A first-generation classification model for lung cancer has now been developed and is performing well. The model is based on gene expression data and is already available for use within Qlucore Insights. The initial model facilitates a division into two main groups, lung adenocarcinoma and lung squamous cell carcinoma, as well as three further groups based on whether the tumor profile corresponds to metastases from breast, colorectal, or kidney cell cancer. The model is designed to utilize tissue fixed with formalin (FFPE), which is a common sample form in the pathological workflow for solid tumors. It is clinically important to distinguish between primary lung cancer tumors and tumors that have metastasized to the lung in order to optimize next step investigations and treatment. Additional model generations are planned with improvements and expansion to support more forms of metastases. The goal is to further develop the solution into Qlucore Diagnostics, enabling CE marking and clinical use.
 
In late 2021, Qlucore, together with Heidelberg University Hospital, received a financial grant from the Eurostars Joint European Programme. This EU programme provides financial support to promote European innovation and competitiveness. The grant amounts to approximately SEK 5.1 million (500,000 euros). The new model is a result of this collaboration.
 
"The collaboration with the team from Heidelberg University Hospital is working very well, and we are pleased to see the first results from the project," says Carl-Johan Ivarsson, CEO of Qlucore.
 
The new classification model is based on Qlucore's knowledge and workflows for developing gene expression-based machine learning models (AI-based) and has been trained on hundreds of carefully selected tissue samples from lung cancer.
 
Precision diagnostics for cancer have advanced rapidly in recent years, driven by next-generation sequencing (NGS). Up until now, the focus has been on mutations and variants in the genetic code, which have been used for patient stratification and decisions on cancer treatment. However, there has been increasing interest in techniques based on measurements of gene activity levels (gene expression) as it provides additional opportunities to describe the type of treatment a patient should receive.
 
Qlucore Insights is intended for research and enables early testing and evaluation. Qlucore Insights is provided to hospitals, clinics, and laboratories via  Qlucore's European sales force.
 
Qlucore Diagnostics fills an important gap in the workflow for clinical precision diagnostics. With disease-specific machine learning models that enable classification and a feature supporting clinical decision-making, including user-friendly 3D visualizations, it becomes possible to predict patients' responses to treatment. Qlucore Diagnostics is the future solution for personalized cancer treatment, and the goal is to obtain CE marking.