Purpose: Colorectal cancers with microsatellite instabilityhigh (MSI-H) status, due to mismatch repair deficiency, are associated with poor patient outcomes after relapse. We aimed to identify novel therapeutic targets for them.Experimental Design: We performed MSI analyses of over 2,800 surgically resected colorectal tumors obtained from consecutive patients treated in Japan from 1998 through June 2016. Whole-exome sequencing, transcriptome sequencing, and methylation analyses were performed on 149 of 162 tumors showing MSI in BAT25 and BAT26 loci. We analyzed patient survival times using Bonferroni-adjusted log-rank tests.Results: Sporadic MSI-H colorectal cancers with promoter methylation of MLH1 (called MM) had a clinicopathological profile that was distinct from that of colorectal cancers of patients with germline mutations (Lynch syndrome, LS-associated) or somatic, Lynch-like mutations in mismatch repair genes. MM tumors had more insertions and deletions and more recurrent mutations in BRAF and RNF43 than LS-associated or Lynch-like MSI-H tumors. Eleven fusion kinases were exclusively detected in MM MSI-H colorectal cancers lacking oncogenic KRAS/BRAF missense mutations and were associated with worse post-relapse prognosis. We developed a simple method to identify MM tumors and applied it to a validation cohort of 28 MSI-H colorectal cancers, identifying 16 MM tumors and 2 fusion kinases.Conclusions: We discovered that fusion kinases are frequently observed among sporadic MM MSI-H colorectal cancers. The new method to identify MM tumors enables us to straightforwardly group MSI-H patients into candidates of LS or fusion kinase carriers.

Every year, approximately 1.2 million cases of colorectal carcinoma (CRC) are newly diagnosed worldwide. Although metastases to distant organs are often fatal complications of CRC, little information is known as to how such metastatic lesions are formed. To reveal the genetic profiles for CRC metastasis, we conducted whole‐exome RNA sequencing on CRC tumors with liver metastasis (LM) (group A, n = 12) and clinical stage‐matched larger tumors without LM (group B, n = 16). While the somatic mutation profiles were similar among the primary tumors and LM lesions in group A and the tumors in group B, the A‐to‐C nucleotide change in the context of “AAG” was only enriched in the LM regions in group A, suggesting the presence of a DNA damage process specific to metastasis. Genes already known to be associated with CRC were mutated in all groups at a similar frequency, but we detected somatic nonsynonymous mutations in a total of 707 genes in the LM regions, but not in the tumors without LM. Signaling pathways linked to such “LM‐associated” genes were overrepresented for extracellular matrix‐receptor interaction or focal adhesion. Further, fusions of the ADAP1 (ArfGAP with dual PH domain 1) were newly identified in our cohort (3 out of 28 patients), which activated ARF6, an ADAP1‐substrate. Infrequently, mutated genes may play an important role in metastasis formation of CRC. Additionally, recurrent ADAP1 fusion genes were unexpectedly discovered. As these fusions activate small GTPase, further experiments are warranted to examine their contribution to CRC carcinogenesis.

Objective: To determine whether the benefits of dapagliflozin in patients with heart failure and reduced ejection fraction (HFrEF) and type 2 diabetes in DAPA-HF varied by background glucose-lowering therapy (GLT). Research design and methods: We examined the effect of study treatment by the use or not of GLT, and by GLT classes and combinations. The primary outcome was a composite of worsening HF (hospitalization or urgent visit requiring intravenous therapy) or cardiovascular death. Results: In the 2139 type 2 diabetes patients, the effect of dapagliflozin on the primary outcome was consistent by GLT use/no use (hazard ratio 0.72 [95%CI 0.58-0.88] versus 0.86 [0.60-1.23]; P-interaction=0.39) and across GLT classes. Conclusions: In DAPA-HF, dapagliflozin improved outcomes irrespective of use/no use of GLT or by GLT type used in patients with type 2 diabetes and HFrEF.

<p>Supplementary Tables S1-S12: Supplementary Table S1: Primer sequences used in this study. Supplementary Table S2: Clinical information of tumours analysed in the present study. Supplementary Table S3: Patients' characteristics. Supplementary Table S4: Likely causal mutations in mismatch repair genes detected in this study. Supplementary Table S5: List of Tier 2 mutations. Supplementary Table S6: Significantly mutated genes identified with MutSigCV. Supplementary Table S7: Genes differentially altered between Lynch syndrome-associated/Lynch-like tumours and MLH1 promoter-methylated tumours. Supplementary Table S8: Recurrent copy number alterations detected with allele-specific copy number analysis. Supplementary Table S9: Pathways detected from top 209 Tier 2 genes with The Database for Annotation, Visualization and Integrated Discovery. Supplementary Table S10: Pathways detected from significantly mutated genes with The Database for Annotation, Visualization and Integrated Discovery. Supplementary Table S11: Detected oncogenic alterations and associated mutations with uncertain significance. Supplementary Table S12: Complementary DNA sequences of fusion points confirmed by reverse-transcription polymerase chain reaction followed by Sanger sequencing.</p>

<div>AbstractPurpose:<p>Colorectal cancers with microsatellite instability–high (MSI-H) status, due to mismatch repair deficiency, are associated with poor patient outcomes after relapse. We aimed to identify novel therapeutic targets for them.</p>Experimental Design:<p>We performed MSI analyses of over 2,800 surgically resected colorectal tumors obtained from consecutive patients treated in Japan from 1998 through June 2016. Whole-exome sequencing, transcriptome sequencing, and methylation analyses were performed on 149 of 162 tumors showing MSI in BAT25 and BAT26 loci. We analyzed patient survival times using Bonferroni-adjusted log-rank tests.</p>Results:<p>Sporadic MSI-H colorectal cancers with promoter methylation of <i>MLH1</i> (called MM) had a clinicopathological profile that was distinct from that of colorectal cancers of patients with germline mutations (Lynch syndrome, LS-associated) or somatic, Lynch-like mutations in mismatch repair genes. MM tumors had more insertions and deletions and more recurrent mutations in <i>BRAF</i> and <i>RNF43</i> than LS-associated or Lynch-like MSI-H tumors. Eleven fusion kinases were exclusively detected in MM MSI-H colorectal cancers lacking oncogenic <i>KRAS</i>/<i>BRAF</i> missense mutations and were associated with worse post-relapse prognosis. We developed a simple method to identify MM tumors and applied it to a validation cohort of 28 MSI-H colorectal cancers, identifying 16 MM tumors and 2 fusion kinases.</p>Conclusions:<p>We discovered that fusion kinases are frequently observed among sporadic MM MSI-H colorectal cancers. The new method to identify MM tumors enables us to straightforwardly group MSI-H patients into candidates of LS or fusion kinase carriers.</p></div>