ECEESPE2025 Poster Presentations Endocrine Related Cancer (76 abstracts)
Radiogenomics pilot study in adrenocortical carcinoma: assessing the relationship between genetic background and computerized tomography texture
Lorenzo Tucci 1 2 3 , Giulio Vara 4 , 5 , Antonio De Leo 1 , 6 , Kassiani Skordilis 7 , Lisa James 7 , Cristina Mosconi 1 , 8 , Balraj Dhesi 9 , Dario De Biase 10 , Juliane Lippert 11 , Abubaker Mohamed 7 , Guido Di Dalmazi 1 , 2 & Cristina L Ronchi 3 & 12
1Alma Mater Studiorum University of Bologna, Medical and Surgical Sciences Department, Bologna, Italy; 2IRCCS Sant’Orsola-Malpighi Polyclinic, Endocrinology and Diabetes Prevention and Care Unit, Bologna, Italy; 3University of Birmingham, Department of Metabolism and Systems Science, Birmingham, United Kingdom; 4Casa di cura madre fortunata Toniolo, Radiology Unit, Bologna, Italy; 5Alma Mater Studiorum University of Bologna, Department of Biomedical and Neuromotor Sciences, Bologna, Italy; 6IRCCS Sant’Orsola Malpighi Polyclinic, Solid Tumor Molecular Pathology Laboratory, Bologna, Italy; 7Queen Elizabeth Hospital Birmingham NHS Trust, Department of Histopathology, Birmingham, United Kingdom; 8IRCCS Sant’Orsola Malpighi Polyclinic, Radiology Unit, Bologna, Italy; 9Queen Elizabeth Hospital Birmingham NHS Trust, Radiology Unit, Birmingham, United Kingdom; 10Alma Mater Studiorum University of Bologna, Department of Pharmacy and Biotechnology, Bologna, Italy; 11University Hospital, University of Wuerzburg, Department of Medicine, Division of Endocrinology and Diabetes, Wuerzburg, Germany; 12Queen Elizabeth Hospital Birmingham NHS Trust, Department of Endocrinology, Birmingham, United Kingdom
JOINT1293
Background: Adrenocortical carcinoma (ACC) is an aggressive cancer with heterogeneous prognosis and few therapeutic options. Somatic genetic alterations have been demonstrated to predict clinical outcomes and have been highlighted as possible therapeutic targets. Radiomics has been used for the prediction of clinically relevant mutations, but not in ACC.
Aim: To predict the presence of key pathogenic gene variants in ACC using radiomics features.
Methods: We retrospectively enrolled 37 ACC patients (25 females) who underwent adrenalectomy at Queen Elizabeth Hospital Birmingham (Birmingham, United Kingdom) and IRCCS Sant’Orsola Malpighi Polyclinic (Bologna, Italy) from 2013 to 2023 with available pre-surgery portal-phase computerized tomography. Targeted Next Generation Sequencing was performed on formalin-fixed paraffin-embed samples using customised panels that included 10 ACC-specific genes (TP53, RB1, CDK4, CTNNB1, APC, ZNRF3, MEN1, TERT, ATM and NF1) designed and assembled in both centres. Sequencing was performed with routinely available techniques at each centre (i. e. Illumina). Radiomics was performed with LifeX software (Lito©, FR). After standardization, features selection was conducted with U Mann Whitney followed by Spearman correlation to avoid collinearity. Predictive models based on radiomics features for the prediction of most frequently mutated genes and mutated beta-catenin pathway genes (BCPG: CTNNB1, ZNRF3, APC), mutated tumour suppressor genes (TSG: TP53, RB1, CDK4), mutated chromatin remodeling genes (CRG: MEN1, TERT), NF1 and ATM were formulated through logistic regression and their diagnostic performances were evaluated with Receiver Operator Characteristic (ROC) curve analysis and negative (NPV) and positive (PPV) predictive factor.
Results: Wild type genotype was observed in 20 samples (54. 1%). Mutations were detected for BCPG (n = 9 (24. 3%): CTNNB1=5 (13. 5%), ZNRF3=3 (8. 1%), APC=2 (5. 4%)), TSG (n = 9 (24. 3%): TP53 =7 (18. 9%), RB1=2 (5. 4%)), CRG (n = 5: MEN1=4 (10. 8%), TERT=1 (2. 7%)), NF1 (n = 4 (10. 8%)). BCPG mutations (P = 0. 020) were predicted by Inverse Variance (Odds ratio (OR)=0. 171 (95% Confidence Interval (95% CI) =0. 034-0. 846), P = 0. 030), Normalised Grey Level Non-Uniformity (Or = 0. 084 (95% CI=0. 007-0. 983), P = 0. 048), with sensitivity=88. 9%, specificity=79. 6%, NPV=81. 3%, PPV=60%. CTNNB1 mutations (P = 0. 002) were predicted by Inverse Variance (Or = 0. 181 (95% CI=0. 039-0. 836), P = 0. 029) and Normalised Grey Level Non-Uniformity (Or = 0. 082 (95% CI=0. 007-0. 959), P = 0. 046), with sensitivity=100%, specificity=81. 8%, NPV=100%, PPV=45. 5%. NF1 mutations (P = 0. 032) were predicted by Difference Entropy (OR=3. 014 (95% CI=1. 000-9. 079), P = 0. 050), with sensitivity=100%, specificity=72. 7%, NPV=100%, PPV=30. 8%.
Conclusion: We present the first radiogenomics study in ACC so far. Radiomics could be a swift and cost-effective analysis useful for early recognition of somatic mutations in BCPG, especially in genes CTNNB1 and NF1. Larger studies might be required to further investigate variant predictions.
Volume 110
Article tools
My recent searches
My recently viewed abstracts
Authors
Endocrine Abstracts
ISSN 1470-3947 (print) | ISSN 1479-6848 (online)
© Bioscientifica 2025 |
Privacy policy |
Cookie settings
BiosciAbstracts
Bioscientifica Abstracts is the gateway to a series of products that provide a permanent, citable record of abstracts for biomedical and life science conferences.
Find out more