Translational Research

Identifying Targets for Therapy

Four projects illustrate the unique scientific capabilities of Dana-Farber and how we are accelerating the pace of translational research and bringing new hope to patients.

Targeting B-cell receptor pathways in diffuse large b-cell lymphoma

Margaret Shipp, PhD 

Diffuse large B-cell lymphomas (DLBCLs) have similar features under the microscope, says physician-scientist Margaret Shipp, MD, of the Department of Medical Oncology. But subtypes of the disease, which is the most common type of non-Hodgkin's lymphoma (NHL), behave differently in response to treatment. Shipp suspected that identifying unique molecular signatures of these tumors might provide insight into their survival pathways. In previous research, her longtime collaborator, Stefano Monti, PhD, of the Broad Institute, had conducted gene expression profiles of DLBCL samples. In interpreting the data, they found a significant subset, named the BCR-type, with a distinctive signature: increased expression of components of the B-cell receptor (BCR) signaling pathway.

A key node in the pathway, spleen tyrosine kinase (SYK), relays signals downstream and plays a major role in low-level, or tonic, BCR signaling. Because SYK has been found to be crucial to survival, Shipp hypothesized that BCR-type tumors might depend on tonic BCR signaling and that a SYK inhibitor – which Rigel Pharmaceuticals had already developed and tested in clinical trials for another indication – might offer a new rational therapy.

To find out, the Shipp group, led by Linfeng Chen, PhD, treated a panel of DLBCL cell lines and primary tumors with the SYK inhibitor and showed that it induced apoptosis by shutting down tonic BCR signaling. Moreover, the responsive cell lines and tumors were identified by transcriptional profiling as the BCR-type, which comprises up to 50 percent of all DLBCLs.

Within months of her preclinical work, Shipp partnered with Dana-Farber oncologist Ann LaCasce, MD, of the Department of Medical Oncology, and Rigel Pharmaceuticals to conduct the first clinical trial of an oral SYK inhibitor in patients with NHL. "Given Margaret's elegant correlative laboratory data, we expected a subset of patients to respond based on their subtype," says LaCasce, who oversaw the Dana-Farber patients on the Phase I/II multicenter trial. "The drug induced a response in 21 percent of patients with DLBCL – exciting results, considering that many of these patients had rapidly advancing disease," says LaCasce. Shipp and pathology colleagues are now searching for biomarkers of BCR-dependent DLBCLs, which will help identify subtypes in real time and predict which patients will respond to this targeted therapy.

Targeting BRAF in pediatric low-grade astrocytomas

Charles Stiles, PhD, and Mark Kieran, MD, PhD 

"Brain tumors have surpassed leukemias as the leading cause of cancer-related death in children," says Charles Stiles, PhD, co-chair of the Department of Cancer Biology. The most common of these tumors are low-grade astrocytomas (LGAs). Some LGAs are curable with surgery and chemotherapy; others, however, arise within inoperable regions of the brain, and the side-effects of cytotoxic drugs in growing children can be severe. In addition, LGAs frequently recur after surgery or drug treatment, and these tumors can be fatal. The Pediatric Low-Grade Astrocytoma Program (PLGA) at Dana-Farber seeks to find nontoxic targeted therapeutics. Stiles and Kieran, along with collaborators Keith Ligon, MD, PhD, and Levi Garraway, MD, PhD, both of the Department of Medical Oncology, have made major progress.

Recent studies from the PLGA Program and other laboratories identified two separate abnormalities in the BRAF oncogene, which together account for as many as 50 percent of pediatric LGAs. To help Stiles and Kieran analyze archival LGA tissue for BRAF abnormalities, Ligon and Garraway adapted genomic technologies to work with the paraffin-embedded samples that comprise the majority of LGAs. Ligon's "paraffin-friendly" fluorescent in situ hybridization (FISH) assay detects the most common BRAF abnormality, while Garraway's OncoMap finds the less common one. In addition, OncoMap can identify point mutations in the other 50 percent of LGAs that are genetically normal for BRAF. Investigators are now working to convert paraffin-based FISH and OncoMap assays into CLIA-certified tests, upon which clinical decisions can be made, says Stiles, who also plans to establish a nationwide BRAF mutation database for pediatric LGAs. "We want to identify the children with BRAF mutations, in the hope that targeted drugs will be available in a few years," he says.

Since BRAF mutations in pediatric LGAs are identical to those found in a high percentage of adult malignant melanomas, data from ongoing Phase I studies of BRAF inhibitors in adults will greatly reduce the lead time to clinical trials in children, explains Stiles. "LGAs grow slowly, and even if they recur after initial treatment, the interval is measured in years," he adds. "For some of these kids, this may be plenty of time."

Overcoming resistance to EGFR inhibitors

Geoffrey Shapiro, MD, PhD talking with patient Shaun Farrell and Alan D'Andrea, MD 

Non-small cell lung cancer is the leading cause of death from cancer in the United States. One of the most common activating mutations found in this type of lung cancer is L858R, located in the tyrosine kinase domain of the epidermal growth factor receptor gene (EGFR). Although patients with EGFR mutations are very responsive to tyrosine kinase inhibitors (TKIs), such as gefitinib and erlotinib, cancer cells ultimately develop resistance to TKIs, primarily through an acquired secondary mutation known as T790M. Overcoming the resistance conferred by T790M has become a major clinical challenge.

In a recent study, Kwok-Kin Wong, MD, PhD, of the Department of Medical Oncology, who specializes in lung cancer models, and clinical investigator Geoffrey Shapiro, MD, PhD, of the same department, teamed up to search for an alternative therapeutic strategy for patients who develop resistance to TKIs. Wong began by genetically engineering mice to express mutant human EGFR with the T790M-L858R compound mutation (hEGFR TL) and then demonstrated that it is both oncogenic and essential to tumor maintenance.

Wong tried treating the hEGFR TL mice with neratinib (HKI-272, from Wyeth), based on studies showing that the newer "irreversible" TKIs could block T790M. If he was expecting a robust response, he was to be disappointed. "It turned out that neratinib alone was not that potent," says Shapiro, whose laboratory collaborated with Wong and conducted parallel experiments in cell lines with similar results. "There was still residual downstream signaling related to PI3K and mTOR," he explains. Subsequently, the two investigators decided to treat the mice with neratinib and the mTOR inhibitor rapamycin. To their great satisfaction, the combination therapy inhibited EGFR and downstream signaling as well, resulting in dramatic tumor regression.

"The results seen in Kwok's mouse models will be extremely useful in predicting outcomes of various treatments as we now move ahead with clinical trials," says Shapiro, who directs the Early Drug Development Center (EDDC) at Dana-Farber, where the majority of Phase I and proof-of-mechanism studies at the Institute are conducted. In fact, EDDC will soon be launching a trial combining neratinib with Wyeth's mTOR inhibitor, temsirolimus. "The mission of the EDDC is to harness as much science as possible from Dana-Farber laboratories and to convert discoveries into trials for our patients," declares Shapiro. "Our work with Kwok is a prime example of that."

A new approach in neuroblastoma

A. Thomas Look, MD and other staff members 

Neuroblastoma (NB), another serious pediatric cancer, is especially dangerous in children over the age of 18 months and those with disseminated disease, says A. Thomas Look, MD, vice chair for research in the Department of Pediatric Oncology. Although survival has improved with intensive chemotherapy, treatment causes long-term effects, adds translational investigator Rani George, MD, PhD, also of Pediatric Oncology, whose laboratory is collaborating with Look's to discover new genetic abnormalities in NB that may lead to better and less toxic treatments.

In previous SNP array analyses of NB tumor samples, George and Look had found that anaplastic lymphoma kinase gene (ALK) was amplified, leading them to wonder whether ALK played a major role in NB. With the help of Matthew Meyerson, MD, PhD, of the Department of Medical Oncology and co-director of the Center for Cancer Genome Discovery, they sequenced ALK in 93 primary tumors from high-risk patients and identified five previously unknown mutations, including F1174L, which was the most common and found to activate signaling by the ALK cell surface receptor.

"These mutations change the structure of the kinase," says Look, "so that the ALK receptor is no longer dependent on a ligand for activation - thereby removing the mechanism for controlling powerful growth signals." He and George hypothesized that when tumor cells mutate and constitutively activate ALK, the receptor becomes oncogenic. To test their hypothesis, George turned to chemical biologist and colleague Nathanael Gray, PhD, of the Department of Cancer Biology and member of the Initiative in Chemical Biology, who had developed a compound, TAE684, which would allow investigators to chemically inhibit ALK in NB cell lines. Not only was the F1174L mutation especially sensitive to the ALK inhibitor, but also siRNA knockdown of ALK with activating mutations, particularly F1174L, caused tumor cell death, "implying that this mutation is one of the major abnormalities mediating growth and proliferation in neuroblastoma cells," explains George.

Fortunately, since drug companies had already introduced an ALK inhibitor in adult clinical trials, a Children's Oncology Group (COG) clinical trial can soon begin enrolling neuroblastoma patients with ALK mutations. "I'm so gratified that all our work in the laboratory has led to a target that we can not only exploit clinically, but also study in more detail to expand our knowledge of neuroblastoma," says George.