Correlation analysis of long non‐coding RNA TUG1 with disease risk, clinical characteristics, treatment response, and survival profiles of adult Ph− Acute lymphoblastic leukemia

Abstract Background Long non‐coding RNA taurine‐upregulated gene 1 (lncRNA TUG1) is reported to be involved in the progression and development of several malignancies; however, its role in Philadelphia chromosome‐negative acute lymphoblastic leukemia (Ph−ALL) is unknown. The present study aimed to explore the correlation of lncRNA TUG1 with disease risk, disease condition, and prognosis of adult Ph−ALL. Methods Total 101 adult Ph− ALL patients and 40 bone marrow (BM) donors were included, followed by detection of BM monocyte cell lncRNA TUG1 expression by reverse transcription‐quantitative polymerase chain reaction. According to the quantiles of lncRNA TUG1 expression in Ph− ALL patients, these patients were divided into four tiers: tier 1 (ranked in 0%~25%), tier 2 (ranked in 25%~50%), tier 3 (ranked in 50%~75%), and tier 4 (ranked in 75%~100%). Results LncRNA TUG1 was upregulated in Ph− ALL patients compared with healthy donors. Further analysis indicated that in Ph− ALL patients, higher lncRNA TUG1 tier was correlated with the presence of central nervous system leukemia, increased white blood cell level, and bone marrow blasts. Furthermore, higher lncRNA TUG1 tier was negatively associated with complete remission (CR) within 4 weeks, total CR, and allogeneic hematopoietic stem cell transplant achievement. In addition, higher lncRNA TUG1 tier was associated with decreased disease‐free survival and overall survival, which was further verified to be an independent factor by Cox's regression analysis. Conclusion lncRNA TUG1 presents potential to be a novel biomarker for disease risk assessment and survival surveillance in Ph− ALL management.


| INTRODUC TI ON
Acute lymphoblastic leukemia (ALL), as a heterogeneous hematologic disease, is characterized by the abnormal proliferation of immature lymphoid cells in the bone marrow, peripheral blood, and organs. 1,2 According to the previous study, approximately 60% of ALL cases are diagnosed at younger than 20 years old with the median age of 15 years, and additionally, ALL is considered as the most common form of childhood leukemia, accounting for 75% of pediatric acute leukemias. 3 The cytogenetics of t(9; 22) chromosomal translocation (also called Philadelphia chromosome (Ph)) is a common chromosomal abnormality in ALL, and there is a wide availability of target agents for Ph − positive (Ph + ) ALL. 3 As for Ph − negative (Ph − ) ALL, despite recent advancement in the treatment options (such as chemotherapy, hematopoietic cell transplant), part of patients with Ph − ALL still suffer from relapsed/refractory disease and presents poor survival outcomes with the 5-year overall survival (OS) rates for high-risk and standard-risk subgroups of approximately 29% and 54%, respectively. 1,4,5 Long non-coding RNA (lncRNA) taurine-upregulated gene 1 (TUG1), located on chromosome 22q12, binds to polycomb repressive complex 2 and thereby has regulatory effect on the p53dependent cell cycle regulatory gene expressions. 6 Additionally, existing evidences demonstrate that lncRNA TUG1 is regarded as a conserved cancer-related lncRNA, being aberrantly expressed in multiple tumor tissues, including pancreatic cancer, ovarian cancer, and colorectal cancer. [7][8][9][10] As for the role of lncRNA TUG1 in hematologic malignancy, lncRNA TUG1 is indicated to be involved in the progression and development of several malignancies, such as multiple myeloma (MM) and chronic lymphocytic leukemia. 11,12 For example, lncRNA TUG1 is highly expressed in MM patients compared with healthy controls, and clinically, it is correlated with increased MM disease stage, and was of value as a biomarker which helps to facilitate MM diagnosis. 11 However, the involvement of lncRNA TUG1 in Ph − ALL has not been determined yet, therefore, we conducted the present study to explore the correlation of lncRNA TUG1 with disease risk, disease condition, and prognosis of adult Ph − ALL.

| Subjects
This study retrospectively reviewed 101 adult Ph − ALL patients treated in our hospital between January 2014 and December 2019.
All analyzed patients met the following criteria: (a) had a diagnosis of primary Ph − ALL by bone marrow morphology, immunology, cytogenetics, and molecular biology (MICM) examinations, (b) aged more than 18 years, (c) had available clinical data and fresh-frozen pretherapy bone marrow (BM) specimen, (d) had complete treatment remission information in previous 4 weeks and in total, (e) had integrated follow-up data that were able to used for assessment of disease-free survival (DFS) and overall survival (OS), and (f) not complicated with other malignancies. Additionally, during the same period, 40 bone marrow donors were included as controls in the current study. This study was approved by Institutional Review Board of our hospital, and written informed consents were collected from all patients (or their families) and healthy donors.

| Date collection
Basic clinical data at diagnosis including age, gender, white blood cell (WBC) count, hemoglobin (HGB), blood platelet (PLT), bone marrow blasts, immunophenotype, and central nervous system leukemia (CNSL) were collected from the Computer-based Patient Record systems (CPRS). Besides, treatment response data including complete remission (CR) within previous 4 weeks and final response status were also extracted from the CPRS. The follow-up data were collected from the visit records to calculate DFS and OS.

| BM sample collection and store
BM samples of Ph − ALL patients were collected before initiation of therapy. After removing of plasma and partial red blood cells, the BM samples were concentrated, followed by adding of dimethyl sulfoxide and TC199 nutrient solution at 4℃; then, the BM samples were put into the polypropylene tube. After that, the tubes were placed in an automatic cooling machine to cool it to −80°C, which were subsequently store in liquid nitrogen until further analysis. The BM samples of health donors were collected when they underwent BM donation, which were treated and stored as described above.

| LncRNA TUG1 determination
The BM sample tube was removed from the liquid nitrogen and put into a 40 ℃ water bath for 2 minutes to melt. After that, 10% serum IMDM medium (Gibco) was added into the tube to dilute slowly, and then, Ficoll lymphocyte separation solution (Sigma-Aldrich) was used to separate BM monocyte cells (BMMCs) for lncRNA TUG1 determination by reverse transcription-quantitative polymerase chain reaction (RT-qPCR). In detail, total RNA was ex-

| Treatment
All patients received the Chinese Acute Lymphoblastic Leukemia Cooperative Group 2008 (CALLG2008) protocol, which was recommend by Chinese expert panel consensus on diagnosis and treatment of adult acute lymphoblastic leukemia. 13 The schedules of CALLG2008 comprised of prophase therapy, induction course, consolidation therapy, maintenance therapy, and central nervous system prophylaxis. Details of the CALLG2008 protocol were listed in Supplementary Table S1. During the treatment, if patients were eligible for transplantation, they underwent human leukocyte antigen (HLA)-matched or haploidentical allogeneic hematopoietic stem cell transplant (allo-HSCT) after 3 to 5 courses of consolidation therapy; if not, patients continued to receive consolidation and maintenance therapy.

| Treatment response evaluation and definition
Morphological analysis of BM cells was performed to evaluate the treatment response on the 28th (±7) day of induction therapy, during which, CR patients were classified as the group with CR within 4 weeks. For patients not achieved CR within 4 weeks, they were given salvage therapy. After induction therapy and salvage therapy, all CR patients were classified as the group with total CR. CR was defined as no circulating blasts or extramedullary disease, trilineage hematopoiesis (TLH) and <5% blasts, absolute neutrophil count (ANC) >1.0 × 10 9 /mL, peripheral blood PLT >100 × 10 9 /mL, and no recurrence for 4 weeks. Relapse was defined as reappearance of blasts in the blood or bone marrow (>5%) or in any extramedullary site after a CR. DFS was defined as the duration from the date of CR to the date of relapse or death in CR status. OS was defined as the duration from the date of diagnosis to the date of death or last follow-up. Outcome was updated on March 31, 2020.

| Statistical analysis
SPSS 24.0 statistical software (IBM) was used for statistical analysis, and GraphPad Prism 8.01 software (GraphPad Software Inc) was used for graphs plotting. Quantitative data were described as mean with standard deviation (SD), or median with interquartile range (IQR). Qualitative data were described as number and percentage (No. (%)). Difference of lncRNA TUG1 expression between Ph − ALL patients and health donors was determined by Wilcoxon rank sum test. According to the quantiles of lncRNA TUG1 expression in all patients, patients were divided into four tiers: tier 1 (whose lncRNA TUG1 expression was ranked in 0%~25% among all patients, 0.765 ≤ Tier 1 < 1.990), tier 2 (whose lncRNA TUG1 expression was ranked in 25%~50% among all patients, 1.990 ≤ Tier 2 < 2.801), tier 3 (whose lncRNA TUG1 expression was ranked in 50%~75% among all patients, 2.801 ≤ Tier 3 < 4.261), and tier 4 (whose lncRNA

| Clinical characteristics of Ph − ALL patients
In Ph − ALL patients, the mean age was 31.7 ± 9.6 years ( Table 1) Table 1.

| Correlation of lncRNA TUG1 with accumulating survival in Ph − ALL patients
We conducted the further analysis to detect the correlation of lncRNA TUG1 with survival profiles in Ph − ALL patients, and found that lncRNA TUG1 tier was negatively associated with accumulating DFS (P < .001) ( Figure 3A) and accumulating OS (P = .014) in Ph − ALL patients ( Figure 3B).

| Subgroup analysis: correlation of lncRNA TUG1 with accumulating survival
In Ph − T-ALL patients, lncRNA TUG1 expression was not corre-

TA B L E 3 Analysis of factors related to DFS
correlated with the presence of CNSL, increased WBC level, and bone marrow blasts. (3) Higher lncRNA TUG1 expression was negatively associated with CR within 4 weeks, total CR, and allo-HSCT.
(4) High lncRNA TUG1 expression was an independent predictive factor for worse DFS and OS.
LncRNAs are a class of non-coding RNAs; furthermore, recent increasing evidence has indicated that lncRNAs are emerging as with normal osteoblastic cell line, and its knockdown suppresses glucose consumption, lactate production, and cell viability of osteosarcoma cells. 16 In another clinical study, lncRNA TUG1 is correlated with worse TNM staging in patients with cervical cancer. 17 As for in hematologic malignancies, lncRNA TUG1 promotes proliferation but inhibits apoptosis in MM via suppressing miR-29b-3p. 18 Nevertheless, the role of lncRNA TUG1 in the pathological progression of Ph − ALL has not been uncovered. Thus, we conducted the present study to investigate the level of lncRNA TUG1 between Ph − ALL patients and healthy donors, and further explore its correlation with disease condition and prognosis in patients with Ph − ALL.
In our present study, we determined the lncRNA TUG1 expres- allo-HSCT is considered to be part of post-remission consolidative therapy, and therefore, for patients with higher lncRNA TUG1 expression, the lower rate of CR was achieved, and the reduced application of allo-HSCT was conducted.
Our study existed some limitations as follows: (a) Firstly, considering that our study was a retrospective single-center study with a relatively small sample size, selection bias and relatively low to bias in the current study. (d) The present study did not include Ph + ALL patients, and therefore, further studies included these patients were needed for validating the clinical role of lncRNA TUG1.
(e) Considering our study was a retrospective study with difference sample size in case group and controls, further prospective studies included the same sample size in case group and control were needed for validation. (f) The patients included were all adult Ph − ALL; however, considering ALL was common among children, and therefore, further studies needed to be conducted for results validation in samples of children.
In conclusion, lncRNA TUG1 is upregulated and correlates with poor disease condition, treatment response, and survival profiles in Ph − ALL patients, implying the potential of lncRNA TUG1 as a useful biomarker in ALL management.