|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Journal of Clinical Oncology, Vol 23, No 8 (March 10), 2005: pp. 1776-1781 © 2005 American Society of Clinical Oncology. DOI: 10.1200/JCO.2005.11.029
Targeted Therapy: Wave of the Future
From the Division of Hematology/Oncology, Department of Medicine, David Geffen/UCLA School of Medicine at Los Angeles; and Genentech Inc, South San Francisco, CA Address reprint requests to Mark D. Pegram, MD, UCLA Center for the Health Sciences, 11-910833 Le Conte Avenue, 11-934 Factor Building, Los Angeles, CA 90095; e-mail: mpegram{at}ucla.edu.
Understanding Gene Alterations in Human Breast Cancer Gene alterations play an important role in the origin and pathogenesis of human breast cancer. Three broad categories of gene changes that appear to contribute to tumor progression include tumor suppressor genes, repair-mutator genes, and oncogenes.1,2 Tumor suppressor genes can be defined as a class of genes whose function is lost as a result of germline or somatic mutation, resulting in tumor development. Repair-mutator genes constitute a subset of the tumor suppressor gene class and are genes involved in DNA repair pathways (such as the DNA mismatch repair genes), whose loss of function likely contributes to cancer via an increased frequency of mutation in other cellular genes involved in growth regulation. Oncogenes are genes directly responsible for cancer progression and often present as altered versions of proto-oncogenes that are normally involved in control of cell growth and differentiation.1-5 In the breast cancer cell, qualitative or quantitative differences are found between the proto-oncogene and its corresponding oncogene. A proto-oncogene can become an oncogene when a mutation in the coding region constitutively activates the biologic activity of the protein product without affecting the total amount of the product. Alternatively, a proto-oncogene can become an oncogene when excess product occurs from amplification (multiple copies) of the gene or from mutation, rearrangement, insertion, or deletion of the regulatory region of the gene.6 The oncogenes are, in turn, involved in the regulation of a complex series of cyclin-dependent kinases and other cell cycle modulators that determine progression through the cell division cycle.4 Breast cancer progression is hypothesized to occur by an accumulated series of genetic and phenotypic changes in pathways regulating cell growth. With classic cytogenetic methods and studies of loss of heterozygosity, gene regions identified as commonly rearranged, amplified, or otherwise altered have been commonly detected at chromosome 1, 3, 6, 7, 8, 9, 11, 13, 15, 16, 17, 18, and 20.4,5 Application of comparative genomic hybridization has also implicated chromosome 10,12, and 22 in the malignant process. As in most human cancers, the most common genetic abnormality in breast cancer is loss of specific chromosome arms. Loss of heterozygosity analysis of polymorphic DNA markers point to chromosomes and subregions of chromosome arms likely to harbor tumor suppressor genes. Loss of heterozygosity generally allows expression of a recessive mutant in an allele of a tumor suppressor gene by removal of a dominant normal allele, as in the case of p53 expression, for example, in the Li Fraumeni syndrome.1-4 The second most common type of cytogenetic alteration in breast cancer appears to be gene amplification.4,6 Karyotype analysis and chromosome in situ hybridization approaches such as comparative genomic hybridization or microfluorescent in situ hybridization point to amplified chromosomal loci likely to harbor oncogenes. The initial step in gene amplification may involve the formation of extrachromosomal, self-replicating units termed double-minute chromosomes. These elements then become permanently incorporated into chromosomal regions and are termed homogeneous staining regions. An amplified genetic unit (amplicon) is initially much larger than the actual size of the principal gene of biologic importance. Irrelevant or silent genes may also be coamplified with one or more expressed genes on an amplicon.4,6
Of the
Insertional Mutagenesis Studies for Identification of New Cellular Oncogenes As Targets for Drug Development The activation of candidate proto-oncogene targets by proviral insertion permits a means of identifying such genes through the isolation of molecular clones, as the integrated viral DNA sequence serves as a marker that can be used to isolate the molecular clone's flanking sequences, including the candidate proto-oncogene target activated by proviral insertion (Fig 1).9-12 A prime example of this approach was demonstrated in 1982 in the laboratory of Harold Varmus. These investigators were able to isolate the wnt-1 gene, which is activated by integration of mouse mammary tumor virus (MMTV) proviral DNA in mouse mammary carcinomas.13 In these classical experiments, an MMTV provirus, together with its flanking DNA sequence, was cloned from a mammary carcinoma phage library using MMTV DNA as a probe. Specific probes for flanking DNA sequences were then utilized to isolate the wnt-1 gene. In this situation, the MMTV long terminal repeat acts as an enhancer to elevate wnt-1 gene expression. Of course, in addition to the mechanism of proto-oncogene activation, integration of proviral DNA within a gene-coding region could result in its inactivation—a methodology that has been used to identify candidate tumor suppressor genes. In summary, (1) MMTV, which does not encode viral oncogenes, acts as a mutagen through random integration in the genome; (2) integration of the virus at multiple sites within individual tumors suggests cooperativity between genes in breast cell neoplastic transformation; and (3) MMTV itself serves as a tag for high throughput polymerase chain reaction amplification, cloning, and sequencing of candidate proto-oncogenes adjacent to the MMTV insertion sites.
Strategy for Identification of Proto-Oncogenes Based on Insertion of MMTV Promoter/Enhancer Elements Recently, methods have been introduced for the rapid identification of proto-oncogenes based on insertion of MMTV promoter/enhancer elements in or near host proto-oncogenes in mammary tumors extracted from mice following neonatal infection with MMTV. With the recent availability of the nearly complete mouse genome sequence, DNA sequences flanking retroviral integration sites can be isolated using high throughput polymerase chain reaction–based methods for rapid identification of candidate disease genes. Common retroviral insertion sites of Moloney murine leukemia virus–induced lymphomas have recently been reported, highlighting the utility of this technology for the rapid identification of candidate proto-oncogenes.12,14-16 These screens have led to the identification of 236 common integration sites and candidate disease genes could be identified in the majority of cases simply by examining the annotated mouse genome sequence.17 It is both reassuring and compelling to note the fact that in many cases, the human orthologs of these genes have turned out to be validated proto-oncogenes in human tumors.17 In many instances, however, when two or more genes are identified by insertional mutagenesis within the same tumor, it implies cooperativity between the genes in tumor induction. Therefore, not all of the genes identified in such screens are necessarily proto-oncogenes, but rather may be playing a supporting role in concert with other oncogenes (as well as inactivation of tumor suppressors) in transformation or tumor progression. Several groups have now initiated exhaustive screens for novel oncogenes in MMTV-induced mouse mammary tumors.18,19 In one such screen, 1,152 mouse genes have been prospectively identified (using curation criteria of occurrence of proviral integration at the same site in at least two independent tumors). Of these, 1,142 human orthologs have been identified through sequence comparisons in human genome databases.19 While comparison of insertional mutagenesis data generated in different laboratories (or even in different mouse strains) may identify unique sets of genes, genes identified in multiple screens using different mouse models are more probable to represent true disease genes.17 This fact has prompted researchers at the Mouse Cancer Genetics Program at the National Cancer Institute to develop the Retroviral Tagged Cancer Gene Database (RTCGD), which at the present time, contains 3,100 retroviral integration site sequences cloned from 17 mouse tumor models by nine different laboratories. Clone lists describe the viral integration site clones, along with the tumor model and tumor type from which they were cloned, candidate disease genes, genomic position, and orientation of the integrated provirus with respect to the candidate genes.17 Researchers can now identify integrations of interest and compare their results with those from multiple tumor models and tumor types using RTCGD.17 Herein lie promising future targets for anticancer drug development for breast cancer as well as many other human malignancies. The task at hand now is to rigorously probe these databases in order to validate which of these genes are playing a role in the pathogenesis of breast cancers so that specific antagonists can be developed for clinical or translational investigation. Sobering is the fact that in such databases, it is typical that at least one quarter of the candidate genes are presently genes of unknown function whose potential role in human malignancy has never been studied. Moreover, it is possible that many more genes with putative function have important biologic activities other than those currently ascribed to the molecule. Therefore, it will still be necessary to conduct careful functional studies of these gene products in order to fully understand their role in human malignancy and to fully exploit therapeutic targeting of these molecules clinically. In the postgenome era, it is apparent that there is still no substitute for cell biology and biochemistry.
Validation of Novel Therapeutic Targets
Parallel Clinical Development of Diagnostic Assays for Novel Therapeutic Targets A Medline literature search using the key word "epidermal growth factor receptor" (EGFR) yields 13,569 citations. And yet despite this intense level of scrutiny and careful scientific study, it was not until 2004 that important mutations in the kinase domain of the EGFR were first reported, which render tumor cells harboring the mutation particularly sensitive to the effects of small molecule tyrosine kinase inhibitors such as gefitinib or erlotinib.20,21 Similarly, mutations in the HER-2 kinase domain have only recently been described, which, it is hoped, may also identify tumors that will be uniquely sensitive to HER-2 kinase-targeted agents.22 Such observations serve as illustrations that highlight and emphasize the level of detail, in terms of understanding molecular biology and structure/function relationships, that we must strive to achieve for future development of molecularly targeted therapeutics. This will require unparalleled sophistication in molecular diagnostics unique to each target to achieve ideal patient selection based on molecular disease pathogenesis. If we do not have a strong commitment for clinical translational research for parallel development of diagnostics along with investigational therapeutics, then clinically important efficacy end points could easily be missed.
To illustrate this point, statistical simulations were performed to assess the impact of patient selection on the ability to observe a hypothetical treatment difference. In this series of simulations (1,000 runs per scenario), the patient population used is an unselected group of newly diagnosed metastatic breast cancer patients treated with first-line chemotherapy. This population is made up of two subgroups that are differentially responsive to the targeted therapy, both of whom have a median survival of 22 months when treated with active chemotherapy. Population A is sensitive to the targeted therapy and will have a Three simulations were performed in which varying proportions of the population A patients were included in the overall breast cancer population. In the three scenarios, the survival outcome was modeled assuming that population A represented 100%, 50%, and 25%, respectively, of the treated population (Fig 2). With decreasing frequency of population A, the survival curves converge and the significant survival benefit that population A derives from the addition of targeted therapy is obscured. These simulations demonstrate the danger of studying targeted therapies in an unselected patient population and highlight the need to understand the scientific basis—and thus the population likely to benefit—for treatment. Patient selection based on early development of molecular diagnostics will be essential to demonstrate the activity of targeted therapies in subset of patients.
In order to anticipate future development of molecularly targeted therapeutic approaches for the prevention or treatment of breast cancer, one needs only to look at past experience with targeted agents in breast cancer, such as those highlighted in this issue of the Journal of Clinical Oncology, in order to understand some of the lessons learned, which are likely to foreshadow future experience with such agents. Arguably, the most important advance in targeted therapy for breast cancer is the ability to target the estrogen receptor. Key to the success of hormonal therapy is the simple yet fundamental observation that there are multiple ways of intercepting estrogen receptor action (eg, inhibition of ligand biosynthesis, receptor antagonism, receptor degradation, and possibly perturbation of cross-talk with cell surface receptors), and that each of these mechanisms has the potential to elicit clinical efficacy, even during the clinical course of an individual patient who may respond to multiple lines of hormonal manipulation. It is entirely possible that this same story will play itself out again with targeted agents directed against HER-2, as well as other newer targets. For instance, it is now well established, based on crystal structure data, that trastuzumab binds to a juxtamembrane epitope of the HER-2 extracellular domain in a region that is normally the site of ectodomain cleavage.23 Such cleavage may, in fact, constitute one mode of activation of HER-2 kinase—a mode that can potentially be blocked by trastuzumab. In contrast to trastuzumab, another HER-2–targeted, humanized, monoclonal antibody, pertuzumab (Omnitarg; Genentech, South San Francisco, CA), binds to a distinct epitope that sterically hinders the HER-2 dimerization motif resulting in blockade of HER-2 dimerization with other HER receptors, thus inhibiting ligand-activated signaling via HER-2–containing heterodimers with EGFR or HER-3. These observations suggest that the two unique anti-HER-2 antibodies have very distinct biologic activities.24 Moreover, the non–cross reactivity of these two antibodies suggests the possibility of using them in combination. In fact preclinical data suggest that combining HER-2–targeted antibodies may be a more effective therapeutic strategy in breast cancer compared with treatment with single anti-HER-2 antibodies alone.25 In addition to antibodies, small molecule inhibitors of HER-2 kinase have recently been developed and are now undergoing clinical testing. Of particular interest is our observation of non–cross resistance between lapatinib (GW572016, a dual EGFR and HER-2 kinase inhibitor) and trastuzumab in cell lines selected for in vitro resistance to trastuzumab.26-28 Indeed, clinical responses have been observed upon lapatinib treatment of patients with prior objective disease progression on trastuzumab-containing regimens.29 Thus, in a fashion analogous to therapeutic targeting of the estrogen receptor, it may be possible to envision clinical use of various HER-2 inhibitors in sequence with resulting serial antitumor efficacy in at least a fraction of patients.28,29 Finally, we hypothesize that hormonal therapy and HER-2–targeted agents will not be the last word in targeted therapy for breast cancer. The clinical translational strategies proposed herein are but a few examples of potential pathways to achieve the goal of rapid identification and validation of novel targets suitable for a new wave of investigational agents with potential for clinical efficacy against breast cancer.
The following authors or their immediate family members have indicated a financial interest. No conflict exists for drugs or devices used in a study if they are not being evaluated as part of the investigation. Employment: Alex Bajamonde, Genentech; Pamela Klein, Genentech; Gwen Fyfe, Genentech. Consultant/Advisory Role: Mark D. Pegram, AstraZeneca, Genentech, GlaxoSmithKline. For a detailed description of these categories, or for more information about ASCO's conflict of interest policy, please refer to the Author Disclosure Declaration and Disclosures of Potential Conflicts of Interest found in Information for Contributors in the front of each issue.
Authors' disclosures of potential conflicts of interest are found at the end of this article.
1. Levine AJ: The genetic origins of neoplasia. JAMA 273:592, 1995 2. Pietras RJ, Pegram MD: Oncogene ativation and breast cancer progression, in Manni A (ed): Contemporary endocrinology: Endocrinology of Breast Cancer. Totowa, NJ, Humana Press Inc, 1999, pp 133-153 3. Li FP, Fraumeni JF Jr: Soft tissue sarcomas, breast cancer and other neoplasms: A familial syndrome? Ann Intern Med 71:747-752, 1969 4. Dickson RB, Lippman ME: Cancer of the breast, in DeVita VT Jr, Hellman S, Rosenberg SA (eds): Cancer: Principles & Practice of Oncology, 5th ed. Philadelphia, PA, Lippincott-Raven Publishers, 1997, pp 1541-1557 5. Hoskins K, Weber BL: Recent advances in breast cancer biology. Curr Opin Oncol 7:495-500, 1995[Medline]
6. Slamon DJ, Godolphin W, Jones LA, et al: Studies of the HER-2 proto-oncogene in human breast and ovarian cancer. Science 244:707-712, 1989 7. Courjal F, Louason G, Speiser P, et al: Cyclin gene amplification and overexpression in breast and ovarian cancers: Evidence for the selection of cyclin D1 in breast and cyclin E in ovarian tumors. Int J Cancer 69:247-253, 1996[CrossRef][Medline] 8. Pauletti G, Dandekar S, Smith K, et al: Chronology of HER-2/neu gene amplification in proliferative and malignant ductal lesions of the breast as determined by fluorescence in situ hybridization (FISH). Proc Am Assoc Cancer Res 38:414-415, 1997 (abstr 2777) 9. DePinho R, Jacks T: A bumper crop of cancer genes. Nat Genet 23:253-254, 1999[CrossRef][Medline] 10. Neil JC, Cameron ER: Retroviral insertion sites and cancer. Cancer Cell 2:253-255, 2002[CrossRef][Medline] 11. Mikkers H, Allen J, Knipscheer P, et al: High-throughput retroviral tagging to identify components of specific signaling pathways in cancer. Nat Genet 32:153-159, 2002[CrossRef][Medline] 12. Suzuki T, Shen H, Akagi K, et al: New genes involved in cancer identified by retroviral tagging. Nat Genet 32,166-174, 2002[CrossRef][Medline] 13. Nusse R, Varmus HE: Many tumors induced by the mouse mammary tumor virus contain a provirus integrated in the same region of the host genome. Cell 31:99-109, 1982[CrossRef][Medline]
14. Hwang HC, Martins CP, Bronkhorst Y, et al: Identification of oncogenes collaborating with p27Kip1 loss by insertional mutagenesis and high-throughput insertion site analysis. Proc Natl Acad Sci U S A 99:11293-11298, 2002 15. Lund AH, Turner G, Trubetskoy A, et al: Genome-wide retroviral insertional tagging of genes involved in cancer in Cdkn2a-deficient mice. Nat Genet 32:160-165, 2002[CrossRef][Medline] 16. Mikkers H, Allen J, Berns A: Proviral activation of the tumor suppressor E2a contributes to T cell lymphomagenesis in EmuMyc transgenic mice. Oncogene 21:6559-6566, 2002[CrossRef][Medline]
17. Akagi K, Suzuki T, Stephens RM, et al: RTCGD: Retroviral tagged cancer gene database. Nucleic Acids Res 32:Database issue D523-527, 2004 18. Theodorou V, Boer M, Weigelt B, et al: Fgf10 is an oncogene activated by MMTV insertional mutagenesis in mouse mammary tumors and overexpressed in a subset of human breast carcinomas. Oncogene 23:6047-6055, 2004[CrossRef][Medline] 19. Ferrick D: The oncogenome: A "fate map" for cancer. California State University Program for Education and Research in Biotechnology, January 2004 Symposium. http://www.csuchico.edu/csuperb/Sym04.html
20. Paez JG, Janne PA, Lee JC, et al: EGFR mutations in lung cancer: Correlation with clinical response to gefitinib therapy. Science 304:1497-1500, 2004
21. Lynch TJ, Bell DW, Sordella R, et al: Activating mutations in the epidermal growth factor receptor underlying responsiveness of non-small-cell lung cancer to gefitinib. N Engl J Med 350:2129-2139, 2004 22. Stephens P, Hunter C, Bignell G, et al: Lung cancer: Intragenic ERBB2 kinase mutations in tumours. Nature 431:525-526, 2004[Medline] 23. Cho HS, Mason K, Ramyar KX, et al: Structure of the extracellular region of HER2 alone and in complex with the Herceptin Fab. Nature 421:756-760, 2003[CrossRef][Medline] 24. Franklin MC, Carey KD, Vajdos FF, et al: Insights into ErbB signaling from the structure of the ErbB2-pertuzumab complex. Cancer Cell 5:317-328, 2004[CrossRef][Medline]
25. Nahta R, Hung MC, Esteva FJ: The HER-2-targeting antibodies trastuzumab and pertuzumab synergistically inhibit the survival of breast cancer cells. Cancer Res 64:2343-2346, 2004 26. Konecny GE, Venkatesan N, Beryt M, et al: Therapeutic advantage of a dual tyrosine kinase inhibitor (GW 2016) in combination with chemotherapy drugs or trastuzumab against human breast cancer cells with HER2 overexpression. Proc Am Assoc Cancer Res 43:1003, 2002 (abstr 4974)
27. Miller KD: The role of ErbB inhibitors in trastuzumab resistance. Oncologist 9:16-19, 2004 (suppl 3)
28. Britten CD: Targeting ErbB receptor signaling: A pan-ErbB approach to cancer. Mol Cancer Ther 3:1335-1342, 2004
29. Burris HA 3rd: Dual kinase inhibition in the treatment of breast cancer: Initial experience with the EGFR/ErbB-2 inhibitor lapatinib. Oncologist 9:10-15, 2004 (suppl 3) Submitted November 10, 2004; accepted November 23, 2004.
Related Editorial
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2005 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
DISCOVERY OF NEW THERAPEUTIC...
CHALLENGES IN DEVELOPING...
25,000 genes contained in the genome of a diploid human breast cancer cell, only a few have been proven to be altered in malignant progression. ErbB2, c-myc, and cyclin D1 are among oncogenes overexpressed and likely involved in the pathogenesis of human breast cancer. With information from studies of clinical cancer specimens, some distinct patterns of gene alteration are beginning to emerge. For example, in the early stages of tumorigenesis, cyclin D gene expression appears to be prevalent in non-comedo ductal carcinoma-in-situ,
