Innovative AI Algorithm Enhances Precision in Chemotherapy Dosing for Colorectal Cancer Patients

Chemotherapy remains a cornerstone in cancer treatment, effectively targeting and halting the progression of various malignancies. Traditionally, dosing chemotherapy relies on the body surface area (BSA), calculated from a patient's weight and height. However, this approach often results in inaccuracies, with approximately 60% of patients receiving either too high or too low a dose. Such miscalculations can lead to severe side effects, treatment discontinuation, and reduced survival rates.

Recognizing the limitations of conventional dosing methods, a collaborative team comprising clinicians, scientists, industry partners, and consumer groups from institutions like the University of Melbourne and WEHI has developed an advanced AI-driven tool called PredicTx. This innovative algorithm utilizes CT scans and detailed body composition features to tailor chemotherapy doses uniquely for each patient.

Colorectal cancer (CRC) is the second leading cause of cancer-related death in Australia, with over 1.9 million new cases globally each year. About half of these patients require chemotherapy, which significantly improves their survival chances. Currently, chemotherapy dosing based on BSA does not account for individual differences in body composition such as fat and muscle mass, leading to potential overdosing or underdosing.

Incorrect dosing poses serious risks, including neuropathy, immunosuppression, cardiac and respiratory complications, and gastrointestinal issues. These adverse effects often compel patients to halt treatment prematurely, adversely affecting outcomes, and in some cases, can be fatal.

To address these challenges, PredicTx aims to refine dosing accuracy by analyzing specific body features from CT scans. The goal is to enable healthcare providers, especially in rural or resource-limited settings, to deliver precise, personalized treatment, thereby reducing toxicity and improving the chances of completing optimal chemotherapy regimens.

By improving dosing accuracy, PredicTx not only aspires to enhance individual patient outcomes but also to increase overall cure rates and quality of life for cancer patients. The project exemplifies successful interdisciplinary collaboration and marks a significant step towards personalized oncology care.

While initially focused on colorectal cancer, the potential applications of PredicTx could extend to other cancer types, ensuring patients everywhere receive the most appropriate treatment dosage at the right time. This advancement underscores the importance of integrating AI technology into clinical practice to transform cancer treatment paradigms.