Bander, Neil H.

5074 Background: PC commonly expresses PSMA and has dose-responsive radiosensitivity. 177Lu-based radioligands usually use an empiric dosing regimen. This 1st dose-escalation study of 177Lu-PSMA-617 employed a dose-dense fractionation schedule intended to allow delivery of higher total doses to overcome radioresistance due to repopulation. Here, we report mature results of the study. Methods: Progressive mCRPC following > 1 potent ARPI (e.g., abi/enza) and chemo (or unfit/refuse taxane) without limit of # prior therapies provided adequate organ function and ECOG PS 0-2. Treatment: single cycle of fractionated dose 177Lu-PSMA-617 on D1 and D15. In Ph I dose-escalation, patients (pts) received one cycle of 7.4 - 22 GBq followed by Simon 2-stage Ph II at 22.2 GBq. PSMA expression was not required for treatment, but pre- and post-treatment 68Ga-PSMA11 PET/CT and/or 177Lu-PSMA-617 SPECT done. Primary efficacy by PSA changes with 2ndary outcomes with CT/bone scans, PSMA PET, circulating tumor cell (CTC; CellSearch) counts, plasma DNA (PCF_SELECT), and survival. Results: Between 1/2017 – 2/2021, 50 pts treated with median PSA 164.3, 35 (70%) CALGB (Halabi) poor risk. 47 (94%) bone, 38 (76%) nodal, 12 (24%) lung, 5 (10%) liver mets (4 additional pts with PSMA PET+ liver mets not apparent on CT). 29 (58%) with prior >2 ARPI, 29 (58%) with >1 chemo, 14 (28%) with Ra-223, 2 (4%) with 177Lu-J591. 27 (54%) treated at 22 GBq. No dose-limiting toxicity during Ph I. 38 (76%) with any PSA decline, 27 (54%) with >50% PSA. Median rPFS 8.3 mo [5.6-16.7], and OS 15.7 mo [10.9-20.6]. Of 17 RECIST measurable, 6 (35.3%) responded, 7 (41.2%) stable, 4 (23.5%) progressed. Of 31 with paired CTC counts, 16 (52%) decreased, 5 (16%) were stable; 10 (32.3%) converted to favorable/undetectable. Plasma DNA with high allele-specific ploidy and ARcopy number gain associated with OS (univariate). Treatment emergent adverse events (TEAE): 32 (64%) dry mouth, 22 (44%) pain flare, 22 (44%) fatigue, 19 (38%) nausea, 15 (30%) anemia, 15 (28%) thrombocytopenia (plts), 9 (18%) neutropenia (ANC). All TEAEs restricted to grade (Gr) 1-2 except 6 (12%) with Gr 3 anemia, 2 (4%) Gr 3/4 plts, 1 (2%) Gr 3 ANC. TEAE incidence and Gr were not related to dose. On multivariable analysis, 177Lu dose administered (p=0.09), CALGB risk (p=0.04), prior chemo (p=0.009), associated with OS; dose (p=0.1), CALGB risk (p=0.2), and AR copy number (p=0.2) associated with PSA50 decline. Baseline PSMA imaging (by imaging score or SUVmax) was not associated with OS or response. Conclusions: A single-cycle of fractionated-dose 177Lu-PSMA-617 is safe. Despite no pre-selection for PSMA expression, most had PSA decline with favorable PFS and OS compared to historical controls and similar to PSMA-selected targeted radionuclide studies administering multiple cycles in a less dose-intense approach. Clinical trial information: NCT03042468 .

Abstract Background: We have previously reported that 177Lu-PSMA-617 given as a single-cycle, dose-intense regimen for patients (pts) with pretreated mCRPC is efficacious even without selection by PSMA (ESMO 2021). Quantitative analyses of pre-treatment PSMA PET may allow for better patient selection and prediction of toxicities. Analyses of sequential PET images may also provide insights that may be associated with longer term outcomes. Our previous analysis of 13 patients demonstrated association of PSMA PET signal with overall survival (OS) and PSA responses, as well as trends with adverse events (Nauseef GU ASCO 2022). Here we present long-term follow up data and expanded imaging correlatives from our completed phase I/II study. MethodsEntry criteria: progressive mCRPC after at least one prior potent AR pathway inhibitor and prior chemotherapy (excepting men ineligible for or refusing of chemotherapy) with allowance for prior radionuclides (incl. Ra223 and PSMA-TRTs). Treatment was fractionated-dose (D1 and D15) 177Lu-PSMA-617 (7.4-22.2 GBq). TRAQinform IQ (AIQ Solutions) was used to identify, track, and quantify regions of interest suspicious of cancer (lesion-ROI) on PSMA PET/CT images and quantify uptake in normal organs outside of lesion-ROI. Measurements included SUVmean across all lesion-ROI, volume of lesion-ROI, and SUVtotal (sum of SUV in all lesion-ROI), as well as changes after treatment, including in PSMA-expressing healthy tissues. Associations with survival were tested via Cox proportional hazard models in univariate analyses and associations with adverse events (AEs) and PSA responses were via assessed via Wilcoxon rank sum tests. Results32 pts (of 50 total in the study between 1/2017-2/2021) had pre- and post-treatment imaging suitable TRAQinform IQ analysis and associated survival analyses. Baseline characteristics were similar to the entire study cohort. PSA response of 50% was observed in 21/32 (66%) of which 16 were confirmed. No imaging variable predicted PSA30 or PSA50, although an association was observed in pretreatment SUVmean and lesion-ROI volume reduction. No imaging variable was significantly associated with OS (e.g., pretx SUVmean HR 0.9, p=0.10, pretx Volume HR 1.00, p=0.052). Pretx SUVmean was associated with mPFS (HR 0.85, 95% CI 0.74, 0.98, p=0.023). Among select AEs, anemia was associated with non-lesion-ROI skeleton SUVtotal in pre- (p=0.027) and post-tx (p=0.007). Higher pre-treatment salivary gland uptake (SUVmean) was associated with more xerostomia (p=0.11). Fatigue was associated with a variety of variables including salivary SUVs (mean, max, and total) and post-tx bowel SUVmax (p=0.030). ConclusionsPre-treatment and pre- to post-treatment 68Ga-PSMA11-PET data analyzed may be associated with outcome. While a strength of this study is the prospective nature of patient enrollment and fixed imaging timepoints, our associations are limited by small sample size. Expansion of this analysis to larger datasets may improve our ability to predict treatment response and toxicity by body-wide PSMA detection. Citation Format: Jones Trevor Nauseef, Charlene Thomas, Michael Sun, Nicole Jacobs, Andrés Ricaurte Fajardo, Sam Ruder, Escarleth Fernandez, Zachary Davidson, Amie Patel, A. Oliver Sartor, Shankar Vallabhajosula, Ana Molina, Cora N. Sternberg, David M. Nanus, Sandra Huicochea Castellanos, Joseph Osborne, Neil H. Bander, Timothy Perk, Scott Tagawa. Quantitative assessment of PSMA imaging before and after 177Lu-PSMA-617 treatment in a Ph I/II trial [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 2 (Late-Breaking, Clinical Trial, and Invited Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(7_Suppl):Abstract nr CT225.

Abstract Tumor cell lysis can be enhanced significantly in vitro when heteroconjugate (HC) antibodies (anti-CD3 x anti-tumor mAb) are used to specifically direct lymphocyte effector cells to the tumor cell target. In order to effectively utilize HC antibodies in an immunotherapy protocol, methods must be identified for the optimum expansion, activation, and retargeting of lymphocyte-effector populations from cancer patients. In this study, we have compared the proliferative responses of different normal and renal cell carcinoma (RCC) patient lymphocyte preparations (PBL, tumor-infiltrating lymphocytes) stimulated in vitro for periods up to 12 days with a variety of growth factor combinations (anti-CD3, rIL-2, rIL-4). These activated lymphocyte preparations were then tested in vitro for their ability to kill RCC tumor cells and tumor cell lines in the presence of HC preparations (anti-CD3 mAb covalently linked to mAb reactive to different RCC tumor-associated Ag). RCC patient PBL cultured with anti-CD3 plus rIL-2 for 12 days resulted in a 3- to 160-fold expansion of effector cells. These cells, as well as tumor infiltrating lymphocytes, when retargeted with appropriate HC antibodies were capable of mediating high levels of killing of autologous tumor cells. No constitutive autologous anti-tumor cell response was detected in the absence of added HC antibodies. Of the five anti-RCC mAb tested (A6H, K29, K20, UR07, and URO 3), HC containing URO 3 x anti-CD3 and K20 x anti-CD3 elicited the highest level of tumor cell lysis by the activated lymphocyte effector cells. Together these results demonstrate that HC antibodies may be a useful imunotherapeutic reagent for directing the killing of RCC tumor cells by autologous lymphocytes.