A brand new machine-learning method for prostate-specific membrane antigen (PSMA) therapy of metastatic castration-resistant prostate most cancers (mCRPC) may estimate radiation dose to tumors and wholesome organs earlier than remedy begins. Utilizing information already obtainable from pre-therapy PET/CT scans, this novel prediction instrument may assist personalize therapy plans, enhance affected person choice, and scale back toxicity danger. This analysis was offered on the Society of Nuclear Drugs and Molecular Imaging 2026 Annual Assembly.
Dosimetry is important for optimizing ⁷⁷Lu-PSMA radiopharmaceutical remedy in mCRPC. At present, post-therapy imaging is often used to calculate dosimetry; nonetheless, it’s time-consuming and resource-intensive. Pre-therapy PET/CT presents a possibility to evaluate potential therapy effectiveness and danger earlier than remedy.
18F-PSMA PET/CT is already routinely carried out and extensively obtainable in prostate most cancers sufferers, however its potential to foretell therapy radiation dose has not beforehand been explored. Our examine sought to find out if data already obtainable from these scans may information therapy planning earlier than remedy begins and help extra customized care.”
Amit Nautiyal, PhD, scientist and Nationwide Institute for Well being and Care Analysis (NIHR) fellow at College Hospital Southampton and the College of Southampton, United Kingdom
On this proof-of-concept examine, 9 sufferers with mCRPC referred for ⁷⁷Lu-PSMA radiopharmaceutical remedy had been included, contributing 57 tumors, 36 salivary glands, and 18 kidneys for evaluation. Researchers developed a machine studying combined results mannequin to foretell absorbed doses in tumors and organs. Predictors included uptake-based PET metrics, radiomic options, and medical biomarkers. Predictive estimates had been in contrast with dosimetry calculated after one cycle of ⁷⁷Lu-PSMA remedy to evaluate accuracy.
The pre-therapy 18F-PSMA PET/CT-based machine studying mannequin confirmed a promising capability to foretell tumor and organ absorbed dose. By combining uptake options, radiomics, and medical biomarkers whereas accounting for patient-level variability, the mannequin reveals potential for utilizing pre-therapy data to foretell post-therapy dosimetry.
“If validated in bigger research, this method could enhance affected person choice and help higher decision-making throughout pre-treatment evaluation, serving to to optimize ⁷⁷Lu-PSMA remedy for particular person sufferers. Extra broadly, it highlights how imaging can transfer past prognosis to actively guiding customized therapy,” stated Nautiyal.
This proof-of-concept analysis is a part of a deliberate five-year program aimed toward amassing extra information and growing a sturdy, validated mannequin. This work was supported by the NIHR within the, United Kingdom. Future work will give attention to bigger, multi-center cohorts to refine pre-therapy absorbed dose predictions and to carry out unbiased validation to help affected person stratification for customized ⁷⁷Lu-PSMA radiopharmaceutical remedy in medical follow.
