Solitary fibrous tumors (SFTs) are NAB2-STAT6 fusion-associated neoplasms. There are several subtypes of NAB2-STAT6 fusions, but their clinical significances are still unclear. Moreover, the mechanisms of malignant progression are also poorly understood. In this study, using 91 SFT cases, we examined whether fusion variants are associated with clinicopathological parameters and also investigated the molecular mechanism of malignant transformation using whole-exome sequencing. We detected variant 1b (NAB2ex4-STAT6ex2) in 51/91 (56%) cases and variants 2a/2b (NAB2ex6-STAT6ex16/17) in 17/91 (19%) cases. The NAB2-STAT6 fusion variant types were significantly associated with their primary site (P < 0.001). In addition, a TERT promoter mutation was detected in 7/73 (10%) cases, and it showed a significant association with malignant SFTs (P = 0.003). To identify molecular changes during malignant progression, we selected an index patient to obtain parallel tissue samples from the primary and metastatic tumors. In the metastatic tissue, 10 unique molecular alterations, including those in TP53 and APAF1, were detected. In vitro functional experiments showed that APAF1 depletion increased the tumor potency of cells expressing NAB2-STAT6 fusion protein under treatment with staurosporine. We found that TP53 immunopositivity (P = 0.006) and loss of APAF1 immunoreactivity (P < 0.001) were significantly associated with malignant SFTs. Our study suggests that dysfunction of TP53 and APAF1 leads to impaired apoptotic function, and eventually contributes toward malignant SFT transformation.Key messages We firstly found that the TERT promoter mutation was strongly associated with malignant SFTs (P = 0.003) and the representative 1b (NAB2ex4-STAT6ex2) or 2a (NAB2ex6-STAT6ex16) fusion variants similarly contribute to tumorigenicity.We also found that TP53 immunopositivity (P = 0.006) and loss of APAF1 immunoreactivity (P < 0.001) were significantly associated with malignant SFTs.Our study suggests that dysfunction of TP53 and APAF1 leads to impaired apoptotic function, and eventually contributes toward malignant SFT transformation. Electronic supplementary materialThe online version of this article (10.1007/s00109-019-01815-8) contains supplementary material, which is available to authorized users.

Formalin fixing with paraffin embedding (FFPE) has been a standard sample preparation method for decades, and archival FFPE samples are still very useful resources. Nonetheless, the use of FFPE samples in cancer genome analysis using next-generation sequencing, which is a powerful technique for the identification of genomic alterations at the nucleotide level, has been challenging due to poor DNA quality and artificial sequence alterations. In this study, we performed whole-exome sequencing of matched frozen samples and FFPE samples of tissues from 4 cancer patients and compared the next-generation sequencing data obtained from these samples. The major differences between data obtained from the 2 types of sample were the shorter insert size and artificial base alterations in the FFPE samples. A high proportion of short inserts in the FFPE samples resulted in overlapping paired reads, which could lead to overestimation of certain variants; >20% of the inserts in the FFPE samples were double sequenced. A large number of soft clipped reads was found in the sequencing data of the FFPE samples, and about 30% of total bases were soft clipped. The artificial base alterations, C>T and G>A, were observed in FFPE samples only, and the alteration rate ranged from 200 to 1,200 per 1M bases when sequencing errors were removed. Although high-confidence mutation calls in the FFPE samples were compatible to that in the frozen samples, caution should be exercised in terms of the artifacts, especially for low-confidence calls. Despite the clearly observed artifacts, archival FFPE samples can be a good resource for discovery or validation of biomarkers in cancer research based on whole-exome sequencing.

Normalization of mRNA levels using endogenous reference genes (ERGs) is critical for an accurate comparison of gene expression between different samples. Despite the popularity of traditional ERGs (tERGs) such as GAPDH and ACTB, their expression variability in different tissues or disease status has been reported. Here, we first selected candidate housekeeping genes (HKGs) using human gene expression data from different platforms including EST, SAGE, and microarray, and 13 novel ERGs (nERGs) (ARL8B, CTBP1, CUL1, DIMT1L, FBXW2, GPBP1, LUC7L2, OAZ1, PAPOLA, SPG21, TRIM27, UBQLN1, ZNF207) were further identified from these HKGs. The mean coefficient variation (CV) values of nERGs were significantly lower than those of tERGs and the expression level of most nERGs was relatively lower than high expressing tERGs in all dataset. The higher expression stability and lower expression levels of most nERGs were validated in 108 human samples including formalin-fixed paraffin-embedded (FFPE) tissues, frozen tissues and cell lines, through quantitative real-time RT-PCR (qRT-PCR). Furthermore, the optimal number of nERGs required for accurate normalization was as few as two, while four genes were required when using tERGs in FFPE tissues. Most nERGs identified in this study should be better reference genes than tERGs, based on their higher expression stability and fewer numbers needed for normalization when multiple ERGs are required.

MET, a cell surface receptor for hepatocyte growth factor, is involved in the development of triple-negative/basal-like breast cancer (TNBC/BLBC). However, its utility as a therapeutic target in this subtype of breast cancer is poorly understood. To evaluate MET fully as a potential therapeutic target for TNBC/BLBC, we investigated the relationship between MET expression and clinical outcomes of patients with breast cancer and the functional effect of MET inhibition. Using automated immunohistochemistry (Ventana), we analyzed MET expression in 924 breast cancer patients with relevant clinicopathologic parameters. BLBC showed the strongest relationship with MET expression (57.5%, p < 0.001). High expression of MET in breast cancer resulted in poor overall survival (p 5 0.001) and disease-free survival (DFS, p 5 0.010). MET expression was relatively high in TNBC cell lines, and the silencing of MET via small interfering RNA reduced cell proliferation and migration. We observed reduced TNBC cell viability after treatment with the MET inhibitor PHA-665752. In the most drug-resistant cell line, MDA-MB-468, which showed elevated epidermal growth factor receptor (EGFR) expression, silencing of EGFR resulted in increased sensitivity to PHA-665752 treatment. We confirmed that PHA-665752 synergizes with the EGFR inhibitor erlotinib to decrease the viability of MDA-MB-468 cells. TNBC patients coexpressing MET and EGFR showed significantly worse DFS than that in patients expressing EGFR alone (p 5 0.021). Our findings strongly suggest that MET may be a therapeutic target in TNBC and that the combined therapy targeting MET and EGFR may be beneficial for the treatment of TNBC/BLBC patients.

The prevalence of METex14 skipping was quite high in East Asian patients without other driver mutations in lung adenocarcinomas. METex14 skipping was associated with old age, the acinar or solid histologic subtype, and high MET immunohistochemical expression. The prognosis of patients with METex14 skipping was similar to that of patients with major driver mutations. siRNA targeting the junction of METex14 skipping could inhibit MET-driven signaling pathways in cells with METex14 skipping.

Surgery and radiation are the current standard treatments for cervical cancer. However, there is no effective therapy for metastatic or recurrent cases, necessitating the identification of therapeutic targets. In order to create preclinical models for screening potential therapeutic targets, we established 14 patient-derived xenograft (PDX) models of cervical cancers using subrenal implantation methods. Serially passaged PDX tumors retained the histopathologic and genomic features of the original tumors. Among the 9 molecularly profiled cervical cancer patient samples, a HER2-amplified tumor was detected by array comparative genomic hybridization and targeted next-generation sequencing. We confirmed HER2 overexpression in the tumor and serially passaged PDX. Co-administration of trastuzumab and lapatinib in the HER2-overexpressed PDX significantly inhibited tumor growth compared to the control. Thus, we established histopathologically and genomically homologous PDX models of cervical cancer using subrenal implantation. Furthermore, we propose HER2 inhibitor-based therapy for HER2-amplified cervical cancer refractory to conventional therapy.

BackgroundTriple-negative breast cancers (TNBCs) and basal-like breast cancers (BLBCs) are known as poor outcome subtypes with a lack of targeted therapy. Previous studies have shown conflicting results regarding the difference of prognostic significance between TNBCs and BLBCs. In this study, we aimed to characterize the prognostic features of TNBCs, in view of BLBCs and quintuple-negative breast cancers (QNBC/5NPs).MethodsUsing tissue microarray-based immunohistochemical analysis, we categorized 951 primary breast cancers into four or five subtypes according to the expression of ER, PR, HER2, and basal markers (CK5/6, EGFR).ResultsThe results of this study showed that both TNBCs and BLBCs were associated with high histological and/or nuclear grades. When the TNBCs are divided into two subtypes by the presence of basal markers, the clinicopathologic characteristics of TNBCs were mainly maintained in the BLBCs. The 5-subgrouping was the better prediction model for both disease free and overall survival in breast cancers than the 4-subgrouping. After multivariate analysis of TNBCs, the BLBCs did not have a worse prognosis than the QNBC/5NPs. Interestingly, the patients with BLBCs showed significant adjuvant chemotherapy benefit. In addition, QNBC/5NPs comprised about 6~8% of breast cancers in publicly available breast cancer datasetsConclusionThe QNBC/5NP subtype is a worse prognostic subgroup of TNBCs, especially in higher stage and this result may be related to adjuvant chemotherapy benefit of BLBCs, calling for caution in the identification of subgroups of patients for therapeutic classification.

In this cohort, approximately 85% of individuals have known driver mutations (EGFR 59.4%, KRAS 7.4%, ALK 7.4%, ERBB2 2.6%, ROS1 2.2%, RET 2.2%, MET 1.8%, BRAF 1.1%, and NRAS 0.4%). Seventy percent of smokers and 90% of nonsmokers had defined oncogenic drivers matching the U.S. Food and Drug Administration-approved targeted therapies.