Improved safety and efficacy of 213Bi-DOTATATE-targeted alpha therapy of somatostatin receptor-expressing neuroendocrine tumors in mice pre-treated with l-lysine

Background Targeted alpha therapy (TAT) offers advantages over current β-emitting conjugates for peptide receptor radionuclide therapy (PRRT) of neuroendocrine tumors. PRRT with 177Lu-DOTATATE or 90Y-DOTATOC has shown dose-limiting nephrotoxicity due to radiopeptide retention in the proximal tubules. Pharmacological protection can reduce renal uptake of radiopeptides, e.g., positively charged amino acids, to saturate in the proximal tubules, thereby enabling higher radioactivity to be safely administered. The aim of this preclinical study was to evaluate the therapeutic effect of 213Bi-DOTATATE with and without renal protection using L-lysine in mice. Tumor uptake and kinetics as a function of injected mass of peptide (range 0.03–3 nmol) were investigated using 111In-DOTATATE. These results allowed estimation of the mean radiation absorbed tumor dose for 213Bi-DOTATATE. Pharmacokinetics and dosimetry of 213Bi-DOTATATE was determined in mice, in combination with renal protection. A dose escalation study with 213Bi-DOTATATE was performed to determine the maximum tolerated dose (MTD) with and without pre-administration of l-lysine as for renal protection. Neutrophil gelatinase-associated lipocalin (NGAL) served as renal biomarker to determine kidney injury. Results The maximum mean radiation absorbed tumor dose occurred at 0.03 nmol and the minimum at 3 nmol. Similar mean radiation absorbed tumor doses were determined for 0.1 and 0.3 nmol with a mean radiation absorbed dose of approximately 0.5 Gy/MBq 213Bi-DOTATATE. The optimal mass of injected peptide was found to be 0.3 nmol. Tumor uptake was similar for 111In-DOTATATE and 213Bi-DOTATATE at 0.3 nmol peptide. Lysine reduced the renal uptake of 213Bi-DOTATATE by 50% with no effect on the tumor uptake. The MTD was <13.0 ± 1.6 MBq in absence of l-lysine and 21.7 ± 1.9 MBq with l-lysine renal protection, both imparting an LD50 mean renal radiation absorbed dose of 20 Gy. A correlation was found between the amount of injected radioactivity and NGAL levels. Conclusions The therapeutic potential of 213Bi-DOTATATE was illustrated by significantly decreased tumor burden and improved overall survival. Renal protection with l-lysine immediately prior to TAT with 213Bi-DOTATATE prolonged survival providing substantial evidence for pharmacological nephron blockade to mitigate nephrotoxicity. Electronic supplementary material The online version of this article (doi:10.1186/s13550-016-0240-5) contains supplementary material, which is available to authorized users.


Background
Targeted alpha therapy (TAT) has shown great promise in the treatment of both micrometastatic [1] and large solid tumors in preclinical and clinical studies [2,3]. Alphaemitters emit high linear energy transfer (LET) α-particles, each causing dense ion pairs (2000-7000) within a relative short path length (50-100 μm) [3]. The radioactive decay of 213 bismuth ( 213 Bi, T 1/2 = 46 min) results in the emission of high-LET α-particles by 213 Bi self and by its daughter 213 Po around 100 keV/μm. Due to the relative short half-life of 213 Bi, 213 Bi can deliver a high radiation dose rate to the target within a relatively short period of time. These physical characteristics make 213 Bi, one of the most commonly used α-emitters for medical applications, with demonstrated promise as TAT in preclinical studies, in vivo imaging, and in clinical treatment of cancer patients [1,4,5].
Peptide receptor radionuclide therapy (PRRT) with radiolabeled somatostatin analogs is commonly employed in patients with inoperable neuroendocrine tumors (NETs) overexpressing somatostatin receptors subtype 2 (SSTR 2 ). Current radiopeptides include 177 Lu-[DOTA 0 ,Tyr 3 ]octreotate ( 177 Lu-DOTATATE) and 90 Y-[DOTA 0 ,Tyr 3 ]octreotide ( 90 Y-DOTATOC). Its efficacy depends on the radiation absorbed dose delivered to the tumor, which depends on SSTR 2 targeting efficiency, clearance kinetics, perfusion, distribution, and tumor mass. High-specific activity radiopeptides are required to deliver adequate radiation absorbed dose to tumors, as the mass of injected peptide is limited by the high affinity but low capacity of SSTR 2expression systems. The mass of injected peptide influences the pharmacokinetics (PK) and absorbed doses in organs and tumors [6]. Therefore, the mass of injected peptide should be optimized to deliver efficacious tumor doses while avoiding toxic absorbed dose to organs, especially to the dose-limiting organs the kidneys and bone marrow [7,8].
Radiolabeled somatostatin analogs are known to accumulate in the renal proximal tubules, due to their net charge, electrostatic forces, and charge distribution from metal-chelation [9,10]. This can result in a high absorbed dose and subsequent renal dysfunction. Co-infusion of Llysine/L-arginine has been shown to reduce renal uptake in patients receiving 177 Lu-DOTATATE or 177 Lu-DOTA-TOC PRRT by 30-50% [11].
Several preclinical studies showed that TAT with 213 Bi resulted in high renal accumulation of radioactivity [12], causing nephrotoxicity and decreased survival without renal protection compared to animals receiving protection [13,14]. Evidence of acute or chronic interstitial nephritis was found in a previous dose escalation study in AR42J tumorbearing rats using 213 Bi-DOTATOC [2]. Nephrotoxicity was observed to be moderate in a clinical trial of 213 Bi-DOTATOC, in combination with renal protection, in patients' refractory to 177 Lu-DOTATATE or 90 Y-DOTATOC PRRT [3]. Conventional approaches to determine kidney function use serum creatinine or nuclear medicine imaging with 99m Tc-MAG3 or 99m Tc-DSMA. However, these approaches are suboptimal to detect early-stage kidney disease. Several renal biomarkers are commercially available to determine acute or chronic kidney injury [15]. However, those biomarkers have not yet been applied in PRRT for detection of nephrotoxicity. Neutrophil gelatinase-associated lipocalin (NGAL) is among the promising renal biomarkers for detection of acute or chronic kidney injury in humans with high specificity and sensitivity [16]. Therefore, NGAL is an interesting renal biomarker to study nephrotoxicity caused by TAT with 213 Bi.
This study aimed to determine the suitability of 213 Bi-DOTATATE for TAT. Administration of 213 Bi-DOTA-TATE was optimized for in vivo applications in AR42J tumor-bearing mice. The rat AR42J tumor is known to express SSTR 2 at high density; this model is commonly used for investigations using somatostatin analogs and PRRT. Additionally, investigations were performed on increasing 213 Bi-DOTATATE's efficacy by using L-lysine as a renal protectant, radiation dosimetry to determine the mean radiation absorbed dose to the tumor and kidney, the resultant dose-effect relation, and a pilot study to evaluate NGAL as a kidney injury biomarker.
Methods 213 Bi-DOTATATE labeling 213 Bi was eluted from a 225 Ac/ 213 Bi generator (Oak Ridge National Laboratory) with 0.1 M/0.1 M HCl/NaI. The resultant elution containing 213 Bi (630-740 MBq) was used for labeling with 10 μg DOTATATE (BioSynthema) in a reaction vial including 0.15 M TRIS buffer and 2.6 mM ascorbic acid at pH 8.4. The reaction was incubated for 5 min at 95°C and cooled to ambient temperature for 2 min before adding 50 mM DTPA [17]. Instant thin-layer chromatography (ITLC-SG, Varian) was performed using 0.9% NaCl as mobile phase to determine the radionuclidepeptide incorporation yield. High-performance liquid chromatography (HPLC, Agilent) was performed to determine the radiochemical purity (RCP) of 213 Bi-DOTA-TATE, being defined as percentage of intact radiopeptide of interest compared to other detectable radioactive compounds in the same HPLC analysis. HPLC was performed using a reverse phase C 18 column (JT Baker, Bakerbond®, 4.6 × 250 mm) eluted with 0.1% TFA and methanol [18]. 111 In-DOTATATE labeling 111 InCl 3 (GE Healthcare) was added to a vial containing 0.03, 0.1, 0.3, 1, or 3 nmol DOTATATE. 115 In(NO 3 ) 3 (0.01 g/L, ICP standard) was added to form a 1:1 M ratio reaction to peptide, and NaOAc 4 M was used to adjust the pH to 4-5. The reaction was heated at 80°C for 20 min and cooled to ambient temperature for 5 min before the addition of DTPA (50 mM) to incorporate potential free 111 In 3+ . Incorporation yields of the labeled peptide were evaluated as described previously.

Animal model
Athymic male nu/nu mice (Tachonic), 6-8 weeks old, were used in all studies. Tumor models were established by inoculating 5 × 10 6 rat pancreatic tumor AR42J cells (American Type Culture Collection) with high SSTR 2 expression into the right hind flank of the animals. After 3 weeks, the tumor size reached approximately 200 mm 3 . All animal experiments were carried out following Institutional Animal Care and Use Committee-approved protocol.
Comparison of biodistribution profiles of 111 In-DOTATATE and 213 Bi-DOTATATE AR42J tumor-bearing animals were used for the comparison of the uptake of 111 In-DOTATATE versus (vs.) 213 Bi-DOTATATE in different organs and tumors. Biodistribution assays were performed with either 111 In-DOTATATE or 213 Bi-DOTATATE (0.3 nmol, n = 3/cohort). Animals were euthanized 10 and 60 min post-injection (p.i.) by CO 2 asphyxiation. Blood samples were collected, and the following organs were harvested and counted in a γcounter (PerkinElmer): tumor, blood, heart, adrenals, kidneys, stomach without content, pancreas, liver, testicles, urinary bladder, femur, femur marrow, pituitary, and muscles. The uptake was expressed as percentage of injected activity per gram of tissue (%IA/g). The actual weight of all organs was used to calculate %IA/g.

Biodistribution 111 In-DOTATATE
Xenograft AR42J nu/nu mice were used to determine PK as a function of injected mass of peptide (n = 4/cohort). Animals were injected intravenously (i.v.) via the tail vein with 0.03, 0.1, 0.3, 1, or 3 nmol (corresponding to 2 × 10 −3 , 7 × 10 −3 , 0.02, 0.07, and 0.22 mg/kg, respectively) of 111 In-DOTATATE (range 0.6-2.9 MBq). Animals were euthanized by CO 2 asphyxiation at 3, 10, 30, and 60 min p.i. Blood samples, organs, and femur-containing femur marrow were harvested and counted as described previously. The uptake was expressed as percentage of injected activity per gram of tissue (%IA/g). and were euthanized at 10 and 60 min p.i (n = 5/cohort). Blood, organs, and femur-containing femur marrow were collected and counted as described previously. The uptake was expressed as percentage of injected activity per gram of tissue (%IA/g). Tumor volume V(t) as a function of time was modeled according to the exponential growth function V(t) = V 0 × e kt , with k the growth constant, related to the doubling time T d by k ¼ ln 2 ð Þ T d . Each individual mouse V(t) in the control group was fitted with the exponential growth function to enable extrapolation of the growth beyond the time when the tumor volume exceeded the maximum. An average control growth curve was obtained by using the mean of the volume data together with the extrapolated growth data to the time points of the last surviving animal. Fitting was also performed for the therapy group with an exponential growth function, where the initial growth rate k 0 slowed down or turned into shrinkage with rate k 0 − k 1 at onset time point T 0 of therapy effect. Regrowth was modeled by exponential growth with rate k 0 − k 1 + k 2 , setting in after the volume nadir time point T 1 . This led to the function V t Maximum tolerated dose (MTD) of 213 Bi-DOTATATE in nontumor-bearing mice in combination of L-lysine MTD was defined as the highest dose given to the animals allowing 100% survival with no significant weight loss >15% throughout the experiment. Nontumor-bearing mice were randomly divided into seven cohorts used to evaluate MTD, six treatments and one control (n = 8/cohort); see Table 1. Cohorts(+) received i.p. injections of L-lysine (35 mg/200 μL) at 2-10 min prior to 213 Bi-DOTATATE administration via i.v. tail vein. Control mice received DOTA-TATE (4 × 0.3 nmol) on four consecutive days. The animals were followed for 90 days. Serum was analyzed for the biomarker neutrophil gelatinase-associated lipocalin (NGAL) using ELISA (R&D Systems, 450 nm). Survival analysis was plotted according to the Kaplan-Meier fit model.

Pharmacokinetics
Saturation of receptor-specific tumor uptake was investigated by determining the kinetics of the tumor uptake with increasing injected mass of peptide. The time-activity curves for the tumor and normal organs were fitted by singleexponential functions using Prism-5 (GraphPad). Goodness of fit was analyzed with the Pearson correlation coefficient R 2 > 0.8. Both F test and the Aikake information criterion were used to decide on the complexity of the curves.

Radiation dosimetry
Cumulated radioactivity in the tumor and normal organs were estimated by integrating the time-activity curves fitted to the 111 In-DOTATATE biodistribution data and folded with the decay curve of 213 Bi and its daughters. Dosimetry was performed according to the MIRD schema by using the spherical nodes S-factors from the Olinda/EXM software [19]. S-factors were interpolated from the actual weight of the organs and tissue. All organs and tissue were assumed to be spherical with a density of 1 kg/m 3 . Mean radiation absorbed doses were obtained as a function-injected mass of peptide, assuming a homogeneous distribution in the tumor and organs. The mean radiation absorbed dose obtained included the cumulative dose of α, β, and γ from all daughters of 213 Bi. Owing to the short path length of α-particles, only the self-dose within each organ was included. The threshold for lethality was determined with renal absorbed dose and injected activity as indicators. Logistic regression analysis was used to determine the LD 50 for presumed renal toxicity-related death.

Statistics
Data analyses, graphs, and calculations were performed in Prism-5. Mann-Whitney t test was used to calculate the significance. The results of statistical tests were considered significant when P was <0.05. Biodistribution data were expressed as mean ± standard deviation (SD) and tumor volume data as mean ± standard error (SEM). Binary logistic analysis (forced entry method) was performed with SPSS software (IBM SPSS statistics, version 20).
Renal activity at 60 min p.i. for 213 Bi-DOTATATE was significantly higher than 111 In-DOTATATE: 17.4 ± 2.2%IA/ Table 1   No differences in tumor uptake were found in animals w/wo L-lysine pretreatment. However, at 60 min p.i., a significant difference in stomach uptake was found in tumorbearing animals with and without L-lysine, 0.9 ± 0.1%IA/g versus 1.6 ± 0.5%IA/g, P = 0.0079. Figure 3 shows the uptake of 213 Bi-DOTATATE w/wo L-lysine pre-injection in different organs in tumor-and nontumor-bearing animals.
In tumor-bearing animals receiving L-lysine, the renal absorbed dose was 0.56 Gy/MBq vs. 1.1 Gy/MBq without L-lysine. In nontumor-bearing animals, the renal absorbed dose was 0.50 Gy/MBq with L-lysine versus 1.0 Gy/MBq without L-lysine; see estimated mean radiation absorbed dose in Additional file 1: Table S7.

MTD in nontumor-bearing mice administration w/wo L-lysine
Renal protection using L-lysine prior to 213 Bi-DOTATATE administration resulted in prolonged survival for both the medium-dose(+) (21.7 ± 1.9 MBq) and high-dose(+) cohorts (28.3 ± 0.8 MBq), Fig. 5. Medium-and high-dose cohorts without L-lysine showed reduced survival rates compared to medium-and high-dose cohorts pre-treated with L-lysine. No animals in high-dose(−) cohort survived beyond 40 days following treatment. No significant difference in survival was observed following low-dose administration of 213 Bi-DOTATATE with or without L-lysine (P = 0.32) or medium-dose with or without L-lysine (P = 0.06). Weight loss was observed in cohorts treated with medium-dose(−), high-dose(−), and high-dose(+) cohorts.
At 90 days post-treatment, all control animals survived. A survival rate of 87.5% was found in the low-dose(−) cohort, 62.5% in the medium-dose(−), and 0% in highdose(−), Fig. 5a. Cohorts receiving L-lysine pre-treatment, Fig. 5b, a very high survival rate was observed: 100% in the low-dose(+) and 100% in the medium-dose(+). In the high-dose(+) cohort, 75% of the animals survived. A median survival of >90 days was found in control and all cohorts except the high-dose without L-lysine (median survival of 24 days, P = 0.0012).
By integrating the radioactivity over time in the kidney, data obtained from biodistribution study w/wo pretreatment of L-lysine, a time-integrated activity coefficient (expressed as min/g tissue) of 6.0 ± 2.4 min/g in mice pretreated with L-lysine and 12.0 ± 3.7 min/g in mice without pre-treatment of L-lysine was found. Based on logistic regression analysis, a LD 50 of 20 ± 8 Gy was found; see Fig. 6. The number of mice that were euthanized within 90 days was indicated as a function of renal absorbed dose obtained from both biodistribution studies w/wo L-lysine.

Discussion
In this preclinical study, TAT with 213 Bi-DOTATATE was systematically studied to understand the injected mass of peptide-dependent uptake, radioactivity-related toxicity, and reduction in tumor burden. 111 In had already been used as a surrogate for 213 Bi earlier in other preclinical studies [20,21]. We demonstrated 111 In is an appropriate surrogate radionuclide for in vivo preclinical studies of PK in tumors allowing the results obtained from 111 In to be used for 213 Bi-dosimetry calculation. Both 213 Bi and 111 In form highly stable complexes with DOTA-somatostatin analogs, including DOTATATE, and show similar affinities for SSTR 2 in tumor.
For PRRT, it is essential to determine the optimal injected mass of radioligand by defining PK of radiopeptides in animal models, given that the injected mass of radioligand influences tumor uptake, the resultant radiation absorbed dose, and eventually the efficacy of the therapy. Moreover, increasing the injected mass of radioligand can diminish the pharmacological selectivity by binding to other SSTRpositive organs [22], which is not beneficial in the case of TAT and may cause off-target toxicities. The optimal injected mass of 111 In-OctreoScan® to obtain the best signal to background ratio for tumor versus other organs was reported as 0.07 nmol in mice (3.5 pmol/g mice) [23]. De Jong et al. showed a "bell-shape" curve for dependent tumor uptake in AR42J tumor-bearing rats as a function of injected mass of peptide, where 0.4 nmol 111 In-DOTATOC (1.8 pmol/g rat) gave the maximum tumor uptake [24].
In this study, the highest absorbed tumor dose (0.66 Gy/MBq) was found at injected mass of peptide of 0.03 nmol (1.07 pmol/g mice). However, lower and more practical specific activity (MBq/nmol) 213  Bi-DOTATATE to determine renal uptake as a significant difference was observed at 60 min p.i. between 111 In-DOTATATE and 213 Bi-DOTATATE. With an absence of SSTR 2 receptors in the kidney, the high renal uptake is not related to SSTR expression. The renal uptake of the labeled peptide is thought to be influenced by the difference in the electrostatic charge of DOTA complex with 111 In and 213 Bi [9,10,25], leading to different interactions with megalin or cubilin [26]. Furthermore, 213 Bi 3+ is known to bind strongly to metallothionein in the kidneys [27], which might lead to a high renal uptake. Apart from high renal uptake, a significantly higher uptake was also found in the pituitary and a higher radioactivity level in plasma. The pituitary gland is a very small organ. During organ harvesting, a systematic uncertainty is introduced by the chance to include surrounding tissue in the weight used for the uptake per gram calculation, resulting to an under-or overestimation of pituitary uptake, which might explain our findings. The high renal uptake and slow clearance rate of 213 Bi-DOTA-TATE indicates tubular reabsorption of 213 Bi-DOTATATE; this might be the cause of higher radioactivity in plasma as well. 111 In-DOTATATE showed a slightly significantly higher uptake than 213 Bi-DOTATATE in pancreas tissue, as yet we do not have an explanation for this difference.
In this study, we were not able to examine the differences in PKs of these radiopharmaceuticals in pituitary, In-DOTATATE still showed to be a proper substitute for tumor uptake, since the PK profile of the tumor uptake was similar to that of 213 Bi-DOTATATE. However, the use of a surrogate radionuclide should be carefully chosen, since each alternate radionuclide has limitations.

A B
Weight loss in animals is often an indicator of toxicity, and the most radiosensitive organs for PRRT are the bone marrow and kidney [8,28,29]. In this study, we observed severe weight loss in 67% of animals exposed to high-dose 213 Bi-DOTATATE (cumulative 33.1 ± 3.7 MBq), within 2 weeks after treatment, indicating acute toxicity. This might be explained by the high renal uptake resulting in a high renal absorbed dose, which increased the risk of acute nephrotoxicity due to limited sublethal damage tissue repair. To investigate acute renal toxicity, a short-term toxicity study over 90 days was performed instead of a follow-up period over 6-12 months, which is commonly performed to investigate long-term chronic nephrotoxicity. A significant reduction of renal activity (50%) was found in animals pretreated with L-lysine in this study. Our findings indicate that pre-treatment with L-lysine improved survival of animals receiving medium-and high-dose 213 Bi-DOTA-TATE resulting from the reduction of renal activity. Song et al. showed in their study a threefold reduction in renal activity following lysine pre-treatment [13]. This result differs significantly from our findings but might be attributed to their method of lysine application used during the therapy procedures, rather than immediately prior. Kobayashi et al. demonstrated that the kidney uptake was influenced by the timing of L-lysine administration [30], such that renal blocking by L-lysine was maximized when i.p. administration of L-lysine was given immediately before administration of the radiolabeled of anti-Tac murine MoAb fragment. In our study, we have chosen to start the therapy 2-10 min after i.p. administration of L-lysine to protect the kidneys, since DOTA-TATE is a relative small molecule and rapidly cleared from the blood. Radioactivity in the blood or uptake of the bone marrow is generally used as an indicator for myelotoxicity. Pre-administration of L-lysine did not significantly affect the radioactivity measured in neither whole blood nor femur uptake in tumor-bearing mice. The mean radiation absorbed dose for whole blood and femur (see Additional file 1) was 3.3 and 1.3 Gy in mice with pre-treatment of L-lysine in the high-dose cohort, whereas without L-lysine, these values were 2.7 and 1.0 Gy. These absorbed doses were lower than the MTD of 25 MBq 213 Bi-DOTA-AMBA in PC3-tumor-bearing mice, corresponding to the MTD at a mean absorbed dose of 4 Gy in the blood [31]. Therefore, we concluded the bone marrow is not a limiting organ in our study.
The LD 50 found for the renal absorbed dose was 20 Gy in this study. Acute renal toxicity at 100-140 Gy was reported by Behr et al., after administration of 90 Y-fab fragments, leading to death of all mice within 2-3 weeks [32]. This corresponds to our observation in cohorts after high-dose 213 Bi-DOTATATE administration, with more than 90% of the animals dead at radiation absorbed dose >28 Gy. Hence, Fig. 6 Dose-effect relation between mean renal radiation absorbed dose and the percentage of mice that died before the end of the experiment at 90 days. LD 50 kidney dose was 20 ± 8 Gy, a threshold LD 5 = 11 ± 4 Gy (P = <0.001). All points represent at least three mice the relative biological effect (RBE) was 4-5 for acute renal toxicity, leading to death within 2-3 weeks, when comparing the absorbed doses in both studies. This estimate for the RBE for 213 Bi-DOTATATE appears to be comparable to the RBE value of 4 used for delayed renal toxicity by Song et al. [13]. Specific uptake in functional units of the kidney might cause changes in radiation absorbed doses to radiation-sensitive structures like the glomeruli that could result in less damage than predicted from whole-organ radiation dosimetry. Small-scale micro-dosimetry using the sub-organ model of Hobbs et al. [33] indicates a possible lowering of the absorbed dose to the glomeruli when 213 Bi is taken up in the proximal tubules by 44% in comparison to homogeneous uptake in the mouse cortex assuming equal kinetics. We found no indication for this sparing effect; otherwise, the RBE would be in the order of 6-8. A direct comparative study would be needed to determine both the RBE and the PK of 213 Bi-and 90 Y-or 177 Lu-labeled peptides inside the kidneys and its functional units.
In this pilot study, NGAL was used as a biomarker to evaluate late-stage renal changes after therapy. NGAL is sensitive to acute kidney injury (AKI) for detection of renal functions in early nephrotoxicity state [16,34]. No nephrotoxicity was found in the low-dose(−) and low-dose(+) cohorts, corresponding to another study done using similar injected mass of radioactivity (MBq) 213 Bi-DOTATATE as TAT in nude mice in two different tumor models wherein nephrotoxicity was investigated by 99m Tc-DMSA as a kidney marker [35]. Overall, NGAL levels were lower in mice pre-treated with L-lysine than mice without pre-treatment at similar dose of 213 Bi-DOTATATE. However, no significant difference was found between those cohorts, which might be explained since NGAL was measured day 90 after TAT, whereby some repair and recovery of the kidney might already occur. Furthermore, the mean renal absorbed dose for the medium-dose(−) and the high-dose(+) cohorts was 23 and 16 Gy, respectively. These absorbed doses were similar to the calculated renal LD 50 , at which 50% of the treated animals would develop acute nephrotoxicity. The sigmoid dose-effect curve for renal toxicity (Fig. 6) shows a steep slope, contributing to a great variation in NGAL values observed at absorbed doses just above and below the LD 50 value. In this study, NGAL proved to be a valuable tool to examine AKI for TAT using 213 Bi as radionuclide supporting its use in future investigations of nephrotoxicity caused by 213 Bi. The use of NGAL as a biomarker of nephrotoxicity is feasible and cost-effective compared to conventional approaches to determine renal functionality in preclinical studies. Creatinine, the most commonly used parameter to determine kidney injury, lacks the ability to evaluate kidney injury at early stages following PRRT. 99m Tc-MAG3 and 99m Tc-DMSA for preclinical applications are invasive, by the use of high radioactivity for imaging, and require additional data extraction and analysis. In addition to NGAL, kidney injury molecule-1 (KIM-1) and cystatine-C are promising biomarkers for both acute and chronic kidney disease [15]. The ability to study early and late kidney injury is essential in TAT, using a combination of both conventional methods, and these commercially available biomarkers could provide more information leading to more understanding of the underlying mechanisms involved in kidney injury after TAT.