OLFML2A Overexpression Predicts an Unfavorable Prognosis in Patients with AML

Background Acute myeloid leukemia (AML) is a malignant clonal disease of the myeloid hematopoietic system. Clinically, standard treatment options include conventional chemotherapy as well as hematopoietic stem cell transplantation. Among them, chemotherapy has a remission rate of 60% to 80% and nearly 50% relapse in consolidation therapy. Some patients have a poor prognosis due to the presence of unfavorable factors such as advanced age, hematologic history, poor prognosis karyotype, severe infection, and organ insufficiency, which cannot tolerate or are not suitable for standard chemotherapy regimens, and scholars have tried to find new treatment strategies to improve this situation. In the pathogenesis and treatment of leukemia, epigenetics has received attention from experts and scholars. Objective To investigate the relationship between OLFML2A overexpression and AML patients. Methods From The Cancer Genome Atlas, researchers used the data of OLFML2A gene to analyze and study the pan-cancer using R language and then divided the high and low levels of this protein into two groups to study its relationship with the clinical characteristics of the disease. The relationship between the high levels of OLFML2A and various clinical features of the disease was studied with emphasis on the relationship between the high levels of OLFML2A and various clinical features of the disease. A multidimensional Cox regression analysis was also performed to study the factors affecting patient survival. The correlation between OLFML2A expression and immune infiltration through the immune microenvironment was analyzed. The researchers then conducted a series of studies to analyze the data collected in the study. The focus was on the relationship between the high levels of OLFML2A and immune infiltration. Gene ontology analysis was also performed to study the interactions between the different genes associated with this protein. Results According to the pan-cancer analysis, OLFML2A was differentially expressed in different tumors. More importantly, the analysis of OLFML2A in the TCGA-AML database revealed that OLFML2A was highly expressed in AML. The researchers found that the high levels of OLFML2A were associated with different clinical features of the disease, and that the expression of the protein was different in different groups. Those patients with the high levels of OLFML2A were found to have substantially longer survival times compared to those with low-protein levels. Conclusions The OLFML2A gene is able to act as a molecular indicator involved in the diagnosis, prognosis, and immune process of AML. It improves the molecular biology prognostic system of AML, provides help for the selection of AML treatment options, and provides new ideas for future biologically targeted therapy of AML.


Introduction
Acute myeloid leukemia (AML) is a malignant disease of the hematological system with strong biological and clinical heterogeneity. Currently, patients are generally stratifed according to their cytogenetic and molecular biological fndings to predict their treatment outcome [1]. In practice, we have found that the prognosis of patients graded according to the current risk stratifcation system still has some variability in treatment outcome and survival time for patients in the same risk stratum, suggesting that we should have a more detailed stratifcation prognosis system. In recent years, studies have found that AML is associated with multiple mutated genes, suggesting that the development of leukemia may be the result of accumulation of mutations in multiple genes [2,3]. In recent years, with the widespread use of high-throughput sequencing, i.e., second-generation sequencing technology, the role of leukemia-related mutated genes in diagnosis and treatment has been gradually highlighted [4]. Applying sequencing technology, we have successively discovered a variety of mutated genes related to the development and treatment of AML, and a large number of studies of these genes in AML have also emerged.
Te OLFML2A (olfactomedin-like 2A) gene is located on chromosome 9q33.3 and encodes a protein also known as "photomedin-1," which belongs to the OLFM (Olfactomedin, OLFM) family IV [5]. Olfactomedin, an exocrine glycoprotein secreted by the epithelial cells of the olfactory organ and deposited in large quantities on its surface, was discovered in 1991 and was the frst member of the OLFM family [6]. More than half of the OLFM family proteins are expressed in the neural tissue [7]. A large body of evidence suggests that OLFM family proteins play an important regulatory role in neurogenesis, neural crest formation, intercellular adhesion, tumor development, and cell cycle regulation [8,9]. Related studies have shown that the OLFM family proteins are key regulatory molecules of cellular signaling pathways such as the Wnt signaling pathway [10,11]. Tere is increasing evidence that the OLFM family proteins play important roles in the normal tissue development and disease development, e.g., myocilin and olfactomedin 2 are key molecules in the development of glaucoma [12], and OLFM4 is associated with the development of common malignancies such as gastric and pancreatic cancers [13,14].
OLFML2A is a member of OLFM family IV, which contains at least eight exons spanning 37.7 kb and encodes a protein with an olfactomedin structural domain at the Cterminus and a unique serine/threonine region that distinguishes it from other proteins in the family, two to three potential glycosylation sites at the N-terminus, and homodimers or oligomers with disulfde bonds. Northernblot of diferent tissue specimens from mice showed that OLFML2A transcription products were not found in the brain tissue [15]. OLFML2A is an exocrine glycoprotein that binds specifcally to chondroitin sulphate-E (CS-E) and heparin [16]. CS-E binds to a number of heparin-binding growth factors, including midkine, Pleiotrophin, several FGFs, and HB-EGF. Specifc binding of OLFML2A to CS-E may promote the local action of growth factors bound to CS-E. To date, the specifc functions of the OLFML2A gene and its encoded protein remain unknown, and its role in the development of AML has not been reported in the literature.
Based on this, from Te Cancer Genome Atlas, researchers analyzed the data using OLFML2A gene data, analyzed and studied pan-cancer using R language, and then divided the high and low levels of the protein into two groups in order to study its relationship with the clinical features of the disease. Finally, it was concluded that the OLFML2A gene, as a molecular indicator, can be involved in the diagnosis, prognosis, and immune process of AML and has the potential to be a reliable prognostic assessment indicator W and a potential therapeutic boot point for AML patients.

Preprocessing of Raw Data.
We collected TCGA-AML expression profles and clinical information from TCGA Genomic Data Commons Data Portal (https://portal.gdc. cancer.gov/). We excluded the insufcient cases or missing data in the later information processing. Te genomic expression information of OLFML2A was calculated from the TCGA database by high-throughput sequencing. Because all information was publicly available, no ethical approval was required.

Pan-Cancer Analysis.
Pan-cancer analysis was performed through the TIMER2 (Tumor Immunology Estimation Resource, version 2) network (https://timer. cistrome.org/) 16 to observe diferences in OLFML2A expression in tumors and nearby normal tumor tissue or particular tumor subtypes in the TCGA program.

Gene Expression Analysis.
We studied the correlation between diferent tissue characteristics and OLFML2A expression through the Wilcoxon rank-sum test.

Survival and Clinical Statistical Analysis.
Te association of OLFML2A with clinical features and overall survival was evaluated using the log-rank tests, Kaplan-Meier survival curves, and one-way Cox analysis. Te correlation between high and low OLFML2A expression and clinical features was researched in this study (age, grade, BM blasts, cytogenetic risk, FLT3 mutation, IDH R132 mutation, IDH R140 mutation, NPM1 mutation, PB blasts, RAS mutation, and WBC count) between OLFML2A mutations. In addition, OLFML2A was split into high and low expression groups. Te OLFML2A expression was judged in relation to overall survival by confrming the high and low OLFML2A expression based on the median. We applied survROC software to measure the accuracy of risk scores on prognosis using time-dependent subject operating characteristic (ROC) curves. We conducted univariate and multivariate analyses of risk scores after adjusting for age, sex, race, BM blasts, PB blasts, Cytogenetic risk, and FAB classifcations. In addition, we analyzed the expression of OLFML2A with diferent clinicopathological features, and we investigated the association of OLFML2A expression with BM cells, cytogenetic risk, FAB classifcation, IDH1 R132 mutation, IDH1 R132 mutation, NPM1 mutation, and race.

Construction of Nomograms.
Since the development of nomograms, they have been used to forecast cancer prognosis. Tis method uses a statistical method to score various factors, such as age, gender, and the TNM stage. It can then produce a total score that provides a personalized estimate of the likelihood of the disease returning. In the study, the researchers used nomograms to predict the likelihood of patients with cancer returning. Te R package rms generated them. Te nomograms were then validated by implementing a series of calibration curves. Subsequently, we utilized the cindex to estimate nomogram accuracy.
2.6. Immunological Analysis. Te researchers then performed a series of studies to analyze the data collected from the studies. Tey frst used a computer program known as the CIBERSORT deconvolution to determine the relative composition of immune cells in each sample. Ten, they performed an immune diferential analysis to study the two groups' immune cell content diferences.

Gene Ontology (GO) and Kyoto Encyclopedia of Genes and
Genomes (KEGG). Te cluster profle package was used for GO and KEGG enrichment analysis and underlying biological pathways that were likely to adjust the cancer development. Tey were also able to identify promising signals that could be linked to the disease.

OLFML2A Had a High Expression in Multiple Cancer
Tissues. To detect the diferential expression of OLFML2A, we frst investigated OLFML2A gene expression in 33 human cancers in TCGA using the TIMER database.

Te Diference in Clinical Characteristics.
We grouped high and low OLFML2A gene expression, and the correlation between OLFML2A gene expression and clinical features were explored. We incorporated age, grade, BM blasts, cytogenetic risk, FLT3 mutation, IDH R132 mutation, IDH R140 mutation, NPM1 mutation, PB blasts, RAS mutation, and WBC count for picture mapping. After analyzing the data collected from the studies, the researchers concluded that the high levels of the OLFML2A were different from the low levels of the protein. Tey also noted that the diference was signifcant when it came to age, BM blasts, and FLT3 mutations (P < 0.05; Figure 2).

Te Prognosis and Diagnosis of OLFML2A Value.
Compared with high-risk patients, according to Kaplan-Meier survival curves, the survival of low OLFML2A expression patients was longer (log-rank test; P < 0.001) (Figure 3(b)). Figure 3(a) shows 0.977, indicating the area under the curve (AUC) value.

Te Diference in Clinicopathological Feature.
Trough the analysis between OLFML2A expression and diferent clinicopathological characteristics, it could be seen that there was a relationship between OLFML2A expression and FAB classifcations, BM blasts, IDH1 R132 mutation, IDH1 R132 mutation, cytogenetic risk, NPM1 mutation, and race ( Figure 4 and Tables 1 and 2). In addition, univariate and multifactorial Cox regression analyses presented that an independent risk factor for AML was OLFML2A expression (Table 3).
OLFML2A is an abnormal protein that can be found in various tissues such as the breast, colon, ovary, and liver [1]. Researchers have also found that high levels of this protein Journal of Oncology are detrimental to patients with acute lymphoblastic leukemia. Te researchers hypothesized that the presence of OLFML2A in these patients could help to predict the likelihood of their cancer recurrence. Tey noted that the high levels of this protein could be a target for novel cancer treatments. In a previous study, researchers found that the presence of OLFML2A in breast cancer cells could hinder the development and metastasis of cancer cells [20]. Te knockdown of OLFML2A in glioma cells inhibits the Wnt/ β-catenin signaling pathway, which leads to upregulation of amyloid precursor protein (APP) expression and a decrease in the degree of stable β-catenin, resulting in having reduced MYC, CD44, and CSKN2A2 expression, thereby inhibiting cell proliferation and promoting apoptosis [21,22].  Journal of Oncology Furthermore, by exploring the signifcance of OLFML2A expression in many clinical parameters, we found an association between OLFML2A and AML survival and clinical features. Tis study performed an immune cell infltration analysis to gain insight into the role of OLFML2A. From Te Cancer Genome Atlas, researchers used OLFML2A gene data to analyze the data. Pan-cancer was analyzed and studied using the R language. Tey then divided the high and low levels of the protein into two groups to study their relationship with clinical features of the disease. Te researchers then conducted a series of studies to analyze the data collected from the study. Tey focused on the relationship between the high levels of OLFML2A and various clinical features of the disease. Tey also performed a multidimensional Cox regression analysis to examine the factors that afect patient survival. We analyzed the correlation between OLFML2A expression and immune     infltration in the immune microenvironment. Te researchers then conducted a series of studies to analyze the data collected from the study. Tey focused on the relationship between the high levels of OLFML2A and immune infltration. Tey also performed gene ontology analysis to examine the interaction between diferent genes associated with the protein. OLFML2A was diferentially expressed in a variety of tumors based on pan-cancer analysis, including the brain cell counting system, DLBC, ESCA, BRCA, CHOL, LGG, COAD, lipocytes, Kipres, GBM, chronic cell count, oligosaccharide nucleic acid, oligos nucleic acid, adipocyte leukocyte leukemia, adipocyte count enzyme, hyaluronidase, cycloplasmic carcinoma, growth hormone, paclitaxel leukocyte leukemia, prostaglandin, cerebroside leukocyte acid, goitre, and auscocis. In addition, the analysis of OLFML2A in the TCGA-AML database revealed that OLFML2A is highly expressed in AML. Te researchers found that high OLFML2A levels were associated with different clinical features of the disease. Tey also noted that protein expression was diferent in diferent groups. Patients with the high levels of OLFML2A were found to survive longer compared to those with low-protein levels. Te researchers found that the high levels of OLFML2A were associated with various clinical features of the disease. Tey also noted that the protein was expressed diferently in diferent groups. Some of these clinical features include BM primitive cells, cytogenetic risk, and IDH1 R132 mutations. Using columnar line graphs, it was possible to predict patient survival based on OLFML2A levels. A relationship was also found between this protein and the growth of acute lymphoblastic leukemia. In the immune microenvironment, the researchers discussed the positive correlation between OLFML2A and various immune cell activities. In parallel, we completed a GO analysis. Te CC terminology encompasses "encapsulated vesicle membrane," "cell-matrix junction," "focal adhesion," "focal adhesion," and "cell-matrix junction." BP terms include "autophagy regulation," "positive regulation of phosphatase activity," "sterol metabolic process," and "phosphorylation."

Conclusion
Te OLFML2A gene is able to act as a molecular indicator involved in the diagnosis, prognosis, and immune process of AML. It improves the molecular biology prognostic system of AML, provides help for the selection of AML treatment options, and provides new ideas for future biologically targeted therapy of AML.

Data Availability
Te experimental data used to support the fndings of this study are available from the corresponding author upon request.

Disclosure
Xuan Lu and Ying Li are common frst authors.

Conflicts of Interest
Te authors declare that they have no conficts of interest regarding this work.