A central goal of NCT is molecularly guided patient stratification as the basis for individualized treatment decisions in somatic and hereditary cancers. By leveraging the expertise of NCT, DKFZ and DKTK in multidimensional tumor characterization and molecular mechanism-based therapy, NCT has developed a standard for (1) comprehensive molecular profiling, (2) clinical interpretation of molecular data, (3) functional analysis of primary patient samples, (4) treatment decision making in molecular tumor boards, (5) longitudinal clinical data collection and (6) clinical trial design and conduct (Figure 2-7).
Joint Aims, Strategy and Perspectives
This pipeline provides a framework for multiple clinical programs and is intertwined with translational research projects centered on understanding the functional consequences of molecular alterations, with the ultimate goal to incorporate functional genomic profiling and ex vivo treatment testing in the precision oncology workflow.
Molecular stratification programs such as MASTER, a registry for young adults with advanced cancer across histologies and adults with rare tumors across age groups, recruit more than 400 patients per year. Potentially actionable findings are discussed in molecular tumor boards, including participants from both NCT sites and all DKTK partner sites, to determine therapeutic choices for individual patients. Towards the goal of systematic clinical translation, the molecular profiling platforms are linked to a range of investigator initiated trials (IITs) at both NCT sites, such as the NCT PMO studies, which are coordinated by the Clinical Trial Center at NCT Heidelberg and the portfolio of the Study Alliance Leukemia coordinated at NCT/UCC Dresden, which includes multicenter IITs across a network of 57 centers in Germany.
Research Profile NCT Heidelberg
At NCT Heidelberg, an integrated concept for molecular stratification of cancer patients has been established that is structured in two tiers and takes advantage of several complementary diagnostic platforms. In Tier 1, a range of tests, including, for example, comprehensive gene panel and targeted RNA sequencing, is available to all patients upon request by the respective entity-specific tumor board.
Tier 2 consists of specialized programs that focus on specific cancer types (CATCH, advanced breast cancer; COGNITION, early-stage breast cancer; N2M2, glioblastoma) or enroll patients across histologies (MASTER, advanced cancer in young adults and rare malignancies; joint activity of NCT Heidelberg and NCT/UCC Dresden) and provide whole-exome/genome and RNA sequencing when standard treatment has failed and routine molecular diagnostics have yielded no actionable target.
The current workflow is being expanded by additional layers of patient characterization, such as genome-wide DNA methylation profiling, which has substantially advanced the diagnosis and classification of brain tumors and sarcomas. To improve clinical translation, a portfolio of molecularly guided clinical trials has been developed, and the biological stratification approach will be extended to additional treatment modalities.
Molecular Stratification NCT Heidelberg - Selected Activities
CATCH / COGNITION – molecular stratification of breast cancer
The personalized oncology registry trials CATCH and COGNITION aim to clinically implement precision oncology for different breast cancer stages by integrative genomic-profiling. CATCH focuses on metastatic breast cancer (recruitment: > 330 patients, first patient first visit: 06/2017) and pilot analyses (n=200 patients), which revealed a clinical implementation of molecular informed therapies of 45% and clinical benefit rates (CBRs) and overall response rates (ORRs) of 33% and 14%, respectively. While precision medicine in metastatic tumors is restricted to prolongation of progression-free survival, the novelty of COGNITION resides in genomic-matched therapeutic interference at an early disease stage, displaying limited tumor heterogeneity (recruitment: > 170 patients, first patient first visit: 04/2019). This strategy holds the prospect to prohibit incurable metastasis and to increase cure rates. Both registry trials serve as a structural framework to facilitate the transition to confirmatory investigational trials.
Schmid et al., Lancet Oncol 2020; Fremd et al., Breast Care 2019
N2M2 – molecular stratification of glioblastoma
In N2M2, newly diagnosed patients with alkylator-resistant (MGMT unmethylated) glioblastoma are assigned to trial groups of this 8-arm umbrella trial based on deep multi-facetted molecular analyses with radiotherapy and temozolomide being the standard of care. Molecular diagnostics, bioinformatic analyses and evaluation in a molecular tumor board are performed within four weeks, allowing a timely initiation of postoperative treatment. The trial is recruiting in 13 NOA sites and has included almost 140 patients in the first 18 months. N2M2 and its accompanying translational research portfolio including the recently awarded SFB UNITE (“Understanding and targeting resistance in glioblastoma“) likely provide the basis for the development of predictive biomarkers and allow for the exploration of new approaches (immunotherapy) and treatments (addressing glioma networks/glio-neuronal synapse).
Venkataramani et al., Nature 2019; Hilf et al., Nature 2019
Tumor classification by DNA methylation profiling
Pathological diagnosis is crucial for the optimal management of cancer patients. For the approximately 100 known Central Nervous System (CNS) tumor types, standardization of the diagnostic process is particularly challenging - with substantial inter-observer variability in the histopathological diagnosis of many tumor types. The comprehensive approach of genome-wide DNA methylation-based classification of all CNS tumors across age groups assist neuropathological diagnostics through robust and reproducible molecular data and demonstrates its application in a routine diagnostic setting with a substantial impact on diagnostic precision. For broader accessibility, a free online classifier tool (www.molecularneuropathology.org) has been designed. The results provide a blueprint for the generation of machine-learning-based tumor classifiers across other cancer entities and may fundamentally transform tumor pathology.
Sahm et al., Lancet Oncol 2017; Capper et al., Nature 2018