Melanoma Program: Research Impact More Than Skin-Deep

November 7, 2019

Representing only 1% of all skin cancers yet accounting for the vast majority of all skin cancer-related deaths, melanoma is both the rarest and deadliest of all skin cancers.1 Advanced melanoma is primarily responsible for the overall mortality rate— five-year survival rates plummet from 98% with localized melanoma to 23% with metastatic melanoma.2 In striking juxtaposition to the centuries-long treatment history of this disease, there were few viable systemic treatment options for melanoma even a decade ago, and the standard of care for unresectable or metastatic melanoma was considered to be enrollment in clinical trials.3,4 Decades of basic research into cancer genomics and immunology enabled recent, crucial advances in immunotherapy and targeted therapy that marked their emergence as the dominant modalities in the treatment of unresectable and metastatic melanoma today. The dramatic shift of feasible therapeutic outcomes in advanced melanoma from palliation for few patients to durable clinical responses for a larger minority and palliation for most has correspondingly shifted the trajectory of melanoma research in these modalities. Parallel and complementary research into understanding response and resistance in immunotherapy and targeted therapy work toward identifying predictive and/or prognostic biomarkers, discovering novel drug targets, and guiding therapeutic combinations and sequencing in order to improve patient outcomes. 

Today, many of the knowledge and treatment gaps in melanoma broadly reflect areas of unmet need across cancer, and insights discovered in melanoma are applicable to other cancer contexts. As DF/HCC Melanoma Program co-leader F. Stephen Hodi, MD (DFCI) noted: “A  lot of the paradigms in melanoma have gone on to be translated in other cancers.” Through diverse and collaborative approaches, DF/HCC Melanoma Program members are investigating treatment response and resistance, developing therapeutic strategies for underserved patient populations, and improving prevention. “There are groups collaborating across hospitals, across departments, across disciplines— zebrafish people who are working with clinicians who are working with people studying mouse models,” said co-leader David E. Fisher, MD, PhD (MGH) about the Melanoma Program. Fisher identified the interdisciplinary and collaborative nature of the program as critical to successful efforts but also inevitable in response to research needs: “The cross-institutional collaborations are extremely strong. That particularly happens when there are screaming questions to ask and the right tools, technologies, and expertise are available to address the questions. The teams come together, and there are countless examples of that in our community.”

Targeted therapy addresses specific mutations in melanoma patients

Targeted therapy intervenes in pathways that specifically contribute to cancer proliferation and survival. The mitogen-activated protein kinase (MAPK) pathway is dysregulated in the vast majority of melanomas, with approximately 50% of melanomas harboring a mutation in BRAF and 30% in NRAS.5 “That became the paradigm or framework of how we thought about trying to tackle melanoma with targeted therapy strategies,” said DF/HCC Developmental Therapeutics Program co-leader and Melanoma Program member Keith Flaherty, MD (MGH), who was involved in the clinical development of first-in-class BRAF inhibitor vemurafenib.6 Efforts to target these mutations have involved the direct inhibition of V600-mutant BRAF as well as indirect strategies to mediate Ras activity through the inhibition of downstream target MEK.  “The biggest issue with these drugs, however,” said Program member Ryan Sullivan, MD (MGH), “is that there remains a significant percentage of patients who will progress despite the fact that the great majority of patients have clinical benefit.” Melanoma Program members are working to improve patient outcomes by addressing resistance to current therapies.

Investigating resistance mechanisms to BRAF-targeted therapy yields additional targets in the MAP kinase pathway

Flaherty believes that “there is still significant room for improvement and optimization in targeting this core pathway that is nearly (but not completely) essential to melanoma formation.” BRAF inhibition can lead to paradoxical activation of the MAPK pathway through compensatory mechanisms as well as the upregulation of various other pro-survival signaling pathways, leading to acquired resistance. Investigations into these resistance mechanisms implicated downstream kinases MEK and ERK. While the clinical benefit of MEK-inhibitor monotherapy is modest, the demonstrated clinical benefit of combining BRAF and MEK inhibition has led to the FDA approval of three combination regimens to date.5,7 Sullivan, collaborating with DF/HCC Developmental Therapeutics co-leaders, Flaherty and Geoffrey Shapiro, MD, PhD (DFCI), and Melanoma Program member James Mier, MD (BIDMC), investigated first-in-class ERK1/2 inhibitor ulixertinib in a Phase I trial in patients with MAPK-mutant advanced solid tumors.8

Targeting proteins involved in apoptosis may improve response to MAP kinase pathway-targeted therapy 

Building on preclinical and mechanistic insights, DF/HCC members are also pursuing targets beyond the MAPK pathway. “In terms of where efforts are focused in going after other potential vulnerabilities in melanoma, most of the laboratory evidence would suggest that even the most promising of those potential targets are not standalone targets in the absence of also targeting the MAPK pathway,” said Flaherty, who is currently leading a Phase I clinical trial evaluating JAK1 inhibitor itacitinib in combination with MAPK pathway-targeted therapy (NCT03272464). Preclinical work by Melanoma Program members Rizwan Haq, MD, PhD (DFCI), Fisher, and Sullivan, with contributions by Lyn Duncan, MD (MGH), Flaherty, and Mier as well as Gastrointestinal Malignancies member James Cusack, MD (MGH), demonstrated that members of the Bcl-2 antiapoptotic protein family can confer resistance to BRAF inhibition in melanoma and that promoting apoptosis through manipulation of the Bcl-2 family could improve response to BRAF-targeted therapy.9,10 Phase I/II studies investigating the combination of Bcl-2/Bcl-XL/Bcl-w inhibitor navitoclax, BRAF inhibitor dabrafenib, and MEK inhibitor trametinib are underway (NCT01989585). Sullivan, who is the principal investigator of this trial at MGH, hopes that this trial will identify a regimen that works better and that analyzing tumors will provide a deeper understanding of how targeting apoptosis affects the tumor environment. In the case that researchers don’t discover dramatic clinical benefit, they’ll wield some knowledge of why and what they can attempt next in terms of targeting apoptosis. 

researchers are developing targeted therapeutic strategies for defined patient populations with unmet needs

Following the development of the core BRAF/MEK inhibitory strategy, progress in improving clinical outcomes with targeted therapies has been incremental. “We haven’t yet seen that next “AHA!” moment. It’s entirely possible that the strategy that we have been pursuing, which is looking for big effects in the overall populations isn’t the way to go. It could be that we need to be more tailored or specific in our strategies for various small subpopulations of patients,” said Flaherty. Patients with uveal melanoma represent one such population low in comparative incidence but high in need. “Eye melanoma is a very unique melanoma that does not respond well to any of the currently available therapies,” said Melanoma Program co-leader Hodi. Haq is currently involved in preclinical efforts to develop therapeutic strategies for uveal melanoma. Uveal melanomas generally lack BRAF mutations, but the majority of them have mutations in G-alpha subunits GNAQ and GNA11 that appear to drive uveal melanoma. Due to the loss-in-function nature of these mutations, direct targeting strategies are ineffective. “Our hope is that we’ll find some way that we can indirectly target GNAQ or GNA11,” said Haq, who is performing CRISPR screening to find ways to destabilize those proteins. Current clinical strategies depend on inhibiting downstream targets activated by mutations in GNAQ/GNA11. Melanoma Program member Elizabeth Buchbinder, MD (DFCI) is leading a phase II clinical trial studying ERK1/2 inhibitor ulixertinib in uveal melanoma (NCT03417739).


Immunotherapy engages anti-tumor immune responses

The clinical development of CTLA-4 antibody ipilimumab,11 which involved Hodi, Kidney Cancer Program leader David McDermott, MD (BIDMC), and Melanoma Program member Donald Lawrence, MD (MGH), for metastatic melanoma pioneered immune checkpoint blockade, marking a turning point in the evolution of immunotherapy in melanoma and other cancers. Antibodies disrupting PD-1/PD-L1 binding, studied by Gordon Freeman, PhD (DFCI) and Arlene Sharpe, MD, PhD (HMS)were rapidly developed for clinical use. Checkpoint inhibitors block co-inhibitory T-cell signaling, favorably modulating anti-tumor immune responses. Immune checkpoint blockade produces lower initial response rates than targeted therapy but durable clinical response even after treatment cessation.12 Melanoma Program members are working to understand response and resistance to immunotherapy due to genetic and immune factors and to translate those insights into therapeutic development.

resistance to immunotherapy can arise from factors intrinsic to melanoma tumors

Tumor mutational burden correlates with clinical response to checkpoint inhibition but has limited clinical utility as a predictive biomarker in microsatellite-stable tumors. In a multi-institutional and multi-disciplinary collaboration, DF/HCC members performed whole-exome sequencing and analyzed 249 tumors from patients treated with immune checkpoint therapies, determining that genomic features such as genetic driver events, intratumoral heterogeneity, and global mutational signatures also contribute to selective response. This work translated prior findings in specific cancers to broader cancer contexts, suggested biomarkers for further study, and described a potential approach for future studies with more statistical power to identify specific predictors of response.13 Broad Institute and DF/HCC members leveraged single-cell technologies to investigate malignant cell states promoting immune evasion in melanoma. This study identified a cell program associated with T-cell exclusion and predictive of resistance to immune checkpoint blockade; repression of this program with cyclin-dependent kinase 4/6 (CDK4/6) inhibitors sensitized melanoma tumors to immune checkpoint blockade in mouse models.14 This further validates the rationale to combine CDK4/6 inhibition with PD-1/PD-L1 checkpoint blockade previously espoused in a collaborative study that defined immunomodulatory functions of CDK4/6 inhibition.15

In melanoma patients that respond to immunotherapy, there exists a small subset who have one or two treatment-resistant lesions. “This kind of pattern implies that there’s a tumor-intrinsic mechanism of resistance,” said Rizwan Haq, MD (DFCI). By studying biopsies from patients with immunotherapy-resistant lesions, the Haq lab seeks to identify putative resistance genes. Immunotherapy is given to immunocompetent mouse models having specific deletions of putative resistance genes. “That has been incredibly useful in dissecting what is actually driving resistance rather than what is associated with resistance,” said Haq. Identifying exact drivers of resistance could potentially enable the development of therapeutic agents that overcome tumor-intrinsic resistance.  

understanding the immune environment can improve response to immune checkpoint blockade by guiding therapeutic strategy

Researchers are also studying the tumor immune microenvironment to better understand resistance to checkpoint inhibition blockade. For example, pathologist and Cancer Immunology Program member Scott Rodig, MD, PhD (BWH) seeks “to identify the types of immune cells that are adjacent to the melanoma, understand whether or not those immune cells are activated or inhibited, and quantify different aspects of how the immune cells are interacting with the melanoma.” Rodig and Hodi discovered that classes of major histocompatibility complex (MHC) confer differential sensitivity to immune checkpoint blockade in previously untreated metastatic melanoma. Response to CTLA-4 blockade is dependent on robust expression of MHC class I by tumor cells, while response to PD-1 blockade requires MHC class II expression and associated interferon γ-mediated inflammation.16 Rodig and Hodi are working closely to verify these findings in additional patient cohorts. More recently, Cancer Immunology Program member W. Nicholas Haining, BCh, BM (DFCI) in collaboration with Cancer Immunology Program leader Arlene Sharpe, MD, PhD (HMS), Hodi, and Rodig, defined a population of tumor-infiltrating lymphocytes, specifically progenitor exhausted CD8+ T cells, that respond favorably to PD-1 checkpoint inhibition and control tumor growth.17 Studies like these further our understanding of predictive biomarkers needed to guide the choice of appropriate therapies for patients as well as suggest potential targets that can be exploited to improve response to checkpoint blockade.

DF/HCC members are developing novel immunotherapies and combinations

Researchers are combining available and experimental therapeutics in novel combinations in an attempt to improve clinical responses. Arising from mechanistic investigations into immunotherapy resistance and increasing mechanistic understanding of immune response are the next generation of immune checkpoint inhibitors. Hodi is involved in a phase I/II clinical trial assessing the safety and efficacy of an inhibitor of T-cell immunoglobulin and mucin-domain-containing-3 (TIM-3) with and without PD-1 inhibitor spartalizumab in patients with advanced solid tumors (NCT02608268). He is also assessing the lymphocyte-activation gene 3 (LAG-3) checkpoint inhibitor relatlimab as monotherapy and in combination with PD-1 inhibitor nivolumab in a phase I/II study of safety, tolerability, and effectiveness in patients with solid tumors (NCT01968109). Hodi and  are studying the effectiveness of nivolumab with or without relatlimab in patients with advanced or unresectable melanoma in a phase II/III clinical trial (NCT03470922). DF/HCC members are also investigating the combination of immune blockade with other immune-modulating therapies. Elizabeth Buchbinder, MD (DFCI) and McDermott are investigating the combination of PD-1 inhibitor pembrolizumab with gene plasmid (NCT03132675) and recombinant interleukin-2 (NCT02748564), respectively.       

personal neoantigen vaccines tune patients’ immune responses to their individual tumors

Personal neoantigen vaccines, which focus patients’ tumor-specific immune responses, are being developed as a treatment strategy for melanoma and other cancers. “Over the last seven years, we have learned from many other studies, not just vaccine work, that neoantigens are important targets of effective anti-cancer responses,” said Patrick Ott, MD, PhD, who is the Clinical Director for both the Melanoma Center and Center for Immuno-Oncology at DFCI. Ott worked with DF/HCC Cancer Immunology Program member Nir Hacohen, MD (MGH) and Leukemia, Lymphoma, and Myeloma Program member Catherine Wu, MD (DFCI) to develop and evaluate personal neoantigen vaccines in previously untreated patients with high-risk, resectable melanoma. They identified tumor-specific mutations by sequencing matched tumor- and normal-cell DNA, developed algorithms to predict which mutated peptides were likely compatible with individual patients’ human leukocyte antigen (HLA) system, and synthesized a cocktail of up to 20 neoantigens to deliver to patients.18 “We reported that this approach was feasible and safe. We showed robust immune responses in patients and that those responses had the right phenotype,” said Ott. Current challenges in cancer immunotherapy that this approach addresses include selectively targeting tumors and targeting tumors with high heterogeneity. Efforts are ongoing to improve the process of neoantigen selection, develop combination strategies with complementary immune therapies, and translate this treatment into other cancer settings, including metastatic melanoma.


Targeted therapy and immunotherapy are complementary treatment strategies

The choice of whether or how to give targeted therapy and immunotherapy in sequence or combination is an important and unresolved consideration in the treatment of BRAF-mutant melanoma patients. While BRAF-targeted therapy yields high initial responses, the long term disease control for patients that do respond to immunotherapy makes it an attractive treatment option. “The success of that approach spreads across melanoma in a way that doesn’t honor the boundaries of how we think about genetic subgroups of melanoma and targeted therapy,” said DF/HCC Melanoma Program member Keith Flaherty (MGH), who chairs the Henri and Belinda Termeer Center for Targeted Therapy at MGH.

Differences in mechanisms of action and patterns of response and resistance provide a rationale for combining treatment strategies.  “There have been various preclinical studies that have shown that combining immunotherapies with BRAF targeted therapy seems to be at least additive, if not synergistic,” said Ryan Sullivan, MD (MGH), who is also a member of the DF/HCC Cancer Immunology Program. DF/HCC members demonstrated that pharmacologic blockade of the MAPK pathway through BRAF or dual BRAF/MEK inhibition yields enhanced melanoma antigen expression and a more favorable tumor microenvironment.19 “We may be creating an environment in the setting of BRAF targeted therapy that will make immunotherapy work better,” said Sullivan. 

A clinical study involving Donald Lawrence, MD (MGH) and F. Stephen Hodi, MD (DFCI) corroborated preclinical findings, reporting an increase in CD8+ T cells, PD-L1 expression, and tumor inflammation gene signature in biopsies of patients concomitantly treated with BRAF inhibitor dabrafenib, MEK inhibitor trametinib, and anti-PD-1 antibody pembrolizumab.20 Sullivan, along with Hodi, reported a phase IB study combining BRAF inhibitor vemurafenib, MEK inhibitor cobimetinib, and anti-PD-L1 antibody atezolizumab. A 28-day run-in period with MAPK-targeted therapy before introducing atezolizumab was established in order to better manage toxicity and exploit the anti-tumor effects of MAPK-targeted therapy.21 “The data are compelling that this approach could, in fact, change the way we think about how to give targeted therapy,” said Sullivan. The two studies were published concurrently in Nature Medicine this June along with the results of a randomized phase II trial comparing dabrafenib and trametinib alone versus in combination with pembrolizumab;22 together, the studies report substantial though manageable toxicity with triple-agent combination therapy and promising efficacy compared to MAPK-targeted therapy alone. 

Given the significant toxicity of concomitant triple therapy and the promising results yielded by leading into combination therapy with a period of MAPK-targeted therapy, Ryan Sullivan, MD, national and MGH site PI, and David McDermott, MD, BIDMC site PI, are coordinating efforts to study the outcomes of a targeted therapy-lead sequencing strategy using dabrafenib and trametinib in patients with BRAF-mutant melanoma and trametinib in patients with BRAF-wild type melanoma plus pembrolizumab (NCT03149029). Biopsies taken at each stage of treatment will inform on correlated changes in the tumor cellular environment.


Preventing advanced stage disease decreases melanoma mortality

Despite the decrease in overall mortality rates since the recent development of systemic treatment modalities, melanoma incidence is steadily increasing. Melanoma Program members are working to prevent disease incidence through sociological approaches to mitigate risk behaviors and biological approaches to interrogate and intervene in carcinogenesis. This research is rich in collaboration, with relationships forming across scientific, professional, and institutional boundaries. For example, translational researcher and clinician David Fisher (MGH) is collaborating with developmental biologist Leonard Zon, MD (BCH) to elucidate mechanisms underlying melanoma formation and with chemical biologist Nathanael Gray, PhD (DFCI) to pharmacologically manipulate melanogenesis. Beyond prevention of disease incidence, there is significant clinical benefit to preventing progression of melanoma to advanced stages. To date, there still remains a significant differential in survival between patients with localized melanoma and patients with metastatic or unresectable melanoma. 

Melanoma Program member Alan Geller, MPH, RN (HSPH) aims to reduce mortality in melanoma through targeted early detection and screening of high-risk individuals. Geller identified men over the age of 50 (who have the fastest growing incidence rate of melanoma1) and adult survivors of childhood cancer (particularly patients who received radiotherapy) as two high-risk patient populations that he studies. “We’ve used a model of patient empowerment, where we’re encouraging and motivating patients to be the squeaky wheel that gets the oil,” said Geller, who advocates for partner- and physician-assisted full-body screening, particularly in high-risk individuals. He continued, “Melanoma writes its message in the skin for all of us to see. We want to take advantage of that— to bring in everybody possible— to make sure that early skin lesions are taken care of before they spread.” By detecting and treating melanoma at stages amenable to localized therapy, patients have a higher likelihood of survival. This benefit of early-stage treatment is being furthered in the clinic by adjuvant treatment with either immunotherapy or MAPK-targeted therapy. Researchers are working to implement preventative measures even at later stages of melanoma. Elizabeth Buchbinder, MD (DFCI) is studying systemic therapy in the neoadjuvant setting for patients with stage III or IV resectable melanoma “to lead to cures in patients that would otherwise have a high risk of their melanoma coming back.” Buchbinder is currently leading a phase II clinical trial evaluating neoadjuvant PD-1 inhibition as monotherapy or in combination with either IDO inhibition or CTLA-4 inhibition (NCT04007588).


Advances in melanoma research are being translated to other cancers

Researchers in the DF/HCC Melanoma Program continue to push forward the leading edge of cancer research and therapeutic development. “Melanoma has not only undergone this exciting revolution in therapy for its own sake,” said Program co-leader Fisher, pausing to acknowledge his excitement and potential bias. “But it’s leading the way, I think, across the much broader world of cancer therapy. Even though success in other diseases has not been as stunning as it has been in melanoma, the lessons are even in the next iterations: understanding success, understanding failure, and working on new combinations in drugs. All of that is happening very closely with melanoma still largely at the forefront.” Forty-four members in the DF/HCC Melanoma Program are collaborating across research and professional boundaries to bridge gaps in knowledge in the community, laboratory, and clinic.

—Written by Katharin Shaw, PhD


1. American Cancer Society. Cancer Facts and Figures 2019. (2019).

2. Siegel, R. L., Miller, K. D. & Jemal, A. Cancer statistics, 2019. Ca Cancer J Clin69, 7–34 (2019).

3. Schvartsman, G. et al. Management of metastatic cutaneous melanoma: updates in clinical practice. Ther Adv Med Oncol11, 1758835919851663 (2019).

4. Bhatia, S., Tykodi, S. S. & Thompson, J. A. Treatment of metastatic melanoma: an overview. Oncol Williston Park N Y23, 488–96 (2009).

5. Luke, J. J., Flaherty, K. T., Ribas, A. & Long, G. V. Targeted agents and immunotherapies: optimizing outcomes in melanoma. Nat Rev Clin Oncol14, 463–482 (2017).

6. Chapman, P. B. et al. Improved Survival with Vemurafenib in Melanoma with BRAF V600E Mutation. New Engl J Medicine364, 2507–2516 (2011). 

7. Sullivan, R. J. & Flaherty, K. T. Resistance to BRAF-targeted therapy in melanoma. Eur J Cancer49, 1297–1304 (2013). 

8. Sullivan, R. J. et al. First-in-Class ERK1/2 Inhibitor Ulixertinib (BVD-523) in Patients with MAPK Mutant Advanced Solid Tumors: Results of a Phase I Dose-Escalation and Expansion Study. Cancer Discov8, 184–195 (2017).

9. Haq, R. et al. BCL2A1 is a lineage-specific antiapoptotic melanoma oncogene that confers resistance to BRAF inhibition. Proc National Acad Sci110, 4321–4326 (2013).

10. Frederick, D. T. et al. Clinical Profiling of BCL-2 Family Members in the Setting of BRAF Inhibition Offers a Rationale for Targeting De Novo Resistance sing BH3 Mimetics. Plos One9, e101286 (2014).

11. Hodi, S. F. et al. Improved survival with ipilimumab in patients with metastatic melanoma. New Engl J Medicine363, 711–723 (2010).

12. Rozeman, E. A. & Blank, C. U. Combining checkpoint inhibition and targeted therapy in melanoma. Nat Med25, 879–882 (2019).

13. Miao, D. et al. Genomic correlates of response to immune checkpoint blockade in microsatellite-stable solid tumors. Nat Genet50, 1271–1281 (2018).

14. Jerby-Arnon, L. et al. A Cancer Cell Program Promotes T Cell Exclusion and Resistance to Checkpoint Blockade. Cell175, 984-997.e24 (2018).

15. Deng, J. et al. CDK4/6 Inhibition Augments Anti-Tumor Immunity by Enhancing T Cell Activation. Cancer Discov8, CD-17-0915 (2017).

16. Rodig, S. J. et al. MHC proteins confer differential sensitivity to CTLA-4 and PD-1 blockade in untreated metastatic melanoma. Sci Transl Med10, eaar3342 (2018).

17. Miller, B. C. et al. Subsets of exhausted CD8 + T cells differentially mediate tumor control and respond to checkpoint blockade. Nat Immunol20, 326–336 (2019).

18. Ott, P. A. et al. An immunogenic personal neoantigen vaccine for patients with melanoma. Nature547, 217–221 (2017).

19. Frederick, D. T. et al. Cancer Therapy: Clinical BRAF Inhibition Is Associated with Enhanced Melanoma Antigen Expression and a More Favorable Tumor Microenvironment in Patients with Metastatic Melanoma. Clin Cancer Res19, 1225–1231 (2013).

20. Ribas, A. et al. Combined BRAF and MEK inhibition with PD-1 blockade immunotherapy in BRAF-mutant melanoma. Nat Med25, 936–940 (2019).

21. Sullivan, R. J. et al. Atezolizumab plus cobimetinib and vemurafenib in BRAF-mutated melanoma patients. Nat Med25, 929–935 (2019).

22. Ascierto, P. et al. Dabrafenib, trametinib and pembrolizumab or placebo in BRAF-mutant melanoma. Nat Med25, 941–946 (2019).