Recruiting PHASE2 INTERVENTIONAL NCT06492486

Changing the dose mid-treatment — an AI workflow registered as radiation

Tata Memorial Centre Updated 2026-08-17

In radiotherapy for glioma, MRI and PET are performed twice during treatment and the dose altered according to what they show, with an AI workflow carrying out the alteration. The registered intervention is radiation.

Trial overview (primary data)

  • StatusRecruiting
  • ConditionsDiffuse Glioma, Glioblastoma, Adaptive Radiotherapy, Artificial Intelligence
  • InterventionsRADIATION: Adaptive radiotherapy
  • SponsorTata Memorial Centre
  • Target enrollment60 participants
  • Period2026-07-27 〜 2028-07-30

Key points

  • Gliomas are common primary brain tumors in adults, standardly treated by surgery followed by radiation and chemotherapy.
  • Standard practice delivers a uniform dose to the disease area with MRI before and after treatment.
  • This trial performs interval imaging with MRI and PET twice during the course of treatment, followed by dose modifications.
  • CT, MRI and PET are combined and specific doses delivered to specific areas based on PET, with modification carried out with the help of artificial intelligence.
  • The phase is 2, target enrolment 60, the sponsor Tata Memorial Centre, with completion expected 30 July 2028.

1Changing the plan while treatment runs

Standard radiotherapy delivers a uniform dose to the disease area, with MRI before and after. What this registration changes is the middle. Twice during the course, MRI and PET are performed and the dose altered according to what they show, and an AI workflow carries out that alteration.

2Standard practice against this trial

The aspectStandard institutional practiceThis trial
The doseUniform across the disease areaSpecific doses to specific areas, based on PET
When imaging happensBefore and after treatmentBefore, twice during, and after
Which imagesMRICT, MRI and PET combined
Dose modificationNoneCarried out with the help of artificial intelligence

What the AI carries is the workflow deciding how the dose changes. The registration states that creating an artificial intelligence framework in radiation oncology promises to improve the quality of workflow, treatment planning and delivery.

3What is expected to improve

  1. 1Tumor controlConcentrating dose on resistant areas is expected to control the tumor better
  2. 2ToxicityTreatment-related toxicities are expected to fall
  3. 3PrecisionDoses to organs at risk are adjusted as the disease changes
  4. 4ComplianceBetter treatment compliance is expected to follow from the above

Raising the dose and lowering it appear together. More where the disease resists, less where tissue must be spared. That is what adaptive radiotherapy means.

4The eight boxes side by side

The registered boxWhat the AI carriedThe registration
No intervention, observationalAnalysing recorded video after the factEmbryo morphokinetics (NCT07611448)
Diagnostic testBuilt into the process of testing itselfCervical cancer screening (NCT06042543)
DeviceAn approved product evaluated in practiceOsteoporosis diagnosis (NCT07620834)
BehaviouralOne part of the machinery supporting a programmeStoma and sleep (NCT07805226)
OtherA tool clinicians consultResidents and language models (NCT07199231)
ProcedureTaking over the operation of image acquisitionSecond-trimester ultrasound (NCT07286591)
DrugDeciding who enters the trialPostoperative opioid use (NCT07672015)
RadiationCarrying the workflow of dose modificationAdaptive radiotherapy for glioma (NCT06492486)

Set out together, what shows is that one thing called medical AI is recorded on the register in eight different ways. And the box settles the framework of assessment that applies. Phases cluster in the drug and radiation boxes and cover only 32 of 776 registrations (4.1%) overall.

Registrations as a device come to 14.6%, while the largest share, 26.3%, sits in the box marked other, which says nothing at all as a classification.

Anyone counting how AI is used in medicine from the register runs first into this instability of classification, because what is recorded is not what an AI does but which box the registrant chose.

Why it matters

One thing called medical AI is recorded on the trial register in eight different ways, and the box chosen settles which framework of assessment applies. Phases cluster in the drug and radiation boxes and reach only 32 of 776 registrations (4.1%); device registrations come to 14.6%, while the largest share sits in other, which classifies nothing. Counting how AI is used in medicine from the register means meeting this instability first.

FAQ

What is adaptive radiotherapy?
Modifying the radiation plan according to changes in disease seen on scans during treatment; here doses are altered from interval MRI and PET.
What does the AI carry?
The registration states that dose modification is done with its help, and that an AI framework promises to improve workflow, treatment planning and delivery.
What benefits are expected?
Better tumor control, reduced treatment-related toxicities, more precise delivery, dose adjustment for organs at risk, and better treatment compliance.

Sources (primary)

Source: ClinicalTrials.gov (U.S. NIH/NLM, public domain). This site does not provide medical advice. Verify the latest and exact details with the official source. This site is not endorsed or certified by the NIH/NLM.

#Clinical trials#Medical AI#Radiotherapy#United States#Trial registration
Disclaimer: This site independently summarizes and classifies information based on official data sources. Always verify the latest and accurate information with the official sources. Content on finance, health, legal, and security is information, not advice. This site is not an official website of the U.S. government.