Eca109 (RRID: CVCL 6898), Siha (RRID: CVCL 0032) and U14 (RRID: CVCL 9U56), cancer cells were purchased from the Cell Bank of Chinese Academy of Sciences (Shanghai, China) and authenticated by the supplier (Shanghai, China). Siha and U14 cultured in Dulbecco’s modified eagle medium (DMEM) (Pricella; PM150210, Wuhan, China), and Eca109 were cultured in the RPMI1640 (Pricella; PM150110, Wuhan, China) medium, supplemented with 10% fetal bovine Serum (Cytiva; SH30401.01, Attleboro, MA, USA) and 1%penicillin–streptomycin (Cytiva; SV30010, Attleboro, MA, USA). All cell lines were validated by STR profiling and tested negative for mycoplasma. Cells were all cultured in a humidified incubator at 37 °C and 5% CO${}_2$.

The following chemotherapeutic agents were used: cisplatin (DDP), carboplatin (CBP), nedaplatin (NDP), lobaplatin (LBP), OXA, paclitaxel (TAX), docetaxel (DOC). All medications donated from the Pharmaceutical Dispensing Center of the Army Medical Center except for the OXA (Selleck; S1224, Houston, TX, USA) and DOC (Selleck; S1148, Houston, TX, USA). To evaluate cytotoxicity of the different chemotherapeutic agents, Eca109, Siha, and U14 cells were seeded in 96-well plates (NEST; 701002, Shanghai, China), incubated overnight. Eca109 cells treated for 48 h with either DDP (0–16 µM), CBP (0–512 µM), NDP (0–128 µM), LBP (0–512 µM), OXA (0–512 µM), TAX (0–32 nM), DOC (0–1600 nM) as single agents. As to Siha cells treated for 48 h with either CDDP (0–80 µM), CBP (0-1200 µM), NDP (0-320 µM), LBP (0-120 µM), OXA (0-640 µM), TAX (0–5160 nM), DOC (0–400 nM). Finally, U14 cells were treated for 24 h with either DDP (0–40 µM), CBP (0-1200 µM), NDP (0-160 µM), LBP (0–120 µM), OXA (0–640 µM), TAX (0–8000 nM), DOC (0–200 nM) as single agents. After treatment, cell counting kit-8 (CCK-8) at 10 µL/well (Dojindo; C0042, Kumamoto, Japan) added for incubation for two hours, and the absorbance measured at 450 nm with a microplate reader (Molecular Devices, Sunnyvale, CA, USA). The inhibitory concentration (IC) of the drug that leads to 50% (IC50) growth inhibition was calculated using the software GraphPad Prism 9.0.0 (9.0.0121) (Dotmatics, Boston, MA, USA). All experiments were performed at least in triplicate. For ECA109 and Siha cells, the treatment was 48 h, while U14 treated for 24 h.

After culture and treatments, cells were collected and centrifuged at 1000 g for 5 min. Then, the cells in 195 µL binding buffer were incubated with 5 µL AnnexinV-Fluorescein Isothiocyanate (FITC) and 10 µL PI for 15 min at room temperature in the dark according to the manufacturer’s instruction (Beyotime, C1062S, Shanghai, China). Cell apoptosis was analyzed by flow cytometer.

Extracellular ATP was measured in the conditioned media (supplemented with heat inactivated FBS) 24 h after chemotherapeutic treatment using the Firefly luciferase ATP assay, according to the manufacturer’s protocol (Beyotime, S20026, Shanghai, China). The bioluminescent signal was measured using a SpectarMaxL plate reader (Molecular Devices, Sunnyvale, CA, USA).

Release of HMGB1 was analyzed 48 h after treatment in the conditioned media using an enzyme-linked immunosorbent assay kit (Animaluni, Human: LV10792, Mouse: LV30652, Shanghai, China) according to the instructions of the manufacturer. The absorption was measured via the SpectarMax iD3 plate reader (Molecular Devices, Sunnyvale, CA, USA).

After culture and treatments, condition media collected by Laboratory Syringe Filters (Merck; SLGPR33RB, Rahway, NJ, USA). It should be noted that the media used to detect supernatant proteins were Serum-free media, this is due to the fact that only Serum-free media can ensure that no other proteins interfere with the collected supernatant. Since the damps molecule itself is minimally expressed in the supernatant, we innovatively used a gradient of drug concentrations in order to be able to observe more significant changes. Briefly, we set the concentration of each drug treated for three cell lines to 0.5 $\times$ IC50, 1 $\times$ IC50, 2 $\times$ IC50, 4 $\times$ IC50, 8 $\times$ IC50 based on the IC50 values of each drug (Supplementary Tables 1,2,3) in different cells determined in the previous period. First, we quantified the amount of protein loaded in each lane utilizing the silver staining kit (Thermo; 24612, Waltham, MA, USA) following the manufacturer’s protocol (Supplementary Fig. 1). Next, Supernatant protein samples (20 µL) were separated by SDS-PAGE (Ace; ET15420Gel, Changzhou, China) and transferred to polyvinylidene fluoride (PVDF) membranes. Next, the membranes were blocked with 10% non-fat milk for one hour and incubated overnight at 4 ℃ with primary antibodies: HSP70 (Cell Signaling; 4872T, Danvers , MA, USA), CRT (Cell Signaling; 12238S, Danvers , MA, USA), Annexin A1 (ANXA 1) (Abcam; ab214486, Cambridge, UK), HMGB1 (Abcam; ab79823, Cambridge, UK). Finally, the membranes were incubated with HRP-conjugated secondary antibodies (Bio-Rad, 1706515, Hercules, CA, USA) at room temperature for one hour. Western blotting analysis was performed using a Bio-Rad chemiluminescence imager (Bio-Rad, Hercules, CA, USA). At last, we performed a grayscale analysis of the individual western-blot bands using the image-lab software (3.0.39529, Bio-Rad, Hercules, CA, USA) and heat maps were drawn by the software GraphPad Prism 9.0.0 (9.0.0121) (Dotmatics, Boston, MA, USA).

We used the classical in vivo experimental model of ICD detection experiment [13]. Concisely, Six-week-old female C57BL/6J wild-type mice were used to generate experimental tumors by grafting a mouse cervical carcinoma cell line (U14). Mice were injected in the groin, according to the experiment, with 2 $\times$ 10${}^6$ U14-control cells and U14-drug handling cells, single IC50 concentration for 24 hours. After seven days, On the contralateral side, we injected the same number of tumor cells without any treatment. On days 1, 4, 7…28, the tumor volume was measured every three days using the following formula: Tumor volume (mm${}^{\mathrm{3}}$) = 0.5 $\mathrm{\times }$ width${}^{\mathrm{2}}$ $\mathrm{\times }$ length. At the end of the experiment, mouse ocular blood collected after anaesthesia with Delivector™ Avertin (20 µL/g.ip) (DW3101, Dowobio, Shanghai, China) and tumors retained for subsequent experiments. Mice are euthanized by way of cervical subluxation when reaching their ethical endpoint (volume $>$1500 mm${}^3$ or deep ulceration).

Tumor tissues prepared as 3 µm thick paraffin sections. For Immunohistochemistry (IHC) staining, the slides were dewaxed with xylene (Sangon Biotech, A530011, Shanghai, China) and rehydrated using graded ethanol (Sangon Biotech, A507050, Shanghai, China). Autoclaved antigen retrieval was performed in sodium citrate HCl buffer (Yun Tai, YTH0057, Shanghai, China), an endogenous H${}_2$O${}_2$ (Sangon Biotech, A001896, Shanghai, China) enzyme blocker (MXB, SP KIT-A3, Fuzhou, China) was used to block the tissues for 10 min, and 5% goat Serum (Sangon Biotech, E510009, Shanghai, China) was used for blocking at 37 ℃ for 1 h. The sections were incubated at 4 ℃ overnight with primary antibodies: anti-CRT rabbit antibody (1:250, Abcam; ab92516, Cambridge, UK), HMGB1 rabbit antibody (1:350, Abcam; ab79823, Cambridge, UK). The HRP-conjugated secondary antibody (Abcam, ab6728, Cambridge, UK) was incubated with the sections for 30 min at 37 ℃. The sections were rinsed with PBS (Pricella; PB180327, Wuhan, China), developed with diaminobenzidine substrate (Abcam, ab64238, Cambridge, UK), and counter-stained with hematoxylin (Abcam, ab220365, Cambridge, UK) for nuclear staining. A light microscope used to observe and capture images of the histopathological changes.

Mouse sera obtained from the in vivo experiments described above, and all experimental steps performed in strict accordance with the kit operating instructions. The DAMPs molecules detected include HSP70 (Animaluni; LV30721, Shanghai, China), CRT (Animaluni; LV31001, Shanghai, China), ANXA 1 (Animaluni; LV31010, Shanghai, China), and HMGB1 (Animaluni; LV30652, Shanghai, China).

Eight patients with histologically diagnosed primary squamous carcinoma who treated at the Oncology Center of the Third Affiliated Hospital of Army Military Medical University selected for the study (Supplementary Table 4). All patients were diagnosed for the first time and did not receive any other treatment regimen, and this time, they treated according to a combination regimen of platinum in combination with a taxanes. To evaluate the promotional effect of the chemotherapy regimen on the DAMPs molecule, we collect peripheral blood from the patients before and during the first cycle after treatment. Serum obtained by high-speed cryo-centrifugation (4 ℃ 1.5 $\times$ 10${}^4$ rpm) for enzyme linked immunosorbent assay (ELISA). Detection of HSP70 (Animaluni; LV10661, Shanghai, China), CRT (Animaluni; LV11352, Shanghai, China), ANXA 1 (Animaluni; LV11124, Shanghai, China), and HMGB1 (Animaluni; LV10792, Shanghai, China) in Serum carried out according to the kit instructions. This study and the use of all clinical materials were approved by the individual Institutional Ethical Committees, and written informed consent was obtained from all participant.

Statistical data expressed as mean $\pm$ standard deviation (SD). One-way analysis of variance used for three groups and more than three groups. All of the statistical analyses assessed by software GraphPad Prism (GraphPad Software, Inc., SanDiego, CA, USA), comparisons among groups were done by the independent sample two-sided Student t-test. The ANOVA performed to evaluate the statistical differences among groups. p-value of 0.05 or less considered as statistical significance.

First, we evaluated the dose-response curves of commonly used TP regimen drugs, namely DDP, CBP, NDP, LBP and OXA, and the taxanes DOC and TAX, in the human SCC cell lines ECA109 and SIHA, and in the mouse cervical SCC cell line U14 (Supplementary Fig. 2A–C). Due to the varying sensitivities of the three SCC cell lines to the chemotherapy agents, different IC50 values were used in subsequent experiments. The most effective drug inhibitory effect in the human SCC lines ECA109 and SIHA occurred after 48 h. In contrast, each drug had a noticeable inhibitory effect on the growth of mouse cervical SCC U14 cells after just 24 h, indicating greater sensitivity of these cells to the drugs. The IC50 values for each chemotherapeutic drug are shown in Tables 1,2,3 for the three SCC cell lines.

To determine the optimal drug concentration for inducing the extracellular release of DAMPs, a gradient treatment approach was employed in which the concentration of each drug was varied. Subsequently, a quantitative assay was used to measure the level of extracellular DAMPs in the supernatants of the three SCC cell lines following treatment with the chemotherapeutic agents. Based on our earlier findings, the supernatant proteins were collected after 48 h of treatment in the ECA109 and SIHA cell lines, and after 24 h of treatment in the U14 cell line.

Previous studies have shown that OXA induced ICD in colorectal tumor cells [14, 15]. However, the role of OXA in esophageal SCC and cervical SCC has yet to be systematically investigated. Here, we found that all three generations of platinum drugs induced release of the DAMP molecules HSP70, ANXA1, HMGB1 and CRT in a dose-dependent manner in the ECA109 cell line (Fig. 1A). The first-generation platinum drug DDP strongly induced the extracellular release of HSP70, ANXA1 and HMGB1. However, CRT induction primarily involved its displacement from the endoplasmic reticulum to the periplasm. The third-generation platinum drug OXA showed comparable results to DDP, with strong induction of HSP70, ANXA1 and HMGB1 release, but limited impact on CRT. However, the second-generation platinum drugs CBP and NDP induced a noticeable release of all four DAMP molecules. CBP induced the highest release at the drug’s half-maximal inhibitory concentration (IC50). Subsequently, the release of DAMP molecules gradually decreased, with HSP70 showing the most reduction. On the other hand, low doses of the third-generation platinum drug LBP induced limited extracellular release of HSP70 and ANXA1. However, a pronounced dose-dependent effect of LBP was observed at twice the IC50 concentration. Similarly, low doses of LBP induced noticeable extracellular release of CRT and HMGB1, with their release peaking at twice the IC50 concentration.

Fig. 1.

Release of HSP70, CRT, ANXA1 and HMGB1 from SCC cell lines after treatment with chemotherapy drugs. (A) Release of HSP70, CRT, ANXA1 and HMGB1 into the supernatant after treatment of ECA109 cells with platinum-based chemotherapeutic agents for 48 h. (B) Release of HSP70, CRT, ANXA1 and HMGB1 into the supernatant after treatment of SIHA cells with platinum-based chemotherapeutic agents for 48 h. (C) Release of HSP70, CRT, ANXA1 and HMGB1 into the supernatant after treatment of U14 cells with platinum-based chemotherapeutic agents for 24 h. (D) Release of HSP70, CRT, ANXA1 and HMGB1 into the supernatant of ECA109, SIHA and U14 cells after treatment with taxane-based chemotherapeutic agents for 48 h or 24 h. HSP70, Heat Shock Protein 70; CRT, Calreticulin; ANXA1, membrane-connected protein A1; HMGB1, High Mobility Group Protein B1; SCC, squamous cell carcinoma.

All three generations of platinum drugs induced the release of DAMP molecules from SIHA cells in a dose-dependent manner (Fig. 1B). However, several of the platinum drugs showed weaker induction of extracellular DAMP release from SIHA cells compared to ECA109 cells. In particular, the first-generation drug DDP and the second-generation drugs CBP and NDP induced less release of extracellular DAMP molecules from SIHA cells compared to ECA 109 cells. In contrast, the third-generation drugs LBP and OXA showed markedly stronger effects in SIHA cells. A more precise representation is provided by the heat map analysis shown in Fig. 2. Among the three generations of platinum drugs, DDP strongly induced the extracellular release of CRT and ANXA1, whereas CBP and NDP showed only limited induction of DAMP release. Both of the third-generation drugs, LBP and OXA, induced strong release of DAMPs, with the exception of CRT.

Fig. 2.

Overview of the normalized values of DAMPs released from the SCC cell lines. (A) ECA109 cells. (B) SIHA cells. (C) U14 cells. DAMPs, damage-associated molecular patterns.

To further examine the effect of ICD in vivo, we evaluated the extracellular release of DAMPs in mouse cervical SCC U14 cells following treatment with platinum drugs. These induced a more pronounced extracellular release in mouse U14 cells compared to SIHA cells (Fig. 1C). Of note, DDP and CBP showed stronger effects on the release of ANXA1 and HMGB1 compared to HSP70 and CRT in U14 cells. NDP had a significant impact on DAMP release, particularly on ANXA1, although the threshold value was reached at twice the IC50 concentration. Overall, LBP had a greater effect on the release of DAMPs compared to OXA. Treatment with LBP at the IC50 concentration led to a pronounced release, particularly for HMGB1, whereas OXA treatment resulted in a significant release of CRT only. The relatively weak induction of CRT observed in the three cell lines may be due to its dual functions. It is generally believed that ecto-CRT can induce ICD in its role as a DAMP. As noted previously, CRT was included as a molecular signature for ICD. However, the presence of CRT in extracellular spaces may also promotetum or recognition by Antigen-presenting cells (APC), thereby contributing to immunological surveillance [16]. Research has shown that the serum CRT level is highly correlated with patient prognosis [17]. Therefore, in order to be more relevant to clinical practice, we chose to detect CRT in the supernatant. Moreover, the expression of ecto-CRT is temporally related, reaching a peak at 48 h and then only gradually increasing in the extracellular space following tumor cell apoptosis [18].

We next evaluated the effect of taxane drugs on the extracellular release of DAMPs in the three SCC cell lines. These drugs induced significant extracellular release of DAMPs in all cell lines (Fig. 1D). However, in SIHA cells the release of CRT was not observed until the dose reached four times the IC50 concentration. DOC showed a diversity of induction across the different cell lines. ECA109 cells showed a peak in the extracellular release of DAMPs at half the IC50 concentration, suggesting high sensitivity to DOC. In contrast, the DOC-induced release of HMGB1 was relatively low in SIHA cells, with release only observed at four times the IC50 concentration. For other DAMP molecules, the peak effect on release was observed at four times the IC50 concentration, followed by a decrease at higher concentrations. A specific threshold effect of DOC may account for this observation in SIHA cells. U14 cells showed comparable results to ECA109 cells, with the peak release of DAMPs occurring at a low drug concentration (0.5 $\times$ IC50). No significant release was observed at higher concentrations.

In summary, the specific induction pattern of each DAMP molecule varied with each drug, while the optimal concentration for induction also differed. However, almost all drugs induced the extracellular release of DAMP molecules.

To visually represent the release of extracellular DAMPs following treatment with each drug, we conducted a grey value analysis of the western blot images shown in Fig. 1. The release of DAMPs from each group was logarithmically transformed using a reference value of one for the blank control. A comprehensive evaluation of the release of DAMPs from the three SCC cell lines is shown in Fig. 2. The selected human and murine SCC cell lines demonstrated significantly elevated levels of HSP70, CRT, ANXA1 and HMGB1 compared to the control group. The release of DAMPs varied according to treatment with either first-, second- or third-generation platinum-based and taxane-based agents. Importantly, the platinum-based agents elicited significantly greater release of DAMPs from esophageal SCC compared to cervical SCC. Additionally, higher doses of platinum-based drugs were required to induce ICD in both the human and murine cervical SCC cell lines.

ICD as a type of regulated cell death that differs from other forms of cell death such as apoptosis, ferroptosis, and necroptosis [19]. We conducted apoptosis assays based on the optimal drug concentrations for induction determined previously. The results are shown in Fig. 3, and changes in cell morphology are documented in Supplementary Fig. 3. The interaction of optimal drug concentrations resulted in a lower proportion of cells in the apoptotic phase, suggesting that ICD is not a straightforward form of apoptosis.

Fig. 3.

Detection of apoptosis following treatment with different chemotherapeutic drugs. (A) Detection of apoptosis in ECA109 cells following treatment with platinum-based drugs and taxane. (B) Proportion of ECA109 cells showing early apoptosis and late apoptosis. (C) Detection of apoptosis in SIHA cells following treatment with platinum-based drugs and taxanes. (D) Proportion of SIHA cells showing early apoptosis and late apoptosis. (E) Detection of apoptosis in U14 cells following treatment with platinum-based drugs and taxanes. (F) Proportion of U14 cells showing early apoptosis and late apoptosis.

We next performed an ATP assay to quantify the release of ATP in conditioned media following treatment of the three SCC cell lines for 24 h with each chemotherapeutic agent. A consistent inhibitory effect of LBP and TAX on the release of ATP was observed in all three cell lines (Fig. 4).

Fig. 4.

Adenosine triphosphate (ATP) release from SCC cell lines after treatment with chemotherapy drugs. The release of ATP from ECA109, SIHA and U14 cells was assessed after 24 h of treatment with the IC50 dose for DDP, CBP, NDP, LBP, OXA, TAX and DOC. (A) ECA109 cells. (B) SIHA cells. (C) U14 cells. Error bars represent the standard deviation. Experiments were performed at least in triplicate. ns p $>$ 0.05, *p 0.05, **p 0.01, ***p 0.001, ****p 0.0001.

Both CBP and DOC induced a significant increase in the secretion of ATP in ECA109 cells, with an approximately 1.5-fold increase compared to the control group (Fig. 4A). OXA caused a slight increase, while DDP and NDP did not induce ATP release. DDP, OXA and DOC induced more release of ATP from SIHA cells compared to CBP and NDP, with the latter two showing minimal ability to induce ATP release (Fig. 4B). Finally, NDP, OXA, and DOC induced significant release of ATP from the U14 cell line compared to the control group (Fig. 4C). In summary, OXA and DOC demonstrated superior induction of ATP release from SCC cell lines.

HMGB1 is known to stabilize DAMPs. We quantified its release from SCC cells using ELISA (Fig. 5). U14 cells were treated with the chemotherapy drugs for 24 h, while ECA109 and SIHA cells were treated for 48 h before collection of the supernatants for analysis. Consistent with the previous findings, significant release of HMGB1 from all three cell lines was observed following drug treatment compared to the control group.

Fig. 5.

Release of high mobility group protein B1 (HMGB1) from SCC cell lines following chemotherapy drug treatment. HMGB1 (pg/mL) in the supernatant was assessed after 48 h of treatment of the ECA109 and SIHA cells, and after 24 h of treatment of the U14 cells. The IC50 dose was used for DDP, CBP, NDP, LBP, OXA, TAX and DOC. (A) ECA109 cells. (B) SIHA cells. (C) U14 cells. **p 0.01, ***p 0.001, ****p 0.0001. Error bars represent the standard deviation. Experiments were performed in triplicate.

In vitro experiments can provide valuable insights into the potential for therapeutic agents to elicit an effective anti-tumor immune response. However, these assays do not take into consideration the complete immune system. To study this, we conducted a vaccination experiment as described earlier. The weight of all mice remained stable throughout the experiment. Fig. 6 illustrates the percentage of tumor-free mice, the rate of tumor growth, and representative tumors in vivo.

Fig. 6.

In vivo vaccination in the C57BL/6J mouse model underscores the immunogenic ability of clinically relevant chemotherapeutics. (A) Schematic diagram of the experimental protocol. (B) Tumor volumes in the xenograft model were measured every three days and are shown at the challenge site. (C) Tumor volumes in the xenograft model were measured every three days and are shown at the rechallenge site. (D) Percentage (%) of tumor-free mice and tumor volume (mm${}^3$) in C57BL/6J mice at 28 days. (E) Tumor volume at the rechallenge site in each treatment group after normalization. Quantitative data from three independent experiments are shown as the mean $\pm$ standard deviation (SD) (error bars). ns p $>$ 0.05, **p 0.01. (F) Tumors at the challenge and rechallenge sites on day 28 are shown in the control mice and in mice treated with DDP, CBP, NDP, LBP, OXA, TAX or DOC. (G) In vivo vaccination showing the immunogenic potential of clinically relevant chemotherapeutic drugs in the mouse model with U14 cells. Resected tumors from the challenge and rechallenge sites at day 28 are shown from the controls and from mice treated with CBP, LBP or TAX. ICD, immunogenic cell death inducer; NICD, non- immunogenic cell death inducer.

All five mice (100%) in the control group vaccinated with untreated cells developed a tumor at the challenge site. In contrast, none of the mice (0%) vaccinated with cells treated with DDP, DOC, NDP or OXA developed a tumor at the challenge site (Fig. 6B–D). Sixty percent of mice vaccinated with carboplatin-treated cells showed tumor development, whereas all mice (100%) vaccinated with LBP- or TAX-treated cells developed a tumor (Fig. 6B–D). Furthermore, at the end of the experiment the tumor volume in the DDP, CBP, NDP, OXA and DOC vaccinated groups was significantly larger than in the control group (Fig. 6E). This result suggests that DDP, CBP, NDP, OXA and DOC are able to induce ICD in mice with cervical SCC.

To further validate and identify the critical molecules involved in ICD, the concentrations of DAMP molecules were evaluated in the Serum of experimental mice (Fig. 7A). DDP was found to induce comparable levels of HSP70, ANXA1 and CRT release into the Serum as the previous in vitro experiments conducted at the cellular level. The known inducer of ICD, OXA, also induced high levels of HSP70, ANXA1, CRT and HMGB1 secretion into the Serum of mice treated with this drug, thus further confirming its effect on ICD in SCC. Mice treated with CBP, NDP or DOC showed significant release of DAMPs, consistent with the results of earlier in vitro experiments. In contrast, LBP and TAX are unable to induce ICD and did not show any significant release of the four DAMP molecules in this mouse model. Although LBP and TAX induced some extracellular release of DAMPs in vitro, they were unable to induce ICD in the complexity of an in vivo environment.

Fig. 7.

Release of DAMP molecules into mouse Serum and tumor tissues. (A) Serum levels of HSP70, ANXA1, CRT, and HMGB1 in C57BL/6J mice treated with chemotherapeutic agents. (B) Immunohistochemical study of CRT and HMGB1 expression and their intracellular localization in representative xenograft tumors (400$\times$ magnification, scale bar = 50 µm). Quantitative data from three independent experiments are shown as the mean $\pm$ SD (error bars). ns p $>$0.05, *p 0.05, **p 0.01, ***p 0.001.

Immunohistochemical evaluation of tumor cells at the site of reinoculation confirmed the Serum test results (Fig. 7B). In the control group, CRT was found to be exclusively intracellular, while HMGB1 was confined to the nucleus, indicating they were unable to induce ICD. In contrast, LBP- and TAX-treated mice showed low expression of these molecules, possibly indicating depletion and insufficient downstream ICD effects. However, CBP-treated mice showed high expression of both CRT and HMGB1, thus confirming the ability of this agent to induce ICD.

DAMPs are typically released at high levels from various types of cancers, including SCC. To extend the current observations on the increased release of DAMPs following chemotherapeutic drug treatment of SCC cell lines, we used ELISA to determine the Serum levels of DAMPs in 8 SCC patients before and after chemotherapy. In four patients, the Serum concentrations of HSP70, CRT, ANXA1 and HMGB1 were significantly higher post-chemotherapy compared to the Pre-chemotherapy SCC patients (Fig. 8A). The other four patients showed increased Serum levels of different DAMPs following chemotherapy. Further analysis revealed that increased DAMP levels were associated with chemotherapeutic stimulation in SCC (Fig. 8B).

Fig. 8.

Elevated Serum levels of CRT and HMGB1 after chemotherapeutic stimulation in SCC patients. (A) Post-chemotherapy Serum levels of HSP70, CRT, ANXA1 and HMGB1 were significantly higher in four SCC patients compared to non-chemotherapy controls, as measured by ELISA. (B) Serum levels of HSP70, CRT, ANXA1 and HMGB1 in non-chemotherapy controls and in 8 post-chemotherapy SCC patients, as measured by ELISA. Quantitative data from three independent experiments are shown as the mean $\pm$ SD (error bars). ns p $>$0.05, *p 0.05, **p 0.01, ***p 0.001, ****p 0.0001. ELISA, enzyme-linked immunosorbent assay.