Abstract

Background

   Carbon ion radiotherapy (CIR) has emerged as a promising therapeutic
   modality for photon-resistant malignancies due to its unique physical
   depth-dose distribution and enhanced radiobiological effectiveness.
   Nevertheless, treatment resistance persists in certain recurrent or
   refractory head and neck squamous cell carcinoma (HNSCC) cases,
   underscoring the need for novel combinatorial strategies. Here, we
   demonstrated the sensitizing effect of targeting discoidin domain
   receptor 1 (DDR1) in HNSCC for CIR.

Methods

   MOC1 and and Cal27 cell lines along with tumor-bearing C57BL/6 mice
   were used for in vitro and in vivo studies. DDR1 was knocked down via
   lentivirus. Cell viability and proliferation were assessed by CCK-8 and
   colony formation assays. Immunogenicity and tumor-infiltrating
   lymphocytes were measured via flow cytometry and immunofluorescence.
   Tumor suppression mechanisms were investigated using RNA sequencing and
   bioinformatics. Ferroptosis markers (lipid peroxidation, iron, ROS)
   were detected using MDA, BODIPY 581/591 C11, FerroOrange, and DCFH-DA
   probes. Upstream ferroptosis mechanisms were analyzed by Western blot,
   co-immunoprecipitation, key molecule modulator administration, and SCD1
   overexpression.

Results

   We demonstrated that targeting DDR1 potentiated CIR by triggering
   ferroptosis-mediated immunogenic cell death, which in turn enhanced
   antitumor immunity. Mechanistically, DDR1 sustained tumor cell survival
   by forming 14–3-3-mediated assembly of a DDR1/14–3-3/Akt ternary
   complex, thereby activating the Akt/mTORC1/SREBP1/SCD1 axis to promote
   monounsaturated fatty acid (MUFA) biosynthesis and suppress
   ferroptosis. Silencing DDR1 disrupted this complex, alleviating
   MUFA-mediated ferroptosis inhibition and subsequently increasing tumor
   immunogenicity. This immunogenic shift facilitated CD8 + T cell
   infiltration and cytotoxicity, amplifying CIR-induced tumor
   suppression. Furthermore, pharmacological inhibition of DDR1 using the
   small-molecule inhibitor 7rh recapitulated these effects, demonstrating
   potent anti-proliferative and ferroptosis-inducing capabilities,
   enhancing CIR sensitivity to better control tumor progression.

Conclusions

   Our findings positioned DDR1 targeting as a therapeutic strategy to
   potentiate CIR through immunogenic ferroptosis induction in HNSCC.

Supplementary Information

   The online version contains supplementary material available at
   10.1186/s12967-025-07062-5.

   Keywords: Carbon ion radiotherapy, Discoidin domain receptor 1,
   Ferroptosis, Immunogenic cell death, PI3K/Akt/mTOR signaling, Head and
   neck squamous cell carcinoma

Introduction

   Head and neck squamous cell carcinoma (HNSCC), a malignancy originating
   from the mucosal epithelium of the oral cavity, oropharynx,
   hypopharynx, or larynx, ranks as the sixth most common cancer globally
   and exhibits pronounced tumor heterogeneity [[44]1–[45]3]. Radiotherapy
   (RT) remains a cornerstone in the management of HNSCC, offering organ
   preservation advantages for both curative and adjuvant purposes.
   Early-stage HNSCC patients treated with definitive RT can achieve
   outcomes comparable to radical surgery while preserving anatomical
   integrity [[46]4]. However, approximately 70% of HNSCC patients present
   with locally advanced disease (stage III-IV) at initial diagnosis.
   Despite aggressive chemoradiotherapy, 15–50% of these patients
   experience locoregional recurrence. For recurrent cases who have
   undergone previous radiotherapy, re-irradiation efficacy remains
   suboptimal, with 5-year survival rates stagnating at 10–30% [[47]5,
   [48]6]. These limitations underscore the urgent need for innovative
   therapeutic strategies to overcome radioresistance and improve survival
   outcomes in HNSCC.

   Carbon ion radiotherapy (CIR), an advanced form of particle therapy,
   has gained prominence as a promising treatment modality owing to its
   distinct physical and biological advantages. The characteristic "Bragg
   peak" energy deposition profile of carbon ions allows for precise tumor
   targeting while sparing surrounding normal tissues from excessive
   radiation exposure. As a high linear energy transfer (LET) radiation
   type, carbon ions generate complex DNA double-strand breaks (DSBs) with
   diminished reliance on oxygen availability and cell cycle phase, which
   effectively counteracts intrinsic radioresistance in hypoxic or
   quiescent tumor populations [[49]7]. Supporting its clinical potential,
   our preliminary data in HNSCC demonstrate favorable
   outcomes—treatment-naïve patients receiving combined CIR achieved a
   3-year local control rate of 83.5%, while re-irradiation with carbon
   ions in recurrent cases yielded 1-year survival and local control rates
   of 95.9% and 84.9%, respectively [[50]8, [51]9]. Despite these
   advances, the 2-year local control rates for carbon ion re-irradiation
   alone remain suboptimal (40–60%) [[52]9, [53]10], underscoring the
   critical need for combined therapeutic strategies to enhance treatment
   efficacy.

   The advent of immune checkpoint inhibitors (ICIs), particularly
   PD-1/PD-L1 inhibitors, has revolutionized oncology. Preclinical and
   early clinical studies demonstrated a synergistic effect between photon
   radiotherapy and ICIs through enhanced tumor antigen release and immune
   activation [[54]11–[55]14]. However, results from a multicenter
   randomized prospective clinical study showed that the combination of
   photon radiotherapy and the ICI pembrolizumab did not significantly
   improve the tumor control and survival of locally advanced HNSCC
   [[56]15], underscoring the need for optimized radio-immunotherapy
   strategies. Intriguingly, emerging evidence indicated that high-LET
   carbon ions might surpass photons in immunomodulatory capacity. By
   inducing immunogenic cell death (ICD), promoting tumor-associated
   antigen release, diversifying T-cell receptor repertoires, and
   reprogramming the immunosuppressive tumor microenvironment (TME), CIR
   could potentiate systemic antitumor immunity and amplify immunotherapy
   efficacy [[57]16–[58]19]. This paradigm positions CIR and immunotherapy
   combination as a compelling therapeutic avenue for refractory or
   recurrent HNSCC.

   Concurrently, the identification of novel molecular targets to optimize
   treatment responses remains critical. Discoidin Domain Receptor 1
   (DDR1), a collagen receptor with tyrosine kinase activity, has garnered
   attention as a regulator of tumor-stroma interactions and immune
   evasion [[59]20, [60]21]. DDR1 activation via cleaved collagen I
   binding triggered NF-κB-NRF2 signaling to enhance macropinocytosis and
   mitochondrial biogenesis, fostering tumor progression in pancreatic
   cancer models [[61]22]. Notably, the DDR1 extracellular domain mediated
   collagen fiber remodeling to restrict immune infiltration in breast
   cancer, while DDR1 inhibition promoted cytotoxic lymphocyte recruitment
   and tumor suppression [[62]23]. Besides, the abnormal increase of DDR1
   level correlated negatively with pro-inflammatory factors across
   multiple malignancies [[63]24, [64]25], positioning DDR1 as a potential
   immunotherapeutic target. However, the tumor-suppressive effect and
   immune response resulting from the combined treatment of targeting DDR1
   and CIR have not been elucidated, particularly in HNSCC.

   Given these knowledge gaps, this study aims to investigate the
   combinatorial effects of DDR1 inhibition and CIR on tumor cell
   eradication and immune regulation in HNSCC, and explored the mechanisms
   underlying their antitumor efficacy, involved in AKT/mTOR signaling and
   immunogenic ferroptosis pathway. This work will establish preclinical
   evidence for a novel therapeutic paradigm with transformative potential
   in HNSCC management.

Materials and methods

Cell culture

   The murine oral squamous cell carcinoma cell line MOC1 (catalog
   #CTCC-001–0916, Zhejiang Meisen Cell Technology Co., Ltd, China) was
   cultured in DMEM/F-12 medium (Gibco) supplemented with 10% fetal bovine
   serum (FBS; Gibco), 10 μg/mL insulin (Sigma-Aldrich), 10 ng/mL
   epidermal growth factor (EGF; PeproTech), 80 ng/mL hydrocortisone
   (Sigma-Aldrich), and 1% penicillin–streptomycin (Gibco). The human
   tongue squamous cell carcinoma cell line Cal27 (catalog #CTCC-003–0022,
   Zhejiang Meisen Cell Technology Co., Ltd, China) was maintained in DMEM
   (Gibco) supplemented with 10% FBS (Gibco) and 1%
   penicillin–streptomycin (Gibco). Both cell lines were maintained at
   37 °C in a humidified 5% CO₂ incubator. The medium was refreshed every
   2 days, and subculturing was performed at 90% confluency using 0.25%
   trypsin–EDTA (Gibco).

Chemicals

   Liproxstatin-1 (Lipro-1) was sourced from Selleck Chemicals (USA).
   Ferrostatin-1 (Fer-1), MK-2206, SC79, MHY1485, stearic acid (SA), and
   7rh were procured from MedChemExpress (MCE, USA). Oleic acid (OA) was
   obtained from Sigma-Aldrich (USA). In vivo dosing regimens for Lipro-1
   and 7rh are detailed in the "Mouse models and treatments" section. For
   in vitro studies, pharmacological agents were introduced 2–3 h
   pre-irradiation with distinct exposure durations: Fer-1 (5 μM) and
   MK-2206 (10 μM) remained in culture for 24 h, whereas 7rh (2 μM) was
   maintained for 12 h before withdrawal. SC79 (15 μM) and MHY1485 (10 μM)
   were removed 2 h post-irradiation. SA and OA (100 μM each) persisted
   until final analysis. Unless otherwise stated, all cellular endpoints
   were assessed at 48 h post-irradiation.

Carbon ion radiotherapy

   Carbon ion radiotherapy was administered through a heavy ion
   synchrotron accelerator (Siemens AG) at Shanghai Proton and Heavy Ion
   Center (SPHIC). In short, carbon ion beams were precisely delivered to
   the target with a homogeneous spread-out Bragg peak with the energy of
   156.3 MeV/u (dose averaged LET 50 keV/μm). For cellular studies,
   culture vessels were mounted on a custom vertical stand orthogonally
   aligned to the beam axis. In animal experiments, tumor-bearing mice
   underwent anesthesia via intraperitoneal pentobarbital sodium
   (50 mg/kg) and were immobilized on a beam-perpendicular platform. A
   tailored protection device ensured localized irradiation to the target
   region while shielding surrounding tissues. It is worth noting that,
   except for the colony formation assay, we used 5 Gy as the irradiation
   dose for cells and animals. Pre-experiment of the colony formation
   assay revealed that 5 Gy irradiation resulted in negligible cell
   survival after 10 days. Consequently, the radiation dose was adjusted
   downward to 2.5 Gy specifically for colony formation assay.
   Furthermore, we also utilized a fractionated irradiation model in both
   parental and DDR1-knockdown cell lines for colony formation assay to
   investigate the radiosensitizing effects of DDR1 inhibition on carbon
   ion radiotherapy. This model employed fractionation schemes of
   2.5 Gy × 2 fractions and 1 Gy × 5 fractions.

Mouse models and treatments

   Syngeneic tumor models were established by subcutaneous implantation of
   DDR1-knockdown or control MOC1 cells in the right hind flank of C57BL/6
   mice (7–8 weeks old). Carbon ion irradiation was initiated when mean
   tumor volume reached 50–100 mm^3. To evaluate ferroptosis-mediated
   tumor suppression, Lipro-1 (10 mg/kg; Selleck) was administered
   intraperitoneally daily from the date of radiotherapy until the end of
   the study. For 7rh intervention, mice received 7rh orally (25 mg/kg;
   MCE) every 48 h starting 7 days pre-irradiation until study
   termination. Tumor progression was monitored via serial caliper
   measurements, with volumes calculated as (length × width^2)/2. Animals
   were humanely euthanized upon reaching the ethical endpoint or at
   predetermined timepoints for tissue analysis. All procedures complied
   with the guidelines of SPHIC Institutional Animal Care and Use
   Committee. The ethics committee allowed a maximum tumor size of 2000
   mm^3.

Statistical analysis

   Statistical analyses were conducted using GraphPad Prism v9.5.1 and
   SPSS v20.0. Group comparisons employed two-tailed unpaired t-tests (two
   groups) or one-way ANOVA (multiple groups), while longitudinal data
   (cell growth curves, tumor volumes) were assessed via two-way ANOVA. A
   significance threshold of p < 0.05 was applied, denoted by asterisks:
   p < 0.05 (*), < 0.01 (**), < 0.001 (***), and < 0.0001 (****).

   The remaining materials and methods can be found in the Supplementary
   File.

Results

DDR1 overexpression correlates with immune evasion and poor prognosis in
HNSCC

   Analysis of data from The Cancer Genome Atlas-head and neck squamous
   cell carcinoma (TCGA-HNSC) dataset revealed a striking upregulation of
   DDR1 in tumor tissues compared to normal tissues (Supplementary
   Fig. 1A, B). ROC curve analysis further demonstrated the diagnostic
   utility of DDR1 in HNSCC, with an area under the curve (AUC) of 0.757
   (Supplementary Fig. 1C). Critically, Kaplan–Meier survival analysis
   showed that patients with high DDR1 expression exhibited significantly
   shorter overall survival (Supplementary Fig. 1D), demonstrating the
   prognostic significance of DDR1.

   Recent studies have emphasized the crucial role of DDR1 in driving
   immune evasion [[65]20, [66]21, [67]23]. To dissect DDR1’s
   immunosuppressive function, we evaluated its correlation with
   immune-related markers. In the TCGA-HNSC dataset, DDR1 expression was
   found to be inversely correlated with the expression of CD8B, GZMB,
   IFNG, and IL12B, as well as scores for T cells, CD8 T cells, and
   cytotoxic cells (Supplementary Fig. 1E-L). Collectively, these findings
   could position DDR1 as a driver of tumor progression and therapy
   resistance by mediating immune evasion in HNSCC, thereby providing a
   rationale for targeting DDR1 to restore antitumor immunity and enhance
   therapeutic outcomes.

DDR1 inhibition sensitizes HNSCC to CIR to suppress tumor malignancy

   While CIR demonstrates promising efficacy in treating HNSCC, its
   limited effectiveness in refractory or recurrent cases underscores the
   need for combination strategies. To investigate whether DDR1 inhibition
   enhanced CIR sensitivity, we generated stable DDR1-knockdown MOC1
   (shDDR1-2, hereinafter referred to as shDDR1) and Cal27 cells using
   lentiviral infection (Fig. [68]1A; Supplementary Fig. 3A). Consistent
   with the hypothesis of this study, colony formation and CCK-8 assays
   revealed that shDDR1 combined with CIR (shDDR1 + CIR) significantly
   reduced cell survival compared to the other three groups
   (Fig. [69]1B-D; Supplementary Fig. 2A-C, 3B-D). Furthermore, we found
   that the combination of DDR1 knockdown and CIR profoundly inhibited the
   migration, invasion, and epithelial-mesenchymal transition (EMT) of
   MOC1 cells (Fig. [70]1E-G).

Fig. 1.

   [71]Fig. 1
   [72]Open in a new tab

   Antitumor effect of DDR1 inhibition combined with carbon ion
   radiotherapy in head and neck squamous cell carcinoma. A MOC1 cells
   were infected with three independent lentiviral particles for DDR1
   knockdown. Western blot analysis of DDR1 protein levels in
   DDR1-knockdown and control cells. B, C Representative images or
   quantitative results of colony formation assay in MOC1 cells after DDR1
   knockdown and carbon ion radiotherapy. Scale bar, 5 mm. D CCK-8 cell
   proliferation assay in MOC1 cells after DDR1 knockdown and carbon ion
   radiotherapy. E Representative images or quantitative results of wound
   healing in MOC1 cells after DDR1 knockdown and carbon ion radiotherapy.
   Scale bar, 50 μm. F Representative images or quantitative results of
   transwell migration and invasion in MOC1 cells after DDR1 knockdown and
   carbon ion radiotherapy. Scale bar, 50 μm. G Western blot analysis of
   EMT markers (E-cadherin, Vimentin, ZEB1 and Slug) in MOC1 cells after
   DDR1 knockdown and carbon ion radiotherapy. H, I Tumor growth and
   images of syngeneic MOC1 subcutaneous tumor models after DDR1 knockdown
   and carbon ion radiotherapy. N = 5 mice/group. J Representative images
   of H&E and immunohistochemistry staining of Ki67. Scale bar, 20 μm. *
   p < 0.05, *** p < 0.001, **** p < 0.0001

   Notably, the strong antitumor effect of inhibiting DDR1 combined with
   CIR was validated in vivo. In the syngeneic MOC1 subcutaneous tumor
   models, shDDR1 + CIR markedly suppressed tumor growth compared to
   monotherapy or control groups (Fig. [73]1H). Immunohistochemistry (IHC)
   further showed reduced Ki67 expression in shDDR1 + CIR tumors,
   indicating diminished proliferative activity (Fig. [74]1J).
   Collectively, these data indicated that targeting DDR1 was a promising
   therapeutic strategy to amplify CIR efficacy in HNSCC.

Combined DDR1 inhibition and CIR promote immunogenic cell death

   Given the immunosuppressive role of DDR1 and its negative correlation
   with CD8 + T cell infiltration (Supplementary Fig. 1), we hypothesized
   that inhibition of DDR1 could further amplify CIR-induced antitumor
   immunity. On this basis, we first conducted in vitro cell experiments
   to investigate whether tumor cells underwent immunogenic cell death.
   Our findings revealed that the shDDR1 + CIR combination treatment
   significantly increased the exposure of damage-associated molecular
   patterns (DAMPs), such as calreticulin (CRT) and heat shock protein 90
   (HSP90), on the surface of tumor cells (Fig. [75]2A, B; Supplementary
   Fig. 4A). Meanwhile, MHC-I on MOC1 cells and HLA-A, B, C on Cal27 cells
   exhibited a similar trend (Fig. [76]2C; Supplementary Fig. 4B, C),
   indicating that the combination of DDR1 inhibition and CIR
   significantly enhanced the immunogenicity of HNSCC cells.

Fig. 2.

   [77]Fig. 2
   [78]Open in a new tab

   Inhibition of DDR1 combined with carbon ion radiotherapy induces
   immunogenic cell death. A–C Flow cytometry analysis of CRT, HSP90, and
   MHC-I expression levels on MOC1 cells after DDR1 knockdown and carbon
   ion radiotherapy. D Immunofluorescence assay of tumor-infiltrating
   CD4 + T cells (red) and CD8 + T cells (green), and immunohistochemistry
   staining of Granzyme B. Scale bar, 20 μm. E Gating strategy for flow
   cytometry analysis of T cell infiltration in tumor tissues. F, G
   Quantitative results of flow cytometry analysis of tumor-infiltrating
   CD8 + T cells and PD1 + CD8 + T cells. N = 6 mice/group. H, I
   Quantitative results of flow cytometry analysis of tumor‐infiltrating
   CD8 + IFN‐γ + cells. N = 6 mice/group. *** p < 0.001, **** p < 0.0001

   Subsequently, we expanded our investigation into the combinatorial
   effects of DDR1 suppression and CIR on TME remodeling through in vivo
   experiment. Immunofluorescence analysis was conducted to analyze immune
   cell infiltration in murine tumor specimens and the results revealed
   pronounced enhancement of CD8 + T lymphocyte infiltration in the
   shDDR1 + CIR cohort (Fig. [79]2D). Flow cytometric quantification
   corroborated the finding, showing a substantial elevation in CD8 + T
   cell infiltration rates in the combination treatment group
   (Fig. [80]2E, F). Also, immunohistochemical and flow cytometric
   analyses demonstrated significantly elevated Granzyme B expression in
   tumor tissues, coupled with enhanced IFN-γ production capacity in
   CD8 + T cells (Fig. [81]2D, H), indicative of augmented cytotoxic
   potential in tumor-infiltrating immune cells. Notably, our
   investigation uncovered that concurrent DDR1 inhibition and CIR
   administration induced PD-1 molecule upregulation on CD8 + T cell
   surfaces (Fig. [82]2E, G), indicating a potential for achieving a
   better antitumor effect when combined with anti-PD-1 immune checkpoint
   inhibitors.

Ferroptosis execution underlies the antitumor effects of DDR1 inhibition
combined with CIR

   To elucidate the molecular mechanism underlying the therapeutic
   efficacy of shDDR1 + CIR combination therapy, we conducted RNA
   sequencing analysis. Comparative transcriptomic profiling between the
   shDDR1 + CIR group and control cohorts (shNC, shDDR1, shNC + CIR)
   identified 906 differentially expressed genes potentially mediating the
   antitumor advantage of the combinatorial treatment (Supplementary
   Fig. 6A). KEGG pathway enrichment analysis revealed involvement of the
   ferroptosis pathway among these candidate genes (Fig. [83]3A).

Fig. 3.

   [84]Fig. 3
   [85]Open in a new tab

   Inhibiting DDR1 sensitizes carbon ion radiotherapy by promoting
   ferroptosis. A KEGG pathway enrichment analysis of differentially
   expressed genes between the combined DDR1 knockdown and carbon ion
   radiotherapy group and other groups in MOC1 cells. B AO/PI staining to
   detect cell death in MOC1 cells after DDR1 knockdown and carbon ion
   radiotherapy. C MDA assay to measure lipid peroxidation levels in MOC1
   cells after treatment. D BODIPY™ 581/591 C11 fluorescent probe to
   detect lipid peroxidation levels in MOC1 cells after treatment. E
   FerroOrange fluorescent probe to detect Fe^2+ levels in MOC1 cells
   after treatment. F DCFH-DA fluorescent probe to detect ROS levels in
   MOC1 cells after treatment. G Transmission electron microscopy
   observation of mitochondrial changes (red arrows) in MOC1 cells after
   treatment. Scale bar, upper 1 μm, lower 500 nm. H Western blot analysis
   of GPX4, SLC7A11, and Hmox1 expression in MOC1 cells after treatment. *
   p < 0.05, *** p < 0.001, **** p < 0.0001

   Ferroptosis, an iron-dependent programmed cell death modality driven by
   lipid peroxide accumulation, was subsequently investigated as a
   potential mechanism. Fluorescence co-staining with acridine
   orange/propidium iodide (AO/PI) demonstrated enhanced cellular
   mortality in the combination treatment group (Fig. [86]3B). Next, we
   found that the shDDR1 + CIR treatment induced maximal elevation of
   malondialdehyde (MDA) and lipid peroxidation compared to the other
   three groups (Fig. [87]3C, D; Supplementary Fig. 5A, B). Untargeted
   lipidomic profiling further revealed substantial enrichment of
   polyunsaturated fatty acid-containing phospholipids (PUFA-PLs) in the
   experimental cohort, particularly ferroptosis-associated
   phosphatidylethanolamines (PEs) bearing arachidonyl (20:4) or adrenyl
   (22:4) acyl chains (Supplementary Fig. 6B). Then, intracellular Fe^2+
   and reactive oxygen species (ROS) levels were detected and the results
   demonstrated the most pronounced accumulation of Fe^2⁺ and ROS in
   shDDR1 + CIR-treated cells (Fig. [88]3E, F; Supplementary Fig. 5C, D).
   We also performed TEM analysis of cells, and found that the
   morphological changes of mitochondria in the shDDR1 + CIR group were
   the most obvious, including mitochondrial atrophy, increased membrane
   density, and reduced or even absent cristae (Fig. [89]3G, indicated by
   red arrows). Moreover, western blot analysis of several
   ferroptosis-related landmark molecules showed significant
   downregulation of GPX4 and SLC7A11 alongside upregulation of Hmox1 in
   the combination group (Fig. [90]3H; Supplementary Fig. 5E).
   Collectively, these findings suggested that combined DDR1 inhibition
   and CIR induced ferroptosis-driven tumor cell death in HNSCC.

   To further validate ferroptosis involvement in this therapeutic
   paradigm, we employed Fer-1 to inhibit ferroptosis at the cellular
   level. The results indicated that the elevations in MDA levels, lipid
   peroxidation, and Fe^2+ levels observed in the shDDR1 + CIR group were
   effectively abrogated upon administration of the Fer-1 (Fig. [91]4A-C).
   Additionally, Fer-1 partially alleviated the significant enhancements
   in CRT, HSP90 and MHC-I levels caused by DDR1 inhibition and CIR
   (Fig. [92]4D-F). We then proceeded to extend our investigation to in
   vivo models, selecting Lipro-1 as a ferroptosis antagonist.
   Immunohistochemical analysis of 4-hydroxynonenal (4-HNE), a pivotal
   byproduct arising from lipid peroxidation, revealed marked elevation in
   the shDDR1 + CIR group, which was significantly mitigated by Lipro-1
   co-treatment (Fig. [93]4G). Notably, Lipro-1 administration
   substantially diminished tumor-infiltrating CD8⁺ T lymphocyte
   populations across all treatment arms, with most pronounced effects
   observed in the shDDR1 + CIR cohort (Fig. [94]4H). Correspondingly, the
   antitumor efficacy of DDR1 inhibition combined with CIR was
   significantly compromised upon ferroptosis blockade (Fig. [95]4I).
   Collectively, these findings substantiated that ferroptosis activation
   served as a critical mechanism through which combined DDR1 suppression
   and CIR elicited immunogenic cell death in HNSCC.

Fig. 4.

   [96]Fig. 4
   [97]Open in a new tab

   The ferroptosis pathway triggered by inhibiting DDR1 in conjunction
   with carbon ion radiotherapy bolsters antitumor immune responses. A-C
   Changes in MDA, lipid peroxidation, and Fe^2+ levels in each group in
   MOC1 cells after Fer-1 treatment. D-F Changes in CRT, HSP90, and MHC-I
   expression levels in each group in MOC1 cells after Fer-1 treatment. G
   Immunohistochemistry staining of 4-HNE in mouse tumor tissues in each
   group with or without Lipro-1 treatment. Scale bar, 20 μm. H Changes in
   tumor-infiltrating CD8 + T cells in each group after Lipro-1 treatment.
   N = 3 mice/group. I Tumor growth in each group with or without Lipro-1
   treatment. N = 5 mice/group. * p < 0.05, ** p < 0.01, *** p < 0.001,
   **** p < 0.0001

DDR1 modulates Akt/mTOR signaling via 14–3-3 interaction to affect carbon ion
radiosensitivity

   We further explored the specific mechanism by which DDR1 inhibition
   combined with CIR induced ferroptosis. Referring back to our KEGG
   pathway enrichment analysis results, we identified a significant
   enrichment of the PI3K-Akt signaling pathway (Fig. [98]3A). Since mTOR
   is a key downstream effector molecule of the PI3K-Akt signaling
   pathway, we first verified through western blot experiments whether the
   PI3K/Akt/mTOR signaling pathway undergoes changes after DDR1 knockdown
   and CIR treatment. The results revealed substantial attenuation of PI3K
   (Tyr458), Akt (Ser473), and mTOR (Ser2448) phosphorylation following
   DDR1 knockdown and CIR exposure (Fig. [99]5A; Supplementary Fig. 7),
   suggesting inactivation of this pathway. To establish functional
   linkage between PI3K/Akt suppression and ferroptosis potentiation, we
   initially employed the Akt inhibitor MK-2206, and observed that MK-2206
   exacerbated lipid peroxidation and MDA accumulation, especially in the
   control group (Fig. [100]5B, C). Conversely, the use of the Akt agonist
   SC79 effectively attenuated the elevation of lipid peroxidation and MDA
   levels caused by DDR1 inhibition and CIR (Fig. [101]5D, E). Also,
   complementary experiments with the mTORC1 activator MHY1485
   demonstrated similar rescue effects, effectively reversing the high
   lipid peroxidation and MDA increases mediated by DDR1 inhibition and
   CIR (Fig. [102]5F, G). These findings demonstrated that PI3K/Akt/mTOR
   pathway inhibition serves as the principal mechanistic conduit through
   which DDR1 suppression and CIR drove ferroptosis.

Fig. 5.

   [103]Fig. 5
   [104]Open in a new tab

   DDR1 regulates carbon ion radiosensitivity by affecting the Akt/mTOR
   pathway through 14–3-3. A Western blot analysis of protein expression
   levels involving PI3K/Akt/mTOR pathway in MOC1 cells after DDR1
   knockdown and carbon ion radiotherapy. B–G Changes in lipid
   peroxidation and MDA levels in each group in MOC1 cells after treatment
   with (B, C) Akt inhibitor MK-2206, (D, E) Akt activator SC79 or (F, G)
   mTOR activator MHY1485. H, I Co-immunoprecipitation of endogenous DDR1,
   Akt and 14–3-3 from whole-cell lysates of MOC1 cells. ** p < 0.01, ***
   p < 0.001, **** p < 0.0001

   To elucidate the DDR1-mediated regulation of carbon ion
   radiosensitivity via the Akt/mTOR axis, we hypothesized a potential
   protein-level interaction between DDR1 and Akt. Endogenous
   co-immunoprecipitation (Co-IP) assays in MOC1 cells confirmed DDR1-Akt
   binding (Fig. [105]5H). Given DDR1’s lack of canonical Akt-binding
   motifs, we investigated intermediary proteins, guided by prior evidence
   implicating 14–3-3 as a linker between DDR1 and Akt signaling
   [[106]26]. Subsequent Co-IP validation demonstrated tripartite complex
   formation involving DDR1, 14–3-3, and Akt (Fig. [107]5H).
   Interestingly, Western blot analysis revealed a significant reduction
   in 14–3-3 protein levels following DDR1 knockdown and CIR treatment
   (Supplementary Fig. 8A). These findings establish a mechanistic axis
   wherein DDR1 engages Akt/mTOR signaling through 14–3-3, thereby
   modulating cellular response to carbon ion irradiation.

Combined DDR1 inhibition and CIR induce ferroptosis by suppressing
mTOR/SREBP1/SCD1-mediated lipogenesis

   Building upon established evidence that PI3K/Akt/mTOR pathway
   inhibition potentiated ferroptosis via SREBP1/SCD1-mediated lipogenic
   regulation [[108]27], we investigated whether DDR1 suppression combined
   with CIR modulated this metabolic axis to drive ferroptosis. We
   performed qPCR and western blot experiments, and the results revealed
   marked downregulation of SREBP1 and its downstream effector SCD1
   following combination treatment (Fig. [109]6A-C). Furthermore, the mTOR
   activator MHY1485 rescued the decreased expression of SREBP1 and SCD1
   induced by DDR1 knockdown and CIR (Fig. [110]6D). SCD1 is an essential
   lipid desaturase capable of converting saturated fatty acids (SFAs)
   into monounsaturated fatty acids (MUFAs) [[111]28]. Therefore, we
   supplemented exogenous MUFA oleic acid (18:1, OA) and SFA stearic acid
   (18:0, SA), and discovered that unlike SA, the addition of OA could
   reduce the elevation of lipid peroxidation levels caused by DDR1
   knockdown and CIR (Fig. [112]6E, F). To further confirm the role of
   SCD1 in ferroptosis, we overexpressed SCD1 in MOC1 cells and observed a
   significant decrease in cellular lipid peroxidation levels
   (Fig. [113]6G). Additionally, CCK-8 proliferation and colony formation
   assays indicated that SCD1 overexpression could enhanced cellular
   viability (Fig. [114]6H-J). In summary, these results indicated that
   inhibiting DDR1 combined with CIR promoted ferroptosis in HNSCC cells
   by suppressing mTOR/SREBP1/SCD1-mediated lipogenesis.

Fig. 6.

   [115]Fig. 6
   [116]Open in a new tab

   Combined DDR1 inhibition and CIR trigger ferroptosis by suppressing
   mTOR/SREBP1/SCD1-mediated lipogenesis. A, B qPCR analysis of SREBF1 and
   SCD1 mRNA levels in MOC1 cells after DDR1 knockdown and carbon ion
   radiotherapy. C Western blot analysis of SREBP1 and SCD1 protein levels
   in MOC1 cells after DDR1 knockdown and carbon ion radiotherapy. D
   Changes in SREBP1 and SCD1 protein expression levels in each group in
   MOC1 cells after MHY1485 treatment. E, F Changes in lipid peroxidation
   levels in each group in MOC1 cells after treatment with oleic acid and
   stearic acid. G Changes in lipid peroxidation levels in each group in
   MOC1 cells after overexpression of SCD1. H CCK-8 cell proliferation
   assay to detect cell viability after overexpression of SCD1 in MOC1
   cells. I, J Colony formation assay to detect cell proliferation ability
   after overexpression of SCD1 in MOC1 cells, with representative images
   J and quantitative results I. ns non-significant, * p < 0.05, **
   p < 0.01, *** p < 0.001, **** p < 0.0001

DDR1 inhibitor 7rh enhances CIR sensitivity to suppress HNSCC tumor growth

   To enhance the clinical translational value of DDR1 as a therapeutic
   target for HNSCC, we employed a highly selective DDR1 inhibitor, 7rh,
   in combination with carbon ion therapy, aiming to observe its
   radiosensitization effect on HNSCC. Our results demonstrated that 7rh
   significantly suppressed HNSCC cell proliferation and enhanced
   radiosensitivity to carbon ion irradiation (Fig. [117]7A, B;
   Supplementary Fig. 9A, B). Mechanistically, 7rh treatment induced
   elevation of lipid peroxidation, Fe^2⁺ accumulation, and ROS levels
   (Fig. [118]7C-E; Supplementary Fig. 9C-E), as well as reduction of PI3K
   (Tyr458), Akt (Ser473) and mTOR (Ser2448) phosphorylation and 14–3-3
   levels (Fig. [119]7F; Supplementary Fig. 8B, 9F), pharmacologically
   confirming that DDR1 inhibition potentiated CIR by suppressing the
   Akt/mTOR pathway to promote ferroptosis, thereby sensitizing cells to
   radiotherapy. Furthermore, in vivo experiments revealed that the
   combination of 7rh and CIR exhibited remarkable increases in
   tumor-infiltrating CD8⁺ T lymphocytes (Fig. [120]7G) along with
   significant tumor growth suppression and regression trends
   (Fig. [121]7H, I). Collectively, these results demonstrated that 7rh
   could effectively promote ferroptosis in tumor cells and enhance the
   antitumor immune response induced by CIR in HNSCC.

Fig. 7.

   [122]Fig. 7
   [123]Open in a new tab

   DDR1 inhibitor 7rh increases CIR sensitivity to suppress tumor growth
   of head and neck squamous cell carcinoma. A Representative images and
   quantitative results of colony formation assay in MOC1 cells after 7rh
   treatment and carbon ion radiotherapy. B CCK-8 cell proliferation assay
   in MOC1 cells after 7rh treatment and carbon ion radiotherapy. C
   BODIPY™ 581/591 C11 fluorescent probe to detect lipid peroxidation
   levels in MOC1 cells after 7rh treatment and carbon ion radiotherapy. D
   FerroOrange fluorescent probe to detect Fe^2+ levels in MOC1 cells
   after 7rh treatment and carbon ion radiotherapy. E DCFH-DA fluorescent
   probe to detect ROS levels in MOC1 cells after 7rh treatment and carbon
   ion radiotherapy. F Western blot analysis of protein expression levels
   involving Akt/mTOR pathway in MOC1 cells after 7rh treatment and carbon
   ion radiotherapy. G Flow cytometry analysis of tumor-infiltrating
   CD8 + T cells in mice after treatment with 7rh and carbon ion
   radiotherapy. N = 5 mice/group. H, I Tumor growth and images of
   syngeneic MOC1 subcutaneous tumor models after treatment with 7rh and
   carbon ion radiotherapy. N = 5 mice/group. J Schematic diagram
   illustrating that the inhibition of DDR1 increases carbon ion
   radiosensitivity by inhibiting Akt/mTOR/SREBP1/SCD1-mediated
   lipogenesis of MUFAs, promoting ferroptosis and immunogenic cell death
   in head and neck squamous cell carcinoma. * p < 0.05, ** p < 0.01, ***
   p < 0.001, **** p < 0.0001

Discussion

   HNSCC is a complex, highly heterogeneous malignancy. Despite
   radiotherapy being a cornerstone treatment, clinical outcomes remain
   suboptimal. CIR offers promise due to its unique physical and
   radiobiological benefits, including the induction of complex DNA damage
   and enhanced antitumor immunity [[124]29, [125]30]. Although clinical
   studies [[126]8–[127]10] showed superior efficacy of CIR, limitations
   still persist, suggesting that CIR-induced immune activation may be
   offset by immunosuppressive mechanisms in HNSCC. Our study established
   DDR1 as a novel therapeutic target to amplify CIR efficacy by
   triggering ferroptosis-mediated immunogenic cell death. As illustrated
   in Fig. [128]7J, we identified a critical molecular mechanism wherein
   DDR1 facilitated Akt/mTOR signaling pathway activation in HNSCC cells
   through 14–3-3-mediated complex formation. Both genetic silencing and
   pharmacological inhibition of DDR1 disrupted the DDR1/14–3-3/Akt
   ternary complex, effectively suppressing mTORC1-dependent regulation of
   SREBP1/SCD1. This metabolic reprogramming attenuated MUFA biosynthesis,
   thereby alleviating MUFA-mediated suppression of ferroptosis. The
   resultant enhancement of ferroptosis augmented tumor immunogenicity,
   promoting CD8 + T cell infiltration and potentiating CIR-mediated
   antitumor effects.

   Bioinformatics analysis of TCGA-HNSC datasets revealed significant DDR1
   overexpression in HNSCC patients, correlating with adverse prognosis
   and inverse associations with cytotoxic immune markers (CD8B, GZMB,
   IFNG and IL12B) and lymphocyte infiltration scores, suggesting that
   DDR1 played a key role in inducing progression and immune escape in
   HNSCC. These findings aligned with prior studies [[129]23,
   [130]31–[131]34] documenting the overexpression of DDR1 and its role in
   therapy resistance across multiple malignancies like colorectal, breast
   and lung cancer, indicating that there might be a conserved mechanism
   across malignancies and DDR1 had the potential to be a cancer target.
   Our data also echoed the study [[132]23] linking DDR1 to T cell
   exclusion in triple-negative breast cancer, although their proposed
   mechanism suggested that the extracellular domain of DDR1, rather than
   its intracellular kinase domain, aligned collagen fibers and then
   impeded immune infiltration. However, Su et al. [[133]22] demonstrated
   no DDR1-dependent alterations in collagen architecture or CD8 + T cell
   infiltration in pancreatic cancer. This means that DDR1 may have
   different regulatory mechanisms in various tumors. Based on the role of
   DDR1 in HNSCC in tumor progression and immune escape, we believed that
   DDR1 could be a promising target for the combined application of CIR,
   thereby improving the therapeutic effect of HNSCC. This hypothesis was
   confirmed by the data collected in this study, which showed that DDR1
   inhibition increased the radiosensitivity of CIR, specifically
   manifesting in the suppression of HNSCC cell proliferation and the
   delay of tumor growth in syngeneic MOC1 subcutaneous tumor models.

   Preclinical and clinical data [[134]11–[135]14] indicated that the
   combination of radiotherapy and immunotherapy had promise for improving
   treatment effectiveness and reducing recurrence by increasing the
   capacity of the immune system to identify and eradicate tumor cells,
   and avoid tumor immune tolerance. Given that CIR had an
   immunomodulatory effect even superior to that of photon radiotherapy
   [[136]16–[137]19], CIR and immunotherapy were believed to have a
   synergistic effect. In fact, the combined therapeutic advantage of CIR
   and ICIs has been confirmed in preclinical studies in several tumors,
   including melanoma and breast cancer [[138]35, [139]36]. Meanwhile,
   relevant multicenter clinical study is also underway [[140]37]. Based
   on the immune regulatory role of DDR1, this study explored the efficacy
   of combining DDR1-targeted therapy with CIR. Surprisingly, the
   knockdown of DDR1 further augmented the tumor cell immunogenicity
   induced by CIR, accompanied by increased intratumoral CD8 + T cell
   infiltration, Granzyme B and IFN-γ production, indicating an
   enhancement in the killing capacity of immune cells. In addition, the
   expression of PD1 on tumor-infiltrating CD8 + T cells increased after
   CIR and rose further upon combined DDR1 inhibition. This suggests that
   PD1 expression could serve as a potential marker for tumor response,
   and combining anti-PD1 therapy holds promise for improving the
   treatment efficacy of HNSCC. We intend to continue exploring this
   research direction in the future.

   A recent study [[141]38] highlighted the pathological relevance of
   diverse programmed cell death ways in HNSCC development, including
   entosis, apoptosis, pyroptosis, autophagy, and ferroptosis. However,
   the predominant cell death mechanism governing therapeutic responses
   remains undefined. Ferroptosis is a novel form of programmed cell
   death, a type of cell death caused by iron-dependent lipid peroxidation
   and excessive ROS production. Previous studies [[142]39–[143]43] have
   demonstrated that ferroptosis served as a crucial mechanism in the
   carcinogenesis of HNSCC, influencing the treatment efficacy, including
   radiotherapy. Song et al. [[144]42] found that radiotherapy promoted
   ferroptosis to exert tumor control effects in HNSCC by disrupting iron
   homeostasis, while glutamine blockade augmented this efficacy. Feng et
   al. [[145]44] demonstrated that SLC7A11-mediated ferroptosis inhibition
   induced NRF2-associated radioresistance in esophageal squamous cell
   carcinoma. However, given the research gap regarding how DDR1 regulates
   ferroptosis and thereby affects the radiosensitivity of carbon ions in
   the fight against HNSCC, this study regarded ferroptosis as a clue to
   explore the potential molecular mechanisms by which inhibiting DDR1
   might increase carbon ion radiosensitivity and promote immunogenic cell
   death in HNSCC. As expected, we demonstrated that DDR1 inhibition
   amplified multiple hallmarks of CIR-induced ferroptosis: intracellular
   Fe^2⁺ accumulation, ROS elevation, lipid peroxidation, and
   characteristic mitochondrial ultrastructural changes. This ferroptotic
   cascade promoted immunogenic cell death, correlating with enhanced
   antitumor immunity, a phenomenon consistent with a recent study
   [[146]45] linking ferroptosis to immunogenic death, which was related
   to endoplasmic reticulum stress.

   The PI3K/Akt/mTOR signaling axis serves as a critical regulator of
   cellular homeostasis, governing fundamental processes including growth,
   proliferation, and metabolic programming. Its dysregulation is
   frequently implicated in tumorigenesis across various malignancies
   [[147]46]. In HNSCC, this pathway emerges as a central oncogenic
   driver, promoting tumor progression, metastatic dissemination, and
   therapeutic resistance. Consequently, targeted inhibition of
   PI3K/Akt/mTOR signaling has gained recognition as a promising
   radiosensitization strategy in HNSCC management [[148]47]. Emerging
   evidence across multiple tumor models has established an inverse
   regulatory relationship between PI3K/Akt/mTOR activation and
   ferroptosis susceptibility [[149]27, [150]48]. Our findings
   corroborated this paradigm, demonstrating that DDR1 inhibition
   potentiated ferroptosis through suppression of PI3K/Akt/mTOR signaling,
   thereby enhancing therapeutic efficacy of CIR. The
   ferroptosis-enhancing effects were markedly amplified by co-treatment
   with the Akt inhibitor MK-2206, while pharmacological activation of Akt
   (SC79) or mTOR (MHY1485) effectively rescued HNSCC cells from DDR1
   knockdown/CIR-induced ferroptosis. These complementary pharmacological
   interventions confirmed the pathway’s pivotal role in modulating
   ferroptosis dynamics. Mechanistic interrogation revealed that DDR1
   engaged with Akt through 14–3-3 scaffold proteins to facilitate mTOR
   activation, thereby orchestrating downstream regulation of both
   ferroptotic sensitivity and radiation response. This molecular
   architecture aligned with previous reports by Vehlow et al. [[151]26,
   [152]49] describing DDR1’s participation in a 14–3-3-BECN1-Akt1 complex
   that potentiated pro-survival signaling through Akt-mTOR axis
   activation. Collectively, our experimental data established
   PI3K/Akt/mTOR pathway inhibition as the principal mechanism underlying
   ferroptosis induction of combined DDR1 blockade and CIR. The 14–3-3
   protein emerged as a critical molecular interface mediating DDR1-Akt
   crosstalk, presenting novel opportunities for therapeutic targeting in
   radiation oncology.

   Ferroptosis initiation and execution demonstrate profound metabolic
   coupling with lipid dynamics, particularly through the intricate
   balance of PUFAs and MUFAs [[153]50, [154]51]. The peroxidation of
   PUFA-PLs by radical species induces catastrophic membrane
   destabilization, constituting a biochemical hallmark of ferroptotic
   death [[155]51, [156]52]. Contrastingly, MUFAs exhibit
   ferroptosis-suppressive activity through dual mechanisms: ROS
   scavenging capacity and competitive displacement of PUFAs from membrane
   phospholipids, with exogenous MUFA supplementation establishing
   ferroptosis-resistant cellular states [[157]53]. The mTORC1 downstream
   effector SREBP1, a master transcriptional regulator of lipid
   biosynthesis, has emerged as both oncogenic driver and prognostic
   biomarker across malignancies [[158]54]. Mechanistically, SREBP1
   orchestrated fatty acid metabolic reprogramming through transcriptional
   activation of SCD1 [[159]55], a pivotal enzyme converting SFAs to
   cytoprotective MUFAs that confer ferroptosis resistance [[160]28].

   Multiple analyses showed that SCD1 overexpression was associated with
   unfavorable outcomes in various tumor types [[161]56–[162]58]. Our
   experimental data corroborated these findings, demonstrating that
   combined DDR1 inhibition and CIR effectively suppressed the SREBP1-SCD1
   axis to potentiate ferroptosis. This pro-death phenotype was reversed
   through either SCD1 overexpression or exogenous OA supplementation,
   confirming the critical anti-ferroptotic role of SCD1 and MUFA
   biosynthesis in HNSCC survival. Importantly, we identified mTORC1
   signaling as the regulatory nexus coordinating this metabolic process,
   as the mTOR agonist MHY1485 effectively rescued the suppression of
   SREBP1/SCD1 protein expression induced by DDR1 knockdown and CIR. These
   findings positioned the PI3K/Akt/mTOR pathway as an essential regulator
   of SREBP1-mediated MUFA lipogenesis, with combinatorial DDR1 inhibition
   and CIR disrupting this pro-survival metabolic circuit to drive
   ferroptotic elimination of tumor cells.

   Our findings established DDR1 as a promising therapeutic target for
   HNSCC, particularly for enhancing CIR efficacy. Based on this, a highly
   selective DDR1 inhibitor 7rh was employed to investigate its ability to
   sensitize tumor cells to CIR. This inhibitor has demonstrated
   significant therapeutic potential in various human cancer models,
   including pancreatic cancer, non-small cell lung cancer, and gastric
   cancer [[163]59–[164]61]. Our results indicated that 7rh effectively
   inhibited tumor growth both in vitro and in vivo, and promoted
   ferroptosis. Notably, 7rh combined with CIR significantly augmented
   CD8 + T cell tumor infiltration, suggesting dual modulation of
   ferroptotic death and antitumor immunity. These findings provided a
   compelling rationale for clinical trials evaluating 7rh and CIR
   combination therapy in HNSCC.

   While mechanistically insightful, this work has three key limitations.
   First, clinical validation of DDR1 inhibition combined with CIR remains
   pending, and such trials would critically bridge preclinical and
   translational research. Second, single-cell profiling could further
   elucidate microenvironmental remodeling by DDR1 blockade and CIR.
   Third, given that DDR1 knockdown attenuated tumor invasion and
   migration, our focus on survival/proliferation mechanisms warrants
   expanded investigations into DDR1’s role in metastatic regulation.
   Addressing these gaps will be prioritized in subsequent studies.

   In summary, this work positioned DDR1 targeting as a novel therapeutic
   strategy to potentiate CIR in HNSCC through immunogenic ferroptosis
   induction. Our findings advocate for clinical exploration of DDR1
   inhibitors as radiosensitizing adjuvants, particularly for refractory
   or recurrent HNSCC where conventional therapies fail.

Supplementary Information

   Below is the link to the electronic supplementary material.
   [165]Supplementary Material 1^ (25.6KB, docx)
   [166]Supplementary Material 2^ (19.2MB, docx)

Acknowledgements