UTMD Suppresses CSF‐1 to Drive Macrophage‐Mediated Ferroptosis in Papillary Thyroid Carcinoma Authors: Yanqiang Niu, Zhenqiang Yao, Minghua Zhang, Jiayan Zhang, Yushan Li, Jinhui Xie, Zubang Zhou Published: 2026 ABSTRACT ABSTRACTThe tumor microenvironment (TME) of Papillary Thyroid Carcinoma (PTC), particularly the role of tumor‐associated macrophages (TAMs), contributes to therapeutic resistance. This study investigated whether ultrasound‐targeted microbubble destruction (UTMD), a non‐invasive physical modality, could reprogram the TME through modulation of TAMs and induction of ferroptosis. Transcriptomic profiling identified CSF‐1 as a top target suppressed by UTMD, which was validated by ELISA, Western blot, and qRT‐PCR. UTMD time‐dependently inhibited CSF‐1 expression and secretion, consequently attenuating PTC‐induced M2 macrophage polarization as evidenced by decreased CD206+ cells and M2 marker expression. Exogenous CSF‐1 rescued this effect, confirming the specificity of the mechanism. Functionally, M1 macrophages selectively induced ferroptotic death in PTC cells, whereas M2 macrophages promoted viability and ferroptosis resistance via GPX4 upregulation and suppression of lipid peroxidation. Critically, UTMD overcame M2‐mediated protection by downregulating CSF‐1, thereby sensitizing PTC cells to ferroptosis both in vitro and in vivo. In a subcutaneous xenograft model, UTMD significantly inhibited tumor growth, reduced intratumoral CSF‐1 and M2‐TAMs, increased M1‐TAMs, and triggered ferroptosis hallmarks including mitochondrial shrinkage and GPX4 downregulation. These findings establish that UTMD suppresses PTC‐derived CSF‐1 to drive TAM repolarization from M2 to M1, consequently sensitizing tumor cells to ferroptosis. This dual reprogramming of immune and metabolic TMEs positions UTMD as a promising non‐pharmacological strategy for refractory PTC.Keywords: colony‐stimulating factor 1, ferroptosis, papillary thyroid carcinoma, tumor microenvironment, ultrasound‐targeted microbubble destruction Abbreviations Adolescents and Young Adults Colony‐stimulating factor 1 granulocyte colony‐stimulating factor Papillary Thyroid Carcinoma reactive oxygen species tumor‐associated macrophages Thyroid carcinoma The tumor microenvironment ultrasound‐targeted microbubble destruction Introduction Thyroid carcinoma (TC) is one of the most common malignancies of the head and neck and arises from thyroid follicular epithelial cells [1]. According to GLOBOCAN estimates, approximately 237,000 new cases of TC were diagnosed among adolescents and young adults (AYAs) worldwide in 2022. TC is now the second most common cancer among female AYAs and the most common among male AYAs, and its global burden continues to increase [2, 3, 4]. Papillary thyroid carcinoma (PTC) is the predominant histological subtype, accounting for nearly 90% of all thyroid cancer cases [5]. Although PTC is generally associated with a favorable prognosis, a subset of patients develops locally invasive disease, lymph node metastasis, or distant dissemination and shows limited response to radioactive iodine therapy and tyrosine kinase inhibitors. Such resistance is closely linked to the dynamic evolution of the tumor microenvironment (TME) and its adaptive immunosuppressive mechanisms [6, 7]. Among the cellular components of the TME, tumor‐associated macrophages (TAMs) have emerged as important regulators of PTC progression because of their plasticity and functional diversity. TAMs are broadly classified into antitumor M1 macrophages and protumorigenic M2‐polarized macrophages. M2‐polarized TAMs promote PTC invasion and metastasis by enhancing angiogenesis, suppressing T‐cell function, and remodeling the extracellular matrix [8]. Clinicopathological studies have shown that increased infiltration of M2‐like TAMs in PTC tissues is associated with tumor recurrence, lymph node metastasis, and poor overall survival. These findings suggest that M2‐TAMs may represent a potential therapeutic target for reversing the immunosuppressive microenvironment and improving treatment response in PTC [9, 10]. Colony‐stimulating factor 1 (CSF‐1) is a key cytokine that regulates TAM polarization [11]. Upon binding to its receptor, CSF‐1R, CSF‐1 activates downstream pathways, including PI3K/Akt and STAT3, leading to upregulation of M2‐associated markers such as CD163, CD206, and Arginase‐1 [12, 13]. In PTC, tumor cells secrete CSF‐1 in a paracrine manner, and this ligand engages CSF‐1R on macrophages and triggers RAS/MAPK and PI3K/AKT signaling [13, 14]. These pathways support macrophage survival and proliferation and drive differentiation toward an M2‐like phenotype. Accordingly, the CSF‐1/CSF‐1R axis has emerged as a promising target for TAM reprogramming and reversal of the immunosuppressive TME. Preclinical studies have shown that small‐molecule inhibitors or neutralizing antibodies against CSF‐1R can shift TAMs toward an M1‐like phenotype and suppress tumor growth, supporting CSF‐1 as a potential therapeutic target [14, 15]. However, systemic inhibition of CSF‐1R is associated with dose‐limiting toxicities, including peripheral edema, elevated liver enzymes, and compensatory release of granulocyte colony‐stimulating factor (G‐CSF) [16, 17]. These limitations highlight the need for strategies that selectively reduce intratumoral CSF‐1 [18]. Ultrasound‐targeted microbubble destruction (UTMD) has recently attracted attention as a non‐invasive physical intervention with demonstrated efficacy in targeted drug delivery and gene transfection [19]. UTMD uses mechanical shear forces and sonoporation generated by microbubble cavitation to transiently disrupt cell membrane permeability and trigger intracellular signaling, including activation of mechanosensitive ion channels, Ca2+ influx, and mitochondrial reactive oxygen species (ROS) production [20, 21, 22]. Notably, a recent study showed that even a single session of UTMD significantly reduced IL‐6, CCL2, and CXCL8 transcription in BRAF‐V600E mutant PTC cells, suggesting that UTMD may directly reprogram the secretory profile of tumor cells [23]. However, whether UTMD can specifically suppress CSF‐1 expression and secretion and thereby interrupt M2 polarization of TAMs remains unclear. Ferroptosis is an iron‐dependent, lipid peroxidation‐driven form of regulated cell death and has emerged as a promising therapeutic target in PTC [24]. Its execution depends on failure of the intracellular antioxidant system and accumulation of lethal lipid peroxides. Distinct macrophage subsets exert opposing effects on ferroptosis. M2‐polarized TAMs enhance tumor cell resistance to oxidative stress and ferroptosis by overexpressing the cystine transporter, secreting ferritin, and facilitating CD36‐mediated lipid uptake. In contrast, M1‐TAMs can sensitize tumor cells to ferroptosis by releasing nitric oxide and reactive oxygen species, potentially acting in synergy with exogenous ferroptosis inducers [25, 26, 27]. Therefore, reducing the M2/M1 ratio may lower the ferroptotic threshold in PTC cells and improve the efficacy of ferroptosis‐based strategies in solid tumors [28, 29]. Although the role of CSF‐1 signaling in TAM polarization is well established, its link to ferroptosis susceptibility in PTC remains unclear. Whether UTMD can coordinate macrophage remodeling and ferroptosis activation through regulation of CSF‐1 secretion has not yet been determined. In this study, we investigated whether UTMD regulates macrophage polarization and ferroptosis in PTC through the CSF‐1 axis. We found that UTMD inhibited CSF‐1 transcription and secretion in PTC cells, disrupted CSF‐1/CSF‐1R signaling, reduced M2‐like TAM polarization, and relieved TAM‐mediated suppression of ferroptosis. Together with the pro‐ferroptotic activity of M1‐like TAMs, these changes promoted lipid peroxidation‐dependent ferroptosis in PTC cells. These findings provide a rationale for targeting the tumor microenvironment in PTC using a non‐pharmacological approach. Materials and Methods Reagents and Antibodies FITC Anti‐Human CD68 (AB‐F1299L), PE Anti‐Human CD86 (E‐AB‐F1012D), and APC Anti‐Human CD206 (E‐AB‐F1161E) were purchased from Elabscience Biotechnology Co. Ltd. (Wuhan, China); Recombinant Human M‐CSF (rhCSF‐1, HY‐P72948), Phorbol 12‐myristate 13‐acetate (PMA, HY‐18739), Interferon‐γ (IFN‐γ, HY‐P4717), Lipopolysaccharide (LPS, HY‐D1056), Interleukin‐4 (IL‐4, HY‐P70445), Interleukin‐13 (IL‐13, HY‐P72795), Ferrostatin‐1 (Fer‐1, HY‐100579), Necrostatin‐1 (Nec‐1, HY‐15760), the pan‐caspase inhibitor Z‐VAD‐FMK (HY‐16658B), the caspase‐1 inhibitor Ac‐FLTD‐CMK (HY‐111675), and the glutathione peroxidase 4 (GPX4) inhibitor RSL3 (HY‐100218A) were purchased from MedChemExpress (MCE, USA). Sulfur hexafluoride‐filled microbubbles (SonoVue) were purchased from Bracco (Milan, Italy). Primary antibodies against CSF‐1 (bs‐4910R) and GPX4 (bs‐34318R) were purchased from Bioss Antibodies (Beijing, China). Antibodies against SLC7A11 (26864‐1‐AP), ACSL4 (22401‐1‐AP), FTH1 (11682‐1‐AP), and TFR1 (10084‐2‐AP) were obtained from Proteintech Group Inc. (Wuhan, China). GSH assay kit and MDA assay kit were purchased from Nanjing Jiancheng (Nanjing, China). Fluo‐4 calcium assay kit and DCFH‐DA‐based ROS assay kit were purchased from Beyotime (Shanghai, China). Cell Culture The human THP‐1 monocytic cells and PTC cell line BCPAP were purchased from Procell (Wuhan, China). All cells were cultured in RPMI 1640 medium, which was prepared with 10% Fetal Bovine Serum, 1% penicillin–streptomycin, and 89% DMEM basal medium, under conditions of 37°C and 5% CO2. All cell lines were authenticated using short tandem repeat (STR) profiling within the past 3 years. All experiments were performed with mycoplasma‐free cells. UTMD Treatment BCPAP cells in the logarithmic growth phase were seeded into six‐well plates at a density of 5 × 105 cells per well. When the cells reached the appropriate growth stage, they were randomly assigned to four groups: Control (untreated), ultrasound treatment (UT), microbubble treatment (MB), and ultrasound‐targeted microbubble destruction (UTMD) groups. Sulfur hexafluoride (SF6) microbubbles were prepared according to the manufacturer's instructions. Briefly, 5 mL of sterile normal saline was added into the vial containing lyophilized microbubble powder, followed by vigorous shaking until a uniform milky suspension was obtained. The prepared microbubble suspension was used within 6 h after reconstitution. For the UT group, cells were exposed to ultrasound irradiation using an ultrasound probe positioned above the culture plate with coupling medium. The ultrasound parameters were set as follows: mechanical index (MI) 1.2, irradiation duration 90 s, duty cycle 20%, and pulse repetition frequency (PRF) 1 kHz. For the MB group, cells were incubated with SF6 microbubble suspension at a final concentration of 10% (v/v) according to the manufacturer's recommended usage conditions, without ultrasound exposure. For the UTMD group, microbubble suspension was added to the culture medium immediately before ultrasound irradiation. Ultrasound‐targeted microbubble destruction was performed using the same ultrasound parameters as the UT group (MI 1.2, irradiation time 90 s, duty cycle 20%, and PRF 1 kHz), with the ultrasound probe positioned above the culture plate to induce microbubble cavitation. After treatment, all groups were incubated for an additional 24 h before subsequent experiments. A representative image of the in vitro UTMD experimental setup is provided in Figure S1. In Vivo Safety Evaluation of UTMD Treatment BALB/c‐nu nude mice (4–6 weeks old, 18–20 g) were obtained from Beijing Vital River Laboratory Animal Technology Co. Ltd. and maintained under specific pathogen‐free conditions. All animal procedures were approved by the Research Ethics Committee of Gansu Provincial Hospital (Approval No. 2026‐044) and conducted in accordance with institutional guidelines. Mice bearing subcutaneous BCPAP xenograft tumors were randomly assigned to four groups: Control, UT, MD, UTMD (n = 8 per group). The same cohort of tumor‐bearing mice used for safety evaluation was subsequently used for the mechanistic and therapeutic efficacy studies described in Section 2.20. Given tissue attenuation differences between the in vitro and in vivo settings, in vivo UTMD parameters were adjusted from the in vitro protocol (Section 2.3). Mice in the UT and UTMD groups received percutaneous ultrasound irradiation directly over the tumor region using an ultrasound probe positioned above the skin with coupling gel, at a mechanical index of 1.5, an irradiation time of 3 min, and a duty cycle of 50%. Mice in the MD and UTMD groups received an intravenous injection of SonoVue microbubble suspension via the tail vein at a dose of 150 μL per mouse, administered immediately prior to ultrasound irradiation. Treatments were administered once weekly for a total of four treatments over a 4‐week period. Mice in the Ctrl group received no treatment. At the end of the treatment period, mice were euthanized by CO2 inhalation, and the liver and kidney were harvested, fixed in 4% paraformaldehyde, embedded in paraffin, and sectioned at 4–5 μm. Sections were stained with hematoxylin and eosin (H&E) and examined under a light microscope to evaluate potential UTMD‐induced histopathological changes in major organs. Measurement of Intracellular Ca2+ Levels Intracellular Ca 2+ levels were measured using a Fluo‐4 calcium assay kit according to the manufacturer's instructions. Briefly, BCPAP cells were incubated with Fluo‐4 AM staining solution at 37°C for 30 min in the dark, following the indicated treatments (Ctrl, UT, MD, or UTMD). Fluorescence images were captured using a fluorescence microscope (Ex/Em = 490/525 nm), and fluorescence intensity was quantified using ImageJ software. Measurement of Intracellular ROS Levels Intracellular reactive oxygen species (ROS) levels were assessed using a DCFH‐DA‐based ROS assay kit. Cells were incubated with 10 μM DCFH‐DA at 37°C for 20 min in the dark following the indicated treatments, washed with serum‐free medium to remove excess probe, and fluorescence images were captured using a fluorescence microscope (Ex/Em = 488/525 nm). Fluorescence intensity was quantified using ImageJ software. Transcriptomic Analysis (RNA‐Seq) Following the designated interventions, total RNA was extracted from cells of the control and U‐M groups. mRNA was enriched using oligo (dT) beads and subsequently fragmented into approximately 300 bp segments. First‐strand cDNA was synthesized using random hexamer primers, followed by second‐strand synthesis. The cDNA libraries were constructed, PCR‐amplified, and size‐selected (~450 bp). Library quality and concentration were assessed using an Agilent 2100 Bioanalyzer. Equimolar amounts of indexed libraries from each sample were pooled, denatured, and subjected to paired‐end (PE) sequencing on an Illumina platform using next‐generation sequencing (NGS) technology. Raw sequencing reads were quality‐filtered to obtain clean data, which were then aligned to the appropriate reference genome. Gene expression levels were quantified based on the alignment results. Differential gene expression analysis between groups was performed, followed by functional enrichment analysis and cluster analysis to identify significantly altered biological pathways and processes. Cell Transfection CSF‐1 overexpressing and empty lentiviral vectors were constructed by Genma Bio (Shanghai, China). For overexpression, BCPAP cells were seeded in 6‐well plates and transfected at 70%–80% confluency. Cells were transfected with a plasmid encoding the human CSF1 gene (pcDNA3.1‐CSF1) using Lipofectamine 3000 reagent, strictly following the manufacturer's protocol. An empty vector was transfected in parallel as a negative control. Cells were harvested 48 h post‐transfection for subsequent functional assays. The transfection efficiency and CSF‐1 overexpression were confirmed by Western blot. In Vitro Differentiation and Polarization of Macrophages Human monocytic THP‐1 cells were differentiated into macrophages (M0) by treatment with 150 nM PMA for 48 h. Subsequently, the M0 macrophages were polarized into distinct phenotypes: for classical M1 polarization, cells were stimulated with a combination of 20 ng/mL IFN‐γ and 100 ng/mL LPS for 24 h; for alternative M2 polarization, cells were stimulated with 20 ng/mL IL‐4 and 20 ng/mL IL‐13 for the same duration. Following polarization, cells were harvested for downstream analyses. The efficiency of polarization was validated by flow cytometry. Co‐Culture System A Transwell co‐culture system was employed to evaluate the paracrine effect of PTC cells on macrophage polarization. M1‐polarized macrophages, generated as described in Section 2.9, were seeded in the upper chambers. The lower chambers contained different groups of BCPAP cells as follows: (1) Control (M1 alone): RPMI‐1640 medium only in the lower chamber; (2) UTMD group: BCPAP cells that had been subjected to UTMD treatment 24 h prior to co‐culture; (3) UTMD + rhCSF‐1 group: UTMD‐treated BCPAP cells with the addition of 50 ng/mL recombinant human CSF‐1. After 48 h of co‐culture, macrophages in the upper chambers were carefully harvested. The impact on macrophage phenotype was assessed through complementary assays. Flow Cytometry Harvested macrophages were washed with ice‐cold PBS. Subsequently, cells were stained with fluorochrome‐conjugated antibodies specific for human CD68, CD86, and CD206 for 30 min at 4°C in the dark. Corresponding isotype control antibodies were used in parallel to define the negative population and set fluorescence thresholds. Cells were analyzed using a flow cytometer, and data were processed using FlowJo 10.8 software. Functional Phagocytosis Assay Following a 48‐h intervention with conditioned medium derived from various BCPAP cell groups, macrophages were incubated with FITC‐labeled dextran (1 mg/mL, molecular weight 40,000; Thermo Fisher Scientific) in serum‐free medium at 37°C for 1 h. The reaction was terminated by placing the cells on ice and extensively washing them with ice‐cold PBS to remove non‐internalized dextran. Subsequently, the cells were collected and immediately subjected to flow cytometric analysis. Phagocytic activity was quantified as the MFI of FITC within the live macrophage population, which was gated as CD68+ cells. Cell Death Mode Analysis by Flow Cytometry Following seeding, BCPAP cells were co‐cultured with M1 macrophages that had been pretreated with specific inhibitors: Ferrostatin‐1 (5 μM), Z‐VAD‐FMK (20 μM), Necrostatin‐1 (30 μM), and Ac‐FLTD‐CMK (50 μM). After 48 h of treatment, both floating and adherent cells were collected. The cells were then stained with FITC‐conjugated Annexin V and PI. Immediately following staining, the cells were subjected to flow cytometric analysis. Cell Proliferation Assays Cell viability was assessed using the CCK‐8 assay. Briefly, cells were seeded in 96‐well plates at appropriate densities and allowed to attach and reach the logarithmic growth phase before treatment. For co‐culture and ferroptosis‐related experiments, BCPAP cells were seeded at a density of 5 × 103 cells per well in 100 μL complete medium. After cell attachment, cells were subjected to the indicated treatments, including co‐culture with different macrophage phenotypes or treatment with the ferroptosis inducer RSL3 (0.5 μM) or the ferroptosis inhibitor Ferrostatin‐1 (1 μM). BCPAP cells cultured alone without additional treatment served as the control group. For co‐culture experiments, cell viability was evaluated at 24, 48, and 72 h after treatment. For ferroptosis induction or inhibition experiments, cells were treated for 24 h before viability assessment. To evaluate the biosafety of the UTMD parameters used in this study, the human embryonic kidney cell line HEK‐293T and the human normal thyroid follicular epithelial cell line Nthy‐ori 3‐1 were used as normal cell models. Cells were seeded at a density of 5 × 103 cells per well in 100 μL complete medium and allowed to attach and reach the logarithmic growth phase. Cells were then randomly divided into four groups: Ctrl, UT, MD, and UTMD. The UT group received ultrasound irradiation alone at a mechanical index of 1.2, irradiation time of 90 s, 20% duty cycle, and pulse repetition frequency of 1 kHz. The MD group was treated with SonoVue microbubble suspension, prepared according to the manufacturer's instructions and added to the culture medium at a final concentration of 10% (v/v) for 24 h. The UTMD group received combined ultrasound and microbubble treatment under the same optimized parameters (mechanical index 1.2, irradiation time 90 s, microbubble concentration 10%, duty cycle 20%, PRF 1 kHz; see Section 2.3). All groups were incubated for 24 h after treatment before viability assessment. At the indicated time points, 10 μL of CCK‐8 reagent was added to each well, followed by incubation at 37°C for 1 h. Absorbance was measured at 450 nm using a microplate reader. Each condition was analyzed using at least three technical replicates, and all experiments were independently repeated three times. Detection of Lipid Peroxidation by C11‐BODIPY Flow Cytometry Following seeding, BCPAP cells were co‐cultured with M1 or M2 macrophages under different intervention conditions. At the end of the treatment period, cells were washed with PBS and subsequently incubated with 2 μM C11‐BODIPY probe (S0043S, Beyotime) in serum‐free medium at 37°C for 30 min in the dark. After incubation, cells were collected, washed twice with PBS to remove unbound probe, and immediately subjected to flow cytometric analysis. Detection of Labile Fe2+ Using RhoNox‐6 Staining Following co‐culture with M2 macrophages in the presence or absence of inhibitors, BCPAP cells were collected for Fe2+detection. Cells were washed with pre‐warmed Hanks' Balanced Salt Solution and incubated with serum‐free medium containing 5 μM RhoNox‐6 probe (S1070S, Beyotime) at 37°C for 30 min in the dark. After staining, cells were trypsinized, collected by centrifugation, resuspended in ice‐cold phosphate‐buffered saline, and analyzed by flow cytometry within 1 h. RhoNox‐6 fluorescence was detected using the appropriate channel according to its emission spectrum. Elisa The concentration of CSF‐1 (EZ0444, Boster) was determined strictly following the manufacturer's instructions and absorbance was measured at 450 nm. Quantitative Real‐Time PCR (qRT‐PCR) Total RNA was extracted from tumor tissues and cells using the Trizol method and reverse‐transcribed into cDNA. Amplification was performed using the SYBR Green method. β‐actin was used as the internal reference gene, and the relative expression levels of target genes were calculated using the 2−ΔΔCt method. Homo sapiens : β‐actin‐F: 5′‐GGATTCCTATGTGGGCGACGA‐3′. β‐actin‐R: 5′‐GCGTACAGGGATAGCACAGC‐3′. CSF‐1‐F: 5′‐CGTTTGCTGAAGGCTTGGAA‐3′. CSF‐1‐R: 5′‐AGGACCCAGGCAAACTTTCA‐3′. Arg1‐F: 5′‐GTCTGTGGGAAAAGCAAGCG‐3′. Arg1‐R: 5′‐CACCAGGCTGATTCTTCCGT‐3′. CD163‐F: 5′‐ATTGCAGTCGGGATCCTTGG‐3′. CD163‐R: 5′‐GGTATCTTAAAGGCTCACTGGGT‐3′. MRC1‐F: 5′‐CCAAACGCCTTCATTTGCCA‐3′. MRC1‐R: 5′‐ACCTTCCTTGCACCCTGATG‐3′ Mus musculus : β‐actin‐F: 5′‐GCCTCACTGTCCACCTTCCA‐3′. β‐actin‐R: 5′‐CATCGTACTCCTGCTTGCTGAT‐3′. CSF‐1‐F: 5′‐CCTTCTTCGACATGGCTGGG‐3′. CSF‐1‐R: 5′‐GTTCTGACACCTCCTTGGCA‐3′. GPX4‐F: 5′‐CCGTCTGAGCCGCTTACTT‐3′. GPX4‐R: 5′‐GTGACGATGCACACGAAACC‐3′. SLC7A11‐F: 5′‐AAATACGGAGCCTTCCACGAG‐3′. SLC7A11‐R: 5′‐CGGTCGTGACTTCCCCTTTG‐3′. ACSL4‐F: 5′‐GTCCGCAAGAGGTAGATCGG‐3′. ACSL4‐R: 5′‐AGGGCAAGAGAAACAGTGGG‐3′. FTH1‐F: 5′‐TTTGCAACTTCGTCGTTCCG‐3′. FTH1‐R: 5′‐CAGGTTGATCTGGCGGTTGA‐3′. TFR1‐F: 5′‐GAGAGTCACCACGCTGAGGC‐3′. TFR1‐R: 5′‐ATGACAATGGTTCCCCACCAA‐3′. Western Blotting Total proteins were extracted by lysing cells with RIPA lysis buffer supplemented with protease and phosphatase inhibitors. The protein lysates were separated by 10% SDS‐PAGE and subsequently transferred onto PVDF membranes. Following blocking with 5% non‐fat milk for 1 h at room temperature, the membranes were incubated with primary antibodies overnight at 4°C. After washing, the membranes were incubated with HRP‐conjugated anti‐mouse or anti‐rabbit secondary antibodies for 1 h at room temperature. Antigen–antibody complexes were finally visualized using an ECL detection system. Protein bands were quantified using ImageJ and normalized to β‐actin. Xenograft Tumor Formation Assay BALB/c‐nu nude mice (4–6 weeks old, 18–20 g) were obtained from Beijing Vital River Laboratory Animal Technology Co. Ltd. and maintained under specific pathogen‐free conditions. All animal procedures were approved by the Research Ethics Committee of Gansu Provincial Hospital (Approval No. 2026‐044) and conducted in accordance with institutional guidelines. Mice were randomly assigned to four groups: oe‐Ctrl, oe‐CSF‐1, U‐M + oe‐Ctrl, and U‐M + oe‐CSF‐1. Cells were harvested and resuspended on ice in a 1:1 mixture of PBS and Matrigel. Under aseptic conditions, 100 μL of the cell suspension was subcutaneously injected into the right axillary region of each mouse using a sterile syringe. Beginning 1 week post‐inoculation, tumor dimensions were measured every 2–3 days with digital calipers, and tumor volume was calculated based on two perpendicular diameters (length [L] and width [W]). At 4 weeks, mice were euthanized by CO2 inhalation, and tumors were collected for weighing, imaging, and further analysis. Transmission Electron Microscopy (TEM) for Ultrastructural Analysis Following primary fixation, tumor tissues were post‐fixed with 1% osmium tetroxide at room temperature for 2 h. The specimens were then dehydrated through a graded series of ethanol (50%, 70%, 90%, 100%) and pure acetone, followed by infiltration and embedding in epoxy resin. Ultrathin sections were mounted on copper grids and double‐stained with uranyl acetate and lead citrate to enhance contrast. Ultrastructural examination was performed using a transmission electron microscope. The morphological hallmarks of ferroptosis were systematically assessed, including mitochondrial shrinkage, increased membrane density, reduction or disappearance of mitochondrial cristae, and intact nuclear membranes. Tumor Histological Analyses Following deparaffinization and antigen retrieval, tissue sections were treated with 3% hydrogen peroxide for 10 min to block endogenous peroxidase activity and then blocked with 5% BSA for 30 min at room temperature. For immunohistochemistry, sections were incubated with anti‐CSF‐1 primary antibody overnight at 4°C, followed by HRP‐conjugated secondary antibody for 30 min at room temperature. Signals were visualized using a DAB substrate kit. For multiplex immunofluorescence, sections were permeabilized and blocked with 5% BSA containing 0.3% Triton X‐100, then sequentially incubated with primary antibodies against CD68, CD86, and CD206 overnight at 4°C, followed by corresponding fluorescent secondary antibodies for 1 h at room temperature in the dark. After staining, sections were dehydrated through graded ethanol, cleared in xylene, and mounted with neutral balsam for immunohistochemistry or antifade mounting medium for multiplex immunofluorescence. Images were captured using a light microscope. Determination of Glutathione (GSH) Levels The glutathione (GSH) content in tumor tissues was measured using a commercial GSH assay kit according to the manufacturer's instructions. Briefly, tumor tissues were homogenized in ice‐cold buffer and centrifuged at 3500 rpm for 10 min. The collected supernatants were used for GSH determination. The assay was based on the reaction between reduced glutathione and 5,5′‐dithiobis‐(2‐nitrobenzoic acid) (DTNB), which produces a yellow‐colored compound. The absorbance was measured at 405 nm using a microplate reader. GSH levels were calculated according to the standard curve and normalized to the total protein concentration determined by a bicinchoninic acid (BCA) protein assay. Determination of Malondialdehyde (MDA) Levels The malondialdehyde (MDA) content in tumor tissues was determined using a commercial MDA assay kit according to the manufacturer's protocol. Briefly, tumor tissue samples were homogenized and centrifuged to obtain supernatants. The MDA levels were determined based on the thiobarbituric acid (TBA) reaction, in which MDA forms a colored MDA–TBA adduct under acidic and high‐temperature conditions. The absorbance was measured at 532 nm using a microplate reader. The MDA concentration was calculated using an MDA standard curve and normalized to the total protein content. Statistical Analysis Statistical analyses were performed using GraphPad Prism 8.0. Data are presented as the mean ± standard deviation (SD) from at least three independent biological replicates for cellular experiments or from the indicated number of independent animals for animal experiments. Comparisons among multiple groups were performed using one‐way analysis of variance (ANOVA) followed by Dunnett's multiple comparisons test when multiple treatment groups were compared with a single control group. For experiments involving two independent variables, two‐way ANOVA followed by Šídák's multiple comparisons test was used for the specified pairwise comparisons. A value of p < 0.05 was considered statistically significant. Results UTMD Safely Suppresses CSF‐1 Secretion in PTC To evaluate the biosafety of the UTMD parameters used in this study prior to subsequent mechanistic investigations, we first assessed the cytotoxicity of ultrasound and microbubble treatment in normal cells and examined the histopathology of major organs in vivo. CCK‐8 assays showed no significant differences in cell viability among the Ctrl, UT, MD, and UTMD groups in either HEK‐293 T or Nthy‐ori 3‐1 cells (Figure 1L), indicating that the UTMD parameters used in this study are not cytotoxic to normal cells. Consistent with this, H&E staining of liver and kidney tissues from mice in the Ctrl, UT, MD, and UTMD groups revealed no obvious histopathological abnormalities, with intact hepatic lobular architecture and preserved glomerular and tubular structures, and no evidence of inflammatory infiltration, necrosis, or structural disruption in the UTMD group compared with the Ctrl group (Figure 1K). These results confirm that the UTMD parameters applied in this study are biosafe, supporting their use in subsequent experiments. UTMD downregulates CSF‐1 expression in PTC cells and tumor tissues. (A) Sankey diagram illustrating the network between differentially expressed genes and their associated signaling pathways following UTMD treatment, with Csf1 highlighted; bubble plot showing KEGG pathway enrichment analysis of the associated genes, with gene ratio on the x‐axis, statistical significance (−log10 p value) indicated by color, and gene count indicated by dot size. (B) Heatmap of differentially expressed immune‐related genes between control (C) and UTMD‐treated (U_M) groups, with Csf1 highlighted. (C) ELISA quantification of CSF‐1 concentration at 0, 12, 24, and 48 h following the indicated treatments (Ctrl, UT, MD, U‐M). (D) Representative images of xenograft tumors from the different treatment groups. (E, F) Representative Western blot image and quantitative analysis of CSF‐1 protein levels in xenograft tumor tissues. (H, G) Representative Western blot image and quantitative analysis of CSF‐1 protein levels in cultured PTC cells. (I, J) qRT‐PCR analysis of CSF1 mRNA levels in xenograft tumor tissues and cultured cells, respectively. (K) Representative H&E staining of liver and kidney tissues from mice in the Ctrl, UT, MD, and UTMD groups, showing no obvious histopathological abnormalities (scale bar: 100 μm). (L) Cell viability of HEK‐293 T and Nthy‐ori 3–1 cells following the indicated treatments (Ctrl, UT, MD, UTMD), assessed by CCK‐8 assay, showing no significant cytotoxicity. (M) Quantitative analysis of relative ROS and Ca2+ levels in BCPAP cells following the indicated treatments. (N) Representative fluorescence images of intracellular ROS (detected using the DCFH‐DA probe) and Ca2+ (detected using the Fluo‐4 AM probe) levels in BCPAP cells following the indicated treatments. Scale bar: 50 μm. Data are presented as mean ± SD (n = 3 independent biological replicates). *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. To assess the effect of UTMD on macrophage‐related signaling in PTC, transcriptomic analysis was performed. UTMD led to the downregulation of a panel of immune‐related genes, among which Csf1 was one of the most prominently suppressed, suggesting reduced CSF‐1 production in PTC cells (Figure 1A,B). To validate this finding, conditioned media and cells were collected at 0, 12, 24, and 48 h following treatment with ultrasound alone, microbubbles alone, or the combined UTMD procedure. ELISA showed that CSF‐1 levels gradually increased over time in the control group, whereas UTMD induced a time‐dependent decline. In contrast, ultrasound or microbubbles alone had minimal effects, indicating that effective suppression of CSF‐1 requires their combined action (Figure 1C). Consistently, representative images of xenograft tumors showed that UTMD markedly reduced tumor size compared with the control group, while ultrasound or microbubbles alone exerted only limited effects (Figure 1D), supporting a stronger antitumor effect of the combined intervention. Further analyses by Western blot and quantitative RT‐PCR demonstrated that, at 48 h, UTMD significantly reduced both CSF‐1 protein and CSF1 mRNA levels in cultured cells and xenograft tissues. This reduction was substantially greater than that observed with either ultrasound or microbubbles alone (Figure 1E–J). Taken together, these results indicate that UTMD suppresses CSF‐1 expression and secretion in PTC through the coordinated physical effects of ultrasound and microbubbles, accompanied by inhibition of tumor growth. To further investigate the upstream mechanism by which UTMD suppresses CSF‐1 transcription, we measured intracellular Ca2+ and ROS levels in BCPAP cells following the indicated treatments. Both intracellular Ca2+ and ROS levels were significantly and progressively increased in the UT, MD, and UTMD groups compared with the Ctrl group, with the UTMD group showing the greatest increase (Figure 1M,N). These findings suggest that UTMD‐induced sonoporation and cavitation elevate intracellular calcium influx and oxidative stress, which may contribute to the transcriptional suppression of CSF‐1. UTMD Reduces M2 Macrophage Polarization by Inhibiting CSF‐1 Secretion M1 and M2 macrophage phenotypes were first established from THP‐1 cells and confirmed by flow cytometry. M0 macrophages were predominantly CD68+CD86−CD206−, M1 macrophages showed high CD86 with minimal CD206, and M2 macrophages showed the opposite pattern (Figure 2A). To assess whether UTMD‐treated PTC cells alter macrophage phenotype through changes in their secretome, M1‐polarized macrophages were exposed to conditioned media from different BCPAP groups. Compared with untreated M1 macrophages, conditioned medium from wild‐type BCPAP cells reduced the M1 phenotype and promoted an M2‐like shift, as shown by decreased CD86 and increased CD206. This finding is consistent with a previous report showing that conditioned medium from BCPAP cells induces M2‐like polarization of human monocytes via a PGE2‐dependent mechanism, further supporting the immunomodulatory role of thyroid cancer cell‐secreted factors in shaping macrophage phenotype [30]. This effect was largely reversed by conditioned medium from UTMD‐treated BCPAP cells, whereas supplementation with exogenous recombinant human CSF‐1 (rhCSF‐1) partially restored the M2‐like phenotype (Figure 2B). Consistently, qPCR showed that wild‐type BCPAP‐conditioned medium increased the expression of M2‐associated genes (Arg1, CD163, and MRC1), while conditioned medium from UTMD‐treated cells reduced their expression; rhCSF‐1 again restored these changes (Figure 2C). Phagocytosis assays showed that M1 macrophages displayed strong phagocytic activity, which was further enhanced by wild‐type BCPAP‐conditioned medium. UTMD‐conditioned medium normalized this response, whereas rhCSF‐1 partially reversed the effect (Figure 2D,E). To further define the role of CSF‐1, we established a stable CSF‐1‐overexpressing BCPAP model. Western blot and qPCR confirmed increased CSF‐1 expression in oe‐CSF‐1 cells (Figure 2F–H). ELISA showed that UTMD reduced CSF‐1 secretion in both control and oe‐CSF‐1 cells, although the absolute level remained higher in the oe‐CSF‐1 group (Figure 2I). Western blot and qPCR further showed that UTMD decreased intracellular CSF‐1 protein and mRNA in all groups, but expression remained higher in UTMD‐treated oe‐CSF‐1 cells than in controls (Figure 2J–L). Finally, co‐culture experiments showed that UTMD significantly reduced M2 polarization, as reflected by the CD86/CD206 ratio, whereas co‐culture with oe‐CSF‐1 cells produced the strongest M2 shift. This effect was attenuated, but not abolished, by subsequent UTMD treatment (Figure 2M–O). UTMD inhibits tumor‐promoting M2 macrophage polarization in PTC by suppressing CSF‐1. (A) Flow cytometric analysis of surface markers confirming successful polarization of THP‐1‐derived macrophages into M0, M1, and M2 phenotypes. (B) Flow cytometric analysis of CD86+ and CD206+ populations in M1 macrophages treated with conditioned media. (C) qPCR analysis of Arg1, CD163, MRC1. (D, E) Representative immunofluorescence images and quantitative analysis of the phagocytic activity of M1 macrophages treated with conditioned media from the indicated groups, assessed by uptake of FITC‐labeled apoptotic tumor cells (green); nuclei were counterstained with DAPI (blue). (F, G) Representative Western blot images and quantitative analysis of CSF‐1 protein levels. (H) qRT‐PCR analysis of CSF1 mRNA levels (I) ELISA analysis of CSF‐1 levels in cell culture supernatants. (J, L) Representative Western blot images and quantitative analysis of intracellular CSF‐1 protein levels. (K) qPCR analysis of CSF1 mRNA levels. (M–O) Flow cytometric analysis of macrophage polarization following co‐culture with BCPAP cells subjected to the indicated treatments. Scale bar: 50 μm. Data are presented as mean ± SD (n = 3, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001). M1 Macrophages Induce BCPAP Cell Death in a Ferroptosis‐Dependent Manner To determine the mode of cell death induced by M1 macrophages in BCPAP cells, Annexin V‐FITC/PI staining was performed in the presence of specific inhibitors. Flow cytometry showed that M1 co‐culture induced marked cell death, which was significantly rescued by the ferroptosis inhibitor Ferrostatin‐1 (Fer‐1), whereas the apoptosis inhibitor Z‐VAD‐FMK, the necroptosis inhibitor Necrostatin‐1 (Nec‐1), and the pyroptosis inhibitor Ac‐FLTD‐CMK produced only minimal effects (Figure 3A). Consistently, CCK‐8 assays showed that M1 macrophages reduced BCPAP cell viability, and this effect was further enhanced by the ferroptosis inducer RSL3. In contrast, M2 macrophages increased cell viability and partially protected BCPAP cells from RSL3‐induced death, whereas Fer‐1 reversed the decrease in viability caused by M1 co‐culture (Figure 3B). C11‐BODIPY staining further showed that M1 macrophages increased lipid peroxidation, whereas M2 macrophages suppressed it. RSL3 further enhanced lipid peroxidation, while Fer‐1 attenuated the effect induced by M1 co‐culture (Figure 3C,D). Western blot analysis showed that GPX4 was downregulated by M1 co‐culture but upregulated by M2 co‐culture. RSL3 further reduced GPX4 expression, whereas Fer‐1 partially restored GPX4 levels in M1‐treated cells (Figure 3E,F). In parallel, M1 co‐culture increased intracellular labile Fe2+ levels, whereas M2 co‐culture reduced them. RSL3 promoted Fe2+ accumulation, while Fer‐1 suppressed the increase induced by M1 macrophages (Figure 3G,H). Together, these findings indicate that M1 macrophage‐induced death of BCPAP cells is predominantly ferroptotic. M1 Macrophages induce BCPAP cell death in a ferroptosis‐dependent manner. (A) Flow cytometry analysis of BCPAP cell death by Annexin V‐FITC/PI staining after treatment with different cell death pathway inhibitors. (B) CCK‐8 assay showing the viability of BCPAP cells co‐cultured with M1 or M2 macrophages, or treated with the ferroptosis inducer RSL3 (0.5 μM, 24 h) or the ferroptosis inhibitor Ferrostatin‐1 (Fer‐1, 1 μM, 24 h), alone or in combination, as indicated. (C) Representative fluorescence images of lipid peroxidation in BCPAP cells detected using the C11‐BODIPY fluorescent probe; reduced (red) and oxidized (green) forms are shown separately, with merged images. (D) Quantitative analysis of the C11‐BODIPY red/green fluorescence ratio shown in (C). (E, F) Representative Western blot images and quantitative analysis of GPX4 protein levels, normalized to β‐Actin. (G, H) Representative fluorescence images and quantitative analysis of intracellular labile ferrous iron (Fe2+) levels detected using RhoNox‐6. Scale bar: 50 μm. Data are presented as mean ± SD. (n = 3, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001). UTMD Promotes PTC Ferroptosis by Downregulating CSF‐1 to Regulate Macrophage Repolarization To determine whether UTMD modulates the M2 macrophage‐mediated protection of PTC cells through the CSF‐1 axis, UTMD‐treated BCPAP cells were co‐cultured with M2 macrophages. Co‐culture with M2 macrophages significantly increased BCPAP cell viability, reduced lipid peroxidation, upregulated GPX4, and decreased intracellular Fe2+ levels, supporting a ferroptosis‐resistant state in PTC cells. UTMD alone produced the opposite effects, with reduced cell viability and enhanced ferroptosis‐related changes. Notably, when UTMD‐treated BCPAP cells were co‐cultured with M2 macrophages, the protective effect of M2 macrophages was markedly attenuated. Supplementation with exogenous rhCSF‐1 partially restored this protection, shifting these changes toward a ferroptosis‐resistant pattern (Figure 4A–G). UTMD counteracts M2 macrophage‐mediated ferroptosis protection in PTC cells by suppressing CSF‐1 signaling. (A) CCK‐8 assay for cell viability. (B, C) Lipid peroxidation measured by C11‐BODIPY fluorescence. (D, E) Expression of GPX4 proteins analyzed by Western blot. (F–I) Intracellular Fe2+ levels detected by FerroOrange probe. (J, K) Western blot analysis of ferroptosis‐associated proteins. Scale bar: 50 μm. Data are presented as mean ± SD. (n = 3, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001). To further define the role of CSF‐1 in this process, we established an oe‐CSF‐1 cell model. FerroOrange staining showed that intracellular Fe2+ levels remained low in both oe‐CSF‐1 cells and the oe‐CSF‐1 + M2 co‐culture group, whereas UTMD significantly increased Fe2+ accumulation. This effect was partially reversed by either M2 co‐culture or CSF‐1 overexpression (Figure 4H,I). Consistently, Western blot analysis showed that GPX4 and SLC7A11 were upregulated by M2 co‐culture or oe‐CSF‐1 treatment, with the strongest increase observed under combined conditions. In contrast, UTMD markedly reduced both proteins, whereas subsequent M2 co‐culture and CSF‐1 overexpression partially restored their expression. ACSL4 showed the opposite pattern (Figure 4J,K). In Vivo Experiments Confirm UTMD Inhibits Tumor Growth by Regulating CSF‐1‐Mediated Macrophage Polarization and Ferroptosis To investigate the effect of UTMD on CSF‐1 secretion in vivo, a subcutaneous xenograft model was established in nude mice followed by randomized UTMD treatment. Representative tumor images, tumor growth curves, and final tumor weights showed that CSF‐1 overexpression significantly promoted PTC progression, whereas UTMD markedly inhibited tumor growth (Figure 5A–C). Transmission electron microscopy revealed that tumor cells in the oe‐CSF‐1 group retained relatively intact mitochondrial structure and showed few ferroptotic features. In contrast, UTMD induced typical ferroptotic morphology, including mitochondrial shrinkage, reduced or lost cristae, and increased membrane density, indicating that UTMD promotes ferroptosis in vivo, while CSF‐1 overexpression attenuates this effect (Figure 5D). Consistent with the in vitro findings, immunohistochemistry confirmed that UTMD significantly reduced CSF‐1 protein expression in tumor tissues (Figure 5E,F). Multiplex immunofluorescence further showed that the oe‐CSF‐1 group exhibited a protumor M2‐polarized microenvironment, with increased CD68+CD206+ M2‐type TAMs and decreased CD68+CD86+ M1‐type TAMs. UTMD reversed this pattern by increasing M1‐type TAMs and reducing M2‐type TAMs, although this immune reprogramming effect was partially weakened by CSF‐1 overexpression (Figure 5G–J). To further substantiate the involvement of ferroptosis in vivo, we measured GSH and MDA levels in tumor tissues. CSF‐1 overexpression significantly increased GSH content and decreased MDA content compared with the oe‐Ctrl group, indicative of preserved antioxidant capacity and reduced lipid peroxidation, whereas UTMD treatment significantly decreased GSH content and increased MDA content relative to the oe‐Ctrl group, consistent with enhanced ferroptotic activity (Figure 5K,L). At both protein and mRNA levels, GPX4, SLC7A11, and FTH1 were highest in the oe‐CSF‐1 group and decreased after UTMD treatment, whereas ACSL4 and TFR1 showed the opposite trend (Figure 5M–O). These findings, together with the GSH and MDA results, provide convergent evidence that UTMD suppresses tumor growth by reducing CSF‐1, remodeling macrophage polarization, and promoting ferroptosis. UTMD inhibits tumor growth by modulating CSF‐1‐mediated macrophage polarization and ferroptosis. (A) Representative photographs of subcutaneous xenograft tumors. (B) Tumor growth curves during the treatment period. (C) Final tumor weights of xenografts. (D) Representative transmission electron microscopy (TEM) images of tumor tissues showing mitochondrial morphology (scale bar: 500 Nm). (E, F) Representative immunohistochemical (IHC) staining and quantitative analysis of CSF‐1 protein expression in tumor tissues (scale bar: 50 μm). (G, H) Representative multiplex immunofluorescence staining for macrophage markers (CD68, CD86, and CD206) and DAPI in tumor tissues (scale bar: 50 μm). (I, J) Quantitative analysis of the proportions of macrophage markers shown in (G) and (H), respectively. (K, L) Measurement of glutathione GSH and MDA levels in tumor tissues. (M) Representative Western blot images showing the expression of ferroptosis‐related proteins, including TFR1, FTH1, ACSL4, SLC7A11, and GPX4, in tumor tissues. (N) Quantitative analysis of protein levels shown in (M), normalized to β‐Actin. (O) Relative mRNA expression levels of ferroptosis‐related genes detected by qRT‐PCR. Data are presented as mean ± SD (n = 3, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001). Discussion This study shows that ultrasound‐targeted microbubble destruction (UTMD), as a non‐invasive physical intervention, suppresses papillary thyroid carcinoma (PTC) progression by downregulating tumor‐derived CSF‐1. This effect contributes to remodeling of the immunosuppressive tumor microenvironment and increases the susceptibility of PTC cells to ferroptosis‐associated cell death. Our findings expand the role of UTMD beyond its conventional use in drug and gene delivery and support its potential as a non‐pharmacological strategy for refractory PTC. UTMD has traditionally been used to improve local delivery efficiency [31, 32, 33]. In the present study, however, UTMD directly altered tumor biology by reducing CSF‐1 transcription and secretion in PTC cells. This effect depended on the synergistic action of ultrasound and microbubbles, suggesting that cavitation‐mediated physical stimulation is required. Previous studies, including work showing UTMD‐induced M1 polarization in pancreatic cancer [34] and reviews describing its capacity to modulate the tumor microenvironment [35], support the broader concept that physical energy can regulate cellular signaling. Our data place CSF‐1 at the center of this response in PTC. To investigate the upstream mechanism underlying this effect, we measured intracellular Ca2+ and ROS levels in BCPAP cells following UTMD treatment and found that both were significantly and progressively increased across the UT, MD, and UTMD groups, with the UTMD group showing the greatest increase. These findings provide direct experimental support for the notion that UTMD‐induced mechanical stress and transient sonoporation trigger calcium influx and reactive oxygen species production [36], which may interfere with transcriptional regulators of CSF‐1 [37, 38, 39]. Nevertheless, the precise downstream signaling cascade linking these Ca2+/ROS changes to CSF‐1 transcriptional suppression was not directly established in this study, and future investigations employing pathway‐specific inhibitors or rescue experiments are warranted to fully delineate this signal transduction cascade, which is crucial for understanding and harnessing the therapeutic potential of UTMD. CSF‐1 is a well‐established driver of TAM polarization toward the M2 phenotype and is associated with poor prognosis in multiple cancers [15, 40, 41, 42, 43]. Our co‐culture experiments showed that CSF‐1 secreted by PTC cells promoted M2 marker expression and impaired macrophage phagocytic activity, supporting a role for CSF‐1 in maintaining an immunosuppressive niche. More importantly, gain‐of‐function and rescue experiments showed that UTMD reduced M2 polarization by suppressing CSF‐1. These observations provide mechanistic evidence that UTMD acts upstream of macrophage reprogramming. This point is clinically relevant because systemic CSF‐1R inhibition, although effective in selected settings such as tenosynovial giant cell tumor [44], is often limited in solid tumors by dose‐dependent toxicities and compensatory granulocyte colony‐stimulating factor release [44, 45]. By acting locally within the tumor, UTMD may reduce protumor signaling at its source while avoiding the adverse effects associated with systemic blockade. The fundamental goal of cancer therapy is the effective eradication of malignant cells. Conventional chemotherapy and radiotherapy primarily aim to eliminate cancer cells by inducing apoptosis. However, the frequent development of apoptosis resistance through mutations often leads to therapeutic failure, highlighting a significant limitation of this approach [46, 47]. Against this backdrop, ferroptosis has emerged as a distinct, iron‐dependent, non‐apoptotic form of cell death. It is driven by the lethal accumulation of lipid peroxides following the failure of cellular antioxidant defenses [48, 49]. In this study, multiple lines of evidence supported the involvement of ferroptosis in PTC cell death, including increased lipid peroxidation detected by C11‐BODIPY staining, reduced GPX4 expression, intracellular Fe2+ accumulation, and, importantly, the partial reversal of cell death by the ferroptosis inhibitor Ferrostatin‐1. The convergence of these biochemical and functional findings is consistent with the current understanding of ferroptosis and supports the conclusion that M1 macrophage‐induced PTC cell death contains a substantial ferroptotic component. This is consistent with emerging evidence that macrophage subsets differentially regulate ferroptosis susceptibility. M1 macrophages can sensitize tumor cells to ferroptosis through mediators such as nitric oxide and interferon‐γ, partly by suppressing SLC7A11 [50]. In contrast, M2 macrophages exert protective effects through several mechanisms, including ferritin secretion and altered lipid handling [51, 52, 53]. Our results extend this framework to PTC and identify CSF‐1 as a mechanistic link between macrophage polarization and ferroptosis sensitivity. By suppressing CSF‐1, UTMD weakened M2‐associated protection, shifted the balance toward an M1‐dominant state, and increased vulnerability to ferroptosis. This provides a mechanistic basis for integrating immune remodeling with ferroptosis‐based therapy in PTC. In vivo experiments supported the in vitro findings. CSF‐1 overexpression accelerated tumor growth and established a protective microenvironment enriched in M2‐like TAMs, accompanied by elevated GPX4 expression and reduced ferroptotic features, resembling a cold tumor phenotype [54, 55, 56]. UTMD reversed these effects by inhibiting tumor growth, reducing intratumoral CSF‐1, promoting M1 polarization, and inducing ferroptosis, as indicated by mitochondrial changes, decreased GPX4, and increased ACSL4 expression and lipid peroxidation. These results suggest that UTMD exerts antitumor effects through coordinated immune and metabolic reprogramming [22, 34, 57, 58]. This dual action may be advantageous in heterogeneous tumors, where single‐target strategies are often undermined by adaptive resistance [22, 59, 60]. As a localized, non‐invasive, and repeatable physical modality, UTMD possesses unique translational potential. This study offers a novel research avenue for addressing the clinical dilemma of Radioactive Iodine‐Refractory (RAIR) and advanced PTC, where treatment options are limited and often accompanied by significant side effects, with the prospect of improving therapeutic efficacy while reducing toxicity in future clinical applications. Despite the encouraging preclinical findings, several practical challenges need to be addressed before UTMD can be translated into clinical applications for thyroid cancer. First, although the superficial anatomical location of the thyroid gland provides a potential advantage for ultrasound‐based interventions, precise and reproducible delivery of acoustic energy remains challenging due to the small size, mobility, and rich vascularization of thyroid lesions, as well as the proximity of critical structures such as the trachea, recurrent laryngeal nerve, and carotid vessels [61, 62]. Therefore, optimization and standardization of ultrasound parameters, including frequency, mechanical index, duty cycle, and exposure duration, will be essential for safe and effective clinical implementation. Second, the availability of clinically applicable therapeutic microbubbles represents another important consideration. Currently approved microbubbles are primarily designed for diagnostic ultrasound imaging, whereas therapeutic UTMD applications often rely on experimentally optimized formulations [63, 64]. The development of clinically compatible microbubble systems with reproducible acoustic properties and appropriate regulatory profiles will be necessary for future translation. Third, compared with systemic CSF‐1/CSF‐1R blockade strategies, UTMD may provide a localized and transient approach to suppress tumor‐derived CSF‐1 while potentially reducing systemic toxicity. However, the therapeutic advantages of UTMD over pharmacological CSF‐1/CSF‐1R inhibitors remain to be directly evaluated in comparative studies [65, 66]. Future investigations using clinically relevant thyroid cancer models and head‐to‐head comparisons with CSF‐1/CSF‐1R‐targeted therapies will be required to determine the translational value of this approach. Several limitations of the present study should nevertheless be acknowledged. First, although we found that UTMD treatment significantly increased intracellular Ca2+ and ROS levels in BCPAP cells, supporting a role for these signaling intermediates in UTMD‐mediated CSF‐1 suppression, the direct causal link between these signaling changes and CSF‐1 transcriptional suppression was not established through pathway‐specific inhibition or rescue experiments (e.g., calcium chelation or ROS scavenging). Future studies employing pathway‐specific inhibitors, genetic manipulation of candidate signaling molecules, and transcriptional regulatory analyses will be required to clarify this process. Second, although the involvement of ferroptosis was supported by C11‐BODIPY‐detected lipid peroxidation, reduced GPX4 expression, intracellular Fe2+ accumulation, and the protective effect of Ferrostatin‐1, we further measured MDA content and GSH levels in xenograft tumor tissues to provide additional biochemical evidence for ferroptosis. To further strengthen the evidence for ferroptosis in vivo, we additionally measured MDA content and GSH levels in xenograft tumor tissues. Compared with the oe‐Ctrl group, CSF‐1 overexpression reduced MDA accumulation and preserved GSH levels, whereas UTMD treatment reversed these alterations by increasing MDA levels and decreasing GSH content (Figure 5L,M). These findings provide additional biochemical evidence supporting enhanced lipid peroxidation and impaired antioxidant capacity following UTMD treatment. Nevertheless, 4‐hydroxynonenal (4‐HNE) and intracellular ROS levels associated with the ferroptotic process were not evaluated in the present study; inclusion of these additional markers in future work would allow a more comprehensive characterization of ferroptotic oxidative damage and antioxidant system disruption, and further clarify the quantitative contribution of ferroptosis to the antitumor effects of UTMD. Third, this study employed a xenograft model established in BALB/c nude mice using the human PTC cell line BCPAP, which requires an immunodeficient host to permit tumor engraftment. Consequently, the lack of mature T lymphocytes in these mice precludes assessment of the full impact of UTMD‐mediated macrophage reprogramming on adaptive anti‐tumor immunity, including T cell infiltration, antigen‐specific responses, and the establishment of immunological memory. Given that TAM reprogramming is known to critically influence T cell recruitment and function within the tumor microenvironment [67, 68], this represents an important limitation of the present study. Future studies using immunocompetent syngeneic mouse models, such as murine thyroid cancer cell lines implanted in immunocompetent BALB/c or C57BL/6 hosts, will be necessary to comprehensively evaluate the interplay between UTMD‐induced innate immune remodeling and adaptive anti‐tumor immunity, and to more fully establish the translational relevance of this therapeutic strategy. Taken together, these limitations highlight several important directions for future investigation. Further studies are needed to elucidate the precise molecular mechanisms linking UTMD‐induced biophysical stimuli to CSF‐1 regulation, to comprehensively characterize the biochemical features of ferroptosis associated with UTMD treatment, and to evaluate the impact of UTMD‐mediated macrophage remodeling on adaptive antitumor immunity using immunocompetent models. Addressing these questions will provide a more comprehensive understanding of the UTMD–CSF‐1–macrophage–ferroptosis regulatory axis and facilitate its future translational development for thyroid cancer therapy. In summary, our results suggest that ultrasound‐targeted microbubble destruction suppresses CSF‐1 secretion in papillary thyroid carcinoma, leading to macrophage repolarization and enhanced ferroptosis sensitivity. This dual modulation of the immune microenvironment and tumor cell vulnerability highlights the potential of this non‐pharmacological approach for the treatment of refractory disease (Figure 6). UTMD inhibits CSF‐1 secretion, reprograms TAMs from M2 to M1, and enhances ferroptosis to suppress PTC progression. Author Contributions Yanqiang Niu: conceptualization, formal analysis, investigation, methodology, validation, visualization, writing – original draft. Zhenqiang Yao: methodology, validation, data curation, formal analysis. Minghua Zhang: methodology, validation, data curation, formal analysis, investigation. Jiayan Zhang: formal analysis, resources, software. Yushan Li: formal analysis, resources, software. Jinhui Xie: methodology, validation, data curation, formal analysis, investigation. Zubang Zhou: conceptualization, funding acquisition, project administration, supervision, writing – review and editing. Funding This work was supported by the Key R&D Program of Gansu Provincial Department of Science and Technology (No. 22YF7FA098 and No. 26YFFA013); the Natural Science Foundation of Gansu Province (No. 24JRRA606); the Internal Scientific Research Fund of Gansu Provincial Hospital (No. 24GSSYA‐3 and No. 20GSSY2‐4). Ethics Statement All animal experiments were conducted in accordance with institutional guidelines and were approved by the Ethics Committee of Gansu Provincial Hospital (Approval No. 2026‐044). Consent All authors consented to the publication of this manuscript. Conflicts of Interest The authors declare no conflicts of interest. Supporting information Figure S1: Representative image of the in vitro UTMD experimental setup. The figure shows the ultrasound‐targeted microbubble destruction (UTMD) apparatus used in this study, including the ultrasound transducer, treatment area, and sample positioning. Detailed treatment parameters are provided in the Materials and Methods section. Acknowledgements The authors have nothing to report. Data Availability Statement The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions. References Associated Data Supplementary Materials Figure S1: Representative image of the in vitro UTMD experimental setup. The figure shows the ultrasound‐targeted microbubble destruction (UTMD) apparatus used in this study, including the ultrasound transducer, treatment area, and sample positioning. Detailed treatment parameters are provided in the Materials and Methods section. Data Availability Statement The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.