A Story of Discovery: KRAS Mouse Model Advances Pancreatic Cancer Research

  • Resize font
  • Print
  • Email
  • Facebook
  • Twitter
  • Google+
  • Pinterest

Key Points

  • Pancreatic ductal adenocarcinoma (PDAC), the most common type of pancreatic cancer, is usually detected at an advanced stage and has a very poor prognosis. Overall, only about 7 percent of patients survive more than 5 years after diagnosis.
  • Before NCI-funded scientists developed the KRAS mouse model to study pancreatic cancer, we knew very little about how to approach treatment of this deadly disease.
  • The best mouse models re-create disease in a way that mimics the disease in humans and help predict what is likely to happen in patients.
  • Thanks to NCI’s development of the KRAS mouse model, researchers around the world have an invaluable tool to study pancreatic cancer, develop ways to detect it earlier, and test new treatments.

Pathway of Discovery

Researchers are constantly seeking new tools to help reveal the complex genetic factors involved in the development of cancers. In this regard, experimental models offer a unique opportunity. A good experimental model allows researchers to investigate genes that are involved in cancer, to study how a specific cancer develops and spreads from one part of the body to another, and to test new treatments and diagnostic biomarkers. The scientific value of these experimental models is significant, as future research builds on the foundations of these models.

Genetically modified mouse model used in cancer research.

Progress against pancreatic cancer has depended on studies with mice genetically modified to mimic the human disease.

Four types of experimental models often used in cancer research include:

  • Cell cultures—using cancerous or normal cells grown in the laboratory and mutating a genes to mimic abnormal growth.
  • Cell line xenografts—implanting established human cancer cell lines under the skin of experimental mice.
  • Patient-derived xenografts (PDX)—implanting a patient’s cancer tissue under the skin of experimental mice.
  • Genetically engineered mouse models—genetically modifying mice to carry mutations in suspect cancer genes to develop particular cancer types. Many of these mouse models re-create disease in a way that closely mimics the disease in humans and helps predict what is likely to happen in patients.

Cancer research in one area can lead to important discovery in another area. The KRAS gene may cause cancer when it is mutated. Mutations of this gene have been found at high rates in pancreatic, lung, and colon cancer. The original mouse models in which cancer is driven by KRAS were developed by Erica Jackson, Ph.D, David Tuveson, M.D., Ph.D., and Tyler Jacks, Ph.D., of MIT. They focused on studying how KRAS gene mutations lead to lung cancer. The inititial research was supported by the NCI Mouse Models of Human Cancers Consortium (MMHCC). Over time, researchers also saw the potential in developing a KRAS mouse model to study pancreatic cancer. The MMHCC provided funding for development of the models and for using them for preclinical studies. The MMHCC has also been instrumental in making the models available to the pancreatic cancer research community.

Before NCI-funded scientists developed KRAS mouse models to study pancreatic cancer, very little was known about how to approach treatment of this deadly disease. This led researchers to explore the possibility that pancreatic cancer may be driven by genetic mutations and tissue-specific contexts.

Researchers hunt for human-relevant PDAC models. PDAC is the most common type of pancreatic cancer. Because of the insidious nature of this disease, it is rarely diagnosed during the early stages when it is most treatable. Diagnosis at the late stage of the disease has a very poor prognosis. Overall, only about 7 percent of patients survive more than 5 years after diagnosis—leaving patients, their families, and doctors with little hope and still searching for answers. However, based on exciting new findings, researchers now believe that mutations in at least two key genes, KRAS and p53, are the major drivers behind pancreatic cancer, with mutant KRAS being essential. In fact, studies have shown that more than 9 in 10 PDAC patients have a mutation in the KRAS gene.

This story of discovery starts with earlier laboratory studies, where researchers used cell cultures and cell line xenograft models to study human pancreatic tumors in mice. Curiously, they found that drugs that were effective treatments in the laboratory did not translate to successful treatment in PDAC patients.

Consequently, researchers continued to work to develop mouse models that more closely reflected the disease process as it occurs in humans. They developed models that could accurately represent how PDAC tumors develop and metastasize, leading to cancers that reflect genetic aberrations and biological properties found in patient cancers. This led researchers down a path to create the KRAS mouse model, a development that marks a turning point in pancreatic cancer research, as researchers could now better study pancreatic cancer, develop ways to detect it earlier, and test new treatments for this deadly disease.

Engineering a mouse model that mirrors human pancreatic cancer. Research is a methodical stepwise process, and over time researchers worked diligently to refine the mouse model. In 2003, Dr. Tuveson and Sunil Hingorani, M.D., Ph.D. and colleagues at the Abramson Cancer Center of the University of Pennsylvania succcessfully developed a mouse model that mirrors human pancreatic cancer. The model reflects pancreatic cancer from its precancerous beginning to the advanced stages by expressing the mutated KRAS gene in pancreatic cells. These genetically engineered mice first develop precancerous growths that then develop into tumors that are very similar to tumors found in patients with PDAC. Although this model was a big step in the right direction, the progression to PDAC occurred with low frequency, hampering its usefulness in studying the advanced cancers found in humans. This drove researchers to develop a more robust mouse model by engineering additional gene mutations found in advanced human PDAC.

In 2005, these same researchers developed a highly improved double-mutation mouse model that has mutations in both the KRAS and p53 genes. These genes are mutated in 90 percent and 75 percent, respectively, of human pancreatic cancer cases. All mice with these genetically engineered KRAS and p53 mutations developed advanced pancreatic cancer, and many harbored cancers that spread to the liver, lung, adrenal glands, diaphragm, peritoneum, and nervous system. These are the same organs to which pancreatic cancer typically spreads in humans. Also, the degree and rate of disease progression more closely followed the course of disease that researchers expected to see in patients. While there are many known mutations that can cause cancer, this exceptional research showed that double mutation of KRAS and p53 can reproduce the entire progression of PDAC in mice. This model is known as the KPC model of pancreatic cancer. In parallel, researchers at the Dana Farber Cancer Institute developed another mouse model that also mutated KRAS, disrupted p53 function, and  modified a third cancer gene. While both models have been used extensively to study PDAC, the KPC model has been most frequently employed for testing therapeutics with the potential for treating patients.

Using the KPC model, researchers also made a key discovery that has expanded the search for effective treatments—that the mutations that directly alter the growth rate in pancreas cells that result in PDAC can also affect the unmutated cells surrounding it, such as connective tissue, blood vessels, and immune cells. The different types of normal cells that surround a tumor are often referred to as the tumor microenvironment. These normal cells, molecules, and blood vessels that surround and feed a tumor cell can have a profound effect on the progression and treatment of the disease, but this tumor microenvironment cannot be replicated in cell culture and PDX models.

Illustration of tumor microenvironment showing how a thick barrier of connective tissue surrounds the tumor, collapses blood vessels, and blocks drug delivery.

Thanks to NCI support, researchers discovered the role of the tumor microenvironment in blocking drug delivery, charting a new course for pancreatic cancer treatment.

Eureka! KPC model opens doors to new, more effective treatments. In 2009, a collaborative study led by Kenneth P. Olive, Ph.D., a postdoctoral fellow in Dr. Tuveson’s lab, used the KRAS-p53 mutant KRAS mouse to show that fewer than normal blood vessels were functional in pancreatic tumors. This is similar to what is seen in human PDAC and may provide one explanation as to why patients respond poorly to chemotherapy.

When pancreatic tumors were treated with the standard chemotherapy of gemcitabine, there was very little response. Further examination of these tumors in the KRAS mouse showed that the blood vessels in these tumors were surrounded by connective tissue that created a thick barrier around the tumors. Initially, researchers thought this was a good thing because it would prevent the cancer from spreading. They actually found that pressure from the thick barrier caused the blood vessels to close, making it impossible for the treatment (gemcitabine) to reach the tumors and preventing it from being effective. In this case, the seemingly normal cells in the tumor microenvironment were keeping the treatment from reaching the tumors. Although further research has shown that the impact of the microenvironment on PDAC treatment is much more complex than originally thought, work in the KPC model helps us to appreciate the complexity of PDAC and has opened doors to improving patient care.

Enhancing Cancer Research

Researchers around the world now enthusiastically embrace the KPC double-gene mutation mouse models to study the processes underlying pancreatic cancer, develop early detection strategies, and test drug treatments. Because they re-create the progression of PDAC from precancerous cells through metastasis, these models allow researchers to explore early events and methods for early detection. Studies are also underway to identify early biomarkers of pancreatic cancer, which could help identify PDAC in people who have no signs or symptoms of the disease. Early detection is key in a better prognosis for patients with this difficult to treat cancer.

The development of this KRAS mouse model was a huge step forward in pancreatic cancer research by allowing the study of the progression of abnormal cell growth, the effects of the microenvironment in which the tumor cells reside, and the genetic processes underlying PDAC in patients. This model also helped bring to light a conundrum researchers faced.  As Dr. Hingorani noted, “One of the things that’s always been confusing is that chemotherapy for this cancer in particular has been extremely ineffective, and we’ve never quite understood why, noted Dr. Hingorani. Only 5 to 10 percent of patients with pancreatic cancer respond to gemcitabine, the standard treatment for this disease. As a result of NCI-funded research, we now  understand why prior treatments failed and have the opportunity to develop alternative treatments.

Turning Discovery Into Health

Researchers, now equipped with this new understanding, have sought to solve how to best unlock the microenvironment to ensure treatment delivery and limit the spread of cancer. In the first study Drs. Olive and Tuveson used the KPC mouse model to identify an important cell to cell signaling pathway involved in creating this barrier. They then sought out dual-pronged therapeutic approaches: targeting the microenvironment that creates this barrier together with the conventional treatment. In this study, KPC mice were treated with gemcitabine (commonly used to treat patients) and an inhibitor of the signaling pathway to reduce the microenvironment barrier to gemcitabine. This increased the number of functional tumor blood vessels, increased the survival of tumor-bearing mice, and reduced the spread of tumors to the liver.

Since this study, numerous investigators have tested similar approaches to “open” the microenvionment barrier to the influx of drugs. Paolo Provenzano, Ph.D., in the Hingorani lab at the Fred Hutchinson Cancer Research Center, and colleagues tried a different, dual-pronged approach. Dr.  Provenzano treated mice with gemcitabine and an enzyme to directly break down the barrier surrounding the tumors. At the end of the study, 100 percent of tumors treated with the combination of gemcitabine plus the enzyme had shrunk in size. Additionally, mice survived nearly 36 days longer after treatment with the gemcitabine/enzyme combination compared with gemcitabine alone; about a 39 percent improvement. This shows that, at least in pancreatic cancer, it is important not only to treat the cancerous cells themselves but also to target the surrounding microenvironment to improve the prognosis in mice, and potentially in human patients.

Much of this research remains in the discovery phase prior to testing in cancer patients. However, Drs. Olive and Tuveson did conduct an initial trial in human patients. Although this study did not yield the expected positive results, as with the early days of organ transplant development, this outcome emphasizes the importance of comparative research to ensure that basic research translates to better clinical outcomes. Ongoing rigorous testing of treatments and treatment strategies and regimes in mouse cancer models is often required.

This prompted NCI’s Center for Cancer Research to launch the Center for Advanced Preclinical Studies (CAPR), led by Terry Van Dyke, Ph.D. Established in 2008, this initiative is committed to accelerating the use of genetic cancer models to guide effective therapeutic and diagnostic studies in humans.

Research to Practice: NCI’s Role

Infographic that says more than 30% of all human cancers are driven by mutations of the RAS genes. It explains that 95% of pancreatic cancers have a KRAS mutation; 45% of colorectal cancers have a KRAS mutation; 35% of Lung cancers have a KRAS mutation; 30% of Acute Myeloid Leukemias have an NRAS mutation; 15% of Melanomas have a NRAS mutation; and 15% of Bladder cancers have an HRAS mutation. Dr. Frank McCormick, the RAS National Program Advisor, is quoted as saying "RAS oncogenes are the worst oncogenes."

Research in understanding pancreatic cancer has made great strides, but still there is no cure. With the support of NCI, researchers now have an invaluable tool that has shifted the way they think about treating pancreatic cancer. Instead of focusing only on the drug and its cancer cell target, researchers are also including the tumor microenvironment along with drug delivery in their treatment strategies, as well as identifying other molecules affected by KRAS and p53 that could be the targets for new drug treatments. Efforts are also ongoing to find better biomarkers to detect PDAC in its early stages, develop imaging tools to detect the smallest tumors, and test gemcitabine and other drugs in combination with novel treatments. Detecting the disease earlier increases the chances of more successful treatment.

More studies are needed, as future research holds even greater promise for the treatment of pancreatic cancer. The KRAS mouse model continues to be an invaluable tool, not only in pancreatic cancers but also in a number of other cancer types. The model is widely used by reseachers to ask key questions about shortcomings of today’s cancer treatment approaches, develop effective strategies to overcome them, and improve patient outcomes.

NCI continues to build on this discovery by recently establishing The RAS Program based at NCI’s Frederick National Laboratory for Cancer Research (FNLCR). The FNLCR is a collaborative effort that works to further understanding for all RAS-driven cancers, including pancreatic cancer. More than 30 percent of all human cancers are caused by mutations of RAS genes, increasing to 95 percent for pancreatic cancer. 

Patients around the world are counting on NCI’s leadership through The RAS Program to solve the enigma of RAS-driven cancers. With a focus on bridging basic research with clinical practice, the FNLCR is a unique resource that partners with others through a “hub and spokes” approach to develop the next generation of tests and treatments to benefit families affected by cancer.

Key Takeaway

NCI’s extraordinary development of a KRAS mouse model gives researchers around the world an invaluable tool to study pancreatic cancer, develop methods for earlier detection, and to test new treatments for this deadly disease.

Selected Resources

Aguirre AJ, Bardeesy N, Sinha M, et al. Activated Kras and Ink4a/Arf deficiency cooperate to produce metastatic pancreatic ductal adenocarcinoma. Genes Dev. 2003 ;17(24):3112-3126.
[PubMed Central]

Hingorani SR, Petricoin EF III, Maitra A, et al. Preinvasive and invasive ductal pancreatic cancer and its early detection in the mouse. Cancer Cell. 2003;4(6):437-450. [PubMed Abstract]

Hingorani SR, Wang L, Multani A, et al. Trp53R172H and KrasG12D cooperate to promote chromosomal instability and widely metastatic pancreatic ductal adenocarcinoma in mice. Cancer Cell. 2005;7(5):469-483. [PubMed Abstract]

National Cancer Institute. Pancreatic cancer. http://cancer.gov/types/pancreatic. 2014.

National Cancer Institute. Surveillance, Epidemiology, and End Results Program. SEER Stat Fact Sheets: Pancreas Cancer. http://seer.cancer.gov/statfacts/html/pancreas.html. 2014.

O’Hagan RC, Heyer J. KRAS mouse model: modeling cancer harboring KRAS mutations. Genes Cancer. 2011;2(3):335-343. [PubMed Central]

Olive KP, Jacobetz MA, Davidson CJ, et al. Inhibition of Hedgehog signaling enhances delivery of chemotherapy in a mouse model of pancreatic cancer. Science. 2009;324(5933):1457-1461. [PubMed Central]  

Provenzano PP, Cuevas C, Chang AE, et al. Enzymatic targeting of the stroma ablates physical barriers to treatment of pancreatic ductal adenocarcinoma. Cancer Cell. 2012 March;21(3):418-29. [PubMed Central]

Van Dyke T. Finding the tumor copycat: approximating a human cancer. Nat Med. 2010;16(9):976-977. [PubMed Central]

  • Posted: October 31, 2014