Advantages
- Capable of determining polyploidy, which is associated with poor tumor prognosis (metastasis, drug resistance).
- Easily determines regional diploidy from pathological images obtained in daily medical practice.
Technology Overview & Background
Efforts have been made to utilize pathological findings and genetic information as predictive markers for anticancer drugs and tumor prognosis. However, reliable prognosis prediction remains difficult due to the diversity and constant evolution of cancer treatment. Across many cancer types, tumor cells frequently undergo genomic polyploidization. Genomic polyploidy in tumor cells is emerging as a novel marker associated with poor prognosis because it leads to chromosomal instability that can drive cancer metastasis and drug resistance. However, conventional methods to assess tumor ploidy, such as flow cytometry, chromosome FISH, and genome sequencing, are complex, costly, and not feasible for routine clinical use.
To address this challenge, researchers have developed the present invention technology that can easily determine tumor ploidy directly from pathological images. This technology can detect the polyploidy-associated features in cancer tissue and classify tumor ploidy by using the Multiplicity Determination Model, a deep learning–based AI system, without the need for special pathological stains. This innovation makes it possible to identify highly malignant, treatment-resistant polyploid cancers from standard pathological images obtained in everyday medical practice. It holds promise as a powerful tool for predicting treatment response and patient prognosis, and it is expected to play an important role as a companion diagnostic method for cancers characterized by chromosomal instability, which is currently under development.
Patent(s)
PCT/JP2023/044155 (published in Japanese as WO2025/182244)
Publication(s)
Matsuura, T., Abe, M. et al. Commun Med (2025) 5, 270.
[Open Access] https://doi.org/10.1038/s43856-025-00967-8
Principal Investigator & Academic Institution
Tomonori Matsumoto, MD, PhD (Associate Professor, The University of Osaka, Japan)
Expectations
TECH MANAGE is seeking a company that is interested in licensing this invention for commercialization and practical use. Direct meetings with our researchers are also available. Please feel free to ask any questions you may have.
We would be happy to consider joint research with the researchers, exclusive evaluation, and options such as preferential negotiation rights prior to licensing.
Project ID:JT-04831