The ability of B-RAF inhibitors to induce unprecedented response rates in B-RAF mutant melanoma represents a watershed moment in the treatment of this dreadful disease. However, the efficacy of B-RAF inhibitors is short-lived as a result of acquired drug resistance. Augmenting the efficacy of B-RAF inhibitors to effect prolonged survival requires its combination with other targeted agents that block escape or compensatory tumor-survival pathways. Combinatorial targeted strategies based on B-RAF inhibitors require a thorough knowledge of resistance mechanisms. Successful combinatorial targeting of melanoma and other cancers is arguably the next frontier in accelerating personalized cancer medicine.
Optimism & opportunities
2010 has been called the ‘Year of Melanoma‘ by cancer scientists and physicians. In 2011, we witnessed the US FDA approval of a type I B-RAF inhibitor (B-RAFi; i.e., PLX4032/vemurafenib/Zelboraf™, Roche, Basel, Switzerland) and an immunomodulatory agent (ipilimumab/Yervoy™, Bristol-Meyers Squibb, NY, USA) for the treatment of advanced melanoma. Although B-RAFi can induce unprecedented response rates in excess of 50% among patients with V600E/KB-RAF melanomas, virtually all of those patients who initially respond later suffer from disease progression (i.e., acquired drug resistance) Citation[1–4]. Thus, overcoming B-RAFi resistance promises to significantly advance melanoma survivability.
The V600B-RAF mutation is the most common single-nucleotide mutation in cancer described to date, with an estimated incidence of 7–8% of all cancers. The fact that it is more prevalent in melanoma (50–60% of metastatic melanoma patients), and that the early clinical experience was highly positive in this indication, argues for studies of B-RAFi resistance in melanoma as being potentially instructive for B-RAFi-based treatment of other cancer types. V600B-RAF mutations can also be found in relatively high frequency in thyroid cancer, and less commonly in colorectal, ovarian, lung and other epithelial cancers. Full appreciation of the impact of this mutation and targeted inhibitors may still be premature. For example, B-RAF mutations were recently found in 100% of hairy-cell leukemias Citation[5].
Understanding the mechanisms of B-RAFi resistance (primary or de novo vs secondary or acquired) operative in patients will require a rigorous and collaborative effort to bank, annotate and analyze patient-derived tissues. With FDA approval of the first B-RAFi and the likely increasing availability of tissues sampled from patients, mechanisms inferred from model systems, such as melanoma cell lines primarily resistant to B-RAFi or those artificially derived to be B-RAFi resistant by continuous drug treatment and selection, can be validated in patient-derived tissues at an ever faster pace. In fact, direct application of discovery platforms (e.g., whole-genome sequencing, whole-exome sequencing, RNA-sequencing) on patient-derived tissues is already feasible and ongoing.
Progress
Few mechanisms of primary resistance to B-RAFi melanoma have been validated to be recurrent and significant in patients. Lack of melanoma tissues, which fail to respond to B-RAFi (defined by not reaching 30% tumor reduction by RECIST 1.1 criteria or tumor size increase at the specified time point of tumor assessment), is no doubt an impediment to progress in this area. In addition, it is yet to be validated that V600EB-RAF melanoma cell lines regarded as primarily resistant to B-RAFi in vitro, as determined typically by a short-course (e.g., 3-day) treatment protocol and an IC50 not reached at low micromolar drug concentrations, truly represent primary melanoma resistance to B-RAFi in the clinic.
Among the proposed mechanisms of acquired B-RAFi resistance, N-RAS mutations (restricted to hotspots) Citation[6], V600EB-RAF truncation (via alternative splicing creating novel exon–exon boundaries) Citation[7], and RTK (PDGFRβ, IGF1-R) Citation[6,8] overexpression have been identified at different study sites from multiple patients who have relapsed on either B-RAFi, vemurafenib/Zelboraf or GSK2118436 (GlaxoSmithKline, London, UK). N-RAS mutations and V600EB-RAF truncation, with the latter resulting in RAS-independent RAF dimerization, reactivate the MAPK pathway in the face of B-RAFi. Preclinical drug treatments predict restoration of B-RAFi sensitivity with MEK inhibitors (MEKi). Alternatively, RTK overexpression activates the PI3K-AKT pathway, serving to provide MAPK-redundant survival signaling. Here, preclinical drug treatments showed restoration of B-RAFi sensitivity with inhibitors of the PI3K–AKT pathway. In addition, although MEKi treatment failed to restore B-RAFi sensitivity in melanoma cell lines with acquired B-RAFi resistance driven by RTK overexpression, MEK inhibition synergized with dual PI3K–mTOR inhibition in growth inhibition and apoptosis induction Citation[9]. Thus, varied mechanisms of acquired B-RAFi resistance shown to recur in clinical samples appear to suggest two general modes of drug escape, namely those that reactivate the MAPK pathway and those that create MAPK-redundant signaling (e.g., RTK–PI3K–AKT signaling). Notably, secondary mutations in V600EB-RAF have not been identified in drug-resistant tumors (as well as in resistant cell lines derived in the laboratory) Citation[6], arguing for the importance of combinatorial targeted strategies to overcome B-RAFi resistance in melanoma.
Collectively, these mechanisms of acquired B-RAFi resistance (N-RAS mutations, V600EB-RAF truncation and RTK overexpression) are estimated to account for approximately 50% of clinical relapses on B-RAFi, with several caveats. In addition to the fact that the total number of tumors (some from the same patients), and especially the number of patients examined, is quite limited, most drug-resistant tumor biopsies have come from patients heavily pretreated with other agents (some mutagenic) prior to B-RAFi therapy. Moreover, most tissues for resistance studies are derived from cutaneous or subcutaneous lesions, raising a potential concern about how representative the spectra of mechanisms may be for visceral progressive melanoma tumors on B-RAFi therapy.
Understanding the full compendium of resistance mechanisms is thus key to designing combinatorial targeted treatments designed to prevent and/or overcome B-RAFi resistance in B-RAF mutant melanomas. Clinical trials designed to prevent (upfront combination) or overcome resistance (after development of acquired B-RAFi resistance) are both needed, the latter especially after the FDA approval of vemurafenib as a monotherapy and the expectation that most vemurafenib-treated patients will develop acquired drug resistance. There is theoretical concern for targeted monotherapy yielding tumors with a potentially more aggressive genetic or epigenetic constitution Citation[10]. Histologic evidence of this has been supported by a recent study in a different cancer. In melanomas with acquired B-RAFi resistance, this concern may be theoretically valid with certain adaptive mechanisms that lead to global gene-expression alterations Citation[6] or with switching of driver oncogenic genes (mutant B-RAF to mutant N-RAS).
Patients who have progressed on B-RAFi and treated subsequently with targeted combination agents (such as B-RAFi with MEKi or MEKi with dual PI3K inhibitors) have provided anecdotal evidence for altered molecular dependencies predicted by preclinical studies. In addition to combinations based on altered tumor dependencies, combinations based further on dynamic signal compensations and recoveries (via transcriptional, translational or post-translational feedbacks) to a matrix of targeted agents should be factored into combinatorial designs. This is an area where proteomic and its variant (e.g., phosphoproteomic, acetyproteomic) profiling could further highlight unexpected B-RAFi-based combinatorial targets.
A combinatorial approach may, in specific settings, even increase drug tolerability by reducing on-target related toxicities. Approximately 20–25% of melanoma patients treated with type I B-RAFi developed cutaneous squamous cell carcinomas (typically of the keratoacanthoma subtype). This is thought to occur via B-RAFi inducing a so-called paradoxical MAPK activation in the context of RAF dimerization driven by upstream RTK or RAS hyperactivity Citation[11]. In a striking parallel to overcoming mutant N-RAS-driven, acquired B-RAFi resistance in melanoma using the B-RAFi and MEKi combination, the same combination may in fact tilt the therapeutic index in favor of efficacy and concomitantly against toxicity Citation[12].
Challenges
Melanoma tumor heterogeneity lies at the heart of the challenge to overcome acquired B-RAFi resistance. This is exemplified at the genomic level by the relatively large number of somatic genetic alterations in metastatic melanomas Citation[13] and at the clinical level by the relatively short time to progression on B-RAFi treatment and the heterogeneous landscape of acquired resistance mechanisms that is emerging. Furthermore, tumors at distinct sites from the same patient may circumvent B-RAF inhibitors via distinct routes. But just as the genome is finite, mechanisms of resistance, de novo or acquired, should also be finite. Failures on targeted B-RAF inhibition afford us the opportunities to understand such finite events.
High-depth, whole-exome sequencing represents a currently cost-permissive approach to dissect the genetic complexity of melanoma tumors with acquired B-RAFi resistance. Such detailed genetic characterization of patient-matched baseline tumors (prior to B-RAFi therapy) may potential pinpoint pre-existing genetic factors that are predictive of specific type of genetic alterations (e.g., copy number gains vs single nucleotide variants) accounting for acquired drug resistance. Recent work with whole-exome sequence highlighted its potential to dissect the temporal relationship of oncogenic events Citation[14]. This type of analysis may help to provide insights into how potentially distinct pathways of tumor progression, up to the point of B-RAFi therapy, may predispose to or even predetermine the specific mechanism of acquired drug resistance under B-RAFi selection. Moreover, whole-exome sequencing can also provide data on genomic DNA copy number variations Citation[15], likely a causative source of acquired drug resistance.
Melanoma tumor heterogeneity, amplified by B-RAFi selective pressure, points to next-generation clinical diagnostics as a key to fully realizing personalized combinatorial therapies based on B-RAFi. Efforts to innovate combinatorial therapeutic strategies must be coupled to an equal effort to innovate diagnostic tools. Here, there are several facets of technologic development worth noting. One involves diagnostic platforms that can maximize the predictive power of baseline tumor biopsies. For instance, single-cell analysis of tumors or other clinical samples (e.g., blood) carrying traces of existing tumors promises to parse tumor heterogeneity and uncover minor genetic subpopulations that represent seeds of eventual resistance mechanisms. Another facet of innovation relates to reliable detection of resistance mechanisms from minute sampling of tumor cells (few thousand tumor cells). Ultimately, minute (and repeated) sampling from minimally invasive procedures to acquire tissues from easily accessible sites (e.g., peripheral blood for capture of circulating tumor cells) will enable real-time monitoring of tumor response (or lack thereof) to regimens of combined targeted agents. Pushing these technologic boundaries and their successful implementation in the clinical setting promise to maximize knowledge gained from clinical trials and advance the full potential of targeted cancer therapy.
Financial & competing interests disclosure
RS Lo acknowledges funding from the following: Burroughs Wellcome Fund, National Cancer Institute (K22CA151638), Melanoma Research Alliance, American Skin Association (Abby S and Howard P Milstein Research Scholar Award), Caltech-UCLA Joint Center for Translational Medicine, Sidney Kimmel Foundation for Cancer Research, Stand Up to Cancer/American Association for Cancer Research (Bud and Sue Selig Innovative Research Grant), Eli and Edythe Broad Center of Regenerative Medicine and Stem Cell Research, the Wesley Coyle Memorial Fund, Ian Copeland Melanoma Fund, Ruby Family Foundation, Shirley and Ralph Shapiro, Louis Belley and Richard Schnarr Fund, and The Seaver Institute. The author discloses a patent application (Patent Application Serial No. PCT/US11/61552). The author has no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.
No writing assistance was utilized in the production of this manuscript.
Additional information
Funding
References
- Chapman PB , HauschildA, RobertC et al. Improved survival with vemurafenib in melanoma with BRAF V600E mutation. N. Engl. J. Med. 364(26) , 2507–2516 (2011).
- Flaherty KT , PuzanovI, KimKB et al. Inhibition of mutated, activated BRAF in metastatic melanoma. N. Engl. J. Med. 363(9) , 809–819 (2010).
- Kefford R , ArkenauH, BrownMP et al. Phase I/II study of GSK2118436, a selective inhibitor of oncogenic mutant BRAF kinase, in patients with metastatic melanoma and other solid tumors. Presented at: 2010 ASCO Annual Meeting: Clinical Science Symposium, Oncogenes in Melanoma: Predictive Factors and Therapeutic Targets. Chicago, IL, USA, 4–8 June 2010 (Abstract 8503).
- Ribas A , KimKB, SchuchterLM et al. BRIM-2: an open-label, multicenter Phase II study of vemurafenib inpreviously treated patients with BRAF V600E mutation-positive metastatic melanoma. Presented at: 2011 ASCO Annual Meeting: Oral Abstract Session, Melanoma/Skin Cancers. Chicago, IL, USA, 4–8 June 2010 (Abstract 8509).
- Tiacci E , TrifonovV, SchiavoniG et al. BRAF mutations in hairy-cell leukemia. N. Engl. J. Med.364(24) , 2305–2315 (2011).
- Nazarian R , ShiH, WangQ et al. Melanomas acquire resistance to B-RAF(V600E) inhibition by RTK or N-RAS upregulation. Nature 468(7326) , 973–977 (2010).
- Poulikakos PI , PersaudY, JanakiramanM et al. Acquired resistance to RAF inhibitors is mediated by splicing isoforms of BRAF (V600E) that dimerize in a RAS independent manner. Nature doi:10.1038/nature10662 (2011) (Epub ahead of print).
- Villanueva J , VulturA, LeeJT et al. Acquired resistance to BRAF inhibitors mediated by a RAF kinase switch in melanoma can be overcome by cotargeting MEK and IGF-1R/PI3K. Cancer Cell 18(6) , 683–695 (2010).
- Shi H , KongX, RibasA, LoRS. Combinatorial treatments that overcome PDGFRβ-driven resistance of melanoma cells to B-RAF(V600E) inhibition. Cancer Res.71(15) , 5067–5074 (2011).
- Sequist LV , WaltmanBA, Dias-SantagataD et al. Genotypic and histological evolution of lung cancers acquiring resistance to EGFR inhibitors. Sci. Transl. Med. 3(75) , 75ra26 (2011).
- Su F , VirosA, MilagreC et al. RAS mutations in cutaneous squamous cell carcinomas with BRAF inhibitors. N. Engl. J. Med. doi: 10.1200/JCO.2011.36.7680 (2011) (Epub ahead of print).
- Infante JR , FalchookGS, LawrenceDP et al. Phase I/II study to assess safety, pharmacokinetics, and efficacy of the oral MEK 1/2 inhibitor GSK1120212 (GSK212) dosed in combination with the oral BRAF inhibitor GSK2118436 (GSK436). Presented at: 2011 ASCO Annual Meeting: Clinical Science Symposium, BRAF: From Biology to Patients. Chicago, IL, USA, 4–8 June 2010 (Abstract CRA8503).
- Pleasance ED , CheethamRK, StephensPJ et al. A comprehensive catalogue of somatic mutations from a human cancer genome. Nature 463(7278) , 191–196 (2010).
- Durinck S , HoC, WangNJ et al. Temporal dissection of tumorigenesis in primary cancers. Cancer Discov. 1(2) , 137–143 (2011).
- Sathirapongsasuti JF , LeeH, HorstBA et al. Exome sequencing-based copy-number variation and loss of heterozygosity detection: exome CNV. Bioinformatics 27(19) , 2648–2654 (2011).