An Epigenetic Approach to Treating Pancreatic Cancer
In This Article
- Models developed by Massachusetts General Hospital investigator Andrew Liss, PhD, are offering insight into mechanisms related to pancreatic cancer lethality
- Researchers are revealing epigenetic mechanisms responsible for allowing cancer cells to optimize their environment to evade therapy, adapt and survive
- Their findings offer new insight into how this cancer evolves and to advance the development of therapeutic strategies to prolong patient survival
Massachusetts General Hospital researchers are engaged in identifying the mechanisms associated with pancreatic cancer—one of the deadliest cancers in the world—to drive the development of effective therapeutic options and improve survival rates. Although accounting for only 3% of all cancers, pancreatic cancer is associated with approximately 7% of cancer-related deaths in the U.S.
Subscribe to the latest updates from Digestive Health Advances in Motion
"Despite advances in our understanding of this disease, its sustained ability to evade diagnosis and treatment remains a major focus of continued research," says Andrew Liss, PhD, a principal investigator in the Division of Gastrointestinal & Oncologic Surgery at Mass General and assistant professor of Surgery at Harvard Medical School. "We are determining how the cancer controls its surroundings and gaining insight into how this can be exploited to improve treatment outcomes."
A Singularly Lethal Form of Cancer
Pancreatic ductal adenocarcinoma (PDAC) accounts for 90% of pancreatic malignancies and has a five-year overall survival rate of approximately 12%. The option for curative therapy is surgical resection, but up to 90% of PDAC patients are diagnosed at an advanced, nonresectable stage. Furthermore, approximately 80% of patients who undergo resection recur with localized or distant disease. The fact that this can occur following successful resection of the primary tumor indicates that the cancer had already spread at the time of surgery.
Screening methods and markers for early diagnosis remain elusive despite advances in the understanding of PDAC onset and progression. Among the complicating factors, the location of the pancreas and proximity to several major blood vessels hinders both diagnostic imaging and surgical interventions. Additionally, patients present with a lack of symptoms or indications that may appear relatively benign.
"The ambiguity of the symptoms and aggressiveness of the cancer toward early metastasis severely limit the therapeutic options in many cases by the time of diagnosis," says Dr. Liss.
Establishing a Protective Barrier
Although many cancers have a stromal fraction that acts as both a scaffold and boundary for the tumor, pancreas cells demonstrate a high degree of plasticity. The result is an injury response capable of rapidly modifying connective tissue within the organ. PDAC formation exacerbates this activity by releasing growth factors capable of transforming host cells into cancer-associated fibroblasts (CAFs).
The crosstalk between cancer cells and CAFs regulates an abnormal collagen accumulation and the formation of a dense fibrotic stroma that acts as a physical barrier. These characteristics also preclude angiogenesis, which hinders drug delivery to the tumor and supports an immunosuppressive environment. The communication between PDAC and CAFs is bidirectional, with CAFs also capable of influencing PDAC cell biology.
The impact of this bidirectional influence results in highly complicated responses to interventions. "We previously showed that co-culture with CAFs or only media from CAF cultures results in PDAC cells acquiring a more aggressive phenotype that supports rapid proliferation and migration," says Dr. Liss. On the other hand, efforts to selectively eliminate CAFs therapeutically can also increase the aggressiveness of the cancer and result in accelerated disease progression.
Findings by Dr. Liss and colleagues identified a family of proteins involved in altering chromosome structure to regulate gene transcription. Inhibiting this activity altered the expression of genes in both cancer cells and stromal cells responsible for the pro-tumor phenotype of the stroma, suggesting an epigenetic component to regulating tumor-stroma crosstalk.
Adapt, Evade, and Survive
Pancreatic cancer cells are broadly divided into classical and basal subtypes; the latter correlates with a more aggressive phenotype, shorter overall survival, and increased resistance to chemotherapy. Although histologically indistinguishable, the two subtypes demonstrate distinct gene expression profiles and activated signaling pathways.
Dr. Liss describes a highly dynamic relationship between the two subtypes. "Both subtypes appear capable of adapting their respective transcription profile to escape the effects of therapy. This includes classical-to-basal transformation and acquisition of chemoresistance in response to treatment both in vitro and in vivo."
Furthermore, experimental outcomes using patient-derived xenograft (PDX) models and PDX-derived cell lines from the Pancreatic Tumor Bank at Mass General revealed contrasts between subtypes according to growth environment. Specifically, 80% of the PDX models presented as classical-subtype tumors, whereas 84% of the cell lines presented a basal-like subtype, with the differences defined by changes in the expression of classical genes. Interestingly, the transplant of cell lines into mice to generate xenograft models re-engages the classical-subtype gene-expression program.
"This plasticity may be a contributing factor in the failure of some therapeutic targets identified in vitro to exhibit similar efficacy in in vivo models and patients," Dr. Liss explains. "Although the mechanisms are elusive, we believe that epigenetics might play a major role in these processes through changes in the global architecture of the genome as an adaptive response."
An Epigenetic Approach to Attacking PDAC
To fit the human genome into a cell nucleus, DNA is wrapped around proteins called histones, forming a DNA–protein complex referred to as chromatin. Chromatin structure can be altered through the addition or removal of chemical groups to histones, as well as by interactions between these groups and regulatory molecules.
This remodeling process results in either an "open" or "closed" structural conformation that allows or restricts access to DNA regions for transcription. As one method of epigenetic regulation, this represents a possible explanation for the plasticity observed in PDAC cells.
"Our models support mechanisms that involve global regulation of cellular phenotype through rapid changes in the expression of certain subsets of genes," advises Dr. Liss.
Although data show that targeting a particular epigenetic regulator can "turn off" gene expression supporting the classical subtype in PDAC cells, Dr. Liss stresses that there are inevitably multiple factors at play, including those involved in regulating the basal subtype.
"The goal is to identify the regulatory pathways that might render the overall tumor microenvironment more susceptible to therapy," he says. "A deeper understanding of the pathways that control the crosstalk between cancer cells and the stroma can potentially pave the way to breakthroughs in how we attack the ability of PDAC to evolve and survive."
Shifting the Treatment Paradigm
Evidence indicates that the PDAC tumor stroma comprises a heterogeneous cell population that tends to be very patient-specific. "We have identified at least four distinct types of stroma that evolve in the PDX models," explains Dr. Liss. "Despite using animals with exactly the same genetic background, xenografts generated in these mice using tumor cells taken from different patients frequently display stroma with completely different characteristics."
This is a significant complicating factor for clinical trials that attempt to assess a particular therapy in a large cohort of patients. "Interpreting outcomes of clinical trials in this area might require a different approach to translating the findings," suggests Dr. Liss. "Given what our models indicate, if the therapy worked in 10% of those patients, it might be worth identifying the factors associated with their response to potentially target other patients with a similar tumor profile."
An Ideal Setting to Address Difficult Problems
The Liss Laboratory is also responsible for operating the Pancreatic Tumor Bank, which houses normal and tumor tissue, and serum and plasma samples from over 3,800 patients. This includes a living tumor bank of xenograft tumors from over 90 patients along with established cell lines from some of these tumors. The bank exists largely through the foresight of Andrew Warshaw, MD, surgeon-in-chief emeritus of the Mass General Department of Surgery.
"As a bench scientist, I am acutely aware of how advantageous it is to have access to this bank and these types of samples," says Dr. Liss. "As a result, I am always open to sharing these resources with people elsewhere doing solid science and looking for access to relevant models."
Dr. Liss also emphasizes how the environment at Mass General fosters close collaboration with colleagues across disciplines. In particular, David T. Ting, MD, contributed to findings related to possible epigenetic regulation of tumor-stroma crosstalk, and work with Ralph Weissleder, MD, PhD, recently revealed tumor cell-derived extracellular vesicles as a possible diagnostic biomarker for early-stage PDAC.
"We have a deep appreciation for the complexity of this problem and are committed to making discoveries that will offer patients a fighting chance at beating the odds associated with this disease."
Refer a patient to the Division of Gastrointestinal & Oncologic Surgery
Learn about the Division of Gastrointestinal & Oncologic Surgery