PSMA PET imaging in metastatic castration-resistant prostate cancer patients treated with 177Lu-PSMA-617: correlation with response, progression free (PFS) and overall survival (OS)

Introduction: Targeted radionuclide therapy (TRT) is becoming an accepted treatment modality for patients with advanced metastatic castration resistant prostate cancer (mCRPC). Prostate specific membrane antigen (PSMA) is a transmembrane enzyme that is selectively expressed in PC, with levels increasing with tumor grade and castration resistance and is a leading target for TRT. Furthermore, PSMA-directed ligands have been specifically co-engineered for imaging, allowing a theranostic approach. While PSMA is expressed to some degree on virtually all PC, there is still significant variability. Unlike many centers which treat only those with strong target imaging, we obtain baseline PSMA imaging as a biomarker, but treat regardless of imaging results. Here, we report the relationship between PSMA expression levels on pre-treatment imaging and response, progression free survival and overall response after PSMA-TRT.

Methods: Eligible subjects with progressive mCRPC despite standard therapy were enrolled in a prospective clinical trial [NCT NCT03042468], where they received ¹⁷⁷Lu-PSMA-617 therapy delivered as a single fractionated cycle (D1 and D15). The initial dose-escalation portion of the study used doses of ¹⁷⁷Lu-PSMA-617 ranging from 7.4 GBq - 22.2 GBq per cycle and when no dose-limiting toxicity was found the phase II portion of the study treated men at 22.2 GBq. Baseline PSMA imaging was performed and all patients were given imaging score (IS) ranging from 0-4 as follows: PET images were scored using mean SUVmax of 5 “hottest” lesions and then comparing it with liver mean SUV. 1=SUVmax < liver mean SUV, 2= SUVmax 1-2.5x liver SUV, 3=SUVmax 2.5-5x liver SUV, 4=SUVmax > 5x liver SUV. Planar scans were scored using the criteria of 0=tumor undetectable, 1=faint tumor activity detectable, 2=strong tumor activity not equal to liver, 3=tumor activity equal to liver, 4=tumor greater than liver. Primary outcome measures included PSA response, progression-free (PFS) and overall survival (OS). The chi-square test was used to assess the association between IS and PSA decline. Kaplan-Meier survival analysis was used to assess PFS/OS and the log-rank test was used to compare PFS/OS between clinical categories of interest. Multivariable logistic (for PSA response) and cox (for PFS/OS) regression were used to control for additional factors.

Results: 44 men (21 at the recommended phase II dose of 22.2 GBq) with median age 69 (range 55-91) and median PSA 182.97 (range 0.89-5541) were treated. The majority (93%) had bone mets, 45% nodal, 18% lung, 9% liver, 9% other visceral metastases. Most (66%) were CALGB (Halabi) poor prognostic group, 30% intermediate risk. 38 (86.4%) patients underwent PSMA PET, and 6 (13.6%) gamma planar imaging. Most (36, 81.8%) patients had IS of 4, 11.4% had IS of 3, 4.6% IS of 2, and 2.3% had IS of 1. In all comers (without selection for PSMA expression, 81.8% of men had PSA decline, with 59.1% having at least 50% PSA decline (66.7% of those treated at 22.2 GBq with >50% PSA decline). On multivariable analysis controlling for dose administered, IS, prior chemotherapy, and CALGB prognostic group, dose administered and IS of 4 were significant for PSA response, PFS, and OS as in the Table below. One of 3 with poor IS and 6 of 9 with poor to intermediate IS had some PSA decline.

Conclusions: Increased PSMA PET avidity as assessed by imaging score (IS) and administered dose correlate with better response and improved progression-free (PSF)and overall survival (OS) following ¹⁷⁷Lu-PSMA-617.