60. Гастроинтестинальная стромальная опухоль

Введение

Gastrointestinal stromal tumors (GISTs) are mesenchymal neoplasms of the gastrointestinal tract. The interstitial cell of Cajal is the normal counterpart of tumor cells. This serves as a pacemaker of gastrointestinal motility, providing an interface between autonomic nerve stimulation and the muscle layer of the gastrointestinal wall.1 GISTs are rare cancers, which were defined as a distinct disease in the 1990s, having been classified within smooth muscle neoplasms for decades from their first description in the 1960s.2,3 Coincidentally, in 2000, they became targetable by new tyrosine kinase inhibitors (TKIs), given the role played by KIT and platelet-derived growth factor receptor α (PDGFRA) in their pathogenesis.4–6 As of today, GISTs serve as an advanced model displaying both potentials and limits of currently available molecularly targeted agents in medical oncology of solid cancers.

From the clinical point of view, surgery is the mainstay treatment when GISTs are localized, and adjuvant therapy is used depending on their risk of relapse. Apparently, adjuvant therapy with TKIs is mainly able to delay relapse, if due to occur, rather than to avoid it, though at the moment, it cannot be ruled out that in some genotypes and/or with longer treatments the cure rate may actually increase. In the advanced disease, TKIs have substantially improved the prognosis of KIT-mutated GISTs and have become standard treatment. They face the major limiting factor of secondary resistance, which affects most patients and is marked by genetic heterogeneity. A minority of GISTs do not harbor mutations to either KIT or PDGFRA genes and thus were formerly referred to as wild type (WT). They are much less amenable to treatment with available TKIs, although their natural history tends to be less aggressive. Their variegated nature adds to the complexity of GISTs as a family of tumors.

In brief, GISTs are more complex than initially believed, whereas targeted therapy has substantially improved their prognosis but is challenged by its apparent inability to eradicate the disease in most cases (even minimum residual disease) and by the heterogeneity of the secondary resistance it often gives rise to in the advanced setting. Intense translational and clinical research is underway and new agents are constantly developed. All this and the rarity of the disease strongly suggest to refer GIST patients to institutions or networks specializing in their treatment and study.

Заболеваемость и этиология

GISTs are rare cancers. Their incidence is suggested to be approximately 1.5 out of 100,000 per year (roughly 5,000 new cases in the United States yearly), with the limitations deriving from the fact that only recently were they identified as a clinicopathologic entity.7,8 However, there are a number of small GISTs that are clinically meaningless and generally go undetected. In addition, microscopic GISTs might be found incidentally in as many as 10% to 25% of stomachs.9,10 The reasons why the vast majority do not give rise to clinically overt diseases are not known, especially considering the fact that most of them harbor the same mutations to KIT and PDGFRA of fully developed diseases, which implies that alterations of these protooncogenes are not solely the pathogenetic mechanisms leading to GISTs.11 Thus, the incidence of histologic GISTs may be much higher than that of clinical cases, which remains low. Prevalence is low as well, although roughly half of clinical GISTs are cured by surgery, and the median survival of advanced GISTs has improved with the use of TKIs and is likely still improving.

GISTs can occur at any age, with a median occurrence at 60 to 65 years. A small minority of GISTs affect children and adolescents: Most of them are WT for KIT and PDGFRA, and some may take place within selected syndromes. In general, GISTs are slightly more incident in males than females. Succinate dehydrogenase (SDH)- deficient GISTs typically occur in young females.

No specific causes are known, although the pathogenesis of KIT- and PDGFRA-mutated GISTs has been elucidated in essence. There are some predisposing conditions for SDH-deficient GISTs, which include the Carney triad (marked by GIST, pulmonary chondromas, and extra-adrenal paragangliomas), the hereditary Carney-Stratakis syndrome (marked by GIST and familial paragangliomas), and neurofibromatosis type 1 (NF1).12–14 Hereditary syndromes driven by germline mutations to KIT or PDGFRA are very rare but well recognized.15,16

Анатомия и патология

More than half of GIST cases arise from the stomach, one-fourth from the small bowel, roughly 5% from the rectum, and a small minority from the esophagus.6 Some GISTs have been labeled as extragastrointestinal, apparently arising from the mesentery, omentum, and retroperitoneum; however, it remains at best unknown whether these are lesions detached from their gastrointestinal origin and/or are metastases from an unknown primary tumor.

Morphologically, GISTs can be made up of spindle cells (in more than two-thirds of cases), epithelioid cells, or both (Fig. 60.1).17 Epithelioid-cell GISTs are more common in the stomach and include those that are PDGFRA mutated and several SDH deficient. Aside from this, there are no major clinical implications in the microscopic aspect of lesions. Importantly, there are no pathologic clues to make a distinction between malignant GISTs and others whose clinical behavior is actually benign. Thus, many GISTs behave as benign diseases as a matter of fact, but this cannot be forecast histologically or molecularly. It follows that all GISTs are currently considered malignant neoplasms, although with a highly variable risk of distant relapse, which is negligible in a significant proportion of them. This is the reason why risk classification systems are generally used in the clinic as prognosticators, being based today on a pathologic factor (i.e., the mitotic count) and two clinical variables (tumor size and tumor site).18–22

Immunohistochemically, the hallmark of most GISTs is their positivity for KIT (CD117) and DOG1 (ANO1) (Fig. 60.2).23–25 A low proportion of GISTs are CD117−, which is typical of PDGFRA-mutated GISTs, but immunohistochemical status does not reflect the mutational status with regard to KIT and PDGFRA, per se, so that it has no concrete predictive value for sensitivity to TKIs. Thus, CD117 has only a meaning in the pathologic differential diagnosis. Given their morphology, GISTs must be differentiated from other soft tissue tumors of the gastrointestinal wall, including those of smooth muscle and neural differentiation, and desmoid-type fibromatosis, glomus tumors, endocrine tumors, melanocytic tumors, lymphomas, etc. Desmin is rarely positive, as opposed to CD34. A negative stain for SDHB identifies the subgroup of SDH-deficient GISTs.26–28

Фигура 60.1 A. Веретено-клеточная гастроинтестинальная стромальная опухоль. B: Эпителиоидно-клеточная гастроинтестинальная стромальная опухоль.

Molecularly, GISTs have become a relatively heterogeneous and complex group of lesions.29 Gain-of-function mutations of the oncogenes located on chromosome 4 (4q12) coding for the type III receptor tyrosine kinases KIT and PDGFRA can be found in approximately 80% of GISTs.5,6 Pathogenetically, they are the drivers of the disease and, therapeutically, underlie the efficacy of currently used TKIs. They are mutually exclusive and result in the constitutive activation of either KIT or PDGFRA, which normally are autoinhibited, being activated by the binding of their respective ligands (i.e., stem-cell factor [Steel factor] and PDGFA). The activation of the receptor binds two molecules of KIT or PDGFRA (dimerization), giving rise to downstream oncogenic signaling, which for both KIT and PDGFRA involves the RAS/MAPK and the phosphatidylinositol 3-kinase (PI3K)/AKT/mammalian target of rapamycin (mTOR) pathways (Fig. 60.3). Mutations can be deletions, insertions, and missense mutations. They affect exon 11 of the KIT oncogene, encoding for the juxtamembrane domain of the KIT receptor, in slightly <70% of GISTs; exon 9 of KIT, encoding for the extracellular domain of the receptor, in <10%; and exon 13 and 17 of KIT, encoding for the intracellular adenosine triphosphate (ATP)- binding pocket and activation loop domains, respectively, in a small minority of GISTs. Approximately 10% of GISTs have mutations homologous to these, which affect PDGFRA (i.e., exons 12, 14, and 18 of the oncogene, with 70% being represented by the exon 18 D842V mutation). The latter is known for its wide lack of sensitivity to approved TKIs, along with a few other rare exon 18 mutations, whereas the deletion of codons 842 to 845 is sensitive. Possibly because of their similarity with different kinds of normal interstitial cell of Cajal, some tumor cell mutations correlate with elective primary sites of origin. In particular, exon 9 mutations of KIT are preferably found in the small bowel, and PDGFRA mutations are found in the stomach.

Approximately 10% to 15% of GISTs are WT for KIT and PDGFRA. They make up a family of tumor subsets with different pathogenetic backgrounds and, to some extent, different natural histories (see Fig. 60.3). Their classification has evolved.6,29,30 In essence, as of today, one may identify (1) SDH-deficient GISTs; (2) NF1- related GISTs; or (3) others, including those with a BRAF V600E or RAS mutation, or an ETV6-NTRK3 gene fusion.30,31 In fact, approximately half of these GISTs are marked by alterations involving the SDH complex, which is crucial for the Krebs cycle and mitochondrial respiratory cell function. Immunohistochemically, these GISTs are negative to SDHB staining. A group of them includes pediatric GISTs and can be associated with the Carney triad.12 In fact, these GISTs tend to arise in children and young adults of the female sex, are gastric and multifocal, can metastasize to lymph nodes, and have a rather indolent evolution. When the Carney triad is fully expressed, it includes GISTs, pulmonary chondromas, and paragangliomas. On the other hand, a group of SDH- deficient GISTs carries germline mutations of the SDHA, SDHB, SDHC, and SDHD units of the SDH complex,29,31 and may be related to the Carney-Stratakis syndrome.13 This is an autosomal dominant condition marked by GIST and paragangliomas. Immunohistochemically, GISTs with SDHA mutations are negative to SDHA staining. The median age of these patients is somewhat higher and the female to male predominance is lower, but the course of disease is indolent as well. Those cases without mutations to the SDH complex are marked by SDHC promoter hypermethylation. Then, SDHB-positive GISTs can occur in the context of NF1, and their pathogenetic mechanism is supposed to be the absence of neurofibromin (i.e., the product of the NF1 gene), which is mutated.32–34 This leads to increased activity of the RAS pathway. GISTs related to NF1 are typically multicentric as well, and have a rather indolent course, but arise from the small bowel. Of course, an NF1 patient may always suffer from a non–NF1-related GISTs. Finally, the remaining SDHB-positive GISTs are probably a basket of different conditions: Some were reported to have the V600E mutation of BRAF35,36 or, more rarely, HRAS, NRAS, PIK3 mutations, or a ETV6-NTRK3 gene fusion.37 All this makes the formerly called WT GISTs a variegated family of tumors, which can now be identified not only through a negative definition (i.e., by the lack of KIT and PDGFRA mutations), but through immunohistochemical or molecular genetics markers, pointing to specific subsets with different natural histories.

Фигура 60.2. Иммуноокрашивание для KIT (CD117) и DOG1.

Фигура 60.3. Сигналинг в KIT/PDGFRA-мутированной и дикого типа гастроинтестинальной стромальной опухоли.

A very rare subset of familial GISTs does exist, being marked by mutations of KIT or PDGFRA affecting the germ line.15,16 They parallel mutations found in sporadic GISTs and lead to the multicentric and multifocal occurrence of GISTs. The behavior of these GISTs is variable (i.e., it is often indolent, but some lesions turn out to become aggressive). Hyperplasia of interstitial cells of Cajal can be found, which may entail altered motility of the gastrointestinal tract. Urticaria pigmentosa and other alterations of skin pigmentation may complete the syndrome.

It is then clear how important genotyping has become for GIST patients. In fact, genotyping has an obvious predictive value, which is crucial for all patients who are candidates for medical therapy, whether in the advanced or in the adjuvant setting. In addition, genotyping confirms the pathologic diagnosis in KIT/PDGFRA-mutated GIST or leads to further pathologic and molecular assessments in formerly called WT GISTs. In fact, and finally, genotyping has wide prognostic implications, particularly in regard to SDH-deficient GISTs. So, in case mutational analysis does not show any mutation to KIT and PDGFRA, the immunohistochemical status of SDH is assessed to single out SDH-deficient patients. These patients can then be genetically counseled in order to proceed to assess mutations in SDHX genes. All SDH-deficient patients have a risk of paragangliomas (which could justify annual whole-body magnetic resonance imaging [MRI]), whereas those with nonsporadic germline SDHX mutations require genetic counseling. For all these reasons, although there are subsets of GISTs with such a low risk of relapse as not to make them candidates for any medical therapy, a mutational analysis is currently felt as a companion to virtually any pathologic diagnosis of GISTs.

Фигура 60.4. Gastrointestinal stromal tumor growing outward of the gastric wall.

Скрининг

GISTs are rare cancers. Therefore, population-based screening policies are unforeseeable. As for all rare cancers, the clinical aim should be a timely diagnosis in the individual patient with symptoms and/or signs of disease. A difficulty thereof is the anatomic tendency of GIST lesions to grow outward from the gastrointestinal wall so that they may go undetected for long periods even when endoscopically explored. However, endoscopic procedures carried out for other reasons may lead to some risk of overdiagnosis, even in such a rare disease, when small gastric lesions are incidentally detected. Some of them will be benign entities, and others will be GISTs unlikely to ever grow as to become clinically relevant. Only a minority of them will turn out to be clinically aggressive GISTs caught in their making.

Диагноз

The outward growth of many GISTs (Fig. 60.4) within the gastrointestinal wall is one of the reasons why several are diagnosed relatively late, either as major abdominal masses or as causes of gastrointestinal bleeding, hemoperitoneum, or perforations (Fig. 60.5). Therefore, as many as one-fourth of GISTs are diagnosed in a clinical emergency, often leading to surgical explorations resulting in the unexpected finding of the disease. One- fourth of GISTs are discovered incidentally during diagnostic assessments (whether an endoscopic procedure, ultrasound, or computed tomography [CT] scan) done for other reasons. The remaining are diagnosed because of symptoms of compression from an abdominal mass, or chronic anemia, fatigue, and the like. Therefore, GISTs should be included in the differential diagnosis of abdominal masses. When their pertinence to the gastrointestinal wall is clear, the possibility of a GIST may be obvious, with a differential diagnosis mainly against epithelial tumors, small bowel endocrine tumors, lymphomas, paragangliomas, etc. Otherwise, retroperitoneal sarcomas and desmoid-type fibromatosis, germ cell tumors, and lymphomas are the main alternatives. Notably, when this is the clinical presentation, surgery is of choice only for some of the possible alternatives within the clinical differential diagnosis. In addition, preoperative treatments may be resorted to even in some of the surgical indications. On top of this, an intraoperative pathologic differential diagnosis is prohibitive. In principle, therefore, a diagnostic core needle biopsy is suggested, allowing pathologic diagnosis and, in the case of GIST, a mutational analysis, prior to any surgical exploration. In the case of gastric or rectal lesions, a biopsy can be carried out by means of endoscopic ultrasound, although, for gastric tumors, the risk of perforation should be factored in depending on the presentation. A CT/ultrasound-guided percutaneous biopsy is the other option, apparently with a negligible risk of dissemination if done at a center of expertise, again factoring in the clinical presentation.38 There may remain some cases in which the difficulty of an endoscopic or percutaneous biopsy and the easiness of a surgical exploration would suggest the latter. In general, however, a biopsy prior to any therapeutic planning can minimize the number of abdominal masses undergoing futile surgery.

Фигура 60.5. Дуоденальная гастроинтестинальная стромальная опухоль с внутриопухолевой перфорацией.

Follow-ups after potentially eradicating surgery are aimed at picking up relapses at an early stage. Local relapses are infrequent and tend to develop outward from the gastrointestinal wall: Therefore, an endoscopy is generally not used as a routine follow-up procedure. A CT scan is the most sensitive exam to pick up peritoneal and liver metastases and is recommended. It can be replaced by MRI, whereas ultrasound is much less sensitive on the peritoneum. The maximum risk interval averages 2 to 3 years after surgery or, if an adjuvant therapy was done, after its completion. Long-term relapses are unlikely, although they are occasionally observed, especially in GISTs with low mitotic rates. All this helps drive rational follow-up policies for potentially cured patients, although there is a lack of any empirical evidence of their effectiveness.39,40

Стадийность

Классическая стадия классификации используется редко. Клиницисты обычно различают локализованное и метастатическое заболевание и, если заболевание локализовано и поддается оперативному лечению, количественно определяют риск рецидива.

Современные классификационные системы риска основаны на комбинации количества митозов, размера опухоли и места происхождения. Число митозов является основным прогностическим фактором, пропорционально коррелирующим с риском рецидива. Его недостатком оказалась его низкая воспроизводимость, но очевидно, что он может быть выше, если патолог осознает его важность в выборе вариантов лечения. Размер опухоли является следующим прогностическим фактором. С одной стороны, он выделяет очень маленькие поражения желудка (<2 см), которые могут подвергаться бдительному наблюдению, если случайно выявляются эндоскопически. С другой стороны, это выделяет поражения, превышающие 5-10 см, которые имеют худший прогноз. Что касается первичного участка, поражения желудка имеют лучший прогноз, чем GIST тонкого кишечника и прямой кишки. Таким образом, сочетание этих трех факторов позволяет прогнозировать риск рецидива, используя такие инструменты, как классификация рисков AFIP (Armed Forces Institute of Pathology), номограмма MSKCC (Memorial Sloan Kettering Cancer Center) или контурные карты. Преимущество контурных карт состоит в том, что они рассматривают как митотическую активность, так и размер опухоли как непрерывные переменные, поскольку они таковы, и точность увеличивается, особенно для случаев со средним риском (Рис. 60.6). Кроме того, проблемы воспроизводимости становятся менее важными, если учитывать количество митозов непрерывную переменную. Кроме того, контурные карты разделяют прогноз поражений, перенесших разрыв опухоли, что является крайне неблагоприятным прогностическим фактором при заболеваниях, анатомически поражающих брюшину.

Фигура 60.6. Контурные прогностические карты при локализованных гастроинтестинальных стромальных опухолях.

The natural history of advanced GISTs is marked by their potential extension to the peritoneum and/or the liver. Thus, a CT scan is the staging procedure of choice to rule out metastatic disease. Lung metastases are rare, with the possible exception of rectal GISTs, although a chest CT scan is generally used to extend the staging workup to lungs and the mediastinum. Bone metastases are possible, but they are usually confined to the very advanced stages of disease so that the skeleton is not routinely assessed in the lack of symptoms.43 Other sites of distant metastases are exceedingly rare. Lymph node regional metastases are not typical of GISTs, as for mesenchymal tumors in general, with the remarkable exception of WT GISTs occurring in children and/or within syndromes. In addition, all syndromic GISTs may be multifocal and multicentric.44,45 This is not tantamount to metastatic spread, being rather a marker of their inherent natural history. All these features of the natural history of GISTs drive staging procedures, in addition to the potential for other syndromic correlates, depending on the presentation.

Менеджмент по этапам

Localized GISTs with no evidence of distant metastases are treated with surgery, followed by adjuvant medical therapy if the risk of relapse is significant. This treatment strategy capitalizes on the consolidated curative potential of surgery and prolongs the relapse-free interval of patients who are not eradicated. When surgery is unfeasible or could be made less mutilating or easier through downsizing, medical therapy is used if the genotype is sensitive to imatinib, possibly followed by surgery and the completion of a medical adjuvant treatment if the risk of relapse is significant. When the disease is metastatic, medical therapy with TKIs is standard treatment and should be maintained indefinitely. Surgery of metastatic residual responding disease can be used when reasonably feasible, but its added value prognostically is unproven. When imatinib fails and/or is ineffective, other available TKIs and judicious use of surgery of limited progression are resorted to (see Table 60.1 for conventionally used agents). This treatment strategy has substantially improved the prognosis of advanced GIST patients by increasing median survival in terms of years if compared to any historical series, with a proportion of patients, limited though it may be, becoming long-term progression-free survivors.

When the disease is localized, surgery is the treatment mainstay. Indeed, all GISTs ≥2 cm in size should be resected when possible because none of them can be considered benign. The management of GISTs <2 cm in size is more questionable.46–48 Although the low risk of progression of many GISTs <2 cm leads to the recommendation of a conservative approach, a reliable mitotic index cannot be determined by biopsy or fine- needle aspiration (FNA), thus preventing the identification of those at higher risk. Therefore, both observation and resection for GISTs 1 to 2 cm can be considered, and the risks and benefits of one versus the other should be discussed with the patient. The endoscopic resection of small gastric GISTs could be an option in these presentations. Risks of perforation may be low, although the decision is made on a case-by-case basis. Regardless of their size, any small GIST that is symptomatic (e.g., bleeding from erosions through the mucosa) or increases in size on serial follow-up should be resected.

Таблица 60.1. Стандартные лекарственные препараты, используемые в настоящее время для гастроинтестинальных стромальных опухолей (GIST)

Иматиниб

• 400 mg by mouth daily
• Possibly 800 mg by mouth daily, in case of:
  • Exon 9 KIT-mutated GIST
  • Progression on imatinib 400 mg

Сунитиниб

• 50 mg by mouth daily for 4 wk every 6 wk
• 5 mg by mouth daily continuously

Регорафениб

• 160 mg by mouth daily for 3 wk every 4 wk

A laparotomic or laparoscopic/laparoscopy-assisted resection of primary GISTs should be performed following standard oncologic principles. On laparotomy/laparoscopy, the abdomen should be thoroughly explored to identify and remove any previously undetected peritoneal metastatic deposits.49–53 Although primary GISTs may demonstrate inflammatory adhesions to surrounding organs, true invasion is not frequent. The goal of surgery is R0 excision. A macroscopically complete resection with negative or positive microscopic margins (R0 or R1 resection, respectively) is associated with a better prognosis than a macroscopically incomplete excision (R2 excision).54–57 Available series have not clearly shown that R1 surgery is associated with a definitely higher risk in terms of survival, although a higher than average risk of local relapse can be anticipated, especially when the lateral margins of a lesion confined to the gastrointestinal wall are positive. Therefore, the decision whether to reexcise a lesion already operated on with microscopically positive margins must be made on an individual basis, aside from the fact that sometimes a reexcision may not be technically foreseeable in the gastrointestinal tract. An exception is GIST of the rectum, where microscopically positive margins are clearly associated with a higher risk of local failure.58,59 In general, local relapse after R0 surgery is very unlikely in GISTs. Of course, the margins of a big lesion toward the peritoneum will not be covered by any clean tissue, and this may well be the main reason for the high peritoneal relapse rate of large tumors even after complete surgery, in the setting, at that point, of a metastatic disease. Tumor rupture or violation of the tumor capsule during surgery are associated with a very high risk of recurrence and therefore should be avoided.42 Some clinicians approach ruptured GISTs as already metastatic, although there may be different kinds of rupture, possibly leading to different risk levels. In addition, there are suggestions that patients with tumor rupture have a prognosis that may still be sensitive to the risk category (i.e., on the basis of mitotic index, size, and site).60 A lymphadenectomy is not routinely required because lymph nodes are rarely involved (in adult patients) and are thus resected only when they are clinically suspect.

In general, surgery is a wedge or segmental resection of the involved gastric or intestinal tract, with margins that can be less wide than for an adenocarcinoma. Sometimes, a more extensive resection (e.g., total gastrectomy for a large proximal gastric GIST, pancreaticoduodenectomy for a periampullary GIST, or abdominoperineal resection for a low rectal GIST) is needed. In the rare syndromic GIST (either SDH deficient or NF1 related), tumors are often multifocal and confined either to the stomach (SDH-deficient GIST) or the small bowel (NF1- related GIST). The extent of surgery should be decided on a case-by-case basis, taking into account the risk of recurrence, the lack of benefit from currently available TKIs, and the actual behavior of the underlying disease.61

Adjuvant medical therapy with imatinib was demonstrated to substantially improve relapse-free intervals, although with a trend to lose the benefit in a time span of 1 to 3 years from the end of therapy.60,62,63 This was shown through randomized trials that compared 1 and 2 years of adjuvant therapy with imatinib versus no adjuvant therapy and 3 years versus 1 year of adjuvant therapy with imatinib. As of today, the suggestion from these studies is that adjuvant therapy with TKIs can delay, but probably not avoid, a relapse, if this is due to occur. This correlated with a survival improvement in one trial63 and with a trend to improvement of a potential surrogate for survival in another,60 where the surrogate was survival free from changing the original TKI—in practice, survival without secondary resistance. In fact, secondary resistance is the limiting factor of TKIs in the advanced setting, so that an adjuvant therapy will be beneficial as long as it either avoids recurrences or at least prolongs freedom from secondary resistance but by no means shortens it. Thus, the risk of any detrimental effect was ruled out for adjuvant therapy durations up to 3 years. In this sense, going beyond 3 years would seem logical, given the tendency to lose the benefit after 1 to 3 years from stopping adjuvant therapy, and in fact, some institutions prefer longer treatment durations for patients who have a high risk of relapse.64,65 However, such a policy needs to be validated by ongoing controlled clinical trials ruling out any adverse effect on secondary resistance. At the moment an uncontrolled study seems to suggest this.66 However, any curative potential of adjuvant therapy has not been demonstrated at the moment, although this cannot be ruled out for specific genotypes (e.g., KIT exon 11 deletions, including those affecting codons 557 and/or 558, which benefit more from 3 years in comparison to 1 year).67 Therefore, adjuvant therapy is recommended as a standard for 3 years and is reserved for patients with a significant risk of relapse, as long as the benefit in absolute terms will be higher as the risk increases, as is the case with all adjuvant therapies. In a sense, the lack of a tangible impact on the long-term relapse rate encourages one to exclude relatively low-risk patients, which is, to some extent, at odds with what is done with adjuvant cytotoxic chemotherapy in some solid cancers. This said, the magnitude of risk that is worth an adjuvant therapy with imatinib for 3 years may well be subject to a shared decision making with the individual patient, and, as a matter of fact, is often placed in the 30% to 50% range. Logically, a benefit can be expected for patients whose genotype is potentially sensitive to imatinib.67 In practice, this leads to the selection of all patients with a KIT-mutated GIST or a PDGFRA-mutated sensitive GIST (with the exception of the D842V exon 18 mutation and the few others that are insensitive in vitro and in vivo to imatinib). Given the benefit shown with the use of a double dose of imatinib (800 mg daily) for advanced GIST patients with an exon 9 KIT-mutated GIST,68 such a dosage can be selected for them, although there is a lack of any positive demonstration in the adjuvant setting. Formerly called WT GISTs are at best much less sensitive to imatinib, and adjuvant studies are lacking with other TKIs, which may be potentially more active. Even more importantly, the natural history of these GISTs is often less aggressive. These are the reasons why most clinicians currently do not select these patients for any adjuvant treatment.

Given the extensive use of adjuvant therapy with imatinib in the high-risk populations and the activity of the drug, several recent multi-institutional retrospective series have questioned the need for extensive resections such as pancreaticoduodenectomy, abdominal perineal resection, or total/proximal gastrectomy, when tumor downsizing can be likely achieved with a preoperative medical treatment. In practice, preoperative imatinib can shrink gastric, periampullary, or rectal GISTs to such an extent as to allow more limited excisions (wedge gastrectomy, excision of periampullary lesions, transanal/perineal resection of rectal GISTs, respectively), and imatinib can then be continued postoperatively to complete the adjuvant treatment (Fig. 60.7). Thus, if extensive surgery is required for complete tumor removal, preoperative imatinib should be considered.69–72 In addition to this, there are some big abdominal masses that may be felt by the surgeon as implying a significant risk of tumor rupture during surgery, which can be treated with preoperative imatinib. Because downsizing is the clinical end point in these cases, the duration of preoperative medical therapy is generally 6 to 12 months, which corresponds to the time interval when the maximum degree of tumor shrinkage was shown to occur in studies on advanced GISTs.73 In addition, mutational status is important in order to select patients likely to respond to imatinib, and tumor response should be monitored closely. Positron emission tomography (PET) scans are a resource because they can demonstrate tumor responsiveness in a matter of weeks.

Фигура 60.7. Tumor shrinkage (A) of a gastric primary gastrointestinal stromal tumor after 12 months of preoperative imatinib, allowing for a sleeve gastrectomy (B) plus splenectomy and liver resection with preservation of most of the stomach after resection.

Syndromic GISTs can present with multifocal and/or multicentric disease, which may imply delicate surgical decisions. Thus, in SDH-deficient GIST and those occurring in NF1 patients, one should take into account the indolent behavior of many lesions and the possible presence of hyperplasia of the interstitial cells of Cajal on one side and the possibility that single lesions may be aggressive on the other. Surgery should judiciously factor in all this. In addition, the relative lack of sensitivity of WT GISTs to available TKIs may suggest surgery more liberally than is currently done with KIT-mutated GISTs.

With regard to the highly rare syndromes of familial GISTs from germline mutations of KIT or PDGFRA, treatment is challenging and may involve resorting to surgery and/or TKIs depending on the behavior and extent of clinically relevant lesions.

When the disease is metastatic or locally advanced, medical therapy is the best choice and is currently based on imatinib continued indefinitely.74–77 However, given the limiting factor of secondary resistance, some clinicians prefer to surgically resect some highly localized first distant relapses, thus delaying the start of imatinib to a subsequent relapse. It is unproven whether this approach may delay progression, its rationale being to delay the onset of time to secondary resistance by delaying the onset of any use of TKIs. Theoretically, the downside may be starting medical therapy with a higher tumor burden, which was shown to be related to a shorter time to secondary resistance to imatinib. In fact, initial tumor burden is virtually the only prognostic factor in the metastatic GIST patient starting imatinib.78 On the other hand, although above 80% overall (considering all types of tumor response), the probability of response is strictly correlated with the mutational status, which is, therefore, the main predictive factor. KIT-mutated GISTs are responsive in most cases, including the most frequent genotype, marked by mutations of exon 11. This applies also to patients who underwent adjuvant imatinib and who did not experience tumor relapse during the adjuvant period so that these patients are currently approached the same way as those who have not been already exposed to imatinib. The standard daily dosage of imatinib is 400 mg. However, there are data derived from retrospective subgroup analyses that suggest that progression-free survival is better with doses higher than 400 mg (i.e., 800 mg daily) for exon 9 KIT-mutated GIST patients.68 Thus, many institutions treat these patients with 800 mg. PDGFRA-mutated GIST patients can be sensitive to imatinib as well, with the remarkable exception of the D842V mutation, which is the most frequent among PDGFRA mutations, and a few others. Nonmutated GISTs are essentially not sensitive to imatinib, which, however, has been reported to be active in occasional patients. Sunitinib and regorafenib are alternative options because they were shown to have some activity in SDH-deficient GISTs.81–83

Generally, the clinical decision in these GISTs, and PDGFRA D842–mutated patients, takes into account the natural history of these subtypes, which is often less aggressive than in KIT-mutated GISTs. However, new agents, such as BLU-285 and crenolanib, are currently under study in PDGFRA D842–mutated patients, so that these should be preferably sent to centers running ongoing trials.84,85

Once started, therapy with anti–tyrosine kinase agents is continued indefinitely. In fact, a discontinuation trial showed that stopping therapy after 1, 3, or 5 years is followed by progression in a matter of months.77 It is true that reestablishing therapy generally leads to a new response, but its quality may be inferior to the previous one.86 In any case, intervals to progression would be remarkably short, and an untenable stop-and-go treatment policy would be the only result.

Imatinib is generally well tolerated, with fatigue, edema, mild diarrhea, and anemia as frequent complaints, along with less frequent toxicities, such as neutropenia, skin rash, and others.75,76,87 Clinical wisdom needs to be exercised in order to maintain dose intensity vis-à-vis side effects. The number of patients truly intolerant to imatinib should be exceedingly low.

Secondary resistance is the limiting factor of imatinib, with a median time to the event averaging 2 years in the frontline advanced setting. Likely, current patients, who are generally put on therapy with lower tumor burdens, may show improved progression-free survival intervals over the earliest series. More importantly, the range of time to secondary resistance is wide, with a limited proportion of patients, averaging 10%, who become long-term progression-free survivors. Currently, there are no known prognostic factors for long-term progression-free survivorship, excluding the mutational status, which affects tumor response to imatinib, and tumor burden at the onset of imatinib therapy, which affects the duration of response. The group of long-term progression-free survivors may thus represent either just the “tail” of a curve driven by the stochastic mechanisms of secondary resistance or the result of specific genomic profiles, still to be elucidated.76

In an attempt to diminish tumor burden, thus potentially prolonging time to secondary resistance, surgery of residual responding disease has been resorted to in many institutions, and its results were retrospectively, but not prospectively, evaluated, with the exception of an underpowered randomized prospective study in patients who had only peritoneal disease.88–92 These case series analyses showed a better prognosis for patients undergoing surgery, but a selection bias might explain the results. A big prospective trial failed to accrue; therefore, the decision whether to surgically excise metastatic lesions responding to imatinib is currently left to a shared decision making with the patient in conditions of uncertainty. Of course, clinical presentations are manifold, and sometimes, the easiness of the surgical resection is the main factor leading to the decision, and vice versa. In general, many institutions currently avoid resorting to major surgery for responding metastases. In any case, only patients amenable to complete resection of all lesions should be candidates for this kind of surgery. In this sense, surgery may be less often indicated in peritoneal compared to liver metastases because the former are frequently underestimated by available imaging modalities and the selection of completely clearable tumors is less feasible. However, the clinician must be aware that imatinib needs to be continued after surgery, even if surgery was complete. In fact, some patients enrolled in the discontinuation trial of imatinib had a complete excision of their metastatic lesions.77 In addition, surgery of metastatic GISTs was never proved to be eradicating in the preimatinib era.93

Progression during therapy with imatinib is often due to secondary resistance, which essentially is marked by the occurrence of new mutations to the same primarily mutated oncogene or, less frequently, oncogene amplifications or alterations of alternative pathways.94,95 Secondary mutations entail such consequences as to lower the binding capacity of the KIT or PDGFRA receptor for imatinib and/or to circumvent its inhibiting action by escape mechanisms. It was demonstrated that secondary resistance can be heterogeneous so that more mutations can be detected in different lesions or even within the same lesion.96,97 Of course, this is a major limiting factor for second-line agents targeting KIT or PDGFRA. Secondary mutations in KIT-mutated GISTs are relatively limited in number, affecting exons 13 and 14, which encode the ATP-binding pocket, or exons 17 and 18, which encode the activation loop. Many resistant PDGFRA-mutated GISTs acquire the D842V mutation, which encodes for the activation loop of the receptor.

Clinically, the progression may be limited in a substantial proportion of cases. This means that progression is radiologically evident only in one or a few lesions, with the others still progressing. A typical clinical pattern is the nodule within the nodule (i.e., a small hyperdensity within a responding hypodense lesion on CT scan).98 Given the scope of activity of further-line therapies, many clinicians now tend to treat limited progression conservatively from the medical point of view, resorting to selected surgical or ablative procedures to get rid of the progressing component of the disease, while continuing imatinib for the remaining.99 This might delay the moment when the first TKI is switched to a second-line one. Although this policy is not based on prospective clinical studies, it makes sense in the economy of advanced GISTs following the introduction of TKIs. Clearly, this does not apply to generalized progressions.

Radiologic progression should be confirmed, taking into account the peculiarities of tumor response patterns in GISTs undergoing TKIs. Furthermore, before attributing progression on frontline imatinib to molecular secondary resistance, one should rule out any lack of patient compliance with therapy, which may often go unnoticed and even be underappreciated by the patient. Another mechanism leading to resistance can lie in changes of the pharmacokinetics of the drug. There is evidence that pharmacokinetics can undergo variations with time, in addition to being variable across individuals.100 This has led to the evaluation of the importance of maintaining target plasma levels of imatinib.101 There are limitations to the standardization of such assessments, and available data pointing to a correlation with progression-free survival are retrospective in nature. Thus, at the moment, we lack any convincing formal demonstration that pharmacokinetics is a factor able to personalize medical therapy (i.e., to drive changes in drug dosages in the lack of evident progression). However, available evidence also suggests that it can well be a variable with many orally administered targeted agents. At least, plasma levels may be assessed in the single patient: in case of clinical progression, to rule out that a major pharmacokinetic issue does exist at that stage; in case of unexpected side effects; and in case of comedications potentially able to interfere with the drug metabolism.

In the case of clinical progression with imatinib 400 mg daily, an option widely used is to increase the dose to 800 mg daily. This proved temporarily successful in a limited proportion of patients crossing over to the higher dose in randomized trials that compared 400 mg with 800 mg as frontline therapy for advanced GISTs.102 When valuing this benefit, one should probably discount patients who had an exon 9 mutation and who started with 400 mg, and possibly some patients failing to comply with therapy, whereas other patients may have benefited due to the correction of pharmacokinetics problems.

Standard second-line therapy is sunitinib, which is not only a tyrosine kinase inhibiting KIT and PDGFRA but also displays antiangiogenic activity by the inhibition of vascular endothelial growth factor receptors (VEGFR) 1, 2, and 3. It was shown to be effective at increasing progression-free survival by 5 months in a randomized trial versus placebo in patients failing (or intolerant) to imatinib.103 Its molecular profile is such as to include activity on exon 9 KIT mutations as well as on secondary mutations of regions coding for the ATP-binding pocket, thus potentially covering mutations that are, respectively, less affected or not affected by imatinib.104,105 What limits the potential of molecular prediction in further-line therapies of GISTs is basically the heterogeneity that often underlies secondary resistance, so that the presence of a secondary mutation is unlikely to be alone, although potentially sensitive to available agents. However, the activity of sunitinib against some secondary mutations and probably its antiangiogenic activity underlie its clinical efficacy after failing to imatinib. Its tolerability profile is less favorable than imatinib, with fatigue and hand–foot syndrome as its main side effects, variable though they may be across patients. Although the clinical trial evaluated a regimen of sunitinib given 50 mg daily for 4 weeks, with a 2-week rest, a continuous regimen with a daily dose of 37.5 mg can be used as well.106

Regorafenib, another TKI with activity on KIT and PDGFRA as well as VEGFR 1, 2, and 3, and thus with antiangiogenic properties, is standard third-line therapy for advanced GIST patients. In fact, it was shown to be effective as a third-line therapy in patients failing both imatinib and sunitinib, by providing a median advantage of 4 months of progression-free survival over placebo in a randomized clinical trial.107 Its tolerability profile is close to sunitinib, with hand–foot syndrome, hypertension, fatigue, and diarrhea as main side effects.108

An observation made in this trial was the persistence of some activity when therapy was carried on beyond progression. In other words, a subset of patients, although arbitrarily selected by investigators on clinical grounds, went on with therapy beyond their first progression, achieving a second progression-free interval that approximated the previous one. This would suggest that a subset of progressing patients have a disease that might be slowed down by continuing the TKI despite the progression. This observation likely applies to all TKIs, at least in selected patients, and is worth testing prospectively by developing criteria to single out those patients who are more likely to benefit. In general, this parallels the clinical feeling that stopping any kind of TKI may accelerate tumor progression, even when resistance to that TKI has been established. Indeed, a criticism that was made to placebo-controlled trials on new TKIs in GIST is the lack of any tyrosine kinase inhibition in the control group, possibly worsening its outcome in comparison to what could happen by continuing the TKI already in use or by rechallenging the disease with a TKI used earlier.

In fact, in anecdotal cases and also in a small randomized clinical trial, it was shown that rechallenging a progressing GIST patient with TKIs used at an earlier stage can be beneficial, at least temporarily.109 In other words, reestablishing imatinib in patients who underwent the drug as frontline therapy and then switched to others results in a benefit in terms of progression-free survival that is not far from what is achievable with further-line agents. One may speculate that there is a process of reexpansion of tumor clones that were sensitive to imatinib and might have been narrowed by the selective pressure of the drug, as long as resistant clones were emerging.

Following third-line therapy, there is no standard option at the moment, aside from rechallenge, and clearly patients are eligible for clinical studies on new agents. In principle, new drugs investigated in currently ongoing trials in patients with secondary resistance include other TKIs targeting KIT and PDGFRA; agents targeting downstream pathways (e.g., PI3K/AKT); agents targeting heat shock proteins, given their chaperone function for KIT and PDGFRA; and agents targeting pathways acquiring expression after resistance (e.g., MET) among others.110 Clearly, agents with a mechanism of action other than imatinib, sunitinib, and regorafenib try to address the limiting factor of the heterogeneous nature of secondary resistance. Combinations of agents with different mechanisms of action are tried as well, although their added toxicity may be prohibitive even when they are reasonably well tolerated as single drugs. Future directions might try to exploit molecular diagnostics such as the liquid biopsy (i.e., the assessment of secondary mutations on circulating DNA shed by tumor cells). The sensitivity of this technique for primary and secondary mutations of GISTs has been demonstrated.111 Thus, it looks promising for the future to allow a degree of molecular personalization of therapy, whether following secondary resistance or before it establishes clinically, as a means to avoid or delay its occurrence. Rotations of TKIs are currently tested, although methods to rationally drive treatment modulation through liquid biopsies are lacking. Actually, the efficacy of treatment beyond progression and of rechallenge suggests that sensitive and resistant clones may fluctuate within the tumor load, depending on the selective pressure they are exposed to. This would be consistent with a kind of a liquid resistance, which, clinically, could be exploited by employing new strategies as from the upfront approach with TKIs.

In the era of immune therapy in medical oncology, evidence of activity of checkpoint inhibitors in GIST has not been provided. On the other hand, programmed cell death protein ligand 1 (PD-L1) expression was shown to be an independent favorable prognostic factor in localized GIST.112 Interestingly, in a mouse model, it was shown that the antitumor activity of imatinib may be potentiated by the immune system and that concomitant immune therapy with checkpoint inhibitors may improve it.113 Results of ongoing attempts are therefore expected.

A methodologic issue in the medical therapy of GISTs with all TKIs lies in the peculiar patterns displayed by tumor response as compared to those observed with standard cytotoxic chemotherapy of solid cancers and lymphomas.114–116 These patterns of tumor response are marked by the possible lack of tumor shrinkage in the face of substantial changes in tumor tissue and tumor metabolic activity. Although most observations derive from imatinib in frontline therapy, in essence, they regard all TKIs, the main difference being possibly the weakness of tumor response to further-line therapies so that some of these aspects may look less clear cut in the further-line therapy setting as opposed to the frontline. Under these patterns of tumor response, first of all, in the presence of symptoms, a subjective response may take place very early. In a matter of days, if not hours, after starting an effective TKI, a symptomatic patient may well feel a clear degree of subjective improvement. As far as imaging is concerned, this is paralleled by metabolic response as assessed through a fluorodeoxyglucose (FDG)-PET scan.117 A positive PET scan may turn negative in a few days. Of course, this does not correspond to the disappearance of the tumor lesion but rather be the consequence of the metabolic switch off that the tumor undergoes when an effective TKI targets its cells. The reverse is true as well, so that any stop of therapy rapidly entails a switch on of functional imaging. Again, this does not correspond to clinical progression, which would follow only for longer interruptions of therapy. This should be taken into account when assessing functional tumor response because, for instance, any lack of compliance with therapy in the days the exam is made might affect metabolic response as detected through PET scanning. For example, metabolic switch on was observed by PET scans performed during the 2-week off interval of therapy with sunitinib. In principle, this adds to the feeling that TKIs need to be maintained in order to preserve tumor response, in the context of a clinically cytostatic effect. It goes without saying that when a PET scan has become negative, the tumor lesion will not be visible to functional imaging, so that a CT scan, an MRI, or an ultrasound needs to be used in order to appreciate the evolution of tumor lesions. The radiologic response, as assessed through CT scans and MRI, is marked by tumor shrinkage and/or changes in tumor tissue. Tumor shrinkage may well appear very early, but, in some cases, it is lacking in the early phases of treatment, or even later on, so that tumor lesions look unchanged dimensionally. Sometimes, tumor size may even increase. In these cases, however, if a response is in place, the radiologic aspect will show substantial changes to the tumor tissue. On a CT scan, this means a decrease in density of responding lesions, with decreased contrast enhancement. On an MRI, it entails an hypointense signaling on T1-weighted images and hyperintense on T2- weighted images and decreased contrast enhancement. These changes are substantial in GIST patients undergoing frontline therapy with imatinib so that recording tumor response is generally unchallenging for the clinician, provided tumor shrinkage is not the only criterion. Signs of nondimensional response may prove less obvious when the tumor response is less clear cut, as with further-line therapy with TKIs. Functional imaging may help, although it is exposed to the same limitations as well if the response is less striking. However, when the response is overt, the main shortfalls of nondimensional tumor response assessments lie in the difficulty to standardize reproducible (i.e., reliable) instruments, as is needed in clinical trials. In this sense, Response Evaluation Criteria in Solid Tumors (RECIST) for tumor response assessment are based on the measurement on one diameter of selected target lesions and, thus, have a good record of reproducibility.118 However, their validity is, by definition, unsatisfactory in the presence of a nondimensional response. Choi criteria were worked out in GISTs to accommodate these patterns of response, by factoring tumor hypodensity on CT scans in addition to a decrease in size.119 Their validity in predicting progression-free survival was demonstrated and compared favorably with RECIST and also paralleled with functional imaging with PET scanning. Then, aside from the need to use easily reproducible instruments in the research setting, for the clinician, the message coming from the GIST model is simple, inasmuch as it points to the existence of nondimensional patterns of tumor response, which can be easily highlighted through CT scans and MRI on one side and through functional imaging with PET scan on the other. One should be aware that the meaning of a radiologic response through CT scans or MRI is deeply different from metabolic response assessed through a PET scan. In fact, a PET scan measures the biologic effect of a TKI on the tumor cells very early but does not necessarily imply any anatomical change in the tumor. On the contrary, CT scans and MRI detect actual changes in the tumor tissue, which correspond to pathologic signs of tumor response. These were found to take shape in terms of a myxoid degeneration widely affecting responding tumor lesions, with signs of apoptosis (Fig. 60.8). A variable proportion of vital cells may be detected, especially to the periphery of lesions (pointing in principle to the prospects of regrowth in case of discontinuation of the TKI). In this sense, a nondimensional tumor response does not mean just absence of progression; indeed, it is all about an actual pathologic response, with major changes to the tumor tissue. Of course, all tumor changes one can see when a tumor response is in place have their counterparts when the tumor progresses. Thus, increased tumor density and contrast enhancement on a CT scan will mark tumor progression, with or without an increase in tumor size. This may well affect just a portion of the tumor lesions, such as its periphery or a small part (as is the case with the nodule within the nodule). In brief, the quality of the tumor tissue should be observed, in addition to its size, in order to detect both response and progression in GISTs undergoing a TKI. Whatever the response pattern, whether dimensional or not, a tumor response on a CT scan or MRI says that the tumor is undergoing pathologic changes that clearly correlate with the prognosis. In fact, both dimensional and nondimensional tumor responses have clearly correlated with improved outcome in clinical trials, as opposed to progression. Only secondary resistance, or treatment interruption, will terminate a dimensional or nondimensional tumor response, with radiologic signs that, as said, will be dimensional or nondimensional as well.

Фигура 60.8. Патологическая опухолевая реакция на повреждение гастроинтестинальной стромальной опухоли при терапии иматинибом.

Паллиативная помощь

The natural history of GISTs that are not cured by initial surgery is dominated by abdominal spread involving the liver and the peritoneum. Liver failure as well as intestinal and urinary obstructions are thus the main palliative challenges. This may well carry the need of palliative surgery in selected patients. Extra-abdominal metastases are occasionally seen, mainly to the bone, and can require palliative irradiation. A systemic sign such as fatigue may add to asthenia induced by anemia as well as directly by TKIs. In fact, the existence of three lines of standard medical therapy, the potentials of rechallenge, and the availability of many agents of interest, either within clinical studies or among TKIs developed for other diseases, lead to treating many GIST patients with molecularly targeted therapy even in the very advanced stages of disease. In this sense, the usual palliative challenges of abdominal malignancies meet the new palliative challenges posed by the use of TKIs, which revolutionized the field of the disease.

Indeed, all phases of treatment of GIST are still a model for the medical oncology of new molecularly targeted agents. This model continues to shed light on their potentials in solid cancers as well as on their current limitations. It also demonstrates how clinical methodology is deeply affected by these agents, not only for medical oncologists but also for all members of the multidisciplinary cancer team, from surgeons to palliative physicians.

Литература

1. Kindblom LG, Remotti HE, Aldenborg F, et al. Gastrointestinal pacemaker cell tumor (GIPACT): gastrointestinal stromal tumors show phenotypic characteristics of the interstitial cells of Cajal. Am J Pathol 1998;152(5):1259– 1269.
2. Martin JF, Bazin P, Feroldi J, et al. Intramural myoid tumors of the stomach. Microscopic considerations on 6 cases. Ann Anat Pathol 1960;5:484–497.
3. Stout AP. Bizarre smooth muscle tumors of the stomach. Cancer 1962;15:400–409.
4. Hirota S, Isozaki K, Moriyama Y, et al. Gain-of-function mutations of c-kit in human gastrointestinal stromal tumors. Science 1998;279(5350):577–580.
5. Rubin BP, Heinrich MC. Genotyping and immunohistochemistry of gastrointestinal stromal tumors: an update. Semin Diagn Pathol 2015;32(5):392–399.
6. von Mehren M, Joensuu H. Gastrointestinal stromal tumors. J Clin Oncol 2018;36(2):136–143.
7. Nilsson B, Bümming P, Meis-Kindblom JM, et al. Gastrointestinal stromal tumors: the incidence, prevalence, clinical course, and prognostication in the preimatinib mesylate era—a population-based study in western Sweden. Cancer 2005;103(4):821–829.
8. Søreide K, Sandvik OM, Søreide JA, et al. Global epidemiology of gastrointestinal stromal tumours (GIST): a systematic review of population-based cohort studies. Cancer Epidemiol 2016;40:39–46.
9. Agaimy A, Wünsch PH, Hofstaedter F, et al. Minute gastric sclerosing stromal tumors (GIST tumorlets) are common in adults and frequently show c-KIT mutations. Am J Surg Pathol 2007;31(1):113–120.
10. Kawanowa K, Sakuma Y, Sakurai S, et al. High incidence of microscopic gastrointestinal stromal tumors in the stomach. Hum Pathol 2006;37(12):1527–1535.
11. Rossi S, Gasparotto D, Toffolatti L, et al. Molecular and clinicopathologic characterization of gastrointestinal stromal tumors (GISTs) of small size. Am J Surg Pathol 2010;34(10):1480–1491.
12. Carney JA. Gastric stromal sarcoma, pulmonary chondroma, and extra-adrenal paraganglioma (Carney Triad): natural history, adrenocortical component, and possible familial occurrence. Mayo Clin Proc 1999;74(6):543–552.
13. Stratakis CA, Carney JA. The triad of paragangliomas, gastric stromal tumours and pulmonary chondromas (Carney triad), and the dyad of paragangliomas and gastric stromal sarcomas (Carney-Stratakis syndrome): molecular genetics and clinical implications. J Intern Med 2009;266(1):43–52.
14. Fuller CE, Williams GT. Gastrointestinal manifestations of type 1 neurofibromatosis (von Recklinghausen’s disease). Histopathology 1991;19(1):1–11.
15. Li FP, Fletcher JA, Heinrich MC, et al. Familial gastrointestinal stromal tumor syndrome: phenotypic and molecular features in a kindred. J Clin Oncol 2005;23(12):2735–2743.
16. Postow MA, Robson ME. Inherited gastrointestinal stromal tumor syndromes: mutations, clinical features, and therapeutic implications. Clin Sarcoma Res 2012;2(1):16.
17. Miettinen MM, Corless CL, Debiec-Rychter M, et al. Gastrointestinal stromal tumors. In: Fletcher CDM, Bridge JA, Hogendoorn PCW, et al., eds. WHO Classification of Tumours of Soft Tissue and Bone. Lyon, France: International Agency for Research on Cancer; 2013:164–167.
18. Miettinen M, Lasota J. Gastrointestinal stromal tumors: pathology and prognosis at different sites. Semin Diagn Pathol 2006;23(2):70–83.
19. Joensuu H. Risk stratification of patients diagnosed with gastrointestinal stromal tumor. Hum Pathol 2008;39(10):1411–1419.
20. Gold JS, Gönen M, Gutiérrez A, et al. Development and validation of a prognostic nomogram for recurrence-free survival after complete surgical resection of localised primary gastrointestinal stromal tumour: a retrospective analysis. Lancet Oncol 2009;10(11):1045–1052.
21. Rossi S, Miceli R, Messerini L, et al. Natural history of imatinib-naive GISTs: a retrospective analysis of 929 cases with long-term follow-up and development of a survival nomogram based on mitotic index and size as continuous variables. Am J Surg Pathol 2011;35(11):1646–1656.
22. Joensuu H, Vehtari A, Riihimäki J, et al. Risk of recurrence of gastrointestinal stromal tumour after surgery: an analysis of pooled population-based cohorts. Lancet Oncol 2012;13(3):265–274.
23. Sarlomo-Rikala M, Kovatich AJ, Barusevicius A, et al. CD117: a sensitive marker for gastrointestinal stromal tumors that is more specific than CD34. Mod Pathol 1998;11(8):728–734.
24. West RB, Corless CL, Chen X, et al. The novel marker, DOG1, is expressed ubiquitously in gastrointestinal stromal tumors irrespective of KIT or PDGFRA mutation status. Am J Pathol 2004;165(1):107–113.
25. Novelli M, Rossi S, Rodriguez-Justo M, et al. DOG1 and CD117 are the antibodies of choice in the diagnosis of gastrointestinal stromal tumours. Histopathology 2010;57(2):259–270.
26. Doyle LA, Nelson D, Heinrich MC, et al. Loss of succinate dehydrogenase subunit B (SDHB) expression is limited to a distinctive subset of gastric wild-type gastrointestinal stromal tumours: a comprehensive genotype-phenotype correlation study. Histopathology 2012;61(5):801–809.
27. Wagner AJ, Remillard SP, Zhang YX, et al. Loss of expression of SDHA predicts SDHA mutations in gastrointestinal stromal tumors. Mod Pathol 2013;26(2):289–294.
28. Ricci R, Dei Tos AP, Rindi G. GISTogram: a graphic presentation of the growing GIST complexity. Virchows Arch 2013;463(4):481–487.
29. Nannini M, Urbini M, Astolfi A, et al. The progressive fragmentation of the KIT/PDGFRA wild-type (WT) gastrointestinal stromal tumors (GIST). J Transl Med 2017;15(1):113.
30. Pantaleo MA, Nannini M, Corless CL, et al. Quadruple wild-type (WT) GIST: defining the subset of GIST that lacks abnormalities of KIT, PDGFRA, SDH, or RAS signaling pathways. Cancer Med 2015;4(1):101–103.
31. Boikos SA, Pappo AS, Killian JK, et al. Molecular subtypes of KIT/PDGFRA wild-type gastrointestinal stromal tumors: a report from the National Institutes of Health Gastrointestinal Stromal Tumor Clinic. JAMA Oncol 2016;2(7):922–928.
32. Kinoshita K, Hirota S, Isozaki K, et al. Absence of c-kit gene mutations in gastrointestinal stromal tumours from neurofibromatosis type 1 patients. J Pathol 2004;202(1):80–85.
33. Maertens O, Prenen H, Debiec-Rychter M, et al. Molecular pathogenesis of multiple gastrointestinal stromal tumors in NF1 patients. Hum Mol Genet 2006;15(6):1015–1023.
34. Gasparotto D, Rossi S, Polano M, et al. Quadruple-negative GIST is a sentinel for unrecognized neurofibromatosis type 1 syndrome. Clin Cancer Res 2017;23(1):273–282.
35. Agaram NP, Wong GC, Guo T, et al. Novel V600E BRAF mutations in imatinib-naive and imatinib-resistant gastrointestinal stromal tumors. Genes Chromosomes Cancer 2008;47(10):853–859.
36. Hostein I, Faur N, Primois C, et al. BRAF mutation status in gastrointestinal stromal tumors. Am J Clin Pathol 2010;133(1):141–148.
37. Brenca M, Rossi S, Polano M, et al. Transcriptome sequencing identifies ETV6-NTRK3 as a gene fusion involved in GIST. J Pathol 2016;238(4):543–549.
38. Eriksson M, Reichardt P, Sundby Hall K, et al. Needle biopsy through the abdominal wall for the diagnosis of gastrointestinal stromal tumour—does it increase the risk for tumour cell seeding and recurrence? Eur J Cancer 2016;59:128–133.
39. Plumb AA, Kochhar R, Leahy M, et al. Patterns of recurrence of gastrointestinal stromal tumour (GIST) following complete resection: implications for follow-up. Clin Radiol 2013;68(8):770–775.
40. D’Ambrosio L, Palesandro E, Boccone P, et al. Impact of a risk-based follow-up in patients affected by gastrointestinal stromal tumour. Eur J Cancer 2017;78:122–132.
41. Amin MB, Edge S, Greene F, et al., eds. AJCC Cancer Staging Manual. 8th ed. New York: Springer; 2017.
42. Hohenberger P, Ronellenfitsch U, Oladeji O, et al. Pattern of recurrence in patients with ruptured primary gastrointestinal stromal tumour. Br J Surg 2010;97(12):1854–1859.
43. Jati A, Tatli S, Morgan JA, et al. Imaging features of bone metastases in patients with gastrointestinal stromal tumors. Diagn Interv Radiol 2012;18(4):391–396.
44. Kaemmer DA, Otto J, Lassay L, et al. The gist of literature on pediatric GIST: review of clinical presentation. J Pediatr Hematol Oncol 2009;31(2):108–112.
45. Kim SY, Janeway K, Pappo A. Pediatric and wild-type gastrointestinal stromal tumor: new therapeutic approaches. Curr Opin Oncol 2010;22(4):347–350.
46. Bruno M, Carucci P, Repici A, et al. The natural history of gastrointestinal subepithelial tumors arising from muscularis propria: an endoscopic ultrasound survey. J Clin Gastroenterol 2009;43(9):821–825.
47. Gill KR, Camellini L, Conigliaro R, et al. The natural history of upper gastrointestinal subepithelial tumors: a multicenter endoscopic ultrasound survey. J Clin Gastroenterol 2009;43(8):723–726.
48. Lim YJ, Son HJ, Lee JS, et al. Clinical course of subepithelial lesions detected on upper gastrointestinal endoscopy. World J Gastroenterol 2010;16(4):439–444.
49. Karakousis GC, Singer S, Zheng J, et al. Laparoscopic versus open gastric resections for primary gastrointestinal stromal tumors (GISTs): a size-matched comparison. Ann Surg Oncol 2011;18(6):1599–1605.
50. Catena F, Di Battista M, Fusaroli P, et al. Laparoscopic treatment of gastric GIST: report of 21 cases and literature’s review. J Gastrointest Surg 2008;12(3):561–568.
51. Bischof DA, Kim Y, Dodson R, et al. Open versus minimally invasive resection of gastric GIST: a multi- institutional analysis of short- and long-term outcomes. Ann Surg Oncol 2014;21(9):2941–2948.
52. Ohtani H, Maeda K, Noda E, et al. Meta-analysis of laparoscopic and open surgery for gastric gastrointestinal stromal tumor. Anticancer Res 2013;33(11):5031–5041.
53. Frankel TL, Chang AE, Wong SL. Surgical options for localized and advanced gastrointestinal stromal tumors. J Surg Oncol 2011;104(8):882–887.
54. DeMatteo RP, Lewis JJ, Leung D, et al. Two hundred gastrointestinal stromal tumors: recurrence patterns and prognostic factors for survival. Ann Surg 2000;231(1):51–58.
55. McCarter MD, Antonescu CR, Ballman KV, et al. Microscopically positive margins for primary gastrointestinal stromal tumors: analysis of risk factors and tumor recurrence. J Am Coll Surg 2012;215(1):53–59.
56. Zhi X, Jiang B, Yu J, et al. Prognostic role of microscopically positive margins for primary gastrointestinal stromal tumors: a systematic review and meta- analysis. Sci Rep 2016;6:21541.
57. Gronchi A, Bonvalot S, Poveda A, et al. Microscopic surgical margins and out come in localized GIST treated within the EORTC STBSG, AGITG, UNICANCER, FSG, ISG, and GEIS randomized trial on adjuvant imatinib. Paper presented at: 22nd Connective Tissue Oncology Society Meeting; November 2017; Maui, HI.
58. Jakob J, Mussi C, Ronellenfitsch U, et al. Gastrointestinal stromal tumor of the rectum: results of surgical and multimodality therapy in the era of imatinib. Ann Surg Oncol 2013;20(2):586–592.
59. Fiore M, Raut CP, Gronchi A. Are we allowed to limit surgical aggressiveness with small rectal gastrointestinal stromal tumors? Ann Surg Oncol 2017;24(5):1153–1156.
60. Casali PG, Le Cesne A, Poveda Velasco A, et al. Time to definitive failure to the first tyrosine kinase inhibitor in localized GI stromal tumors treated with imatinib as an adjuvant: a European Organisation for Research and Treatment of Cancer Soft Tissue and Bone Sarcoma Group Intergroup randomized trial in collaboration with the Australasian Gastro-Intestinal Trials Group, UNICANCER, French Sarcoma Group, Italian Sarcoma Group, and Spanish Group for Research on Sarcomas. J Clin Oncol 2015;33(36):4276–4283.
61. Mussi C, Schildhaus HU, Gronchi A, et al. Therapeutic consequences from molecular biology for gastrointestinal stromal tumor patients affected by neurofibromatosis type 1. Clin Cancer Res 2008;14(14):4550–4555.
62. Dematteo RP, Ballman KV, Antonescu CR, et al. Adjuvant imatinib mesylate after resection of localised, primary gastrointestinal stromal  tumour:  a  randomised,  double-blind,  placebo-controlled    Lancet 2009;373(9669):1097–1104.
63. Joensuu H, Eriksson M, Sundby Hall K, et al. Adjuvant imatinib for high-risk GI stromal tumor: analysis of a randomized trial. J Clin Oncol 2016;34(3):244–250.
64. Benjamin RS, Casali PG. Adjuvant imatinib for GI stromal tumors: when and for how long? J Clin Oncol 2016;34(3):215–218.
65. Blanke CD, DeMatteo RP. Duration of adjuvant therapy for patients with gastrointestinal stromal tumors: where is goldilocks when we need her? JAMA Oncol 2016;2(6):721–722.
66. Raut CP, Espat NJ, Maki RG, et al. Adjuvant imatinib (IM) for patients (pts) with primary gastrointestinal stromal tumor (GIST) at significant risk of recurrence: PERSIST-5 planned 3-year interim analysis. J Clin Oncol 2015;33(15 Suppl):10537.
67. Joensuu H, Wardelmann E, Sihto H, et al. Effect of KIT and PDGFRA mutations on survival in patients with gastrointestinal stromal tumors treated with adjuvant imatinib: an exploratory analysis of a randomized clinical trial. JAMA Oncol 2017;3(5):602–609.
68. Gastrointestinal Stromal Tumor Meta-Analysis Group. Comparison of two doses of imatinib for the treatment of unresectable or metastatic gastrointestinal stromal tumors: a meta-analysis of 1,640 patients. J Clin Oncol 2010;28(7):1247–1253.
69. Fiore M, Palassini E, Fumagalli E, et al. Preoperative imatinib mesylate for unresectable or locally advanced primary gastrointestinal stromal tumors (GIST). Eur J Surg Oncol 2009;35(7):739–745.
70. Rutkowski P, Gronchi A, Hohenberger P, et al. Neoadjuvant imatinib in locally advanced gastrointestinal stromal tumors (GIST): the EORTC STBSG experience. Ann Surg Oncol 2013;20(9):2937–2943.
71. Koontz MZ, Visser BM, Kunz PL. Neoadjuvant imatinib for borderline resectable GIST. J Natl Compr Canc Netw 2012;10(12):1477–1482.
72. Wang D, Zhang Q, Blanke CD, et al. Phase II trial of neoadjuvant/adjuvant imatinib mesylate for advanced primary and metastatic/recurrent operable gastrointestinal stromal tumors: long-term follow-up results of Radiation Therapy Oncology Group 0132. Ann Surg Oncol 2012;19(4):1074–1080.
73. Tirumani SH, Shinagare AB, Jagannathan JP, et al. Radiologic assessment of earliest, best, and plateau response of gastrointestinal stromal tumors to neoadjuvant imatinib prior to successful surgical resection. Eur J Surg Oncol 2014;40(4):420–428.
74. Joensuu H, Roberts PJ, Sarlomo-Rikala M, et al. Effect of the tyrosine kinase inhibitor STI571 in a patient with a metastatic gastrointestinal stromal tumor. N Engl J Med 2001;344(14):1052–1056.
75. Blanke CD, Rankin C, Demetri GD, et al. Phase III randomized, intergroup trial assessing imatinib mesylate at two dose levels in patients with unresectable or metastatic gastrointestinal stromal tumors expressing the kit receptor tyrosine kinase: S0033. J Clin Oncol 2008;26(4):626–632.
76. Casali PG, Zalcberg J, Le Cesne A, et al.; for European Organisation for Research and Treatment of Cancer Soft Tissue and Bone Sarcoma Group, Italian Sarcoma Group, and Australasian Gastrointestinal Trials Group. Ten-year progression-free and overall survival in patients with unresectable or metastatic GI stromal tumors: long-term analysis of the European Organisation for Research and Treatment of Cancer, Italian Sarcoma Group, and Australasian Gastrointestinal Trials Group Intergroup phase III randomized trial on imatinib at two dose levels. J Clin Oncol 2017;35(15):1713–1720.
77. Le Cesne A, Ray-Coquard I, Bui BN, et al.; for French Sarcoma Group. Discontinuation of imatinib in patients with advanced gastrointestinal stromal tumours after 3 years of treatment: an open-label multicentre randomised phase 3 trial. Lancet Oncol 2010;11(10):942–949.
78. Van Glabbeke M, Verweij J, Casali PG, et al. Initial and late resistance to imatinib in advanced gastrointestinal stromal tumors are predicted by different prognostic factors: a European Organisation for Research and Treatment of Cancer-Italian Sarcoma Group-Australasian Gastrointestinal Trials Group study. J Clin Oncol 2005;23(24):5795–5804.
79. Heinrich MC, Owzar K, Corless CL, et al. Correlation of kinase genotype and clinical outcome in the North American Intergroup Phase III Trial of imatinib mesylate for treatment of advanced gastrointestinal stromal tumor: CALGB 150105 Study by Cancer and Leukemia Group B and Southwest Oncology Group. J Clin Oncol 2008;26(33):5360–5367.
80. Heinrich MC, Rankin C, Blanke CD, et al. Correlation of long-term results of imatinib in advanced gastrointestinal stromal tumors with next-generation sequencing results: analysis of phase 3 SWOG Intergroup Trial S0033. JAMA Oncol 2017;3(7):944–952.
81. Janeway KA, Albritton KH, Van Den Abbeele AD, et al. Sunitinib treatment in pediatric patients with advanced GIST following failure of imatinib. Pediatr Blood Cancer 2009;52(7):767–771.
82. Rutkowski P, Magnan H, Chou AJ, et al. Treatment of gastrointestinal stromal tumours in paediatric and young adult patients with sunitinib: a multicentre case series. BMC Cancer 2017;17(1):717.
83. Ben-Ami E, Barysauskas CM, von Mehren M, et al. Long-term follow-up results of the multicenter phase II trial of regorafenib in patients with metastatic and/or unresectable GI stromal tumor after failure of standard tyrosine kinase inhibitor therapy. Ann Oncol 2016;27(9):1794–1799.
84. Heinrich MC, Griffith D, McKinley A, et al. Crenolanib inhibits the drug- resistant PDGFRA D842V mutation associated with imatinib-resistant gastrointestinal stromal tumors. Clin Cancer Res 2012;18(16):4375–4384.
85. Evans EK, Gardino AK, Kim JL, et al. A precision therapy against cancers driven by KIT/PDGFRA mutations. Sci Transl Med 2017;9(414):eaao1690.
86. Patrikidou A, Chabaud S, Ray-Coquard I, et al. French Sarcoma Group. Influence of imatinib interruption and rechallenge on the residual disease in patients with advanced GIST: results of the BFR14 prospective French Sarcoma Group randomised, phase III trial. Ann Oncol 2013;24(4):1087–1093.
87. Joensuu H, Trent JC, Reichardt P. Practical management of tyrosine kinase inhibitor-associated side effects in GIST. Cancer Treat Rev 2011;37(1):75–88.
88. DeMatteo RP, Maki RG, Singer S, et al. Results of tyrosine kinase inhibitor therapy followed by surgical resection for metastatic gastrointestinal stromal tumor. Ann Surg 2007;245(3):347–352.
89. Sym SJ, Ryu MH, Lee JL, et al. Surgical intervention following imatinib treatment in patients with advanced gastrointestinal stromal tumors (GISTs). J Surg Oncol 2008;98(1):27–33.
90. Raut CP, Wang Q, Manola J, et al. Cytoreductive surgery in patients with metastatic gastrointestinal stromal tumor treated with sunitinib malate. Ann Surg Oncol 2010;17(2):407–415.
91. Mussi C, Ronellenfitsch U, Jakob J, et al. Post-imatinib surgery in advanced/metastatic GIST: is it worthwhile in all patients? Ann Oncol 2010;21(2):403–408.
92. Du CY, Zhou Y, Song C, et al. Is there a role of surgery in patients with recurrent or metastatic gastrointestinal stromal tumours responding to imatinib: a prospective randomised trial in China. Eur J Cancer 2014;50(10):1772– 1778.
93. Bamboat ZM, DeMatteo RP. Metastasectomy for gastrointestinal stromal tumors. J Surg Oncol 2014;109(1):23– 27.
94. Gramza AW, Corless CL, Heinrich MC. Resistance to tyrosine kinase inhibitors in gastrointestinal stromal tumors. Clin Cancer Res 2009;15(24):7510–7518.
95. Wang WL, Conley A, Reynoso D, et al. Mechanisms of resistance to imatinib and sunitinib in gastrointestinal stromal tumor. Cancer Chemother Pharmacol 2011;67(Suppl 1):S15–S24.
96. Liegl B, Kepten I, Le C, et al. Heterogeneity of kinase inhibitor resistance mechanisms in GIST. J Pathol 2008;216(1):64–74.
97. Wardelmann E, Merkelbach-Bruse S, Pauls K, et al. Polyclonal evolution of multiple secondary KIT mutations in gastrointestinal stromal tumors under treatment with imatinib mesylate. Clin Cancer Res 2006;12(6):1743–1749.
98. Shankar S, vanSonnenberg E, Desai J, et al. Gastrointestinal stromal tumor: new nodule-within-a-mass pattern of recurrence after partial response to imatinib mesylate. Radiology 2005;235(3):892–898.
99. Fairweather M, Balachandran VP, Li GZ, et al. Cytoreductive surgery for metastatic gastrointestinal stromal tumors treated with tyrosine kinase inhibitors: a 2-institutional analysis. Ann Surg 2017 [Epub ahead of print].
100. Judson I, Ma P, Peng B, et al. Imatinib pharmacokinetics in patients with gastrointestinal stromal tumour: a retrospective population pharmacokinetic study over time. EORTC Soft Tissue and Bone Sarcoma Group. Cancer Chemother Pharmacol 2005;55(4):379–386.
101. Demetri GD, Wang Y, Wehrle E, et al. Imatinib plasma levels are correlated with clinical benefit in patients with unresectable/metastatic gastrointestinal stromal tumors. J Clin Oncol 2009;27(19):3141–3147.
102. Zalcberg JR, Verweij J, Casali PG, et al.; for EORTC Soft Tissue and Bone Sarcoma Group, the Italian Sarcoma Group; Australasian Gastrointestinal Trials Group. Outcome of patients with advanced gastro-intestinal stromal tumours crossing over to a daily imatinib dose of 800 mg after progression on 400 mg. Eur J Cancer 2005;41(12):1751–1757.
103. Demetri GD, van Oosterom AT, Garrett CR, et al. Efficacy and safety of sunitinib in patients with advanced gastrointestinal stromal tumour after failure of imatinib: a randomised controlled trial. Lancet 2006;368(9544):1329–1338.
104. Heinrich MC, Maki RG, Corless CL, et al. Primary and secondary kinase genotypes correlate with the biological and clinical activity of sunitinib in imatinib-resistant gastrointestinal stromal tumor. J Clin Oncol 2008;26(33):5352–5359.
105. Reichardt P, Demetri GD, Gelderblom H, et al. Correlation of KIT and PDGFRA mutational status with clinical benefit in patients with gastrointestinal stromal tumor treated with sunitinib in a worldwide treatment-use trial. BMC Cancer 2016;16:22.
106. George S, Blay JY, Casali PG, et al. Clinical evaluation of continuous daily dosing of sunitinib malate in patients with advanced gastrointestinal stromal tumour after imatinib failure. Eur J Cancer 2009;45(11):1959–1968.
107. Demetri GD, Reichardt P, Kang YK, et al. Efficacy and safety of regorafenib for advanced gastrointestinal stromal tumours after failure of imatinib and sunitinib (GRID): an international, multicentre, randomised, placebo- controlled, phase 3 trial. Lancet 2013;381(9863):295–302.
108. Grothey A, George S, van Cutsem E, et al. Optimizing treatment outcomes with regorafenib: personalized dosing and other strategies to support patient care. Oncologist 2014;19(6):669–680.
109. Kang YK, Ryu MH, Yoo C, et al. Resumption of imatinib to control metastatic or unresectable gastrointestinal stromal tumours after failure of imatinib and sunitinib (RIGHT): a randomised, placebo-controlled, phase 3 trial. Lancet Oncol 2013;14(12):1175–1182.
110. Szucs Z, Thway K, Fisher C, et al. Promising novel therapeutic approaches in the management of gastrointestinal stromal tumors. Future Oncol 2017;13(2):185–194.
111. Demetri GD, Jeffers M, Reichardt P, et al. Mutational analysis of plasma DNA from patients (pts) in the phase III GRID study of regorafenib (REG) versus placebo (PL) in tyrosine kinase inhibitor (TKI)-refractory GIST: correlating genotype with clinical outcomes. J Clin Oncol 2013;31:10503.
112. Bertucci F, Finetti P, Mamessier E, et al. PDL1 expression is an independent prognostic factor in localized GIST. Oncoimmunology 2015;4(5):e1002729.
113. Balachandran VP, Cavnar MJ, Zeng S, et al. Imatinib potentiates antitumor T cell responses in gastrointestinal stromal tumor through the inhibition of Ido. Nat Med 2011;17(9):1094–1100.
114. Benjamin RS, Debiec-Rychter M, Le Cesne A, et al. Gastrointestinal stromal tumors II: medical oncology and tumor response assessment. Semin Oncol 2009;36(4):302–311.
115. Dimitrakopoulou-Strauss A, Ronellenfitsch U, Cheng C, et al. Imaging therapy response of gastrointestinal stromal tumors (GIST) with FDG PET, CT and MRI: a systematic review. Clin Transl Imaging 2017;5(3):183–197.
116. Tirumani SH, Jagannathan JP, Krajewski KM, et al. Imatinib and beyond in gastrointestinal stromal tumors: a radiologist’s perspective. AJR Am J Roentgenol 2013;201(4):801–810.
117. Van den Abbeele AD. The lessons of GIST—PET and PET/CT: a new paradigm for imaging. Oncologist 2008;13(Suppl 2):8–13.
118. Eisenhauer EA, Therasse P, Bogaerts J, et al. New response evaluation criteria in solid tumours: revised RECIST guideline (version 1.1). Eur J Cancer 2009;45(2):228–247.
119. Choi H, Charnsangavej C, Faria SC, et al. Correlation of computed tomography and positron emission tomography in patients with metastatic gastrointestinal stromal tumor treated at a single institution with imatinib mesylate: proposal of new computed tomography response criteria. J Clin Oncol 2007;25(13):1753–1759.