Splenic Uptake Variability Challenges Standardized [18F]PSMA-1007 PET Interpretation

A multicenter study finds that 15% of patients show higher tracer uptake in the spleen than parotid glands, complicating lesion scoring.

For Doctors in a Hurry

Clinicians lack standardized reference values for prostate-specific membrane antigen PET imaging, complicating the interpretation of tumor uptake levels.
The researchers analyzed 102 PET/CT scans from patients with biochemical recurrence to establish normal uptake ranges in reference tissues.
Median maximum standardized uptake values were 15.8 for the spleen and 26.6 for the parotid glands across the cohort.
The study found that 15 percent of patients exhibited higher splenic uptake than parotid gland uptake, complicating standardized scoring protocols.
Physicians should exercise caution when using splenic uptake for lesion categorization until further research clarifies these unexpected interpatient variations.

Standardizing Interpretation in PSMA PET Imaging

Positron emission tomography (PET) targeting prostate-specific membrane antigen (PSMA) has emerged as a standard clinical tool for evaluating biochemical recurrence, defined as a rising prostate-specific antigen level following primary treatment, in prostate cancer patients [1]. Meta-analytic data involving 8,119 patients demonstrate that PSMA PET achieves positivity rates of 48% even at low prostate-specific antigen (PSA) levels between 0.2 and 0.5 ng/ml, frequently localizing disease that conventional imaging fails to detect [1, 2]. To maintain diagnostic consistency, clinicians utilize standardized reporting frameworks such as the Prostate Cancer Molecular Imaging Standardized Evaluation (PROMISE), which employs reference tissues to categorize tracer uptake intensity [3]. However, significant variability exists in how these scans are performed and reported across different centers, which may affect the reproducibility of lesion scoring [4]. A recent multicenter analysis now examines how fluctuations in these reference standards might impact the accuracy of clinical staging and subsequent management decisions.

Quantifying Reference Tissue Variability

The researchers conducted a multicenter analysis of 102 [18F]PSMA-1007 PET/CT scans from patients with biochemical recurrence across three hospitals to establish a reliable baseline for tracer distribution. The study investigated the normal range and interpatient variability of tracer uptake in reference tissues, which is essential for the consistent application of the PROMISE framework. To quantify these levels, the authors measured standardized uptake values (SUV), a metric that accounts for the injected dose and patient body weight to allow for comparisons between different patients. They specifically recorded the SUV max (the highest radioactivity concentration measured in a single pixel within a region of interest) and the SUV mean (the average radioactivity concentration within that entire region) for the blood pool, liver, spleen, and parotid glands. All results were reported as a median with the 5th and 95th percentiles to define the expected clinical range, providing a statistical window for what constitutes typical physiological uptake.

Clinical Implications for Lesion Categorization

Because [18F]PSMA-1007 has primarily hepatobiliary excretion, meaning it is cleared through the liver and bile rather than the kidneys, the liver often exhibits high and variable background activity that can interfere with the detection of nearby lesions. Consequently, the study suggests using the spleen as a reference tissue instead of the liver for this specific radiopharmaceutical. The data showed that the median SUV max of the blood pool was 1.2 (5th to 95th percentile range: 0.6 to 2.2). In the spleen, the median SUV max was 15.8 (range: 8.9 to 26.2), while the median SUV mean was 11.9 (range: 6.6 to 20.7). These values provide a baseline for clinicians to determine whether a suspected lesion's uptake is significantly higher than normal physiological background, which is the primary method for identifying metastatic sites. The parotid glands, which serve as the high-intensity reference point in standardized scoring, demonstrated a median SUV max of 26.6 (5th to 95th percentile range: 14.5 to 40.4) and a median SUV mean of 19.3 (range: 11.2 to 32.1). The wide distribution between the 5th and 95th percentiles across all tissues underscores the significant variability between individual patients, suggesting that a "one size fits all" approach to uptake intensity may be flawed.

The study identified a significant deviation from the expected hierarchy of tracer distribution that directly impacts the utility of standardized scoring systems. In 15 out of 102 patients (15%), the SUV max was higher in the spleen than in the parotid glands, while 20 out of 102 patients (20%) exhibited an SUV mean in the spleen that exceeded the parotid glands. This inversion of uptake levels is clinically significant because the PROMISE framework relies on the parotid glands serving as the highest-intensity reference point for PSMA expression. When splenic uptake surpasses that of the parotid glands, the intended application of the PROMISE framework scoring cannot be applied as designed, as the reference scale for determining lesion intensity becomes inconsistent. The researchers noted that the reasons for this unexpectedly high splenic uptake relative to the parotid glands are currently unexplained, lacking a clear physiological or pathological driver in the patient cohort. For the practicing clinician, this variability means that a notable subgroup (15%) of patients may present with scans where the standard benchmarks for lesion categorization are unreliable. Because these reference tissues are used to determine whether a suspicious finding represents low, intermediate, or high PSMA expression, an atypical splenic baseline could lead to the mischaracterization of metastatic disease. Until further studies clarify the mechanisms behind this uptake pattern, clinicians are advised to exercise caution when interpreting [18F]PSMA-1007 PET scans in patients where the spleen appears more intense than the salivary glands to avoid potential staging errors.

Study Info

Reference tissue uptake of [18F]PSMA-1007 in positron emission tomography of recurrent prostate cancer

Bendik Skarre Abrahamsen, Ingerid Skjei Knudtsen, Andreas Julius Tulipan, Eivor Hernes, et al.

Journal European Radiology

Published April 23, 2026