The therapeutic potential of targeting minimal residual disease in melanoma

Abstract

Background: Cutaneous melanoma is a lethal form of skin cancer with morbidity and mortality rates highest amongst European, North American and Australasian populations. The developments of targeted therapies (TTs) directed at the oncogene BRAF and its downstream mediator MEK, and immune checkpoint inhibitors (ICI), have revolutionized the treatment of metastatic melanoma, improving patient outcomes. However, both TT and ICI have their limitations. Although TTs are associated with high initial response rates, these are typically short-lived due to resistance. Conversely, although ICIs provide more durable responses, they have lower initial response rates. Due to these distinct yet complementary response profiles, it has been proposed that sequencing ICI with TT could lead to a high frequency of durable responses whilst circumventing the toxicity associated with combined ICI + TT treatment. However, several questions remain unanswered, including the mechanisms underpinning this synergy and the optimal sequencing strategy. The key to determining this is to uncover the biology of each phase of the therapeutic response.

Aims and methods: In this review, we show that melanoma responds to TT and ICI in three phases: early response, minimal residual disease (MRD) and disease progression. We explore the effects of ICI and TT on melanoma cells and the tumour immune microenvironment, with a particular focus on MRD which is predicted to underpin the development of acquired resistance in the third phase of response.

Conclusion: In doing so, we provide a new framework which may inform novel therapeutic approaches for melanoma, including optimal sequencing strategies and agents that target MRD, thereby ultimately improving clinical outcomes for patients.

Keywords: BRAF/MEK inhibitors; acquired resistance; combination treatment; cross-resistance; immune checkpoint inhibitors; intrinsic resistance; melanoma; minimal residual disease.

FIGURE 1

Response of cutaneous BRAF‐mutant melanoma to targeted therapy: (A) phases of the therapeutic response to targeted therapy. Cutaneous melanoma responds to targeted therapy in three phases. The first phase is defined by initial tumour regression upon exposure to therapy. The second phase is defined by minimal residual disease (MRD), whereby residual cancer cells persist whilst under therapeutic pressure, due to acquired genetic and non‐genetic changes. The third phase is defined by disease progression, whereby persistent cells of the MRD expand due to a growth and survival advantage; (B) non‐genetic mechanisms of acquired resistance. Melanoma cells undergo epigenetic and transcriptomic changes which result in the establishment of distinct melanoma phenotypes, thereby allowing cells to adapt to stressors within the tumour microenvironment. The treatment naïve phase is initially dominated by the pigmented differentiated cell phenotype. Upon therapeutic exposure to BRAFi + MEKi, non‐genetic changes result in the emergence of de‐differentiated cell phenotypes, including NCSCs and invasive cells, which ultimately drive resistance to targeted therapy. Genetic mechanisms of persistence may also exist; (C) changes within the tumour immune microenvironment (TIME) during each phase of the therapeutic response. The TIME plays a vital role in dictating therapeutic responses to targeted therapy. Treatment naïve melanoma is classically considered to be ‘immunologically hot’, dominated by immune stimulatory cell types, with marginal infiltrates of immune suppressive cells. Upon exposure to BRAFi + MEKi, this immunostimulatory TIME is further enriched and is characterized by increased infiltrates of cytotoxic Cd8+ T‐cells and increased antigen presentation through the upregulation of MHC I. Little is known about the composition of the TIME during the MRD phase. During the acquired resistance phase, the balance is tipped in favour of an immunosuppressive TIME characterized by increased T‐reg cells and reduced cytotoxic T‐cells and antigen‐presenting cells.

FIGURE 2

Response of cutaneous melanoma to immune checkpoint inhibition. Melanoma responds to immune checkpoint inhibitors (ICI) in three phases. In the first phase, tumours with sensitivity to ICI demonstrate early disease control, as shown by a reduction in tumour load. Biomarkers predictive of ICI sensitivity include an immunostimulatory tumour immune microenvironment (TIME) characterized by low infiltrates of T‐regulatory cells and high infiltrates of effector and memory B and T‐cells, and monocytes. Conversely, biomarkers predictive of primary ICI resistance include an immunosuppressive TIME defined by high levels of immunosuppressive T‐regulatory cells, TGF‐B and IL‐10 expression. In addition, although primary non‐responders are enriched for immunosuppressive signalling pathways such as the innate α‐PD1 resistance signature (IPRES), primary responders are enriched for immunostimulatory signalling pathways resulting in increased MHC, IFN‐g and IFN‐a expressions. Mutations in genes such as B2M also drive a high tumour mutational burden in primary responders, resulting in the generation of neoantigens which are targeted by the immune system. In the second phase, NGFR^hi tumour cells are recruited and persist whilst under therapeutic pressure. This is aided by multiple genomic changes, including beta‐2‐microglobulin (B2M) loss of heterozygosity (LOH), CDKN2A and phosphatase and tensin homolog (PTEN) deletion, which cover a survival and growth advantage. In the third phase, tumours with acquired resistance have been found to be enriched for NGFR^hi tumour cells, defects in IFN‐γ signalling and antigen presentation pathways, and an immunosuppressive TIME defined by high expression levels of TGF‐B and IL‐10, and high levels of T‐regulatory cells, the emergence and suppressive functions of which are induced by the V‐domain I_g suppressor of T‐cell activation (VISTA) ligand.

FIGURE 3

Cross‐resistance between immunotherapy and targeted therapy (TT): (A) tumours that progress on TT are enriched for an immunosuppressive tumour immune microenvironment (TIME) characterized by high infiltrates of myeloid‐derived suppressor cells (MDSCS) and low infiltrates of CD8+ T‐cells and CD103+ DCs, all of which hamper therapeutic responses to subsequent immunotherapy; (B) preliminary results from the DREAMseq trial indicate improved 2‐year overall survival (OS) and progression‐free survival (PFS) rates with first‐line α‐CTLA4 + α‐PD1 followed by BRAFi + MEKi compared with first‐line BRAFi + MEKi followed by α‐CTLA4 + α‐PD1; (C) tumours that progress on immune checkpoint inhibitors (ICI) are found to be enriched for NGFR^hi cells and an Innate Α‐PD1 Resistance (IPRES) Signature but express low levels of MHC and CD8+ T‐cells, all of which hamper therapeutic responses to second‐line TT.