After successful dosimetric beam commissioning, quality assurance measurements performed during a 24-month period show very stable beam characteristics, which are therefore suitable for performing safe and accurate patient treatments.

Objectives: (1.1) to evaluate the association between baseline 18F-FDG PET/CT semi-quantitative parameters of the primary lesion with progression free survival (PFS), overall survival (OS) and response to immunotherapy, in advanced non-small cell lung carcinoma (NSCLC) patients eligible for immunotherapy; (1.2) to evaluate the application of radiomics analysis of the primary lesion to identify features predictive of response to immunotherapy; (1.3) to evaluate if tumor burden assessed by 18F-FDG PET/CT (N and M factors) is associated with PFS and OS. Materials and Methods: we retrospectively analyzed clinical records of advanced NCSLC patients (stage IIIb/c or stage IV) candidate to immunotherapy who performed 18F-FDG PET/CT before treatment to stage the disease. Fifty-seven (57) patients were included in the analysis (F:M 17:40; median age = 69 years old). Notably, 38/57 of patients had adenocarcinoma (AC), 10/57 squamous cell carcinoma (SCC) and 9/57 were not otherwise specified (NOS). Overall, 47.4% patients were stage IVA, 42.1% IVB and 8.8% IIIB. Immunotherapy was performed as front-line therapy in 42/57 patients and as second line therapy after chemotherapy platinum-based in 15/57. The median follow up after starting immunotherapy was 10 months (range: 1.5–68.6). Therapy response was assessed by RECIST 1.1 criteria (CT evaluation every 4 cycles of therapy) in 48/57 patients or when not feasible by clinical and laboratory data (fast disease progression or worsening of patient clinical condition in nine patients). Radiomics analysis was performed by applying regions of interest (ROIs) of the primary tumor delineated manually by two operators and semi-automatically applying a threshold at 40% of SUVmax. Results: (1.1) metabolic tumor volume (MTV) (p = 0.028) and total lesion glycolysis (TLG) (p = 0.035) were significantly associated with progressive vs. non-progressive disease status. Patients with higher values of MTV and TLG had higher probability of disease progression, compared to those patients presenting with lower values. SUVmax did not show correlation with PD status, PFS and OS. MTV (p = 0.027) and TLG (p = 0.022) also resulted in being significantly different among PR, SD and PD groups, while SUVmax was confirmed to not be associated with response to therapy (p = 0.427). (1.2) We observed the association of several radiomics features with PD status. Namely, patients with high tumor volume, TLG and heterogeneity expressed by “skewness” and “kurtosis” had a higher probability of failing immunotherapy. (1.3) M status at 18F-FDG PET/CT was significantly associated with PFS (p = 0.002) and OS (p = 0.049). No significant associations were observed for N status. Conclusions: 18F-FDG PET/CT performed before the start of immunotherapy might be an important prognostic tool able to predict the disease progression and response to immunotherapy in patients with advanced NSCLC, since MTV, TLG and radiomics features (volume and heterogeneity) are associated with disease progression.

To perform a systematic review on the research on the application of artificial intelligence (AI) to imaging published in Italy and identify its fields of application, methods and results. Materials and Methods: A Pubmed search was conducted using terms Artificial Intelligence, Machine Learning, Deep learning, imaging, and Italy as affiliation, excluding reviews and papers outside time interval 2015-2020. In a second phase, participants of the working group AI4MP on Artificial Intelligence of the Italian Association of Physics in Medicine (AIFM) searched for papers on AI in imaging. Results: The Pubmed search produced 794 results. 168 studies were selected, of which 122 were from Pubmed search and 46 from the working group. The most used imaging modality was MRI (44%) followed by CT(12%) ad radiography/mammography (11%). The most common clinical indication were neurological diseases (29%) and diagnosis of cancer (25%). Classification was the most common task for AI (57%) followed by segmentation (16%). 65% of studies used machine learning and 35% used deep learning. We observed a rapid increase of research in Italy on artificial intelligence in the last 5 years, peaking at 155% from 2018 to 2019.

To test the hypothesis that irradiated volume of specific subregions of pelvic active bone marrow as detected by (18)FDG-PET may be a predictor of decreased blood cells nadirs in anal cancer patients undergoing concurrent chemoradiation, we analyzed 44 patients submitted to IMRT and concurrent chemotherapy. Several bony structures were defined: pelvic and lumbar-sacral (LSBM), lower pelvis (LPBM) and iliac (IBM) bone marrow. Active BM was characterized employing (18)FDG-PET and characterized in all subregions as the volume having standard uptake values (SUVs) higher than SUVmean. All other regions were defined as inactive BM. On dose-volume histograms, dosimetric parameters were taken. Endpoints included white blood cell count (WBC), absolute neutrophil count (ANC), hemoglobin (Hb) and platelet (Plt) nadirs. Generalized linear modeling was used to find correlations between dosimetric variables and blood cells nadirs. WBC nadir was significantly correlated with LSBM mean dose (β = -1.852; 95 % CI -3.205/-0.500; p = 0.009), V10 (β = -2.153; 95 % CI -4.263/-0.721; p = 0.002), V20 (β = -2.081; 95 % CI -4.880/-0.112; p = 0.003), V30 (β = -1.971; 95 % CI -4.748/-0.090; p = 0.023) and IBM V10 (β = -0.073; 95 % CI -0.106/-0.023; p = 0.016). ANC nadir found to be significantly associated with LSBM V10 (β = -1.878; 95 % CI -4.799/-0.643; p = 0.025), V20 (β = -1.765; 95 % CI -4.050/-0.613; p = 0.030) and IBM V10 (β = -0.039; 95 % CI -0.066/-0.010; p = 0.027). Borderline significance was found for correlation between Plt nadir and LSBM V30 (β = -0.056; 95 % CI -2.748/-0.187; p = 0.060), V40 (β = -0.059; 95 % CI -3.112/-0.150; p = 0.060) and IBM V30 (β = -0.028; 95 % CI -0.074/-0.023; p = 0.056). No inactive BM subsites were found to be correlated with any blood cell nadir. (18)FDG-PET is able to define active bone marrow within pelvic osseous structures. LSBM is the strongest predictor of decreased blood cells nadirs in anal cancer patients undergoing concurrent chemoradiation.

Purpose: to investigate the role of selective avoidance of hematopoietically active BM within the pelvis, as defined with 18FDG-PET, employing a targeted IMRT approach, to reduce acute hematologic toxicity (HT) profile in anal cancer patients undergoing concurrent chemo-radiation. Methods: a one-armed two-stage Simon’s design was selected to test the hypothesis that BM-sparing approach would improve by 20% the rate of G0–G2 (vs. G3–G4) HT, from 42% of RTOG 0529 historical data to 62% (α = 0.05 and the β = 0.20). At the first stage, among 21 enrolled patients, at least 9 should report G0–G2 acute HT to further proceed with the trial. We employed 18FDG-PET to identify active BM within the pelvis. Acute HT was assessed via weekly blood counts and scored as per the Common Toxicity Criteria for Adverse Effects version 4.0. Results: from December 2017 to October 2019, 21 patients were enrolled. Maximum observed acute HT comprised 9% rate of ≥G3 leukopenia and 5% rate of ≥G3 neutropenia and anemia. Overall, only 4 out of 21 treated patients (19%) experienced ≥G3 acute HT. Conversely, 17 patients (81%) experienced G0–G2 events, way above the threshold set by the trial design. Conclusion: 18FDG-PET-guided BM-sparing IMRT was able to reduce acute HT in anal cancer patients treated with concomitant chemo-radiation. These results prompted us to conclude the second part of this prospective phase II trial.