Radionuclide‑drug conjugates in lung cancer: advances in precision therapy and clinical translation Authors: Junzhi Liu, Xiaoyu Zuo, Xue Yang, Xinyue Wang, Lu Gan, Jianxin Xue Published: 2026 ABSTRACT AbstractPurposeLung cancer is the leading cause of cancer‑related mortality worldwide. Current therapies continue to face challenges such as drug resistance and tumour heterogeneity. Radionuclide‑drug conjugates (RDCs) represent an emerging theranostic platform designed to precisely irradiate tumours. This review aims to systematically outline the landscape of advances in RDCs for lung cancer, and provides a forward‑looking perspective on next‑generation RDCs.MethodsThis analysis was based on preclinical and clinical data retrieved from PubMed and ClinicalTrials.gov, with all records reviewed up to January 2026. Studies were exhaustively surveyed across most radionuclides relevant to RDCs in lung cancer, following the periodic table to ensure comprehensive coverage.ResultsIn lung cancer, a total of 66 RDCs have been screened, with 30 having entered early‑phase clinical trials. Among completed trials, RDCs underwent a transition from initial 131I/90Y‑labelled antibodies toward 177Lu‑labelled somatostatin receptor (SSTR)‑targeting peptides. SSTR remains the dominant target, with a notable shift from agonists to antagonists. Meanwhile, fibroblast activation protein (FAP), epidermal growth factor receptor (EGFR), and programmed death‑ligand 1 (PD‑L1) are receiving growing attention. Peptides and antibodies are equally employed, with bispecific antibodies (bsAbs), single‑domain antibodies (sdAbs), and cyclic peptides advancing rapidly. Furthermore, nanoparticles (NPs) offer versatile platforms, and pretargeting or dual‑targeting strategies are being developed to improve both efficacy and safety. Crown chelators and bipyridine derivatives provide more stable chelating options. Although 177Lu remains the mainstay, α emitters and emerging mixed‑decay radionuclides like 161Tb are gaining ground. Combination therapies are also being investigated to enable first‑line application.ConclusionDespite recent progress in RDC development, challenges such as off‑target toxicity, radiation resistance, and radionuclide production remain. Next‑generation RDCs hold promise to overcome these barriers by novel radionuclides, personalised dosimetry, multifunctional delivery platforms, and multidrug combination strategies. Collectively, these innovations will propel RDCs into a personalised, precision theranostic platform for lung cancer.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12943-026-02695-6.Keywords: Lung cancer, Radionuclide‑drug conjugates, Targeted radionuclide therapy, Radiotheranostics, Radiopharmaceuticals, Radioimmunoconjugates. Background Lung cancer remains the primary cause of cancer‑related mortality globally, accounting for 12.4% of newly diagnosed cancer cases and 18.7% of cancer‑related deaths [1]. Non‑small cell lung cancer (NSCLC) constitutes over 85% of lung cancer cases [2]. For patients with early‑stage or resectable locally advanced NSCLC, surgical resection continues to be the preferred treatment modality. Those inoperable may receive alternative curative‑intent therapies, such as definitive radiotherapy, platinum‑based chemotherapy, or concurrent chemoradiotherapy [3, 4]. Treatment for metastatic NSCLC primarily relies on platinum‑based doublet chemotherapy, molecularly targeted agents, immunotherapy, and systemic combination regimens [5]. In contrast, small cell lung cancer (SCLC) is characterised by high aggressiveness, early dissemination, and poor prognosis, presenting a limited role for surgical intervention [6]. The standard treatment for limited‑stage SCLC mainly involves radiotherapy combined with etoposide‑platinum chemotherapy, whereas extensive‑stage SCLC (ES‑SCLC) is commonly treated with etoposide‑platinum chemotherapy plus anti‑PD‑L1 or anti‑PD‑1 monoclonal antibody immunotherapy [7]. Although these therapies have significantly improved survival outcomes for lung cancer patients, partial therapeutic challenges persist. These include radiotherapy resistance, chemotherapy‑induced toxicity, acquired resistance to targeted therapies, and a scarcity of druggable targets [8–12]. Consequently, there is an urgent need to explore novel therapeutic strategies enabling early detection, precise diagnosis, and personalised treatment. Targeted radionuclide therapy (TRT) is a precision therapeutic paradigm guided by molecular imaging. Its core principle lies in conjugating both diagnostic and therapeutic radionuclides to the same targeting carrier, thereby enabling radiotheranostics [13]. In the diagnostic phase, molecular probes radiolabelled with diagnostic positron emitters (e.g., 68Ga, 18F) or γ emitters (e.g., 99mTc) allow PET/SPECT imaging, which facilitates precise assessment of tumour volume and distribution and patient risk stratification [14]. In the therapeutic phase, the same targeted carrier is used to deliver therapeutic radionuclides (e.g., 177Lu, 225Ac, and 211At). Their emitted β or α particles induce direct effects of ionising radiation (e.g., DNA single‑ and double‑strand breaks and mitochondrial oxidative stress) and indirect radiobiological effects (e.g., bystander, crossfire, and abscopal effects), enabling precise tumour targeting and eradication [15] (Fig. 1). Schematic illustration of how RDCs enable radiotheranostics. The targeting carrier (e.g., peptides, small molecules, antibodies) is labelled with either diagnostic (e.g., 68Ga, 99mTc) or therapeutic (e.g., 177Lu, 225Ac) radionuclides, enabling tumour visualisation, patient stratification, and subsequent targeted radiotherapy via precise DNA damage. Created with Microsoft PowerPoint and BioGDP.com [259] The concept of TRT originated in the 1940s with the first use of 131I for differentiated thyroid carcinoma [16]. To date, more than ten radiopharmaceuticals, like 90Y‑ibritumomab tiuxetan, have been approved [17–19]. They remain clinically limited due to poor targeting specificity, biosafety, and manufacturing constraints [17, 20, 21]. The maturation of targeted delivery platforms has since driven the rapid development of radionuclide‑drug conjugates (RDCs) and advanced the clinical practice for TRT [22]. The canonical structure of an RDC consists of four key components: a targeting carrier, a linker, a chelator, and a radionuclide [23]. With the exception of a few radionuclides (e.g., 131I and 223Ra) exhibit intrinsic organ tropism, most require a specific targeting carrier, typically a small molecule, peptide, or antibody, to achieve selective accumulation in tumour tissue [24]. The linker and chelator connect the carrier to the radionuclide, ensuring stability and reducing off‑target toxicity caused by radionuclide dissociation [25]. Two landmark RDC approvals have recently shaped the field. In 2018, the FDA approved 177Lu‑DOTATATE for somatostatin receptor (SSTR)‑positive gastroenteropancreatic neuroendocrine tumours (GEP‑NETs). Subsequently, in 2022, 177Lu‑PSMA‑617, targeting prostate‑specific membrane antigen (PSMA), was approved for PSMA‑positive metastatic castration‑resistant prostate cancer [26]. These approvals not only validate the clinical feasibility of theranostics but also expand the scope of TRT. Building on this foundation, a series of novel RDCs targeting lung cancer is accelerating into the research and development pipeline. The emergence of RDCs provides a practical and innovative strategy for achieving whole‑cycle management of lung cancer, encompassing precise diagnosis, staging, treatment, and response monitoring. The preclinical and clinical landscape of RDCs in lung cancer The first section of this review comprehensively outlines the overview of RDCs in lung cancer. It begins with an exhaustive search of PubMed and ClinicalTrials.gov across all radionuclides based on the periodic table, identifying a total of 66 RDCs (see Additional file 1). The following then discusses these RDCs in turn according to their structural components (Fig. 2). It provides a foundational framework for understanding RDC mechanisms, optimising delivery strategies, and guiding their future clinical translation in lung cancer. Structural Components of RDCs. A Radionuclides used in RDCs for lung cancer therapy. Therapeutic radionuclides comprise β emitters (e.g., 177Lu, 131I, 90Y, 188Re, 47Sc, 89Zr, 153Sm) and α emitters (e.g., 225Ac, 211At, 227Th, 212Pb, 213Bi). B Carriers for targeted delivery. They encompass peptides, small molecules, mAbs, cyclic peptides, NPs, sdAbs, and bsAbs. C Bifunctional chelators and linkers. Chelators are categorised into general‑purpose (e.g., DOTA, DTPA and their derivatives) and specific chelators (e.g., HEPA, DFO, PSC, 3,2‑HOPO). Linkers are primarily based on EB and PEG motifs. Abbreviations: mAb, monoclonal antibodies; NPs, nanoparticles; sdAbs, single‑domain antibodies; bsAbs, bispecific antibodies; EB, Evans blue; PEG, polyethylene glycol. Created with Microsoft PowerPoint Radionuclides Therapeutic radionuclides are classified into three categories based on their decay modes: β emitters, α emitters, and Auger electron (AE) emitters [27] (Fig. 2A). β emitters, such as 177Lu, 90Y, and 131I, release electrons through β⁻ decay, converting neutrons into protons accompanied by electron antineutrino production. Their radiation induces predominantly reparable single‑strand DNA breaks [28]. Owing to their relatively low linear energy transfer (LET), long tissue penetration, and capacity for crossfire effects, β emitters are well‑suited for treating large‑volume tumours. A corresponding limitation is their non‑specific cytotoxicity [29]. α emitters, such as 225Ac, 213Bi, and 212Pb, release helium nuclei through α decay, inducing irreparable double‑strand DNA breaks and mitotic chromosomal abnormalities [30]. The high LET and short range of α particles result in more precise cell killing with minimal exposure to adjacent healthy tissues, making α emitters particularly effective against small tumours and micrometastases [31]. AE emitters, such as 161Tb and 125I, release low‑energy electrons through electron capture. Their extremely short path length necessitates internalisation and close proximity to nuclear DNA to deliver a highly localised, high‑LET dose. It makes them particularly suitable for eradicating minimal residual lesions [32]. Currently, very few studies choose AE emitters in lung cancer. β emitters β emitters are the most extensively utilised class of therapeutic radionuclides. They decay by emitting β particles, exhibiting an LET of approximately 0.2 keV/µm and a tissue penetration range of 1–10 mm [33]. In lung cancer, the most frequently employed β emitters are, in order of use: 177Lu, 131I, 90Y, 188Re, 47Sc, and 153Sm. Among these, 177Lu is the predominant choice, featured in 26 RDCs (see Additional file 1). It decays with a half‑life of 6.65 days, emitting both β particles and γ rays. Its β particles have a maximum energy of 497 keV, a mean path length in tissue of 0.16 mm, and can penetrate approximately 2 mm in tissue [34]. 177Lu has moderate particle energy, a half‑life well‑matched to the pharmacokinetics of peptides and antibodies, and a suitable penetration depth. Therefore, it particularly suitable for the targeted radiotherapy of medium‑volume tumours while sparing surrounding healthy tissue [35]. Other β emitters present distinct profiles based on their physical characteristics, such as particle energy and tissue penetration range. 131I and 47Sc, with their shorter penetration ranges, are suited for treating smaller tumours. However, 131I emits high‑energy and abundant γ rays, potentially posing more severe radiation hazards [36]. 47Sc possesses a shorter half‑life and tissue penetration range compared to 177Lu, making it a candidate for micrometastatic disease, but its production relies on high‑energy proton accelerators and remains costly and inefficient [37, 38]. In contrast, 90Y and 188Re deliver higher‑energy β particles with longer ranges, enabling coverage of larger tumour volumes. 90Y emits no γ radiation, necessitating the use of 86Y or 111In for diagnostic imaging [39]. 188Re has an extremely short half‑life (17.0 h), restricting its applicability to rapidly biotargeting agents [40]. 89Zr is commonly used as a PET tracer due to its high‑resolution imaging resulting from its relatively low average positron energy [41]. A preclinical study evaluated the imaging and therapeutic potential of 89Zr‑labelled MDX‑1338 in NSCLC models [42]. Additionally, a clinical case reported a patient with progressive lung metastases from fibrosarcoma treated with 153Sm‑FAPI‑46. It notes that this radiopharmaceutical’s low molar activity and potential contamination from the daughter nuclide 154Eu could limit broader application [43]. α emitters α particles are helium nuclei with a charge of + 2. Their LET ranges from 50 to 230 keV/µm, with a short tissue penetration range of approximately 50–100 μm. Consequently, these properties confer highly specific and potent cytotoxicity against small tumours [44]. However, the energy imparted to the recoil atom during α decay reaches up to 100 keV, vastly exceeding typical chemical bond energies and leading to the release of daughter isotopes [45]. They can non‑specifically irradiate surrounding tissues through successive decay chains, contributing to off‑target toxicity [46]. Therefore, only a limited number of α emitters are suitable for therapy. In lung cancer, the most frequently employed α emitters in descending order of application are 225Ac, 211At, 227Th, 212Pb, and 213Bi. 225Ac is the most commonly used, having been employed in 9 studies (see Additional file 1). Its relatively long half‑life (9.92 days) facilitates radiopharmaceutical synthesis and interregional transport aligning well with the pharmacokinetics of antibodies and peptides [47]. With an LET of 80–230 keV/µm and a tissue penetration range of 50–100 μm, 225Ac delivers 4 consecutive α particles before reaching a stable nuclide [48]. This decay cascade enhances cytotoxicity by inducing dense DNA damage, making it particularly effective against micrometastatic disease. However, the release of daughter isotopes (e.g., 213Bi, 209Pb) during decay may cause adverse toxicity [49]. As a decay daughter of 225Ac, 213Bi also exhibits significant therapeutic potential. Kratochwil C et al. reported using 213Bi‑DOTATOC to treat 7 patients with differentiated neuroendocrine tumours (NETs), including one pulmonary case. Nevertheless, the extremely short half‑life of 213Bi (45.6 min) remains a major practical barrier for production, radiolabelling, and administration cycles [50]. In contrast, 211At is a radioactive halogen with a half‑life of only 7.2 h. It emits a single α particle per decay and leads directly to stable 207Pb, thereby circumventing the issue of toxicity arising from radioactive daughter isotopes [51]. However, its clinical translation is hindered by in vivo instability due to dehalogenation and limited availability, as production requires high‑energy cyclotrons [52]. Both 227Th and 212Pb deliver mixed α particles and β particles. Specifically, 227Th first undergoes α decay to 223Ra before emitting β particles, while 212Pb decays through β particles to 212Bi followed by emitting α particles. This mechanism enables the simultaneous targeting of both macroscopic tumours and microscopic metastases [53]. Furthermore, the long half‑life of 227Th may prolong systemic exposure and increase toxicity risks [54]. For 212Pb, a primary safety concern is the substantial γ rays from its decay daughter 208Tl, which necessitates stringent radiation safety measures [55]. Carriers The efficacy and safety of RDCs are largely determined by their carriers, which govern key pharmacokinetic properties, including in vivo distribution and tumour‑targeting specificity. Current carriers encompass antibodies, peptides, small molecules, and nanoparticles (NPs). Full‑length antibodies offer high specificity, but their slow clearance often leads to haematotoxicity. Smaller derivatives, such as single‑domain antibodies (sdAbs) and bispecific antibodies (bsAbs), clear faster but remain a concern of renal accumulation. Peptides provide rapid tissue penetration and are clinically validated for SSTR targeting, yet they show lower affinity. Small‑molecule carriers display rapid clearance, deep penetration, and high tumour‑to‑background ratios. However, their application in lung cancer remains limited, and most studies have focused on FAPIs. Nanoparticle‑based systems can co‑deliver radionuclides and act as radiosensitisers. Regarding clinical development, completed trials document a shift from antibodies toward SSTR‑targeting peptides, which currently represent the dominant carrier class. Small‑molecule carriers remain less common and are mainly in early‑phase evaluation. Antibody derivatives (sdAbs and bsAbs), leveraging pretargeting and dual‑targeting strategies, are now entering early clinical trials. By contrast, most nanoparticle‑based systems are still preclinical. Advances in drug delivery systems have driven the development of carriers with smaller molecular weights, higher specificity, and more rapid clearance (Fig. 2B). These improvements have not only enhanced therapeutic efficacy and reduced off‑target toxicity but also broadened the clinical potential of TRT. Antibodies Antibodies and their derivatives continue to be the most widely employed carriers for RDCs in lung cancer. Among radioimmunoconjugates (RICs), most utilise IgG monoclonal antibodies (mAbs). However, the prolonged in vivo clearance of full‑length antibodies leads to increased off‑target toxicity, hindering their widespread clinical use [56]. Emerging antibody derivatives such as sdAbs and bsAbs, along with NPs, offer promising avenues to overcome these challenges. SdAbs are minimal antigen‑binding domains derived from heavy chain antibodies, characterised by low molecular weight (~ 15 kDa), high affinity and specificity, and low immunogenicity [57, 58]. Consequently, their favourable pharmacokinetic profile includes excellent tumour penetration and rapid clearance from blood, making them ideal carriers for RDCs. In a study by Hannelore C et al., the RIC (sdAb 4AH29 labelled with 225Ac or 131I) showed potent tumour targeting and rapid renal clearance in animal models. After injection of 225Ac‑DOTA‑4AH29, tumour uptake reached 3.3 ± 1.1%IA/g at 3 h, with a renal clearance rate of 11.5 ± 0.5%IA/g at the same time point [59]. Clinically, two sdAb‑based radiopharmaceuticals, 177Lu‑RAD204 and 225Ac‑ABD147, are in Phase I clinical trials for NSCLC and SCLC, respectively (NCT06305962, NCT06736418). However, the small size of sdAbs also leads to substantial renal accumulation, resulting in unnecessary nephrotoxicity while enhancing blood clearance [60]. In view of this, approaches such as incorporating albumin‑binding moieties and utilising red blood cell hitch‑hiking technology have been shown to reduce nephrotoxicity [61, 62]. BsAbs represent an emerging class of antibodies featuring two antigen‑binding sites capable of binding two distinct antigens, offering broad prospects for developing novel targeted therapies [63]. One strategy focuses on dual‑targeting (simultaneously targeting two tumour‑associated antigens) to enhance tumour specificity and broaden patient eligibility. A representative example involves the ²²⁵Ac‑FPI‑2068, which targets tumours expressing EGFR and/or c‑MET, is currently in a Phase I trial for multiple solid tumours, including NSCLC (NCT06147037). Another approach employs pretargeting strategies to improve pharmacokinetics and reduce radiation exposure to normal tissues compared with directly labelled antibodies [64]. Taking 177Lu‑DOTA‑GD2‑SADA as an example: it utilises a bsAb (anti‑GD2 × anti‑DOTA) fused to SADA domains. The tetrameric GD2‑SADA construct initially binds GD2‑expressing tumours. After unbound GD2‑SADA tetramers dissociate into dimers or monomers and are rapidly cleared, 177Lu‑DOTA is infused to bind the anti‑DOTA arm of the tumour‑localised bsAb, delivering radiation precisely while minimising systemic exposure by shortening radionuclide residence time in the body [65]. A Phase I trial of this drug in patients with recurrent or metastatic solid tumours, including SCLC, is underway (NCT05130255) [66]. 177Lu‑IMP288‑TF2 is another pretargeting drug with a similar mechanism [67]. A Phase I/II trial in patients with SCLC and NSCLC (NCT01221675) showed enhanced pharmacokinetics and limited haematotoxicity, though antitumour efficacy requires further validation [68]. In addition, combining NPs with mAbs as a radionuclide delivery system helps sustain optimal drug concentrations while reducing systemic toxicity [69]. For instance, the nanoparticle‑based conjugate 177Lu‑HApZr‑CTX, currently in preclinical development, exhibits accelerated and more specific uptake in pulmonary tumours compared with mAbs alone [70]. Concurrently, when exposed to X‑rays, metal NPs significantly elevate intracellular reactive oxygen species (ROS), potentially enhancing radiotherapy efficacy [71]. Therefore, nanoparticle‑based RDCs hold promise as both radiosensitisers and nanocarriers for therapeutic radionuclide delivery, achieving dual effects of antitumour activity and radiotherapy sensitisation. Peptides and small molecules Compared with antibodies, peptides typically show lower affinity and shorter half‑lives in vivo, but demonstrate higher efficiency in tissue penetration and cellular internalisation [72]. Among peptide radionuclide‑drug conjugates (PRCs), SSTR remains the predominant target, with a shift in ligand design from agonists toward antagonists. SSTR antagonists can bind to additional binding sites, resulting in greater tumour uptake and a higher tumour‑to‑background ratio (TBR) [73]. Cyclic peptides, owing to their rigid conformation, exhibit stronger binding and greater stability than linear peptides, demonstrating broad development potential. Examples include the cyclic RGD conjugate 177Lu‑DOTA‑E (cRGDfK)², 177Lu‑DPI‑4452, and 177Lu‑FAP‑2286 [74, 75]. Additionally, a smaller subset of RDCs in lung cancer employ small‑molecule ligands, primarily fibroblast activation protein inhibitors (FAPIs) such as FAPI‑46 and FAPT [43, 76]. Notably, synergistic targeting strategies combining peptides and small molecules into hybrid ligands may help overcome the limitations of single‑modality ligands. For example, 177Lu‑DOTA/Alexa647‑HA100‑N uses RGD peptide‑modified HA to enhance tumour uptake through dual synergistic targeting, addressing the relatively low specificity of HA alone for CD44 [77]. Similarly, 131I‑RGD‑BSA‑PCL potentiates the antitumour efficacy of liposomes by conjugating RGD peptides to PEG‑modified long‑chain liposomes (BSA‑PCL) [78]. Chelators and linkers In RDCs, the formation of stable coordination complexes between radionuclides and targeting carriers primarily relies on bifunctional chelating agents (BFCAs). An ideal BFCA should possess favourable metal‑complexation kinetics, high selectivity for the conjugated radionuclides, and excellent thermodynamic and stability [79]. BFCAs are broadly categorised into two main structural classes: macrocyclic chelators, exemplified by DOTA, and acyclic chelators, represented by DTPA. Recent years have seen the emergence of novel chelators tailored for specific radionuclides, including HEHA, DFO, PSC, and 3,2‑HOPO. Furthermore, certain radionuclides (e.g., 131I, 188Re, and 211At) can be stably conjugated to carriers via direct labelling methods without requiring chelators. Additionally, BFCAs are typically connected with the carriers through covalent bonds. They can also incorporate pharmacokinetic modifier linkers (PKM linkers) to improve pharmacokinetics and tumour targeting capabilities [80] (Fig. 2C). The macrocyclic chelator DOTA and its derivatives are the most commonly used BFCAs in RDCs for lung cancer. DOTA forms stable chelates with various radionuclides. Its derivative DOTA‑MMA has also been shown to yield highly thermodynamically stable conjugates with radionuclides such as 177Lu and 225Ac [81]. This further illustrates the modifiability of the DOTA to accommodate diverse carriers and radionuclides. Nevertheless, DOTA still presents certain limitations, such as reduced thermodynamic stability and slowed chelation kinetics when chelating with large metal ions [79]. Moreover, the coupling reaction between DOTA and radionuclides requires heating to 100 °C for 5–30 min under mildly acidic conditions (pH 4.0‑6.0), which demands adequate thermal stability from the carriers [82]. Another common acyclic chelator, DTPA, was developed earlier and is frequently paired with the widely used radionuclide 90Y in various RDC drugs, such as 90Y‑MAb‑6, 90Y‑DTPA‑cT84.66, 90Y‑mCC49, and 90Y‑hMN‑14 [83–85]. Compared with the macrocyclic DOTA, the chainlike DTPA rapidly forms complexes under mild conditions (room temperature, pH 4.5–5.5), thereby simplifying the radiolabelling process. However, its in vivo stability is generally inferior to that of DOTA [82]. To address this limitation, the derivative DTPA‑CHX‑A″ increases ligand rigidity by incorporating a cyclohexyl moiety, significantly improving its stability [82, 86]. This modified chelator has been utilised in the construction of three RDCs: 177Lu‑DTPA‑CHX‑A“‑αCD133, 177Lu‑DTPA‑CHX‑A”‑SC16, and 47Sc‑DTPA‑CHX‑A"‑cetuximab [87–89]. Additionally, several other chelating agents are employed with specific radionuclides due to their distinct properties. HEHA was initially designed for the large radiometal 225Ac. An HEHA‑based RDC, 225Ac‑HEHA‑PD‑L1‑i, remains under preclinical investigation [90]. The chelator DFO, used in 89Zr‑DFO‑CXCR4‑mAb, is an acyclic hexadentate ligand that rapidly and efficiently forms stable complexes with 89Zr at room temperature. It is currently the most widely used chelator for 89Zr [42, 91]. 227Th‑3,2‑HOPO‑anetumab employs the octadentate chelator 3,2‑HOPO, which forms stable thorium complexes [92, 93]. For another α emitter, 212Pb, Li M et al. developed a Pb‑specific chelator (PSC) composed of carboxyl and acetamide donor groups [94]. 212Pb‑PSC‑PEG2‑TOC, utilising this chelator, is currently being evaluated in three Phase I/IIa clinical trials (NCT05636618, NCT06479811, NCT06148636). However, not all radionuclides require chelating agents. 131I, for instance, can be covalently linked to carriers via direct labelling methods such as chloramine‑T oxidation or through tyrosine residues, as seen with 131I‑ERIC‑1 and 131I‑cNGEGQQc. Alternatively, it can be conjugated to sdAbs using chelating agents like GMIB [95–97]. 188Re also enables direct labelling of carriers, which streamlines manufacturing but presents challenges such as limited site specificity and reduced stability [98–100]. Furthermore, 211At structurally resembles 131I and currently lacks a chelator capable of providing sufficient in vivo stability [79]. All three 211At‑labelled RDCs in lung cancer (211At‑ADIPA, 211At‑SAB‑Oct, and 211At‑M201B) employ an unactivated aryl‑based astatine labelling strategy. Under the oxidation of N‑chlorobutyridine, 211At is introduced onto the aromatic ring of the tributyltin precursor via electrophilic substitution, followed by conjugation with the targeting ligand [101–103]. RDCs, particularly PRCs, commonly face pharmacokinetic challenges. Rapid renal clearance leads to insufficient tumour uptake and restricted radiation delivery, frequently necessitating repeated administration or higher administered doses to achieve therapeutic efficacy [104, 105]. The introduction of PKM linkers between the chelators and the carriers has therefore become a critical and effective strategy. In the field of lung cancer, two main types of PKM linkers are polyethylene glycol (PEG) and Evans Blue dye (EB). Both 131I‑RGD‑BSA‑PCL and 177Lu‑FAPT employ a PEG linker to enhance their hydrophilicity, in vivo stability, and radiochemical yield [76, 78, 106]. In contrast, EB reversibly binds to albumin in vivo, thereby prolonging circulation time and increasing tumour accumulation [107]. This linker has been successfully applied in drug candidates such as 177Lu‑EB‑RGD and 225Ac‑EBTATE [108, 109]. The development and implementation of PKM linkers are instrumental in reducing systemic toxicity due to allowing for lower administered doses. Targets Target selection is a crucial step in developing RDCs, significantly impacting both drug design and translation prospects. Lung cancer‑specific targets are mainly categorised into three types based on their spatial distribution and mechanisms: cell membrane targets, extracellular matrix targets, and intracellular targets. Among these, tumour cell membrane targets, particularly SSTRs, have been the most extensively studied (see Additional file 1). This section will systematically elaborate on the application strategies and developmental potential of different target categories, comprehensively illustrating the broad prospects of RDCs in precision lung cancer therapy (Fig. 3). Radiobiological mechanisms and common targets of RDCs in lung cancer. Abbreviations: VEGFR, vascular endothelial growth factor receptor; SMRC, small‑molecule radionuclide conjugates; PRC, peptide radionuclide conjugates; ARC, antibody radionuclide conjugates; BsAbs, bispecific antibodies; DLL3, Delta‑like ligand 3; GPCR, G protein‑coupled receptor; SSTR, somatostatin receptor; CXCR4, C‑X‑C chemokine receptor type 4; CCK2R, cholecystokinin‑2 receptor; ROBO1, roundabout guidance receptor 1; MSLN, mesothelin; CDH‑3, cadherin‑3 (P‑cadherin); EGFR, epidermal growth factor receptor; EGP‑1, epithelial glycoprotein‑1; CEA, carcinoembryonic antigen; NCAM1, neural cell adhesion molecule 1; PD‑L1, programmed death‑ligand 1; FAP, fibroblast activation protein; CAIX, carbonic anhydrase IX; PSMA, prostate‑specific membrane antigen; GD2, disialoganglioside GD2. Created with Microsoft PowerPoint and BioGDP.com [259] G protein‑coupled receptors (GPCRs) SSTRs are GPCRs that regulate cell proliferation and hormone secretion, with five distinct subtypes identified [110]. Among these, SSTR2 is the most frequently expressed subtype in NETs of the lung [111]. Somatostatin and its synthetic analogues act as agonists that bind specifically to SSTR2, enabling the precise delivery of RDCs to tumour cells via endocytosis for targeted radiotherapy [112]. The octreotide derivatives DOTATOC and DOTATATE are the most commonly used somatostatin analogues in clinical practice. 90Y‑DOTATOC, developed by Novartis, was the first SSTR‑targeted RDC to enter clinical trials. However, its use is associated with substantial risks of transient haematotoxicity and permanent nephrotoxicity [113]. Two clinical trials of 90Y‑DOTATOC targeting SCLC (NCT00006370, NCT00006368) have been completed, but results remain unpublished. Furthermore, an exploratory pilot study suggested that the antitumour activity of 90Y‑DOTATOC in SCLC was relatively limited [114]. Owing to the toxicity profile of 90Y, 177Lu has emerged as the preferred radionuclide for subsequent clinical development. A comparative cohort study in patients with advanced NETs showed no significant difference in median overall survival between those treated with 177Lu‑DOTATOC and 90Y‑DOTATOC. However, 177Lu‑DOTATOC demonstrating lower haematotoxicity [115]. In a phase II trial, the outcomes (disease control survival, DCS: 34.5 months, objective response rate, ORR: 54.2%) suggested that ¹⁷⁷Lu‑DOTATOC might be more effective than any other current treatment for metastatic NETs (including NETs of the lung) [116]. Additionally, a phase III trial comparing the impact on survival and quality of life of 177Lu‑DOTATOC with everolimus in patients with NETs of the lung and thymus is currently ongoing (NCT05918302). Beyond 177Lu‑DOTATOC, two additional DOTATOC‑based RDCs include 213Bi‑DOTATOC and 212Pb‑PSC‑PEG2‑TOC. Three phase I or I/II trials of the latter are currently underway (NCT05636618, NCT06479811, NCT06148636). DOTATATE represents another conjugation of the chelator DOTA with octreotide. The therapeutic potential of both 177Lu‑DOTATATE and 225Ac‑DOTATATE in SCLC has been demonstrated in preclinical studies [117–119]. A completed phase I trial (NCT03325816) indicated that 177Lu‑DOTATATE and nivolumab exhibited synergistic effects in treating NETs of the lung [120]. Current research is centred on assessing the efficacy and safety of RDCs, both as monotherapies and in combination with chemotherapy or immunotherapy for ES‑SCLC (NCT05142696, NCT05595460). Notably, Zhang J et al. reported a patient with refractory thymic NET resistant to 177Lu‑DOTATOC (β‑radiation) successfully treated with 225Ac‑DOTATATE [121]. Furthermore, conjugating EB to DOTATATE (225Ac‑EBTATE) can prolong its circulation time, thereby enhancing tumour uptake and retention [108]. The SSTR‑targeted RDCs discussed thus far are all based on agonists. In lung cancer, JR11 and LM3 are two recently developed SSTR antagonists. Phase I clinical trials of 177Lu‑DOTA‑LM3 and 177 Lu‑DOTA‑JR11 (NCT02592707, NCT02609737) demonstrated superior biodistribution and higher tumour radiation doses relative to agonists, despite differences in their clinical safety profiles [122–124]. This property not only improves the accuracy of tumour imaging but also enhances therapeutic efficacy [125]. Furthermore, a phase I/II clinical trial is currently investigating 225Ac‑DOTA‑JR11, an α emitter SSTR antagonist, in patients with ES‑SCLC (NCT06939036). Researchers are also actively exploring novel peptide ligands targeting SSTR beyond octreotide. Qin S et al. introduced D‑amino acids into the octreotide scaffold to construct 211At‑SAB‑Oct, a modification that enhances resistance to enzymatic degradation and prolongs in vivo retention [103]. P2045, an 11‑amino acid somatostatin analogue, has been conjugated with 188Re to form 188Re‑P2045, which has advanced to clinical evaluation [126]. A completed phase I trial confirmed the feasibility of 188Re‑P2045 for advanced lung cancer. Among the eight patients enrolled, five achieved stable disease lasting at least 8 weeks, although all of them ultimately experienced progressive disease [127]. Regrettably, a phase I/II dose‑escalation trial (NCT02030184) evaluating 188Re‑P2045 both as monotherapy and in combination with topotecan for SCLC was withdrawn due to the departure of the principal investigator. The status of another clinical trial involving 188Re‑P2045 remains unknown (NCT00100256). Within the GPCR family, targets such as the cholecystokinin‑2 receptor (CCK2R) and the C‑X‑C chemokine receptor type 4 (CXCR4) also serve as key molecular targets in lung cancer. CCK2R, a receptor for gastrin and cholecystokinin, is commonly expressed in SCLC (57%), but is relatively rare in NSCLC [128]. 177Lu‑DOTA‑MGS5 has been evaluated in first‑in‑human dosimetry studies involving patients with SCLC, positioning it for clinical trials [129]. CXCR4 is a cell surface receptor specific for the C‑X‑C motif chemokine ligand 12 (CXCL12) and has a key role in tumour adhesion, invasion, survival and growth [130, 131]. The CXCR4‑targeting monoclonal antibody MDX‑1338 and the peptide pentixafor have demonstrated clinical activity in haematological malignancies, although their efficacy and safety in solid tumours require further investigation [132, 133]. 89Zr‑MDX‑1338 and 212Pb/203Pb‑pentixather have each demonstrated favourable therapeutic responses in xenograft models of NSCLC and SCLC, respectively [42, 134]. Among them, ²¹²Pb‑pentixather is poised to enter clinical trials, with an ongoing phase I study (NCT05557708) currently recruiting participants to evaluate its safety in patients with pulmonary atypical carcinoid tumours. Fibroblast activation protein (FAP) FAP is overexpressed in cancer‑associated fibroblasts (CAFs) within the tumour stroma and has a crucial role in tumour cell growth, invasion and migration [135]. A systematic review found FAP expression to be upregulated in nearly 100% of NSCLC cases and in 67% of SCLC and LCNC cases [136]. Following the first‑generation small‑molecule inhibitor FAPI‑04, which demonstrated clinical potential, its optimised derivative FAPI‑46 enables labelling with multiple isotopes and has a relatively short half‑life [137]. Kratochwil C et al. reported a case of pulmonary metastases from fibrosarcoma treated with 153Sm‑FAPI‑46 [43]. Furthermore, Huang J et al. developed 177Lu‑DOTA‑FAPT by designing a GlcP‑PEG2 hydrophilic linker, which showed improved pharmacokinetics and favourable tumour uptake compared with ¹⁷⁷Lu‑FAPI‑04 [76]. Its paired diagnostic RDC showed high sensitivity for detecting lymph node metastases in NSCLC [76]. Consequently, FAPI PET/CT could offer a more precise imaging modality for preoperative staging, treatment selection and prognosis prediction. Currently in clinical development, a promising small‑molecule inhibitor, FAP‑2286, exhibits enhanced binding stability and superior detection of metastatic lesions, benefiting from its cyclic structure [138]. ¹⁷⁷Lu‑FAP‑2286 has shown efficacy and a manageable safety profile in patients with advanced lung cancer, and a phase I/II trial is ongoing in a broader population of advanced solid tumours (NCT04939610) [75]. Beyond small molecules, antibodies represent another important class of FAP‑targeting ligands. Ceuppens H et al. developed 131I/225Ac‑DOTA‑4AH29 based on the single‑chain antibody 4AH29 [59]. Notably, both 225Ac‑DOTA‑4AH29 and 177Lu‑FAP‑2286 have shown pronounced antitumour activity and substantial clinical potential in combination with ICIs in lung cancer [59, 139]. Additionally, 131I‑sibrotuzumab demonstrated an acceptable safety profile in a phase I dose‑escalation study (NCT02209727) involving patients with NSCLC [140]. Another FAP‑targeting RDC, ¹⁷⁷Lu‑NNS309, is currently being evaluated in a phase I trial (NCT06562192). Delta‑like ligand 3 (DLL3) DLL3 is an inhibitory Notch ligand aberrantly expressed in up to 85% of patients with SCLC, where it promotes tumour cell proliferation, migration and invasion [141]. In SCLC xenograft models, the ¹⁷⁷177Lu‑labelled anti‑DLL3 antibody SC16 (177Lu‑DTPA‑CHX‑A″‑SC16) demonstrated high DLL3‑specific tumour uptake and mild transient haematotoxicity [87]. However, Lakes AL et al. found that the overall efficacy of 225Ac/177Lu‑DOTA‑MMA‑SC16.56 was inferior to pyrrolobenzodiazepine‑based antibody–drug conjugates due to poor serum stability [81]. A phase Ia/b dose‑escalation trial is currently underway to evaluate the safety and initial efficacy of 225Ac‑ABD147 in patients with SCLC or pulmonary large‑cell neuroendocrine carcinoma (LCNEC) who have progressed after platinum‑based chemotherapy (NCT06736418). Epidermal growth factor receptor (EGFR) EGFR is a receptor tyrosine kinase that, upon ligand binding, dimerises and activates downstream signalling cascades that regulate cell growth, proliferation and survival [142]. Mutated EGFR serves as both a prognostic marker and a predictor of resistance in lung cancer. EGFR tyrosine kinase inhibitors (TKIs) are now standard first‑line therapies for advanced NSCLC [11, 143]. Although mAbs such as cetuximab have limited efficacy as monotherapies, their conjugation with radionuclides to form RDCs enables synergistic antitumour effects and represents an emerging strategy to overcome EGFR resistance [144]. In preclinical studies, a single injection of 188Re‑cetuximab improved tumour growth inhibition (MGI = 0.049) and prolonged median survival (62.50 days) compared with controls [145]. Concurrently, Kim D‑M et al. demonstrated that 47Sc‑DTPA‑cetuximab exhibited superior therapeutic efficacy against NSCLC cells relative to cetuximab monotherapy. The study also suggested that RUNX3 modulates tumour sensitivity to 47Sc‑DTPA‑cetuximab. This modulation can occur through regulation of DNA damage and repair mechanisms, thereby counteracting resistance to EGFR‑targeted therapy [88]. Furthermore, a cetuximab‑conjugated hydroxyapatite‑zirconium NP (HApZr‑CTX) has been developed as a delivery system for 177Lu, with evidence of radiosensitisation following X‑ray irradiation [70]. In addition, a phase Ia/b dose‑escalation trial is currently evaluating 225Ac‑FPI‑2053 in patients with various solid tumours, including NSCLC (NCT06147037). Programmed death‑ligand 1 (PD‑L1) PD‑L1 is a transmembrane immune checkpoint protein that suppresses T cell responses via PD‑1 engagement and serves as both a predictive biomarker and a therapeutic target in lung cancer [146, 147]. Luna‑Gutiérrez M et al. synthesised two radiolabelled PD‑L1 inhibitory peptides (177Lu‑DOTA‑PD‑L1‑i and 225Ac‑HEHA‑PD‑L1‑i), both of which could potentially enhance the antitumour efficacy of immune checkpoint inhibitors (ICIs) [90]. Among these, 177Lu‑DOTA‑PD‑L1‑i is considered more promising for clinical translation due to its simpler synthesis and superior stability. Additionally, an ongoing phase 0/I first‑in‑human study is evaluating the safety and tolerability of 177Lu‑RAD204 in patients with solid tumours to determine the maximum tolerated dose (MTD) and recommended dose (NCT06305962). Integrins Integrins are a family of heterodimeric transmembrane receptors that mediate cell adhesion, migration and signalling between the extracellular matrix and immunoglobulin superfamily molecules [148]. α3β1 mediates adhesion of epithelial cells to basement membrane laminins [149]. αvβ3 is highly expressed on tumour neovasculature and correlates with poor prognosis [150, 151]. αvβ6 is selectively upregulated in epithelial malignancies, where it cooperates with fibrotic microenvironment remodelling to facilitate metastasis [152, 153]. Several integrin‑targeted RDCs have demonstrated antitumour activity in NSCLC xenograft models, although conjugated to different forms of RGD peptides. These include the α3β1‑targeting RDCs 131I‑cNGEGQQc and 131I‑RGD‑BSA‑PCL, as well as the αVβ3‑targeting RDCs 177Lu‑DOTA‑E(cRGDfK)₂ and 177Lu‑EB‑RGD [74, 78, 95, 109]. Another αvβ6‑targeting RDC, 177Lu‑DOTA‑ABM‑5G, is being evaluated in a phase I trial for metastatic NSCLC (NCT06228482). Carcinoembryonic antigen (CEA) CEA is a glycoprotein broadly expressed across multiple tumour types and serves as a key biomarker for monitoring disease recurrence and progression in lung cancer [154, 155]. Three CEA‑targeted RDCs are currently under evaluation in phase I or I/II clinical trials for lung cancer: 90Y‑DTPA‑cT84.66 (NCT00738452), 90Y‑hMN‑14 (NCT00006458, NCT00006347), and 177Lu‑IMP‑288 (NCT01221675). Cluster of differentiation (CD) CD molecules are cell surface markers expressed on immune cells of distinct lineages and play a key role in immune cell interactions within the tumour microenvironment [26]. Several CD molecules, including CD56, CD133, CD44, and CD13, have emerged as relevant targets for TRT in lung cancer. CD56 (NCAM1) and CD133 (prominin‑1) are overexpressed in most SCLCs [156, 157]. Both the CD56‑targeting RDC 131I‑ERIC1 and the CD133‑targeting RDC 177Lu‑DTPA‑A″‑CHX‑αCD133 have demonstrated marked tumour growth inhibition in SCLC xenograft models [89, 96]. Additionally, a first‑in‑human trial of the CD44v6‑targeting RDC 177Lu‑AKIR001 is underway (NCT06639191). Hyaluronic acid (HA), a primary ligand of CD44, exhibits limited tumour targeting specificity in vivo owing to non‑specific binding [158]. To address this, Li X et al. functionalised HA with asparagine‑glycine‑arginine (NGR) peptides, which target CD13 [77]. This dual‑targeting agent, 177Lu‑DOTA/Alexa647‑HA100‑N, simultaneously engages CD13 and CD44 and represents a promising strategy for TRT. Similarly, 225Ac‑FPI‑2068, which dually targets EGFR and c‑MET, is currently under phase I trial in patients with various solid tumours, including NSCLC (NCT06147037). Intracellular targets Tumour necrosis therapy (TNT) is a targeted radiotherapy approach that uses mAbs to recognise intracellular antigens exposed after tumour cell death, thereby directing RDCs to necrotic areas to kill adjacent cancer cells [159]. 131I‑chTNT has been approved by China’s National Medical Products Administration as the world’s first marketed radioimmunotherapy drug for lung cancer, indicated for advanced lung cancer unresponsive to or recurrent after chemoradiotherapy [160]. Staudacher, Alexander H et al. radiolabelled the anti‑La/SSB antibody DAB4 (APOMAB) with 227Th and demonstrated that 227Th‑APOMAB preferentially targets hypoxic tumour cells within necrotic regions [161]. Combining this RDC with chemotherapy effectively overcame chemoradiotherapy resistance. Other studies have explored small‑molecule ligands as alternatives to mAbs to reduce immunogenicity. Prohy, a polycyclic aromatic anthraquinone from Hypericum perforatum, binds selectively to necrotic tumour tissue [162]. Combretastatin‑A4‑phosphate (CA4P) is a vascular disrupting agent that rapidly induces widespread ischaemic necrosis, thereby creating an ideal target for TNT [163]. Liu X et al. investigated the antitumour efficacy of 131I‑prohy combined with CA4P in NSCLC xenograft models and found that CA4P markedly enhanced tumour uptake of 131I‑prohy within necrotic areas [164]. Acrolein is a toxic aldehyde that enters the body both exogenously (for example, through combustion of organic matter or high‑temperature cooking) and endogenously [165]. It is significantly overexpressed in most cancer cells but present at very low levels in normal tissues [166]. Ode Y et al. developed 211At‑ADIPA, which accumulates specifically in tumours and reacts selectively with endogenous acrolein [102]. By covalently binding to organelles, it can anchor α particles within cancer cells, thereby killing them while minimising damage to healthy tissues. Besides, this compound holds promise for targeting diverse cancer types regardless of subtype. Vascular‑related targets The growth and metastasis of lung cancer are highly dependent on angiogenesis, making anti‑angiogenic therapy a key strategy for controlling disease progression [167]. Bevacizumab, an FDA‑approved humanised mAb, blocks angiogenesis by specifically binding to vascular endothelial growth factor (VEGF) [168]. Direct labelling of bevacizumab with 188Re has been shown to enhance its antitumour efficacy while reducing side effects in preclinical NSCLC models [98]. MAb 201B is another mAb targeting pulmonary thrombomodulin. When conjugated with α emitters such as 225Ac or 211At, it enables precise targeting of perivascular tumours in the lung, as exemplified by 225Ac‑HEHA‑MAb 201B and 211At‑SAPS‑MAb 201B [101, 169]. Others In addition to the targets discussed above, several emerging candidates are currently under preclinical investigation or in early‑phase clinical trials. 90Y‑MAb‑6 and 90Y‑FF‑21,101, both targeting P‑cadherin, displayed potent antitumour activity in mouse xenograft models [85, 170]. 90Y‑FF‑21,101 further completed pharmacokinetic and dosimetry studies in cynomolgus monkeys, enabling estimation of a safe dose for human trials [170]. A completed phase I trial (NCT02454010) of ⁹⁰Y‑FF‑21,101 confirmed its favourable safety profile and provided early evidence of antitumour activity [171]. Notably, its non‑myeloablative properties support potential combination strategies with immunotherapies or other treatment modalities. Extracellular matrix targets, such as tenascin‑C and glucuronidase, also have been explored. Tenascin‑C is a hexameric extracellular matrix glycoprotein that plays a key role in cell proliferation and migration [172]. 131I‑F16SIP, a 131I‑labelled conjugate of the anti‑tenascin‑C minibody F16SIP, has entered phase I/II clinical trials for a range of malignancies, including lung cancer (NCT01240720). Furthermore, İnce İ et al. developed 131I‑TQGMNP, a novel RDC that conjugates magnetic nanoparticles (MNPs) and the thymoquinone glucuronide (TQG) for theranostic applications in β‑glucuronidase‑rich tumours such as lung cancer [173]. Additional RDCs with promising preclinical profiles in lung cancer include ROBO1‑targeting 90Y‑anti‑ROBO1 IgG, mesothelin‑targeting 227 Th‑3,2‑HOPO‑anetumab and epithelial glycoprotein 1‑targeting 177Lu-DOTA-RS7 [36, 92, 174]. Moreover, several other RDCs have advanced to phase I or I/II trials, including 90Y‑mCC49 targeting tumour‑associated glycoprotein 72, 177Lu‑DOTA‑complex targeting GD2 (NCT05130255), 177Lu‑DPI‑4452 targeting carbonic anhydrase IX (NCT05706129) and 177Lu‑J591 targeting prostate‑specific membrane antigen (NCT00967577). Completed clinical trials A total of 30 RDCs in lung cancer have entered clinical trials, with 11 having been completed (Table 1). Early clinical trials primarily employed 131I‑ or 90Y‑labelled monoclonal antibodies. While these agents demonstrated antitumour activity, their development was hindered by significant haematotoxicity due to prolonged circulation and frequent immunogenic responses. For instance, phase I trials of 131I‑sibrotuzumab and 90Y‑DTPA‑cT84.66 reported immunogenic responses in 32.0% (human anti‑human antibody, HAHA) and 33.3% (human anti‑chimeric antibody, HACA) of patients, respectively [83, 140]. Similarly, the clinical application of 131I‑chTNT, although approved in China, has also been limited by its chimeric antibody [160]. Other early‑phase trials focused on different objectives: for example, one phase I trial determined the maximum tolerated dose (MTD) of 90Y‑mCC49 to be 14 mCi/m², while another provided preliminary efficacy data for 188Re‑P2045 in advanced lung cancer (OS: 11.5 months, EFS: 2.9 months, N = 15) [84, 127]. Table 1Summary of completed clinical trials on RDCs in lung cancerRadionuclideRDCTargetCarrierPhase and IDParticipantsOutcome177Lu177Lu‑DOTATATE [120]SSTR2PeptideI/IINCT03325816NETs of the lungN = 9/9ORR: 14.3%, DCR:42.9%AEs of G3 or more (55.6%): lymphopenia 44.4%, pneumonitis 11.1%, anaemia 11.1%, and thrombocytopenia 11.1%177Lu‑DOTATOC [116]SSTR2PeptideIINCT04790708NETs of the lungN = 7/561ORR: 0%, DCR: 46.2%, PFS: 6 months (95% CI: 2.9‑9.0), OS: 13.3 months (95% CI: 2.6‑23.9)AEs of G3 or more: only anaemia 1.8%177Lu‑DOTA‑JR11 [122, 124]SSTR2PeptideI/IINCT02592707NETs of the lungN = 7/40ORR: 21.1% (95% CI: 9.6–37.3), DCR: 94.7% (95% CI: 82.3–99.4), PFS: 28.1 months (95% CI: 19.4–non‑estimable)AEs of G3 or more (42.5%): lymphopenia 7.5%, thrombocytopenia 7.5%, neutropenia 7.5%INCT02609737NETs of the lungN = 1/20ORR: 45.0%, DCR: 85.0%, PFS: 21.0 months (95% CI: 13.6‑NR)AEs of G3 or more (20.0%): thrombocytopenia 20.0%, haemoglobin toxicity 10.0%, neutropenia 15.0%, leukopenia 15.0%177Lu‑DOTA‑LM3 [123]SSTR2PeptideIUndisclosedNENs of the lungN = 4/51ORR: 36.2%, DCR: 85.1%AEs of G3 or more (15.7%): lymphopenia, 9.8%, thrombocytopenia 5.9%177Lu‑J591PSMAMAbINCT00967577NSCLCUndisclosed177Lu‑IMP288 + TF2 [68]CEABsAbI/IINCT01221675NSCLC and SCLCN = 9/9AEs of G3 or more (0%): only G1 thrombocytopenia 22.2%, G2 anaemia 11.1%131I131I‑F16SIPTenascin‑CMAbI/IINCT01240720Lung cancerUndisclosed131I‑chTNT [160]Necrotic tumoursMAbApprovedLung cancerN = 107/107ORR: 34.6%, DCR: 89.7%AEs of G3 or more: thrombocytopenia 13.1%, neutropenia 6.5%, haemoglobin toxicity 2.8%131I‑sibrotuzumab [140]FAPMAbINCT02209727NSCLCN = 6/26DCR: 15.4%AEs of G3 or more 0%, HAHA reaction 32.0%188Re188Re‑P2045[127]SSTR2PeptideINCT00100256SCLC and NSCLCN = 15/15OS: 11.5 months, EFS: 2.9 monthsAEs of G3 or more (13.3%): lymphopenia 6.7%, symptomatic toxicity 6.7% (abdominal pain, hyponatraemia, and sudden death)90Y90Y‑DOTATOCSSTR2PeptideIINCT00006370SCLCUndisclosedINCT00006368Lung cancerUndisclosed90Y‑DTPA‑cT84.66 [83]CEAMAbINCT00738452NSCLCUndisclosedINCT00645060Lung cancerN = 7/36DCR: 63.9%AEs of G3 or more: leukopenia 27.8%, lymphopenia 30.6%; HACA reaction 33.3%90Y‑mCC49 [84]TAG‑72MAbIUndisclosedLung cancerN = 34DCR: 38.2%The MTD of 90Y‑mCC49/IFN: 14 mCi/m2, the MTD of paclitaxel co‑administration: 60 mg/m2Abbreviations: SSTR Somatostatin receptor, CEA Carcinoembryonic antigen, PSMA Prostate‑specific membrane antigen, TAG‑72 Tumour‑associated glycoprotein 72, FAP Fibroblast activation protein, bsAb Bispecific antibody, mAb Monoclonal antibody, NETs Neuroendocrine tumours, NENs Neuroendocrine neoplasms, NSCLC Non‑small cell lung cancer, SCLC Small cell lung cancer, ORR Objective response rate, DCR Disease control rate, OS Overall survival; progression‑free survival, EFS Event‑free survial, AEs Adverse events, MTD Maximum tolerated dose, HACA Human anti‑chimeric antibody, HAMA Human anti‑mouse antibody1 7/56 indicates that among the 56 enrolled participants, 7 were diagnosed with lung cancer. The same applies hereafter Summary of completed clinical trials on RDCs in lung cancer I/II NCT03325816 NETs of the lung N = 9/9 ORR: 14.3%, DCR:42.9% AEs of G3 or more (55.6%): lymphopenia 44.4%, pneumonitis 11.1%, anaemia 11.1%, and thrombocytopenia 11.1% II NCT04790708 NETs of the lung N = 7/561 ORR: 0%, DCR: 46.2%, PFS: 6 months (95% CI: 2.9‑9.0), OS: 13.3 months (95% CI: 2.6‑23.9) AEs of G3 or more: only anaemia 1.8% I/II NCT02592707 NETs of the lung N = 7/40 ORR: 21.1% (95% CI: 9.6–37.3), DCR: 94.7% (95% CI: 82.3–99.4), PFS: 28.1 months (95% CI: 19.4–non‑estimable) AEs of G3 or more (42.5%): lymphopenia 7.5%, thrombocytopenia 7.5%, neutropenia 7.5% I NCT02609737 NETs of the lung N = 1/20 ORR: 45.0%, DCR: 85.0%, PFS: 21.0 months (95% CI: 13.6‑NR) AEs of G3 or more (20.0%): thrombocytopenia 20.0%, haemoglobin toxicity 10.0%, neutropenia 15.0%, leukopenia 15.0% I Undisclosed NENs of the lung N = 4/51 ORR: 36.2%, DCR: 85.1% AEs of G3 or more (15.7%): lymphopenia, 9.8%, thrombocytopenia 5.9% I NCT00967577 I/II NCT01221675 NSCLC and SCLC N = 9/9 I/II NCT01240720 Lung cancer N = 107/107 ORR: 34.6%, DCR: 89.7% AEs of G3 or more: thrombocytopenia 13.1%, neutropenia 6.5%, haemoglobin toxicity 2.8% I NCT02209727 NSCLC N = 6/26 DCR: 15.4% AEs of G3 or more 0%, HAHA reaction 32.0% 188Re‑P2045 [127] I NCT00100256 SCLC and NSCLC N = 15/15 OS: 11.5 months, EFS: 2.9 months AEs of G3 or more (13.3%): lymphopenia 6.7%, symptomatic toxicity 6.7% (abdominal pain, hyponatraemia, and sudden death) II NCT00006370 I NCT00006368 I NCT00738452 I NCT00645060 Lung cancer N = 7/36 DCR: 63.9% AEs of G3 or more: leukopenia 27.8%, lymphopenia 30.6%; HACA reaction 33.3% I Undisclosed Lung cancer N = 34 DCR: 38.2% The MTD of 90Y‑mCC49/IFN: 14 mCi/m2, the MTD of paclitaxel co‑administration: 60 mg/m2 Abbreviations: SSTR Somatostatin receptor, CEA Carcinoembryonic antigen, PSMA Prostate‑specific membrane antigen, TAG‑72 Tumour‑associated glycoprotein 72, FAP Fibroblast activation protein, bsAb Bispecific antibody, mAb Monoclonal antibody, NETs Neuroendocrine tumours, NENs Neuroendocrine neoplasms, NSCLC Non‑small cell lung cancer, SCLC Small cell lung cancer, ORR Objective response rate, DCR Disease control rate, OS Overall survival; progression‑free survival, EFS Event‑free survial, AEs Adverse events, MTD Maximum tolerated dose, HACA Human anti‑chimeric antibody, HAMA Human anti‑mouse antibody 1 7/56 indicates that among the 56 enrolled participants, 7 were diagnosed with lung cancer. The same applies hereafter In recent years, 177Lu‑labelled SSTR‑targeting peptides have emerged as the predominant RDCs in clinical trials for lung cancer. The leading agent, 177Lu‑DOTATATE, expanded indications in NETs of the lung, showing modest antitumour activity when combined with nivolumab (objective response rate, ORR: 14.3%; disease control rate, DCR: 42.9%; N = 9) [120]. Another drug, 177Lu‑DOTATOC, has demonstrated a favourable systemic toxicity profile, with only 1.8% of patients experiencing grade ≥ 3 adverse events (AEs) in a phase II trial. However, its efficacy in the subgroup of patients with NETs of the lung was limited (ORR: 0%; DCR: 46.2%, N = 7) [116]. Furthermore, a notable trend is the shift from SSTR agonists to antagonists. Phase I trials of the antagonists 177Lu‑DOTA‑LM3 and 177Lu‑DOTA‑JR11 have shown improved stability and antitumour efficacy compared with agonists, with no nephrotoxicity reported [122–124]. However, 177Lu‑DOTA‑JR11 was associated with a higher haematotoxicity risk, with 42.5% of patients experiencing grade ≥ 3 AEs, mainly including lymphopenia, thrombocytopenia, and neutropenia [124]. In addition, the pre‑targeting strategy using 177Lu‑IMP288 combined with TF2 held promise in reducing drug retention time and limiting systemic toxicity, with no grade ≥ 3 AEs reported in the trial [68]. Nevertheless, current clinical evidence is limited by small sample sizes and a predominance of retrospective analyses. Many trials enrolled broad populations of NETs, and the resulting subgroup analyses for NETs of the lung are often limited by the heterogeneous patients. Moreover, most of these studies are early‑phase, single‑arm trials with exploratory endpoints. The original reports frequently present only estimated rates (e.g., ORR, DCR) without confidence intervals or P‑values, which further limits the statistical reliability of the available data. Therefore, it needs large‑scale prospective studies specifically targeting lung cancer to further evaluate the therapeutic potential and safety of RDCs. Furthermore, higher‑level evidence‑based studies are warranted to validate the efficacy and safety of RDCs in lung cancer. Prospects and challenges for next‑generation RDC development This section explores the development directions and translational challenges of next‑generation RDCs in lung cancer. It begins by analysing the main dose‑limiting toxicities of current RDCs in the clinic and discusses how to simultaneously improve efficacy and safety through personalised dosimetry and novel chelators. The focus then shifts to innovations in radionuclides and delivery systems. Targeted alpha therapy (TAT) is highlighted as a key driver of next‑generation RDCs. Emphasis is placed on how novel antibody derivatives, pretargeting strategies and multifunctional nanocarriers can address major challenges. These challenges include daughter nuclide redistribution, dose uniformity, and in vivo stability. Furthermore, achieving radiotheranostics in lung cancer also requires rational design of theranostic radionuclide pairs, sustainable production chains, and regulatory frameworks aligned with green nuclear medicine. Finally, combining TRT with chemotherapy, targeted therapy or immunotherapy offers considerable promise. Such combinations can overcome tumour heterogeneity and resistance while also remodelling the immune microenvironment. These strategies lay the groundwork for more diverse and integrated treatment approaches in lung cancer (Fig. 4). Next‑generation RDCs: challenges and future perspectives. A Adverse effects and mitigation strategies. Major ones include nephrotoxicity, haematotoxicity, and radiotoxicity from free radionuclides released via demetallisation. Countermeasures involve AI‑driven individualised dosimetry. B Combination therapies. RDCs synergise with chemotherapy, targeted therapy, immunotherapy, and isotope combinations (concurrent or sequential). C Next‑generation RDC design. Future development should prioritise novel radionuclides (e.g., TAT, AE emitters), stable chelators, advanced carriers (e.g., DARPins, bicyclic peptides, pretargeting systems), and scalable production with globally harmonised regulations. Abbreviations: TAT, targeted alpha therapy; AI, artificial intelligence; AE, Auger electron; ICD, immunogenic cell death; DARPin, designed ankyrin repeat protein; PNA, peptide nucleic acid; PROTACs, proteolysis‑targeting chimaeras. Created with Microsoft PowerPoint and BioGDP.com [259] Adverse reactions and safety RDCs offer high tumour selectivity with fewer side effects than conventional radiotherapy or chemotherapy. However, minimising radiation exposure to dose‑limiting organs such as the kidneys and bone marrow remains a crucial challenge. Prolonged retention in specific organs, influenced by carrier architecture and radionuclide properties, underlies these adverse effects. For instance, peptide‑based RDCs tend to accumulate in the renal proximal tubules, causing nephrotoxicity. Antibody‑based RDCs, by contrast, have long blood circulation times, leading to haematotoxicity. In addition, free radiometals released from unstable RDCs can be sequestered in the liver, potentially causing necrotic liver injury. α emitters have a short tissue range and limited distribution, therefore generally exhibiting lower haematotoxicity and nephrotoxicity than β emitters. However, α‑emitter conjugates are typically more challenging to stabilise. This may increase the risk of free radiometal release and greater hepatotoxicity. For any candidate, matching the biological half‑life of the carriers with the physical half‑life of the radionuclides is essential to maximise tumour dose while minimising off‑target accumulation. Nephrotoxicity is a primary concern for many RDCs, particularly peptide‑based agents such as somatostatin analogues. This reflects their clearance pathway: RDCs are primarily eliminated via glomerular filtration, and then reabsorbed and retained in proximal tubular cells [175, 176]. Key measures to reduce nephrotoxicity include adequate hydration, plasma expanders, radioprotectants such as amifostine, and co‑infusion of amino acids. Rigorous renal function monitoring is also essential [177]. Of note, α emitters have an extremely short tissue range and may not reach the glomerulus after tubular reabsorption, potentially conferring lower nephrotoxicity than β emitters. Furthermore, oral chelators such as DMPS, combined with diuretics, have been shown to enhance renal clearance of 225Ac daughter nuclides (e.g., 213Bi), thereby limiting accumulated renal damage [178]. Retrospective studies with long‑term follow‑up have not revealed significant chronic nephrotoxicity based on serum creatinine. However, this marker is suboptimal for assessing glomerular filtration rate (GFR), and imaging‑based renal evaluation is lacking [179]. Prospective, multicentre studies incorporating more appropriate biomarkers are therefore needed to confirm the long‑term renal safety of TRT. In addition to nephrotoxicity, haematotoxicity represents another major dose‑limiting factor. RDCs circulating in the peripheral blood can irradiate haematopoietic stem cells, leading to thrombocytopenia and potentially inducing clonal haematopoiesis [180]. Although rare, TRT may also increase the long‑term risk of secondary myeloid neoplasms such as acute myeloid leukaemia and myelodysplastic syndromes [181]. Antibody‑based RDCs are more frequently associated with haematotoxicity owing to their prolonged circulation in the bloodstream. Conversely, α emitters confer a lower risk of haematotoxicity due to their short tissue range and limited blood exposure [182]. Selecting carriers and radionuclides with appropriate circulation kinetics, together with monitoring predictive biomarkers, is essential to minimise the risk of TRT‑related myeloid neoplasms. Radiotoxicity can also stem from the intrinsic instability of RDCs, not only from physiological uptake in healthy tissues. When RDCs are labelled with radiometals, endogenous metal ions can replace the radionuclide through demetallisation. The released radioactive metals then redistribute freely in the body, leading to non‑specific radiation exposure [183]. This principle, however, can also be exploited therapeutically. Aminocarboxylic acid chelators with multiple carboxyl groups, such as Ca‑DTPA and Ca‑TTHA, have been shown to significantly enhance the excretion of free 225Ac, offering an effective strategy to reduce systemic radiation exposure [182]. Furthermore, radionuclides may also undergo transchelation, in which they are captured by endogenous metalloproteins to form new organometallic complexes. This process can lead to chelator exchange and potentially result in necrotic liver injury [80]. Therefore, developing more stable and selective metal‑chelator complexes is essential for limiting radionuclide release and off‑target toxicity. In addition to the primary toxicities discussed above, TRT may also induce other complications. For instance, radionuclide‑induced tumour lysis may precipitate carcinoid crisis in patients with NETs, characterised by severe flushing and diarrhoea. This can be mitigated by prophylactic octreotide and, where appropriate, debulking surgery [184]. Other reported complications include Gitelman‑like electrolyte disturbances and hypothyroidism [185, 186]. These diverse potential adverse effects highlight the need for comprehensive, individualised monitoring and management in patients undergoing TRT. Therefore, individualised dosimetry is likely to be key to optimising the efficacy and safety of TRT. By quantifying actual absorbed doses in tumours and healthy tissues through imaging, this approach enables precise adjustment of administered activity based on marked differences in tracer uptake. It also allows regimens to be tailored to interpatient variability in target expression, thereby maximising tumour radiation while sparing normal tissues. Innovations in drug delivery systems Traditional carriers Carrier engineering is central to next‑generation RDCs, governing tumour targeting, pharmacokinetics, and off‑target toxicity. Although mAbs are widely used, their clinical utility is hampered by prolonged blood circulation, suboptimal tumour penetration, and potential off‑target effects. Small molecules and peptide drugs, by contrast, are primarily constrained by nephrotoxicity and insufficient in vivo stability. An ideal carrier should combine high affinity, rapid clearance, deep tissue penetration, and high stability. Novel antibody derivatives are gaining momentum. SdAbs offer short circulation half‑lives and excellent tissue penetration, whereas bsAbs can reduce haematotoxicity through pretargeting strategies. Additionally, designed ankyrin repeat proteins (DARPins) are small, antibody‑mimetic scaffolds that combine rapid clearance with high thermal stability. Preclinically, 177Lu‑ or 212Pb‑labelled DARPins have shown promising tumour targeting and therapeutic efficacy [187]. However, their low molecular weight poses a risk of renal accumulation, which will require further optimisation through surface engineering of the DARPin scaffold. Peptide‑based RDCs have also evolved from cyclic to bicyclic formats. Bicyclic peptides retain the high permeability and excellent modifiability of their predecessors while acquiring antibody‑like affinity and selectivity [188]. Nevertheless, solubility and optimal cyclisation site selection remain key challenges. Furthermore, the potential of peptide multimerisation is exemplified by 177Lu‑labelled RGD trimers, particularly those enriched in tyrosine. They achieved superior tumour uptake with minimal non‑specific organ retention but face pronounced renal retention as a key pharmacokinetic hurdle to clinical translation as αvβ6‑targeted RDCs [189, 190]. Dual‑targeting and pretargeting The emergence of bsAbs has accelerated the clinical translation of dual‑targeting strategies and broadened the potential for combining TRT with other treatment modalities. Amivantamab, which targets both EGFR and MET, is already approved for EGFR‑mutated NSCLC [191]. Meanwhile, ivonescimab, a PD‑1×VEGF bsAb, combines immune checkpoint blockade with anti‑angiogenic therapy and has shown promising synergistic activity [192]. Applying such bsAbs to RDC development could steer TRT towards greater precision and mechanistic diversity. Two primary strategies are being explored for bsAb‑based delivery. The first is dual epitope or dual antigen recognition on tumour cells to enhance specificity. By engaging two distinct targets simultaneously, this design increases tumour binding avidity and reduces the risk of off‑target binding to healthy tissues. It may also facilitate receptor‑mediated internalisation, improving intracellular payload delivery. However, this strategy can be compromised by heterogeneous antigen expression, which may allow tumour cell escape. In addition, off‑target toxicity may occur if either antigen is also present at low levels in normal organs. The second strategy is pretargeting. In this approach, bsAbs are designed with one arm binding tumour antigens first and the other capturing subsequently administered radioligands. For example, the bsAb TF2 (anti‑CEACAM5 × anti‑HSG) is infused first and binds to tumours. After unbound TF2 clears, 177Lu‑IMP288 (a hapten peptide carrying two HSG residues) is administered. It binds with the prelocalised TF2 and kills tumour cells, while the excess peptide is rapidly excreted. This strategy minimises systemic radiation exposure by shortening radionuclide residence time in the body. This pretargeting strategy has been validated by 177Lu‑DOTA‑GD2‑SADA and 177Lu‑IMP288‑TF2, which achieved high TBR and reduced off‑target radiation exposure [66, 68]. Dual‑targeting and pretargeting can also be achieved without bsAbs, including PNA‑based pretargeting, dual‑ligand conjugation, and the integration of a hypoxia‑sensitive moiety into FAPI tracers. An unconventional pretargeting strategy employs peptide nucleic acid (PNA) technology. In this system, an antibody–PNA conjugate is administered first as the primary tumour‑targeting agent, followed by a radiolabelled complementary PNA probe that forms a stable hybrid at the tumour site [193]. Compared with bsAb‑based approaches, PNA probes offer faster blood clearance, higher TBR and lower immunogenicity, owing to their small size. For dual-targeting, Jin W et al. chemically conjugated a PSMA ligand with the SSTR2‑targeting peptide JR11 to generate dual‑targeted RDCs ([68Ga] Ga‑1 and [68Ga] Ga/[177Lu] Lu‑2) [194]. In another approach, a hypoxia‑sensitive 2‑nitroimidazole moiety (NI) was introduced into FAPI‑04, enabling simultaneous targeting of FAP‑expressing stroma and the hypoxic tumour microenvironment [195]. By enabling synergistic precision through dual‑targeting and improved pharmacokinetics through pretargeting, these strategies are advancing next‑generation RDCs. Emerging delivery systems: NPs and affibody Beyond continuously optimised traditional carriers, NPs offer a versatile platform with advantages including excellent stability, synergistic drug loading, radiosensitisation, and theranostic potential. Firstly, NPs can serve as stable nanocarriers that physically encapsulate radionuclides. For example, lanthanide phosphate nanoparticles (LnPO₄ NPs) retained approximately 90% of 225Ac daughter radionuclides such as 213Bi at the tumour site 24 h after administration, thereby reducing systemic toxicity caused by radionuclide redistribution [82]. Furthermore, emerging nanoliposome‑based delivery strategies can improve the intratumoural distribution of α emitters, thereby addressing irradiation heterogeneity [196]. Secondly, NPs can act as radiosensitizers. Certain metal‑based NPs effectively generate secondary ions and AE upon radiation exposure and catalyse ROS production, thereby amplifying radiation‑induced cytotoxicity [197]. Furthermore, some pH‑responsive NPs can improve tumour‑specific drug release and accumulation by exploiting the tumour microenvironment [198]. In addition, the multifunctional loading capacity of NPs enables combination therapy. For instance, HoIG‑Pt NPs co‑encapsulate cisplatin and 165Ho, resulting in antitumour efficacy superior to either agent alone [199]. Similarly, AFt‑Rh4 NPs induce immunogenic cell death and dendritic cell maturation, remodelling the tumour immune microenvironment to overcome multidrug resistance and suppress metastasis in NSCLC [200]. Finally, paramagnetic metal ions such as Gd3+ and Mn2+ confer MRI contrast properties. By altering the relaxation rates of surrounding water protons in response to changes in the tumour microenvironment, these agents enable real‑time visualisation of drug accumulation and release, thereby supporting image‑guided, personalised TRT [201–203]. Affibodies are an emerging class of non‑immunoglobulin scaffold proteins that hold promise as carriers owing to their high specificity, low immunogenicity, and rapid blood clearance [204]. Clinical development has so far focused mainly on diagnostic applications and production costs remain high. Continued advances in protein engineering and manufacturing may establish affibodies as key components of next‑generation RDCs. Novel chelating agents The selection and design of chelators also remain critical to next‑generation RDC development. Traditional macrocyclic chelators such as DOTA, NOTA, and their derivatives are widely used for complexing various metal ions. However, they exhibit slow complexation kinetics with larger lanthanide and actinide radionuclides, often requiring elevated temperatures [82, 205]. This factor limits their applicability to short‑lived nuclides and heat‑sensitive biological carriers. Acyclic chelators such as DTPA enable rapid labelling, but their insufficient kinetic inertness can lead to transchelation [206]. In recent years, a new generation of chelating agents, represented by crown chelators and bipyridine derivatives, has emerged with improved radiochemical performance. Building on traditional macrocyclic frameworks, crown chelators such as H2macropa and H4crown enable rapid and efficient 225Ac labelling under mild conditions (15 min, room temperature, low concentration) [207]. In parallel, novel nonmacrocyclic chelators are developed by introducing strong bidentate donor groups (e.g., picolinates) onto the terminal amines of the backbone to replace acetate groups. Classic symmetric designs (e.g., H₄py4pa, H₄noneunpa) maintain symmetric substitution at the terminal amines. In contrast, H₄noneunpaX adopts an inverted configuration, linking paired donor groups to the same nitrogen atom to form an asymmetric linear backbone. This structural design eliminates the need for protecting group strategies and facilitates the introduction of bifunctional chelating moieties. It thereby simplifies radiopharmaceutical synthesis and optimises pharmacokinetic properties, making it suitable for emerging radionuclides 225Ac/155Tb [208]. Within the acyclic class, bipyridine‑derived chelators have emerged as attractive candidates for divalent and trivalent metal ions. They combine the rapid complexation kinetics of acyclic systems with the high stability and inertness typical of macrocycles [209]. The rigidity of the bipyridine scaffold renders it mostly unaffected by the ionic radius of the metal, enabling the formation of structurally matched theranostic radionuclide pairs provided the metal ions share the same charge. Furthermore, its open cavity is well suited to accommodate large ions such as lanthanides and actinides, while the donor groups on the side arms can be modified to optimise coordination. For example, H2bispa2 (L1) features picolinate side arms, whereas H2bispox2 (L2) replaces them with 8‑hydroxyquinoline. This substitution retains the scaffold rigidity while forming stronger coordination bonds. The increased lipophilicity of L2 may also enhance tumour uptake and reduce non‑specific binding [210]. L1‑based RDCs targeting SSTR2 demonstrated antitumour efficacy and tumour selectivity in xenograft models [209]. On this basis, researchers replaced the side arm of L2 with terpyridine to generate L3 to further optimise coordination towards the daughter radionuclides (213Bi and 209Pb). However, it significantly reduced chelation efficiency for Ac3+, Lu3+, and La3+, highlighting the trade‑offs and complexity in conformational design. In contrast, L2 efficiently chelates a broad range of radionuclides under mild conditions, demonstrating its potential for TAT and multimodal theranostic applications [211]. While next‑generation chelators are being developed, established agents are also expanding their applications. For example, the octadentate ligand 3,2‑HOPO, originally designed for 227Th, has been demonstrated to effectively chelate 43Sc, 44Sc, and 47Sc [212]. Radionuclides Targeted alpha therapy (TAT) The choice of therapeutic radionuclide is central to balancing antitumour efficacy with toxicity and represents a key consideration in next‑generation RDC design. β emitters have long tissue penetration and a pronounced crossfire effect, but carry a higher risk of non‑specific radiation damage. In contrast, a single α particle offers higher LET and a shorter tissue range, inducing irreparable DNA double‑strand breaks [213]. Their cytotoxicity is mostly unaffected by oxygen‑mediated damage and cell cycle status, enabling potent direct killing. Their limited crossfire effect restricts uniform irradiation in large or heterogeneous tumours. However, α particles can trigger bystander effects, damaging adjacent non‑irradiated cells even at low absorbed doses and extending efficacy within heterogeneous regions [214]. Moreover, α particles can induce immunogenic cell death and elicit systemic antitumour immune responses via abscopal effects, providing a rationale for combination with ICIs [215]. Notably, α emitters retain cytotoxicity against hypoxic cells, overcoming a key mechanism of radioresistance, and have shown potential to reverse resistance to β emitters‑based TRT [88, 216]. These unique physical and biological properties position TAT as a particularly promising direction for next‑generation RDC development. Despite its therapeutic promise, the clinical translation of TAT faces several challenges. First, certain α emitters such as 225Ac and 227Th generate long‑lived daughter radionuclides whose in vivo redistribution can increase haematological toxicity. Moreover, the high‑energy nuclear recoil associated with α decay may disrupt chelator–metal bonds, leading to radionuclide release and systemic exposure [79]. Developing more robust delivery systems is therefore critical. Potential strategies include several approaches. Nanoparticle carriers can confine daughter nuclides. Novel crown and bipyridine chelators offer improved labelling efficiency and stability for α emitters. Aminocarboxylic acid chelators facilitate clearance of free radionuclides via reverse transmetallation [182]. Second, the short tissue range of α particles can result in heterogeneous dose distribution within tumours, limiting efficacy against large or heterogeneous lesions. Nanoliposome delivery strategies offer a promising solution. Studies have demonstrated that nanoliposomes can release highly diffusible 225Ac‑DOTA within the tumour stroma, enabling more uniform radionuclide distribution. Furthermore, their prolonged adhesion to the extracellular matrix also extends irradiation time, better suppressing solid tumour progression with reduced dosage [196]. Leveraging improving drug distribution, nanoliposomes can form complementary hybrid carriers with antibodies to achieve synergistic killing of both macroscopic tumour regions and microscopic infiltrates [217]. Despite these technical advances, key practical bottlenecks continue to impede the routine clinical translation of TAT. Personalised dosimetry remains a major challenge in TAT. Firstly, most therapeutic α emitters have low γ photon yields (often < 1%) and complex energy spectra, resulting in poor signal‑to‑noise ratios and limited quantitative reliability on SPECT/CT [218]. Furthermore, the few clinically feasible quantitative imaging approaches, such as bremsstrahlung SPECT, are either technically demanding or not widely available. Consequently, treatment planning for TAT often relies on empirical weight‑based dosing rather than truly absorbed‑dose‑guided individualisation, limiting the ability to optimise the therapeutic window for each patient. In recent years, however, 149Tb has emerged as a promising theranostic radionuclide for TAT owing to its unique decay properties. Its extremely short half‑life (4.1 h) and daughter‑free decay chain reduce the risk of systemic toxicity [219]. More importantly, 149Tb simultaneously emits α particles (3.97 MeV, 16.7%) for therapy and positrons (730 keV, 7.1%) for imaging, enabling real‑time PET tracking during therapy and supporting the development of true ‘Alpha‑PET’ radiotheranostics [220, 221]. Secondly, the high LET and short distance of α particles lead to highly heterogeneous dose deposition at the microscopic level, which conventional macroscale dosimetry models cannot accurately reflect. Encouragingly, microdosimetric approaches based on Monte Carlo simulations, combined with artificial intelligence (AI), are increasingly enabling precise modelling of nuclide distribution and energy deposition, paving the way for personalised TAT [222]. However, these methods are computationally intensive and rely on patient‑specific input data that are seldom obtainable. Their lack of standardisation and the difficulty of real‑time dose verification currently hinder widespread clinical adoption. AE emitters AE emitters demonstrate tremendous potential for precisely eliminating microscopic residual lesions, owing to their ultrashort range (< 500 nm) and ability to localise high‑energy deposition within the cell nucleus. Their high LET (4–26 keV/µm) confers a relative biological effectiveness comparable to that of α particles, enabling potent tumour cell killing [223]. However, the level of lethal DNA damage by AEs depends critically on their proximity to nuclear DNA [224]. Therefore, delivery systems must not only ensure tumour specificity but also enable efficient localisation of radionuclides into the nucleus. In response, several strategies are under investigation, including cell‑penetrating peptides, nuclear localisation signals (NLS), DNA intercalators (e.g., IUdR), and ligands targeting nuclear proteins (e.g., PARP1) [225–227]. Among these, neuropeptide Y (NPY) conjugates represent specialised carriers engineered for 161Tb, incorporating modifications like NLS to enhance nuclear delivery, though their metabolic stability requires further optimisation [228]. Currently, most research on AE emitters focuses on diagnostic performance and pharmacokinetics [229]. Although 125I‑HuBA‑1‑3D and 64Cu‑CB‑TE2A‑2PA‑NRA have been used to validate Delta‑like 1 homolog (DLK1) and neurotensin receptor 1 (NTSR1) as potential therapeutic targets in lung cancer, their therapeutic potential remains to be further evaluated [230, 231]. When selecting radionuclides, prioritise those that generate abundant AE with minimal photon emission. Excessive photons contribute to off‑target toxicity, whereas insufficient imageable photon emission compromises image quality [223]. However, pure AE emitters such as 119Sb are limited by complex production and radiochemical separation, as well as the lack of stable chelating systems. Chelators based on thiol donors are therefore a promising direction, given their high affinity for Sb(III) [232]. In addition, current dosimetric methods cannot reliably assess Auger electron energy deposition at the nanoscale, further hindering clinical translation. Auger electron therapy offers a novel approach to intracellular precision killing. Advancing nuclear delivery systems, simplifying production and radiolabelling of emitters such as 165Er and 119Sb, and establishing nanodosimetric methods are all crucial steps. These advances could position these radionuclides as key components of next‑generation RDCs for treating micrometastases and residual disease. Theranostics For novel RDCs to fully realise their clinical potential, theranostic integration is essential. Ideally, the same targeting ligand paired with matched diagnostic and therapeutic radionuclides would enable pre‑treatment target verification and patient stratification. It would also support individualised dosimetry based on tracer uptake and real‑time monitoring of target modulation during therapy. Post‑treatment assessment of dose delivery and residual disease could also be achieved. To enable radiotheranostics, an ideal pair of diagnostic and therapeutic RDCs should share comparable pharmacokinetic profiles while also being cost‑effective [233]. However, conventional strategies often employ different chelators for diagnostic and therapeutic radionuclides, leading to divergent biodistribution and compromising radiotheranostics. Widely used clinical pairs, such as 68Ga‑ and 177Lu‑labelled agents, often require different chelators and exhibit divergent biodistribution, which compromises the accuracy of dosimetry extrapolation and treatment guidance. These limitations have motivated the development of better‑matched theranostic radionuclide pairs and alternative chelator strategies. Using theranostic radionuclide pairs composed of different isotopes of the same element, such as 64Cu/67Cu, 44Sc/47Sc, and 155Tb/161Tb, can effectively circumvent this issue. These radionuclides emit both therapeutic β particles and γ photons or positrons suitable for imaging. Furthermore, using same-element radionuclide pairs enables PET/SPECT‑based dosimetry before treatment and real‑time monitoring during therapy. Among these, the 155Tb/161Tb has attracted particular attention. It shows greater potential than 67Cu or 47Sc for eradicating microscopic residual lesions due to its unique decay scheme, which combines medium‑energy β particles with AE and α particles [234]. Although same‑element theranostic pairs are conceptually ideal, their variety remains extremely limited. To overcome this constraint, an alternative strategy involves developing hybrid chelators capable of binding multiple radionuclides. For example, 3p‑C‑NETA combines a macrocyclic NODA backbone with flexible tridentate arms, enabling radiolabelling with diverse nuclides such as 18F, 161Tb, and 213Bi [235]. Another approach integrates two distinct chelators within a single construct. For instance, covalently linking DFO and DOTA via an L‑lysine unit forms the trifunctional chelator D2, enabling selective coordination of 89Zr and 177Lu [236]. This design achieves regioselective metal binding and expands the applicability of theranostic radionuclide pairs. Taken together, these theranostic advances support imaging‑guided TRT across its entire course. Before treatment, matched diagnostic scans help verify target expression and generate personalised dosimetry estimates. During therapy, serial imaging can track receptor occupancy and detect early signs of response or resistance. After treatment, post‑therapy scans enable absorbed‑dose verification and residual disease assessment. Future imaging tracers are expected to offer higher sensitivity for detecting microscopic residual disease and enable real‑time quantification of target dynamics. Their integration with artificial intelligence further allows automated dosimetry and response prediction. Multi‑isotope PET tracers that simultaneously map multiple targets may further refine patient selection and combination strategies. Production and regulation The future development of lung cancer RDCs needs to depend on a stable supply of therapeutic radionuclides and regulatory compliance. Currently, mainstream radionuclides such as 177Lu, 225Ac and 212Pb are produced mainly by nuclear reactors, cyclotrons, and generators. 177Lu is the most widely used therapeutic radionuclide and is produced mainly in nuclear reactors. However, global production capacity is increasingly strained by anticipated demand from expanding indications. Furthermore, both 177Lu and the emerging nuclide 161Tb require 176Yb as a target material, which is currently sourced almost exclusively from Russia, posing supply chain disruption risks [237]. For α emitters such as 225Ac, production is hampered by complex processes, the scarcity of precursor 229Th, and inconsistent yield and purity [46]. Alternative methods, such as the 226Ra(p,2n)225Ac reaction using cyclotrons, offer promise for large‑scale production with improved economic viability [238]. Furthermore, high‑energy accelerators and novel linear accelerators (e.g., Rhodotron) are also being deployed to scale up production of therapeutic nuclides, including 225Ac, 67Cu and 211At [239]. Notably, short‑half‑life radionuclides (e.g., 228Th/212Pb) necessitate more efficient transport and distribution networks. Local generator systems provide a practical solution by enabling regional production, reducing transport distance, and improving cost‑effectiveness [240]. At the regulatory level, RDCs face multiple challenges, including inconsistent global standards, uncoordinated approval processes leading to delayed market entry, and radioactive waste disposal [241, 242]. Addressing these issues will require promoting global regulatory harmonisation, along with continued efforts to secure radionuclide supply chains. These production and regulatory gaps translate into tangible clinical barriers. Limited production capacity can delay patient treatment and restrict clinical trial recruitment. Divergent national regulatory standards slow the initiation of multi‑centre studies and complicate the development of uniform treatment protocols. In addition, the requirement for dedicated radiopharmaceutical waste management and specialised on‑site infrastructure limits the number of centres capable of offering TRT, particularly in low‑resource regions. For short‑half‑life radionuclides, the reliance on local generator systems, while practical, demands technical expertise and quality control that are not yet universally available. Overcoming these translational hurdles is therefore essential to move novel RDCs from investigational platforms to widely accessible clinical options. Combination therapy TRT has shown considerable promise as a monotherapy for lung cancer. Moreover, TRT‑based combination therapy is essential for overcoming challenges such as tumour heterogeneity, acquired resistance, and immunosuppressive microenvironments. Combining TRT with other therapies enhances antitumour efficacy through radiosensitisation while also remodelling the tumour microenvironment and overcoming resistance, thereby comprehensively suppressing lung cancer progression and metastasis. Combination chemotherapy Chemotherapy, a cornerstone of lung cancer treatment, has shown synergistic effects when combined with TRT. Chemotherapeutic agents such as platinum compounds and etoposide enhance radiosensitivity by inducing DNA damage and apoptosis, while TRT provides targeted irradiation that can eliminate chemotherapy‑resistant tumour cells [243]. Therefore, RDCs based on low‑LET β emitters are often combined with drugs that interfere with DNA synthesis or integrity, including capecitabine, gemcitabine and others [83, 244, 245]. In contrast, high‑LET α emitters induce irreparable DNA double‑strand breaks, enabling direct killing of hypoxic cells. This mechanism extends the range of synergistically compatible chemotherapeutics [246]. Tumours resistant to radiotherapy and chemotherapy are often characterised by necrotic areas adjacent to hypoxic regions. Hypoxic cells are recognised as key drivers of relapse, progression and therapy resistance in solid tumours [161]. Combining TAT with chemotherapy therefore offers a theoretical strategy to overcome such resistance. By employing carriers for RDCs that target necrotic tumour zones, chemotherapy‑induced necrosis may provide additional binding sites for TAT, enhancing radionuclide accumulation. Simultaneously, TAT directed to necrotic areas can also eliminate adjacent hypoxic cells, countering resistance while amplifying efficacy. For example, 227Th‑DAB4 combined with chemotherapy effectively controlled chemotherapy‑resistant LL2 tumours in mice, with higher radiation doses deposited near hypoxic cells within necrotic regions [161]. These preclinical findings accelerate the advancement of TAT–chemotherapy combinations into clinical trials for lung cancer. Combination targeted therapy Similarly, targeted therapies that are now integral to standard‑of‑care lung cancer treatment, such as EGFR and poly(ADP‑ribose) polymerase (PARP) inhibitors, also can act as radiosensitisers by inducing DNA damage [145]. Although EGFR inhibitors often show initial efficacy, acquired resistance remains a major obstacle to long‑term clinical benefit. EGFR‑targeting RDCs, formed by conjugating agents such as cetuximab with radionuclides, induce greater DNA damage than cetuximab alone. This is accompanied by RUNX3 activation, enhancing tumour suppression and potentially counteracting resistance [88]. In addition, the combination of 177Lu‑DOTATOC with PARP inhibitors has demonstrated synergistic antitumour activity and favourable safety profiles in preclinical SCLC models [247]. Several clinical trials are currently evaluating SSTR2‑targeting RDCs in combination with PARP inhibitors in patients with NETs of the lung (NCT05870423, NCT05053854). Combination immunotherapy ICIs are now a critical therapeutic modality in lung cancer. TRT can enhance their efficacy through immunomodulatory effects, such as promoting chemokine expression, upregulating co‑stimulatory molecules, and inducing immunogenic cell death [248]. Preclinically, 225Ac‑DOTA‑4AH29 upregulated PD‑1, PD‑L1, and LAG‑3, reducing immunosuppression [59]. Moreover, 211At‑SAB‑Oct increased calreticulin and MHC‑I expression, modulating the tumour immune microenvironment and activating endogenous antitumour immunity [103]. A phase I/II trial (NCT03325816) has confirmed the feasibility of combining 177Lu‑DOTATATE with nivolumab [120]. Notably, PD‑L1 expression is dynamic and can be monitored by SPECT/CT or PET/CT, supporting the development of PD‑L1‑targeting RDCs such as 177Lu‑DOTA‑PD‑L1‑i and 225Ac‑HEHA‑PD‑L1‑i [90]. These RDCs offer both theranostic potential and a pathway toward image‑guided, personalised immunotherapy combinations. Isotope combination Combining isotopes, concurrent or sequential administration of radionuclides with distinct physical properties, offers unique advantages. Rational pairing of nuclides with different physical properties may improve dose coverage and therapeutic outcomes across lesions of varying size and distribution. The combination of 90Y and 177Lu demonstrated significantly prolonged survival compared with 90Y monotherapy, without a significant increase in systemic toxicity during sequential administration [249, 250]. 90Y‑DOTATOC is being used less frequently due to nephrotoxicity, while 177Lu‑DOTATATE has become more prevalent. Future studies should systematically compare the efficacy differences between 177Lu‑DOTATATE monotherapy and various isotope combinations in lung cancer. Additionally, sequential isotope use, particularly alternating α emitters and β emitters, represents greater developmental potential. A first-in-human study documented effective disease control with 213Bi‑DOTATOC in 7 patients with advanced NETs refractory to β radiation (90Y/177Lu‑DOTATOC), with only mild haematotoxicity [50]. Combined or sequential α/β radionuclide therapy holds broad translational potential in lung cancer, especially for overcoming resistance and eradicating residual disease. Further clinical investigations are warranted to systematically evaluate the feasibility, optimal sequences, and adverse reactions of isotope combinations in lung cancer. Other combination therapies Beyond these strategies, additional combination approaches are broadening the scope of TRT in lung cancer. Radiation‑induced DNA damage activates PARP‑1 and depletes nicotinamide adenine dinucleotide (NAD) [251]. Therefore, using RDCs to inhibit nicotinamide phosphoribosyltransferase (NAMPT), which participates in NAD metabolism, represents a viable strategy for radiosensitisation [252]. Low‑dose TRT combined with CAR‑T cell therapy also has shown potential in treating metastatic solid tumours. Studies indicate that 1–2 Gy of β‑radiation can effectively stimulate immune activation while minimising impairment of CAR‑T cell function [253]. However, the underlying synergistic mechanisms and clinical benefits warrant further investigation. Additionally, proteolysis‑targeting chimaeras (PROTACs) induce targeted protein degradation by recruiting E3 ubiquitin ligases, but their clinical potential is limited by poor pharmacokinetics and off‑target toxicity [254]. PROTAC‑mediated degradation of BRD4 can downregulate RAD51AP1, a protein associated with radiosensitivity, consequently enhancing TRT efficacy. Concurrently, TRT can offset the pharmacokinetic limitations of PROTACs [255]. This bidirectional synergy allows dose reduction of both agents, thereby limiting off‑target toxicity. Although this strategy was demonstrated in preclinical prostate cancer models, further studies are needed to extend its application to lung cancer and advance its clinical evaluation. In terms of clinical translation, several trials (NCT05142696, NCT05595460) are assessing the safety and preliminary efficacy of 177Lu‑DOTATATE combined with standard‑of‑care regimens (such as carboplatin, etoposide, and atezolizumab) to advance RDCs into first‑line therapy. Concurrently, research is actively exploring treatment sequencing and indication expansion. A retrospective study reported that neoadjuvant TRT enabled curative surgery in 26.3% of patients with locally advanced NETs [256]. Furthermore, ongoing research continues to evaluate RDCs for alleviating secondary endocrine symptoms associated with NETs (e.g., hyperreninaemia, hypokalaemia) and its application in specific subtypes, including pulmonary carcinoids [257, 258]. However, co‑administered drugs can alter RDC biodistribution through pharmacokinetic and pharmacodynamic interactions. Highly protein‑bound chemotherapeutics may displace RDCs from plasma proteins, transiently raising the free radionuclide fraction and shifting organ exposure. Anti‑angiogenic agents modify tumour vascular permeability. Depending on the degree of vessel normalisation, they may either improve or reduce RDC delivery. Some immunotherapy agents upregulate or downregulate target receptors on tumour cells, directly affecting radionuclide targeting. At the renal clearance level, co‑administered agents can compete for tubular transport pathways, changing the excretion rate of peptide‑based RDCs and influencing both nephrotoxicity risk and systemic residence time. Free radiometal release may also be influenced by compounds that chelate metal ions or alter redox balance. These multifaceted interactions highlight the importance of incorporating biodistribution monitoring and pharmacokinetic profiling into the design of future TRT combination trials. While combination therapy shows promise in clinical settings, further research is needed to address key issues like systemic toxicity, optimal sequencing, and personalised dosing. Moreover, because most research has focused on NETs of the lung, the findings need to be applied more broadly to lung cancer. Conclusion RDCs, as the enabling agents of TRT, represent a profound transformation in radiotheranostics for lung cancer. Currently, several RDCs have entered early‑phase clinical trials for lung cancer, with 11 completed. Early‑stage studies focused on 131I/90Y‑labelled antibodies but were limited by haematotoxicity and immunogenicity. Recent efforts have shifted toward 177Lu‑labelled SSTR‑targeting peptides and emerging antagonists, which show improved efficacy but variable haematotoxicity. However, clinical translation is still constrained by dose‑limiting toxicity, inadequate radionuclide supply, lack of global regulatory harmonisation, and limited clinical evidence. The development of next‑generation RDCs hinges on a multidimensional innovation framework. It mainly centres on three key directions: radionuclide selection, drug delivery and targeting, and combination therapy. Radionuclide selection is governed by their decay properties. β emitters, with their longer tissue penetration range and crossfire effect, are suitable for larger tumours. α emitters, by contrast, deliver high energy over a short range and retain activity in hypoxic regions, rendering them ideal for micrometastases and overcoming resistance. AE emitters offer potential for intranuclear killing of minimal residual disease. Notably, mixed‑decay radionuclides like 161Tb, combining α, β and AE emission, are emerging as promising multifunctional isotopes. Equally central to progress are advanced drug delivery systems. This begins with the exploration of lung cancer‑specific targets (e.g., SSTR, FAP, EGFR, and DLL3) and novel carriers with optimised pharmacokinetics (e.g., sdAbs, bsAbs, and NPs). Alongside carrier engineering, more stable chelators (e.g., bipyridine derivatives) are being developed to improve stability. Additional targeting advances include pretargeting strategies to reduce exposure of normal tissues and dual‑targeting carriers to broaden applicability and address tumour heterogeneity. Matching the half‑life of radionuclides to the pharmacokinetics of carriers is equally important to ensure sufficient time for production and delivery while minimising systemic toxicity from prolonged retention. Ultimately, personalised dosimetry platforms integrating multimodal imaging and AI are also crucial to translate these delivery advances into precise, individualised administration. Combination strategies represent the third major direction. These approaches are being actively explored to overcome tumour heterogeneity, acquired resistance and immunosuppressive microenvironments. RDC‑based combinations with chemotherapy, targeted therapy, and immunotherapy are being evaluated in clinical settings to enhance radiosensitisation and antitumour efficacy. However, more large‑scale, multicentre, prospective trials designed for lung cancer are required to validate the efficacy and safety of RDC monotherapy and combination therapy. Furthermore, concurrent or sequential regimens of dual isotopes provide a promising strategy to cover lesions of varying sizes and distributions. Only through sustained innovation and rigorous clinical translation can RDCs realise their potential to improve long‑term outcomes in lung cancer. Supplementary Information Supplementary Material 1 Acknowledgements Not applicable. Abbreviations Auger electron Adverse events Artificial intelligence Antibody radionuclide conjugates Bifunctional chelating agents Bispecific antibody Combretastatin‑A4‑phosphate Cancer‑associated fibroblasts Carbonic anhydrase IX Cholecystokinin‑2 receptor Cluster of differentiation Cadherin‑3 (P‑cadherin) Carcinoembryonic antigen C‑X‑C motif chemokine ligand 12 C‑X‑C chemokine receptor type 4 Designed ankyrin repeat protein Disease control rate Disease control survival Delta‑like ligand 3 Evans Blue dye Event‑free survival Epithelial glycoprotein‑1 Epidermal growth factor receptor Extensive‑stage small cell lung cancer Fibroblast activation protein Fibroblast activation protein inhibitor Disialoganglioside GD2 Gastroenteropancreatic neuroendocrine tumours Glomerular filtration rate G protein‑coupled receptors Hyaluronic acid Human anti‑chimeric antibody Human anti‑human antibody Human anti‑mouse antibody Hydroxyapatite‑zirconium nanoparticle conjugated with cetuximab Immune checkpoint inhibitors Large‑cell neuroendocrine carcinoma Linear energy transfer Monoclonal antibody Major histocompatibility complex class I Magnetic nanoparticle Magnetic resonance imaging Mesothelin Maximum tolerated dose Nicotinamide adenine dinucleotide Nicotinamide phosphoribosyltransferase Neural cell adhesion molecule 1 Neuroendocrine neoplasms Neuroendocrine tumours Asparagine‑glycine‑arginine 2‑nitroimidazole Nuclear localisation signal Nanoparticles Neuropeptide Y Non‑small cell lung cancer Neurotensin receptor 1 Objective response rate Overall survival Poly (ADP‑ribose) polymerase Programmed death‑1 Programmed death‑ligand 1 Polyethylene glycol Positron emission tomography Progression‑free survival Pharmacokinetic modifier linker Peptide nucleic acid Peptide radionuclide conjugates Peptide radionuclide‑drug conjugates Proteolysis‑targeting chimaeras Prostate‑specific membrane antigen Radionuclide‑drug conjugate Arginine‑glycine‑aspartic acid Radioimmunoconjugate Roundabout guidance receptor 1 Reactive oxygen species Self‑assembling and disassembling Small cell lung cancer Single‑domain antibody Small‑molecule radionuclide conjugates Single photon emission computed tomography Somatostatin receptor Somatostatin receptor subtype 2 Targeted alpha therapy Tumour‑to‑background ratio Tyrosine kinase inhibitors Tumour necrosis therapy Thymoquinone glucuronide Targeted radionuclide therapy Vascular endothelial growth factor Vascular endothelial growth factor receptor Authors’ contributions J.L. and X.Z. contributed equally to this manuscript. Conceptualisation: J.L., X.Z., X.Y., J.X.; investigation: J.L., X.Z.; visualisation: J.L., W.Y.; supervision: J.X.; writing original draft: J.L., X.Z.; writing-review & editing: J.L., X.Y., L.G., J.X.; all authors read and approved the final manuscript. Funding This work was supported by the National Science Foundation of China (Grant No. 82373021); the Science and Technology Department of Sichuan Province (2025ZNSFSC0045); the Natural Science Foundation of Sichuan Province (2026NSFSCZY0147) and the 1.3.5 Project for Disciplines of Excellence (ZYYC23010, ZYYC23008) from West China Hospital. Data availability All data supporting the findings of this study are available within the paper and its Supplementary Information. Declarations Ethics approval and consent to participate Not applicable. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Junzhi Liu and Xiaoyu Zuo contributed equally to this work. References Associated Data Supplementary Materials Supplementary Material 1 Data Availability Statement All data supporting the findings of this study are available within the paper and its Supplementary Information.