Fullbeam PET monitoring in particle therapy with the

Full-beam PET monitoring in particle therapy with the INSIDE scanner: first measurements Piergio Cerello (INFN, Torino) on behalf of the INSIDE collaboration N. Belcari M. Morrocchi M. G. Bisogni M. A. Piliero N. Camarlinghi G. Pirrone A. Del Guerra V. Rosso E. Kostara G. Sportelli P. Cerello S. Coli V. Ferrero E. Fiorina G. Giraudo F. Pennazio C. Peroni A. Rivetti M. Rolo R. Wheadon E. De Lucia R. Faccini P. M. Frallicciardi M. Marafini C. Morone V. Patera L. Piersanti A. Sarti A. Sciubba C. Voena F. Ciciriello F. Corsi F. Licciulli C. Marzocca G. Matarrese G. Battistoni M. Cecchetti F. Cappucci S. Muraro P. Sala INSIDE coordinator: M. G. Bisogni partners: M. Ciocca, M. Donetti, A. Mairani

The Project INnovative Solutions for In-beam Dosim. Etry in Hadrontherapy DOSE PROFILER IN-BEAM PET HEADS @ P. Cerello MEDAMI 2016, Ajaccio, May 3 rd 2016 Range Verification in Particle Therapy with sub-mm precision Designed to: q be integrated in the CNAO gantry q be operated in-beam q provide a real-time feedback on the particle range

The Project Real time monitoring of Bragg Peak depth with passive signal by secondary particles with a sub-mm precision IN-BEAM PET HEADS b+ activity distribution J Pawelke et al. , Proceedings IBIBAM, 26. -29. 09. 2007, Heidelberg PET image reconstruction P. Cerello MEDAMI 2016, Ajaccio, May 3 rd 2016 DOSE PROFILER prompt secondary particles emission L. Piersanti et al. Phys. Med. Biol. 59 1857 track and interaction point reconstruction

in-beam PET heads 10 x 20 x 5 cm 3 Distance from the isocenter=25 cm 256 LFS pixelated crystals (3 x 3 x 20 mm 3) coupled to MPPCs (Multi Pixel Photon Counters, Si. PMs). FE board (Tofpet ASIC) In-beam PET module P. Cerello MEDAMI 2016, Ajaccio, May 3 rd 2016

in-beam PET heads @ @ January 2016 P. Cerello MEDAMI 2016, Ajaccio, May 3 rd 2016 February 2016

The System Specifications IN-BEAM PET HEADS DOSE PROFILER Signal β+ decay Charged secondary particles Acquisition phases in-spill, inter-spill and after-treatment In-spill q face to face heads Position q perpendicular to beam axis q fixed wrt the isocenter q forward direction q 60 o wrt the beam axis. Distance from beam isocenter 25 cm about 40 cm Output 3 D PET image Emission point distribution P. Cerello MEDAMI 2016, Ajaccio, May 3 rd 2016

Beam test May 2015 @ phantom 25 cm Monoenergetic 1 D proton treatments on PMMA phantoms Single spot @ E = 68 / 72 / 84 Me. V 2 E+11 protons, 102 / 98 / 183 spill, 1. 96 E+9 / 2. 04 E+9 / 1. 09 E+9 protons/spill p beam PMMA PET modules TOT raw [ns] P. Cerello chip 7, channel 40 DATA ANALYSIS q Online/offline analysis q Automated energy window selection q Automated time delay calibration q 3 D PET image reconstruction q Profile analysis MEDAMI 2016, Ajaccio, May 3 rd 2016

Beam test May 2015 @ Turning on INSIDE: mono-energetic treatments in-spill inter-spill Range Uncertainty difference Theoretical 3. 7 mm - Inter-spill 3. 6 mm 0. 1 mm In-spill 3. 8 mm 0. 3 mm First 10 spill, inter-spill 3. 0 mm 0. 8 mm Simulation 3. 6 mm 0. 1 mm P. Cerello 1 D proton treatment Monoenergetic @ 68 / 72 Me. V p beam PET reconstructed images p beam MEDAMI 2016, Ajaccio, May 3 rd 2016 z (cm)

Beam test May 2015 @ Turning on INSIDE: rejecting the background for in-beam data Beam microstructure: 690 ns microbunches P. Cerello MEDAMI 2016, Ajaccio, May 3 rd 2016 Background rejection: Neutrons and prompt photons

Beam test May 2015 @ Turning on INSIDE: inter-spill vs. in-spill profile reconstruction Inter-spill image In-spill profile distal fall-off about 2 mm longer!!!! - prompt photon-induced pair production - 12 N contribution P. Cerello why? MEDAMI 2016, Ajaccio, May 3 rd 2016

Beam test Feb 2016 @ Turning on INSIDE: Calibration 18 F-FDG point-like source Coincidence Time Resolution: with beam : 480 ps with 68 Ge rods: 385 ps - Calibration not optimal - CTR not critical P. Cerello MEDAMI 2016, Ajaccio, May 3 rd 2016 68 Ge rods

Beam test Feb 2016 @ Turning on INSIDE: rate (singles) Peak to valley ~ 13 Turning on INSIDE: rate (coincidences) Peak to valley ~ 3 (decreases with integration time) P. Cerello MEDAMI 2016, Ajaccio, May 3 rd 2016

Beam test Feb 2016 @ Turning on INSIDE: Bragg Peak Depth Validation Bragg peak 77 Me. V Bragg peak 105 Me. V Expected difference: 31. 2 mm Measured difference: (30. 2 +/- 0. 3) mm P. Cerello MEDAMI 2016, Ajaccio, May 3 rd 2016 beam in

Beam test Feb 2016 @ Turning on INSIDE: image & profile reconstruction A B Data E 1 = 77 Me. V E 2 = 105 Me. V P. Cerello D beam Simulation E 1 = 77 Me. V E 2 = 105 Me. V MEDAMI 2016, Ajaccio, May 3 rd 2016 Treatment plan

Beam test Feb 2016 @ Turning on INSIDE: carbon beam profile P. Cerello MEDAMI 2016, Ajaccio, May 3 rd 2016

Beam test Feb 2016 @ Turning on INSIDE: carbon beam profile beam Fall-off distance: (82. 6 ± 0. 1) mm P. Cerello MEDAMI 2016, Ajaccio, May 3 rd 2016

Beam test Feb 2016 @ Turning on INSIDE: carbon beam profile Fall-off distance: (82. 6 ± 0. 1) mm P. Cerello MEDAMI 2016, Ajaccio, May 3 rd 2016

Beam test Feb 2016 @ P. Cerello MEDAMI 2016, Ajaccio, May 3 rd 2016

Beam test Apr 2016 @ Turning on INSIDE: how stable are we? 10 x 20 cm 3 PMMA phantom 144. 41 Me. V proton beam (Data) Activity profile along z (all image) (8. 35 ± 0. 01) cm (20. 57± 0. 01) cm Fall-off distance: (12. 22 ± 0. 02) cm P. Cerello MEDAMI 2016, Ajaccio, May 3 rd 2016

Beam test Apr 2016 @ Turning on INSIDE: how stable are we? 10 x 12 cm 3 homogeneous activity distribution (Simulation) Activity profile along z (all image) (7. 15 ± 0. 01) cm (19. 26± 0. 01) cm Fall-off distance: (12. 11 ± 0. 02) cm P. Cerello MEDAMI 2016, Ajaccio, May 3 rd 2016

Beam test Apr 2016 @ Turning on INSIDE: how stable are we? Is there a bias when changing the position in the field of view? Color codes refer to different positions along the lateral coordinate (y) Data Simulation P. Cerello MEDAMI 2016, Ajaccio, May 3 rd 2016

Beam test Apr 2016 @ Turning on INSIDE: how stable are we? Is there a bias when changing the position in the field of view? Color codes refer to different positions along the height coordinate (x) Data Simulation P. Cerello MEDAMI 2016, Ajaccio, May 3 rd 2016

@ … on the way to testing with patients P. Cerello MEDAMI 2016, Ajaccio, May 3 rd 2016

Simulations for testing with patients Turning on INSIDE: towards testing with patients 25 cm p beam PMMA p beam Monoenergetic 1 D proton treatment: single spot @ E=68 Me. V 2. E+11 protons 102 spill 1. 96 E+9 average protons/spill P. Cerello MEDAMI 2016, Ajaccio, May 3 rd 2016

Simulations for testing with patients Turning on INSIDE: towards testing with patients Correction for not simulated isotope production (86% during treat. , 95% after treat. ) Quantitative agreement between simulation and real data profiles The MC simulation is a reliable tool to evaluate the performance of the full in-beam PET system P. Cerello MEDAMI 2016, Ajaccio, May 3 rd 2016

Simulations for testing with patients Turning on INSIDE: towards testing with patients Time pattern and Expected Rates P. Cerello MEDAMI 2016, Ajaccio, May 3 rd 2016

Simulations for testing with patients Turning on INSIDE: towards testing with patients FLUKA-based simulation layout with a patient CT PET heads Accelerator Nozzle P. Cerello MEDAMI 2016, Ajaccio, May 3 rd 2016

Conclusions & Prospects • • • Successful data acquisition during beam delivery • use with Cyclotrons! Real time reconstruction and analysis Real time Bragg peak position validation • Hypofractionation • Secondary tumour treatments In-spill data filter In-spill data bias? to be further investigated On the way to testing with patients • Simulations essential to provide correction map & priors for real-time comparison This work has been supported by Ministero dell’Istruzione, dell’Università e della Ricerca of the Italian government under the program PRIN 2010 -2011 project nr. 2010 P 98 A 75 and by European Union’s Seventh Framework Program for research, technological development and demonstration under grant agreement no 317446 (INFIERI) P. Cerello MEDAMI 2016, Ajaccio, May 3 rd 2016