Higher 9.1 GBq Lutetium-177 PSMA Activity Leaves Tumor and Kidney Doses Unchanged

Patients with advanced prostate cancer who received a 23 percent higher injected activity of the radiopharmaceutical [177Lu]Lu-PSMA-I&T did not accumulate proportionally different radiation doses per unit of activity in their tumors or kidneys, according to a retrospective single-center study published in EJNMMI Physics. The analysis included 76 patients with PSMA-avid metastatic castration-resistant prostate cancer, of whom 54 were injected with 7.4 ± 0.1 GBq and 22 with 9.1 ± 0.3 GBq. Because absorbed dose per gigabecquerel remained comparable between the two activity levels, the findings suggest that within this range the dose deposition scales linearly with the amount of activity injected, without saturation, redistribution, or nonlinear pharmacokinetic behavior.

The standard activity of 7.4 GBq every six weeks traces back to the VISION trial era, when the therapy was approved on the basis of hematologic safety considerations, since blood count nadirs typically occur three to six weeks after administration of molecular radiopharmaceutical therapy. That regimen was chosen using dosimetry data and experience from the NETTER-1 trial. In the absence of organ-specific tolerance data for radiopharmaceutical therapy, administered activities have historically been constrained by absorbed dose limits borrowed from external beam radiation therapy, such as kidney thresholds of 18 to 23 Gy. Using a published average of 4.7 Gy per cycle of [177Lu]Lu-PSMA-I&T, only three or four cycles would reach such thresholds, leading the authors to argue that these limits may be overly conservative given the low-dose-rate exposure of radiopharmaceutical therapy compared with the high dose rate of external beam treatment.

Higher activities have already been explored elsewhere. German compassionate use studies reported activities up to 9.7 GBq per cycle, Australian clinical trials up to 8.7 GBq per cycle, and a phase 1 dose-escalation study, NCT03042468, safely administered up to 22.2 GBq per cycle with promising early signals of efficacy and tolerability. Yet a survey of 95 international theranostic centers found the mean standard injected activity for 177Lu-PSMA therapy to be 7.3 GBq per cycle, with a range of 5.5 to 11.1 GBq, and a recent meta-analysis covering injected activities from 3.6 to 7.5 GBq found that tumor absorbed dose did not change with activity. Direct comparisons across those studies are complicated by differences in imaging protocols, lesion segmentation methods, and dose calculation approaches, which all shape the reported estimates.

To address that gap, the new study compared two activity levels within a single center using one standardized dose estimation method. All patients underwent quantitative SPECT/CT imaging at 24, 48, and 72 hours after the first cycle on a Siemens Symbia system, with images reconstructed using a maximum-a-posteriori ordered-subset expectation-maximization algorithm that incorporated CT-based attenuation correction, Monte Carlo-based scatter correction, and resolution modeling. Lesions were segmented automatically with a 3.0 SUV threshold on the 24-hour images, while the functional units of the kidneys, the cortex and medulla, were segmented manually on the CT component. Time-activity curves for each volume of interest were fit with a mono-exponential function, and the time-integrated activity was convolved with a voxel S-value using a predefined 177Lu kernel to yield absorbed doses in gray. No spill-over or partial volume corrections were applied, and lesions smaller than 3 mL as well as rib lesions were excluded from dose calculations because of partial volume effects and registration or motion errors.

The two cohorts differed substantially in disease burden. Median total tumor volume was 189.8 mL, ranging from 4.2 to 2182.5 mL, in the 7.4 GBq group, compared with 947.3 mL, ranging from 21.7 to 4308.9 mL, in the 9.1 GBq group, a statistically significant difference. Because the higher-activity patients carried more disease, the researchers stratified their analysis by tumor volume, comparing patients with total tumor volumes of 100 to 380 mL and of 500 to 2000 mL. The 100 to 380 mL stratum contained 19 patients in the lower-activity group and 5 in the higher-activity group, while the 500 to 2000 mL stratum contained 11 patients in each arm.

Absorbed dose per gigabecquerel across the whole tumor burden was only slightly greater in the 9.1 GBq cohort, at 0.95 ± 0.42 versus 0.92 ± 0.52 Gy/GBq, a difference that was not statistically significant. Kidney doses were likewise comparable, at 0.29 ± 0.08 Gy/GBq in the higher-activity group versus 0.33 ± 0.13 Gy/GBq in the standard group. The hottest lesion in each patient received 1.37 ± 0.61 versus 1.30 ± 0.74 Gy/GBq, and the coldest lesion 0.40 ± 0.16 versus 0.50 ± 0.32 Gy/GBq, neither difference reaching significance. One exception emerged: bone lesions in the 9.1 GBq cohort received a significantly higher absorbed dose per gigabecquerel, while lymph node lesions showed no difference between the activity levels. Within the high tumor volume stratum of the 7.4 GBq group, bone lesions received significantly lower dose per gigabecquerel, and lymph node analysis in that stratum was limited to very small numbers of patients.

The study also tested for the tumor sink effect, a hypothesized phenomenon in which a large tumor burden would sequester the radiopharmaceutical and thereby lower the dose delivered per gigabecquerel to normal organs such as the kidneys. Plotting kidney absorbed dose per gigabecquerel against total tumor volume revealed no increase in kidney dose with decreasing tumor volume, indicating a lack of a clinically relevant tumor sink effect in patients with tumor volumes below 2000 mL. However, only four patients in the study had more than 2000 mL of disease, which prevented a sufficiently powered analysis of the very high burden range, and the authors note that a comprehensive analysis including more patients with increased disease burdens is required before drawing conclusions.

Beyond dosimetry, the study captured a clinically meaningful difference in prostate-specific antigen response. PSA values measured four weeks after the first cycle showed a median percent change from baseline of −1 percent, ranging from −86 to 371 percent, in the 7.4 GBq group, compared with −32 percent, ranging from −65 to 65 percent, in the 9.1 GBq group, a statistically significant difference. A PSA decline of at least 50 percent was achieved by 7 of 53 patients, or 13 percent, in the standard-activity group and 4 of 20 patients, or 20 percent, in the higher-activity group. Three patients could not be evaluated for PSA because their values were recorded only as greater than 1000 ng/dL. The authors caution that tumor absorbed dose is governed by heterogeneous microenvironmental and histopathologic factors such as perfusion variability, receptor density, necrosis, and fibrosis, which were not assessed at the tumor level, so the absence of change in dose per gigabecquerel does not inform tumor biological dose-response, and changes were interpreted globally with respect to PSA.

Safety findings come with important caveats. No adverse renal side effects were observed after the first cycle in either cohort, and no acute toxicities were found in the higher-activity group following the first administration. However, dosimetry was performed exclusively after the first treatment cycle, and because patients undergo multiple cycles at six-week intervals with evolving tumor biology and organ kinetics, this restriction limits assessment of chronic toxicity. Bone marrow, salivary glands, and lacrimal glands were outside the scope of the investigation, and accurate computation of absorbed dose to these organs needs to be addressed before toxicity thresholds can be defined. The authors also acknowledge limitations from the SUV-based thresholding segmentation approach, the lack of partial volume correction, and the use of only mono-exponential time-activity fits.

Based on their findings, the researchers suggest that future activity escalation studies exceeding 9.1 GBq could be warranted, provided other critical organs are carefully evaluated. Ongoing clinical trials are exploring whether greater injected activities can improve tumor control and further define toxicity thresholds, including LUMOD-ID, which tests modified [177Lu]Lu-DOTATATE delivery based on individualized dosimetry, and PSMA-Boost, which investigates intensified [177Lu]Lu-PSMA-617 treatment for patients with low PSMA-expressing disease. Other prospective trials are examining different aspects of regimen design, such as the time interval between consecutive cycles in FLEX-MRT and the total number of cycles in RIALTO. The authors are currently investigating the tumor sink effect further using a larger, bi-institutional cohort of patients.

Subject of Research: Absorbed dose comparison between 7.4 and 9.1 GBq injected activities of [177Lu]Lu-PSMA-I&T in metastatic castration-resistant prostate cancer

Article Title: Influence of increased injected activities up 9.0 GBq [177Lu]Lu-PSMA-I&T on absorbed doses to lesions and kidneys

Article References: Ells, Z., Hempel, L., Ells, J., Wu, M., Scheifele, M., Schoell, M., Siegmund, S. C., Zacherl, M., Werner, R. A., Calais, J., Dahlbom, M., Böning, G., Holzgreve, A., Delker, A., & Unterrainer, L. M. (2026). Influence of increased injected activities up 9.0 GBq [177Lu]Lu-PSMA-I&T on absorbed doses to lesions and kidneys. EJNMMI Physics, 13(1), Article 92. https://doi.org/10.1186/s40658-026-00932-x

Image Credits: AI Generated

DOI: 10.1186/s40658-026-00932-x

Keywords: 177Lu-PSMA-I&T, radiopharmaceutical therapy, absorbed dose, dosimetry, metastatic castration-resistant prostate cancer, kidney absorbed dose, tumor sink effect, total tumor volume, SPECT/CT, activity escalation, PSA response, prostate cancer