ESTRO 2020 Abstract book

S112 ESTRO 2020

We enrolled 213 consecutive patients (age 31, range 15- 77) affected with HL and treated at our institution between 2010 and 2018. Overall, 179 patients (84%) received mediastinal RT according to the disease extension. Many patients had a huge mediastinal involvement with 51.2% of them with a bulky lesion at the time of diagnosis (109/213). All patients received a combined-modality treatment, with CT followed by RT. Median prescription dose to the PTV was 30 Gy (range 20- 40). The treatment target was delineated according to ISRT and INRT principles. Structures as lungs, breasts, heart and cardiac substructures (chambers, valves and coronary arteries) were all contoured as OARs. For each of them, mean (Dmean) and maximum doses (Dmax) were calculated (Table 1). A predictive model for CAD was estimated, according to previously published reports ( van Nimwegen et al., JCO, 2016) . Results The clinical outcome of our cohort was extremely good and comparable to literature data. The PTV coverage was in respect of ICRU criteria with a median V95% = 95.1% (Table 1). The adoption of an IMRT/VMAT planning kept low doses (Dmean and Dmax reported) to all organs at risk, as detailed on the attached table. In particular, mean heart dose was only 5.1 Gy in our population. The median relative risk of CAD of our patients, compared to healthy people, was only 1.79 (CI 95%: 1.00-6.48), which is significantly lower to the 5-fold risk reported in historical reports (Figure 1).

OC-0202 Combining hyperthermia or a VDA with checkpoint inhibitors to increase tumour immunogenicity P.B. Elming 1 , T.R. Wittenborn 1 , M.R. Horsman 1 1 Aarhus University Hospital, Dept. of Experimental Clinical Oncology, Aarhus, Denmark Purpose or Objective Although checkpoint inhibitors (CIs) have improved outcome for some cancer patients, many patients do not respond. Our pre-clinical studies were designed to test whether treatments like hyperthermia or vascular disrupting agents (VDAs) that kill tumour cells both directly due to cytotoxicity and indirectly via the induction of vascular damage could improve tumor response to CIs. Material and Methods Male or female CDF1 mice were inoculated on the right rear foot with a C3H mammary carcinoma and treated when tumours reached a volume of 200mm 3 . These treatments included local water bath heating (42.5°C for 1 hour) applied on day 0; intra peritoneal (i.p.) injection with the VDA OXi4503 (25 mg/kg) on days 0, 3, 7 and 10; and anti-CTLA-4 (10 mg/kg; i.p.) on days 1, 4, 8 and 11. Tumour size was measured daily, with the endpoint being the time to reach five times treatment volume (TGT5). Tumours were also excised upto 10 days after treatment with heat or OXi4503 and histologically assessed for CD4, and CD8 expression, and HSP70 expression assessed using ELISA. Statistical comparisons were performed using an unpaired t-test (significance level of P<0.05). Results Control untreated tumours had a mean (+ 1 S.E.) TGT5 of 6.6 days (+ 0.2). No significant change was found following treatment with anti-CTLA-4, the TGT5 being 7 days (+0.6). A significant increase was obtained with heating or OXi4503, the respective TGT5 values being 11.5 days (+ 0.6) and 18.6 days (+ 1). When heat or OXi4503 were combined with anti-CTLA4 an even greater effect was observed. With heating the TGT5 values was 18.0 days (+ 3.5). When combined with OXi4503 it was only 19.0 days (+ 1.3), but this did not include the measurements in 3 tumour which were completely controlled. Quantification of CD4 and CD8 expression showed a significant drop of 40- 60 % one day after injecting OXi4503 and around a 70 % drop 3 days after heating. However, these values had returned to pre-treatment levels by 3 to 4 and 7 to 10 days, respectively. HSP70 expression 4 hours after heating was not significantly different from control values, but at 24 hours there was a significant elevation (P=0.004). Conclusion Our C3H mammary carcinoma model did not respond to treatment with anti-CTLA-4, consistent with it being non- immunogenic. However, when treated with agents that induce significant cell kill both directly and indirectly via the induction of vascular damage, the tumour became sensitive to anti-CTLA-4 treatment. Surprisingly, rather than seeing an increase in CD4 or CD8 expression there was a transient decrease with both heating and OXi4503, returning to pre-treatment levels by the end of the anti- CTLA-4 treatment period. HSP70 did increase after applying heat. Thus, we do have possibilities to improve the anti-tumor effect of CIs, but the mechanism remains unclear. Supported by grants from the Danish Cancer Society and the Danish Council for Independent Research: Medical Sciences OC-0203 Eliminating tumour hypoxia to improve the impact of immunotherapy D. Marcus 1 , A. Van der Wiel 1 , R. Biemans 1 , N. Lieuwes 1 , A. Heyerick 2 , J. Smaill 3 , A. Patterson 3 , J. Theys 1 , A. Yaromina 1 , P. Lambin 1 , L. Dubois 1 1 GROW - School for Oncology and Developmental Biology-

Conclusion Our study shows that the application of modern RT planning and delivery techniques, combined with a detailed contouring of the heart and of the cardiac substructures favors a reduction of the risk of RT-related CAD in patients affected with HL.

Proffered Papers: Proffered papers 9: Radiation effects on stem cells and the microenvironment