Comparison of 18F-DCFPyL and 68Ga-PSMA-11 for 177Lu-PSMA-617 therapy patient selection

Purpose 68Ga-PSMA-11 is recommended for the selection of patients for treatment in the package insert for 177Lu-PSMA-617. We aimed to compare imaging properties and post-treatment outcomes from radioligand therapy (RLT) of patients selected with 68Ga-PSMA-11 and 18F-DCFPyL. Methods We retrospectively evaluated 80 patients undergoing PSMA RLT, who had pretreatment imaging using either 68Ga-PSMA-11 or 18F-DCFPyL. For both groups, we compared the biodistribution and lesion uptake and the PSA response to treatment. Results Both agents had comparable biodistribution. Patients initially imaged with 18F-DCFPyL had a higher PSA response (66% vs. 42%), and more patients had a PSA50 response (72% vs. 43%) compared to patients imaged with 68Ga-PSMA-11. Conclusion 18F-DCFPyL and 68Ga-PSMA-11 had comparable biodistribution and lesion uptake. Patients imaged with 18F-DCFPyL demonstrated clinical benefit to PSMA RLT comparable to those imaged with 68Ga-PSMA-11, and either agent can be used for screening patients.

Introduction 177 Lu-PSMA-617 radioligand therapy (RLT) has been shown to improve clinical outcomes and have a favorable safety profile in men with metastatic castration resistant prostate cancer (mCRPC) (1)(2)(3)(4)(5).In the phase 3 VISION study, 177 Lu-PSMA-617 RLT was shown to prolong overall survival (OS) and improve quality of life measures in patients with mCRPC relative to best supportive care (3), while the Phase 2 TheraP Trial demonstrated that 177 Lu-PSMA-617 resulted in a higher rate of PSA decline relative to cabazitaxel chemotherapy (4).When compared to conventional imaging, prostate-specific membrane antigen (PSMA)-based positron emission tomography (PET) has higher detection rates and greater diagnostic accuracy for patients with initial high risk, biochemically recurrent or persistent prostate cancer, and mCRPC (6)(7)(8)(9).In this theranostic approach, PSMA PET is used for the screening of patients to demonstrate the presence of PSMA expression, which makes them eligible for PSMA RLT (10)(11)(12).
Despite the market availability of 18 F-DCFPyL, the 177 Lu-PSMA-617 (vipivotide tetraxetan) package insert specifically recommends selecting patients for treatment using 68 Ga-PSMA-11, as the imaging agent and the utility of other PSMA ligands to select patients remains unclear.We retrospectively evaluated patient outcomes and imaging properties of 68 Ga-PSMA-11 and 18 F-DCFPyL in patients undergoing PSMA RLT in order to help determine if 18 F-DCFPyL is appropriate to use for patient selection.

Study population
In this study, we retrospectively screened individuals who underwent pre-treatment PET imaging with either 68 Ga-PSMA-11 or 18 F-DCFPyL before 177 Lu-PSMA-617 RLT at our institution from October 2021 to April 2023.The selection of radiopharmaceutical was determined by availability at each imaging center.Included patients had PSMA PET performed within 6 months prior to the first cycle of 177 Lu-PSMA-617 RLT.This study was approved by the institutional review board, and informed consent was waived.

Image interpretation
Each PSMA PET scan was interpreted using Visage (Visage Imaging).Five regions were recorded for the presence of prostate cancer including prostate bed (T), osseous (M1b), pelvic nodes (N), extrapelvic nodes (M1a), and visceral metastases (M1c).Maximum standardized uptake values (SUVmax) were recorded for osseous metastases, extrapelvic nodes, and visceral metastases with the highest uptake.Additionally, SUV was also recorded for physiological uptake in the liver (SUVmean) and parotid glands (SUVmax).

Response to RLT
Serum PSA levels served as the standard of reference for response assessment to 177 Lu-PSMA-617 RLT (17)(18)(19).The maximum decline in PSA that occurred anytime during or within 12 weeks of completion of RLT was taken for PSA response analysis.Baseline serum PSAs were drawn on the day of cycle 1 of RLT treatment, and the best PSA response during RLT was assessed for each patient.A decline of 50% from baseline PSA was defined as PSA50 response.

Statistical plan
Descriptive statistics in the form of median (interquartile) for continuous variable and count (percentage) for the binary variables was used to describe quantitative variables from the clinical data.While comparing the baseline characteristics between the groups imaged with 68 Ga-PSMA-11 and 18 F-DCFPyL; Student's t-test was used for the continuous variables and Fisher exact test was used for the discrete variables.A Student's t-test was conducted to assess the relationship between the lesion SUV and organ uptake between the groups imaged with 68 Ga-PSMA-11 and 18 F-DCFPyL.For the comparison of SUV and response, the median SUVmax was used to split the population evenly.p <0.05 was considered significant.A comparison of the best overall post-treatment PSA relative to baseline PSA was made between 68 Ga-PSMA-11 and 18 F-DCFPyL using Student's t-test.The maximum decline in PSA during RLT was reported for each patient using waterfall plots.

Patient characteristics
Of the 80 patients who received 177 Lu-PSMA-617 therapy at our institution from June 2022 to June 2023, 47 patients received 68 Ga-PSMA-11 and 33 patients received 18 F-DCFPyL for pre-treatment PET imaging.The patients in both these groups were similar for age, Gleason score, and pre-therapy PSA levels, and prior treatments.Patients imaged using 68 Ga-PSMA-11 had higher rates of prior radiation therapy.The two groups had a similar distribution of disease in the prostate/prostate bed, and metastatic disease to lymph nodes (N1), bone (M1a), soft tissue (M1b), and distant organs (M1c) (Table 1).

Semi-quantitative comparison of metastatic lesions on pretreatment PET imaging
In comparing the highest SUVmax lesion for the three metastatic sites between the 68 Ga-PSMA-11 and 18 F-DCFPyL groups, no statistically significant difference was observed between the extrapelvic lymph nodes (p=0.33) or osseous lesions (p=0.39),respectively.For visceral metastatic lesions, 18 F-DCFPyL had a higher uptake than 68 Ga-PSMA-11 (median = 11.2 (9.0-20.9)for 68 Ga-PSMA-11 = versus median = 28.7 (23.9-32.2) for 18 F-DCFPyL, p=0.04;Table 2).The median SUVmax across the overall population was 34.2 and was used to divide patients into two groups: those with high uptake (SUVmax > 34.2) and those with low uptake (SUVmax < 34.2).There was a trend to a higher PSA response in patients imaged with 18 F-DCFPyL compared to 68 Ga-PSMA-11, which was not statistically significant (Supplementary Table S2).

Discussion
This is the first report of outcomes in patients treated with PSMA RLT, who were selected with 18 F-DCFPyL.We demonstrated that the biodistribution between the two agents was identical and showed that the PSA50 response in patients selected with 18 F-DCFPyL was higher than in patients selected for treatment with 68 Ga-PSMA-11.Although prior work has focused on the diagnostic utility of 18 F-DCFPyL, our results demonstrate that it is appropriate to use it for PSMA RLT screening.
The PSA response to RLT was higher in patients who underwent pretreatment PET imaging with 18 F-DCFPyL than those with 68 Ga-PSMA-11.This was unexpected, and it is unclear from our small patient numbers if this finding is generalizable.Overall, our results indicate that patients selected with 18 F-DCFPyL appear to benefit at least equally to PSMA RLT, which is consistent with guidelines that indicate that either agent can be used for patient selection (11,12).
As has been previously reported, the biodistributions were similar between the two agents (20,21).The SUVmax of metastatic lesions are comparable for extrapelvic lymph nodes and osseous sites.While 18 F-DCFPyL outperforms 68 Ga-PSMA-11 in having higher uptake in visceral lesions, this is limited by the number of lesions included, and intra-patient comparison is needed to confirm that there is in fact higher uptake in visceral lesions.
This work has focused on the difference between 18 F-DCFPyL and 68 Ga-PSMA-11, but with the recent approval of rhPSMA-7.3, it is uncertain how our results can be extrapolated to include this newer radiopharmaceutical, especially given the partial hepatobiliary clearance seen with rhPSMA-7.3.Although rhPSMA-7.3has been shown to have lower urinary excretion, which may lead to enhance visualization of local recurrence, the liver uptake is higher than in the other two agents.At this time, it is unclear what the threshold should be for patient selection when using rhPSMA-7.3.
Our study has several limitations.First, this was a retrospective study.This study is additionally subject to various confounders inherent to its lack of intra-patient comparison of the two imaging agents.Specifically, mCRPC exhibits significant heterogeneity, and our analysis only considers the lesion with highest avidity, which may not fully represent the disease burden.The 18 F-DCFPyL group received more treatment cycles, and their visceral metastatic lesions showed higher radiotracer avidity, which would be expected to impact the PSA Waterfall plots of PSA response to RLT in patients with pretreatment PET imaging with 68 Ga-PSMA-11 (A) and 18 F-DCFPyL (B).
Yadav et al. 10.3389/fonc.2024.1382582 Frontiers in Oncology frontiersin.orgresponse rate.In the absence of head-to-head trials, comparisons of reported radiopharmaceutical performance on outcomes should be interpreted with caution due to the significant impact of differing patient populations, treatment cycles, end points, scanning protocols, scanning equipment, and readers.

Conclusion
Patients imaged with 18 F-DCFPyL demonstrated clinical benefit to PSMA RLT in this retrospective study.As previously shown, the physiological biodistribution and lesion uptakes at metastatic sites are comparable for both agents.Patients selected with 18 F-DCFPyL had higher PSA50 responses, but comparison to 68 Ga-PSMA-11 is limited given the many confounders.Although 68 Ga-PSMA-11 was utilized for patient selection in clinical trials of 177 Lu-PSMA-617, screening patients can be done using either of radiopharmaceuticals and should not be limited to 68 Ga-PSMA-11.