Lesion Absorbed Dose-Response Relationship in Patients with Metastatic Castration-Resistant Prostate Cancer Undergoing [177Lu]Lu-PSMA-617 Radiopharmaceutical Therapy

Abstract

The relationship between lesion absorbed dose (AD) and response in patients with metastatic castration-resistant prostate cancer undergoing [¹⁷⁷Lu]Lu-PSMA-617 radiopharmaceutical therapy (RPT) remains poorly understood. The objective of this work was to investigate the AD-response relationship at both the patient and lesion levels. Methods: Sixty-five patients underwent serial SPECT/CT imaging after receiving 7.31 ± 0.27 GBq of [¹⁷⁷Lu]Lu-PSMA-617. Single-time-point (STP) (Hänscheid approximation at 72 h) and multiple-time-point voxelwise dosimetry were performed. Patient response was evaluated by changes in serum prostate-specific antigen level before and after cycle 1 of RPT. The response of individual lesions was evaluated by the change in the SUV_max before, during, and after RPT with [⁶⁸Ga]Ga-PSMA-11 PET/CT. Results: Baseline PET and SPECT lesion SUV_max were strongly correlated (Pearson r, 0.74; n = 1364 lesions). Kidney ADs were relatively low (0.28 ± 0.12 Gy/GBq). No significant decrease in estimated glomerular filtration rate was observed 1 y after RPT. On average, STP dosimetry underestimated the AD by 8%. A moderate negative relationship was observed between the mean lesion AD for an individual patient (Spearman ρ, -0.33; n = 63) and lesion (Spearman ρ, -0.30; n = 681) responses. Patients receiving a higher mean AD (>7.5 Gy) had a significantly better prostate-specific antigen response (median, 70% vs. -5%; P < 0.001; unpaired t test) and longer biochemical progression-free survival (median, 4.1 mo vs. 1.6 mo; P = 0.005; unpaired t test) compared with patients whose mean AD was less than 7.5 Gy, respectively. Conclusion: A moderate AD-response relationship was observed in patients with metastatic castration-resistant prostate cancer undergoing [¹⁷⁷Lu]Lu-PSMA-617 RPT. The feasibility of STP dosimetry facilitates its implementation for treatment personalization. Kidney ADs may be reduced with abundant hydration.

Keywords: [177Lu]Lu-PSMA-617; [68Ga]Ga-PSMA-11; dose–response relationship; metastatic prostate cancer.

Graphical abstract

FIGURE 1.

Consolidated Standards of Reporting Trials (CONSORT) diagram.

FIGURE 2.

Relationships between baseline [⁶⁸Ga]Ga-PSMA-11 PET SUV_max and [¹⁷⁷Lu]Lu-PSMA-617 SPECT SUV_max (n = 1364 lesions) (A) and VIP as calculated from baseline PET and voxelwise dose maps (n = 65 patients) (B). Linear fit and identity links displayed in blue and gray, respectively.

FIGURE 3.

Feasibility of STP dosimetry. AD_max (A) and AD_mean (B) as calculated with either multiple-time-point (MTP) dosimetry or Hänscheid approximation using 72-h data point (STP72). LoA = limits of agreement.

FIGURE 4.

AD–response relationship on patient level. (A) Mean AD over all lesions from individual patient vs. percentage change (%Δ) in PSA after cycle 1 of [¹⁷⁷Lu]Lu-PSMA-617 RPT (n = 63; posttherapy PSA data not available for 2 patients). (B) Corresponding relationship with metric that incorporates AD_mean and PSA doubling time before RPT (n = 58; PSA doubling time not available for 5 patients).

FIGURE 5.

Outcomes for patients dichotomized based on mean AD from all lesions (cohort median = 7.5 Gy). (A) Best PSA response; P < 0.001 (unpaired t test). Kaplan–Meier estimator results for bPFS survival (B) and overall survival (C). PFS = progression-free survival.

FIGURE 6.

(A) Lesion AD_max vs. %Δ(PSMA PET SUV_max) between baseline (BL) and early follow-up PET (681 lesions and 37 patients). (B) Tumor control probability vs. lesion AD_max from cycle 1. Tumor control was defined as decrease of ≥50% in PSMA PET SUV_max between BL and early follow-up PET. Odds ratio, 1.02; 95% CI, 1.01–1.03; P < 0.001.

FIGURE 7.

PET data from 62-y-old patient with baseline PSA of 102 ng/mL and PSA doubling time of 2.1 mo who underwent [⁶⁸Ga]Ga-PSMA-11 PET 5 d before initiation of RPT. Follow-up [⁶⁸Ga]Ga-PSMA-11 PET scans were performed after cycles 2, 4, and 6. Lesions, sized between 6.1 and 87.7 cm³ (median, 41.1 cm³), had AD of 3.36–15.40 Gy/GBq (median, 8.12 Gy/GBq) in cycle 1. Mean lesion AD was 13.6 Gy, and bPFS was 14.0 mo. (A) Maximum-intensity-projection images for baseline PET, [¹⁷⁷Lu]Lu-PSMA-617 SPECT, and 3 follow-up PET scans, acquired 71, 159, and 246 d after initiation of RPT. (B) SUV_max correlation between baseline [⁶⁸Ga]Ga-PSMA-11 PET and [¹⁷⁷Lu]Lu-PSMA-617 SPECT. (C) [⁶⁸Ga]Ga-PSMA-11 SUV_max over time for all lesions with baseline SUV_max >100 (remained visible and quantifiable at all time points). These lesions had an AD_max of 33–153 Gy in cycle 1.