Melanoma is the eighth most common cancer. It accounts for approximately 5% of all new cancer cases, and incidence is rising steadily.1,2 The incidence of cutaneous melanoma in the Spanish population is 8.7 cases per 100 000 population.3 Survival rates for early-stage melanoma are higher than 95%, but prognosis worsens considerably for disseminated forms, which have a 5-year-survival rate of less than 20%.2 Chemotherapy is the most widely accepted treatment for advanced cutaneous melanoma, but response rates are lower than 15% and the impact on survival outcomes is minimal.4

The emergence of immunotherapy marked an important change in the search for alternative treatments. The association between cancer cells and the immune system has been recognized for decades, and cutaneous melanoma is probably one of the most immunogenic tumors that exist. The interaction between the immune system and melanoma is dynamic, however, and tumor cells use various mechanisms to avoid recognition and destruction by the immune system, favoring disease progression.5

Treatment strategies aimed at boosting the immune response through vaccines and biochemotherapy were attempted for several years, but the results were disappointing.6,7 A more recent approach, however, that of using monoclonal antibodies directed against immune checkpoint inhibitors, has been met with enthusiasm.8

In contrast to other cancers, numerous findings, both clinical and histological, support the involvement of the immune system in the fight against melanoma.9

An inflammatory infiltrate within and around the tumor, sometimes accompanied by tumor regression, is observed in the majority of melanomas. Signs of regression are observed in up to 35% of cases, and tumor regression is actually 6 times more common in melanoma than in other types of tumors.10

Another more recently described finding is a correlation between the density of high endothelial venules and lymphocytic infiltration in association with greater tumor regression.11 An association has also been observed between tumor regression and melanomas with a lower mitotic or proliferative rate, i.e., melanomas with a better prognosis12 (Fig. 1).

Figure 1.

Image showing regression in melanoma. A, Clinical image showing a central region with loss of pigment. B, Dermoscopic features of the lesion showing a multicomponent pattern with a central white area. C and D, Histologic images showing a dense lymphocytic infiltrate in the dermis without epidermal involvement, together with melanophages and fibrosis in the dermis (hematoxylin-eosin, original magnification ×10 [C] and ×20 [D]).

(0.46MB).

The appearance of a halo around nevi provides further evidence of autoimmune reactivity and this halo can eventually cause the nevus to disappear completely. Vitiligo has also been observed in 2% to 7% of advanced melanomas following systemic therapy and has been linked to a better prognosis.13

Although the above findings and observations only serve to generate hypotheses, numerous studies have demonstrated the participation of the immune system in regressive melanoma. Findings include predominant infiltration of tumors by T lymphocytes, which are mediators of the innate immune system,14 and oligoclonal expansion of T lymphocytes in melanomas with extensive areas of regression.15 Adoptive cell therapy with T lymphocytes isolated from regressive melanomas has also been show to generate cytolytic activity against autologous melanomas.16

Whatever the case, the development of melanoma implies a breakdown in the body's defense system. This immune system failure is mediated by both factors that are intrinsic to the tumor cells and impaired immune defenses, which reduce the ability of the immune system to exert its effector function.

Suboptimal activation of specific T lymphocytes due to weak or absent expression of antigens in melanoma cells is known to result in immune response failure.17 The expression of endothelial adhesion molecules in the tumor is also reduced, and in addition, certain cells in the tumor infiltrate, such as macrophages, release growth factors, cytokines, and suppressors of immune response.18,19

The antitumor immune response alone thus is not sufficient to produce the elimination of tumor cells, but it can favor response to immunotherapy.

Two signals are needed for the activation of T lymphocytes. First, the T-cell receptor (TCR) recognizes the antigen presented by dendritic cells through the major histocompatibility complex, and second, the cluster of differentiation (CD) antigen 28 expressed on the surface of T lymphocytes produces a costimulatory signal by binding to its ligands B7-1 or B7-2. Without this second signal, the T lymphocytes would not respond to the stimulus, resulting in anergy.20

Activated T lymphocytes must then migrate to the site of the tumor, where they will act to destroy the tumor cells.21 These intratumoral cells, however, often exhibit decreased proliferation, cytokine production, and cytolytic activity, although outside the tumor microenvironment, they have been shown to mount powerful specific responses, suggesting the existence of signals in this microenvironment that block lymphocyte function.22

The activation of TCR also produces signals that stimulate immune checkpoint inhibitors. Cytotoxic T-lymphocyte associated antigen 4 (CTLA-4) and programmed cell death protein 1 (PD-1) are cell surface receptors involved in the regulation of activated T lymphocytes. They belong to the B7-CD28 superfamily and by binding to their ligands, they trigger inhibitory pathways resulting in reduced T-lymphocyte activity.23

Cytotoxic T-Lymphocyte Associated Antigen

Activated T lymphocytes express CTLA-4. This receptor is expressed approximately 24 to 48hours after activation and it binds to B7 with a greater infinity than to CD28, providing an inhibitory signal to T lymphocytes.24

CTLA-4 is a powerful coinhibitor that has a role in the development of tolerance from early on. Studies of CTLA-1-/− mice have shown that the absence of CTLA-4 results in a massive infiltrate of autoreactive T lymphocytes in diverse tissues, with death occurring within 2 to 3 weeks25,26 (Fig. 2).

Figure 2.

Diagram showing T lymphocyte activation. Activation of T lymphocyte after interaction between TCR and HLA and between costimulatory molecules B7 and CD28 (left). Inhibition of response following binding of CTLA-4 to B7 (right).

TCR indicates T-cell receptor; CD28, cluster of differentiation 28; MHC, major histocompatibility complex; CTLA-4, cytoxic T-lymphocyte associated protein 4.

(0.25MB).

PD-1 and PD Ligand 1

PD-1 is a transmembrane receptor of the CD28 family that is normally expressed in activated CD4+ and CD8+ lymphocytes.27 It has an immunosuppressive function, which under normal conditions prevents an excessive immune response to autologous cells. PD-1 expression is induced by the activation of T lymphocytes and is interrupted when the antigen is eliminated by the immune response. If this response is inadequate, however, PD-1 expression persists, generating a phenotype of ineffective T lymphocytes.28

PD-1 has 2 ligands: PD-ligand 1 (PD-L1), also known as B7-H1 and CD274, and PD-L2 (B7-DC or CD273). PD-L1 is expressed in more cells than PD-L2. When PD-L1 is expressed on the surface of tumor cells and binds to PD-1 on T lymphocytes, it activates a signaling cascade that suppresses TCR activation, thereby preventing the release of growth factors and survival signals.29 Evidently, thus, binding of PD-1 expressed on the surface of T lymphocytes to its ligand has a key role in blocking the immune response (Fig. 3).

Figure 3.

Diagram showing PD-1 and PD-L1 activity. Activation of T lymphocyte by interaction between stimulatory and costimulatory molecules (left). Inhibition of T cell following binding of PD-1 to PD-L1 (right).

APC indicates antigen-presenting cell; MHC, major histocompatibility complex; TCR, T-cell receptor; CD28, cluster of differentiation 28; PD-1, programmed death protein 1; PD-L1, programmed death protein ligand 1.

(0.26MB).

Numerous treatment strategies aimed at boosting the antitumor immune response have been developed since the 1980s, but they have had limited success.16 Recent years, however, have witnessed the emergence of a new strategy based on the blockage of receptors that inhibit the antitumor immune response rather than on the direct reactivation of the immune system. The antibodies used in these treatments bind not to the tumor cells but to the lymphocytes in order to stimulate a response.

CTLA-4 Inhibition

T lymphocytes express CTLA-4 on their surface within 2 to 3 days of activation and as a consequence exhibit reduced activity.30 In theory, thus, CTLA-4 blockade would allow specific T cells to maintain their activity.

The above theory was proven by the use of fully humanized monoclonal antibodies targeting CTLA-4, such as ipilimumab. By interfering with CTLA-4-B7 binding, the inhibitory stimulus disappears, and the lymphocytes maintain their activity.31 While this interaction means that the immune system will continue to fight against the tumor cells, it also means that it will target nontumor cells (Fig. 4).

Figure 4.

Diagram showing activity of ipilimumab. T-lymphocyte inhibition through binding between CTLA-4 and B7 (left). Activation of T lymphocyte through CTLA-4 blockade following binding between CTLA-4 and the monoclonal antibody ipilimumab.

TCR indicates T-cell receptor; CD28, cluster of differentiation 28; MHC, major histocompatibility complex; CTLA-4, cytoxic T-lymphocyte associated protein 4.

(0.26MB).

PD-1 and PD-L1 Blockade

PD-1 needs to bind to its ligands in order to exert its inhibitory effect. Most tissues, including tumor cells, express PD-L1. PD-L2, by contrast, is expressed only on dendritic cells, macrophages, mast cells, and B lymphocytes.32 PD-L1 expression on tumor cells can be mediated by oncogenic processes secondary to activating mutations, or activated by interferon released by specific T lymphocytes infiltrating the tumor.33,34 PD-L1 expression, however, can apparently only be induced if T lymphocytes expressing PD-1 exist. This phenomenon is known as adaptive immune resistance35 (Fig. 5).

Figure 5.

Interaction between PD-1 and PD-L1 leading to T-lymphocyte inhibition (left). Binding between anti PD-1 drugs such as nivolumab or pembrolizumab to PD-1 on the surface of the lymphocytes, resulting in their activation and hence the generation of the antitumor response (right).

APC indicates antigen-presenting cell; MHC, major histocompatibility complex; TCR, T-cell receptor; CD28, cluster of differentiation 28; PD-1, programmed death protein 1; PD-L1, programmed death protein ligand 1.

(0.27MB).

The above pathway can be inhibited by blocking PD-1, preventing it from binding to PD-L1 or PD-L2, or blocking PD-L1, impeding thus its interaction with PD-1.

Unlike CTLA-4-deficient mice, PD-1- and PD-L1-deficient mice do not die early, but rather progressively develop multiple autoimmune disorders.36,37 This would explain the wide range of adverse effects observed in patients treated with immunomodulators.

CTLA-4 and PD-1 exert an inhibitory function through different pathways, and therefore blockade of both receptors should favor the destruction of tumor cells by the immune system. This theory was first proven in animal models66 and later in humans. To date, randomized clinical trials have investigated the combined use of nivolumab and ipilimumab and shown higher response rates than those achieved with either treatment alone.67–69 Nonetheless, over half of the patients treated developed serious adverse effects. The most common serious effects were colitis, diarrhea, and elevated liver enzymes, and most cases required treatment with systemic corticosteroids.68,69

The combined use of nivolumab and ipilimumab was approved by the FDA in September 2015 and is currently pending approval by the EMA.

Following the success of immunotherapy, research has now turned to the search for biomarkers capable of identifying patients with a greater chance of responding to treatment.

Expression of PD-L1 in pretreatment tumor samples has shown the clearest association with response to date, with an increased response to anti PD-1 and anti PD-L1 antibodies observed in PD-L1-positive patients.34,70,71 No association has been found between PD-L1 expression and response to ipilimumab and nivolumab combination therapy.68

PD-L1 has several disadvantages in terms of its potential as a biomarker. Its expression is highly variable, even in samples from the same patient, and in addition, although greater clinical benefits have been observed in PD-L1-positive patients, a considerable number of PD-L1-negative patients also respond to treatment.60,70

Interferon released by CD8+ cells infiltrating the tumor margin and parenchyma is another potential confounder, as it induces PD-L1 expression in tumor cells (adaptive immune response). PD-L1 expression is thus an indirect marker of antigen-specific T lymphocyte activation. Tumors with higher levels of PD-L1 and PD-1 respond better to treatment, but specific CD8+ lymphocytes must first be present.72

Generally speaking, the best biomarker currently available for predicting response to any immunomodulatory drug is the presence of CD8+ at the tumor margin.72

Treatment of melanoma has changed significantly since the introduction of immunotherapy scarcely 5 years ago, and survival rates are much higher than those obtained with previous treatments.

Nevertheless, the discovery of any new treatment necessitates new studies to better understand the underlying mechanisms of action. Research in the field of immunotherapy and melanoma should focus on identifying criteria for selecting the most suitable candidates for treatment and exploring which drugs or combinations of drugs produce the best responses. The effectiveness of immunomodulatory drugs could be even further enhanced by combining their use with other approaches, such as targeted therapy and oncolytic viral therapy.

In conclusion, boosting the immune response is a key factor in the treatment of melanoma and will probably form part of all treatment regimens yet to come.

The authors declare that they have no conflicts of interest.