Investigator

Luca Boldrini

Medical Doctor · Fondazione Policlinico Universitario “A. Gemelli” IRCCS, Radiation Oncology

Research Interests

LBLuca Boldrini
Papers(2)
Deep-Learning to Pred…Low Tesla magnetic re…
Collaborators(8)
Tina PasciutoVincenzo ValentiniAngelo MinucciAntonio PirasBruno FiondaCamilla NeroFrediano InzaniGiuditta Chiloiro
Institutions(3)
Agostino Gemelli Univ…Universit Cattolica D…Università degli Stud…

Papers

Deep-Learning to Predict BRCA Mutation and Survival from Digital H&E Slides of Epithelial Ovarian Cancer

BRCA 1/2 genes mutation status can already determine the therapeutic algorithm of high grade serous ovarian cancer patients. Nevertheless, its assessment is not sufficient to identify all patients with genomic instability, since BRCA 1/2 mutations are only the most well-known mechanisms of homologous recombination deficiency (HR-d) pathway, and patients displaying HR-d behave similarly to BRCA mutated patients. HRd assessment can be challenging and is progressively overcoming BRCA testing not only for prognostic information but more importantly for drugs prescriptions. However, HR testing is not already integrated in clinical practice, it is quite expensive and it is not refundable in many countries. Selecting patients who are more likely to benefit from this assessment (BRCA 1/2 WT patients) at an early stage of the diagnostic process, would allow an optimization of genomic profiling resources. In this study, we sought to explore whether somatic BRCA1/2 genes status can be predicted using computational pathology from standard hematoxylin and eosin histology. In detail, we adopted a publicly available, deep-learning-based weakly supervised method that uses attention-based learning to automatically identify sub regions of high diagnostic value to accurately classify the whole slide (CLAM). The same model was also tested for progression free survival (PFS) prediction. The model was tested on a cohort of 664 (training set: n = 464, testing set: n = 132) ovarian cancer patients, of whom 233 (35.1%) had a somatic BRCA 1/2 mutation. An area under the curve of 0.7 and 0.55 was achieved in the training and testing set respectively. The model was then further refined by manually identifying areas of interest in half of the cases. 198 images were used for training (126/72) and 87 images for validation (55/32). The model reached a zero classification error on the training set, but the performance was 0.59 in terms of validation ROC AUC, with a 0.57 validation accuracy. Finally, when applied to predict PFS, the model achieved an AUC of 0.71, with a negative predictive value of 0.69, and a positive predictive value of 0.75. Based on these analyses, we have planned further steps of development such as proving a reference classification performance, exploring the hyperparameters space for training optimization, eventually tweaking the learning algorithms and the neural networks architecture for better suiting this specific task. These actions may allow the model to improve performances for all the considered outcomes.

Low Tesla magnetic resonance guided radiotherapy for locally advanced cervical cancer: first clinical experience

Objective: Magnetic resonance–guided radiotherapy (MRgRT) represents an innovative approach for personalized radiotherapy treatments and its applications are being explored in various anatomical sites to fully understand its potential advantages. This study describes the first clinical experience of MRgRT application in patients with locally advanced cervical cancer (LACC) undergoing neoadjuvant chemoradiotherapy. The feasibility of the technique is evaluated and its toxicity profile and clinical outcomes are reported. Methods: Patients with LACC (International Federation of Gynecology and Obstetrics stage IIA–IVA) undergoing neoadjuvant chemoradiotherapy (CRT) on a 0.35T Tri-60-Co hybrid unit (ViewRay) were retrospectively compared with randomly selected patients treated with a standard linear accelerator. Total prescribed dose was 50.6 Gy (2.3 Gy/fraction) to planning target volume 1 (PTV1) and 39.6 Gy (1.8 Gy/fraction) to PTV2, delivered using a simultaneous integrated boost. Surgery was performed 8 weeks after the end of CRT. The effect of magnetic resonance guidance on replanning approaches, treatment-related toxicities, and pathologic response were assessed for each patient. Patient outcomes were noted and dosimetric comparisons performed between the 2 arms. Results: Nine patients with LACC treated from May 2018 to November 2018 were retrospectively enrolled and their records compared with the records of an equivalent cohort of randomly selected patients. Five replanning cases were performed in the MRgRT group and 0 in the linear accelerator group. Acute G1–G2 gastrointestinal toxicities were observed in 33.3% of MRgRT patients and in 55.5% of linear accelerator patients; acute G1–G2 genitourinary toxicities in 22.2% and 33.3%, respectively. No G3 toxicity was found except for neutropenia in 2 patients. No differences were observed in pathologic response between the 2 groups. Conclusions: Despite the retrospective nature of the observations and the low number of enrolled patients, the application of MRgRT in LACC appears to be safe and feasible with a favorable toxicity profile and response rates comparable to gold standard, supporting the setup of larger prospective studies to investigate the potentialities of this new technology.

97Works
2Papers
8Collaborators
NeoplasmsLung NeoplasmsPrognosisBiomarkers, TumorNeoplasm Recurrence, LocalNeoplasm StagingCarcinoma, Non-Small-Cell LungCirculating Tumor DNA

Positions

2016–

Medical Doctor

Fondazione Policlinico Universitario “A. Gemelli” IRCCS · Radiation Oncology

2012–

Resident

Università Cattolica del Sacro Cuore Facoltà di Medicina e Chirurgia · Radiation Oncology

Education

2011

Medical doctor

Università Cattolica del Sacro Cuore Facoltà di Medicina e Chirurgia · Radiation Oncology

2005

Liceo Classico Graduation

Collegio San Giuseppe Istituto de Merode · Liceo Classico

Country

IT

Keywords
radiotherapyradiation oncologycontouringsegmentation