Background: The prediction of the regrowth potential of pituitary adenomas after surgery is challenging. The genome-wide DNA methylation profiling of pituitary adenomas may separate adenomas into distinct methylation classes corresponding to histology-based subtypes. Specific genes and differentially methylated probes involving regrowth have been proposed, but no study has linked this epigenetic variance with regrowth potential and the clinical heterogeneity of nonfunctioning pituitary adenomas. This study aimed to investigate whether DNA methylation profiling can be useful as a clinical prognostic marker. Methods: A DNA methylation analysis by Illumina’s MethylationEPIC array was performed on 54 pituitary macroadenomas from patients who underwent transsphenoidal surgery during 2007–2017. Twelve patients were excluded due to an incomplete postoperative follow-up, degenerated biobank-stored tissue, or low DNA methylation quality. For the quantitative measurement of the tumor regrowth rate, we conducted a 3D volumetric analysis of tumor remnant volume via annual magnetic resonance imaging. A linear mixed effects model was used to examine whether different DNA methylation clusters had different regrowth patterns. Results: The DNA methylation profiling of 42 tissue samples showed robust DNA methylation clusters, comparable with previous findings. The subgroup of 33 nonfunctioning pituitary adenomas of an SF1-lineage showed five subclusters with an approximately unbiased score of 86%. There were no overall statistically significant differences when comparing hazard ratios for regrowth of 100%, 50%, or 0%. Despite this, plots of correlated survival estimates suggested higher regrowth rates for some clusters. The mixed effects model of accumulated regrowth similarly showed tendencies toward an association between specific DNA methylation clusters and regrowth potential. Conclusion: The DNA methylation profiling of nonfunctioning pituitary adenomas may potentially identify adenomas with increased growth and recurrence potential. Larger validation studies are needed to confirm the findings from this explorative pilot study.
背景：预测垂体腺瘤术后再生长潜力具有挑战性。垂体腺瘤的全基因组DNA甲基化分析可将腺瘤分为与组织学亚型相对应的不同甲基化类别。已有研究提出涉及再生的特定基因和差异甲基化探针，但尚无研究将这种表观遗传变异与非功能性垂体腺瘤的再生长潜力及临床异质性联系起来。本研究旨在探讨DNA甲基化谱是否可作为临床预后标志物。方法：采用Illumina MethylationEPIC芯片对2007年至2017年间接受经蝶手术的54例垂体大腺瘤样本进行DNA甲基化分析。其中12例患者因术后随访不完整、生物库储存组织退化或DNA甲基化质量较低而被排除。为定量测量肿瘤再生长率，我们通过年度磁共振成像对肿瘤残余体积进行三维体积分析，并采用线性混合效应模型检验不同DNA甲基化簇是否具有不同的再生长模式。结果：42例组织样本的DNA甲基化谱分析显示稳定的DNA甲基化簇，与既往研究结果一致。其中33例SF1谱系非功能性垂体腺瘤亚组可进一步分为五个亚簇，其近似无偏评分达86%。比较100%、50%或0%再生长风险比时，总体未发现统计学显著差异。尽管如此，相关生存估计图显示某些簇的再生长率较高。累积再生长的混合效应模型同样提示特定DNA甲基化簇与再生长潜力之间存在关联趋势。结论：非功能性垂体腺瘤的DNA甲基化谱分析可能识别具有较高生长和复发潜力的腺瘤，但需更大规模的验证研究来确认本探索性先导研究的结果。
Genome-Wide DNA Methylation Profiling as a Prognostic Marker in Pituitary Adenomas—A Pilot Study
肿瘤(癌症)患者之家