New imaging tool predicts breast cancer treatment success before therapy begins
Researchers at The Hong Kong Polytechnic University have created a noninvasive imaging biomarker called the Tumour Heterogeneity Index (THI) that uses MRI data to forecast how well breast cancer patients will respond to AKT inhibitors. Patients with a high THI showed a 56.1% rate of pathological complete response, significantly better than those with low THI scores. This approach could enable earlier, more personalized treatment decisions and reduce exposure to ineffective therapies.
Researchers at The Hong Kong Polytechnic University (PolyU), led by Professor Cai Jing and Dr. Teng Xinzhi from the Department of Health Technology and Informatics, have developed the Tumour Heterogeneity Index (THI), a noninvasive imaging biomarker derived from dynamic contrast-enhanced MRI (DCE-MRI) data. The tool was tested in a study involving patients with high-risk, early-stage breast cancer who were receiving neoadjuvant chemotherapy combined with the AKT inhibitor MK2206. Results showed that patients with a high THI had a 56.1% pathological complete response (pCR) rate—significantly higher than those with low THI scores—demonstrating its predictive power for treatment success. The study, published in *Breast Cancer Research*, highlights THI’s ability to outperform traditional immunohistochemistry (IHC)-based testing in identifying suitable candidates for AKT inhibitor therapy and assessing the tumor microenvironment, including immune cell infiltration and vascular patterns. This advancement addresses a longstanding challenge: despite MK2206 showing promise in the I-SPY2 trial, its effectiveness varies widely across aggressive subtypes like triple-negative breast cancer, making pre-treatment prediction essential.
The development of THI builds on prior limitations in cancer treatment personalization, where biopsy-based methods offered only localized, potentially biased tissue samples and could not capture tumor-wide heterogeneity. By leveraging DCE-MRI—already standard
The introduction of THI could significantly shift how oncologists approach treatment selection for breast cancer patients, particularly those with aggressive subtypes where therapy response is highly variable. By enabling noninvasive prediction of drug efficacy before treatment begins, it may reduce the use of ineffective therapies—such as AKT inhibitors in non-responders—thereby minimizing patient exposure to side effects and lowering healthcare costs. This could also alleviate pressure on genetic sequencing, which remains expensive and inaccessible in many settings, potentially democratizing access to precision medicine. A comparable precedent is the adoption of radiogenomics in lung cancer, where imaging features were linked to specific gene mutations, guiding therapy choices without invasive testing. If validated broadly, THI may influence pharmaceutical development by helping companies identify responsive patient populations earlier in clinical trials, improving trial efficiency