The new study, published in Cancer Reports, centers on a secreted signaling protein called Dickkopf-1, or DKK1, which is best known as an antagonist of the Wnt/beta-catenin pathway, a cascade central to embryonic development and frequently hijacked by tumors. Drawing on data from The Cancer Genome Atlas for head and neck squamous cell carcinoma, the researchers found that DKK1 transcript levels were significantly elevated in tumor tissues compared with adjacent healthy tissue, and that higher expression tracked with more advanced clinical stage, higher histological grade, and lymph node spread. Most strikingly, patients with high DKK1 expression had markedly worse overall survival, with a hazard ratio of 2.190, meaning their risk of death was more than double that of patients with low expression.

To confirm these population-level signals at the tissue level, the team analyzed twelve paired samples of oral cancer and matched normal tissue obtained from patients at Guangdong Provincial People’s Hospital and Southern Medical University. In ten of the twelve cases, both messenger RNA and protein levels of DKK1 were higher in the tumor than in the adjacent normal margin, and immunohistochemical staining visually confirmed the marked accumulation of the protein in malignant cells. The clinical cohort, which included seven men and five women with a median age of 65, spanned early and advanced disease stages in equal measure, lending weight to the conclusion that DKK1 upregulation is a general feature of oral cancer rather than a quirk of one subgroup.

The researchers then turned to laboratory models to test whether DKK1 actually drives drug resistance rather than merely accompanying it. They exposed two human oral cancer cell lines, CAL-27 and SCC-9, to gradually increasing doses of cisplatin over three to six months, generating resistant sublines whose half-maximal inhibitory concentrations had roughly doubled or tripled, from about 17 micromolar in the parental cells to 35.46 and 47.11 micromolar in the resistant derivatives. Notably, cisplatin itself pushed DKK1 expression upward in a dose-dependent manner, suggesting the drug inadvertently feeds the very defense mechanism that undermines it. When the team silenced DKK1 using lentiviral short hairpin RNAs, the resistant cells regained their sensitivity to cisplatin, demonstrating that the protein is not just a bystander but an active participant in chemoresistance.

The key to understanding how DKK1 protects tumor cells lay in an unusual form of regulated cell death known as ferroptosis. Unlike apoptosis, ferroptosis is an iron-dependent process driven by the runaway oxidation of lipids in cell membranes, ultimately rupturing them. Because cancer cells already live under high oxidative stress, they are particularly vulnerable to any collapse of their antioxidant defenses. When the researchers profiled how the resistant cells died under cisplatin treatment, they found that blocking apoptosis, necroptosis, or autophagy with specific inhibitors did little to change the outcome, but blocking ferroptosis with the drug ferrostatin-1 rescued cell viability almost completely. The resistant cells were also cross-resistant to erastin, a classic ferroptosis inducer, cementing the link between ferroptosis evasion and chemotherapy failure.

With ferroptosis identified as the critical death pathway, the team searched for the molecular switch that DKK1 uses to suppress it. Genetic screening of the resistant cells after DKK1 knockdown revealed a strong downregulation of SLC3A2, a structural subunit of the system Xc- cystine transporter that imports the amino acid building blocks cells need to manufacture glutathione, the cell’s principal antioxidant. Without SLC3A2, the glutathione supply dwindles, lipid peroxides accumulate, and ferroptosis proceeds. Consistent with this, analysis of the TCGA cohort showed SLC3A2 was elevated in head and neck tumors, correlated with histological grade, and predicted worse survival, and its expression was strongly positively correlated with DKK1 across the dataset.

Rescue experiments sealed the causal chain. When the researchers forced SLC3A2 overexpression in cells where DKK1 had been silenced, the ferroptotic markers reversed: lipid peroxidation measured by C11-BODIPY staining dropped, intracellular ferrous iron detected by FerroOrange fluorescence declined, reactive oxygen species and malondialdehyde levels fell, and glutathione was replenished. Most importantly, the cells regained their resistance to cisplatin, proving that SLC3A2 sits downstream of DKK1 as the functional mediator of ferroptosis suppression. The team then traced the signaling route upstream, discovering that DKK1 restrains beta-catenin-driven STAT3 activity, and that activated STAT3 binds directly to the SLC3A2 promoter, as shown by chromatin immunoprecipitation and luciferase reporter assays. When DKK1 is removed, beta-catenin rises, STAT3 phosphorylation increases, and STAT3 enrichment at the SLC3A2 promoter represses its transcription, dismantling the antioxidant shield.

The final and most clinically compelling test came in living animals. The researchers implanted cisplatin-resistant CAL-27 cells into immunodeficient mice and treated the resulting tumors with cisplatin at 2.5 milligrams per kilogram twice weekly. Tumors in which DKK1 had been genetically silenced shrank dramatically and grew more slowly than controls, but when the team simultaneously forced SLC3A2 overexpression, the protective effect largely vanished and the tumors rebounded. Immunohistochemical staining for 4-hydroxynonenal, a reliable marker of lipid peroxidation, showed that ferroptosis was indeed reactivated in the DKK1-depleted tumors receiving cisplatin, and that SLC3A2 overexpression blunted this response. The in vivo data thus mirror the cell culture findings and establish the DKK1/beta-catenin/STAT3/SLC3A2 axis as a genuine driver of treatment failure in a whole-organism setting.

The implications reach beyond oral cancer. DKK1 has already been implicated in resistance to bortezomib in multiple myeloma, sorafenib in liver cancer, paclitaxel in breast cancer, and even immunotherapy through its suppression of antitumor immune cells in the tumor microenvironment. A monoclonal antibody against DKK1, known as DKN-01, is currently in Phase II clinical trials, primarily for gastric and gynecologic cancers, and the new findings suggest that patients with oral cancer could one day benefit from similar strategies, potentially combined with cisplatin or ferroptosis-inducing agents. The authors caution that their clinical cohort was small, that ferroptosis is governed by many parallel pathways they did not explore, and that cancer-associated fibroblasts in the tumor microenvironment may secrete additional DKK1 that cooperates with the tumor’s own supply, a possibility that will require further study in more sophisticated models.

Nevertheless, the study delivers a clear conceptual advance: it identifies an upstream, secreted, and therefore druggable regulator of ferroptosis resistance in cisplatin-treated oral cancer. By showing that a single signaling axis connects a Wnt pathway antagonist to the cystine transporter that feeds cellular antioxidant defenses, the work reframes chemoresistance not as an impenetrable wall of redundant mechanisms but as a chain of dependencies with identifiable weak links. If future trials confirm that blocking DKK1 can strip away the ferroptosis shield in patients, a decades-old chemotherapy drug could regain its potency against one of the most treatment-resistant cancers of the head and neck.

Subject of Research: DKK1-mediated suppression of ferroptosis through SLC3A2 as a mechanism of cisplatin resistance in oral squamous cell carcinoma

Article Title: DKK1 Promotes Cisplatin Resistance in Oral Squamous Cell Carcinoma via SLC3A2‐Dependent Ferroptosis Suppression

Article References: Liu, Z., Chen, X., Lin, Y., & Zhu, L. (2026). DKK1 Promotes Cisplatin Resistance in Oral Squamous Cell Carcinoma via SLC3A2 ‐Dependent Ferroptosis Suppression. Cancer Reports, 9(10), Article e70715. https://doi.org/10.1002/cnr2.70715

Image Credits: AI Generated

DOI: 10.1002/cnr2.70715

Keywords: oral squamous cell carcinoma, cisplatin resistance, DKK1, ferroptosis, SLC3A2, system Xc-, Wnt/beta-catenin, STAT3, glutathione, lipid peroxidation, chemotherapy, TCGA