Mechanisms of Disease: Oncogene Addiction-A Rationale for Molecular Targeting in Cancer Therapy

I Bernard Weinstein; Andrew K Joe


Nat Clin Pract Oncol. 2006;3(8):448-457. 

In This Article

Future Directions and Clinical Applications

As emphasized earlier, human cancers display multiple genetic and epigenetic abnormalities. Furthermore, it is now apparent that these abnormalities frequently differ between different types of cancer, and also between subsets of the same type of cancer. In view of this complexity, how can we identify the specific oncogene or oncogenes that have a critical role in maintaining the malignant phenotype in these different cancer types, or within individual cases? In other words, how do we identify the Achilles' heel in specific cancers so that each patient can be treated with the appropriate molecular targeted agent? At the present time there are no methods to fully assess the total circuitry that controls cell proliferation, differentiation and apoptosis in normal or cancer cells. Advances in network theory, systems biology, and computer modeling, which are discussed below, may eventually make this possible.

Currently, several empiric approaches can be used to help identify the Achilles' heel of specific types of human cancer. One approach is to use high-throughput screening of thousands of compounds in chemical libraries to identify specific compounds that preferentially inhibit in vitro growth or induce apoptosis in specific types of human cancer cells. A related approach that is being rapidly expanded is to use a library of siRNAs—low-molecular weight RNAs that are taken up by cells and inhibit the expression of specific genes—to identify which genes are required to maintain the proliferation and/or survival of specific types of cancer cells. Once such genes are identified, drugs can be designed to target the related proteins.[52] A recent study in mice suggests that it might become feasible to administer to patients a specific siRNA preparation that knocks down the expression of a critical oncogene in the tumor, thus providing a novel approach for delivering cancer therapy.[53]

In addition, specific criteria might be used to assist in identifying genes that are most likely to be critical for maintaining the malignant phenotype. For example, oncogenes that are mutated early in the multistage process of tumor development might be favored candidates because they had a critical role in determining subsequent aspects of the abnormal circuitry in the evolving cancer cells. Oncogenes that are mutated, and not simply overexpressed, might also be more likely targets for therapy since they reflect the 'hard-wiring' of cancer cells, rather than epigenetic abnormalities. Mutated oncogenes might therefore be more likely to be present in the stem-cell population of tumors rather than just in the progeny cells. In addition, mutated oncogenes might be more likely to have qualitatively different roles than oncogenes that are only overexpressed. This aspect is exemplified by the properties of a mutated EGFR in NSCLC cells.[49] Specific cancers display increased expression of genes that normally have critical roles in stem cells, or during normal tissue development and differentiation. These include the genes that encode proteins involved in the Wnt, Hedgehog, TGFβ/BMP, Notch, Snail, Slug, and MITF signaling pathways.[54,55,56,57,58] Specific cancers could be highly dependent on, (i.e. addicted to) one of these factors. For example, inhibition of the Hedgehog pathway in murine medulloblastoma blocked proliferation and inhibited tumorigenesis.[57] The transcription factor MITF is involved in normal melanogenesis and is overexpressed in human melanoma; inhibition of its expression in melanoma cell cultures markedly inhibited cell growth.[56] Recent studies indicate that Snail, a transcription factor that has a major role during embryonic mesoderm development, also has a role in the recurrence of human breast cancers.[13] It might, therefore, also be a critical target for therapy in these cancers. It is possible that dependence on a specific oncogene might be different in the stem cells than in the progeny cells in a given tumor, because of differences in their intracellular circuitry. Optimal therapy would then require developing molecular targeted agents that specifically target the critical oncogene in the stem cells of a specific cancer. Further characterization of stem cells in specific tumors should clarify this aspect of oncogene addiction, and the potential limitations of specific molecular targeted agents.

'Network theory' and the emerging field of systems biology may eventually provide methods for conceptualizing and analyzing the entire circuitry of specific types of normal and cancer cells, and thus facilitate identification of specific pathways of oncogene addiction in different types of cancer cells. Networks in mammalian cells share similar principles with networks that are relevant to engineering and other disciplines. Thus, by viewing the cancer cell as a complex or complicated system,[59] genes and proteins as 'nodes',[60,61] and groups of interacting proteins in a signal transduction pathway as 'modules', engineering tools and concepts can be used to analyze large datasets to generate an holistic view of the regulatory network of normal and cancer cells. Furthermore, a concept called Boolean genetic network theory can be used to study the cancer cell as a dynamic system.[62] By analyzing the myriad of individual and interacting genes in a cancer cell in a Boolean fashion (i.e. active = 'on', inactive = 'off'), the overall activity state ('attractor state') of specific cancer cells (e.g. proliferation, apoptosis, or differentiation) might be determined or predicted. Cells are described as being in a dynamic equilibrium between these attractor states, and changes in cell state ('hysteresis') can be described as responses to various external and internal stimuli. Thus, in the future it could be possible to use network theory to describe and predict the existence of an 'addicted' attractor state, and to also predict the specific oncogene responsible for this addicted state.[59,62]

Advances in network theory and systems biology are being facilitated by powerful microarray and proteomic methods that compare expression profiles of thousands of genes and proteins between normal tissues, cancers, and subtypes of specific cancers. Recent data are providing insights into signaling pathways and networks characteristic of specific types of cancer; this is discussed in a review by Baak et al.[63] Hopefully, this information will facilitate the identification of pathways of oncogene addiction characteristic of specific cancers or their subtypes, and thus guide the use of specific molecular targeted agents in specific patients. A more detailed understanding of these networks might also provide insights into tumor resistance and the appropriate choice of combination therapy for specific patients.

The concept of oncogene addiction is also relevant to tumor-stromal interactions, tumor invasion and tumor metastasis. Thus, activated oncogenes like EGFR or Ras stimulate signaling pathways in cancer cells that cause increased expression of matrix metalloproteinases (which enhance tumor invasion) and of the angiogenic factor VEGF.[64,65] Inactivation of these genes in cancer cells can inhibit tumor invasion, angiogenesis, and even recurrence. The neovasculature associated with tumor-induced angiogenesis is abnormal compared with normal vasculature.[66] The circuitry that regulates the growth and function of endothelial cells in this neovasculature might, therefore, differ from that of normal vasculature[41,42,43]. Thus, it could be possible to develop agents that preferentially target tumor vasculature; bevacizumab provides a promising example, and its use has led to improved disease-free and overall survival rates in patients with colon cancer.[42]

It should be emphasized that although the concept of oncogene addiction may apply to a given cancer at a particular time or stage, it is apparent from some of the mouse model experiments[67] ( Table 1 ) and from clinical experience with molecular targeted agents ( Table 3 ) that cancers can 'escape' from a given state of oncogene addiction. Presumably, this finding reflects the genomic instability of cancers, and could also reflect epigenetic changes in gene expression that lead to an altered state of cell circuitry; for example, ongoing changes in DNA methylation and chromatin structure in a cancer cell population. It is not known whether this escape leads to secondary addiction to another oncogene or to the growth of a population of 'non-addicted' cancer cells. For these reasons, as well as the likelihood of heterogeneity within tumors, it is unlikely that the use of a single molecular targeted agent will achieve long lasting remissions or cures in human cancers, especially for late-stage disease. Combination therapy will, therefore, be required, which raises several unresolved questions. Can such combinations be rationally designed? Should the individual agents act on the same molecular target but by different mechanisms, or on different targets in the same pathway, or should each agent target a different pathway or cellular mechanism? The first approach could prevent emergence of the types of resistance recently seen with imatinib[44] and gefitinib.[45] The second approach also seems rational because it is likely that cancer cells may be addicted to specific signaling pathways rather than a single oncogene. A drug combination that targets proteins that function at different stages in the same pathway may, therefore, be more likely than a single drug to inactivate that pathway.[68] The third approach can also be justified because of the disturbance of multiple pathways in cancer cells. It is becoming increasingly apparent that certain molecular targeted agents are actually promiscuous, i.e. they target more than one molecule, and this multiple targeting could enhance their therapeutic efficacy.[69] For example, the drug sorafenib seems to have activity against renal carcinomas by targeting both a mutant Raf oncogene and the VEGF receptor, thus inhibiting both cancer cell proliferation and angiogenesis.[55] The compound sunitinib apparently exerts antitumor activity by targeting multiple protein kinases.[55] It may therefore be useful to further exploit this principle by using so-called 'multi-targeting' agents. HSP90 is a molecular chaperone that stabilizes various proteins. Geldanamycin[70] and related compounds inhibit HSP90 and may therefore, increase the degradation of mutant oncogenic proteins. Conversely, inhibitors of the proteosome-ubiquination proteolysis pathway such as bortezomib[71] might stabilize and increase the activity of cellular proteins, including tumor suppressors (e.g. p53). These two types of agents could, therefore, also be relevant to the concept of oncogene addiction and might preferentially target mutant oncogenic and tumor suppressor proteins.

Clinical studies indicate that the efficacy of certain molecular targeted agents can be enhanced by combining them with cytotoxic agents, i.e. agents that often act by inhibiting DNA or chromosomal replication. Trastuzumab that targets HER2 can improve response and survival rates if given in combination with paclitaxel to patients with metastatic breast cancer.[28] The combination of bevacizumab or cetuximab with cytotoxic chemotherapy agents can also improve response rates in patients with metastatic colon and breast cancer, respectively.[40,41] Furthermore, when bevacizumab was added to a combination chemotherapy regimen it improved overall survival rates in patients with metastatic colon cancer.[42] As with chemotherapy, the efficacy of targeted therapy is likely to be greater in patients with minimum residual disease. Thus, treatment with trastuzumab after adjuvant chemotherapy significantly improves disease-free survival in patients with early-stage breast cancer.[29]

In this review we have emphasized the roles of dominant-acting oncogenes that enhance cell proliferation and cell survival. The disordered circuitry of cancer cells, however, is also a consequence of inactivation or loss of expression of tumor suppressor genes, which normally inhibit proliferation or enhance apoptosis. Experimental studies indicate that reintroducing a wild-type tumor suppressor gene (e.g. those encoding p53, Rb, or APC) into human cancer cells where the respective endogenous gene was inactive usually caused marked inhibition of growth, induction of apoptosis and/or inhibition of tumorigenesis in mice.[2,3] These results would not be expected if cancer cells evolved simply through the stepwise addition of genetic abnormalities, because then correction of just one mutagenic event should have only a modest inhibitory effect. Thus, some cancer cells seem to be hypersensitive to the growth-inhibitory effects of specific tumor suppressor genes. We postulate that, like oncogene addiction, this effect also reflects the bizarre circuitry of cancer cells, and have termed this phenomenon 'tumor suppressor gene hypersensitivity'.[2,3] This phenomenon can also be exploited in cancer treatment by gene therapy, for example by the use of an adenovirus that encodes a normal p53 protein[72] or by targeting a downstream signaling pathway that was activated as a result of loss of activation of a tumor suppressor gene. Alternatively, if a tumor suppressor gene is inactivated by the process of DNA methylation, which frequently occurs in cancer cells (e.g. p16INK4gene), drugs that cause demethylation of cellular DNA could switch such genes back on and thereby inhibit tumor growth.[73] This mechanism seems to be the means by which 5-azacytidine and zebularine inhibit tumor growth.[73,74] Drugs that inhibit histone deacetylase enzymes and thereby enhance gene expression, such as depsipeptide or suberoylanilide hydroxamic acid, might also exert antitumor effects by reactivating silenced tumor suppressor genes.[75] Synergistic or additive effects on tumor growth inhibition might be obtained by combining drugs that exploit both oncogene addiction and tumor suppressor gene hypersensitivity.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.