Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat

The DOI matches its Crossref record; 32 of the 32 references it cites were found in a registry or an index; 26 of the 33 sources its claims cite were checked for support.

Bioinformatics-based identification of glycolysis-related signatures associated with drug resistance and prognosis in lung adenocarcinoma

Open access

Abstract

Background Drug resistance and poor clinical outcomes in lung adenocarcinoma (LUAD) necessitate robust biomarkers for personalized therapy. Glycolysis reprogramming is a hallmark of cancer, but its clinical utility remains incompletely defined. Methods We integrated TCGA and GEO transcriptomic data with Weighted gene co−expression network analysis (WGCNA), least absolute shrinkage and selection operator (LASSO), and multivariate Cox regression to construct a glycolysis−related prognostic signature. A nomogram combining the risk score with clinicopathological factors was developed. Drug sensitivity was predicted using the pRRophetic algorithm. qRT−PCR and xenograft models using A549 and cisplatin−resistant A549/DDP cells validated the expression of candidate genes. Results Patients stratified by glycolysis-related risk scores exhibited significantly distinct survival outcomes, and the glycolysis-based signature functioned as an independent prognostic factor for overall survival in LUAD. The nomogram demonstrated robust predictive performance and effectively estimated patient sensitivity to three commonly used conventional chemotherapeutic agents. In vitro and in vivo studies using A549 cells and their cisplatin-resistant derivative A549/DDP revealed aberrant expression of VIPR1 , ADRB2 , RXFP1 , PDGFB , WNT3A , and SPRY1 in the resistant cell line and in corresponding xenograft tumor tissues. These findings suggest that glycolytic activity is closely associated with both drug resistance and clinical prognosis in LUAD. Conclusions This study identifies a glycolysis-related gene signature with demonstrable utility for prognostic stratification and therapeutic response prediction in LUAD. The proposed integrative model holds promise for enhancing precision treatment decision-making through optimized risk assessment and rational selection of chemotherapeutic regimens.

Overview

CoChat Summary

TL;DR

This bioinformatics study built a 19-gene glycolysis-related prognostic signature for lung adenocarcinoma (LUAD) and validated it across TCGA and GEO cohorts, showing independent prognostic value and a nomogram with improved discrimination. It also linked the signature to predicted chemosensitivity and experimentally examined six glycolysis-related genes in cisplatin-resistant A549/DDP cells and xenografts.

Key findings

  • A glycolysis-related risk score based on 19 prognostic genes stratified patients into high- and low-risk groups with significantly different overall survival in TCGA-LUAD, GSE72094, and GSE14814 (external log-rank P = 0.04244); time-dependent AUCs were 0.765/0.729 at 3/5 years in TCGA, 0.657/0.730 in GSE72094, and 0.608/0.613 in GSE14814.
  • The risk score was an independent prognostic factor in multivariate Cox analysis: TCGA-LUAD risk score HR = 3.848 and stage HR = 2.068; across LUAD patients, risk score HR = 6.129 and stage HR = 2.930 (P < 0.01). The nomogram achieved AUC = 0.712, outperforming stage alone (0.587) and risk score alone (0.680).
  • pRRophetic predicted significant IC50 differences between risk groups for cisplatin, gemcitabine, and doxorubicin, but not gefitinib; the high-risk subgroup had lower predicted IC50 values for the three significant agents.
  • In cisplatin-resistant cells, IC50 values were 137.63 μM for A549/DDP and 73.17 μM for H460/DDP versus 18.51 μM for A549 and 21.36 μM for H460. qRT-PCR showed elevated WNT3A, VIPR1, ADRB2, SPRY1, PDGFB, and RXFP1 in resistant cells, with WNT3A, ADRB2, and SPRY1 significant; in A549/DDP-derived tumors, VIPR1, SPRY1, PDGFB, and RXFP1 were significantly higher.

Methods

  • Integrated TCGA-LUAD (n = 456) and GEO datasets GSE72094 (n = 369) and GSE14814 (n = 133), requiring complete survival/clinical information and survival time ≥ 90 days; expression data were log2-transformed, quantile-normalized, and batch-corrected with ComBat.
  • Used univariate Cox, LASSO with ten-fold cross-validation (λ_min = 0.01215), WGCNA (pink module, 314 genes, correlation −0.54 with glycolysis, P < 0.05), and multivariate Cox to build a 19-gene risk score; median risk score defined high/low groups; evaluated with Kaplan-Meier, timeROC, nomogram with bootstrap internal validation, DESeq2, GISTIC 2.0, ChAMP, and pRRophetic.
  • Validated candidate genes by qRT-PCR in A549, H460, A549/DDP, and H460/DDP cells and in a xenograft model using 4-week-old male BALB/c nude mice (n = 6 per group, repeated twice) inoculated with A549 or A549/DDP cells (4×10^6 cells); no chemotherapy was administered.

Limitations

  • Stated: the study relied primarily on retrospective public datasets, which may introduce selection bias; the dataset provided only transcriptomic data, so DNA methylation and CNV could not be directly analyzed for the 19 core prognostic genes; validation in large, well-characterized clinical cohorts and prospective/functional studies is still needed.
  • Stated/evident: the in vitro IC50 results contradicted the pRRophetic prediction that high-risk patients have lower IC50 values, and the paper explains this as a possible baseline-sensitivity versus acquired-resistance paradox.
  • Evident: the xenograft experiment administered no chemotherapy and found no significant differences in tumor volume, tumor weight, or body weight between A549 and A549/DDP groups; external validation detected only 13 of the 19 prognostic genes; and the figure legend notes that methylation/CNV analyses were genome-wide and did not specifically interrogate the 19-gene signature.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.3389/fonc.2026.1871582

Cites

32 of 32 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Figures

Figure 1 Glycolysis is a major risk factor for overall survival in patients with lung adenocarcinoma. (A) Univariate Cox regression analysis, with the pathway name in the first column, the p-value results in the second column, and the HR value in the third column, with values in parentheses denoting 95% confidence interval results. Risk factors are HR>1, and inhibitors are HR<1. Red dots and vertical lines reflect the log-partial likelihood function value + standard error corresponding to each in (B) The LASSO regression screening. The Log(λ) one standard error from the minimum is shown as the rightmost dashed line; (C) Cox multivariate findings, with values in parentheses denoting results with 95% confidence intervals. The first column of the results lists the pathway name, the second lists the p-value result, and the third lists the HR. HR<1: inhibitory factors, HR>1: risk factors.
Figure 1 Glycolysis is a major risk factor for overall survival in patients with lung adenocarcinoma. (A) Univariate Cox regression analysis, with the pathway name in the first column, the p-value results in the second column, and the HR value in the third column, with values in parentheses denoting 95% confidence interval results. Risk factors are HR>1, and inhibitors are HR<1. Red dots and vertical lines reflect the log-partial likelihood function value + standard error corresponding to each in (B) The LASSO regression screening. The Log(λ) one standard error from the minimum is shown as the rightmost dashed line; (C) Cox multivariate findings, with values in parentheses denoting results with 95% confidence intervals. The first column of the results lists the pathway name, the second lists the p-value result, and the third lists the HR. HR<1: inhibitory factors, HR>1: risk factors.CC BY · PubMed Central
Figure 2 Prognostic genes involved in glycolysis in lung adenocarcinoma. The WGCNA hierarchical clustering in (A) displays each module; (B) Correlation analysis between the GLYCOLYSIS pathway and modules, with each cell including the correlation coefficient and the outcomes of a statistical test (numbers in brackets) between the two. Statistics were judged significant at P < 0.05. Vertical lines and red dots represent of Lasso regression screening genes; (C) show the value of the log partial likelihood function plus standard error for each λ. The Log(λ) one standard error from the minimum is shown by the dashed line on the right; (D) Regression coefficients for 19 prognostic genes.
Figure 2 Prognostic genes involved in glycolysis in lung adenocarcinoma. The WGCNA hierarchical clustering in (A) displays each module; (B) Correlation analysis between the GLYCOLYSIS pathway and modules, with each cell including the correlation coefficient and the outcomes of a statistical test (numbers in brackets) between the two. Statistics were judged significant at P < 0.05. Vertical lines and red dots represent of Lasso regression screening genes; (C) show the value of the log partial likelihood function plus standard error for each λ. The Log(λ) one standard error from the minimum is shown by the dashed line on the right; (D) Regression coefficients for 19 prognostic genes.CC BY · PubMed Central
Figure 3 Model construction and validation of glycolysis-related prognostic genes. (A) The TCGA-LUAD survival curves for various risk categories; (B)Survival curves of different risk groups in GSE72094; (C)ROC curves were used to assess lung cancer survival rates at three and five years in the TCGA; (D)ROC curves were used to assess lung cancer survival rates at 3- and 5-year intervals in GSE72094. AUC values range from 0 to 1, with higher values indicating better predictive performance.
Figure 3 Model construction and validation of glycolysis-related prognostic genes. (A) The TCGA-LUAD survival curves for various risk categories; (B)Survival curves of different risk groups in GSE72094; (C)ROC curves were used to assess lung cancer survival rates at three and five years in the TCGA; (D)ROC curves were used to assess lung cancer survival rates at 3- and 5-year intervals in GSE72094. AUC values range from 0 to 1, with higher values indicating better predictive performance.CC BY · PubMed Central
Figure 4 Comparison of the prognostic and predictive power of glycolysis-related prognostic risk scores with traditional features. (A) The cox multivariate results of Age, Gender, Stage, and riskScore in the training set of the TCGA, where the first column is the prognostic indicator, the second column is the result of the P value, and the third column is the HR value, in parenthesis. The figures show outcomes with a 95% confidence level. The risk factor is HR>1, the inhibiting factor is HR<1; (B) The cox multivariate results of Age, Gender, Stage, and riskScore in the validation set GSE72094, where the first column is the prognostic indicator, the second column is the result of the P value, and the third column is the HR value, in parenthesis. The figures show outcomes with a 95% confidence level. The risk factor is HR>1, the inhibiting factor is HR<1; (C) Training set TCGA multi-index ROC curve. (D) Validation set GSE72094 multi-index ROC curve.
Figure 4 Comparison of the prognostic and predictive power of glycolysis-related prognostic risk scores with traditional features. (A) The cox multivariate results of Age, Gender, Stage, and riskScore in the training set of the TCGA, where the first column is the prognostic indicator, the second column is the result of the P value, and the third column is the HR value, in parenthesis. The figures show outcomes with a 95% confidence level. The risk factor is HR>1, the inhibiting factor is HR<1; (B) The cox multivariate results of Age, Gender, Stage, and riskScore in the validation set GSE72094, where the first column is the prognostic indicator, the second column is the result of the P value, and the third column is the HR value, in parenthesis. The figures show outcomes with a 95% confidence level. The risk factor is HR>1, the inhibiting factor is HR<1; (C) Training set TCGA multi-index ROC curve. (D) Validation set GSE72094 multi-index ROC curve.CC BY · PubMed Central
Figure 5 Nomogram drawing and evaluation. (A) A prognostic nomogram model that projects overall survival at 3 and 5 years based on risk levels and other clinical data; (B) Nomogram and multi-index ROC curve; (C) Calibration curves for estimating the total survival time across three and five years.
Figure 5 Nomogram drawing and evaluation. (A) A prognostic nomogram model that projects overall survival at 3 and 5 years based on risk levels and other clinical data; (B) Nomogram and multi-index ROC curve; (C) Calibration curves for estimating the total survival time across three and five years.CC BY · PubMed Central
Figure 6 Multi−omics characterization of high− and low−risk subgroups. (A) Volcano plot of differentially expressed genes (DEGs) between risk groups; red = upregulated, blue = downregulated (padj < 0.05, |log2FC| > 1); (B) GO enrichment analysis of DEGs; (C) KEGG pathway enrichment analysis of DEGs; (D) Methylation differential analysis; red = hypermethylated, blue = hypomethylated in high−risk group; (E) Copy number variation (CNV) analysis in high−risk subgroup; (F) CNV analysis in low−risk subgroup. (D–F) represent genome−wide analyses and do not specifically interrogate the 19−gene signature, as methylation and CNV data were not available for the same cohort. (G) Methylation GO enrichment analysis; (H) Methylation KEGG enrichment analysis.
Figure 6 Multi−omics characterization of high− and low−risk subgroups. (A) Volcano plot of differentially expressed genes (DEGs) between risk groups; red = upregulated, blue = downregulated (padj < 0.05, |log2FC| > 1); (B) GO enrichment analysis of DEGs; (C) KEGG pathway enrichment analysis of DEGs; (D) Methylation differential analysis; red = hypermethylated, blue = hypomethylated in high−risk group; (E) Copy number variation (CNV) analysis in high−risk subgroup; (F) CNV analysis in low−risk subgroup. (D–F) represent genome−wide analyses and do not specifically interrogate the 19−gene signature, as methylation and CNV data were not available for the same cohort. (G) Methylation GO enrichment analysis; (H) Methylation KEGG enrichment analysis.CC BY · PubMed Central
Figure 7 Predicted drug sensitivity between high‑ and low‑risk groups. Estimated IC50 values for cisplatin, gefitinib, gemcitabine, and doxorubicin were calculated using the pRRophetic algorithm. Lower IC50 in the high‑risk group indicates enhanced predicted baseline chemosensitivity. P<0.05, significant. These predictions are derived from treatment‑naïve transcriptomic profiles and do not represent acquired resistance.
Figure 7 Predicted drug sensitivity between high‑ and low‑risk groups. Estimated IC50 values for cisplatin, gefitinib, gemcitabine, and doxorubicin were calculated using the pRRophetic algorithm. Lower IC50 in the high‑risk group indicates enhanced predicted baseline chemosensitivity. P<0.05, significant. These predictions are derived from treatment‑naïve transcriptomic profiles and do not represent acquired resistance.CC BY · PubMed Central
Figure 8 In vitro and in vivo validation of glycolysis−related candidate genes. (A) Different concentrations of cisplatin were used on A549, and A549/DDP cells and CCK-8 calculated IC50 after 24h to detect drug resistance; (B) H460 and H460/DDP cells were treated with different concentrations of cisplatin, and cck8 calculated IC50 after 24 h to detect drug resistance; (C, D) mRNA expression levels of WNT3A, VIPR1, ADRB2, SPRY1, PDGFB and RXFP1 detected by qRT-PCR in A549/A549/DDP (C) and H460/H460/DDP (D) cells detected by qRT−PCR, *p<0.05, **p<0.01, ***p <0.001; (E–G) Tumor volume and tumor weight changes in mice; (H) Body weight changes in mice; (I) qRT−PCR analysis of WNT3A, VIPR1, ADRB2, SPRY1, PDGFB and RXFP1 mRNA levelsin A549− and A549/DDP−derived tumor tissues. Data are shown as mean ± SD (n = 3 technical replicates; n = 6 biological replicates per group in vivo). *p<0.05, **p<0.01, ***p<0.001. Statistical comparisons were analyzed based on one‐way analysis of variance (ANOVA) followed by Dunnett’s test. A p‐value of < 0.05 indicates a significant difference.
Figure 8 In vitro and in vivo validation of glycolysis−related candidate genes. (A) Different concentrations of cisplatin were used on A549, and A549/DDP cells and CCK-8 calculated IC50 after 24h to detect drug resistance; (B) H460 and H460/DDP cells were treated with different concentrations of cisplatin, and cck8 calculated IC50 after 24 h to detect drug resistance; (C, D) mRNA expression levels of WNT3A, VIPR1, ADRB2, SPRY1, PDGFB and RXFP1 detected by qRT-PCR in A549/A549/DDP (C) and H460/H460/DDP (D) cells detected by qRT−PCR, *p<0.05, **p<0.01, ***p <0.001; (E–G) Tumor volume and tumor weight changes in mice; (H) Body weight changes in mice; (I) qRT−PCR analysis of WNT3A, VIPR1, ADRB2, SPRY1, PDGFB and RXFP1 mRNA levelsin A549− and A549/DDP−derived tumor tissues. Data are shown as mean ± SD (n = 3 technical replicates; n = 6 biological replicates per group in vivo). *p<0.05, **p<0.01, ***p<0.001. Statistical comparisons were analyzed based on one‐way analysis of variance (ANOVA) followed by Dunnett’s test. A p‐value of < 0.05 indicates a significant difference.CC BY · PubMed Central

Verification

Claims the paper makes

12 Valid · 1 Not valid · 4 Not checked · 71 Own results and interpretations

48 more claims

Create a free account to see all 88 claims and to verify the 3 references still open.

See all claims — free account
  • Probably not a substantial claim. 16
  • It may state two separate claims as one. 24

Create a free account to see these claims.

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

2 of 2

Citations

Cites

Show all 32 references

Metadata

Authors
Zheng, Q., Liu, Y., Li, T., Zhang, X., Geng, Y., Chen, Z., Zhou, J., Fan, W., Wang, Y., Zhang, L., & Cao, Y.
Affiliations
  • University of Science and Technology of China (CN)
  • Anhui Medical University (CN)
  • Anhui Institute of Medicine (CN)
Corresponding author
  • Wang, Yong
  • Zhang, Lei
  • Cao, Yuzhu
Venue
Frontiers in Oncology
Published
2026-07-15
Volume
16
Pages
1871582-1871582
ISSN
2234-943X (ISSN-L)
OpenAlex
W7168364447
Cited by
0
References
32
FWCI
0.00
Type
journal article
Language
en
Open access
Locations
  • Frontiers in Oncology (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion)
Funders
  • National Natural Science Foundation of China (52373160)
  • Key Technologies Research and Development Program of Anhui Province (201904a07020092, 202304295107020050)
  • University Natural Science Research Project of Anhui Province (2208085QH238, 2208085MH252)
Topics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Cancer Research › Cancer, Hypoxia, and Metabolism
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Cancer Research › Ferroptosis and cancer prognosis
  • Health Sciences › Medicine › Oncology › Lung Cancer Research Studies
Keywords
lung adenocarcinoma, glycolysis, drug resistance, prognostic gene signature, cisplatin resistance, chemotherapy response prediction, nomogram, VIPR1, ADRB2, RXFP1, PDGF-B, Wnt3a, SPRY1, overall survival, cancer metabolism

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

The claim verdicts this report counts are experimental: their accuracy is not yet measured.

A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Cited sources are checked with a language model’s help (gpt-6-sol).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Bioinformatics-based identification of glycolysis-related signatures associated with drug resistance and prognosis in lung adenocarcinoma. https://app.cochat.ai/papers/10.3389/fonc.2026.1871582
The paper
Zheng, Q., Liu, Y., Li, T., Zhang, X., Geng, Y., Chen, Z., Zhou, J., Fan, W., Wang, Y., Zhang, L., & Cao, Y. (2026). Bioinformatics-based identification of glycolysis-related signatures associated with drug resistance and prognosis in lung adenocarcinoma. Frontiers in Oncology. https://doi.org/10.3389/fonc.2026.1871582

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.