Deciphering the phenotypic responses to 177Lu-based radioligand therapy to identify efficacious combinations Authors: Marco Ranzani, Alessandro Cicconi, Guillermo Sastre-Moreno, Asmita Thapa, Elise Simon, Josefine Reber, Joseph Barlow, Marie Laure Boursier, Aurora Cerutti, Luisa Deberle, Luigi Del Pozzo, Elias Elinati, Lerin Geo, Vera Grinkevich, Charlene Hartnagel, Sravanth K. Hindupur, Katjusa Koler, Suraj Menon, Joana Neves, Barbara Schacher, Nan Shao, Eeson Rajendra, Markus Reschke, Tobias Schmelzle, Graeme C. M. Smith, Helen M. R. Robinson, Marta Cortes-Cros Published: 2026 ABSTRACT PurposeRadioligand therapy (RLT) delivers radionuclides, such as Lutetium-177 (177Lu), to selectively target cancer cells. Despite the favourable clinical outcomes of 177Lu-based RLT, complete cure remains infrequent. A comprehensive understanding of its mode of action (MoA) is essential for strategic design of novel combination approaches that could improve therapeutic responses.MethodsDNA damage response (DDR) markers were profiled by immunofluorescence in cancer cell lines treated with DOTA-chelated 177Lu ([177Lu]Lu-DOTA). Multiple DDR-deficient isogenic cell lines and DDR inhibitors were screened in viability assays with [177Lu]Lu-DOTA to identify sensitizers. 177Lu-RLT combination with an inhibitor of the non-homologous end joining (NHEJ) core factor DNA-PK, and its effect on viability and cell death, were further characterised in vitro and in vivo.Results[177Lu]Lu-DOTA treatment induced multiple DDR biomarkers indicative of DNA double-strand break (DSB) repair and genomic instability. Viability assays performed with DDR-deficient isogenic models and DDR inhibitors demonstrated that loss of DSB repair through NHEJ elicits the strongest sensitisation to [177Lu]Lu-DOTA. Consistently, DNA-PK inhibition strongly sensitised PSMA-positive prostate cancer cells to [177Lu]Lu-PSMA-617, exacerbated apoptosis and cell cycle arrest and improved efficacy in vivo, demonstrating good tolerability.ConclusionThese results provide novel mechanistic insights into 177Lu-based RLT and pinpoint NHEJ as a key pathway for the repair of 177Lu-induced DNA damage. DNA-PK inhibition strongly sensitises cancer cells to 177Lu-based RLT and results in stronger potentiation compared with other tested DDR inhibitor combinations, warranting the clinical exploration of 177Lu-based RLT and DNA-PK inhibitor (DNA-PKi) combination.Supplementary InformationThe online version contains supplementary material available at https://doi.org/10.1007/s00259-026-08092-x. Introduction Radioligand therapy is a therapeutic modality that utilizes radionuclide-conjugated molecules (including peptides, small molecules and antibodies) that target tumour-specific surface receptors, delivering the radioactive payload with high precision (reviewed in [1, 2]). Unlike earlier forms of radiation therapy that relied on the natural affinity of certain radioactive elements to human tissues (e.g. Iodine-131 to the thyroid gland or Radium-223 to bone), radioligand therapy (RLT) leverages tumour-specific or tumour-overexpressed surface antigens to selectively target cancer cells while minimizing damage to healthy tissues [1, 2]. Among the different approaches in RLT, the use of radiolabelled peptides and small molecules provides advantageous pharmacokinetic properties, low immunogenicity and a differentiated approach compared with alternatives such as radiolabelled antibodies (reviewed in [3]). The emergence of RLT as a groundbreaking therapeutic approach was marked by the FDA approval of [177Lu]Lu-DOTATATE, (commercially known as Lutathera®), targeting SSTR2 positive neuroendocrine tumours, and [177Lu]Lu-PSMA-617, (commercially known as Pluvicto®), designed for the treatment of PSMA-positive metastatic castration resistant prostate cancer (mCRPC). The therapeutic efficacy of [177Lu]Lu-DOTATATE and [177Lu]Lu-PSMA-617, which demonstrated substantial improvement of overall survival and progression free survival [4, 5], has sparked great interest for clinical and pre-clinical investigation on RLT, aiming to identify novel tumour targets, to assess the efficacy of different radionuclides and to explore combination therapies. Despite its clinical effectiveness, RLT is rarely curative [4, 5]. Both [177Lu]Lu-DOTATATE and [177Lu]Lu-PSMA-617 rely on the β emitter Lutetium-177 (177Lu) as the radioactive payload, and it is unclear how other radionuclides, such as the α-emitter Actinium-225 (225Ac), may influence the therapeutic responses, with several ongoing clinical trials aiming to elucidate this (e.g. AcTION trial NCT04597411 and TATCIST trial NCT05219500, reviewed in [6]). Considering that DNA damage is one of the most cytotoxic effects of ionizing radiation (IR) [7], improved therapeutic outcomes could also be achieved by combination with inhibitors of the DNA damage response (DDR), an intricate network of pathways that detect, signal and repair DNA damage (reviewed in [8]). Poly(ADP-ribose) polymerase (PARP) inhibitors, such as Olaparib and Talazoparib, which hinder single-strand break (SSB) repair by trapping PARP1/2 (reviewed in [9]), can sensitize some cancer cells to 177Lu-based RLT [10–12], with the combination with Olaparib being under clinical investigation (LuPARP trial NCT04375267 [13]), . However, sensitization to 177Lu-based RLT is not universally observed with PARP inhibitors [14, 15]. DNA-PKcs, the catalytic subunit of the DNA-PK (DNA-dependent protein kinase) complex and a core factor of the double-strand break (DSB) repair pathway non-homologous end joining (NHEJ) [16], was identified as a strong sensitizer to [177Lu]Lu-DOTA-octreotate in a genome-wide CRISPR screen performed with SSTR2 positive cells [17]. In agreement with these findings, the inhibition of DNA-PK sensitizes neuroendocrine tumour cells to [177Lu]Lu-DOTATATE in vitro and in vivo [18], as well as prostate and renal models to 177Lu-based RLT in vivo [19]. Furthermore, DNA-PK inhibition also sensitizes prostate xenografts to α-emitting radioligand conjugates [20]. Notwithstanding this evidence, the understanding of the phenotypic responses to 177Lu-based RLT, to the associated DNA damage and the pathways that mediate its repair remains limited, particularly in prostate cancer cells. This knowledge is of fundamental relevance to unveil potential vulnerabilities to 177Lu-based RLT and guide rational design of combination therapies. Herein, we have investigated the cellular phenotypes elicited by 177Lu-based RLT in prostate cancer models, comparatively studied 177Lu and X-ray induced DDR markers, and inquired DNA repair pathways by genetic and pharmacological impairment to identify effective combinations. Materials and methods Radiolabeling No-carrier-added (n.c.a.) [177Lu]LuCl3 (in 0.04 M HCl) and either labelling buffer 1 (28 mg gentisic acid, 925 mg sodium acetate and dissolved in 25 mL traceselect water 325 µL HCl 30% is added) or labelling buffer 2 (0.25 M sodium acetate buffer (sodium acetate/acetic acid) pH 5.0, containing 1 mg/mL ascorbic acid), were mixed in a total reaction volume of 200–300 µL. After measuring the activity, the right amount of PSMA-617 or DOTA was added. The reaction mixture was stirred at 95 °C, 450–600 RPM for 5–15 min. The incorporation was monitored by radio-HPLC and radio-TLC. The final product was diluted with either formulation buffer (900 mg sodium ascorbate, 4.5 mg DTPA and dissolved in 25 mL traceselect water) or PBS/saline to an activity concentration needed for in vitro (60 MBq/nmol [177Lu]Lu-PSMA-617 and 50 MBq/nmol [177Lu]Lu-DOTA, respectively) or in vivo (100 MBq/nmol for efficacy and 40 MBq/nmol for biodistribution of [177Lu]Lu-PSMA-617). Cell culture and media All cell lines were grown in standard culture conditions (5% CO2, 37 °C). Details on cell lines origin, media used and seeding densities for each assay are shown in Sup.Table 1. Additional details on the cell lines used and the generation of PSMA-expressing models are provided in Supplementary methods. Viability assays Cells were seeded in multi-well 96 plates and treated as described below. The viability readout was performed 8 days after the start of the treatment, adding the CellTiter-Glo reagent (Promega, cat. G7573) to the samples and reading the resulting luminescence at 560 nm with a GloMax Explorer Microplate Reader (Promega). Raw values were normalized for the average readout of the vehicle-treated samples, plotted in GraphPad Prism and fitted using the log(inhibitor) vs. response – Variable slope (four parameters) function to calculate the half maximal effective concentration (EC50). For combinations, the potentiation factor (PF) was calculated as the EC50 ratio between samples treated with 177Lu alone and samples treated with 177Lu and chemotherapies. For isogenic models, the PF was calculated as the EC50 ratio between the parental cell line and the KO model. Incucyte analysis Incucyte Caspase 3/7 dye (Sartorius, cat. 4440) and Incucyte Cytotox Red Dye (Sartorius, cat. 4632) were added to the cells in multi-well 96 plates after radioligand washout and, if needed, after chemotherapies re-drugging, at the final concentration of 5 µM and 0.5 µM, respectively. The plates were imaged every 4 h with an Incucyte S3 equipped with a SX5 G/R Optical Module using the adherent cell-by-cell analysis, with exposure times of 300ms and 1000ms for green and red channel, respectively. Confluency was measured as the percentage of phase area containing cells. Apoptotic and dying cells were defined as those cells with green and red fluorescence values, respectively, above a defined threshold, and their percentage on the total number of cells per well was calculated. Raw values were normalized by dividing for the average percentage of apoptotic/dying cells in the vehicle treated samples at matched time points. The obtained values were plotted against each 4 h-time point in scatter plots or against 24 h-interval time points in histograms. Uptake and internalization 300.000 cells of LNCaP clone FGC or 150.000 cells of the other cell lines (DU145, DU145 PSMA High, PC-3 and PC-3 MP9) were seeded in duplicates per condition in 12 well plates and treated with 37.5 kBq of [177Lu]Lu-PSMA-617 (60 MBq/nmol) or an equal volume of vehicle control (11.9–14.9 µM ascorbic acid, 493–597 µM sodium acetate buffer in PBS) for 24 h. After treatment, cells used for uptake analyses were washed with ice-cold PBS and directly lysed in 1 M NaOH. Cells used for internalization analyses were treated with a glycine stripping buffer (0.05 M glycine, 100mM NaOH, pH 2.8) for 10 min and subsequently washed with PBS prior to lysis in 1 M NaOH. For each sample, activity measured using a wizard gamma counter (Revvity) and protein concentration was determined by bichinconinic acid assay (BCA). For each sample, measured activity was compared to the total applied activity to calculate the percentage of total applied activity, normalized to fold difference of protein concentration of each sample relative to its vehicle control and corrected by subtracting the activity measured in the vehicle control. Immunofluorescence Immunofluorescence was performed on cells grown in multi-well 96 plates compatible with the Operetta CLS system (Cell Carrier-96, Ultra Microplates, PerkinElmer, cat. 6055300). At the appropriate time points, cells were fixed with 4% PFA for 10 min at room temperature, blocked for 1 h at room temperature in 0.5% BSA/0.5% Triton X-100/PBS and incubated overnight at 4 °C with the primary antibodies diluted in blocking buffer (see Sup.Table 2 for details on antibodies and dilutions used). The following day, cells were incubated for 1 h at room temperature with the appropriate secondary antibodies and DAPI (4’, 6-diamino-2’-phenylindole, dihydrochloride, Life Technologies, cat. 62248) diluted in blocking buffer, then washed and kept in PBS for imaging. Images were acquired with the Operetta CLS High-Content Analysis System (Revvity, aka PerkinElmer) using a 20X air objective, and adjusting exposure times to optimize the intensity of each DDR marker analysed. Cells and relative DDR foci were detected and analysed using the Harmony software to obtain the percentage of cells with at least 5 (RAD51) or 10 (γ-H2AX, 53BP1, pATM) foci. Micronuclei were defined as foci in the DAPI channel located within a 10 μm radius outside the nucleus (nuclear band) and nuclei with at least 2 micronuclei within the nuclear band were quantified. Micronuclei were not assessed in LNCaP clone FGC due to the cells morphology affecting the detection capabilities of the analysis pipeline. The results were normalised for the vehicle-treated samples included in the same plate as internal controls to obtain the induction fold change. Fold change from technical triplicates were averaged for each independent experiment and plotted in histograms showing the mean and standard error (SEM) for all the experiments performed. Cell cycle analysis Cell cycle analysis was performed on immunofluorescence samples using the Click-iT EdU Cell Proliferation Kit for Imaging, Alexa Fluor 647 (Life Technologies, cat. C10340). Cells were incubated with 10 µM EdU for 30 min at 37 °C, followed by fixation with 4% PFA for 10 min at room temperature. Detection of EdU was performed according to the kit manufacturer’s instructions, after which cells were incubated with DAPI in IF blocking buffer for 1 h at room temperature. Images were acquired as described above and the Harmony software was used for the analysis. Cell cycle profiles were obtained plotting EdU intensity (y axis) vs. DAPI intensity (x axis) for each cell. Threshold values in both the EdU and the DAPI axis were empirically set based on cell cycle profiles in control samples, allowing to distinguish cells in G1 (low EdU, low DAPI intensity), S (high EdU staining) and G2/M (low EdU, high DAPI intensity) phase. Radioligand treatment for in vitro assays Cells were treated with radioligands 24 h after seeding, with three samples drugged for each dose (three technical replicates). For monotherapy treatment, a 1:2 dilution curve of the radioligand preparation was diluted in the appropriate amount of cell media and added to each well to reach the required final concentration. As control, an equal amount of radioligand vehicle (either DOTA or PSMA-617) was added to the cells. [177Lu]Lu-DOTA was left with the cells until the endpoint, while for treatment of prostate models with [177Lu]Lu-PSMA-617, the media containing the radioligand-conjugate was removed after 24 h of treatment, the cells were washed with PBS and fresh media was added. Cells were left to grow under standard conditions until the experimental endpoint. For combination of [177Lu]Lu-DOTA with chemotherapies, a dilution curve of the appropriate concentration of compound and either a fixed dose of radioligand or an equivalent volume of vehicle was supplemented to the cells and left in the samples until the endpoint. For combinations of [177Lu]Lu-PSMA-617 and chemotherapies, a dilution curve of radioligand and either a fixed dose of compound or an equivalent volume of DMSO was added to the cells at the appropriate concentrations. After washout of the radioligand conjugate, cells were re-drugged with the anchor compound and left to grow under standard conditions until the endpoint. A minimum of two independent experiments (biological replicates) were performed for each assay in each cell lines, details on the number of replicates are specified in the individual figure legends. The described treatments were used for viability assays, Incucyte analysis, immunofluorescence and cell cycle analysis. The compounds used are listed in Sup.Table 3. X-ray treatment for in vitro assays Cells were treated with the defined X-ray dose 24 h after seeding. The irradiation was performed with a CellRad X-ray cabinet (Faxitron), with a dose rate of 1.02 Gy/min. After irradiation, cells were kept in standard growth conditions until the experimental endpoint. In vivo experiments Xenograft studies were performed at Novartis, approved by the Cantonal Veterinary Office of Basel Stadt, Switzerland, and in strict adherence to the Swiss Federal Animal Welfare Act and Ordinance. All animals had access to food and water ad libitum and were identified with transponders. They were housed in a specific-pathogen-free facility in IVC racks under a 12 h–12 h light–dark cycle. To establish the cell-line-derived xenograft model, 6–7 week old female athymic nude (Crl: NU(NCr)-Foxn1nu) mice from Charles River were engrafted subcutaneously with 3 million tumour cells in HBSS/Matrigel 1:1. PC-3-derived xenograft model is androgen-independent and is widely used in female immunodeficient mice for subcutaneous tumor growth and treatment-response studies. Additionally, female mice were chosen to reduce housing-related aggression and wound complications, in order to improve animal welfare and consistency of tumour measurement. Compound formulation and animal treatment AZD7648 was prepared for dosing as homogenous suspensions in 0.5% MC + 0.1% Tween 80. Fresh suspensions were prepared once every 7 days and stored at RT. For the biodistribution AZD7648 was administers orally 10 min before the i.v. injection of 4 MBq [177Lu]Lu-PSMA-617 (0.1 nmol per single injection). For the efficacy study AZD7648 was administered orally at a volume of 10 mL/kg, qd for 5 days. [177Lu]Lu-PSMA-617 was prepared in PBS and was administered i.v. at a volume of 100µL with a total activity of 10 MBq (0.1 nmol per single injection). On the first study day AZD7648 was administers orally 10 min before the i.v. injection of 4 MBq [177Lu]Lu-PSMA-617. Vehicle control was administered once with 100 µL PBS i.v. Biodistribution experiments The biodistribution study was performed when tumours were approximately 200 mm3 in volume. Tumour volumes were measured with callipers and determined according to the formula: length x diameter2 x π ⁄ 6. Values are mean ± SEM. Mice were randomized into groups of three, and freshly prepared test articles were injected through the lateral tail vein or gavage in a volume of 10 mL/kg. 4 h and 24 h post radioactive-agent dose administration, mice were euthanized, and organs (tumour, kidneys, liver and spleen) were collected, weighed and placed inside counting vials. Each tissue sample was counted for the activities of radioelement using a gamma-counter. Samples of the injectate were used as decay correction standards. Final bar graphs are expressed as % injected dose per gram of tissue (%ID/g). Evaluation of antitumour activity Efficacy treatments were initiated when the mean tumour volumes were approx. 210 mm3 (26 days post tumour cells injection) in mice. Values are mean ± SEM; sample size, (n = 8 mice per group). Tumour volumes were measured with callipers and determined according to the formula: length x diameter2 x π ⁄ 6. Carry forward was applied to the mean graph when animals were taken down and mean tumour graph was stopped when less than four animals remained in each group. Body weights and tumour volumes were recorded two to three times a week. FFPE tumour samples staining Tumour samples were collected from animals at 6 h, 24 h and 120 h after the start of the treatment. Immediately after dissection, tumours were fixed in 10% (vol/vol) neutral buffered formalin for 24 h at room temperature, rinsed in PBS, processed for dehydration, cleared, and embedded in paraffin. 3-µm sections were prepared and processed for immunohistochemistry or immunofluorescent staining. For immunohistochemistry a Discovery Ultra fully automated system (Ventana Medical Systems) was used: sections were deparaffinized, treated with Cell Conditioning #1 antigen retrieval reagent, then incubated for 60 min at room temperature with the PSMA (EP192) Rabbit Monoclonal primary antibody, followed by incubation with ready-to-use OmniMap anti rabbit antibody, ChromoMap DAB Kit, counterstained with Hematoxilin / Bluing reagent, dehydrated and cover slipped. Images were acquired using the Aperio AT2 slide scanner (Leica) and assessed by a certified pathologist. For immunofluorescent staining the slides were incubated at 63 °C for 1 h, de-waxed with 3 successive short incubation with xylene (Scientific Laboratory Supplies, cat. 534056) at room temperature and re-hydrated with decreasing concentrations of ethanol (100%, 95%, 70%, 30%). The samples were then washed with Dako wash buffer (Agilent, cat. S300685-2), heated with a microwave oven in Dako target retrieval solution pH 9 (Agilent, cat. S236784-2) for 9 min to expose the antigens, and then left to cool down at room temperature for at least 30 min. After blocking with 0.5% BSA/0.5% Triton X-100/PBS for 1 h at room temperature, samples were incubated with the primary antibodies diluted in blocking buffer at 4 °C overnight (see Sup.Table 2). The following day, the incubation with secondary antibodies was performed for 1 h at room temperature (see Sup.Table 2), followed by a 10 min incubation with DAPI (4’, 6-diamino-2’-phenylindole, dihydrochloride, Life Technologies, cat. 62248) diluted 1:5000 in PBS. Samples were then dehydrated with increasing concentrations of ethanol (70%, 90%, 100%), and slides were mounted with coverslips using an antifade mountant and left to cure at room temperature overnight protected from light. Images were acquired using the Operetta CLS High-Content Analysis System equipped with a slide holder and a 40x water objective. Image acquisition and analysis were performed as described above. Statistical analysis The statistical analysis for immunofluorescence and Incucyte monotherapy assays was performed using one-way ANOVA followed by Dunnett’s multiple comparison test against control samples (vehicle-treated or non-irradiated). For in vitro combination experiments (immunofluorescence, cell cycle analysis at 2.5 MBq/mL and Incucyte analysis) two-way ANOVA and Šídák multiple comparison test was used to detect differences across the time points assessed between samples treated with 177Lu alone or with the combination of 177Lu and DNA-PKi. For cell cycle analysis at different doses (Fig. 4n-o, Sup. Figure 4g), the analysis was performed using two-way ANOVA and Šídák multiple comparison test to assess differences in the distribution of each cell cycle phase at the different doses compared with vehicle-treated samples. For in vivo experiments, unpaired t test with Two-stage step-up (Benjamini, Krieger, and Yekutieli) was used to determine significant differences in tumour growth between experimental groups and controls, while the statistical significance of survival curves was verified using Log-rank (Mantel-Cox) test. In the immunofluorescence from ex-vivo samples, statistical significance was calculated at each time point using two-way ANOVA and Dunnett’s multiple comparison test between each treatment and vehicle-control samples. All statistical tests used are indicated in each individual figure legend. Analysis was performed using GraphPad Prism version 10. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. Results Profiling DDR marker induction by [177Lu]Lu-DOTA To elucidate the cellular responses triggered by 177Lu-based RLT, we treated three cancer cells lines with [177Lu]Lu-DOTA and analysed the induction of DDR makers at different time points by immunofluorescence (Fig. 1a). These markers included γH2AX for DSBs, 53BP1 for NHEJ, RAD51 for homologous recombination (HR), XRCC1 for SSB repair and phospho-ATM (ataxia telangiectasia mutated kinase) for DSB signalling and checkpoint activation. We used [177Lu]Lu-DOTA, a chelated, non-capturable agent present during the entire duration of the experiment, to avoid potential bias due to differential uptake across cell lines. We argue that these conditions represent a clinically relevant scenario, as 177Lu-activity can persist in mouse tumour xenografts for multiple days [21]. For comparison, we also exposed the same cell lines to X-rays, a well characterized form of IR used for external beam radiation therapy (EBRT), at a dose showing a comparable effect on cell viability (Sup. Figure 1a-b). Both EBRT and [177Lu]Lu-DOTA induced γH2AX, 53BP1, RAD51, pATM foci, XRCC1 and micronuclei accumulation, albeit with different kinetics (Fig. 1b-g and Sup. Figure 1c-g). While EBRT caused a fast induction that rapidly decreased over time, [177Lu]Lu-DOTA mediated a slower accumulation (Fig. 1b-g and Sup. Figure 1d and 1f), reflecting the respective acute or prolonged nature of these treatments. Taken together, these results demonstrate that 177Lu elicits DDR markers typically associated with SSBs (XRCC1) and DSBs (γH2AX foci), however, it is unclear whether these forms of DNA damage are induced as a direct or indirect effect of [177Lu]Lu-DOTA radiation. 177Lu-associated damage triggers biomarkers of cell-cycle checkpoint activation (pATM) and DSB repair by both NHEJ (53BP1 foci) and homologous recombination (HR, RAD51 foci) but can also persist and cause segregation defects in the form of micronuclei. Fig. 1[177Lu]Lu-DOTA and X-ray EBRT induce markers of DNA damage with distinct kinetics. (a) Experimental outline. (b) Representative immunofluorescence images of DLD-1 cells treated with X-ray and [177Lu]Lu-DOTA. Scale bar: 50 μm. c-g) Quantification of the DDR marker induction in DLD-1, represented as the fold change in the percentage of cells with at least 10 (c, d, e) or 5 (f) DDR foci or with at least 2 micronuclei (g) relative to the untreated cells, at the different timepoints. Histograms represent cells treated with [177Lu]Lu-DOTA (blue bars) or X-ray (red bars) at indicated dose. Comparable results were obtained in LNCaP and DU145 PSMA high (Sup. Figure 1). Statistical analysis: one-way ANOVA and Dunnett’s multiple comparison test against vehicle-treated or non-irradiated samples for [177Lu]Lu-DOTA and X-ray treatments, respectively. Each bar represents the average of 2 independent experiments ± SEM, with each value being the average of 3 technical replicates [177Lu]Lu-DOTA and X-ray EBRT induce markers of DNA damage with distinct kinetics. (a) Experimental outline. (b) Representative immunofluorescence images of DLD-1 cells treated with X-ray and [177Lu]Lu-DOTA. Scale bar: 50 μm. c-g) Quantification of the DDR marker induction in DLD-1, represented as the fold change in the percentage of cells with at least 10 (c, d, e) or 5 (f) DDR foci or with at least 2 micronuclei (g) relative to the untreated cells, at the different timepoints. Histograms represent cells treated with [177Lu]Lu-DOTA (blue bars) or X-ray (red bars) at indicated dose. Comparable results were obtained in LNCaP and DU145 PSMA high (Sup. Figure 1). Statistical analysis: one-way ANOVA and Dunnett’s multiple comparison test against vehicle-treated or non-irradiated samples for [177Lu]Lu-DOTA and X-ray treatments, respectively. Each bar represents the average of 2 independent experiments ± SEM, with each value being the average of 3 technical replicates Sensitisation to [177Lu]Lu-DOTA by loss or inhibition of DDR pathways To further characterize the cellular responses to 177Lu-based RLT, we investigated which DDR pathways predominantly repair 177Lu-associated DNA damage and promote cell viability. To this aim, we performed viability assays with [177Lu]Lu-DOTA and a collection of seventeen isogenic models deficient for key factors encompassing multiple DDR pathways, namely, NHEJ, HR, Fanconi anaemia (FA), G2/M checkpoint, base excision repair (BER), nucleotide excision repair (NER), and mismatch repair (MMR) (Fig. 2a-d and Sup. Figure 2a-n). We observed strong sensitisation in models deficient in NHEJ, HR, the DDR kinase ATM and FA (Fig. 2d). Remarkably, the models deficient in the NHEJ core factors PRKDC, which codes for the catalytic subunit of DNA-PK, and NHEJ1 were the most sensitive (Fig. 2a, d and Sup. Figure 2a). We also evaluated whether clinically relevant DDR inhibitors can sensitize three cancer cell lines to [177Lu]Lu-DOTA. In line with our previous results, DNA-PK inhibition consistently sensitized all cell lines to [177Lu]Lu-DOTA, with two of the tested cell lines showing strong sensitization (Fig. 2e-g). ATM inhibition showed more modest combination potential and only displayed strong sensitization in DLD-1 cells (Fig. 2e-g). Finally, no sensitisation was observed using 4 different PARP inhibitors including olaparib, veliparib, talazoparib and AZD5305 (Fig. 2e-g and Sup. Figure 2o-r). Altogether, our data suggest that DSBs, especially those repaired by NHEJ, are a highly cytotoxic form of DNA damage caused by 177Lu. The combination with inhibitors of NHEJ (DNA-PK) and DSB repair signalling (ATM) strongly sensitises cells to [177Lu]Lu-DOTA, albeit to a lesser extent with the latter. Fig. 2Loss or inhibition of key DDR factors sensitizes cells to [177Lu]Lu-DOTA, with the greatest sensitisation observed with DNA-PK impairment. a-c) Survival curve from viability assays with isogenic models carrying deficiencies in representative DDR proteins. Each point shows mean viability ± SEM at the indicated doses. Representative curves from N = 3 (a, b) and N = 6 (c) independent experiments. d) Representation of the potentiation factor (PF, Y axis) observed in isogenic models for the indicated DDR factors after [177Lu]Lu-DOTA continuous treatment. Each bar represents mean ± SEM of PF calculated from at least N = 2 biological replicates. e-g) Survival curves from viability assay with DLD-1 (e), PC-3 (f) and 22Rv1 (g) cell lines treated with different, indicated DDR inhibitors either alone (blue curves) or in combination with [177Lu]Lu-DOTA (red curves). Representative data from at least N = 2 biological replicates. Combinations with additional DDR inhibitors are shown in Sup. Figure 2 Loss or inhibition of key DDR factors sensitizes cells to [177Lu]Lu-DOTA, with the greatest sensitisation observed with DNA-PK impairment. a-c) Survival curve from viability assays with isogenic models carrying deficiencies in representative DDR proteins. Each point shows mean viability ± SEM at the indicated doses. Representative curves from N = 3 (a, b) and N = 6 (c) independent experiments. d) Representation of the potentiation factor (PF, Y axis) observed in isogenic models for the indicated DDR factors after [177Lu]Lu-DOTA continuous treatment. Each bar represents mean ± SEM of PF calculated from at least N = 2 biological replicates. e-g) Survival curves from viability assay with DLD-1 (e), PC-3 (f) and 22Rv1 (g) cell lines treated with different, indicated DDR inhibitors either alone (blue curves) or in combination with [177Lu]Lu-DOTA (red curves). Representative data from at least N = 2 biological replicates. Combinations with additional DDR inhibitors are shown in Sup. Figure 2 Phenotypic responses and combination potential of [177Lu]Lu-PSMA-617 in prostate cancer cell lines We next sought to validate our observations using [177Lu]Lu-PSMA-617, given its clinical relevance for the treatment of PSMA-positive prostate cancer. We analysed 3 PSMA-positive prostate cancer cell lines: one naturally PSMA-expressing model, LNCaP clone FGC (hereafter referred to as LNCaP), and two engineered models, DU145 PSMA High and PC-3 MP9, which were transduced to express PSMA. In clear contrast to the PSMA-negative parental cell lines, we confirmed PSMA expression in the transduced models DU145 PSMA High and PC-3 MP9 by immunofluorescence and flow cytometry, which was comparable to the naturally expressing model LNCaP (Fig. 3a and Sup. Figure 3a). Similarly, we also confirmed a comparable [177Lu]Lu-PSMA-617 uptake and similar reduction in viability among these PSMA positive cell lines (Fig. 3b and Sup. Figure 3b). In agreement with our previous results using [177Lu]Lu-DOTA, treatment with [177Lu]Lu-PSMA-617 caused a dose-dependent accumulation of γH2AX, 53BP1, pATM, RAD51 and micronuclei (Fig. 3c-g). However, in these experiments, the levels of the DDR markers decreased at 96 h compared to 24 h, with the anticipated exception of micronuclei (Fig. 3c-g). This suggests repair of the 177Lu-induced damage in this experimental setting where [177Lu]Lu-PSMA-617 was washed out 24 h after treatment (Sup. Figure 3c). By time-course live imaging analysis, we demonstrated that [177Lu]Lu-PSMA-617 caused a significant reduction of cell confluency and prompted cell death and apoptosis (Fig. 3h-j and Sup. Figure 3d). This indicates that [177Lu]Lu-PSMA-617 not only hinders cell growth but also increases cytotoxicity. Finally, in agreement with our previous results, 2 different DNA-PK and ATM inhibitors sensitised the prostate cancer cell lines to [177Lu]Lu-PSMA-617 (Fig. 3k-m and Sup. Figure 3e-g). Compared to our previous experiments with [177Lu]Lu-DOTA, ATM inhibition showed stronger sensitisation in this setup, however, the DNA-PK inhibitor (DNA-PKi) demonstrated more consistent sensitisation to [177Lu]Lu-PSMA-617 (Fig. 3k-m and Sup. Figure 3e-f). Fig. 3[177Lu]Lu-PSMA-617 elicits dose-dependent induction of DDR markers, apoptosis and cell death, and is effectively potentiated by DNA-PK inhibition in prostate cancer models. (a) PSMA expression by immunofluorescence (Y axis). Representative data from one out of 2 biological replicates. (b) [177Lu]Lu-PSMA-617 uptake and internalization in prostate cancer cell lines. Data are shown as the percentage of detected activity relative to the total activity added, normalized to the fold difference of protein concentration relative to the corresponding vehicle controls and corrected by subtraction of activity detected in the vehicle controls. (N = 3, independent experiments, mean ± SD, each dot represents the mean of two technical replicates). c-g) Dose-dependent induction of DDR markers in DU145 PSMA High at 24 h (blue bars) and 96 h (red bars) detected by immunofluorescence assay. Each bar represents mean ± SEM of 2 independent experiments, each with 3 technical replicates, normalized for vehicle-treated samples at matched time points. Statistical analysis: one-way ANOVA followed by Dunnetts’s multiple comparison test against vehicle-treated cells at each time point. h-j) Incucyte assay showing the reduction of cell confluency (h), and the induction of Caspase-3/7 (apoptosis, i) and cytotoxicity (cell death, j) in DU145 PSMA High cell line treated with [177Lu]Lu-PSMA-617. The results are representative of N = 2 biological replicates. Each point represents mean ± SEM from 3 technical replicates. Statistical analysis: one-way ANOVA and Dunnett’s multiple comparison test against vehicle-treated cells at the endpoint. k-m) Survival curve from viability assays for the combination of [177Lu]Lu-PSMA-617 with the DNA-PK inhibitor AZD7648 in 3 PSMA+ prostate cancer cell lines. Representative data from at least N = 3 independent experiments [177Lu]Lu-PSMA-617 elicits dose-dependent induction of DDR markers, apoptosis and cell death, and is effectively potentiated by DNA-PK inhibition in prostate cancer models. (a) PSMA expression by immunofluorescence (Y axis). Representative data from one out of 2 biological replicates. (b) [177Lu]Lu-PSMA-617 uptake and internalization in prostate cancer cell lines. Data are shown as the percentage of detected activity relative to the total activity added, normalized to the fold difference of protein concentration relative to the corresponding vehicle controls and corrected by subtraction of activity detected in the vehicle controls. (N = 3, independent experiments, mean ± SD, each dot represents the mean of two technical replicates). c-g) Dose-dependent induction of DDR markers in DU145 PSMA High at 24 h (blue bars) and 96 h (red bars) detected by immunofluorescence assay. Each bar represents mean ± SEM of 2 independent experiments, each with 3 technical replicates, normalized for vehicle-treated samples at matched time points. Statistical analysis: one-way ANOVA followed by Dunnetts’s multiple comparison test against vehicle-treated cells at each time point. h-j) Incucyte assay showing the reduction of cell confluency (h), and the induction of Caspase-3/7 (apoptosis, i) and cytotoxicity (cell death, j) in DU145 PSMA High cell line treated with [177Lu]Lu-PSMA-617. The results are representative of N = 2 biological replicates. Each point represents mean ± SEM from 3 technical replicates. Statistical analysis: one-way ANOVA and Dunnett’s multiple comparison test against vehicle-treated cells at the endpoint. k-m) Survival curve from viability assays for the combination of [177Lu]Lu-PSMA-617 with the DNA-PK inhibitor AZD7648 in 3 PSMA+ prostate cancer cell lines. Representative data from at least N = 3 independent experiments Characterisation of DNA-PK inhibitor and [177Lu]Lu-PSMA-617 combination in prostate cancer models The observation that genetic and pharmacological suppression of DNA-PK activity strongly sensitises cells to 177Lu-based RLT prompted us to evaluate how this combination modulates the previously characterised cellular responses to [177Lu]Lu-PSMA-617. By performing immunofluorescence experiments, we observed that DNA-PK inhibition strongly exacerbated γH2AX, 53BP1, pATM, RAD51 foci and micronuclei accumulation by 177Lu (Fig. 4a-j and Sup. Figure 4a-b), while not inducing any of the tested DDR markers as a single agent (Sup. Figure 4c). Moreover, DNA-PKi combination with [177Lu]Lu-PSMA-617 prevented repair of 177Lu-associated damage when we compared DDR marker levels at 24 h and 96 h timepoints (Fig. 4a-j and Sup. Figure 4a-b). Combination with DNA-PK inhibition also significantly increased [177Lu]Lu-PSMA-617-mediated apoptosis and cell death in prostate cancer cell lines (Fig. 4k-m and Sup. Figure 4d). Similarly, we observed that [177Lu]Lu-PSMA-617 doses that showed a modest effect on DU145 cells as monotherapy led to a complete lack of detectable growth when combined with a DNA-PK inhibitor, even over long periods of time (3 weeks) (Sup. Figure 4e-f). Collectively, these data suggest that the impairment of NHEJ is highly cytotoxic for [177Lu]Lu-PSMA-617-treated cells. Finally, we analysed the effect of [177Lu]Lu-PSMA-617 in cell-cycle progression in DU145 PSMA High and PC-3 MP9 cells. We observed that [177Lu]Lu-PSMA-617 treatment caused G2/M cell-cycle arrest, which was again significantly exacerbated by addition of DNA-PKi (Fig. 4n-o and Sup. Figure 4 g-h), suggesting that [177Lu]Lu-PSMA-617 can lead to defective mitotic progression, possibly compromising viability. Fig. 4DNA-PK inhibition exacerbates [177Lu]Lu-PSMA-617-induced DDR marker accumulation, cell cycle perturbation, apoptosis, and cell death. a-j) Induction of DDR markers in DU145 PSMA High treated with [177Lu]Lu-PSMA-617 alone (blue bars) or in combination with DNA-PKi (red bars) measured by immunofluorescence assay at 24 h (a-e) or 96 h (f-j). Each bar shows mean ± SEM from 3 independent experiments, each performed in three technical replicates. k-m) Incucyte assays showing the decrease of confluency (k), induction of apoptosis (l) and cell death (m) in DU145 PSMA High cells treated with 1.25 MBq/mL of [177Lu]Lu-PSMA-617 alone (blue bars) or in combination with DNA-PKi (red bars) over the time (X-axis). Representative data from N = 2 independent experiments. n-o) Analysis of cell cycle by immunofluorescence assay in DU145 PSMA High cells treated with [177Lu]Lu-PSMA-617 alone (left bars) or in combination with DNA-PKi (right panels) at 24 h (n) or 96 h (o). Histograms show mean ± SEM from 3 independent repeats, with each experiment consisting of 6 technical replicates. Statistical analysis: two-way ANOVA followed by multiple comparison test with Šidák correction DNA-PK inhibition exacerbates [177Lu]Lu-PSMA-617-induced DDR marker accumulation, cell cycle perturbation, apoptosis, and cell death. a-j) Induction of DDR markers in DU145 PSMA High treated with [177Lu]Lu-PSMA-617 alone (blue bars) or in combination with DNA-PKi (red bars) measured by immunofluorescence assay at 24 h (a-e) or 96 h (f-j). Each bar shows mean ± SEM from 3 independent experiments, each performed in three technical replicates. k-m) Incucyte assays showing the decrease of confluency (k), induction of apoptosis (l) and cell death (m) in DU145 PSMA High cells treated with 1.25 MBq/mL of [177Lu]Lu-PSMA-617 alone (blue bars) or in combination with DNA-PKi (red bars) over the time (X-axis). Representative data from N = 2 independent experiments. n-o) Analysis of cell cycle by immunofluorescence assay in DU145 PSMA High cells treated with [177Lu]Lu-PSMA-617 alone (left bars) or in combination with DNA-PKi (right panels) at 24 h (n) or 96 h (o). Histograms show mean ± SEM from 3 independent repeats, with each experiment consisting of 6 technical replicates. Statistical analysis: two-way ANOVA followed by multiple comparison test with Šidák correction DNA-PK inhibitor combination with [177Lu]Lu-PSMA-617 in vivo We next sought to evaluate DNA-PKi combination with [177Lu]Lu-PSMA-617 in vivo in representative PSMA-positive mouse tumour xenograft models of PC-3 MP9 (Sup. Figure 5a). We confirmed efficient tumour accumulation of [177Lu]Lu-PSMA-RLT and rapid clearance from peripheral organs. Notably, the tumour uptake of [177Lu]Lu-PSMA-617 RLT was approximately 9% ID/g at 4 h post-injection and decreased to approximately 6% ID/g at 24 h (Sup.Table 4), which is consistent with prior reports in similar models [18, 20]. Importantly, the co-administration of a DNA-PKi did not alter the biodistribution or tumour retention kinetics of the radioligand therapy (Sup. Figure 5b, Sup.Table 4). Similarly, analysis of DNA-PK inhibition pharmacokinetics demonstrated a distribution profile in line with previous findings (Sup. Figure 5c-d [18]). The therapeutic protocol for the combination treatment involved a single injection of [177Lu]Lu-PSMA-617 administered 10 min after the initial dose of the DNA-PKi. The DNA-PKi was subsequently dosed once daily (qd) for five consecutive days. In vivo assessments did not reveal any significant differences between the tumour growth rates in groups treated with DNA-PKi or [177Lu]Lu-PSMA-617 RLT as monotherapies. However, the combination of DNA-PKi with [177Lu]Lu-PSMA-617 RLT led to a significantly greater reduction in tumour growth compared to either single agent treatment (Fig. 5a). Furthermore, this combination therapy produced a significant improvement in animal survival outcomes relative to monotherapies (Fig. 5b). Specifically, median survival was 26.5 days for the vehicle group and increased to 30 days and 33 days for the DNA-PKi and [177Lu]Lu-PSMA-617 RLT single agent groups, respectively. Interestingly, the combination therapy extended median survival to more than 37 days, demonstrating enhanced efficacy. Importantly, the absence of significant weight loss in treated animals suggested good tolerability of this combination treatment in vivo (Fig. 5c). Finally, immunofluorescence analysis of mouse tumour xenograft samples showed that the combination of DNA-PKi with [177Lu]Lu-PSMA-617 caused an increased accumulation of γH2AX foci at the 120 h timepoint compared to [177Lu]Lu-PSMA-RLT alone (Fig. 5d-e), confirming the persistence of DNA damage as a result of the combination. Taken together, our results demonstrate that the combination of DNA-PKi not only strongly potentiates [177Lu]Lu-PSMA-617 better than other tested combinations in vitro, but is efficacious and well-tolerated in vivo. These findings highlight the potential of the DNA-PKi and [177Lu]Lu-PSMA-617 RLT combination as a promising therapeutic strategy for PSMA-positive tumours. Fig. 5[177Lu]Lu-PSMA-617 combination with AZD7648 in PC-3 MP9 PSMA tumor bearing female nude mice. (a) Tumor growth over the time (n = 8). The statistical significance of differential findings between experimental groups and controls was determined by unpaired t test with Two-stage step-up (Benjamini, Krieger, and Yekutieli). (b) Mouse survival curve. Statistical significance was determined by Log-rank (Mantel-Cox) test. (c) Mouse body weight over the time indicates that the combination is well tolerated. (d) Representative images of immunofluorescent staining for γ-H2AX in slides from tumors harvested at 120 h after administration of the treatment. Scale bar: 50 μm. (e) Quantification of γ-H2AX foci induction in tumor samples harvested at 6 h, 24 h and 120 h after animal treatment. Data is represented as fold change in the percentage of cells with at least 5 foci relative to vehicle-treated samples. The histogram represents average and standard error of quantification from 3 different tumors for each treatment group. Statistical analysis: two-way ANOVA followed by Dunnett’s multiple comparison test [177Lu]Lu-PSMA-617 combination with AZD7648 in PC-3 MP9 PSMA tumor bearing female nude mice. (a) Tumor growth over the time (n = 8). The statistical significance of differential findings between experimental groups and controls was determined by unpaired t test with Two-stage step-up (Benjamini, Krieger, and Yekutieli). (b) Mouse survival curve. Statistical significance was determined by Log-rank (Mantel-Cox) test. (c) Mouse body weight over the time indicates that the combination is well tolerated. (d) Representative images of immunofluorescent staining for γ-H2AX in slides from tumors harvested at 120 h after administration of the treatment. Scale bar: 50 μm. (e) Quantification of γ-H2AX foci induction in tumor samples harvested at 6 h, 24 h and 120 h after animal treatment. Data is represented as fold change in the percentage of cells with at least 5 foci relative to vehicle-treated samples. The histogram represents average and standard error of quantification from 3 different tumors for each treatment group. Statistical analysis: two-way ANOVA followed by Dunnett’s multiple comparison test Discussion In this work we have investigated the cellular phenotypes induced by 177Lu treatment and the pathways that repair 177Lu-associated damage. β− emitters, such as 177Lu, are considered to elicit several forms of DNA damage including SSBs and DSBs, the latter at lower frequency (reviewed in [22]). This DNA damage may originate either from direct interaction of the β− particles with the DNA or indirectly through generation of reactive oxygen species (ROS) that damage the DNA [22]. Nonetheless, cellular analyses of the DNA damage caused by 177Lu are, to our knowledge, limited, particularly in prostate cancer models. Arguably, our understanding of 177Lu mechanism of action is partially derived from that of photon-based ionizing radiation (reviewed in [23]) and evidence demonstrating that linear energy transfer (LET) of IR correlates with the complexity of the associated DNA damage [24–26]. Our analysis of DDR marker induction shows that [177Lu]Lu-DOTA, as anticipated, induces markers indicative of both SSBs and DSBs, similarly to X-ray irradiation, albeit with different kinetics (Fig. 1 and Sup. Figure 1d-f). However, this comparison is solely phenomenological, as the treatments are not matched for the absorbed dose, which is a limitation of our study. The different kinetics observed likely reflect the treatment regimens, chosen to mimic the acute nature of EBRT (X-rays) and the more prolonged persistence of 177Lu-based RLT in tumours (as evidenced in preclinical studies such as [21]). Nevertheless, our data indicate that [177Lu]Lu-DOTA and X-ray irradiation elicit similar forms of DNA damage. Induction of SSBs by β emitters raised the hypothesis that PARP inhibitors of SSB repair could sensitise cancer cells to 177Lu-based RLT. This hypothesis was particularly attractive due to the availability of approved PARP inhibitors such as olaparib and talazoparib, which trap PARP1/2 at SSBs and lead to single-ended DSB generation by replication fork collapse [9]. These replication-associated DSBs are repaired by BRCA1/2-mediated HR, rendering PARP inhibitors synthetically lethal to BRCA1/2-deficient tumours [9]. PARP inhibitors can sensitise tumour cells to 177Lu-based RLT [10–12], which prompted clinical evaluation. However, this combination is not universally efficacious [14]. In fact, the prostate cancer model PC-3-PIP is not sensitised to [177Lu]Lu-PSMA-I&T by treatment with 3 different PARP inhibitors, namely, olaparib, veliparib and talazoparib [15]. Similarly, our experiments show lack of in vitro sensitisation to [177Lu]Lu-DOTA by combination with 4 different PARP inhibitors (olaparib, veliparib, talazoparib and AZD5305) in colorectal and prostate cancer cells (Fig. 2e-g and Sup. Figure 2o-q). This suggests that inhibition of SSB repair and PARP-trapping, leading to DSB generation through replication fork collapse, is not consistently detrimental to 177Lu-treated cancer cells. Intriguingly, the determinants of why only certain cancer models are sensitive to PARP inhibitor and 177Lu-based RLT combination remains to be characterised, being critical to identify patient populations that could benefit from the combination. Among the pathways we interrogated using DDR-deficient isogenic models and DDR inhibitors, our data indicate that NHEJ deficiency elicits the strongest sensitisation to [177Lu]Lu-DOTA (Fig. 2 and Sup. Figure 2). Compared to the other major pathways devoted to DSB repair in mammalian cells, NHEJ, also dubbed classical NHEJ (cNHEJ), stands out for repairing the DSB by direct synapsis requiring minimal end-processing (reviewed in [16, 27]). In agreement with our data, NHEJ deficiency causes the strongest sensitisation to photon-based ionizing radiation in whole genome CRISPR-Cas9 screens [28], reflecting the cytotoxicity of DSBs. Similarly, genome-wide CRISPR-Cas9 screens with [177Lu]Lu-DOTA-octreotate identified DNA-PKcs, encoded by the PRKDC gene, as the strongest hit [17]. NHEJ is a flexible pathway that only strictly requires a small number of factors, known as “core”. These include DNA-PKcs, a member of the PIKK family (phosphoinositide 3-kinase (PI3K)-related kinase) that binds the Ku70/80 heterodimer to form the DNA-PK holoenzyme [27]. Our data demonstrate that DNA-PK inhibition enhances the sensitivity to [177Lu]Lu-DOTA treatment in all the tested cell lines (Fig. 2e-g), in contrast to PARP inhibitors (Fig. 2e-g and Sup. Figure 2o-q). Moreover, DNA-PK inhibitors strongly sensitise PSMA-expressing prostate cancer cells to [177Lu]Lu-PSMA-617, both in vitro and in vivo (Fig. 3k-m, Sup. Figure 3e and Fig. 5a). Collectively, these data suggest that disruption of NHEJ-mediated repair of DSBs strongly potentiates the treatment of cancer cells with 177Lu-based RLT. Accordingly, inhibition of ATM, another kinase belonging to the PIKK family that is involved in sensing and signalling for DSB repair to promote both NHEJ and HR (reviewed in [29]) also resulted in some sensitisation to [177Lu]Lu-DOTA or [177Lu]Lu-PSMA-617 in vitro, but less consistently than DNA-PKi (Fig. 2e-g and Sup. Figure 3f), arguably reflecting its less prominent role in NHEJ. Despite the consistent sensitization to 177Lu-based RLT observed through genetic and pharmacological disruption of NHEJ across the different models used in this study, further analysis of broader cell line panels could better address how the response to 177Lu-based RLT may be modulated by cell line-intrinsic biological differences, such as proliferation rates and receptor expression, this being another limitation of our study. We demonstrate that exposure to [177Lu]Lu-PSMA-617 is cytotoxic for prostate cancer cells, leading to increased apoptosis and cell death (Fig. 3h-j and Sup. Figure 3d). Combination with DNA-PKi prevents repair of [177Lu]Lu-PSMA-617-associated damage (Fig. 4a-j and Sup. Figure 4a-b) and robustly exacerbates cytotoxicity (Fig. 4k-m and Sup. Figure 4d). Interestingly, [177Lu]Lu-PSMA-617 treatment caused G2/M accumulation that was also aggravated by combination with DNA-PKi (Fig. 4n-o and Sup. Figure 4 g-h). This suggests that impairment of NHEJ by inhibition of DNA-PK results in the accumulation of DSBs that hinder progression through mitosis and lead to cell death. Further studies are necessary to delineate the exact mechanisms of 177Lu-induced cell death and how they may be modulated by combination with DNA-PK inhibitors. Additionally, the in vivo efficacy and tolerability of this combination could be further investigated to address the effects of immune response and normal tissue radiosensitisation, which cannot be determined in nude mice xenografts. Nevertheless, by characterising the repair of 177Lu-RLT-induced DNA damage, we conclude that the combination with DNA-PKi consistently sensitises cancer cells to 177Lu-based RLT in vitro and we provide evidence of the potential of this combinatorial treatment in vivo. Our findings support the clinical exploration DNA-PKi and 177Lu-based RLT combination and provide a mechanistic framework for further characterization. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 1