Abstract
Background
It has been demonstrated that aberrant expression of serum microRNAs is potential markers for the prognostic prediction of acute myeloid leukemia (AML). However, the clinical significance of serum miR‐22 remained uncovered. In this study, we aimed to explore the potential prognostic value of serum miR‐22 for AML.
Methods
Blood samples were collected from 124 patients with AML and 60 healthy individuals. Serum miR‐22 level was detected by quantitative reverse transcription‐polymerase chain reaction (qRT‐PCR), and its potential clinical value was investigated.
Results
Our results showed that serum miR‐22 expression was significantly downregulated in AML subjects compared to healthy controls. Serum miR‐22 levels were lowest in AML patients with M4/M5 subtypes, and low serum miR‐22 expression occurred more frequently in AML patients with higher white blood cell counts or poor cytogenetic risk. Receiver operating characteristic (ROC) analysis revealed that serum miR‐22 well differentiated AML cases from healthy controls. In addition, serum miR‐22 downregulation was closely associated with worse clinical features and shorter survival. Low serum miR‐22 expression was confirmed to be an independent predictor for overall survival and relapse‐free survival in AML patients. Moreover, the expression level of serum miR‐22 was dramatically increased following treatment. In addition, serum miR‐22 levels were significantly higher in AML patients achieving complete remission (CR) than those without CR.
Conclusion
Collectively, serum miR‐22 might serve as a novel and promising prognostic biomarker for AML.
Keywords: acute myeloid leukemia, complete remission, prognosis, serum miR‐22
1. INTRODUCTION
Acute myeloid leukemia (AML) is an aggressive hematopoietic stem cell malignancy that is characterized by the clonal proliferation of myeloid precursors. 1 Abnormal accumulation of leukemic blasts in the bone marrow, blood, and other tissues results in significant reductions of normal blood cells. 2 AML patients can be divided into 3 risk‐based categories according to cytogenetic information: poor, intermediate, and favorable. 3 Though treatment of this disease has made huge progress over the past years, the 5‐year overall survival rate of AML patients remains unfavorable, ranging from 11%‐55%. 4 , 5 Therefore, identifying novel and reliable prognostic biomarkers for risk stratification of AML patients is critical for selecting the optimal therapeutic strategies and improving the clinical outcome. 6 , 7
MicroRNAs (miRNAs) are small (19‐25 nucleotide), noncoding RNAs that control the gene expression at the post‐transcriptional level, leading to target mRNAs degradation or translational inhibition. 8 , 9 Increasing evidence has demonstrated that miRNA dysregulation is associated with the initiation and progression of cancer. 10 In addition, miRNAs are highly stable in the biofluids such as serum, plasma, and saliva. 11 These features of miRNAs enable them to become the promising biomarkers for the diagnosis and prognosis of cancer.
MiR‐22, located in chromosome 17p13, has been found to act as a tumor suppressor in AML. For instance, Jiang et al showed miR‐22 expression was significantly decreased in AML patients, and miR‐22 upregulation markedly inhibited leukemic cell viability in vitro and suppressed leukemia progression in vivo. 12 Similarly, miR‐22 was demonstrated to play a tumor suppressive role in AML development. 13 However, the clinical significance of serum miR‐22 in AML remains uncertain. The aim of our study was to elucidate the potential prognostic value of serum miR‐22 in AML.
2. MATERIALS AND METHODS
2.1. Ethics statement
The current study was approved by the Ethics Committee of Panyu Central Hospital, and written informed consent was obtained from all participants. All specimens were handled and made anonymous according to the ethical and legal standards.
2.2. Patients and sample collection
A total of 124 patients with newly diagnosed AML and 60 healthy individuals as controls were enrolled in this study. The diagnosis and classification of AML patients was made according to French‐America‐British (FAB) and World Health Organization criteria. The patients’ characteristics are summarized and presented in Table 1. The AML patients were stratified into three risk categories based on the cytogenetics (poor, intermediate, and favorable). The detailed information about the cytogenetic information of AML patients was shown in Supplementary Table S1. Complete remission (CR) was defined by <5% blast cells in the bone marrow and normalization of the peripheral blood counts at four weeks after starting induction therapy, as well as no residual evidence of extramedullary disease. Overall survival (OS) was defined as the time from the diagnosis to death due to any cause or the last follow‐up. Relapse‐free survival (RFS) was defined as the time from the achievement of CR to relapse or the last follow‐up. No patient lost to follow‐up, and the median follow‐up time was 23 months (range 5.4‐60 months). All the AML patients received similar chemotherapy treatment. The induction chemotherapy was administered to the AML patients based on their clinical status, and the details of the therapeutic protocols were referred to NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines) for AML. The expression level of serum miR‐22 was detected in AML patients both before and after the treatment.
Table 1.
The association between serum miR‐22 and the clinicopathological parameters of AML
| Characteristics | Number | Serum miR‐22 | P | |
|---|---|---|---|---|
| Low expression | High expression | |||
| Age (years) | ||||
| <60 | 83 | 35 (42.2%) | 48 (57.8%) | .227 |
| ≥60 | 41 | 22 (53.7%) | 19 (46.3%) | |
| Gender | ||||
| Male | 75 | 33 (44.0%) | 42 (56.0%) | .587 |
| Female | 49 | 24 (48.9%) | 25 (51.1%) | |
| BM Blasts (%) | ||||
| <50 | 56 | 22 (39.3%) | 34 (60.7%) | .176 |
| ≥50 | 68 | 35 (51.5%) | 33 (48.5%) | |
| WBC counts (×109/L) | ||||
| <10 | 54 | 19 (35.2%) | 35 (64.8%) | .034 |
| ≥10 | 70 | 38 (54.3%) | 32 (45.7%) | |
| FAB subtype | ||||
| M0 | 10 | 4 (40.0%) | 6 (60.0%) | .202 |
| M1/M2 | 72 | 29 (40.3%) | 43 (59.7%) | |
| M4/M5 | 42 | 24 (57.1%) | 18 (42.9%) | |
| Platelet counts (× 109/L) | ||||
| <50 | 63 | 24 (38.1%) | 39 (61.9%) | .074 |
| ≥50 | 61 | 33 (54.1%) | 28 (45.9%) | |
| Cytogenetics | ||||
| Favorable | 49 | 15 (30.6%) | 34 (69.4%) | .008 |
| Intermediate | 58 | 30 (51.7%) | 28 (48.3%) | |
| Poor | 17 | 12 (70.6%) | 5 (29.4%) | |
| Complete Remission | ||||
| Yes | 55 | 21 (38.2%) | 34 (61.8%) | .120 |
| No | 69 | 36 (52.2%) | 33 (47.8%) | |
For the serum sample collection, fasting peripheral vein blood was obtained from 124 AML patients and 60 controls. The samples were centrifugated at 2500 g for 10 minutes, followed by centrifugation at 16 000 g for 10 minutes. The serum samples were stored at −80°C until further processing.
2.3. Total RNA extraction and quantitative RT‐PCR
Total RNA was isolated from serum with the miRNeasy Serum/Plasma kit (Qiagen) according to the manufacturer's instructions. The RNA concentration was measured with a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific). Cel‐miR‐39 was used as the spiked‐in control. 14 , 15 The cDNA was reverse‐transcribed from total RNA using the TaqMan MicroRNA Reverse Transcription Kit (Applied Biosystems, Foster City, CA, USA). The amplification of cDNAs was performed on the 7500 Real‐Time PCR system (Applied Biosystems). The relative serum miR‐22 expression levels were normalized to cel‐miR‐39 and calculated by the 2–ΔΔ C t method.
2.4. Statistical analysis
Student's t test or one‐way ANOVA was conducted to compare the difference in serum miR‐22 concentrations between two or three groups, respectively. The chi‐square test was used to analyze possible associations between serum miR‐22 expression and clinical variables. Receiver operating characteristic (ROC) curve was performed, and area under the ROC curve (AUC) was used to assess the discriminative power of serum miR‐22. The Kaplan‐Meier method was performed to construct the survival curves of AML patients, and the difference in survival curves was compared by log‐rank test. Multivariate Cox proportional hazards regression analysis was used to identify the independent prognostic factors. All analyses were performed with the GraphPad Prism 5 for Windows (GraphPad software, San Diego, CA, USA) and SPSS 16.0 software (SPSS Inc). P value <.05 was considered statistically significant.
3. RESULTS
3.1. Serum miR‐22 was downregulated in AML patients
The serum miR‐22 expression levels were measured in 124 AML subjects and 60 normal controls by qRT‐PCR. As shown in Figure 1A, serum miR‐22 levels were significantly lower in AML patients than in the healthy controls (P < .0001). In addition, high serum miR‐22 expression occurred more frequently in AML patients with M1/M2 subtypes compared with those with M4/M5 subtypes (P = .016). Moreover, AML patients with white blood cells (WBC) <10 × 109/L had higher serum miR‐22 expression than those with WBC ≥ 10×109/L (P = .026, Figure 1C). Furthermore, serum miR‐22 levels were significantly reduced in poor cytogenetic risk group compared to intermediate cytogenetic risk subgroup (P = .024) and favorable risk cytogenetic group (P = .0002), and significant lower serum miR‐22 expression was observed in intermediate cytogenetic risk group than in favorable cytogenetic risk group (P = .035, Figure 1D). ROC analysis showed that serum miR‐22 well differentiated AML subjects from normal controls with AUC value of 0.830; the specificity and the sensitivity were 78.3% and 76.6%, respectively (Figure 2).
Figure 1.
Reduced serum miR‐22 in AML patients. A, Serum miR‐22 levels were significantly reduced in AML patients compared to controls. B, Serum miR‐22 levels were significantly downregulated in M4/M5 subtypes compared to M1/M2 subtypes. C, Serum miR‐22 levels were significantly downregulated in AML patients with higher WBC counts. D, Serum miR‐22 levels were gradually downregulated in AML patients with favorable risk cytogenetic group, intermediate risk cytogenetic group, and unfavorable risk cytogenetic group
Figure 2.
Serum miR‐22 exhibited good performance for discriminating AML from normal controls
3.2. Correlation between serum miR‐22 expression and clinical characteristics
All AML subjects were classified into high serum miR‐22 expression group (n = 67) and low serum miR‐22 expression group (n = 57) using the mean value of serum miR‐22 level. As shown in Table 1, significant associations were found between low serum miR‐22 expression and WBC counts (P = .034) and cytogenetics (P = .008). However, no significant correlation was found between serum miR‐22 and age (P = .227), gender (P = .587), bone marrow blasts (P = .176), FAB subtype (P = .202), platelet counts (P = .074) and CR (P = .120) (Table 2).
Table 2.
Multivariate analysis of the independent prognostic factors for AML
| Characteristics | Risk Ratio | 95% CI | P |
|---|---|---|---|
| OS (all AML, N = 124) | |||
| WBC counts (× 109/L) | 2.46 | 1.40‐3.68 | .020 |
| Cytogenetics | 2.82 | 1.51‐4.32 | .013 |
| Serum miR‐22 | 2.73 | 1.44‐4.17 | .015 |
| RFS (all AML, N = 124) | |||
| WBC counts (× 109/L) | 2.61 | 1.47‐4.05 | .018 |
| Cytogenetics | 3.02 | 1.65‐4.58 | .011 |
| Serum miR‐22 | 3.23 | 1.84‐4.97 | .007 |
3.3. Low serum miR‐22 expression was associated with worse survival in AML
Kaplan‐Meier method and log‐rank test were used to analyze the association between serum miR‐22 expression and survival. Both the OS (P = .0211, Figure 3A) and RFS (P = .0013, Figure 3B) rates were markedly lower in AML cases in low serum miR‐22 expression group (n = 57) than in high serum miR‐22 expression group (n = 67).
Figure 3.
The associations between serum miR‐22 level and survival of AML. A, The AML patients in the low serum miR‐22 group had worse OS than those in the high serum miR‐22 group. (B) The AML patients in the low serum miR‐22 group had worse RFS than those in the high serum miR‐22 group
3.4. Serum miR‐22 was an independent prognostic factor for both OS and RFS
Multivariable analysis showed that WBC counts (RR = 2.46; 95% CI = 1.40‐3.68, P = .020), cytogenetics (RR = 2.82; 95% CI = 1.51‐4.32, P = .013), and serum miR‐22 expression (RR = 2.73; 95% CI = 1.44‐4.17, P = .015) were significant prognostic factors for poorer OS. Similarly, WBC counts (RR = 2.61; 95% CI = 1.47‐4.05, P = .018), cytogenetics (RR = 3.02; 95% CI = 1.65‐4.58, P = .011), and serum miR‐22 expression (RR = 3.23; 95% CI = 1.84‐4.97, P = .007) were also demonstrated to be independent prognostic markers for RFS in AML patients.
3.5. The changes in serum miR‐22 expression in AML patients following treatment
On the 28th day of the first chemotherapy cycle, blood samples were obtained from all the AML patients and serum miR‐22 levels were detected for monitoring treatment response. Of 124 AML patients, 55 patients achieved a CR while 69 cases failed to achieve a CR. We found serum miR‐22 levels were remarkably increased both in the subgroup of patients achieving CR (P = .0002, Figure 4A) and the subgroup of patients failing to achieve a CR (P = .0169, Figure 4B). Also, AML patients achieving CR exhibited significant higher serum miR‐22 expression than those without CR achievement for both pre‐treated and post‐treated blood samples (P < .0001, Figure 4C‐4D). The data demonstrated serum miR‐22 might be used to monitor the treatment response of AML patients.
Figure 4.
Serum miR‐22 was sensitive for monitoring therapeutic responses. A, For the AML cases achieving CR, serum miR‐22 levels were markedly elevated in the post‐treated blood samples compared to the pre‐treated blood samples. B, For the AML cases not achieving CR, serum miR‐22 levels were increased in the post‐treated blood samples compared to the pre‐treated blood samples. C‐D, For both pre‐treated and post‐treated blood samples, serum miR‐22 levels were significantly higher in AML achieving CR than in those not achieving CR
4. DISCUSSION
In the present study, we found that serum miR‐22 was significantly lower in AML patients compared to healthy individuals. ROC analysis showed that serum miR‐22 exhibited good performance for discriminating AML patients from healthy controls. The possible reason accounting for the reduced serum miR‐22 level in AML patients might be that the cancer cells synthesized and secreted much less miR‐22 into the peripheral blood. The AML patients with low serum miR‐22 level might receive more aggressive therapeutic strategies. In addition, we found that low serum miR‐22 expression was significantly correlated with aggressive clinicopathological variables and shorter survival. The multivariate analysis demonstrated that low serum miR‐22 expression was an independent risk factor for both OS and RFS, suggesting that the serum miR‐22 level might be closely correlated with AML progression and serve as a promising prognostic indicator for AML. Interestingly, serum miR‐22 levels were significantly increased in AML patients receiving treatments, especially those achieving CR. These findings suggest that serum miR‐22 level is sensitive for monitoring the therapeutic responses. Collectively, low serum miR‐22 might act as a promising biomarker for monitoring the initiation and development of AML. Therefore, the results of our study were consistent with previous studies that miR‐22 played a tumor suppressive role in AML. 12 , 13
Except for AML, the tumor suppressive role of miR‐22 has also been reported in many other types of cancers. For instance, Wan et al found that miR‐22 expression was markedly reduced in epithelial ovarian cancer tissues compared to normal tissues, and downregulation of miR‐22 was closely associated with worse clinical parameters. 16 In breast cancer (BC), miR‐22 expression was significantly downregulated in BC tissues and cells, and ectopic expression of miR‐22 significantly inhibited the tumorigenesis and enhanced radiosensitivity of BC cells. 17 , 18 Similarly, miR‐22 expression was decreased in gastric cancer (GC) tissues, and its downregulation was associated with aggressive phenotype of GC and reduced survival. Restoration of miR‐22 remarkably suppressed cancer cell growth, migration, and invasion in vitro. 19 , 20 In hepatocellular carcinoma (HCC), reduced miR‐22 expression occurred more frequently in HCC tissues and cell lines. In vitro and in vivo analysis showed that miR‐22 overexpression significantly inhibited carcinogenesis of HCC by regulating CD147. 21 , 22 In bladder cancer, overexpression of miR‐22 markedly reduced MAPK1 and Snail expression, and attenuated cancer cell proliferation, migration, and invasion. 23
Interestingly, miR‐22 might also act as an oncomiR in cancer progression. Serum miR‐22 was overexpressed in patients with papillary thyroid cancer, and its upregulation was strongly associated with metastasis. 24 The controversial findings reflect that the role of miR‐22 in tumorigenesis might be cancer‐type specific, and the hidden molecular mechanisms about the role of miR‐22 in tumorigenesis need to be further explored.
One possible limitation of our study was the relatively small sample size. Larger independent cohorts are warranted to validate the clinical significance of serum miR‐22 in AML. In addition, serum miR‐22 might also be deregulated in other malignancies or other human diseases. Therefore, serum miR‐22 should be combined with other known biomarkers and the clinical information to accurately predicts the prognosis of AML. Moreover, the molecular mechanisms for the role of miR‐22 in AML progression need deeper investigation.
In conclusion, our study has demonstrated that the serum miR‐22 expression is markedly downregulated in AML. In addition, downregulation of serum miR‐22 is significantly associated with aggressive clinical variables and poor prognosis of AML. Therefore, serum miR‐22 might serve as a promising prognostic biomarker for AML.
Supporting information
Table S1
Qu H, Zheng G, Cheng S, et al. Serum miR‐22 is a novel prognostic marker for acute myeloid leukemia. J Clin Lab Anal. 2020;34:e23370 10.1002/jcla.23370
REFERENCES
- 1. Estey EH. Acute myeloid leukemia: 2013 update on risk‐stratification and management. Am J Hematol. 2013;88(4):318‐327. [DOI] [PubMed] [Google Scholar]
- 2. Ferrara F, Schiffer CA. Acute myeloid leukaemia in adults. Lancet. 2013;381(9865):484‐495. [DOI] [PubMed] [Google Scholar]
- 3. Gregory TK, Wald D, Chen Y, Vermaat JM, Xiong Y, Tse W. Molecular prognostic markers for adult acute myeloid leukemia with normal cytogenetics. J Hematol Oncol. 2009;2:23. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Estey EH. Treatment of acute myeloid leukemia. Haematologica. 2009;94(6):10‐16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Tallman MS, Gilliland DG, Rowe JM. Drug therapy for acute myeloid leukemia. Blood. 2005;106(4):1154‐1163. [DOI] [PubMed] [Google Scholar]
- 6. Lin SY, Miao YR, Hu FF, et al. A 6‐Membrane protein gene score for prognostic prediction of cytogenetically normal acute myeloid leukemia in multiple cohorts. J Cancer. 2020;11(1):251‐259. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Zeng M, Ding S, Zhang H, Huang Q, Ren Y, Guo P. Predictive value of integrin α7 for acute myeloid leukemia risk and its correlation with prognosis in acute myeloid leukemia patients. J Clin Lab Anal. 2019;e23151. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Chen H, Liu X, Jin Z, et al. A three miRNAs signature for predicting the transformation of oral leukoplakia to oral squamous cell carcinoma. Am J Cancer Res. 2018;8(8):1403‐1413. [PMC free article] [PubMed] [Google Scholar]
- 9. Garofalo M, Croce CM. Role of microRNAs in maintaining cancer stem cells. Adv Drug Deliv Rev. 2015;81:53‐61. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Feng B, Zhang K, Wang R, Chen L. Non‐small‐cell lung cancer and miRNAs: novel biomarkers and promising tools for treatment. Clin Sci (Lond). 2015;128(10):619‐634. [DOI] [PubMed] [Google Scholar]
- 11. Mitchell PS, Parkin RK, Kroh EM, et al. Circulating microRNAs as stable blood‐based markers for cancer detection. Proc Natl Acad Sci U S A. 2008;105(30):10513‐10518. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Jiang X, Hu C, Arnovitz S, et al. MiR‐22 has a potent anti‐tumour role with therapeutic potential in acute myeloid leukaemia. Nat Commun. 2016;7:11452. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Shen C, Chen MT, Zhang XH, et al. The PU.1‐modulated microRNA‐22 is a regulator of monocyte/macrophage differentiation and acute myeloid leukemia. PLoS Genet. 2016;12(9):e1006259. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Kroh EM, Parkin RK, Mitchell PS, Tewari M. Analysis of circulating microRNA biomarkers in plasma and serum using quantitative reverse transcription‐PCR (qRT‐PCR). Methods. 2010;50(4):298‐301. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Marzi MJ, Montani F, Carletti RM, et al. Optimization and standardization of circulating microRNA detection for clinical application: the miR‐test case. Clin Chem. 2016;62(5):743‐754. [DOI] [PubMed] [Google Scholar]
- 16. Wan WN, Zhang YQ, Wang XM, et al. Down‐regulated miR‐22 as predictive biomarkers for prognosis of epithelial ovarian cancer. Diagn Pathol. 2014;9:178. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Zhang X, Li Y, Wang D, Wei X. MiR‐22 suppresses tumorigenesis and improves radiosensitivity of breast cancer cells by targeting Sirt1. Biol Res. 2017;50(1):27. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Song YK, Wang Y, Wen YY, Zhao P, Bian ZJ. MicroRNA‐22 suppresses breast cancer cell growth and increases paclitaxel sensitivity by targeting NRAS. Technol Cancer Res Treat. 2018;17:1533033818809997. [DOI] [PMC free article] [PubMed] [Google Scholar] [Retracted]
- 19. Zuo QF, Cao LY, Yu T, et al. MicroRNA‐22 inhibits tumor growth and metastasis in gastric cancer by directly targeting MMP14 and Snail. Cell Death Dis. 2015;6:e2000. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Wang W, Li F, Zhang Y, Tu Y, Yang Q, Gao X. Reduced expression of miR‐22 in gastric cancer is related to clinicopathologic characteristics or patient prognosis. Diagn Pathol. 2013;8:102. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Chen M, Hu W, Xiong CL, et al. MiR‐22 targets YWHAZ to inhibit metastasis of hepatocellular carcinoma and its down‐regulation predicts a poor survival. Oncotarget. 2016;7(49):80751‐80764. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Luo LJ, Zhang LP, Duan CY, et al. The inhibition role of miR‐22 in hepatocellular carcinoma cell migration and invasion via targeting CD147. Cancer Cell Int. 2017;17:17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Xu M, Li J, Wang X, et al. MiR‐22 suppresses epithelial–mesenchymal transition in bladder cancer by inhibiting Snail and MAPK1/Slug/vimentin feedback loop. Cell Death Dis. 2018;9(2):209. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Wang D, Guo C, Kong T, Mi G, Li J, Sun Y. Serum miR‐22 may be a biomarker for papillary thyroid cancer. Oncol Lett. 2019;17(3):3355‐3361. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Table S1