WP6 – Monitoring cell-free DNA in plasma and tracking tumour mutations in tissue- and sequential plasma samples in patients with newly diagnosed brain cancer
Impact/relevance
Evaluating treatment response in glioblastoma (GBM), can be challenging. Evaluation with liquid biopsies have great potential due to the minimal invasive procedure. 1) To detect a relapse earlier and hence change treatment strategies before the patient deteriorates too much to receive further treatment. 2) To aid in the difficult decision with pseudo progression or not. 3) To evaluate treatment response and hence reduce the use of MRIs. 4) To use circulating cell free tumor DNA (ctDNA) in a dynamic targetable treatment approach.
Background
Brain cancer, including GBM is a heterogeneous tumour and can evolve both spatially and temporally thereby influencing choice of treatment. It is possible to evaluate the evolution of tumour mutations through sequential biopsies obtained by intracerebral surgery.
Due to the high-risk nature of the procedure, this approach is only performed when having therapeutic consequences. Pseudo progression is seen in approximately 20%, but only surgery with histopathological examination can either verify or reject the diagnosis, indicating that relapse surgery might be futile for a subgroup of patients. A blood sample is more accessible than a tumour biopsy and contains cell-free DNA (cfDNA) from both degraded tumour- and normal cells. CtDNA represents only cfDNA from tumour tissue. CtDNA has been identified in approximately 93% of cancer patients, a correlation between ctDNA and tumour progression has been identified in other cancer types than brain cancer and it is possible to detect a progression earlier by ctDNA than by radiologic assessment. Furthermore, by treating in a dynamic, targetable approach, according to if a mutation is present or not, has improved survival in lung cancer patients. At Rigshospitalet, Denmark, we have performed a feasibility study with eight newly diagnosed patients with GBM with collection of blood samples and measurement of cfDNA throughout a treatment period. We defined the optimal time point for collecting the first sample, identified a correlation between fluctuation of cfDNA and the clinical course and found implications that cfDNA measurements could aid in the decision of performing relapse surgery or not.
Aim
The aims of this study are (i) to investigate if the findings with cfDNA can be replicated in a larger cohort of GBM and (ii) to investigate the feasibility of tracking tumour specific mutations in GBM in a subgroup of 20 patients with combined tissue and sequential plasma samples (ctDNA). We also plan to send eight plasma samples and sequential tumour samples to Deutsches Krebsforschungszentrum (DKFZ) in Heidelberg, Germany. Theses eight patient samples are from a previous collected cohort, the Copenhagen Glioblastoma Cohort.
Methods
The project is planned as a national Danish study with participation from the Oncology departments of Aalborg, Aarhus, Odense and Rigshospitalet.
For aim (i) we plan to include all newly diagnosed patients with GBM, receiving oncological treatment. Peripheral blood will be collected at first visit to the Department of Oncology and at fixed time points until progression. Peripheral blood will be collected in cell-stabilizing Blood Collection Tubes (BCT; Streck Laboratories, Omaha, NE, USA). CfDNA will be extracted from 4 ml plasma using the QIAsymphony Circulating DNA Kit (Qiagen, Hilden, Germany) according to the manufacturer’s instructions using an elution volume of 60µl. Quantification of cfDNA will be performed using the dsDNA HS Assay Kit on a Qubit Fluorometer (Thermo Fisher Scientific, Waltham, MA).
For aim (ii) we will track tumor mutations in the plasma. Each mutation will be patient specific, based on the WGS results. A patient specific DNA probe based on the mutation we plan to follow, will be used and analyzed with ddPCR.
For the eight samples send to DKFZ, we will perform low coverage WGS on all cfDNA samples to estimate tumour content. Good quality samples will be chosen for a deeper sequencing and results will be compared to the mutations identified in the tissue of each patient.
Expected outcome
If the present study can find the same correlation between cfDNA and clinical course as we did in the pilot study, we plan a phase II study to validate the findings, with the perspective to implement cfDNA in daily clinical evaluation of GBM patients.