ESTRO 38 Abstract book

S183 ESTRO 38

adoptive immunotherapy, a ratio of conventional T cells (Tcons) and T regulatory cells (Tregs). Material and Methods From July 2015 to July 2018 26 AML patients (14 in first and 10 in second complete remission, 2 in partial remission) and 2 patients with myelodysplasia underwent haploidentical HSCT. There were 15 male and 13 female; median age was 60 years (range 42-70). Composite comorbidity/age scores were 1/2 in 10 patients and ≥ 3 in 18. TMLI target volumes were skeletal bones, major lymph node chains and spleen. TMLI was delivered by helical tomotherapy in 4 and half days (from day -15 to day -11) in 2 daily fractions of 1.5 Gy (TMI) and 1.3 Gy (TLI) (total doses 13.5Gy and 11.7Gy respectively). Chemotherapy schedule included tiothepa 2.5 mg/kg on days -10 and -9; fludarabine 30 mg/m 2 from days -10 to -6; cyclophosphamide 15 mg/kg on days -8 and -7. Haploidentical grafts consisted of 10x10 6 /kg purified CD34+cells, 1x10 6 /kg Tcons, 2 x10 6 /kg Tregs. No post- transplant immunosuppression was given. Results TMLI-related acute toxicity was G1-G2 in 23 patients, G3 in 4 and G4 in 1. All patients sustained primary full-donor type engraftment. Grade II-IV acute GvHD developed in 8 patients (29%) and chronic GvHD in none. Transplant- related causes of death in 6 patients were veno-occlusive disease (1), hemorrhage (1) sepsis (2) and acute GvHD (3). Immune reconstitution was good, with peripheral blood T cells rapidly increasing. Relapse did not occur in any patients. Twenty two (79%) patients are alive and relapse- free at a median follow up of 19 months (3-40 months). Conclusion This transplant strategy (TMLI, low doses of tiothepa, fludararabine, cyclophosphamide and adoptive immunotherapy with Tregs-Tcons) was successful in haploidentical transplantation for elderly and young unfit AML patients. TMI provides strong myeloablation and TLI provides efficient immunosuppression. The conditioning regimen was associated with low toxicity and mortality. The appropriate Tcon/Treg ratio was confirmed to exert a powerful T-cell dependent GvL effect with a low incidence of GvHD. PV-0368 Persistence of late substantial patient reported symptoms (LAPERS): A report from the EMBRACE study A.S. Vittrup 1 , L.U. Fokdal 1 , R. Pötter 2 , S.M. Bentzen 3 , J.C. Lindegaard 1 , A. Sturdza 2 , B. Segedin 4 , K. Bruheim 5 , I.M. Jürgenliemk-Schulz 6 , U. Mahantshetty 7 , B. Rai 8 , C. Haie-Meder 9 , R. Cooper 10 , S. Marit 11 , F. Huang 12 , E. Van der Steen-Banasik 13 , E. Villafranca 14 , R.A. Nout 15 , K. Tanderup 1 , K. Kirchheiner 2 1 Aarhus University Hospital, Department of Oncology, Aarhus, Denmark ; 2 Medical University of Vienna/General Hospital of Vienna, Department of Radiation Oncology- Comprehensive Cancer Center, Vienna, Austria; 3 University of Maryland School of Medicine, Department of Epidemiology and Public Health, Baltimore, USA ; 4 Institute of Oncology Ljubljana, Department of Radiotherapy, Ljubljana, Slovenia; 5 The Norwegian Radium Hospital- Oslo University Hospital, Department of Oncology, Oslo, Norway ; 6 University Medical Centre Utrecht, Department of Radiation Oncology, Utrecht, The Netherlands ; 7 Tata Memorial Hospital, Department of Radiation Oncology, Mumbai, India ; 8 Postgraduate Institute of Medical Education and Research, Department of Radiotherapy and Oncology, Chandigarh, India ; 9 Gustave-Roussy, Department of Radiotherapy, Villejuif, France; 10 St James's University Hospital, Leeds Cancer Centre, Leeds, United Kingdom ; 11 St. Olavs Hospital, Clinic of Oncology and Women's Clinic, Trondheim, Norway; 12 Cross Cancer Institute and University of Alberta, Department of Oncology, Edmonton, Canada ; 13 Radiotherapiegroep Arnhem, Department of

Radiotherapy, Arnhem, The Netherlands ; 14 Hospital of Navarra, Department of Radiation Oncology, Pamplona, Spain ; 15 Leiden University Medical Center, Department of Radiation Oncology, Leiden, The Netherlands Purpose or Objective Incidence methods and prevalence rates are commonly used statistical approaches for summarizing late side effects after radiotherapy. Incidence methods capture the first occurrence of an event; but are unable to distinguish transient from persistent symptoms. Prevalence rates contain information on persistence of symptoms in the cohort; but not on an individual patient level. A novel method to identify patients with LA te, PER sistent, S ubstantial and likely treatment related symptoms (LAPERS) is applied on patient reported outcomes (PRO) from the prospective, observational, and longitudinal study on MRI image-guided, adaptive brachytherapy in locally advanced cervical cancer (EMBRACE study). Material and Methods PRO (EORTC QLQ-C30 + CX24) were analyzed in 657 out of 1416 patients within the EMBRACE study who had a valid baseline, 3 months’ assessment and at least 3 late follow ups (6 months and ongoing). A LAPERS event for an individual patient was defined when the median scoring over late follow-ups was “quite a bit” or “very much” (substantial symptoms). For organ-related symptoms (e.g. urinary frequency) baseline morbidity was taken into account by requiring the median to be worse than the minimum of baseline and 3 months scoring (treatment- related); whereas for more unspecific symptoms (e.g. tiredness) baseline correction was not applied. Crude incidences and median prevalence rates of substantial symptoms from 6 to 60 months were calculated in the same cohort and put into perspective with LAPERS via ratio calculations. Results Median follow-up was 42 months (IQR 30-59). Table 1 presents the outcome of the different methods for reporting. LAPERS was ≥10% in 10 out of 31 symptoms; amongst them swelling in one or both legs, tiredness, urinary frequency and trouble sleeping. LAPERS was below 2% in 8 symptoms for example pain/burning feeling passing urine and blood in stools. LAPERS/crude incidence ratios lower than 0.1 was seen in vaginal symptoms, pain/burning feeling passing urine, blood in stools, nausea and vomiting; indicating that less than 10% of patients experiencing substantial symptoms did so persistently. LAPERS/median prevalence ratios lower than 0.4 was seen in the same symptoms; indicating that the patients experiencing substantial symptoms do not consist of the same individuals over time. LAPERS/median prevalence ratios close to 1 (e.g. swelling in one or both legs) indicate that patients experiencing substantial symptoms over time