New AI Tool Accurately Assesses Cancer Aggressiveness, Enhancing Treatment Options

Wed 16th Jul, 2025

In a significant advancement for oncology, researchers from Brazil and Poland have developed an innovative artificial intelligence (AI) tool capable of predicting the aggressiveness of certain cancer types. This groundbreaking study, published in the journal Cell Genomics, has the potential to transform cancer diagnosis and treatment methodologies.

The AI-driven model assesses tumor aggressiveness by analyzing specific proteins present in cancer cells. It generates a 'stemness index' ranging from 0 to 1, where 0 signifies low aggressiveness and 1 indicates high aggressiveness. A higher index correlates with increased drug resistance and a greater likelihood of cancer recurrence.

The stemness index reflects the extent to which tumor cells resemble pluripotent stem cells, which are capable of developing into various cell types within the body. As tumors progress, cancerous cells become increasingly dissimilar to their tissue of origin, exhibiting self-renewal capabilities and an undifferentiated state.

Utilizing data from the Clinical Proteomic Tumor Analysis Consortium (CPTAC), the researchers focused on 11 different cancer types, including breast, ovarian, lung, kidney, uterine, brain, head and neck, colon, and pancreatic cancers. They meticulously analyzed over 1,300 samples to develop the protein expression-based stemness index, known as PROTsi.

By integrating PROTsi with proteomic data derived from 207 pluripotent stem cell lines, the team identified specific proteins that contribute to the aggressive nature of certain tumors. These proteins may serve as promising targets for future therapies, thus enhancing personalized treatment approaches for cancer patients.

Many of the identified proteins are already recognized as targets for existing cancer drugs, highlighting the tool's potential to inform new therapeutic strategies. Researchers emphasized that their findings lay the groundwork for clinical applications, pointing to the importance of aligning tumor characteristics with treatment options.

The study's lead researchers expressed their commitment to advancing cancer treatment methodologies, noting that the tool not only aids in clinical development but also facilitates personalized therapy based on individual tumor profiles.

Recent statistics underscore the urgency of this research. The World Health Organization (WHO) reported that 40 individuals are diagnosed with cancer every minute globally. Cancer remains a leading cause of death, particularly among younger populations. A 2023 study highlighted a 79% increase in early-onset cancer cases among adults under 50 between 1990 and 2019, alongside a 28% rise in cancer-related fatalities.

In Brazil, estimates from the National Cancer Institute (INCA) suggest approximately 704,000 new cancer cases annually from 2023 to 2025, with non-melanoma skin cancer, breast cancer, and prostate cancer being the most prevalent. The PROTsi tool has shown consistent efficacy in distinguishing between tumor and non-tumor samples and is particularly effective in identifying high-grade tumors in various cancer types.

The researchers continue to refine their model, aiming to enhance predictive accuracy across diverse cancer types. Their work represents a promising leap forward in understanding tumor progression and resistance mechanisms, ultimately aspiring to improve treatment outcomes and patient quality of life.


More Quick Read Articles »