Scientists decode 'software' instructions of aggressive leukemia cells

The research, published in Cancer Discovery, is the first study to show how these three different forms of white blood cell cancer are epigenetically programmed by several different molecules controlling cascading biological networks that manipulate normal gene function, directing cancer development and growth.

"Epigenetic programming is the software that is written on to human DNA, which can be viewed as its hard drive. This programming contains the instructions that determine how cells including leukemia cells function and cause disease," says the study's lead investigator, Dr. Ari Melnick, associate professor of medicine and director of the Raymond and Beverly Sackler Center for Biomedical and Physical Sciences at Weill Cornell Medical College.

"Finding the instructions that ultimately lead to cancer development, and to the especially bad outcome seen in patients with these different forms of ALL, is especially urgent. Epigenetic instructions are contained in many chemical layers. Our study is the first to integrate the decoding of many layers simultaneously, which has enabled us to unlock some of the mysteries explaining the malignant and aggressive behavior of these leukemias," says Dr. Melnick, who is also a hematologist-oncologist at NewYork-Presbyterian Hospital/Weill Cornell Medical Center.

Abnormal Epigenetic Programming Leads to Poor Outcomes

The research team examined abnormalities in the "software" epigenetic programming that leads to the three forms of adult B-acute lymphoblastic leukemia (B-ALL), the most common form of ALL. These forms are BCR-ABL1-positive B-ALL, E2A-PBX1-positive B-ALL, and MLLr-B-ALL. These three B-ALL subtypes feature mutations of different master regulatory genes which force bone marrow cells to produce cancer-promoting proteins. Long-term survival is less than 40 percent among these patients.

"Similar to normal tissue, we believe that tumors may be dependent on specific patterns of epigenetic programming -- especially in B-ALL, where studies suggests epigenomic programming is globally disrupted," Dr. Melnick says. "Our goal was to identify epigenetically modified genes and the molecular machines that cause them to become abnormally programmed."

To that end, the research team performed DNA methylation and gene expression profiling on 215 adult B-ALL patients enrolled in the ECOG E2993 clinical trial, a multi-center and multi-national study, testing different forms of treatment in patients with ALL.

Researchers identified core epigenetic gene signatures that were associated with abnormal fusion proteins. In the case of BCR-ABL1-positive B-ALL, they found that the most deregulated gene network centered around an extraordinarily epigenetically deregulated gene they identified as interleukin-2 receptor alpha, which encodes a protein called CD25.

"Among patients who had BCR-ABL1-positive B-ALL, it was those with aberrant epigenetic programming of CD25 that had significantly worse outcome," says Dr. Melnick. "It's the patients that have this programming glitch that do really poorly."

Although the researchers don't yet know what CD25 does, and why it is important, they say CD25 will be a useful biomarker to test for patients that are at highest risk for poorer outcome.

Dr. Melnick stresses that therapeutic antibodies to the CD25 protein already exist that can, theoretically, destroy leukemic cells expressing this protein. The research team showed that using the CD25 antibody successfully killed BCR-ABL1-positive B-ALL in laboratory experiments. "One could potentially conceive of a human clinical trial where those antibodies are used to attack these cancerous cells," he says.

Researchers also discovered what is writing the bad software in the other two B-ALL subtypes. In both cases, the abnormal cancer proteins E2A-PBX1 and MLLr turn out to be directly involved in altering the epigenetic programming of leukemic cells. Remarkably, MLLr epigenetically turns on a powerful cancer protein called BCL6. In the study, drugs developed by Dr. Melnick that block BCL6 activity potently killed and suppressed the ALL cells in patients enrolled in this clinical trial, which warrants the testing of BCL6 inhibitors in this aggressive form of ALL.

"This study links the direct actions of oncogenic fusion proteins with disruption of epigenetic regulation that leads to abnormal production of cancer-driving genes," Dr. Melnick says. "It potentially provides us with a biomarker for cancer outcomes as well as potential treatments in these aggressive forms of leukemia."

This research study was supported by the Chemotherapy Foundation, Burroughs Wellcome Foundation, The Leukemia & Lymphoma Society and the Sackler Center for Biomedical and Physical Sciences at the Weill Cornell Medical College.

Study co-authors include Dr. Sarah Brennan, Yushan Li, Dr. Chuanxin Huang, Yuan Xin, Dr. Monica L. Guzman and Dr. Olivier Elemento from Weill Cornell; Dr. Huimin Geng, formerly of Weill Cornell and now at University of California, San Francisco; Dr. Janis Racevskis and Dr. Elisabeth Paietta from Albert Einstein College of Medicine; Dr. Thomas A. Milne, Dr. Wei-Yi Chen, Dr. Debabrata Biswa, Dr. C. David Allis and Dr. Robert G. Roeder from Rockefeller University; Christian Hurtz, Dr. Soo-Mi Kweon, Dr. Seyedmehdi Shojaee and Dr. Markus Müschen from the University of Southern California, Los Angeles; Lynette Zickl and Dr. Donna Neuberg from the Dana Farber Cancer Institute, Boston; Dr. Rhett P. Ketterling and Dr. Mark R. Litzow from the Mayo Clinic, Rochester, Minnesota; Dr. Selina M. Luger from the University of Pennsylvania; Dr. Martin S. Tallman from the Memorial Sloan-Kettering Cancer Center; Dr. Jacob M. Rowe from the Rambam Medical Center in Haifa, Israel; and Dr. Hillard Lazarus from Case Western Reserve University in Cleveland, Ohio.