Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Dosimetric benefit of MR-guided online adaptive radiotherapy in different tumor entities: liver, lung, abdominal lymph nodes, pancreas and prostate

Dosimetric benefit of MR-guided online adaptive radiotherapy in different tumor entities: liver,... Background: Hybrid magnetic resonance (MR)-Linac systems have recently been introduced into clinical practice. The systems allow online adaption of the treatment plan with the aim of compensating for interfractional anatomical changes. The aim of this study was to evaluate the dose volume histogram (DVH)-based dosimetric benefits of online adaptive MR-guided radiotherapy (oMRgRT ) across different tumor entities and to investigate which subgroup of plans improved the most from adaption. Methods: Fifty patients treated with oMRgRT for five different tumor entities (liver, lung, multiple abdominal lymph nodes, pancreas, and prostate) were included in this retrospective analysis. Various target volume (gross tumor vol- ume GTV, clinical target volume CTV, and planning target volume PTV ) and organs at risk (OAR) related DVH param- eters were compared between the dose distributions before and after plan adaption. Results: All subgroups clearly benefited from online plan adaption in terms of improved PTV coverage. For the liver, lung and abdominal lymph nodes cases, a consistent improvement in GTV coverage was found, while many fractions of the prostate subgroup showed acceptable CTV coverage even before plan adaption. The largest median improve- ments in GTV near-minimum dose (D ) were found for the liver (6.3%, p < 0.001), lung (3.9%, p < 0.001), and abdomi- 98% nal lymph nodes (6.8%, p < 0.001) subgroups. Regarding OAR sparing, the largest median OAR dose reduction during plan adaption was found for the pancreas subgroup (-87.0%). However, in the pancreas subgroup an optimal GTV coverage was not always achieved because sparing of OARs was prioritized. Conclusion: With online plan adaptation, it was possible to achieve significant improvements in target volume cov- erage and OAR sparing for various tumor entities and account for interfractional anatomical changes. Keywords: Online MRI guided radiotherapy, Plan adaption, MRgOART , Online adaptive RT, MR-guided RT Background Various inter- and intra-fractional anatomical changes in patient anatomy pose a major challenge for the safe *Correspondence: Lukas.Nierer@med.uni-muenchen.de and successful treatment application in modern abla- Department of Radiation Oncology, University Hospital, LMU Munich, tive image-guided radiotherapy (RT). Typical examples Marchioninistr. 15, 81377 Munich, Germany of such changes in patient geometry are different organ Full list of author information is available at the end of the article © The Author(s) 2022. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/. The Creative Commons Public Domain Dedication waiver (http:// creat iveco mmons. org/ publi cdoma in/ zero/1. 0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Nierer et al. Radiation Oncology (2022) 17:53 Page 2 of 14 fillings of bladder, stomach or rectum, breathing-related using the MRIdian system (ViewRay Inc., Oakwood Vil- motion, peristalsis, cardiac motion, tumor response lage, OH, USA) with a step-and-shoot intensity modu- (shrinkage), or organ and patient weight changes [1]. lated radiotherapy (IMRT) technique in the thoracic or Numerous motion patterns of organs at risk (OAR), tar- abdominal region according to the institutional oMRgRT get volumes or quantification of motion amplitudes can clinical protocol. The commercially available system con - be found in the literature [2–6]. sists of a hybrid MR-Linac and an integrated treatment These types of anatomical changes occur on various planning system (TPS). Dose prescription referred to time-scales, ranging from seconds to weeks, and can either the 65%, 80% or 95% isodose. Two patients were potentially be accounted for via tumor-tracking or gat- treated for two lesions simultaneously (patients 35 and ing techniques and plan adaption strategies [1]. Although 49) and one patient was treated for three lesions simul- the technical implementation of such advanced RT tech- taneously (patient 42). In those three patients, the mul- niques can be challenging, several adaptive RT (ART) tiple lesions were treated with one single treatment plan. approaches have found their way into clinical routine [1– u Th s, all OAR constraints of the corresponding frac - 7]. The feasibility and clinical benefit of offline ART using tions were counted as if these patients only had a single computed tomography (CT) or magnetic resonance lesion. Parameters for target volumes (both gross tumor imaging (MRI) has been demonstrated [8–12]. Strategies volume (GTV) and planning target volume (PTV)) were for online ART based on in-room cone-beam computed evaluated separately for each lesion. A total of 265 online tomography (CBCT) have also been proposed [13–21]. adapted fractions were analyzed. Only fractions which A newly developed commercial CBCT-based system were adapted were considered in the analysis. even allows for fast online ART in the clinical routine The mean percentage of adapted fractions per patient (Ethos : Varian Medical Systems, Palo Alto, CA, USA) was 86% (range 15% to 100%) and 79% of all fractions [22]. In this context, combined hybrid MR-Linac systems were adapted in total. Table 2 (Results) shows the portion (e.g. MRIdian : ViewRay Inc., Oakwood Village, USA; or of adapted plans and characteristics of the online adapted Unity : Elekta AB, Stockholm, Sweden) have the advan- plans for each subgroup. tage of superior soft-tissue contrast and dose-free intra- fractional imaging (where available, also with real-time oMRgRT workflow tumor tracking and gated RT) in addition to a full online The oMRgRT workflow was similar to that described by ART workflow. After the first patient was treated on such Bohoudi et al. [2]. For initial treatment planning, a plan- a hybrid device in 2017 [23], combined MR-guided RT ning MR and CT were acquired using the same patient systems for online ART (“MR-Linacs”) are now com- setup. The planning CT was acquired immediately after mercially available and becoming increasingly popular the MR. The CT was registered using deformable image [24]. The feasibility of online magnetic resonance guided registration (DIR) to the MR to obtain electron density ART (oMRgRT) has already been demonstrated [25–31] values for dose calculation. GTV, clinical target volume and initial studies have reported its dosimetric and clini- (CTV) and OAR delineation was performed on the MR. cal benefits in a wide range of indications [32–46]. The In the TPS, Boolean operations of regions of interest aim of the present study was to evaluate the potential (ROIs; e.g. subtraction or margin expansion of struc- of treatment adaption in oMRgRT in terms of improved tures) can be performed and stored as so-called “rules”. target volume coverage and OAR sparing across five dif - Such rules were defined for the automatic generation ferent tumor entities, which are frequently treated with of the PTV (expansion of the GTV) and derived struc- oMRgRT. tures were defined at the treating physician’s discretion to reduce the contouring effort during online adap - Methods tion. After dose prescription and contour delineation, a Patients baseline treatment plan was generated analogous to the Overall, 50 patients treated between 01/2020 and workflow in conventional RT. All plans were generated 11/2021 were included in this retrospective analysis. All as step-and-shoot IMRT via inverse planning (6 MV patients were treated for one out of five different tumor flattening filter free beam; 1.5 mm calculation grid size entities, which are typical indications for oMRgRT (liver, with isotropic voxels; 1.0% Monte Carlo dose calcula- lung, abdominal lymph nodes, pancreas or prostate). tion uncertainty) and the maximum number of multi Ten patients per entity were randomly selected from all leaf collimator (MLC) segments was limited, depend- patients of the respective subgroups, who successfully ing on the complexity of the plan. This segment num - completed their RT treatment in the given period. Table 1 ber limit of the baseline plan was subsequently used provides an overview of patient and treatment character- for online plan adaption. For each treatment fraction, istics. All patients received hypofractionated oMRgRT a balanced steady-state free precession (bSSFP) pulse N ierer et al. Radiation Oncology (2022) 17:53 Page 3 of 14 Table 1 Patient characteristics Patient Nr Nr. of lesions Nr. of adapted Total nr. of fx Fraction dose Total dose (Gy) Prescription (%) Group fx (Gy) 1 1 3 20 3 60 95 Prostate 2 1 14 20 3 60 95 Prostate 3 1 5 5 7.25 36.25 95 Prostate 4 1 7 20 3 60 95 Prostate 5 1 20 20 3 60 95 Prostate 6 1 19 20 3 60 95 Prostate 7 1 20 20 3 60 95 Prostate 8 1 16 20 3 60 95 Prostate 9 1 9 20 3 60 95 Prostate 10 1 5 5 7 35 95 Prostate 11 1 2 5 8 40 95 Pancreas 12 1 4 5 8 40 95 Pancreas 13 1 5 5 8 40 80 Pancreas 14 1 5 5 8 40 80 Pancreas 15 1 5 5 8 40 80 Pancreas 16 1 5 5 8 40 80 Pancreas 17 1 5 5 8 40 80 Pancreas 18 1 5 5 8 40 80 Pancreas 19 1 5 5 6.6 33 80 Pancreas 20 1 5 5 8 40 80 Pancreas 21 1 3 3 12.5 37.5 65 Liver 22 1 3 3 12.5 37.5 65 Liver 23 1 3 3 12.5 37.5 65 Liver 24 1 3 3 12.5 37.5 65 Liver 25 1 3 3 15 45 65 Liver 26 1 3 3 12.5 37.5 65 Liver 27 1 2 3 15 45 65 Liver 28 1 3 3 12.5 37.5 65 Liver 29 1 3 3 15 45 65 Liver 30 1 2 3 12.5 37.5 65 Liver 31 1 5 5 7 35 95 Lymph nodes 32 1 4 5 5 25 80 Lymph nodes 33 1 2 5 8 40 95 Lymph nodes 34 1 4 5 6.4 32 80 Lymph nodes 35 2 5 6 6 36 80 Lymph nodes 36 1 9 10 4 40 95 Lymph nodes 37 1 5 5 7 35 95 Lymph nodes 38 1 3 5 7 35 95 Lymph nodes 39 1 4 5 7 35 80 Lymph nodes 40 1 5 5 6 30 80 Lymph nodes 41 1 3 5 10 50.0 95 Lung 42 3 3 3 13.5 40.5 65 Lung 43 1 2 3 13.5 40.5 65 Lung 44 1 3 3 13.5 40.5 65 Lung 45 1 3 3 13.5 40.5 65 Lung 46 1 2 3 13.5 40.5 65 Lung 47 1 3 3 13.5 40.5 65 Lung 48 1 3 3 13.5 40.5 65 Lung 49 2 3 3 13.5 40.5 65 Lung 50 1 2 3 13.5 40.5 65 Lung Nierer et al. Radiation Oncology (2022) 17:53 Page 4 of 14 Table 1 (continued) The dose prescription refers to the corresponding isodose sequence 3D setup MRI scan was acquired for transla- the target volume showed a breathing-related motion tional patient setup correction (couch shift). For more were treated using a breath-hold technique (mostly deep information about the MR pulse sequences and the inspiration breath hold). However, patients treated with technical design of the MRIdian system, refer to Klüter a free breathing approach (in cases of very limited tumor et  al. [47]. The MRI of the baseline plan was then reg - motion, e.g. most prostate cases) were also treated using istered via DIR to the volumetric setup MRI of the day the automated gating function in order to ensure that the and all target structures, OARs and the electron den- target was positioned within tolerance boundaries during sity of the planning CT were propagated onto the setup treatment application. MRI. All contours were edited (if necessary), a track- All baseline plans were validated dosimetrically with ing contour was defined and the baseline plan was cal - an ionization chamber and/or diode detector array (Arc- culated on the MRI (more precisely the synthetic CT) Check-MR; Sun Nuclear Corporation, Melbourne, FL, of the day, which results in the so-called predicted USA) prior to the first fraction. dose (baseline plan calculated on the anatomy of the day with updated structures). This dose distribution Extraction of DVH and plan parameters shows the dose of a single non-adapted fraction and is Several dose volume histogram (DVH) parameters the basis to decide whether to adapt a plan or not. In were extracted from the TPS for the predicted (non- case of a subsequent plan adaption, the initial predicted adapted) and re-optimized scenarios for all fractions: dose corresponds to the dose distribution prior to plan the dose to 98%, 95%, 50% and 2% of the volume of the adaption. For most OARs, a time-saving and practi- PTV (PTV D = near minimum dose, PTV D , PTV 98% 95% cal partial re-contouring approach was used, in which D = median dose, PTV D = near maximum dose) 50% 2% the OARs were edited only in the close surrounding of and the mean PTV dose (PTV D ). All parameters mean the PTV (PTV + 3.0  cm), where the highest dose gra- were also reported for the GTV. In prostate cases, the dients occur. This approach was described by Bohoudi CTV was reported, as no GTV was defined. Further - et al. [2], while Ahunbay et al. described a similar basic more, the volume of the PTV (V ) was extracted. Out PTV concept [48]. When the decision was made to adapt of usually multiple, patient-specific OAR constraints pre - the plan of the current fraction, the plan parameters scribed in the TPS, three OAR constraints were chosen and dose constraints of the baseline plan were used for OARs, which were closest to the PTV and as a result as a starting point for dose optimization. A treatment received the highest maximum dose values. Only OAR plan was adapted if either the target coverage or OAR constraints were chosen, which were related to structures constraints of the predicted plan were not fulfilled, or that were updated (in the PTV + 3.0  cm region; see last a combination of both. Therefore, the planning goal sub-section). The individual OAR dose constraints (near was always to achieve optimal target coverage while maximum dose or dose-to-volume constraints) depend respecting all OAR constraints. Online plan adaption on the dose prescription and the individual case and were was performed either as re-optimization with the same defined by a senior physician based on the applicable objectives of the baseline plan or as full re-optimization guidelines. with modified objectives and/or plan parameters. The Technical plan parameters like the net beam-on time dose distribution of the online adapted plan, calculated (BOT), the number of segments, the number of Monitor on the current synthetic CT (based on the MR of the Units (MU) and the number of beams were read out from day) with updated structures is referred to as re-opti- treatment plan documentation files (Table 2; Results). mized dose. All dose calculation settings for the re- optimized dose were the same as for the predicted dose Comparison of DVH parameters, statistical analysis (as defined in the baseline plan). After plan adaption, and definition of dosimetric endpoints the dose was verified for QA using a secondary Monte First the OAR related DVH parameters for the predicted Carlo code before treatment. and re-optimized doses of each adapted fraction were For tumor tracking via a 2D bSSFP cine MRI sequence, compared. the tracking structure was propagated onto a 2D cine Second, the same comparison was made for target MRI slice and a gating ROI was created by expansion volume (PTV, CTV or GTV) related DVH parameters. of the tracking structure. These structures were subse - These parameters were systematically compared pairwise quently used for online beam gating. All patients in which N ierer et al. Radiation Oncology (2022) 17:53 Page 5 of 14 to quantify the DVH-based dosimetric benefit of online was performed via paired Wilcoxon signed-rank test. A plan adaption for each subgroup separately. Primary significance level of α = 5% was used. dosimetric endpoints were chosen as follows: increase in GTV (all cases expect prostate) or CTV (prostate cases) D98%, increase in GTV or CTV D95%, increase in GTV Results mean dose, reduction in OAR exposure. The GTV (or Characteristics of the online adapted plans are shown in CTV) near-minimum parameters (D98% and D95%) Table 2. were chosen because underdose (e.g. dose drop at the Figure  1 shows an example DVH of a single adaptive surface or cold spots) within the GTV/CTV are likely to fraction of an abdominal lymph node case and illustrates affect the local control (LC). The GTV (or CTV in pros - the large potential of online ART to increase target cov- tate cases) mean dose was chosen because there is evi- erage and OAR sparing with the oMRgRT technique. For dence for a predictive value of this parameter in terms example, the GTV D improved by more than 10.0% 95% of improved LC in liver [49] and lung tumors [50]. Box- and the duodenum D was reduced by about one third. 50% plots were generated to visualize the data. Therefore, due Figure  2 indicates for each subgroup the change of to the different dose prescriptions, data was normalized. OAR exposure when plans were adapted. The three most The D and D dose values were normalized to the frequently considered OARs per region were bowel, duo- 98% 95% prescribed dose (PD) for non-homogenous stereotac- denum, stomach (for liver), lung, heart, esophagus (for tic prescriptions (65% or 80% prescription isodose PI) lung), bowel, duodenum, spinal cord (for lymph nodes), or 0.95 × PD for homogenous prescriptions (95% PI). duodenum, stomach, bowel (for pancreas), and rectum, The D and D dose values were normalized to the bladder, femur (for prostate). The largest median dose 50% mean PD. The D was normalized to PD/PI (thus PD/0.8 for reduction of OARs adjacent or close to the PTV was 2% a prescription to the 80% isodose) for stereotactic pre- found for the pancreas subgroup (-87.0%). This dose scriptions or to PD for homogenous prescriptions. Since reduction was significant, as well as the smaller dose all normalization factors depend solely on the plan- reduction achieved in the lymph-nodes subgroup. Small ning aim for the PTV, these factors may differ between but significant median increased OAR doses were found patients but the same factors are applied to all fractions for liver and lung and no statistically significant differ - of a single patient. The dose normalization was done to ence was found for prostate. All p-values and median achieve comparable data for the generation of analysis percent changes are shown in the first row of Table  3. For figures. Statistical analysis or the calculation of percent a more detailed insight in the effect on individual OARs, changes of dose values is not affected by this normaliza - Table  4 provides mean and median percent changes of tion. Statistical analysis of the DVH-based parameters Table 2 Characteristics of the online adapted plans Liver Lung Lymph nodes Pancreas Prostate Adapted fractions (%) 93.3 84.4 82.1 92.0 69.4 Mean BOT (min) 7.1 9.1 4.5 6.2 1.9 Min. BOT (min) 4.9 4.5 2.5 3.2 1.1 Max. BOT (min) 11.3 17.7 9.4 9.3 4.4 Mean number of beams 12 10 16 15 14 Min. number of beams 9 8 9 13 9 Max. number of beams 16 15 19 17 21 Mean number of segments 33 26 72 70 54 Min. number of segments 13 9 12 46 33 Max. number of segments 54 40 100 95 129 Mean MU 4269.2 5439.9 2652.2 3688.4 1139.4 Min. MU 2950.3 2675.2 444.3 1906.7 649.0 Max. MU 6776.0 10,612.0 5607.9 5547.1 2624.5 Mean V (cm ) 38.4 15.2 65.5 251.1 114.7 PTV Min. V (cm ) 7.3 3.9 1.9 59.6 67.9 PTV Max. V (cm ) 109.9 32.0 291.3 455.8 192.6 PTV BOT = beam-on time, MU = number of monitor units, V = volume of the PTV PTV Nierer et al. Radiation Oncology (2022) 17:53 Page 6 of 14 PTV re. PTV pr. Stomach re. Stomach pr. Duodenum re. Duodenum pr. GTV re. GTV pr. Bowel re. Bowel pr. CBD* re. CBD* pr. 01234567 8 Dose [Gy] Fig. 1 Exemplary cumulative DVH of fraction 2 of patient 32 under the adaptive workflow (abdominal lymph node treatment with a dose prescription of 5 × 5.0 Gy to the 80% isodose); re-optimized (re., solid lines) and predicted (pr., dashed lines). Target coverage increased after plan adaption while OAR exposure could be reduced. *Common bile duct Maximum Q3 Median Q1 -50 Minimum -100 Outlier (1.5 × IQR) -150 -200 Fig. 2 Boxplots of percent changes of OAR DVH parameters re-optimized versus predicted for all subgroups. The negative (lower) half corresponds to fractions where OAR exposure was reduced when adapted and the positive (upper) half corresponds to increased OAR exposure when adapted. Q1: first quartile, Q3: third quartile, IQR: interquartile range OAR dose parameters of three most frequently consid- For the target volume DVH parameters, the largest ered OARs per region. changes were found for D and D for the PTV and 98% 95% GTV/CTV when comparing the re-optimized with the Liver Lung Lymph nodes Pancreas Prostate Change of OAR dose constraints [%] Volume [%] N ierer et al. Radiation Oncology (2022) 17:53 Page 7 of 14 Table 3 p-values and median percent changes [p value/median change (%)] of DVH OAR and target volume parameters when comparing the re-optimized versus the predicted dose distributions Liver Lung Lymph nodes Pancreas Prostate OAR 0.000/6.9 0.004/5.9 0.001/−4.5 0.000/−87.0 0.135/0.8 PTV D 0.000/3.8 0.006/1.0 0.000/3.0 0.041/0.8 0.000/0.9 mean PTV D 0.712/0.1 0.131/−0.7 0.757/0.0 0.002/−0.9 0.093/−0.2 2% PTV D 0.001/2.6 0.065/0.0 0.003/0.9 0.196/−0.5 0.021/0.3 50% PTV D 0.000/17.3 0.000/6.0 0.000/9.4 0.000/5.7 0.000/2.8 95% PTV D 0.000/25.5 0.000/7.8 0.000/15.6 0.000/11.0 0.000/5.8 98% GTV or CTV D * 0.004/1.0 0.012/0.9 0.000/1.4 0.454/−1.3 0.007/0.4 mean GTV or CTV D 0.872/0.0 0.042/−0.4 0.949/−0.2 0.176/−0.8 0.176/−0.1 2% GTV or CTV D 0.040/0.6 0.127/0.6 0.020/1.0 0.164/−1.4 0.185/0.2 50% GTV or CTV D 0.001/4.6 0.000/3.1 0.000/4.9 0.946/−0.3 0.000/0.7 95% GTV or CTV D 0.000/6.3 0.000/3.9 0.000/6.8 0.589/0.5 0.000/1.3 98% Significant differences are highlighted bold *GTV for all subgroups except prostate and CTV for prostate Table 4 Mean and median percent changes when comparing the re-optimized versus the predicted dose distributions [mean change (%)/median change (%)] of the most frequently used OAR dose parameter for each of the three most frequently considered OARs per region Liver D (bowel) D (duodenum) D (stomach) max max max 18.2/8.5 35.6/50.8 11.2/14.0 Lung V15 (lung left or right) D (heart) D (esophagus) Gy max max 9.0/5.9 18.3/0.0 3.6/8.2 Lymph nodes V20 (bowel) V18 (duodenum) D (spinal canal) Gy Gy max − 10.8/− 27.8 − 43.9/− 48.2 − 2.9/− 4.4 Pancreas V33 (duodenum) V33 (stomach) V33 (bowel) Gy Gy Gy − 94.8/− 97.1 − 84.0/− 99.1 − 83.4/− 98.6 Prostate V40 (rectum) V40 (bladder) D (femur left or right) Gy Gy max 6.8/0.3 9.3/7.4 − 0.7/− 1.2 predicted doses (Fig.  3). All PTV D and D median were found for prostate, compared to the liver, lung and 98% 95% dose values increased significantly across all subgroups. lymph node subgroups. All p-values and median percent More importantly, the GTV/CTV D and D values increases are shown in Table 3. 98% 95% increased significantly, except for pancreas, where no sig - Additional target dose volume parameters are shown nificant difference was found (see Table  3). This means in Fig.  4. The near maximum doses D did not show 2% that the target volume dose coverage increased signifi - significant changes in most cases. Significant but small cantly in most cases when adapting. The largest median median near maximum dose reductions inside the PTV increases of PTV D were found in liver, lung and or GTV were found for PTV D (pancreas) and GTV 95% 2% lymph nodes. The largest median increases in GTV D D (lung). When looking at the boxplots (Fig.  4), it can 95% 2% were also found in the same subgroups. Although a rela- be seen that a few cases of the lung, lymph node and pan- tively large significant increase was found for PTV D creas subgroups showed predicted near maximum doses, 95% in pancreatic cases, no significant increase was found for which exceeded 10% of the ideally achieved maximum the pancreas GTV D and GTV D . Despite being dose (mostly outliers). A slight reduction of these high 98% 95% statistically significant, smaller median increases in PTV near maximum doses was achieved when adapting. After D , PTV D , GTV/CTV D and GTV/CTV D plan adaption, all non-outlier near-maximum values 98% 95% 98% 95% Nierer et al. Radiation Oncology (2022) 17:53 Page 8 of 14 Liver LungLymph nodes PancreasProstate Liver Lung Lymph nodes PancreasProstate pr.re. pr.re. pr. re.pr. re.pr. re. pr. re.pr. re.pr. re.pr. re.pr. re. 150 150 140 140 130 130 120 120 110 110 100 100 90 90 80 80 70 70 60 60 50 50 40 40 Liver Lung Lymph nodesPancreasProstate Liver Lung Lymph nodes PancreasProstate pr.re. pr.re. pr. re. pr. re. pr.re. pr.re. pr.re. pr.re. pr. re.pr. re. Fig. 3 Boxplots for all subgroups of DVH target volume dose values D and D for PTV and GTV/CTV (*GTV for all cases except prostate; CTV in 98% 95% case of prostate) for the predicted (pr.) and re-optimized (re.) dose distributions. Dose normalized to the ideally achieved PTV encompassing dose (see sub-section “Comparison of DVH parameters and statistical analysis”) PTV D and GTV D exceeded the ideally achieved PTV coverage, similar to that of the lung cases. With 2% 2% maximum dose by less than 10%. PTV and GTV/ CTV 93.3% of all fractions adapted, liver showed the highest median (D ) and mean (D ) doses show an inverse portion of adapted plans (similar to pancreas with 92.0%). 50% mean behaviour. For these values, slight median increases were This means that the initial target volume coverage using found in most cases, except for the pancreas subgroup, the base plan was not ideal in almost every fraction. After where no significant changes were found for PTV D , plan adaption, PTV D was > 97.0% of the PD in all frac- 50% 98% GTV D and GTV D (Table 3). tions. Even though the initial median PTV coverage was 50% mean worse compared to the lung cases, the initial GTV cov- Discussion erage was similar, which means that the PTV, designed Liver for liver cases in breath hold technique, worked very well. For the liver subgroup, the initial predicted median PTV When comparing the normalized percent values of Fig. 3 coverage (here PTV D and PTV D ) before plan between these two (or any other) subgroups, especially 98% 95% adaption was worse compared to lung and prostate but for the GTV, one has to bear in mind that all liver cases was comparable to lymph nodes and pancreas. One rea- had a stereotactic prescription to the 65% isodose, but son might be that less complex shaped (sphere-like) tar- only 8/10 lung cases had the same prescription. No sig- gets, as small lung lesions (lung: mean V = 15.2 cm vs. nificant change in the PTV or GTV near-maximum dose PTV liver: mean V = 38.4 cm , see Table 2) or prostate tar- was found, just like for most other subgroups. Regarding PTV gets, are easier to cover with default baseline plans. After OAR sparing, Fig.  2 shows a more or less symmetrical plan adaption, a largely improved PTV coverage was distribution around zero for liver, but with a significantly found, which resulted in a close-to-ideal post-adaption increased median OAR dose. This is because OAR dose PTV D [%] PTV D [%] 95% 98% GTV or CTV* D [%] GTV or CTV* D [%] 98% 95% N ierer et al. Radiation Oncology (2022) 17:53 Page 9 of 14 160 160 150 150 140 140 130 130 120 120 110 110 100 100 90 90 80 80 Liver Lung Lymph nodesPancreas Prostate LiverLungLymph nodes Pancreas Prostate pr.re. pr.re. pr. re. pr. re.pr. re. pr.re. pr. re. pr. re. pr. re. pr.re. 130 130 125 125 120 120 115 115 110 110 105 105 100 100 95 95 LiverLung Lymph nodes Pancreas Prostate LiverLungLymph nodes Pancreas Prostate pr.re. pr.re. pr. re. pr. re.pr. re. pr.re. pr.re. pr.re. pr.re. pr.re. Liver LungLymph nodesPancreas Prostate LiverLungLymph nodes Pancreas Prostate pr.re. pr.re. pr.re. pr.re. pr.re. pr.re. pr.re. pr.re. pr.re. pr.re. Fig. 4 Boxplots for all subgroups of DVH target volume dose values D, D and D for PTV and GTV or CTV (*GTV for all cases except prostate; 50% 2% mean CTV in case of prostate) for the predicted (pr.) and re-optimized (re.) dose distributions. D and D normalized to the prescribed dose. D 50% mean 2% normalized to the ideally achieved maximum dose (see sub-section “Comparison of DVH parameters and statistical analysis”) limits were, on average, not fully reached prior to adap- OARs were found for liver. During the optimization pro- tion. This tendency can also be seen when looking at the cess, the OAR exposure was fully exploited and brought exposure changes of the three most frequently consid- closer to the dose limits, in order to achieve a very good ered OARs (bowel, duodenum, stomach, see Table  4), target coverage without violation of OAR constraint. where mean and median increases in dose for all these With a mean number of 33 segments, adapted liver plans PTV D [%] PTV D [%] 50% PTV D [%] 2% Mean GTV or CTV* D [%] GTV or CTV* D [%] GTV or CTV* D [%] Mean 50% 2% Nierer et al. Radiation Oncology (2022) 17:53 Page 10 of 14 were simpler compared to those of lymph node, pancreas (Table 4). Without online plan adaption, OAR dose limits and prostate cases. would have been frequently violated. Lung Prostate For lung cases, the initial PTV coverage of the predicted For most prostate cases, the initial PTV and especially base plan was better compared to liver, but could still be the CTV coverage were acceptable. In only a few frac- significantly improved and resulted in a near-optimal tions very insufficient initial target volume coverage was PTV coverage after plan adaption. The initial GTV cov - found (Fig. 3). For these fractions, large improvements in erage was similar to that of liver, but could still be sig- PTV and even CTV coverage were made when re-opti- nificantly improved. With 84.4% of all fractions adapted, mizing the plans. On average, no significant change of lung showed similar adaption rates as lymph node cases. OAR exposure was achieved with online plan adaption. A Regarding OAR sparing and change of PTV near-maxi- mean of 54 segments were needed to achieve acceptable mum doses, similar findings were made as for the liver plans, which indicates plans of medium complexity. Only subgroup. The mean PTV volume of lung cases was the 69.4% of fractions were adapted. smallest of all subgroups (15.2 cm ) and so was the mean number of segments (26), indicating easy-to-adapt, sim- Comparison between tumor entities ple treatment plans. All subgroups clearly benefited from online plan adap - tion in terms of improved PTV coverage. The improved Abdominal lymph nodes target dose varied between the different tumor entities. For the abdominal lymph nodes subgroup, initial PTV To estimate the absolute dose changes of the target vol- coverage was not ideal but could be efficiently improved umes achieved by online plan adaption, it is possible to with online plan adaption, which was performed in multiply the total prescribed dose of any dose prescrip- 82.1% of fractions. As for to the liver and lung cases, a tion (Table  1) with the corresponding median percent relatively large median increase (6.8%) in GTV D was change of any DVH parameter of interest (Table  3) and 98% found without an increase in the near-maximum dose. with the percentage of adapted fractions of the cor- The median OAR dose could be reduced significantly responding patient subgroup (Table  2, line 1). Obvi- by − 4.5% (Table 3). A larger reduction of OAR dose was ously, this formula can only provide a rough estimate of only observed in the pancreas subgroup. The lymph node absolute dose changes and does not consider individual treatment plans had an average number of 72 segments, patients. Based on the underlying idea of the PTV target which is similar to that of pancreas (70). This shows that volume concept, one would assume that the GTV/CTV lymph node treatment plans were usually highly modu- coverage physically achieved during dose application is lated, similar to the pancreas plans. of higher prognostic value than the PTV coverage, since the sole purpose of the PTV margins is to guarantee the Pancreas GTV/CTV coverage with some degree of confidence. Plan adaptation resulted in a significant improvement Under this assumption it is possible to identify subgroups in PTV near minimum dose. Regarding the GTV cover- that might benefit more from online plan adaption than age, no significant improvement could be made on aver - others. In the present study, the three subgroups liver, age, when performing online plan adaption. Although the lung, and abdominal lymph nodes had the greatest ben- average re-optimized (and even predicted) GTV coverage efit from online plan adaption in terms of improved was acceptable, some cases were observed where suf- GTV coverage. After plan adaption, all three subgroups ficient GTV coverage could not be achieved, even after showed excellent GTV coverage (D and D ). In 98% 95% plan adaption, because sparing of OARs was prioritized. addition, a large portion of fractions (> 82%) required This can be seen in Fig.  3, when looking at the relatively re-optimization in all three subgroups, indicating that large interval between the first and third quartiles com - most of these fractions can be significantly improved by pared to those of the abdominal lymph nodes and pros- online plan adaption. Since small values of GTV D 98% tate cases. The large portion of adapted fractions (92.0%) and D indicate insufficient GTV coverage, these indi - 95% indicates the need for plan adaptation to reduce OAR ces can be considered the most predictive (of all DVH doses to meet clinically acceptable OAR dose levels. parameters examined) in terms of improved tumor con- OAR doses could be reduced in more than 75.0% of all trol probability (TCP) and possibly LC when comparing OAR constraints of all adapted fractions of the pancreas non-adapted and online adapted fractions. However, it cases and the median OAR dose reduction was − 87.0% remains unclear if the significant increases in GTV near- (Fig.  2). The V33 of duodenum, stomach and bowel minimum and mean doses will translate into a detectable Gy was reduced on average by more than 80% respectively improved LC. N ierer et al. Radiation Oncology (2022) 17:53 Page 11 of 14 In addition to the significantly improved GTV cover - several fractions among the pancreas cases were found, age in the liver, lung and abdominal lymph nodes sub- where GTV dose de-escalation was necessary due to groups when using plan adaption, these subgroups might OAR proximity of the tumor. Regnery et al. [52] prospec- also largely benefit from the breath-hold and automated tively compared predicted versus adapted dose distribu- beam gating capabilities of the oMRgRT system, since tions in 154 online adapted fractions in 21 lung tumor no internal target volume concept is needed. An internal patients. The higher adaption rate of 93.3% compared target volume concept would increase the total irradiated to 84.4% in the current study can be explained with the volume [51], especially for these cases, where breathing- large number of ultracentral lung tumors in the cohort related motion of target volumes can be frequently seen. of Regnery et  al., where OAR violations are more likely In this study, the influence of beam-gating was not inves - to occur due to adjacency OARs. Regnery et  al. found tigated and no motion range assessment was made for a large increase in the minimum biologically effective the target volumes. That is beyond the scope of this study. dose (BED) of the PTV and a moderate increase in the It was found for pancreas, that the plan adaption capa- minimum BED of the GTV. We observed the same ten- bilities were largely used to reduce OAR doses to an dency when considering PTV and GTV D or D . In 98% 95% acceptable level. Although the re-optimized GTV cov- the same study only small increases in mean BED inside erage was acceptable in most cases, the improvement in the PTV and GTV were found, which is also in accord- GTV coverage via online plan adaption was not as large ance with our findings of PTV and GTV D . El Bared mean as in the other subgroups, as OAR sparing was prior- et al. [35] evaluated the dosimetric benefits in 10 patients itized. In contrast, in the prostate subgroup, only 69.4% treated for unresectable pancreatic cancer on a cobalt-60, of all fractions were considered for re-optimization. For 0.35  T MRI system when performing online plan adap- most of these fractions, the initial CTV coverage was tion and reported outcome. Although comparability to quite good but could still be slightly improved when our study is limited due to a fundamental discrepancy in adapted. Ultimately, few fractions were found, where the technical design and beam quality (tri-source cobalt-60 initial PTV and CTV coverage was unacceptable. For vs. 6 MV flattening filter free Linac), El Bared also found these fractions, online re-optimization resulted in an improved PTV coverage when performing online plan excellent CTV coverage. In summary, for prostate cases, adaption. However, the influence on GTV coverage was the benefit of online re-optimization was found to be not evaluated. Placidi et  al. [37] found a similar trend of not as systematic as for the liver, lung, abdominal lymph improved PTV coverage in 8 pancreatic cancer patients nodes and pancreas cases. In each subgroup, at least also treated on the cobalt-60 system and similarly one of the primary dosimetric endpoints defined under reported increased CTV dose after plan adaption. The “Methods” (significant increase in GTV/CTV near-min - influence on GTV coverage was not evaluated. Mayinger imum and mean dose, and significantly reduced OAR et al. [53] analyzed online adapted treatment plans of 15 exposure) was achieved (Table  3). Except for pancreas, patients with liver metastases and found improved PTV all other subgroups met all primary endpoints related to coverage in cases where the target volume was in close improved GTV/CTV dose. proximity to OARs. The influence on GTV coverage was To the best of our knowledge, up to now, no attempt not evaluated in detail. In the present study, we did not has been made to quantify the influence of MR-guided stratify patients for adjacency of OARs, but when look- online plan adaption on DVH-related parameters and ing at the PTV mean and near-minimum doses, we also systemically compare the results between multiple sub- found a significant increase after plan adaption in liver groups with different tumor entities typically treated on cases. Padgett et al. [43] artificially created adapted plans integrated MR-Linac systems. Our intent was to pro- for 10 patients with liver cancers on the cobalt-60 system vide information for a more informed decision making and compared the results to the non-adapted plans and when assigning patients to the (still very limited access) also found improved PTV and GTV coverage as well as MR-Linac. Henke et  al. [40] analyzed 81 online adapted a reduced number of OAR violations (duodenum, bowel fractions (20 patients) in a patient cohort of mixed oli- and stomach) after plan adaption. We observed the same gometastatic or unresectable abdominal malignancies trend regarding target volumes and reported small mean (hepatic lesions, adrenal metastasis, pancreatic adenocar- and median dose increases for OARs like duodenum, cinoma and lymph node metastases). The overall adap - bowel and stomach. This is no contradiction to the find - tion rate (83.5%) was comparable to our study, although ings of Padgett et  al. since OAR sparing was prioritized less liver fractions (66.0%) were adapted compared to over target volume coverage in our study and no hard our liver subgroup (93.3%). Similar to the present study, OAR constraints were violated during plan adaption. Nierer et al. Radiation Oncology (2022) 17:53 Page 12 of 14 References Conclusions 1. Hunt A, Hansen VN, Oelfke U, Nill S, Hafeez S. Adaptive radiotherapy All subgroups clearly benefited from online plan adap - enabled by MRI guidance. Clin Oncol. 2018;30:711–9. https:// doi. org/ 10. tion in terms of improved PTV coverage. Moreover, 1016/j. clon. 2018. 08. 001. 2. Bohoudi O, Bruynzeel AME, Senan S, Cuijpers JP, Slotman BJ, Lagerwaard for the liver, lung and abdominal lymph node cases, a FJ, Palacios MA. Fast and robust online adaptive planning in stereotactic systematic improvement in GTV coverage was found, MR-guided adaptive radiation therapy (SMART ) for pancreatic cancer. resulting in excellent target coverage after re-optimiza- Radiother Oncol. 2017;125:439–44. https:// doi. org/ 10. 1016/j. radonc. 2017. 07. 028. tion in most fractions. In combination with the breath- 3. Heerkens HD, van Vulpen M, van den Berg CAT, Tijssen RHN, Crijns SPM, hold-based technique, these subgroups can fully exploit Molenaar IQ, et al. MRI-based tumor motion characterization and gating the potential of oMRgRT systems. In the pancreatic can- schemes for radiation therapy of pancreatic cancer. Radiother Oncol. 2014;111:252–7. https:// doi. org/ 10. 1016/j. radonc. 2014. 03. 002. cer subgroup, online plan adaption resulted in largely 4. Heijkoop ST, Langerak TR, Quint S, Bondar L, Mens JWM, Heijmen BJM, decreased OAR doses but the target coverage could not Hoogeman MS. Clinical implementation of an online adaptive plan-of- always be improved due to the limiting OAR constraints. the-day protocol for nonrigid motion management in locally advanced cervical cancer IMRT. Int J Radiat Oncol Biol Phys. 2014;90:673–9. https:// While many fractions of the prostate subgroup could, in doi. org/ 10. 1016/j. ijrobp. 2014. 06. 046. theory, also be effectively treated without plan adaption, 5. Liu F, Erickson B, Peng C, Li XA. Characterization and management of most fractions still showed improved PTV coverage and interfractional anatomic changes for pancreatic cancer radiotherapy. Int J Radiat Oncol Biol Phys. 2012;83:e423–9. https:// doi. org/ 10. 1016/j. ijrobp. few fractions even showed large CTV coverage improve- 2011. 12. 073. ments after online plan re-optimization. 6. McPartlin AJ, Li XA, Kershaw LE, Heide U, Kerkmeijer L, Lawton C, et al. MRI-guided prostate adaptive radiotherapy—a systematic review. Radio- Acknowledgements ther Oncol. 2016;119:371–80. https:// doi. org/ 10. 1016/j. radonc. 2016. 04. Not applicable. 7. Pathmanathan AU, van As NJ, Kerkmeijer LGW, Christodouleas J, Lawton Authors’ contributions CAF, Vesprini D, et al. Magnetic resonance imaging-guided adaptive LN extracted technical plan parameters, performed data analysis and drafted radiation therapy: a “game changer” for prostate treatment? Int J Radiat the manuscript. VM helped with data analysis. CE, JB, PT and RB performed Oncol Biol Phys. 2018;100:361–73. https:// doi. org/ 10. 1016/j. ijrobp. 2017. extraction of dose-related data. GL and SC supervised the study. SC designed 10. 020. the study. CE., MN, VM, GL, CK, MR, CB and SC reviewed the manuscript. All 8. Ahunbay EE, Peng C, Chen G-P, Narayanan S, Yu C, Lawton C, Li XA. An authors read and approved the final manuscript. on-line replanning scheme for interfractional variations. Med Phys. 2008;35:3607–15. https:// doi. org/ 10. 1118/1. 29524 43. Funding 9. Ahunbay EE, Peng C, Holmes S, Godley A, Lawton C, Li XA. Online adap- The Department of Radiation Oncology of the University Hospital, LMU tive replanning method for prostate radiotherapy. Int J Radiat Oncol Biol Munich has received research grants from ViewRay Inc. (Oakwood Village, OH, Phys. 2010;77:1561–72. https:// doi. org/ 10. 1016/j. ijrobp. 2009. 10. 013. USA). ViewRay was not involved and had no influence on the study design, 10. Foroudi F, Wong J, Kron T, Rolfo A, Haworth A, Roxby P, et al. Online adap- the collection or analysis of data, on the writing of the manuscript, or the deci- tive radiotherapy for muscle-invasive bladder cancer: results of a pilot sion to submit the manuscript for publication. CE received funding from the study. Int J Radiat Oncol Biol Phys. 2011;81:765–71. https:// doi. org/ 10. German Cancer Aid [Mildred Scheel Stipendienprogramm für Krebsforschung, 1016/j. ijrobp. 2010. 06. 061. 2018 (57468956)]. 11. Li Y, Hoisak JDP, Li N, Jiang C, Tian Z, Gautier Q, et al. Dosimetric benefit of adaptive re-planning in pancreatic cancer stereotactic body radiotherapy. Availability of data and materials Med Dosim. 2015;40:318–24. https:// doi. org/ 10. 1016/j. meddos. 2015. 04. Not applicable. 12. Mohan R, Zhang X, Wang H, Kang Y, Wang X, Liu H, et al. Use of deformed Code availability intensity distributions for on-line modification of image-guided IMRT to Not applicable. account for interfractional anatomic changes. Int J Radiat Oncol Biol Phys. 2005;61:1258–66. https:// doi. org/ 10. 1016/j. ijrobp. 2004. 11. 033. 13. Castelli J, Simon A, Louvel G, Henry O, Chajon E, Nassef M, et al. Impact of Declarations head and neck cancer adaptive radiotherapy to spare the parotid glands and decrease the risk of xerostomia. Radiat Oncol. 2015;10:6. https:// doi. Ethics approval and consent to participate org/ 10. 1186/ s13014- 014- 0318-z. This study was conducted according to the guidelines of the Declaration of 14. Guckenberger M, Wilbert J, Richter A, Baier K, Flentje M. Potential of Helsinki, and approved by the Ethics Committee of LMU (LMU 20-291, 13 May adaptive radiotherapy to escalate the radiation dose in combined 2020). radiochemotherapy for locally advanced non-small cell lung cancer. Int J Radiat Oncol Biol Phys. 2011;79:901–8. https:// doi. org/ 10. 1016/j. ijrobp. Consent for publication 2010. 04. 050. Not applicable. 15. Henke LE, Kashani R, Hilliard J, DeWees TA, Curcuru A, Przybysz D, et al. In silico trial of MR-guided midtreatment adaptive planning for hypof- Competing interests ractionated stereotactic radiation therapy in centrally located thoracic SC received speaker fees/travel support from ViewRay Inc. (Oakwood Village, tumors. Int J Radiat Oncol Biol Phys. 2018;102:987–95. https:// doi. org/ 10. OH, USA). All other authors declare that they have no competing interests. 1016/j. ijrobp. 2018. 06. 022. 16. de Jong R, Visser J, Crama KF, van Wieringen N, Wiersma J, Geijsen ED, Author details Bel A. Dosimetric benefit of an adaptive treatment by means of plan Department of Radiation Oncology, University Hospital, LMU Munich, Mar- selection for rectal cancer patients in both short and long course chioninistr. 15, 81377 Munich, Germany. German Cancer Consortium (DKTK), radiation therapy. Radiat Oncol. 2020;15:13. https:// doi. org/ 10. 1186/ Partner Site Munich, Munich, Germany. s13014- 020- 1461-3. 17. Mohamed ASR, Bahig H, Aristophanous M, Blanchard P, Kamal M, Ding Y, Received: 23 December 2021 Accepted: 27 February 2022 et al. Prospective in silico study of the feasibility and dosimetric advan- tages of MRI-guided dose adaptation for human papillomavirus positive N ierer et al. Radiation Oncology (2022) 17:53 Page 13 of 14 oropharyngeal cancer patients compared with standard IMRT. Clin Transl 34. Menten MJ, Wetscherek A, Fast MF. MRI-guided lung SBRT: present and Radiat Oncol. 2018;11:11–8. https:// doi. org/ 10. 1016/j. ctro. 2018. 04. 005. future developments. Phys Med. 2017;44:139–49. https:// doi. org/ 10. 18. Møller DS, Holt MI, Alber M, Tvilum M, Khalil AA, Knap MM, Hoffmann L. 1016/j. ejmp. 2017. 02. 003. Adaptive radiotherapy for advanced lung cancer ensures target coverage 35. El-Bared N, Portelance L, Spieler BO, Kwon D, Padgett KR, Brown KM, Mel- and decreases lung dose. Radiother Oncol. 2016;121:32–8. https:// doi. lon EA. Dosimetric benefits and practical pitfalls of daily online adaptive org/ 10. 1016/j. radonc. 2016. 08. 019. MRI-guided stereotactic radiation therapy for pancreatic cancer. Pract 19. Nijkamp J, Pos FJ, Nuver TT, de Jong R, Remeijer P, Sonke J-J, Lebesque JV. Radiat Oncol. 2019;9:e46–54. https:// doi. org/ 10. 1016/j. prro. 2018. 08. 010. Adaptive radiotherapy for prostate cancer using kilovoltage cone-beam 36. Hassanzadeh C, Rudra S, Bommireddy A, Hawkins WG, Wang-Gillam A, computed tomography: first clinical results. Int J Radiat Oncol Biol Phys. Fields RC, et al. Ablative five-fraction stereotactic body radiation therapy 2008;70:75–82. https:// doi. org/ 10. 1016/j. ijrobp. 2007. 05. 046. for inoperable pancreatic cancer using online MR-guided adaptation. 20. Oh S, Stewart J, Moseley J, Kelly V, Lim K, Xie J, et al. Hybrid adaptive Adv Radiat Oncol. 2020. https:// doi. org/ 10. 1016/j. adro. 2020. 06. 010. radiotherapy with on-line MRI in cervix cancer IMRT. Radiother Oncol. 37. Placidi L, Romano A, Chiloiro G, Cusumano D, Boldrini L, Cellini F, et al. 2014;110:323–8. https:// doi. org/ 10. 1016/j. radonc. 2013. 11. 006. On-line adaptive MR guided radiotherapy for locally advanced pancreatic 21. van de Schoot AJAJ, de Boer P, Visser J, Stalpers LJA, Rasch CRN, Bel A. cancer: clinical and dosimetric considerations. Techn Innov Patient Sup- Dosimetric advantages of a clinical daily adaptive plan selection strategy port Radiat Oncol. 2020;15:15–21. https:// doi. org/ 10. 1016/j. tipsro. 2020. 06. compared with a non-adaptive strategy in cervical cancer radiation 001. therapy. Acta Oncol. 2017;56:667–74. https:// doi. org/ 10. 1080/ 02841 86X. 38. Rudra S, Jiang N, Rosenberg SA, Olsen JR, Roach MC, Wan L, et al. Using 2017. 12879 49. adaptive magnetic resonance image-guided radiation therapy for treat- 22. Yoon SW, Lin H, Alonso-Basanta M, Anderson N, Apinorasethkul O, ment of inoperable pancreatic cancer. Cancer Med. 2019;8:2123–32. Cooper K, et al. Initial evaluation of a novel cone-beam CT-based semi-https:// doi. org/ 10. 1002/ cam4. 2100. automated online adaptive radiotherapy system for head and neck can- 39. Kerkhof EM, Raaymakers BW, van der Heide UA, van de Bunt L, cer treatment—a timing and automation quality study. Cureus. 2020;12: Jürgenliemk-Schulz IM, Lagendijk JJW. Online MRI guidance for healthy e9660. https:// doi. org/ 10. 7759/ cureus. 9660. tissue sparing in patients with cervical cancer: an IMRT planning study. 23. Inc., Viewray. First patients treated with ViewRay’s MRIdian Linac System Radiother Oncol. 2008;88:241–9. https:// doi. org/ 10. 1016/j. radonc. 2008. at Henry Ford Health System. 2017.04. 009. 24. Corradini S, Alongi F, Andratschke N, Belka C, Boldrini L, Cellini F, et al. 40. Henke L, Kashani R, Robinson C, Curcuru A, DeWees T, Bradley J, et al. MR-guidance in clinical reality: current treatment challenges and Phase I trial of stereotactic MR-guided online adaptive radiation therapy future perspectives. Radiat Oncol. 2019;14:92. https:// doi. org/ 10. 1186/ (SMART ) for the treatment of oligometastatic or unresectable primary s13014- 019- 1308-y. malignancies of the abdomen. Radiother Oncol. 2018;126:519–26. 25. Acharya S, Fischer-Valuck BW, Kashani R, Parikh P, Yang D, Zhao T, et al. https:// doi. org/ 10. 1016/j. radonc. 2017. 11. 032. Online magnetic resonance image guided adaptive radiation therapy: 41. Henke LE, Olsen JR, Contreras JA, Curcuru A, DeWees TA, Green OL, et al. first clinical applications. Int J Radiat Oncol Biol Phys. 2016;94:394–403. Stereotactic MR-guided online adaptive radiation therapy (SMART ) for https:// doi. org/ 10. 1016/j. ijrobp. 2015. 10. 015. ultracentral thorax malignancies: results of a phase 1 trial. Adv Radiat 26. Fischer-Valuck BW, Henke L, Green O, Kashani R, Acharya S, Bradley Oncol. 2019;4:201–9. https:// doi. org/ 10. 1016/j. adro. 2018. 10. 003. JD, et al. Two-and-a-half-year clinical experience with the world’s first 42. Mittauer KE, Hill PM, Geurts MW, de Costa A-M, Kimple RJ, Bassetti MF, magnetic resonance image guided radiation therapy system. Adv Radiat Bayouth JE. STAT-ART: the promise and practice of a rapid palliative single Oncol. 2017;2:485–93. https:// doi. org/ 10. 1016/j. adro. 2017. 05. 006. session of MR-guided online adaptive radiotherapy (ART ). Front Oncol. 27. Güngör G, Serbez İ, Temur B, Gür G, Kayalılar N, Mustafayev TZ, et al. Time 2019;9:1013. https:// doi. org/ 10. 3389/ fonc. 2019. 01013. analysis of online adaptive magnetic resonance-guided radiation therapy 43. Padgett KR, Simpson G, Asher D, Portelance L, Bossart E, Dogan N. Assess- workflow according to anatomical sites. Pract Radiat Oncol. 2020. https:// ment of online adaptive MR-guided stereotactic body radiotherapy of doi. org/ 10. 1016/j. prro. 2020. 07. 003. liver cancers. Phys Med. 2020;77:54–63. https:// doi. org/ 10. 1016/j. ejmp. 28. Lagerwaard F, Bohoudi O, Tetar S, Admiraal MA, Rosario TS, Bruynzeel A. 2020. 07. 027. Combined inter- and intrafractional plan adaptation using fraction par- 44. Palacios MA, Bohoudi O, Bruynzeel AM, van Sörsen de Koste JR, Cobussen titioning in magnetic resonance-guided radiotherapy delivery. Cureus. P, Slotman BJ, et al. Role of daily plan adaptation in MR-guided stereotac- 2018;10: e2434. https:// doi. org/ 10. 7759/ cureus. 2434. tic ablative radiation therapy for adrenal metastases. Int J Radiat Oncol 29. Paulson ES, Ahunbay E, Chen X, Mickevicius NJ, Chen G-P, Schultz C, et al. Biol Phys. 2018;102:426–33. https:// doi. org/ 10. 1016/j. ijrobp. 2018. 06. 002. 4D-MRI driven MR-guided online adaptive radiotherapy for abdominal 45. Vestergaard A, Hafeez S, Muren LP, Nill S, Høyer M, Hansen VN, et al. The stereotactic body radiation therapy on a high field MR-Linac: implemen- potential of MRI-guided online adaptive re-optimisation in radiotherapy tation and initial clinical experience. Clin Transl Radiat Oncol. 2020;23:72– of urinary bladder cancer. Radiother Oncol. 2016;118:154–9. https:// doi. 9. https:// doi. org/ 10. 1016/j. ctro. 2020. 05. 002.org/ 10. 1016/j. radonc. 2015. 11. 003. 30. Raaymakers BW, Jürgenliemk-Schulz IM, Bol GH, Glitzner M, Kotte ANTJ, 46. van Timmeren JE, Chamberlain M, Krayenbuehl J, Wilke L, Ehrbar S, Bogo- van Asselen B, et al. First patients treated with a 1.5 T MRI-Linac: clinical wicz M, et al. Treatment plan quality during online adaptive re-planning. proof of concept of a high-precision, high-field MRI guided radiotherapy Radiat Oncol. 2020;15:203. https:// doi. org/ 10. 1186/ s13014- 020- 01641-0. treatment. Phys Med Biol. 2017;62:L41–50. https:// doi. org/ 10. 1088/ 1361- 47. Klüter S. Technical design and concept of a 0.35 T MR-Linac. Clin Transl 6560/ aa9517. Radiat Oncol. 2019;18:98–101. https:// doi. org/ 10. 1016/j. ctro. 2019. 04. 007. 31. Tetar SU, Bruynzeel AM, Lagerwaard FJ, Slotman BJ, Bohoudi O, Palacios 48. Ahunbay EE, Li XA. Gradient maintenance: a new algorithm for fast online MA. Clinical implementation of magnetic resonance imaging guided replanning. Med Phys. 2015;42:2863–76. https:// doi. org/ 10. 1118/1. 49198 adaptive radiotherapy for localized prostate cancer. Phys Imaging Radiat 47. Oncol. 2019;9:69–76. https:// doi. org/ 10. 1016/j. phro. 2019. 02. 002. 49. Andratschke N, Parys A, Stadtfeld S, Wurster S, Huttenlocher S, Imhoff D, 32. Bruynzeel AME, Tetar SU, Oei SS, Senan S, Haasbeek CJA, Spoelstra FOB, et al. Clinical results of mean GTV dose optimized robotic guided SBRT et al. A prospective single-arm phase II study of stereotactic magnetic- for liver metastases. Radiat Oncol. 2016;11:74. https:// doi. org/ 10. 1186/ resonance-guided adaptive radiotherapy for prostate cancer: early toxic-s13014- 016- 0652-4. ity results. Int J Radiat Oncol Biol Phys. 2019. https:// doi. org/ 10. 1016/j. 50. Baumann R, Chan MKH, Pyschny F, Stera S, Malzkuhn B, Wurster S, et al. ijrobp. 2019. 08. 007. Clinical results of mean GTV dose optimized robotic-guided stereotactic 33. Nicosia L, Sicignano G, Rigo M, Figlia V, Cuccia F, de Simone A, et al. Daily body radiation therapy for lung tumors. Front Oncol. 2018;8:171. https:// dosimetric variation between image-guided volumetric modulated arc doi. org/ 10. 3389/ fonc. 2018. 00171. radiotherapy and MR-guided daily adaptive radiotherapy for prostate 51. Finazzi T, Palacios MA, Haasbeek CJA, Admiraal MA, Spoelstra FOB, cancer stereotactic body radiotherapy. Acta Oncol. 2020;60:1–7. https:// Bruynzeel AME, et al. Stereotactic MR-guided adaptive radiation therapy doi. org/ 10. 1080/ 02841 86X. 2020. 18210 90. for peripheral lung tumors. Radiother Oncol. 2020;144:46–52. https:// doi. org/ 10. 1016/j. radonc. 2019. 10. 013. Nierer et al. Radiation Oncology (2022) 17:53 Page 14 of 14 52. Regnery S, Buchele C, Weykamp F, Pohl M, Hoegen P, Eichkorn T, et al. Adaptive MR-guided stereotactic radiotherapy is beneficial for ablative treatment of lung tumors in high-risk locations. Front Oncol. 2022. https:// doi. org/ 10. 3389/ fonc. 2021. 757031. 53. Mayinger M, Ludwig R, Christ SM, Dal Bello R, Ryu A, Weitkamp N, et al. Benefit of replanning in MR-guided online adaptive radiation therapy in the treatment of liver metastasis. Radiat Oncol. 2021;16:84. https:// doi. org/ 10. 1186/ s13014- 021- 01813-6. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in pub- lished maps and institutional affiliations. Re Read ady y to to submit y submit your our re researc search h ? Choose BMC and benefit fr ? Choose BMC and benefit from om: : fast, convenient online submission thorough peer review by experienced researchers in your field rapid publication on acceptance support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year At BMC, research is always in progress. Learn more biomedcentral.com/submissions http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Radiation Oncology Springer Journals

Dosimetric benefit of MR-guided online adaptive radiotherapy in different tumor entities: liver, lung, abdominal lymph nodes, pancreas and prostate

Loading next page...
 
/lp/springer-journals/dosimetric-benefit-of-mr-guided-online-adaptive-radiotherapy-in-jBitKZk59X
Publisher
Springer Journals
Copyright
Copyright © The Author(s) 2022
eISSN
1748-717X
DOI
10.1186/s13014-022-02021-6
Publisher site
See Article on Publisher Site

Abstract

Background: Hybrid magnetic resonance (MR)-Linac systems have recently been introduced into clinical practice. The systems allow online adaption of the treatment plan with the aim of compensating for interfractional anatomical changes. The aim of this study was to evaluate the dose volume histogram (DVH)-based dosimetric benefits of online adaptive MR-guided radiotherapy (oMRgRT ) across different tumor entities and to investigate which subgroup of plans improved the most from adaption. Methods: Fifty patients treated with oMRgRT for five different tumor entities (liver, lung, multiple abdominal lymph nodes, pancreas, and prostate) were included in this retrospective analysis. Various target volume (gross tumor vol- ume GTV, clinical target volume CTV, and planning target volume PTV ) and organs at risk (OAR) related DVH param- eters were compared between the dose distributions before and after plan adaption. Results: All subgroups clearly benefited from online plan adaption in terms of improved PTV coverage. For the liver, lung and abdominal lymph nodes cases, a consistent improvement in GTV coverage was found, while many fractions of the prostate subgroup showed acceptable CTV coverage even before plan adaption. The largest median improve- ments in GTV near-minimum dose (D ) were found for the liver (6.3%, p < 0.001), lung (3.9%, p < 0.001), and abdomi- 98% nal lymph nodes (6.8%, p < 0.001) subgroups. Regarding OAR sparing, the largest median OAR dose reduction during plan adaption was found for the pancreas subgroup (-87.0%). However, in the pancreas subgroup an optimal GTV coverage was not always achieved because sparing of OARs was prioritized. Conclusion: With online plan adaptation, it was possible to achieve significant improvements in target volume cov- erage and OAR sparing for various tumor entities and account for interfractional anatomical changes. Keywords: Online MRI guided radiotherapy, Plan adaption, MRgOART , Online adaptive RT, MR-guided RT Background Various inter- and intra-fractional anatomical changes in patient anatomy pose a major challenge for the safe *Correspondence: Lukas.Nierer@med.uni-muenchen.de and successful treatment application in modern abla- Department of Radiation Oncology, University Hospital, LMU Munich, tive image-guided radiotherapy (RT). Typical examples Marchioninistr. 15, 81377 Munich, Germany of such changes in patient geometry are different organ Full list of author information is available at the end of the article © The Author(s) 2022. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/. The Creative Commons Public Domain Dedication waiver (http:// creat iveco mmons. org/ publi cdoma in/ zero/1. 0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Nierer et al. Radiation Oncology (2022) 17:53 Page 2 of 14 fillings of bladder, stomach or rectum, breathing-related using the MRIdian system (ViewRay Inc., Oakwood Vil- motion, peristalsis, cardiac motion, tumor response lage, OH, USA) with a step-and-shoot intensity modu- (shrinkage), or organ and patient weight changes [1]. lated radiotherapy (IMRT) technique in the thoracic or Numerous motion patterns of organs at risk (OAR), tar- abdominal region according to the institutional oMRgRT get volumes or quantification of motion amplitudes can clinical protocol. The commercially available system con - be found in the literature [2–6]. sists of a hybrid MR-Linac and an integrated treatment These types of anatomical changes occur on various planning system (TPS). Dose prescription referred to time-scales, ranging from seconds to weeks, and can either the 65%, 80% or 95% isodose. Two patients were potentially be accounted for via tumor-tracking or gat- treated for two lesions simultaneously (patients 35 and ing techniques and plan adaption strategies [1]. Although 49) and one patient was treated for three lesions simul- the technical implementation of such advanced RT tech- taneously (patient 42). In those three patients, the mul- niques can be challenging, several adaptive RT (ART) tiple lesions were treated with one single treatment plan. approaches have found their way into clinical routine [1– u Th s, all OAR constraints of the corresponding frac - 7]. The feasibility and clinical benefit of offline ART using tions were counted as if these patients only had a single computed tomography (CT) or magnetic resonance lesion. Parameters for target volumes (both gross tumor imaging (MRI) has been demonstrated [8–12]. Strategies volume (GTV) and planning target volume (PTV)) were for online ART based on in-room cone-beam computed evaluated separately for each lesion. A total of 265 online tomography (CBCT) have also been proposed [13–21]. adapted fractions were analyzed. Only fractions which A newly developed commercial CBCT-based system were adapted were considered in the analysis. even allows for fast online ART in the clinical routine The mean percentage of adapted fractions per patient (Ethos : Varian Medical Systems, Palo Alto, CA, USA) was 86% (range 15% to 100%) and 79% of all fractions [22]. In this context, combined hybrid MR-Linac systems were adapted in total. Table 2 (Results) shows the portion (e.g. MRIdian : ViewRay Inc., Oakwood Village, USA; or of adapted plans and characteristics of the online adapted Unity : Elekta AB, Stockholm, Sweden) have the advan- plans for each subgroup. tage of superior soft-tissue contrast and dose-free intra- fractional imaging (where available, also with real-time oMRgRT workflow tumor tracking and gated RT) in addition to a full online The oMRgRT workflow was similar to that described by ART workflow. After the first patient was treated on such Bohoudi et al. [2]. For initial treatment planning, a plan- a hybrid device in 2017 [23], combined MR-guided RT ning MR and CT were acquired using the same patient systems for online ART (“MR-Linacs”) are now com- setup. The planning CT was acquired immediately after mercially available and becoming increasingly popular the MR. The CT was registered using deformable image [24]. The feasibility of online magnetic resonance guided registration (DIR) to the MR to obtain electron density ART (oMRgRT) has already been demonstrated [25–31] values for dose calculation. GTV, clinical target volume and initial studies have reported its dosimetric and clini- (CTV) and OAR delineation was performed on the MR. cal benefits in a wide range of indications [32–46]. The In the TPS, Boolean operations of regions of interest aim of the present study was to evaluate the potential (ROIs; e.g. subtraction or margin expansion of struc- of treatment adaption in oMRgRT in terms of improved tures) can be performed and stored as so-called “rules”. target volume coverage and OAR sparing across five dif - Such rules were defined for the automatic generation ferent tumor entities, which are frequently treated with of the PTV (expansion of the GTV) and derived struc- oMRgRT. tures were defined at the treating physician’s discretion to reduce the contouring effort during online adap - Methods tion. After dose prescription and contour delineation, a Patients baseline treatment plan was generated analogous to the Overall, 50 patients treated between 01/2020 and workflow in conventional RT. All plans were generated 11/2021 were included in this retrospective analysis. All as step-and-shoot IMRT via inverse planning (6 MV patients were treated for one out of five different tumor flattening filter free beam; 1.5 mm calculation grid size entities, which are typical indications for oMRgRT (liver, with isotropic voxels; 1.0% Monte Carlo dose calcula- lung, abdominal lymph nodes, pancreas or prostate). tion uncertainty) and the maximum number of multi Ten patients per entity were randomly selected from all leaf collimator (MLC) segments was limited, depend- patients of the respective subgroups, who successfully ing on the complexity of the plan. This segment num - completed their RT treatment in the given period. Table 1 ber limit of the baseline plan was subsequently used provides an overview of patient and treatment character- for online plan adaption. For each treatment fraction, istics. All patients received hypofractionated oMRgRT a balanced steady-state free precession (bSSFP) pulse N ierer et al. Radiation Oncology (2022) 17:53 Page 3 of 14 Table 1 Patient characteristics Patient Nr Nr. of lesions Nr. of adapted Total nr. of fx Fraction dose Total dose (Gy) Prescription (%) Group fx (Gy) 1 1 3 20 3 60 95 Prostate 2 1 14 20 3 60 95 Prostate 3 1 5 5 7.25 36.25 95 Prostate 4 1 7 20 3 60 95 Prostate 5 1 20 20 3 60 95 Prostate 6 1 19 20 3 60 95 Prostate 7 1 20 20 3 60 95 Prostate 8 1 16 20 3 60 95 Prostate 9 1 9 20 3 60 95 Prostate 10 1 5 5 7 35 95 Prostate 11 1 2 5 8 40 95 Pancreas 12 1 4 5 8 40 95 Pancreas 13 1 5 5 8 40 80 Pancreas 14 1 5 5 8 40 80 Pancreas 15 1 5 5 8 40 80 Pancreas 16 1 5 5 8 40 80 Pancreas 17 1 5 5 8 40 80 Pancreas 18 1 5 5 8 40 80 Pancreas 19 1 5 5 6.6 33 80 Pancreas 20 1 5 5 8 40 80 Pancreas 21 1 3 3 12.5 37.5 65 Liver 22 1 3 3 12.5 37.5 65 Liver 23 1 3 3 12.5 37.5 65 Liver 24 1 3 3 12.5 37.5 65 Liver 25 1 3 3 15 45 65 Liver 26 1 3 3 12.5 37.5 65 Liver 27 1 2 3 15 45 65 Liver 28 1 3 3 12.5 37.5 65 Liver 29 1 3 3 15 45 65 Liver 30 1 2 3 12.5 37.5 65 Liver 31 1 5 5 7 35 95 Lymph nodes 32 1 4 5 5 25 80 Lymph nodes 33 1 2 5 8 40 95 Lymph nodes 34 1 4 5 6.4 32 80 Lymph nodes 35 2 5 6 6 36 80 Lymph nodes 36 1 9 10 4 40 95 Lymph nodes 37 1 5 5 7 35 95 Lymph nodes 38 1 3 5 7 35 95 Lymph nodes 39 1 4 5 7 35 80 Lymph nodes 40 1 5 5 6 30 80 Lymph nodes 41 1 3 5 10 50.0 95 Lung 42 3 3 3 13.5 40.5 65 Lung 43 1 2 3 13.5 40.5 65 Lung 44 1 3 3 13.5 40.5 65 Lung 45 1 3 3 13.5 40.5 65 Lung 46 1 2 3 13.5 40.5 65 Lung 47 1 3 3 13.5 40.5 65 Lung 48 1 3 3 13.5 40.5 65 Lung 49 2 3 3 13.5 40.5 65 Lung 50 1 2 3 13.5 40.5 65 Lung Nierer et al. Radiation Oncology (2022) 17:53 Page 4 of 14 Table 1 (continued) The dose prescription refers to the corresponding isodose sequence 3D setup MRI scan was acquired for transla- the target volume showed a breathing-related motion tional patient setup correction (couch shift). For more were treated using a breath-hold technique (mostly deep information about the MR pulse sequences and the inspiration breath hold). However, patients treated with technical design of the MRIdian system, refer to Klüter a free breathing approach (in cases of very limited tumor et  al. [47]. The MRI of the baseline plan was then reg - motion, e.g. most prostate cases) were also treated using istered via DIR to the volumetric setup MRI of the day the automated gating function in order to ensure that the and all target structures, OARs and the electron den- target was positioned within tolerance boundaries during sity of the planning CT were propagated onto the setup treatment application. MRI. All contours were edited (if necessary), a track- All baseline plans were validated dosimetrically with ing contour was defined and the baseline plan was cal - an ionization chamber and/or diode detector array (Arc- culated on the MRI (more precisely the synthetic CT) Check-MR; Sun Nuclear Corporation, Melbourne, FL, of the day, which results in the so-called predicted USA) prior to the first fraction. dose (baseline plan calculated on the anatomy of the day with updated structures). This dose distribution Extraction of DVH and plan parameters shows the dose of a single non-adapted fraction and is Several dose volume histogram (DVH) parameters the basis to decide whether to adapt a plan or not. In were extracted from the TPS for the predicted (non- case of a subsequent plan adaption, the initial predicted adapted) and re-optimized scenarios for all fractions: dose corresponds to the dose distribution prior to plan the dose to 98%, 95%, 50% and 2% of the volume of the adaption. For most OARs, a time-saving and practi- PTV (PTV D = near minimum dose, PTV D , PTV 98% 95% cal partial re-contouring approach was used, in which D = median dose, PTV D = near maximum dose) 50% 2% the OARs were edited only in the close surrounding of and the mean PTV dose (PTV D ). All parameters mean the PTV (PTV + 3.0  cm), where the highest dose gra- were also reported for the GTV. In prostate cases, the dients occur. This approach was described by Bohoudi CTV was reported, as no GTV was defined. Further - et al. [2], while Ahunbay et al. described a similar basic more, the volume of the PTV (V ) was extracted. Out PTV concept [48]. When the decision was made to adapt of usually multiple, patient-specific OAR constraints pre - the plan of the current fraction, the plan parameters scribed in the TPS, three OAR constraints were chosen and dose constraints of the baseline plan were used for OARs, which were closest to the PTV and as a result as a starting point for dose optimization. A treatment received the highest maximum dose values. Only OAR plan was adapted if either the target coverage or OAR constraints were chosen, which were related to structures constraints of the predicted plan were not fulfilled, or that were updated (in the PTV + 3.0  cm region; see last a combination of both. Therefore, the planning goal sub-section). The individual OAR dose constraints (near was always to achieve optimal target coverage while maximum dose or dose-to-volume constraints) depend respecting all OAR constraints. Online plan adaption on the dose prescription and the individual case and were was performed either as re-optimization with the same defined by a senior physician based on the applicable objectives of the baseline plan or as full re-optimization guidelines. with modified objectives and/or plan parameters. The Technical plan parameters like the net beam-on time dose distribution of the online adapted plan, calculated (BOT), the number of segments, the number of Monitor on the current synthetic CT (based on the MR of the Units (MU) and the number of beams were read out from day) with updated structures is referred to as re-opti- treatment plan documentation files (Table 2; Results). mized dose. All dose calculation settings for the re- optimized dose were the same as for the predicted dose Comparison of DVH parameters, statistical analysis (as defined in the baseline plan). After plan adaption, and definition of dosimetric endpoints the dose was verified for QA using a secondary Monte First the OAR related DVH parameters for the predicted Carlo code before treatment. and re-optimized doses of each adapted fraction were For tumor tracking via a 2D bSSFP cine MRI sequence, compared. the tracking structure was propagated onto a 2D cine Second, the same comparison was made for target MRI slice and a gating ROI was created by expansion volume (PTV, CTV or GTV) related DVH parameters. of the tracking structure. These structures were subse - These parameters were systematically compared pairwise quently used for online beam gating. All patients in which N ierer et al. Radiation Oncology (2022) 17:53 Page 5 of 14 to quantify the DVH-based dosimetric benefit of online was performed via paired Wilcoxon signed-rank test. A plan adaption for each subgroup separately. Primary significance level of α = 5% was used. dosimetric endpoints were chosen as follows: increase in GTV (all cases expect prostate) or CTV (prostate cases) D98%, increase in GTV or CTV D95%, increase in GTV Results mean dose, reduction in OAR exposure. The GTV (or Characteristics of the online adapted plans are shown in CTV) near-minimum parameters (D98% and D95%) Table 2. were chosen because underdose (e.g. dose drop at the Figure  1 shows an example DVH of a single adaptive surface or cold spots) within the GTV/CTV are likely to fraction of an abdominal lymph node case and illustrates affect the local control (LC). The GTV (or CTV in pros - the large potential of online ART to increase target cov- tate cases) mean dose was chosen because there is evi- erage and OAR sparing with the oMRgRT technique. For dence for a predictive value of this parameter in terms example, the GTV D improved by more than 10.0% 95% of improved LC in liver [49] and lung tumors [50]. Box- and the duodenum D was reduced by about one third. 50% plots were generated to visualize the data. Therefore, due Figure  2 indicates for each subgroup the change of to the different dose prescriptions, data was normalized. OAR exposure when plans were adapted. The three most The D and D dose values were normalized to the frequently considered OARs per region were bowel, duo- 98% 95% prescribed dose (PD) for non-homogenous stereotac- denum, stomach (for liver), lung, heart, esophagus (for tic prescriptions (65% or 80% prescription isodose PI) lung), bowel, duodenum, spinal cord (for lymph nodes), or 0.95 × PD for homogenous prescriptions (95% PI). duodenum, stomach, bowel (for pancreas), and rectum, The D and D dose values were normalized to the bladder, femur (for prostate). The largest median dose 50% mean PD. The D was normalized to PD/PI (thus PD/0.8 for reduction of OARs adjacent or close to the PTV was 2% a prescription to the 80% isodose) for stereotactic pre- found for the pancreas subgroup (-87.0%). This dose scriptions or to PD for homogenous prescriptions. Since reduction was significant, as well as the smaller dose all normalization factors depend solely on the plan- reduction achieved in the lymph-nodes subgroup. Small ning aim for the PTV, these factors may differ between but significant median increased OAR doses were found patients but the same factors are applied to all fractions for liver and lung and no statistically significant differ - of a single patient. The dose normalization was done to ence was found for prostate. All p-values and median achieve comparable data for the generation of analysis percent changes are shown in the first row of Table  3. For figures. Statistical analysis or the calculation of percent a more detailed insight in the effect on individual OARs, changes of dose values is not affected by this normaliza - Table  4 provides mean and median percent changes of tion. Statistical analysis of the DVH-based parameters Table 2 Characteristics of the online adapted plans Liver Lung Lymph nodes Pancreas Prostate Adapted fractions (%) 93.3 84.4 82.1 92.0 69.4 Mean BOT (min) 7.1 9.1 4.5 6.2 1.9 Min. BOT (min) 4.9 4.5 2.5 3.2 1.1 Max. BOT (min) 11.3 17.7 9.4 9.3 4.4 Mean number of beams 12 10 16 15 14 Min. number of beams 9 8 9 13 9 Max. number of beams 16 15 19 17 21 Mean number of segments 33 26 72 70 54 Min. number of segments 13 9 12 46 33 Max. number of segments 54 40 100 95 129 Mean MU 4269.2 5439.9 2652.2 3688.4 1139.4 Min. MU 2950.3 2675.2 444.3 1906.7 649.0 Max. MU 6776.0 10,612.0 5607.9 5547.1 2624.5 Mean V (cm ) 38.4 15.2 65.5 251.1 114.7 PTV Min. V (cm ) 7.3 3.9 1.9 59.6 67.9 PTV Max. V (cm ) 109.9 32.0 291.3 455.8 192.6 PTV BOT = beam-on time, MU = number of monitor units, V = volume of the PTV PTV Nierer et al. Radiation Oncology (2022) 17:53 Page 6 of 14 PTV re. PTV pr. Stomach re. Stomach pr. Duodenum re. Duodenum pr. GTV re. GTV pr. Bowel re. Bowel pr. CBD* re. CBD* pr. 01234567 8 Dose [Gy] Fig. 1 Exemplary cumulative DVH of fraction 2 of patient 32 under the adaptive workflow (abdominal lymph node treatment with a dose prescription of 5 × 5.0 Gy to the 80% isodose); re-optimized (re., solid lines) and predicted (pr., dashed lines). Target coverage increased after plan adaption while OAR exposure could be reduced. *Common bile duct Maximum Q3 Median Q1 -50 Minimum -100 Outlier (1.5 × IQR) -150 -200 Fig. 2 Boxplots of percent changes of OAR DVH parameters re-optimized versus predicted for all subgroups. The negative (lower) half corresponds to fractions where OAR exposure was reduced when adapted and the positive (upper) half corresponds to increased OAR exposure when adapted. Q1: first quartile, Q3: third quartile, IQR: interquartile range OAR dose parameters of three most frequently consid- For the target volume DVH parameters, the largest ered OARs per region. changes were found for D and D for the PTV and 98% 95% GTV/CTV when comparing the re-optimized with the Liver Lung Lymph nodes Pancreas Prostate Change of OAR dose constraints [%] Volume [%] N ierer et al. Radiation Oncology (2022) 17:53 Page 7 of 14 Table 3 p-values and median percent changes [p value/median change (%)] of DVH OAR and target volume parameters when comparing the re-optimized versus the predicted dose distributions Liver Lung Lymph nodes Pancreas Prostate OAR 0.000/6.9 0.004/5.9 0.001/−4.5 0.000/−87.0 0.135/0.8 PTV D 0.000/3.8 0.006/1.0 0.000/3.0 0.041/0.8 0.000/0.9 mean PTV D 0.712/0.1 0.131/−0.7 0.757/0.0 0.002/−0.9 0.093/−0.2 2% PTV D 0.001/2.6 0.065/0.0 0.003/0.9 0.196/−0.5 0.021/0.3 50% PTV D 0.000/17.3 0.000/6.0 0.000/9.4 0.000/5.7 0.000/2.8 95% PTV D 0.000/25.5 0.000/7.8 0.000/15.6 0.000/11.0 0.000/5.8 98% GTV or CTV D * 0.004/1.0 0.012/0.9 0.000/1.4 0.454/−1.3 0.007/0.4 mean GTV or CTV D 0.872/0.0 0.042/−0.4 0.949/−0.2 0.176/−0.8 0.176/−0.1 2% GTV or CTV D 0.040/0.6 0.127/0.6 0.020/1.0 0.164/−1.4 0.185/0.2 50% GTV or CTV D 0.001/4.6 0.000/3.1 0.000/4.9 0.946/−0.3 0.000/0.7 95% GTV or CTV D 0.000/6.3 0.000/3.9 0.000/6.8 0.589/0.5 0.000/1.3 98% Significant differences are highlighted bold *GTV for all subgroups except prostate and CTV for prostate Table 4 Mean and median percent changes when comparing the re-optimized versus the predicted dose distributions [mean change (%)/median change (%)] of the most frequently used OAR dose parameter for each of the three most frequently considered OARs per region Liver D (bowel) D (duodenum) D (stomach) max max max 18.2/8.5 35.6/50.8 11.2/14.0 Lung V15 (lung left or right) D (heart) D (esophagus) Gy max max 9.0/5.9 18.3/0.0 3.6/8.2 Lymph nodes V20 (bowel) V18 (duodenum) D (spinal canal) Gy Gy max − 10.8/− 27.8 − 43.9/− 48.2 − 2.9/− 4.4 Pancreas V33 (duodenum) V33 (stomach) V33 (bowel) Gy Gy Gy − 94.8/− 97.1 − 84.0/− 99.1 − 83.4/− 98.6 Prostate V40 (rectum) V40 (bladder) D (femur left or right) Gy Gy max 6.8/0.3 9.3/7.4 − 0.7/− 1.2 predicted doses (Fig.  3). All PTV D and D median were found for prostate, compared to the liver, lung and 98% 95% dose values increased significantly across all subgroups. lymph node subgroups. All p-values and median percent More importantly, the GTV/CTV D and D values increases are shown in Table 3. 98% 95% increased significantly, except for pancreas, where no sig - Additional target dose volume parameters are shown nificant difference was found (see Table  3). This means in Fig.  4. The near maximum doses D did not show 2% that the target volume dose coverage increased signifi - significant changes in most cases. Significant but small cantly in most cases when adapting. The largest median median near maximum dose reductions inside the PTV increases of PTV D were found in liver, lung and or GTV were found for PTV D (pancreas) and GTV 95% 2% lymph nodes. The largest median increases in GTV D D (lung). When looking at the boxplots (Fig.  4), it can 95% 2% were also found in the same subgroups. Although a rela- be seen that a few cases of the lung, lymph node and pan- tively large significant increase was found for PTV D creas subgroups showed predicted near maximum doses, 95% in pancreatic cases, no significant increase was found for which exceeded 10% of the ideally achieved maximum the pancreas GTV D and GTV D . Despite being dose (mostly outliers). A slight reduction of these high 98% 95% statistically significant, smaller median increases in PTV near maximum doses was achieved when adapting. After D , PTV D , GTV/CTV D and GTV/CTV D plan adaption, all non-outlier near-maximum values 98% 95% 98% 95% Nierer et al. Radiation Oncology (2022) 17:53 Page 8 of 14 Liver LungLymph nodes PancreasProstate Liver Lung Lymph nodes PancreasProstate pr.re. pr.re. pr. re.pr. re.pr. re. pr. re.pr. re.pr. re.pr. re.pr. re. 150 150 140 140 130 130 120 120 110 110 100 100 90 90 80 80 70 70 60 60 50 50 40 40 Liver Lung Lymph nodesPancreasProstate Liver Lung Lymph nodes PancreasProstate pr.re. pr.re. pr. re. pr. re. pr.re. pr.re. pr.re. pr.re. pr. re.pr. re. Fig. 3 Boxplots for all subgroups of DVH target volume dose values D and D for PTV and GTV/CTV (*GTV for all cases except prostate; CTV in 98% 95% case of prostate) for the predicted (pr.) and re-optimized (re.) dose distributions. Dose normalized to the ideally achieved PTV encompassing dose (see sub-section “Comparison of DVH parameters and statistical analysis”) PTV D and GTV D exceeded the ideally achieved PTV coverage, similar to that of the lung cases. With 2% 2% maximum dose by less than 10%. PTV and GTV/ CTV 93.3% of all fractions adapted, liver showed the highest median (D ) and mean (D ) doses show an inverse portion of adapted plans (similar to pancreas with 92.0%). 50% mean behaviour. For these values, slight median increases were This means that the initial target volume coverage using found in most cases, except for the pancreas subgroup, the base plan was not ideal in almost every fraction. After where no significant changes were found for PTV D , plan adaption, PTV D was > 97.0% of the PD in all frac- 50% 98% GTV D and GTV D (Table 3). tions. Even though the initial median PTV coverage was 50% mean worse compared to the lung cases, the initial GTV cov- Discussion erage was similar, which means that the PTV, designed Liver for liver cases in breath hold technique, worked very well. For the liver subgroup, the initial predicted median PTV When comparing the normalized percent values of Fig. 3 coverage (here PTV D and PTV D ) before plan between these two (or any other) subgroups, especially 98% 95% adaption was worse compared to lung and prostate but for the GTV, one has to bear in mind that all liver cases was comparable to lymph nodes and pancreas. One rea- had a stereotactic prescription to the 65% isodose, but son might be that less complex shaped (sphere-like) tar- only 8/10 lung cases had the same prescription. No sig- gets, as small lung lesions (lung: mean V = 15.2 cm vs. nificant change in the PTV or GTV near-maximum dose PTV liver: mean V = 38.4 cm , see Table 2) or prostate tar- was found, just like for most other subgroups. Regarding PTV gets, are easier to cover with default baseline plans. After OAR sparing, Fig.  2 shows a more or less symmetrical plan adaption, a largely improved PTV coverage was distribution around zero for liver, but with a significantly found, which resulted in a close-to-ideal post-adaption increased median OAR dose. This is because OAR dose PTV D [%] PTV D [%] 95% 98% GTV or CTV* D [%] GTV or CTV* D [%] 98% 95% N ierer et al. Radiation Oncology (2022) 17:53 Page 9 of 14 160 160 150 150 140 140 130 130 120 120 110 110 100 100 90 90 80 80 Liver Lung Lymph nodesPancreas Prostate LiverLungLymph nodes Pancreas Prostate pr.re. pr.re. pr. re. pr. re.pr. re. pr.re. pr. re. pr. re. pr. re. pr.re. 130 130 125 125 120 120 115 115 110 110 105 105 100 100 95 95 LiverLung Lymph nodes Pancreas Prostate LiverLungLymph nodes Pancreas Prostate pr.re. pr.re. pr. re. pr. re.pr. re. pr.re. pr.re. pr.re. pr.re. pr.re. Liver LungLymph nodesPancreas Prostate LiverLungLymph nodes Pancreas Prostate pr.re. pr.re. pr.re. pr.re. pr.re. pr.re. pr.re. pr.re. pr.re. pr.re. Fig. 4 Boxplots for all subgroups of DVH target volume dose values D, D and D for PTV and GTV or CTV (*GTV for all cases except prostate; 50% 2% mean CTV in case of prostate) for the predicted (pr.) and re-optimized (re.) dose distributions. D and D normalized to the prescribed dose. D 50% mean 2% normalized to the ideally achieved maximum dose (see sub-section “Comparison of DVH parameters and statistical analysis”) limits were, on average, not fully reached prior to adap- OARs were found for liver. During the optimization pro- tion. This tendency can also be seen when looking at the cess, the OAR exposure was fully exploited and brought exposure changes of the three most frequently consid- closer to the dose limits, in order to achieve a very good ered OARs (bowel, duodenum, stomach, see Table  4), target coverage without violation of OAR constraint. where mean and median increases in dose for all these With a mean number of 33 segments, adapted liver plans PTV D [%] PTV D [%] 50% PTV D [%] 2% Mean GTV or CTV* D [%] GTV or CTV* D [%] GTV or CTV* D [%] Mean 50% 2% Nierer et al. Radiation Oncology (2022) 17:53 Page 10 of 14 were simpler compared to those of lymph node, pancreas (Table 4). Without online plan adaption, OAR dose limits and prostate cases. would have been frequently violated. Lung Prostate For lung cases, the initial PTV coverage of the predicted For most prostate cases, the initial PTV and especially base plan was better compared to liver, but could still be the CTV coverage were acceptable. In only a few frac- significantly improved and resulted in a near-optimal tions very insufficient initial target volume coverage was PTV coverage after plan adaption. The initial GTV cov - found (Fig. 3). For these fractions, large improvements in erage was similar to that of liver, but could still be sig- PTV and even CTV coverage were made when re-opti- nificantly improved. With 84.4% of all fractions adapted, mizing the plans. On average, no significant change of lung showed similar adaption rates as lymph node cases. OAR exposure was achieved with online plan adaption. A Regarding OAR sparing and change of PTV near-maxi- mean of 54 segments were needed to achieve acceptable mum doses, similar findings were made as for the liver plans, which indicates plans of medium complexity. Only subgroup. The mean PTV volume of lung cases was the 69.4% of fractions were adapted. smallest of all subgroups (15.2 cm ) and so was the mean number of segments (26), indicating easy-to-adapt, sim- Comparison between tumor entities ple treatment plans. All subgroups clearly benefited from online plan adap - tion in terms of improved PTV coverage. The improved Abdominal lymph nodes target dose varied between the different tumor entities. For the abdominal lymph nodes subgroup, initial PTV To estimate the absolute dose changes of the target vol- coverage was not ideal but could be efficiently improved umes achieved by online plan adaption, it is possible to with online plan adaption, which was performed in multiply the total prescribed dose of any dose prescrip- 82.1% of fractions. As for to the liver and lung cases, a tion (Table  1) with the corresponding median percent relatively large median increase (6.8%) in GTV D was change of any DVH parameter of interest (Table  3) and 98% found without an increase in the near-maximum dose. with the percentage of adapted fractions of the cor- The median OAR dose could be reduced significantly responding patient subgroup (Table  2, line 1). Obvi- by − 4.5% (Table 3). A larger reduction of OAR dose was ously, this formula can only provide a rough estimate of only observed in the pancreas subgroup. The lymph node absolute dose changes and does not consider individual treatment plans had an average number of 72 segments, patients. Based on the underlying idea of the PTV target which is similar to that of pancreas (70). This shows that volume concept, one would assume that the GTV/CTV lymph node treatment plans were usually highly modu- coverage physically achieved during dose application is lated, similar to the pancreas plans. of higher prognostic value than the PTV coverage, since the sole purpose of the PTV margins is to guarantee the Pancreas GTV/CTV coverage with some degree of confidence. Plan adaptation resulted in a significant improvement Under this assumption it is possible to identify subgroups in PTV near minimum dose. Regarding the GTV cover- that might benefit more from online plan adaption than age, no significant improvement could be made on aver - others. In the present study, the three subgroups liver, age, when performing online plan adaption. Although the lung, and abdominal lymph nodes had the greatest ben- average re-optimized (and even predicted) GTV coverage efit from online plan adaption in terms of improved was acceptable, some cases were observed where suf- GTV coverage. After plan adaption, all three subgroups ficient GTV coverage could not be achieved, even after showed excellent GTV coverage (D and D ). In 98% 95% plan adaption, because sparing of OARs was prioritized. addition, a large portion of fractions (> 82%) required This can be seen in Fig.  3, when looking at the relatively re-optimization in all three subgroups, indicating that large interval between the first and third quartiles com - most of these fractions can be significantly improved by pared to those of the abdominal lymph nodes and pros- online plan adaption. Since small values of GTV D 98% tate cases. The large portion of adapted fractions (92.0%) and D indicate insufficient GTV coverage, these indi - 95% indicates the need for plan adaptation to reduce OAR ces can be considered the most predictive (of all DVH doses to meet clinically acceptable OAR dose levels. parameters examined) in terms of improved tumor con- OAR doses could be reduced in more than 75.0% of all trol probability (TCP) and possibly LC when comparing OAR constraints of all adapted fractions of the pancreas non-adapted and online adapted fractions. However, it cases and the median OAR dose reduction was − 87.0% remains unclear if the significant increases in GTV near- (Fig.  2). The V33 of duodenum, stomach and bowel minimum and mean doses will translate into a detectable Gy was reduced on average by more than 80% respectively improved LC. N ierer et al. Radiation Oncology (2022) 17:53 Page 11 of 14 In addition to the significantly improved GTV cover - several fractions among the pancreas cases were found, age in the liver, lung and abdominal lymph nodes sub- where GTV dose de-escalation was necessary due to groups when using plan adaption, these subgroups might OAR proximity of the tumor. Regnery et al. [52] prospec- also largely benefit from the breath-hold and automated tively compared predicted versus adapted dose distribu- beam gating capabilities of the oMRgRT system, since tions in 154 online adapted fractions in 21 lung tumor no internal target volume concept is needed. An internal patients. The higher adaption rate of 93.3% compared target volume concept would increase the total irradiated to 84.4% in the current study can be explained with the volume [51], especially for these cases, where breathing- large number of ultracentral lung tumors in the cohort related motion of target volumes can be frequently seen. of Regnery et  al., where OAR violations are more likely In this study, the influence of beam-gating was not inves - to occur due to adjacency OARs. Regnery et  al. found tigated and no motion range assessment was made for a large increase in the minimum biologically effective the target volumes. That is beyond the scope of this study. dose (BED) of the PTV and a moderate increase in the It was found for pancreas, that the plan adaption capa- minimum BED of the GTV. We observed the same ten- bilities were largely used to reduce OAR doses to an dency when considering PTV and GTV D or D . In 98% 95% acceptable level. Although the re-optimized GTV cov- the same study only small increases in mean BED inside erage was acceptable in most cases, the improvement in the PTV and GTV were found, which is also in accord- GTV coverage via online plan adaption was not as large ance with our findings of PTV and GTV D . El Bared mean as in the other subgroups, as OAR sparing was prior- et al. [35] evaluated the dosimetric benefits in 10 patients itized. In contrast, in the prostate subgroup, only 69.4% treated for unresectable pancreatic cancer on a cobalt-60, of all fractions were considered for re-optimization. For 0.35  T MRI system when performing online plan adap- most of these fractions, the initial CTV coverage was tion and reported outcome. Although comparability to quite good but could still be slightly improved when our study is limited due to a fundamental discrepancy in adapted. Ultimately, few fractions were found, where the technical design and beam quality (tri-source cobalt-60 initial PTV and CTV coverage was unacceptable. For vs. 6 MV flattening filter free Linac), El Bared also found these fractions, online re-optimization resulted in an improved PTV coverage when performing online plan excellent CTV coverage. In summary, for prostate cases, adaption. However, the influence on GTV coverage was the benefit of online re-optimization was found to be not evaluated. Placidi et  al. [37] found a similar trend of not as systematic as for the liver, lung, abdominal lymph improved PTV coverage in 8 pancreatic cancer patients nodes and pancreas cases. In each subgroup, at least also treated on the cobalt-60 system and similarly one of the primary dosimetric endpoints defined under reported increased CTV dose after plan adaption. The “Methods” (significant increase in GTV/CTV near-min - influence on GTV coverage was not evaluated. Mayinger imum and mean dose, and significantly reduced OAR et al. [53] analyzed online adapted treatment plans of 15 exposure) was achieved (Table  3). Except for pancreas, patients with liver metastases and found improved PTV all other subgroups met all primary endpoints related to coverage in cases where the target volume was in close improved GTV/CTV dose. proximity to OARs. The influence on GTV coverage was To the best of our knowledge, up to now, no attempt not evaluated in detail. In the present study, we did not has been made to quantify the influence of MR-guided stratify patients for adjacency of OARs, but when look- online plan adaption on DVH-related parameters and ing at the PTV mean and near-minimum doses, we also systemically compare the results between multiple sub- found a significant increase after plan adaption in liver groups with different tumor entities typically treated on cases. Padgett et al. [43] artificially created adapted plans integrated MR-Linac systems. Our intent was to pro- for 10 patients with liver cancers on the cobalt-60 system vide information for a more informed decision making and compared the results to the non-adapted plans and when assigning patients to the (still very limited access) also found improved PTV and GTV coverage as well as MR-Linac. Henke et  al. [40] analyzed 81 online adapted a reduced number of OAR violations (duodenum, bowel fractions (20 patients) in a patient cohort of mixed oli- and stomach) after plan adaption. We observed the same gometastatic or unresectable abdominal malignancies trend regarding target volumes and reported small mean (hepatic lesions, adrenal metastasis, pancreatic adenocar- and median dose increases for OARs like duodenum, cinoma and lymph node metastases). The overall adap - bowel and stomach. This is no contradiction to the find - tion rate (83.5%) was comparable to our study, although ings of Padgett et  al. since OAR sparing was prioritized less liver fractions (66.0%) were adapted compared to over target volume coverage in our study and no hard our liver subgroup (93.3%). Similar to the present study, OAR constraints were violated during plan adaption. Nierer et al. Radiation Oncology (2022) 17:53 Page 12 of 14 References Conclusions 1. Hunt A, Hansen VN, Oelfke U, Nill S, Hafeez S. Adaptive radiotherapy All subgroups clearly benefited from online plan adap - enabled by MRI guidance. Clin Oncol. 2018;30:711–9. https:// doi. org/ 10. tion in terms of improved PTV coverage. Moreover, 1016/j. clon. 2018. 08. 001. 2. Bohoudi O, Bruynzeel AME, Senan S, Cuijpers JP, Slotman BJ, Lagerwaard for the liver, lung and abdominal lymph node cases, a FJ, Palacios MA. Fast and robust online adaptive planning in stereotactic systematic improvement in GTV coverage was found, MR-guided adaptive radiation therapy (SMART ) for pancreatic cancer. resulting in excellent target coverage after re-optimiza- Radiother Oncol. 2017;125:439–44. https:// doi. org/ 10. 1016/j. radonc. 2017. 07. 028. tion in most fractions. In combination with the breath- 3. Heerkens HD, van Vulpen M, van den Berg CAT, Tijssen RHN, Crijns SPM, hold-based technique, these subgroups can fully exploit Molenaar IQ, et al. MRI-based tumor motion characterization and gating the potential of oMRgRT systems. In the pancreatic can- schemes for radiation therapy of pancreatic cancer. Radiother Oncol. 2014;111:252–7. https:// doi. org/ 10. 1016/j. radonc. 2014. 03. 002. cer subgroup, online plan adaption resulted in largely 4. Heijkoop ST, Langerak TR, Quint S, Bondar L, Mens JWM, Heijmen BJM, decreased OAR doses but the target coverage could not Hoogeman MS. Clinical implementation of an online adaptive plan-of- always be improved due to the limiting OAR constraints. the-day protocol for nonrigid motion management in locally advanced cervical cancer IMRT. Int J Radiat Oncol Biol Phys. 2014;90:673–9. https:// While many fractions of the prostate subgroup could, in doi. org/ 10. 1016/j. ijrobp. 2014. 06. 046. theory, also be effectively treated without plan adaption, 5. Liu F, Erickson B, Peng C, Li XA. Characterization and management of most fractions still showed improved PTV coverage and interfractional anatomic changes for pancreatic cancer radiotherapy. Int J Radiat Oncol Biol Phys. 2012;83:e423–9. https:// doi. org/ 10. 1016/j. ijrobp. few fractions even showed large CTV coverage improve- 2011. 12. 073. ments after online plan re-optimization. 6. McPartlin AJ, Li XA, Kershaw LE, Heide U, Kerkmeijer L, Lawton C, et al. MRI-guided prostate adaptive radiotherapy—a systematic review. Radio- Acknowledgements ther Oncol. 2016;119:371–80. https:// doi. org/ 10. 1016/j. radonc. 2016. 04. Not applicable. 7. Pathmanathan AU, van As NJ, Kerkmeijer LGW, Christodouleas J, Lawton Authors’ contributions CAF, Vesprini D, et al. Magnetic resonance imaging-guided adaptive LN extracted technical plan parameters, performed data analysis and drafted radiation therapy: a “game changer” for prostate treatment? Int J Radiat the manuscript. VM helped with data analysis. CE, JB, PT and RB performed Oncol Biol Phys. 2018;100:361–73. https:// doi. org/ 10. 1016/j. ijrobp. 2017. extraction of dose-related data. GL and SC supervised the study. SC designed 10. 020. the study. CE., MN, VM, GL, CK, MR, CB and SC reviewed the manuscript. All 8. Ahunbay EE, Peng C, Chen G-P, Narayanan S, Yu C, Lawton C, Li XA. An authors read and approved the final manuscript. on-line replanning scheme for interfractional variations. Med Phys. 2008;35:3607–15. https:// doi. org/ 10. 1118/1. 29524 43. Funding 9. Ahunbay EE, Peng C, Holmes S, Godley A, Lawton C, Li XA. Online adap- The Department of Radiation Oncology of the University Hospital, LMU tive replanning method for prostate radiotherapy. Int J Radiat Oncol Biol Munich has received research grants from ViewRay Inc. (Oakwood Village, OH, Phys. 2010;77:1561–72. https:// doi. org/ 10. 1016/j. ijrobp. 2009. 10. 013. USA). ViewRay was not involved and had no influence on the study design, 10. Foroudi F, Wong J, Kron T, Rolfo A, Haworth A, Roxby P, et al. Online adap- the collection or analysis of data, on the writing of the manuscript, or the deci- tive radiotherapy for muscle-invasive bladder cancer: results of a pilot sion to submit the manuscript for publication. CE received funding from the study. Int J Radiat Oncol Biol Phys. 2011;81:765–71. https:// doi. org/ 10. German Cancer Aid [Mildred Scheel Stipendienprogramm für Krebsforschung, 1016/j. ijrobp. 2010. 06. 061. 2018 (57468956)]. 11. Li Y, Hoisak JDP, Li N, Jiang C, Tian Z, Gautier Q, et al. Dosimetric benefit of adaptive re-planning in pancreatic cancer stereotactic body radiotherapy. Availability of data and materials Med Dosim. 2015;40:318–24. https:// doi. org/ 10. 1016/j. meddos. 2015. 04. Not applicable. 12. Mohan R, Zhang X, Wang H, Kang Y, Wang X, Liu H, et al. Use of deformed Code availability intensity distributions for on-line modification of image-guided IMRT to Not applicable. account for interfractional anatomic changes. Int J Radiat Oncol Biol Phys. 2005;61:1258–66. https:// doi. org/ 10. 1016/j. ijrobp. 2004. 11. 033. 13. Castelli J, Simon A, Louvel G, Henry O, Chajon E, Nassef M, et al. Impact of Declarations head and neck cancer adaptive radiotherapy to spare the parotid glands and decrease the risk of xerostomia. Radiat Oncol. 2015;10:6. https:// doi. Ethics approval and consent to participate org/ 10. 1186/ s13014- 014- 0318-z. This study was conducted according to the guidelines of the Declaration of 14. Guckenberger M, Wilbert J, Richter A, Baier K, Flentje M. Potential of Helsinki, and approved by the Ethics Committee of LMU (LMU 20-291, 13 May adaptive radiotherapy to escalate the radiation dose in combined 2020). radiochemotherapy for locally advanced non-small cell lung cancer. Int J Radiat Oncol Biol Phys. 2011;79:901–8. https:// doi. org/ 10. 1016/j. ijrobp. Consent for publication 2010. 04. 050. Not applicable. 15. Henke LE, Kashani R, Hilliard J, DeWees TA, Curcuru A, Przybysz D, et al. In silico trial of MR-guided midtreatment adaptive planning for hypof- Competing interests ractionated stereotactic radiation therapy in centrally located thoracic SC received speaker fees/travel support from ViewRay Inc. (Oakwood Village, tumors. Int J Radiat Oncol Biol Phys. 2018;102:987–95. https:// doi. org/ 10. OH, USA). All other authors declare that they have no competing interests. 1016/j. ijrobp. 2018. 06. 022. 16. de Jong R, Visser J, Crama KF, van Wieringen N, Wiersma J, Geijsen ED, Author details Bel A. Dosimetric benefit of an adaptive treatment by means of plan Department of Radiation Oncology, University Hospital, LMU Munich, Mar- selection for rectal cancer patients in both short and long course chioninistr. 15, 81377 Munich, Germany. German Cancer Consortium (DKTK), radiation therapy. Radiat Oncol. 2020;15:13. https:// doi. org/ 10. 1186/ Partner Site Munich, Munich, Germany. s13014- 020- 1461-3. 17. Mohamed ASR, Bahig H, Aristophanous M, Blanchard P, Kamal M, Ding Y, Received: 23 December 2021 Accepted: 27 February 2022 et al. Prospective in silico study of the feasibility and dosimetric advan- tages of MRI-guided dose adaptation for human papillomavirus positive N ierer et al. Radiation Oncology (2022) 17:53 Page 13 of 14 oropharyngeal cancer patients compared with standard IMRT. Clin Transl 34. Menten MJ, Wetscherek A, Fast MF. MRI-guided lung SBRT: present and Radiat Oncol. 2018;11:11–8. https:// doi. org/ 10. 1016/j. ctro. 2018. 04. 005. future developments. Phys Med. 2017;44:139–49. https:// doi. org/ 10. 18. Møller DS, Holt MI, Alber M, Tvilum M, Khalil AA, Knap MM, Hoffmann L. 1016/j. ejmp. 2017. 02. 003. Adaptive radiotherapy for advanced lung cancer ensures target coverage 35. El-Bared N, Portelance L, Spieler BO, Kwon D, Padgett KR, Brown KM, Mel- and decreases lung dose. Radiother Oncol. 2016;121:32–8. https:// doi. lon EA. Dosimetric benefits and practical pitfalls of daily online adaptive org/ 10. 1016/j. radonc. 2016. 08. 019. MRI-guided stereotactic radiation therapy for pancreatic cancer. Pract 19. Nijkamp J, Pos FJ, Nuver TT, de Jong R, Remeijer P, Sonke J-J, Lebesque JV. Radiat Oncol. 2019;9:e46–54. https:// doi. org/ 10. 1016/j. prro. 2018. 08. 010. Adaptive radiotherapy for prostate cancer using kilovoltage cone-beam 36. Hassanzadeh C, Rudra S, Bommireddy A, Hawkins WG, Wang-Gillam A, computed tomography: first clinical results. Int J Radiat Oncol Biol Phys. Fields RC, et al. Ablative five-fraction stereotactic body radiation therapy 2008;70:75–82. https:// doi. org/ 10. 1016/j. ijrobp. 2007. 05. 046. for inoperable pancreatic cancer using online MR-guided adaptation. 20. Oh S, Stewart J, Moseley J, Kelly V, Lim K, Xie J, et al. Hybrid adaptive Adv Radiat Oncol. 2020. https:// doi. org/ 10. 1016/j. adro. 2020. 06. 010. radiotherapy with on-line MRI in cervix cancer IMRT. Radiother Oncol. 37. Placidi L, Romano A, Chiloiro G, Cusumano D, Boldrini L, Cellini F, et al. 2014;110:323–8. https:// doi. org/ 10. 1016/j. radonc. 2013. 11. 006. On-line adaptive MR guided radiotherapy for locally advanced pancreatic 21. van de Schoot AJAJ, de Boer P, Visser J, Stalpers LJA, Rasch CRN, Bel A. cancer: clinical and dosimetric considerations. Techn Innov Patient Sup- Dosimetric advantages of a clinical daily adaptive plan selection strategy port Radiat Oncol. 2020;15:15–21. https:// doi. org/ 10. 1016/j. tipsro. 2020. 06. compared with a non-adaptive strategy in cervical cancer radiation 001. therapy. Acta Oncol. 2017;56:667–74. https:// doi. org/ 10. 1080/ 02841 86X. 38. Rudra S, Jiang N, Rosenberg SA, Olsen JR, Roach MC, Wan L, et al. Using 2017. 12879 49. adaptive magnetic resonance image-guided radiation therapy for treat- 22. Yoon SW, Lin H, Alonso-Basanta M, Anderson N, Apinorasethkul O, ment of inoperable pancreatic cancer. Cancer Med. 2019;8:2123–32. Cooper K, et al. Initial evaluation of a novel cone-beam CT-based semi-https:// doi. org/ 10. 1002/ cam4. 2100. automated online adaptive radiotherapy system for head and neck can- 39. Kerkhof EM, Raaymakers BW, van der Heide UA, van de Bunt L, cer treatment—a timing and automation quality study. Cureus. 2020;12: Jürgenliemk-Schulz IM, Lagendijk JJW. Online MRI guidance for healthy e9660. https:// doi. org/ 10. 7759/ cureus. 9660. tissue sparing in patients with cervical cancer: an IMRT planning study. 23. Inc., Viewray. First patients treated with ViewRay’s MRIdian Linac System Radiother Oncol. 2008;88:241–9. https:// doi. org/ 10. 1016/j. radonc. 2008. at Henry Ford Health System. 2017.04. 009. 24. Corradini S, Alongi F, Andratschke N, Belka C, Boldrini L, Cellini F, et al. 40. Henke L, Kashani R, Robinson C, Curcuru A, DeWees T, Bradley J, et al. MR-guidance in clinical reality: current treatment challenges and Phase I trial of stereotactic MR-guided online adaptive radiation therapy future perspectives. Radiat Oncol. 2019;14:92. https:// doi. org/ 10. 1186/ (SMART ) for the treatment of oligometastatic or unresectable primary s13014- 019- 1308-y. malignancies of the abdomen. Radiother Oncol. 2018;126:519–26. 25. Acharya S, Fischer-Valuck BW, Kashani R, Parikh P, Yang D, Zhao T, et al. https:// doi. org/ 10. 1016/j. radonc. 2017. 11. 032. Online magnetic resonance image guided adaptive radiation therapy: 41. Henke LE, Olsen JR, Contreras JA, Curcuru A, DeWees TA, Green OL, et al. first clinical applications. Int J Radiat Oncol Biol Phys. 2016;94:394–403. Stereotactic MR-guided online adaptive radiation therapy (SMART ) for https:// doi. org/ 10. 1016/j. ijrobp. 2015. 10. 015. ultracentral thorax malignancies: results of a phase 1 trial. Adv Radiat 26. Fischer-Valuck BW, Henke L, Green O, Kashani R, Acharya S, Bradley Oncol. 2019;4:201–9. https:// doi. org/ 10. 1016/j. adro. 2018. 10. 003. JD, et al. Two-and-a-half-year clinical experience with the world’s first 42. Mittauer KE, Hill PM, Geurts MW, de Costa A-M, Kimple RJ, Bassetti MF, magnetic resonance image guided radiation therapy system. Adv Radiat Bayouth JE. STAT-ART: the promise and practice of a rapid palliative single Oncol. 2017;2:485–93. https:// doi. org/ 10. 1016/j. adro. 2017. 05. 006. session of MR-guided online adaptive radiotherapy (ART ). Front Oncol. 27. Güngör G, Serbez İ, Temur B, Gür G, Kayalılar N, Mustafayev TZ, et al. Time 2019;9:1013. https:// doi. org/ 10. 3389/ fonc. 2019. 01013. analysis of online adaptive magnetic resonance-guided radiation therapy 43. Padgett KR, Simpson G, Asher D, Portelance L, Bossart E, Dogan N. Assess- workflow according to anatomical sites. Pract Radiat Oncol. 2020. https:// ment of online adaptive MR-guided stereotactic body radiotherapy of doi. org/ 10. 1016/j. prro. 2020. 07. 003. liver cancers. Phys Med. 2020;77:54–63. https:// doi. org/ 10. 1016/j. ejmp. 28. Lagerwaard F, Bohoudi O, Tetar S, Admiraal MA, Rosario TS, Bruynzeel A. 2020. 07. 027. Combined inter- and intrafractional plan adaptation using fraction par- 44. Palacios MA, Bohoudi O, Bruynzeel AM, van Sörsen de Koste JR, Cobussen titioning in magnetic resonance-guided radiotherapy delivery. Cureus. P, Slotman BJ, et al. Role of daily plan adaptation in MR-guided stereotac- 2018;10: e2434. https:// doi. org/ 10. 7759/ cureus. 2434. tic ablative radiation therapy for adrenal metastases. Int J Radiat Oncol 29. Paulson ES, Ahunbay E, Chen X, Mickevicius NJ, Chen G-P, Schultz C, et al. Biol Phys. 2018;102:426–33. https:// doi. org/ 10. 1016/j. ijrobp. 2018. 06. 002. 4D-MRI driven MR-guided online adaptive radiotherapy for abdominal 45. Vestergaard A, Hafeez S, Muren LP, Nill S, Høyer M, Hansen VN, et al. The stereotactic body radiation therapy on a high field MR-Linac: implemen- potential of MRI-guided online adaptive re-optimisation in radiotherapy tation and initial clinical experience. Clin Transl Radiat Oncol. 2020;23:72– of urinary bladder cancer. Radiother Oncol. 2016;118:154–9. https:// doi. 9. https:// doi. org/ 10. 1016/j. ctro. 2020. 05. 002.org/ 10. 1016/j. radonc. 2015. 11. 003. 30. Raaymakers BW, Jürgenliemk-Schulz IM, Bol GH, Glitzner M, Kotte ANTJ, 46. van Timmeren JE, Chamberlain M, Krayenbuehl J, Wilke L, Ehrbar S, Bogo- van Asselen B, et al. First patients treated with a 1.5 T MRI-Linac: clinical wicz M, et al. Treatment plan quality during online adaptive re-planning. proof of concept of a high-precision, high-field MRI guided radiotherapy Radiat Oncol. 2020;15:203. https:// doi. org/ 10. 1186/ s13014- 020- 01641-0. treatment. Phys Med Biol. 2017;62:L41–50. https:// doi. org/ 10. 1088/ 1361- 47. Klüter S. Technical design and concept of a 0.35 T MR-Linac. Clin Transl 6560/ aa9517. Radiat Oncol. 2019;18:98–101. https:// doi. org/ 10. 1016/j. ctro. 2019. 04. 007. 31. Tetar SU, Bruynzeel AM, Lagerwaard FJ, Slotman BJ, Bohoudi O, Palacios 48. Ahunbay EE, Li XA. Gradient maintenance: a new algorithm for fast online MA. Clinical implementation of magnetic resonance imaging guided replanning. Med Phys. 2015;42:2863–76. https:// doi. org/ 10. 1118/1. 49198 adaptive radiotherapy for localized prostate cancer. Phys Imaging Radiat 47. Oncol. 2019;9:69–76. https:// doi. org/ 10. 1016/j. phro. 2019. 02. 002. 49. Andratschke N, Parys A, Stadtfeld S, Wurster S, Huttenlocher S, Imhoff D, 32. Bruynzeel AME, Tetar SU, Oei SS, Senan S, Haasbeek CJA, Spoelstra FOB, et al. Clinical results of mean GTV dose optimized robotic guided SBRT et al. A prospective single-arm phase II study of stereotactic magnetic- for liver metastases. Radiat Oncol. 2016;11:74. https:// doi. org/ 10. 1186/ resonance-guided adaptive radiotherapy for prostate cancer: early toxic-s13014- 016- 0652-4. ity results. Int J Radiat Oncol Biol Phys. 2019. https:// doi. org/ 10. 1016/j. 50. Baumann R, Chan MKH, Pyschny F, Stera S, Malzkuhn B, Wurster S, et al. ijrobp. 2019. 08. 007. Clinical results of mean GTV dose optimized robotic-guided stereotactic 33. Nicosia L, Sicignano G, Rigo M, Figlia V, Cuccia F, de Simone A, et al. Daily body radiation therapy for lung tumors. Front Oncol. 2018;8:171. https:// dosimetric variation between image-guided volumetric modulated arc doi. org/ 10. 3389/ fonc. 2018. 00171. radiotherapy and MR-guided daily adaptive radiotherapy for prostate 51. Finazzi T, Palacios MA, Haasbeek CJA, Admiraal MA, Spoelstra FOB, cancer stereotactic body radiotherapy. Acta Oncol. 2020;60:1–7. https:// Bruynzeel AME, et al. Stereotactic MR-guided adaptive radiation therapy doi. org/ 10. 1080/ 02841 86X. 2020. 18210 90. for peripheral lung tumors. Radiother Oncol. 2020;144:46–52. https:// doi. org/ 10. 1016/j. radonc. 2019. 10. 013. Nierer et al. Radiation Oncology (2022) 17:53 Page 14 of 14 52. Regnery S, Buchele C, Weykamp F, Pohl M, Hoegen P, Eichkorn T, et al. Adaptive MR-guided stereotactic radiotherapy is beneficial for ablative treatment of lung tumors in high-risk locations. Front Oncol. 2022. https:// doi. org/ 10. 3389/ fonc. 2021. 757031. 53. Mayinger M, Ludwig R, Christ SM, Dal Bello R, Ryu A, Weitkamp N, et al. Benefit of replanning in MR-guided online adaptive radiation therapy in the treatment of liver metastasis. Radiat Oncol. 2021;16:84. https:// doi. org/ 10. 1186/ s13014- 021- 01813-6. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in pub- lished maps and institutional affiliations. Re Read ady y to to submit y submit your our re researc search h ? Choose BMC and benefit fr ? Choose BMC and benefit from om: : fast, convenient online submission thorough peer review by experienced researchers in your field rapid publication on acceptance support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year At BMC, research is always in progress. Learn more biomedcentral.com/submissions

Journal

Radiation OncologySpringer Journals

Published: Mar 12, 2022

Keywords: Online MRI guided radiotherapy; Plan adaption; MRgOART; Online adaptive RT; MR-guided RT

References