OBJECTIVES: Categorization and analysis of positive and negative decision drivers for Health Technology Assessment (HTA) by Gemeinsame Bundesausschuss (G-BA) to understand evidence requirements in Germany. Non-Small Cell Lung Cancer (NSCLC) was chosen for this research because of dramatic changes in the treatment algorithms over the past 5 years. METHODS: All appraisal results of medicines in NSCLC published by G-BA between Jan 2013 and Feb 2018 were analyzed using primary data from the G-BA website. Additional information was obtained from the global HTA database Evalumade. RESULTS: We found 20 assessments for 12 medicines and 46 subgroups. Fourteen subgroups were given a positive rating (major, considerable or minor added benefit) because of one or more supportive pieces of evidence in the four major domains (mortality, morbidity, QoL and safety). Morbidity benefit contributed to the positive rating in many cases (12 out of 14 subgroups). Thirty-one subgroups received no added benefit and one subgroup less benefit. About one third of them were due to an inappropriate comparator. Other negative drivers were 1) small number of patients in a trial, 2) no relevant data generated, 3) potential bias of the study and/or endpoint, 4) unaccepted historical or indirect comparison and 5) transferability of the data to the German population. CONCLUSIONS: We found that generating convincing access evidence in NSCLC for Germany has involved: 1) including sufficient number of patients in a trial to prepare for extensive subgroup analyses or keeping a trial population homogeneous, 2) designing a study to show patient relevant benefits in at least one of the four domains (particularly morbidity and mortality) and 3) designing a trial that includes a patient population that is generalizable to the German population. It remains the responsibility of pharmaceutical companies whether to take into account German access evidence requirements as well as other country requirements in drug development decisions.