INTRODUCTION — Stromal or mesenchymal neoplasms affecting the gastrointestinal (GI) tract have undergone a striking evolution in how they are perceived and classified over the last 30 years. A significant breakthrough occurred with the identification of near-universal expression of the CD117 antigen by these tumors (now called gastrointestinal stromal tumors [GISTs]). The other group of spindle cell neoplasms arising in the GI tract (which are analogous to soft tissue tumors throughout the rest of the body and include lipomas, schwannomas, hemangiomas, usual leiomyomas and the malignant counterpart, leiomyosarcomas) is typically CD117-negative [1].
The CD117 molecule is part of the KIT (c-kit) receptor, a membrane tyrosine kinase that is a product of the KIT protooncogene. In 80 percent of cases, KIT overexpression is the result of an activating mutation in the KIT protooncogene. Although the majority of GISTs are KIT positive, some KIT negative GISTs have activating mutations in a related tyrosine kinase receptor, platelet-derived growth factor receptor alpha (PDGFRA). The current view is that the overwhelming majority of GI tract mesenchymal tumors fall into the GIST category; they are identifiable by KIT immunoreactivity or the presence of activating mutations in KIT or PDGFRA. (See "Clinical presentation, diagnosis, and prognosis of gastrointestinal stromal tumors", section on 'Pathogenesis'.)
Prior to the year 2000, there was no known effective therapy for unresectable or metastatic GISTs. It has long been appreciated that GI tract sarcomas have lower response rates to chemotherapy than other sites of soft tissue sarcomas, indicating a higher rate of primary resistance to chemotherapy in these tumors [2-4].
Treatment of GISTs was revolutionized by the finding that mutational activation of KIT or PDGFRA stimulated growth of these cancer cells. This led to effective systemic therapies in the form of small molecule inhibitors of the receptor tyrosine kinases. Imatinib (Gleevec), the prototype drug, was originally approved for the treatment of chronic myeloid leukemia (CML), a disorder in which an aberrant tyrosine kinase results from molecular rearrangement. (See "Initial treatment of chronic myeloid leukemia in chronic phase".)
It subsequently became evident that molecularly targeted therapy with imatinib induced dramatic, rapid, and sustained clinical benefit in GISTs as well. These agents block signaling via KIT or PDGFRA by binding to the ATP-binding pocket required for phosphorylation and activation of the receptor. Other tyrosine kinase inhibitors (TKIs) have been identified that block several tyrosine kinase targets, including KIT (referred to as multitargeted TKIs). Some data suggest an anti-GIST immune response is associated with good clinical outcomes in patients with GIST on imatinib [5].
Here we will discuss TKI therapy for patients with advanced/metastatic GISTs. Molecular classification, clinical presentation, diagnosis, management of localized GISTS, the role of surgery in patients with metastatic disease, and the use of preoperative as well as adjuvant imatinib are addressed in separate topic reviews. (See "Clinical presentation, diagnosis, and prognosis of gastrointestinal stromal tumors" and "Local treatment for gastrointestinal stromal tumors, leiomyomas, and leiomyosarcomas of the gastrointestinal tract" and "Adjuvant and neoadjuvant therapy for gastrointestinal stromal tumors".)
INITIAL TREATMENT BASED ON MUTATION STATUS
Imatinib for KIT and most PDGFRA mutant GIST — Imatinib is used as initial therapy for patients with KIT and most platelet-derived growth factor receptor alpha (PDGFRA) mutant GIST, although GISTs associated with PDGFRA D842V are an exception (algorithm 1). (See 'Avapritinib for PDGFRA D842V mutant tumors' below.)
An assessment of tumor mutational status using DNA sequencing techniques is advised during the initial evaluation of patients with advanced or metastatic disease because clinical responses to imatinib (and other TKIs) correlate with tumor genotype. For most patients, systemic treatment with imatinib can be initiated empirically while awaiting confirmation of tumor mutational status. Treatment may be subsequently modified once tumor mutational status becomes available. However, for those with histologies suggestive of imatinib resistance (eg, succinate dehydrogenase [SDH] deficiency or neurofibromatosis 1 [NF1]-related disease), referral to a tertiary care center for clinical trials is warranted, rather than empiric imatinib. (See "Clinical presentation, diagnosis, and prognosis of gastrointestinal stromal tumors", section on 'Diagnosis' and 'SDH-deficient tumors and those associated with NF1' below.)
Efficacy — Since the initial report of a patient with rapidly progressive GIST who exhibited a dramatic and sustained response within one month of starting imatinib [6], many studies have confirmed the utility of this agent in advanced disease (table 1) [7-11]:
●In a phase II study, for example, 147 patients received either 400 or 600 mg of imatinib daily; 54 percent had radiographic responses documented within six months, with no significant difference between the two doses [8]. Positron emission tomography (PET) scanning proved to be a sensitive and reliable response indicator, with markedly diminished uptake, compared with baseline studies, seen in responders as early as 24 hours after the first dose of imatinib. (See 'Assessing response to therapy' below.)
Adverse events were common but generally mild. These included grade 1 or 2 nausea or diarrhea in approximately one-half, fluid retention (predominantly periorbital) in three-fourths, muscle cramps and fatigue (40 and 35 percent, respectively), and gastrointestinal (GI) or intra-abdominal hemorrhage in 5 percent. (See 'Side effects and their management' below.)
A sizeable subset of patients in this trial survived long-term on first-line imatinib. In a preliminary report of long-term follow-up of a cohort of 56 patients who continued to take imatinib beyond three years, 26 (18 percent of the initial cohort) remained on continuous imatinib since study entry at a median follow-up of 9.4 years [12]. The overall likelihood of remaining progression-free at nine years or beyond was dependent on tumor size at initial diagnosis and ranged from 29 percent in those with an initial tumor bulk of <39.1 mm2 to 3 percent in those with an initial tumor bulk >262.6 mm2. The corresponding rates of overall survival (OS) were 58 and 23 percent, respectively.
A similar fraction of patients with metastatic disease who achieved long-term disease control was reported in an analysis derived from the SWOG phase III S0033 trial, which studied two different dose levels of imatinib (see 'Impact of dose' below). In a preliminary report presented at the 2014 American Society of Clinical Oncology (ASCO) annual meeting, 180 of the 695 eligible patients (26 percent) survived eight years or longer, and the estimated 10-year survival rate was 22 percent [13]. Among the 137 long-term survivors, imatinib was the sole therapy administered continuously to 49 percent, while 39 percent received subsequent systemic agents including sunitinib and sorafenib.
●Following the introduction of imatinib, the median survival of patients with advanced GIST increased from an average of 18 to 57 months in the trial with the longest follow-up to date [14].
●Despite the high efficacy of imatinib in most patients with metastatic/inoperable GISTs, complete responses are rare overall (less than 10 percent), and most patients who initially respond eventually acquire resistance via additional mutations in KIT, discussed below. The median time to progression is approximately two to three years [8,10,15,16], although it is longer in some series [17]. A subset of patients with metastatic GIST experiences durable responses and long-term survival with imatinib treatment [18]. Factors influencing the duration of disease control are not well understood [16]. (See 'Treatment of disease refractory to imatinib' below.)
Assessing response to therapy — The optimal method for establishing response to tyrosine kinase inhibitors (TKIs) such as imatinib is evolving.
Positron emission tomography scanning — PET scans appear to identify a greater number of responses and at an earlier time than computed tomography (CT) scans [19-23]. With functional imaging such as PET, responses can be observed within 24 hours of starting therapy, although this has little practical impact on the typical patient with metastatic GIST. In addition to detecting primary imatinib resistance, PET can aid in the detection of secondary resistance, as does detection of the "nodule within a tumor" as a sign of drug resistance. (See 'RECIST versus Choi criteria' below.)
If employed, a rigorous standardized approach to patient preparation and study technique is needed for reproducible results, as emphasized by consensus guidelines on use of fluorodeoxyglucose (FDG)-PET for response assessment [24-26].
The added benefit for dual modality PET/CT imaging in most cases with advanced GIST is unclear, and PET scans are rarely used to assess the response to therapy in patients with advanced disease, unless in the context of a clinical trial. Newly proposed CT criteria using either no growth in tumor size or a combination of tumor density and size criteria have shown a close correlation with the predictive value results of FDG-PET, and serial CT or MRI scanning at three- to six-month intervals is more often recommended during therapy. (See 'Cross-sectional imaging' below and 'RECIST versus Choi criteria' below.)
One clinical scenario where baseline and follow-up PET scan might prove useful is for a patient with a borderline resectable GIST or a potentially resectable tumor that requires extensive organ disruption who is being treated with initial imatinib [22,27]. In these situations, early assessment of treatment response might provide an opportunity to shift to an alternative therapy (eg, resection or sunitinib) if imatinib is ineffective. (See "Adjuvant and neoadjuvant therapy for gastrointestinal stromal tumors".)
An important point is that patient symptoms also improve rapidly in the setting of response, decreasing to some degree the need to obtain imaging shortly after starting neoadjuvant imatinib.
Cross-sectional imaging — Conventional CT scans with intravenous contrast enhancement are a good standard of care for radiographic assessment of most patients with metastatic GIST. Contrast-enhanced magnetic resonance imaging (MRI) also provides similar information to that obtained with contrast-enhanced CT scans.
Certain finer points are worth keeping in mind based on the experience with metastatic GIST patients since the introduction of imatinib. For example, GISTs will infrequently increase in size during early treatment as a consequence of intratumoral hemorrhage or myxoid degeneration. A decrease in tumor density as seen on CT (the corollary of decreased FDG uptake on PET, see above) is an important early clinical marker of antitumor activity [28,29], as is decreased contrast uptake on standard CT or MRI scans. Once tumors become hypodense (cystic), the size of the lesions may decrease slowly and eventually stabilize.
Late responses are often seen in patients who initially have stable disease, and survival in those with stable disease parallels that of patients with an objective response [14,30]. The median time to achieve an objective response is four months, while maximal response may take six months or even longer [10].
As a result of these issues, response to imatinib is frequently defined as absence of progression at the time of the first formal disease re-evaluation (typically two to three months after starting therapy) [31]. Clear-cut evidence of progression at this time point is considered initial (primary) resistance, while progression or relapse after a period of stable or responding disease is referred to as late (secondary) resistance. Initial resistance was seen in 12 percent of 934 patients in a randomized European trial exploring two different doses of imatinib (discussed below) and was more likely in patients with lung but not liver metastases (41 percent) [15]. The management of patients with imatinib-refractory disease is discussed below. (See 'Treatment of disease refractory to imatinib' below.)
For patients receiving imatinib for advanced disease, the optimal interscan interval is not established. In keeping with guidelines from the National Comprehensive Cancer Network (NCCN) [27], we perform history and physical examination, and repeat cross-sectional imaging of the abdomen and pelvis every three to six months. For patients who have low-volume or no residual detectable metastatic disease who continue to have an excellent response to imatinib, we transition to every six months imaging after three years, and then to annual imaging after six or more years.
RECIST versus Choi criteria — As noted above, radiographic response to TKIs is often indicated by an early decrease in tumor density, followed by slow tumor regression. This pattern of response is not well suited to the use of standard Response Evaluation Criteria In Solid Tumors (RECIST), which is based upon tumor measurements (table 2) [32,33].
Likewise, disease progression in patients with GIST may also fail to be captured by standard RECIST. Tumor progression may manifest as new or enlarging tumor masses, as partial to complete filling-in of a previously hypodense lesion, or as a hyperdense "nodule-within-a-mass" pattern [34]. The importance of identifying this latter situation is that successful ablation of such resistant clones has been reported, with continued imatinib sensitivity of areas of disease that remain [35].
An alternative set of response evaluation criteria has been proposed, the so-called Choi criteria [36,37]. Investigators at MD Anderson initially showed that a 10 percent decrease in unidimensional tumor size or a 15 percent decrease in tumor density on contrast-enhanced CT scans (as reflected by differences in x-ray attenuation between a given material and water, expressed in Hounsfield units) correlates well with PET scan findings and is a better predictor of response to therapy (as judged by time to tumor progression) than standard RECIST [36].
These investigators compared the Choi criteria with standard (version 1.0 (table 2)) RECIST in 98 patients (40 in a training set and 58 a test set) receiving imatinib for advanced GIST who had CT scans eight weeks after starting therapy [37]. The test set had 28 (48 percent) good responders by RECIST, compared with 49 (84 percent) good responders by Choi criteria. Even when the 98 patients were analyzed together, the response group by RECIST did not correlate significantly with either disease-specific survival or time to tumor progression, whereas the Choi response group did correlate with both endpoints. The authors concluded that tumor response for GIST should preferentially be categorized by the Choi rather than the RECIST.
Impact of dose — At least two randomized trials have failed to show improved OS for higher imatinib doses (800 mg daily) versus standard dosing of imatinib (400 mg daily) [10,11]. In both trials, patients with disease progression in the standard-dose arm were allowed to cross over to high-dose therapy. The European trial demonstrated a modestly but significantly higher progression-free survival (PFS) with the 800 mg dose with median 25-month follow-up, which was no longer evident with longer follow-up (median 40 months), and which did not lead to a significant improvement in OS [10]. The American trial showed no advantage for higher dose therapy in terms of either PFS or OS [11]. Both trials indicated more side effects from higher-dose therapy.
A meta-analysis of both trials came to the following conclusions [38]:
●At a median follow-up of 45 months, compared with standard-dose imatinib, a modest PFS advantage was seen with higher dose imatinib (hazard ratio [HR] for progression 0.89, 95% CI 0.79-1), but OS and best response (51 versus 54 percent) were similar.
●The presence of a KIT exon 9 mutation was the only significant predictive factor for benefit from higher doses. Among the patients with an exon 9 mutation, compared with standard-dose imatinib, higher dose imatinib was associated with improved PFS (HR 0.58, 95% CI 0.38-0.91) and overall response rates (47 versus 21 percent), but not OS. In the absence of such mutations, no difference in these outcomes was observed between the treatment arms.
Pharmacokinetic variability — Another possible explanation for the failure to demonstrate benefit from higher imatinib doses in both trials is interpatient variability in pharmacokinetic exposure [39,40]:
●In a study of 73 patients who were randomly assigned to 400 or 600 mg of imatinib daily for advanced GIST, there was a 10-fold variance in trough levels with either dose (from 414 to 4182 ng/mL) [39]. Clinical outcomes were correlated with trough levels at steady state. Trough values below 1100 ng/mL were associated with a significantly shorter time to tumor progression and a lower rate of clinical benefit as compared with higher trough levels.
●Others have shown a significant correlation between low imatinib trough levels and prior major gastrectomy, higher creatinine clearance, and high serum albumin levels [40].
Pharmacokinetic variability may also contribute to acquired drug resistance [41]. In a small population-based pharmacokinetic study in patients with GIST, imatinib clearance increased after long-term treatment (>1 year), which reduced systemic exposure by 42 percent compared with the start of treatment [42]. These findings were confirmed in a subsequent prospective population pharmacokinetic study of 50 patents with GIST being treated with imatinib [43]. After 90 days of treatment, there was a significant decrease in imatinib exposure of 29 percent compared with baseline. However, there appeared to be no statistically significant effect of pharmacokinetic variability on PFS in the subset of patients treated with imatinib for advanced incurable disease.
It is not yet clear whether increased imatinib clearance is a significant factor in the amelioration of imatinib toxicity that occurs with time and/or has any impact on disease control; further work is required to ascertain the clinical implications of these observations.
Influence of mutations on response to therapy — An assessment of mutation status is advised for patients with metastatic disease because clinical responses to imatinib (and other TKIs) correlate with tumor genotype, which in turn impacts prognosis and treatment (algorithm 1). (See 'KIT mutations' below and 'PDGFRA mutations' below.)
Approximately 10 percent of patients with GIST have primary resistance to imatinib, defined as progression within the first six months of treatment. Many of these resistant tumors lack mutations in KIT or PDGFRA, or they harbor a PDGFRA D842V mutation. While we generally do not offer initial therapy with imatinib to these patients, the one notable exception is those with advanced spindle cell tumors of the GI tract. Imatinib may be effective in patients with these tumors, which appear histologically compatible with a diagnosis of GIST, even if they are immunohistochemically KIT negative, and these patients may be offered a trial of imatinib. (see "Clinical presentation, diagnosis, and prognosis of gastrointestinal stromal tumors", section on 'Other risk factors')
The general approach to patients with a D842V mutation or a succinate dehydrogenase (SDH)-deficient GIST is discussed below. (See 'Avapritinib for PDGFRA D842V mutant tumors' below and 'SDH-deficient tumors and those associated with NF1' below.)
KIT mutations — Imatinib is approved in the United States for the treatment of all KIT-expressing metastatic GISTs, regardless of mutation status. However, in our view and that of others [44], assessment of mutation status is advised for most patients being treated for metastatic disease, given the predictive and prognostic information that is provided [45]. Specific mutations in KIT (exon 11 versus exon 9) correlate with clinical response to imatinib (algorithm 1).
For patients with a KIT exon 9 mutation, consensus-based guidelines from the NCCN suggest initiating therapy for unresectable or metastatic disease with a higher dose of imatinib (800 mg daily) [27]. Similarly, the European Society for Medical Oncology (ESMO) recommends mutation testing for all patients and considers this higher dose of imatinib to be a standard treatment for patients with a KIT exon 9 mutation [44]. Although KIT exon 9 mutations are associated with an increased response rate to higher dose imatinib, OS is not improved. (See 'Impact of dose' above.)
In the United States, for patients where the KIT mutational status is unknown, one treatment approach is to start with imatinib 400 mg daily and increase the dose to 800 mg daily if there is no response. Initiation of second-line sunitinib is a reasonable alternative. Whether outcomes are better with early institution of sunitinib following progression on imatinib 400 mg daily is not known. (See 'Sunitinib' below.)
Data are as follows [46-50]:
●In one study of 127 patients with GISTs receiving imatinib, activating mutations in KIT and PDGFRA were found in approximately 88 and 5 percent, respectively [46]. All KIT mutant isoforms were associated with a response to imatinib, while only a subset of PDGFRA mutants was imatinib sensitive. (See 'PDGFRA mutations' below.)
Among the patients with KIT mutations, those with an exon 11 mutation had a substantially greater likelihood of a partial response compared with patients with either an exon 9 mutation or no detectable mutation in either KIT or PDGFRA (84 versus 48 and 0 percent, respectively), and they had a longer time to treatment failure as well.
●Data from two randomized trials (one from the United States Intergroup and another from the European Organisation for the Research and Treatment of Cancer [EORTC]), along with a meta-analysis of these two trials, suggest that higher daily doses of imatinib may preferentially benefit those with exon 9 mutations [38,49,51]. Data from the individual trials are discussed below, and data from the meta-analysis are reviewed above. (See 'Impact of dose' above.)
In the United States Intergroup trial comparing two doses of imatinib in 324 patients, patients whose tumors expressed an exon 11 mutant isoform were more likely to have an objective response to imatinib compared with those with an exon 9 isoform or those who had no kinase mutations (72 versus 44 and 45 percent, respectively) [49]. Patients with an exon 11 mutation also had a significantly longer time to disease progression (25 versus 17 and 13 months, respectively) and median OS (median 60 versus 38 and 49 months, respectively).
In the EORTC trial of 58 patients whose tumors expressed an exon 9 mutant KIT protein, an initial daily imatinib dose of 800 mg resulted in a significantly superior PFS (HR for progression 0.39) compared with 400 mg/day [51]. In contrast, the time to progression was not affected by the initial dose in patients with an exon 11 KIT mutation or wild-type KIT. There were no corresponding differences in OS between low-dose and high-dose initial therapy in patients with exon 9 mutations [51].
PDGFRA mutations — Platelet-derived growth factor receptor alpha (PDGFRA) mutations in exons 12, 14, and 18 occur in approximately 10 to 20 percent of patients with GIST, with a higher frequency among tumors that are KIT negative [48]. However, not all PDGFRA activating mutations demonstrate the same biologic response to imatinib. In particular, the D842V mutation in exon 18 confers relative primary resistance to imatinib (algorithm 1) [52-54].
By contrast, other PDGFRA mutations appear to render GIST tumors sensitive to imatinib [46,48,55,56]. As an example, in one observational study of approximately 60 patients with PDGFRA mutant GIST, the response rate to imatinib was 0 percent among those with a D842V substitution versus 39 percent for those with non-D842V PDGFRA mutations [52].
Further details and available treatment options for patients with advanced and metastatic GISTs that harbor the PDGFRA D842V mutation are discussed separately. (See "Clinical presentation, diagnosis, and prognosis of gastrointestinal stromal tumors", section on 'PDGFRA mutations' and 'Avapritinib for PDGFRA D842V mutant tumors' below.)
Duration of therapy — The optimal duration of imatinib therapy for responding patients with advanced/metastatic disease was addressed in a French trial that randomly assigned patients with advanced GIST and no disease progression after one year of imatinib to continuous treatment or interruption until disease progression [57]. The study was stopped prematurely after only 58 patients had been randomized when it became evident that the risk of progression was significantly higher if therapy was interrupted, even in completely responding patients. In the initial report of the 32 patients who interrupted therapy, 26 progressed and needed retreatment, compared with only 8 of 26 in the continuous therapy arm (81 versus 31 percent). The corresponding median PFS durations were 18 versus 6 months. There were no significant differences between the groups with respect to survival, incidence of imatinib resistance, or quality of life. Drug reintroduction achieved tumor control in 24 of 26 of the patients whose therapy was interrupted.
The study was subsequently amended to allow randomization after three or five years of therapy. In the final subset of 71 patients who remained free of progression after one (n = 32), three (n = 25), or five (n = 14) years of therapy and who were randomized to discontinue therapy, 51 patients had restarted imatinib upon documentation of progressive disease [58]. While 18 progressed only within known lesions, 33 (65 percent) had new lesions, with concurrent progression in the known lesions in 17. Only eight (42 percent) of the patients who had been in complete remission at randomization and 12 (52 percent) of the patients who had been in partial remission at randomization achieved a new complete or partial response, respectively, after reintroduction of imatinib.
Thus, interruption of imatinib results in rapid progression in most patients with advanced GIST, and it cannot be recommended unless there is significant toxicity or an intervening complication, such as the need to treat another diagnosis (eg, another cancer), in the same patient. Continuous therapy until disease progression (or lifelong if disease does not progress) is advised.
Side effects and their management — Imatinib is generally well tolerated; most side effects are less than grade 2, and the majority of patients can continue treatment without interruption. In general, the side effect profile tends to improve with prolonged therapy [59]. (See "Initial treatment of chronic myeloid leukemia in chronic phase", section on 'Imatinib' and "Initial treatment of chronic myeloid leukemia in chronic phase", section on 'Managing toxicity'.)
Minimal data have been published on managing the side effects of imatinib [45]. The most common side effects reported in GIST patients receiving imatinib are fluid retention, diarrhea, nausea, fatigue, muscle cramps, abdominal pain, and rash. These and other potentially more severe but less common side effects and proposed management strategies are summarized below.
●Fluid retention – Fluid retention with peripheral edema, and occasionally pleural effusion and ascites may be more common in older patients and in those with cardiac disease. Periorbital edema is more common and often does not respond to diuretics, but may improve with dietary salt restriction. It tends to be most prominent in the morning and decreases in intensity if the patient is upright during the day. (See "Ocular side effects of systemically administered chemotherapy", section on 'Imatinib'.)
In patients treated with imatinib in National Cancer Institute (NCI)-sponsored trials, the rate of grade 3/4 edema considered likely related to imatinib was 1.3 percent.
●Muscle cramps – Muscle cramps are perhaps the most bothersome long-term symptom associated with imatinib, most commonly affecting calves, feet, and hands. There is no definitive treatment, although anecdotally some patients have obtained benefit from treatment with calcium or magnesium supplements, or the use of quinine.
●Nausea, vomiting, dyspepsia – Nausea and vomiting are generally not a problem when the drug is taken with food, which does not diminish absorption. Dyspepsia and GI upset may also be lessened by taking the drug with meals. Symptomatic treatment with antacids or proton pump inhibitors may be needed.
●Abdominal discomfort and diarrhea – Flatulence and mild abdominal discomfort are common. Diarrhea is usually grade 1 to 2 but is occasionally more severe. Loose stools can usually be managed with loperamide or atropine sulfate/diphenoxylate hydrochloride.
●Rash – Skin rash is usually maculopapular and mild. It often resolves with continued treatment. Severe cases may respond to a short course of high-dose glucocorticoids [60]. Some patients who flare on redosing may benefit from desensitization, starting with doses as low as 25 mg daily, using cetirizine as premedication. (See "Cutaneous adverse events of molecularly targeted therapy and other biologic agents used for cancer therapy", section on 'Maculopapular eruption'.)
●Hematologic toxicity – Mild anemia is common with chronic use or higher doses [61]. Treatment with erythropoietin may benefit the occasional patient with more severe anemia. Macrocytosis, with elevation of the mean corpuscular volume, is observed frequently; the mechanism is unknown.
Neutropenia is rare. Patients may safely continue the drug as long as the absolute neutrophil count is ≥1000 cells/microL. Withholding the drug usually leads to resolution of neutropenia, often within days. Reinitiation of the same drug dose is often accomplished without reoccurrence; dose reduction may be needed if the patient continues to experience significant neutropenia.
●Hypophosphatemia – Hypophosphatemia is reported, especially with higher doses [62,63]. Routine monitoring of phosphate levels has been suggested by the manufacturer [64].
●Gynecomastia – Gynecomastia was reported in 7 of 38 men receiving imatinib for chronic myeloid leukemia (CML) and associated with reduced levels of free testosterone. Gynecomastia is not yet reported in men receiving imatinib for GIST.
●Lung and liver toxicity – Lung toxicity and elevation of liver function tests (LFTs) have only occasionally been reported. Fatal hepatotoxicity was noted in a patient in blast crisis who was taking large doses of acetaminophen concurrently, and it is recommended that acetaminophen use be avoided, if possible. At least some data suggest that concomitant administration of imatinib and corticosteroids may permit continued imatinib treatment in patients who develop LFT abnormalities [65].
●Cardiac toxicity – The issue of cardiac toxicity is unresolved. In 2006, there was a report of severe heart failure developing in 10 individuals receiving imatinib, although the disease for which treatment was being given was not specified [66]. Subsequent experience suggests that among patients receiving imatinib for CML, there is a risk for heart failure, but the risk is fairly low and probably limited to those with pre-existing heart disease. (See "Initial treatment of chronic myeloid leukemia in chronic phase", section on 'Imatinib'.)
Whether imatinib causes cardiac toxicity in patients treated for GIST is unclear; the following information on this issue is available:
•A report from the EORTC study of imatinib 400 versus 800 mg noted no excess of clinical cardiac events in the study population who were treated for a median of 24 months [67]. (See 'Impact of dose' above.)
•A retrospective review of the MD Anderson experience of 219 patients enrolled in clinical trials of imatinib for GIST over a six-year period identified potential cardiac adverse events in 18 (8 percent) [68]. The adverse events included dyspnea, chest pain, edema, pleural effusion, ascites, cardiac ischemia and arrhythmia. Thirteen had risk factors for coronary artery disease or established heart disease. None of the patients had evidence of pulmonary vascular congestion on chest radiography. Of the eight patients who underwent echocardiography or radionuclide ventriculography, only one had a left ventricular ejection fraction <50 percent. All patients were able to continue imatinib with dose adjustment and toxicity-specific management.
•In the landmark phase III placebo-controlled adjuvant imatinib trial, there was no excess of cardiotoxicity in the group that received one year of postoperative imatinib. (See "Adjuvant and neoadjuvant therapy for gastrointestinal stromal tumors", section on 'Phase III trials'.)
Until further information becomes available, the following actions would be prudent:
•As part of the informed consent process, patients taking imatinib should be notified that heart failure may be a rare but potentially serious adverse event.
•Patients currently taking imatinib should be monitored for signs and symptoms of left ventricular systolic dysfunction.
•Heart failure should be considered in the differential diagnosis of any patient experiencing peripheral edema while receiving this agent.
●GI bleeding – Patients with large bulky tumors have a 5 percent risk of tumor hemorrhage not associated with thrombocytopenia. These patients should be closely monitored for a drop in hemoglobin in the first four to eight weeks of therapy [45]. If the hemoglobin level acutely decreases ≥2 g/dL, imatinib should be temporarily withheld until blood counts have stabilized, with transfusion if needed. Emergent surgical intervention may be needed if bleeding does not resolve. However, emergency surgery is uncommonly needed. It was required in three of 232 patients in one report (1.3 percent) [69].
If life-threatening side effects, such as GI bleeding, develop that cannot be managed with maximal supportive care, a switch to sunitinib should be considered. (See 'Treatment of disease refractory to imatinib' below.)
●Ocular toxicity – In addition to periorbital edema, imatinib is associated with epiphora (excess tearing), spontaneous conjunctival/subconjunctival hemorrhage, and rare instances of retinal hemorrhage and optic neuritis. (See "Ocular side effects of systemically administered chemotherapy", section on 'Imatinib'.)
Exceptions
Avapritinib for PDGFRA D842V mutant tumors — For patients with symptomatic and/or rapidly progressive disease harboring a platelet-derived growth factor receptor alpha (PDGFRA) exon 18 D842V mutation, we recommend avapritinib over either imatinib or observation in the setting of initial therapy. These tumors often demonstrate primary resistance to imatinib, whereas the response rate with avapritinib is high (approximately 90 percent) (algorithm 1).
For those with the PDGFRA D842V mutation who exhibit asymptomatic and/or indolent disease, a period of observation is preferable to immediate therapy with avapritinib in order to avoid treatment-related toxicities, such as potential cognitive impairment. In these patients, avapritinib may be initiated at the onset of symptomatic and/or rapidly progressive disease. Ripretinib, another agent targeting the PDGFRA D842V mutation, is approved for patients who have progressed on multiple TKIs. (See 'Ripretinib' below.)
This is in contrast to the approach for patients with GISTs harboring a PDGFRA exon 18 mutation other than D842V. Although avapritinib is active for such patients, other agents (including imatinib, sunitinib, and regorafenib) also are effective, and avapritinib is typically reserved for use after progression on these therapies (algorithm 1). (See 'PDGFRA mutations' above and 'Treatment of disease refractory to imatinib, sunitinib, and regorafenib' below.)
Avapritinib is administered at 300 mg daily until disease progression, intolerance, or unacceptable toxicity occurs. Avapritinib is associated with central nervous system toxicities, including cognitive impairment; mood, speech, or sleep disorders; dizziness; hallucinations; and intracranial hemorrhage. Patients must be closely monitored for these central nervous system toxicities, with appropriate dose reductions or permanent discontinuation of avapritinib as clinically indicated [70]. Of note, it has been our experience that some cognitive impairment from avapritinib may persist despite dose modification or discontinuation of treatment, and further follow-up is needed to determine the duration and potential reversal of symptoms. (See "Neurologic complications of cancer treatment with molecularly targeted and biologic agents", section on 'Dasatinib, imatinib, and avapritinib'.)
Avapritinib is also teratogenic; therefore, females of reproductive potential or males with female partners of reproductive potential should use effective contraception while on therapy as well as for six weeks after the final dose is administered.
The efficacy of avapritinib was evaluated in a single-arm, open-label, phase I clinical trial (NAVIGATOR) [71,72]. In this study, a cohort of 56 patients with GIST containing the PDGFRA exon 18 D842V mutation received avapritinib at 300 or 400 mg daily. Patients with a D842V mutation were not required to have received prior systemic therapy. After a median follow-up of approximately 28 months, overall responses for those with the D842V mutation were seen in 51 patients (91 percent), including seven complete responders (13 percent) and 44 partial responders (79 percent). One-year PFS and three-year OS rates were 83 and 61 percent, respectively.
Grade ≥3 toxicities included anemia (28 percent), hyperbilirubinemia (9 percent), fatigue (9 percent), abdominal pain (6 percent), diarrhea (5 percent), cognitive impairment (5 percent), edema (2 percent), and pleural effusion (2 percent) [73].
Based on these phase I results, the US Food and Drug Administration (FDA) granted approval to avapritinib for patients with unresectable or metastatic GISTs with a PDGFRA exon 18 mutation [73]. Further randomized trials comparing avapritinib with other standard therapies are needed to confirm these results.
Limited data suggest that dasatinib and the investigational agent crenolanib (a TKI with specificity for PDGFRA, platelet-derived growth factor beta [PDGFRB], and fms-like tyrosine kinase 3 [FLT3]) may also be active against tumors with a PDGFRA D842V mutation [74-76]. These results require further investigation. (See 'Investigational therapies' below.)
SDH-deficient tumors and those associated with NF1 — Many tumors that lack mutations in KIT and PDGFRA (so-called wild-type tumors) are deficient in expression of one or more subunits of the genes that encode the proteins that form the succinate dehydrogenase (SDH) protein complex. SDH-deficient tumors have a high rate of primary resistance to various TKIs but a comparatively indolent course [77,78]. The best way to treat these patients is not established. Although they are refractory to imatinib, they may have some responsiveness to sunitinib or regorafenib. Referral to an expert center is preferred. These patients are also good candidates for clinical trials (algorithm 1). (See 'Sunitinib' below and 'Regorafenib' below.)
As an example, in one report of 95 patients with KIT/PDGFRA wild-type GIST (84 of which were SDH deficient), only 1 of 49 patients treated with imatinib had a partial response, and there were only four objective responses to sunitinib (one complete, three partial) [77]. Of the 63 patients with SDH mutant GIST followed for a median of six (range 1 to 44) years, only three had died (from 8 to 24 years after initial diagnosis); in contrast, 3 of 11 (27 percent) patients with SDH-competent wild-type GIST had died of progressive disease with a median follow-up of eight years (range 2 to 17). (See "Clinical presentation, diagnosis, and prognosis of gastrointestinal stromal tumors", section on 'KIT/PDGFRA wild-type GISTs'.)
Patients with neurofibromatosis type 1 (NF1), an inherited cancer predisposition syndrome that is caused by a mutation in the NF1 gene that encodes neurofibromin, have an increased risk of GIST. GISTs that arise in these patients may be KIT expressing, but they lack mutations in KIT and PDGFRA. These tumors rarely respond to imatinib [79]. (See "Clinical presentation, diagnosis, and prognosis of gastrointestinal stromal tumors", section on 'Pathogenesis'.)
ROLE OF SURGERY — The role of surgery in patients with metastatic disease is addressed separately. (See "Local treatment for gastrointestinal stromal tumors, leiomyomas, and leiomyosarcomas of the gastrointestinal tract", section on 'Role of surgery in patients with metastatic disease'.)
TREATMENT OF DISEASE REFRACTORY TO IMATINIB — Dose escalation is an option for patients with primary as well as secondary (late) resistance to standard doses of imatinib. Patients who are intolerant of imatinib are better served by switching to an alternative tyrosine kinase inhibitor (TKI), sunitinib.
Dose escalation of imatinib — Dose escalation may be considered in patients started on imatinib 400 mg daily who, after careful review of radiologic studies, are judged to have clear evidence of disease progression. The efficacy of this approach was shown in follow-up reports from both the American and European randomized dose-finding studies described above [11,80] (see 'Impact of dose' above):
●In the European study described above [10], 247 of the 473 patients randomly assigned to low-dose therapy progressed, and 133 were crossed over to higher dose therapy (400 mg twice daily) [80]. There were three confirmed partial responses, and 36 patients had prolonged periods of stable disease. Anemia and fatigue increased on the higher dose, but 18 percent were still alive and progression-free at one year.
●Similar results were noted in the American trial [11]. Of the 164 patients progressing on low-dose therapy, 133 crossed over to 800 mg daily. Following crossover, 31 percent of assessable patients (n = 117) achieved either an objective response (n = 3) or stable disease (n = 33). The median durations of progression-free survival (PFS) and overall survival (OS) after crossover were 5 and 19 months, respectively.
Increasing the dose is unlikely to benefit patients who progress rapidly (within two months) after starting therapy.
Alternative therapies — Emerging data suggest that selection of clones with secondary KIT mutations that cluster in the ATP-binding pocket and the activation loop of the kinase domain is the most common mechanism of secondary imatinib resistance [81-87]. Possible methods of overcoming imatinib resistance include resection of metastases to remove resistant clones, the use of more-powerful selective TKIs, multitargeted inhibitors, or combination therapies utilizing selective inhibitors of the tyrosine kinase receptor itself coupled with selective inhibitors of downstream pathways. In vitro studies suggest that the choice of TKI for imatinib-refractory GIST might depend, at least in part, on the specific mutation responsible for the acquisition of resistance [87,88]. However, these data require validation in human subjects before they can be applied to clinical practice.
Metastasectomy — Surgical resection may be considered in selected patients. In general, resection appears to benefit responding patients (ie, those who have a partial response, stable disease, or focal progression while receiving imatinib, and possibly those with isolated sites of progression) but has little to offer those who experience generalized or multifocal disease progression. The goal of metastasectomy is to halt disease progression by eliminating resistant clones. The appropriate timing of surgical intervention is not established. This subject is discussed in detail separately. (See "Local treatment for gastrointestinal stromal tumors, leiomyomas, and leiomyosarcomas of the gastrointestinal tract", section on 'Role of surgery in patients with metastatic disease'.)
Sunitinib
Efficacy — The multitargeted TKI sunitinib is active in imatinib-refractory or intolerant patients [89-95]. An international phase III trial of sunitinib versus placebo in 312 patients with refractory disease definitively established the role of sunitinib in this setting [92]. Patients demonstrating progression (by Response Evaluation Criteria In Solid Tumors [RECIST] version 1.0) while on placebo crossed over to the active treatment arm. In the latest update, at a median follow-up of 42 months, despite a low objective response rate in the sunitinib group (7 percent partial response), median time to tumor progression, the primary endpoint, was fourfold higher as compared with the placebo group (27 versus 6 weeks) [94]. Although survival was significantly better with sunitinib in the initial report [92], over time, as expected, OS converged (median 73 versus 65 weeks) [94]. The median number of weeks on treatment was 22.
As with imatinib, the clinical activity of sunitinib is significantly influenced by the specific mutation type [91,96]. This was shown in a phase I/II trial performed in 97 patients with metastatic, imatinib-refractory or intolerant GISTs [91]. Clinical benefit (partial response or stable disease for longer than six months) was significantly higher for those with a primary KIT 9 exon (58 percent) or wild-type KIT/platelet-derived growth factor receptor alpha (PDGFRA) mutation (56 percent) than for those with a KIT exon 11 mutation (34 percent). The same pattern was seen for PFS and OS. Following progression on imatinib, patients with KIT exon 9 mutation or a PDGFRA mutation had a median time to progression of 19 months, while for those with exon 11 mutations, it was only five months. There was also a correlation between secondary mutations and response to sunitinib. Both PFS and OS were significantly longer for patients with secondary KIT exon 13 or 14 mutations than for those with exon 17 or 18 mutations (7.8 versus 2.3 months).
As with imatinib, PET scans can permit an earlier assessment of response than conventional CT scans [90,97] but are rarely used outside of the context of a clinical trial, except in patients who cannot tolerate intravenous contrast or magnetic resonance imaging (MRI) as other means of evaluating metastatic GISTs radiologically. (See 'Assessing response to therapy' above.)
Sunitinib is approved in the United States for the treatment of imatinib-refractory or intolerant advanced GISTs. The approved dose is 50 mg daily for four of every six weeks. However, continuous daily dosing (37.5 mg daily) appears similarly safe and effective [98].
Resistance to sunitinib shares similar pathogenetic mechanisms to those identified in imatinib failure, with acquisition of secondary mutations after an extended initial response to the drug [99].
Side effects and management — Sunitinib can cause fatigue, nausea, vomiting, anemia, neutropenia, diarrhea, abdominal pain, mucositis, anorexia, hypothyroidism, and discoloration of skin and hair. Other less common toxicities include bleeding, fever, hypertension, hand-foot skin reaction, myelosuppression, proteinuria and other forms of renal toxicity, elevation in serum amylase and lipase, and reduced left ventricular ejection fraction/clinical heart failure. Most sunitinib-related toxicities can be managed symptomatically or with temporary withdrawal or dose reduction. In severe cases, the drug may need to be discontinued.
The following represents a brief synopsis of the most common toxicities seen with sunitinib:
●Renal toxicity – Small molecule TKIs that target the vascular endothelial growth factor receptors (VEGFR) are associated with proteinuria, which is rarely nephrotic range (>3.5 g/24 hours) and, even more rarely, associated with the nephrotic syndrome. Hypertension frequently accompanies proteinuria. The factors associated with both the occurrence and the severity of proteinuria in patients treated with sunitinib are incompletely characterized.
There are no published guidelines for managing albuminuria during sunitinib treatment; however, good clinical practice dictates obtaining a baseline urinalysis with periodic assessment of proteinuria. Treatment interruption is appropriate management for patients who develop moderate to severe proteinuria (≥3 g/24 hours). (See "Toxicity of molecularly targeted antiangiogenic agents: Non-cardiovascular effects", section on 'Proteinuria/nephrotic syndrome'.)
●Hematologic toxicity – Sunitinib may cause myelosuppression. Sunitinib should be held if the absolute neutrophil count is ≤1000 cells/mm3. Recurrent episodes of grade 3 or 4 neutropenia or thrombocytopenia should prompt dose reduction to 37.5 or 25 mg daily [45]. Anemia, if acute, should prompt interruption of sunitinib and a search for a source of bleeding or evidence for hemolysis. (See "Toxicity of molecularly targeted antiangiogenic agents: Non-cardiovascular effects", section on 'Myelosuppression'.)
Rarely, sunitinib has been associated with a thrombotic microangiopathy (TMA), which may include a thrombotic thrombocytopenic purpura (TTP) or hemolytic uremic syndrome (HUS)-type picture with microangiopathic hemolysis and renal failure. Hypertension is present in most cases. Withdrawal of sunitinib is crucial because this toxicity is potentially life-threatening. Other aspects of management are presented separately. (See "Drug-induced thrombotic microangiopathy (DITMA)", section on 'Management'.)
●Hypothyroidism – Hypothyroidism is a frequent complication of prolonged sunitinib therapy. In one series, 15 of 42 euthyroid patients with intact thyroid glands (36 percent) became hypothyroid (as defined as a persistently elevated level of thyroid stimulating hormone [TSH]) while receiving sunitinib for advanced GIST [100]. The risk increased with longer duration of therapy (18, 29, and 90 percent in patients treated for 36, 52, and 96 weeks, respectively). The mean time to development of hypothyroidism was 50 weeks. Because of the high prevalence of hypothyroidism, regular surveillance of TSH levels is warranted during sunitinib therapy. We and others [101] suggest that thyroid function be evaluated at baseline and monitored at monthly intervals. (See "Toxicity of molecularly targeted antiangiogenic agents: Non-cardiovascular effects", section on 'Thyroid dysfunction' and "Diagnosis of and screening for hypothyroidism in nonpregnant adults" and "Treatment of primary hypothyroidism in adults".)
Another potential mechanism for hypothyroidism that should be considered in patients being treated with antiangiogenic TKIs for advanced GIST is consumptive hypothyroidism due to excessive degradation of thyroid hormone. However, this appears to be caused by overexpression of the thyroid hormone inactivating enzyme type 3 iodothyronine deiodinase (D3) within large GISTs and is not a drug-related adverse effect. This subject is discussed in detail separately. (See "Disorders that cause hypothyroidism", section on 'Consumptive hypothyroidism' and "Clinical presentation, diagnosis, and prognosis of gastrointestinal stromal tumors", section on 'Clinical presentation'.)
●Hypertension and cardiac toxicity – Patients should be closely monitored for hypertension and evidence of cardiac dysfunction while receiving sunitinib. (See "Toxicity of molecularly targeted antiangiogenic agents: Cardiovascular effects", section on 'Hypertension' and "Toxicity of molecularly targeted antiangiogenic agents: Cardiovascular effects", section on 'Left ventricular dysfunction and myocardial ischemia'.)
The frequency of these events in patients treated with sunitinib for GIST can be illustrated by the following:
•In a study of 75 patients treated with sunitinib for imatinib-resistant GISTs, 47 percent developed hypertension (defined as a blood pressure >150/100 mmHg) during therapy [102]. In addition, 10 of 36 patients (28 percent) who were serially assessed had a ≥10 percent decline in left ventricular ejection fraction (LVEF) during therapy, and clinical heart failure developed in six (8 percent). Heart failure and decreased LVEF generally resolved with medical management and withdrawal of sunitinib.
•Somewhat lower rates of treatment-related hypertension (20 percent, any grade), decreased ejection fraction (8 percent), and left ventricular dysfunction (2 percent) were reported in the longitudinal report of the phase III trial of sunitinib versus placebo discussed above [94].
●Gastrointestinal (GI) bleeding and/or bowel perforation – Patients who are treated with sunitinib for advanced GIST can develop GI bleeding or a bowel perforation. In one series, emergency surgery for hemorrhage, tumor perforation, or abscess was required in 4 of 43 patients during second-line therapy with sunitinib [69]. (See "Toxicity of molecularly targeted antiangiogenic agents: Non-cardiovascular effects", section on 'Bleeding' and "Toxicity of molecularly targeted antiangiogenic agents: Non-cardiovascular effects", section on 'Intestinal perforation/fistula formation'.)
●Potential for delayed wound healing – Impaired wound healing (and reopening of previously healed wounds) has been observed following treatment with all TKIs that target VEGF, including sunitinib. However, at least some data support the view that wound healing complications are not more common after extensive GIST resections in patients on sunitinib as compared with imatinib [103]. (See "Toxicity of molecularly targeted antiangiogenic agents: Non-cardiovascular effects", section on 'Delayed wound healing'.)
Because of the potential for wound healing problems, many clinicians suggest interruption of therapy for at least one week before surgery, and not reinitiating until adequate wound healing has occurred. However, given that these tumors can progress quite rapidly off therapy, it is our practice to continue sunitinib until three to four days prior to surgery and to resume the drug at the first postoperative visit. (See "Toxicity of molecularly targeted antiangiogenic agents: Non-cardiovascular effects", section on 'Delayed wound healing'.)
Is there a role for ripretinib? — We do not offer ripretinib to patients with disease refractory to imatinib. In a randomized phase III trial, ripretinib was better tolerated but did not improve PFS over sunitinib [104]. Ripretinib is indicated for disease refractory to multiple TKIs (imatinib, sunitinib, and regorafenib). (See 'Ripretinib' below.)
Ripretinib was evaluated in a randomized phase III trial (INTRIGUE) of 453 patients with advanced GISTs who experienced disease progression on or were intolerant to imatinib [104]. In this study, ripretinib resulted in similar PFS compared with sunitinib in those with a KIT exon 11 mutation (median 8.3 versus 7 months, HR 0.88, 95% CI 0.66-1.16) and among the entire study population (median PFS 8 versus 8.3 months, HR 1.05, 95% CI 0.82-1.33). Compared with sunitinib, ripretinib had higher objective response rates (24 versus 15 percent) and lower grade ≥3 treatment-related toxicities (27 versus 55 percent).
TREATMENT OF DISEASE REFRACTORY TO IMATINIB AND SUNITINIB
Regorafenib — For patients with imatinib- and sunitinib-refractory GIST, we recommend regorafenib rather other tyrosine kinase inhibitors (TKIs), as this approach improved PFS in a randomized phase III trial [105]. Additionally, regorafenib may confer greater benefit among such patients harboring mutations in KIT exon 11 or succinate dehydrogenase (SDH) [106].
Regorafenib is an orally active TKI that is structurally similar to sorafenib and targets a variety of kinases including KIT, platelet-derived growth factor receptor (PDGFR), and VEGFR.
Based on data from phase II trials [106,107], regorafenib was evaluated in a double-blind, placebo-controlled phase III trial (GRID) conducted in 199 patients with metastatic or unresectable GIST refractory to or intolerant of sunitinib. Patients were randomly assigned to best supportive care plus either regorafenib (160 mg once daily for three of every four weeks) or placebo [105]. Compared with placebo, regorafenib improved progression-free survival (PFS; 4.8 versus 0.9 months, HR 0.27, 95% CI 0.19-0.39) but not overall survival (OS; HR 0.77, 95% CI 0.42-1.41). The lack of OS benefit was likely due to the crossover design, as a majority (85 percent) of patients in the placebo group received regorafenib upon disease progression.
Grade ≥3 toxicities for regorafenib included hypertension (23 percent), hand-foot skin reaction (20 percent), and diarrhea (5 percent). Treatment-related toxicity with regorafenib is discussed in more detail separately. (See "Toxicity of molecularly targeted antiangiogenic agents: Non-cardiovascular effects" and "Toxicity of molecularly targeted antiangiogenic agents: Cardiovascular effects".)
Based on these data, the US Food and Drug Administration (FDA) approved regorafenib for treatment of metastatic or unresectable GIST no longer responsive to imatinib and sunitinib.
TREATMENT OF DISEASE REFRACTORY TO IMATINIB, SUNITINIB, AND REGORAFENIB
Ripretinib — For patients with GIST (irrespective of D842V mutation status) who have progressed on or are intolerant of three or more tyrosine kinase inhibitors (TKIs), including imatinib, we suggest ripretinib over other available TKIs, as this agent improved survival and was well tolerated in a placebo-controlled randomized trial [108].
Ripretinib is administered at an initial dose of 150 mg daily until disease progression, intolerance, or unacceptable toxicity occurs [109].
Management considerations specific to use of ripretinib are as follows:
●The use of concomitant strong and moderate CYP3A inducers (table 3) with ripretinib should be avoided.
If the concomitant use of moderate CYP3A inducers cannot be avoided, the ripretinib dose should be increased to 150 mg twice daily during the co-administration period, with monitoring for efficacy and toxicity. Ripretinib can be resumed at 150 mg daily two weeks after the moderate CYP3A inducer is stopped [110]. (See "Drugs and the liver: Metabolism and mechanisms of injury".)
●We obtain a baseline echocardiogram or MUGA scan prior to initiation of therapy and as clinically indicated afterwards; ripretinib should be discontinued in those who develop grade ≥3 left ventricular diastolic dysfunction on treatment. (See "Toxicity of molecularly targeted antiangiogenic agents: Cardiovascular effects", section on 'Specific VEGFR tyrosine kinase inhibitors' and "Cardiotoxicity of cancer chemotherapy agents other than anthracyclines, HER2-targeted agents, and fluoropyrimidines", section on 'Ripretinib'.)
●Patients should receive regular dermatologic evaluation, as ripretinib is associated with the development of primary cutaneous malignancies (such as squamous cell carcinoma and melanoma), alopecia, and hand-foot skin reaction syndrome. (See "Hand-foot skin reaction induced by multitargeted tyrosine kinase inhibitors".)
●Patients of reproductive potential and their partners should use effective contraception during therapy and for at least one week after treatment completion.
●Ripretinib should be held for at least one week prior to elective surgeries and at least two weeks after major surgery to reduce the risk of wound healing complications.
●The approach to monitoring and managing hypertension in patients treated with ripretinib is similar to that used with antiangiogenic agents and is discussed separately. (See "Toxicity of molecularly targeted antiangiogenic agents: Cardiovascular effects", section on 'Hypertension'.)
Ripretinib, a "switch control" inhibitor, targets multiple molecular alterations present in GIST, including several KIT mutations (exon 9, 11, 13, 14, 17, and 18), by stabilizing the KIT molecule into an inactive form. Ripretinib also targets PDGFRA mutations in exon 18, including the D842V resistance mutation, and the D816V secondary resistance mutation in exon 17, which is also present in most patients with systemic mastocytosis. (See "Advanced systemic mastocytosis: Management and prognosis".)
In a phase III trial (INVICTUS), 129 patients with advanced GIST either refractory to or intolerant of imatinib, sunitinib, and regorafenib were randomly assigned to ripretinib or placebo [108]. At median follow-up of 6 months, ripretinib improved median overall survival (OS; 15 versus 6 months, hazard ratio [HR] 0.36, 95% CI 0.20-0.62) relative to placebo. Ripretinib also improved progression-free survival (PFS; six versus one month; six-month PFS 51 versus 3 percent, HR 0.15, 95% CI 0.09-0.25) in all patients regardless of mutation status [108] and in each subgroup with specific KIT mutations (exons 11, 9, 13, and 17) relative to placebo [111]. Objective response rates were also higher for ripretinib compared with placebo (9 versus 0 percent). Ripretinib was well tolerated, as the number of patients who discontinued or interrupted treatment due to adverse events was similar between ripretinib and placebo. Grade ≥3 treatment-related toxicities included anemia (9 percent), abdominal pain (7 percent), hypertension (7 percent), and decreased ejection fraction (3 percent). The most common side effect was alopecia (52 percent), which is unique to ripretinib relative to other TKIs used for GIST.
Based on these data, ripretinib is approved by the US Food and Drug Administration (FDA) for patients with advanced GIST who have received three or more TKIs, including imatinib [110].
Alternative targeted agents — For patients who are ineligible for or intolerant of ripretinib, alternative options include nilotinib, sorafenib, pazopanib, or pimitespib (where available). Avapritinib is an option for those with the PDGFRA D842V mutation who have progressed on imatinib and sunitinib.
Limited data are available on the efficacy of these alternative TKIs to ripretinib for imatinib- and sunitinib-refractory GISTs [75,112-122]. These agents have not been directly compared, and none are specifically approved for refractory disease. The choice between these agents is based on prior therapy, tumor mutational status, pharmacokinetics, tolerance, as well as patient and clinician preference. Patients previously treated with imatinib, sunitinib, and regorafenib are more likely to progress and have less durable treatment responses when these alternative agents are used beyond third-line therapy. Therefore, their use is best considered a "holding strategy," and clinical trial enrollment is encouraged, where available.
Nilotinib — For frail patients who poorly tolerated imatinib, sunitinib, or regorafenib therapy, nilotinib may be better tolerated and, consequently, more likely to be administered at therapeutic doses. Although limited, the available data suggest that nilotinib is not an effective agent in patients whose tumors have a KIT exon 9 mutation [123].
Nilotinib is a more potent second-generation TKI that targets KIT and PDGF receptors; it is not more efficacious than imatinib as first-line therapy [123], although efficacy in imatinib-refractory cases has been shown in several reports [116,117,124]. Benefit in patients refractory to both imatinib and sunitinib is less certain [118,119,122]. In a phase III trial, 248 patients with advanced GIST following prior imatinib and sunitinib failure were randomly assigned to nilotinib (400 mg twice daily) or best supportive care with or without imatinib or sunitinib [122]. Compared with best supportive care, nilotinib improved PFS based on local investigator assessment (median 119 versus 70 days) but not when it was based on blinded central radiology review (median 109 versus 111 days).
As has been shown with imatinib, prior gastrectomy may result in markedly lower bioavailability of a variety of other TKIs such as nilotinib [125]. (See 'Pharmacokinetic variability' above.)
Sorafenib — For those with poor tolerance of regorafenib alone, sorafenib may be helpful, given the similarity of both agents. The half-life of sorafenib is shorter than that of regorafenib, enabling more rapid resolution of toxicity and facilitating titration to the patient's own maximally tolerated dose. However, for those with disease progression on regorafenib, sorafenib is unlikely to be helpful.
The efficacy of sorafenib (a TKI that inhibits KIT, vascular endothelial growth factor receptor [VEGFR], and platelet-derived growth factor receptor beta [PDGFRB]) was addressed in a multicenter phase II trial involving patients with either imatinib (n = 6) or imatinib and sunitinib-refractory (n = 32) GIST [113]. In preliminary results, the disease control rate was 68 percent, and median PFS was 5.2 months. The most common grade 3 toxicities were hand-foot syndrome (45 percent) and hypertension (21 percent).
Pazopanib — For those with KIT wild-type GIST, pazopanib may be helpful, as it is a potent inhibitor of vascular endothelial growth factor (VEGF) signaling.
Pazopanib has activity in patients with imatinib and sunitinib-refractory disease [120,121]. The efficacy of pazopanib, compared with best supportive care (BSC) alone, was addressed in a small randomized phase II PAZOGIST trial conducted in 81 patients who were refractory to imatinib and sunitinib [121]. Compared with BSC alone, pazopanib improved PFS (median 3.4 versus 2.3 months, HR 0.59, 95% CI 0.37-0.96). Although there were no objective responses, 84 percent had stable disease (compared with 70 percent with BSC only). However, toxicity rates were high; among the 76 patients who eventually were treated with pazopanib (which included 36 who crossed over from the BSC group), 72 percent experienced grade ≥3 adverse events, including serious adverse events (eg, pulmonary embolism in 9 percent).
Avapritinib — Avapritinib is an option for those patients with a PDGFRA D842V mutation who have progressed on imatinib and sunitinib. In this subset of patients, avapritinib improved PFS and objective response rates (ORR) compared with regorafenib in a phase III trial [126]. For other patients with a PDGFRA exon 18 mutation other than D842V, we reserve the use of avapritinib for those with disease that has progressed on all three agents (imatinib, sunitinib, and regorafenib). For all other patients with such treatment-refractory disease lacking a PDGFRA exon 18 mutation, including D842V, we do not suggest the use of avapritinib because it confers no additional clinical benefit compared with regorafenib and is associated with central nervous system toxicity.
In an open-label phase III trial (VOYAGER), 476 patients with unresectable or metastatic GIST previously treated with imatinib and one or two additional TKIs were randomly assigned to either avapritinib or regorafenib [126]. A majority of patients (95 percent) also previously received sunitinib, but none received regorafenib or more than three different TKIs. In the entire study population, compared with regorafenib, avapritinib improved ORR (17 versus 7 percent) but had similar PFS (median 4.2 versus 5.6 months, HR 1.25, 95% CI 0.99-1.57), disease control rates (42 versus 46 percent), and overall survival (OS) at one year (68 versus 67 percent). Among the 13 patients with PDGFRA D842V-mutant tumors, avapritinib improved PFS over regorafenib (median not reached versus 4.5 months) and ORR (43 versus 0 percent). In contrast, for patients without a PDGFRA D842V mutation, regorafenib improved PFS compared with avapritinib (median 3.9 versus 5.6 months, HR 1.34, 95% CI 1.06 to 1.69). The rates of grade ≥3 treatment-related toxicity were similar between the two treatment arms (55 versus 58 percent).
The efficacy of avapritinib as initial therapy in those with the PDGFRA D842V mutation is discussed above. (See 'Avapritinib for PDGFRA D842V mutant tumors' above.)
Pimitespib — Pimitespib, a heat-shock protein 90 inhibitor, is another effective option (where available) that improves PFS and OS in patients with GIST refractory to multiple TKIs.
In a randomized, double-blind phase III trial (CHAPTER-GIST-301) of 86 patients with advanced GIST refractory to imatinib, sunitinib, and regorafenib, pimitespib improved PFS (median 3 versus 1 months, HR 0.51, 95% CI 0.30-0.87) and cross-over-adjusted OS (median 14 versus 8 months, HR 0.42, 95% CI 0.21-0.85) compared with placebo [127]. Grade ≥3 toxicities included diarrhea (14 percent), anemia (5 percent), renal impairment (3 percent), and decreased appetite and malaise (2 percent each).
Pimitespib is available in Japan, where it has regulatory approval for the treatment of progressive GIST [128].
Other agents
●Ponatinib – Ponatinib is an oral TKI with potent preclinical activity against primary KIT exon 11 mutations and clinically important secondary resistance mutations, including A loop mutations that are resistant to imatinib and sunitinib [129]. In a phase II study of ponatinib after failure of standard approved TKIs, among patients with a primary KIT exon 11 mutation, 11 of 22 (50 percent) had stable disease or better at 16 weeks [130]. For patients with a primary KIT exon 9 mutation, 3 of 11 (27 percent) had stable disease or better at 16 weeks. This was a heavily pretreated patient group; 74 percent had four or more prior regimens.
Patients considered for ponatinib should be carefully selected because of the risk of serious arterial thrombotic events that accumulate over time, as has been seen in patients with chronic myelogenous leukemia treated with ponatinib [131]. These findings led to more restrictive eligibility criteria and a black box warning from the FDA. (See "Toxicity of molecularly targeted antiangiogenic agents: Cardiovascular effects", section on 'VEGFR tyrosine kinase inhibitors'.)
●Dasatinib – Dasatinib is an active agent as well in this setting [75,112,132]. In a phase II study of dasatinib (70 mg orally twice daily) in 50 patients with advanced GIST who were refractory to imatinib, 80 percent of whom were also refractory to sunitinib, 12 had a partial response by Choi criteria, and this was sustained for at least eight weeks in five [75]. Among patients with a mutation in KIT exon 11, median PFS and OS were 2.7 and 19.6 months, respectively; median PFS for those with a mutation in PDGFRA exon 18 was 22 months, and median OS was not reached. It should be noted, however, that the Choi criteria have not been validated in patients treated with any TKI beyond imatinib, and none of the above phase II trials of any TKI other than imatinib have been evaluated using Choi criteria. (See 'RECIST versus Choi criteria' above.)
Imatinib or sunitinib rechallenge — For patients who are refractory to multiple TKIs (including imatinib, sunitinib, and regorafenib) imatinib or sunitinib rechallenge is a preferred strategy over discontinuing a TKI altogether, even in the face of worsening disease burden. It is hypothesized that retreatment with imatinib or sunitinib provides inhibition of the bulk of disease clones that retain sensitivity to either drug; however, the relatively short duration of benefit suggests that TKI-resistant clones continue to progress during treatment. Data that support a modest benefit for imatinib or sunitinib retreatment are as follows:
●A Korean trial randomly assigned 81 patients with advanced GIST following failure of all available TKIs to imatinib rechallenge (400 mg once daily) or placebo [133]. Median PFS was significantly greater for those patients who received imatinib (1.8 versus 0.9 months), and the disease control rate at 12 weeks was 32 percent (versus 5 percent in the placebo group). With 93 percent of the placebo patients crossing over to active treatment, median OS was similar in both groups (8.2 versus 7.5 months in the imatinib and placebo groups, respectively).
●An Italian retrospective of 82 patients with advanced GIST previously treated with imatinib, sunitinib, or regorafenib included 74 who were retreated with imatinib (dose 400 mg daily) and eight who were retreated with sunitinib (dose individualized according to clinician choice) [134]. In preliminary results available in abstract form, the median time to progression in retreated patients was 5.4 months, and OS was 10.4 months. A correlation between mutational status and response rate, time to progression, or OS was not found.
INVESTIGATIONAL THERAPIES — Clinical trials may be offered to both patients with treatment-naïve disease and those who have progressed on standard therapies. Patients who are interested in clinical trials should be referred to centers of excellence specializing in the treatment of GIST.
●For previously untreated patients, investigational agents include the combination of imatinib plus binimetinib, an MEK inhibitor [135].
●For patients with treatment-refractory GIST, investigational agents include immunotherapy, TKIs [136], and combination therapies that exploit resistance pathways [137].
ROLE OF RADIATION THERAPY — For patients with progressive symptomatic soft tissue metastases despite treatment with multiple TKIs, radiation therapy (RT) is a reasonable option, although response rates are limited for this approach.
Given the limited duration of benefit for second-line treatment with a TKI (approximately five to six months), most patients with advanced GIST will eventually require palliative care. Current treatment guidelines do not address radiotherapy as a therapeutic option or consider it only for palliation of the rare patient with painful bone metastases [44,138]. GIST has been historically considered radiotherapy resistant or minimally responsive [139-142], although selected patients may benefit [143-145]. A prospective study of 25 patients progressing at intra-abdominal soft tissue or liver sites during or after TKI therapy demonstrated a low objective response rate to radiotherapy (only two [8 percent] achieved a partial remission), but 20 (80 percent) had stable target lesion size for three or more months (median duration of stabilization 16 months), and this was fourfold longer than the median time to GIST progression at any site.
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Gastrointestinal stromal tumors".)
INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5th to 6th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10th to 12th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.
Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)
●Basics topics (see "Patient education: Soft tissue sarcoma (The Basics)")
SUMMARY AND RECOMMENDATIONS
●The majority of mesenchymal neoplasms affecting the gastrointestinal (GI) tract are gastrointestinal stromal tumors (GISTs). Approximately 95 percent of these tumors overexpress KIT (which can be identified by immunohistochemical staining), and 85 percent have mutations in the KIT protooncogene that leads to constitutive activation of KIT, a receptor tyrosine kinase. A subset of GISTs that lack KIT mutations has activating mutations in a related receptor tyrosine kinase, platelet-derived growth factor receptor alpha (PDGFRA). (See 'Introduction' above.)
●An assessment of mutation status in individual GISTs is advised for most patients being treated for metastatic disease because of the prognostic and predictive information that can be derived from such an assessment. (See 'Influence of mutations on response to therapy' above.)
●For patients with advanced unresectable GIST, we recommend an orally active tyrosine kinase inhibitor (TKI) as initial therapy (algorithm 1) (Grade 1A).
•For patients with KIT exon 11 mutations, initial treatment is with imatinib, typically at 400 mg daily. However, for those with KIT exon 9 mutations, we suggest imatinib at 800 mg daily rather than 400 mg daily (Grade 2C). (See 'KIT mutations' above.)
•For patients with a PDGFRA exon 18 D842V mutation and symptomatic and/or rapidly progressive disease, we suggest initial treatment with avapritinib rather than other TKIs or observation (Grade 2C). However, for those with asymptomatic and/or indolent disease, a period of observation is preferable to immediate therapy with avapritinib to avoid treatment-related toxicities, such as potential cognitive impairment. Although avapritinib has not been directly compared with other TKIs, these tumors often demonstrate primary resistance to imatinib. (See 'Avapritinib for PDGFRA D842V mutant tumors' above.)
•For patients with a PDGFRA mutation other than D842V, we suggest initial treatment with imatinib rather than other agents, given the recognized imatinib sensitivity of this subtype of GIST. (See 'PDGFRA mutations' above.)
•For patients with succinate dehydrogenase (SDH)-deficient tumors, we refer patients for clinical trials, as these tumors are minimally responsive to imatinib and other agents. For those who are ineligible or who decline such trials, we offer sunitinib or regorafenib, as these agents have some limited efficacy in this setting. (See 'SDH-deficient tumors and those associated with NF1' above.)
●The optimal means of response assessment in patients receiving TKIs is unclear. Radiographic response is often indicated by an early decrease in tumor density on contrast-enhanced computed tomography (CT) scan, followed by slow tumor regression. This pattern of response is not well suited to the use of standard response criteria. Because of this, TKIs should be continued indefinitely in the absence of overt radiographic progression (Grade 1B). (See 'Assessing response to therapy' above and 'Duration of therapy' above.)
Although positron emission tomography (PET) scans appear to identify a greater number of responses and at an earlier time than CT, PET is not routinely indicated for response assessment. (See 'Positron emission tomography scanning' above.)
However, one clinical scenario where baseline and follow-up PET scan might prove useful is for a patient with borderline resectable GIST or a potentially resectable tumor that requires extensive organ disruption who is being treated with initial imatinib. Early assessment of treatment response provides the opportunity to shift to an alternative therapy (eg, resection or sunitinib) if imatinib is ineffective. (See "Adjuvant and neoadjuvant therapy for gastrointestinal stromal tumors", section on 'Neoadjuvant therapy'.)
●The role of surgery in patients with locally unresectable, nonmetastatic GISTs and in those with potentially resectable metastatic disease who respond to imatinib is evolving. This approach has little to offer patients with extensive disease progression while receiving imatinib. (See "Local treatment for gastrointestinal stromal tumors, leiomyomas, and leiomyosarcomas of the gastrointestinal tract", section on 'Role of surgery in patients with metastatic disease'.)
●Imatinib does not appear to benefit non-GIST GI mesenchymal tumors, which are treated similarly to other advanced soft tissue sarcomas. (See "Surgical treatment and other localized therapy for metastatic soft tissue sarcoma".)
●For patients who are intolerant of imatinib or who become refractory to treatment, we recommend a trial of the multitargeted TKI sunitinib (Grade 1B). Thyroid function should be evaluated at baseline and monitored at frequent intervals during therapy. (See 'Treatment of disease refractory to imatinib' above.)
An alternative approach, increasing the dose of imatinib to 800 mg daily, may be useful in patients who have clearly progressive disease while receiving 400 mg daily doses and who are tolerating the drug reasonably well. However, this approach is unlikely to benefit patients who have primary resistance to treatment (eg, within two months) after starting therapy. (See 'Dose escalation of imatinib' above.)
●For patients with imatinib- and sunitinib-refractory GIST, we recommend regorafenib rather than other TKIs (Grade 1B). (See 'Treatment of disease refractory to imatinib and sunitinib' above.)
●For patients with GIST (irrespective of PDGFRA D842V mutation status) refractory to or intolerant of imatinib, sunitinib, and regorafenib, we suggest the use of ripretinib over other available TKIs (Grade 2C). (See 'Treatment of disease refractory to imatinib, sunitinib, and regorafenib' above and 'Ripretinib' above.)
•For those who are ineligible for or intolerant of ripretinib, options include nilotinib, sorafenib, pazopanib, pimitespib (where available), or rechallenge with either imatinib or sunitinib. Avapritinib is an option for patients with the PDGFRA D842V mutation who have progressed on imatinib and sunitinib. (See 'Alternative targeted agents' above and 'Imatinib or sunitinib rechallenge' above.)
●For patients with progressive symptomatic soft tissue metastases despite treatment with multiple TKIs, radiation therapy is a reasonable option. (See 'Role of radiation therapy' above.)
ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges George Demetri, MD, who contributed to earlier versions of this topic review.
