143P - A comparison of treatment recommendations by molecular tumor boards worldwide

Abstract

Background: Precision oncology holds promise to improve patient outcome. Yet, a benefit of personalized therapy in comparison to standard therapy has not been shown in prospective trials. We asked molecular tumor boards (MTB) worldwide to make treatment recommendations for sample patients to assess standards and differences in the precision oncology approach. Methods: 4 fictional patients with various degrees of genomic information (mutation, gene expression, copy number and gene fusion data, somatic and germline events) were created. A questionnaire was designed to identify methods and structures of the respective molecular tumor board’s recommendation process. 29 molecular tumor boards from 9 countries were identified and asked to participate in the survey between August 2016 and March 2017. A qualitative interpretation of the results was performed. Results: 5 MTBs from 4 countries completed the questionnaire and provided therapy recommendations. An identical treatment recommendation by all 5 MTB was not made for any one of the patients. In only one patient an overlapping treatment recommendation was made by three MTBs. Heterogeneity was larger for patients with more complex genomic information. The availability of clinical trials did not influence the recommendation heterogeneity. The setup of MTBs showed similarities in participating specialties (e.g. medical oncology, pathology, molecular biology), duration of discussion, testing methods and number of discussed patients. Differences were seen in the interpretation process. Further differences were identified in the interpretation of germline aberrations and interpretation of variant allele frequency. Comments by MTBs helped to identify minimum reporting standards for genetic testing. Conclusions: Several differences in treatment recommendations were observed. In cases with obvious recommendations there was higher concordance among assessments. However, differences in treatment recommendations were observed in complex cases, and such heterogeneity contributes to the complexity of precision oncology. Larger studies are necessary for rational and stepwise development of principles and practice of MTB procedures. Legal entity responsible for the study: Damian Rieke Funding: None Disclosure: U. Keilholz: Speaker honoraria and advisory board roles are with AstraZeneca, Bristol-Myers Squibb, Merck, MSD, Pfizer, Novartis, Innate. Corporate-sponsored research with AstraZeneca and Merck. All other authors have declared no conflicts of interest.