Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2022 Feb 14:13:838021.
doi: 10.3389/fgene.2022.838021. eCollection 2022.

Prognostication of Pancreatic Cancer Using The Cancer Genome Atlas Based Ferroptosis-Related Long Non-Coding RNAs

Jiayu Li  1 , Jinghui Zhang  2 , Shuiliang Tao  3 , Jiaze Hong  1 , Yuyan Zhang  2 , Weiyan Chen  3
Affiliations

Prognostication of Pancreatic Cancer Using The Cancer Genome Atlas Based Ferroptosis-Related Long Non-Coding RNAs

Jiayu Li et al. Front Genet. .

Abstract

Background: Long non-coding RNAs (lncRNAs) are key regulators of pancreatic cancer development and are involved in ferroptosis regulation. LncRNA transcript levels serve as a prognostic factor for pancreatic cancer. Therefore, identifying ferroptosis-related lncRNAs (FRLs) with prognostic value in pancreatic cancer is critical. Methods: In this study, FRLs were identified by combining The Cancer Genome Atlas (TCGA) and FerrDb databases. For training cohort, univariate Cox, Lasso, and multivariate Cox regression analyses were applied to identify prognosis FRLs and then construct a prognostic FRLs signature. Testing cohort and entire cohort were applied to validate the prognostic signature. Moreover, the nomogram was performed to predict prognosis at different clinicopathological stages and risk scores. A co-expression network with 76 lncRNA-mRNA targets was constructed. Results: Univariate Cox analysis was performed to analyze the prognostic value of 193 lncRNAs. Furthermore, the least absolute shrinkage and selection operator and the multivariate Cox analysis were used to assess the prognostic value of these ferroptosis-related lncRNAs. A prognostic risk model, of six lncRNAs, including LINC01705, AC068620.2, TRAF3IP2-AS1, AC092171.2, AC099850.3, and MIR193BHG was constructed. The Kaplan Meier (KM) and time-related receiver operating characteristic (ROC) curve analysis were performed to calculate overall survival and compare high- and low-risk groups. There was also a significant difference in survival time between the high-risk and low-risk groups for the testing cohort and the entire cohort, with AUCs of .723, .753, respectively. Combined with clinicopathological characteristics, the risk model was validated as a new independent prognostic factor for pancreatic adenocarcinoma through univariate and multivariate Cox regression. Moreover, a nomogram showed good prediction. Conclusion: The signature of six FRLs had significant prognostic value for pancreatic adenocarcinoma. They may be a promising therapeutic target in clinical practice.

Keywords: ferroptosis; long non-coding RNA; nomogram; pancreatic adenocarcinoma; risk model.

PubMed Disclaimer

Conflict of interest statement

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Figures

FIGURE 1
FIGURE 1
Flow chart of the study.
FIGURE 2
FIGURE 2
Selection of FRLs using LASSO and multivariate Cox regression. (A)Eight FRLs based LASSO cross validation plot. (B) LASSO coefficient of eight FRLs in PAAD. (C) Multivariate Cox regression showing six FRLs (LINC01705, AC068620.2, TRAF3IP2-AS1, AC092171.2, AC099850.3, and MIR193BHG) associated with OS in TCGA-PAAD.
FIGURE 3
FIGURE 3
KM survival curves for the six prognostic FRLs. Three FRLs (LINC01705, AC099850.3 and MIR193BHG) were independent unfavorable factors and three lncRNAs (AC068620.2, TRAF3IP2-AS1, and AC092171.2) were independent favorable factors for PAAD.
FIGURE 4
FIGURE 4
Risk score model development and validation. (A) Trianing cohort risk score distribution of a patient with PAAD based on FRLs. (B) Trianing cohort scatter plots showing the association between the OS and the risk score in PAAD patients according to prognostic features of FRLs. (C) Trianing cohort heatmap showing three unfavorable genes (LINC01705, AC099850.3, and MIR193BHG) with high expression in high-risk patients, contray to the expression of the three favorable genes (AC068620.2, TRAF3IP2-AS1, and AC092171.2). (D) Trianing cohort KM survival curve analysis. (E) Testing cohort area under the ROC curve based on FRLs-based prognostic features at 12 months. (F) Testing cohort risk score distribution of a patient with PAAD based on FRLs. (G) Testing cohort scatter plots showing the association between the OS and the risk score in PAAD patients according to prognostic features of FRLs. (H) Testing cohort heatmap showing three unfavorable genes (LINC01705, AC099850.3, and MIR193BHG) with high expression in high-risk patients, contray to the expression of the three favorable genes (AC068620.2, TRAF3IP2-AS1, and AC092171.2). (I) Testing cohort KM survival curve analysis. (J) Testing cohort area under the ROC curve based on FRLs-based prognostic features at 12 months. (K) Entire cohort risk score distribution of a patient with PAAD based on FRLs. (L) Entire cohort scatter plots showing the association between the OS and the risk score in PAAD patients according to prognostic features of FRLs. (M) Entire cohort heatmap showing three unfavorable genes (LINC01705, AC099850.3, and MIR193BHG) with high expression in high-risk patients, contray to the expression of the three favorable genes (AC068620.2, TRAF3IP2-AS1, and AC092171.2). (N) Entire cohort KM survival curve analysis. (O) Entire cohort area under the ROC curve based on FRLs-based prognostic features at 12 months.
FIGURE 5
FIGURE 5
Estimated prognostic accuracy of FRLs in patients with PAAD. (A) Univariate Cox regression showing that the age, N-stage, and risk score were associated with OS (p < .05). (B) Multivariate Cox regression showing that the age and the risk score (p < .01) were independent prognostic indicators of OS in patients with PAAD. (C) ROC curve showing that the risk score has the highest prognostic accuracy.
FIGURE 6
FIGURE 6
Survival rates of PAAD patients with high- and low-risk patients with PAAD in the subgroups based on clinicopathological characteristics. (A) Subgroup of age <65 years. (B) Subgroup of age ≥65 years. (C) Male subgroup. (D) Female subgroup. (E) G1 and G2 subgroups. (F) G3 and G4 subgroups. (G) T1 and T2 subgroups. (H) T3 and T4 subgroups. (I) N0 subgroup. (J) N1 subgroup.
FIGURE 7
FIGURE 7
Construction and validation of the nomogram. (A) Prognostic nomogram based on the risk score using the prognostic FRLs signature and the age to predict 12-, 18-, and 24-month survival in patients with PAAD. Calibration curves corrected for deviations in agreement between the predicted and observed survival rates at (B) 12, (C) 18 and (D) 24-months.
FIGURE 8
FIGURE 8
Ferroptosis-related lncRNA-mRNA co-expression network and functional enrichment. (A) LncRNA-mRNA network showing 76 lncRNA-mRNA co-expression pairs formed between six FRLs and 61 mRNAs. The yellow rectangles denote FRLs, and the blue rectangles denote mRNAs. (B) GO analysis showing enriched biological function of these mRNAs co-expressed with six FRLs. (C) KEGG pathway analysis showing enriched signaling pathways.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Badgley M. A., Kremer D. M., Maurer H. C., DelGiorno K. E., Lee H.-J., Purohit V., et al. (2020). Cysteine Depletion Induces Pancreatic Tumor Ferroptosis in Mice. Science 368 (6486), 85–89. 10.1126/science.aaw9872 - DOI - PMC - PubMed
    1. Bannon M. J., Savonen C. L., Jia H., Dachet F., Halter S. D., Schmidt C. J., et al. (2015). Identification of Long Noncoding RNAs Dysregulated in the Midbrain of Human Cocaine Abusers. J. Neurochem. 135 (1), 50–59. 10.1111/jnc.13255 - DOI - PMC - PubMed
    1. Bhan A., Soleimani M., Mandal S. S. (2017). Long Noncoding RNA and Cancer: A New Paradigm. Cancer Res. 77 (15), 3965–3981. 10.1158/0008-5472.CAN-16-2634 - DOI - PMC - PubMed
    1. Du C., Zhang J.-l., Wang Y., Zhang Y.-y., Zhang J.-h., Zhang L.-f., et al. (2020). The Long Non-coding RNA LINC01705 Regulates the Development of Breast Cancer by Sponging miR-186-5p to Mediate TPR Expression as a Competitive Endogenous RNA. Front. Genet. 11, 779. 10.3389/fgene.2020.00779 - DOI - PMC - PubMed
    1. Fabbri C., Serretti A. (2016). Genetics of Long-Term Treatment Outcome in Bipolar Disorder. Prog. Neuro-Psychopharmacology Biol. Psychiatry 65, 17–24. 10.1016/j.pnpbp.2015.08.008 - DOI - PubMed

LinkOut - more resources