Department of Pathology, University Medical Center Utrecht, Utrecht University, Heidelberglaan 100, 3584 CX, Utrecht, The Netherlands.
Department of Medical Oncology, University Medical Center Utrecht, Utrecht University, Utrecht, The Netherlands.
Cell Oncol (Dordr). 2021 Dec;44(6):1339-1349. doi: 10.1007/s13402-021-00641-w. Epub 2021 Oct 16.
Investigation of nipple aspirate fluid (NAF)-based microRNAs (miRNAs) as a potential screening tool for women at increased risk of developing breast cancer is the scope of our research. While aiming to identify discriminating NAF-miRNAs between women with different mammographic densities, we were confronted with an unexpected confounder: NAF sample appearance. Here we report and alert for the impact of NAF color and cloudiness on miRNA assessment.
Seven classes of NAF colors coupled with cloudiness appearance were established. Using 173 NAF samples from 154 healthy women (19 samples were bilaterally collected), the expression of 14 target and 2 candidate endogenous control (EC) miRNAs was investigated using Taqman Advanced miRNA assays to identify significant differential expression patterns between color-cloudiness classes. Inter- and intra-individual variation of miRNA expression was analyzed using the coefficient of variation (CV).
We found that between the seven NAF classes, fold change miRNA expression differences ranged between 2.4 and 19.6 depending on the interrogated miRNA. Clear NAF samples exhibited higher miRNA expression levels compared to cloudy NAF samples with fold change differences ranging between 1.1 and 6.2. Inter-individual and intra-individual miRNA expression was fairly stable (CV < 15 %), but nevertheless impacted by NAF sample appearance. Within NAF classes, inter-individual variation was largest for green samples (CV 6-15 %) and smallest for bloody samples (CV 2-6 %).
Our data indicate that NAF color and cloudiness influence miRNA expression and should, therefore, be systematically registered using an objective color classification system. Given that sample appearance is an inherent feature of NAF, these variables should be statistically controlled for in multivariate data analyses. This cautionary note and recommendations could be of value beyond the field of NAF-miRNAs, given that variability in sample color and cloudiness is likewise observed in liquid biopsies such as urine, cerebrospinal fluid and sputum, and could thereby influence the levels of miRNAs and other biomarkers.
研究乳头吸出液（NAF）中的 microRNAs（miRNAs）作为一种有发展为乳腺癌风险的女性潜在筛查工具，这是我们研究的范围。在旨在确定不同乳房密度女性之间有区别的 NAF-miRNAs 时，我们遇到了一个意想不到的混杂因素：NAF 样本的外观。在这里，我们报告并提醒 NAF 颜色和混浊度对 miRNA 评估的影响。
建立了七种 NAF 颜色类别和混浊外观。使用来自 154 位健康女性的 173 个 NAF 样本（19 个样本双侧采集），使用 Taqman 高级 miRNA 检测试剂盒检测 14 个靶标和 2 个候选内源性对照（EC）miRNAs 的表达，以确定颜色-混浊度类别之间的显著差异表达模式。使用变异系数（CV）分析 miRNA 表达的个体内和个体间变异性。
我们发现，在七种 NAF 类别之间，根据所检测的 miRNA，miRNA 表达差异的倍数变化范围在 2.4 到 19.6 之间。与混浊的 NAF 样本相比，清澈的 NAF 样本的 miRNA 表达水平更高，倍数变化差异在 1.1 到 6.2 之间。个体内和个体间的 miRNA 表达相当稳定（CV<15%），但仍受 NAF 样本外观的影响。在 NAF 类别内，绿色样本的个体间变异性最大（CV 6-15%），而血性样本的个体间变异性最小（CV 2-6%）。
我们的数据表明，NAF 的颜色和混浊度会影响 miRNA 的表达，因此，应该使用客观的颜色分类系统系统地记录这些信息。鉴于样本外观是 NAF 的固有特征，在多元数据分析中应进行统计控制这些变量。鉴于尿液、脑脊液和痰液等液体活检中同样观察到样本颜色和混浊度的变化，因此这种警示和建议可能具有超出 NAF-miRNAs 领域的价值，并且可能影响 miRNA 和其他生物标志物的水平。
