Zeitschrift fur medizinische Physik · 2026
Limitations of Mono-Exponential Individual Fitting as a Reference Model for Single-Time-Point Dosimetry in [177Lu]Lu-PSMA-617 Therapy.
СтатьяНазвание и аннотация: Перевод пока не подготовлен
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Fiarka Fulki, Hakim Assyifa Rahman, Budiansah Indra, Wicaksono Abdurrahman Aziz, Jabar Jaja M, Peters Steffie M B, Prive Bastiaan M, Nagarajah James, Konijnenberg Mark W, Siregar Syahril, Hardiansyah Deni, Glatting Gerhard
Аннотация
PURPOSE: To examine whether mono-exponential individual fitting (Mono-IFIT) is an adequate reference model for validating single-time-point (STP) renal dosimetry in [177Lu]Lu-PSMA-617 therapy, and to compare the resulting STP accuracy rankings with those obtained using population-based model selection with non-linear mixed-effects modelling (PBMS NLMEM).
METHODS: Kidney biokinetic data were analysed using a training dataset of 500 virtual patients generated from a six-parameter sum-of-exponentials function previously identified within the PBMS NLMEM framework, and an external testing dataset comprising 10 clinical patients with low-volume metastatic hormone-sensitive prostate cancer. Reference absorbed doses were calculated with Mono-IFIT and PBMS NLMEM. Four STP methods (STPNLMEM, STPMLR, STPH, and STPM) were evaluated at (47.7±2.2) h post-injection. Accuracy was assessed using relative deviation (RD), mean absolute percentage error (MAPE), and root mean square error (RMSE).
RESULTS: The selected PBMS NLMEM model provided a better description of the renal time-activity data than Mono-IFIT according to predefined goodness-of-fit criteria and Akaike-weight-based model selection. STP performance rankings depended strongly on the chosen reference model. Against Mono-IFIT, STPM and STPH showed the smallest deviations (mean RD -2.3% and 3.3%; MAPE 3.5% and 4.5%; RMSE 4.6% and 5.3%, respectively), whereas STPNLMEM showed the largest deviation (mean RD 16.0%; MAPE 16.0%; RMSE 17.9%). Against PBMS NLMEM, STPNLMEM showed the closest agreement (mean RD -2.5%; MAPE 6.6%; RMSE 9.3%), whereas STPMLR, STPH and STPM underestimated absorbed dose (mean RD -14.1%, -13.1% and -17.8%; MAPE 15.2%, 13.1% and 17.8%; RMSE 17.3%, 14.8% and 19.1%, respectively).
CONCLUSION: The apparent accuracy of STP methods is highly dependent on the reference model. For [177Lu]Lu-PSMA-617 renal dosimetry, Mono-IFIT should not be considered the preferred reference for method validation. In this dataset, PBMS NLMEM provided the better reference framework, and STPNLMEM showed the closest agreement with it. To ensure broader applicability, this approach should also be evaluated for salivary glands and tumours, which may require tailored fitting strategies, thereby supporting the development of a generalisable STP method for personalised treatment. Prospective validation in larger cohorts remains necessary.
METHODS: Kidney biokinetic data were analysed using a training dataset of 500 virtual patients generated from a six-parameter sum-of-exponentials function previously identified within the PBMS NLMEM framework, and an external testing dataset comprising 10 clinical patients with low-volume metastatic hormone-sensitive prostate cancer. Reference absorbed doses were calculated with Mono-IFIT and PBMS NLMEM. Four STP methods (STPNLMEM, STPMLR, STPH, and STPM) were evaluated at (47.7±2.2) h post-injection. Accuracy was assessed using relative deviation (RD), mean absolute percentage error (MAPE), and root mean square error (RMSE).
RESULTS: The selected PBMS NLMEM model provided a better description of the renal time-activity data than Mono-IFIT according to predefined goodness-of-fit criteria and Akaike-weight-based model selection. STP performance rankings depended strongly on the chosen reference model. Against Mono-IFIT, STPM and STPH showed the smallest deviations (mean RD -2.3% and 3.3%; MAPE 3.5% and 4.5%; RMSE 4.6% and 5.3%, respectively), whereas STPNLMEM showed the largest deviation (mean RD 16.0%; MAPE 16.0%; RMSE 17.9%). Against PBMS NLMEM, STPNLMEM showed the closest agreement (mean RD -2.5%; MAPE 6.6%; RMSE 9.3%), whereas STPMLR, STPH and STPM underestimated absorbed dose (mean RD -14.1%, -13.1% and -17.8%; MAPE 15.2%, 13.1% and 17.8%; RMSE 17.3%, 14.8% and 19.1%, respectively).
CONCLUSION: The apparent accuracy of STP methods is highly dependent on the reference model. For [177Lu]Lu-PSMA-617 renal dosimetry, Mono-IFIT should not be considered the preferred reference for method validation. In this dataset, PBMS NLMEM provided the better reference framework, and STPNLMEM showed the closest agreement with it. To ensure broader applicability, this approach should also be evaluated for salivary glands and tumours, which may require tailored fitting strategies, thereby supporting the development of a generalisable STP method for personalised treatment. Prospective validation in larger cohorts remains necessary.