Thomas Willigenburg MRI-GUIDED STEREOTACTIC RADIOTHERAPY for localised prostate cancer
MRI-GUIDED STEREOTACTIC RADIOTHERAPY for localised prostate cancer Thomas Willigenburg
MRI-guided stereotactic radiotherapy for localised prostate cancer PhD thesis, Utrecht University, The Netherlands © Thomas Willigenburg, 2023. All rights served. No part of this thesis may be reproduced, stored or transmitted in any way or by any means without the prior written permission from the author. The copyright of the papers that have been published or have been accepted for publication has been transferred to the respective journals. The research described in this thesis was performed at the Department of Radiation Oncology of the University Medical Centre Utrecht. Financial support for publication of this thesis was kindly provided by the Department of Radiation Oncology of the University Medical Centre Utrecht, Elekta B.V., and RT-Idea EU B.V. The research described in this thesis was supported by ZonMw IMDI/LSHTKI Foundation (The Netherlands, project number 104006004). Cover design and layout: Thomas Willigenburg Printing: Ridderprint | www.ridderprint.nl ISBN: 978-94-6458-786-9
MRI-guided stereotactic radiotherapy for localised prostate cancer MRI-gestuurde stereotactische bestraling voor gelokaliseerde prostaatkanker (met een samenvatting in het Nederlands) Proefschrift ter verkrijging van de graad van doctor aan de Universiteit Utrecht op gezag van de rector magnificus, prof. dr. H.R.B.M. Kummeling, ingevolge het besluit van het college voor promoties in het openbaar te verdedigen op donderdag 26 januari 2023 des middags te 4.15 uur door Thomas Willigenburg geboren op 25 oktober 1993 te Eemnes
Promotor: Prof. dr. ir. J.J.W. Lagendijk Copromotoren: Dr. J.R.N. van der Voort van Zyp Dr. J.C.J. de Boer Beoordelingscommissie: Prof. dr. P.J. van Diest (voorzitter) Prof. dr. R.M. Pijnappel Prof. dr. B.W. Raaymakers Prof. dr. R.J.A. van Moorselaar Dr. A.C. Tree
Voor mama
TABLE OF CONTENTS 1 General introduction and thesis outline 9 PART I 2 Evaluation of daily online contour adaptation by radiation therapists for prostate cancer treatment on an MRI-guided linear accelerator 35 3 Fast and accurate deformable contour propagation for intrafraction adaptive MRI-guided prostate radiotherapy 51 4 Seminal vesicle intrafraction motion during the delivery of radiotherapy sessions on a 1.5 T MR-Linac 65 5 Clinical application of a sub-fractionation workflow for intrafraction re-planning during prostate radiotherapy treatment on a 1.5 T MR-Linac: a practical method to mitigate intrafraction motion 93 6 MRI-guided adaptive radiotherapy for prostate cancer: the first results from the MOMENTUM study, an international registry for the evidence-based introduction of MRI-guided adaptive radiotherapy 109 7 Accumulated bladder wall dose is correlated with patient-reported acute urinary toxicity in prostate cancer patients treated with stereotactic, daily adaptive MRI-guided radiotherapy 131 PART II 8 Development and internal validation of multivariable prediction models for biochemical failure after MRI-guided focal salvage high-dose-rate brachytherapy for radiorecurrent prostate cancer 157 9 Focal salvage treatment for radiorecurrent prostate cancer: an MRI-guided stereotactic body radiotherapy versus high-dose-rate brachytherapy planning study 181 PART III 10 Summary 199 11 General discussion and future perspectives 211 APPENDICES Nederlandse samenvatting (Summary in Dutch) 233 Publications 245 Dankwoord (Acknowledgements) 249 Curriculum vitae 255
1 General introduction and thesis outline
General introduction and thesis outline 11 The present thesis evaluates technological and clinical aspects of magnetic resonance (MR) imaging (MRI)-guided radiotherapy for the treatment of primary localised and locally recurrent prostate cancer. Considering the broadness of the topics discussed in this thesis, the following introduction offers a comprehensive overview of prostate cancer, with a focus on MRI-guided radiotherapy treatment for primary prostate cancer. Furthermore, MRI-guided focal salvage radiotherapy for the treatment of locally recurrent prostate cancer after primary radiotherapy is introduced. Epidemiology of prostate cancer Prostate cancer is the second most common cancer in men and the fifth leading cause of cancer death among men worldwide.1 In The Netherlands, prostate cancer is the most common type of cancer for men above 45 years of age, with about 12.500 new cases in 2020 (Figure 1).2 Approximately 1 in 9 men are diagnosed with prostate cancer over the course of their lifetime.3 The incidence rates vary greatly between countries, which is most likely due to international differences in diagnostic practices.4 In the late 1980s and early 1990s, incidence rates rapidly increased due to the introduction of prostate-specific antigen (PSA) testing, thereby detecting asymptomatic (preclinical) prostate cancer.5 In recent years, incidence rates have declined in many countries due to recommendations against the routine use of PSA screening. Whereas incidence rates continue to decrease in some countries, incidence rates have stabilised in recent years in others, including the United States, Denmark, and Norway.4–10 On the other hand, in many countries, such as China and Eastern Europe, incidence rates continue to increase, potentially due to increased awareness and improvements in healthcare systems in these countries.10,11 The increase in incidence was accompanied by an increase in prostate cancer-specific mortality in the early 1990s, after which the mortality rates steadily dropped, most likely as a result of advances in treatment and earlier detection.12–14 There is still little known about the aetiology of prostate cancer, although some risk factors have been established. These include age (prostate cancer is most frequently diagnosed among men aged 65-74 and is rarely seen in men below 40 years of age), family history of prostate cancer (especially when diagnosed before the age of 65), genetic mutations such as BRCA1 and BRCA2, and genetic conditions such as Lynch syndrome.1 Furthermore, Western African ancestry is thought to modulate the prostate cancer risk, with black males having the highest incidence rates.15 Due to early diagnosis, most patients (75-80%) have localised disease at presentation.16 Around 20% of the prostate cancer patients present withmetastasised disease, of whom approximately 65%with regional lymph node involvement only and 35%with distant metastases.16,17 Prostate cancer survival is high – in fact, prostate cancer survival is the highest of all cancers in the United States – with a 5year relative survival for all stages combined of 98%.14 However, the stage at diagnosis significantly impacts the relative survival, with excellent 5-year relative survival rates in patients with localised or regional disease (99.3-100%) and much worse 5-year relative survival rates of 32.3% in patients with distant metastases at presentation.17 1
Chapter 1 12 Symptoms, diagnosis, and staging Early-stage prostate cancer usually causes no symptoms and therefore is only diagnosed after finding an elevated serum PSA ‘by accident’ in the majority of these patients. PSA is often measured when men present with urinary symptoms, which are often caused by benign prostate hyperplasia. More advanced prostate cancer, however, can present with genitourinary (GU) symptoms, such as weak or interrupted urine flow, difficulty with starting or stopping urination, frequent urination (most notably during the night, i.e. nocturia), and pain or burning sensation during urination. Metastasised disease can present with pain due to spreading of cancer to the bones. Regarding the value of PSA screening in asymptomatic men, there is still an ongoing debate. The arguments against the routine screening of PSA levels in males include the risk of over-diagnosis and unnecessary treatment of indolent tumours, consequently leading to financial, emotional, and physical burden.8,18 However, decisions against routine testing were largely based on criticised clinical trial data with insufficient follow-up time.8,14,19 Arguments in favour of routine PSA measurements include the risk of delaying diagnosis and consequently diagnosing more advanced disease, thereby negatively affecting survival rates.20,21 Because of an increase in the number of patients presenting with distant-stage prostate cancer since 2010, recommendations regarding PSA screening have changed.22,23 The current clinical guidelines recommend shared decision-making for PSA testing in men, balancing the benefits and harms based on individual characteristics (i.e. age, family history for prostate cancer, black males) and preferences.24 More recent evidence suggests that the long-term benefit of PSA screening is underappreciated, especially when considering Figure 1 – Trends in cancer incidence rates among males in the Netherlands. Image adapted from: Netherlands Cancer Registry (NCR), Netherlands Comprehensive Cancer Organisation (IKNL).
General introduction and thesis outline 13 advances in preventing over-detection and reducing over-treatment by implementing active surveillance strategies in patients with low-risk prostate cancer.19,25,26 Historically, physical examination in the form of digital rectal examination (DRE) was the most important tool clinicians had to diagnose prostate cancer. Still, current guidelines suggest performing DRE to establish the clinical tumour stage (T-stage).27,28 Four main prostate cancer Tstages are distinguished (Figure 2).27 Nowadays, diagnostic imaging plays an important role in the diagnosis and accurate staging of prostate cancer.29 The last years, diagnostic imaging has vastly improved with the introduction of multiparametric magnetic resonance imaging (mp-MRI) and prostate-specific membrane antigen (PSMA) positron emission tomography (PET) computed tomography (CT) imaging (PSMA-PET/CT). mp-MRI has emerged as an important diagnostic tool in early diagnosis of clinically significant prostate cancer by facilitating targeted prostate biopsies.30 Pre-biopsy use of MRI can reduce the number of unnecessary biopsies, minimise over-diagnosis of clinically insignificant prostate cancer, and improve risk stratification of patients, warranted that high-quality MRI scans are used and radiologists are adequately trained to deliver reliable reports.31,32 Whereas PSMA-PET/CT is already widely applied in case of (suspected) recurrent prostate cancer, its role in primary prostate cancer diagnosis and risk classification is still partly undetermined. Further research is required and currently underway, but there may be a role for PSMA-PET/CT in characterising tumour biology, complementing mp-MRI diagnostics, and identifying lymph node metastases in high-risk prostate cancer patients.33,34 Figure 2 – Four main prostate cancer tumour stages. 1
Chapter 1 14 Besides DRE and radiological examination, histopathological verification is obtained through systematic biopsies. Currently, the European Association of Urology (EAU) guidelines recommend mp-MRI-targeted biopsies, as this improves the detection of International Society of Urological Pathology (ISUP) grade ³ 2 prostate cancer.28 Histopathology is needed for verification of the tumour type (most often adenocarcinoma) and for establishing the Gleason score.35 The Gleason score consists of two scores for the two most dominant growth patterns. This ranges from one (well differentiated) to five (poorly differentiated). The two scores result in a sum score (Table 1). In case only a single dominant growth pattern is present, this is counted twice. In the original grading system, sum scores ranged from two to ten. Currently, sum scores 2-5 are no longer assigned.36 Five grade groups, as suggested by Epstein et al.36, can be distinguished based on the sum score (Table 1). Gleason sum score Grade group £ 6 1 3 + 4 = 7 2 4 + 3 = 7 3 8 4 9 or 10 5 The combination of the clinical T-stage, Gleason grade group, and serum PSA level allows for stratification of patients into risk groups (Table 2). These prognostic groups are based on the risk of cancer recurrence after treatment. Several risk group stratification guidelines have been established, with minor differences in the definition of intermediate- and high-risk disease.28,37–39 In recent years, specific risk groups have been identified – such as favourable and unfavourable intermediate-risk – to aid treatment individualisation (Table 2).37 Treatments for primary localised prostate cancer Several curative treatment options exist for patients with localised, non-metastatic prostate cancer, depending on the risk classification. In case of active treatment, whole-gland treatment is generally offered to patients in the form of surgery (radical prostatectomy [RP]) or radiotherapy. Both treatments appear to have equivalent outcomes with respect to disease control in selected patients.40 Nevertheless, there are differences in terms of expected side effects and potential complications. In (very) low-risk patients, active surveillance is the preferred strategy, as these indolent tumours may not need active treatment right away – or ever. Long-term data in predominantly low-risk patients shows that 10-year prostate cancer-specific mortality is very low (approximately 1%) and does not differ between patients who initially undergo active treatment (RP or radiotherapy) or active surveillance (with or without active treatment at a later point in time).40 However, disease progression occurs – not unexpected – more frequently in patients who undergo Table 1 – Gleason sum scores and grade groups.
General introduction and thesis outline 15 NCCN risk group Clinical/pathological features Very low-risk T1c AND grade group 1 AND PSA < 10.0 ng/mL AND < 3 positive biopsy fragments/cores with £ 50% cancer in each AND PSA density < 0.15 ng/mL/g. Low-risk T1-T2a AND grade group 1 AND PSA < 10.0 ng/mL AND does not qualify for Very low-risk. Intermediate-risk Intermediate-risk factors: - T2b-T2c - Grade group 2-3 - PSA 10.0-20.0 ng/mL Favourable No more than one intermediate-risk factor (see left) AND grade group 1-2 AND < 50% positive biopsy cores. Unfavourable Two or more intermediate-risk factors (see left) AND/OR grade group 3 AND/OR ³ 50% positive biopsy cores. High-risk T3a OR grade group 4-5 OR PSA > 20.0 ng/mL AND does not qualify for Very high-risk. Very high-risk T3b-T4 OR primary Gleason pattern 5 OR > 1 High-risk feature OR > 4 cores with grade group 4-5. active surveillance.40 In patients with high-risk disease on the other hand, aggressive multi-modal treatment should be provided and mostly consists of either radiotherapy with systemic androgen deprivation therapy (ADT) and/or a brachytherapy (see ‘Radiotherapy techniques’) tumour boost or surgery with additional radiotherapy to the prostate bed or ADT.28 Patients with localised intermediate-risk prostate cancer are usually treated with radiotherapy (with or without ADT and/or a (brachytherapy) boost) or RP (with or without pelvic lymph node dissection [PLND]).28 With RP, the entire prostate gland and seminal vesicles are surgically removed. This surgery is nowadays often performed using a laparoscopic, robot-assisted approach (robot-assisted radical prostatectomy [RARP]). This approach usually shows low complication rates (< 5%).41 Potential complications for RP include excessive blood loss needing blood transfusion, organ injury, infection, and anastomotic leakage.41 However, post-operative comorbidity is much more common and can have a major impact on quality of life.42 This mainly consists of urinary incontinence (up to 21% at one year post-surgery) and erectile dysfunction (up to 75% at one year post-surgery) and the incidences depend, apart from patient characteristics, on the surgical technique, the surgeon, and hospital volumes.28,42,43 In addition to surgery and radiotherapy, several investigational treatments for localised prostate cancer exist. These include freezing (cryotherapy) and heating (high-intensity focused ultrasound [HIFU]) of the tumour and prostate.28 For cryotherapy, cryo-needles are inserted into the prostate under ultrasound-guidance. HIFU, on the other hand, uses focused ultrasound as a therapeutic mean. Although outcomes in small, single arm case-studies have been published, directly comparative data with surgery and/or radiotherapy is scarce.44 Table 2 – National Comprehensive Cancer Network (NCCN) prostate cancer risk groups. 1
Chapter 1 16 Radiotherapy treatment Radiotherapy uses ionising radiation to cause damage to double-stranded DNA within cancer cells, thereby depriving tumour cells of their reproductive potential and inducing cell death (apoptosis).45 The goal of radiotherapy is to stop cancer cell reproduction and to kill cancer cells, while sparing healthy surrounding tissue as much as possible. This can be achieved to a certain extent, as cancer cells are generally more sensitive to DNA damage caused by radiation compared to healthy tissues.45,46 Nevertheless, the tolerance of normal tissues surrounding the tumour limits the dose that can be prescribed. With modern day radiotherapy techniques, the therapeutic ratio – the balance between cure and toxicity – has significantly improved.47 Radiotherapy techniques Radiotherapy can be administered in two ways: either using external radiotherapy (i.e. external beam radiation therapy [EBRT]) or internal radiotherapy (i.e. brachytherapy [BT]). With EBRT, the patient is treated on a linear accelerator (often called linac). In short, the linac accelerates electrons that collide with a heavy metal, thereby producing high-energy photons. These photons are then directed at the target and enter the patient through the skin. EBRT can also be proton-based, with theoretical benefits in terms of organ-sparing compared to photon-based radiotherapy.48 Nevertheless, these benefits have not (yet) been demonstrated in clinical practice.48 Brachytherapy, on the other hand, uses the insertion of radioactive seeds into the tumour to deliver radiation. Low-dose-rate brachytherapy (LDR-BT) uses permanent insertion of radioactive seeds, whereas high-dose-rate brachytherapy (HDR-BT) uses temporary insertion of a radioactive source with a much higher dose rate (³ 12.0 Gy/h compared to £ 2.0 Gy/h for LDR-BT).49 The biggest differences between EBRT and brachytherapy can be observed in the dose distributions.49,50 With EBRT, the dose fall-off is less steep compared to brachytherapy. The further away from the centre of the target (iso-centre), the lower the dose is. Because the dose fall-off is much steeper with brachytherapy, this leads to less radiation dose in surrounding healthy tissues. Furthermore, the high-dose areas are often much larger with brachytherapy, as EBRT is mostly limited by the radiation dose received by healthy tissues. Modern day EBRT techniques, such as intensity-modulated radiation therapy (IMRT) and volumetricmodulated arc therapy (VMAT), allow the radiation beam to be shaped to closely follow the form of the target. In addition, with these techniques the dose is delivered in multiple smaller beams from different angles using a cross-fire technique.51 This way, a conformal dose distribution is achieved that is optimised to treat the target and spare healthy tissues as much as possible. For prostate cancer treatment, the radiation dose is delivered to the patient in multiple treatment fractions on separate days (fractionated treatment). With conventional CT-guided radiotherapy, a single treatment plan is created before treatment based on the pre-treatment CT scan. However, the position of the prostate within the pelvic area can change significantly between fractions (interfraction) due to e.g. organ motion and tension of the pelvic muscles. Therefore, the position of the prostate is checked prior to each treatment fraction. To be able to verify the position of the prostate on conventional cone-beam (CB) CT-guided systems during each fraction, fiducial gold
General introduction and thesis outline 17 markers (fiducials) are inserted into the prostate prior to treatment. These fiducials can be visualised prior to dose delivery. This way, corrections for (large) interfraction shifts can be performed. Hypofractionated radiotherapy Over the years, radiotherapy schemes have changed significantly for prostate cancer patients. Whereas 10 years ago patients received no less than 35 treatment fractions, nowadays low- and intermediate-risk patients are commonly treated with only five fractions.52,53 Although patients are treated with fewer fractions, the dose per fraction has also increased significantly from in general 2.0 Gray (Gy) to ³ 7.0 Gy for some indications. These shifts find their origin in the fact that prostate cancer appears particularly sensitive to radiation and more specifically, to the dose delivered per treatment fraction. Radiosensitivity is often expressed by the alpha/beta ratio (α/β). For prostate cancer, the α/β is estimated to be as low as 1.5 Gy.54 Most cancers actually have a much higher α/β, often around 10 Gy for rapidly proliferating tumours.55 These are more sensitive to a higher overall radiation dose, whereas a low α/β means that the tumour is more sensitive to the radiation dose per treatment fraction and less to the total delivered dose. With hypofractionation, the treatment is delivered in less fractions than with ‘conventional’ fractionation (i.e. 20 compared to 39 fractions for prostate cancer). With terms such as ultra-hypofractionation and extreme hypofractionation, mostly treatment delivery in five or less fractions with a high fractional dose (³ 7.0 Gy) is implied. This is also referred to as stereotactic body radiation therapy (SBRT). Because organs-at-risk (OARs) generally have a higher α/β compared to prostate cancer tissue, it was hypothesised that larger fractional doses might not only improve oncological outcomes, but also increase the therapeutic ratio (the ratio between oncological outcomes and toxicity).54 The vast improvement in accuracy of radiotherapy treatment delivery with modern day techniques, and therefore better sparing of OARs, has created room for ultra-hypofractionated radiotherapy. The last years, results from several trials have been published that directly compared conventional fractionated radiotherapy with five-fraction SBRT for the treatment of localised prostate cancer, including the randomised HYPO-RT-PC and PACE-B trials.56,57 No significant differences were observedwith respect to oncological outcomes, with a 5-year failurefree survival rate of 84% in both treatment groups in the HYPO-RT-PC trial.56 Furthermore, in the HYPO-RT-PC trial weak evidence was found for higher acute grade ³ 2 urinary toxicity in the SBRT group, although this was statistically non-significant (28% for SBRT versus 23% for conventional fractionated radiotherapy in 39 fractions).56 A systematic review and meta-analysis by Jackson et al.58 confirmed the excellent oncological outcomes after SBRT, with an overall 5- and 7-year biochemical recurrence-free survival (bRFS) of 95.3% and 93.7%, respectively. In addition, severe (grade ³ 3) late GU and gastrointestinal (GI) toxicity was estimated to be 2.0% and 1.1%, respectively.58 Not only is SBRT considered effective and safe for the treatment of low- and intermediate-risk prostate cancer, it also positively impacts patient burden and departmental capacity, as patients need to make fewer visits to the hospital compared to conventional hypofractionated therapy.59 However, with increasing fractional doses and fewer fractions to deliver the complete dose, delivery accuracy becomes even more important, especially in the light of intrafraction target and OAR motion.60–62 With higher fractional doses, the overall daily treatment time is prolonged due to longer beam-on times, thus allowing more intrafraction motion and deformations to occur.62 Intrafraction 1
Chapter 1 18 motion can significantly impact the dose delivered to the target as well as the OARs, thereby potentially increasing the risk of toxicity and lowering the effectiveness of treatment.63 With fewer fractions, there is less room for delivery errors to be blurred over the course of treatment. Therefore, optimal image-guidance before and during treatment as well as intrafraction adaptation possibilities are desired. MRI-guided radiotherapy Over the past decades, image-guided radiotherapy (IGRT) has evolved immensely. Technological developments have eventually led to the clinical introduction of MRI-guided radiotherapy using MRI-guided linear accelerator systems (MR-Linac). Currently, two systems are on the market: the 0.35 Tesla (T) MRIdian (ViewRay Inc., Oakwood, U.S.A.) and the 1.5 T Unity (Elekta AB, Stockholm, Sweden). The idea for integrating a linac with a 1.5 T MRI scanner originated in Utrecht.64–66 The 1.5 T MR-Linac system was developed at the University Medical Centre Utrecht (UMC Utrecht, Utrecht, The Netherlands), together with Philips (Best, The Netherlands) and Elekta AB (Stockholm, Sweden). It consists of a 1.5 T Philips MRI scanner surrounded by a 7 megavolt (MV) linac, mounted on a ringshaped gantry. The Unity MR-Linac provides diagnostic quality MR imaging before and during treatment. In May 2017, the first patient was treated on the system.67 In Utrecht, the first prostate cancer patient was treated on an MR-Linac in February 2019 with 20 fractions of 3.1 Gy.62 The greatest benefit of MRI-guided linacs over CT-guided linacs is the superior soft-tissue contrast, allowing clear visualisation of the target and OARs.59 Because of the superior contrast, no fiducials are needed to verify the position of the prostate. Also, tattoos for patient positioning are no longer required with the daily plan adaptation. Additionally, continuous 3D imaging is possible and no ionising radiation is used to obtain MR images, allowing unlimited imaging without additional harm to the patient.61,62 Treatment preparation for MRI-guided SBRT Prior to the actual dose delivery (online phase), the treatment is prepared (offline phase) by acquiring pre-treatment imaging of the target area, generally consisting of CT and/or mp-MRI scans. For most prostate cancer patients at the UMC Utrecht, an MRI-only workflow is applied, i.e. without obtaining a planning CT scan but instead using a pseudo-CT scan created from dedicated MR images.68,69 The treating physician identifies and delineates the target and relevant OAR structures on the MRI scan (Figure 3). For the treatment of prostate cancer, the Gross Tumour Volume (GTV) is delineated. The GTV encompasses the visible tumour on mp-MRI. To cover any potential microscopic tumour spread, the GTV is enlarged to create the Clinical Target Volume (CTV). The CTV includes the prostate body and the GTV with a 4 mm margin, excluding the bladder and rectum. Depending on the location of the tumour and the risk classification, the seminal vesicles are (partly) included in the CTV. To account for geometric inaccuracies and uncertainties, including machine uncertainties and patient set-up errors, the CTV is enlarged by a 5 mm isotropic margin to create the Planning Target Volume (PTV). In addition to the target, the bladder, rectum, and several bony structures (e.g. femur heads) are delineated and identified as OAR. Using the pre-treatment imaging and delineations, a baseline radiotherapy treatment plan is created. The treatment plan is optimised to ensure adequate target coverage without violating the dose constraints for the OARs.
General introduction and thesis outline 19 Online clinical workflow During the actual treatment delivery (online phase), the workflow consists of several steps (Figure 4). For prostate cancer SBRT treatment, a so-called Adapt-to-Shape (ATS) workflow is used, in which a daily treatment plan is created for the anatomy of the day, using a new MRI scan with updated contours.70 Before treatment can be started, the patient is positioned on the treatment couch. After patient positioning, an initial daily MRI scan is acquired to assess the daily anatomy (step 2). Using deformable image registration (DIR), this daily MRI scan is registered to the pre-treatment planning MRI and the target and OAR contours are warped from the pre-treatment image to the daily MRI (step 3). An operator – either a radiation therapist or a physician – checks the contours andmanually adapts them wherever needed (step 4). Using the daily MRI and the updated contours, a new daily treatment plan is created (step 5). Just before the end of dose calculation, a second daily MRI scan is acquired – a so-called position verification (PV) scan – to verify the position of the target (step 6). In case any major shifts have occurred between the initial daily MRI and the PV MRI scan, a virtual couch shift (VCS) or Adapt-to-Position (ATP) can be applied to re-align the position.70,71 After approval of the treatment plan, treatment delivery is started with simultaneous acquisition of 3D cine-MR images to track the prostate motion (step 7).62 The duration of one treatment fraction is approximately 45 min (including patient positioning), which is well tolerated by patients. The current, commercially available Unity workflows allow for interfraction adaptation of the treatment plan according to the anatomy of the day. Ultimately, these machines will enable realtime adaptation during treatment delivery, to counteract intrafraction motion and deformations. However, at the moment of writing, real-time intrafraction adaptive treatments are not yet clinically available on 1.5 T MR-Linac systems. Figure 3 – Transversal MR image with the Gross Tumour Volume (GTV), Clinical Target Volume (CTV), and Planning Target Volume (PTV) contours for primary prostate cancer SBRT. 1
Chapter 1 20 MOMENTUM study and Utrecht Prostate Cohort study The chapters in the first part of this thesis are based on manuscripts that use data from two prospective cohort studies. All patients, including prostate cancer patients at the UMC Utrecht, who are treated on one of the MR-Linacs at a participating centre, are asked to participate in the MOMENTUM (‘Multi-OutcoMe EvaluatioN of radiation Therapy Using the MR-Linac’) study (NCT04075305).72 The MOMENTUM study is an international, prospective registry that was established with the aim of facilitating evidence-based clinical implementation of the 1.5 T Unity MR-Linac. Within the registry, clinical and technical data is collected at baseline, during treatment, and during follow-up. Since February 2020, all patients who are diagnosed with primary localised prostate cancer at the UMC Utrecht or St. Antonius hospital (Utrecht and Nieuwegein, The Netherlands) are asked to participate in the prospective Utrecht Prostate Cohort (UPC) study (NCT04228211). Whereas the MOMENTUM study solely includes patients treated on an MR-Linac, the UPC study additionally includes patients treated with other primary treatments or who undergo an active surveillance strategy. Like the MOMENTUM study, the UPC study prospectively collects clinical patient and treatment data at baseline and during follow-up in patients with primary localised prostate cancer. Besides physician-reported outcomes, patient-reported outcomes on toxicity and quality of life are collected using multiple questionnaires. Both cohorts allow evaluation of current treatments and outcomes as well as the initiation of (randomised) trials that investigate a new treatment. Furthermore, the imaging and technical data collected in the studies can be used in many different ways, for example to study technological innovations. Figure 4 – Overview of the clinical MR-Linac workflow for prostate cancer patients. Legend: CT = Computed Tomography. MRI = Magnetic Resonance Imaging. PV = Position Verification. ATP = Adapt-to-Position.
General introduction and thesis outline 21 Locally recurrent prostate cancer Although recurrence rates after primary prostate cancer are very low in low-risk patients, still up to 50% of the high-risk patients develop a (biochemical) recurrence within 10 years after treatment.73 The management of recurrent prostate cancer remains controversial, as the natural course is very heterogeneous.74 In some patients, the cancer remains indolent for many years, without becoming clinically relevant. In others, the cancer progresses rapidly, causing morbidity and mortality. Generally, biochemical recurrence precedes both progression to distant metastases and prostate cancer-specific mortality.73,75 As with primary prostate cancer, it remains important to weigh the potential benefits and harms of treatment. Nowadays, many patients with biochemical recurrence are still treated with (deferred) ADT.76 However, ADT is a systemic treatment and is associated with significant side effects, such as hot flushes, osteoporosis, loss of libido, erectile dysfunction, and therefore deterioration of quality of life.77 Moreover, tumour progression under hormonal treatment (i.e. castration resistant prostate cancer) usually occurs within 1-3 years after start of hormonal treatment, requiring further systemic treatment with for example second-line hormonal treatment or chemotherapy.29,78–80 Local treatment for recurrent prostate cancer Since after radiotherapy the recurrent lesion (radiorecurrent prostate cancer) is often confined to the prostate, local therapy is an attractive treatment option with the aim of postponing toxic systemic treatment. Various salvage treatment modalities, such as salvage radical prostatectomy, brachytherapy, HIFU, and cryotherapy have been investigated for patients with local, organconfined recurrences.81–84 In the early days, salvage treatment was aimed at treating the entire prostate gland (so-called whole-gland treatment). These treatments were associated with high rates of urinary incontinence, impotence, fistulae, and urethral strictures.85–88 Focal salvage treatment Prostate tumours seem to recur most often at the site of the primary dominant lesion (the so-called index lesion) and are mostly unifocal.89,90 Because of this, focal salvage is an attractive treatment option for patients with a local, intraprostatic recurrence. Due to improved imaging techniques, such as mp-MRI and PSMA-PET/CT, focal treatments have become available over the last two decades. With focal salvage treatment, only a part of the prostate, such as the tumour or the quadrant or hemi-prostate containing the MRI- and PSMA-PET/CT-visible tumour, is targeted. The advantage of focal salvage over whole-gland salvage treatment is the reduced risk of (severe) side effects.82 Due to the lower treatment volume, OARs can be spared more optimally. Multiple focal salvage techniques have been investigated over the last years in – often small and retrospective – single arm cohort studies. Currently, focal salvage treatment is performed using a range of modalities, mostly still within clinical trials. These include brachytherapy91–99, cryotherapy100–106, and HIFU107–110. As the studies often included a small sample size, and because of heterogeneity with respect to inclusion criteria, the results are difficult to compare. Overall, focal salvage techniques seem to have similar outcomes with respect to cancer control when compared to whole-gland 1
Chapter 1 22 salvage treatment. However, the incidence of moderate and severe (grade ³ 3) toxicity often seems to be lower. A recent review showed that grade 3 GU toxicities ranges from 0-10% for all focal salvage modalities.83 With respect to oncological outcomes, results seem comparable between the different focal salvage modalities. For example, for focal salvage cryotherapy, biochemical recurrencefree survival (bRFS) ranges from 48.1-95.3% at 1 year, to 48.1-72.4% at 2 years, and 46.5-54.4% at 5 years.100–106,111 For focal salvage high-dose-rate brachytherapy (FS-HDR-BT), bRFS ranges from 95.2% at 1 year, to 61-92% at 2 years, and 41.8-61% at 3 years.95,97–99,112 bRFS seems somewhat higher (61% at 3 years) in the study by Murgic et al.95, who used two fractions of 13.5 Gy, compared to those investigating a single fraction HDR treatment (41.8% and 46% at 3 years).98,99 However, as stated before, differences in inclusion criteria and thus risk groups make it difficult to draw definite conclusions regarding differences in efficacy from these small studies. MRI-guided focal salvage high-dose-rate brachytherapy At the Department of Radiation Oncology of the UMC Utrecht, several prospective studies were initiated to investigate the efficacy and safety of MRI-guided FS-HDR-BT, including the PRECISE study (NTR7014). In these studies, patients were treated in a single fraction with a dose of 19.0 Gy to the CTV. The CTV consisted of the mp-MRI visible tumour (GTV) with a 5 mm isotropic expansion, staying within the prostate boundaries and excluding the urethra (Figure 5). The FS-HDR-BT treatment is extensively described in earlier publications.92 In short, pre-treatment imaging (3 T MRI and PSMA-PET/CT) is acquired to identify the target and OARs and to create the Figure 5 – Transversal image of contours used for focal salvage high-dose-rate brachytherapy (FS-HDR-BT) treatment of a lesion in the right hemi-prostate. The Gross Tumour Volume (GTV), Clinical Target Volume (CTV), prostate, urethra, and rectum contours are displayed.
General introduction and thesis outline 23 pre-treatment plan. During the FS-HDR-BT procedure, patients undergo spinal anaesthesia. Brachytherapy catheters are then inserted into the prostate via the perineum under ultrasoundguidance. After catheter insertion, an intraoperative MRI scan is acquired to verify the position of the brachytherapy catheters and to update the treatment plan accordingly. After this phase, the dose is delivered to the patient by transporting a radioactive iridium-192 source from the HDR afterloader into the prostate. Generally, patients can leave the hospital the same day. From these studies, as well as from other FS-HDR-BT studies, we have seen thus far very promising results with regard to GU and GI toxicity of this focal treatment.113 Although new-onset grade 2 GU toxicity is reported to be up to 40%, grade 3 GU and GI toxicity is generally low (£ 10%).95,98,99,113,114 Nevertheless, initial analysis of oncological outcomes showed a bRFS rate at 2.5 years of 51%, leaving room for treatment improvement and/or better selection of patients.112 MRI-guided focal salvage stereotactic body radiation therapy Several small, retrospective studies have also investigated partial prostate re-irradiation of locally recurrent prostate cancer using SBRT.83,115–121 Most of these studies report on SBRT treatment using a 5-7 fraction scheme with a fractional dose of around 6-7 Gy. Again, oncological outcomes are comparable to other focal salvage techniques. Overall, bRFS ranges from 68-85.7% at 1 year116,117,119, to 40-73% at 2 years115,116,118, and 55% at 3 years118. Furthermore, higher-grade GU and GI toxicity seems low, with up to 2% acute grade 3 GU toxicity, up to 10% late grade 3 GU toxicity, no grade 4 GU toxicity, and no acute or late grade 3 or higher GI toxicity.83 Grade 2 GU and GI toxicity is reported to be up to 25%, with most studies reporting values between 0-10%.83 However, again follow-up is predominantly short, and the study populations are quite heterogeneous, sometimes also including patients with prostate bed irradiation. Furthermore, the retrospective nature of these studies could have led to an underestimation of treatment-related toxicity. According to the EAU-ESTRO-SIOG guidelines, there is currently no role for salvage SBRT due to dose limitations to surrounding OARs and therefore limited chance of cure with conventional EBRT techniques.29 Still, because SBRT is a less invasive treatment compared to FS-HDR-BT, it is potentially an attractive treatment option. This is especially true when fractionated treatment is preferred. However, to safely deliver a high fractional dose of e.g. 13.5 Gy or 19.0 Gy, more accurate dose delivery is needed than is possible with conventional CBCT linacs. With CBCT-guided radiotherapy, error margins often need to be ³ 3 mm to assure adequate target coverage. In addition, when no online adaptation (during beam-on time) is possible, these margins need to incorporate all the expected intrafraction motion. For these high fractional doses, delivery times are well over 10 min and thus the expected intrafraction motion is quite large.61 Because of this and because of the less steep dose fall-off with EBRT in comparison with brachytherapy, the dose to the surrounding OARs is substantial. This is especially true when no adjustments during treatment can be performed to counteract the effect of intrafraction motion and deformations. In these instances, OAR constraints might be violated, or the target dose cannot be achieved. The only option to reach the target dose while respecting the OAR constraints, is to deliver the dose more accurately with small error margins (e.g. £ 2 mm). 1
Chapter 1 24 Thesis outline This thesis covers a wide range of topics with a common denominator: MRI-guided radiotherapy for the treatment of prostate cancer. It provides evaluations of technological and clinical aspects of MRIguided radiotherapy in patients with primary localised and locally recurrent prostate cancer. Part I of this thesis focuses on the clinical introduction, technological developments, and clinical evaluation of MRI-guided SBRT using a 1.5 T MR-Linac for the treatment of primary localised prostate cancer. In February 2019, the first prostate cancer patient was treated on an MR-Linac at the Department of Radiation Oncology of the UMC Utrecht. During the first year, the online clinical workflow was fine-tuned. To reduce the workload of physicians, some of the work was carried onto radiation therapists (RTTs). This included the check and manual adaptation of contours. In chapter 2, the feasibility of this RTT-led workflow for prostate cancer treatment is assessed, with a focus on the clinical quality of the contours used for the actual treatment of patients. Since manual contour adaptation for daily adaptive treatment is quite labour-intensive and timeconsuming, our work has focussed on optimising the online clinical workflow. By improving the interfraction propagated target and OAR contours, the workload can be reduced. Furthermore, fast, online adaptive workflows can be enabled with accurate intrafraction contour propagation. In chapter 3, we assess the clinical quality and usability of propagated contours that were created by a deformable image registration algorithm. This work paves the way for exploring intrafraction adaptive workflows. So far, only low- and intermediate-risk patients have been treated with SBRT using an online adaptive ATS workflow on the MR-Linac. Patients with high-risk disease often have more extensive disease with involvement of the seminal vesicles. Currently, these patients are still treated in ³ 20 fractions. Before progressing to SBRT-like treatments in these patients, information on intrafraction seminal vesicle motion during treatment is warranted. At our department, a soft tissue-tracking algorithm was developed that uses 3D cine-MR images to track the motion of the prostate during radiotherapy treatment. In chapter 4, a similar method is described to determine the intrafraction motion of the seminal vesicles from 3D cine-MR images using this soft tissue-tracking algorithm. This work serves as a basis for future MR-Linac treatment of high-risk prostate cancer patients. Real-time, intrafraction adaptive treatments are – at the moment of writing – not yet clinically available on 1.5 T MR-Linac systems. Therefore, intrafractionmotion can still have a significant impact on the accuracy of dose delivery. This is especially true for SBRT treatments with relatively long beam-on times. In chapter 5, a new intrafraction adaptive workflow for 1.5 T MR-Linac systems is presented. The sub-fractionation workflow enables the delivery of multiple treatment plans within a single treatment session. This allows the delivery of a fraction in multiple parts (sub-fractions), e.g. a fraction of 7.25 Gy can be delivered in two sub-fractions of 3.625 Gy each. For each sub-fraction, the plan is updated based on the latest anatomy and this way, systematic (drift) intrafraction motion can potentially be counteracted. For efficiency, imaging and treatment planning are performed
General introduction and thesis outline 25 simultaneously with dose delivery. In chapter 5, the workflow is tested in 15 prostate cancer patients and the effects on intrafraction prostate motion and required error margins are assessed and discussed. At the moment of writing this thesis, more than 650 prostate cancer patients treated with SBRT on a 1.5 T MR-Linac have been included in the MOMENTUM study. In chapter 6, the initial 12-month outcomes are presented for patients treated with MRI-guided SBRT (5 x 7.25 Gy) on a 1.5 T MR-Linac using daily imaging and treatment plan optimisation. PSA kinetics, physician-reported toxicity, and patient-reported outcomes were analysed in the first 425 patients. These results provide early insight into the short-term outcomes of MRI-guided prostate cancer SBRT and can be used to guide further research. Toxicity after radiotherapy can have significant, detrimental effects on quality of life. Since lowergrade (grade 1-2) urinary toxicity is still quite common after prostate cancer SBRT in the first months following treatment, there remains room for improvement. By correlating the dose that is delivered to specific organs or (sub)structures with outcomes (toxicity), we can potentially improve the radiotherapy plan to lower the risk of toxicity. However, data on the association between toxicity and the actual delivered dose is scarce. The data that is obtained through MR-Linac systems (including images and daily treatment plans) allows for a precise estimation of the actual delivered dose. In chapter 7, we assess the correlation between the accumulated dose received by the bladder and bladder wall and patient-reported acute urinary toxicity, and we propose new bladder (wall) dose constraints. In part II of this thesis, MRI-guided radiotherapy treatment for locally recurrent prostate cancer after primary radiotherapy is discussed. Although the toxicity rates after MRI-guided FS-HDR-BT seemvery favourable, biochemical failure after treatment occurs frequently and often within two years after treatment. Currently, it is unclear which patients are likely to benefit most from this focal salvage treatment. In chapter 8, the development and internal validation of two prediction models for biochemical failure after FS-HDR-BT treatment, in patients with locally radiorecurrent prostate cancer, is described. These models can be used to counsel patients before and during treatment. For easy use in clinical practice, two interactive nomograms were established. Since FS-HDR-BT is an invasive treatment – patients need to undergo spinal anaesthesia and hollow needles are inserted into the prostate – not all patients are eligible for treatment. MR-Linac systems could open up possibilities for a comparable (i.e. single fraction focal treatment with a high dose), non-invasive treatment using high-precision SBRT. In chapter 9, the planning feasibility of single fraction (19.0 Gy) SBRT treatment for locally recurrent prostate cancer on a 1.5 T MR-Linac is investigated. Chapter 10 provides a summary of the presented work. Chapter 11 presents a general discussion of MRI-guided (stereotactic) radiotherapy for the treatment of localised primary and recurrent prostate cancer and describes the future perspectives. 1
Chapter 1 26 References 1. Sung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A, et al. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin 2021;71:209–49. 2. IKNL Nederlandse Kanker Registratie. NKR Cijfers. Accessed via: https://iknl.nl/nkr-cijfers (accessed June 1, 2022). 3. American Cancer Society. Cancer Facts & Figures 2021. Accessed via: https://www.cancer.org/content/dam/cancerorg/research/cancer-facts-and-statistics/annualcancer-facts-and-figures/2021/cancer-facts-andfigures-2021.pdf (accessed June 1, 2022). 4. Zhou CK, Check DP, Lortet-Tieulent J, Laversanne M, Jemal A, Ferlay J, et al. Prostate cancer incidence in 43 populations worldwide: An analysis of time trends overall and by age group. Int J Cancer 2016;138:1388– 400. 5. Center MM, Jemal A, Lortet-Tieulent J, Ward E, Ferlay J, Brawley O, et al. International variation in prostate cancer incidence and mortality rates. Eur Urol 2012;61:1079–92. 6. Kvåle R, Auvinen A, Adami H-O, Klint A, Hernes E, Møller B, et al. Interpreting trends in prostate cancer incidence and mortality in the five Nordic countries. J Natl Cancer Inst 2007;99:1881–7. 7. Zargar H, van den Bergh R, Moon D, Lawrentschuk N, Costello A, Murphy D. The impact of the United States Preventive Services Task Force (USPTSTF) recommendations against prostate-specific antigen (PSA) testing on PSA testing in Australia. BJU Int 2017;119:110–5. 8. Moyer VA. Screening for prostate cancer: U.S. Preventive Services Task Force recommendation statement. Ann Intern Med 2012;157:120–34. 9. Tikkinen KAO, Dahm P, Lytvyn L, Heen AF, Vernooij RWM, Siemieniuk RAC, et al. Prostate cancer screening with prostate-specific antigen (PSA) test: a clinical practice guideline. BMJ 2018;362:k3581. 10. Culp MB, Soerjomataram I, Efstathiou JA, Bray F, Jemal A. Recent Global Patterns in Prostate Cancer Incidence and Mortality Rates. Eur Urol 2020;77:38– 52. 11. Seraphin TP, Joko-Fru WY, Kamaté B, Chokunonga E, Wabinga H, Somdyala NIM, et al. Rising Prostate Cancer Incidence in Sub-Saharan Africa: A Trend Analysis of Data from the African Cancer Registry Network. Cancer Epidemiol Biomarkers Prev 2021;30:158–65. 12. Tsodikov A, Gulati R, Heijnsdijk EAM, Pinsky PF, Moss SM, Qiu S, et al. Reconciling the Effects of Screening on Prostate Cancer Mortality in the ERSPC and PLCO Trials. Ann Intern Med 2017;167:449–55. 13. Etzioni R, Tsodikov A, Mariotto A, Szabo A, Falcon S, Wegelin J, et al. Quantifying the role of PSA screening in the US prostate cancer mortality decline. Cancer Causes Control 2008;19:175–81. 14. Siegel RL, Miller KD, Fuchs HE, Jemal A. Cancer Statistics, 2021. CA Cancer J Clin 2021;71:7–33. 15. Rebbeck TR, Devesa SS, Chang B-L, Bunker CH, Cheng I, Cooney K, et al. Global patterns of prostate cancer incidence, aggressiveness, and mortality in men of african descent. Prostate Cancer 2013;2013:560857. 16. National Cancer Institute. Surveillance, Epidemiology, and End Results (SEER) Program. Cancer Stat Facts: Prostate cancer. Accessed via: https://seer.cancer.gov/statfacts/html/prost.html (accessed July 5, 2021). 17. Siegel DA, O’Neil ME, Richards TB, Dowling NF, Weir HK. Prostate Cancer Incidence and Survival, by Stage and Race/Ethnicity - United States, 2001-2017. MMWR Morb Mortal Wkly Rep 2020;69:1473–80. 18. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2016. CA Cancer J Clin 2016;66:7–30. 19. Shoag JE, Nyame YA, Gulati R, Etzioni R, Hu JC. Reconsidering the Trade-offs of Prostate Cancer Screening. N Engl J Med 2020;382:2465–8. 20. Sheng IY, Wei W, Chen Y-W, Gilligan TD, Barata PC, Ornstein MC, et al. Implications of the United States Preventive Services Task Force Recommendations on Prostate Cancer Stage Migration. Clin Genitourin Cancer 2021;19:e12–6. 21. Hugosson J, Roobol MJ, Månsson M, Tammela TLJ, Zappa M, Nelen V, et al. A 16-yr Follow-up of the European Randomized study of Screening for Prostate Cancer. Eur Urol 2019;76:43–51. 22. Fenton JJ, Weyrich MS, Durbin S, Liu Y, Bang H, Melnikow J. Prostate-Specific Antigen-Based Screening for Prostate Cancer: Evidence Report and Systematic Review for the US Preventive Services Task Force. JAMA 2018;319:1914–31. 23. Grossman DC, Curry SJ, Owens DK, Bibbins-Domingo K, Caughey AB, Davidson KW, et al. Screening for Prostate Cancer: US Preventive Services Task Force Recommendation Statement. JAMA 2018;319:1901– 13.
www.ridderprint.nlRkJQdWJsaXNoZXIy MTk4NDMw