Search

Objective We present the Berlin-Tübingen-Oncology corpus (BRONCO), a large and freely available corpus of shuffled sentences from German oncological discharge summaries annotated with diagnosis, treatments, medications, and further attributes including negation and speculation. The aim of BRONCO is to foster reproducible and openly available research on Information Extraction from German medical texts. Materials and Methods BRONCO consists of 200 manually deidentified discharge summaries of cancer patients. Annotation followed a structured and quality-controlled process involving 2 groups of medical experts to ensure consistency, comprehensiveness, and high quality of annotations. We present results of several state-of-the-art techniques for different IE tasks as baselines for subsequent research. Results The annotated corpus consists of 11 434 sentences and 89 942 tokens, annotated with 11 124 annotations for medical entities and 3118 annotations of related attributes. We publish 75% of the corpus as a set of shuffled sentences, and keep 25% as held-out data set for unbiased evaluation of future IE tools. On this held-out dataset, our baselines reach depending on the specific entity types F1-scores of 0.72–0.90 for named entity recognition, 0.10–0.68 for entity normalization, 0.55 for negation detection, and 0.33 for speculation detection. Discussion Medical corpus annotation is a complex and time-consuming task. This makes sharing of such resources even more important. Conclusion To our knowledge, BRONCO is the first sizable and freely available German medical corpus. Our baseline results show that more research efforts are necessary to lift the quality of information extraction in German medical texts to the level already possible for English.

Plasmablastic lymphoma (PBL) represents a rare and aggressive lymphoma subtype frequently associated with immunosuppression. Clinically, patients with PBL are characterized by poor outcome. The current understanding of the molecular pathogenesis is limited. A hallmark of PBL represents its plasmacytic differentiation with loss of B-cell markers and, in 60% of cases, its association with Epstein-Barr virus (EBV). Roughly 50% of PBLs harbor a MYC translocation. Here, we provide a comprehensive integrated genomic analysis using whole exome sequencing (WES) and genome-wide copy number determination in a large cohort of 96 primary PBL samples. We identify alterations activating the RAS-RAF, JAK-STAT, and NOTCH pathways as well as frequent high-level amplifications in MCL1 and IRF4. The functional impact of these alterations is assessed using an unbiased shRNA screen in a PBL model. These analyses identify the IRF4 and JAK-STAT pathways as promising molecular targets to improve outcome of PBL patients., Plasmablastic lymphoma (PBL) is an aggressive lymphoma subtype characterized by poor prognosis but the molecular knowledge of the disease is limited. Here, the authors perform whole exome sequencing and copy number determination of primary samples highlighting IRF4 and JAK-STAT pathways as therapeutic targets for PBL.

Purpose Precision oncology holds the promise of improving patient outcome. It is based on the idea that the testing of genomic biomarkers can lead to the recommendation of a treatment option tailored to the specific patient. To derive treatment recommendations from molecular profiles, interdisciplinary molecular tumor boards (MTBs) have been established recently in many academic institutions. The recommendation process in MTBs, however, has not been well defined, which limits applicability to larger clinical trials and patient populations. Methods We created four fictional patients on the basis of recent real cases with genomic information on mutations, fusions, copy numbers, and gene expression. We identified 29 tumor boards from nine countries worldwide and asked them to provide treatment recommendations for the sample patients. In addition, a questionnaire regarding the setup and methods used by MTBs was circulated. Results Five MTBs from four countries provided treatment recommendations and answered the questionnaire. For one patient, three tumor board treatment recommendations were identical, and two tumor boards had identical treatment strategies for the other three patients. There was heterogeneity in the interpretation of tumor and germline aberrations as well as in standards of prioritization. Conclusion Differences in the interpretation and recommendation process contribute to heterogeneity in MTB recommendations. Additional comparative analyses of recommendations could help improve rational decision making and lead to standardization.

Purpose Despite promising achievements in precision cancer medicine (PCM), participating patients are still faced with manifold uncertainties, especially regarding a potential treatment benefit of molecular diagnostics (MD). Hence, MD poses considerable challenges for patient information and communication. To meet these challenges, healthcare professionals need to gain deeper insight into patients’ subjective experiences. Therefore, this qualitative study examined information aspects of MD programs in cancer patients. Methods In two German Comprehensive Cancer Centers, 30 cancer patients undergoing MD participated in semi-structured interviews on information transfer and information needs regarding MD. Additionally, patients provided sociodemographic and medical data and indicated their subjective level of information (visual analogue scale, VAS, 0–10). Results On average patients had high levels of information (mean = 7, median = 8); nevertheless 20% (n = 6) showed an information level below 5 points. Qualitative analysis revealed that patients show limited understanding of the complex background of MD and have uncertainties regarding their personal benefit. Further, patients described unmet information needs. Existential threat in awaiting the results was experienced as burdensome. To withstand the strains of their situation, patients emphasized the importance of trusting their physician. Conclusion The challenges in PCM consist in providing unambiguous information, especially concerning treatment benefit, and providing guidance and support. Therefore, psycho-oncology needs to develop guidelines for adequate patient communication in order to help healthcare providers and cancer patients to handle these challenges in the developing field of PCM.

Background Reliable and reproducible interpretation of molecular aberrations constitutes a bottleneck of precision medicine. Evidence-based decision management systems may improve rational therapy recommendations. To cope with an increasing amount of complex molecular data in the clinical care of patients with cancer, we established a workflow for the interpretation of molecular analyses. Methods A specialized physician screened results from molecular analyses for potential biomarkers, irrespective of the diagnostic modality. Best available evidence was retrieved and categorized through establishment of an in-house database and interrogation of publicly available databases. Annotated biomarkers were ranked using predefined evidence levels and subsequently discussed at a molecular tumour board (MTB), which generated treatment recommendations. Subsequent translation into patient treatment and clinical outcomes were followed up. Results One hundred patients were discussed in the MTB between January 2016 and May 2017. Molecular data were obtained for 70 of 100 patients (50 whole exome/RNA sequencing, 18 panel sequencing, 2 immunohistochemistry (IHC)/microsatellite instability analysis). The MTB generated a median of two treatment recommendations each for 63 patients. Thirty-nine patients were treated: 6 partial responses and 12 stable diseases were achieved as best responses. Genetic counselling for germline events was recommended for seven patients. Conclusion The development of an evidence-based workflow allowed for the clinical interpretation of complex molecular data and facilitated the translation of personalized treatment strategies into routine clinical care. The high number of treatment recommendations in patients with comprehensive genomic data and promising responses in patients treated with combination therapy warrant larger clinical studies.

3564 Background: The clinical interpretation of molecular data is a bottleneck of precision oncology. High-dimensional molecular data, such as RNA sequencing (RNA-seq) and whole-exome sequencing (WES), will likely increase the complexity of clinical interpretation. We compared the recommendations by two molecular tumor boards (MTBs) that independently interpreted the same high-dimensional molecular profiles. Methods: Patients with advanced solid tumors, no available standard therapy, an ECOG performance status of 0-1, and available fresh-frozen tissue underwent WES of tumor tissue and normal blood as well as RNA-seq of tumor tissue within the MASTER (Molecularly Aided Stratification for Tumor Eradication Research) precision oncology program of the German Cancer Consortium (DKTK). Data from 46 patients (WES and RNA-seq, n = 41; WES alone, n = 5) were independently discussed by two MTBs. Treatment recommendations were compared with regard to levels of evidence, therapeutic baskets, and types of biomarkers. Results: A total of 51,610 aberrations (median, 393 per patient) were considered for clinical interpretation (34,314 mutations/single-nucleotide variants, 7,115 mRNA expression changes, 6,144 DNA copy number variations, 4,037 gene fusions). 110 and 132 treatment options were identified by the two MTBs, respectively, with an overall agreement rate of 44.1%. The highest agreement rates were identified for treatment options based on clinical levels of evidence (Level 1, 60%; Level 2, 49.6%) and for poly(ADP-ribose)-polymerase inhibition (57.1%). The lowest agreement rates were identified when MTBs opted for traditional chemotherapy (0%), combination therapies (6.9%), therapies based on preclinical levels of evidence (Level 3, 35.9%; Level 4, 32%), and MAPK inhibition (35%). Similar agreement rates, ranging from 39% (gene fusions) to 54% (loss of heterozygosity), were observed for different types of biomarkers. Conclusions: Reproducible, evidence-based annotation of high-dimensional molecular data is feasible. Our experience provides a basis for ongoing harmonization and standardization efforts within the MTBs of the DKTK.

Background Therapeutic options for neuroendocrine neoplasms (NEN) are limited, and only few NEN patients have been treated according to their individual genomic profile. We report the molecular and clinical characteristics of a large NEN cohort that was studied by prospective whole-genome (WGS), whole-exome (WES), and transcriptome sequencing (TS) within the NCT/DKTK MASTER precision oncology program. Methods Between 2013 and 2018, a total of 115 patients with advanced, heavily pretreated NEN were enrolled. Primary sites were gastrointestinal (n = 30), pancreatic/hepatic (n = 33), pulmonary (n = 20), genitourinary (n = 13), head/neck (n = 6), and other regions (n = 11). Grading according to the 2017 and 2019 WHO classification was neuroendocrine tumor (NET) G1, n = 12; NET G2, n = 31; NET G3, n = 12; neuroendocrine carcinoma, n = 36; mixed neuroendocrine-non-neuroendocrine neoplasm, n = 12; and other, n = 11. Results Of WES (n = 72) or WGS (n = 44) and TS (n = 92) were discussed in a multidisciplinary molecular tumor board. Recommendations were given in 103 cases (94%) for the following potential treatment alternatives: tyrosine or serine/threonine kinase inhibition (n = 67), PARP inhibition (n = 41), immunotherapy (n = 33), mTOR inhibition (n = 30), CDK4/6 inhibition (n = 19), MEK inhibition (n = 12), platinum-based chemotherapy (n = 7), BET inhibition (n = 7), anti-claudin18.2 (n = 5) or anti-DLL3 antibody treatment (n = 8). Somatic hypermutation (>10 mutations/megabase) was seen in 8.6% of patients. Treatment options could be implemented in 35 patients, of which 31 were evaluable for response: partial response, n = 10; mixed response, n = 1; stable disease, n = 11; progressive disease, n = 9; resulting in an overall response rate of 32% and a disease control rate of 71%. Conclusions Comprehensive genomic and transcriptomic characterization provides new insight into the molecular pathogenesis of NEN and creates therapeutic opportunities in a significant proportion of NEN patients who have exhausted standard treatment. Legal entity responsible for the study NCT Heidelberg & German Cancer Consortium. Funding German Cancer Research Center. Disclosure S. Kreutzfeldt: Research grant / Funding (institution), Travel / Accommodation / Expenses: PharmaMar; Honoraria (self): Novartis; Research grant / Funding (institution): Roche; Travel / Accommodation / Expenses: Pfizer. C.E. Heilig: Honoraria (self), Research grant / Funding (institution), Travel / Accommodation / Expenses: PharmaMar; Honoraria (institution), Travel / Accommodation / Expenses: Lilly; Travel / Accommodation / Expenses: Novartis; Travel / Accommodation / Expenses: Celgene; Travel / Accommodation / Expenses: Teva; Travel / Accommodation / Expenses: BMS; Research grant / Funding (institution): AstraZeneca; Research grant / Funding (institution): Pfizer. All other authors have declared no conflicts of interest.

Therapeutic options for neuroendocrine neoplasms (NEN) are limited. Within the MASTER program, a multi-institutional registry trial for prospective stratification of younger adults with advanced-stage cancer across all histologies and patients with rare tumors conducted under the auspices of NCT Heidelberg/Dresden and the German Cancer Consortium, we apply prospective whole-exome/genome sequencing (WES/WGS) and RNA sequencing (RNA-seq) to determine therapeutic choices for individual patients who have exhausted standard therapy options. We here report the clinical and molecular characteristics of the NEN cohort within this study. Between 2013 and 2018, 108 patients (male, n=65; female, n=43) were enrolled. Histologies according to the 2017 WHO Classification were neuroendocrine tumor grade 1 (NET G1), n=8; NET G2, n=31; NET G3, n=9; neuroendocrine carcinoma, n=49; and mixed neuroendocrine/non-neuroendocrine neoplasm, n=11. Primary tumor sites were gastrointestinal tract, n=30; pancreas, n=31; thorax, n=21; genitourinary system, n=11; head and neck, n=5; and other regions, n=10. All patients had advanced-stage disease and had received a median of 2 prior lines of systemic therapy. WES, WGS, and RNA-seq were performed in 69, 40, and 87 patients, respectively. Clinical evaluation of germline and somatic molecular data (single-nucleotide variants, small insertions and deletions, copy number variations, mutational burden, mutational signatures, homologous recombination deficiency scores, gene expression patterns, etc.) from 105 patients by a dedicated molecular tumor board yielded evidence-based recommendations for clinical management in 91 cases (87%). Treatment recommendations were grouped as follows: PARP inhibition, n=35; immunotherapy, n=27; mTOR inhibition, n=22; CDK4/6 inhibition, n=15; tyrosine or serine/threonine kinase inhibition, n=57 (ALK, n=1; ERBB, n=6; FGFR, n=10; MET, n=5; RET, n=12; VEGFR, n=4, MEK, n=9; other, n=10), DNA-crosslinking chemotherapy, n=16; anti-claudin18.2 antibody, n=5; BET inhibition, n=5; and DLL3 antibody, n=5. As of November 2018, at least 18 patients had received molecularly guided treatment (PARP inhibition, immunotherapy, mTOR inhibition, tyrosine or serine/threonine kinase inhibition) of which 11 were evaluable for response (partial response, n=4; stable disease, n=2; progressive diseases, n=5). Twenty-seven patients died before therapy could be started, 16 are currently receiving other regimens, and for the remaining patients no follow-up data are available yet. In conclusion, comprehensive molecular profiling offers valuable insight into to the genomic and transcriptomic landscape of NEN and creates additional therapeutic opportunities in a subset of patients. Citation Format: Simon Kreutzfeldt, Leonidas Apostolidis, Malgorzata Oles, Peter Horak, Christoph E. Heilig, Christoph Heining, Barbara Hutter, Laura Gieldon, Barbara Klink, Mario Lamping, Damian T. Rieke, Sebastian Uhrig, Henning Jann, Ulrich F. Pape, Albrecht Stenzinger, Eva C. Winkler, Bertram Wiedenmann, Dirk Jäger, Benedikt Brors, Evelin Schröck, Ulrich Keilholz, Marianne Pavel, Hanno Glimm, Stefan Fröhling. Clinical relevance of comprehensive genomic analysis in patients with advanced-stage neuroendocrine neoplasms: Results from the MASTER trial of the German Cancer Consortium [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 919.

e17577 Background: Salivary gland tumors (SGT) represent a rare and heterogeneous group of malignancies. No standard treatment exists in the advanced situation and the prognosis is poor. We here report characteristics and clinical outcomes of patients with SGT discussed at the Charité molecular tumor board (MTB). Methods: Patients with advanced cancer and no curative treatment option were discussed at the Charité MTB. Eligible patients underwent fresh tissue sampling and subsequent whole exome (WES) and RNA sequencing (RNA-seq) and immunohistochemical analyses (EGFR, HER2, AR as well as validation tests) or panel sequencing. Results from molecular testing were discussed at the MTB and patients were followed-up after recommendations were made. Results: 24 patients (median age 56 years, 13 male, 11 female) with advanced SGT were presented at the MTB between 2016 and 2019 (9 adenoidcystic carcinomas, 5 adenocarcinomas, 3 mucoepidermoid, 2 carcinosarcoma, 5 miscellaneous). WES/RNA sequencing was performed on tumor tissue from 16 patients. 2 patients were not included in the sequencing program and WES/RNA-Seq was ongoing for another 4 patients at the time of analysis. For another 2 patients, panel sequencing and IHC analysis, respectively was done. Results from analyses were discussed and a median of 2 recommendations, ranked by priority according to prespecified evidence levels, were made for 17 patients, each. Most commonly proposed treatment options by the MTB were FGFR inhibitors in 6 patients, mTOR or PARP inhibitors in 5 each, EGFR, HDAC inhibitors or antiandrogen therapy in 4, each. Treatments following MTB recommendations were initiated in 8 patients, 1 of which received a second recommended therapy after progression (antiandrogen therapy in 4, EGFR inhibitor in 2, a PDGFR, mTOR and PARP inhibitor in 1, each). A clinical benefit (CR = 1; Mixed Response = 1, SD = 3) was achieved in 5 patients, including a complete response in a patient with a metastatic adenocarcinoma of the parotid gland, treated with antiandrogen therapy. Conclusions: Precision oncology represents a feasible treatment strategy in patients with advanced SGT and shows early evidence of activity in a subset of patients. These results suggest further exploration of personalized therapy in these hard-to-treat tumors.

e24164Background: Precision Oncology programs hold promise to improve the treatment of cancer patients. The interpretation and prioritization of complex molecular data from comprehensive analyses h...

9566Background: There is a lack of active treatment against mUM. Such “hard-to-treat” tumour might benefit from treatment decisions driven by a complete genomic and transcriptomic analysis program....