The primary finding of the present study is the successful construction of a prediction model for short-term and long-term prognosis in AML patients based on cuproptosis-related lncRNA signature. The identification of cuproptosis-related lncRNAs and the development of a prediction model might facilitate decision making in the establishment and implementation of AML treatment strategies in clinical practice and drug development.
In the present study, 454 cuproptosis-related lncRNAs were extracted from the TCGA-LAML data. Subsequently, LASSO, univariable and multivariable Cox regression analyses were performed to identify the cuproptosis-related lncRNAs associated with OS of AML patients. Seven lncRNAs were utilized in the construction of a prognostic model for AML prognosis prediction. Evidence exists that one of them, PSMB8-AS1, regulates cell proliferation, apoptosis, and radio-resistance in glioblastoma [27]. In addition, the modulation of miR-574-5p/RAB10 expression by PSMB8-AS1 was found to promote the proliferation of glioma cell [28]. In other studies, the preferential expression of the gamma-globin genes was regulated by NFE4, which is an indispensable component in the prognostic model of clear cell renal cell carcinoma [29, 30]. LINC00989 was confirmed to be associated with the OS prognosis of patients with breast cancer and hepatocellular carcinoma [31, 32]. These findings suggest that cuproptosis might play biological roles in solid tumors other than AML. In addition, the differential expression of the predictive lncRNAs was validated in clinical samples. To the best of our knowledge, the functional role of LINC02062, AC006460.2, AL353796.1, and AC000120.1 has not been reported. Nevertheless, further studies are warranted to unveil their functional roles in cancer treatment.
To validate the clinical significance of the proposed signature, the AML patients were divided into a high- and low-risk groups based on their risk scores. The low-risk group had a better prognosis than the high-risk group in all cohorts. Previous study established that the long-term survival of AML declined with age [5], which is in agreement with the findings of the present study. The AUC of the risk score was 0.846, indicating a good performance in predicting AML prognosis, which was further confirmed by the C-index curves. In addition, the stratified survival analysis revealed that the established model had good performance in different age groups and gender distribution, and can thus have a wide application in clinical practice. To further validate the utility of the model, we performed PCA, whose results showed that the lncRNAs used in the constructed model had the most pronounced distinction, indicating that they could be used to discriminate among patients from different risk groups. These findings suggest that the model established by the seven cuproptosis-related lncRNAs was reliable in predicting the prognosis of AML.
To determine the possible mechanism involved in AML survival, we performedseveral functional enrichment analyses. GSEA demonstrated that the high-risk group was enriched in the immune-associated pathways. Interestingly, all the immune-related functions and 13 of the 16 immune-related cells were significantly enriched in the ssGSEA analysis. Among these, all the immune-related functions and cells were overexpressed in the high-risk group. T-cell co-inhibition is a vital element contributing to immune function suppression by providing inhibitory signals to activated T cells [33, 34]. Relative study showed that immune-linked processes including T-cell co-stimulation and antigen presentation had a significant correlation with post-transplantation relapses in AML [35]. Moreover, macrophages and Tregs were significantly upregulated in the high-risk group. CD4 + regulatory T cells (Tregs) are highly immune suppressive and considered as pivotal regulators of immune escape for inhibiting the proliferation and function of immune killer cells through cellular contact and suppressive cytokine production [36]. Macrophages that reside within the tumor microenvironment are known as tumor-associated macrophages (TAMs), promoting immune suppression through enhanced angiogenesis, metastasis and chemoresistance [37]. The increase in Tregs and TAMs is associated with a poorer survival in AML [38, 39]. Furthermore, T- and NK-cell exhaustion and dysfunction, which contribute to immune disorder and tumor immune escape [40], were correlated with therapeutic reactivity, high risk for relapse, and unfavorable prognosis of AML [41]. Therefore, we speculated that cuproptosis-related lncRNAs may modulate the tumor microenvironment and promote tumor immune evasion through inhibitory immune cells, which result in favor leukemia survival. And it could be reasonable to infer that a possible tight connection might also exist between cuproptosis and tumor immunity in AML.
This study is not without limitations. Because no other databases have available lncRNA expression for our obtained lncRNAs, the utility of this model was confirmed by a validation cohort in one database. Although the differential expression of the prognostic lncRNAs was validated by qRT-PCR analysis, more prospective investigations are needed to validate its predictive power. Since the role of cuproptosis in AML has not been reported before, in future studies we will further explore their correlations and undermine potential mechanisms.