Towards an optimal clinical protocol for the treatment of moving targets with pencil beam scanned proton therapy

(1)

University of Groningen

Towards an optimal clinical protocol for the treatment of moving targets with pencil beam scanned proton therapy

Ribeiro, Cássia O.

DOI:

10.33612/diss.126443635

IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below.

Document Version

Publisher's PDF, also known as Version of record

Publication date:

2020

Link to publication in University of Groningen/UMCG research database

Citation for published version (APA):

Ribeiro, C. O. (2020). Towards an optimal clinical protocol for the treatment of moving targets with pencil beam scanned proton therapy. University of Groningen. https://doi.org/10.33612/diss.126443635

Copyright

Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).

Take-down policy

If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.

Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.

(2)

(3)

(4)

(5)

SUMMARY

Radiotherapy, as one of the main treatment modalities for patients with moving targets, such as abdominal (liver) and thoracic (lung and oesophagus) cancers, was explored throughout the several chapters of this thesis. More conformal radiotherapy techniques are especially attractive for these patients due to the need to limit the dose to the existent critical organs-at-risk (OARs) adjacent to their tumours. The clinical challenge of radiotherapy for this patient group remains mostly in their tumour motion, caused by respiration.

This thesis focuses on the treatment of moving targets with proton radiotherapy. State-of-the-art photon radiotherapy techniques, such as volumetric modulated arc therapy (VMAT), were used as baseline comparison. A dosimetric comparison between VMAT and the more conformal proton plans obtained with pencil beam scanned proton therapy (PBS-PT) delivery, such as intensity-modulated proton therapy (IMPT), is carried out in Chapter 4. A nominal (error-free) dosimetric comparison between VMAT and IMPT treatment plans was performed for thoracic indications treated in our photon clinic at the University Medical Center Groningen (UMCG). All nominal treatment plans of these patients were clinically accepted, with sufficient target coverage. The dosimetric sparing of OARs dose of IMPT relative to VMAT was clear for ten non-small cell lung cancer (NSCLC) and ten oesophageal cancer patients, especially concerning mean lung and heart doses.

From VMAT to IMPT, nominal Dmean(lungs-minus-gross tumour volume [lungs-GTV]) decreased 2.75 ± 0.56 GyRBE

and 3.76 ± 0.92 GyRBE over all lung and oesophageal cancer patients, respectively. A more pronounced reduction was verified for Dmean(heart): 5.38 ± 7.36 GyRBE (lung cases) and 9.51 ± 2.25 GyRBE (oesophagus cases). The OARs dosimetric superiority demonstrated from VMAT to IMPT is substantial for the rationale of PBS-PT for the treatment of thoracic cancer patients.

(6)

Summary Due to the potential clinical benefits of the treatment of moving targets with PBS-PT, a planning protocol for thoracic indications was designed for the UMCG Proton Therapy Center (GPTC). Before starting to actually perform these treatments at our proton clinic, this protocol needed to be evaluated (with representative patient population) for the possible dosimetric disturbances occurring when treating moving targets with PBS-PT. In fact, these uncertainty effects contribute to the hampering of PBS-PT treatments for moving targets in proton centres worldwide providing this modern technology. The aim of Chapter 2 was to incorporate all these uncertainties in one method, in a comprehensive, but also accurate manner. This developed method, called the 4D robustness evaluation method (4DREM), was then used to evaluate PBS-PT plans for thoracic indications. Briefly, the 4DREM was implemented in-house, by scripting capabilities of our treatment planning system RayStation (RaySearch Laboratories, Stockholm, Sweden). It incorporates setup and range errors, potential machine errors, anatomy changes, breathing motion, and interplay effects occurring during treatment delivery. Treatment-plan specific PBS-PT machine log files (by performing plan dry runs) and extensive 4DCT imaging data were used. The specificities of this method are thoroughly explained in Chapter 2, and its applicability to PBS-PT treatment plans created for two thoracic indications (a NSCLC and an oesophageal cancer patients) is demonstrated.

In Chapter 4, the 4DREM has been used to assess two IMPT optimisation strategies: 3D robust optimisation (IMPT_3D) and 4D robust optimisation (IMPT_4D). The analysis in terms of plan robustness of these optimisation strategies is particularly relevant for the clinical workflow at the GPTC, since they differ considerably in both manual work and pro- cessing time between each other. Only in computation time, IMPT_4D takes ≈ 7 hours, contrasting with the ≈ 3 hours of IMPT_3D. For most thoracic tumours, our IMPT_3D planning protocol showed to be robust and clinically suitable.

Target coverage robustness of IMPT_3D was sufficient for

(7)

18/20 patients. Averaged dose in heart, lung, spinal cord, and oesophagus over all 4DREM scenarios changed only slightly for both IMPT_3D and IMPT_4D. At least for the group of NSCLC and oesophageal cancer patients included in this study, relative to IMPT_3D, no gain in IMPT_4D was observed. Nevertheless, accurate patient positioning throughout the treatment course remain compulsory for both optimisation approaches. Additionally, verified target motion variability, or anatomical changes, in general, along the beam path, can demand adaption. All these aspects emphasize the importance of repeated or daily volumetric imaging in the proton workflows for moving targets.

As described in both Chapter 2 and Chapter 4, the 4DREM makes use of extensive 4DCT data sets. 4D imaging is man- datory to perform 4D dose calculations, which result in 4D dose distributions encompassing the dosimetric impacts caused by the patient’s breathing motion. A 4D dose calculation workflow relies on deformable image registration (DIR) to estimate the motion induced geometric differences between two scans. In Chapter 3, six DIR algorithms: ANA- CONDA, Morfeus, B-splines, Demons, CT Deformable, and Total Variation, were respectively applied to nine 4DCT data sets of liver cancer patients. Most of these DIR algorithms are available in commercial systems currently used in clinical radiotherapy practice. With the availability of a ground truth (GT) motion, we quantified both the geometric and the dosimetric error provided by DIR. The uncertainty in motion estimated by different DIR methods, and their con- sequent influence in the 4D dose calculations, was inspected in Chapter 3. Differences in 4D dose distributions among different DIR algorithms, and compared to the results using GT, were pronounced. Up to 40 % of clinically relevant dose calculation points showed dose differences of 10 % or more between the GT. Differences in V95(clinical target volume [CTV]) reached up to 11.34 ± 12.57 %. However, the dosimetric errors became in general less substantial when applying multiple-field plans or using rescanning. Intrinsic geometric errors by DIR can then influence the clinical evaluation of

(8)

Summary PBS-PT for abdominal tumours. It is, therefore, important to have insights on the magnitude of the dosimetric errors provided by DIR. That is why we recommend the use of an error bar for correctly interpreting individual 4D dose distributions, when using any DIR algorithm available. The analysis performed in Chapter 3 is based on single-field uniform dose (SFUD) plans for liver cancer patients. IMPT plans, or even the application of DIR to other moving indications, could modify these error bars.

DIR is also used for the assessment of the influence of anatomical changes over the radiotherapy treatment frac- tions. To determine if adaptation is needed, the planned treatment dose recalculated on repeated CTs can be warped back onto the planning CT and subsequently accumulated using DIR. In Chapter 6, the dosimetric impact (both in target and OARs) of using six different clinically available DIR algorithms for dose accumulation was investigated for lung cancer proton therapy. In this chapter, seven locally advanced NSCLC patients (with nine repeated CTs) planned with IMPT under breath-hold conditions were included.

Therefore, here breathing motion is negligible and only the inter-fractional anatomical variations were quantified. The V95(PTV) of accumulated doses decreased by 16 % on aver- age over all patients, with variations due to DIR selection of 8.7 %. The mean doses to the ipsilateral lung and heart showed variations of 1.8 % and 8.5 % due to DIR, respectively.

Additionally, an increased value compared to the planned mean doses of on average 2.3 % (ipsilateral lung) and 3.4 % (heart) was verified. A separation of the dose effects caused by anatomical changes and DIR uncertainty showed a good agreement between the dose degradation caused by anatomical changes and predicted from the average of all DIRs (differences of only 1.6 % in V95(PTV) and 3.4 % in Dmean(heart)).

The dose degradations caused by anatomical changes were more pronounced than the variations introduced by different DIR algorithms. However, substantial differences between different DIR algorithms of the fraction and accumulated doses were found. Since averaged results from several DIRs

(9)

provided a good estimation of the dose degradation caused by anatomical changes, using multiple DIR algorithms is a valid approach to reduce the uncertainty.

At the GPTC, we have a Proteus^®Plus (PPlus) PBS-PT machine from IBA (Louvain-la-Neuve, Belgium). For our clinical protocol of PBS-PT for moving targets, according to previous evidence, and due to our machine type, we apply scaled five times layered rescanning as a default for motion mitigation technique to all our delivered treatment plans.

Other common IBA PBS-PT technology available in proton centres worldwide is Proteus^®One (POne). In Chapter 5, the performances of the beam delivery approaches provided by PPlus and POne, together with the scaled layered rescanning possibility by PPlus, were assessed using the 4DREM. Five patients (with their IMPT_4D plans) from the ten NSCLC patients of Chapter 4 were included in this study. The five patients selected were the lung patients with larger CTV motion, OARs involvement, and we also tried to have varied primary tumour locations. The achievable performances of PPlus with rescanning and POne, in terms of robustness for both target (coverage and homogeneity) and OARs dose, were quite comparable. In general, when compared to PPlus without rescanning, target homogeneity proved to be superior for POne in all cases (on average a 0.41 GyRBE

enhancement). Within the PPlus plans, irradiation times from no rescanning to rescanning increased on average 79 %.

From no rescanned PPlus to POne, field delivery times also increased, but more slightly. Fields in POne took on average 10.1 s (32 %) longer to be delivered than PPlus without any rescanning. The POne plan delivery times were, however, below delivery times of PPlus plans with rescanning.

The work presented in this thesis contributed to the clinical implementation of thoracic treatments at the GPTC.

This was done by (1) implementation of a comprehensive evaluation of a planning protocol for moving targets, using patient and machine specific information and (2) subse- quent application for representative patients of the patient population at our department, using clinically approved

(10)

Summary delineations and treatment plans. A separate analysis of the uncertainty provided by DIR in radiotherapy applications was also successful. Future work will consist mainly in assessing our planning protocol for patients with larger CTV motion amplitudes. This will allow us to extend our findings concerning the need of IMPT_4D, or any additional motion mitigation techniques, to a larger scope of thoracic indications. With the tools developed and experience gained since the treatment of our first thoracic indication at the GPTC, our ultimate goal is to extend the treatments to all moving indications.

(11)

SAMENVATTING

Radiotherapie is één van de belangrijkste behandelmogelijk- heden voor patiënten met kanker. In deze thesis wordt de bestralingsbehandeling van patiënten onderzocht waarbij het tumorvolume beweegt, zoals een tumor gelegen in de buik (leverkanker) of borstkas (longen slokdarmkanker). Omdat deze tumoren omringd worden door kritische organen zijn meer conforme bestralingstechnieken van uiterst belang, zodat dosis in deze structuren zo laag mogelijk blijft. De beweging van de tumor, veroorzaakt door ademhaling, vormt de grootste uitdaging voor de klinische toepassing van meer conforme bestralingstechnieken.

Deze thesis richt zich op de behandeling van bewegende tumoren met protonentherapie. Ter vergelijking werd volumetric-modulated arc therapy (VMAT), de meest geavan- ceerde fotonentechniek, gebruikt als gouden standaard. In hoofdstuk 4 werd een nominale (fout-vrije) dosimetrische vergelijking uitgevoerd tussen VMAT en de meer conforme plannen van pencil beam scanned proton therapy (PBS-PT), zoals intensity-modulated proton therapy (IMPT). Bestra- lingsplannen van tien niet-kleincellige longen tien slokdarmkanker patiënten werden geanalyseerd. De patiënten werden eerder bestraald met fotonen in het Universitair Medisch Centrum Groningen (UMCG). Alle bestralingsplannen waren klinisch acceptabel met adequate dosisafdekking.

In alle patiënten bood IMPT een duidelijk voordeel ten opzichte van VMAT wat betreft dosis in de kritieke structuren (vooral gemiddelde hart- en longdosis). De nominale gemiddelde dosis (Dgem) in de longen zonder het tumorvolume (lungs-GTV) was 2.75 ± 0.56 GyRBE en 3.76 ± 0.92 GyRBE lager voor IMPT dan voor VMAT in respectievelijk de longen slokdarmkanker patiënten. In het hart werd een grotere reductie waargenomen: 5.38 ± 7.36 GyRBE (longkanker pati- enten) en 9.51 ± 2.25 GyRBE (slokdarmkanker patiënten). Deze aangetoonde dosisreductie in de kritieke structuren is van wezenlijk belang voor de rechtvaardiging van PBS-PT voor patiënten met een tumor in de thorax.

(12)

Samenvatting Door het potentiële klinische voordeel van PBS-PT bij bewegende tumoren, werd een planningsprotocol opgesteld in het UMCG Protonen Therapie Centrum (GPTC). Voordat met PBS-PT gestart kon worden in onze protonenkliniek, diende de invloed van beweging op de dosisdistributie onderzocht te worden. De onzekerheden van bestraling van bewegende doelvolumen belemmert de behandeling van deze tumoren met PBS-PT wereldwijd. Het doel van hoofdstuk 2 was om alle mogelijke onzekerheden te kunnen evalueren met een uitgebreide en nauwkeurige methode. Deze ontwikkelde methode, de 4D-robuustheid evaluatie methode (4DREM) genoemd, werd vervolgens gebruikt om PBS-PT plannen voor thoracale indicaties te evalueren. De 4DREM werd geïmplementeerd in ons bestralingsplanningsysteem RayS- tation (RaySearch Laboratories, Stockholm, Sweden). De evaluatiemethode houdt rekening met positionerings- en bereikfouten, eventuele toestelfouten, anatomische veranderingen, ademhalingsbeweging- en interplay-effecten die mogelijk optreden tijdens de bestraling. Bestralingsplan-specifieke PBS-PT log files van het toestel (verkregen door een dry run van het bestralingsplan) en uitgebreide 4DCT beeld- vorming werden gebruikt in de evaluatie. Gedetailleerde informatie over de methode en de toepassing op PBS-PT bestralingsplannen van een slokdarm- en een longkanker patiënt is opgenomen in hoofdstuk 2.

In hoofdstuk 4 werd de 4DREM gebruikt om de robuustheid van twee verschillende IMPT optimalisatiestrategieën te beoordelen: de 3D-robuuste optimalisatie (IMPT_3D) en de 4D-robuuste optimalisatie (IMPT_4D). Deze ro- buustheidsanalyse is vooral belangrijk voor de klinische implementatie, omdat er een aanzienlijk verschil bestaat in zowel manueel werk als verwerkingstijd tussen de beide optimalisatiestrategieën. Dit verschil is al duidelijk in alleen de berekeningstijd; IMPT_4D duurt ≈ 7 uur en IMPT_3D duurt ≈ 3 uur. De plannen gemaakt met het IMPT_3D planningsprotocol waren robuust en klinisch acceptabel. Het doelvolume werd adequaat bestraald in 18/20 patiënten. Het verschil in de gemiddelde dosis van het hart, de longen, het

(13)

ruggenmerg en de slokdarm van alle 4DREM scenario’s was klein voor zowel IMPT_3D als IMPT_4D. In deze groep slokdarm- en longkanker patiënten was er geen voordeel aantoonbaar voor IMPT_4D ten opzichte van IMPT_3D.

Nauwkeurige positionering van de patiënt blijft essentieel, ongeacht de optimalisatiestrategie. Planaanpassing kan nodig zijn als de beweging van het doelvolume verandert of als anatomische veranderingen in de bundelrichting optreden.

Al deze aspecten benadrukken hoe belangrijk 4D-beeldvor- ming is bij protonenbestraling van bewegende doelvolumen.

Zoals beschreven in zowel hoofdstuk 2 en hoofdstuk 4 maakt de 4DREM gebruik van uitgebreide 4D-data. 4D-beeld- vorming is noodzakelijk om 4D-dosisberekeningen uit te kunnen voeren. In de 4D-dosisberekening wordt de invloed van de ademhalingsbeweging van de patiënt op de dosisdistributie meegenomen. Hiervoor wordt deformable image registration (DIR) gebruikt, die de geometrische verschillen inschat die veroorzaakt worden door beweging. In hoofdstuk 3 werden zes DIR algoritmen (ANACONDA, Morfeus, B-splines, Demons, CT Deformable en Total Variation) toegepast op negen 4DCTs van negen leverkanker patiënten. De meeste van deze DIR algoritmen zijn beschikbaar in commer- ciële systemen die gebruikt worden in de hedendaagse radiotherapie praktijk. De werkelijke beweging (GT) was bekend, waardoor we zowel de geometrische als de dosimetrische fout van de DIR konden bepalen. De onzekerheid in de beweging, ingeschat door de verschillende DIR algoritmen, en de invloed hiervan op de 4D-dosisberekeningen werd onderzocht in hoofdstuk 3. Verschillen in 4D-dosisdistributies tussen de verschillende DIR algoritmen en de resultaten van de GT waren groot. Dosisverschillen van meer dan 10 % vergeleken met de GT werden vastgesteld in 40 % van klinisch relevante do- sispunten. Verschillen in de V95 van het clinical target volume (CTV) waren maximaal 11.34 ± 12.57 %. Over het algemeen werden deze dosisverschillen echter kleiner als meerdere bestralingsvelden of rescanning gebruikt werden. Intrin- sieke geometrische fouten, veroorzaakt door DIR, kunnen van invloed zijn op de klinische evaluatie van PBS-PT voor

(14)

Samenvatting tumoren in het abdomen. Het is daarom belangrijk inzicht te hebben op de grootte van de dosimetrische fout veroorzaakt door DIR. Daarom raden wij aan een foutmarge te gebruiken om de individuele 4D-dosisdistributies te interpreteren bij het gebruik van alle DIR algoritmen. De uitgevoerde analyse in hoofdstuk 3 is gebaseerd op single-field uniform dose (SFUD) plannen voor leverkanker patiënten. Bij IMPT, of zelfs de toepassing van het DIR algoritme op andere bewegende doelvolumen, zouden de foutmarges anders kunnen zijn.

DIR wordt ook gebruikt om de invloed van anatomische veranderingen te beoordelen tijdens de bestralingsbehandeling. Om te beoordelen of een planaanpassing nodig is, wordt aan de hand van DIR de herberekende dosis op de nieuwe CT gekopieerd naar de plannings CT, waar vervolgens de dosis wordt geaccumuleerd. In hoofdstuk 6 werd de invloed van zes verschillende klinisch beschikbare DIR algoritmen op de dosisaccumulatie onderzocht bij protonentherapie van longkanker (voor zowel het doelvolume als kritieke structuren).

In dit hoofdstuk bestond de onderzoekspopulatie uit zeven patiënten met een lokaal uitgebreid niet-kleincellig longcar- cinoom. Voor elke patiënt werd een IMPT plan gemaakt. Voor de 10 gemaakte CT’s diende de patiënt de adem in te houden, waardoor de ademhalingsbeweging verwaarloosbaar was en alleen gekeken werd naar de anatomische veranderingen tussen de fracties. De V95(PTV) van de geaccumuleerde dosis daalde met gemiddeld 16 % bij alle patiënten, met variaties veroorzaakt door het gekozen DIR algoritme van 8.7 %.

Verschillen in de gemiddelde dosis van de ipsilaterale long en het hart waren respectievelijk 1.8 % en 8.5 % bij toepassing van de verschillende DIR algoritmen. Daarnaast waren deze dosissen met gemiddeld 2.3 % (ipsilaterale long) en 3.4 % (hart) hoger in vergelijking met de geplande dosis.

Om de invloed van het DIR algoritme te evalueren werden de effecten op de dosis gescheiden op basis van de oorzaak (anatomische veranderingen of het DIR algoritme). De ver- andering in dosis kon beter worden voorspeld aan de hand van het gemiddelde van alle DIR algoritmen (verschillen van 1.6 % in de V95(PTV) en 3.4 % in Dgem(hart)). Anatomische

(15)

veranderingen hadden meer invloed op de dosisdistributie dan de variaties die de DIR algoritmen veroorzaakten. De verschillen waren echter aanzienlijk tussen de evaluatie per fractie en de geaccumuleerde dosis van alle fracties. Omdat de gemiddelde resultaten van meerdere DIR algoritmen de dosisreductie door anatomische veranderingen goed kon voorspellen, is het gebruik van meerdere algoritmen een goede manier om een nauwkeuriger resultaat te verkrijgen.

In het GPTC hebben we een Proteus^®Plus (PPlus) PBS-PT toestel van IBA (Louvain-la-Neuve, België). Dit type toestel maakt het mogelijk layer rescanning toe te passen. Layer rescanning heeft in studies aangetoond mogelijk storende effecten bij de bestraling, als gevolg van ademhaling van de patiënt, te verminderen. In het klinische protocol van PBS-PT voor bewegende doelvolumen passen we dit standaard vijf keer toe voor alle bestralingen. Een ander IBA PBS-PT toestel in protonencentra wereldwijd is de Pro- teus^®One (POne). In hoofdstuk 5 werd de 4DREM ingezet om de verschillende bestralingstechnologieën van beide toe- stellen te beoordelen, in combinatie met de geschaalde layer rescanning mogelijkheden van PPlus. Vijf IMPT_4D plannen van de tien longkanker bestralingsplannen uit hoofdstuk 4 werden gebruikt in deze studie. Deze vijf patiënten werden geselecteerd op basis van grotere beweging van het CTV, de betrokkenheid van kritieke structuren en een brede variatie met betrekking tot de locatie van de primaire tumor. PPlus met rescanning en POne presteerden vergelijkbaar wat betreft robuustheid, voor zowel het doelvolume (dosisafdekking en homogeniteit) als dosis in de kritieke structuren. De dosis in het doelvolume was bij alle patiënten homogener in de POne plannen vergeleken met de PPlus plannen zonder rescanning (gemiddeld 0.41 Gy verbetering). De bestraling van de PPlus plannen duurde gemiddeld 79 % langer wanneer rescanning toegepast werd. De bestralingstijd van de velden was gemiddeld 10.1 s (32 %) langer met de POne dan met de PPlus plannen zonder rescanning. De bestralingstijden met POne waren echter korter dan de bestralingstijden van PPlus plannen met rescanning.

(16)

Samenvatting Het werk in deze thesis heeft bijgedragen aan de klinische implementatie van PBS-PT bij tumoren in de thorax. Deze bijdrage bestond uit (1) implementatie van een uitgebreide evaluatie van een planningsprotocol voor bewegende doelvolumen, waar gebruik gemaakt wordt van patiënt- en toestel-specifieke informatie en vervolgens (2) de toepassing hiervan op een representatieve patiëntengroep op onze af- deling, met gebruik van klinisch geaccepteerde intekeningen en bestralingsplannen. Daarnaast werd een analyse van de onzekerheden bij DIR algoritmen uitgevoerd. In de toekomst zal onderzoek zich vooral richten op de evaluatie van het planningsprotocol voor patiënten met grotere bewegings- waarden van het CTV, om de resultaten voor een grotere patiëntengroep te laten gelden. Daarnaast worden andere methoden om de invloed van beweging te verminderen, zoals IMPT_4D, verder onderzocht. Met de ontwikkelde methode, en de ervaring opgedaan sinds de eerste 4D-indi- catie in het GPTC is ons uiteindelijke doel alle bewegende doelvolumen met protonen te kunnen behandelen.

(17)

ACRONYMS

3D Three-dimensional

4D Four-dimensional

4DCBCT Four-dimensional cone beam computed tomography

3DCT Three-dimensional computed tomography 4DCT Four-dimensional computed tomography 4DMRI Four-dimensional magnetic resonance

imaging

4DREM Four-dimensional robustness evaluation method

4DVwamax Voxel-wise worst-case maximum over four-dimensional accumulated scenarios

4DVwamin Voxel-wise worst-case minimum over four-dimensional accumulated scenarios

ASTRO American Society for Radiation Oncology av_pCT Averaged planning computed tomography CBCT Cone beam computed tomography

CT Computed tomography

CTV Clinical target volume

DDVH Dose-deviation-volume histogram DIBH Deep inspiration breath-hold DIR Deformable image registration DVF Deformation vector field DVH Dose-volume histogram

EE_pCT End-of-exhalation planning computed tomography phase

ESTRO European SocieTy for Radiotherapy and Oncology

fVOI Field-specific volume of interest

GPTC University Medical Center Groningen Proton Therapy Center

GT Ground truth

GT-DVFs Ground truth deformation vector fields GTV Gross tumour volume

HI Homogeneity index

iCTV Internal clinical target volume

(18)

Acronyms IMPT Intensity-modulated proton therapy

IMPT_3D Three-dimensional robust optimised intensity-modulated proton therapy plan

IMPT_4D Four-dimensional robust optimised intensity- modulated proton therapy plan

IMRT Intensity-modulated radiation therapy LLL Left lower lobe

LUL Left upper lobe MHD Mean heart dose

MLD Mean lungs-minus-gross tumour volume dose MOD Mean oesophagus dose

MRI Magnetic resonance imaging NSCLC Non-small cell lung cancer

OAR Organ-at-risk

PBS-PT Pencil beam scanned proton therapy PD Prescribed dose

POne Proteus^®One PPlus Proteus^®Plus

PTCOG Particle Therapy Co-Operative Group PTV Planning target volume

QA Quality assurance

RBE Relative biological effectiveness RLL Right lower lobe

RML Right middle lobe ROI Region of interest RUL Right upper lobe SCLC Small cell lung cancer SD Standard deviation SFUD Single-field uniform dose TPS Treatment planning system VMAT Volumetric modulated arc therapy Vwmax Voxel-wise worst-case maximum Vwmin Voxel-wise worst-case minimum

(19)

PUBLICATIONS

Journal articles

1. M. O’Halloran, S. Lohfeld, G. Ruvio, J. Browne, F.

Krewer, C. O. Ribeiro, V.C. Inácio Pita, R.C. Conceição, E. Jones, M. Glavin. Development of anatomically and di- electrically accurate breast phantoms for microwave imaging applications. Conference on Radar Sensor Technology XVIII, part of SPIE (The International Society for Optical Engineering) Defense + Security, Baltimore, Maryland, USA, May 2014.

2. T.-T. Zhai, L.V. van Dijk, B.-T. Huang, Z.-X Lin, Z.-X. Lin, C. O. Ribeiro, C.L. Brouwer, S.F. Oosting, G.B. Halmos, M.J.H. Witjes, J.A. Langendijk, R.J.H.M. Steenbakkers, N.M. Sijtsema. Improving the prediction of overall survival for head and neck cancer patients using image biomarkers in combi- nation with clinical parameters. Radiotherapy and Oncology 2017; 124(2): 256-262.

3. C. O. Ribeiro, A. Knopf, J.A. Langendijk, D.C. Weber, A.J. Lomax, Y. Zhang. Assessment of dosimetric errors induced by deformable image registration methods in 4D pencil beam scanned proton treatment planning for liver tumours. Radio- therapy and Oncology 2018; 128(1): 174-181.

4. C. O. Ribeiro, A. Meijers, E.W. Korevaar, C.T. Muijs, S. Both, J.A. Langendijk, A. Knopf. Comprehensive 4D robustness evaluation for pencil beam scanned proton plans.

Radiotherapy and Oncology 2019; 136: 185-189.

5. F. Guerreiro, C. Zachiu, E. Seravalli, C. O. Ribeiro, G.O.

Janssens, M. Ries, B. Denis de Senneville, J.H. Maduro, C.L. Brouwer, E.W. Korevaar, A.C. Knopf, B.W. Raaymak- ers. Evaluating the benefit of PBS vs. VMAT dose distributions in terms of dosimetric sparing and robustness against inter-fraction anatomical changes for pediatric abdominal tumors. Radiother- apy and Oncology 2019; 138: 158–165.

6. L. Nenoff, C. O. Ribeiro, M. Matter, L. Hafner, M. Jo- sipovic, J.A. Langendijk, G.F. Persson, M. Walser, D.C.

Weber, A.J. Lomax, A. Knopf, F. Albertini, Y. Zhang.

Deformable image registration uncertainty for inter-fractional

(20)

Publications dose accumulation of lung cancer proton therapy. Accepted in Radiotherapy and Oncology.

7. C. O. Ribeiro, S. Visser, E.W. Korevaar, N.M. Sijtsema, R.M. Anakotta, M. Dieters, S. Both, J.A. Langendijk, R.

Wijsman, C.T. Muijs, A. Meijers, A. Knopf. Towards the clinical implementation of intensity-modulated proton therapy for thoracic indications: evaluation of 3D vs. 4D robust optimi- sation by means of patient and machine specific information.

Submitted to Radiotherapy and Oncology.

8. C. O. Ribeiro, J. Terpstra, G. Janssens, J.A. Langendijk, S. Both, C.T. Muijs, R. Wijsman, A. Knopf, A. Meijers.

Evaluation of intrinsic ‘rescanning’ (pulsed beam) versus scaled rescanning (continuous beam) for the treatment of moving targets with pencil beam scanned proton therapy. To be submitted.

9. R.M. Anakotta, H.P. van der Laan, S. Visser, C. O. Ribeiro, M. Dieters, J.A. Langendijk, S. Both, E.W. Korevaar, N.M.

Sijtsema, A. Knopf, C.T. Muijs. Weekly robustness evaluation of Intensity-Modulated Proton Therapy for oesophageal cancer.

Submitted to Radiotherapy and Oncology.

10. S. Visser, H. Neh, C. O. Ribeiro, E.W. Korevaar, A.

Meijers, B. Poppe, S. Both, J.A. Langendijk, C.T. Muijs, A. Knopf. Assessment of a Diaphragm Override Strategy for Robustly Optimized Proton Therapy Planning for Esoph- ageal Cancer Patients. Submitted to Radiotherapy and Oncology.

Grants

ESTRO Mobility Grant (October, 2016). Assessment of dosim- etric errors induced by deformable image registration for 4D treatment planning using 4DCT-MRI data sets. Center for Proton Therapy at Paul Scherrer Institut (PSI), Villigen, Switzerland.

International conference abstracts and presentations (first author only)

1. Poster. ESTRO 2017 (Vienna, Austria). Comprehensive pro- spective evaluation tool for treatments of thoracic tumours with scanned protons. C. O. Ribeiro, A. Meijers, G. Janssens, J.

Widder, J.A. Langendijk, E.W. Korevaar, A. Knopf.

(21)

2. Poster. PTCOG 2017 (Yokohama, Kanagawa, Japan).

Assessment of dosimetric uncertainties induced by deformable image registration methods in 4D proton treatment planning for liver tumours. C. O. Ribeiro, A. Knopf, A.J. Lomax, J.A.

Langendijk, D.C. Weber, Y. Zhang.

3. Poster. 4D Workshop 2017 (Vienna, Austria). Dosimetric uncertainties induced by deformable image registration methods in 4D pencil beam scanned proton treatment planning. C. O.

Ribeiro, A. Knopf, J.A. Langendijk, D.C. Weber, A.J. Lo- max, Y. Zhang.

4. Oral presentation. ESTRO 2018 (Barcelona, Spain). How reliable are deformable image registration methods for scanned proton 4D dose calculations? C. O. Ribeiro, A. Knopf, J.A.

Langendijk, D.C. Weber, A.J. Lomax, Y. Zhang.

5. Poster discussion. ASTRO 2018 (San Antonio, Texas, USA). Comprehensive 4D robustness evaluation for pencil beam scanned proton plans. C. O. Ribeiro, A. Meijers, E.W. Kor- evaar, S. Both, J.A. Langendijk, A. Knopf.

6. Poster discussion. PTCOG 2019 (Manchester, UK). A comparison of various scanned proton pencil beam delivery strategies for the treatment of moving targets. C. O. Ribeiro, D. Dumont, J. Terpstra, G. Janssens, J.A. Langendijk, S.

Both, E. Sterpin, A. Knopf, A. Meijers.

7. Poster presentation. ASTRO 2019 (Chicago, Illinois, USA). Towards the Clinical Implementation of Pencil Beam Scanned Proton Therapy (PBS-PT) for Thoracic Indications:

A Comprehensive 4D Robustness Evaluation Based on Patient and Machine Specific Information. C. O. Ribeiro, S. Visser, A. Meijers, E.W. Korevaar, R. Wijsman, C. T. Muijs, S.

Both, J.A. Langendijk, A. Knopf.

8. Oral presentation. PTCOG 2020 (Taipei, Taiwan). Patient and machine specific evaluation of 3D versus 4D robust optimis- ation in intensity-modulated proton therapy (IMPT) for thoracic indications. C. O. Ribeiro, S. Visser, E.W. KYeorevaar, N.M.

Sijtsema, S. Both, J.A. Langendijk, R. Wijsman, C.T. Muijs, A. Meijers, A. Knopf.

(22)

Acknowledgements

ACKNOWLEDGEMENTS

During the four years of my PhD I have been through very hard work, trips, and amazing experiences. Looking back, I would not change my decision to come to Groningen to start this adventure for nothing. The most essential part of these last four years of my life was definitely the people I had the chance to coexist during these times, both at work and in my personal life. This people were definitely what made this PhD journey so fulfilling for me in all aspects.

My connexion with the UMCG started already on the 1st of April of 2013 (so more than seven years ago) when I sent an e-mail to Hans to ask for opportunities to perform an internship at the Radiation Oncology department. I did not know him and I sent him the e-mail because he was the head of the department. It was a long shot. However, I was surprised with the kind and open response. That is why my first acknowledgment has to be for him. Hans, without your willingness to receive me for the first time here in Gronin- gen, I would not have done my internship here, and would most probably not be here today with a PhD book ready to be printed. Additionally, your role as my PhD promotor contributed substantially to the quality of all the research I performed over the last years. Besides your intelligence, sim- plicity and work ethic, I have always admired your openness to other cultures. I am sure this characteristic of you will continue to attract highly talented people from around the world to our department.

From my previous experiences at the UMCG before my PhD, there are many people to consider, who luckily somehow also played a role on the research performed over the last four years. Marianna and Roel, as supervisors of my master internship, they also contributed to the possibility for me to stay at the department and do my PhD, which I would not have started if I had not had the best experience by working with them. Marianna and Roel, thank you for your research guidance and for opening me the door to the radiotherapy world.

(23)

I remember very well when Filipa texted me saying: ‘Oh you are so lucky, I just found out that your PhD supervisor will be Antje!’. At that time, I did not know much about Antje or her work, since during my master internship I was working in a different field than protons. However, it did not take me much time to realise how lucky I was indeed. Antje, working with a highly talented and nice person as you could not have been a better experience. Besides your knowledge and motivation, I have to emphasize your personality. Your humbleness and encouragement are some of your many valuable characteristics that I admire. Those made my work move forward, and not the other way around. Working with you made me grow not only as a scientist, but also as a person.

It is very nice to have a ‘boss’ with whom you relate to in real life. I cannot forget to thank you for keeping up with me all these years, and with my more stressful periods. However, I am afraid you will not get rid of me so soon (still lucky me!).

I feel like you are still there for me for all my questions and dilemmas in life and career prospects, and I really appreciate that. Most importantly, thank you so much for being the greatest supervisor I could ask for!

I would like to express my sincere gratitude to Arturs.

Despite the fact that he did not officially had to, he was present in almost all my projects of my PhD. Arturs, I feel very lucky to have worked with you during all these years.

Your availability, hard work, efficiency, objectiveness, and smart thinking, are just some of your characteristics that make you a great colleague and professional. Thank you for being part of my PhD, thank you for all the exceptional work you performed at our department, and thank you for your contribution to our field. You definitely left a legacy at the UMCG. You (together with your smile and sarcastic humour) are already being missed! I am very glad I was able to be part of your journey at the UMCG and I really hope our paths will again cross each other in the near future.

My co-supervisor and another medical physicist that contributed substantially to my work (and still does), Erik, I appreciate your guidance and help during the four years

(24)

Acknowledgements of my PhD. Your calmness is something I always admired, especially when I had my stressful moments during my PhD.

I am also very grateful to Stefan. Thank you for all your input during our meetings, all your thorough revisions performed in our papers, and fruitful discussions. It is really efficient and productive when, as a PhD student, you are able to work with a multidisciplinary and talented team. I would also like to thank Aart for the attention and support given to my future endeavours.

Sabine Visser, I told you (while lying ) that I will write a whole page of acknowledgments for you. Well, to be honest that will not be a difficult task. I could write a lot about the amazing experience I had by working with you. I think I never met a co-worker that was a better match. We think together, and most importantly, we challenge and we listen to each other. I think working with you is a success also due to the fun we have together. I don’t remember seeing you sad or annoyed. Of course, as anyone else, you went through difficult times, but you were somehow always able to not bring your sadness or frustrations to your work and colleagues. This is a very wise way to live life, and this makes you a very special person. Sabine, my colleague, my paranymph, and my friend, you are a big part of my life and also of this thesis. I appreciate all the moments we had together and your joy in life makes you and everything you do outstanding.

Never change these characteristics of you, so that I can keep annoying you, hopefully for many many years.

I would like to give a special acknowledgement to all the medical doctors of our department that contributed to my work and my publications during my PhD. Particularly, Robin, Kristel, Melissa, and Margriet. Your support on delineations, your attendance on several meetings for in- spection of treatment plans, and your thorough revisions of my papers, were substantial for making my research more clinically meaningful and of higher quality!

One of the most fulfilling experiences for me during my PhD was my research stay at PSI. Besides all the interesting work done there, I was able to meet very nice people and get

(25)

to know the pioneering centre of scanned proton therapy.

I would like to thank all the amazing researchers I shared the office with there and have been able to meet again at conferences, and all the people from the clinic that showed me around. A special acknowledgement goes to Ye, Tony, and Damien, who contributed to the success of my stay and all the resultant publications. Ye, was really nice working with you and being supervised by you. Thank you for all your support and guidance in our project! After my research stay I was able to continue with the collaboration project between UMCG and PSI, thanks to Lena, Francesca, and Ye. A special recognition goes to you Lena and all your hard work in making this follow-up collaboration productive and efficient. It was very nice working with you and having you with us at the UMCG!

From IBA I cannot forget to acknowledge Guillaume for all his contribution to the work of my thesis. Additionally, also from IBA, I thank Carolina for the data acquired. Another appreciation goes to Mark and Djamal from Mirada, who were extremely cooperative and responsive for projects of my PhD.

In Terracotta I had the chance to work with a few internship students: Tom, Elyse, Jorvi, Damien, and Hendrike. Thank you all very much for your good work and contribution to my research! I would also like to give a special acknowledgement to Ronald. Your contribution was exceptional for all the work conducted in robustness evaluation at our department.

As soon as I started my PhD, I was already surrounded by very nice people. New PhDs started that, besides providing fruitful scientific discussions, definitely made the whole work environment more fun for me. Lydia, my ‘work wife’, my first long term roommate in Terracotta, thank you for all the moments we shared. Lisa, I always enjoyed chatting with you in the evenings. Gracias por todos los momentos agradables y encuentros fuera del trabajo. Qué bueno es tener una amiga como tu! Elske, my friend, my second roommate and second ‘work wife’, it was very pleasant to have you at my side daily and all the amazing moments we shared together

(26)

Acknowledgements will stay in my memory. You always make me laugh. Tian- Tian, my first foreigner colleague at the department, it was really nice to experience the tumour image biomarker world with you during my internship and have your company at the department as well afterwards for my PhD. Sanne, Daan and Veerle, you definitely contributed to this joyful environment at work and also to very fun gatherings outside work.

Lydia, Lisa, Elske, Tian-Tian, Sanne, Daan, and Veerle, you definitely made my life in Groningen during my PhD more exciting. Bedankt allemaal voor de geweldige momenten samen (lunchen, congressen, reizen, Sinterklaas, bruiloften, etentjes en pubquizen). Hopelijk: to be continued!!

A recognition goes to all the people who worked in Ter- racotta during the time I was sitting there and definitely gave the Villa some more life for me: Roel, Dan, Charlotte, Gabriel, Jeffrey, Vincent, Heleen, Sander, Arjen, Gert, Linda, Martijn, Dirk, Witold, Amir, Jan, Dave, Peter, among others. A special acknowledgement goes to Gabriel, my office neighbour in Terracotta, who very often had to bear with me and Sabine and sometimes only with me asking him questions, and also did the favour of delivering a few plans at the GPTC for me. Thank you (and Vivian) for all the Brazilian food that made my days in Groningen sweeter and tasting like home. Gabriel a tua energia positiva é muito contagiante e faz falta na gente, vem para o GPTC!

More recently, in the era of the research office at the GPTC, the fun still continues. To my new office mates Yong (my new funny neighbour behind my back), Jiapan (my new friendly neighbour right at my side), Hans Paul, Wouter, Makbule, Bas, and Adrian, thank you for making me company daily at work and for all the pleasant coffee breaks and gatherings outside the office. You are all able to make my working days more pleasant and rewarding. Makbule, one special acknowledgement goes to you for all your support when I needed it. I really enjoyed our moments together at the ASTRO last year.

I would like to give a word of appreciation to everyone working at the department and that somehow assisted me

(27)

during my PhD: Lies, Linda, Hilda, Tenny, Patrick, Lysbert, Chakir, Mathilde, Erik, Peter (Verschuur), Peter (Vonck), Ricardo, and many others.

I highly appreciate the availability and contribution of the assessment committee members of my thesis: Prof. dr.

Sytze Brandenburg, Prof. dr. Bas W. Raaymakers, and Prof.

dr. Joachim Widder.

Now the two people that contributed to my PhD book pro- duction process: Joanna and Jorge. Joanna thank you very much for your patience and for listening to all my requests for the layout, you did an amazing job. Jorge obrigada pela tua incrível criatividade e trabalho de excelência no conceito do meu livro. It was great working with both of you!

I dedicate this book to my family, distributed in different parts of the world. A very especial acknowledgement goes to my Groningen family, without whom I could never go through so many years performing this work apart from my other family. Actually, one of the things I appreciate the most of going away from Portugal was to get to know you and have all the experiences I lived with you. Bego, Maria, Mohammad (Moooo), Raouf, Marta, Elias, Joshua, Giulia, Carmen, and Amr, my Groningen family, I hope we will keep being a family, wherever the destiny brings us. I am thankful everyday to have met you and have you in my life. There are no words to describe how much you all meant for me these years of hard work of my PhD. Our long night outs, long talks, and adventures were the best experiences I had. Bego, thanks to you our journey all together started (at a Dutch course of the University of Groningen). Maria, I feel I am the luckiest to have met such a kind and amazing person as you in the beginning of my PhD in Groningen. I always say, you are definitely one-of-a-kind, I don’t think they make people like you anymore. Having you as my friend is definitely a blessing.

Marta, Mohammad, Raouf, Elias, and Joshua came to my life through Maria. Coming through Maria, I could not except less than the best friends ever I could ask for. Marta, gracias por todos los momentos que hemos pasado juntas y por todas las aventuras en Warhol . Moooo I appreciate

(28)

Acknowledgements all the times we were together and all the moments you registered from us. Raouf, thank you for putting my spirit always up and for the amazing parties. Elias, my dear friend, the most like me, you are such an amazing person. I am so grateful for our friendship. At this moment, I even miss our legendary arguments. Joshua, the only Dutch of the group, the tallest, the only one putting some order, you definitely fill my life with joy. Amr, our last male acquisition, thank you for adding so much to the group. Giulia and Carmen, you came to Groningen and joined us almost in the last year of my PhD. Thank you for making this last and more stressful year the most fulfilling one. Carmen, Carmencita, my friend, my colleague, my paranymph, and my neighbour, qué bueno tenerte siempre muy cerca de mi y agradezco mucho que seas la amiga maravillosa y agradable que eres. Giulia, Bella, our moments together are just unforgettable for me in all aspects, you are so sweet, you are the cherry on top of the cake of my life in Groningen! Bego, Maria, Moooo, Raouf, Marta, Elias, Joshua, Giulia, Carmen, and Amr, the complete Groningen family, I love you all <3. Also to the other friends of this family that crossed my path in Groningen, thank you for all our fun moments together!

I would like to recognise and give my word of appreciation to all the other friends that I met here in Groningen, and that even after leaving Groningen, we maintain the contact and the friendship. Natalia, my gym and Ibiza partner, it is just impressive how we immediately hit off and became friends. Sinto muito a sua falta em Groningen amiga, mas eu sei que você vai ter muito sucesso em Barcelona. Admiro muito a sua determinação! Xu and Eric, my two also PhD friends in Groningen, thank you for our dinners together.

Xu, I especially miss our matches at the tennis court to free our minds of our days of intense work on our papers.

During these four years of PhD in Groningen, I felt I did not leave home entirely, mainly due to all my University friends that also came to do their PhD in the Netherlands and made me company for so many weekends and getaways:

Filipa, Joli, Joana, João (Guimas), Iva, and others. Filipa, not

(29)

only I had the chance to have you in the Netherlands near me these years, as I also had the opportunity to work with you. It is always a pleasure to work with someone as dedicated as you and to be able to share my ideas and thoughts of our crazy proton world with you. Joana, you were the only one with courage enough to come to the far North, and I am glad I was able to share so many moments with you (and Charles). Thank you all for making me feel at home in the Netherlands!

My best friend Patricia (and her family), since I came to the Netherlands, always made sure I had a home in Amsterdam.

Thank you Patricia for being always there for me. A very special thanks goes to my good friend Lucy (and also Rick) for one of the greatest trips of my life and for receiving me at your home in Zaandam for many weekends during my PhD. Zé, pai dos ‘Tugas em Groningen’, your joyful spirit is able to make people smile and feel good while around you.

Thank you for organising all the events of the Portuguese community. You definitely make Groningen a home for all the amazing people that I met during my PhD (Mafalda, Tiago, Pedro, Ana, Rita, Filipe, Cindy, Mandy, Xavier, Inês, Hugo, Adriana, and many more).

Filipe, my almost neighbour in Lisbon, who I came to meet here in Groningen. Not a day goes by that I do not think about how lucky I am to have met you and have you in my life, always supporting me, respecting me, and loving me. Thank you for all your patience during the last more stressful moments of my PhD. Thank you for being so special to me, and for always being there for me. Destiny brought us together here in Groningen, and I am looking forward to what is yet to come, for me and you, together, hopefully for many many many many years <3.

To all my friends/family I left in Portugal and Brazil, I would like to thank you all for your support. This book is also dedicated to all of you. Especially to my girls Rita, Maria, Inês (Cardoso), Inês (Lourenço), Carlota, Luísa, Catarina, Bea, Inês (Conceição), and my sister Carina, thank you. I could not have done this without your support, even being

(30)

Acknowledgements far away from me. Whenever you came to visit me or I had the chance to travel to meet you, I felt fulfilment and that gave me strength for my work and other more though periods of my life here in Groningen.

Finally, I would like to give a very especial acknowledgement to the most amazing people and couple I know: my parents (Paulo e Vera), who are my role models in life. Thank you for your love and, frankly, just thank you for who you are.

I could never have done this without you, not only because you made me, but due to your unconditional support and encouragement at all times. Thank you for all the opportunities you provided to me in life, thank you for giving me the chance to explore the world and different cultures. Most importantly, thank you for making me be a better person everyday.

Cássia

Voor mijn familie Para a minha família

(31)

CURRICULUM VITAE

Cássia Oraboni Ribeiro was born in January 7, 1991 in Curitiba, Brazil. In 1992 she moved with her parents and sister to Portugal (where she lived six years in Porto and nine months in Lisbon). In 1999 she moved to San- tiago, Chile, where she attended four years of basic school in Colegio Los Andes. In 2002, she moved back to Lisbon, where she followed her last years of basic school in Escola Básica de São Julião da Barra. She obtained her high school diploma from Colégio Mira Rio in 2010. In the same year she

started the bachelor and master program in Biomedical Engi- neering and Biophysics at the University of Lisbon. As part of her bachelor education, she carried out during three months an internship in Electrical and Electronic Engineering at the National University of Ireland Galway (NUIG). Her bachelor research project entitled ‘Prevention of unnecessary biopsies by the classification of breast tumours using microwave imaging’, was supervised by Dr. Martin O’Halloran and Prof.

dr. Eduardo Ducla-Soares. She started her master education with specialisation in Radiation Therapy in 2013, and as part of it, she carried out during nine months an internship at the department of Radiation Oncology of the University Medical Center Groningen (UMCG). Her master research project, entitled ‘Identification and application of image biomarkers for the prediction of radiotherapy treatment response in head and neck cancer patients’, was supervised by Dr. ir. Marianna Sijtsema, Dr. Roel Steenbakkers, Prof.

dr. Hans Langendijk, and Prof. dr. Luis Peralta. During her internship, she followed a course of Medical Physics for Radiation Oncology. After her master graduation in 2015, she attended an intensive Dutch language course at the University of Amsterdam. In February, 2015, she started

(32)

Curriculum vitae her PhD research in proton therapy at the department of Radiation Oncology of the UMCG, under the supervision of Prof. dr. Antje Knopf, Prof. dr. Hans Langendijk, and Dr.

ir. Erik Korevaar. At the beginning of her PhD, she followed an ESTRO particle therapy course in Kraków, Poland. This thesis presents the work developed during the four years of her doctorate, focused on the clinical implementation of pencil beam scanned proton therapy for the treatment of moving targets. Currently, Cássia is working as a clinical fellow at the same department. In this new position, while still conducting research, she aims to develop and optimise clinical protocols and workflows for the application of proton therapy for all moving targets.