Clinical Limitations in Thromboprophylaxis for DLBCL Patients

In patients with Diffuse Large B-cell Lymphoma (DLBCL), Venous Thromboembolism (VTE) is a life-threatening complication with an incidence rate of approximately 16% within two years. Currently, in clinical practice, prophylactic treatment is administered using Low Molecular Weight Heparin (LMWH) such as enoxaparin or Direct Oral Anticoagulants (DOACs) such as apixaban. However, due to the lack of clear criteria for determining which patients should initiate prophylactic anticoagulation, unnecessary bleeding side effects and cases of failed prophylaxis have frequently occurred. This is a significant unmet medical need in the area of Cancer-Associated Thrombosis (CAT) management, as it reduces the quality of life for cancer patients and increases additional healthcare costs.

Clear Limitations of Existing Risk Prediction Models

To identify these risks, existing tools such as the Khorana Score, Throly Score, and Model IX (published in 2020) have been primarily used. However, these existing models were developed using patient populations including relatively low-risk, indolent lymphoma patients, which has led to a critical limitation when applied to the high-risk DLBCL patient population: reduced predictive performance. These models are based solely on static variables such as blood counts and general condition at the time of diagnosis, and do not reflect the dynamic changes experienced by patients. This has limited clinicians' ability to accurately determine whether or not to initiate anticoagulation, and the need for a DLBCL-specific model has been consistently raised throughout the industry.

PREDIC-TO Study Completes Large-Scale Data Validation

To address these issues, the research team from the Fondazione IRCCS San Gerardo dei Tintori in Italy and the University of Milano Bicocca conducted and completed the PREDIC-TO observational study (NCT06694779) on a large cohort of 1,504 DLBCL patients. The research team compared the performance of the existing three major prediction models based on the patients' actual treatment data, and comprehensively tracked and collected dynamic risk factors (Dynamic Risk Factors) that appear during the treatment process, such as the insertion of a Central Venous Catheter (CVC) and the administration of chemotherapy. This study goes beyond simple statistical comparisons and utilizes large-scale Real-World Data (RWD) to maximize the statistical precision of the prediction model and lay the foundation for deriving new, dedicated algorithms.

Personalized Anticoagulation Therapy and Changes in Future Treatment Protocols

The completion of this observational study will be an important milestone in revolutionizing the standard treatment protocol for DLBCL patient management in the future. The developed DLBCL-specific model will contribute to the establishment of guidelines for personalized thromboprophylaxis tailored to the individual treatment pathways of patients, and will significantly improve patient safety by preventing unnecessary misuse of anticoagulants. This will play a key role in increasing adherence to anticancer treatment and reducing side effects, in line with the continued growth of the global DLBCL therapeutics market. Ultimately, it is expected to lead to revisions in standard clinical guidelines by organizations such as the National Comprehensive Cancer Network (NCCN) and the American Society of Hematology (ASH), thereby realizing the value of preventive medicine.