A Prospective Controlled Trial of Large Language Model–Based Simplification of Oncologic PET/CT Reports for Patients with Cancer

A Prospective Controlled Trial of Large Language Model–Based Simplification of Oncologic PET/CT Reports for Patients with Cancer - SIMPLE-PET Trial

Status

Active, not recruiting

Phases

Unknown

Study type

Observational

Source

DRKS

Registry ID

DRKS00039421

Enrollment

400

Registered

2026-05-07

Start date

2026-07-01

Completion date

Unknown

Last updated

2026-06-01

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

Improved Understanding Through AI-Simplified Reports. The study cohort comprises patients with metastatic prostate cancer (ICD-10: C61), neuroendocrine tumors (ICD-10: C7A), malignant melanoma (ICD-10: C43), and adenocarcinomas of the esophagogastric junction (ICD-10: C15–C16). C61 C73-C75 C43 C15 C16

Interventions

Group 1: After written informed consent, eligible oncology patients are randomized in a 1:1 ratio to either the intervention or control group. The study is based on a PET/CT examination that has alrea

no follow-up is planned.

Eligibility

Sex/Gender

All

Age

18 Years to No maximum

Inclusion criteria

Inclusion criteria: - Age = 18 years - Histologically or clinically confirmed oncological disease - PET/CT examination performed as part of routine clinical care - Availability of at least one prior PET/CT with a complete report - Ability to complete the questionnaire independently - Sufficient understanding of the study content

Exclusion criteria

Exclusion criteria: - Lack of valid informed consent - Cognitive, psychological, or physical impairments preventing participation - Insufficient language proficiency without adequate translation support - Missing or incomplete report documentation - No prior PET/CT examination - Prior participation in the study

Design outcomes

Primary

Measure	Time frame
Primary Endpoint Patient-reported comprehension of the PET/CT report (primary endpoint) Definition: Composite score derived from multiple Likert-scaled items (e.g., comprehensibility, clarity, ease of understanding). Scale: Continuous mean score; higher values indicate better subjective understanding. Timepoint of assessment: Single assessment immediately after reading the respective report. Data collection: Standardized paper-based questionnaire; subsequent pseudonymized transfer into the study database.	—

Measure

Time frame

Primary Endpoint Patient-reported comprehension of the PET/CT report (primary endpoint) Definition: Composite score derived from multiple Likert-scaled items (e.g., comprehensibility, clarity, ease of understanding). Scale: Continuous mean score; higher values indicate better subjective understanding. Timepoint of assessment: Single assessment immediately after reading the respective report. Data collection: Standardized paper-based questionnaire; subsequent pseudonymized transfer into the study database.

—

Secondary

Measure	Time frame
Secondary Endpoints 1. Cognitive load Likert-based score assessing perceived mental effort. Assessed immediately after reading the report. 2. Reading time Time (in minutes) required to read the report (self-reported or measured). Continuous variable. 3. Perception of the report Likert scales assessing usefulness, clarity, and trust. Analyzed as a composite score. 4. Anxiety/concern after reading Likert scale assessing emotional response. 5. Objective text characteristics Readability metrics (Flesch Reading Ease, Flesch-Kincaid Index). Word count of the report. 6. Physician-documented corrections of the LLM version Number and severity of content-related deviations.	—

Measure

Time frame

Secondary Endpoints 1. Cognitive load Likert-based score assessing perceived mental effort. Assessed immediately after reading the report. 2. Reading time Time (in minutes) required to read the report (self-reported or measured). Continuous variable. 3. Perception of the report Likert scales assessing usefulness, clarity, and trust. Analyzed as a composite score. 4. Anxiety/concern after reading Likert scale assessing emotional response. 5. Objective text characteristics Readability metrics (Flesch Reading Ease, Flesch-Kincaid Index). Word count of the report. 6. Physician-documented corrections of the LLM version Number and severity of content-related deviations.

—

Countries

Germany

Contacts

Public ContactLisa Steinhelfer

Institut für diagnostische und interventionelle Neuroradiologie, TUM Klinikum Rechts der Isar

lisa.steinhelfer@tum.de+498941407064

Outcome results

None listed

Source: DRKS (via WHO ICTRP) · Data processed: Jun 11, 2026