Describe Typical Gray Scale/Color Doppler Ultrasound Features of Gestational Trophoblastic Neoplasia at the US Examination, To Assess if There Are Differences at the Baseline US Scan Between Low-risk and High-risk Patients, To Identify Ultrasound Predictors of Resistance to First-line Single Drug Chemotherapy in Low-risk GTN Patients, To Investigate if Machine Learning (ML) Algorithms (Radiomics, Deep Neural Networks (DNN), and Other Machine Learning Algorithms
Conditions
Keywords
Gestational Trophoblastic Neoplasia, Ultrasound, Gestational
Brief summary
Gestational trophoblast diseases are characterized by abnormal proliferation of trophoblastic tissue, which can occur consequent to any pregnancy event. Pre-malignant forms (partial hydatiform mola and complete mola) and malignant forms are distinguished, the latter also known as gestational trophoblast neoplasms (invasive hydatiform mola, choriocarcinoma, placental site trophoblastic tumor, and epithelioid trophoblastic tumor). These are neoplasms associated with good prognosis, amenable to conservative treatment and highly sensitive to chemotherapy. The identification of ultrasonographic and echofluximetric features typical of malignant forms as well as the identification of ultrasonographic parameters predictive of chemoresistance to single-drug treatments could help improve their management.
Detailed description
Primary aim 1. To describe typical gray scale and color Doppler ultrasound features of Gestational Troph-oblastic Neoplasia at the baseline US examination; the results will be presented as in Table 4 2. To assess if there are differences at the baseline US scan between low-risk and high-risk patients and, in the low-risk group, between responders and non-responders to the single drug therapy; the results will be presented as in Table 5 3. To identify ultrasound predictors of resistance to first-line single drug chemotherapy in low-risk GTN patients; statistical analysis will be performed using logistic regression mod-eling 4. To investigate if machine learning (ML) algorithms (radiomics, deep neural networks (DNN), and other machine learning algorithms), applied on ultrasound images, can dis-criminate between low-risk patients that will or will not develop chemoresistance to single-drug therapy.To investigate if ML algorithms, combined with the conventional risk-score improve prediction of resistance to first-line single drug chemotherapy in low risk GTN pa-tients. Secondary aims: I. To describe any longitudinal changes in ultrasound characteristics of Gestational Tropho-blastic Neoplasia during treatment. I. To describe any longitudinal changes in ultrasound characteristics of Gestational Tropho-blastic Neoplasia after stopping treatment, when the uterine lesion is still visible despite the normalization of hCG.
Interventions
None listed
Sponsors
Fondazione Policlinico Universitario Agostino Gemelli IRCCS
Study design
Observational model
OTHER
Time perspective
PROSPECTIVE
Eligibility
Sex/Gender
FEMALE
Age
18 Years to No maximum
Healthy volunteers
No
Inclusion criteria
* Patients with a diagnosis of Gestational Trophoblastic Neoplasia (GTN), after any kind of pregnancy: 1. Histological diagnosis of Placental Site Trophoblastic Tumor or Epithelioid Trophoblastic Tumor 2. International Federation of Gynecology and Obstetrics (FIGO) 2002 criteria for the diagnosis of Gestational Trophoblastic Neoplasia I. Histological evidence of invasive mole or choriocarcinoma II. Rising hCG after evacuation of hydatidiform mole, i.e. two consecutive rises in hCG of 10% or greater over at least 2 weeks (at least 10% between day 1 and 7 and then again at least 10% between day 7 and 14) III. Plateau of hCG after evacuation of hydatidiform mole, i.e. four or more equivalent values of hCG over at least 3 weeks (days 1, 7, 14 and 21) IV. The hCG level remains elevated for 6 months or more after evacuation even if decreasing 3. Charing Cross Hospital (UK) additional criteria for the diagnosis of Gestational Trophoblastic Neoplasia: I. Heavy vaginal bleeding or evidence of gastrointestinal or intraperitoneal hemorrhage in patients with histological diagnosis of GTD II. Serum hCG concentration of 20 000 IU/L or more 4 weeks or more after evacuation of a mole, because of the risk of uterine perforation III. Evidence of metastases in brain, liver, or gastrointestinal tract, or radiological opacities larger than 2 cm on chest radiograph in patients with GTD. Referral Centers for Gestational Trophoblastic Disease with at least 5 cases of GTN per year, that routinely use ultrasound in clinical evaluation, could participate in the study adhering to the: * BASIC PROTOCOL if they are able to guarantee only the baseline ultrasound evaluation at the diagnosis but NOT the follow-up ultrasound scans during and after chemotherapy * FULL PROTOCOL if they are able to guarantee both the baseline and all the follow-up ultrasound examination according to the complete protocol schedule.
Exclusion criteria
*
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| typical features of Gestational Trophoblastic Neoplasia | 30 minutes | To describe typical gray scale and color Doppler ultrasound features of Gestational Trophoblastic Neoplasia at the baseline US examination |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| differences at the baseline US scan between low-risk and high-risk patients | 30 minutes | US evaluation |
Other
| Measure | Time frame | Description |
|---|---|---|
| changes characteristics of Gestational Trophoblastic Neoplasia during treatment | 30 minutes | US longitudinal follow-up |
Countries
Italy
Contacts
Primary ContactAntonia Carla Testa
Outcome results
None listed