The advent of molecular biology and molecular profiling in clinical medicine has transformed our understanding of childhood leukemia. As a result, we are now empowered to shift away from the classification of hematologic malignancy based on microscopic appearance towards a new paradigm of diagnosis and treatment focused on specific molecular mechanisms of pathogenesis, or alterations detected in both leukemia cells and cells representing an individual’s heritable constitution. The heritable, constitutional or germ-line contribution to the development of childhood leukemia is a phenomenon increasingly recognized in childhood cancer. Studies to date have revealed the constitutional basis for cancers in subtypes of leukemia, however there is a clear missing heritability fraction given a high frequency of families without an identifiable genetic etiology. Given a lack of awareness and an incomplete germline evaluation, pediatric oncologists are unable to take avail of complete information pertaining to a child’s predisposition to developing leukemia when planning therapeutics and guiding. Knowledge of germ line alterations may direct patient care, and enable genetic counseling for patients and their families. My research focuses on identifying the heritable underpinnings of childhood leukemia.
Cells within a tumor must acquire nutrients from their environment and convert these nutrients into the cellular components necessary to support continued growth. This set of processes is broadly referred to as tumor metabolism. We are interested in understanding how tumor metabolism is distinct from the metabolism of normal tissues with the hope of identifying those genes or pathways upon which cancer cells are particularly dependent for survival. Recently, we have become fascinated by how tumors utilize one key metabolite, the amino acid serine. We found that the production of serine is activated in several cancer types, including breast cancer of the basal type, a particularly difficult to treat form of breast cancer. Cancer cells use this serine for various purposes, including the production of DNA. In this proposal we will evaluate the anti-cancer effect of inhibiting utilization of serine in a mouse model of basal breast cancer that recapitulates many aspects of the human disease. Furthermore, we will use a novel technology permitting editing of the cancer genome to determine whether perturbing serine utilization uncovers additional dependencies which can be the target of future anti-cancer therapies.
Cancer cells express mutated proteins that are distinct from the proteins in non-cancerous cells, known as “tumor-specific antigens.” Over a century ago, scientists reasoned that our immune system (T cells) should be able to recognize these mutated proteins as “foreign” and eliminate cancer cells. While we find tumor-specific T cells within tumors, these T cells are not functional, allowing cancers to grow unimpeded. Our goal is to understand why tumor-specific T cells are dysfunctional and develop strategies to reprogram these tumor-specific T cells to fight cancer.
Using genetic cancer mouse models, we found that during tumor development, tumor-specific T cells become dysfunctional because the genes and pathways needed for normal T cell function are dysregulated. All cells in our body, including T cells, contain two layers of information encoding each cell’s characteristics. The first layer is the genome and consists of the DNA nucleotide sequence, the second layer is the epigenome and consists of chemical modifications to DNA or to the scaffold proteins associated with DNA. The genome and the epigenome together determine a T cell’s properties. Because functional and non-functional T cells in our model have identical genomes, tumor-induced loss of function must result from epigenomic changes. We propose to define the epigenome modifications that render tumor-specific T cells non-functional and test strategies to reverse this “code of dysfunction” so that we can reprogram tumor-specific T cells for human cancer therapy.
Funded in Collaboration with Stand Up To Cancer (SU2C)
Preclinical and clinical studies have informed the development of increasingly effective cancer therapies that can lead to dramatic clinical responses. However, in the majority of cases, patients subsequently develop therapeutic resistance. Although recent studies by our labs and by others have elucidated mechanisms of resistance, the complex series of genetic and epigenetic events that drive therapeutic resistance have not been well delineated, there are few therapeutic options to prevent resistance. We have chosen acute myeloid leukemia (AML) to investigate the dynamics of therapeutic response and resistance for several important reasons. First, there is abundant evidence that targeted therapies (tyrosine kinase inhibitors) can induce substantive clinical responses, including complete remissions, in patients with genotypically defined disease subsets. Moreover, combination chemotherapy can induce a high rate of complete response similar to that observed with molecularly targeted therapies. Second, our team proposes to perform genomic and transcriptional/epigenomic studies of serially obtained clinical isolates from patients before therapy, at the time of maximal response, and at therapeutic relapse. Our clinical sites have established a robust infrastructure to obtain clinical samples at the different time points. This sampling approach, coupled with state-of-the-art genomic, transcriptional, and functional studies, will address questions that are central to the fields of cancer biology, modeling, and cancer therapeutics, and – most importantly will allow us to test models of the evolution of drug resistance and novel therapeutic approaches that can then be rapidly translated to the clinical context.
Tumors consist not only of cancer cells, but also stromal and immune cells that constitute the tumor microenvironment (TME). Cancer cells can take on dramatically different properties based on the microenvironment. The clinical impact of the TME is only becoming appreciated in recent years. In many different cancer types, including breast cancer (BC), tumors with higher stromal fractions portend worse clinical outcomes. In contrast, tumors infiltrated by CD8 T cells have better clinical outcomes. Hence, tumors behave differently based on the collective behavior of the microenvironment. We will leverage biotechnology advances in sequencing single cells to better understand the important determinants of the coevolution between the adaptive immune response and the tumor. By tracking the spatial geometry of cells in tumor samples we hope to better understand the TME and ultimately determine which genetic factors can be best exploited for therapeutic intervention.
Funded in Collaboration With Stand Up To Cancer (SU2C)
The so called targeted therapies are effective in tumors that strictly depend on a given protein or cellular signaling (the target) for growth and survival. Hyperactivation of the PI3K pathway is frequent in breast cancers and its pharmacologic inhibition showed clinical responses. However, these molecules alone cannot elicit a durable inhibition of tumor growth because the tumor can adapt and compensate the inhibition of the pathway.
Thus, targeting these compensatory mechanisms in combination with the PI3K pathway would in principle lead to stronger and more durable antitumor activity.
In this proposal we aim to validate in the laboratory theoretical predictions of successful drug combinations. These predictions are obtained from mathematical models developed from what is currently known about the perturbations of the PI3K/AKT signaling network in response to different inhibitors of the pathway. In addition, we plan to test therapeutic combinations based on genomic analyses from tissue samples of breast cancer patients treated with PI3K inhibitors.
Taken together, our results should provide the rationale to test novel and more effective therapeutic options for patients with hyperactivation of the PI3K pathway.
Funded in Collaboration With Stand Up To Cancer (SU2C)
The study encompasses multiple directions. First, genome of cancer cells acquires mutations at a higher rate compared to benign cells. Some of these novel mutations affect proteins synthesized within the cell, and these modified proteins (tumor neoantigens) may interact with immune system. We identify these novel neoantigens and study their interaction with immune cells in the tumor microenvironment. The other direction is quantification of non-coding RNAs, in particular, some repeat RNAs, expressed by cancer cells. We focus on the mechanisms of expression of these RNAs, their immunogenic properties and their interaction with tumor microenvironment. Understanding these topics would open the door towards unleashing immune response against pancreatic tumors.
Funded in Collaboration with Stand Up To Cancer (SU2C)
Pancreatic ductal adenocarcinoma (PDAC) is a frequent cause of cancer death in the United States; it currently is the fourth most common cause of cancer death and is expected to become the second most common cause of cancer death within the next five years. Unlike virtually all other major cancers, pancreas cancer is both increasing in incidence and has shown essentially no improvement in five year survival over the past two decades. The exceptional lethality of pancreas cancer is multifactorial, resulting from an intrinsically aggressive biology, lack of effective means of early detection, and poor responsiveness to systemic chemotherapy. Clearly novel approaches to this disease are needed.
Although there have been anecdotal reports of responses to immune-based therapies in pancreas cancer, activation of cellular immunity using checkpoint inhibitors, vaccine strategies and transfer of genetically modified T cells has not been shown to be generally effective. We have assembled a team of physicians, cancer immunobiologists, computational biophysicists, and engineers to better understand the unique immunological microenvironment of pancreatic cancer, develop the technologies needed to take advantage of therapeutic vulnerabilities, and to form a multi-institutional clinical consortium to readily implement these strategies to help change the course of this deadly disease.
The development of drug resistance is a major challenge in cancer therapy. Mantle cell lymphoma (MCL), like many other human cancers, remains incurable mainly due to acquired drug resistance. Ibrutinib received approval from the U.S. Food and Drug Administration (FDA) for treatment of MCL in November 2013, and has shown promise in treating many MCL patients. Unfortunately, about one-third of patients are resistant to ibrutinib and many patients become resistant after the initial response. The tumors grow faster than before and there are no effective therapeutic options. The underlying mechanism is unknown. By longitudinal genomic and RNA sequencing analysis of both MCL tumors and healthy tissue we have identified a relapse-specific genetic mutation, C481S, in Burton’s Tyrosine Kinase (BTK), which ibrutinib specifically targets. This is the first identified mutation specific to MCL patients who relapse from ibrutinib after a durable response. However, this BTK mutation is not found in MCL patients who do not respond to ibrutinib or become resistance after transient response, suggesting two patterns of ibrutnib resistance.in MCL.
Ibrutinib resistance appears to correlate to an increased activation of a number of other molecular mechanisms known to contribute to lymphoma growth, among them the protein CDK4, and signaling along the PI3K-AKT pathway. We further discovered that targeting CDK4 with palbociclib (PD 0332991), a selective CDK4-inhibitor, made the MCL tumor cells sensitive to inhibitors of PI3K regardless of BTK mutation. These findings suggest that a combination therapy of palbociclib and PI3K inhibitor may overcome ibrutinib resistance. To address this exciting possibility, we will 1) investigate the mechanism by which inhibition of CDK4 sensitizes lymphoma cells to PI3K inhibitor, focusing on the disruption of glucose metabolism based on preliminary evidence; and 2) determine the clinical efficacy of overriding ibrutinib resistance by dual targeting of CDK4 and PI3K in a Phase I clinical trial in MCL, and discover genes and pathways that discriminate sensitivity and resistance by integrative longitudinal genomic and RNA sequencing analysis of serial biopsies before, during and after therapy.
This novel study not only suggests new approaches for treating MCL but also has implications for treatment of other B cell lymphomas, such as chronic lymphocytic leukemia and a diverse group of non-Hodgkin lymphomas. It is also exciting because CDK4 is a new kind of drug target; it controls the cell cycle, which is a central cancer pathway. As such, targeting CDK4 is not just important for MCL but for many forms of cancer. For example, when combined with letrozole, palbociclib more than doubled the progression free survival of metastatic breast cancer patients. Since PI3K is commonly mutated or over-activated in human cancers, including breast cancer, the palbociclib and PI3K combination represents a novel therapy for other human cancers as well.
