Chemoradiotherapy, Locally Advanced Cervical Cancer, Machine Learning, Radiomics, Ultrasound Imaging, Uterine Cervical Neoplasms
Conditions
Keywords
Locally Advanced Cervical Cancer, Cervical Cancer, Uterine Cervical Neoplasms, Radiomics, Ultrasound, Ultrasound Imaging, Machine Learning, Gynecologic Oncology
Brief summary
This is a retrospective, multicenter observational study aimed at evaluating the role of ultrasound-based radiomics in patients with locally advanced cervical cancer (LACC). The study will analyze pre-treatment ultrasound images to identify radiomic features that may predict treatment response and disease recurrence. A total of 220 patients treated with exclusive chemoradiotherapy or neoadjuvant chemoradiotherapy followed by radical surgery between 2011 and 2024 will be included. Using clinical and imaging data, machine learning models will be developed to distinguish between responders and non-responders, and to identify patients at higher risk of relapse. The goal is to improve personalized care in LACC by integrating radiomic analysis into treatment planning and follow-up strategies.
Interventions
OTHERRadiomic Analysis of Ultrasound Images
Quantitative analysis of pre-treatment ultrasound images of the primary cervical tumor to extract radiomic features. These features will be used to develop and validate machine learning models for predicting treatment response and disease relapse in patients with locally advanced cervical cancer (LACC).
Sponsors
Fondazione Policlinico Universitario Agostino Gemelli IRCCS
Study design
Observational model
COHORT
Time perspective
RETROSPECTIVE
Eligibility
Sex/Gender
FEMALE
Age
18 Years to No maximum
Healthy volunteers
No
Inclusion criteria
Female patients aged ≥18 years Histologically confirmed diagnosis of locally advanced cervical cancer (FIGO 2018 stage IB3-IVA, excluding IIA1) Histologic subtypes: squamous cell carcinoma, adenocarcinoma, or adenosquamous carcinoma Underwent transvaginal or transrectal ultrasound prior to treatment At least one pre-treatment DICOM ultrasound image of the primary tumor available Treated with either exclusive chemoradiotherapy or neoadjuvant chemoradiotherapy followed by radical surgery Completed at least 12 months of follow-up after primary treatment Signed informed consent (or equivalent declaration)
Exclusion criteria
Age \<18 years Only printed ultrasound images available Ultrasound images with poor tumor visualization or with text/markers obscuring the tumor
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Performance of ultrasound-based radiomic models in predicting treatment response in patients with locally advanced cervical cancer. | Up to 12 months after primary treatment. | Evaluation of the diagnostic accuracy (AUC, sensitivity, specificity, F1-score) of radiomic models based on pre-treatment ultrasound images in distinguishing responder vs. non-responder patients to primary treatment (exclusive chemoradiotherapy or neoadjuvant chemoradiotherapy followed by radical surgery). Models will be developed and validated using retrospective data. |
Countries
Italy
Outcome results
None listed