ESTRO 2020 Abstract book

S586 ESTRO 2020

PO-1014 Mitigating breathing motion with nasal high flow therapy (NHFT) in lung cancer: A prospective study S. Peeters 1 , S. Canisius 1 , D. Tissen 1 , E. Van Enckevort 1 , E. Rousch 1 , M. Öllers 1 , G. Bosmans 1 , K. Verhoeven 1 , C. Overhof 1 , J. Van Loon 1 , D. De Ruysscher 1 , G. Vilches- Freixas 1 1 MAASTRO Clinic, Radiation Oncology, Maastricht, The Netherlands Purpose or Objective Reducing breathing motion in RT is an attractive strategy to reduce margins, to better spare normal tissues. Also proton RT might benefit from this, as anatomical shifts caused by breathing can lead to severe over- and underdosage. The objective of this prospective study is to investigate the feasibility of treating NSCLC patients in moderate deep inspiration breath hold (mDIBH) using a nasal high-flow therapy (NHFT) device in order to mitigate the breathing motion. Material and Methods Stage III NSCLC patients were prospectively included and treated with photon RT during mDIBH using a NHFT device (AIRVO ® ). This device administers heated humidified air (flow 40 l/min, 80% O2) through a nasal interface. The breath hold (BH) with visual feedback through goggles was monitored by optical surface scanning (C-RAD ® ), with a gating window of 3mm. During training session patients were instructed to hold their breath as long as possible, 2x without (BH_w/o), and 2x with NHFT (BH_NHFT); the maximum BH length is reported. PlanningCT and treatment were done in mDIBH; a back-up plan in free breathing (FB) was available. Every session, matching was performed through a CBCT in BH. During treatment, patients were instructed to keep the BH as long as they wish; the beam delivery was automatically gated with the breathing signal. Subjective tolerance was measured with a questionnaire (5-points scale: 0 (no) to 5 (maximal discomfort)) at training session, first and last treatment fraction. Results Nine patients were included (5 males/4 females). Mean age was 68 years (range 60-74y). Mean FEV1 was 63% (Fig1). Five patients had concurrent chemoRT (60Gy/2Gy). One patient was excluded after training session because of metastatic disease on planningCT. Six out of 8 patients (75%) completed the whole treatment as planned. Two patients asked for a FB back-up treatment for 2 and 3 fractions, at a dose of 40Gy and 54Gy, respectively, because of bad condition due to side effects caused by chemoRT. At the training session, the mean length of BH_w/o was 43s (range 15-85s;SD 21s), and of BH_NHFT 106s (range 32-220s;SD 62s) (Fig1). With NHFT, the BH duration was significantly prolonged by a mean factor of 3.0 (range 1.1-9.2;SD 2.7)(p=0.02). The mean session time from entering to leaving the room was 23min (SD 2,7) . The length of the BHs during the treatment sessions (mean 48s, SD 15s) allowed performing the CBCT in one BH for all patients and, for some, to deliver more than one beam in one BH . Subjective tolerance was very good (Fig2).

observed in 53 patients and 23 of them had local failure. Finally, 88 patients were dead. The 3-year LPFS, PFS, and OS for all patients were 69.2%, 57.1 % and 72.9%, respectively. The 3-year LPFS, PFS, and OS in non-anemia and anemia groups as follows: 3-year LPFS: 75.6% and 54.9 % (p < 0.05), 3-year PFS: 64.0 % and 41.7 % (p < 0.05), and 3-year OS: 79.2% and 58.9% (p < 0.05), respectively. In multivariate analysis, anemia was a significant factor for LPFS, PFS and OS. Conclusion The pretreatment anemia is a prognostic factor of LPFS, PFS and OS in stage I NSCLC patients treated with SBRT. The pretreatment anemia indicates poor clinical outcomes in these patients. PO-1013 A cost-effectiveness analysis of consolidation immunotherapy with durvalumab in stage III NSCLC. C. Panje 1 , J.E. Lupatsch 2 , M. Barbier 2 , E. Pardo 3 , M. Lorez 4 , K.J. Dedes 5 , D.M. Aebersold 6 , L. Plasswilm 7 , O. Gautschi 8 , M. Schwenkglenks 9 1 Cantonal Hospital St. Gallen, Radiation Oncology, St. Gallen, Switzerland ; 2 Swiss Group for Clinical Cancer Research Coordinating Centre and University of Basel, Institute of Pharmaceutical Medicine, Bern and Basel, Switzerland ; 3 Cantonal Hospital Lucerne, Medical Oncology, Lucerne, Switzerland ; 4 National Institute for Cancer Epidemiology and Registration NICER, Statistics, Zurich, Switzerland ; 5 University Hospital Zurich, Gynecology, Zurich, Switzerland ; 6 University Hospital Bern, Radiation Oncology, Bern, Switzerland ; 7 Cantonal Hospital St. Gallen and University Bern, Radiation Oncology, St. Gallen and Bern, Switzerland ; 8 Cantonal Hospital Lucerne and University Bern, Medical Oncology, Lucerne and Bern, Switzerland ; 9 University of Basel, Institute of Pharmaceutical Medicine, Basel, Switzerland Purpose or Objective Consolidation immunotherapy with the programmed death ligand 1 (PD-L1) inhibitor durvalumab has shown to improve survival in patients with stage III non-small cell lung cancer (NSCLC) responding to radiochemotherapy according to the PACIFIC trial. However, consolidation immunotherapy for up to twelve months leads to a considerable cost increase. The aim of this study was to assess the cost-effectiveness of durvalumab in Switzerland based on the most recent PACIFIC survival follow up. Material and Methods We constructed a Markov model based on the 3-year follow-up data of the PACIFIC trial and compared consolidation durvalumab against observation. We used published utility values and assessed costs for treatment strategies from the perspective of the Swiss healthcare payer. Cost-effectiveness was tested both in the intention- to-treat (ITT) population of the PACIFIC trial unselected for PD-L1 tumor expression and in patients with PD-L1 expressing tumors. Results In the unselected / PD-L1 positive patients, durvalumab showed an incremental effectiveness of 0.74 / 1.17 QALY and incremental costs of Swiss France (CHF) 67,688 / 83,023, resulting in ICERs of CHF 91,354 / 70,700 per QALY gained, respectively. The most influential factors for the ICER were the utility prior to first progression, costs for consolidation therapy with durvalumab, and the hazard ratio for overall survival under durvalumab versus observation. Durvalumab was cost-effective in 70% of the simulations in probabilistic sensitivity analysis. Conclusion Assuming a willingness-to-pay threshold of CHF 100,000 per QALY gained, consolidation durvalumab is likely to be cost-effective both in patients with inoperable stage III NSCLC unselected for PD-L1 status and in patients with PD- L1 expressing tumors in Switzerland.