Impact of blood kinetics fittings and simplification methods on bone marrow dosimetry for 177Lu-PSMA-617: Early experience from the Canadian Cancer Trials Group PR21 trial (NCT 04663997)

Introduction: Bone marrow (BM) dosimetry poses an essential yet challenging process in ¹⁷⁷Lu-PSMA-617 radiopharmaceutical therapy (RPT) of metastatic prostate cancer. Estimation of BM absorbed doses needs to account for various contributions from red marrow itself (self-dose), bone, other high-activity uptake organs (e.g. kidneys) and the remainder of the patients’ body. BM self-dose, the main contribution to the total BM dose, can be commonly estimated based on blood-sampling by measuring the activity concentration in the blood at different time points. However, studies have shown that the blood biodistribution follows a bi- or tri-phase exponential of clearance kinetics (including a fast phase at the beginning of the activity injection). To accurately estimate BM self-dose, multiple (≥4) blood sample collections must be considered. To decrease patient burden and simplify the dosimetry procedure, single time point (STP) dosimetry has been proposed; however, the accuracy of STP BM dose is still under discussion. We aimed to investigate BM self-dose accuracy using different kinetic curve fittings for simplified BM dosimetry.

Methods: Eighteen patients with first cycle of treatment were included in this study. The injected activity was 7.40±0.74 GBq. Blood samples were collected at 15 min (t1), 24 h (t2) and 48 or 72 h (t3) post injection (p.i.). Both bi- and tri-phase clearance were considered by combining mono-exponential fittings for different time intervals (0-t1, t1-t2, t2-t3-inf). To explore tri-phase kinetics, simulations were performed to determine the time threshold between the 1^st and 2^nd phase by testing wide ranges of effective half-lives (1^st phase: =5min-2.5h; 2^nd phase: 2.5-15h). To simplify the process, two out of three blood samples were used to test the accuracy of single-phase clearance. All the fitting methods were summarized in Table 1. Differences in blood-based self-dose obtained from different curve fittings and phase contributions were studied. Furthermore, STP dosimetry based on the bi-/tri-phase fitting (with fixed T_eff for each phase)were developed and the corresponding results were compared with published STP methods (proposed by Hänscheid and Madsen, respectively).

Results: Our simulation shows that the 1^st clearance phase for tri-phase kinetics would be dominant at less than 15min p.i. with T_eff < 2.5h. The T_eff’s of 2^nd and 3^rd phases were 5.1±0.7h and 15.0±4.8h, with a switch from 2^nd to 3^rd phase at around 24h p.i.. The dose differences between tri- and bi-phases was < 1%. Considering single-phase clearance, mono-exponential fitting using the data acquisition at t1 and t2 provided the most accurate dose, which is 8.0±3.4% smaller than those from using bi-phase fittings. 95% or higher accuracy of STP doses could be achieved based on a fixed bi-phase curve fittings with either prior T_eff for each patient or using population-based mean T_eff for each phase, under the assumption that the time threshold between two phases was set to 24h p.i.. However, mono-exponential fitting with single-phase clearance would result in large errors using published STP methods.

Conclusions: We investigated the impact of blood sample acquisition time points and kinetic curve fittings on the BM dosimetry (blood-based method) for ¹⁷⁷Lu-PSMA-617 RPT. We conclude that bi-phase clearance must be considered to accurately estimate BM dose. STP methods with single-mono-exponential clearance kinetics are inaccurate for BM dosimetry. However, STP methods that use prior information from the first cycle to determine T_eff and allow for fixed bi-phase kinetics modeling would result in a reasonable BM dose accuracy.

• Download figure
• Open in new tab
• Download powerpoint

• Download figure
• Open in new tab
• Download powerpoint