Melanoma Prognostics and Personalized Therapeutics at a Crossroad

COMMENTARY

NKG2D ligands and MHC class I. J Immunother 35:245–53 Paschen A, Sucker A, Hill B et al. (2009) Differential clinical significance of individual NKG2D ligands in melanoma: soluble ULBP2 as an indicator of poor prognosis superior to S100B. Clin Cancer Res 15:5208–15 Rosental B, Appel MY, Yossef R et al. (2012) The effect of chemotherapy/radiotherapy on cancerous pattern recognition by NK cells. Curr Med Chem 19:1780–91 Ugurel S, Schadendorf D, Pfo¨hler C et al. (2006) In vitro drug sensitivity predicts response and survival after individualized sensitivity-

directed chemotherapy in metastatic melanoma: a multicenter phase II trial of the Dermatologic Cooperative Oncology Group. Clin Cancer Res 12:5454–63 Vetter CS, Groh V, thor Straten P et al. (2002) Expression of stress-induced MHC class I related chain molecules on human melanoma. J Invest Dermatol 118:600–5 Walter S, Weinschenk T, Stenzl A et al. (2012) Multipeptide immune response to cancer vaccine IMA901 after single-dose cyclophosphamide associates with longer patient survival. Nat Med; e-pub ahead of print 29 July 2012

See related article on pg 509

Melanoma Prognostics and Personalized Therapeutics at a Crossroad Roger S. Lo1,2 Melanoma has recently emerged as a poster child for targeted therapies and immunotherapies, with game-changing BRAF and immune checkpoint inhibitors now in clinical trials and with approved clinical indications. One highly anticipated use of novel therapeutic agents is in the adjuvant setting. Adjuvant BRAF and/or immune checkpoint inhibition may positively affect the survival of melanoma patients diagnosed at earlier stages but still at high risk for postsurgical relapses. BRAF V600 mutations and, potentially, melanoma-associated immunity are predictive biomarkers for responses to these novel therapies. Emerging evidence points to these predictive biomarkers doubling as prognostic biomarkers for highrisk stage III patients, promising to help stratify these patients for the application of novel adjuvant therapies. Journal of Investigative Dermatology (2013) 133, 292–295. doi:10.1038/jid.2012.386

Inhibition of BRAF or immune checkpoint in melanoma

Clinical exploitation of BRAF oncogene addiction and tumor immunogenicity in advanced melanoma has recently culminated in the Food and Drug Administration (FDA)’s approval of a BRAF inhibitor (BRAFi), vemurafenib, and an immune checkpoint inhibitor, ipilimumab. This unprecedented success

has led to paradigm shifts in therapeutic concepts (and in the design of clinical trials). In the case of adenosine triphosphate-competitive small-molecule inhibitors of mutant BRAF, success has meant that the presence of BRAF V600 mutations can predict partial (in the majority of those treated) or complete (in a minority of those treated) melanoma shrinkage in 50–60% of patients (Chapman

1

Department of Medicine, Division of Dermatology, David Geffen School of Medicine, University of California, Jonsson Comprehensive Cancer Center, Los Angeles, California, USA and 2Department of Molecular and Medical Pharmacology, David Geffen School of Medicine, University of California, Jonsson Comprehensive Cancer Center, Los Angeles, California, USA Correspondence: Roger S. Lo, Department of Medicine, Division of Dermatology, David Geffen School of Medicine, University of California, Jonsson Comprehensive Cancer Center, 10833 Le Conte Avenue, Los Angeles, California 90095-1750, USA. E-mail: [email protected]

292

Journal of Investigative Dermatology (2013), Volume 133

et al., 2011; Sosman et al., 2012). In other words, the BRAF V600 mutation is a therapeutic biomarker. Across the board, in all melanoma patients with BRAF V600 mutations, it is reasonable to believe that the timing of intervention at an advanced disease stage with overt systemic involvement limits the initial extent and the eventual durability of clinical responses. Although not statistically significant, lower serum lactate dehydrogenase before vemurafenib treatment was associated with higher overall responses (Jeffrey A. Sosman and Antoni Ribas, personal communication). Thus, BRAF inhibition earlier in the course of disease may diminish the frequency of partial initial responses and the almost universal occurrence of acquired (secondary) drug resistance. The mechanisms of BRAFi resistance are varied, but often involve reactivation of the mitogenactivated protein kinase (MAPK) signaling pathway, with concomitant activation of the RTK–PI3K–AKT signaling cascade (Nazarian et al., 2010; Villanueva et al., 2010; Poulikakos et al., 2011; Wagle et al., 2011; Shi et al., 2012). On the basis of the studies of BRAFi resistance (Nazarian et al., 2010), clinical trials are underway to test combinations of targeted agents (e.g., against BRAF and its downstream kinase MAPK/ERK kinase (MEK1/2) to prevent or delay acquired BRAFi resistance. In terms of immune checkpoint inhibitors, the first success was with ipilimumab, a monoclonal antibody targeting the T-cell surface molecule called cytotoxic T-lymphocyte antigen 4 (CTLA-4). CTLA-4 is a negative regulatory protein that downregulates the priming phase (i.e., lymph node) of antitumor immune responses. Clinical inhibition of CTLA-4 by ipilimumab has delivered objective response rates of about 10–15% among metastatic melanoma patients. Although this rate is lower than that achieved with BRAFi, responses tend to be more durable. In T cells, CTLA-4 competes with the costimulatory receptor CD28 for binding to CD80 (B7-1) and CD86 (B7-2), which are expressed by antigenpresenting cells. Other related T-cellnegative regulatory receptors with homology to the costimulatory CD28 receptor include program death-1 (PD-1), B7-H4, and B- and T-lymphocyte atten-

COMMENTARY

Clinical Implications 

Inhibitors of mutant BRAF and tumor-associated immune checkpoints have shown unprecedented clinical activities against advanced melanoma.



Clinical trials are currently testing BRAF and immune checkpoint inhibitors in the adjuvant setting in earlier-stage patients at high risk of postsurgical disease recurrence.



Stratification of high-risk patients on the basis of prognostic biomarkers that are mechanistically linked to therapeutic biomarkers may identify those patients most receptive to adjuvant BRAF and/or immune checkpoint inhibitors.

uator. Unlike CTLA-4, which counters T-cell interaction with antigen-presenting cells during the priming stage, PD-1 counters T-cell interaction with tumor cells during the effector stage, via interaction with the PD-1 ligand, PD-L1, which is expressed by tumor cells or the tumor microenvironment. Early clinical data with other immune checkpoint inhibitors, namely monoclonal antibodies against the PD-1 receptor (Topalian et al., 2012) and its ligand PD-L1/B7-H1 (Brahmer et al., 2012), has jolted the melanoma field and beyond, with 20–30% objective response rates among patients with metastatic melanoma, renal cell carcinoma, and even non-small-cell lung carcinoma, which is notoriously resistant to other immunotherapies. The durability of responses in the majority of patients treated with monoclonal antibodies against CTLA-4, PD-1, and PD-L1 contrasts with the shorter durabilities of responses to BRAFis. However, in contrast to the deployment of BRAFis guided by detection of BRAF V600 mutations, there is currently no prospectively tested therapeutic biomarker that guides the deployment of immune checkpoint inhibitors. Emerging evidence supports an immuneactive tumor microenvironment (as indicated by immune gene signature, infiltrate, or paradoxical elaboration of negative feedback PD-L1 upregulation) potentially correlating with improved melanoma patient survival or response to immune checkpoint inhibitors. Stage III melanoma: prognostic factors meet therapeutic biomarkers

Therapeutic biomarkers have the potential to double as prognostic and

predictive factors (Figure 1). A prognostic factor offers information on the likely course of a cancer (or any other disease) in an untreated patient, providing information on patient outcomes independent of subsequent therapy. Among low-risk or early-stage melanoma patients, for instance, well-defined prognostic factors would be helpful in selecting patients for adjuvant systemic clinical trials, which, in the case of melanoma where options are few, should more often than not be recommended. On the other hand, a predictive factor offers information to stratify subsets of patients who are more likely to respond to a given therapy. Thus, predictive factors imply a rational ability to select a specific therapy, thereby improving the anticipated outcome for that subpopulation of patients. Clinicopathologic features have dominated as prognostic factors for primary melanoma, with molecular features producing less useful results. Recent data suggest that BRAF mutations may be associated with worse survival in stage IV melanomas (Long et al., 2011). What about stage III melanoma (metastasis to lymph nodes or in-transit metastasis)? In stage III melanoma, the 5-year and 10year survival varies from 80 to 40% and from 70 to 25%, respectively, owing to some patients being at a high risk of recurrence and death even after surgery for resectable cases. Currently, the only FDA-approved adjuvant treatments for such patients are high-dose interferon alpha-2b and pegylated interferon. There are few data specifically pertaining to the adjuvant use of any therapy, including interferons, for patients at the highest recurrence risk category (stage

IIIb and c). Clinical trials are underway to accrue high-risk patients with resected stage IIIc/IV disease and prospectively randomize them to high-dose interferon versus ipilimumab. Hence, prognostic and predictive factors for stage III melanoma (and specifically for stage IIIb and c patients) would help stratify patients further for adjuvant therapeutic trials. It is interesting to note that in this issue Mann et al. (2012) report on the outcome (time from nodal resection to death) of a group of melanoma patients with surgically resected macroscopic nodal metastasis (stage IIIb and IIIc). They examined this outcome by multivariate analysis with respect to a detailed series of clinicopathologic variables and molecular features (NRAS and BRAF mutation status and expression signatures) and their association with good versus poor prognosis groups. The investigators hypothesized that the conventional primary melanoma prognostic factors (Breslow’s thickness, mitotic rate, ulceration; 2009 revised American Joint Committee on Cancer staging criteria), which portend eventual metastasis, may not be similarly strong prognostic factors once the disease has progressed overtly to involve lymph node metastasis. Indeed, they found that either NRAS or BRAF mutations (versus those wild type in both genes) were associated independently with a worse prognosis (survivalo1 year). In a separate report, stage III melanoma patients with BRAF V600 mutation have also been shown to fare worse in terms of overall survival compared with those with BRAF wild type status (Moreau et al., 2012). In addition, Mann et al. (2012) show that a gene expression signature comprising 46 genes and enriched for biological pathways consistent with immune activation independently predict a better prognosis (survival 41 year). Importantly, this gene signature was also found to be associated with good prognosis in two independent stage III melanoma data sets. Adjuvant therapy using BRAFi and immune checkpoint inhibitors

The first targeted therapy, trastuzumab (which targets overexpression or amplification of the epidermal growth factor receptor 2 or human epidermal growth www.jidonline.org

293

COMMENTARY

Prognosis

Prediction

Prediction

ICi

BRAFi

Prediction

or

Prediction

BRAFi + ICi

Figure 1. Mutant BRAF and immune response in melanoma: prognostic and predictive biomarkers. In melanoma macroscopically metastatic to lymph nodes, disease recurrence and poor survival after complete surgical resection are associated with BRAF (and NRAS) mutations and the lack of an immune response or signature, both of which serve as independent prognostic biomarkers for this high-risk group of patients. Prognostic biomarkers indicate the likely course of disease without treatment after surgical resection of melanoma metastatic to a lymph node(s). The upward arrow indicates those patients who will likely relapse after surgery. BRAF (V600E/K) mutations may serve as predictive biomarkers for adjuvant therapeutic responses to a BRAF inhibitor (BRAFi) or the combination of a BRAFi with MAPK/ERK kinase inhibitor (both clinical trials are underway). Predictive biomarkers identify a subset of patients most likely to respond to the indicated adjuvant therapy (right, left, and downward arrows). The presence of a tumor-associated immune response (or a more specific surrogate marker such as PD-L1 expression) may serve as a predictive biomarker for adjuvant therapeutic response to an immune checkpoint inhibitor (ICi). The combination of BRAFi and ICi in an adjuvant setting may benefit those patients with BRAF mutant melanomas showing evidence of tumor-associated immunity. Alternatively, according to one hypothesis, BRAFi may induce tumor-associated immunity, thereby predicting responses to the combination of BRAFi and ICi. Red circle, BRAF (V600E/K) melanomas; white circle, BRAF wild-type melanomas; yellow halo, melanoma-associated immunity.

factor receptor 2), was originally developed for advanced breast cancer, but it has now become a standard adjuvant therapy for human epidermal growth factor receptor 2-positive early-stage breast cancer. There is no reason why BRAFi and immune checkpoint inhibitors cannot do the same for the adjuvant treatment of high-risk early-stage or intermediate-stage melanoma. The reality of long-term treatment (perhaps up to 1 year or longer after surgery) and issues of drugrelated toxicities should not dissuade the use of these highly active (and paradigmchanging) therapies for melanoma, in order to improve progression-free survival or overall survival. Intervention when the tumor burden is low, informed by prognostic factors that mechanistically relate to therapeutic biomarkers, may well be a recipe for success. Yes, grade 3 and grade 4 toxicities of BRAFi’s and immune checkpoint inhibitors have made headlines. For instance, 294

BRAFis, including vemurafenib and dabrafenib, accelerate the formation of cutaneous squamous cell carcinomas. This has been attributed to a pharmacologic property termed ‘‘paradoxical MAPK pathway activation’’ in BRAF wild-type tissues and cells harboring activation signals upstream of RAF. In addition, ipilimumab can lead to serious side effects, such as colitis, presumably attributable to drug-induced autoimmunity. However, unlike drug toxicities in the past, rational strategies to overcome drug toxicities are becoming available. The combination of BRAF and MEK inhibition to prevent cutaneous squamous cell carcinomas (Su et al., 2012) and targeting of alternative immune checkpoints with higher specificity to produce antitumor (rather than anti-host) immunity are two successful examples. CONFLICT OF INTEREST

The authors state no conflict of interest.

Journal of Investigative Dermatology (2013), Volume 133

REFERENCES Brahmer JR, Tykodi SS, Chow LQ et al. (2012) Safety and activity of anti-PD-L1 antibody in patients with advanced cancer. N Engl J Med 366:2455–65 Chapman PB, Hauschild A, Robert C et al. (2011) Improved survival with vemurafenib in melanoma with BRAF V600E mutation. N Engl J Med 364:2507–16 Long GV, Menzies AM, Nagrial AM et al. (2011) Prognostic and clinicopathologic associations of oncogenic BRAF in metastatic melanoma. J Clin Oncol 29:1239–46 Mann GJ, Pupo GM, Campain AE et al. (2012) BRAF mutation, NRAS mutation, and the absence of an immune-related expressed gene profile predict poor outcome in patients with stage III melanoma. J Invest Dermatol 133:509–17 Moreau S, Saiag P, Aegerter P et al. (2012) Prognostic value of BRAF (V600) mutations in melanoma patients after resection of metastatic lymph nodes. Ann Surg Oncol; e-pub ahead of print 7 July 2012 Nazarian R, Shi H, Wang Q et al. (2010) Melanomas acquire resistance to B-RAF(V600E) inhibition by RTK or N-RAS upregulation. Nature 468:973–7

COMMENTARY

Poulikakos PI, Persaud Y, Janakiraman M et al. (2011) RAF inhibitor resistance is mediated by dimerization of aberrantly spliced BRAF(V600E). Nature 480:387–90 Shi H, Moriceau G, Kong X et al. (2012) Melanoma whole-exome sequencing identifies (V600E)BRAF amplification-mediated acquired B-RAF inhibitor resistance. Nat Commun 3:724 Sosman JA, Kim KB, Schuchter L et al. (2012) Survival in BRAF V600-mutant advanced

Su F, Viros A, Milagre C et al. (2012) RAS mutations in cutaneous squamous-cell carcinomas in patients treated with BRAF inhibitors. N Engl J Med 366:207–15

Villanueva J, Vultur A, Lee JT et al. (2010) 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: 683–95

Topalian SL, Hodi FS, Brahmer JR et al. (2012) Safety, activity, and immune correlates of antiPD-1 antibody in cancer. N Engl J Med 366:2443–54

Wagle N, Emery C, Berger MF et al. (2011) Dissecting therapeutic resistance to RAF inhibition in melanoma by tumor genomic profiling. J Clin Oncol 29:3085–96

melanoma treated with vemurafenib. N Engl J Med 366:707–14

www.jidonline.org

295