Exploring the causal relationship between hemoglobin and pancreatic cancer and its potential mechanisms through bioinformatics and Mendelian randomization
Aug 5, 2025Discover oncology
Possible links and biological pathways between hemoglobin levels and pancreatic cancer
Elevated hemoglobin (HGB) levels are associated with a reduced risk of pancreatic adenocarcinoma (PAAD) (β_IVW = -0.40, OR_IVW = 0.66, p = 0.013).
Genetic analysis suggests a negative correlation between HGB levels and the development of PAAD.
Seven key genes linked to HGB levels were identified as potential independent risk factors for poor prognosis in PAAD.
Patients with high HGB scores based on these genes showed significantly poorer overall survival compared to those with low scores (p < 0.0001).
The demonstrated strong predictive accuracy for 1-, 3-, and 5-year overall survival with AUC values of 0.77, 0.79, and 0.91, respectively.
Validation with external datasets confirmed the model's robustness and reliability in predicting overall survival in PAAD patients.
AI simplified
BACKGROUND: Abnormal hemoglobin (HGB) levels and the onset of malignant tumors have attracted substantial clinical interest. PAAD, a highly fatal malignancy of the digestive system, warrants further investigation regarding its potential link with HGB levels. To explore the genetic relationship between the two, we employed in conjunction with transcriptomic analysis to probe their underlying connection.
METHODS: A combined approach utilizing Mendelian randomization (MR) and transcriptomics was adopted to examine the genetic association between HGB levels and PAAD, along with possible mechanistic pathways. Based on GWAS datasets derived from European populations, MR analysis was conducted to evaluate the causal relationship between HGB levels and the risk of PAAD. To test the reliability of the results, heterogeneity and directional pleiotropy were evaluated using the MR-Egger intercept test, Cochran's Q test, and leave-one-out analysis. Transcriptomic datasets from TCGA and GEO were then integrated to identify differentially expressed genes, followed by functional enrichment analysis. LASSO regression was subsequently applied to select characteristic genes and construct a , which was then subjected to validation.
RESULTS: MR analysis revealed a negative association between HGB levels and the development of PAAD. Genetically, elevated HGB levels were linked to a reduced risk of PAAD (β_IVW = - 0.40, OR_IVW = 0.66, 95% CI = 0.48-0.92, p = 0.013). Using the PAAD dataset, seven key genes (DNMT3A, TFCP2L1, PPARGC1A, GSTA5, BICC1, NRG4, BCL2L13) were identified through LASSO regression, and HGB scores were computed based on their expression. Kaplan-Meier survival curve analysis indicated that patients with high scores exhibited significantly poorer overall survival (OS) than those in the low-score group (p < 0.0001). The scoring model demonstrated high predictive accuracy for 1-, 3-, and 5-year OS, with AUC values of 0.77, 0.79, and 0.91, respectively. Multivariate Cox regression and prognostic modeling of the seven genes showed that, apart from NRG4, the remaining six were independent risk factors associated with unfavorable prognosis in PAAD (all p < 0.05). The model yielded a C-index of 0.72, reflecting strong predictive power. Column-line plots further confirmed the model's effective performance for predicting 1-, 3-, and 5-year OS. Validation with the GSE85916 and TCGA-PAAD dataset demonstrated consistent robustness of the model in forecasting OS in PAAD patients, reinforcing its reliability and potential applicability.
CONCLUSIONS: This study identified a genetic causal relationship between HGB levels and the risk of PAAD. Through transcriptomic analysis, we constructed a prognostic model based on HGB-associated key genes. The model displayed reliable predictive capacity and offers new perspectives for clinical strategies aimed at preventing PAAD.
Key numbers
0.66
Odds Ratio for PAAD Risk
Odds ratio derived from analysis.
0.77
1-Year OS AUC
Area under the curve for the 1-year overall survival prediction.
0.91
5-Year OS AUC
Area under the curve for the 5-year overall survival prediction.
Full Text
We can’t show the full text here under this license. Use the link below to read it at the source.
