxyglucose) demonstrated by positron emission tomography scanning after 4 weeks of imatinib 400 mg/d.66,67 In the phase I trial, up to 800 mg of imatinib daily (given as 400 mg twice daily) was well tolerated. The most common side effects—rash, edema, diarrhea, nausea, and vomiting—were usually mild, were manageable, and tended to diminish over time; myelosuppression was uncommon.13 A phase II study, in which the two currently approved imatinib doses of 400 and 600 mg daily were used, similarly showed sustained objective tumor responses in more than half of patients and showed freedom from progression in an additional third.15 Eighty-eight percent of patients were alive after 1 year of treatment with imatinib (the median duration of survival is not yet defined). The side-effect profile was similar to that in the phase I study, except that gastrointestinal or peritoneal hemorrhage occurred in 5% of patients, probably as a result of imatinib-induced tumor necrosis.
|
EMERGING APPROACHES
The availability, for the first time, of a systemic therapy that has proven effective for the treatment of advanced GIST raises the issue of its possible use as an adjuvant or neoadjuvant agent in patients who are, or might be, candidates for resectional surgery. Currently, imatinib is approved for use only in patients with unresectable or metastatic malignant GIST. With respect to dose response, the ongoing phase III trials with a high-dose (800 mg/d) imatinib arm are indicating that some patients who crossed over to the higher dose from the initial 400-mg daily dose because of treatment failure or progression have experienced disease stabilization or response. A few of these patients remain on treatment 12 months or more after crossover.69,70 Adjuvant and neoadjuvant trials of imatinib are also under way (Table 4). Current and historical data suggest that certain high-risk GIST patients might warrant candidacy for imatinib therapy in conjunction with surgery.
|
Hohenberger et al.72 recently reported their experience of operating on 18 GIST patients with residual tumor masses after treatment with imatinib. This proved to be a major undertaking, involving procedures such as hepatic resection (n = 3), multivisceral resection (n = 10), and pelvic exenteration (n = 2). However, seven of eight patients in ongoing partial remission at the time of surgery achieved a histologically complete R0 resection, and of these, only one patient had disease progression at the time of reporting. The results in patients with progressive disease at the time of surgery were relatively poor. These data confirm the value of combining surgery with imatinib in responding patients.
Surgery and Imatinib: Preliminary Considerations
It is clear that most patients with malignant disseminated or unresectable GIST benefit from treatment with imatinib. The question remains as to defining profiles that can stratify patients into high-risk groups who have operative disease and may experience improved outcome with the addition of imatinib to surgical resection. Some evidence suggests that this may be feasible and can be approached through consideration of risk stratification and mutational status.
Risk Stratification
The consensus approach to evaluation of risk based on GIST size and mitotic rate is receiving validation in research studies.9 Significant correlations have been found between a tumor size of 5 cm or more and increased mitotic activity (generally, more than 5 mitoses per 50 HPFs), on the one hand, and an increased risk of adverse outcome (e.g., local recurrence, metastasis, and reduced survival) on the other.31,37,52,73 Surgical candidates with a primary GIST that has size and mitotic characteristics indicating probable aggressive behavior should be considered for enrollment in an adjuvant clinical trial of this agent, where available. The choice of management for patients undergoing resection of an intermediate-risk GIST is likely to present challenges. A recent population-based study that included 51 patients with intermediate-risk primary GISTs (<5 cm and 6–10 mitoses per 50 HPFs, or 5–10 cm and <5 mitoses per 50 HPFs) identified between 1983 and 2001 and treated surgically found that those patients had no tumor recurrences or metastases, no tumor-related deaths, and no differences in survival compared with the age- and sex-matched general population.73 Whether these preliminary findings will be confirmed at other centers and in larger groups of patients awaits further investigation.
The significantly increased risk of GIST spread and shortened survival in cases of perforation, tumor rupture, or incomplete resection supports a potential role for adjuvant imatinib administration in patients with these risk factors. Patients who have undergone surgery for primary GIST and return for resection of completely excisable locally recurrent or metastatic tumors may constitute another high-risk group in which the benefit from adjuvant imatinib therapy should be studied.
Mutational Status
Since the discovery of gain-of-function mutation of KIT as an important event in malignant transformation, increased interest has focused on potential associations between mutational status and outcomes. Evidence to date suggests that the presence of a KIT mutation per se is not a prognostic indicator. Activating KIT mutations have been found in approximately 90% of GISTs and seem to be acquired very early in the development of most of these tumors.74 Initial reports that patients with mutation-positive GISTs were more likely to have malignant disease and a worse prognosis than those with mutation-negative tumors have not been confirmed by subsequent investigations.75–77 Mutations of KIT have been found in GISTs across the risk spectrum, with no significant difference in the types or incidences of mutations in apparently benign and malignant GISTs.74,78–80
However, evidence of differences between the various types of KIT mutations may turn out to be relevant to prognosis and management in GIST. Singer et al.81 reported that patients whose GIST exhibited missense mutations of exon 11, which encodes the KIT juxtamembrane domain, had a 5-year recurrence-free survival rate of 89%, compared with 40% for patients with tumors demonstrating other KIT mutation types (P = .03). The most significant of the 4 independent predictors of disease-free survival in this study of 48 patients who underwent resection for their GIST (42 with total excision) were mixed spindle-cell/epithelioid histology (hazard ratio [HR], 21; P = .0001) and more than 15 mitoses per 30 HPFs (HR, 18; P = .0001), followed by a deletion/insertion mutation of exon 11 (HR, 4; P = .0006) and male sex (HR, 3; P = .05).
Exon 11 mutations account for approximately 70% of KIT mutations in GISTs. Cases involving exon 9, 13, or 17 (extracellular region and kinase domain mutations), as well as GISTs with no detectable mutation, are encountered infrequently.7,82 There are studies in addition to that of Singer et al.81 suggesting that a mutation in exon 9 or 13 is associated with an unfavorable prognosis, although the evidence with respect to exon 9 seems conflicting.82–84
It is clear, however, that advanced GISTs with exon 11 mutations show significantly better responses to imatinib therapy. In an examination of GISTs from 121 patients enrolled in a phase II trial of imatinib, a KIT exon 11 mutation was associated with a significantly higher rate of partial response to imatinib (72%) than either an exon 9 mutation (31.6%; P = .0033) or no detectable mutation (11.8%; P < .0001).85–88 The duration of response to imatinib was also longer in patients with an exon 11 mutation than in those without one. Despite these differences, in vitro analysis demonstrated that the kinase activity of all mutant forms of KIT was equally sensitive to inhibition by clinically relevant concentrations of imatinib.
The evidence of high rates of response to imatinib in GIST patients with a tumor harboring a KIT exon 11 mutation suggests a potential clinical trial design with imatinib as first-line therapy upon diagnosis of a GIST with this genotype before surgery. Optimization of both surgical management and imatinib therapy in such patients may offer the best opportunity for eradicating the disease, but as yet, the additional value of imatinib in this setting is unknown. In patients whose tumor is known to have other KIT mutations or no detectable mutation, an early and aggressive surgical approach, if feasible, would seem to be the preferred strategy. Recently, oncogenic activating mutations of PDGF-R were detected in 14 (35%) of 40 GISTs that lacked KIT mutations.64 Because the PDGF-R tyrosine kinases are also targets for inhibition with imatinib, the drug should not be withheld from patients whose GIST either seems to be KIT negative or expresses wild-type KIT. A clinical study correlating outcome with tumor genotype testing in GIST patients with metastatic or unresectable tumors, based on the initial phase II imatinib treatment trial, has been completed and will be reported in the near future.
Rendering GIST Resectable
The possibility of cure afforded by surgery provides a rationale for using imatinib neoadjuvantly for tumor debulking in cases, for example, of marginally resectable primary GIST. Imatinib treatment in patients who present with inoperable malignant GIST might enable them to undergo successful resection after a reduction in tumor size or spread. Most patients who responded to imatinib in the phase I and II clinical trials had a decrease of at least 50% in the size of their GIST.13,15 Pharmacologic debulking with imatinib may also be a possible neoadjuvant strategy to optimize the timing of surgery and avoid emergency operations, with the attendant risk of complications, in patients with large GISTs that predispose them to potentially life-threatening acute events, such as hemorrhage or tumor rupture. In some patients with multiple-site locally recurrent or metastatic GIST whose tumor showed a mixed pattern of response to imatinib therapy, surgical resection of unresponsive lesions has been performed.71
It is conceivable that if imatinib can improve the outcome of surgery, surgery might enhance the results of imatinib therapy. The extent to which strategies combining the use of imatinib and surgery in treating GIST are feasible in actual practice awaits elucidation in clinical trials.
For all patients receiving imatinib for any indication, maintenance of an optimal dosage is essential. A dosage increase to 600 or 800 mg/d should be considered for patients who show signs of GIST progression after an initial response to therapy at a recommended starting dosage of 400 or 600 mg/d. The optimal duration of treatment has not yet been defined, although at this time it seems to be prudent to continue imatinib therapy in the absence of a definitive demonstration, by a sensitive method, that there is no residual disease. There is evidence suggesting that early initiation of imatinib therapy may increase its effectiveness,50 underscoring the importance of prompt and accurate GIST diagnosis followed by timely implementation of appropriate management.
Preliminary Recommendations
Until the trial results of the adjuvant and neoadjuvant use of imatinib are analyzed and reported, the following suggestions may be helpful in guiding the current approach to the surgical management of GIST:
CONCLUSIONS
The introduction of imatinib has changed the paradigm for the management of GIST, a disease resistant to all radiotherapeutic and previously available systemic treatments. For patients with resectable GISTs, surgery continues to be the first-line treatment of choice. However, the curative potential of surgery is seldom realized in practice, because up to 90% of patients with GIST have a recurrence after complete resection and because only 65% of those with the best prognosis are alive 5 years after successful primary surgery. An important approach to the prolongation of survival in GIST is to improve the outcome of surgery. The demonstrated ability of imatinib to induce significant reductions in tumor size and to control overtly malignant metastatic disease in most patients with advanced GIST may expand the opportunities for surgeons to treat more GIST patients, decrease the incidence of postoperative GIST recurrence and spread, and thereby extend lives. Data from current and future trials of adjuvant and neoadjuvant imatinib therapy in patients undergoing GIST resection are expected to begin to shed light on some key questions: Can imatinib improve operability, reduce the morbidity of surgery, and increase progression-free and overall survival? Can we predict which surgical patients will benefit from adjuvant or neoadjuvant GIST administration? What would be the optimal dosage and duration of therapy?
When mesenchymal neoplasms are detected in the gastrointestinal tract and adjacent sites, it is important to consider the diagnosis of GIST and to ensure that an immunohistochemical assay for KIT (CD117) is performed in making the differential diagnosis. A GIST diagnosis should prompt immediate initiation of appropriate treatment: resection, systemic therapy with imatinib if metastatic or unresectable, or possibly both in the case of marginal resectability. In the future, risk profile and mutation type are likely to be taken into account in determining the role of adjuvant and neoadjuvant use of imatinib, depending on the results of the appropriate clinical trials. Perforation, tumor rupture, and incomplete resection are clearly associated with significantly reduced postoperative survival and should figure prominently in the assessment of risk. The possibility of genotyping GIST patients with reference to KIT mutation before consideration of treatment may enhance the physician’s ability to predict response to available therapies.
FOOTNOTES
Gastrointestinal Stromal Tumor (GIST) is a well defined pathological entity characterized by the expression of KIT. The success of imatinib in the management of metastatic GIST has prompted consideration for its use in the adjuvant and neoadjuvant setting.
Received for publication September 12, 2003. Accepted for publication January 28, 2004.
REFERENCES
This article has been cited by other articles:
 |
|
 |
|
  B. K. P. Goh, P. K. H. Chow, W.-M. Yap, S. M. Kesavan, I.-C. Song, P. G. Paul, B.-S. Ooi, Y.-F. A. Chung, and W.-K. Wong Which Is the Optimal Risk Stratification System for Surgically Treated Localized Primary GIST? Comparison of Three Contemporary Prognostic Criteria in 171 Tumors and a Proposal for a Modified Armed Forces Institute of Pathology Risk Criteria Ann. Surg. Oncol., August 1, 2008; 15(8): 2153 - 2163. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Rutkowski, Z. I. Nowecki, W. Michej, M. Debiec-Rychter, A. Wozniak, J. Limon, J. Siedlecki, U. Grzesiakowska, M. Kakol, C. Osuch, et al. Risk Criteria and Prognostic Factors for Predicting Recurrences After Resection of Primary Gastrointestinal Stromal Tumor Ann. Surg. Oncol., July 1, 2007; 14(7): 2018 - 2027. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Bonvalot, H. Eldweny, C. Le Pechoux, D. Vanel, P. Terrier, A. Cavalcanti, C. Robert, N. Lassau, and A. Le Cesne Impact of Surgery on Advanced Gastrointestinal Stromal Tumors (GIST) in the Imatinib Era Ann. Surg. Oncol., December 1, 2006; 13(12): 1596 - 1603. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. S. Benjamin, C. D. Blanke, J.-Y. Blay, S. Bonvalot, and B. Eisenberg Management of Gastrointestinal Stromal Tumors in the Imatinib Era: Selected Case Studies Oncologist, January 1, 2006; 11(1): 9 - 20. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M B Loughrey, C Mitchell, G B Mann, M Michael, and P M Waring Gastrointestinal stromal tumour treated with neoadjuvant imatinib J. Clin. Pathol., July 1, 2005; 58(7): 779 - 781. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
