Ogura Koichi, Hosoda Fumie, Nakamura Hiromi, Hama Natsuko, Totoki Yasushi, Yoshida Akihiko, Ohashi Shoko, Rokutan Hirofumi, Takai Erina, Yachida Shinichi, Kawai Akira, Tanaka Sakae, Shibata Tatsuhiro
Division of Cancer Genomics, National Cancer Center Research Institute, 5-1-1 Tsukiji, Chuo-ku, Tokyo, 104-0045, Japan.
Department of Musculoskeletal Oncology, National Cancer Center Hospital, Tokyo, Japan.
Genes Chromosomes Cancer. 2017 Oct;56(10):711-718. doi: 10.1002/gcc.22469. Epub 2017 Jul 27.
Recurrent H3F3A and IDH2 mutations have been reported in giant cell tumor of bone (GCTB). However, the reported incidences have varied, and other molecular genetic alterations have not been identified due to the small number of cases analyzed with comprehensive methods. Moreover, the relative sensitivities of Sanger sequencing and next-generation sequencing (NGS) for the detection of H3F3A mutations in DNA extracted from archival formalin-fixed paraffin-embedded (FFPE) samples for clinical diagnosis have not been assessed. To address these issues, we conducted whole-exome sequencing of 7 GCTBs and integrated the previously published genomic sequencing data of 6 GCTBs. We subsequently performed targeted sequencing of an additional 39 GCTBs, including 2 atypical cases and an extremely rare case of primary malignant transformation of GCTB. We also evaluated the sensitivity of Sanger sequencing for detecting H3F3A mutations in FFPE samples that are usually used for clinical diagnosis. H3F3A glycine hotspot mutations were the most frequently detected mutations (96%) in the 52 GCTBs by NGS. Of the 50 hotspot mutations, p.G34W was observed in 48 cases and p.G34L/G34R was detected in one. One of two atypical GCTB cases with wild-type H3F3A had a H3F3B mutation (p.G34V). Other mutated genes were not recurrent. Sanger sequencing did not detect H3F3A mutations in 10 of 15 H3F3A NGS mutation-positive FFPE samples. In conclusion, we confirmed that H3F3A is the most frequently mutated GCTB driver gene, and that H3F3A mutations are not present in atypical GCTBs. Sanger sequencing was much less sensitive than targeted NGS for detecting H3F3A mutations in FFPE samples.
复发性H3F3A和IDH2突变已在骨巨细胞瘤（GCTB）中被报道。然而，报道的发生率有所不同，并且由于采用综合方法分析的病例数量较少，尚未鉴定出其他分子遗传改变。此外，尚未评估桑格测序法和新一代测序（NGS）对从存档福尔马林固定石蜡包埋（FFPE）样本中提取的用于临床诊断的DNA检测H3F3A突变的相对敏感性。为了解决这些问题，我们对7例GCTB进行了全外显子测序，并整合了之前发表的6例GCTB的基因组测序数据。随后，我们对另外39例GCTB进行了靶向测序，包括2例非典型病例和1例极其罕见的GCTB原发性恶性转化病例。我们还评估了桑格测序法对通常用于临床诊断的FFPE样本中检测H3F3A突变的敏感性。通过NGS在52例GCTB中检测到的最常见突变是H3F3A甘氨酸热点突变（96%）。在50个热点突变中，48例观察到p.G34W，1例检测到p.G34L/G34R。2例H3F3A野生型的非典型GCTB病例中有1例发生了H3F3B突变（p.G34V）。其他突变基因未复发。在15个H3F3A NGS突变阳性的FFPE样本中，桑格测序法未检测到其中10个样本中的H3F3A突变。总之，我们证实H3F3A是GCTB中最常发生突变的驱动基因，并且非典型GCTB中不存在H3F3A突变。在FFPE样本中检测H3F3A突变时，桑格测序法的敏感性远低于靶向NGS。
