The integration of prostate-specific membrane antigen positron emission tomography/computed tomography into clinical practice has substantially changed how metastatic castration-sensitive prostate cancer is detected and classified. However, the implications of this transition for real-world treatment selection and survival interpretation remain incompletely understood.
Findings from the Dutch CAPRI-3 registry showed that the use of PSMA PET/CT for staging metastatic castration-sensitive prostate cancer increased rapidly between 2016 and 2021. Patients staged with PSMA PET/CT generally had a lower disease burden and more favorable clinical characteristics than those staged using conventional imaging alone.
The study also identified important differences in treatment patterns between the imaging groups. Nevertheless, the investigators emphasized that the longer observed survival among patients undergoing PSMA PET/CT should not be interpreted as evidence that the imaging modality itself improves survival.
The article, titled “Real-world use of PSMA imaging in metastatic castration-sensitive prostate cancer: Findings from the CAPRI-3 registry,” was published in the European Journal of Nuclear Medicine and Molecular Imaging on July 10, 2026.
Authors: K. J. M. van der Velden, D. Bosch, Z. Vassilev, A. M. Bergman, A. C. M. van den Bergh, M. J. van der Doelen, A. J. M. van den Eertwegh, J. Lavalaye, R. J. A. van Moorselaar, D. M. Somford, M. Tascilar, A. Raval, C. A. Uyl-de Groot, M. C. P. Kuppen, and I. M. van Oort.
Evaluating PSMA PET/CT in Routine Clinical Practice
CAPRI-3 is an ongoing observational, retrospective, multicenter registry that includes patients diagnosed with metastatic castration-sensitive prostate cancer across 19 hospitals in the Netherlands.
The current analysis identified 4,601 patients diagnosed between 2016 and 2021. Imaging information at the time of metastatic disease diagnosis was available for 3,644 patients, who formed the population used for comparisons of imaging and treatment patterns.
Staging performed within 90 days of diagnosis was classified as either:
- Conventional imaging alone, including CT, bone scintigraphy, and/or MRI
- PSMA PET/CT, performed either alone or in combination with conventional imaging
The primary objective was to describe changes in imaging utilization at the diagnosis of metastatic castration-sensitive prostate cancer. Secondary objectives included evaluating systemic treatment patterns and overall survival.
Disease volume was classified according to the CHAARTED criteria. When findings differed between imaging modalities, the final assessment made by the local multidisciplinary team was used.
PSMA PET/CT Use Increased From 8% to 62%
Of the 3,644 patients with available imaging data, 2,312, or 63%, were staged using conventional imaging alone. The remaining 1,332 patients, or 37%, underwent PSMA PET/CT, either with or without conventional imaging.
The use of PSMA PET/CT increased markedly over the study period. It was used in 8% of patients in 2016 and 62% in 2021. Over the same period, the proportion staged exclusively with conventional imaging decreased from 92% to 38%.
Adoption occurred in both synchronous and metachronous metastatic disease. Among patients with metachronous disease, PSMA PET/CT use increased from 17% in 2016 to 63% in 2021. Among patients with synchronous disease, its use increased from 4% to 61%.
The investigators noted that uptake accelerated around several important clinical and regulatory milestones, including the 2019 European Association of Urology guideline recommendation, the 2020 Amsterdam EAU/EANM consensus panel, and European Medicines Agency approval. However, the trend toward greater PSMA PET/CT use had already begun before these milestones.
A Lower-Burden Population
Patients staged with PSMA PET/CT represented a clinically distinct population from those staged using conventional imaging alone. The median age was 72 years in the PSMA PET/CT group and 76 years in the conventional imaging group. Patients undergoing PSMA PET/CT also had fewer comorbidities and lower Gleason scores at their initial diagnosis.
Metachronous metastatic disease was more common in the PSMA PET/CT group, occurring in 29% of patients compared with 22% in the conventional imaging group. Nodal-only disease was also more frequent, at 47% versus 24%. High-volume disease was identified in 27% of patients staged with PSMA PET/CT and 45% of those staged with conventional imaging alone. Bone metastases were reported in 47% and 63%, respectively.
The PSMA PET/CT group also had lower biochemical markers of tumor burden. Median prostate-specific antigen levels were 23.5 ng/mL in the PSMA PET/CT group and 85.1 ng/mL in the conventional imaging group, while median alkaline phosphatase levels were 87 U/L and 130 U/L, respectively.
Treatment Shifts
Androgen deprivation therapy-based treatment remained the predominant approach, used alone or with local therapy in 67.1% of patients. Overall use was similar between the conventional imaging and PSMA PET/CT groups, at 66.2% and 65.6%, respectively, although treatment selection differed.
Patients staged with PSMA PET/CT more frequently received androgen deprivation therapy combined with local treatment, at 42.3% compared with 20.3% following conventional imaging. Among patients with synchronous disease, androgen deprivation therapy monotherapy was used in 27% and 50%, respectively. Taxane-based intensification was more common after conventional imaging, at 22.3% versus 15.1%, whereas androgen receptor pathway inhibitor use was higher in the PSMA PET/CT group, at 5.6% versus 3.7%. Local treatment alone and no treatment were also more frequent following PSMA PET/CT staging.
Among patients with low-volume disease, local therapy was used in 62.9% of the PSMA PET/CT group and 38.0% of the conventional imaging group, most commonly in combination with androgen deprivation therapy. In high-volume disease, intensified systemic therapy was used in 48.4% of the PSMA PET/CT group and 40.0% of the conventional imaging group.
The investigators suggested that these differences may reflect earlier detection, patient selection, evolving treatment standards, and the higher prevalence of nodal-only and lower-burden disease in the PSMA PET/CT cohort.
Interpreting Survival
Median overall survival in the entire registry population was 49.8 months. The descriptive median overall survival was 78.3 months among patients staged with PSMA PET/CT and 40.0 months among those staged using conventional imaging alone.
These results require careful interpretation. PSMA PET/CT was adopted more recently, resulting in shorter follow-up and a higher proportion of censored patients in that group. In addition, the groups differed substantially in age, comorbidities, metastatic presentation, disease volume, PSA levels, and treatment selection.
No formal between-group survival comparison was conducted because of these imbalances. The investigators therefore cautioned that the observed difference should not be attributed to PSMA PET/CT itself. Instead, it likely reflects a combination of lead-time bias, selection bias, and stage migration.
Understanding the Will Rogers Phenomenon
The greater sensitivity of PSMA PET/CT can identify metastatic lesions in patients who would previously have been classified as having nonmetastatic disease based on conventional imaging.
This redistribution is known as the Will Rogers phenomenon. Patients with relatively limited disease are moved into the metastatic category, improving the apparent prognosis of that group. At the same time, the remaining conventionally defined nonmetastatic population may also appear to have better outcomes because patients with previously undetected metastases have been removed.
In the CAPRI-3 analysis, stage migration contributed to the identification of a metastatic population with more nodal-only disease, lower biochemical tumor burden, and a more favorable prognosis, while patient selection may partly explain the younger age of the PSMA PET/CT cohort.
Because landmark trials establishing contemporary standards of care generally enrolled patients staged using conventional imaging, it remains uncertain whether the same treatment benefits and intensification strategies apply to patients with PET-only metastatic disease, who may have a more indolent disease course.
The authors also noted that the CHAARTED definition of high- and low-volume disease was not developed for PSMA PET/CT. PSMA PET-specific definitions of tumor burden may ultimately be needed to improve prognostic stratification and patient selection in future trials.
Study Limitations
The study was limited by its retrospective registry design. Identifiable imaging data were unavailable in the registry for 957 patients because of interoperability problems. A quality-control audit indicated that 97.5% of cases classified as having missing imaging had undergone imaging, often at referring hospitals, but the information could not be captured by the automated extraction system.
The investigators also lacked consistent information about why a particular imaging modality was selected and could not consistently distinguish standard-of-care scans from trial-related scans. This limited their ability to separate the effects of physician preference from institutional protocols. The intent of radiotherapy for bone metastases was not documented, preventing a definitive distinction between palliative radiotherapy and metastasis-directed treatment.
Despite these limitations, the study included a large, unselected population treated across 19 hospitals and provides an extensive view of the real-world transition from conventional imaging to PSMA PET/CT.
Takeaway
PSMA PET/CT use in metastatic castration-sensitive prostate cancer increased rapidly. Patients staged with PSMA PET/CT were younger and had lower disease burden and more frequent nodal-only involvement. The longer survival observed in the PSMA PET/CT cohort should not be interpreted as a benefit caused by the imaging modality, as stage migration, earlier detection, and patient selection likely contributed to the apparent difference.
The optimal management of patients whose metastatic disease is detected only by PSMA PET/CT remains uncertain. Whether this population may be suitable for treatment de-escalation requires prospective validation before implementation in clinical practice.
The full article is available in the European Journal of Nuclear Medicine and Molecular Imaging.
