Pancreatic cancer is one of the most difficult cancers to treat. For this reason, every new discovery in this field attracts considerable attention, especially when the results appear particularly promising. In recent months, a study led by Mariano Barbacid’s group at the Spanish National Cancer Research Centre sparked significant discussion: a combination of three drugs appeared to induce complete and durable regression of pancreatic cancer in mouse models.
The study “A targeted combination therapy achieves effective pancreatic cancer regression and prevents tumor resistance”, published on December 11, 2025, in Proceedings of the National Academy of Sciences (PNAS), described a strategy based on three molecular targets connected to KRAS biology, one of the proteins most frequently altered in pancreatic cancer. The combination included three drugs: daraxonrasib, afatinib, and SD36. In the preclinical models used, including mouse models, this combination led to marked tumor regression and, in some cases, to the absence of resistance or relapse for prolonged periods.
Scientifically, the rationale was interesting. Pancreatic cancer is often driven by alterations in a specific protein known as KRAS, but targeting a single component of the tumor network is rarely enough: cancer cells can adapt, activate alternative pathways, and develop resistance. The idea behind the study was therefore to block multiple tumor survival pathways at the same time. In simple terms, instead of closing only one door to the tumor, the aim was to close several emergency exits at once.
The problem is that, a few months later, the study was retracted. The retraction did not directly concern the experimental demonstration of the therapy, but rather a fundamental aspect of scientific research: transparency. According to the PNAS statement, the article was withdrawn because of a relevant undeclared conflict of interest. In particular, some of the authors had financial interests linked to a biotechnology company connected to the commercial development of therapeutic strategies in this area.
This point is important: a conflict of interest does not automatically mean that a study is false. However, it does mean that readers, the journal, and the scientific community need to know about it. When a researcher or research group may benefit financially from the development of a therapy, this information must be clearly disclosed. Not because it makes the data useless, but because it allows readers to evaluate the work with the appropriate level of transparency.
The case became even more delicate because the study had been presented publicly with considerable media attention. Some messages surrounding the discovery suggested that a cure for pancreatic cancer might be close. This generated enthusiasm, but also very high expectations among patients and their families. According to El PaĂs, a Spanish newspaper, many people with pancreatic cancer reportedly contacted the research centre asking for information or access to the therapy, even though the treatment was still in a preclinical phase and was not available to patients.
This is where an essential distinction is needed: curing a tumor in mice does not mean curing it in human beings. Mouse models are indispensable tools in cancer research. They allow scientists to study biological mechanisms, test hypotheses, and understand whether a therapeutic strategy deserves further investigation. But a mouse is not a patient. The immune system is different, drug metabolism is different, clinical complexity is different, and human tumors are often much more heterogeneous and complex. Many therapies that work extraordinarily well in animal models later fail to demonstrate the same efficacy in clinical trials.
The Barbacid case therefore reminds us of two lessons. The first concerns science: even the most promising results must go through a long process involving independent validation, toxicology studies, phase I, II, and III clinical trials, and comparison with already available therapies. The second concerns communication: when we talk about cancer, every word matters. This is especially true when the audience is not made up only of researchers, but also of patients, families, and people looking for hope.
In conclusion, the retracted study does not necessarily erase the scientific interest of the proposed strategy. The combination of multiple molecular targets in pancreatic cancer remains an important and potentially promising area. However, this case shows how essential it is to distinguish between preclinical promise and demonstrated clinical benefit. Cancer research needs enthusiasm, but also rigor, transparency, and responsibility.
Â
