Radiomics-based prediction of two-year clinical outcome in locally advanced cervical cancer patients undergoing neoadjuvant chemoradiotherapy.

PURPOSE: The aim of this study is to determine if radiomics features extracted from staging magnetic resonance (MR) images could predict 2-year long-term clinical outcome in patients with locally advanced cervical cancer (LACC) after neoadjuvant chemoradiotherapy (NACRT). MATERIALS AND METHODS: We retrospectively enrolled patients with LACC diagnosis who underwent NACRT followed by radical surgery in two different institutions. Radiomics features were extracted from pre-treatment 1.5 T T2w MR images. The predictive performance of each feature was quantified in terms of Wilcoxon-Mann-Whitney test. Among the significant features, Pearson correlation coefficient (PCC) was calculated to quantify the correlation among the different predictors. A logistic regression model was calculated considering the two most significant features at the univariate analysis showing the lowest PCC value. The predictive performance of the model created was quantified out using the area under the receiver operating characteristic curve (AUC). RESULTS: A total of 175 patients were retrospectively enrolled (142 for the training cohort and 33 for the validation one). 1896 radiomic feature were extracted, 91 of which showed significance (p < 0.05) at the univariate analysis. The radiomic model showing the highest predictive value combined the features calculated starting from the gray level co-occurrence-based features. This model achieved an AUC of 0.73 in the training set and 0.91 in the validation set. CONCLUSIONS: The proposed radiomic model showed promising performances in predicting 2-year overall survival before NACRT. Nevertheless, the observed results should be tested in larger studies with consistent external validation cohorts, to confirm their potential clinical use.

The role of apparent diffusion coefficient (ADC) in the evaluation of lymph node status in patients with locally advanced cervical cancer: our experience and a review.

Purpose: To evaluate the role of apparent diffusion coefficient (ADC) value measurement in the diagnosis of meta-static lymph nodes (LNs) in patients with locally advanced cervical cancer (LACC) and to present a systematic review of the literature. Material and methods: Magnetic resonance imaging (MRI) exams of patients with LACC were retrospectively eva-luated. Mean ADC, relative ADC (rADC), and correct ADC (cADC) values of enlarged LNs were measured and compared between positron emission tomography (PET)-positive and PET-negative LNs. Comparisons were made using the Mann-Whitney U-test and Student's t-test. ROC curves were generated for each parameter to identify the optimal cut-off value for differentiation of the LNs. A systematic search in the literature was performed, exploring several databases, including PubMed, Scopus, the Cochrane library, and Embase. Results: A total of 105 LNs in 34 patients were analysed. The median ADC value of PET-positive LNs (0.907 × 10-3 mm2/s [0.780-1.080]) was lower than that in PET-negative LNs (1.275 × 10-3 mm2/s [1.063-1.525]) (p < 0.05). rADC and cADC values were lower in PET-positive LNs (rADC: 0.120 × 10-3 mm2/s [-0.060-0.270]; cADC: 1.130 [0.980-1.420]) than in PET-negative LNs (rADC: 0.435 × 10-3 mm2/s [0.225-0.673]; cADC: 1.615 [1.210-1.993]) LNs (p < 0.05). ADC showed the highest area under the curve (AUC 0.808). Conclusions: Mean ADC, rADC, and cADC were significantly lower in the PET-positive group than in the PET-negative group. The ADC cut-off value of 1.149 × 10-3 mm2/s showed the highest sensitivity. These results confirm the usefulness of ADC in differentiating metastatic from non-metastatic LNs in LACC.

CT findings after pelvic exenteration: review of normal appearances and most common complications.

The aim of this review is to illustrate normal computed tomography (CT) findings and the most common complications in patients who underwent pelvic exenteration (PE) for advanced, persistent or recurrent gynecological cancers. We review the various surgical techniques used in PE, discuss optimal CT protocols for postsurgical evaluation and describe cross-sectional imaging appearances of normal postoperative anatomic changes as well as early and late complications. The interpretation of abdominopelvic CT imaging after PE is very challenging due to remarkable modifications of normal anatomy. After this radical pelvic surgery, the familiarity with expected CT appearances is crucial for diagnosis and appropriate management of potentially life-threatening complications in patients who underwent PE.

MRI in the Evaluation of Locally Advanced Vulvar Cancer Treated with Chemoradiotherapy and Vulvar Cancer Recurrence: The 2021 Revision of FIGO Classification and the Need for Multidisciplinary Management.

Magnetic resonance imaging (MRI) plays an essential role in the management of patients with locally advanced vulvar cancer (LAVC), who frequently benefit from a multidisciplinary approach. Accordingly, chemoradiotherapy (CRT) with radical or neoadjuvant intent seems to provide a better quality of life and less morbidity than extensive surgery alone. In this overview, we discuss the role of MRI in the post-CRT assessment of LAVC, emphasizing the evaluation of primary tumor response. In order to assess treatment response and select candidates for post-CRT local excision, the MRI findings are described according to signal intensity, restricted diffusion, enhancement, and invasion of adjacent organs. We also focus on the role of MRI in detecting vulvar cancer recurrence. It occurs in 30-50% of patients within two years after initial treatment, the majority appearing near the original resection margins or in ipsilateral inguinal or pelvic lymph nodes. Finally, we describe early and delayed complications of CRT, such as cellulitis, urethritis, vulvar edema, bone changes, myositis, and fistulization. By describing the role of MRI in assessing LAVC response to CRT and detecting recurrence, we hope to provide suitable indications for a personalized approach.

CT-Based Radiomics and Deep Learning for BRCA Mutation and Progression-Free Survival Prediction in Ovarian Cancer Using a Multicentric Dataset.

PURPOSE: Build predictive radiomic models for early relapse and BRCA mutation based on a multicentric database of high-grade serous ovarian cancer (HGSOC) and validate them in a test set coming from different institutions. METHODS: Preoperative CTs of patients with HGSOC treated at four referral centers were retrospectively acquired and manually segmented. Hand-crafted features and deep radiomics features were extracted respectively by dedicated software (MODDICOM) and a dedicated convolutional neural network (CNN). Features were selected with and without prior harmonization (ComBat harmonization), and models were built using different machine learning algorithms, including clinical variables. RESULTS: We included 218 patients. Radiomic models showed low performance in predicting both BRCA mutation (AUC in test set between 0.46 and 0.59) and 1-year relapse (AUC in test set between 0.46 and 0.56); deep learning models demonstrated similar results (AUC in the test of 0.48 for BRCA and 0.50 for relapse). The inclusion of clinical variables improved the performance of the radiomic models to predict BRCA mutation (AUC in the test set of 0.74). CONCLUSIONS: In our multicentric dataset, representative of a real-life clinical scenario, we could not find a good radiomic predicting model for PFS and BRCA mutational status, with both traditional radiomics and deep learning, but the combination of clinical and radiomic models improved model performance for the prediction of BRCA mutation. These findings highlight the need for standardization through the whole radiomic pipelines and robust multicentric external validations of results.

Evaluation of early regression index as response predictor in cervical cancer: A retrospective study on T2 and DWI MR images.

BACKGROUND AND PURPOSE: Early Regression Index (ERITCP) is an image-based parameter based on tumor control probability modelling, that reported interesting results in predicting pathological complete response (pCR) after pre-operative chemoradiotherapy (CRT) in rectal cancer. This study aims to evaluate this parameter for Locally Advanced Cervical Cancer (LACC), considering not only T2-weighted but also diffusion-weighted (DW) Magnetic Resonance (MR) images, comparing it with other image-based parameters such as tumor volumes and apparent coefficient diffusion (ADC). MATERIALS AND METHODS: A total of 88 patients affected by LACC (FIGO IB2-IVA) and treated with CRT were enrolled. An MRI protocol consisting in two acquisitions (T2-w and DWI) in two times (before treatment and at mid-therapy) was applied. Gross Tumor Volume (GTV) was delineated and ERITCP was calculated for both imaging modalities. Surgery was performed for each patient after nCRT: pCR was considered in case of absence of any residual tumor cells. The predictive performance of ERITCP, GTV volumes (calculated on T2-w and DW MR images) and ADC parameters were evaluated in terms of area (AUC) under the Receiver Operating Characteristic (ROC) curve considering pCR and two-years survival parameters as clinical outcomes. RESULTS: ERITCP and GTV volumes calculated on DW MR images (ERIDWI and Vmid_DWI) significantly predict pCR (AUC = 0.77 and 0.75 respectively) with results superior to those observed considering T2-w MR images or ADC parameters. Significance was also reported in the prediction of 2-years local control and disease free-survival. CONCLUSION: This study identified ERITCP and Vmid as good predictor of pCR in case of LACC, especially if calculated considering DWI. Using these indicators, it is possible to early identify not responders and modifying the treatment, accordingly.

MRI Staging in Locally Advanced Vulvar Cancer: From Anatomy to Clinico-Radiological Findings. A Multidisciplinary VulCan Team Point of View.

MR imaging provides excellent spatial and contrast resolution to stage locally advanced vulvar cancer (LAVC) for tumor and nodal evaluation in order to facilitate the planning of treatment. Although there are no standard indications for how to estimate the clinical stage of International Federation of Gynecology and Obstetrics at diagnosis, MR imaging can depict the tumor and its extension to the vulvar region and adjacent organs, such as the vagina, urethra, and anus. Optimizing the MR imaging protocol and technique is fundamental for correct staging. The aim of this overview was to focus on the role of MR imaging in LAVC staging. We define vulvar anatomy and corresponding MR imaging findings, MR imaging protocol, and technique. Moreover, we describe the MR imaging findings of LAVC with example cases stage by stage. Key imaging findings based on signal intensity, diffusion restriction, and enhancement are portrayed to correctly identify and stage vulvar cancer. A structured report for LAVC staging is reported in order to give all necessary information to the clinicians and to facilitate MR imaging comprehension.

Pretreatment MRI Radiomics Based Response Prediction Model in Locally Advanced Cervical Cancer.

The aim of this study was to create a radiomics model for Locally Advanced Cervical Cancer (LACC) patients to predict pathological complete response (pCR) after neoadjuvant chemoradiotherapy (NACRT) analysing T2-weighted 1.5 T magnetic resonance imaging (MRI) acquired before treatment start. Patients with LACC and an International Federation of Gynecology and Obstetrics stage from IB2 to IVA at diagnosis were retrospectively enrolled for this study. All patients underwent NACRT, followed by radical surgery; pCR-assessed on surgical specimen-was defined as absence of any residual tumour. Finally, 1889 features were extracted from MR images; features showing statistical significance in predicting pCR at the univariate analysis were selected following an iterative method, which was ad-hoc developed for this study. Based on this method, 15 different classifiers were trained considering the most significant features selected. Model selection was carried out using the area under the receiver operating characteristic curve (AUC) as target metrics. One hundred eighty-three patients from two institutions were analysed. The model, showing the highest performance with an AUC of 0.80, was the random forest method initialised with default parameters. Radiomics appeared to be a reliable tool in pCR prediction for LACC patients undergoing NACRT, supporting the identification of patient risk groups, which paves treatment pathways tailored according to the predicted outcome.