Complete Tumour Regression in Pancreatic Cancer Mouse Models: What Mariano Barbacid’s Latest Study Shows

30th Jan 2026

Complete Tumour Regression in Pancreatic Cancer Mouse Models: What Mariano Barbacid’s Latest Study Shows

Pancreatic ductal adenocarcinoma (PDAC) remains one of the most lethal human cancers, with five-year survival rates still in the single digits. Despite decades of research, meaningful therapeutic breakthroughs have been rare, particularly for tumours driven by oncogenic KRAS mutations, which are found in over 90% of cases and underpin aggressive tumour biology and resistance to therapy. [1]

In a recent preclinical study, Mariano Barbacid and his team at the Spanish National Cancer Research Centre (CNIO) reported a result long considered extremely difficult in pancreatic cancer research: complete regression of established pancreatic tumours in mouse models using a targeted triple-therapy approach. [2]

While these findings do not represent a clinical cure, they provide evidence that even highly aggressive, KRAS-driven cancers can be rendered vulnerable under specific molecular and signalling conditions.

Why pancreatic cancer has been so resistant to treatment

Pancreatic cancer’s poor prognosis is due to multiple factors, including a dense, desmoplastic tumour microenvironment that can limit drug penetration, early metastatic dissemination, and molecular drivers such as KRAS that are challenging to target directly. These features make conventional chemotherapy and targeted agents less effective in many patients. [1]

What Barbacid’s team demonstrated in mice

Barbacid’s group employed a triple combination strategy designed to inhibit multiple signalling nodes in the KRAS-driven network in mouse models of pancreatic ductal adenocarcinoma. In these preclinical models, the approach was associated with:

  • Complete regression of established PDAC tumours
  • Durable responses without evident drug resistance
  • Well-tolerated treatment in mice (as reported by the authors/institution)

The study was published in the Proceedings of the National Academy of Sciences (PNAS). [2] [3]

Why this result is scientifically significant

The significance of this work lies not only in the outcome but in the experimental design and validation:

  • Use of genetically engineered mouse models designed to better reflect key aspects of human disease
  • Multi-node blockade of KRAS-associated signalling pathways rather than reliance on single-agent targeting
  • Quantitative validation using molecular and histological readouts

This methodological rigour strengthens confidence that the observed tumour regressions reflect genuine biological dependencies rather than model-specific artefacts. [3]

Common markers used to validate tumour regression in PDAC models

Rigorous validation of tumour response in preclinical pancreatic cancer studies typically includes established markers such as:

  • KRAS pathway activation markers (e.g. ERK, phospho-ERK)
  • Proliferation markers (e.g. Ki-67)
  • Apoptosis markers (e.g. cleaved caspase-3)
  • Stromal markers (e.g. α-SMA)
  • Immune cell markers (e.g. CD3, F4/80)

Accurate detection of these markers by validated immunohistochemistry (IHC) or immunofluorescence (IF) is essential for interpreting regression outcomes, especially in fibrotic murine pancreatic tissue.

What the study does not yet demonstrate

Despite the promising results in mice, key translational uncertainties remain:

  • Human clinical efficacy has not yet been established
  • Human PDAC may adapt via compensatory pathways not fully captured in mouse systems
  • Safety, dosing, and feasibility in patients remain to be determined

Translation from preclinical success to human benefit has historically been challenging in PDAC, warranting cautious interpretation of even striking mouse results. [2]

Implications for preclinical cancer research

This study underscores the importance of model relevance and reagent validation in translational cancer research. Poorly validated antibodies or unreliable marker performance can lead to misleading conclusions—particularly when evaluating complex endpoints such as tumour regression and microenvironment remodelling.

Supporting reproducible cancer research

At St John’s Laboratory, we provide rigorously validated antibodies for immunohistochemistry and immunofluorescence in mouse tumour models, designed to support high-confidence data generation and enhance reproducibility in preclinical oncology research.