Page 38 - TD-3-3

Tumor Discovery LncRNA HA117 in osteosarcoma regulation

pathway enrichment analysis, using the human genome from biopsy procedures. We analyzed samples from
genes provided by DAVID as the background list. The 15 patients, consisting of 10 non-chemotherapy and five
statistical significance of enriched GO terms and Reactome chemotherapy biopsies. Among these patients, 10 (67%)
pathways was assessed using the hypergeometric test and were male and 5 (33%) were female. The mean age of
corrections were made using the Benjamini-Yekutieli the patients was 29.7 years, with an age range from 8 to
method to control the FDR. The P-values and FDR were 80 years.
calculated to indicate the significance of gene enrichment
and to adjust the P-values, respectively. Only GO terms 3.2. Summary of the 51 osteosarcoma sequencing
and Reactome pathways with a P < 0.1 were output by samples
DAVID, and significantly enriched terms or pathways As a recently discovered lncRNA, HA117 was not included
were defined based on a P-value cutoff below 0.05. The R in the hg19 annotation database. Therefore, we merged
ggplot2 package was utilized to visualize the results of HA117 with other mRNAs to construct a custom database
38
GO and Reactome enrichment analyses. for read mapping. The summary of the characteristics of the
51 samples and genome mapping results are described in
2.5. Survival analysis of 11 HA117 target genes Table S3. After quality control of the readings, it was found
Since gene expression data and corresponding survival that the number of reads in FFPE samples was relatively
data for osteosarcoma patients are not available in publicly low, ranging from 7.24 to 27.35 million, while the number
accessible databases, we utilized data from sarcoma of reads for the 36 fresh samples ranged from 56.46 to
patients available in the Kaplan–Meier plotter database 38.69 million, which was higher than that of FFPE samples.
39
for survival analysis. The Kaplan–Meier plotter (https:// Based on alignments of clean reads to the hg19 genome
kmplot.com/analysis/) is a publicly accessible database that and custom gene sets, we found the genome alignment
offers gene expression data to evaluate survival outcomes rates of FFPE samples were 50.24% – 55.25%. This lower
in patients with various cancer types, including sarcoma. alignment rate may be attributed to the degradation of RNA
In this study, we downloaded gene expression data and in the samples stored in paraffin for extended periods. The
survival information for 259 sarcoma patients from the alignment rates for fresh samples were significantly higher
TCGA and GEO databases. From FFPE samples, we than the FFPE samples, ranging from 72.38% to 79.79%.
identified 11 DEGs that are also predicted target genes for In addition, the number of expressed genes was higher in
HA117. These DEGs were utilized for survival analysis to fresh samples compared to FFPE samples, with an average
compare patients with low and high gene expression levels. of 18,967 genes expressed in fresh samples and only 7033
Statistical significance was assessed using the hazard ratio genes expressed in FFPE samples.
(HR) with a 95% confidence interval (CI) and the log-rank
P-value. 3.3. DEGs and HA117 analysis in osteosarcoma
To identify significant DEGs in osteosarcoma, we
3. Results performed differential analysis using edgeR software
on fresh tumor tissues and their paired samples, fresh
3.1. Sample characteristics
samples with and without chemotherapy, as well as
A total of 51 whole transcriptome sequencing samples were FFPE samples with and without chemotherapy. First,
downloaded for this study, comprising 36 fresh samples and by comparing tumors (T group) with normal (N group)
15 FFPE samples. The 36 fresh samples were obtained from tissues in all fresh samples (N vs. T), we identified 2794
osteosarcoma and adjacent tissues of 18 patients, while the significant DEGs based on the criteria |log2fold|>1 and
15 FFPE samples were obtained only from osteosarcoma FDR < 0.001. Among these, 1104 genes were up-regulated,
tissues. The characteristics of osteosarcoma patients in the and 1960 genes were down-regulated (Figure 1A). The five
present study are listed in Table S2. The 36 fresh samples most up-regulated genes in osteosarcoma were COL2A1,
were obtained from 18 Vietnamese osteosarcoma patients COL9A1, PENK, CAPN6, and CHRDL2. The five most
who underwent surgical resection of the affected bone. The down-regulated genes were DEFA1, PRG4, DEFA3,
mean age of these patients was 18.8 years, ranging from DEFA1B, and MPO (Table S4). Next, to investigate the
7 to 52 years. The gender distribution included 12 males effect of chemotherapy on gene expression in tumors,
and six females, indicating more males than females in the we compared tumor tissues in fresh samples between
samples. These cases mainly involved the tibia (six cases), the non-chemotherapy group (No-ch-T group) and
the femur (10 cases), and the humerus (two cases). Among the chemotherapy group (Ch-T group) (No-ch-T vs.
these patients, 15 received chemotherapy, and three did Ch-T). Using the same screening criteria, we identified
not receive chemotherapy. All FFPE samples were obtained a total of 2704 DEGs, with 1214 up-regulated genes

Volume 3 Issue 3 (2024) 4 doi: 10.36922/td.3670