嗜鉻細胞瘤和副神經節瘤的次世代定序基因檢測

Abstract

「嗜鉻細胞瘤與副神經結瘤」是罕見惡性之內分泌腫瘤，遺傳學證實兩者為相同之疾病，在最新第五版WHO內分泌和神經內分泌腫瘤已被重新分類為「副神經結瘤」。此腫瘤具有獨特之臨床與生化特徵，其分泌之兒茶酚胺作用於血管系統會造成高血壓危象與危害生命之心臟病變。具有高達40% 之生殖細胞系遺傳之易感性，且與多種遺傳型家族性症候群相關。 自2016年實驗室前輩們建立了「嗜鉻細胞瘤與副神經結瘤」以雜合捕獲為基礎之標的次世代定序基因檢測平台，於臨床確診收案之周邊血液基因檢測已可達到34% 之檢出率。然而，對這三分之二之的家族中首位確診病患的收案檢測結果為「未發現」或「臨床意義尚未確認」會影響基因功能之變異位點之個案，因此無法釐清其真正致病之基因變異、預測疾病進程、特定分子分群相關之預後與遺傳風險。此研究首要是提高致病變異之檢出率，使用福馬林固定石蠟包埋之腫瘤組織，利用聲能聚焦處理的技術來萃取DNA和RNA，將萃取出的DNA以廣泛包含36個基因區域標的嗜鉻細胞瘤與副神經結瘤之基因套組來捕獲和擴增建庫，使用Illumina NovaSeq為主的基因定序平台，再以基因體分析工具包Mutect2執行體細胞變異位點偵測之數據分析。其次，本實驗收案周邊血檢測結果為帶有「確定」或「可能」會影響影響生殖細胞系基因功能之變異位點之個案，一方面可提供使用福馬林固定石蠟包埋之組織檢測方法之確效參考，另一方面一旦有體細胞變異之發現，可能有助於預後之預測與精準醫療之應用。再者，就福馬林固定石蠟包埋之腫瘤組織在目前新建立之檢測方法與次世代定序之平台上，所提供之效能是否優於周邊血加以評估。 研究收案總數有62位是家族中首位確診病患，在有開刀可取得福馬林固定石蠟包埋之腫瘤組織也完成數據分析有26位，當中包含11位周邊血檢測結果為帶有「確定」或「可能」會影響影響生殖細胞系基因功能之變異位點，而且沒有發現鑲嵌現象。有5位在周邊血生殖細胞系為「未發現」或「臨床意義尚未確認」之，但在腫瘤組織被檢測出「可能」會影響基因功能之變異位點，因此為體細胞變異；包含2位HRAS [c.182A>G (p.Q61R)] 以及各1位的IDH2 [c.514A>G (p.R172G)]、EPAS1 [c.1592C>T (p.P531L)]、與TP53 [c.632_672+12del (p.X211_splice)]。在檢測有結果的16位中，有13位在分子分群是屬於假性低氧的分子群。在11位周邊血檢測結果為帶有「確定」或「可能」會影響生殖細胞系基因功能之變異位點，其腫瘤組織中也呈現生殖細胞系之「確定」與「可能」會影響基因功能之變異位點，包含有7位SDHD [c.3G>C (p.0)]、1位SDHB [c.136C>T (p.R46*)]、與3位VHL 呈現各自之變異位點 [c.500G>A (p.R167Q)]、[c.482G>A (p.R161Q)] 與 [c.277G>T (p.G93C)]。在已保存21年的福馬林固定石蠟包埋腫瘤組織，除了檢測出SDHD生殖系變異，也呈現腫瘤變異等位基因頻率為7% 的KMT2D [c.2506C>T (p.Q836*)] 體細胞變異。周邊血的檢出率為42%，福馬林固定石蠟包埋之腫瘤組織檢出率為62%，且此兩種檢體所檢測出之生殖細胞系變異位點是一致的。並且在最低腫瘤比例30% 與最久21年之福馬林固定石蠟包埋之腫瘤組織皆可測到「確定」或「可能」致病變異之存在。 此研究發現這5位罹病個案之體細胞之「可能」致病變異之發現，對於提供進一步遺傳諮詢來闡明致病原因是重要的。當中一位檢測出TP53之體細胞變異之結果，可能是可以解釋其腫瘤對肝臟和骨頭轉移的高侵襲性行為。在11位腫瘤組織呈現與周邊血生殖細胞系之變異一致，這結果確效了使用福馬林固定石蠟包埋之組織與對應之檢測流程。其中7位為SDHD之結果，可能是奠基者效應，導致整體分布呈現16位中有13位為假性缺氧為主的分子群現象。在其中一位生殖細胞系變異SDHD中，腫瘤組織發現有亞群之KMT2D體細胞變異，有可能會影響其臨床表現與預後。由於致病變異位點之檢出率由周邊血的42% 提升至福馬林固定石蠟包埋之腫瘤組織的62%，因此福馬林固定石蠟包埋之腫瘤組織整體效能優於周邊血。此研究之限制在於無法完全排除生殖細胞系鑲嵌現象遺傳給下一代之可能性，遺傳諮詢門診與針對該致病體細胞變異之位點執行試管嬰兒之胚胎著床前基因檢測是可能之解決方案。 使用腫瘤組織之致病變異檢測率優於周邊血，腫瘤組織可測得體細胞和生殖細胞系變異，其中又以假性缺氧分子群為主。次世代定序與腫瘤組織之結合不僅優化了基因驅動之基礎，可作為與特定分子群的預後預測與遺傳諮詢之依據，更可作為在精準醫療之新利基，應用於蓬勃發展的不定腫瘤類型之標靶治療用藥。因此建議在嗜鉻細胞瘤和副神經節瘤之基因檢測，可優先採用腫瘤組織的檢體。

Pheochromocytoma and paraganglioma (PPGL), the same disease demonstrated by genetics and reclassified as “paraganglioma” in the latest WHO Endocrine and Neuroendocrine Tumours (5th ed.), are rare malignant endocrine tumors. PPGL manifests unique clinical and biochemical characteristics. Catecholamines secreted by tumors acting on the vascular system can cause hypertensive crisis and life-threatening cardiomyopathy. PPGL has up to 40% genetic predisposition and is associated with multiple hereditary syndromes. Since 2016, the Hybrid capture-based targeted Next Generation Gene Sequencing (NGS) platform for the genetic test of PPGL established by the laboratory's predecessors has achieved a 34% detection rate for peripheral blood (PB). Nevertheless, the results of the two-thirds of proband enrollment revealed “not found” or “a variant of uncertain significance (VUS)”. Hence, the causal variants of the disease, the prediction of disease progression, the prognosis related to specific molecular clusters, and the risk of inheritance were unable to be determined. First and foremost, to increase the detection rate of causal variants, this study was conducted using formalin-embedded paraffin-embedded (FFPE) tumor tissue, the technology of Adaptive Focused Acoustics for extraction of DNA and RNA, the comprehensive PPGL-targeted gene panel composed of 36 genes for capture and enrichment of DNA for library construction, the Illumina NovaSeq as prime sequencing platform, and the GATK Mutect2 for data analysis of somatic variant calling. Next, the probands that had pathogenic or likely pathogenic variants of PB, for one thing, which could provide references for the validation of the assay using FFPE tissue; for another, which might be beneficial in prognosis prediction and application of precision medicine. Lastly, whether the effectiveness of FFPE tissue was superior to PB on the newly established method and NGS platform would be evaluated. The total number of enrollments is 62 probands. The data analysis was accomplished in 26 probands in which FFPE tumor tissue could be available after surgery. Eleven probands among them showed pathogenic or likely pathogenic variants of PB, and without mosaicism. Five probands, with the results of “not found” or “VUS” of PB, were detected the presence of somatic likely pathogenic variants, including 2 HRAS [c.182A>G (p.Q61R)], 1 IDH2 [c.514A>G (p.R172G)], 1 EPAS1 [c.1592C>T (p.P531L)] and 1 TP53 [c.632_672+12del (p.X211_splice)] in the FFPE tumor tissue. Among 16 probands with "pathogenic" or "likely pathogenic" variants, 13 probands were classified as the pseudohypoxia molecular cluster. Eleven probands, with the germline "pathogenic" or "likely pathogenic" variants of PB, were demonstrated "pathogenic" or "likely pathogenic" variants of germline in the FFPE tumor tissue as well, including 7 SDHD [c.3G>C (p.0)], 1 SDHB [c.136C>T (p.R46*)], and 3 VHL exhibiting different codons, such as [c.500G>A (p.R167Q)]、[c.482G>A (p.R161Q)] and [c.277G>T (p.G93C)]. The KMT2D somatic variation with a 7% frequency of tumor variant allele was identified additionally in the background of germline variation of SDHD from the 21-year-old FFPE tumor tissue. The detection rate of causal variants of PB was 42%, and that of the FFPE tissue was 62%. Moreover, the codons of germline variants in both specimens were identical. The FFPE tumor tissue with a minimum tumor cell content of 30% and a maximum storage time of 21 years, could be demonstrated the presence of "pathogenic" or "likely pathogenic" variants. The findings of somatic "likely pathogenic" variants of these 5 probands are important in genetic consultation to clarify the cause of the disease. The detection of TP53 somatic variant in one proband without germline variant might give an explanation of the highly aggressive behavior of the tumor metastasizing to the liver and bone. In 11 probands, the tumor tissues displayed germline variation consistent with PB, which validated the feasibility of using FFPE tumor tissue and the corresponding assay. Seven probands of SDHD germline variant possessed the founder effect, which contributed to an overall distribution of 13 out of 16 probands with the dominant pseudohypoxia molecular cluster. The presence of the subgroup of KMT2D somatic variant in one proband with the germline variation of SDHD may have affected the clinical behavior and prognosis. The effectiveness of FFPE tumor tissue was superior to that of PB because the detection rate of causal variants increased from 42% in PB to 62% in FFPE tumor tissue. This study was limited by the fact that the possibility of germline mosaicism being passed on to the next generation cannot be completely ruled out. Genetic consultation and preimplantation genetic testing for this causative somatic variant would be the possible solutions. Tumor tissue showed higher detection rates of causal variants than peripheral blood. Among the somatic and germline variants demonstrated by tumor tissue, pseudo-hypoxic is the dominant molecular cluster. The combination of NGS and PPGL tumor tissue not only optimizes the genetic-driven basis for prognostic predictions of specific molecular clusters and genetic counseling but also serves as a new niche in precision medicine for targeted therapy in the burgeoning field of tumor-agnostic drugs. Therefore, it is recommended that genetic testing for PPGL be prioritized with the use of tumor tissue.