Gastrointestinal stromal tumors (GIST) are the most common malignant sarcomas of the gastrointestinal tract. Inhibitors of the mutated driver proteins KIT and PDGFRA can temporarily control disseminated GIST disease but cannot cure it. Recently, the investigators were able to elucidate the two main resistance mechanisms-the coexistence of ATP-binding pocket (AP) and activation loop (AL) mutations, as well as AP/AL combination mutations. Determining these resistance factors requires methods that are not yet established in routine clinical practice. The investigators hypothesize that identifying these factors and creating resistance profiles in the context of the anatomical location of metastatic lesions can open up new, clinically relevant therapeutic options, including local therapies. The aim of this project is to evaluate the practicability of these methods in everyday clinical use. An AI-supported image analysis and specialized molecular methods will be combined into a comprehensive resistance profile and will be provided to the treating physicians to potentially guide treatment decisions.

Gastrointestinal stromal tumors are among the most common sarcomas and typically harbor activating mutations of the KIT or PDGFR receptor tyrosine kinases [1-3]. While KIT/PDGFRA-directed therapies have revolutionized treatment, most patients eventually progress [4, 5]. Those progressing on the approved last-line therapy (ripretinib or avapritinib) face a dismal outcome with a poor median overall survival of approximately 6 months. The investigators have recently shown that KIT and PDGFRA remain the dominant oncogenic drivers throughout all treatment lines, including in patients failing 4th or later line ripretinib or avapritinib [6, 7]. Strategies that effectively target KIT (and PDGFRA) remain the most promising treatment strategies regardless of treatment line. Our recent findings further deepened our understanding of kinase inhibitor escape which could greatly help to adapt salvage treatments in the future.

Two types of secondary resistance mutations emerge following selective pressure from kinase inhibitors: Mutations involving either the ATP-binding pocket (AP) or the activation loop (AL) of KIT. The investigators have previously shown that sunitinib and ripretinib exhibit a highly differential sensitivity profile with sunitinib being a potent inhibitor of AP mutations and ripretinib being a highly potent inhibitor of AL mutations [6]. A recent substudy of the INTRIGUE trial (2nd-line sunitinib vs ripretinib in GIST) revealed that liquid biopsies can be highly predictive markers of response by determining heterogeneity of secondary resistance [8]. Patients that display secondary resistance mutations restricted to either the AP or the AL subdomain of KIT show marked clinical benefit to a single drug treatment. However, none of these drugs can control both types of mutations at the same time hence AL-mutations are selected for during sunitinib treatment and AP mutations during ripretinib treatment. Knowledge about the status of heterogeneity could inform about the selection of drugs.

In addition, a novel type of resistance, AP/AL mutations, representing compound mutations that involve both the ABP and the AL on the same allele (in cis) were recently described [7]. Next generation sequencing (NGS) panels fail to detect these mutations in clinical practice, as their identification requires allele-specific, long-read sequencing. AP/AL mutations were observed in half of all patients progressing on ripretinib and preclinical models showed that these mutations confer a highly resistant phenotype that cannot be controlled by any approved drug or any drug currently tested in clinical trials. Notably, the same is true in PDGFRA-mutant GIST with the common D842V primary (AL-) mutation where secondary mutations typically result in AP/AL mutations [6]. Local treatments (radiotherapy, surgery, ablation) currently represent the only approach to counter these hyper-resistant clones.

The investigators conclude that two fundamental mechanisms determine resistance in refractory GIST:

1. Intra-patient heterogeneity of secondary resistance defined by the presence of separate clones harboring ATP-binding pocket (AP) and activation loop (AL) mutations which cannot be inhibited by mono-therapy [8]
2. Compound-mutations involving secondary and tertiary mutations that involve the AP and the AL in cis. [6, 7]

The investigators hypothesize that implementing these recent research insights into a resistance profiler could help to develop adaptive treatment strategies in the future, particularly utilizing local treatment approaches, in patients currently lacking other treatment options. This image-based visualization platform aims to integrate longitudinal image analysis, functional imaging at time of inclusion, treatment-related parameters as well as liquid and tumor biopsies. The combined information will provide a visualization of TKI resistance in a more accessible, anatomically annotated way (as opposed to a pure paper-report of multiple biopsies) which will then be provided to a regular sarcoma expert tumorboard

Inclusion Criteria:

• Advanced gastrointestinal stromal tumors history of exposure and progression or intolerance to imatinib, sunitinib, regorafenib and ripretinib
• Clinical indication for PET-CT and tumor biopsy (optional) (sarcoma center tumorboard protocol)
• Age 18 years and older
• FFPE tumor tissue available
• Previous imaging studies available (CT/MRI/PET Abdomen, ideally at baseline, best response and progression)
• ECOG ≤ 2
• Progressing tumor lesion(s) of which one is accessible for biopsy

Exclusion Criteria:

• contraindication for systemic TKI treatment
• contraindication for contrast-enhanced both CT and MRI imaging.
• GIST without KIT or PDGFRA mutations
• pregnant or breastfeeding women
• patient that - to the investigator's discretion - are not able or willing to comply with the provided protocol and visit schemes

Patients with advanced gastrointestinal stromal tumors history of exposure and progression or intolerance to imatinib, sunitinib, regorafenib and ripretinib