Disturbing the Redox Balance Using Buthionine Sulfoximine Radiosensitized Somatostatin Receptor-2 Expressing Pre-Clinical Models to Peptide Receptor Radionuclide Therapy with 177Lu-DOTATATE

Simple Summary Peptide receptor radionuclide therapy with 177Lu-DOTATATE is an efficient treatment for patients suffering from metastasized neuroendocrine tumours. Nevertheless, suboptimal effects have been observed in the majority of patients. Hence, strategies to improve 177Lu-DOTATATE efficacy are desirable. Lu-177 induces oxidative stress, eventually leading to tumour cell death. Inhibition of the antioxidant defence mechanisms, using buthionine sulfoximine (BSO), represents an attractive strategy to increase 177Lu-DOTATATE efficacy. In cells and an animal model, the combination of 177Lu-DOTATATE and BSO was more effective than 177Lu-DOTATATE alone. In addition, it did not result in additional toxicity. Targeting the antioxidant defence system opens new safe treatment combination opportunities with 177Lu-DOTATATE. Abstract Peptide receptor radionuclide therapy with 177Lu-DOTATATE improves the outcome of patients with somatostatin receptor (SSTR)-expressing neuroendocrine tumours. Nevertheless, stable disease has been the main response pattern observed, with some rare complete responses. Lu-177 exerts about two-thirds of its biological effects via the indirect effects of ionizing radiation that generate reactive oxygen species, eventually leading to oxidative damage and cell death. This provides a rationale for targeting the antioxidant defence system in combination with 177Lu-DOTATATE. In the present study, the radiosensitizing potential and the safety of depleting glutathione (GSH) levels using buthionine sulfoximine (BSO) during 177Lu-DOTATATE therapy were assessed in vitro and in vivo using a xenograft mouse model. In vitro, the combination resulted in a synergistic effect in cell lines exhibiting a BSO-mediated GSH decrease. In vivo, BSO neither influenced 177Lu-DOTATATE biodistribution nor induced liver, kidney or bone marrow toxicity. In terms of efficacy, the combination resulted in reduced tumour growth and metabolic activity. Our results showed that disturbing the cell redox balance using a GSH synthesis inhibitor increased 177Lu-DOTATATE efficacy without additional toxicity. Targeting the antioxidant defence system opens new safe treatment combination opportunities with 177Lu-DOTATATE.


Introduction
One of the major successes in the field of nuclear medicine has been the radiotheranostic approach in neuroendocrine tumours (NET) by selectively targeting somatostatin receptor subtype 2 (SSTR2) with the radiolabelled somatostatin analogue (SSA) 177 Lu-DOTATATE. The landmark phase III randomized NETTER-1 clinical trial has validated the superiority of 177 Lu-DOTATATE over long-acting SSAs in terms of progression-free survival and quality of life [1,2] and has led to its regulatory approval (i.e., Lutathera ® ), which consequently highly impacted NET patient's management. Nevertheless, stable disease has been the main response pattern observed, with some rare complete responses (1% in the NETTER-1 trial) [1,3,4]. Diverse optimization strategies have therefore been investigated to improve patient outcomes after Lu-177-based radionuclide therapy. These strategies include dosimetry-based treatment individualization, new radiopharmaceuticals and combination therapies with radiosensitizing molecules [5][6][7][8][9], including the ones that target the DNA repair mechanisms [10][11][12]. Other potential targets for radiosensitization could be antioxidant defences. Cellular enzymatic and non-enzymatic detoxification systems exist to cope with excessive reactive species production from endogenous and exogenous sources, such as ionizing radiation (IR) [13]. As a β-emitter with a low linear energy transfer (LET), Lu-177 exerts about two-thirds of its biological effects via the indirect effects of IR that generate reactive oxygen species (ROS), eventually leading to oxidative damage and cell death. In our previous work, we have shown that antioxidant defences were significantly lower in cell lines with a higher sensitivity to 177 Lu-DOTATATE compared to cell lines with a lower sensitivity to 177 Lu-DOTATATE [14]. Consequently, targeting the antioxidant defences might represent a novel promising radiosensitizing strategy for peptide receptor radionuclide therapy (PRRT) with 177 Lu-DOTATATE.
Among the variety of cellular antioxidants, glutathione (GSH) is a highly abundant thiol-containing tripeptide (L-glutamate, cysteine and glycine) that plays a prominent role in redox homeostasis. It acts directly as an oxidant scavenger and indirectly as a cofactor of several enzymes such as glutathione peroxidase, a first-line defence antioxidant [15]. Increased GSH levels have been implicated in tumour cell resistance to several chemotherapeutics (alkylating agents [16,17], platinum-containing compounds [18,19] and anthracyclines [20]) via the elimination of ROS, among other mechanisms [21,22]. Therefore, strategies to deplete intracellular GSH levels have been developed. L-Buthionine-(S,R)-Sulfoximine (BSO) is a potent and specific inhibitor of γ-glutamylcysteine synthetase (γ-GCS) [23], the rate-limiting factor in the biosynthesis of GSH. BSO causes intracellular GSH depletion by preventing its de novo synthesis [24,25], thereby enhancing sensitivity to various anti-cancer agents in vitro and in vivo [26,27]. In the clinical setting, BSO was mainly studied in combination with the alkylating agent melphalan [25].
In this study, we assessed the radiosensitizing potential of depleting GSH using BSO in combination with 177 Lu-DOTATATE in vitro using a panel of human cancer cell lines expressing SSTR as well as in vivo using a multiple myeloma xenograft mouse model. Indeed, several evidence from the literature on SSTR expression in multiple myeloma cell lines [14,28,29] and patients [30][31][32][33] led us to consider multiple myeloma as a potential new indication for PRRT with 177 Lu-DOTATATE. Our results showed that disturbing the cell redox balance using a GSH synthesis inhibitor increased 177 Lu-DOTATATE efficacy without additional toxicity.

Production of 177 Lu-DOTATATE
The radiopharmacy facility in the nuclear medicine department at Institut Jules Bordet produced 177 Lu-DOTATATE for clinical use, as previously described by [38]. Labelling was carried out with a fully automated process using a synthesis module with disposable cassettes (Modular Pharmlab, Eckert & Ziegler, Hopkinton, MA, USA). Nine GBq of noncarrier added 177 LuCl3 (EndolucinBeta ® , Ph. Eur, ITM, Garching/Munich, Germany) and 150 µg DOTATATE (Bachem AG, Bubendorf, Switzerland) in sodium ascorbate/acetate buffer were heated for 20 min at 80 • C. The obtained raw radioactive solution was purified using solid-phase extraction on a C18 cartridge. The radiolabeled peptide was then eluted with 1 mL of 50% ethanol, followed by 19 mL of saline and subsequent sterile filtration over a 0.22 µm filter (included in the disposable cassette). All quality controls were performed according to the European Pharmacopoeia, allowing the conditional release of the radiopharmaceutical after appearance, pH, radiochemical purity (specification: >95%) and pyrogenicity testing, and subsequent final release after sterility results.

In Vitro Treatments
Cells were seeded in 12-well plates (Corning ® CellBIND ® Multiple Well Plate, Merck, Rahway, NJ, USA) at different densities (HBL and COLO-677: 1000 cells/well; MM162: 2000 cells/well; EJM: 4000 cells/well; MIA-PACA-2 and HT-29: 200 cells/well) and, at least 4 h later, pre-treated or not with BSO (10 −7 M, L-buthionine sulfoximine, B2515, Merck). A low to non-toxic concentration of BSO was chosen to have minimal effect on cell survival as a monotherapy. The next day, cells were exposed to 177 Lu-DOTATATE, as previously described [14]. Five MBq of 177 Lu-DOTATATE were added in each well (four replicates) for 4 h. Subsequently, as well as three days later, the radioactive medium was removed and replaced with fresh medium with or without BSO. A crystal violet assay was performed on day 10.
The coefficient of drug interaction (CDI) was used to analyse the interactions between 177 Lu-DOTATATE and BSO.

Cell Survival Assay
Cell survival was assessed with crystal violet. The culture medium was removed, and cells were gently washed with phosphate-buffered saline (PBS), fixed in 1% glutaraldehyde (Merck) in PBS for 15 min and stained with 0.1% crystal violet (Sigma) in water for 30 min. The plates were washed under running tap water and subsequently lysed with 0.2% Triton X-100 (Roche, Basel, Switzerland) in water for 90 min. The associated absorbance was measured at 590 nm using a BioTek ® 800™ TS Absorbance Reader (Hong Kong).

Animals and Treatments
Mice experiments were approved by the animal ethics committee of Université Libre de Bruxelles (Comité d'Ethique du Bien-Etre Animal (CEBEA)) and performed according to European guidelines. Female athymic nude Foxn1 nu mice (Charles River) (age = 5 to 6 weeks) were subcutaneously injected in the right flank with 1 × 10 6 EJM cells suspended in 100 µL of PBS and matrigel (1:1; Cultrex™ Basement Membrane Extract, Type 3, R&D Systems). The tumours were allowed to grow for 4 weeks before initiating treatments. At a tumour size of approximately 0.5 cm 3 , mice were randomized into 4 treatment groups of 10 mice: control, 177 Lu-DOTATATE, BSO and 177 Lu-DOTATATE + BSO. BSO (10 mM, L-buthionine sulfoximine, B2515, Merck) was administered via drinking water [39] for 3 weeks, starting the day before a single intravenous (IV) injection of 30 MBq 177 Lu-DOTATATE or saline. All mice were monitored three times a week for body weight and tumour growth using a calliper. Tumour volume (mm 3 ) was calculated using the formula π 6 × length × width × height [40]. The relative tumour volume was represented and calculated as (tumour volume on measured day)/(tumour volume measured at baseline). Mice were sacrificed using cervical dislocation under anaesthesia once tumours exceeded the maximal ethical volume. Another group of mice receiving 177 Lu-DOTATATE alone or in combination with BSO was sacrificed 24, 72 and 168 h post-injection (p.i.) for ex vivo biodistribution studies and early glutathione quantification (4 mice per group per time point) (Figure 1).

Cell Survival Assay
Cell survival was assessed with crystal violet. The culture medium was removed, and cells were gently washed with phosphate-buffered saline (PBS), fixed in 1% glutaraldehyde (Merck) in PBS for 15 min and stained with 0.1% crystal violet (Sigma) in water for 30 min. The plates were washed under running tap water and subsequently lysed with 0.2% Triton X-100 (Roche, Basel, Switzerland) in water for 90 min. The associated absorbance was measured at 590 nm using a BioTek ® 800™ TS Absorbance Reader (Hong Kong).

Animals and Treatments
Mice experiments were approved by the animal ethics committee of Université Libre de Bruxelles (Comité d'Ethique du Bien-Etre Animal (CEBEA)) and performed according to European guidelines. Female athymic nude Foxn1 nu mice (Charles River) (age = 5 to 6 weeks) were subcutaneously injected in the right flank with 1 × 10 6 EJM cells suspended in 100 µL of PBS and matrigel (1:1; Cultrex™ Basement Membrane Extract, Type 3, R&D Systems). The tumours were allowed to grow for 4 weeks before initiating treatments. At a tumour size of approximately 0.5 cm 3 , mice were randomized into 4 treatment groups of 10 mice: control, 177 Lu-DOTATATE, BSO and 177 Lu-DOTATATE + BSO. BSO (10 mM, Lbuthionine sulfoximine, B2515, Merck) was administered via drinking water [39] for 3 weeks, starting the day before a single intravenous (IV) injection of 30 MBq 177 Lu-DOTA-TATE or saline. All mice were monitored three times a week for body weight and tumour growth using a calliper. Tumour volume (mm 3 ) was calculated using the formula π 6 × length × width × height [40]. The relative tumour volume was represented and calculated as (tumour volume on measured day)/(tumour volume measured at baseline). Mice were sacrificed using cervical dislocation under anaesthesia once tumours exceeded the maximal ethical volume. Another group of mice receiving 177 Lu-DOTATATE alone or in combination with BSO was sacrificed 24, 72 and 168 h post-injection (p.i.) for ex vivo biodistribution studies and early glutathione quantification (4 mice per group per time point) ( Figure 1).  18 F-FDG PET/CT to assess metabolic tumour response. Tumour growth was also followed using a calliper. Another group of mice receiving 177 Lu-DOTATATE alone or in combination with BSO was sacrificed on days 1, 3 and 7 p.i. for ex vivo biodistribution studies and early glutathione quantification. All animals were continuously monitored for body weight. IV = intravenous, p.o. = per os. . Mice had a baseline and a post-treatment 18 F-FDG PET/CT to assess metabolic tumour response. Tumour growth was also followed using a calliper. Another group of mice receiving 177 Lu-DOTATATE alone or in combination with BSO was sacrificed on days 1, 3 and 7 p.i. for ex vivo biodistribution studies and early glutathione quantification. All animals were continuously monitored for body weight. IV = intravenous, p.o. = per os.

Ex Vivo Biodistribution Studies and Time-Activity Curves
The amount of radioactivity was measured in different organs at 24, 72 and 168 h p.i. Animals were sacrificed, and then the organs were collected and weighed, and radioactivity was measured using a gamma counter (2480 WIZARD2, Perkin Elmer, Waltham, MA, USA). The biodistribution data were used to determine the kinetics of the activity in organs. Tissue uptake was calculated as the percentage of injected activity per gram of tissue (%IA/g). The measured activity data as a function of time were fitted with a single exponential curve using the least-square regression method with GraphPad Prism version 7.01.

Glutathione Quantification
Total glutathione was assessed in cell lines and mice tissues using the Quantification kit for oxidized and reduced glutathione (Sigma), according to the manufacturer's instructions. Cells in culture were harvested, washed and then lysed with freeze-thaw cycles and subsequently mixed with 5% 5-sulfosalicylic acid. After centrifugation at 8000× g for 10 min, the supernatant was collected, and the 5-sulfosalicylic acid concentration was reduced to 0.5% with water for glutathione quantification. Tumour, liver and kidneys were harvested, snapfrozen in liquid nitrogen and stored at −70 • C until assayed for glutathione. Tissues were homogenized in 5% 5-sulfosalicylic acid in lysing matrix D tubes (MP Biomedicals) containing ceramic beads using a FastPrep-24 homogenizer (MP Biomedicals, Irvine, CA, USA) and centrifuged at 8000× g for 10 min. The supernatant was collected, and the 5-sulfosalicylic acid concentration was reduced to 0.5% with water for glutathione quantification. The buffer solution was added to samples (either from cell lines or tissues) (in triplicates) in a 96-well plate. After 1 h incubation at 37 • C, the substrate (5,5 -dithiobis (2-nitrobenzoic acid)) (DTNB), coenzyme and enzyme (glutathione reductase) were added into the wells, followed by 10 min incubation at 37 • C. Absorbance was read at 405 nm using a microplate reader (Thermo Labsystems Multiskan EX, ThermoFisher, Waltham, MA, USA). Total glutathione concentration was determined using a calibration curve.

In Vivo Imaging and 18 F-FDG PET/CT Images Analysis
Baseline 18 F-FDG PET/CT imaging was performed 4 to 5 days before 177 Lu-DOTATATE injection and repeated 3 weeks p.i. to assess metabolic tumour response. Studies were conducted using a µPET-CT (nanoScan PET-CT, Mediso, Budapest, Hungary) on all mice (40 mice, 10 per group) at baseline and on the remaining mice 17 days post-treatment (25 mice, 4-7 mice per group). First, 18 F-FDG was injected in the tail vein, and then the mice were imaged 1 h p.i. Animals were scanned under isoflurane anaesthesia (induction: 3% isoflurane, 3L O 2 ; maintenance: 1.5% isoflurane, 1.5 L O 2 ) and kept at 37 • C using a thermoregulation unit (Minerve) during image acquisition. PET/CT image analysis and quantification were performed using MIM software. A three-dimensional anatomical contour of the tumour (volume of interest (VOI)) was manually drawn on the merged PET/CT images, and a threshold of 15% of the VOI maximum uptake value was applied to assess tracer uptake. To quantify radioactivity within the tumour, corresponding PET uptake values were normalized to the injected radioactivity per body weight. Changes in 18 F-FDG uptake before and after treatment are represented as fractional increase (value-baseline/baseline) in total lesion glycolysis (TLG) (TLG = metabolic tumour volume × SUVmean) (∆TLG). Changes in anatomical tumour volume before and after treatment are represented as fractional increase (value-baseline/baseline) in tumour volume (∆Vol T ).

Ex Vivo Tissue Preparation and Hematoxylin and Eosin (H & E) Staining
Liver and kidneys were harvested, fixed in PAXgene Tissue FIX (Qiagen, Hilden, Germany) for 1 day at room temperature and kept at 4 • C in PAXgene Tissue STABILIZER (Qiagen) until they were embedded in paraffin for histological studies. Staining was carried out on 4 µm paraffin-embedded tissue sections subsequently deparaffinized in xylene and rehydrated in decreasing alcohol concentrations. Slides were incubated in Mayer's hematoxylin for 6 min and subsequently in eosin for 3 min. Slides were mounted using Tissue-Tek Prisma (Sakura, Alphen aan den Rijn, The Netherlands). Digital scans of slides were acquired with Nanozoomer NDP (Hamamatsu Photonics, Shizuoka, Japan) and analysed by an experienced clinical pathologist.

Bone Marrow Assessment
At the time of sacrifice, bone marrow (BM) was harvested from both femurs (containing hematopoietically active marrow) of the animals. Femurs were excised from each mouse, and then the attached muscle was trimmed from the periosteal surface and the bone was opened at the distal end. BM was isolated using the cytocentrifuge method [41]. In brief, the femur was placed in a 0.5 mL Eppendorf tube pre-punctured with a needle at the base of the tube to form an opening. The 0.5 mL tube was nested within a 1.5 mL Eppendorf tube, and both were centrifuged at 10,000 rpm for 15 s to form a BM pellet at the bottom of the 1.5 mL Eppendorf tube yet to retain the femur within the 0.5 mL tube.

•
Bone marrow smears and cytological assessment.
BM smears were prepared using the push-slide technique [42]. In brief, one drop of BM was placed at one extremity of a glass slide and was then spread in an even film using a second glass slide (held at an angle) as a spreader. The spreader slide was slid forward the length of the slide to produce a smear. Slides were fixed in absolute methanol for 5 min and then air-dried. May-Grunwald-Giemsa staining was used for routine evaluation of BM cytomorphology. Smears were incubated for 10 min in the May-Grunwald stain (50%) diluted with pH 6.8 water, followed by a 25 min incubation step in the Giemsa stain (10%) diluted with pH 6.8 water. Slides were then rinsed 3 times with pH 6.8 water before being air-dried. A qualitative and quantitative cytological evaluation of all the major lineages was performed by an experienced clinical pathologist on 300 to 500 cells [42].
BM samples were assayed for apoptosis. A 10 min incubation step at 37 • C with NH4Cl was performed for red blood cell lysis. Samples were then centrifuged for 5 min, and the supernatant containing lysed red blood cells was discarded. Samples were then incubated with Annexin-V Binding Buffer (ThermoFisher, Waltham, MA, USA), Annexin V (BD Biosciences, Franklin Lakes, NJ, USA) and 7-AAD (BD Biosciences, Franklin Lakes, NJ, USA) for 15 min at room temperature. Annexin Buffer was added to samples that were subsequently analysed using a MACSQuant Analyzer 10 Flow Cytometer (Miltenyi Biotec, Bergisch Gladbach, Germany).

Statistical Analyses
Statistical analyses were performed with GraphPad Prism 7.01 (GraphPad Software). After normality verification using the Shapiro-Wilk test, a parametric t-test was performed when comparing 2 means. A two-way ANOVA with a Turkey's multiple comparison test was used to assess differences in apoptotic stages between treatment groups. A one-way ANOVA or Kruskal-Wallis test was used to assess differences in animal groups. Outliers were identified using the Grubb's method and excluded from statistical analysis.

BSO Did Not Influence 177 Lu-DOTATATE Biodistribution nor Pharmacokinetics In Vivo
An ex vivo biodistribution study was performed to evaluate the impact of BSO on 177 Lu-DOTATATE body distribution. The activity concentrations in normal tissues are shown in Table S1. Twenty-four hours p.i., the highest uptake of 177 Lu-DOTATE alone or in combination was found in kidneys and urine, reflecting the known urinary excretion of 177 Lu-DOTATATE as well as partial reabsorption of the radiotracer. In addition, 177 Lu- and GEP (MIA-PACA-2 and HT-29) cell lines exposed to BSO. Total glutathione quantification was performed using cell lines exposed to 10 −7 M BSO for 18 h. Results are expressed as a percentage of the non-treated counterpart (black dotted line) and are represented as mean ± SEM (n = 6 from 2 independent experiments). *** p ≤ 0.001; ** p ≤ 0.01; * p ≤ 0.05; NS = non-statistically significant.

BSO Did Not Influence 177 Lu-DOTATATE Biodistribution nor Pharmacokinetics In Vivo
An ex vivo biodistribution study was performed to evaluate the impact of BSO on 177 Lu-DOTATATE body distribution. The activity concentrations in normal tissues are shown in Table S1. Twenty-four hours p.i., the highest uptake of 177 Lu-DOTATE alone or in combination was found in kidneys and urine, reflecting the known urinary excretion of 177 Lu-DOTATATE as well as partial reabsorption of the radiotracer. In addition, 177 Lu-DOTATATE accumulated in SSTR2-expressing organs such as the pancreas and stomach. No statistically significant differences in the 177 Lu-DOTATATE uptake between the two treatment groups ( 177 Lu-DOTATATE alone and its combination with BSO) were observed in any of the investigated organs/tissues and time points (Figure 4).  (HBL and MM162), multiple myeloma (COLO-677 and EJM) and GEP (MIA-PACA-2 and HT-29) cell lines exposed to BSO. Total glutathione quantification was performed using cell lines exposed to 10 −7 M BSO for 18 h. Results are expressed as a percentage of the non-treated counterpart (black dotted line) and are represented as mean ± SEM (n = 6 from 2 independent experiments). *** p ≤ 0.001; ** p ≤ 0.01; * p ≤ 0.05; NS = non-statistically significant.

BSO Did Not Influence 177 Lu-DOTATATE Biodistribution nor Pharmacokinetics In Vivo
An ex vivo biodistribution study was performed to evaluate the impact of BSO on 177 Lu-DOTATATE body distribution. The activity concentrations in normal tissues are shown in Table S1. Twenty-four hours p.i., the highest uptake of 177 Lu-DOTATE alone or in combination was found in kidneys and urine, reflecting the known urinary excretion of 177 Lu-DOTATATE as well as partial reabsorption of the radiotracer. In addition, 177 Lu-DOTATATE accumulated in SSTR2-expressing organs such as the pancreas and stomach. No statistically significant differences in the 177 Lu-DOTATATE uptake between the two treatment groups ( 177 Lu-DOTATATE alone and its combination with BSO) were observed in any of the investigated organs/tissues and time points (Figure 4). Time-activity curves were equivalent between both groups for the liver and kidneys, with overlapping 95% confidence bands, demonstrating similar pharmacokinetics ( Figure  5). No differences in effective half-life between the groups were observed in the kidneys

BSO Reduced Glutathione Levels in the Liver, Kidneys and Tumour
Total glutathione levels were checked in the liver, kidneys and tumour. Liver glutathione levels in the combination group were 30% lower than in the 177 Lu-DOTATATE group, and this reduction was maintained over time (although, at the statistical limit on
BSO was administered via drinking water; therefore, mice water consumption was monitored to verify it was not influenced by the presence of BSO. There were no significant differences in water intake between the different treatment groups: control: 5.0 ± 0.6 mL/day/mouse; BSO: 5.6 ± 0.6 mL/day/mouse; 177 Lu-DOTATATE: 5.2 ± 0.5 mL/day/mouse; and 177 Lu-DOTATATE + BSO: 5.1 ± 0.7 mL/day/mouse) ( Figure S1). This resulted in a BSO dose of 2 mmol/kg/day.

The Combination of 177 Lu-DOTATATE with BSO Did Not Result in Signs of Additional Hemato-, Nephro-or Hepatotoxicity
The evaluated hematotoxicity in mice sacrificed post-177 Lu-DOTATATE injection (29 to 38 days), revealed no significant difference in the number of BM apoptotic cells among the different treatment groups (control: 28.6 ± 3.2%; 177 Lu-DOTATATE: 33.0 ± 2.8%; 177 Lu-DOTATATE + BSO: 30.6 ± 2.6%) ( Figure 8A). Furthermore, the assessment of the relative proportion of the different BM lineages revealed no significant differences between treatment groups for any of the erythroid, granulocytic or lymphoid lineages ( Figure 8B-D). In addition, the qualitative assessment of BM smears did not show any morphological abnormalities in the major BM lineages, including the megakaryocytes population.

The Combination of 177 Lu-DOTATATE with BSO Did Not Result in Signs of Additional Hemato-, Nephro-or Hepatotoxicity
The evaluated hematotoxicity in mice sacrificed post-177 Lu-DOTATATE injection (29 to 38 days), revealed no significant difference in the number of BM apoptotic cells among the different treatment groups (control: 28.6 ± 3.2%; 177 Lu-DOTATATE: 33.0 ± 2.8%; 177 Lu-DOTATATE + BSO: 30.6 ± 2.6%) ( Figure 8A). Furthermore, the assessment of the relative proportion of the different BM lineages revealed no significant differences between treatment groups for any of the erythroid, granulocytic or lymphoid lineages ( Figure 8B-D). In addition, the qualitative assessment of BM smears did not show any morphological abnormalities in the major BM lineages, including the megakaryocytes population. Analysis of the H&E-stained slides of the kidneys revealed no morphological alterations and no inflammatory infiltrates in the renal cortex of any of the mice of the four treatment groups (Figure 9). Similarly, the liver tissues showed no signs of toxicity ( Figure  10). Analysis of the H&E-stained slides of the kidneys revealed no morphological alterations and no inflammatory infiltrates in the renal cortex of any of the mice of the four treatment groups (Figure 9). Similarly, the liver tissues showed no signs of toxicity ( Figure 10).

The Combination of 177 Lu-DOTATATE with BSO Reduced Tumour Growth
The efficacy of 177 Lu-DOTATATE (mean injected activity: 29.5 ± 0.4 MBq) and its combination with BSO was assessed using the EJM multiple myeloma xenograft model. The mean tumour size at the time of treatment administration was 487 ± 295 mm 3 , with no

The Combination of 177 Lu-DOTATATE with BSO Reduced Tumour Growth
The efficacy of 177 Lu-DOTATATE (mean injected activity: 29.5 ± 0.4 MBq) and its combination with BSO was assessed using the EJM multiple myeloma xenograft model. The

The Combination of 177 Lu-DOTATATE with BSO Reduced Tumour Growth
The efficacy of 177 Lu-DOTATATE (mean injected activity: 29.5 ± 0.4 MBq) and its combination with BSO was assessed using the EJM multiple myeloma xenograft model. The mean tumour size at the time of treatment administration was 487 ± 295 mm 3 , with no significant differences in the tumour volume between treatment groups. Longitudinal assessment of tumour growth revealed an increase in subcutaneous tumour volume in all treatment groups with different tumour growth rates ( Figure 11A,C). Tumour growth slowed down with an increased doubling time induced with 177 Lu-DOTATATE alone (9.8 days) and further enhanced in combination with BSO (12.4 days) compared to the control (7.4 days) or BSO alone (7.5 days) ( Figure 11C). Three weeks after the start of the treatment (day 45 post-inoculation), the relative tumour volume was significantly lower in the combination group (4.4, IQR: 2.7) compared to the control (8.4, IQR: 6.2, p = 0.03) and BSO (7.6, IQR: 7.3, p = 0.02) groups but not to 177 Lu-DOTATATE alone (6.1, IQR: 5.3, p = 0.32) ( Figure 11B). This was corroborated by a significantly smaller ∆Vol T , measured with CT, in the combination group compared to the control group (4.5, IQR: 3.8 versus 11.1, IQR: 8.3, p = 0.04), but not compared to the other groups (BSO: 9.1, IQR: 9.3, p = 0.15; 177 Lu-DOTATATE: 7.5, IQR: 11.9, p = 0.27) ( Figure 11D).

The combination of 177 Lu-DOTATATE with BSO Reduced 18 F-FDG Metabolic Activity
The effects of 177 Lu-DOTATATE and BSO on tumour metabolism were assessed using the changes in 18 F-FDG uptake (mean injected activity: 4.44 ± 0.30 MBq). At baseline, there was no significant difference in TLG among the four treatment groups. A statistically significant lower ∆TLG was observed in the combination group (3.2, IQR: 1.9) compared to the control (8.5, IQR: 4.4, p = 0.02) and BSO (8.8, IQR: 9.3, p = 0.04) groups but not to 177 Lu-DOTATATE alone (4.8, IQR: 7.9, p = 0.20) (Figure 12). before and after treatment, measured on the CT part of the PET/CT, represented as a fractional increase in tumour volume (ΔVolT). Only significant differences are indicated, where * p ≤ 0.05.

Discussion
177 Lu-DOTATATE has significantly changed the therapeutic landscape of SSTR-expressing NETs, resulting in increased survival and quality of life for NET patients [1]. Nevertheless, complete responses are rare, and optimization strategies are being investigated to improve the outcome for 177 Lu-DOTATATE-treated patients. Potentiating the effects of 177 Lu-DOTATATE, through its combination with radiosensitizing molecules, appears to represent a promising approach [5][6][7][8]. By the same token, targeting specific molecular pathways, based on a robust radiobiology rationale, could result in desired synergistic effects. In this study, we assessed the radiosensitizing potential of GSH depletion using BSO both in vitro, using different cancer cell lines, and in vivo, using an SSTR-expressing multiple myeloma xenograft model. The multiple myeloma model was chosen for the reason that it is a radiosensitive malignancy expressing SSTR that could allow us to consider it as a novel indication for PRRT.
In our SSTR-expressing human cancer cell line panel, BSO had a synergistic effect when combined with 177 Lu-DOTATATE (Table 1), which is conceivably explained by the

Discussion
177 Lu-DOTATATE has significantly changed the therapeutic landscape of SSTR-expressing NETs, resulting in increased survival and quality of life for NET patients [1]. Nevertheless, complete responses are rare, and optimization strategies are being investigated to improve the outcome for 177 Lu-DOTATATE-treated patients. Potentiating the effects of 177 Lu-DOTATATE, through its combination with radiosensitizing molecules, appears to represent a promising approach [5][6][7][8]. By the same token, targeting specific molecular pathways, based on a robust radiobiology rationale, could result in desired synergistic effects. In this study, we assessed the radiosensitizing potential of GSH depletion using BSO both in vitro, using different cancer cell lines, and in vivo, using an SSTR-expressing multiple myeloma xenograft model. The multiple myeloma model was chosen for the reason that it is a radiosensitive malignancy expressing SSTR that could allow us to consider it as a novel indication for PRRT.
In our SSTR-expressing human cancer cell line panel, BSO had a synergistic effect when combined with 177 Lu-DOTATATE (Table 1), which is conceivably explained by the decreased glutathione levels (up to −20%) (Figure 3). Conversely, in cells without a significant decrease in glutathione levels, the radiosensitization effect was absent (Figure 2). Similarly, in vivo, in our multiple myeloma model, BSO-mediated radiosensitization to 177 Lu-DOTATATE resulted in a significant decrease in tumour growth rate, tumour volume and tumour 18F-FDG uptake compared to the control and BSO groups (Figures 11 and 12).
Once more, the observed BSO radiosensitizing effect was associated with glutathione depletion under BSO administration (Figure 7).
The rationale to target GSH was based on our previous work suggesting a key role for cellular antioxidant defences in 177 Lu-DOTATATE radioresistance [14], and BSO was chosen for its potency in the inhibition of GSH as well as its favourable toxicity profile. The current study confirms the potency of BSO with tumour GSH depletion of~80%, which is consistent with studies using comparable BSO dosing schedules in mice resulting in an 80 to 95% reduction in various tumour types [22,43]. In terms of toxicity, in our in vivo model, BSO (2 mmol/kg/day) was well tolerated and did not lead to water consumption differences between the treatment groups ( Figure S1), which are consistent with similar observations from other groups [39,44]. Of note, in the clinical setting, early-phase clinical studies on solid tumours (melanoma, ovarian, pancreas, breast, colon, renal and lung cancer) appeared safe, with only grade 1-2 nausea and vomiting caused by BSO alone [45][46][47][48]. Nevertheless, combined with melphalan, grade 3-4 myelosuppression was observed in heavily pretreated patients [48]. Whether this degree of myelosuppression was related to the addition of BSO was uncertain [25]. In a phase I trial, BSO was shown to significantly impact melphalan pharmacokinetics, with decreased melphalan clearance and volume of distribution compared to melphalan given alone [46]. In our study, BSO administration did not result in a change in 177 Lu-DOTATATE biodistribution and pharmacokinetics nor increased toxicity of the combination compared to 177 Lu-DOTATATE alone. Indeed, the results of the biodistribution of 177 Lu-DOTATATE were in line with previous studies [49,50] and were not affected by BSO (Figures 4 and 5, Table S1).
Although 177 Lu-DOTATATE is a well-tolerated treatment with moderate toxicity, side effects may occur mainly in the BM and to a much lesser degree in the kidneys [51,52], making them the organs at risk (OAR). Therefore, BSO may potentially radiosensitize these OAR and consequently exacerbate 177 Lu-DOTATATE-associated adverse events. Furthermore, while the liver is not fundamentally considered as an OAR for 177 Lu-DOTATATE treatment alone, when combined with BSO, due to the critical role of GSH in detoxification functions within the hepatic cells, its depletion may induce 177 Lu-DOTATATE radiosensitization as well as impact normal hepatic function, potentially turning the liver into an AOR. As also reported by other groups [24,39,53], we demonstrated that BSO decreased kidney and liver glutathione levels ( Figure 6), with a greater decrease for the former reflecting the faster turnover of GSH in the kidneys. Of note, the GSH level fluctuation over time observed in our results may be related to GSH circadian rhythm [39,54]. Nevertheless, at the time of sacrifice, no signs of kidney ( Figure 9) or liver ( Figure 10) toxicity were observed in any of the treatment groups, even in the absence of a nephroprotective solution. Our results are consistent with the preclinical findings of Sun et al., who investigated a series of biochemical indicators of liver toxicity and found no differences in mice that received 30 mM BSO via drinking water compared to the control, suggesting that BSO was an effective and nontoxic method for lowering GSH levels [44]. Additionally, Skapek et al. showed that BSO-mediated GSH depletion prior to treatment with melphalan did not result in enhanced organ toxicity compared to melphalan alone in nude mice [55]. Their BSO dosing schedule consisted of seven doses of intraperitoneal BSO injections (2.5 mmol/kg) at 12 h intervals and concomitant oral administration via drinking water (20 mM), an even higher dose of BSO compared to our study. Furthermore, the Bailey et al. study on 28 patients receiving BSO alone or in combination with melphalan did not report any renal or hepatic toxicity [46].
The most common toxicity with 177 Lu-DOTATATE is haematological, with about 10% of patients developing grade 3/4 subacute haematological toxicities [1,3,52]. In our study, no BM toxicity was observed in the 177 Lu-DOTATATE group, and no negative impact of the addition of BSO on the hematopoietic system was observed, as shown by the preserved apoptosis levels, cell morphology and cell count. Except for some BM samples, the percentages of the major BM lineages were indeed within reported ranges for C57BL/6J mice [56] (Figure 8). Albeit the performed evaluation of BM was partial and not fully differential, by grouping several developmental stages, it is considered adequate to characterize the major cytological findings in the BM [42]. Furthermore, Svensson et al. showed no acute or long term (sacrifice at 6 months p.i.) BM toxicity in nude mice treated with higher activities of 177 Lu-DOTATATE (90 MBq), as measured by leucocyte count [57]. In our study, the follow-up time for toxicity assessment was limited, and possible long-term adverse effects may be encountered. Additionally, one should be cautious interpreting BM toxicity results in the preclinical setting as commonly used models (including our model) do not receive prior lines of treatments as those mainly found in the clinical setting. Furthermore, for multiple myeloma patients, an increased BM toxicity might be expected as a result of limited BM reserves (from previous sequential myelotoxic treatments) but also of higher BM absorbed doses of 177 Lu-DOTATATE (due to the intramedullary localisation of lesions).
Apart from being safe, the combination of PRRT and BSO was shown to be effective, achieving a significant decrease in tumour growth rate, tumour volume and tumour 18F-FDG uptake compared to the control and BSO but not to 177 Lu-DOTATATE alone (Figures 11 and 12). The low therapeutic effect of 177 Lu-DOTATATE in our in vivo mouse model could be justified by the fact that animals received a single administration of the radiotracer (as opposed to up to four injections in clinical practice) and that the mean tumour volume at baseline was quite high, with a great variation within groups (487 ± 295 mm 3 ). For the latter, different groups have shown the influence of tumour size on PRRT response, with the effects of 111In-DTPA-octreotide [58] or 177 Lu-DOTATATE [59] being less pronounced in rats bearing large CA20948 pancreatic tumours (>8 cm 2 for 111In-DTPAoctreotide and >1 cm 2 for 177 Lu-DOTATATE) compared to small ones (≤1 cm 2 ). Several reasons were considered to explain this observation and could be relevant to our model, such as less accessible area (due to poor vascularization and enhanced interstitial pressure) or less effective treatment (due to hypoxic regions or loss of SSTR expression following tumour dedifferentiation). In addition, Schmitt et al. found an inverse linear relationship between tumour size and 177 Lu-DOTATATE uptake in a small cell lung cancer mouse model [49]. Nevertheless, despite these drawbacks, a sub-lethal dose of BSO combined with a sub-lethal activity of 177 Lu-DOTATATE led to a significant efficacy of the treatment combination, highlighting the radiosensitizing potential of BSO.
Although we investigated GSH depletion-mediated radiosensitization of 177 Lu-DOTATATE using a multiple myeloma xenograft mouse model, the first application of the treatment combination in the clinic would likely be with the approved indication of 177 Lu-DOTATATE, namely NETs. A superior radiosensitization effect may even be expected in NETs given their higher GSH levels and their relatively radioresistant nature compared to multiple myeloma [14]. Nevertheless, multiple myeloma could represent a novel indication for PRRT with 177 Lu-DOTATATE. Indeed, several in vitro and in vivo studies, including our own, have reported the expression of functional SSTR in multiple myeloma [14,[28][29][30][31][32]. We have launched a phase-II trial to assess SSTR2 expression by 68Ga-DOTATATE PET/CT in symptomatic relapsing and refractory multiple myeloma patients (SCARLET trial-NCT04379817). Potential tumour responses may be anticipated in this radiosensitive malignancy and should be assessed with a representative BM environment, which was lacking in our mouse model.
The optimal combination treatment schedule would still have to be determined, which would be the one allowing the use of the minimal effective dose of BSO, i.e., the dose necessary to achieve sufficient tumour GSH depletion to have an efficient tumour radiosensitization, without excessive healthy tissue toxicity. The rapid clearance of BSO (estimated half-life in plasma of less than 2 h [46]) and the rapid rate of GSH turnover [60] would allow one to closely control GSH levels with BSO and allow for a GSH depletion only during tumour irradiation with 177 Lu-DOTATATE, limiting the potential toxicities.

Conclusions
In conclusion, we investigated for the first time the potential for disturbing the redox balance by decreasing GSH levels with BSO as a new radiosensitizing strategy for 177 Lu-DOTATATE in vitro and in vivo. We provided promising efficacy and safety data to support clinical trials using BSO in combination with 177 Lu-DOTATATE. Efficient tumour GSH depletion was achieved, without significant BSO-related toxicity, resulting in an increased therapeutic effect of 177 Lu-DOTATATE. Furthermore, we provided the first therapeutic data about 177 Lu-DOTATATE in a potential novel indication, multiple myeloma.

Data Availability Statement:
The data presented in this study are available in this article.