R Danieli¹*, Y Dewaraja², D Raspanti³, M Ferrari⁴, F Botta⁵, G Glatting⁶, M Cremonesi⁷, (1) Institut Jules Bordet, Brussels, Brussels-Capital Region, BE, (2) University of Michigan, Ann Arbor, MI, (3) Temasinergie S.p.A, Faenza, Emilia-Romagna, IT, (4) European Institute of Oncology IRCCS, Milano, Lombardia, IT, (5) European Institute of Oncology IRCCS, Milano, Lombardia, IT, (6) Universitat Ulm, Ulm, BW, DE, (7) European Institute of Oncology IRCCS, Milano, Lombardia, IT

• 
  R Danieli

SU-E-202-4 (Sunday, 7/10/2022) 1:00 PM - 2:00 PM [Eastern Time (GMT-4)]

Room 202

Purpose: Dosimetry in Peptide-Receptor Radionuclide Therapy (PRRT) is increasingly recognized for treatment optimization. However, the standardization of the absorbed dose methodology still represents a major effort for medical physicists. The use of different methods for fitting and integrating the time-activity curve (TAC) represents a considerable source of variability. This study compares methods commonly provided for computing time-integrated activity coefficients (TIACs) in 177Lu-PRRT.

Methods: Five methods were selected from those applied by used commercial/free software or distinguished authors, including mono-exponential, two-exponential (with uptake phase) fitting, trapezoids and one (PHYS) using physical decay after the last experimental time-point (TP). Publicly available (SNMMI Dosimetry Challenge) quantified SPECT/CTs of two patients treated with 177Lu-DOTATATE were used, acquired at 4, 24, 96, 120 (Patient-4) or 196 (Patient-6) hours post-injection. TACs for lesions and kidneys were obtained with MIM SurePlan MRT. Python was used for customized fitting and integration. Coefficients of variation (CoV) and percentage differences between TIACs obtained with different methods, percentage contributions to TIAC of partial (0-TP1;TP1-TP2;TP2-TP3;TP3-TP4;TP4-infinity) and extrapolated areas (0-TP1+TP4-infinity) were computed.

Results: For Patient-4/Patient-6, respectively: maximum CoV in TIACs were 18%/10% for kidneys and 17%/11% for lesions. Excluding PHYS, CoV reduced to 2%/6% for kidneys and 19%/9% for lesions. Maximum TIAC variations with [without] PHYS were: kidneys 41%/31% [4%/15%]; lesions 32%/30% [32%/18%]. Percentage extrapolated areas ranged 19%-41%/6%-23% [19%-22%/9%-15%] for kidneys; 49%-72%/12%-36% [49%-72%/12%-27%] for lesions.

Conclusion: This study showed a consistent variability due exclusively to using different methods for computing TIACs. Such variability depends both on the pharmacokinetics profile (presence/absence of uptake phase) and the time sampling (e.g. last TP position). With TP1=4h, an uptake phase is not infrequent: mono-exponential fitting considering all TPs could be inappropriate or even physically incorrect. When TP4=120h, extrapolated areas could be >50%. TAC extrapolation using physical decay demands awareness. Proper recommendations and software alerts are warranted.

Funding Support, Disclosures, and Conflict of Interest: Yuni Dewaraja is a Consultant for MIM Software Inc.

Keywords

Not Applicable / None Entered.

Taxonomy

Not Applicable / None Entered.

Contact Email