Nuclear Medicine

SSTR PET tracers are interchangeable for meningioma, while PRRT offers modest disease control

Among 4,136 SSTR PET/CTs, 152 patients with 171 SSTR-positive meningiomas showed high tumour-to-meningeal contrast with all tracers. [18F]AlF-NOTA-octreotide, [68Ga]Ga-DOTATATE, and [68Ga]Ga-DOTATOC had robust uptake without significant quantitative differences across delineation methods. Incidental SSTR-avid meningiomas occurred in 3.7% of patients scanned for unrelated indications, relevant for incidentaloma interpretation and follow-up advice. In 12 heavily pretreated patients, PRRT achieved 50% disease control, median PFS 8 months, and median OS 20 months. 1

Repeat 18F-flotufolastat PET has substantial yield after an initially negative scan in BCR

In 101 men with biochemical recurrence after prostatectomy and an initial negative 18F-flotufolastat PET, 57% had lesions on the second scan. Detection rates rose with higher PSA, greater absolute and relative ΔPSA, higher PSA velocity, and shorter PSA doubling time. ROC analysis suggested PSA > 0.82 ng/mL, absolute ΔPSA > 0.50 ng/mL, and relative ΔPSA > 100% as useful cut-offs. PSA velocity > 0.30 ng/mL/year and PSA doubling time ≤ 17 months also discriminated PET-positive from PET-negative patients. 2

82Rb PET perfusion, flow, and calcium jointly refine coronary event risk

Among 5,285 patients undergoing stress–rest 82Rb PET/CT, 683 cardiac events occurred during roughly 43 months of follow-up. In patients with prior CAD, angina, stress TPD ≥ 5%, reduced MFR (< 2), and impaired coronary flow capacity independently predicted events. In patients without known CAD, diabetes, CAC ≥ 400, abnormal stress TPD, reduced MFR, and impaired coronary flow capacity were independent predictors. Weibull modeling showed highest event probability when known CAD, abnormal stress TPD, reduced MFR, and impaired coronary flow capacity coexisted. 3

Machine learning defines a bilirubin threshold for REILD risk in HCC SIRT

In 138 HCC patients treated with SIRT, machine-learning models consistently highlighted baseline bilirubin as the dominant predictor of REILD. A bilirubin threshold of 26.5 µmol/L (1.55 mg/dL) was associated with increased probability of radioembolization-induced liver disease. In 136 patients analysed for response, tumor dose was the feature most frequently selected by models, though overall predictive accuracy was limited. Findings support using a tighter bilirubin cut-off for patient selection and emphasise that tumor dose alone cannot reliably predict response. 4