Overview

Cancer emerges as a worldwide health emergency, standing as a primary cause of both mortality and morbidity across the human population (Ferlay 2020). According to the World Health Organization (WHO), in 2020, there were approximately 18.1 million new cancer cases, resulting in 10 million deaths. The WHO predicts a staggering increase in cancer cases over the next two decades, estimating around 28 million cases by 2040. Furthermore, projections indicate 16.3 million deaths, marking a 63.7% increase from the statistics recorded in 2020 (Sung et al. 2021). Thus, these findings and estimates highlight the urgent need for effective cancer treatment strategies. A range of treatment modalities, such as surgery, radiation therapy (RT), chemotherapy, immunotherapy, targeted therapy, stem cell transplantation, natural antioxidants, nanoparticles, ablation therapy, radionics, chemodynamic therapy, ferroptosis and sonodynamic therapy, as well as multidisciplinary approaches, are utilized in the battle against cancer (Debela et al. 2021; Pucci et al. 2019; Arruebo et al. 2011; Moo et al. 2018). Among these treatments, RT is a potent method, especially when combined with other treatment modalities. In principal, RT should be effective against all types of tumor cells. However, individual cells within tumors exhibit varying degrees of sensitivity to radiation, resulting in diverse treatment outcomes (Wang et al. 2018; Baskar et al. 2012).

Biological responses triggered by radiation exposure depend on energy deposition per unit length of radiation track (keV/μm) known as linear energy transfer (LET), along with the cellular and tissue water activity and the composition of the extracellular matrix (ECM). Low-LET radiation (γ-rays and X-rays) is termed "sparsely ionizing" because it causes scattered ionizing events in various directions across space. In contrast, high-LET radiation, exemplified by protons (≤ 50 keV/µm), carbon ions (C-ions) (≤ 200 keV/µm), and alpha-emitting radionuclides (60 to 110 keV/μm), is referred to as "densely ionizing" because its ionizing effect predominantly occurs in the narrow region at the end of the primary path (Bragg peak). This results in more complex and clustered DNA damage, characterized by the presence of more than two lesions per DNA helical turn, including strand breaks, base damage, and abasic sites (Mladenova, et al. 2022; Timm et al. 2018; Kruijff et al. 2015). This feature contributes to its greater relative biological effectiveness (RBE) in causing DNA damage and cell death (Paganetti et al. 2002; Oden et al. 2017; Ray et al. 2018; Busato et al. 2022). Modern therapy techniques utilize this principle by employing spread-out Bragg peak (SOBP) (Lin et al. 2018; Akagi et al. 2023; Darafsheh et al. 2021). This approach involves employing multiple charged particle beams with partly overlapping bragg peaks, allowing for the precise delivery of effective doses to the tumor in a three-dimensional fashion. This method reduces harm to healthy tissue situated before and after the tumor, in contrast to conventional low-LET photon therapy (γ-rays and X-rays) (Ray et al. 2018).

High-LET radiation induces dense ionization events that result in a substantial formation of complex DNA damage including lethal lesions such as base alterations, single-strand breaks (SSBs), apurinic/apyrimidinic site (AP sites), double-strand breaks (DSBs), and chromosomal aberrations (Nikjoo et al. 1998; Watanabe et al. 2015; Wilkinson et al. 2023). These lesions, collectively referred to as "clustered lesions," occur in very close proximity, typically within a few nanometers (Hada and Georgakilas 2008; Sage and Shikazono 2017). Among all types of damage, clustered lesions present a significant challenge for DNA repair machinery of the cell and explain why high-LET ionizing radiation has higher RBE and highly effective, despite not necessarily causing a greater overall number of DSBs compared to low-LET radiation (Georgakilas et al. 2013; Lorat et al. 2015; Lorat et al. 2016).

Nevertheless, despite progress in radiation techniques and treatment strategies, the emergence of radiation resistance remains a significant obstacle to enhancing treatment outcomes (Busato et al. 2022; Barcellini 2022). Radiation resistance can arise from several sources, including tumor heterogeneity (differences between tumors of identical types in distinct patients, or difference among cancerous cells within a singular tumor, or difference between a primary and a secondary tumor cells), the tumor microenvironment, TLS, RGS, EMT, and epigenetic alterations within cancer cells (Dagogo-Jack and Shaw 2018). Understanding the mechanisms that drive the development of resistance is essential for surmounting this obstacle in both traditional and novel forms of high-LET RT (Sato et al. 2019). Here, we discuss radiation resistance strategies, repair approaches for clustered DNA lesions, and the possible potential for mutagenesis against clustered DNA lesions. Furthermore, we suggest that clinicians exercise the caution when deciding to use high-LET RT as a treatment for tumors instead of low-LET RT, as there may high risk of tumor resistance developing in the medium to longer term.

Role of DNA damage repair and other molecular mechanisms in radiation resistance against low-LET radiation

Ionizing radiation (IR) exerts a direct impact on the structure of DNA by introducing DNA breaks, specifically DSBs. Additionally, it triggers secondary indirect effects such as the production of reactive oxygen species (ROS), which in turn oxidize proteins and lipids. These ROS also contribute to various forms of DNA damage, including the creation of abasic sites and SSBs (Borrego-Soto et al. 2015). Out of the diverse types of DNA damage caused by radiation therapy, DSBs are particularly harmful (Huang and Zhou 2020; Deng, et al. 2022). There are four primary DNA repair pathways: recombination repair, mismatch repair (MMR), base excision repair (BER), and nucleotide excision repair (NER). Recombination repair deals SSBs and DSBs, while mismatch repair fixes replication errors. BER addresses oxidation-induced base damage and removes AP sites, while NER corrects bulky adducts and cross-linked DNA damage.

Among these, the DSBs repair pathways, namely, nonhomologous end joining (NHEJ) involving DNA-PKs and Ku heteroduplex proteins and homologous recombination (HR) catalyzed by Rad51 and regulated by ATM and ATR kinases, are vital for mending radiation-induced DNA breaks (Dietlein et al. 2014; Dietlein and Reinhardt 2014; Visnes et al. 2018; Zhu et al. 2009). In response to DSBs caused by low-LET radiation, sensor proteins such as Rad24p (Aylon and Kupiec 2003), NBS1/hMRE11/hRAD50 complex (Zhou and Paull 2013), γH2AX (Georgoulis, et al. 2017), Ku (Abbasi et al. 2021), MDC1 (Ruff et al. 2020), and 53BP1 (Noordermeer et al. 2018), initiate downstream signaling responses mediated by sensor kinases (DNA-PKs, ATM, and ATR) (Mladenov et al. 2019a; Mladenov et al. 2019b; Li, et al. 2021; Smith et al. 2010; Goodarzi et al. 2003). Together, this leads to an initiation of the DNA damage response (DDR) triggers a cascade of complex reactions, encompassing cell cycle arrest and DNA repair processes (Marechal and Zou 2013; Cuadrado et al. 2006; Weber and Ryan 2015). In response to low-LET RT, the expression levels of different DNA damage sensors (γH2AX, 53BP1, NBS1, BRCA1/2, and Ku) have the potential to act as predictive biomarkers for assessing the outcomes of RT in cancer patients (Liu, et al. 2020; Stover et al. 2016). Replication protein A (RPA) is a crucial eukaryotic protein involved in binding to ssDNA and has roles in DNA replication, DNA damage repair, replication stress, and cell cycle regulation. RPA may contribute to radiation resistance by utilizing the ATR/Chk1 pathway to alleviate replication stress (Glanzer et al. 2014; Zou et al. 2006). Inhibition of RPA negatively affects the time bound damaged DNA repair, rendering cells radiation-sensitive. XRCC1 (X-ray repair cross-complementing 1), a key BER protein stands out as a significant predictive biomarker for radiation therapy, and mutations within the XRCC1 gene have the potential to heighten sensitivity to ionizing radiation (Hanssen-Bauer et al. 2012; Caldecott 2019; Niu et al. 2013; Eckelmann, et al. 2020). Additionally, p53 gene (a tumor suppressor gene) expression status in cancerous cells profoundly influences the efficacy of RT (Kong et al. 2021).

In addition to DDR mechanisms, poly(ADP-ribosyl)ation (PARylation) and replication stress management are key fundamental processes that play crucial roles in safeguard of genome integrity against DNA damage. PARylation is a posttranslational modification that occurs in response to DNA damage, particularly in nuclear proteins such as histones. It serves to enhance the survival of damaged proliferating cells (Kamaletdinova et al. 2019). PAR is produced from NAD + through the action of Poly (ADP-ribose) Polymerases (PARPs), a group of 18 proteins encoded by different genes (Alemasova and Lavrik 2019). These PARPs are linked to promoting tolerance to RT (Seyedin et al. 2020; Higuchi et al. 2015). After exposure to ionizing radiation, PARP-1 activates the AMPK/mTOR pathway, leading to processes such as autophagy and reduced sensitivity to radiation (Bi et al. 2018; Rose et al. 2020).

Replication stress can arise from various sources, including genome integrity perturbation by clustered DNA damage, conflicts between replication and transcription machinery, and oncogenic stress (Nickoloff et al. 2022). Certain difficult to repair clustered DNA damages activate DNA damage tolerance (DDT) pathways, where specialized polymerases like REV1, POL ζ, POL η, POL κ, and POL i bypass damage to facilitate continuous replication and radioresistance with possibilities to introduce mutations (Chatterjee and Walker 2017). Furthermore, if these underlying issues, such as DNA damage, are not promptly resolved, nucleases facilitate the restart of the fork. Several nucleases, including MUS81 (Marini et al. 2023; Chen et al. 2021), FEN1 (Zhang et al. 2022a; Vaitsiankova et al. 2022; Yang et al. 2022a), EEPD1 (Wu et al. 2015), Artemis (Betous et al. 2018), Metnase, EXO1 (Tomimatsu et al. 2017), DNA2-BLM, MRE11 (Vertemara and Tisi 2023), CtIP (Mozaffari et al. 2021), SLX1-SLX4 (Wyatt et al. 2013; Sarbajna et al. 2014), XPG (Trego et al. 2016), and XPF-ERCC1-SLX4 (Betous et al. 2018; Xu et al. 2021) are implicated in replication stress management and play roles in promoting the repair and restart of replication forks (Nickoloff et al. 2022).

Modulating the cell cycle regulation and distribution by eukaryotic cells is another crucial adaptation for the protection of genome integrity against DNA damage (Morales-Ramirez et al. 2022; Jackson 2002). Upon detecting difficult to repair DNA damage cells may either undergo apoptosis or enter a state of senescence. Conversely, when damage is reparable, cells prolong their cell cycle to facilitate DNA repair and accomplish it. DNA damage induces cell cycle arrest, particularly through the activation of G2/M checkpoints, which are orchestrated by ATM and ATR (Marechal and Zou 2013; Cuadrado et al. 2006; Barnaba and LaRocque 2021; Pai, et al. 2021). These checkpoints play pivotal roles in preventing cells from entering the mitotic phase when DNA integrity is compromised.

Moreover, resistance to low-LET radiotherapy in cancer involves various pathways, such as autophagy (Chaachouay et al. 2011; Chen et al. 2015; Chen et al. 2011; Khan et al. 2020), the NF‐κB pathway (Deorukhkar and Krishnan 2010; Li and Sethi 2010), Akt/cyclin D1/CDK4 signaling pathway (Shimura 2017; Shimura et al. 2011; Shimura et al. 2012), and the Wnt/β‐catenin pathway (Yu et al. 2021; Liu et al. 2022). Furthermore, factors contributing to this resistance encompass tumor genetics and epigenetics, tumor metabolism, the tumor microenvironment, and the presence of nonmalignant cells, including fibroblast-associated cancer cells, tumor-infiltrating lymphocytes, endothelial cells, macrophage-associated cancer cells, and cancer stem cells (Busato et al. 2022; Rycaj and Tang 2014; Peitzsch et al. 2019; Huang et al. 2023; Wu et al. 2023a) (for a more comprehensive understanding, readers are encouraged to consult the cited references). Thus, in a wider perspective, cancer cells have the capability to build resistance against traditional low-LET RT using various mechanisms.

High-LET radiation causes difficult to repair clustered DNA damage

High-LET radiation produces clustered DNA damage resulting in complex clustered DNA lesion sites (CDS) or multiple damage sites (MDS). These sites include clustered DSBs together with base lesions or AP sites, as well as clusters of non-DSB damage comprising base lesions and ssDNA breaks (Sage and Shikazono 2017; Georgakilas et al. 2013; Goodhead and Nikjoo 1989; Sage and Harrison 2011; Eccles et al. 2011; Sai et al. 2023; Bukowska and Karwowski 2018). At the chromosomal level, chromatin compaction reduces the incidence of DNA damage caused by ionizing radiation (Tang et al. 2019). Nevertheless, repair of high-LET-induced heterochromatic region DSBs take substantial period while DSBs caused by low-LET ionizing radiation are promptly restored within 24 h, irrespective of chromatin compaction status (Lorat et al. 2015; Roobol, et al. 2020). The DNA damages caused by low-LET radiation usually dispersed throughout the genome, and repair mechanisms are highly efficient at addressing such damage (Chatterjee and Walker 2017; Tian et al. 2015). However, repair of CDS and MDS showed slower repair kinetics, resulting in inaccurate repair and chromothripsis (a solitary catastrophic incident results in the extensive reorganization of genetic material within one or few number of chromosomes), or sometimes not at all repaired (Sage and Shikazono 2017; Cortes-Ciriano et al. 2023; Voronina et al. 2020; Kozmin, et al. 2021; Danforth et al. 2022). This distinction is a pivotal reason why high-LET carbon ions exhibit a 2–threefold greater RBE than low-LET ionizing radiation (Okayasu 2012; Allen et al. 2011; Tsujii et al. 2004; Okada et al. 2010; Paganetti 2022). From this viewpoint, the heightened cytotoxicity of high-LET ionizing radiation leverages the innate shortcomings of DNA repair systems when confronted with clustered lesions. This limitation attributed to the absence of natural selective pressure that drives the development of cellular repair systems capable of efficiently repairing clustered DNA damage (Lorat et al. 2015; Asaithamby and Chen 2011). Furthermore, from a clinical perspective, the efficacy of carbon-ion radiation therapy (CIRT) relies less on oxygen compared to low-LET RT (Antonovic et al. 2014; Strigari et al. 2018). As a result, CIRT presents as an efficient therapeutic option for hypoxic tumors, found in melanoma, head and neck cancer, and pancreatic cancer, which typically demonstrate resistance to traditional radiotherapy. (Allen et al. 2011; Kamada et al. 2015; Tinganelli and Durante 2020).

Clustered complex DNA damage can be broadly divided into two primary categories: DSBs and non-DSB oxidative clustered DNA lesions (OCDLs) (Hada and Georgakilas 2008). OCDLs can be further classified into two categories: bistranded clustered lesions (where multiple lesions are closely located on both DNA strands) and tandem clustered lesions (where multiple lesions are closely positioned on the same DNA strand). Several factors influence the repair phenomenon of clustered complex damage, including the type of lesion, lesion proximity, and lesion orientation. Synthetic oligonucleotides have been employed to study OCDL repair with known DNA lesions using mammalian nuclear or whole-cell extracts and purified BER proteins (Bellon et al. 2009; Kazmierczak-Baranska, et al. 2021; Mourgues et al. 2007; Lomax et al. 2004a; Harrison et al. 1999; Eot-Houllier et al. 2005; Boguszewska, et al. 2021; Karwowski 2019, 2021). The repair efficiency of bistranded OCDLs decreases when lesions are within five bases of an AP site or SSB (Mourgues et al. 2007; Lomax et al. 2004b). Lesions such as 8-oxoG, thymine glycol (Tg), and 5,6-dihydrothymine (DHT) reduce repair efficiency and may generate DSBs by replication (Bellon et al. 2009; Lomax et al. 2004a). Repair mechanisms also differ depending on the orientation of the bistranded lesions. In cases where lesions are in the negative orientation (with the opposing strand lesion positioned 5′ to the base opposite the reference lesion), both long and short patch BER processes are engaged. Conversely, in the positive orientation (where the opposing strand lesion is positioned 3′ to the base opposite the reference lesion), short-patch BER is the dominant mechanism (Harrison et al. 1999; Eot-Houllier et al. 2005; Paap et al. 2008). The presence of nearby lesions interferes with the initial phase of the BER pathway, which involves the excision of damaged bases by DNA glycosylases. For instance, an AP site or SSB on opposite DNA strand can significantly hinder the excision of base lesions (Georgakilas et al. 2004). The rate of base lesion excision is compromised when another lesion is nearby, mainly due to interference with glycosylase binding (Harrison et al. 1999; Eot-Houllier et al. 2005; Paap et al. 2008; Singh and Das 2013). Tandem OCDLs, also exhibit compromised repair, similar to bistranded clusters (Venkhataraman et al. 2001; Jiang et al. 2009).

To deal with DSBs generated directly by high-LET radiation tracks or DSBs converted by OCDLs, error-prone processes surpass high-fidelity homologous recombination (HR) repair (Mladenova, et al. 2022). Moreover, increased DNA end resection and DNA binding characteristics of Mre11 and Ku are crucial for repair pathway decisions for complex lesions (Yajima et al. 2013; Averbeck et al. 2014; Hays et al. 2020; Wang et al. 2010). Additionally, it has been suggested that the ATM protein, partners with the Artemis repair protein, plays a pivotal role in addressing complex and "dirty / clustered" DSBs introduced by high-LET radiation (Lobrich and Jeggo 2005; Xue et al. 2009). The results also indicate that ATM reacts differently to DSBs caused by high-LET radiation than to those induced by low-LET radiation. For instance, high-LET radiation significantly reduces ATM-specific phosphorylation of pATF2 and pSMC1. In contrast, the absence of ATM results in a noticeable decrease in the phosphorylation of DNA-PKC at Thr-2609 and adversely impacts the formation of Rad51 foci (Okayasu 2012; Whalen et al. 2008). Together, these findings suggest that the presence of clustered DNA lesions, whether bistranded or tandem, or DSBs, can hinder and/or alter the repair machinery's approach to repairing them, potentially leading to an increased lifetime of these lesions or conversion to more severe DNA lesions such as DSB, which poses a difficult task for the repair machinery.

Clustered DNA damage implicated by high-LET radiation has high mutagenic potential

The biological ramifications of OCDLs in mammalian cells are complex and influenced by various factors. The damage caused by OCDLs is primarily repaired through BER pathway (Bellon et al. 2009; Mourgues et al. 2007; Byrne et al. 2009; Cunniffe et al. 2014a), and studies have revealed that specific glycosylases, such as MutY, may determine the mutagenic outcome of clustered base damage (Bellon et al. 2009; Kazmierczak-Baranska, et al. 2021; Byrne et al. 2009; Shikazono et al. 2006). Furthermore, bigger risks arise when bistranded / tandem cluster lesions are converted to DSBs. Studies indicate that levels of repair enzymes, DNA replication status, density of non-DSB damage, and the complexities of the BER pathway may dictate whether these lesions are converted into DSBs or undergo repair without such conversion (Sage and Shikazono 2017; Nikitaki et al. 2016). A firefly luciferase reporter assay uncovered that bistranded clusters featuring opposing furans could transition into DSBs within mammalian cells, with the predominant activity observed from the class II AP endonuclease targeting these lesions. In mammalian cells, other bistranded clusters, such as uracil or 8-oxoG, are not converted to DSBs (Malyarchuk et al. 2009). Additionally, there is only scant evidence suggesting that clustered uracil lesions may be converted into DSBs in mammalian cells (Malyarchuk and Harrison 2005). Why bistranded furans clusters are converted to DSBs more frequently than uracil or 8-oxoG lesions is explained by the concept of “passing the baton”. This involves the cooperation of short BER pathway enzymes, such as DNA glycosylase, XRCC1, AP endonuclease (Ape1), DNA pol β, and DNA ligase 1, which interact to transfer repair intermediates to subsequent repair enzymes, preventing the premature release of repair intermediates. However, disruptions to this seamless transfer, or "passing the baton", can occur when dealing with damage types such as clustered furans that necessitate the involvement of long BER pathway instead short BER pathway enzymes (Wilson and Kunkel 2000).

Together, Bistranded / tandem cluster lesions remain unrepaired, they may encounter replication forks during DNA replication, potentially causing replication-induced DSBs or DNA mispairing, which can eventually lead to collapse of replication or incorporation of adaptive mutations (Sage and Shikazono 2017; Sage and Harrison 2011; Eccles et al. 2011; Kozmin, et al. 2021; Cunniffe et al. 2014a; Cunniffe et al. 2014b; Hsu et al. 2022; Naldiga et al. 2020; Shikazono and Akamatsu 2020). These attributes were experimentally validated in proton-irradiated K-rasLA1 mice (Luitel et al. 2018) and high-LET space-irradiation-exposed Bcl2 transgenic C57BL/6 mice (Xie et al. 2020), where it has been observed that a higher incidence of increased carcinoma and DNA replication stress promotes lung carcinogenesis, respectively.

Does clustered DNA damage introduced by high-LET radiation effectively thwart the radioresistance of cancer cells?

Advancements in research on synthetic clustered DNA lesions have deepened our understanding of the biological impacts linked to these lesions within cellular or tissue environments (Lomax et al. 2004a; Shikazono et al. 2006; Malyarchuk et al. 2004; Malyarchuk et al. 2003; Budworth and Dianov 2003). The clustered DNA lesions exhibit a broad spectrum of diversity and undergo various processing mechanisms influenced by factors such as base modifications, the distance between lesions, and the presence of strand breaks. The mutation spectrum associated with non-DSB clustered lesions encompasses various genetic alterations, including deletions, base substitutions, and insertions/deletions (indels) of 1–2 base pairs, primarily focused on the lesions within these clusters (Mladenova, et al. 2022; Kozmin, et al. 2021; Malyarchuk et al. 2009, 2008). The sequential handling of lesions within a non-DSB cluster results in the formation of SSBs as intermediate repair products. These SSBs can hinder the excision or repair of other lesions within the cluster. Consequently, the repair ability of clustered DNA lesions is diminished, leading to elevated lifetime of these lesions. Notably, the detrimental effects of repair intermediates can be exacerbated during replication, potentially leading to the generation of DSBs from single-stranded DNA gaps (Malyarchuk et al. 2009; Harper et al. 2010; Cannan and Pederson 2016; Kumar et al. 2023; Datta et al. 2005; Wang et al. 2014). While most DSBs caused by low-LET radiation are swiftly repaired during the G1 phase, mainly through the canonical nonhomologous end-joining (cNHEJ) pathway. The role of NHEJ-specific DNA-PKcs in repairing high-LET radiation-induced DSBs in tumor cells has shown (Anderson et al. 2010; Liu et al. 2018). The DSBs caused by high-LET radiation, particularly complex DSBs, exhibit slower repair kinetics and extend throughout the cell cycle and a substantial portion of DSBs undergo repair during the late S-phase and G2-phase through homologous recombination by altering cell cycle checkpoint regulations (Okayasu 2012; Okayasu et al. 2006; Gerelchuluun et al. 2015; Mohammadian Gol et al. 2019). Additionally, complex DSBs repaired slowly and inaccurately through alternative (Alt-EJ) or micro-homology mediated (MMEJ and SSA) DNA end joining, often resulting in deletions ranging from a few base pairs to several hundred bases (Scully et al. 2019; Wang and Xu 2017; Sallmyr and Tomkinson 2018; Hanscom and McVey 2020; Mladenova et al. 2022).

Moreover, high-LET radiation exposure leads to an increased occurrence of ssDNA gaps owing to the clustered pattern of DNA damage. The introduction of persistent ssDNA gaps can trigger cell death pathways. Therefore, many tumor cells have developed mechanisms to avoid and escape ssDNA gaps, such as slowing fork movement (Peng et al. 2018; Cong, et al. 2021), reversing replication forks for repair (Bai, et al. 2020), blocking fork elongation, increasing the presence of RPA, and gaining access to translesion synthesis (TLS) to ensure continuous replication without gaps (Igarashi et al. 2023; Fu et al. 2024; Nayak, et al. 2020; Nayak, et al. 2021; Tonzi and Huang 2019; Saldanha, et al. 2023; Wu et al. 2023b; Li et al. 2019; He et al. 2022; Park et al. 2022; Venkadakrishnan et al. 2023). All of these replication gap suppression (RGS) mechanisms involve distinct players that facilitate ssDNA gap avoidance and may contribute to tumor resistance to high-LET radiation. Recent research has highlighted the role of TLSs in adaptive mutability to evade targeted drugs against EGFR/BRAF (Crisafulli and Siravegna 2023; Russo et al. 2019). The unrepaired complex DSBs lead to cell death, primarily because cells are unable to complete the mitotic process. However, a significant fraction of cells proceed through the cell cycle checkpoint despite the presence of unrepaired clustered DNA lesions, leading to an increased occurrence of chromatid and chromosomal aberrations including breaks in chromatids and chromosomes (Mladenova, et al. 2022; Sadeghi Moghadam et al. 2023). Notably, dynamic movement of DSB repair foci toward less densely packed chromatin, and their aggregation with other repair centers may predispose them to erroneous rejoining, particularly during the late S and G2-phases. Consequently, compared with those exposed to low-LET radiation, cells that survive exposure to high-LET radiation exhibit significant translocations, sister chromatid exchanges, and gross chromosomal abnormalities (Mladenova, et al. 2022; Timm et al. 2018). This leads to more pronounced and intricate reorganization of the chromosomes. Additionally, a study underscores the particular types of DNA damage caused by radiation that evade repair mechanisms, resulting in clustered mutations in germ cells (Adewoye et al. 2015). Thus, ionizing radiation, particularly high-LET radiation, causes additional ssDNA gaps, mutations, and complex clustered DNA damage, leading to the killing of the majority of tumor cells (Du et al. 2022). However, a small fraction of radioresistant tumor cell can survive with increased chromosomal abnormalities and adaptive clustered mutations resulting in radiation resistance against high-LET RT (Sage and Shikazono 2017; Sage and Harrison 2011; Adewoye et al. 2015; Nickoloff et al. 2020).

High-LET radiation therapy: harnessing its potential with caution

RT stands as an invaluable asset in cancer treatment, providing benefits to approximately half of all cancer patients. The clinical applications of high-LET radiation have seen a worldwide expansion, notably in proton and hadron therapy (Mohamad, et al. 2017; Yang et al. 2018; Hayashi et al. 2019; Tomizawa et al. 2023; Shiba et al. 2022a; Shiba et al. 2023; Musha et al. 2022; Kiseleva, et al. 2022). Despite the increasing utilization of high-LET radiation in clinics, there remains a lack of comprehensive understanding regarding the biological mechanisms underlying its effects. The remarkable feature of high-LET radiation lies in its ability to generate clustered OCDLs sites, and DSBs. The present therapeutic strategy involving the use of high-LET radiation combined with inhibitor therapy has shown initial benefits. Clinical evidence supports its effectiveness in treating various nonsquamous cell histologies, including adenocarcinoma, malignant melanoma, adenoid cystic carcinoma, hepatoma, early-stage and locally advanced non-small cell lung cancer, and bone/soft tissue sarcoma. (Mohamad, et al. 2017; Yang et al. 2018; Hayashi et al. 2019; Tomizawa et al. 2023; Shiba et al. 2022a; Shiba et al. 2023; Musha et al. 2022; Kiseleva, et al. 2022). Synergistic approach of high-LET RT alongside targeted inhibitors of DNA repair pathways (such as BER, HR, and NHEJ) have been adopted (Cesaire et al. 2019; Srivastava et al. 2018; Hirai et al. 2012; Fujisawa et al. 2015; Ma et al. 2015; Bright et al. 2022). Alternatively, targeting immune checkpoint inhibitors (anti-PD-1/PD-L1 and anti-CTLA4) is explored to counteract resistance (Zhang et al. 2022b; Mondini et al. 2020; Bernal, et al. 2024). Furthermore, high-LET RT option considered effective against hypoxic tumors (Kabakov and Yakimova 2021). However, cancer cells adapted to hypoxic conditions, often due to deregulation of hypoxia-inducible factor-1 (HIF-1) (Yeo et al. 2017; Dongre and Weinberg 2019), or heat shock factor 1 (HSF1) / heat shock protein 90 (HSP90) (Dai 2018). Consequently, small-molecule inhibitors of HIF-1 (Gameiro et al. 2016; Zhang et al. 2015), HSF1 (Schilling et al. 2015), or HSP90 (Kudryavtsev et al. 2017) demonstrate a radiosensitizing effect on hypoxic tumors treated with high-LET radiation (Li et al. 2016; Lee et al. 2016). Additionally, RT combined with inhibitors targeting the PI3K/AKT/mTOR or HIF-1α pathways has been shown to enhance the radiosensitivity of endometrial cancer (Miyasaka et al. 2015; Song et al. 2024). This enhancement is achieved through the reducing autophagy, induction of apoptosis, inhibition of EMT, as well as suppression of NHEJ and HR repair mechanisms (Mardanshahi et al. 2021; Chang et al. 2014; Chang et al. 2013). Thus, combined approach may induce systemic antitumor immune responses, presenting a promising strategy for treating metastatic and difficult to treat cancer.

Although high-LET RT has been shown to be beneficial over low-LET RT, apprehension and caution is warranted. Existing experimental evidence suggests that exposure to high-LET radiation is associated with an increased likelihood of cancer incidents (Luitel et al. 2018; Xie et al. 2020; Datta et al. 2013; Patel et al. 2020). Furthermore, high-LET RT has been shown to induce elevated levels of mutations, genomic instability, and chromosomal aberrations (Masumura et al. 2002; Yatagai et al. 2002), as well as elevated oxidative stress (Suman et al. 2018; Suman et al. 2013). Moreover, predictions from space exploration studies indicate a two-fold increase in excess relative risk (ERR) for male liver and female breast cancer due to high linear energy transfer (LET) radiation (Cucinotta 2022). High LET radiation also constitutes the primary source of uncertainty in galactic cosmic ray (GCR) studies (Cucinotta et al. 2020). The radioresistance observed against high-LET RT may be attributed to various damage bypass or avoidance mechanisms (Igarashi et al. 2023; Fu et al. 2024; Nayak, et al. 2020, 2021; Tonzi and Huang 2019; Saldanha, et al. 2023; Wu et al. 2023b; Li et al. 2019; He et al. 2022; Park et al. 2022; Venkadakrishnan et al. 2023; Nusawardhana et al. 2024), induction of autophagy (Chaachouay et al. 2011; Chen et al. 2015; Chen et al. 2011; Khan et al. 2020), RGS mechanisms (Cantor 2021; Cantor and Calvo 2017; Cong et al. 2024), replication fork reversal/slow movement (Zellweger et al. 2015; Bi 2015), EMT activation (Wu et al. 2023a), release of exosomes (Jokar, et al. 2022; Li et al. 2021a), and epigenetic changes (Kennedy et al. 2018; Perdyan et al. 2024; Tomsia et al. 2024) (Fig. 1). Consequently, the mutagenic potential of high-LET radiation appears to be transiently or persistently heightened. Concerns also arise regarding the delayed repair of clustered DNA damage, which could lead to mutations and the generation of DSBs from halted replication forks, particularly in rapidly dividing tumor cells. Additionally, clustered DSBs may undergo adaptive mutagenic repair, resulting in large and intricate deletions, as observed in cultured cells exposed to high-LET radiation (Michalettou, et al. 2021). Furthermore, genome loss contributes to the genomic instability of tumor cells, potentially leading to the death of the majority of tumor cells. However, rare surviving tumor cells may repopulate tumors with persistent growth due to adaptive mutability induced by high-LET radiation treatment, akin to bacterial cells. In bacterial cells, extensive generation of ssDNA gaps and unrepairable or challenging-to-repair DNA damage activate SOS responses, inducing adaptive mutagenesis (McKenzie et al. 2000; Fuchs 2016). Hence, it is crucial to carefully assess the potential of individual patient cancer cells to activate adaptive mutability and other molecular mechanisms responsible to bypass clustered DNA damage, thereby enhancing survival strategies. A comprehensive understanding of these bypass or survival strategies against clustered DNA damage is imperative on large cohorts studies over extended period for anticipating appropriate therapeutic outcomes from high-LET RT.

Fig. 1
figure 1

The biological effects and possible radiation resistance mechanisms against high-LET radiotherapy (RT). Cellular responses to both low and high-LET radiation offer valuable insights into the biological ramifications. Low-LET radiation (X-rays, γ-rays) causes sparsely distributed DNA damage (DSBs, SSBs, and oxidized bases) (1) in the genome, while high-LET radiation (α-particles, carbon ions, and protons) causes densely distributed clustered complex DNA damage (2). With increasing linear energy transfer (LET) (3), the density of clustered DNA damage increases markedly (4), while the reduction in radiation toxicity to healthy tissues is observed (5). However, the relative biological effectiveness (RBE) improves substantially (6). The cell killing and tumor control by high-LET RT are threatened by numerous molecular mechanisms (7). The possible mechanisms could be mutations incorporation, slow repair and altered cell cycle regulation, replication gap suppression (RGS), reversal or slow fork progression, translesion synthesis (TLS), epithelial‐mesenchymal transition (EMT) activation, release of exosomes, and epigenetic changes, which might confer radioresistance to tumor cells against high-LET RT, and may potentially, reduce the effectiveness of high-LET RT

Furthermore, high-LET RT combined with various inhibitors approach may also pose risks to individual long-term survival due to increased genomic instability and genome restructuring. These consequences could lead to the resurgence of more aggressive tumor growth, as the tumor cells may develop adaptive resistance. Moreover, late toxicity, secondary cancers, and other adverse effects such as cataracts, fibrosis, vascular damage, immunological changes, endocrine disruptions, neurodegeneration could manifest over time (Ramaekers et al. 2011; Yang et al. 2022b; Nakashima et al. 2022; Hanna, et al. 2020; Shiba et al. 2022b; Chung et al. 2013). Additionally, while the increased energy deposition of high-LET radiation in nearby healthy tissue may not be significantly raised, the ramifications of this alteration remain undetermined (Blakely and Chang 2004). Unlike proton therapy, CIRT exhibits a fragmentation tail, wherein nuclear fragments contribute to the dose beyond the intended target area. This phenomenon introduces additional uncertainty, particularly in tissues situated distally from the target site (Malouff et al. 2020). While existing studies suggest that high-LET radiotherapy propensity to induce secondary malignancies is comparatively low compared to photon therapy, the presence of the fragmentation tail raises concerns. It is plausible that the development of secondary, radiation-induced cancers in adjacent healthy tissue may be facilitated by mutations in tumor suppressor genes (Chung et al. 2013; Dracham et al. 2018; Facoetti et al. 2019; Aherne and Murphy 2018; Kraus and Combs 2019; Yock and Caruso 2012; Mohamad et al. 2019).

Therefore, despite the increasing volume of related research and the proliferation of proton beam therapy (PBT) and/or CIRT centers, the clinical advantages of hadron-therapy treatments compared to conventional photon RT remain uncertain (Jefferson, et al. 2019). Ongoing clinical research may not resolve this uncertainty thus far due to a lack of consensus on appropriate study designs and insufficient collaboration among centers to develop comprehensive research protocols (Apisarnthanarax et al. 2018; Li et al. 2021b). Consequently, there has been a proliferation of small, inadequately designed, and poorly reported studies on this topic. These limitations may raise serious questions for the use of PBT and CIRT in experimental treatments, necessitating full disclosure of the risks and uncertainties to patients considering these therapies (Jefferson, et al. 2019; Goetz et al. 2019; Mishra et al. 2017; Fossati et al. 2018). Hence, it is imperative for clinicians and researchers to meticulously assess the therapeutic advantages of high-LET RT in a more comprehensive manner, considering the long-term suppression of tumor growth while minimizing detrimental health consequences. Researchers should focus on understanding cellular adaptive responses and strive to devise radiotherapy devices capable of inducing sparsely distributed clustered DNA damage instead of densely distributed DNA damage, thereby circumventing extreme cellular adaptive mechanisms such as RGS and the utilization of TLS for DNA damage bypass. Consequently, the notion of eradicating cancer cells without triggering their heightened adaptive mutability and facilitating DNA damage bypass should be prioritized.

Conclusion

RT is a crucial therapeutic tool of cancer treatment, benefiting approximately half of cancer patients worldwide. High-LET radiation, notably in proton and hadron therapy, has gained global traction. However, a thorough comprehension of its biological mechanisms remains elusive. High-LET RT induces clustered DNA lesions through precision delivery of tumor radiation doses, potentially enhancing its tumor cell killing potential. Exploiting repair process vulnerabilities, such as BER and HR, alongside high-LET therapy, shows promise in augmenting tumor cell eradication. Nonetheless, caution is warranted due to cellular adaptive mutability supported by various molecular mechanisms (RGS, TLS, autophagy, EMT, release of exosomes, and epigenetic changes) which may lead to long-term toxicities and tumor relapse in the medium to longer term. Despite ongoing research, the clinical superiority of hadron therapy (high-LET) over conventional photon therapy (low-LET) remains inadequate. Addressing these uncertainties requires robust research protocols and transparent communication with patients. Therefore, a balanced evaluation of high-LET RT benefits and risks is imperative for informed clinical decisions.