Prediction of Clinical Response and Outcome in Uterine Cervix Cancer

Completed

• Conditions: Cervix Cancer

• Registration Number: NCT01764217
• Lead Sponsor: Maastricht Radiation Oncology

Brief Summary

Observational study based on the routine clinical treatment and diagnostic course, to correlate imaging features with outcome objectives. Outcome will evaluated as clinical response to the standard treatment and as recurrence and survival in the follow up. The study hypothesis is that data extracted form FDG-PETCT used in the routine clinical practice can predict outcomes following standard treatment.

Detailed Description

This study will prospectively collect patients undergoing to the standard diagnostic and treatment protocol in Maastro Clinic. Any difference in the normal procedure will be adopted. The aim is to extrapolate form the PET images some features of the metabolic tumor activity to associate with different outcomes and tumor behaviours.

Study Timeline

• Study Start: 2006-07
• Primary Completion: 2006-07
• Study First Submitted: 2012-11-12
• Study Completion: 2013-01
• Study First Submitted QC: 2013-01-07
• Study First Posted: 2013-01-09
• Status Verified: 2016-03
• Last Update Submitted: 2016-03-29
• Last Update Posted: 2016-03-30

Recruitment & Eligibility

• Status: COMPLETED
• Sex: Female
• Target Recruitment: 100

Inclusion Criteria

• Histologically confirmed cervix carcinoma (all subtypes)
• Tumor Stages FIGO IB - IVA
• Scheduled for primary curative radiotherapy (either or not combined with chemotherapy or hyperthermia)
• pre treatment FDG PETCT
• The patient is willing and capable to comply with study procedures
• 18 years or older
• Written informed consent to the treatment

Exclusion Criteria

• Recent (< 3 months) myocardial infarction
• Uncontrolled infectious disease
• Pregnant or breast feeding and/or not willing to take adequate contraceptive measures during the study
• Previous surgery to the Cervix
• Previous radiation to the Cervix

Study & Design

• Study Type: OBSERVATIONAL
• Study Design: Not specified

Primary Outcome Measures

Name	Time	Method
Change of SUV-related tumor characteristics predicting recurrence	Changes of parameters will be calculated on the pretreatment scan, than in average 2.5 months after the last radiotherapy session, than at least each 6 month for the first 2 years, eventually shortening the interval if clinically needed	* Standard Uptake Value Max (SUV- defined as the ratio of tissue radioactivity concentration (e.g. in MBq/kg=kBq/g) at time t, c(t), and the injected activity ( in MBq) at the time of injection (t=0) divided by the body weight in kg),; * Metabolic Volume (MV) calculated in cc:volume of the evaluable metabolic activity on the PET scan calculated in a specific Region of Interest (ROI) semiautomatically delineated on the primary tumor in the uterine Cervix

Secondary Outcome Measures

Name	Time	Method
Interobserver variability of Gross Tumor Volume (GTV) contours	GTV's will be delineated 2 weeks after the end of accrual	* GTV volume in cc contoured by 5 different observers on pretreatment scan: the difference in cc between each contour obtained will be scored; * GTV Overlapping fraction rate: the overlapping rate of GTV volume between contours
Change of SUV-related tumor characteristics predicting clinical overall response	Changes of parameters will be calculated on the pretreatment scan, and in average 2.5 months after the last radiotherapy session.	* Standard Uptake Value Max (SUV- defined as the ratio of tissue radioactivity concentration (e.g. in MBq/kg=kBq/g) at time t, c(t), and the injected activity ( in MBq) at the time of injection (t=0) divided by the body weight in kg),; * Ratio of Pre/Post treatment SUV MAX,
Radiomics Features	Radiomics features will be evaluated on the preteratment CT-fdg PET scan in average at least 2 weeks after the end of the accrual.	* We will apply a high throughput approach to convert medical images to minable data, where it is hypothesized that it will improve tumor characterization and treatment outcome prediction.; * Extracted imaging features consist firstly of global properties, providing information on the first order histogram of voxel intensity values within the tumor VOI.; * Local and regional textural features describing patterns and spatial distribution of voxel intensities, are calculated from respectively gray level co-occurrence and gray level run-length matrix representations. Images will be discretized before texture analysis, which allows for a direct comparison of all calculated textural features between patients. Co-occurrence and gray level run-length matrices are determined considering 26-connected voxels (i.e. voxels were considered to be neighbors in all 13 directions in three dimensions) and a distance of 1 between consecutive voxels. Features derived from the co-occurrence and gray

Trial Locations

Locations (1)

Philippe Lambin; 🇳🇱 Maastricht, Limburg, Netherlands