Objectives: Adverse Events (AEs) associated with existing and future treatments in Non-Small Cell Lung Cancer (NSCLC) are frequent, and may impact should be accounted for in health economic models. In addition to utility decrements, appropriate costing of AE management is needed. In the case of NSCLC which affects 45 200 patients in France, published data are scarce and not always complete. It was therefore thought of importance to assess resource use and costs associated with AEs. MethOds: When looking at the grade 3 and4 AEs appearing for at least 1% of patients in the clinical trials for erlotinib (TITAN, LUX-LUNG 8), ramucirumab plus docetaxel (REVEL), docetaxel and pembrolizumab (KEYNOTE 010) a list of 20 grade 3 and 4 AEs was identified as relevant for Health Economic analysis. Among them, 7 did not had cost data available in the literature. Three French oncology experts were consulted. A questionnaire was sent before the meeting with the experts, asking for, according to the grade, insight on resource use (hospitalisation, follow-up visits, diagnosis exams, treatments). Results were discussed during the meeting in order to reach a consensus. Total cost per AE, expressed in 2016 Euros, was then calculated from a national health insurance perspective based on public and private weighted tariffs and 2014 PMSI database. Results: The mean AE cost was € 206 for thrombocytopenia, € 263 for ocular toxic effect, € 2,332 for arthralgia, € 3,282 for venous thromboembolism, € 3,732 for pulmonary bleeding, € 5,176 for dehydration, € 5,751 for pneumonia, infiltration or pneumonitis. Hospitalization costs corresponded to 46% to 100% of total AE cost. These AE costs were in line with the costs of other grade ¾ AEs previously published for NSCLC therapies. cOnclusiOns: Among the seven AEs, four appear to have an important economic impact, with a management cost of at least € 3,000.

Orphan drugs were identified through transparency committee opinions. RESULTS: In France, 58 efficiency reports were published by the HAS between 2014 and 2018 including 12 orphan drugs in different therapeutic area: haematology treatments (N¼4), cardiology treatments (N¼2), oncology treatment (N¼1), pneumology treatment (N¼1), endocrinology treatment (N¼1) and a musculoskeletal treatment (N¼1). ICERs of orphan drugs ranged from 33,127V/QALY to 2,661,514V/QALY. Among the 4 heamatology drugs, only defibrotide was costeffective with an ICER of 33,273V/QALY. For cardiology treatments, macitentan and riociguat indicated in pulmonary arterial hypertension had ICERs of 85,359V/ QALY and 108,876V/QALY, respectively. However, only two ICERs of orphan drugs from French cost-effectiveness analyses have been criticized. These drugs were riociguat with an ICER of 108 876V/QALY and nusinersen sodium with an ICER of 2,661,514V/QALY. Both ICERs were considered extremely high by the CEESP despite no characterized level of ICERs.OBJECTIVES: Corticosteroid-refractory (SR) graft versus host disease (GvHD) is one of the most severe and life-threatening complications of allogeneic hematopoietic stem cells transplantation. To date, there are no standards for the treatment of this complication and few pharmacoeconomic analyses of various methods of treatment. METHODS: The data was collected retrospectively from medical records for 4 patients with acute SR GvHD treated with ruxolitinib, 8 patients with chronic SR GvHD treated with ruxolitinib, 8 patients with acute SR GvHD treated with etanercept, and 16 patients with chronic SR GvHD treated with extracorporeal photopheresis. We analysed effectiveness of the treatments based on overall response (OR) and failure-free survival (FFS), and performed cost minimization analysis with probabilistic sensitivity analysis. OR was assessed on day 28 for acute SR GvHD and after 12 weeks for chronic SR GvHD. RESULTS: There was no difference in 1-year FFS between ruxolitinib and control group in acute (50% vs 67%, p¼0.90) and chronic SR GvHD (88% vs 88%, p¼1.0). OR was faster in ruxolitinib group (100% vs 71%, p¼0.035), however we considered these effects comparable. Cost-minimization analysis demonstrated that six-month treatment with ruxolitinib was more expensive compared to the control group: 5 160 685 vs 3 626 654 rubles/patient for acute SR GvHD (ruxolitinib is more expensive by 43%), and 2 222 354 vs 1 681 221 rubles/ patient for chronic SR GvHD (ruxolitinib is more expensive by 32%) accordingly. CONCLUSIONS: We showed that ruxolitinib was economically unprofitable comparing to other frequently used therapies (etanercept and extracorporeal photopheresis). The power of this study does not allow to draw conclusions regarding the effectiveness of the therapies and further studies are required to to resolve this issue. However, probabilistic sensitivity analysis confirmed stability of the estimated costs of treatment.

2011 to 2017. Outcome parameters included surgical and post-operative costs, quality adjusted life years (QALY), and incremental cost per QALY gained or the incremental cost effectiveness ratio (ICER). A sensitivity analysis was carried out in order to propagate the uncertainty of the estimations to the results of the model. We use a multivariate and stochastic sensitivity analysis performed by 5000 Monte Carlo simulations. The cost-effectiveness plane was used to represent all pairs of solutions of the model. RESULTS: A total of 28 RDP and 26 LDP have been included. Conversion rate resulted to be significative higher in the LDP (3.6% vs 19.2%; p ¼ 0.04). The overall rate of pancreatic leak was 10.7% in the RDP group and 15.4% in the LDP group (p > 0.5). The mean number of hospital stay days was significative higher in the LDP (8.9 days vs 16.9 days, p ¼ 0.03). The mean operative time was higher in the RDP (294 vs 241 min; p ¼ 0.02). The overall mean total cost was similar in both groups (RDP: 9198.64Vversus LDP: 9399.74V; P > 0.5). Mean QALYs at 1 year for RDP (0.622) was higher than that associated with LDP (0.60025) (p>0.5). At a willingness-to-pay threshold of 20000 V and 30000 V, there was a 63.58% and 76.69% probability that RDP was cost-effective relative to LDP. CONCLUSIONS: RDP procedure appears to be cost-effective compared with LDP.