Potassium channel gene KCNJ11 predicts glioma survival and chemotherapy response

A gene best known for controlling insulin release from pancreatic beta cells may hold a surprising clue to how long patients with brain cancer survive — and whether a standard chemotherapy drug will actually work for them. In a retrospective study spanning nearly two decades of clinical care, researchers at Fujian Cancer Hospital report that mRNA levels of KCNJ11, the potassium inward rectifier channel subfamily J member 11 gene, fall steadily as gliomas become more malignant. Patients whose tumors expressed high levels of the transcript lived markedly longer than those with low expression, and — most strikingly — only the high-expression group appeared to derive a survival benefit from postoperative temozolomide, the alkylating agent that anchors adjuvant therapy for most glioma patients. The findings, published in Frontiers in Oncology, position KCNJ11 as both an independent protective prognostic marker and a potential predictive biomarker for chemotherapy sensitivity, a combination that could eventually reshape how clinicians stratify patients for individualized adjuvant treatment.

Glioma is the most prevalent primary malignant intracranial neoplasm in adults, and despite a decade of progress in molecular classification its outlook remains grim. Biomarkers such as isocitrate dehydrogenase (IDH) mutation, combined 1p/19q chromosomal co-deletion, MGMT promoter methylation, TERT promoter mutation, EGFR amplification, TP53 and PTEN somatic mutations, MET gene amplification and ATRX mutation now inform diagnosis and prognosis under the 2021 WHO integrated molecular classification. Yet glioblastoma, the most aggressive subtype, still carries a five-year overall survival rate of just 5 percent. High-throughput transcriptomic profiling has become a mainstream strategy for hunting new biomarkers that can stratify prognosis and guide targeted therapy, and ion channels have attracted particular translational attention. Membrane potassium channels modulate critical tumor biological behaviors including cell proliferation and apoptotic signaling, and voltage-gated potassium channels, transient receptor potential channels and ABC transporters have already been validated as promising diagnostic markers and therapeutic targets across multiple human malignancies.

KCNJ11 sits on chromosome 11p15.1 and encodes Kir6.2, a 390-amino-acid protein produced from a single exon. Kir6.2 assembles with the sulfonylurea receptor SUR1, encoded by ABCC8, at a 4:4 stoichiometry to form the ATP-sensitive potassium (KATP) channel complex that mediates glucose-stimulated insulin secretion in pancreatic beta cells. When glucose metabolism raises the intracellular ATP/ADP ratio, KATP channels close, depolarizing the membrane, opening voltage-gated calcium channels, flooding the cell with calcium and triggering insulin granule exocytosis. Gain- or loss-of-function variants in KCNJ11 or ABCC8 disrupt this channel and cause congenital diabetes or persistent hyperinsulinemic hypoglycemia. Epidemiological links between chronic hyperinsulinemic states and malignancies such as colorectal carcinoma have kept the gene on the translational radar, and reduced KCNJ11 protein abundance has been reported in breast carcinoma, where high expression predicts improved clinical outcomes, partially regulated by gene methylation. In hepatocellular carcinoma, KCNJ11 potassium channel mRNA declines during cirrhosis-to-cancer transformation — a tumor-suppressive expression pattern. The team’s earlier work showed that ABCC8 mRNA independently predicts glioma prognosis and temozolomide response, but the biological function and clinical significance of KCNJ11 in glioma had never been reported.

To fill that gap, the investigators analyzed 516 consecutive glioma patients treated at the Neuro-Oncology Surgery Department of Fujian Cancer Hospital between January 2006 and January 2025. Tumor specimens were snap-frozen in liquid nitrogen immediately after resection, total RNA was extracted and quality-screened with an RNA integrity number of at least 6.8, and transcriptome sequencing was performed on Illumina HiSeq 2000, 2500 or 4000 platforms. Raw reads were mapped to the hg19 human reference genome with STAR 2.5.2b and quantified with RSEM 1.2.31, which normalizes first against transcript length to eliminate gene-length bias and then against total library fragment counts to correct inter-sample sequencing-depth variation, yielding FPKM expression values that were log2-transformed for all downstream analyses. Clinical variables — gender, age at diagnosis, primary or recurrent tumor status, histological subtype, WHO 2021 histological grade, IDH mutation, 1p/19q co-deletion, MGMT promoter methylation, postoperative radiotherapy and chemotherapy regimens, and survival outcome — were retrospectively collected, with follow-up censored at January 31, 2026. KCNJ11 expression was dichotomized at the median FPKM value, with a complementary sensitivity analysis entering expression as a continuous variable.

The expression pattern tracked malignancy with remarkable fidelity. KCNJ11 levels ranked oligodendroglioma highest, astrocytoma intermediate and glioblastoma lowest: 90.4 percent of oligodendrogliomas showed high expression, versus 50.9 percent of astrocytomas and only 23.9 percent of glioblastomas. High expression dominated WHO grade 2 tumors (73.1 percent) but fell to 58.6 percent in grade 3 and 23.9 percent in grade 4. IDH-mutant tumors carried higher transcript levels than IDH-wildtype cases (68.8 versus 25.4 percent high), 1p/19q co-deleted tumors outstripped non-deleted ones (93.9 versus 37.4 percent), and MGMT-methylated tumors exceeded unmethylated ones. Under the WHO 2021 integrated molecular classification, IDH-mutant 1p/19q co-deleted oligodendroglioma showed the highest expression, followed stepwise by IDH-mutant astrocytoma, IDH-wildtype astrocytoma and IDH-wildtype glioblastoma, in which only 18.4 percent of tumors expressed high levels. Patients receiving postoperative chemoradiotherapy displayed lower KCNJ11 expression than those without adjuvant treatment, and single-variable stratifications further associated older age, recurrent glioma and MGMT-unmethylated status with reduced transcription. Expression showed no significant correlation with gender, age dichotomization or recurrence status overall.

Kaplan–Meier analysis demonstrated that patients with high intratumoral KCNJ11 mRNA expression enjoyed consistently prolonged overall survival across all histological and molecular glioma subgroups. Univariate Cox regression flagged age, tumor recurrence, histopathological type, WHO grade, IDH mutation, 1p/19q co-deletion, MGMT methylation and KCNJ11 expression as survival-related factors, while postoperative chemotherapy was not significant in univariate analysis (P = 0.230) but emerged after adjustment. In the multivariate model, age (hazard ratio 1.015, 95 percent CI 1.005–1.025), tumor recurrence (HR 2.357, 95 percent CI 1.833–3.031) and WHO histological grade (HR 2.256, 95 percent CI 1.542–3.301) were independent risk factors, while IDH mutation (HR 0.717, 95 percent CI 0.521–0.986), 1p/19q co-deletion (HR 0.448, 95 percent CI 0.283–0.709), postoperative chemotherapy (HR 0.588, 95 percent CI 0.418–0.826) and KCNJ11 mRNA expression (HR 0.687, 95 percent CI 0.507–0.929) were independent protective determinants of overall survival. Variance inflation factors all stayed below 5, excluding severe multicollinearity among the inter-correlated clinical and molecular covariates, and the continuous-variable sensitivity analysis confirmed the median-split results.

The chemosensitivity finding is the study’s most consequential. Because comparing significance across separate subgroups cannot serve as formal evidence of interaction, the team constructed a multivariate Cox model for the full cohort containing a KCNJ11-chemotherapy interaction term. The omnibus test confirmed good model performance (−2 log-likelihood 2671.138, χ² 234.031, df 11, P < 0.001), and the interaction term reached statistical significance (P = 0.030, Exp(B) = 0.944, 95 percent CI 0.924–0.980), indicating that the prognostic effect of postoperative chemotherapy on overall survival was modified by KCNJ11 expression status. Stratified Cox regression sharpened the picture: in the high-expression subgroup, temozolomide chemotherapy remained an independent protective factor (HR 0.471, 95 percent CI 0.276–0.806, P = 0.006), whereas in the low-expression subgroup chemotherapy lost independent prognostic value (P = 0.186), with age, tumor recurrence and WHO grade dominating as major risk factors instead. In practical terms, only patients whose tumors expressed high KCNJ11 levels appeared to gain a survival benefit from standard adjuvant temozolomide.

External validation replicated the prognostic association in three independent public cohorts from China and the United States: 603 glioma cases from The Cancer Genome Atlas, 284 cases from the Repository for Molecular Brain Neoplasia Data (REMBRANDT) and 264 cases from the GSE16011 dataset in NCBI Gene Expression Omnibus. In all three datasets, high KCNJ11 expression predicted significantly longer median overall survival (log-rank χ² 79.317, P < 0.001; χ² 16.108, P < 0.001; χ² 6.580, P = 0.010, respectively). The authors note that KCNJ11 single-nucleotide polymorphisms are associated with adult metabolic syndrome — central obesity, hypertension, hyperglycemia and dyslipidemia — and that insulin resistance and compensatory hyperinsulinemia directly accelerate proliferation of epithelial malignancies such as colorectal and breast cancer, while body weight and obesity modulate KCNJ11 transcription itself. Diabetic breast cancer patients exhibit higher mortality and inferior prognosis, and translational experiments have verified reduced KCNJ11 protein in breast carcinoma tissues. The glioma data now extend this tumor-suppressive expression pattern to brain cancer, where no prior research had reported KCNJ11 expression or clinical relevance.

The authors are careful about the limits of their evidence. The analysis relies solely on transcriptomic mRNA expression data; further cellular, animal and tissue protein validation is required to elucidate the upstream transcriptional regulatory network and downstream tumor-suppressive signaling pathways mediated by KCNJ11 in glioma. Although three public cohorts confirmed the prognostic association, those datasets lack complete standardized postoperative chemotherapy information, so the formal KCNJ11-chemotherapy interaction test could not be replicated externally — the predictive conclusion rests exclusively on the single-center retrospective discovery cohort, and multi-center prospective studies with complete treatment records are needed before clinical translation can be considered. Chemotherapy was entered into Cox regression as a fixed binary covariate rather than a time-dependent variable; although assignment was determined shortly after surgery with no prolonged immortal-time window, theoretical immortal-time bias cannot be completely excluded in a retrospective observational design. The median-split cut-off is arbitrary, and mild correlation among covariates persists despite acceptable variance inflation values. Still, the message is clear: KCNJ11 mRNA expression falls as gliomas turn malignant, marks longer survival independently of traditional risk factors, and may identify which patients truly benefit from temozolomide.

Subject of Research: Neuro-oncology

Article Title: Potassium inward rectifier channel subfamily J member 11 mRNA expression in glioma and its significance in predicting prognosis and chemotherapy sensitivity

Article References: Zhou, K., Zhou, B., Gu, Q., Lu, T., Ke, Q., Liu, X., Xu, J., & Zhang, X. (2026). Potassium inward rectifier channel subfamily J member 11 mRNA expression in glioma and its significance in predicting prognosis and chemotherapy sensitivity. Frontiers in Oncology, 16, Article 1949969. https://doi.org/10.3389/fonc.2026.1949969

Image Credits: AI Generated

DOI: 10.3389/fonc.2026.1949969

Keywords: glioma, KCNJ11, temozolomide, biomarker, prognosis, chemotherapy sensitivity, potassium channels, Potassium, inward, rectifier, channel, subfamily