Current Molecular Profile of Gastrointestinal Stromal Tumors and Systemic Therapeutic Implications

Simple Summary Gastrointestinal stromal tumors (GISTs) are malignant mesenchymal tumors classified primarily as soft tissue sarcomas and characteristically arise from the intestinal pacemaker cells of Cajal responsible for the gastrointestinal tract motility. They comprise a heterogenous group of tumors due to a variety of molecular alterations, mostly KIT (60–70%) or PDGFRA (10–15%) mutations, but also more rare alterations, including inactivation of the NF1 gene, mutations in the succinate dehydrogenase (SDH), BRAF, and RAS genes, and also gene fusions. Here, we review the most recent data on the molecular profile of GISTs and highlight systemic therapeutic implications according to distinct GIST molecular subtypes. Abstract Gastrointestinal stromal tumors (GISTs) are malignant mesenchymal tumors arising from the intestinal pacemaker cells of Cajal. They compose a heterogenous group of tumors due to a variety of molecular alterations. The most common gain-of-function mutations in GISTs are either in the KIT (60–70%) or platelet-derived growth factor receptor alpha (PDGFRA) genes (10–15%), which are mutually exclusive. However, a smaller subset, lacking KIT and PDGFRA mutations, is considered wild-type GISTs and presents distinct molecular findings with the activation of different proliferative pathways, structural chromosomal and epigenetic changes, such as inactivation of the NF1 gene, mutations in the succinate dehydrogenase (SDH), BRAF, and RAS genes, and also NTRK fusions. Currently, a molecular evaluation of GISTs is imperative in many scenarios, aiding in treatment decisions from the (neo)adjuvant to the metastatic setting. Here, we review the most recent data on the molecular profile of GISTs and highlight therapeutic implications according to distinct GIST molecular subtypes.


Introduction
Gastrointestinal stromal tumors (GISTs) are malignant mesenchymal tumors classified primarily as soft tissue sarcomas [1] and characteristically arise from the intestinal pacemaker cells of Cajal responsible for the gastrointestinal tract motility [2]. GISTs are the most frequent type of sarcomas and also the most frequent sarcoma of the gastrointestinal tract [3]. Most GISTs are sporadic, solitary tumors, and more are routinely found in the stomach or small intestines (60-65% and 20-35% of cases, respectively) [2]. Primary GISTs of the rectum can be found in 3-5% of cases, and an even smaller portion of cases can be found in the colon, esophagus, and peritoneal cavity. The most common metastatic sites are the liver and peritoneal cavity, mostly due to anatomic contiguity [4].
Given the different subtypes of GISTs, we recommend the following stepwise fashion of testing ( Figure 1) whenever systemic treatment is indicated and next-generation sequencing (NGS) is not promptly available.
Conversely, the group of SDH-deficient GISTs includes wild-type GISTs in association with CT, CSS, and sporadic pediatric and young adult GISTs.
Given the different subtypes of GISTs, we recommend the following stepwise fashion of testing ( Figure 1) whenever systemic treatment is indicated and next-generation sequencing (NGS) is not promptly available. Legend: Initial test for the presence of KIT and PDGFRA mutations. If negative, test for SDH status by IHC. If SDH-proficient, patients should be tested for other genetic alterations. If SDH-deficient, test for mutations in SDH A/B/C/D. When mutated, consider germline testing for CSS. If not mutated, consider SDH C promoter methylation testing.

SDH-Competent GIST with either KIT or PDGFRA Mutation
KIT codifies the c-Kit protein (also known as stem cell factor (SCF) receptor or CD117), a receptor tyrosine kinase (RTK) with five Ig-like extracellular (EC) domains (therefore, class III RTK) with 145 kDa and 976 amino acid residues in normal splicing [22]. In humans, four isoforms of c-Kit have been described so far, with the GNNK-negative isoform showing greater kinase activity and downstream signaling. The first three EC domains are essential for SCF binding and the other two for receptor homodimerization. The EC domains are connected by a single spanning transmembrane helix to the intracellular juxtamembrane (JM) domain, which works as a modulator of the signal transduction. The tyrosine kinase domain is split in two functional lobes (N-terminal and C-terminal), and in the enzyme inactive state, the JM domain inserts between the two lobes, sterically blocking the conformational changes needed for the activation loop of the catalytic unit to assume its active form. Finally, following the C-terminal lobe of the tyrosine kinase domain is the C-terminal intra-cellular tail. Upon dimeric SCF binding to adjacent c-Kit molecules, reorientation of D4-5 EC domains leads to receptor homodimerization. The following conformational changes lead to sequential trans-phosphorylations in the JM domain, activation loop, kinase insert region, and C-terminal tail [22]. These phosphorylated Legend: Initial test for the presence of KIT and PDGFRA mutations. If negative, test for SDH status by IHC. If SDH-proficient, patients should be tested for other genetic alterations. If SDH-deficient, test for mutations in SDH A/B/C/D. When mutated, consider germline testing for CSS. If not mutated, consider SDH C promoter methylation testing.

SDH-Competent GIST with either KIT or PDGFRA Mutation
KIT codifies the c-Kit protein (also known as stem cell factor (SCF) receptor or CD117), a receptor tyrosine kinase (RTK) with five Ig-like extracellular (EC) domains (therefore, class III RTK) with 145 kDa and 976 amino acid residues in normal splicing [22]. In humans, four isoforms of c-Kit have been described so far, with the GNNK-negative isoform showing greater kinase activity and downstream signaling. The first three EC domains are essential for SCF binding and the other two for receptor homodimerization. The EC domains are connected by a single spanning transmembrane helix to the intra-cellular juxtamembrane (JM) domain, which works as a modulator of the signal transduction. The tyrosine kinase domain is split in two functional lobes (N-terminal and C-terminal), and in the enzyme inactive state, the JM domain inserts between the two lobes, sterically blocking the conformational changes needed for the activation loop of the catalytic unit to assume its active form. Finally, following the C-terminal lobe of the tyrosine kinase domain is the C-terminal intra-cellular tail. Upon dimeric SCF binding to adjacent c-Kit molecules, reorientation of D4-5 EC domains leads to receptor homodimerization. The following conformational changes lead to sequential trans-phosphorylations in the JM domain, activation loop, kinase insert region, and C-terminal tail [22]. These phosphorylated tyrosine residues serve as substrate docking sites for signal transduction. The following signaling pathways are activated by c-Kit: mitogen-activated protein kinase (MAP kinase), phosphatidylinositol 3-kinase/protein kinase B (PI3K/AKT), phospholipase C gamma (PLC-gamma), Janus kinase/signal transducer and activator of transcription (JAK/STAT), and Scr kinase pathways [22,23]. The specific phosphorylated tyrosine residues determine which signaling pathways are activated.
Primary KIT mutations occur in roughly 70% of GISTs [24]. These gain-of-function mutations lead to constitutive activation of the receptor and are commonly found in two hot spots ( Figure 2): membrane-proximal EC domain (exon 9; 9-10% of all GISTs) and intra-cellular JM domain (exon 11; 60% of all GISTS). GISTs with exon 11 mutations can occur throughout the gastrointestinal tract, and those with exon 9 mutations arise primarily in the small or large bowel [2]. Importantly, the prognosis is dependent on the specific type of mutation within exon 11, with tumors displaying deletions in codons 557 and 558 presenting more aggressive clinical behavior. Primary mutations in exons 8 (membraneproximal EC domain), 13 (ATP-binding site), and 17 (activation loop) are rare (<1% of all GISTs) [24].
Another important and common molecular alteration and drive mutation in GISTs is PDFGA mutation. As the KIT gene, PDGFRA is located on chromosome 4, encodes a type III RTK, and is homologous to KIT, presenting a similar structure and downstream pathways, such as RAS/RAF/MAPK and PI3K [14,25,26]. Importantly, KIT and PDGFRA mutations are considered to be mutually exclusive. PDGFRA mutations, once present, trigger a ligand-independent phosphorylation and uninhibited cellular proliferation [15].
PDGFRA mutations are mainly encountered in exon 18 (activation loop of the receptor; 13-14%) and less frequently in exon 12 (JM domain; 0.6-2%) or exon 14 (ATP-binding site; 1.4%) [2] (Figure 2). The PDGFRA exon 18 D842V missense mutation can be found in around 50-70% of PDFRA-mutated GISTs and 8% of all GISTs [27] and promotes the stabilization of the kinase conformation in its active form, therefore conferring resistance of this molecular subtype to the most commonly used treatments [28,29]. The majority of PDFGRA-mutated GISTs arise from the stomach (15-18%) and the small intestine (5-7%). They classically have a more indolent behavior [2] and favorable prognosis compared to the more commonly encountered KIT mutations.

Therapeutic Implications of GIST with KIT or PDGFRA Mutations
The main therapeutic strategy to treat GISTs with either KIT or PDGFRA mutations has been the use of TKIs, especially imatinib. Initially developed to target the BCR-ABL translocation of chronic myelogenous leukemia [30], it was shown to be active against most GISTs [31]. Thus, imatinib is considered one of the first targeted therapies for the treatment of solid tumors.
The initial treatment of high-risk (>50% recurrence risk) GISTs in the adjuvant setting is considered to be a daily 400 mg dose of imatinib for a period of ≥3 years [32][33][34]. The recommended duration of treatment was defined based on evidence that, with 3 years compared with 1 year of therapy, recurrence-free survival (RFS) at 5 years was 71.1% versus 52.3% and overall survival (OS) was 91.9% versus 82.3%, respectively [32]. However, studies have been conducted aiming to define the appropriate duration of adjuvant treatment with imatinib and possibly expanding the duration for a period of 5 years [35], and yet, other studies are underway in order to determine the efficacy of extending the duration of adjuvant treatment (NCT02413736, NCT02260505).
In the advanced/metastatic disease setting, imatinib is started at a 400 mg daily dose, thus achieving response rates and progression-free survival (PFS) rates of up to 40 months and 70%, respectively, in patients with the most common KIT and PDGFRA mutations [36][37][38][39][40]. However, regardless of such outcomes with first-line imatinib, subsequent lines of therapy present only modest benefits, and wild-type KIT/PDGFRA GISTs typically present with early progression with imatinib [36,41].

Therapeutic Implications of GIST with KIT or PDGFRA Mutations
The main therapeutic strategy to treat GISTs with either KIT or PDGFRA mutations has been the use of TKIs, especially imatinib. Initially developed to target the BCR-ABL translocation of chronic myelogenous leukemia [30], it was shown to be active against most GISTs [31]. Thus, imatinib is considered one of the first targeted therapies for the treatment of solid tumors.
The initial treatment of high-risk (> 50% recurrence risk) GISTs in the adjuvant setting is considered to be a daily 400 mg dose of imatinib for a period of ≥3 years [32][33][34]. The recommended duration of treatment was defined based on evidence that, with 3 years compared with 1 year of therapy, recurrence-free survival (RFS) at 5 years was 71.1% versus 52.3% and overall survival (OS) was 91.9% versus 82.3%, respectively [32]. However, studies have been conducted aiming to define the appropriate duration of adjuvant treatment with imatinib and possibly expanding the duration for a period of 5 years [35], and yet, other studies are underway in order to determine the efficacy of extending the duration of adjuvant treatment (NCT02413736, NCT02260505).
In the advanced/metastatic disease setting, imatinib is started at a 400 mg daily dose, thus achieving response rates and progression-free survival (PFS) rates of up to 40 months and 70%, respectively, in patients with the most common KIT and PDGFRA mutations [36][37][38][39][40]. However, regardless of such outcomes with first-line imatinib, subsequent lines of therapy present only modest benefits, and wild-type KIT/PDGFRA GISTs typically present with early progression with imatinib [36,41].
Tumors with deletions of codons 557 and 558 present more aggressive clinical behavior than those with exon 11 missense mutations. However, such mutations display significant responses to imatinib. Furthermore, patients' genetics seems to modify the chance Tumors with deletions of codons 557 and 558 present more aggressive clinical behavior than those with exon 11 missense mutations. However, such mutations display significant responses to imatinib. Furthermore, patients' genetics seems to modify the chance of tumor response to imatinib. Among patients with exon 11 mutant GISTs, differences in clinical response to imatinib have been shown to be associated with single-nucleotide polymorphisms in genes related to the absorption, distribution, metabolism, and excretion (ADME) process [42]. GISTs with exon 9 typically present lower response to imatinib when compared to those with exon 11 mutations. Nonetheless, in the metastatic setting, this can be partially circumvented by doubling the dose of imatinib (800 mg/day). Data from randomized trials suggest that the higher dose of imatinib is associated with an improved objective response rate (47 vs. 21%) and PFS (HR = 0.57; p = 0.017) [39]. While some GISTs with primary exon 13 mutations are sensitive to imatinib, most primary exon 17 mutations confer primary resistance do imatinib [2].
Most PDGFRA mutations are sensitive to imatinib therapy (exons 12, 14, and 18). However, PDGFRA exon 18 D842V mutation can also present with primary resistance to imatinib [41], as well as other TKIs [14,24,43]. Yet, due to the more indolent nature of this GIST subtype, patients may present with stable disease while on therapy [40].
Of note, around 10% of patients diagnosed with advanced GIST can present with primary resistance to imatinib with disease progression within the first 6 months of treatment. This primary resistance to treatment is linked to the GIST genotype [36,38,39,43] and is found to be more frequent in KIT exon 9 mutations and in wild-type KIT/PDGFRA GISTs.
Acquired KIT and PDGFRA mutations occur in up to 90% of patients with metastatic GIST treated with imatinib and are the main cause of secondary resistance [44]. In most situations, imatinib therapy induces tumor volume reduction upon inducing cellular apoptosis. However, a fraction of tumor cells enters a quiescent, non-proliferative state and may acquire secondary genetic mutations, rendering such cells imatinib-resistant, leading to disease progression. Such mutations confer an evolutionary advantage under the selective pressure of imatinib exposure and cluster into two hot regions of the KIT gene: the ATPbinding pocket (exons 13 and 14 of KIT and exon 14 of PDGFRA) and the activation loop (exons 17 and 18 of KIT and exon 18 of PDGFRA) [2,[45][46][47]. Therefore, treatment sequencing with other TKIs becomes necessary to overcome disease progression due to such mutations. c-Kit molecules with secondary mutations in the ATP-binding pocket are relatively sensitive to sunitinib [48], ponatinib [49], ripretinib [17,50], and avapritinib [51,52], but not to regorafenib or sorafenib [2], as shown in Table 1. Conversely, receptors carrying acquired mutations in the activation loop can be targeted with regorafenib [53], sorafenib [54], nilotinib [55], ponatinib [49], dasatinib (NCT01643278), ripretinib [50], or avapritinib [51]. The one exception is the D816V mutation in exon 17, which is resistant to all TKIs with the exception of ponatinib, ripretinib, and avapritinib [44,56,57].
The PDGFRA-mutant GISTs' resistance mechanism profile is less detailed compared to KIT, and little is known beyond primary or secondary PDGFRA D482V mutations [58]. However, D842V has been shown to be homologues to the KIT exon 17 D816V mutation [24] and, as such, demonstrates a similar treatment sensitivity profile with resistance to the majority of the approved TKIs for the treatment of advanced GISTs, harboring a very poor prognosis. Yet, in 2020, the NAVIGATOR trial demonstrated the activity of avapritinib, a fourth-generation TKI designed specifically to inhibit PDGFRA D842V in advanced GISTs with such mutation with an objective response rate of 84% (7% complete responses) [51]. However, resistance mechanisms to avapritinib in PDGFRA D842V mutated GISTs have been demonstrated and cause avapritinib binding blockage due to substitutions in exons 13, 14, and 15 (V658A, N659K, Y676C, and G680R) [58]. The checkmarks indicate drug sensitivity, and the crosses indicate drug resistance.

SDH-Competent GIST without KIT and PDGFRA Mutation NF1-Mutant GIST
Neurofibromatosis 1 (NF1) is an autosomal dominant disorder caused by inherited or de novo germline mutations within the NF1 tumor suppressor gene. It is clinically characterized by multiple café-au-lait macules, intertriginous freckling, multiple cutaneous neurofibromas, plexiform neurofibromas, malignant peripheral nerve sheath tumors and, sometimes, learning disability or behavioral problems [59,60]. The product of the NF1 gene is the large and multifunctional cytoplasmic protein neurofibromin, which belongs to a family of guanosine-triphosphate-hydrolase (GTPase)-activating proteins that stimulate intrinsic GTPase activity in the Ras family [59].
Patients with germline inactivation of the NF1 gene have a propensity to develop several tumors through acquired inactivation of the normal NF1 allele, and it has been estimated that 7% of individuals with NF1 develop GIST during their lifespan, which corresponds to a 34-fold higher risk than the average population [59,61]. NF1-associated GISTs, which correspond to 1% of all GISTs, are commonly multicentric and predominantly located in the small intestine [62,63] as shown in Table 2. Although most cases of NF1related GISTs are germinative, somatic inactivating NF1 mutations may also occur as an oncogenic mechanism for a small subset of wild-type GISTs [62].
NF1-related GISTs typically express KIT, DOG1, and SDHB by IHC [64]. They also frequently demonstrate loss of heterozygosity (LOH) at 14q and 22q, which may contribute to the early phase of tumor development [65]. In addition, genomic profiling studies of adult GISTs have revealed that NF1 alterations are enriched in GISTs located at the duodenal-jejunal flexure. These tumors may harbor concurrent activating KIT and/or inactivating Notch pathway mutations [63].

Therapeutic Implications of NF1-Mutant GISTs
Since most NF1-mutant GISTs do not harbor KIT mutations, they are typically insensitive to imatinib and surgery remains the mainstay therapy for these patients. Although, so far, there are no data on target-specific drugs for NF1-related GISTs, MEK inhibitors have pointed towards clinical efficacy in other NF1-associated tumors. Given the key role of abnormal RAF/MEK/ERK pathway signaling in neurofibromas, a phase I study of selumetinib (a MEK inhibitor) in children/young adults with NF1-related inoperable plexiform neurofibromas has demonstrated a median maximal decrease of 24% in the volume of the tumors [66]. Currently, a phase II trial with a different MEK inhibitor is ongoing for adolescents and young adults with NF1-associated plexiform neurofibromas (NCT02096471).

Therapeutic Implications of BRAF Mutant GISTs
BRAF V600E mutation in GISTs has been shown to confer resistance to imatinib and sunitinib [71]. Since BRAF V600E mutations are currently considered an agnostic genetic alteration and seem to respond to BRAF inhibitors regardless of the primary tumor type, we recommend a BRAF inhibitor with or without a MEK inhibitor to patients with unresectable or metastatic BRAF mutant GISTs. The combination of dabrafenib (BRAF inhibitor) and trametinib (MEK inhibitor) has recently received FDA approval for any BRAF mutant tumor, regardless of the origin site. The first case of a patient with a BRAF mutant GIST who responded during 8 months to dabrafenib was reported in 2013 [72]. This patient had already failed to receive benefit from imatinib, sunitinib, and a MEK inhibitor.
Interestingly, a complete radiological response with first-line regorafenib was reported in a 51-year-old woman with a BRAF-mutated GIST [73]. Regorafenib has a wide spectrum of action, with MAPK signaling pathway blockade at different levels. There is an ongoing phase II, single-arm, trial evaluating regorafenib in the first-line setting for advanced KIT/PDGFRA wild-type GISTs (NCT02638766).
The role of kinase inhibitors targeting BRAF is yet unknown in the adjuvant setting for this subset of patients. Therefore, we recommend no systemic therapy following R0 resections, even for those with high-risk BRAF mutant disease.

GISTs with NTRK Fusion
TRK fusions arise from aberrant rearrangements of neurotrophic receptor kinase (NTRK) genes 1-3 with several gene partners. They are constitutively active ligandindependent oncogenic drivers in different solid tumors, including GISTs. Until recently, only ETV6-NTRK3 gene fusions were described in GISTs. However, NTRK 1 fusion has been recently reported [74].

Therapeutic Implications of GISTs Harboring NTRK Fusions
NTRK fusions are considered an agnostic feature that renders tumors sensitive to TRK inhibitors independently of the primary cancer. In the metastatic setting, larotrectinib, a TRK inhibitor, demonstrated great activity against solid tumors harboring NTRK fusions [75]. In a pooled analysis from three phase I/II clinical trials, 159 patients with different tumor types harboring NTRK fusions were treated with larotrectinib. The objective response rate was 79%. Indeed, four patients with GISTs were included, and all of them achieved an objective response [75]. In another pooled analysis from three different studies of individuals with NTRK fusions, three patients with GIST were treated with larotrectinib and all responded, including one with complete response [76].

GISTs with FGFR Pathway Alterations
Fibroblast growth factor receptor (FGFR) fusions, mutations, and ligand overexpression represent the most common molecular alterations in quadruple WT GISTs, suggesting a possible pathogenetic role [77].
Although very promising, FGF/FGFR pathway inhibition in GISTs is still far from clinical practice.

Other Rare Mutations
KRAS-mutated GISTs are extremely rare, and the clinicopathologic features still lack complete comprehension [87]. Yet another extremely rare GIST subtype are the PIK3CA mutant GISTs. Currently, the most robust study involving imatinib-naïve GIST patients, 10 (mostly localized tumors) out of 529 GISTs harbored PIK3CA mutations, which were associated with larger and aggressive tumors [88] ]. Given the limited number of patients, further studies with longer follow up are required.

SDH-Deficient GIST
Wild-type GISTs are the primary form of GIST in children (85%) and only occasionally occur in adults (15%) [89]. They are often localized in the stomach, multifocal, with lymph node involvement, indolent, and primarily affect young females [90]. Approximately half of pediatric WT-GISTs are SDH-deficient, which have loss-of-function of the SDH complex subunit (SDHA, SDHB, SDHC, or SDHD). This loss of function occurs either due to a mutation in one of the SDHx genes or due to epigenetic silencing [91,92].
The SDH complex is a mitochondrial complex that participates in the Krebs cycle in the conversion of succinate to fumarate (succinate reductase). Loss of the mitochondrial SDH complex through mutations in SDHA, SDHB, SDHC, or SDHD causes the accumulation of succinate, resulting in the overexpression of hypoxia-inducible factor (HIF) proteins and increased transcription of HIF-1a-regulated genes. It also leads to decreased DNA demethylation. Hence, an increase in DNA methylation and activation of insulin-like growth factor 1 receptor (IGF1R) and vascular endothelial growth factor receptor (VEGFR) are the molecular characteristics of SDH-deficient GISTs [93].
SDH-deficient GISTs have different clinical courses, molecular characteristics, prognoses, and therapeutic responses when compared to other GISTs. SDHA is the most commonly mutated (30%) in all SDH-deficient GISTs [94]. Immunohistochemical analysis for SDHB is appropriate to diagnose SDH deficiency in KIT/PDGFRA wild-type GISTs, as the absence of SDHB can indicate the deficiency of any of the SDHx genes. However, not all SDHB-immunonegative GISTs harbor an SDH gene mutation; these tumors may have other epigenetic and genetic defects in the SDH pathway [8].
Approximately 5% of patients with GISTs have genetic syndromes associated with the development of these tumors, which include primary familial GIST syndrome (heritable mutations in either the KIT or PDGFRA genes), NF1, CSS (SDHx genes), and CT (hypermethylation of the SDHC promoter). CSS, also referred to as the Carney-Stratakis dyad, is a rare inherited predisposition syndrome caused by germline mutations in the SDHB/C or D subunits, which presents with the dyad of GISTs and paragangliomas [95,96]. The Carney triad is an extremely rare nonhereditary syndrome consisting of GISTs, paragangliomas, and pulmonary chondromas caused by epigenetic inactivation (hypermethylation) of the SDHC gene with functional impairment of the SDH complex [95,97,98].
Patients diagnosed with SDH-deficient GISTs and those who have NF1 mutations should be referred for genetic counseling and appropriate germline testing. Furthermore, patients with SDH-deficient GISTs, or known SDH mutations, could have paraganglioma associated (Carney-Stratakis dyad or CT), and therefore, serum/urine catecholamine/metanephrine testing should be considered prior to surgery [99].

Therapeutic Implications of SDH-Deficient GISTs
An observational study evaluated the clinical and molecular features of KIT/PDGFRA WT-GISTs in a cohort (patients < 19 y with GIST or > 19 y with WT-GIST), which was characterized by IHC for SDHB, sequencing of SDH genes, determination of SDHC promoter methylation, as well as germline testing of SDH genes for some of them (consented). Among the 95 patients analyzed (70% female; median age: 23 years; 19% with syndromic GIST), three molecular subtypes were defined: 11 had SDH competency (which included mutations in NF1; BRAF; other rare mutations/fusions; or no identified), and 84 had SDH deficiency (75% due to SDH mutations and 25% due to SDHC promoter hypermethylation). SDH mutations were often germline (82%) among 38 tested patients. In this cohort, despite the non-standardized retrospective assessment of treatment, only 1 of 49 patients treated with imatinib and 4 of 38 patients treated with sunitinib had an objective response [7]. SDH-deficient GIST is an indolent disease, and most patients have a good median survival of approximately 10 years even though with disease progression [100]. This sub-type of GISTs is known to be generally unresponsive to TKIs. In the analysis of molecular alterations associated with the treatment benefit of adjuvant imatinib in the ACOSOG Z9001 study, KIT/PDGFRA WT tumors were uncommon (9 in the placebo and 6 in the imatinib group). As a result, the therapeutic impact of adjuvant imatinib could not be considered definitive in this subset and remains controversial [101]. The National Comprehensive Cancer Network (NCCN) guidelines version 2.2022 [99] and the European Society for Medical Oncology (ESMO) guidelines 2021 [3] do not support routine clinical use of imatinib adjuvant for WT-GISTs. In the advanced scenario, the best way to treat these patients is not established, and personalized treatment is recommended [3]. Although refractory to imatinib, they may have some responsiveness to sunitinib [102], regorafenib [103], Pazopanib [85], Nilotinib, and Linsitinib [104]. However, the data on TKIs in patients with SDH-deficient GISTs is extremely limited. Referral to a clinical trial is a reasonable consideration for them.
Molecular characterization demonstrated that temozolomide (TMZ) elicits DNA damage and apoptosis in SDH-deficient GISTs with a 40% rate of objective responses among five patients treated with this alkylating agent [105]. Based on this early efficacy data, a phase II study of single-agent TMZ (NCT03556384) in patients with SDH-mutant GISTs is ongoing.

Conclusions
Recent advances in molecular profiling of GISTs brought significant value to prognostication and therapeutic decisions in clinical practice. New drugs have been approved for advanced GISTs with either the KIT or PDGFRA mutation. Nonetheless, there are significant unmet needs for the subset of patients with KIT/PDGFRA wild-type GISTs.

Conflicts of Interest:
The authors declare no conflict of interest.