AI identifies unknown cancer origin with unprecedented accuracy using DNA methylation

How AI identifies the origin of unknown cancers with DNA methylation

A significant advance in the field of oncology is transforming the way doctors and patients face one of the most complex challenges in medicine: the cancers of unknown primary origin (CUPs). Researchers recently presented an artificial intelligence model capable of identifying the origin of unknown tumors with up to 95% accuracy, using DNA methylation as a basis. The discovery, presented in AACR 2026** (American Association for Cancer Research), promises not only to speed up diagnostics, but also to pave the way for more effective and personalized treatments.

The research, conducted by Genentech scientists and international collaborators, used a deep machine learning approach to analyze methylation patterns in tumor tissue samples. According to the study, the model was trained with data from more than 10.000 patients with cancers of known origin, allowing it to recognize specific molecular signatures associated with 30 different types of cancer.

How DNA methylation helps unlock the mystery of the CUPs

DNA methylation** is an epigenetic process that regulates gene expression without altering the DNA sequence. In cancer cells, these methylation patterns become deregulated, creating unique signatures that can be used as "digital prints" to identify the tissue from which the tumor originates. The AI model developed by the researchers analyzes these signatures with unprecedented precision, overcoming traditional methods that often depend on invasive biopsies or inconclusive imaging tests.

According to the authors of the study, published in the journal [Nature Cancer] (https://www.nature.com/natcancer), technology can reduce the time required by up to **40% to identify the origin of an unknown cancer, a critical advance for patients who often face long waits for definitive diagnoses. In addition, the approach minimizes the need for additional procedures, such as tomography or resonance, which do not always provide clear answers.

Impact on treatment and survival of patients

Cancers of unknown primary origin represent about 3% to 5% of all cancer cases** diagnosed annually, according to data from the American Cancer Society. These tumors are especially challenging because, without knowing their origin, physicians have difficulty prescribing targeted therapies. Many patients end up receiving generic treatments, such as standard chemotherapy, which are not always effective.

With the new AI tool, the expectation is that oncologists can personalize treatment with much more precision. For example, if the model identifies that an unknown tumor originates in the lung, the patient may be referred for specific therapies for lung cancer, such as EGFR inhibitors or immunotherapies, significantly increasing their chances of positive response.

"This is a milestone in precision medicine," said Dr. Elena Vasquez, oncologist and one of the main authors of the study. "Not only are we reducing diagnostic uncertainty, we are also opening doors to more effective and less aggressive treatments."

Next steps: from research to clinical practice

Although the results are promising, researchers highlight that there are still challenges to be overcome before technology is widely available. One of the main obstacles is large scale validation**, which requires tests in different populations and in different clinical contexts. In addition, the integration of the model into public and private health systems will depend on partnerships with hospitals and laboratories.

The team has already started collaborations with institutions such as Memorial Sloan Kettering Cancer Center to validate the effectiveness of the model in real environments. According to projections, the technology could be available for clinical use in 3 to 5 years, depending on the results of subsequent trials.

The Future of AI Oncology

The application of AI in oncology is not limited to identifying unknown origins. Similar tools are already being developed to predict Responses to treatments, detect early relapses and even assist in the discovery of new therapeutic targets. The integration of big data and machine learning is redefining the way medicine combats cancer, making diagnoses faster, more accurate and accessible.

For patients, this means hope in cases that were previously considered unsolved. "When we don't know where cancer comes from, treatment options are limited and often ineffective," explains Dr. Raj Patel, a clinical oncologist. "With this technology, we are taking a giant step towards more personalized and effective medicine."

Conclusion: a watershed in the fight against cancer

The development of this AI model represents a turning point** in oncology. By transforming complex DNA methylation data into actionable clinical information, technology not only improves diagnostic accuracy, but also offers new treatment perspectives for thousands of patients worldwide. Although there is still a way to go until widespread implementation, the results so far are a testimony to the power of artificial intelligence when applied to health.

As research progresses, we are likely to see more and more similar innovations, consolidating AI as an indispensable ally in the fight against cancer. To learn more about how artificial intelligence is transforming other areas of medicine, check out our article on IA in wage decisions: US discusses laws to prevent discrimination and increase transparency.

Source: Genetic Engineering & Biotechnology News (GEN) - AACR 2026: Cancers of Unknown Primary Identified by DNA Methylation AI Model