Gastrointestinal stromal tumors: assessment of mutations in tumors and in circulating tumor DNA and measurement of TKI plasma exposure to optimize treatment

Gastrointestinal stromal tumors (GISTs) belong to the sarcoma group and are
characterized by oncogenic mutations in the c-KIT, PDGFRA, BRAF and NF-1 genes
that drive tumor growth. Since tyrosine kinase inhibitors (TKIs) have become
available, the median survival of GIST patients increased from 9 months to over
5 years. Consequently, this rare disease has become a role model for other
targeted therapies. However, response to TKIs is extremely heterogeneous: ~15%
of the patients experience no benefit from imatinib, whereas ~17% of the
patients enjoy stable disease for over 9 years. Treatment failure due to
primary and secondary resistance is caused in part by mutations in oncogenic
genes that cause change in drug sensitivity. A new technique, using circulating
tumor DNA, has enabled us to assess mutations in a simple blood sample obtained
from patients on treatment, and thus detect new mutations early in the course
of the disease. Also, differences in pharmacokinetic drug behavior add to the
observed heterogeneity, and may cause resistance due to drug underexposure and
thereby proliferation of the least sensitive tumor cells. This offers the
opportunity to optimize and personalize targeted treatment for individual GIST
patients by timely treatment adaptation based on early detection of secondary
TKIs resistance mutations. Achieving this urgently requires data on daily
clinical practice, including prospective serial mutation analysis and serial
drug plasma concentration measurement. At a fundamental level this will also
help to unravel the driving factors behind primary and secondary TKIs
resistance in this model disease.

Primary objectives:
• To assess whether secondary GIST mutations can be found in circulating tumor
DNA of patients with progressive disease on TKI treatment (according to RECIST
1.1 on computer tomography), whereas they are NOT present in the patients that
have no progressive disease after the same time of TKI treatment
• To establish whether these secondary mutations can be detected some time (> 3
months) before progressive disease is assessed according to RECIST 1.1 on
computer tomography
Secondary objective:
• To assess whether TKI pharmacokinetics play a role in de development of
secondary TKI resistance (progression during TKI treatment)

In the bio-databank data of patients treated with imatinib (standard first line
treatment) as well as other TKIs will be included. The focus of this project is
to develop a model predicting secondary imatinib resistance based on tumor
genotype (serial secondary mutation analysis and tumor mutation analysis at
progression) and patient phenotype (TKI pharmacokinetics, multi-morbidity).

The treatment of Dutch GIST patients is centralized: almost all patients are
referred to one of the five collaborating centers forming the Dutch GIST
consortium, UMCG, NKI-AvL, Radboud UMC, Erasmus MC and LUMC. To further
optimize treatment for all patients, these centers have implemented a
standard-of-care diagnostic and treatment plan that assures collection of
homogenous phenotypic and treatment data for the bio-databank. The consortium
is supported by and works in close collaboration with the Dutch sarcoma and
GIST patient organizations.
A prospective, longitudinal bio-databank will be set up. Data regarding
multi-morbidity, drug pharmacokinetics and serial tumor genotypic data will be
collected prospectively from all (new) GIST patients during TKI treatment. Our
standard-of-care plan includes primary tumor mutation analysis, performed by
pathology laboratories on site. At each follow up visit during treatment, blood
will be collected to assess TKI plasma exposure and to perform mutation
analysis on circulating tumor DNA. All patients will be followed for tumor
RECIST 1.1 progression assessed by CT scans and asked to undergo a tumor biopsy
at progression to detect secondary resistance mutations.
The development of a model predicting secondary imatinib resistance based on
patient phenotype and tumor genotype, will be achieved by analyzing GIST
patients with progressive disease on imatinib (index patients; n=30) in our
bio-databank, in which a secondary mutation is detectable in the ctDNA.. These
patients will be matched 1:1 with non-progressive patients treated for the same
duration as the index patients. Regarding the index patients, next-generation
gene-targeted mutation analysis will be performed on archival tumor material
and on a tumor biopsy at progression to identify patient*s unique secondary
mutations. The mutations that will be studied are: KIT exon 9, exon 11, exon
13, exon 14, exon 17 and exon 18; PDGFRA exon 12, exon 14 and exon 18 and BRAF
exon 10 en exon 15.
In-depth analysis regarding mutation analysis in circulating tumor DNA and
imatinib drug concentration assessment will be performed for these 60 patients.

GIST patients will be asked to provide 30ml blood that will be collected in
four Na-EDTA blood collection tubes at every routine outpatient visit. All
patients will undergo a routine vena puncture for blood collection, whereby
this extra blood collection will not impose an extra burden or risk for them.
Moreover, patients will be asked to undergo a tumor biopsy at progression of
disease as assessed by CT-scan. This is routine clinical practice as is
described in the standard-of-care diagnostic and treatment plan of the Dutch
GIST Consortium. Metastatic lesions of GIST are mostly localized in the liver
and/or intra-abdominal. The risks of a biopsy are the pain of the procedure and
the chance of bleeding.