Results from a prospective registry of 18F-Fluciclovine PET/CT use in prostate cancer management: a cautionary lesson for implementation of PSMA PET.

BACKGROUND: 18F-Fluciclovine positron emission tomography (PET) was FDA-approved in the U.S. in 2016 and was the most sensitive imaging modality for prostate cancer (PC) until the approval of prostate-specific membrane antigen (PSMA) PET in 2020. However, providers' reasons for ordering 18F-Fluciclovine PET/CT (FluPET) in practice and impact on patient care remain poorly defined. This prospective registry at a tertiary academic center describes patterns of FluPET use and outcomes prior to the FDA approval of PSMA PET in December 2020. METHODS: Providers ordering FluPET for patients with PC were surveyed before, ≤2 weeks after, and ≥1 year after imaging to assess reasons for obtaining FluPET, projected treatment plan, changes in plan due to FluPET findings, and toxicity attributable to the change in treatment plan. Baseline patient characteristics, FluPET results, and longitudinal outcomes were collected. RESULTS: Between 12/2018-09/2021, 62 patients with localized PC (8.1%), biochemical recurrence (BCR; 80.6%), non-metastatic castration-resistant PC (CRPC) (3.2%), metastatic castration-sensitive PC (3.2%), or metastatic CRPC (4.8%) were enrolled and underwent FluPET. Most scans (90.3%) were performed prior to the FDA approval of PSMA PET 12/2020. FluPET was most often obtained to guide local salvage or metastasis-directed therapies (90.3%); other reasons (non-exclusive) were initial staging (9.6%) or clarifying equivocal lesions from other imaging (9.6%). FluPET detected ≥1 PC lesion in 74.2% of patients. After FluPET, 48.4% of providers reported changing treatment plans, which was more likely when FluPET was positive (60.9% vs 12.5%, P<0.001), and often involved initiation of systemic therapy (19.4%). Treatment changes were reported in 57% of patients with BCR1 and 48.2% of patients with BCR2. In contrast, only 20% of patients with distant metastatic disease had a change in treatment. Among patients in the BCR1 and BCR2 cohort, treatment plan changes were associated with a median time to next treatment that was not reached after a median follow-up of 67.6 months. There was no statistically significant difference in overall survival between patients with biochemical recurrence (BCR) who did and did not have a treatment plan change. A year after FluPET, reported potential toxicities from treatment plan changes were minimal. CONCLUSION: FluPET was utilized across the disease spectrum of PC, primarily to guide local salvage or metastasis-directed therapies, given its improved sensitivity for detecting prostate bed recurrence due to the slow physiologic excretion of 18F-fluciclovine. Notably, a positive FluPET frequently prompted initiation of systemic therapy; however, the clinical benefit of such management remains uncertain. Moreover, providers often selected multiple, sometimes conflicting, treatment plans following FluPET, reflecting uncertainty in translating imaging findings into definitive management decisions. A larger, prospective registry using PSMA PET with a requirement for providers to select a single post-scan treatment strategy is warranted to better assess whether imaging-guided treatment changes improve clinical outcomes.