ESTRO 2024 Congress report
There is a lot to digest in upper digestive tract radiotherapy for RTTs, physicians and physicists! ESTRO 2024 in Glasgow demonstrated that there is significant progress in upper GI radiation oncology which is mainly driven by advances in technology but also in biology. Advances in strategies for more effective dose delivery, better protection of organs at risk and integration into multidisciplinary treatment strategies are predicted to increase the relative importance of upper GI radiotherapy compared to other body sites. Keeping pace with these developments is key to providing top-notch therapy for our patients.
Oesophagus and gastric tumours
On the topic of oesophagus and gastric tumours, 35 total abstracts were presented within the upper GI tract at ESTRO 2024, mainly directed towards oesophageal cancer.
Radiation therapy for oesophageal cancer plays a critical role in contemporary treatment regimens, though patient selection remains crucial due to the risk of toxicity. This was also reflected in the number of abstracts addressing the issue of toxicity, with particular interest in cardiac toxicity.
In a French retrospective cohort of 129 patients, left atrium doses of ≥15Gy were associated with the development of grade 3+ major cardiovascular events (1149). Additionally, in an 88-patient retrospective Belgian cohort, the mean dose to the left ventricle correlated with postoperative cardiac complications after voxel-based toxicity analysis and NTCP modelling (model AUC=0.72 in validation) (1945). Furthermore, an interim analysis of the prospective, observational CLARIFY study, executed on 103 oesophageal cancer and 68 lung cancer patients, showed that left ventricle global longitudinal strain decreased more after photon than after proton radiotherapy. This decrease was related to the mean heart dose, in the photon therapy group (β=-0.483, p=0.014) (779). These findings highlight the potential advantage of proton therapy in reducing toxicity for oesophageal cancer radiotherapy. However, to maintain this advantage (e.g. mean heart dose reduction of on average 1.2 Gy, p<0.001) and ensure target coverage, it seems necessary to adapt the majority (23 out of 40) of proton therapy plans throughout the treatment course (2201).
While we await results from the NRG-GI006 and PROTECT phase 3 trials, prospective data presented at ESTRO 2024 from The Netherlands already support the benefits of proton therapy in reducing toxicity (3536). Their data highlighted a significant reduction in postoperative pneumonia risk from 22% (propensity score matched historical photon cohort) to 8% proton therapy. This reduction in pulmonary complications could be due to the reduction of mean lung and heart dose, and a reduced exposure to low doses in both organs, as indicated by principal component analysis (438). Additionally, the hospital stays for patients undergoing trimodal treatment were reduced from 19 days with photons to 13 days with protons. Lastly, proton therapy for oesophageal cancer also reduces lymphopenia, which is often linked to survival outcomes. A large retrospective dataset of 1337 patients suggested that the best radiation-induced lymphopenia parameter to distinguish survival outcomes for oesophageal cancer is the absolute lymphocyte count at the beginning of week 3 of chemoradiotherapy (threshold of 0.5 x 103/μL, p<0.0001) (923).
A study explored the use of dual-tracer imaging with FAPI-46-PET/CT and FDG-PET/CT to improve radiotherapeutic management of oesophageal cancer (EC). Among 32 patients, primary tumours were detected in all FAPI-46 scans and 94% of FDG scans. The functional tumour volumes (FTVs) were generally larger with FAPI-46 and dual-tracer imaging compared to FDG alone. This imaging method led to nodal upstaging in 38% of patients and identified new metastases in 6% of cases. Consequently, radiotherapy fields were adjusted in 16% of patients and treatment regimens changed in 9%. The findings suggest that FAPI-46 PET/CT significantly impacts EC management and highlight the need for further prospective studies.
OMEC
The OligoMetastatic Esophagogastric Cancer (OMEC) project aims to establish clinical practice guidelines for the definition, diagnosis, and treatment of esophagogastric oligometastatic disease (OMD) across Europe. Developed following AGREE II and GRADE principles, these guidelines are based on systematic reviews, clinical case discussions, and a Delphi consensus study, involving 69 experts from 49 cancer centres across 16 countries. Moderate quality evidence led to weak recommendations: OMD is defined as having 1 organ with ≤3 metastases or 1 involved extra-regional lymph node station, with no disease progression after systemic therapy. 18F-FDG-PET/CT imaging is recommended for staging and restaging. Treatment varies based on the disease-free interval (DFI), with systemic therapy followed by local treatment for synchronous or short DFI cases, and upfront local treatment for longer DFI cases. These guidelines aim to standardise clinical trial criteria and reduce treatment variations. This is expected to result in a better understanding of OMEC and identifying more tailored therapeutic approaches identifying the right subgroups of patients to overcome obstacles such as recently presented at the ASCO annual meeting on the RENAISSANCE trial.
Pancreas
Though not predominant, there were 55 contributions on pancreatic cancer (PDAC) including one symposium, five proffered papers, six mini orals, six poster discussions and a large number of digital posters. Much of the work centred around stereotactic body radiation therapy (SBRT) and MR-LINAC. In an RTT symposium, a new workflow for MR-LINAC treatment (3434) was defined for PDAC making use of over 100 CT scans which were deformably registered to an MR-scan to generate an electron density map. Overall agreement of organs at risk was good providing an accuracy of <2% and for most cases of planning target volume (PTV) prescription as well and it was estimated that 80% of the patients may only need an MRI-simulation appointment without CT scans. In a similar context, two proffered papers reported on technical aspects, namely improved image quality with breath-hold CBCT for pancreatic cancer and daily organs at risk variations and the impact on dose target coverage in PDAC (2898; 1989).
From the clinical side, neoadjuvant FOLFIRINOX and chemoradiation (CRT) in borderline resectable and locally advanced PDAC (50% each) were investigated in a 65 patients phase II trial, all of whom had FDG-PET/CT prior to treatment (2954). About one-third of the patients had progressive disease and those were excluded from CRT, and 40 patients remained for CRT. All of the patients with radical resection were complete (R0). Median overall survival after resection was 28.2 months versus 13.2 months (p = 0.004). The observation of high complete resection rates is well in line with the recently reported CONKO-007 trial (Fietkau, ASCO 2023).
A very novel approach is adjuvant SBRT in patients with PDAC and high-risk features 4-6 weeks after resection. Preliminary toxicity results of the phase 2 SPARTA trial on 42 patients were reported (1954). SBRT consisted of five fractions and a dose prescription of 40 Gy to the high-risk region as an SIB and 30 Gy with further expansion of the high-risk volume. No grade 4 toxicities were observed in the preliminary analysis and grade 2 toxicity was well below 10%.
In the palliative setting, the impact of coeliac plexus radiosurgery on survival in 90 evaluable patients of a single-arm phase 2 trial was analysed post-hoc (893). Single fraction SBRT was performed in patients with severe retroperitoneal pain syndrome. A positive pain response rate of 53% has been reported earlier, and on univariate analysis positive pain response was associated with an HR of 0.63 (p = 0.04) for overall survival which was 122 days at the median (IQR 77-190 days). Only lower body mass index (BMI) was another significant factor (p = 0.03). Both factors remained significant at multivariate analysis with positive pain response showing an HR of 0.58 (p = 0.01).
Finally, the immuno-molecular modulation was analysed in 50 patients with localised PDAC and resection after neoadjuvant SBRT (765). Patients were treated with SIB to the tumour-vessel interface up to a maximum dose of 53 Gy in 0.05 mL in five fractions. Seventeen patients had a primary resection, 17 induction chemotherapy with FOLFIRINOX only and 16 additionally SBRT. RNA sequencing and immunohistochemistry were performed on paraffin-embedded tissue. Among other factors of immune modulation, at IHC, expression levels of PD-L 1 were significantly increased while PD-1 expression was significantly decreased. In conclusion, of the three groups only patients treated with SBRT were modulated towards a more favourable prognostic immunophenotype which could be therapeutically exploited.
Liver
In hepatocellular carcinoma (HCC) there were 21 contributions including one symposium, two preferred papers, two mini orals, three poster discussions and a large number of digital posters discussing various innovative treatments and methodologies highlighting ongoing research and clinical trials to improve outcomes for patients with advanced liver cancer.
The Phase Ib ‘Departure’ trial presented by Makishima et al. investigated the safety and efficacy of combining Durvalumab (an immune checkpoint inhibitor) with Tremelimumab and particle therapy (carbon ion radiotherapy) for advanced HCC with macrovascular invasion (MVI). The study hypothesises that this combination could provide survival benefits due to the synergistic effects of immunotherapy and high-precision radiotherapy. Initial results indicate manageable safety profiles with no dose-limiting toxicities in the early stages. Serious adverse events were more common in the combination therapy group. Preliminary data show promising progression-free survival (PFS) and overall survival (OS) rates, suggesting potential efficacy in targeting HCC with MVI.
Chan et al. evaluated the impact of severe lymphopenia, induced by SBRT, on OS in HCC patients. Data from 107 patients reveal that severe lymphopenia post-SBRT correlates with significantly poorer survival outcomes. Factors like baseline liver function, tumour stage, and radiotherapy parameters were analysed. The findings suggest that minimising low-dose radiation exposure to the liver during SBRT could preserve lymphocyte counts and improve patient survival.
In a national pattern of care analysis Deseyne et al. examined the results of SBRT for primary liver tumours and metastases across multiple centres. Analysis of 409 liver lesions treated with SBRT indicates a steady increase in its use and a correlation between higher radiation doses and better local control (LC) and OS. Factors such as tumour volume and prior systemic therapy significantly influenced outcomes. The study underscores the importance of personalised treatment planning to optimise SBRT efficacy.
A propensity score-matched cohort study by Liu et al. compared the outcomes of PBT and TACE in BCLC stage B HCC patients. Results from 275 matched patients show that PBT offers a significant survival advantage over TACE, with higher 1-year and 2-year OS rates. The findings advocate for considering PBT as a viable alternative to TACE for intermediate-stage HCC due to its superior efficacy in tumour control and survival.
Suh et al. evaluated in a prospective trial the impact of Continuous Positive Airway Pressure (CPAP) on reducing tumour motion and liver dose during radiotherapy for liver tumours. Preliminary results suggest that CPAP, compared to conventional abdominal compression, effectively minimises tumour movement and liver radiation exposure, potentially improving the precision and safety of radiotherapy for liver cancer.
Collectively these studies reflect significant strides in HCC treatment, emphasising the integration of advanced radiotherapy techniques and immunotherapy to enhance therapeutic efficacy and patient survival.
Patterns of care and outcome of liver SBRT: results from a multicentre national quality project
A multicentre national project in Belgium investigated the patterns of care and factors influencing LC and OS for patients with liver tumours treated with stereotactic body radiotherapy (SBRT). Data from 353 patients treated at 14 centres between August 2013 and December 2019 showed an increase in SBRT use and higher dose prescriptions over time. Colorectal adenocarcinoma was the most common primary tumour. After a median follow-up of 23.6 months, 1-year and 2-year OS rates were 74% and 50%, respectively. Factors positively influencing OS included better performance status (PS), smaller PTV, and higher biologically effective dose (BED10). For LC, better outcomes were associated with smaller PTV, higher BED10, and the absence of prior systemic therapy. While certain technical parameters initially appeared influential, multivariable analysis revealed that PTV volume and BED10 were the most significant factors.
ESMO-ESTRO consensus recommendations on the safety of combining targeted therapies with radiotherapy
The joint ESMO-ESTRO initiative aims to enhance cancer treatment outcomes by combining targeted therapies and immunotherapy with radiotherapy (RT) while addressing the potential increase in toxicity. Given the scarcity of high-quality toxicity data and the absence of international guidelines, the initiative conducted systematic literature reviews and involved 60 international experts in a modified Delphi consensus process. The reviews covered the safety of combining common systemic therapies with RT, resulting in a database of 376 analysed articles. The first Delphi round achieved significant consensus on most statements, with further review planned for unresolved items. The final recommendations will guide clinicians in making informed decisions and minimising the risk of undertreatment and unexpected toxicity.
RTT
RTT skills development and role extension were a common thread throughout ESTRO 2024. For example, the implementation of an RTT-led online OAR re-delineation training programme for pancreatic adaptive radiotherapy (ART) on the MR-Linac using a multi-stage mix of didactic and practical training (1344). To date, one RTT (‘trainer RTT’) has completed the online stage of the programme, three the initial offline stage, with 7# for 12 patients having RTT-led OAR red-delineation. Building on this, the implementation of a dedicated advanced practice RTT role significantly reduced contouring times for online abdominal ART in comparison to a clinician, at <10 mins vs >20 mins (1355).
With regard to IGRT developments, an assessment of multimodality imaging for pancreatic radiotherapy planning shows that the addition of MRI does not automatically improve target volume delineation consistency and that MRI sequence optimisation as well as a robust training programme is required (3105). Targets delineated on MRI were smaller than those on breath hold CT, which can lead to underestimation of the true target volume size. MRI sequence optimisation as well as a robust training programme were identified as potential solutions for this issue.
Work evaluating CBCT image quality in breath hold for six pancreatic cancer patients was found to be markedly improved on visual assessment in comparison to free-breathing CBCT (2898). Another study of 16 upper abdominal radiotherapy patients including the pancreas found that patient factors, including male sex, cirrhosis and high BMI were found to be independent risk factors for increased free-breathing respiratory motion in using cine MRI (968). The mean craniocaudal motion was 19.56 mm and 12.55 mm for males and females, respectively. A progressive increase in craniocaudal motion was observed among patients with normal, overweight, and obese BMI, with measurements of 12.41mm, 19.44mm, and 23.46mm, respectively. A similar pattern was observed for mediolateral motion. This suggests that for patients with these risk factors who cannot tolerate one motion management strategy (i.e., breath hold) for upper abdominal SBRT, an alternative should be offered instead of treatment in free-breathing with an ITV approach.
Finally, the measured volume of highly variable critical OAR (particularly stomach and duodenum) within a 2cm margin around the PTV was found to be positively associated with reduced target coverage (1989). This work allowed the development of a prospective target dose coverage prediction model based on the daily location of anatomy to facilitate effective online adaptive plan review.
Advances in Upper GI Tumour Physics
During ESTRO 2024, 14 abstracts were presented focusing on upper GI tumours, particularly emphasising improvements in the accuracy and precision of dose delivery to tumours in the upper abdomen, such as the pancreas and liver. Accurate delivery in these regions is complicated by daily anatomical variations and respiratory movements.
For pancreatic tumours, treatment planning is challenging due to the proximity of radiosensitive bowel structures. However, significant correlations were found between the geometrical overlap of bowel structures and the PTV, as well as the maximum achievable target coverage, significantly aiding treatment planning on CyberKnife systems (2066). Additionally, online adaptive radiotherapy using an MR-Linac was shown to effectively target the tumour dose while sparing surrounding organs at risk. A critical aspect here is the assessment of accumulated dose throughout radiotherapy. Deformable image registration (DIR) is a useful tool for this, with recent evidence suggesting that an organ-weighted approach provides a geometrical advantage over global DIR strategies (2197).
In the field of liver SBRT, several studies highlighted the complexities of intrafraction motion management, which is crucial for patients undergoing proton therapy due to dose deposition.) X appears feasible, enabling accurate proton dose delivery (1618). Australian research groups have made strides in liver SBRT using conventional linear accelerators. Initial results from a multicentre trial demonstrated that real-time plan adaptations could be made based on the movement of gold markers using a kilovoltage X-ray system, leading to improved outcomes (1080). Moreover, emerging evidence suggests that tumour tracking might also be feasible using radiopaque contrast injected during TACE, which remains visible on imaging for months and can be tracked using deep learning segmentation techniques (1501).
Conclusion
In summary, ESTRO 2024 was highly interesting and informative for cutting-edge upper GI radiotherapy in esophagogastric cancer, pancreatic cancer, primary and secondary liver cancer as well as OMD. Upper GI is an interdisciplinary challenge for all occupational groups and optimal results can only be achieved when physicians, physicists and RTTs click like clockwork.
Authors:
Thomas B. Brunner, Graz, Austria
Eleni Gkika, Bonn, Germany
Dr Gert Meijer, Medical Physicist, UMC Utrecht, The Netherlands
Mairead Daly, RTT, The University of Manchester and The Christie Hospital, Manchester, UK
Pieter Populaire, MD, UZ Leuven, Leuven, Belgium
Dr Tiuri Kroese, In Training Radiation Oncologist, USZ, Zurich, Switzerland