Carolyn A. Felix, M.D.
Associate Professor, Dept of Pediatrics

Cancer Biology Program

Address

Abramson Research Center, Room 902B
3615 Civic Center Blvd
Philadelphia, PA 19104-4318

Office tel.: 215 590-2831
Lab tel.: 215 590-4540
Fax: 215 590-3770
E-mail: felix@email.chop.edu

Link(s)

Dr. Felix's Children's Hospital of Philadelphia Page

EDUCATION

Boston College: BS (Biology/Chemistry), 1977.

Boston University School of Medicine: MD (Medicine), 1981.

RESEARCH INTERESTS

• Infant and chemotherapy-related leukemias with MLL (Mixed Lineage Leukemia; Myeloid Lymphoid Leukemia) gene translocations.

Key words: leukemia, infant, DNA topoisomerase II, epipodophyllotoxin, MLL gene, translocation.

Search PubMed for articles

DESCRIPTION OF RESEARCH

The research activities of my laboratory are directed at solving the problems of leukemias in infants and leukemias caused by chemotherapeutic DNA topoisomerase II inhibitors. My research group has shown that these leukemias have similar chromosomal translocations of the MLL gene at chromosme band 11q23. While most children with leukemia can be cured; the treatment options for patients with MLL-rearranged leukemias, which include intensive chemotherapy and hematopoietic stem cell transplantation, are associated with excessive toxicity and a poor prognosis. The therapies have lagged far behind the scientific advances of the new genomics era. New risk factors and insights on the DNA damage leading to the translocations have emerged from the work of my laboratory, the mission of which is to build bridges between basic science and the bedside that will accelerate targeted prevention and more efficacious, less toxic, targeted new treatments for these high-risk patients.

The research of my laboratory has pioneered new panhandle PCR methodology for the detection of MLL translocations. The MLL gene can be fused with one of many different partner genes to form translocations. My laboratory has discovered a large proportion of the known partner genes of MLL. In studies of AML in infant twins we have shown that the translocations and, therefore the DNA damage leading to the translocations, occur in the prenatal period. In the chemotherapy-related cases we have shown that the translocations can be present early during the primary cancer treatment. Work in my laboratory using an in vitro assay has suggested that DNA topoisomerase II, an essential protein in all cells, may be involved in the chromosomal breakage that leads to translocations. These studies are important because the chemotherapy drugs associated with this form of leukemia are DNA topoisomerase II inhibitors and, in leukemia in infants, classical epidemiology has suggested associations of maternal prenatal consumption of certain dietary DNA topoisomerase II inhibitors with an increased risk. Currently we are developing an experimental model that exploits human bone marrow progenitor cells and novel DNA array technology in order to mimic the damage to the MLL gene and localize the DNA topoisomerase cleavage complexes that are formed in vivo. In other work, we have determined that genetic variations in the detoxifying pathways NQO1 and CYP3A4, respectively, confer susceptibility to infant and treatment-related leukemias with MLL translocations. In the infant leukemias, the prenatal period when the translocation must occur holds possibilities for prevention if there are relevant exposures in high-risk individuals. In the treatment-related leukemias, the identification of high-risk individuals could lead to alterations in chemotherapy dose or schedule and preventative strategies for the future. In addition to experiments that are focused on prevention, other efforts in my laboratory are directed at developing better treatments. We are using gene expression profiling and we are planning profiling experiments via large-scale proteomics to understand the influence of the partner genes of MLL on disease biology and identify new drug targets. We have also shown that leukemias in infants have increased Bcl-2 expression, and other current research involves testing an anti-sense compound directed at this central anti-apoptotic protein, which confers chemotherapy resistance. Therefore, the research of my laboratory is relevant to new gene discovery and to diagnosis, prognosis and monitoring in patients with leukemia, and may lead to strategies for treatment and prevention.

RECENT PUBLICATIONS

Raffini, L.J., Slater, D.J., Rappaport, E.F., Lo Nigro, L., Cheung, N.-K.V., Biegel, J.A., Nowell, P.C., Lange, B.J., Felix, C.A.: Panhandle and reverse panhandle PCR enable cloning of der(11) and der(other) genomic breakpoint junctions of MLL translocations and identify complex translocation of MLL, AF-4 and CDK6. Proc Natl Acad Sci USA, 99:4568-4573, 2002.

Leonard, D.G.B., Travis, L.B., Addya, K., Dores, G.M., Holowaty, E.J., Bergfeldt,K., Malkin, D., Kohler, B.A., Lynch, C.F., Wiklund, T., Stovall, M., Hall, P., Pukkala, E., Slater , D.J., Felix, C.A.: p53 mutations in leukemia and myelodysplastic syndrome after ovarian cancer. Clin Cancer Res, 8:973-985, 2002.

Slater, D.J., Hilgenfeld, E., Rappaport, E.F., Shah,N., Meek, R.G., Williams, W.R., Lovett, B.D., Osheroff, N., Autar, R.S., Ried, T., Felix, C.A.: MLL-SEPTIN6 fusion recurs in novel translocation of chromosomes 3, X, and 11 in infant acute myelomonocytic leukemia and in t(X;11) in infant acute myeloid leukemia, and MLL genomic breakpoint in complex MLL-SEPTIN6 rearrangement is a DNA topoisomerase II cleavage site. Oncogene 21: 4706-4714, 2002.

Smith, M.T., Wang, Y., Skibola, C., Slater, D.J., Lo Nigro, L., Lange, B.J., Nowell, P.C., Felix, C.A.: Low NAD(P)H:Quinone Oxidoreductase (NQO1) activity is associated with increased risk of leukemia harboring MLL gene rearrangements in infants and young children. Blood, 100:4590-4593, 2002

Whitmarsh, R., Saginario, C., Zhuo, Y., Hilgenfeld, E., Rappaport, E.F., Megonigal, M.D., Carroll, M., Liu, M., Osheroff, N., Cheung,N.-K.V., Ried, T., Knutsen, T., Blair, I.A., Felix, C.A.: Reciprocal DNA topoisomerase II cleavage events at 5'-TATTA-3' sequence in MLL and AF-9 create homologous single-stranded overhangs that anneal to form der(11) and der(9) genomic breakpoint junctions in treatment-related AML without further processing (In press).

Lab

ROTATION PROJECTS FOR 2006-2007

• MLL translocation breakpoint cloning MLL fuses with many different partner genes to form translocations. My laboratory developed four different panhandle PCR approaches to isolate MLL genomic breakpoint junctions and MLL chimeric transcripts, and identified many of the 40 known partner genes of MLL. We found hCDCrel, a member of the SEPTIN family, as a new partner gene of MLL in identical, non-constitutional t(11;22)(q23;q11.2) translocations in acute myeloid leukemia (AML) of infant twins. This work is pertinent to the natural history of AML in infants because it showed that the MLL translocation was an in utero event. We also identified another SEPTIN family member SEPTIN6, CDK6, ALKALINE CERAMIDASE, MYO1F and RPS3, as new partner genes of MLL in leukemias in infants. In DNA topoisomerase II inhibitor-related AML, we found that MLL can fuse with itself to form tandem duplications, and identified the new partner genes GAS7(Growth Arrest Specific 7) and GMPS (Guanosine 5’-monophosphate synthetase). Another finding was that MLL translocations could be detected early during treatment. This project involves the molecular cloning of additional MLL genomic breakpoint junctions in infant and treatment-related leukemias via panhandle PCR approaches and is expected to lead to new partner gene discovery. NAD(P)H: quinone oxidoreductase (NQO1) detoxifies benzoquinone and other naturally-occurring simple quinones. In a recent study we showed that an inactivating NQO1 polymorphism is a risk factor in the infant cases, validating a prior study in a British population. Another aspect of this work on infants involves determining whether the inactivating NQO1 polymorphism is associated with specific DNA damage patterns at the genomic sequence level. The epipodophyllotoxins are the anticancer drugs most often associated with leukemia with MLL translocations. We found that the genotype of cytochrome P450 3A4 (CYP3A4) modulates the risk. CYP3A4 converts epipodophyllotoxin to a catechol metabolite, which is readily oxidized to a quinone, both potential genotoxins. In the chemotherapy related leukemias we are interested in determining whether the genomic breakpoint junction sequences hold similar clues to the etiology of the damage.
• Leukemia Etiology Since chemotherapy that disrupts DNA topoisomerase II is associated with MLL translocations, and epidemiological associations of maternal prenatal dietary DNA topoisomerase II inhibitor exposures are observed in infant cases, research on the role of DNA topoisomerase II in producing translocations is central to our work. In vitro assays of DNA topoisomerase II cleavage identified relationships of functional DNA topoisomerase II cleavage sites in MLL and partner genes with the translocation breakpoints. This led to a model in which DNA topoisomerase II mediates chromosomal breakage that results in translocations, DNA topoisomerase II inhibitors contribute to the breakage, and translocations form when the breakage is repaired. The CYP3A4 genotype-leukemia association ushered in the finding that etoposide metabolites can damage MLL and partner genes in a DNA topoisomerase II-dependent manner. Similar studies on naturally-occurring DNA topoisomerase II inhibitors and compounds in the NQO1 pathway are in progress. This project involves utilization of DNA topoisomerase II cleavage assays to map DNA topoisomerase II cleavage complexes induced by chemotherapeutic and naturally-occurring DNA topoisomerase II inhibitors in MLL and in its partner genes in vitro and development of DNA oligonucleotide arrays to locate DNA-DNA topoisomerase II complexes induced by DNA topoisomerase II inhibitors in hematopoietic stem cells in vivo. These experiments to understand leukemia etiology hold future hope for targeting prevention if the relevant damaging agents are identified.
• Pharmacogenomic Study of Leukemia Risk Assessment We have shown that the genotoxic catechol and quinone metabolites of etoposide induce DNA topoisomerase II cleavage at the sites of translocation breakpoints in MLL and partner genes in chemotherapy-related leukemias as described above. In the next experiments, we returned from the in vitro model to the patient. In pharmacokinetic studies of children receiving anticancer treatment with etoposide, which were performed in collaboration with Dr. Ian Blair in the Center for Cancer Pharmacology, we observed accumulation of the genotoxic catechol metabolite of etoposide from the first day to the last day of conventional multiple-day, bolus etoposide infusions. This project on leukemia risk assessment is a pharmacogenomic study directed at examining the relationship of CYP3A4 genotype to etoposide metabolism, with the future goal of rational dose and/or schedule alterations to reduce the risk.
• Leukemia Biomarkers This project involves testing CYP3A4 genotype as a host-specific biomarker, and developing panhandle PCR approaches to utilize MLL translocations as disease-specific biomarkers to predict the development of leukemia as a secondary cancer in children with primary neuroblastoma, where the dose-intensive treatment is associated with a high incidence of this treatment complication. Identification of relevant biomarkers to identify at-risk populations should enable individualization of the chemotherapy to reduce the risk.
• Gene Expression Profiles and Proteomic Profiles of MLL-rearranged leukemias We are using gene expression profiles ascertained by oligonucleotide array experiments to understand the influence of the partner genes of MLL on biology and prognosis, and we have found that leukemias with different partner genes can easily be distinguished. Through collaboration with the national pediatric cooperative group (Children's Oncology Group) we recently acquired specimens from a set of uniformly treated infants all diagnosed with acute lymphoblastic leukemia (ALL) for gene expression profiling. Gene expression profiling is beginning to set a precedent for the identification of significant biomarkers via global profiling for use as diagnostic and prognostic tools. However, because there are ~30,000 human genes but up to ~7 different proteins are produced per gene, and also because there is accumulating evidence for lack of correlation between transcript profiles and protein production, activation state, post-translational modification and/or localization, the incorporation of proteomics will be an essential and complementary adjunct to our genomics research. In the work of this project, transcript profiles will be mined and correlations will be examined between protein and transcript profiles of leukemias with MLL translocations.
• Bcl-2 Antisense Therapy for Infant Leukemias with MLL Translocations Recently we found high levels of the anti-apoptotic Bcl-2 protein in cell lines with MLL translocations, and we have promising new data on the cytotoxicity of a Bcl-2 antisense compound (GenasenseTM; Genta Inc.) in these cell lines in tissue culture. This project involves identifying synergistic combinations of the Bcl-2 antisense compound GenasenseTM and conventional cytotoxic anticancer drugs of different classes against MLL-rearranged leukemia cell lines grown in culture, and then determining whether the synergistic combinations of GenasenseTM and cytotoxic drugs defined in the in vitro model result in improved survival of mice with xenografts of leukemias with MLL translocations.

Lab personnel:

Charles Saginario, Ph.D., Research Associate
Diana Slater, Research Technician
Christos Kolaris, Research Technician
Dana Pietrzak, Research Technician
Matthew Colombo, Research Technician