Koki Uchiyama

Mr. Koki Uchiyama

orcid

Research Focus & Metrics

Academic Career Length

1Year (2026-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
18 September 2026 in Genes
Verified
Journal Article
Open access
PDF file
KUKoki Uchiyama
Koki Uchiyama
HNHinano Nishikubo
Hinano Nishikubo
TSTomoya Sano
Tomoya Sano
YFYukina Furukawa
Yukina Furukawa
DMDONGHENG MA
DONGHENG MA
MYMasakazu Yashiro
Masakazu Yashiro
Background: Hereditary cancer predisposition syndromes show a high risk of cancer development with germline pathogenic variants. Organ-specific surveillance has been recommended according to the causative gene. However, even among carriers of the same germline variants, developing cancer types are not always uniform. Tumor–normal paired comprehensive genomic profiling (CGP) enables the simultaneous assessment of germline and somatic variants in same patient. Correlations between germline variants and cancer types have been reported; however, the significance of combinations of a germline variant and somatic variants remain unclear in the development of cancer types. Methods: In this study, we analyzed the correlation between pancreatic cancer and somatic variant patterns in patients with hereditary cancer predisposition syndrome, especially in patients with germline variants on BRCA1, BRCA2, or ATM. Results: Here, we show the characteristic pancreatic cancer patterns according to co-existing somatic variants in patients with germline pathogenic variants. The tumor–normal paired genetic data from 18,732 patients who underwent tumor–normal paired CGP were analyzed, including 853 germline-positive patients registered in a Japanese database. The combination of the somatic TP53 variant and KRAS variant was closely correlated with the development of pancreatic cancer in patients with germline BRCA1, BRCA2, or ATM. In particular, the somatic KRAS variant was significantly associated with pancreatic cancer. Conclusions: These findings suggest that pancreatic cancer of hereditary cancer syndromes might be associated with co-existing somatic KRAS variants within the pancreas.

15 November 2026 in Genes
Verified
Journal Article
Open access
PDF file
DMDONGHENG MA
DONGHENG MA
TSTomoya Sano
Tomoya Sano
HNHinano Nishikubo
Hinano Nishikubo
YFYukina Furukawa
Yukina Furukawa
KUKoki Uchiyama
Koki Uchiyama
MYMasakazu Yashiro
Masakazu Yashiro
Background: KRAS mutations occur in over 90% of pancreatic ductal adenocarcinomas (PDAC), and KRASG12D is the most frequent variant. G12D-selective inhibitors are now in clinical development, but computational pathology has so far predicted mutation status only at the gene level, leaving allele-specific prediction unexplored. Because genomic testing can be time-consuming and costly, a tool that predicts KRASG12D mutation status from routine hematoxylin and eosin (H&E) images could help identify patients for confirmatory molecular testing and potential treatment with KRASG12D inhibitors. Here we present a multi-scale framework that predicts KRASG12D status from H&E whole-slide images, with integration of computed tomography (CT). Methods: Two public cohorts (TCGA, CPTAC) formed the training set, and an independent external cohort (Osaka Metropolitan University Hospital) served as validation. We developed a dual-magnification pipeline. In the high-magnification branch, a pathology foundation model (CONCH v1.5) was fine-tuned with low-rank adaptation (LoRA) and an attention-based multiple-instance learning head on 20× patches; in the low-magnification branch, a generative image model encoded a whole-slide overview. The two branches were integrated by late fusion into a multi-scale pathomics model, which was further combined with a radiomics model of CT. Results: The multi-scale pathomics model reached a pooled internal test area under the receiver operating characteristic curve (AUC) of 0.611 and an external AUC of 0.701. In the external patients with CT, the combined model reached an AUC of 0.772, compared with 0.753 for pathomics alone. Conclusions: KRASG12D status leaves a signal in routine H&E images and CT of PDAC that deep learning can detect and that transfers to an independent institution. After prospective validation, the framework might help prioritize patients for KRAS testing.