Kidney dosimetry in 777 patients during 177Lu-DOTATATE therapy: aspects on extrapolations and measurement time points

Abstract

Purpose: Fractionated peptide receptor radionuclide therapy (PRRT) with ¹⁷⁷Lu-DOTATATE is increasingly applied as an effective treatment for patients with disseminated neuroendocrine tumors. In parallel to dose planning before external beam radiation therapy, dosimetry is also needed to optimize PRRT to the individual patient. Accordingly, absorbed doses to organs at risk need to be calculated during PRRT, based on serial measurements of radioactivity distribution utilizing SPECT/CT. The dosimetry should be based on as few measurements as possible, while still retaining reliable results. The main aim of the present work was to calculate the fractional contribution of the extrapolations of the curve fits for the absorbed dose calculations to the kidneys. The secondary aim was to study agreement between absorbed dose (AD) and the effective half-life (t_eff) for the kidneys, estimated by means of measurements at one or two time points, in comparison to our current method employing three time points.

Methods: In 777 patients with disseminated neuroendocrine tumors undergoing PRRT, SPECT/CT over the abdomen was acquired at 1, 4, and 7 days after ¹⁷⁷Lu-DOTATATE infusion. The absorbed dose to the kidneys was calculated from SPECT/CT radioactivity distribution data, and the t_eff and fractional contributions of the extrapolations were estimated, utilizing data from one, two, and three time points, respectively.

Results: The fractional contributions from extrapolations before day 1 measurement and after day 7 measurement were approximately 26% and 11%, respectively. The mean differences in absorbed dose, based on one, two, and three time points were small, but with high method dependence for individual patients. The differences in estimated t_eff were small when it was based on measurements at days 1 and 7, but high for days 1 and 4 time points.

Conclusion: When assessing simplifications of methods for calculation of the absorbed dose to the kidneys, it was of the uttermost importance to incorporate the fractional contribution for the extrapolations included in the reference method. Measurements at an early and a late time point were found most important. An intermediate measurement contributes with an idea of the goodness of the fit.

Keywords: 177Lu-DOTATATE; Dosimetry; Extrapolation; Kidney; Neuroendocrine tumors; Simplifications.

Fig. 1

Examples of fractional contributions on a typical curve for the kidney for a (fc_0-24) the extrapolated portion of the curve from time = 0 to the measurement at day 1, b (fc_168-∞) the portion of the curve after the day 7 measurement (to infinity), c (fc_{0-24 + 168-∞}) the sum of fc_0-24 and fc_168-∞, d (fc_96-∞) the portion of the curve after the day 4 measurement (to infinity), e (fc_{0-24 + 96-∞}) the sum of fc_0-24 and fc_96-∞, and (fc_0-8) the fractional contribution of the first 8 h

Fig. 2

Fractional contribution of absorbed dose calculations for the left kidney to the total number of decays

Fig. 3

Histograms of the absorbed doses using our standard method AD_147, based on SPECT/CT measurements at days 1, 4, and 7, for the a left kidney and b right kidney

Fig. 4

Bland-Altman plots of the percent difference versus the mean for the left kidneys in our standard method AD₁₄₇ versus the simplified methods a AD₁₄, b AD₁₇, c AD_4/52, and d AD_4/H

Fig. 5

Bland-Altman plots of the percent difference versus the mean for the left kidneys in the simplified method AD₁₄ versus the single time point methods a AD_4/52 and b AD_4/H

Fig. 6

Histograms of the effective half-lives in the absorbed dose calculation using our standard method (AD₁₄₇) for the a left kidney and b right kidney and using the simplified method AD₁₄ for the c left kidney and d right kidney

Fig. 7

Bland-Altman plot of the effective half-life (t_eff) for the reference method using data from 1, 4, and 7 days (t_eff147) versus the simplified methods a using data from 1 and 4 days (t_eff14) and b using data from 1 and 7 days (t_eff17)