CTLA-4, a B7/CD28 family member, is a coinhibitory receptor expressed on the surface of T-cells that eventually inhibits T-cells, and it is expressed by regulatory T-cells (Tregs) (14). Discovered in 1987, it was considered to function as a negative regulator of T-cell activation until the mid-1990s (15-17).

When CTLA-4 is expressed on the surface of CD4⁺ and CD8⁺ T-cells, the binding affinity increases and is higher to CD80 and CD86, which are the costimulatory receptors present on antigen-presenting cells (APCs) compared to CD28 which is another costimulatory receptor (18). The expression of CTLA-4 increases when there is an activation of T-cell receptors and the release of cytokines such as IL-12 and interferon (IFN)-γ. This upregulation creates a feedback inhibition loop on T-effector cells which are activated, leading to CTLA-4 acting as a natural ‘brake’ on CD4⁺ and CD8⁺ T-cell activation induced by APCs, as shown in Fig. 3.

Tregs also play a key role in maintaining immune homeostasis by inhibiting excessive immune responses. One of the mechanisms through which Tregs suppress effector T-cell activity is via CTLA-4 signaling (19). Two anti-CTLA-4 drugs have been studied in patients with melanoma: i) Ipilimumab, the first ICI evaluated and approved for the treatment of melanoma is a fully human immunoglobulin anti-CTLA-4 monoclonal antibody (20,21); ii) tremelimumab, a fully human immunoglobulin anti-CTLA-4 monoclonal antibody which is still under investigation (12).

There are two major mechanisms through which these drugs act. First, the inhibition of CTLA-4 signaling in cytotoxic T-cells that specifically target tumors can directly affect these cells by enabling them to evade a state of anergy and enter an active proliferative effector phase. Once activated, these effector T-cells are more likely to penetrate the tumor and exhibit direct cytotoxic effects on tumor cells, while also releasing cytokines such as IL-2 and IFN-γ to stimulate an immunogenic tumor microenvironment. Thus, by blocking the CTLA-4 pathway, T-cells that were previously inactive can become activated and effectively target the tumor cells, causing a more powerful immune response against the cancer. This new approach holds promise as a potential immunotherapy for the treatment of cancer (22).

The second major mechanism driving these drugs is the blocking of CTLA-4 signaling in Tregs, which may impair their ability to halt the activity of effector T-cells. This inhibition of CTLA-4 signaling can either cause a decrease in the number of Tregs or reduce their function without affecting their population size. Therefore, blocking CTLA-4 on Tregs may disrupt this suppression and lead to increased immune activation against tumor cells (19,23,24).

Patients with melanoma are treated with the primary aim of suppressing the molecular interplay between the melanoma cells and immune effector cells. Ipilimumab, which has mainly been approved for the treatment of more advanced stages, such as unresectable or metastatic melanoma, has been shown to be associated with a marked overall survival rate confirmed from a phase 3 clinical trial (25). The interference of ipilimumab on CTLA-4 expressed on the subset of tumor-specific T-cell proliferation and B7 molecules on antigen-presenting cells is expected to prevent tumor development (26).

PD-1 is a protein present on the surface of T-cells, B-cells and natural killer (NK) cells. It functions as an inhibitory molecule by binding to PD-L1 (or B7-H1) and PD-L2 (B7-H2). PD-L1 is expressed in numerous types of tissue, including hematopoietic cells and certain tumors such as melanoma, where they are expressed in 40-50% of cases. PD-L2 is mainly expressed in hematopoietic cells. The binding of PD-1 to PD-L1/2 inhibits the death of tumor cells and promotes the conversion of T-effector cells into Tregs, while also inducing exhaustion in peripheral T-effector cells, as illustrated in Fig. 4 (27,28).

PD-1 and/or PD-L1 are also expressed on cells, such as NK cells, monocytes and dendritic cells (27,28). The PD-1 pathway operates through various mechanisms, such as reducing the activity of T-cells during an inflammatory response, increasing the proliferation and suppressive activity of Tregs, and reducing the lytic activity of B-cells and NK cells (29).

The affinity between PD-1 and PD-L1 is 3-fold stronger than the affinity between PD-1 and PD-L2. When PD-L1 binds with PD-1 on T-cells, it results in T-cell exhaustion, dysfunction, neutralization and the production of IL-10 within the tumor mass. This process allows tumors that overexpress PD-L1 to protect themselves from being attacked and killed by CD8⁺ cytotoxic T-cells (30).

Pro-effector cytokines, namely IL-12 and IFN-γ, can upregulate the expression of PD-1 and PD-L1/L2, which helps to prevent excessive T-effector cell activity. It is worth noting that PD-L1 has also been shown to inhibit CD80, indicating the existence of complex interactions between CTLA-4, PD-1 and other pathways (31,32).

The PD-1 and PD-L1 antibody inhibitors were created with the aim of preventing the PD-1 or PD-L1 side from functioning, reactivating T-cells and promoting an immune response against cancer cells (13).

Based on promising results from clinical trials, antibodies that inhibit PD-1 (such as pembrolizumab, nivolumab-IgG4 fully humanized and dostarlimab), as well as those that inhibit PD-L1 (such as avelumab, atezolizumab, and durvalumab) are being evaluated for use in melanoma cases and various other malignancies (NCT04020809, NCT04274816, NCT03313206, NCT03842943 and NCT05928962). However, it is not yet known which inhibitor, PD-1 or PD-L1, is more efficient (13). Of these, nivolumab and pembrolizumab are the two major FDA-approved anti-PD-1 monoclonal antibodies available for the treatment of advanced and metastatic melanomas.

Melanoma cells exhibit increased levels of PD-L1, which promotes the apoptosis of the likewise increased levels of T-cells (33). It has also been found that the circulating melanoma antigen-specific T-cells and tumor-infiltrating lymphocytes express PD-1 abnormally. It is considered that melanoma cells are capable of initiating, as well as sustaining PD-1 signals, T-cell fatigue and dysfunction (33). Hence, by blocking PD-1 in patients with melanoma, one could possibly restore abnormal activation and signaling and eventually recover the immune effect. Pembrolizumab or lambrolizumab were used in unresectable or metastatic melanoma in the study by Hamid et al (34), in an aim to elucidate the effects of PD-1 medications in melanoma.

However, due to the heterogeneous nature of tumors, the expression of PD-L1 is not uniform throughout. The extent of PD-L1 expression can differ in various locations within the tumor, resulting in varying levels of PD-L1 in immunohistochemical staining. Moreover, the effectiveness of PD-L1/PD-1 inhibitors can also be influenced by several other factors, such as the type of cancer, the patient's immune system and the genetic profile of the tumor. Thus, a more in-depth understanding of these factors is essential for developing effective treatment strategies that consider the heterogeneity of tumors and the variability in PD-L1 expression (35).

The combination ICI therapy used in the treatment of patients with metastatic melanoma primarily involves CTLA-1 and PD-1 inhibitors. This amplified the inhibitions that can be simultaneously intervened during different phases of the interaction among melanoma cells and the immune system. This, for example, includes anti-CTLA-4 inhibiting the priming stage at the same time anti-PD-1 inhibits the effector stage (36,37). It has also been noted that the use of anti-CTLA-4 inhibitors results in an increased expression of PD-1; hence, using combination therapy results in a more robust treatment response in patients with melanoma (38).

An increased understanding of immunological mechanisms has led to the identification of additional potential targets for checkpoint inhibition in the treatment of cancer. Some of these potential targets include BTLA, VISTA, TIM-3, CD47 and LAG-3. i) The blockade of BTLA has been shown to enhance New York esophageal squamous cell carcinoma 1) specific CD8⁺ T-cell function and enhance the efficacy of anti-PD-1 (39-41). ii) VISTA blockade has been shown to increase T-cell infiltration and function in tumors, thereby reducing tumor growth (10,42). iii) TIM-3 blockade causes T-helper-1 cell hyperproliferation and cytokine release, leading to tumor shrinkage in a mouse model when combined with anti-CTLA-4 or anti-PD-1 (43-47). iv) Targeting CD47 with a humanized anti-CD47 monoclonal antibody in combination with rituximab has shown to lead to objective responses in half of the heavily pretreated patients with relapsed or refractory non-Hodgkin's lymphoma, including a complete response in more than one-third of patients (48). v) An immune pathway known as LAG-3 has been identified as a potential complement to the PD-1/PDL1 pathway in enhancing the immune response against cancer. LAG-3 is an immune checkpoint receptor that regulates the function of T-cells. BMS-986016 is a therapy that targets LAG-3 and is currently under investigation in combination with nivolumab, which targets PD-1, to enhance the immune response against cancer cells. The combination of these two therapies has the potential to create a synergistic effect, leading to improved treatment outcomes for patients with cancer (49).

Ipilimumab is a fully human monoclonal antibody developed to antagonize CTLA-4. A clinical study was conducted on patients who had unresectable stage III or IV melanoma, where they were randomly assigned in a 3:1:1 ratio to receive ipilimumab (3 mg/kg) plus the glycoprotein 100 (gp100) vaccine, ipilimumab alone, or gp100 alone. The patients who received ipilimumab plus gp100 had a longer median overall survival rate of 10 months compared to 6.4 months for those who received gp100 alone, with a hazard ratio (HR) for mortality of 0.68 and a statistically significant P-value of <0.001. The median overall survival rate of patients who received ipilimumab alone was 10.1 months (HR, 0.66; P=0.003) compared to those who received gp100 alone (20).

In the study by Robert et al (25), patients with untreated metastatic melanoma were randomly assigned to receive either ipilimumab plus dacarbazine or dacarbazine alone. A significantly longer median overall survival (11.2 months) was found in the ipilimumab plus dacarbazine group compared to the dacarbazine alone group (9.1 months). In a follow-up maintenance study, patients who received ipilimumab plus dacarbazine had a higher 5-year survival rate (18.2%) compared to those who received dacarbazine alone (8.8%). These findings suggest that ipilimumab in combination with dacarbazine may be an effective treatment option for metastatic melanoma patients (76).

In a trial for ipilimumab as an adjuvant treatment performed by Eggermont et al (77), patients with stage III cutaneous melanoma who had undergone complete resection were randomly administered either ipilimumab (10 mg/kg) (n=475) or a placebo (n=476). Following a median follow-up of 5.3 years, the ipilimumab group had a 5-year recurrence-free survival rate of 40.8%, while the placebo group had a rate of 30.3% (HR, 0.76; P<0.001). In terms of the 5-year overall survival rate, the ipilimumab group had a rate of 65.4% compared to 54.4% in the placebo group (HR, 0.72; P=0.001) (77).

In the study by Ascierto et al (78), patients with unresectable stage III or IV melanoma were randomly assigned to receive either 10 or 3 mg/kg ipilimumab. The median overall survival rate was 15.7 months for the 10 mg/kg group and 11.5 months for the 3 mg/kg group (HR, 0.84; P=0.04). Overall, their study suggested that ipilimumab treatment improved the survival outcomes of patients with unresectable stage III or IV melanomas, with higher doses of the drug (10 mg/kg) leading to an improved overall survvial compared to lower doses (3 mg/kg) (78).

Nivolumab is a monoclonal antibody that inhibits the interaction of PD-1 with PD-L1. The clinical study performed by Robert et al (79) compared the efficacy of nivolumab with the standard therapy of dacarbazine. In their study, patients who had metastatic melanoma without a BRAF mutation were randomly divided into two groups (1:1 with nivolumab at 3 mg/kg once every 2 weeks (n=210) and dacarbazine (n=208). The survival rate at 1 year was 72.9% in patients treated with nivolumab compared to 42.1% in patients who were assigned dacarbazine (HR, 0.42, P<0.001). The objective response rate was 40% with nivolumab compared to 13.9% with dacarbazine (odds ratio, 4.06; P<0.001) (79). Another randomized controlled trial was carried out between 2012-2014 on patients with advanced melanoma who progressed after ipilimumab therapy or a combination of ipilimumab and a BRAF inhibitor if they were found to be positive for a V600E mutation (80). That study assessed the role of nivolumab as a second-line treatment in the management of patients with advanced melanoma. Patients were divided into three groups in a 2:1 pattern where one group (n=272) received nivolumab at 3 mg/kg once every 2 weeks and another group (n=133) received the investigator's choice of chemotherapy (ICC), which was either dacarbazine or paclitaxel plus carboplatin (80). An interim analysis of that study found that, in the first 120 patients of the nivolumab group, 38 patients (31.7%) experienced confirmed objective responses, whereas only 5 out of the 47 patients (10.8%) receiving the ICC treatment exhibited similar responses. Subsequently, upon further analysis of that trial, it was revealed that the median overall survival rate of patients who received nivolumab was 16 months, while for those who received ICC, it was 14 months (81). The HR was 0.95, indicating that nivolumab did not improve the survival rate of patients who had ipilimumab-refractory metastatic melanoma when compared to ICC. However, nivolumab had a higher overall response rate of 27% vs. 10% for ICC, and the median duration of response was also longer for nivolumab at 32 months compared to 13 months for ICC (77). Hence, nivolumab exhibiting a higher overall response rate and a longer duration of response suggests that it may be a more effective treatment option for some patients (81). Another study was also carried out to compare the efficacy of nivolumab compared to ipilimumab as an adjuvant therapy in patients who had resected advanced melanoma. In patients with stage III or stage IV melanoma, adjuvant therapy was administered with either nivolumab (n=453) or ipilimumab (n=453) and follow-up was performed after 18 months (82). The 12-month rate of recurrence-free survival was significantly higher in the nivolumab group at 70.5%, vs. 60.8% in the ipilimumab group with a HR of 0.65 (P<0.001). It was also noted that treatment-related adverse events were 14.4% for patients treated with nivolumab and 45.9% for those treated with ipilimumab. Therefore, patients who received ipilimumab therapy experienced more severe side-effects than those who received nivolumab therapy. This suggests that nivolumab may be a more effective and tolerable treatment option for patients with stage IIIB, IIIC, or IV melanoma following surgical resection (82).

Pembrolizumab is a monoclonal antibody which functions by blocking the PD-1 on T-cells and allowing these T-cells to identify and kill cancer cells. Similar to nivolumab, pembrolizumab was also compared with ICC in ipilimumab-refractory melanoma. A randomized controlled study was conducted on patients with advanced melanoma and have progressed even after receiving ipilimumab and/or standard BRAF therapy (83). Patients were divided into three groups as follows: One group (n=181) received 10 mg/kg pembrolizumab, one group (n=180) received 2 mg/kg pembrolizumab, and another group (n=179) received ICC. The 6-month progression-free survival rate was found to be 38% in patients treated with pembrolizumab at 10 mg/kg (HR, 0.5 vs. ICC; P<0.0001), 34% in the 2 mg/kg group (HR, 0.57 vs. ICC; P<0.0001) and 16% in the ICC group (83). Another study was conducted by Robert et al (84), this time comparing pembrolizumab with ipilimumab. Patients with advanced melanoma were divided at a 1:1:1 ratio to receive pembrolizumab at 10 mg/kg once every 2 weeks or pembrolizumab at 2 mg/kg once every 3 weeks or four doses of ipilimumab at 3 mg/kg for once every 3 weeks. An interim analysis was performed which revealed that the 6-month progression-free survival of the patients treated with pembrolizumab once every 2 weeks was 47.3% (HR, 0.58 vs. ipilimumab; P<0.001), 46.4% for those treated with pembrolizumab once every 3 weeks (HR, 0.58 vs. ipilimumab; P<0.001) and 26.5% for those treated with ipilimumab (84). A final analysis revealed that the median overall survival rate was not reached in both pembrolizumab groups; however, it was noted to be 16 months in the ipilimumab group (HR, 0.68 for pembrolizumab once every 2 weeks vs. ipilimumab, P=0.0009; and HR, 0.68 for pembrolizumab once every 3 weeks vs. ipilimumab, P=0.0008) (85). Similarly, in the study by Robert et al (84) the 24-month overall survival rate was 55% in the group treated once every 2 weeks, 55% in the group treated once every 3 weeks and 43% in the ipilimumab group. Not only do nivolumab and pembrolizumab prolong overall survival, but they also maintain the quality of life of patients with melanoma (86,87). These findings have led to the FDA approval of pembrolizumab for ipilimumab and/or BRAF inhibitory refractory advanced melanoma. A summary of the comparison among ICIs and their outcomes in patients is presented in Table III.

The role of serum biomarkers in the early detection of melanoma is described in Table IV (88-106).

i) Tumor infiltrating lymphocyte (TIL) patterns are often divided into grades, such as ‘absent’, which is no presence of any lymphocytes within the tumor, ‘non-brisk’, which suggest few foci of lymphocytes within the tumor, or ‘brisk’, which is a large diffuse infiltration of lymphocytes within the tumor (107). As demonstrated by Clark et al (107) in 1989, as well as by others, the presence of brisk TILs in a vertical growth pattern is often associated with a favorable disease-specific survival and overall survival rate after non-brisk and absent patterns of TILs (107,108).

ii) Histotype: The majority of melanoma histotypes are not considered prognostic when looked at individually from tumor thickness, and are therefore not included in the American Joint Committee on Cancer staging system (90,109,110). However, a nodular melanoma is an independent predictor which can be used for the measurement of recurrence and its association with mortality due to melanoma (111).

iii) Digital images trained from AI: New advancements have allowed for the development of deep learning-based biomarkers, which can help to stratify the stages of melanoma into risk groups, and thus associate disease-specific survival with two independent validating cohorts to accurately predict the prognosis of patients with early-stage melanoma (112).

iv) Melanoma cell adhesion molecule (MCAM): Expressed in 80% of metastatic tumors, MCAM is a cell adhesion marker (113). Those who are positive for MCAM have significantly worse 5-year survival rates than those who are negative for MCAM, and there is an inverse association between the amount of marker expressed and survival (114,115).

v) Ki-67: Ki-67 is a unique nuclear antigen that can function as a marker for cellular proliferation during the active phase of the cell cycle (116). For melanomas who have a thickness <1 mm, the risk of metastasis increases with the expression of Ki-67 and an increased mitotic rate (117). However, with the increasing thickness of melanomas, Ki-67 can serve as a more effective prognostic marker than the mitotic rate, and is often associated with ulceration within the tumor, necrosis, higher level Clark's level of invasion, and even vascular invasion (118). In addition, with recurrent melanomas, higher values of Ki-67 exhibit an independent association with a decreased overall survival (119).

vi) Lymphatic invasion: In research on primary melanomas with a thickness >1 mm, D2-40 staining was assessed for lymphatic invasion, which is an antibody against sialoglycoprotein that selectively attaches on endothelial cells of lymphatic vessels and helps detect sentinel lymph node metastasis (120-122).

vii) Osteopontin: Overexpressed in numerous visceral malignancies, osteopontin is known as an integrin-binding protein and used as a biomarker to measure tumor progress and metastasis (123-125). It functions as an independent predictor for the prognosis of melanoma and was found to be associated with increased sentinel lymph node positivity in a cohort of 345 patients who had primary melanoma detected using immunohistochemical analysis (126).

viii) Driver mutations: It has been found that BRAF and NRAS are associated with a significantly lower melanoma-specific survival in high-risk tumors, such as a stage >2(127). NF1 mutations has also been found to be associated with a lower disease-specific survival and overall survival (128). However, further research is required to identify patients with BRAF mutations and uncover the role of BRAF mutations in directing the treatment strategy.

Predictive markers. The role of predictive markers in melanoma and its clinical importance is summarized in Table V (61,129-151).

Diarrhea is the most frequent irAE with incidences between 10 and 13% (164).

In decreasing order, the first endocrine system that is most affected by ICIs is the thyroid gland (typically hypothyroidism observed following a transient thyroiditis-induced thyrotoxicosis) followed by the rest of the endocrine organs. The median time frame from the start of the treatment to the development of thyroid symptoms, most commonly hypothyroidism, is 6 weeks, followed by pituitary (hypophysitis), adrenals (primary adrenal insufficiency) and β-cells of the pancreas (insulin deficient diabetes, analogous to type 1 diabetes) (170). These are different from the side-effects brought on by conventional cytotoxic chemotherapy or even more recent molecular-targeted medicines, which infrequently result in endocrine dysfunction (171).

Non-specific adverse events such as maculopapular rash, pruritus, psoriasiform, eczematous and lichenoid dermatosis are among the most prevalent (172,173). Compared to anti-PD-1 monotherapy, the maculopapular rash phenotype is more prevalent when CTLA-4 inhibition is implemented (21). Bullous pemphigoid, vitiligo-like skin hypopigmentation/depigmentation and alopecia are other less-common irCAEs (174,175). Although severe reactions, such as Stevens-Johnson syndrome, toxic epidermal necrolysis and drug reaction with eosinophilia and systemic symptoms are uncommon, cutaneous consequences are typically self-limiting (174-176). Early diagnosis and the administration of corticosteroids or antitumor necrosis factor-agents are the foundation of treatment algorithms for irCAEs (176,177). However, the use of corticosteroids before or after ICI initiation may result in a diminished antitumor efficacy. Anti-CTLA-4 and anti-PD1 therapy have both been associated with reports of vitiligo (178). The occurrence of skin hypopigmentation or depigmentation such as vitiligo has been linked to an extensive anticancer benefit from drug treatment in patients with melanoma. Vitiligo has been proven as a positive predictive factor in measuring the tumor response to treatment. In comparison with the general population, patients with melanoma have a 10-fold increased incidence of drug-related cutaneous hypopigmentation and depigmentation (179). Since the PD-L1:PD1 pathway mostly regulates the peripheral tolerance of melanosomal proteins (such as tyrosinase and TRP-2), the interference of PD-1 signaling may result in autoimmune vitiligo (180). This offers a reasonable explanation for the onset and durability of depigmentation in patients receiving immunotherapy.

Case series studies have demonstrated that patients develop organizing pneumonia, diffuse alveolar damage, acute respiratory distress syndrome (ARDS) and non-specific interstitial pneumonia, which is then managed by intravenous and oral steroids (181-185).

A previous meta-analysis revealed adverse effects associated with the use of anti-PD-1/PD-L1 monoclonal antibodies for malignancies with an increased incidence of pancreatitis, and increased levels of liver enzymes, such as aspartate aminotransferase and alanine transaminase, elevated creatinine levels, nephritis and renal failure (164).