5 ALK-Related Questions for D. Ross Camidge, MD, PhD and Fredika M. Robertson, PhD

D. Ross Camidge, MD, PhD

Director, Thoracic Oncology Clinical Program, University of Colorado, Denver, CO

Fredika M. Robertson, PhD

Professor, Department of Experimental Therapeutics, Division of Cancer Medicine, University of Texas MD Anderson Cancer Center, Houston, TX

D. Ross Camidge, MD, PhD, has served as a principal investigator on numerous national and international clinical trials, including studies that helped pave the way for the approval of crizotinib (Xalkori).

Fredika M. Robertson, PhD, focuses on identifying molecular targets for metastasis. She helped identify dysregulation of anaplastic lymphoma kinase (ALK) signaling in patients with inflammatory breast cancer, in collaboration with Massimo Cristofanilli, MD, chairman of the Department of Medical Oncology at Fox Chase Cancer Center in Philadelphia, and Emanuel F. Petricoin III, PhD, co-director of the Center for Applied Proteomics and Molecular Medicine at George Mason University in Fairfax, Virginia.

1

What role does the ALK signaling pathway play in normal cells?

Camidge: ALK is a developmentally regulated kinase that has a proven role in the development of visceral muscle patterning in Drosophila and in the development of the visual system in mice. Native ALK in adult humans is detectable within the intestine, central nervous system (CNS), and testes, but its function remains largely unknown.

Robertson: The receptor tyrosine kinase ALK is a member of the insulin receptor family and has a role primarily in normal embryonic neural development and differentiation, with limited distribution in adult mammals. The ligands for ALK include midkine and pleiotrophin and, based on a demonstrated role for these ligands in neural development and angiogenesis, ALK has been suggested to play a role in these processes through ALK phosphorylation, associated with activation of PI3K, Akt, and MAP kinase.

2

How is ALK signaling implicated in the development of cancer?

Camidge: Activation of ALK in association with neoplasia has been described in two main ways. Most commonly, the 3’ end of the gene containing the kinase domain is fused with the 5’ end of a different gene. The ALK is then expressed as part of a fusion protein, driven by the 5’ gene’s promoter that is active in the cancer tissue. The 5’ fusion partner protein acts as a dimerization motif mediating the activation of ALK.

This mechanism was first identified in some lymphomas but has since been described in subsets of multiple different cancers, often with different 5’ partners, including inflammatory myofibroblastic tumors (IMTs) and non-small cell lung cancer (NSCLC). More rarely, ALK can be activated by a primary mutation occurring in the kinase domain of full-length ALK, examples of which have been described in neuroblastoma and anaplastic thyroid cancer, among other cancers. These same kinase domain mutations and others can also occur as secondary events in cancers activated by ALK gene rearrangements, but in these situations the mutations are acting as a mechanism of acquired resistance to an ALK inhibitor.

Robertson: ALK is one of the only receptor tyrosine kinases identified to have oncogenic roles in both hematopoietic and solid tumors. ALK was first discovered based on its fusion with the nucleophosmin gene, designated as NPM-ALK, which is prevalent in anaplastic lymphoma. Additional fusions of ALK with other genes have been identified, perhaps most notably with echinoderm microtubuleassociated protein 4 (EML4-ALK), found to be present in a subgroup of patients with NSCLC who are very sensitive to the small-molecule ALK inhibitor crizotinib. The responsiveness of this subgroup of NSCLC patients, who typically have a very poor prognosis, led to the FDA approval of this ALK- targeted therapy, and also to the approval of the tandem EML4-ALK diagnostic assay.

There are also a range of diverse ALK genetic abnormalities reported in different tumor types, including neuroblastoma, IMT, ovarian carcinoma, pediatric and adult renal cell carcinomas, undifferentiated thyroid cancers, and esophageal carcinomas. ALK genetic abnormalities and signaling activation can occur via multiple pathways in these tumors that may include missense point mutations, gene amplification, fusions of ALK with different genes including tropomyosin 3/4, clathrin heavy chain gene, Ran binding protein 2, FN1, vinculin, and dysregulation of the TPM4-ALK fusion oncoprotein.

Our recent studies in inflammatory breast cancer (IBC) demonstrated the activation of the ALK signaling pathway characterized by ALK phosphorylation, with downstream activation of biochemically linked proteins including JAK/STAT3, Akt, mTOR, AMP kinase, and PDK1 in preclinical models of IBC. Based on these studies, we evaluated IBC patients for ALK abnormalities and found a prevalence of increased ALK gene expression, increased copy numbers of ALK, and in some cases, gene amplification with no detection of ALK fusions. This resulted in evaluation of ALK genetic abnormalities in IBC patients, and their enrollment in a phase I clinical trial with the small-molecule inhibitor, LDK378 (NCT01283516). Other clinical trials enrolling IBC patients with ALK-positive tumors are now being planned.

3

How has this been exploited in the development of ALK-targeted anticancer therapies?

Camidge: Recognizing that these abnormalities can be tested for—for example, by using a break apart FISH probe that binds upstream and downstream of the common breakpoint in ALK for rearrangements, or sequencing the kinase domain looking for activating mutations—patients can be preselected for treatment with experimental (and now at least one licensed) ALK inhibitor. This approach of preselecting patients with ALK-positive NSCLC early in drug development led to the rapid licensing of crizotinib within less than three years from when the first ALKpositive patient was treated.

Robertson: Currently, there are two primary approaches being taken to develop ALK-targeted therapies. The first approach is to develop small-molecule inhibitors of receptor tyrosine kinases, such as crizotinib, which is an inhibitor of ALK and another receptor tyrosine kinase, c-Met. The second approach is to develop therapeutic antibodies. There are several groups developing antibodies directed against ALK.

4

What are the key ALK-targeted therapeutic agents currently available and under development?

Camidge: Crizotinib is now licensed by the FDA. This license is conditional pending the results of randomized studies including PROFILE 1007 comparing crizotinib with standard secondline chemotherapies (docetaxel or pemetrexed) for NSCLC. Several other experimental ALK inhibitors including Novartis’ LDK378 and Ariad’s AP-26113 are in early-phase clinical trials. Experimental Hsp90 inhibitors also seem to have activity against ALK gene-rearranged NSCLC, and studies in crizotinib-naïve and crizotinib- resistant patients are under way. Finally, there are some conflicting retrospective data on whether ALK-positive NSCLC may also be particularly sensitive to pemetrexed (a licensed cytotoxic), and the results of the pemetrexed arm in the PROFILE 1007 study are eagerly awaited to confirm or deny this in a data-rich setting.

Robertson: There are a number of ALK-targeted therapeutic antibodies currently under development, including those from the laboratory of Dr Marc Vigny (Université Pierre et Marie Curie, Paris) and Delenex Therapeutics AG (Zurich, CH). There are also numerous small-molecule inhibitors of ALK, including the FDA-approved agent crizotinib, CH5424802, and LDK378 in phase I/II clinical trials, and GSK1838705A, NSM-E628, and AP-26113, among others undergoing preclinical evaluation. These agents are derived from numerous different compound classes including the aminopyridines, pyrazolopyrimidines, and acylaminoindazoles, and are being evaluated in a variety of different ALK-driven tumor types.

5

What challenges do researchers face in the further development of ALK-targeted therapies?

Camidge: The challenges are both practical and scientific. Practical challenges include finding the relevant subsets of patients who are ALK-positive in the first place, and making sure they can get to a trial site to have access to an experimental therapy, in addition to conducting trials looking for the activity of a new drug in the crizotinib-naïve setting when crizotinib is now a licensed drug and there is so much “buzz” about it.

Scientific challenges include the fact that the CNS penetration of crizotinib is very low, and progression in the brain is a major problem: Will new agents have activity in the CNS, and will they even address this within their study designs? In addition, systemic progression on crizotinib occurs through multiple different resistance mechanisms including ALK kinase domain mutations, copy number gain of rearranged ALK, and the selection of other driver abnormalities in the same or different cells within the tumor.

Finally, although the side effects of crizotinib are generally mild, they are not zero. For example, recently a rapid drop in testosterone levels in men on crizotinib has been described, which rises again once the drug is discontinued. As crizotinib may be taken for many months, the chronic consequences of this side effect may turn out to be significant in some men, including the chronic effects of low testosterone on energy levels, bone and muscle mass, mood, and libido.

Like many other tyrosine kinase inhibitors, crizotinib also affects a number of other targets including c-Met and ROS1. Therefore, it is not yet clear whether newer ALK inhibitors will have the same or a different side effect profile.

Robertson: There are two primary issues that we face in the further development of ALK-targeted therapies.

The first is to identify the optimal strategy for overcoming the development of acquired therapeutic resistance by patients treated with small-molecule ALK inhibitors, which is a predicted occurrence that commonly occurs with targeted therapies. One option to overcome resistance is to combine an ALK small-molecule inhibitor, such as crizotinib, with ALK therapeutic antibodies. Another approach is to use therapeutics that target heat shock protein 90, which operates as a molecular chaperone for ALK. These approaches are both being evaluated for their efficacy.

The second challenge is to develop alternative approaches to identify those patients, regardless of their tumor type, who may respond to ALK-targeted therapies. There is an immediate need for additional diagnostic tests that will allow identification of those patients with ALK abnormalities that are not detected by the FDA-approved tandem diagnostic, which is the Vysis ALK Break Apart FISH Probe kit.