ESTRO 2020 Abstract Book

S280 ESTRO 2020

subtypes of SGC, which makes each subtype even rarer. Recognition of, and differentiation between these different subtypes is notoriously difficult and different subtypes exhibit different clinical features adding up to the complexity of the disease. For localized and resectable disease, surgical resection with or without postoperative radiotherapy is the cornerstone of treatment. In case of local recurrent or metastatic (R/M) disease, systemic treatment is challenging, but urgent given the prognosis of this disease stage. Lumped for all types of SGC with distant metastases (71% of the patients presenting with recurrent disease) the median overall survival is 15 months with overall survival rates at 1 and 5 years of 54.5% and 14.8%, respectively. This however, varies widely between different subtypes. For example, in adenoid cystic carcinoma (AdCC) median overall survival of several years in distant metastatic disease has been reported. This contrasts with salivary duct carcinoma (SDC), an aggressive subtype of SGC, in which median overall survival for R/M disease receiving best supportive care is only 5 months. The clinicopathological diversity of the disease justifies therapy tailored to the specific SGC subtype, highlighting the importance of adequate pathological examination (e.g. subtype, stage, growth pattern), preferably performed by a salivary gland expert pathologist. However, rarity of SGC and its extensive heterogeneity hinders large-scale patient accrual in prospective trials and difficulties in correct histopathological subtyping of SGC endanger homogeneity of cohorts. Therefore, performance of clinical trials in SGC is challenging. This is reflected in the limited amount of studies performed on classic chemotherapeutic agents in SGC. The use of chemotherapy in R/M SGC is poorly studied and is unable to drastically change outcomes in most R/M SGC. Survival rates and limited benefit of chemotherapy emphasize that there is an unmet need for new therapeutic strategies for patients with R/M SGC. The paucity of treatment options may be reduced by mapping tumor characteristics and unraveling genetic aberrations in search for possible targets for systemic therapies. By doing so, salivary gland cancer patients could also share in the benefits of the therapeutic advances made in more common malignancies, especially since the body of evidence for presence of such targets in different histological subtypes is increasing. Several different subtypes of SGC have characteristics targetable with systemic treatment. Especially in AdCC and SDC, the histological subtypes of SGC with the largest burden of R/M disease, promising therapeutic strategies are available. These new developments in the treatment of (R/M) salivary gland cancer, i.e. targeted therapy and hormonal treatment will be discussed. Also, other new treatment options, i.e. immunotherapy with checkpoint inhibitors, will be shortly discussed. Highlights: - Salivary gland cancer is a highly heterogenous disease consisting of 22 subtypes - In several of the subtypes novel systemic treatment strategies can be rationalized - Immunohistochemistry and genomic profiling can identify targeted therapy options - This approach can significantly alter the prognosis of salivary gland cancer patients

Symposium: Target definition and dose prescription in the era of dose painting and probabilistic planning

SP-0507 Integrating contouring uncertainties in target definition V. Grégoire 1 1 Centre Léon Bérard, Radiation Oncology, Lyon, France Abstract text The definition of the Target Volumes has evolved with time. In 1978, ICRU #29 defined for the first time the term “Target Volume” combining both the macroscopic disease and its microscopic extension. In the subsequent ICRU reports #50 (1993) and #62 (1999), a distinction was introduced between the GTV for the Gross (visible or palpable) Tumor Volume and the CTV for Clinical Target Volume encompassing both the macroscopic (if any) and the microscopic tumor burden. More recently, the ICRU report #83 (2010) summarised the practical implementation of these concepts in the 3D imaging configuration. However, the CTV was still defined as a finite volume typically encompassing anatomic compartments in which a given amount of cancer cells was thought to be homogeneously distributed. In 2020, we now have evidences that neither the GTV nor the CTV are homogenous structures, but rather are volumes with an heterogenous distribution of target cell (stem cells) densities. Molecular imaging with PET or MRI can clearly depict the heterogeneous nature of GTV. For various tumor types such as head and neck, lung, brain and prostate, three-dimentional pathological examination of surgical specimens have also identified that microscopic cell density typically decreases as a function of the distance from the border of the GTV; such distance varies according to the histologic types and the organ in which tumor developed. Altogether, these data should prompt the clinicians to revisit the practical implementation of the target volume concept to better integrate the notion of probability of target cells presence within the complex anatomic compartments surrounding the GTV. The identification of such heterogeneous tumor cell distribution, could serve as the basis for varying dose prescription. SP-0508 Integrating target delineation decisions in interactive treatment planning T. Bortfeld 1 , N. Shusharina 1 1 mass. General Hospital, Radiation Oncology- Division Of Radiation Biophysics, Boston- Ma, Usa Abstract text Interactive treatment planning has been achieved through fast re-optimization and interactive navigation based on pre-calculated Pareto-optimal plans, using multi-criteria optimization (MCO). The interactive part is limited to changing dose tolerance levels in organs-at risk (OAR) and target volumes that were defined beforehand. It is desirable to generalize the MCO concept to explore the tradeoff between expanding the clinical target volume (CTV) for better coverage of potential tumor spread in surrounding healthy tissues, and sparing of OARs. In the first part of this talk we will present rapid methods to automatically expand the CTV beyond the gross tumor volume (GTV), while respecting automatically defined (with deep learning) anatomic barriers to tumor spread. The computational method is based on the Dijkstra shortest path algorithm. In the second part we will translate calculated distances from the GTV into