Characterization of Neuroblastoma Tumor Behavior by Molecular Imaging

神經母細胞瘤（neuroblastoma；NB）起源於交感神經，好發於嬰幼兒。過半病童診斷時已有多處轉移，預後不佳。因此，臨床上需使用多種影像檢查來進行診斷、分期與追蹤。本研究欲以分子影像探討神經母細胞瘤的腫瘤行為特徵。首先，我們運用氟多巴正子造影（18F-FDOPA PET）在身體各部位診斷神經母細胞瘤；其敏感度達97.6%（95%信賴區間87.4%–99.9%），特異性87.5%（47.3%–99.7%），可與目前黃金標準的MIBG核子醫學掃描（123I-MIBG scintigraphy）或去氧葡萄糖正子造影（18F-FDG PET）互補。接下來，由2007年至2014年新診斷神經母細胞瘤的個案中，分析主要腫瘤FDG與FDOPA正子影像的最大標準吸收值（SUVmax），發現年齡較大、分期較嚴重、MYCN放大等較惡性特徵者，皆具有FDG吸收值較高、FDOPA吸收值較低的現象（所有P值 < 0.02）。利用接收者操作特徵曲線（receiver operating characteristics）統計，我們將FDG吸收值 ≥ 3.31及FDOPA吸收值 < 4.12定義為「正子影像極高危險群（PET ultra-high-risk）」，可區分出節段性染色體變化或MYCN致癌基因放大等高度惡性的基因體類型，為無事件存活率較差的獨立預測因子（多變數分析之風險比值 [HR] = 4.9 [1.9–30.1]，P = 0.012），且較不易以外科手術完整切除（gross total resection；切除率46%相對於100%，P = 0.04）。此外，FDG與FDOPA吸收值的比值（G:D）與糖解作用HK2激酶基因表現呈正相關（Spearman’s ρ = 0.86，P < 0.0001）、與多巴代謝酵素DDC基因表現呈負相關（ρ = -0.58；P = 0.02），表示PET-UHR腫瘤很可能具有較強的糖解作用活性及較差的神經代謝分化。最後，我們整合小動物超音波與正子造影，以Th-MYCN基因轉殖小鼠發展臨床前試驗平臺，發現其腫瘤75%為FDG吸收值較高、FDOPA吸收值較低，與人類神經母細胞瘤帶有MYCN放大者相似。利用此影像評估與追蹤平臺，我們驗證了BRD4表觀基因標靶藥物可強力抑制Th-MYCN小鼠腫瘤生長，但維他命A酸分化治療則無效。總結來說，FDG與FDOPA功能性分子影像可評估神經母細胞瘤的腫瘤行為、補強現有的危險群分類系統、發展為重要的臨床前評估工具，未來可望轉譯為運用分子影像發展個人化標靶治療的新穎治療策略。Neuroblastoma (NB) is a devastating tumor that usually occurs in young children. More than 50% of patients are diagnosed with metastatic diseases or other high-risk features, who suffer from a dismal prognosis. Therefore, a multitude of imaging studies are required for the staging and surveillance of NB. This study aims to characterize NB tumor behavior using molecular imaging modalities. First, we showed that positron emission tomography (PET) with 18F-fluoro-dihydroxyphenylalanine (FDOPA), a radiolabeled amino acid targeting catecholamine metabolism, had a sensitivity of 97.6% (87.4%–99.9%) and a specificity of 87.5% (47.3%–99.7%) and may compliment current 123I-metaiodobenzylguanidine (123I-MIBG) and 18F-fluorodeoxyglucose (FDG) PET scans. Next, NB patients undergoing paired FDG and FDOPA PET scans at diagnosis during 2007–2014 (N = 42; median age, 2.0 years; 28 boys and 14 girls;) were evaluated for the maximum standardized uptake value (SUVmax) of FDG or FDOPA by the primary tumor, revealing that NB with older age, advanced stages, or MYCN amplification showed higher FDG and lower FDOPA SUVmax (all P < 0.02). Receiver operating characteristics analysis identified FDG SUVmax ≥ 3.31 and FDOPA SUVmax < 4.12 as an ultra-high-risk feature (PET-UHR) that distinguished the most unfavorable genomic types, i.e. segmental chromosomal alterations and/or MYCN amplification, at a sensitivity of 81.3% (54.4%–96.0%) and a specificity of 93.3% (68.1%–99.8%). Considering with age, stage, MYCN status, and anatomical image-defined risk factor, PET-UHR was an independent predictor of inferior event-free survival (multivariate hazard ratio, 4.9 [1.9–30.1]; P = 0.012) and was associated lower likelihood of gross total resection of the main tumor (46% vs. 100%; P = 0.04). Meanwhile, the ratio between FDG and FDOPA SUVmax (G:D) correlated positively with hexokinase 2 (HK2; Spearman’s ρ = 0.86, P < 0.0001) and negatively with DOPA decarboxylase (DDC; ρ = -0.58, P = 0.02) gene expression levels, suggesting higher glycolytic activity and less catecholaminergic differentiation in NB tumors taking up higher FDG and lower FDOPA. Finally, we incorporated small-animal ultrasound screening and PET imaging to build up a preclinical trial platform. We characterized the tumor latency and progression in the hemizygous Th-MYCN mice, a genetically-engineered mouse model of NB. In parallel to human NB with MYCN amplification, the majority (75%) of Th-MYCN murine NB tumors showed higher uptake of FDG and lower uptake of FDOPA. Using the imaging platform, We have validated the potent therapeutic effect of epigenetic inhibition of MYCN transcription by targeting the bromodomain BRD4 reader protein, while 13-cis-retinoic acid showed no anti-tumor activity in this model. In conclusion, the functional FDG and FDOPA PET imaging characterizes the NB tumor behavior; complements the current risk stratification systems of NB; and serves as an important part of preclinical trails. The results of this study can be translated into critical clinical knowledge of molecular imaging of NB, and may pave the way of individualized target therapy through molecular imaging