Oncogene addiction: pathways of therapeutic response, resistance, and road maps toward a cure
2015; Springer Nature; Volume: 16; Issue: 3 Linguagem: Inglês
10.15252/embr.201439949
ISSN1469-3178
AutoresRaymond Pagliarini, Wenlin Shao, William R. Sellers,
Tópico(s)PARP inhibition in cancer therapy
ResumoReview13 February 2015free access Oncogene addiction: pathways of therapeutic response, resistance, and road maps toward a cure Raymond Pagliarini Raymond Pagliarini Department of Oncology, Novartis Institutes for BioMedical Research, Cambridge, MA, USA Search for more papers by this author Wenlin Shao Wenlin Shao Department of Oncology, Novartis Institutes for BioMedical Research, Cambridge, MA, USA Search for more papers by this author William R Sellers Corresponding Author William R Sellers Department of Oncology, Novartis Institutes for BioMedical Research, Cambridge, MA, USA Search for more papers by this author Raymond Pagliarini Raymond Pagliarini Department of Oncology, Novartis Institutes for BioMedical Research, Cambridge, MA, USA Search for more papers by this author Wenlin Shao Wenlin Shao Department of Oncology, Novartis Institutes for BioMedical Research, Cambridge, MA, USA Search for more papers by this author William R Sellers Corresponding Author William R Sellers Department of Oncology, Novartis Institutes for BioMedical Research, Cambridge, MA, USA Search for more papers by this author Author Information Raymond Pagliarini1, Wenlin Shao1 and William R Sellers 1 1Department of Oncology, Novartis Institutes for BioMedical Research, Cambridge, MA, USA *Corresponding author. Tel: +1 617 871 7069; E-mail: [email protected] EMBO Reports (2015)16:280-296https://doi.org/10.15252/embr.201439949 See the Glossary for abbreviations used in this article. PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info Abstract A key goal of cancer therapeutics is to selectively target the genetic lesions that initiate and maintain cancer cell proliferation and survival. While most cancers harbor multiple oncogenic mutations, a wealth of preclinical and clinical data supports that many cancers are sensitive to inhibition of single oncogenes, a concept referred to as ‘oncogene addiction’. Herein, we describe the clinical evidence supporting oncogene addiction and discuss common mechanistic themes emerging from the response and acquired resistance to oncogene-targeted therapies. Finally, we suggest several opportunities toward exploiting oncogene addiction to achieve curative cancer therapies. Glossary ADC antibody–drug conjugate AKT Ak thymoma kinase, key member of the PI3K pathway ALK anaplastic lymphoma kinase, activated by gene translocation in cancer APL acute promyelocytic leukemia AR androgen receptor, lineage driver of prostate cancer ATP adenosine triphosphate ATRA all-trans retinoic acid BCR-ABL B-cell receptor–Abelson kinase, oncogenic fusion protein that drives CML BRAF v-Raf murine sarcoma viral oncogene homolog B, activated by point mutations in cancer BRD4 bromodomain containing 4, chromatin modulator activated by translocation in cancer BTK Bruton's tyrosine kinase, key member of oncogenic B-cell receptor signaling CCLE cancer cell line encyclopedia CGP cancer genome project CML chronic myelogenous leukemia CRAF v-Raf murine sarcoma viral oncogene homolog B CRISPR/CAS9 clustered regularly interspaced short palindromic repeats/Cas9 nuclease; eukaryotic gene editing technology derived from a prokaryotic viral immune editing system DM1 maytansine derivative, toxic payload often linkered to ADC's DNA deoxyribonucleic acid DUSP dual specificity phosphatase, negative regulator of MAPK pathway EGFR epidermal growth factor receptor, activated by point mutation and small deletions in cancer ER estrogen receptor, lineage driver for breast and other cancers ERK extracellular signal-regulated kinase, member of MAPK pathway EZH2 enhancer of zeste homolog 2, activated by point mutation in cancer FGFR fibroblast growth factor receptor, activated by gene fusion and point mutation in cancer FLT3 Fms-like tyrosine kinase 3, activated by ITD and point mutation in cancer GIST gastrointestinal stromal tumor HER2 human epidermal growth factor receptor 2 HER3 human epidermal growth factor receptor 3 IDH1 isocitrate dehydrogenase 1, activated by point mutation in cancer IDH2 isocitrate dehydrogenase 2, activated by point mutation in cancer IGF-1R insulin-like growth factor 1 receptor, feedback activator of oncogenic signaling IL-6 interleukin 6 cytokine ITD internal tandem duplication, a common mutation of the FLT3 oncogene KIT v-kit Hardy–Zuckerman 4 feline sarcoma viral oncogene homolog KRAS Kirsten RAS viral oncogene homolog, activated by point mutation in cancer MAP2K1 mitogen-activated protein kinase kinase 1 MAP2K2 mitogen-activated protein kinase kinase 2 MEK MAPK/ERK kinase, alternate common name for MAPK2K1, MAPK2K2, and/or MAPK pathway MAPK mitogen-activated protein kinase MITF microphthalmia-associated transcription factor, lineage driver for melanoma MYB v-myb avian myeloblastosis viral oncogene homolog NRAS neuroblastoma RAS viral oncogene homolog, activated by point mutation in cancer NSCLC non-small cell lung cancer NSD2 Wolf–Hirschhorn syndrome candidate 1 PI3K phosphoinositide 3-kinase (pathway) PIK3CA phosphatidylinositol-4,5-bisphosphate 3-kinase, catalytic subunit alpha, activated by point mutation in cancer PTEN phosphatase and tensin homolog, tumor suppressor and negative regulator of PI3K pathway RAS rat sarcoma viral oncogene; generic name for KRAS, NRAS, HRAS oncogenes, and/or the signaling pathway RNAi RNA interference ROS1 ROS proto-oncogene 1 RTK receptor tyrosine kinase SPRY Sprouty homologs (e.g., SPRY1, SPRY2), negative regulators of RTK signaling TERT telomerase reverse transcriptase, activated by promoter point mutations in cancer Introduction Cancer is a disease resulting from the acquisition of somatic genetic alterations. The results of extensive cancer genome sequencing and myriad preclinical in vitro and in vivo functional studies have underscored that cancers are initiated and maintained by recurrent ‘driver’ oncogene and/or tumor suppressor gene mutations. Established cancers in humans harbor, on average, approximately 30–60 mutations capable of altering protein function, with cancers such as melanoma bearing roughly 200 protein function-altering mutations per tumor 1. A key goal of cancer therapeutics development is to selectively target somatic cancer mutations—however, targeting all of these alterations in any one cancer seems a daunting task. Although cancer develops through progressive gene mutations that activate a variety of oncogenic functions, compelling evidence from preclinical studies, and most importantly from cancer patients treated with oncogene-targeted therapeutics, suggests that cancer cell survival relies on relatively few key genetic driver events. The term ‘oncogene addiction’ was coined to describe this phenomenon of exquisite cancer cell dependence on individual oncogenes to sustain the malignant phenotype 2. Clinical evidence for oncogene addiction As we focus herein on the clinical evidence for oncogene addiction, we direct the reader to excellent reviews of the preclinical data supporting oncogene and non-oncogene addiction 34. A prime clinical example of oncogene addiction is in CML. CML is driven by the BCR-ABL mutant oncogene, produced as a result of chromosome 9:22 translocation, otherwise known as the ‘Philadelphia’ chromosome 56. While preclinical studies provided evidence that BCR-ABL was a bona fide oncogene both in vitro and in vivo 78, addiction of CML to BCR-ABL was demonstrated in patients through the profound clinical responses attained with the kinase inhibitor imatinib, which targets BCR-ABL. This addiction was further reinforced by the description of genetic mechanisms of resistance that largely led to reactivation of BCR-ABL kinase activity (described in more detail below). These observations in aggregate provided a transformative proof-of-concept for oncogene-targeted cancer therapy 9. As summarized in Table 1 (and referenced therein), the strategy of targeting mutant oncogenic kinases has now been repeated many times over in a variety of cancer types. The common theme from these studies is a marked improvement in initial patient responses when oncogene-targeted therapies, tested in the correct oncogene-mutated patient population, are compared head-to-head with prior standard of care therapeutics. Table 1. Examples of approved oncogene-targeted therapies and observed resistance mechanisms in patients Target/indication Inhibitor(s) Observed clinical responses Resistance mechanisms Secondary oncogene mutation Pathway mutations Bypass BCR-ABL mutant CML Imatinib, nilotinib, dasatinib, ponatinib Complete cytogenetic responses: 65–80% 9112113166 T315I and other mutations, BCR-ABL amplification 26167168 SRC family upregulation FGF2/FGFR3 activation 168169 KIT mutant GIST Imatinib 53.7% partial response in patients with refractory disease 170 KIT mutations (e.g., V654A, T670I) or amplification 171172 PDGFRA mutation 172 Rhabdomyosarcomatous differentiation 109 BRAF mutant melanoma Vemurafenib, dabrafenib 45–51% response rate; benefits observed versus prior standard of care 5859173 P61-BRAF splice variant 28; BRAF amplification 174 NRAS, NF1, MAP2K1, MAP2K2 mutation MITF amplification 313233175176 PI3K pathway mutations; CRAF, RTK, COT, AXL upregulation 293133110175176177 EGFR mutant NSCLC Gefitinib, erlotinib, afatinib 9–13 months progression-free survival; 73.7% response rate for gefitinib; benefit versus standard chemotherapy 178179180181182 T970M mutation (+/− gene amplification ~40–65%); other EGFR point mutations (~1–2%) 105183184185 PIK3CA mutation, BRAF mutation (~1%) 105185 MET or HER2 amplification, histologic transformation (EMT, SCLC ~12–14%) 3738105185186 EGFR-amplified colorectal cancer Cetuximab, panitumumab Improvements in progression-free survival versus best supportive care 187 KRAS, BRAF, PIK3CA, PTEN mutation 187 ALK-translocated NSCLC Crizotinib, ceritinib, alectinib 55–65% response rate; improved response rate versus standard chemotherapy 116117188 L1196M, I1171T, V1180L, and other mutations, with or without amplification (~28–65%) 185189190191 KRAS mutation 191 KIT amplification, EGFR upregulation or mutation, IGF-1R upregulation 190191192 HER2/ERBB2-amplified breast cancer Trastuzumab, lapatinib, pertuzumab Trastuzumab: 33% combined complete and partial response rate 193; Lapatinib: 39% partial response rate 194 Trastuzumab epitope mutations: p95-HER2, D16 27195 PIK3CA/PTEN mutation 196 EGFR, HER3, HER4, IGF-1R, MET upregulation and heterodimerization 195196 ROS1-translocated NSCLC Crizotinib 72% objective response rate 197 G2032R mutation 198 RET mutant medullary thyroid carcinoma (MTC) Vandetanib 46% objective response rate in patients with hereditary MTC harboring RET mutation 199 Retinoic acid receptor (RARA)-translocated APL ATRA Complete response rates of > 90%; superior to prior chemotherapy regimens 200 Ligand-binding domain mutation (~40%) 11201 AR-positive castration-resistant prostate cancer Enzalutamide 18.4-month overall survival, 54% PSA reduction 202 F876L mutation 99100101, AR-V7 ligand-binding domain truncation 203 GR upregulation 203 ER-positive metastatic breast cancer Tamoxifen, toremifene, fulvestrant, letrozole, anastrozole, exemestane Tamoxifen: approximately 50% drop in mortality with 10 years of treatment 204 Ligand-binding domain mutation 30205 The oncogene-addicted phenotype is not unique to mutated kinases. One of the earliest examples of targeted therapy (albeit one where clinical efficacy was established prior to molecular cloning of the causative oncogene) was the use of ATRA in APL. APL bears characteristic translocations affecting the retinoic acid receptor, generating fusion proteins such as PML-RARA that interfere with normal cell differentiation 10. ATRA binds to the ligand-binding domain of PML-RARA, which inhibits its oncogenic function 11. An additional example is the use of antiandrogens for the treatment of prostate cancers, which are ‘lineage-addicted’ 12 to AR and bear recurrent AR amplifications or mutations upon resistance to first-line therapies 1314. Finally, recent cancer genome sequencing has revealed a prevalent novel class of mutated oncogenes involved in the regulation of epigenetic states 15. Examples include oncogenic point mutations or chromosomal translocations affecting EZH2, NSD2, BRD4, IDH1, and IDH2. Given the aforementioned history of cancer addiction to mutated driver oncogenes, as well as emerging preclinical studies demonstrating the dependence on these oncogenes for tumor maintenance, drugs targeting these lesions have been rapidly developed 1617181920. Several of these have already entered early clinical investigation, with encouraging initial responses 21222324. Together, these striking results demonstrate that the concept of oncogene addiction indeed translates into clinical responses. Therapeutic resistance reveals oncogenic pathway addiction Despite the robust initial clinical responses described above, chronic exposure to most targeted therapeutics often gives way to relapse, and cures remain elusive. Does this argue against oncogene addiction? Answers lie in the observed clinical mechanisms of resistance to oncogene-targeted therapeutics. As detailed below, three common themes emerge upon resistance to many oncogene-targeted therapies; these themes demonstrate that most cancers retain an underlying addiction to oncogene-induced signaling pathways, if not a monolithic addiction to the originally mutated oncogene. Secondary alterations of the oncogene drug target Single-agent BCR-ABL inhibition often results in cancer cell apoptosis and profound long-term responses 925; however, a significant fraction of patients show resistance to existing therapies. The main observed mechanism of resistance to BCR-ABL inhibition is the acquisition of second-site mutations in BCR-ABL itself 26. Predominant among these is mutation of the ATP-binding pocket at the ‘gatekeeper’ residue threonine 315. Mutation at this site prevents optimal binding of imatinib and other inhibitors, while still allowing ATP hydrolysis, and hence restoring BCR-ABL signaling in the presence of inhibitors (Fig 1A). Treatment of lung cancers with drugs targeting mutant EGFR, ALK, and ROS1 also results in a significant fraction of resistant disease bearing second-site oncogene mutations that restore oncogene function in the presence of drug. These acquired mutations often occur within the highly conserved gatekeeper residue (Table 1). The HER2 oncogene commonly develops resistance to the humanized HER2 antibody trastuzumab in a slightly different fashion; in this case, the trastuzumab-binding epitope is lost, while oncogene function is retained 27. BRAF obtains resistance to kinase inhibitors, at least in part, through either kinase amplification or truncations that further activate kinase activity 2829. Outside of kinases, AR, PML-RAR, and ER (a lineage driver for many breast cancers) acquire mutations in their ligand-binding domains that reduce or abrogate drug efficacy (Table 130) while restoring oncogene function. The common theme of treatment-acquired secondary oncogene alterations is that they provide resistance to therapy while reinstating oncogene function—this clinically observed resistance mechanism makes the most compelling argument for oncogene addiction. Figure 1. Common mechanisms of resistance to oncogene-targeted therapeutics(A) Second-site mutations can reinstate oncogene function while abrogating inhibitor activity, as exemplified by BCR-ABL gatekeeper mutations as an inhibitor resistance mechanism. (B) Mutations in oncogene pathway components can reinstate pathway signaling despite continued oncogene inhibition, as exemplified by MAP2K1 mutations as a resistance mechanism for BRAF inhibitors. (C) Mutational or non-mutational activation of bypass signaling pathways can render cancer cells independent of the original oncogene, as exemplified by MET activation as a resistance mechanism for EGFR inhibition. Download figure Download PowerPoint Activating mutations in oncogenic pathway components Acquired resistance to oncogene-targeted drugs also occurs via mutation of alternate components of oncogene-induced signaling pathways. For example, mutant BRAF signals through the MAPK signaling pathway to promote melanoma growth. As such, one key resistance mechanism to BRAF inhibitors such as vemurafenib is the acquisition of activating mutations in other known MAPK signaling pathway components such as NRAS 31, or more rarely MAP2K1, and MAP2K2 3132; loss of function mutations in the negative MAPK pathway regulator NF1 3133; or amplification and activation of the MAPK pathway target gene MITF, a lineage driver of melanoma 31. All of these mutations restore MAPK oncogenic pathway signals despite continued pharmacological inhibition of mutant BRAF (Fig 1B). This theme recurs in NSCLC, where activation of RTKs such as EGFR and ALK are key driver events. RTKs signal through several intracellular pathways, including the MAPK and PI3K pathways. As such, acquired resistance to HER2-, EGFR-, and ALK-targeted therapies includes selection for activating mutations in the MAPK or PI3K pathways (Table 1 and references therein). As with second-site mutation of the oncogene itself, resistance mutations in key members of an oncogenic signaling pathway highlights that many cancers retain dependence upon specific oncogenic pathways, if not always the oncogene itself. Induction of bypass pathway signaling The third common theme in acquired resistance is the induction of bypass signaling pathways. A key example is observed in resistance to BRAF inhibitors. RAF family proteins normally function as dimers; common oncogenic mutations in BRAF (e.g., V600E) allow monomeric BRAF proteins to activate downstream signaling pathways in the absence of upstream activating signals 28. This activity is blocked by selective BRAF kinase inhibitors. However, CRAF, another key RAF isoform, can still activate the MAPK pathway in the presence of upstream pathway signals. This commonly occurs through the induction of growth factor-dependent and/or RAS-dependent signals 34. Such upstream activating signals are often paradoxically induced by oncogene inhibition, as discussed in more detail below. Despite mutant BRAF inhibition, upstream pathway activation can still signal through CRAF to reinstate MAPK pathway signaling. This is facilitated through the activation of CRAF homodimers, or through CRAF:BRAF heterodimers that are stabilized in the presence of some BRAF inhibitors 343536. Induction of oncogene bypass signaling is not unique to BRAF inhibitors. As discussed in more detail below, acquired resistance to EGFR-, HER2-, and ALK-targeted therapeutics commonly occurs through the upregulation or amplification of alternate RTK's (Table 1). This bypasses the need for the mutated oncogene, but often reinstates the original downstream signaling pathways. A key example is the selection for activated MET signaling as a resistance mechanism to EGFR-targeted therapies 3738 (Fig 1C). As another recent example, resistance to the AR inhibitor enzalutamide can be caused by glucocorticoid receptor (GR)-dependent bypass of AR signaling 39. While many bypass resistance mechanisms reinstate the same signaling pathways originally activated by the oncogene, this is not always the case. For acquired resistance to BRAF inhibitors in melanoma, mutational activation of the PI3K-PTEN-AKT pathway has been identified as a prevalent mechanism in bypassing tumor dependence on MAPK signaling 2940. In summary, despite the apparent mechanistic diversity of acquired resistance to oncogene-targeted therapies, the three major themes of resistance outlined above demonstrate that most cancers retain addiction to specific oncogene-activated pathway signals. This suggests that the key dependencies of cancer cells remain tractable despite acquired resistance and that a better knowledge of resistance mechanisms can lead to rational therapeutic strategies that reduce or prevent resistance in the clinic. ‘Oncogenic shock’ as a model to understand response versus resistance to therapy While many oncogene-addicted cancers show striking initial responses to targeted therapies, the heterogeneity of response within and across cancers must be noted. Why is the proportion of response and resistance so different between different oncogenes—for example, why are durable single-agent responses often seen with BCR-ABL inhibition in CML, while inhibitors of FLT3-ITD in AML appear to only provide transient benefits 4142? Similarly, why does inhibition of the same oncogene have divergent responses in different cancer types—as exemplified by a ~50% response rate to BRAF inhibitors in BRAF mutant melanoma, but a less than 5% response rate in BRAF mutant colorectal cancers 43? Such inconsistencies could be explained by an inability to achieve complete and sustained target inhibition in different tumor types, due to pharmacological limitations across different drugs and among different patient populations 44. However, emerging data demonstrate that intrinsic biological differences across oncogenes and tumor types also exist. A useful paradigm to understand the biological diversity of responses to oncogene inhibition is that of ‘oncogenic shock’ 4546. This hypothesis builds on the knowledge that activated oncogenes promote proliferation and survival, but at the same time paradoxically activate signals that promote arrest or apoptosis 4748495051. Upon acute inactivation of oncogene signaling, the timing of how these two pathways respond may differ for different oncogenes, or in different contexts. If the oncogenic pathway is quickly blocked by a drug, while the paradoxical oncogene-activated growth inhibitory pathway is slow to turn off, then apoptosis, or oncogenic shock, prevails. Conversely, if the paradoxical growth inhibitory signals from the oncogene can quickly reset, this provides a scenario where cells may survive to become resistant to oncogene inhibition. This differential in pathway response, presumably due to differences in the turnover of signaling proteins such as phosphatases that negatively regulate discreet prosurvival or proapoptotic pathways 46, may explain why some oncogenes show more profound responses than others upon acute inhibition. What are the mechanisms that allow oncogenic shock phenotypes to occur—or to be bypassed—and are they common among cancers? An early view into the mechanism of paradoxical oncogene-induced growth inhibitory pathways in cancer cells was afforded by a study of the differential sensitivity of BRAF mutant versus RTK-activated cancer cells to MAPK pathway inhibition 52. This study suggested that mutant BRAF activates a unique ERK-dependent transcriptional output, including the upregulation of DUSP phosphatases and the SPRY family of secreted RTK inhibitory proteins, both of which negatively regulate MAPK pathway signaling 5354. BRAF inhibition blocks MAPK-dependent growth signaling, but also shuts off MAPK-dependent SPRY expression, which relieves SPRY-dependent inhibition of HER-family-, FGFR-, and/or IGF-1R-dependent responsiveness to exogenous growth factors 5556. Such feedback appears to be particularly active in BRAF mutant colorectal cancers via the rapid activation of EGFR upon BRAF inhibition 57. While this explains the limited efficacy of BRAF inhibitors in this indication, it also provides the rationale for dual inhibition of BRAF and EGFR in BRAF mutant colorectal cancers 4357. In melanoma, paradoxical feedback pathway activation (as well as most other clinically observed resistance mechanisms) reinstates MAPK signaling, providing the rationale for dual BRAF/MEK inhibition 5859. While BRAF signaling inhibition has become a paradigm for paradoxical feedback pathway activation, oncogenic alterations in EGFR, HER2, ALK, and MET also function through a MAPK-dependent feedback pathway to block IL-6-facilitated activation of STAT3 and PI3K pathway-mediated survival signals 60. Also like BRAF, BCR-ABL and FLT3-ITD oncogenes can block the expression of growth factor receptors via a MAPK pathway-dependent feedback mechanism 61. The contrasting behavior of BCR-ABL- and FLT3-ITD-dependent feedback responses noted in an isogenic cell background 61 is of particular interest for the oncogenic shock hypothesis: pulsed inhibition of BCR-ABL rapidly shuts down BCR-ABL-dependent downstream survival signaling (including MAPK pathway signaling), but BCR-ABL- and MAPK pathway-dependent inhibition of normal growth factor-dependent signaling is slow to revert back to its basal state. This creates a window of time where no prosurvival signals are present, resulting in apoptosis. While pulsed FLT3 inhibition in FLT3-ITD mutant expressing cells similarly inhibits oncogenic signaling, MAPK pathway-dependent negative feedback is rapidly lost and growth factor-dependent signaling pathway signaling is quickly restored. In this setting, there is not enough time for apoptosis to be induced before the FLT3-ITD-inhibited cells restore functional growth factor receptor-dependent signaling. Together, these data provide evidence that oncogene-dependent feedback inhibition of growth factor-dependent signaling may be pervasive across many cancers (even BCR-ABL mutant CML). These studies furthermore suggest that the turnover rate of these feedback mechanisms in different cancers dictates the fine line between oncogenic shock versus the activation of bypass resistance mechanisms (e.g., BCR-ABL inhibition in CML versus BRAF inhibition in colorectal cancer). Finally, the data implicate the MAPK pathway as a key node regulating the oncogene-induced feedback inactivation of growth factor receptor signaling (Fig 2). Figure 2. Oncogenic shock versus feedback reactivation of growth factor receptor signaling(A) Many oncogenes actively suppress growth factor receptor (GF-R)-dependent signaling in addition to activating oncogenic pathway signaling. (B) Oncogenic shock may predominate when GF-R survival signaling is slow to reinstate after oncogene inhibition (tortoise), creating a window where cells have no prosurvival signals. (C) Bypass resistance may predominate when GF-R responsiveness is quickly reactivated upon oncogene inhibition (hare). Download figure Download PowerPoint While MAPK pathway-dependent feedback may commonly attenuate oncogenic shock responses across cancers, alternate mechanisms have been reported. First, altered epigenetic regulation can generate ‘drug-tolerant’ states that allow for the survival of small subpopulations within otherwise treatment-sensitive cancer cells; despite this alternate mechanism of resistance, survival is still generated through the upregulation of growth factor receptors such as IGF-1R 62. Second, PI3K-AKT-dependent feedback pathways have also been reported 6364—once again, resistance is driven through relief of oncogene-induced negative feedback regulation of growth factor receptors. A particularly interesting case is in prostate cancer—oncogenic PI3K and AR activation exists in a vicious cycle where inhibition of either pathway results in feedback upregulation of the other, via a mechanism that induces EGFR family RTK signaling 65. As above, knowledge of these feedback pathways sheds light onto rationally designed therapeutic combinations to prevent resistance—in this case, dual inhibition of AR and PI3K signaling pathways shuts down paradoxical bypass signaling and achieves remarkable efficacy in preclinical models. A Road map for targeting oncogene addiction The clinical benefits observed with agents targeting mutated oncogenes provide hope that the ‘one-step remedy’ to oncogene addiction initially proposed by Weinstein 2 may be attainable for some cancers. However, the common patterns of resistance to oncogene-targeted therapies must be anticipated and intercepted in order to achieve deep and sustained clinical benefit, and ultimately cures. The road map to curative therapy will require a rationally designed ‘one-two punch’ with combinations of targeted agents, rather than a one-step remedy. Below, we offer some suggestions to reach this goal. Genetically define all cancers Classical characterization of cancer subtypes include distinctions based on tissue type, histology, pathology, and level of differentiation, characteristics that are open to biased interpretation 66. Targeted therapies require a genetic definition for the patient populations most likely to respond. The most straightforward example is that of CML, where the BCR-ABL fusion defines the disease better than any histologic distinction. This is precisely because the genetic lesion directly impacts the therapeutic strategy (i.e., sensitivity to BCR-ABL inhibitors). This paradigm extends to other tumor types with more heterogeneous genetic etiology. Clinical trials of the EGFR inhibitor gefitinib in otherwise non-selected NSCLC patients originally failed to show strong differences in overall survival. It was not until the ‘outlier’ patients who responded to therapy 67 were retrospectively analyzed for EGFR mutations that the patterns of response in NSCLC could be rationalized 686970. EGFR mutant NSCLC is therefore a key clinical subtype. This initial finding now extends to other mutations in NSCLC that are paired with targeted therapeutics (e.g., ALK translocations and ceritinib). An initially homogeneous histological subtype of lu
Referência(s)