Craif announces joint research results on urological cancers (renal cell carcinoma, prostate cancer, urothelial carcinoma) at EAU 2026 (European Association of Urology)

Bio-AI startup Craif Inc. (Headquarters: Shinjuku-ku, Tokyo; Representative Director CEO: Ryuichi Onose; hereinafter "Craif") jointly announced research results on urological cancers (renal cell carcinoma, prostate cancer, urothelial carcinoma) with Project Assistant Professor Yuto Baba (at the time of the conference) and Senior Lecturer Nobuyuki Tanaka of the Department of Urology, Keio University School of Medicine, at "EAU 2026 (European Association of Urology)", one of the world's largest urology conferences held in London, UK in March 2026. The research, titled "Machine learning-based analysis of urinary exosomal microRNA: Detection, prognostic evaluation, and immune profiling of urological cancers," was presented.

■ About EAU 2026 (European Association of Urology)

This conference is one of the world's largest and most authoritative international conferences in the field of urology, where urologists, researchers, and experts from medical device and pharmaceutical companies from around the world gather to discuss the latest research findings and clinical data. It presents the latest research and clinical trial results across a wide range of urological diseases, including urological cancers, and shares important evidence that leads to updates in clinical guidelines and standard treatments, significantly contributing to the advancement of global urological medicine and the improvement of clinical practice.

• Period: Friday, March 13, 2026 – Monday, March 16, 2026

• Venue: London, UK

• Official Website: https://eaucongress.uroweb.org/

■ Key Research Findings

• Highly accurate detection model for urological cancers using a painless urine test

Aiming to develop a non-invasive testing system for urological cancers, we focused on microRNAs contained in urinary exosomes and performed Small RNA-seq analysis using urine samples from 380 urological cancer patients (renal cell carcinoma, prostate cancer, urothelial carcinoma, including 61.5-81.3% up to Stage 2) and 204 healthy adults. By constructing diagnostic models to distinguish renal cell carcinoma, prostate cancer, and urothelial carcinoma using machine learning algorithms, high diagnostic accuracy was achieved for all three in distinguishing them from healthy controls: renal cell carcinoma (AUC=0.92), prostate cancer (AUC=0.92), and urothelial carcinoma (AUC=0.96).

• Application to Prognostic Prediction

The prognostic model constructed using urinary microRNAs accurately predicted the presence or absence of recurrence for renal cell carcinoma (AUC=0.89), prostate cancer (AUC=0.78), and urothelial carcinoma (AUC=0.74) over a 48-month follow-up period. This model included microRNAs reported to be associated with tumor metastasis, immunosuppressive environments, and treatment response.

• Application to the Evaluation of Tumor Immune Status

To evaluate the tumor immune status, which is related to prognosis and treatment response to immune checkpoint inhibitors, we analyzed the association between urinary microRNAs and 102 cases of urothelial carcinoma for which immunohistochemical information on CD3, CD8, FOXP3, and PD-L1 in tumor tissue was available. As a result, the possibility of estimating the tumor's immune status with a certain accuracy (AUC 0.65-0.75) solely from urinary microRNA information was demonstrated.

■ Collaborative Research Overview

In this study, we developed a technology for the early detection of urological cancers (renal cell carcinoma, prostate cancer, urothelial carcinoma) by analyzing microRNAs in exosomes contained in urine. The constructed model demonstrated high screening accuracy for all cancer types and suggested its potential utility in prognosis, recurrence prediction, and evaluation of tumor immune status.

■ Terminology

• CD3: A protein that serves as a marker for T cells, a type of immune cell. A high number of CD3-positive cells around a tumor indicates a large accumulation of T cells, providing a clue as to the extent of the body's immune response against cancer.

• CD8: A marker for T cells that directly attack cancer cells. A higher number of CD8-positive cells within a tumor suggests that the body's immune system may be fighting the cancer.

• FOXP3: A marker for cells called "regulatory T cells" that suppress immune function. High FOXP3 expression may indicate an environment where cancer can more easily evade immune attack.

• PD-L1: A protein expressed on cancer cells and other cells, which plays a role in weakening the function of immune cells. It is considered an important indicator related to the effectiveness of immune checkpoint inhibitors.

■ About Craif

Craif (pronounced: Kuraifu) is a bio-AI startup founded in 2018, dedicated to the early detection of cancer. By fusing its proprietary analysis technology platform "NANO IP®︎ (NANO Intelligence Platform)," which precisely detects diverse biomarkers such as DNA and microRNA from bodily fluids including urine, with AI technology, Craif is developing innovative tests that enable ultra-early detection, early treatment, and early recovery from cancer. By broadly delivering the power of biotechnology and AI to society, the company promotes its vision of "realizing a society where people live out their full natural lifespan."

【Company Profile】

Company Name: Craif Inc. (Pronunciation: Kuraifu, English Name: Craif Inc.)

Representative: Ryuichi Onose, Representative Director CEO

Establishment: May 2018

Capital: 100 million yen (as of April 1, 2025)

Business: Research and development of next-generation tests for early detection of diseases and realization of personalized medicine, primarily in the oncology field, and provision of the urinary cancer test "MySignal series."

Headquarters: THE PORTAL iidabashi B1F, 8-30 Shinkogawacho, Shinjuku-ku, Tokyo

URL: https://craif.com/