Novel technologies in radiotherapy in the Nordic countries - report from the NACP2020/21 conference

As for many other conferences in the 2020-21 season, the extended travel limitations due to the COVID-19 pandemic implied that the Nordic Association for Clinical Medical Physics meeting NACP2020 had to be first postponed, and then reinvented in an online format as the NACP2020/21 conference.

It was decided to considerably shorten the program from an intended 2½ full days physical meeting down to 3 half days online format. Fortunately, this change was well received and the number of submitted abstracts at 79 was slightly larger than both the canceled 2020 physical edition (66 abstracts) and the 2017 edition [1] (70 abstracts). The online format does also allow participation where the attendee would otherwise not be available for travel for the full conference, for example due to clinical duties, but where participation in select topics of the meeting would be relevant. The prior NACP in Oslo in 2017 had 156 participants and 33 vendor representatives registered. In comparison, the virtual meeting of 2021 had more than 250 registered participants (plus some 50 company representatives) of which a cross sectional count on a set of parallel sessions saw 192 simultaneous active participants, suggesting that the online format did attract a higher active attendance than prior meetings.

The program contained a mix of live parallel sessions, prerecorded plenary presentations and virtual poster “walks”.

Keynote presenter Søren Bentzen described the resulting discussions with the following words It’s hard to match the atmosphere of a face-to-face meeting. But of all the virtual meetings, I attended over the last 15 months, I have to compliment the NACP for being the most successful in terms of creating interactive sessions that actually were interactive!

We believe that careful selection of session moderators and grouping of poster presenters with similar research topics in order to create a subspeciality meeting space of an appropriate size and friendly atmosphere was beneficial for this goal. We hope that this could also inspire future physical meetings.

The theme of the conference was “Novel medical technologies - applications, assessment and implementations”, and the content covered all three specialties of clinical medical physics; radiation oncology, radiology, and nuclear medicine physics. The three specialties were represented both in the plenary presentations and in parallel sessions throughout the three days of the conference.

The topics of submitted abstracts within radiation oncology represent current trends in the community in the Nordic countries and were generally focused on development and implementation of new technologies. They can be roughly divided under headlines, as covered in the sections below, with the addition of a section on more general medical physics topics of interest.

Artificial intelligence in radiotherapy

No conference in medical physics without discussions of machine learning and artificial intelligence. The conference kicked off on this topic with presentations of Marc Kachelriess and Mads Nielsen who both provided the audience with an overview of current and upcoming state-of-the-science [2,3] also including notes of caution with regards to appropriate uses of AI in image processing and the risks thereof.

Several attendees presented their work on AI and machine learning. One treatment site where the optimal radiotherapy approach has been debated fairly extensively is in breast cancer, in particular where the internal mammary chain needs irradiation with robustness toward anatomical changes, but still sparing the heart. Volumetric modulated arc is a challenging technique in these cases, but Engstrøm et al. [4] presented an automation of the process with robustness and faster delivery with fewer breath holds than conventional therapy. Another application with much promise to soon impact practice is the use of AI for segmentation purposes where Ren et al. [5] presented promising initial work on segmentation of the target. Lorenzen et al. present a study of inter-observer variability of normal tissue segmentations that may serve as a benchmark for automated procedures for normal tissue segmentation [6].

Another application of deep learning models from AI is for acceleration of complex processing tasks, where a network is trained from a large number of time consuming detailed calculations in order to learn a fast prediction. This was utilized by Pettersen team al to classify track quality in proton CT [7] albeit the effect on image quality was minor.

Intra- and inter-fractional motion and adaptation

The field of motion management has been a consistent component of the radiotherapy field for the past 20+ years, and continues to be a subject of new developments and research. In Damkjaer et al. [8], a simple but clever design for a novel external marker for measurement of respiratory motion is presented, featuring a very small footprint and ease of use. The magnitude and reproducibility of internal lung tumor motion was measured over the entire treatment for a large cohort of patients treated with SBRT in Nielsen et al. [9], demonstrating a high degree of tumor motion reproducibility.

Whereas the focus of research has traditionally mostly been on intra-fractional respiratory motion management, this has shifted somewhat toward management of inter-fractional variations. An example is the paper from Nielsen et al. [9] referenced above, and several additional abstracts at the conference focused solely on quantification of interfractional changes for various clinical sites. This focus is likely also related to the increased use and development of adaptive radiotherapy. MR based radiotherapy was also a key discussion point, obviously inspired by the acquisition of MR accelerators at several hospitals. This included the use of MR accelerators for treatment of moving tumors, but also use of MR only radiotherapy planning in, for example, prostate radiotherapy.

Proton therapy

As evidenced by a substantial number of abstracts related to proton therapy, this is still a hot topic reflecting the increasing use and availability of this treatment modality in the Nordic countries. A wide range of topics were covered, from technologically focused studies to investigations of the biological effects of protons. With the growing maturity of clinical proton operations, studies exhibit a move toward more focus on clinical applications as for instance compared to the NACP2017 conference. Invited speaker Antje Knopf (Cologne University Hospital) gave an overview of the challenges related to proton treatment of moving targets in the thorax, and presented both novel and practical solutions and data [10]. In Rønde et al. [11], the organ sparing potential related to the difficult case of pelvic re-irradiation in recurring rectal cancer is discussed, showing that significant organ sparing can be achieved with IMPT as compared to VMAT in the intermediate and low dose regions. The organ sparing potential of proton therapy in the treatment of pediatric brain cancers is investigated in Toussaint et al. [12], where the dose to neurovascular structures is correlated with tumor location, and a delineation atlas is evaluated. Further abstracts at the conference related further organ sparing studies, investigations of biological effect, and clinical trial design for photon-proton comparison.

General medical physics contributions

Bringing together researchers and clinicians from both nuclear medicine, imaging, and radiotherapy, broader topics of interest to all were also presented at NACP 2020/21. In particular, the keynote lectures covered topics of general interest to the medical physics community, primarily in relation to novel advanced imaging technologies and their application. Daniela Thorwarth from the University of Tübingen was invited to give a plenary talk on Functional imaging to personalize radiotherapy which highlights the interdisciplinarity of the field. An insight into the large potential of the new generation of PET scanners with total body coverage was presented by invited speaker Simon Cherry (UC Davis, CA) which was inspiring both for diagnostics and radiotherapy [13]. An update on the latest development of spectral [14], photon counting [14] and phase contrast CT [15] was given by Cynthia McCollough (Mayo Clinic, Rochester) Anais Viry (Lausanne University Hospital) and Martin Bech (Lund University).

The submitted abstracts to the conference also covered a broad scope of content of general medical physics interest. This included topics in advanced imaging and analysis, with a more specific dive into hypoxia PET and the stability requirements necessary for the putative use of hypoxia PET for radiation dose painting. Wright et al. gave her presentation in a PET/SPECT specific parallel session and has published the detailed results in this issue of Acta [16]. Dosimetry is also a field relevant across disciplines, and dosimetry focused presentations were covering aspects of e.g., dose measurements in 3 D, in preclinical protons studies, for SPECT. Small fields measurements is a general challenge in radiotherapy where inappropriate methodology unfortunately has lead to systematic mistreatments in recent time. Therefore, it is of relevance to the field to have detailed dosimetry studies of available detector systems as in the paper by Partanen et al [17] In the current volume.

Finally, a keynote lecture was dedicated to discussion of the challenges related to translation of technologies into clinical practice by Søren Bentzen (University of Maryland). Here, there was emphasis on evidence generation with discussion of methodology for complex interventions that could be used to increase the level of evidence for the modern technologies [18–20]. As is clear not least from this conference, new technologies are developed and implemented at a fast pace in current radiotherapy. It is therefore important to establish frameworks and consensus on how to evaluate the impact of technical innovations that can be used in practice.

Disclosure statement

No potential conflict of interest was reported by the author(s).