Reference Organ Selection for F-18 PSMA PET Scan Interpretation
For F-18 PSMA PET scans, the parotid gland is the preferred reference organ for interpreting PSMA-avid lesions, as it demonstrates superior test-retest repeatability and lower inter-reader variability compared to the spleen.
Evidence Supporting Parotid Gland as Reference
Repeatability and Reliability Data
The parotid gland demonstrates high repeatability for SUVmean measurements (wCOV 9.0-14.3%) on F-18-DCFPyL PET scans, making it a reliable reference standard 1
Minimal inter-reader variability exists for whole parotid gland measurements (bias ±0.31 for PET/CT and ±0.24 for PET/MR), supporting its use as a consistent reference organ 2
The parotid gland shows consistent physiologic PSMA-ligand uptake with mean SUVmax of 12.3 ± 3.9, providing a stable background for comparison 3
Tumor-to-Background Ratio Performance
Tumor-to-parotid ratios demonstrate significantly higher values compared to other reference organs (p = 0.008), improving lesion detection sensitivity 4
The parotid gland provides better tumor-to-background discrimination than spleen (p = 0.002) or kidney (p = 0.006) for F-18-DCFPyL imaging 4
Limitations of Spleen as Reference
The spleen shows acceptable repeatability (wCOV 9.0-14.3% for SUVmean) but does not offer superior performance over the parotid gland 1
Inter-subject variability is higher for parotid gland compared to liver or blood pool, but this does not diminish its utility as a reference standard when using whole-gland measurements 2
Practical Implementation Guidelines
Measurement Technique
Use whole parotid gland ROI with 40% SUV threshold methodology rather than small 1-cm spherical ROIs, as the latter show greater inter-reader variability 2
Measure SUVmean rather than SUVmax for reference organ quantification, as SUVmean demonstrates superior repeatability for large organs including parotid glands 1
Alternative Reference Organs
The liver can serve as an acceptable alternative reference organ, showing minimal inter-reader variability (bias ±0.82 for PET/CT) and high repeatability 2
Blood pool measurements show excellent inter-reader agreement but may not provide optimal tumor-to-background ratios for all lesion types 2
Clinical Context Considerations
For standardized reporting frameworks (PROMISE, E-PSMA), both liver and parotid glands are utilized as reference organs for patient selection and scan interpretation 1
The parotid gland's consistent high PSMA-ligand uptake (SUVmax 5.2-22.9 range) makes it particularly useful for identifying true PSMA-avid lesions versus background activity 3
Important Caveats
Avoid using small ROIs in the parotid gland, as these introduce significant measurement variability that can compromise interpretation accuracy 2
Be aware that parotid gland uptake can be affected by prior radiation therapy or salivary gland pathology, which may necessitate using liver as the reference organ in these specific cases 3
For F-18-fluciclovine (a different amino acid tracer), the parotid gland is specifically recommended as the reference organ due to very low physiological brain uptake, though this tracer is distinct from PSMA agents 5