Craif Announces Joint Research Results on Early Detection and Prognosis Prediction of Lung Cancer Using Urinary MicroRNA at AACR Annual Meeting 2026
Craif Inc. announced joint research results at the AACR Annual Meeting 2026 on the early detection and prognosis prediction of lung cancer using urinary microRNA. This research, conducted in collaboration with Tokyo Jikei University School of Medicine and Higashi Osaka Medical Center, demonstrated high-precision diagnosis and recurrence risk prediction through a non-invasive urine test, highlighting the potential for a 'Single-assay concept' in cancer care.
📋 Article Processing Timeline
- 📰 Published: April 30, 2026 at 19:00
- 🔍 Collected: April 30, 2026 at 10:31
- 🤖 AI Analyzed: April 30, 2026 at 21:28 (10h 57m after Collected)
Bio-AI startup Craif Inc. (Location: Shinjuku-ku, Tokyo; Representative Director CEO: Ryuichi Onose; hereinafter "Craif") presented joint research results on lung cancer with Associate Professor Yu Fujita, Director of the Department of Next-Generation Drug Discovery Research, Research Center for Medical Science, Tokyo Jikei University School of Medicine, and Dr. Takashi Nojiri, Director of Respiratory Surgery, Higashi Osaka Medical Center, at the "AACR Annual Meeting 2026 (American Association for Cancer Research Annual Meeting)" held in San Diego, USA, in April 2026. This research demonstrated the possibility of highly accurate early detection of lung cancer through a urine test that does not require blood sampling or hospital visits, and the potential to predict postoperative recurrence risk from urine collected before surgery.
■ About AACR Annual Meeting 2026 (American Association for Cancer Research Annual Meeting)
This conference is one of the world's largest and most authoritative academic gatherings for cancer research, bringing together doctors, researchers, and specialists from medical device and pharmaceutical companies from around the world to discuss the latest cancer research findings and clinical data. It significantly contributes to the advancement of global cancer research and the improvement of clinical practice.
Presentation Date: Sunday, April 19, 2026
Location: San Diego, California, USA
Official Website: https://www.aacr.org/meeting/aacr-annual-meeting-2026/
■ Key Research Findings
Highly Accurate Early Detection of Lung Cancer via Urine Test Without Blood Sampling
Aiming to develop a non-invasive testing system for lung cancer, the study focused on microRNAs contained in urinary exosomes. Small RNA-seq analysis was performed on 278 lung cancer patients and 213 non-cancer controls (total 491 subjects), and a diagnostic model for identifying lung cancer was constructed using a machine learning algorithm. The diagnostic accuracy (AUC) in an independent test set was 0.941 (95%CI: 0.899–0.983), with a sensitivity of 88.2% (95%CI: 73.4–95.3%) and specificity of 87.0% (95%CI: 75.6–93.6%) for early-stage cancer (Stage 0/I). Furthermore, age-matching analysis and multivariate analysis confirmed that the results were not affected by age, gender, BMI, smoking history, etc.
Prediction of Postoperative Recurrence Risk from Pre-Surgical Urine
For 76 patients with surgically resected Stage I-II lung cancer, a Cox regression model was constructed using pre-surgical urinary microRNA expression data. A prognostic prediction panel consisting of three types of microRNAs (hsa-miR-181a-5p, hsa-miR-185-5p, hsa-miR-934) was developed. A significant stratification was confirmed between the high-risk and low-risk recurrence groups, with a hazard ratio of 8.3 (95%CI: 1.9–37.0) for recurrence-free survival (ROC AUC at 3 years: 0.796). This demonstrated the potential to assess recurrence risk from urine before surgery.
■ Joint Research Overview
This study is a multi-center case-control study using urine samples collected from four facilities in Japan. For 278 lung cancer patients and 213 non-cancer controls, an early detection model was constructed and validated by combining Small RNA-seq analysis of microRNAs contained in urinary extracellular vesicles (EVs) and machine learning. Additionally, a prognostic microRNA panel was constructed using pre-surgical urinary microRNA expression data, demonstrating the potential to stratify recurrence risk. These findings indicate the possibility of achieving "early detection of lung cancer" and "recurrence risk assessment" integrally through urine collection that does not require blood sampling or hospital visits (Single-assay concept).
■ Glossary
Extracellular Vesicles (Exosomes): Small, sac-like particles secreted by cells, used for intercellular communication within the body. Extracellular vesicles contain substances such as microRNAs and are attracting attention as important biomarkers for cancer diagnosis.
MicroRNA: Very small RNA molecules within cells that regulate gene function. In cancer cells, the types and amounts of microRNAs change, making them potentially useful for early disease detection and prognosis prediction.
Machine Learning: A type of AI (Artificial Intelligence) technology that learns patterns from large amounts of data to make future predictions or classifications. In this research, it was used to highly accurately determine the presence or absence of disease using microRNA data for cancer diagnosis.
AUC (Area Under the Curve): An indicator of diagnostic accuracy, ranging from 0 to 1. A value closer to 1 indicates higher diagnostic performance.
Sensitivity and Specificity: Sensitivity indicates the proportion of people with the disease who test positive, while specificity indicates the proportion of people without the disease who test negative.
Hazard Ratio: An indicator showing the ratio of the risk of an event (such as recurrence) occurring between two groups.
■ About AACR Annual Meeting 2026 (American Association for Cancer Research Annual Meeting)
This conference is one of the world's largest and most authoritative academic gatherings for cancer research, bringing together doctors, researchers, and specialists from medical device and pharmaceutical companies from around the world to discuss the latest cancer research findings and clinical data. It significantly contributes to the advancement of global cancer research and the improvement of clinical practice.
Presentation Date: Sunday, April 19, 2026
Location: San Diego, California, USA
Official Website: https://www.aacr.org/meeting/aacr-annual-meeting-2026/
■ Key Research Findings
Highly Accurate Early Detection of Lung Cancer via Urine Test Without Blood Sampling
Aiming to develop a non-invasive testing system for lung cancer, the study focused on microRNAs contained in urinary exosomes. Small RNA-seq analysis was performed on 278 lung cancer patients and 213 non-cancer controls (total 491 subjects), and a diagnostic model for identifying lung cancer was constructed using a machine learning algorithm. The diagnostic accuracy (AUC) in an independent test set was 0.941 (95%CI: 0.899–0.983), with a sensitivity of 88.2% (95%CI: 73.4–95.3%) and specificity of 87.0% (95%CI: 75.6–93.6%) for early-stage cancer (Stage 0/I). Furthermore, age-matching analysis and multivariate analysis confirmed that the results were not affected by age, gender, BMI, smoking history, etc.
Prediction of Postoperative Recurrence Risk from Pre-Surgical Urine
For 76 patients with surgically resected Stage I-II lung cancer, a Cox regression model was constructed using pre-surgical urinary microRNA expression data. A prognostic prediction panel consisting of three types of microRNAs (hsa-miR-181a-5p, hsa-miR-185-5p, hsa-miR-934) was developed. A significant stratification was confirmed between the high-risk and low-risk recurrence groups, with a hazard ratio of 8.3 (95%CI: 1.9–37.0) for recurrence-free survival (ROC AUC at 3 years: 0.796). This demonstrated the potential to assess recurrence risk from urine before surgery.
■ Joint Research Overview
This study is a multi-center case-control study using urine samples collected from four facilities in Japan. For 278 lung cancer patients and 213 non-cancer controls, an early detection model was constructed and validated by combining Small RNA-seq analysis of microRNAs contained in urinary extracellular vesicles (EVs) and machine learning. Additionally, a prognostic microRNA panel was constructed using pre-surgical urinary microRNA expression data, demonstrating the potential to stratify recurrence risk. These findings indicate the possibility of achieving "early detection of lung cancer" and "recurrence risk assessment" integrally through urine collection that does not require blood sampling or hospital visits (Single-assay concept).
■ Glossary
Extracellular Vesicles (Exosomes): Small, sac-like particles secreted by cells, used for intercellular communication within the body. Extracellular vesicles contain substances such as microRNAs and are attracting attention as important biomarkers for cancer diagnosis.
MicroRNA: Very small RNA molecules within cells that regulate gene function. In cancer cells, the types and amounts of microRNAs change, making them potentially useful for early disease detection and prognosis prediction.
Machine Learning: A type of AI (Artificial Intelligence) technology that learns patterns from large amounts of data to make future predictions or classifications. In this research, it was used to highly accurately determine the presence or absence of disease using microRNA data for cancer diagnosis.
AUC (Area Under the Curve): An indicator of diagnostic accuracy, ranging from 0 to 1. A value closer to 1 indicates higher diagnostic performance.
Sensitivity and Specificity: Sensitivity indicates the proportion of people with the disease who test positive, while specificity indicates the proportion of people without the disease who test negative.
Hazard Ratio: An indicator showing the ratio of the risk of an event (such as recurrence) occurring between two groups.