Because of the high dependence across features extracted from images based on the same filter type, the number of redundant features may be reduced in future studies.References 1.

ConclusionFRED was commissioned and validated against the experimental data acquired during the start-up of PBT centre in Krakow.A fast procedure for phase space library implementation was developed.FRED passed clinical acceptance tests required to admit certified TPS for clinical use.Once commissioned and tested with patient CT data, FRED can be used for fast recalculation of clinical treatment plans.

Purpose or ObjectiveGenerally, patient with locally advanced pancreatic cancer (LAPC) is treated with chemotherapy alone or Xray radiotherapy (RT) concurrent with gemcitabine (GEM).Median survival time (MST) of standard treatment is about 14 months.Recently, there was a report of carbon ion radiotherapy(C-ion RT)with concurrent usage of GEM for LAPC which showed 2-year overall survival (OS) rate and MST 48% and 22.5 months, that is better outcome compared with other treatment for LAPC.However, there is no report for C-ion RT combined with other chemotherapy.On the other hand, postoperative adjuvant chemotherapy with S-1 significantly extended both overall and relapse-free survival of Japanese patients with resected pancreatic cancer compared with GEM.[OM1] [Office2] Thus, we hypothesize that C-ion RT concurrent S-1 (tegafur/gimeracil/oteracil potassium) is more effective for LAPC and started prospective phase II study in our institute. Material and MethodsEligibility criteria: (1) Pathologic confirmation of pancreatic invasive ductal carcinomas or clinical diagnosis by imaging.(2) Absence of distant metastasis (patients with para-aortic lymph node metastasis were eligible) (3) Unresectable primary tumors due to T4 disease based on the 7th edition of TNM classification, involving either the celiac axis or the superior mesenteric artery.(4) Without gastrointestinal ulcer (stomach, duodenum) (excluding repaird ulcer) (5) Age at registration is from 20 to 80 years old (6) ECOG performance status (PS) of 0 to 2. (7) Without the surgical resection history to a pancreatic cancer.Without the radiotherapy to a pancreatic cancer.Treatment: Prescribed doses were 55.2 Gy [relative biological effectiveness (RBE)] in 12 fractions.Concurrent S-1 was administered orally during 28 days, the dose levels were 80 mg/-1.00m 2 , 100 mg/1.00-1.50m 2 ,120 mg/1.50-m 2 per day.Clinical outcome measures: The primary endpoint was overall survival (OS).The secondary endpoints were local control (LC) rate, progression free survival (PFS) rate and adverse effects.Toxicity was graded using Common Terminology Criteria for Adverse Events, version 4.0 (CTCAE 4.0). ResultsThere were 13 patients enrolled this study from January 2016 to December 2017 in our institute.MST was 23.4 months. 1 and 2-year OS was 100% and 38%.1 and 2-year LC are 80% and 28%. 1 and 2-year PFS rate are 60% and 49%.There were occurrences of acute toxicity, leukopenia grade2 in 2 patients and grade 3 in 1 patient, anemia grade2 in 2 patients and grade 3 in 1 patient, thrombocytopenia Grade 2 in 1 patient.There is no patient with Grade 4 or higher.There were no occurrences of late toxicity of grade 3 or higher. ConclusionInitial results show that C-ion RT combined with S-1 appears to be tolerated and not inferior compared with GEM for LAPC.Although there are still few cases, C-ion RT combined with S-1 is also considered to be one of the options of treatment for LAPC.

SUMMARY: 3D rotational angiography provides remarkable spatial resolution for cerebrovascular disorders; however, it cannot be integrated directly into gamma knife planning due to the discrepancy of DICOM “tag” information, and most physicians still cannot benefit from 3D rotational angiography. Here, we describe a simple and easy technique to enable the integration of 3D rotational angiography.

Radiomics analysis using advanced medical imaging technologies provides opportunities to detect the tumor characteristics such as its histology non-invasively. To assess the potential ability of radiomics to predict tumor characteristics, this study analyzes the association of CT-based radiomics features and histology of early-stage non-small cell lung cancer (ES-NSCLC) in a Japanese cohort. In this analysis, we included 40 ES-NSCLC patients treated by stereotactic ablative radiotherapy (SABR). Of them, 22 patients have adenocarcinoma (AC), and the remains have squamous cell carcinoma (SC). To acquire the robust radiomics features taking into account interobserver contouring variations, four volume-of-interests (VOIs) were created on the maximum exhale phase of four-dimensional planning CT (4DCT); an original gross tumor volume (GTV) used in actual SABR planning and three retrospective GTVs contoured by two radiation oncologists and one medical physicist. For all patients, a total of 416 quantitative radiomics features were extracted from each GTV were then analyzed in the machine-learning based classification. To select optimal features, the classification of ES-NSCLC histology (AC/SC) was first assessed by the univariate manner with significance and area under the receiver operating characteristic (ROC) curve (AUC) for all above features. Twenty-eight (70%) of 40 patients was sampled as the training cohort, whereas the rest twelve patients (30%) was assigned into the test cohort. The prediction performance of selected radiomics features based on the training cohort was validated by the AUC value of test cohort. To evaluate multivariate model performance using cross validation, such a random sampling and validation was repeated 30 times and the average AUC and its standard deviation was evaluated. In our univariate analysis, we observed four features having significant association with histological subtypes. For the test cohort, averaged AUC among four VOIs with four features found to be 0.707±0.011, and almost independent of the choice of feature evaluation index.Despite the difference in the class distributions between training and validation dataset, the multivariate machine learning models achieved high prediction accuracy. For ES-NSCLC Japanese patients, radiomics analysis was able to predict ES-NSCLC histology by using multiple GTVs contouring on 4DCT. The high prediction accuracy of non-invasive histology evaluation by the current advanced machine-learning approaches is believed to be the promising clinical application.

Purpose or ObjectiveTo investigate the prognostic factors in patients with advanced squamous cell carcinoma of the head and neck (SCCHN) who received a novel weekly induction chemotherapy (IndCT) followed by local therapy (surgery/radiotherapy). Material and MethodsFifty patients with stage III/IV SCCHN were enrolled.Outpatient IndCT consisted of an uniform 4-drug regimens (cisplatin 60 mg/m2, day 1; docetaxel 50 mg/m2, day 8; 5-fluorouracil 2500 mg/m2 + leucovorin 250 mg/m2, day 15; epirubicin 30 mg/m2 + methotrexate 30 mg/m2, day 22; repeated every 4 weeks for 3-4 cycles).After finishing IndCT, local therapy including surgery, radiotherapy, concurrent chemoradiotherapy, or bio-radiotherapy was administered.Univariate and multivariate Cox proportional hazard model were used to identify significant prognostic factors.Analyzed variables included patient's characteristics (age, gender, performance status), tumor factors (primary site, pathological differentiation, T-stage, N-stage), treatment factors (chemotherapy, surgery, radiotherapy) and pre-treatment FDG PET scan parameters (SUVmax of primary tumor, metabolic tumor volume [MTV], total lesion glycolysis [TLG]). ResultsAfter a median follow-up of 25 months, 13 patients experienced locoregional recurrence, 1 distant metastasis, and 1 both locoregional recurrence and distant metastasis.Kaplan-Meier survival analysis revealed TLG (> vs. < 520), MTV (> vs. < 70), SUVmax (> vs. < 14.0) of primary tumor, and chemotherapy duration affected overall survival significantly (P=0.0004,P=0.0059, P=0.0209, and P=0.0039, respectively).Only TLG (P=0.0102) and MTV (P=0.0164) were significant factors for locoregional failure-free survival.In univariate analysis, chemotherapy duration (P=0.0101),SUVmax of the primary tumor (P=0.0279),MTV (P=0.0094), and TLG (P=0.0011) were significant predictors for death; MTV (P=0.0236) and TLG (P=0.0159) were significant factors in predicting locoregional recurrence.In multivariate analysis, both TLG and MTV revealed as an independent predictor for death (P=0.0057 and P=0.0156) and locoregional recurrence (P=0.0157 and P=0.0225).nguage:EN-US; total lesion glycolysis [TLG]).Conclusion TLG and MTV were the most important prognostic factors in predicting mortality and locoregional recurrence in patients with advanced SCCHN treated by IndCT followed by surgery/radiotherapy.

________________________________________________________________________________Follow up was with CXR at 6months followed by CT at 6 months and clinical follow up, 3 monthly.Results: 167 patients with stage IA-IIB disease treated.55% histologically proven.There were 4 (2.4%) radiologically confirmed local recurrences giving a local control rate of 97.6%.Median survival was 43.2months.3 year Overall Survival was 56.4% (see Fig 1).Treatment was well tolerated with minimal G3 toxicity (5 patients). Conclusion:Our results suggest that SABR for medically inoperable NSCLC can be safely and effectively implemented in a non-academic institution with appropriate equipment and training.Clinical outcomes are comparable with internationally published series [3}, with encouraging 3yr OS rate of 56%.Toxicity is minimal.Longer term follow-up is required to confirm findings and provide data regarding longterm toxicity.

RT appears to be the current standard of care.

Some recent studies reported that radiation dose of subventricular zone (SVZ) is associated with recurrence pattern and survival of malignant glioma patients. Glioma cells are believed to originate from cancer stem cells (CSCs) which are transformed neural stem cells in SVZ. Our previous study showed that SVZ-contiguous tumors at initial presentation are prone to recur from SVZ-contiguous region, and high dose radiation therapy is associated with decreased possibility of SVZ-contiguous recurrence. The purpose of this study was to evaluate the prognostic value of radiation dose to SVZ in malignant glioma patients. Consecutive 74 histologically proven malignant glioma patients who received radiation therapy at our hospital from 1992 to 2013 were included. Fifty-four of 74 patients' dosimetries were accessible. Radiation oncologists contoured SVZ on computed tomography to calculate the delivered radiation dose. Univariate and multivariate analyses were performed to examine the prognostic factors for progression free survival (PFS) and overall survival (OS). In multivariate analysis, we included 9 covariates into Cox proportional hazards model: age, sex, histology, extent of resection, ipsilateral SVZ dose, clinical target volume (CTV) volume, Karnofsky Performance Status (KPS), localization of the lesion, and MIB-1 index. Patients' median age was 60.5 years old (range, 5-83). There were 46 male and 28 female patients. There were 58 glioblastoma and 16 anaplastic astrocytoma patients. Fifty-seven patients received surgery. Prescription doses were 60 Gy and 80 Gy in 55 and 18 patients, respectively. Eighteen patients have no evidence of recurrence at the time of analysis. Median follow-up, PFS, and OS periods were 666, 428 and 802 days, respectively. The median of mean ipsilateral SVZ, contralateral SVZ, and CTV doses were 58.2, 44.1, and 60.4 Gy, respectively. On multivariate analysis only MIB-1 index was significant to OS and extent of surgery was significant to PFS. In a subgroup analysis of 58 glioblastoma patients, none of 8 factors was significant to OS while higher ipsilateral SVZ dose and less than total resection were associated with shorter PFS. Higher radiation doses to ipsilateral SVZ did not affect PFS or OS in malignant glioma patients on multivariate analysis. Higher radiation doses were associated with decreased PFS in glioblastoma patients.