The purpose of the paper was to optimize a direct amplification protocol for quantification of microRNA (miR, miRNA)-21 in serum, investigate the stability of miR-21 during freeze-thaw cycling of serum, and to investigate the sensitivity and specificity of this assay for the diagnosis of breast cancer. Levels of miR-21 in formalin-fixed paraffin-embedded (FFPE) tumor and normal adjacent specimens were also compared.
Direct serum assay for microRNA-21 concentrations in early and advanced breast cancer.
Asaga S, Kuo C, Nguyen T, Terpenning M, Giuliano AE, Hoon DS
Purpose
Conclusion
miR-21 was only amplified directly in serum when the serum was treated with Tween-20 in the absence of Proteinase K. miR-21 levels were strongly correlated specimens in which RNA was extracted before analysis and serum specimens treated with 2.5% Tween-20 but not subjected to RNA extraction but no correlation was observed when treated with 1% Tween-20. The direct amplification assay required a minimum input of 0.625 µL of serum and but was not affected by use of DNAse and up to four freeze-thaw cycles of serum. Normalized miR-21 levels were significantly higher in breast cancer patients than controls regardless of stage and RNA extraction, but when amplified directly from serum, miR-21 levels were higher in the serum of stage IV cancer patients than stage I, II, or III patients. Importantly, the miR-21 direct amplification assay had a specificity of 75% and a sensitivity of 67% for diagnosis of locoregional breast cancer and a specificity of 86% and sensitivity of 59% for diagnosis of stage IV breast cancer versus stage I, II, or III. Levels of miR-21 were also found to be higher in FFPE breast tumor than in normal adjacent tissue.
Studies (1)
Purpose
The purpose of the study was to optimize a direct amplification protocol for quantification of miR-21 in serum, investigate the stability of miR-21 during freeze-thaw cycling of serum, and to investigate the sensitivity and specificity of this assay for the diagnosis of breast cancer. Levels of miR-21 in FFPE tumor and normal adjacent specimens were also compared. Blood was collected from 102 patients with breast cancer and 20 healthy women in red tiger-top gel separator tubes. After clotting for 2-5 h, serum was obtained by centrifugation (details not included) and use of a 13-mm serum filter and then stored frozen at -80°C. RNA was extracted from serum using TRI reagent BD and reverse transcribed using miR-21 specific primers. In 12 cases, RNA extraction was not performed and instead serum was mixed with: 1) nothing, 2) Proteinase K, 3) 1% Tween-20 and Proteinase K, 4) 2.5% Tween-20 and Proteinase K, 5) 1% Tween-20, or 6) 2.5% Tween-20 before reverse transcription and then centrifugation at 9000 x g for 5 min before real-time PCR amplification. The effect of DNAse treatment was tested on eight serum specimens. RNA was extracted from fresh 10 µm sections of 14 matched archival FFPE breast tumor and normal adjacent specimens using a modification of the RNAWiz Isolation Kit and quantified using Quant-iT RiboGreen RNA Assay Kit. miR-21was quantified by real-time RT-PCR and was normalized to RNU6B (tissues) or miR-16 (serum). The stability study was conducted using four serum specimens that were subjected to up to four cycles of freezing at -80°C and thawing at 23°C.
Summary of Findings
miR-21 was only amplified directly in serum when the serum was treated with Tween-20 in the absence of Proteinase K. miR-21 levels -21 levels were strongly correlated specimens in which RNA was extracted before analysis and when directly amplified from serum after treatment with 2.5% Tween-20 (r=0.796) but no correlation was observed when amplified directly from serum treated with 1% Tween-20(r=-0.064). Using direct amplification of serum treated with 2.5% Tween-20, there was no difference in miR-16 normalized miR-21 levels when 1.25 µL or 0.625 µL of serum were used but miR-16 was not detected when 0.3125 µL of serum was used. The mean standard deviation (SD) of the quantification cycle (Cq) values in this assay ranged from 0.42-0.59. The authors report that DNAse treatment had little effect on the Cq values (mean of 0.75 cycles). The normalized levels of miR-21 were not significantly affected by up to four freeze-thaw cycles. No difference in miR-16 Cq values were found among serum from healthy controls and different stages of breast cancer regardless of if RNA was extracted prior to analysis or amplified directly from serum. In contrast, normalized miR-21 levels were significantly higher in serum from breast cancer patients than controls, regardless of stage and RNA extraction (P<0.001). While no differences in serum miR-21 levels were found among the different cancer stages when RNA was extracted from serum, significantly higher normalized miR-21 levels were found in the serum of stage IV cancer patients than stage I (P=0.035 and P<0.001), II (P=0.0034 and P<0.001), and III (P=0.0014 and P=0.0006) patients when amplified directly from serum in two different cohorts. Importantly, the miR-21 direct amplification assay had a specificity of 75% and a sensitivity of 67% for diagnosis of locoregional breast cancer and a specificity of 86% and sensitivity of 59% for diagnosis of stage IV breast cancer versus stage I, II, or III. Levels of miR-21 were found to be higher in FFPE breast tumor than in normal adjacent tissue.
Biospecimens
FluidSerumTissueBreastPreservation Types
FrozenFormalinDiagnoses
NeoplasticCarcinomaNeoplasticNormal AdjacentNormalPlatforms
RNAReal-time qRT-PCRPre-analytical Factors
Classification Pre-Analytical Factor Value(s) Preaquisition Diagnosis/ patient condition Breast CancerHealthyPreaquisition Prognostic factor Stage I breast cancerStage II breast cancerStage III breast cancerStage IV breast cancerBiospecimen Aliquots and Components Aliquot size/volume 0.3125 µL0.625 µL1.25 µLStorage Freeze/thaw cycling 1 cycle2 cycles3 cycles4 cyclesAnalyte Extraction and Purification Analyte isolation method Addition of Tween-20TRI reagent BDAnalyte Extraction and Purification Nucleic acid digestion DNAse treatedNot treatedAnalyte Extraction and Purification Protein digestion Proteinase KAnalyte Extraction and Purification Protein solubilization 1% Tween-202.5% Tween-20NoneReal-time qRT-PCR Specific Targeted nucleic acid RNU6BmiR-16miR-21