The SoraTram trial is a basket study designed to evaluate a new concept for treating a wide range of tumors with kinase-inactivating BRAF mutations. The BRAF enzyme is an important link in the RAS/RAF/MEK/ERK signaling chain that promotes cell division and, as a result of mutations, is present in a hyperactive state in many tumors. Drugs that inhibit the hyperactive oncoprotein BRAF V600E are already used to treat a number of cancers. The sequencing of tumor genomes has also identified increasing numbers of non-V600E mutations. In most cases, the clinical significance of these and their susceptibility to drugs have not yet been described. A sub-group of these non-V600E BRAF mutations inhibit the enzyme activity of the BRAF kinase. Paradoxically, however, these inactive BRAF mutants activate the ERK signaling chain and stimulate the onset of tumors. Preclinical models have shown that sorafenib, a drug that was approved several years ago for the treatment of kidney and liver cancer, can block the paradoxical effect of kinase-inactive BRAF mutants. The combination of sorafenib and the inhibitor trametinib is even more effective at blocking tumor growth. The effectiveness of this SoraTram combination has already been demonstrated in an index case with a metastatic melanoma. (1) 

The SoraTram study, which is planned to run at all DKTK sites, aims to

• identify patients in the DKTK MASTER network with tumors harboring non-V600 BRAF mutations 

• characterize mutations in terms of their susceptibility to BRAF, CRAF and MEK inhibitors and combinations of these

• treat 30 DKTK MASTER patients harboring non-V600 BRAF mutations that respond to CRAF/MEK inhibition following in vitro testing, with sorafenib and trametinib in a prospective multicenter Phase II trial

Comprehensive tumor sequencing in the DKTK network, especially through the MASTER program, is currently identifying many novel BRAF mutations whose significance remains unclear. These will be validated and verified during the study, both in terms of their kinase-inactivating potential and their oncogenic potential and susceptibility to pharmacological intervention. The cataloging of all non-V600E mutations can then be used for personalized therapy recommendations.

(1) Hoefflin R, Geißler AL, Fritsch R, Claus R, Wehrle J, Metzger P, et al. Personalized clinical decision making through implementation of a molecular tumor board – a German single-center experience.  JCO Precision Oncology. 2018; DOI: 10.1200/PO.18.00105 JCO Precision Oncology - published online August 16, 2018

Despite increasing application and success of personalized treatment in medical oncology, little progress has been made in personalized surgical cancer therapy. The ARMANI trial presents the first prospective, randomized trial to evaluate effectiveness and safety of molecular-guided resection in patients with colorectal liver metastasis (CRLM). While CRLM might be removed independently of the liver’s segmental borders, retrospective data favor anatomic resections in the subgroup of patients with a mutation in the RAS oncogene. Therefore, the ARMANI trial will test the hypothesis, if an anatomic resection (AR) improves long-term outcome vs. a non-anatomical resection (NAR) in patients undergoing surgery for RAS-mutated CRLM. The trial will be carried out among 11 high-volume centers of hepato-biliary surgery in Germany. A total of 220 patients will be enrolled and randomized in a 1:1 ratio to undergo an AR vs. NAR. The primary endpoint is intrahepatic disease-free survival (iDFS). In addition, the study will provide important data on perioperative outcomes and quality of life for both surgical techniques. Given the trend among liver surgeons to aim for parenchymal-sparing operations to preserve liver parenchyma, a positive trial will be practice changing and present the first piece of high-level evidence on benefits of personalized surgical therapy guided by the tumor’s mutational profile in patients with CRLM.

The German-wide INFORM (for children with relapsed cancers) and MASTER (for young adults with refractory cancers and patients with rare tumors) registry studies stratify patients based on genome and transcriptome sequencing signatures as well as epigenetic profiling for experimental therapies. Both programs follow their patients to obtain a comprehensive assessment about the usefulness of broad individualized molecular profiling in patient care. However, data for the functionally and therapeutically important layer of the proteome is currently not included.

Aim of the INFORM / MASTER project is to perform comprehensive quantitative proteome expression profiling in a set of initially 200 sarcoma patients included in the INFORM or MASTER program. The profiling data will be discussed in the molecular tumor board, and treatment recommendations based on the newly integrated proteogenomic profiling data will be compared to “conventional” genomic profiling. By this approach, information will be obtained, if a) proteomic profiling is feasible in the context of large-scale personalized patient stratification programs, b) the additional proteomic information adds to or changes current recommendation practice in molecular tumor boards, c) recommendations based on proteomes or proteogenomes correlate stronger with response to selected experimental therapies in sarcomas than those based on genomics/epigenomics/transcriptomics alone (exploratory) and d) which recurrently altered non-genomically activated signalling pathways do exist in sarcomas.

Comprehensive molecular diagnosis using whole genome and RNA sequencing of patients with advanced cancers as part of the NCT/DKTK MASTER program led to the identification of cancer-driving NRG1 fusions in a significant proportion of KRAS wild type pancreatic cancers. NRG1 rearrangements drive cancer development through ErbB receptor-mediated signaling pathway activation, as was shown when patients with NRG1-rearranged pancreatic cancer responded to ErbB inhibitor treatment.

NRG1 fusions are very rare but have now been detected in a broad range of malignant diseases. During a Phase II study, we plan to investigate the effectiveness of the pan-Erb inhibitor afatinib in advanced NRG1-rearranged malignant cancers once standard treatment has failed, at a number of centers in Germany. Patients with NRG1-positive tumors who meet the eligibility criteria can take part in the trial and will receive afatinib monotherapy.

The multi-center first-user study to investigate the safety, tolerance and effectiveness of the bispecific PSMAxCD3 antibody CC-1 in patients with prostate cancer is being carried out at University Hospital Tübingen in its Clinical Collaboration Unit Translational Immunology, and at other DKTK sites in Germany.

Bispecific antibodies are protein molecules with two different binding sites. In the case of the bispecific PSMAxCD3 antibody CC-1, one binding site addresses the prostate-specific membrane antigen (PSMA), which presents on the surface of aggressive prostate cancer cells. The other arm binds to a protein that is responsible for activating defense cells called T cells. When the antibody binds to the two binding sites, it activates the body’s immune response. In addition, CC-1 has a special characteristic: it also binds to the tumor’s blood vessels, creating a dual anti-tumor effect.

A serious side effect of the bispecific antibodies currently available is excessive activation of the immune system. This causes cytokine release syndrome (CRS). CRS can lead to a wide range of symptoms, including in particular fever and cardiovascular problems. When CRS occurs, the patient is usually treated with tocilizumab, an antibody that lessens the excessive immune system response.

For the study, the CC-1 antibody was optimized to minimize the risk of an unwanted immune activation. In addition, patients participating in the trial are given tocilizumab as a preventative measure to start with, in order to prevent CRS occurring in the first place.

Clonal hematopoiesis is a hallmark of hematologic malignancy, but can also be seen in healthy elderly people. This so-called “clonal hematopoiesis of indeterminate potential” (CHIP) is considered as a risk factor for stem cell-driven hematological diseases, including the myelodysplastic syndromes (MDS) and acute myeloid leukemia (AML). In this project, DKTK scientists join forces to investigate the role of the bone/bone marrow niche in the evolution of CHIP in elderly patients. The aim of the interdisciplinary study is to comprehensively characterize all components of the stem cell niches in the bone marrow in blood-healthy elderly people and patients with MDS/AML. For this purpose, an extensive collection of biosamples from the BoHemE study (Bone and Hematology in the Elderly) is available. With the help of high-throughput technologies, the CHOICE team will trace the molecular and cellular changes in the hematological compartment and the bone marrow microenvironment in CHIP towards MDS/AML. The researchers aim to define factors and key mechanisms that drive clonal hematopoiesis and malignant evolution of the mutant blood stem cells. The project also aims to elucidate the relationship between bone health and impaired hematopoiesis. Functional investigations using patient-derived xenograft models and 3D cell culture models are planned to evaluate the impact of an altered microenvironment on stem cell clonal evolution. The findings from the CHOICE project could help to better predict and perhaps prevent the transformation of a pre-leukemic clone to leukemia in CHIP patients.

A particularly aggressive form of breast cancer is the so-called triple negative breast cancer (TNBC). Patients with this type of tumor usually receive treatment with chemotherapy before surgical removal. While in some cases this treatment sequence leads to a massive reduction in the size of the tumor, it offers little or no benefit to other patients. In the past, attempts have been made to identify subgroups of TNBC based on RNA sequencing that respond differently to chemotherapy. However, these postulated groups are still not clinically applied in the diagnostic routine. 

A new method to classify malignant tumor diseases is the analysis of genome wide DNA methylation. DNA methylation is a dynamic modification of genetic information that can influence gene activity. By using artificial intelligence, complex patterns can be recognized in the DNA methylation profile of tumors, which can provide information about the origin of a tumor cell. In recent studies, new, prognostically relevant tumor subgroups have been found, particularly in the area of brain tumors. 

In cooperation with the German Breast Group (GBG), the extensive analysis of the DNA methylation profile of 400 patients from the GeparOcto (G8) study will be used to identify possible new subclasses of TNBCs and evaluate their significance in terms of predicting the response to chemotherapy. This analysis will be complemented by RNA sequencing in order to be able to make a comparison with the groups postulated so far.