New AI Model PROTsi Predicts Aggressiveness of Tumors Using Protein Markers

Researchers from Poznan University of Medical Sciences in Poland and the University of São Paulo in Brazil have unveiled a groundbreaking artificial intelligence (AI) model named PROTsi, designed to predict the aggressiveness of various tumors by analyzing protein expression levels. This innovative approach, detailed in a study published in the journal Cell Genomics on July 25, 2025, aims to enhance cancer diagnosis and treatment personalization amid rising global cancer incidence rates.

The PROTsi model, developed using data from the Clinical Proteomic Tumour Analysis Consortium (CPTAC), utilizes a stemness index that ranges from zero to one. A higher score indicates a tumor's closer resemblance to pluripotent stem cells, which correlates with increased aggressiveness and potential drug resistance. The research team analyzed over 1,300 samples from 11 different cancer types, including breast, ovarian, lung, kidney, uterine, brain, head and neck, colon, and pancreatic cancers. By comparing these tumor samples to 207 pluripotent stem cell samples, the researchers identified key proteins that drive tumor aggressiveness and could serve as targets for future therapies.

Professor Tathiane Malta from the Multiomics and Molecular Oncology Laboratory at the Ribeirão Preto Medical School of the University of São Paulo highlighted the significance of their findings. "Many of these proteins are already targets of drugs available on the market for cancer patients and other diseases. They can be tested in future studies based on this identification," she explained. The PROTsi model demonstrated robust predictive capabilities during validation, effectively distinguishing between tumor and non-tumor samples and setting a foundation for more accurate cancer evaluations.

"We sought to build a model that can be applied to any cancer, but we found that it works better for some than for others. We’re making a data source available for future work," added Malta. The model showed exceptional performance particularly in specific cancer types, such as uterine, head, neck, pancreatic, and pediatric brain tumors.

The implications of PROTsi extend beyond mere classification; they pave the way for potential therapeutic advancements. By identifying proteins associated with stemness, the model could contribute to the development of more effective therapies tailored to specific tumor characteristics. Renan Santos Simões, co-first author of the study, emphasized the collaborative nature of this research and its potential impact on patient care. "Science advances slowly, carefully, and is built by many hands. It’s gratifying to realize that we’re contributing to this process. That’s what motivates us: knowing that what we do today can make a real difference for patients, improving treatments and quality of life," he stated.

As the research team at the University of São Paulo continues to refine and enhance the PROTsi model, they aim to further translate molecular data into actionable clinical insights. This innovation reflects a significant step forward in the integration of AI into oncology, promising to make cancer treatment more precise, targeted, and effective.

In conclusion, the PROTsi model stands as a testament to the potential of artificial intelligence in transforming cancer research and treatment. As researchers build upon this foundation, the prospect of more personalized and effective cancer therapies becomes increasingly attainable, offering hope to patients worldwide. The exploration of AI in oncology is still in its early stages; however, tools like PROTsi are indicative of a future where treatment can be tailored to the unique molecular fingerprints of individual tumors, thereby improving patient outcomes on a global scale.