Abstract
241044
Introduction: Eligibility criteria for PSMA-RLT are currently debated, particularly with respect to the role of FDG-PET in patient selection. We are proposing new imaging eligibility criteria for PSMA-RLT based on dual-tracer FDG/PSMA-PET that do not rely on lesion size on CT. The aims of this analysis were to compare these new criteria with those of the TheraP and VISION trials when applied in a prospective cohort of patients with metastatic castration-resistant prostate cancer (mCRPC) and to correlate the RLT eligibility with overall survival (OS).
Methods: Ninety-eight patients with progressive mCRPC and at least three metastases on conventional imaging were enrolled in 3TMPO, a multicenter prospective cohort study investigating the intrapatient intermetastatic heterogeneity with multi-tracer PET/CT (NCT04000776). All participants underwent 18F-FDG and 68Ga-PSMA-617 PET/CT within ten days. Lesions that were analyzed had a molecular tumor volume ≥1 cc on either PET when segmented using a threshold of 1.5x liver SUVmean. For each lesion and tracer, the ratio of SUVpeak to liver SUVmean (SUVR) was computed. Eligibility to RLT was defined as having at least one PSMA+ lesion and no FDG+/PSMA- lesion. We assessed different positivity thresholds defining RLT eligibility: 1) the protocol's prespecified threshold of SUVR ≥1.5 for both tracers (3TMPOA); and 2) post-hoc, relaxed thresholds of SUVR ≥2.0 for FDG and of SUVR ≥1.0 for PSMA (3TMPOB). RLT eligibility according to TheraP criteria (i.e. ≥1 lesion with SUVmax ≥20; all CT-measurable lesions/nodes ≥10 mm with SUVmax ≥10; no FDG+ lesion with PSMA SUVmax <10 or FDG>PSMA) and VISION (i.e. ≥1 lesion with PSMA uptake > liver; no bone superscan; no CT-measurable soft-tissue lesion ≥10 mm or node ≥25 mm with PSMA uptake < liver) was determined independently by two expert readers who reached a consensus on discordant cases. OS were derived from Kaplan-Meier curves and log-rank test was used for comparisons.
Results: Application of the 3TMPOA RLT eligibility criteria, i.e. excluding patients harboring FDG+(SUVR≥1.5)/PSMA-(SUVR<1.5) lesions, resulted in 52 (53%) participants being deemed eligible to PSMA-RLT. Application of the relaxed 3TMPOB RLT eligibility criteria, i.e. excluding patients harboring FDG+(SUVR≥2.0)/PSMA-(SUVR<1.0) lesions, increased this number to 69 (70%). In comparison, 39 (40%) and 75 (77%) participants would have been deemed eligible to RLT based on TheraP and VISION criteria, respectively. After a median follow-up of 12.3 mo. (IC 95% CI: 12.1-12.4 mo.), the median OS of the cohort was 10.2 mo. (95% CI: 8.5-11.8 mo.). For all criteria sets, the median OS of RLT-eligible participants was superior to that of ineligible ones, with significant to near-significant reductions in risk of death (HR=0.38-0.63, p=0.0002-0.10; Fig. 1). Of note, the median OS of the fifteen 3TMPOA-eligible/TheraP-ineligible participants was not reached (Fig. 2A), while that of the two 3TMPOA-inelegible/TheraP-eligible participants was 5.2 mo. Similarly, the median OS of the four 3TMPOB-eligible/VISION-ineligible participants was 9.8 mo., while that of the ten 3TMPOB-ineligible/VISION-eligible participants was 6.1 mo. (Fig. 2B).
Conclusions: We propose two novel dual FDG/PSMA PET-based RLT eligibility criteria sets - one more selective and one more permissive - that would both allow more mCRPC patients to receive PSMA-RLT than the TheraP criteria yet would still exclude those patients with poorly targeted, significantly hypermetabolic lesions (including bone metastases) otherwise permitted by the VISION criteria and associated with a poor prognosis. Our pragmatic dual-PET quantitative criteria sets thus appear reasonably balanced between the most frequently cited ones, and their adoption in clinical trials would allow to prospectively fill the knowledge gap regarding the benefits - or the lack thereof - of PSMA-RLT in patients harboring FDG+ lesions with borderline PSMA expression.
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.