Predictive Factors for Response to Lutetium-177 PSMA Radioligand Therapy in Patients With Metastatic Castration-Resistant Prostate Cancer

BACKGROUND AND RATIONALE Radioligand therapy (RLT) with [177Lu]Lu-PSMA-617 (Pluvicto) is an approved treatment for metastatic castration-resistant prostate cancer (mCRPC) after androgen receptor pathway inhibitors and taxane chemotherapy. In published phase 3 data, a PSA decline of 50% or more was seen in fewer than half of treated patients. Baseline PSMA PET/CT imaging biomarkers have shown predictive and prognostic value, but data are heterogeneous across studies. Post-therapy SPECT/CT dosimetry is less studied, and studies integrating diagnostic PET biomarkers, post-therapy dosimetry, and clinical-biochemical markers in one multivariable model are lacking.

HYPOTHESIS Baseline 68Ga-PSMA PET/CT imaging biomarkers, post-therapy 177Lu-PSMA SPECT/CT dosimetry parameters, and clinical and biochemical markers assessed before and during treatment, alone and in combination, predict response to RLT and progression-free survival in patients with mCRPC.

OBJECTIVES

1. To assess the predictive value of baseline 68Ga-PSMA PET/CT imaging biomarkers for response to 177Lu-PSMA RLT.
2. To assess the relationship between post-therapy 177Lu-PSMA SPECT/CT dosimetry parameters and baseline 68Ga-PSMA PET/CT biomarkers, and their predictive value for treatment response.
3. To assess the predictive value of clinical, quality of life, biochemical, and molecular markers observed before and during treatment.
4. To assess the applicability of treatment response criteria and their prognostic value for progression-free and overall survival.
5. To develop a multivariable predictive model integrating imaging, dosimetric, clinical, and biochemical markers and to assess its accuracy in predicting treatment outcomes.

DESIGN Single-center observational cohort study at the Department of Nuclear Medicine, Lithuanian University of Health Sciences Hospital Kaunas Clinics, with two data collection streams: a prospective stream (newly treated patients, with informed consent) and a retrospective stream (up to approximately 20 patients who started RLT before study start; waiver of informed consent granted). The same data are collected in both streams. Treatment is given according to the approved Summary of Product Characteristics (SmPC), independently of study participation. Planned total sample: 70 patients.

TREATMENT (STANDARD CARE) [177Lu]Lu-PSMA-617 7400 MBq intravenously every 6 weeks (plus or minus 1 week), up to 6 cycles, with dose modification, delay, or discontinuation per SmPC. Androgen deprivation therapy with GnRH analogues is continued.

68Ga-PSMA-11 PET/CT Baseline PET/CT before the first cycle and follow-up PET/CT approximately 4 to 8 weeks after the last cycle. Images are analyzed according to PROMISE V2 (miTNM staging, miPSMA score). Quantitative parameters: SUVmax, SUVmean, PSMA-positive tumor volume (PSMA-TV), and number of lesions, recorded by region (local disease, regional lymph nodes, distant metastases) where technically feasible.

POST-THERAPY SPECT/CT DOSIMETRY Quantitative three-timepoint SPECT/CT after every cycle on a cadmium-zinc-telluride (CZT) SPECT/CT camera, per EANM/SNMMI and EANM Dosimetry Committee recommendations: 3 to 4 hours (biodistribution and extravasation check), 24 hours, and 120 hours post-infusion. Absorbed doses (Gy) per cycle are calculated for kidneys, salivary glands, bone marrow, and tumor lesions, plus other parameters available in the dosimetry software.

LABORATORY AND CLINICAL DATA Blood tests (complete blood count, ALT, AST, albumin, total bilirubin, creatinine, creatinine clearance, alkaline phosphatase, LDH, PSA, CRP, ferritin; testosterone at baseline only) at baseline, 10 to 11 days before each subsequent cycle, 2 to 3 weeks after each cycle, 4 to 6 weeks after the last cycle, and every 4 to 6 months during follow-up. Clinical data include demographics, Gleason/ISUP grade, prior treatments, ECOG performance status, pain score (NRS 0 to 10), metastatic pattern, PSA doubling time, comorbidities, and subsequent treatments after RLT.

ADDITIONAL (NON-STANDARD) STUDY PROCEDURES

1. Circulating tumor DNA (ctDNA) quantification: 10 ml blood drawn during the same venipuncture as routine blood tests, 2 to 3 weeks after each cycle.
2. SF-36 health-related quality of life questionnaire (validated Lithuanian version) before cycle 1 and after cycles 2, 4, and 6 (or at treatment completion), completed during hospitalization.

FOLLOW-UP Patients are followed within standard oncological follow-up every 4 to 6 months until study end. The last enrolled patient is followed for at least 6 months.

SAMPLE SIZE Calculated with G*Power 3.1.9.7 (chi-square test of independence, df = 1), effect size w = 0.35, based on reported PSA50 response differences of 32 to 39 percentage points between favorable and unfavorable baseline PSMA PET/CT biomarker groups, alpha = 0.05, power = 0.80: minimum n = 65. With approximately 10% allowance for missing data, 70 patients are planned.

STATISTICAL ANALYSIS Normality is tested with the Kolmogorov-Smirnov test. PSA50 response rate is reported with 95% CI. Associations with PSA50 response are assessed by logistic regression and predictive accuracy by ROC analysis (AUC with 95% CI). Associations with PFS and OS are assessed by Cox proportional hazards regression (HR with 95% CI); PFS and OS are estimated by Kaplan-Meier with log-rank tests. Correlations between dosimetry and PET/CT parameters are assessed by Pearson or Spearman coefficients. Agreement between response criteria is assessed by Cohen's kappa. Significant predictors enter multivariable models (logistic regression for PSA50, Cox regression for survival), with up to 3 independent variables per the EPV of 10 or more rule. SF-36 change from baseline and its relation to response are assessed; ctDNA analysis is exploratory. Two-sided p < 0.05 is significant. Analyses are performed in SPSS.