Session Item

The UK Proton Overseas Programme (POP) began in 2008, sending approved eligible patients abroad for proton beam therapy (PBT) funded by the National Health Service (NHS). POP patient outcomes data have been collected since 2008, now held in the Proton Clinical Outcomes Unit (PCOU) of The Christie NHS Foundation Trust. The POP will continue until the imminent second NHS centre is at full capacity. Outcomes are reported and analysed here for POP patients diagnosed with non-central nervous system (non-CNS) tumours treated from 2008 to Sept 2020.

All non-CNS tumour files for treatments as of 30/09/2020 were interrogated for follow-up information (latest dates and status), and type (following CTCAE v4) and time of onset of any late (>90 days post-PBT completion) grade 3-5 toxicities. Where a patient’s available data was inadequate, their local UK follow-up centre was requested to update via a specific outcomes form. Univariate and multivariate Cox regression analysis was performed for survival and toxicity modelling.

495 patients were analysed (436 for toxicity). Median follow-up was 2.1 years (0–9.3 years). Median age was 11 years (0–69 years). 70.3% of patients were paediatric (<16y). Rhabdomyosarcoma (RMS) and Ewing sarcoma were the most common diagnoses (42.6% and 34.1%). 51.3% of treated patients were for Head & Neck (H&N) tumours, then Abdo/Pelvis (31.9%). The median PBT dose was 51Gy (range 20.4-76Gy), increasing with cohort age. At last known follow-up, 86.1% of all patients were alive, with 2- and 5-yr survival rates of 88.3% and 82.1% respectively. Equivalent local control rates were 88.8% and 82.6%. The highest mortality risk was for spindle cell sarcoma (60%). Mortality and local control were worse for adults (³25 years) than for the younger groups.

The grade 3 toxicity rate was 12.6%, with median onset of 2.3 years. Most were in the H&N region in paediatric RMS patients. However, in relative terms, the adult cohort rate was higher (19.2%) than paediatric (14%) or TYA (5.6%), likely from higher doses received. Cataracts (30.5%) were most common, then musculoskeletal deformity (10.1%) and premature menopause (10.1%). Three paediatric patients (1-3 y at treatment) experienced secondary malignancy and 2 were found to have Li Fraumeni syndrome. Seven grade 4 toxicities occurred (1.6%), all in the H&N region and most in paediatric RMS patients. Six related to eyes (cataracts, retinopathy, scleral disorder) or ears (hearing impairment). Multivariate Cox regression analysis showed significance for increasing dose related to worse survival (p=0.006) and adult cohort grade 3 toxicity probability being higher than the paediatric cohort (p<0.001).

This large well-defined cohort of patients, diagnosed with non-CNS tumours undergoing multimodality therapy including PBT, demonstrates good local control, survival and acceptable toxicity rates, with some significant multivariate data analysis findings.