- 23 例儿童肿瘤（18 CNS / 5 非 CNS）并行 Illumina EPIC 与 Oxford Nanopore（ONT）甲基化分析；CNS 例中两平台谱高度相关（1 例离群后剔除）。
- crossNN 在家族层级与组织学（无 NGS）一致；CNV 高度一致；MGMT 启动子状态一致率 94%（16/17）。
- ONT 可实现当日、样本级的家族层级分类；EPIC 在类别层级仍略占优；ONT 依赖新鲜冰冻 DNA，且分类器多围绕芯片 CpG 位点构建。
摘要
DNA 甲基化谱可使儿童中枢神经系统（CNS）肿瘤获得更精确的分类。Oxford Nanopore Technologies（ONT）可提供当日、单样本甲基化读出，但其与 Illumina EPIC 芯片在常规诊断任务中的一致性仍未完全明确。我们对 23 例儿童肿瘤（18 例 CNS，5 例非 CNS）分别以 EPIC 芯片与 ONT 进行谱分析。两平台甲基化谱均用 crossNN（脑模型或泛癌模型）分类；ONT 数据另用 Rapid-CNS2 与 Sturgeon 分类。比较内容包括：（i）分类器与整合组织学（无 NGS）在家族/类别层级的一致性；（ii）超过平台特异性得分阈值的通过率；（iii）拷贝数变异（CNV）与 MGMT 启动子甲基化状态的跨平台一致性。
在 CNS 病例中，ONT 与 EPIC 甲基化谱呈强相关，仅 1 例离群（P2）并自后续分析中剔除。两平台比较显示：（a）以 crossNN 对 CNS 肿瘤作分子分类时，全部病例在家族层级与组织学（无 NGS）一致；（b）拷贝数谱跨平台高度一致；（c）MGMT 启动子甲基化状态在 94% 病例（16/17）匹配。在仅用 ONT 数据比较 ONT 专用分析流程时，Rapid-CNS2 在类别层级赋值最可靠，与组织病理诊断一致率 94%（16/17），略高于 crossNN 与 Sturgeon。在非 CNS 肿瘤中，泛癌模型输出置信度低，与组织学（无 NGS）一致性差（仅 1/5 一致），提示对这些实体尚不具备常规应用条件。
总之，ONT 可实现当日、临床可靠的家族层级 CNS 肿瘤分类，并与芯片高度一致，而 EPIC 在类别层级仍略占优势。ONT 的关键限制在于依赖新鲜冰冻 DNA，以及分类器多围绕芯片来源 CpG 位点构建，而非原生基于 ONT 数据训练的模型。
Abstract
DNA methylation profiling enables precise classification of pediatric central nervous system (CNS) tumors. Oxford Nanopore Technologies (ONT) offers same-day, single-sample methylation readouts, but its concordance with Illumina EPIC arrays in routine diagnostic tasks remains incompletely defined. We profiled 23 pediatric tumors (18 CNS, 5 non-CNS) by EPIC arrays and ONT. Methylation profiles from both platforms were classified with crossNN (brain model or pan-cancer model); ONT data were additionally classified with Rapid-CNS2 and Sturgeon. We compared (i) classifier agreement with integrated histology (w/o NGS) at family/class levels, (ii) pass-rate above platform-specific score cutoffs, (iii) cross-platform concordance of copy-number variation (CNV), and MGMT promoter methylation status. In CNS cases, ONT and EPIC methylation profiles demonstrated strong correlation, except for a single outlier (P2), which was excluded from further analysis. Comparative assessment of the two platforms showed that: (a) Molecular classification of CNS tumors using the crossNN classifier was consistent with histology (w/o NGS) at the family level in all cases. (b) Copy-number profiles showed high concordance between platforms. (c) MGMT promoter methylation status matched in 94% of cases (16/17). When comparing ONT-specific analysis pipelines using the ONT data, the Rapid-CNS2 pipeline yielded the most reliable class level assignments with 94% (16/17) concordance with the histopathological diagnosis, which marginally exceeded the crossNN and sturgeon classifiers. In non-CNS tumors, the pan-cancer model produced low-confidence outputs with poor agreement with histology (w/o NGS) (only 1/5 concordant), indicating limited readiness for these entities. In conclusion, ONT enables same-day, clinically reliable family-level CNS tumor classification with high concordance to arrays, while EPIC retains a modest class-level edge. A key limitation of ONT is its reliance on fresh-frozen DNA and on classifiers originally built around array-derived CpG sites, rather than on models developed natively from ONT data.