An MRI-based measure of tumor variation
Researchers at The Hong Kong Polytechnic University have developed an imaging biomarker intended to help predict whether some patients with breast cancer are likely to respond to a targeted drug treatment before that treatment begins.
The measure, called the Tumour Heterogeneity Index, or THI, draws on radiomics: computational analysis of medical images that extracts patterns not readily apparent through conventional visual review. The team derived the index from dynamic contrast-enhanced magnetic resonance imaging, known as DCE-MRI.
The premise is straightforward but potentially important. Two breast tumors that appear similar in size can have different internal imaging characteristics. Those differences may reflect biological variation within and between tumors. If imaging can capture variation associated with treatment response, clinicians may gain useful information without first requiring an invasive tissue-based assessment specifically for that purpose.
The researchers say their index enables noninvasive assessment of a patient's likely response to AKT inhibitors. Their study focused on people with high-risk, early-stage breast cancer receiving the investigational AKT inhibitor MK2206 in addition to neoadjuvant chemotherapy, the treatment given before surgery.
Why treatment prediction matters
Choosing cancer therapy is rarely a simple matter of identifying a drug that can work. Treatments can bring adverse effects, consume valuable time and affect the sequence of care leading to surgery. A tool that improves the ability to identify patients most likely to benefit could support more tailored treatment decisions at an earlier stage.
In the reported analysis, patients with a high THI had an estimated pathological complete response rate of 56.1%. Pathological complete response is a postsurgical finding indicating that no invasive cancer is detected in the tissue assessed after preoperative treatment. The source material describes this rate as significantly higher than the rate observed among patients with a low THI.
That result does not mean the index by itself establishes which treatment an individual should receive. It does, however, suggest that image-derived indicators of tumor heterogeneity may add clinically relevant information beyond familiar features such as tumor size and location.
For patients, the long-term value of this approach would be a more informed discussion of likely benefit before beginning therapy. For clinicians, it could offer another input when weighing the possible gains of adding a targeted drug to chemotherapy. And for health systems, better matching of treatment to likely response could eventually reduce unnecessary exposure to side effects.
Using an existing imaging method differently
DCE-MRI is already widely used in cancer diagnosis and treatment planning. It uses contrast-enhanced MRI scans to provide information about tumors. The PolyU team extended that imaging data into a predictive analysis by applying advanced radiomics techniques.
Radiomics typically converts features within scans into quantitative data that can be analyzed for associations with diagnosis, prognosis or treatment response. In this case, the research team used subtle image differences to construct the THI. The source notes that these differences can be difficult to identify through conventional observation.
The goal is not to replace clinical expertise or standard pathology. Rather, the work points toward a future in which routine imaging may contribute more directly to individualized treatment planning. A noninvasive predictor could be especially useful before therapy, when care teams are deciding how aggressively to treat a tumor and which combination of treatments to consider.
The treatment under study
The study examined response to MK2206, an investigational AKT inhibitor. AKT is part of the AKT/mTOR/PI3K signaling pathway, which the researchers describe as important in promoting tumor growth and survival. That pathway has long been considered a therapeutic target in cancer treatment.
The reported finding concerns the combination of MK2206 with neoadjuvant chemotherapy in the specific study population, not all breast cancers or all AKT-targeting drugs. That distinction matters. Biomarkers often perform differently across patient groups, disease stages, drug regimens and clinical settings.
The source also says MK2206 had previously been shown to improve pathological complete response in the context examined by the study. THI is therefore being investigated as a way to distinguish patients who may be more likely to receive that benefit.
What needs to happen next
The result is promising, but it should be understood as research rather than a new standard of care. The supplied report does not establish that THI is ready for broad clinical deployment, nor does it provide the validation details needed to determine how reliably the approach would generalize across hospitals, MRI systems or patient populations.
Further work would need to test the index in additional cohorts and determine how it performs alongside established clinical and biological measures. Researchers and clinicians would also need to assess practical questions: whether the analysis can be standardized, how quickly results can be returned, and whether use of the measure improves real-world treatment decisions and patient outcomes.
Even so, the study illustrates an important direction for precision oncology. Medical images contain more information than can be captured by looking only at a tumor's dimensions. If carefully validated, tools that quantify that information could help move treatment selection toward a more individualized model—one based not only on where a tumor is and how large it is, but also on characteristics associated with how it may respond.
This article is based on reporting by Medical Xpress. Read the original article.
Originally published on medicalxpress.com
