Page 29 - TD-1-2

Tumor Discovery Prognostic biomarkers in pancreatic cancer

high-risk and low-risk subgroups. A differential analysis PAAD patients were constructed (Figure 11A). Based on
of tumor mutational burden was then performed and the the correlation plots, the observed and predicted rates of
tumor mutational burden (TMB) was then calculated from survival in PAAD patients at 1, 3, and 5 years showed good
TGCA somatic mutation data. The low-risk group had agreement (Figure 11B).
lower TMB than the high-risk group (Figure 7F). Next, a
survival analysis of TMB was performed. Figure 8A shows 4. Discussion
that the survival of the low TMB group was better than that Pancreatic cancer is the main cause of cancer-related death
of the high TMB group, and then combined with the TMB worldwide and has been a serious threat to human life
with the patient risk score for survival analysis, Figure 8B and health due to its insidious onset, strong invasiveness,
shows that patients with low TMB and low-risk score were poor prognosis, and high mortality rate [2,18,19] . Through
found to have a higher probability of survival. further research, it has been found that the disorder of

3.5. Identification of potential drugs for prognostic m6A methylation modification regulation may affect the
models processing, degradation, and translation of mRNA, resulting
in the activation of oncogenes and the inactivation of
To explore potential drugs for the treatment of PAAD tumor suppressor genes, and the occurrence, development,
patients with our prognostic model, we used the pRRophetic and drug resistance of malignant tumors. The occurrence
algorithm to estimate treatment response based on the half- of m6A is closely related, and m6A changes play a crucial
maximal inhibitory concentration (IC50) of each drug in the role in carcinogenesis and tumor progression .
[8]
Genomics of Cancer Drug Sensitivity (GDSC) database. We
screened for six drugs with significantly different estimated M6A plays a post-transcriptional modification role in
IC50s between the high- and low-risk groups, and the low- eukaryotic mRNAs and lncRNAs, such as in regulating mRNA
risk group was found to be more sensitive to most of the transcription, splicing and translation, as well as affecting the
potential drugs. Figure 9 shows six potential drugs that can structure and function of lncRNAs with extensive regulatory
[11]
be used for further analysis of PAAD patients. roles . M6A regulators can modify specific lncRNAs,
and lncRNAs can maintain malignancy in various tumors
3.6. Independent prognostic analysis of through transcriptional, epigenetic, and post-transcriptional
prognostic models and assessment of clinical levels [10,20] . The role of m6A-regulated lncRNAs may be
features of PAAD critical for the proliferation and migration of cancer cells .
[11]
Studies have reported that m6A methylation modification
We performed univariate and multivariate Cox regression of lncRNA can affect the occurrence and development of
analyses to assess whether risk models for m6A-related tumors, and m6A modification can also affect the formation
LncRNAs had independent prognostic features of of RNA-DNA triple helix, in which one lncRNA binds to this
PAAD. Based on Figure 10A, first in the univariate Cox series through the Hoogsteen base pair in the main groove
regression analysis, the HRs for the risk score and 95% of double-stranded DNA. In addition, m6A may also affect
confidence interval (CI) were 1.181 and 1.097−1.271, the reciprocal site between lncRNA and specific DNA [21,22] .
respectively (p < 0.001). Based on Figure 10B, HR was Both m6A and lncRNA are important regulators of PAAD
1.162 in multivariate Cox regression analysis, 95% CI was occurrence. However, studies on their roles and biological
1.074−1.257 (P < 0.001). A concordance index analysis of mechanisms in PAAD progression are still relatively
the risk score was then performed and it was found that lacking [13,17] . In this study, an independent prognostic model
the concordance index of the risk score was consistently
greater than other clinical factors over time, suggesting based on m6A-related lncRNA was constructed, inspired by
that the risk class could better predict the prognosis the functions of m6A and lncRNA in PAAD.
of PAAD patients (Figure 10C). Thereafter, the area In this work, 14056 m6A-associated lncRNAs were
under the ROC curve (AUC) analysis of risk grades was identified from the TCGA dataset to explore the prognostic
performed (Figure 10D and E), and the AUCs of risk score functions of m6A-associated lncRNAs. After confirming
grades were also shown to be higher than those of other the prognostic value of m6A-related lncRNAs in the
clinical features, indicating that the prognostic risk model TCGA dataset, five of them were selected to construct
constructed in this study was relatively reliable. m6A-related lncRNA prognostic models to predict the
survival of PAAD patients. Model validation for clinical
3.7. Construction and evaluation of prognostic grouping was also performed, the risk scores for each
nomograms patient in the training group and across the entire set
Nomograms including risk classes and clinical were calculated, and principal component analysis was
characteristics to predict 1-, 3-, and 5-year survival in performed to validate the prognostic model, all of which

Volume 1 Issue 2 (2022) 12 https://doi.org/10.36922/td.v1i2.165