Oral Presentation 37th TROG Cancer Research Annual Scientific Meeting 2025

Imaging biomarkers and quantitative imaging – the new standard for clinical trial end points? (#49)

Annette Haworth 1 2 , Yu-Feng Wang 1 2 3 , Sirisha Tadimalla 1 2 , Lois Holloway 3 , Martin A Ebert 4 , Joel Poder 5 , Hayley M Reynolds 6
  1. Sydney West Radiation Oncology, Western Sydney LHD, Westmead, New South Wales, Australia
  2. University of Sydney, Camperdown, NSW, Australia
  3. Ingham Institute for Applied Medical Research, Liverpool, New South Wales, Australia
  4. School of Physics, Mathematics and Computing, The University of Western Australia, Crawley, WA, Australia
  5. St George Hospital Cancer Care Centre, Kogarah, New South Wales, Australia
  6. Auckland Bioengineering Institute, The University of Auckland, Auckland, ., New Zealand

Background: A reliable measure of treatment effectiveness is critical for evaluation of clinical trial success. Subjective imaging assessment, surrogates that poorly correlate with meaningful clinical endpoints, and endpoints that take a long time to achieve are undesirable. Quantitative imaging has the potential to provide objective treatment response measures, as it allows the extraction of measurable data from medical images and can characterise biological processes. However, the use of quantitative imaging as a reliable metric in multi-centre trials poses challenges. The Sequential Imaging for Biologically Targeted Radiation Therapy (SI-BiRT) trial investigated the challenges in developing treatment response imaging biomarkers for prostate cancer.

Aims: To define the challenges of developing robust imaging biomarkers for treatment response.

Methods: The SI-BiRT(1) trial recruited 21 participants from 3 hospitals who were imaged using MRI and PET before, and at 18-months post-radiation therapy. Additional MRI data were collected at 6- and 12-months post treatment. Reproducibility and repeatability of the data were assessed through phantom studies and test-retest studies involving healthy volunteers and prostate cancer patients. Blood-based PSA measurements followed standard clinical protocols.

Results: All patients were imaged using the same 3T MRI scanner across all time points. Phantom studies confirmed inter-scanner reproducibility and intra-scanner stability. Test-retest studies revealed variability in the quantitative imaging parameters, with uncertainty influenced by factors such as the location within the prostate. The study identified multiple quantitative MRI parameters that were robust to measurement uncertainties and statistically different to baseline values across the measurement timepoints. Correlation with clinical outcomes is pending.

Conclusions: Quantitative imaging offers potential to reveal biological characteristics that could provide early, accurate assessments of treatment response. Clinical validation is essential, with future trials offering the opportunity to generate the necessary data.