Artificial Intelligence-Guided Radiotherapy Planning for Glioblastoma

NCT06657027 | Not Yet Recruiting

• 1 other identifier: PI-24-563-H
• Study Type: observational
• Target: 40
• Locations: 0 countries
• Sites: N/A

Brief Summary

The ARTPLAN-GLIO study aims to evaluate the feasibility and effectiveness of integrating artificial intelligence in personalized radiotherapy planning for glioblastomas. On the basis of previous work by our group, where a predictive model was developed from radiological characteristics extracted from MR images, this project will evaluate the use of tumor infiltration probability maps in radiotherapy planning. Currently, radiotherapy treatment uses margins defined by population studies, without considering the individual characteristics of the patients. Although 80% of recurrences occur in peritumoral areas close to the surgical margins, treatment volumes are not customized owing to the lack of techniques that distinguish between edema and infiltrated tumor tissue. Our recurrence probability maps address this limitation and could improve radiation planning. In this study, the volumes and doses of radiotherapy were adjusted according to the predictions of the model, with a focus on high-risk areas to optimize local control and reduce toxicity in healthy tissues. Survival results will be compared between patients treated with personalized AI-guided radiotherapy and a historical cohort with standard treatment. In addition, the safety of the approach will be evaluated by adverse event analysis. Finally, an accessible online platform with the potential to transform glioblastoma treatment and improve patient survival will be developed to implement this predictive model.

Conditions

Glioblastoma

Keywords

radiotherapy; oncology; artificial intelligence; glioblastoma; personalized radiotherapy; precision medicine

Outcome Measures

Primary Outcomes (1)

• Feasibility of AI-Guided Radiotherapy for Glioblastoma

  The primary outcome of the study is to assess the feasibility of integrating an AI-based predictive model into radiotherapy planning for patients with glioblastoma. The model uses radiomic features derived from multiparametric MRI to generate tumor infiltration probability maps, which guide the personalized adjustment of treatment volumes and doses. Feasibility will be determined by evaluating the successful integration of the AI model into clinical practice, the precision of the model in identifying areas of tumor infiltration, and the ability to implement personalized treatment plans in a routine clinical setting.

  12 months after the start of radiotherapy for the last enrolled patient.

Secondary Outcomes (3)

• Progression-Free Survival (PFS) at 1 Year

  12 months after the start of radiotherapy for each patient.

• Overall Survival (OS)

  24 months after the start of radiotherapy for each patient.

• Quality of Life

  12 months after the start of radiotherapy for each patient.

Study Arms (1)

AI-Guided Radiotherapy Cohort

This cohort includes patients with newly diagnosed IDH wild-type glioblastoma, grade 4, according to the 2021 WHO classification of Central Nervous System Tumors. Patients in this group will undergo personalized radiotherapy guided by artificial intelligence (AI) and multiparametric MRI, using predictive models to adjust treatment volumes and doses according to areas of tumor infiltration. The AI model, developed from radiomic characteristics of postoperative MRI, predicts tumor recurrence and infiltration, enabling targeted dose escalation to high-risk areas while minimizing radiation exposure to healthy tissues.

Eligibility Criteria

• Age: 15 Years+
• Sex: all
• Healthy Volunteers: No
• Age Groups: Child (0-17), Adult (18-64), Older Adult (65+)
• Sampling Method: Non-Probability Sample

Study Population

The study population consists of adult patients with newly diagnosed IDH wild-type glioblastoma, grade 4, as classified by the World Health Organization (WHO) 2021 Central Nervous System Tumor guidelines. Eligible participants must have undergone maximum safe tumor resection and be scheduled to receive radiotherapy.

You may qualify if:

• Patients with a recent diagnosis of IDH wild-type glioblastoma, grade 4 according to the Central Nervous System Tumors classification of the World Health Organization of 2021.
• Ability to undergo MRI studies.
• Performance status with Karnofsky Performance Status (KPS) ≥ 60.
• Life expectancy ≥ 12 weeks.
• Laboratory results within the following ranges, obtained in the 14 days prior to enrollment:
• Leucocitos ≥ 3,000/µL.
• Absolute neutrophils ≥ 1,500/µL.
• Plaquetas ≥ 75,000/µL.
• Hemoglobin ≥ 9.0 g/dL (transfusion is allowed to reach the minimum level).
• Glutamic-oxaloacetic transaminase (SGOT) ≤ 2 times the upper limit of normal.
• Bilirubin ≤ 2 times the upper limit of normal.
• Creatinina ≤ 1.5 mg/dL.
• Women of childbearing age must present a negative pregnancy test ≤ 14 days prior to enrollment.
• Ability to understand and sign the informed consent.
• Willingness to refrain from other cytotoxic or noncytotoxic therapies against the tumor during the protocol.

You may not qualify if:

• Presence of pacemakers, neurostimulators, cochlear implants, metal in ocular structures, or work history that compromise safety in MRI.
• Significant medical illnesses that may compromise tolerance to treatment, at the discretion of the investigator.
• History of invasive cancer in the last 3 years, with few exceptions.
• Active infections or serious intercurrent illnesses.
• Previous treatments with cytotoxic, noncytotoxic, experimental agents, or cranial radiation therapy.
• Maximum radiation target volume (GTV3) greater than 65 cc.

Sponsors & Collaborators

• Hospital del Rio Hortega: lead

Related Publications (1)

• Cepeda S, Luppino LT, Perez-Nunez A, Solheim O, Garcia-Garcia S, Velasco-Casares M, Karlberg A, Eikenes L, Sarabia R, Arrese I, Zamora T, Gonzalez P, Jimenez-Roldan L, Kuttner S. Predicting Regions of Local Recurrence in Glioblastomas Using Voxel-Based Radiomic Features of Multiparametric Postoperative MRI. Cancers (Basel). 2023 Mar 22;15(6):1894. doi: 10.3390/cancers15061894.

  PMID: 36980783 BACKGROUND

MeSH Terms

Conditions

Glioblastoma; Neoplasms

Condition Hierarchy (Ancestors)

Astrocytoma; Glioma; Neoplasms, Neuroepithelial; Neuroectodermal Tumors; Neoplasms, Germ Cell and Embryonal; Neoplasms by Histologic Type; Neoplasms, Glandular and Epithelial; Neoplasms, Nerve Tissue

Central Study Contacts

Santiago Cepeda Principal Investigator, MD., PhD

CONTACT

+34983420400; scepedac@saludcastillayleon.es

Olga Esteban Co-PI, MD

CONTACT

+34983420400; oestebans@saludcastillayleon.es

Study Design

• Study Type: observational
• Observational Model: COHORT
• Time Perspective: PROSPECTIVE
• Sponsor Type: OTHER
• Responsible Party: PRINCIPAL INVESTIGATOR
• PI Title: Attending Neurosurgeon

Study Record Dates

• First Submitted: October 15, 2024
• First Posted: October 24, 2024
• Study Start: January 1, 2025
• Primary Completion (Estimated): December 31, 2026
• Study Completion (Estimated): June 30, 2027
• Last Updated: October 24, 2024

Record last verified: 2024-10