Identifying best practices for RTTs and dosimetrists: a Delphi consensus study on standardizing workflows in proton therapy Authors: J.J. Jacobs, E. van Weerd, A.M. Moerman, Y.L.B. Klaver, C. Xavier, M. Høyer, E.G.C. Troost, V. Nguyen, L. Jørgensen, M. Zorneth, S. Kelly, J. Thiele, N. Bizzocchi, H. Bijwaard, M.S. Hoogeman Published: 2026 ABSTRACT Highlights•Experts identified 24 best practices for RTTs across seven workflows.•Specialized training for RTTs in proton therapy is needed.•RTTs are capable of independent decision-making using protocols.•Multidisciplinary involvement in guideline development is essential. Introduction Proton therapy (PT) has become increasingly available across Europe, with more than 25 operational centers, 18 of which opened in the last decade [1]. However, with this substantial growth of PT, variations in practices across centers and countries have been observed [2], [3], [4], [5]. These variations may negatively affect clinical outcomes and patient experience. Additionally, they may limit the generalizability of results of clinical studies across centers and hamper combined analyses of results. Introduction of guidelines can reduce these variations, and studies have emphasized the need for consensus guidelines in PT in order to facilitate this and improve clinical practice [4], [5]. A recently conducted scoping review identified guidelines and recommendations from the past decade related to workflows performed by radiation therapists (RTTs)1 in PT [6]. The review revealed major gaps in existing guidelines, including the absence of RTT-focused guidelines and a lack of detailed guidance on operational procedures for RTT-specific tasks in clinical practice [5], [7], [8], highlighting the need for further development of standardized practices. This is particularly important because RTTs play a crucial role in the multidisciplinary treatment through the preparation and delivery of radiation treatments. A high level of their expertise and technical skills is pivotal in the safe and effective delivery of PT [9], [10]. To further develop standardized practices, a Delphi consensus study was performed to identify best practices for workflows in PT in which RTTs play a leading role. Results of the scoping review and a two-day RTT workshop served as input for this study [6]. This article presents the results of the consensus study, with the ultimate goal of establishing guidelines, as part of the Erasmus+ “Towards a Sustainable RTT Network” (TaSeRnet) project, to harmonize and improve the quality of PT practices in Europe [11]. Methods and materials A modified Delphi consensus methodology was employed involving RTTs, hereafter referred to as experts, in three rounds of online questionnaires [12], [13]. The overall process, including the preparation and conclusion phases, is illustrated in Fig. 1.Fig. 1Schematic overview of the overall process and timeline.Fig. 1 Schematic overview of the overall process and timeline. Expert selection Potential experts were identified from proton therapy centers (PTCs) meeting the following criteria: 1) at least one operational gantry for treating patients, 2) the use of pencil beam scanning technique, 3) in-room imaging available for setup and verification, 4) PTC > 1 year of clinical operation (e.g., treating patients), and 5) institutes located in Europe. All PTCs meeting the inclusion criteria were invited to participate in the Delphi study, using a European Particle Therapy Network (EPTN) mailing list, personal contacts, and contact details provided on institutional websites. Information about the TaSeRnet project, as well as the Delphi study procedure and timeline was provided. Contacts were encouraged to share this information within their network to maximize inclusion of potential experts. Participants were invited to complete a questionnaire providing details about their working experience in radiotherapy and PT, country, institute, fields, and level of expertise (novice, intermediate, advanced). No predefined criteria were used for assigning this level; experts self-assigned their level. The selection of participants was left to the discretion of each center, based on availability and interest, to ensure practical feasibility. The goal was to include an equal number of experts from each institution. Initially, a maximum of five participants per institute was planned; however, this was reduced to three following review of the registrations. To ensure balance, experts from overrepresented institutions were excluded, while underrepresented centers were encouraged to recruit additional experts. Results from the scoping review, along with additional information on the Delphi consensus method, were provided as background to the experts concurrently with the distribution of the first questionnaire containing the statements. Preparation of statements The scoping review was performed to identify knowledge gaps in guidelines related to seven predefined workflows performed by RTTs working in PT [6]. These included: 1) patient simulation, 2) image processing, 3) treatment planning (TP), 4) setup and verification, 5) quality assurance (QA), 6) treatment execution and 7) evaluation during treatment course and were obtained from the TaSeRnet project proposal. Additionally, a two-day RTT workshop was organized to explore opportunities for harmonization and identify topics within these seven workflows for potential best practices. In preparation for the workshop, a questionnaire was distributed to RTTs working in European PTCs to gather input on potential topics for best practices (supplementary material). During the workshop, 32 RTTs from nine European PTCs (The Netherlands, United Kingdom, Switzerland, Germany, Norway, and Denmark), and one educational institute from The Netherlands participated in group discussions to identify potential topics for best practices within each workflow. To enhance interactivity, participants were divided into smaller groups. A carousel method was employed to ensure that each participant could provide input on all seven workflows. The topics were ranked and presented at the end of the workshop. Statements were developed by JJ, EW, MiH, AM, CX, and YK using a stepwise approach informed by the workshop outcomes and scoping review results. During a structured meeting, statements were generated and refined through collective discussion and continuous alignment with predefined objectives from the TaSeRnet proposal. The workshop output, prioritized topics for best practices, was compared with these objectives and scoping review findings. Particular emphasis was placed on their applicability to the professional practice of RTTs with a specific focus on PT. Based on this process, selected topics were converted into statements. The statements were subsequently reviewed for clarity, neutrality, and consistency by VN, LJ, MZ, SK, JT, and NB. The final list of 26 statements was approved by a multidisciplinary review panel (MoH, ET, HB, MiH). After each round, responses were summarized and anonymized, and statements not reaching consensus were revised as necessary. Revised statements were reviewed in the same manner during subsequent rounds. Formulators of the statements did not complete the questionnaires. One reviewer also participated in the Delphi study. Consensus process A 5-point Likert scale (strongly disagree, disagree, neutral, agree, strongly agree) was used to score responses. A “no opinion” option was included to prevent experts from being forced to express an opinion on a topic outside their expertise as well as a neutral option for those neither agreeing nor disagreeing. “No opinion” votes were excluded from the consensus analysis. Additionally, experts could comment on each statement, to share concerns, suggestions, or potential improvements. Prior to the start of the study, consensus (C) was defined as > 66% agreement or disagreement (including strong agreement or disagreement), with strong consensus (SC) as ≥ 80% [12], [14]. In each round, experts were asked to submit their responses within 10 working days. After Round 1, statements without consensus were shared in Round 2, along with the Round 1 percentages. Statements on which consensus was reached were not included in the second round. For the final round, the remaining statements, along with the panel’s agreement level and anonymized Round 2 comments, were shared with the experts to provide additional insight into experts’ considerations. Results Forty experts from 15 PTCs in Europe (Table 1) participated in the Delphi study, including 35 RTTs, three oncology nurses specialized in radiotherapy and two dosimetrists. Of the 40 experts, 36 provided more detailed information regarding years of expertise (median years of experience: radiotherapy 8.5 (range 3–19); proton therapy 5 (range 3–6)) and level of expertise (Fig. 2). Response rates were 40/40 (100%), 35/40 (87.50%) and 37/40 (92.50%) for the three rounds, respectively. Statements from each round are available in supplementary material.Table 1Distribution of experts by country, with the numbers in brackets indicating the number of centers per country.CountryNumber of participantsAustria (1)3Belgium (1)3Denmark (1)1France (2)6Germany (2)6Italy (1)3Sweden (1)2Switzerland (1)2The Netherlands (3)9United Kingdom (2)5 Table 1Distribution of experts by country, with the numbers in brackets indicating the number of centers per country.CountryNumber of participantsAustria (1)3Belgium (1)3Denmark (1)1France (2)6Germany (2)6Italy (1)3Sweden (1)2Switzerland (1)2The Netherlands (3)9United Kingdom (2)5 Distribution of experts by country, with the numbers in brackets indicating the number of centers per country. In total, 24 out of 26 statements reached consensus after three rounds. The following sections present the results, including references to the corresponding statement numbers listed in Table 2. Table 2 further provides full statement descriptions, workflow category, percentages and final results. Detailed response distributions are available in supplementary material.Table 2An overview of the results after three Delphi rounds, including the statement number, workflow domain, and final consensus. Disagreement includes 'strongly disagree' and 'disagree', while agreement includes 'strongly agree' and 'agree'. * = Consensus is derived from disagreement with a negative statement.Statement numberStatementWorkflowDisagree (%)Neutral (%)Agree (%)Final consensus1There should be a pre scanning protocol for checking CT scan parameters and treating physician instructions by RTTs.Patient simulation11%8%81%Strong consensus2After scanning, the CT-scan should be checked by an RTT according to a post scanning protocol on quality, including (motion) artefacts and image contrast and patient specific instructions (e.g. bladder filling), prior to image processing.Patient simulation9%12%79%Consensus3Because of the physics characteristics of protons, optimal patient positioning is of utmost importance. Therefore, to ensure optimal positioning, a re-fit of immobilization devices should be performed right before the acquisition of the CT scan.Patient simulation10%21%69%Consensus4Based on a protocolized CT quality check, RTTs should be able to independently decide whether a re-scan is necessary.Patient simulation12%6%82%Strong consensus5Before the first treatment fraction, the practical feasibility of every treatment plan (regardless of the complexity), should be tested in a dry-run, without a patient present. This minimally includes collision check and feasibility of table rotations and chosen table inserts.Patient simulation62%8%30%No consensus6All patients should undergo a dry-run, as a separate appointment, before the first treatment fraction, to achieve optimal patient preparation and cooperation. This minimally includes setting up the patient on the treatment couch, checking gantry and table rotation of the treatment plan and feasibility of image acquisition.Patient simulation86%0%14%Strong consensus7For treatment sites where surface guidance is used for patient setup in proton therapy, the use of tattoos becomes obsolete.Patient simulation17%14%69%Consensus8An RTT or dosimetrist does not need specific knowledge on how an AI system works, when using AI toolboxes for auto-segmentation.*Image processing76%10%14%Consensus9Regardless of the high conformality and precision in proton therapy, critical OARs could be segmented by the RTT or dosimetrist, if they are reviewed by a physician.Image processing26%0%74%Consensus10There should be standardized window level settings for delineating foreign materials for proton therapy treatment planning. We hereby assume that this is feasible with the various hardware and software configurations.Image processing4%20%76%Consensus11It is best practice to individually determine the optimal beam angles, in order to minimize the impact of range uncertainties, instead of using a fixed beam setup per treatment site.Treatment planning12%11%77%Consensus12A condition for achieving best practice in treatment planning is the availability of an international standardized list of foreign materials (e.g. for dental implants, surgical devices) with designated HU overrides.Treatment planning3%3%94%Strong consensus13An RTT and dosimetrist working in treatment planning should have a clear understanding of the impact of uncertainties on the dose distribution and how to minimize it.Treatment planning3%0%97%Strong consensus14RTTs and dosimetrists should take an active role (not responsibility) to share knowledge by organizing peer-reviews of treatment plans outside their own institute, but within the same country.Treatment planning9%20%71%Consensus15RTTs and dosimetrists should take an active role (not responsibility) to share knowledge by organizing peer-reviews of treatment plans outside their own institute, but within Europe.Treatment planning8%31%61%No consensus16RTTs and dosimetrists should take an active role in sharing knowledge by organizing interdisciplinary reviews of treatment plans within their institute.Treatment planning3%3%95%Strong consensus17Specific training in communication for RTTs working with children (patients under 18 years, treated with and without anesthesia) is necessary to accomplish an effective set-up and positioning.Setup & verification12%15%73%Consensus18In IGPT the automatic match is considered best; RTTs should not manually adjust the registration after the automatic match.*Setup & verification91%6%3%Strong consensus19Independent decision making by RTTs, using an RTT-only traffic light protocol, is a prerequisite for an efficient workflow on the gantry. Such a protocol should at a minimum, define action levels and outline the corresponding actions. A traffic light protocol is an action level-based protocol that help RTTs make decisions in response to anatomical changes.Setup & verification3%3%94%Strong consensus20RTTs should be able to decide independently (without consultation of a medical physicist or radiation oncologist) to rescan a CBCT, based on a protocol balancing ALARA and treatment accuracy.Setup & verification16%8%76%Consensus21There should be a standardized protocol for RTTs or dosimetrists checking the treatment plan which should include checks on HU overrides and plan robustness. The check in this statement refers to the RTT/dosimetrist check on the treatment plan. It assumes that the medical physicist will carry out the final check on the treatment plan.Quality assurance17%10%73%Consensus22An RTT responsible for the treatment execution should be knowledgeable of the impact of uncertainties on the dose distribution.Treatment execution0%0%100%Strong consensus23The details on patient positioning (such as the use of help structures, mask fitting and shoulder position) should be recorded and transferred in a standardized fashion for the treatment on the gantry.Treatment execution5%3%93%Strong consensus24The RTT has an active role in monitoring the overall treatment course and identifying the need for plan evaluation.Evaluation during treatment course6%6%89%Strong consensus25For specific tumor sites, RTTs or dosimetrists should be able to independently execute and evaluate a recalculation on a repeat-CT, and take appropriate actions, based on a traffic light protocol. Such a protocol should at a minimum, define action levels and outline the corresponding actions.Evaluation during treatment course10%17%73%Consensus26RTTs should be involved in meetings evaluating decisions regarding plan adaptation.Evaluation during treatment course6%8%86%Strong consensus Table 2An overview of the results after three Delphi rounds, including the statement number, workflow domain, and final consensus. Disagreement includes 'strongly disagree' and 'disagree', while agreement includes 'strongly agree' and 'agree'. * = Consensus is derived from disagreement with a negative statement.Statement numberStatementWorkflowDisagree (%)Neutral (%)Agree (%)Final consensus1There should be a pre scanning protocol for checking CT scan parameters and treating physician instructions by RTTs.Patient simulation11%8%81%Strong consensus2After scanning, the CT-scan should be checked by an RTT according to a post scanning protocol on quality, including (motion) artefacts and image contrast and patient specific instructions (e.g. bladder filling), prior to image processing.Patient simulation9%12%79%Consensus3Because of the physics characteristics of protons, optimal patient positioning is of utmost importance. Therefore, to ensure optimal positioning, a re-fit of immobilization devices should be performed right before the acquisition of the CT scan.Patient simulation10%21%69%Consensus4Based on a protocolized CT quality check, RTTs should be able to independently decide whether a re-scan is necessary.Patient simulation12%6%82%Strong consensus5Before the first treatment fraction, the practical feasibility of every treatment plan (regardless of the complexity), should be tested in a dry-run, without a patient present. This minimally includes collision check and feasibility of table rotations and chosen table inserts.Patient simulation62%8%30%No consensus6All patients should undergo a dry-run, as a separate appointment, before the first treatment fraction, to achieve optimal patient preparation and cooperation. This minimally includes setting up the patient on the treatment couch, checking gantry and table rotation of the treatment plan and feasibility of image acquisition.Patient simulation86%0%14%Strong consensus7For treatment sites where surface guidance is used for patient setup in proton therapy, the use of tattoos becomes obsolete.Patient simulation17%14%69%Consensus8An RTT or dosimetrist does not need specific knowledge on how an AI system works, when using AI toolboxes for auto-segmentation.*Image processing76%10%14%Consensus9Regardless of the high conformality and precision in proton therapy, critical OARs could be segmented by the RTT or dosimetrist, if they are reviewed by a physician.Image processing26%0%74%Consensus10There should be standardized window level settings for delineating foreign materials for proton therapy treatment planning. We hereby assume that this is feasible with the various hardware and software configurations.Image processing4%20%76%Consensus11It is best practice to individually determine the optimal beam angles, in order to minimize the impact of range uncertainties, instead of using a fixed beam setup per treatment site.Treatment planning12%11%77%Consensus12A condition for achieving best practice in treatment planning is the availability of an international standardized list of foreign materials (e.g. for dental implants, surgical devices) with designated HU overrides.Treatment planning3%3%94%Strong consensus13An RTT and dosimetrist working in treatment planning should have a clear understanding of the impact of uncertainties on the dose distribution and how to minimize it.Treatment planning3%0%97%Strong consensus14RTTs and dosimetrists should take an active role (not responsibility) to share knowledge by organizing peer-reviews of treatment plans outside their own institute, but within the same country.Treatment planning9%20%71%Consensus15RTTs and dosimetrists should take an active role (not responsibility) to share knowledge by organizing peer-reviews of treatment plans outside their own institute, but within Europe.Treatment planning8%31%61%No consensus16RTTs and dosimetrists should take an active role in sharing knowledge by organizing interdisciplinary reviews of treatment plans within their institute.Treatment planning3%3%95%Strong consensus17Specific training in communication for RTTs working with children (patients under 18 years, treated with and without anesthesia) is necessary to accomplish an effective set-up and positioning.Setup & verification12%15%73%Consensus18In IGPT the automatic match is considered best; RTTs should not manually adjust the registration after the automatic match.*Setup & verification91%6%3%Strong consensus19Independent decision making by RTTs, using an RTT-only traffic light protocol, is a prerequisite for an efficient workflow on the gantry. Such a protocol should at a minimum, define action levels and outline the corresponding actions. A traffic light protocol is an action level-based protocol that help RTTs make decisions in response to anatomical changes.Setup & verification3%3%94%Strong consensus20RTTs should be able to decide independently (without consultation of a medical physicist or radiation oncologist) to rescan a CBCT, based on a protocol balancing ALARA and treatment accuracy.Setup & verification16%8%76%Consensus21There should be a standardized protocol for RTTs or dosimetrists checking the treatment plan which should include checks on HU overrides and plan robustness. The check in this statement refers to the RTT/dosimetrist check on the treatment plan. It assumes that the medical physicist will carry out the final check on the treatment plan.Quality assurance17%10%73%Consensus22An RTT responsible for the treatment execution should be knowledgeable of the impact of uncertainties on the dose distribution.Treatment execution0%0%100%Strong consensus23The details on patient positioning (such as the use of help structures, mask fitting and shoulder position) should be recorded and transferred in a standardized fashion for the treatment on the gantry.Treatment execution5%3%93%Strong consensus24The RTT has an active role in monitoring the overall treatment course and identifying the need for plan evaluation.Evaluation during treatment course6%6%89%Strong consensus25For specific tumor sites, RTTs or dosimetrists should be able to independently execute and evaluate a recalculation on a repeat-CT, and take appropriate actions, based on a traffic light protocol. Such a protocol should at a minimum, define action levels and outline the corresponding actions.Evaluation during treatment course10%17%73%Consensus26RTTs should be involved in meetings evaluating decisions regarding plan adaptation.Evaluation during treatment course6%8%86%Strong consensus An overview of the results after three Delphi rounds, including the statement number, workflow domain, and final consensus. Disagreement includes 'strongly disagree' and 'disagree', while agreement includes 'strongly agree' and 'agree'. * = Consensus is derived from disagreement with a negative statement. Experts agreed on the need for protocols across workflows. These include a pre-scanning protocol to check scan parameters (S1) and a post-scanning protocol to evaluate acquired images (S2). In QA, a standardized protocol for treatment plan checking is needed (S21). Furthermore, consensus was reached on the need for protocols supporting independent decision-making in patient simulation (S4), setup and verification (S19, S20) and evaluation during treatment course (S25). Consensus was reached that there is a need for specialized training for RTTs working in PT. RTTs need to understand the impact of uncertainties in treatment execution (S22) and knowledge how to minimize these uncertainties working in TP (S13). Specific communication training is necessary when working with children (S17). Furthermore, knowledge about the functioning of artificial intelligence (AI) systems when using AI toolboxes is needed (S8). To share knowledge, experts reached consensus about having an active role in organizing interdisciplinary reviews of treatment plans within their institute (S16) and organizing peer reviews of a treatment plan within their own country (S14). However, no consensus was reached on RTTs and dosimetrists taking an active role to share knowledge at the European level (S15). Regarding patient positioning, consensus highlighted performing a re-fit of immobilization devices to ensure optimal positioning (S3). There is no need for all patients to undergo a dry run as a separate appointment to achieve optimal patient preparation and cooperation (S6). On the other hand, no consensus was reached on the need for a dry-run to assess the practical feasibility of each treatment plan without a patient present (S5). Additionally, if surface guidance is used for patient setup, tattoos become obsolete (S7). Experts reached consensus on standardized workflows, including the use of an international standardized list of foreign materials (S12) and standardized window level setting for delineation (S10). Furthermore, the recording of details on patient positioning should follow a standardized format (S23). Regarding evaluation during treatment course, consensus was reached that RTTs should actively monitor the overall treatment course and identify the need for plan evaluation (S24). Furthermore, RTTs should be involved in meetings evaluating decisions regarding plan adaptation (S26). Remaining results highlight consensus on practical procedures. Specifically, RTTs could segment critical organs of interest (OOI), provided they are reviewed by a physician (S9). In TP, beam angles should be individually determined rather than using a fixed beam setup for each treatment site (S11). For image-guided proton therapy (IGPT), consensus was reached that manual adjustments after the automatic match should be made if necessary (S18). Finally, comments indicated that certain best practices within the various workflow steps, are considered as the result of multidisciplinary decision-making. Discussion To our knowledge, this is the first study presenting consensus and recommendations on best practices for RTTs working in PT. In this study, international expert consensus was reached for 24 of 26 statements after three Delphi rounds, addressing best practices within seven PT workflows in which RTTs play a leading role. However, consensus was not reached on two statements regarding patient simulation and TP. Several statements emphasize the importance of standardized protocols regarding image quality assessment for planning CTs and in TP and QA. Although this principle applies broadly to radiotherapy, the physical properties of protons make small errors in patient positioning or anatomical variations particularly impactful, potentially leading to significant dose deviations [15]. Therefore, robust protocols are especially critical in PT compared with photon therapy. Similar considerations apply to statements concerning knowledge and education, where limited training exists for RTTs working in PT. Coffey et al. [10] identified that such gaps can compromise treatment quality and limit the adoption of innovative techniques. Consistent with the findings of this study, these observations underscore the necessity of comprehensive education for RTTs in PT. In response, the TaSeRnet project will develop an advanced training module specifically targeting RTTs working in PT. Experts emphasized the importance of knowledge sharing and their role in facilitating this within their institution and nationally. However, no consensus was reached on participation in European-level peer review activities, as this was considered less feasible due to equipment differences between centers. Nevertheless, literature underlines the importance of peer review processes to achieve high-quality and consistent outcomes [16], [17]. Furthermore, Stock et al. demonstrated the potential for harmonization of TP in a European multicenter setting, despite these differences [18]. Although conducted in a multidisciplinary context, this study suggests that peer review and knowledge sharing at the European level may still be feasible, even with such variations. Although no consensus was achieved on verifying the practical feasibility of each treatment plan with a dry run without the patient present, expert feedback suggests that agreement may be achievable. Both experts who agreed and disagreed with the statement noted in their comments that this may not be necessary for standard patients, for whom protocols exist. Despite efforts to clarify this statement in consecutive rounds, these comments were observed throughout all three rounds. Further investigation is recommended to determine whether consensus can be reached on the necessity of performing a dry run without a patient present for each case. In the analysis of the results, the “no opinion” votes were excluded from the consensus analysis, which poses a potential risk of introducing bias. To explore this further, additional analysis of the “no opinion” votes was conducted to investigate whether trends existed between the voters' expertise levels and their responses to the associated statement. The analysis revealed that these responses were distributed across a range of expertise levels. This indicates that the lack of opinion was not limited to participants with novice or no level of expertise but also occurred among experts in the field. However, absolute numbers were low, making it difficult to draw firm conclusions regarding the reasons behind selecting “no opinion”. A strength of this study is the use of the Delphi methodology given its key role in developing best practices in areas with limited research [12], [13], [19]. Additionally, combining results from the scoping review, an in-person workshop, and structured multidisciplinary meetings to formulate and validate statements, ensures the validity of our results. We believe that the expert opinions in this study provide a reliable representation of European expert perspectives. Although experts had substantial radiotherapy experience, PT-specific experience was relatively limited, reflecting the fact that most PTCs have not been in clinical operation for an extended period [20], [21]. Most experts reported intermediate or advanced expertise across workflows, with a sufficient number of experts contributing to each step (Fig. 2). No consensus on a fixed number of experts has been reported; however, the literature suggests a minimum of 10–15 experts, which has been met in this study [12], [13], [19].Fig. 2Detailed information about the experts' level of expertise related to the seven workflows, with numbers indicating the absolute number of experts.Fig. 2 Detailed information about the experts' level of expertise related to the seven workflows, with numbers indicating the absolute number of experts. The fields of expertise least represented in the voting panel were TP and QA. This can be explained by the limited number of dosimetrists among the experts, and the fact that TP tasks are not a core role of RTTs in some countries [22]. In these countries, medical physicists are more closely involved in TP, and the same holds true for QA. Although these findings are relevant, we still consider our results on best practices to be reliable because statements regarding these fields of expertise are more general and are less directly related to the specific practices of TP and QA. This study has several limitations. PTCs from Spain, Poland, and the Czech Republic were not represented in this study, due to non-responses during the recruitment process. Furthermore, because of the European focus of the TaSeRnet project, consensus reached about best practices might not reflect the opinion of RTTs working in PTCs outside Europe. Differences in regulations, training, and healthcare infrastructure should be considered when interpreting the results. Although the expert panel was deliberately composed exclusively of RTTs, we recognize that this could be considered a limitation. However, a multidisciplinary team, including medical physicists and radiation oncologists, contributed to the statement development process and final set of statements, ensuring broader perspectives. Comments showed that some best practices are viewed as multidisciplinary decisions (e.g., statement 4, 9, 20, 24, 25, 26), despite being common tasks or responsibilities of RTTs [23]. Involving radiation oncologists and medical physicists in the development of best practices into guidelines increases the likelihood of their effectiveness and successful implementation. The statements are intentionally formulated in a general manner, serving as a foundation for the development of more specific guidelines for RTTs working in PT across Europe, as part of the TaSeRnet project. While the primary focus of this study was on best practices in PT, the resulting statements could potentially also apply for RTTs working with carbon ions. Regarding photon therapy, neither the scoping review nor this study considered existing guidelines in this field. The focus was instead on workflow elements that could potentially differ compared to photon therapy, such as different or additional requirements for patient positioning or image quality. This is due to the increased sensitivity of proton dose distributions to uncertainties in the beam path, for example caused by positioning devices or CT artefacts. Nevertheless, existing photon therapy literature should be reviewed and adapted if possible when developing guidelines. Based on these consensus-based best practices, future project activities include refining, validating, and implementing our results on best practices. The results from this study have guided the development of an ESTRO guideline, intended to be evidence-based and currently being developed. Furthermore, the consensus statements are also used to inform the design of an advanced course for RTTs working in PT. Conclusion This Delphi consensus study presents the first European consensus on best practices within workflows in PT, in which RTTs play a leading role. Experts reached consensus on 24 out of 26 statements. The study highlights the need for specialized training for RTTs working in PT across Europe, and RTT-only workflows based on protocols. Although statements encompass workflows where RTTs play a key role, multidisciplinary involvement is necessary to further develop these best practices into practical guidelines. The consensus reached in this study on best practices provide a solid foundation to create guidelines for RTTs working in PT across Europe. This will contribute to harmonizing and improving the quality of PT practices throughout Europe. Declaration of competing interest This research was conducted as part of the “Towards a Sustainable RTT network” (TaSeRnet) project, which is financed by the European Union under grant agreement number 2023-1-NL01-KA220-HED-000158704. The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Jenneke Jacobs reports financial support was provided by European Union. Eva van Weerd reports financial support was provided by European Union. Catarina Xavier reports financial support was provided by European Union. Morten Hoyer reports financial support was provided by European Union. Esther Troost reports financial support was provided by European Union. Lene Jorgensen reports financial support was provided by European Union. Julia Thiele reports financial support was provided by European Union. Harmen Bijwaard reports financial support was provided by European Union. Mischa Hoogeman reports financial support was provided by European Union. Yvonne Klaver declares to be part of the Varian Medical Systems FlashForward Consortium: Chair of Clinical Investigations Working Group, Member of the Scientific Committee. Esther Troost declares to be a member of the scientific advisory board of IBA. Mischa Hoogeman reports research collaborations with Accuray Inc (Sunnyvale, CA, USA), Elekta AB (Stockholm, Sweden), Varian, a Siemens Healthineers Company (Palo, Alto, CA, USA), Raysearch (Stockholm, Sweden), a test and feedback agreement with Siemens Healthineers (Erlangen, Germany), and reports participation in a ThinkTank meeting for Accuracy Inc. - all paid to the institute. If there are other authors, they declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.