Immunotherapy of Metastatic Melanoma

Abstract

Immunotherapy is part of the new treatments that significantly improved the prognostic of metastatic melanoma patients. The article reviews briefly the old immunotherapeutic approaches e.g., interferon-𝛼2 and interleukin-2, and focuses on immune checkpoint inhibitors such as anti-CTLA-4 inhibitors and anti-PD-1 inhibitors in monotherapy or in combination (dual immune blockade). We detailed the results from CheckMate and KEYNOTTE clinical trials that lead to US Food and Drug Administration and European Medicines Agency approvals of the new agents for the treatment of advanced melanoma. The chapter concentrates on the algorithms for BRAF wild-type and BRAF mutated metastatic melanoma treatments, according to American (NCCN) and European (ESMO) guidelines. We underlined the first line, second line, and subsequent lines of treatment for both melanoma subtypes and for particular cases, such as in-transit metastasis or brain metastasis. A special attention was paid to treatment options for early and late disease progression (primary and acquired resistance after adjuvant therapy). Unfortunately, the new immune agents produce a higher toxicity rate, mainly immune adverse events. Also, these drugs can interact with the gut microbiome and with antibiotics, decreasing the efficacy of immune therapy. Finally, we review the new directions for immune therapy e.g., new immune combinations, the association of immune and targeted therapies, and adoptive cellular therapy with tumor-infiltrating lymphocytes, interleukin-2, and anti-PD-1.

1. Introduction

For more than a century, it is a known fact that cancer is an inflammatory disease and that immunotherapy (IT) can be used as a strategy for fighting it. Coley’s toxin, utilized as early as 1893, can be considered the first IT approach in cancer [1].

At the beginning, the research was focused on the activation of the immune response using antitumor vaccines or direct stimulation with recombinant cytokines, e.g., interferons and interleukin-2 [2, 3].

Interferon alpha-2 (IFN-𝛼2) was the first immunotherapeutic agent approved by the US Food and Drug Administration (FDA) in 1995, for adjuvant treatment of stages IIB-III melanoma patients, based on the results of the ECOG EST 1684 trial [4]. Previous outcomes of phase I/II studies demonstrated a tumor response rate of ~16%, but with a modest median duration of response (of ~4 months) for the treatment of disseminated melanoma [5, 6]. A meta-analysis of 13 trials published in 2018 showed a median relapse-free survival of 2.2 years (1.2–3.3 years) for the patients who received different IFN-𝛼2b regimens, compared with 1.9 months for the patients that did not receive any adjuvant treatment for stages II and III [7].

The second cytokine approved by FDA in 1998 for the treatment of metastatic melanoma was Interleukin 2 (IL-2), due to its proven potential for durable disease control [8]. The administration of two cycles of high-dose IL-2 (HD-IL-2), each of them receiving 600,000 to 720,000 IU/Kg/per dose intravenously, every 8 hours, for up to a maximum of 14 doses per cycle, leads to clinical responses in ~16% of patients, including ~6% who had complete responses [9].

Unfortunately, the responses were infrequent and associated with severe side effects, especially for HD-IL-2, such as capillary leak syndrome (with hypotension, pulmonary edema, and renal failure), hepatic, gastrointestinal, endocrine, and cutaneous toxicities, arrhythmias, and psychiatric disturbances [10]. These toxicities generally resolve in a few days after stopping HD-IL-2 therapy, but the mortality rate related to this treatment is 1–2% [11].

The combination of cytokines, IFN-𝛼 and/or IL-2, with chemotherapeutic agents (e.g., cisplatin, vinblastine, and dacarbazine), also named bio-chemotherapy, can enhance the response rates, but at the cost of significantly increased toxicity. Multiple prospective randomized clinical trials failed to demonstrate significant improvement in survival compared to chemotherapy alone [12].

The understanding of the mechanisms through which the immune system fights against cancer represents one of the greatest breakthroughs in medicine over the last 15 years [13].

The interaction between Cytotoxic T Lymphocyte-associated antigen 4 (CTLA-4) and Programmed cell death 1 (PD-1) receptors and their ligands, discovered by the teams that won the 2018 Nobel Prize for Medicine, led by James Allison [14, 15] and Tasuku Honjo [16, 17], became the foundation of the development of immune checkpoint inhibitors (ICI).

Immune checkpoints (IC) are negative regulators of T-cell activation. Along with co-stimulatory molecules, they have an important role in maintaining self-tolerance.

2. CTLA-4 inhibitors

The anti-CTLA-4 monoclonal antibodies (mAbs) ipilimumab, a fully human IgG1, and tremelimumab, a fully human IgG2, were the first IC blocking drugs to enter clinical trials in oncology; however, the only one approved by FDA for metastatic melanoma was ipilimumab, in March 2011, initially as a single-therapy.

The approval of ipilimumab as monotherapy for unresectable stage III or stage IV melanoma was based on the results of two clinical trials, CA 184–002 [18] and CA 184–024 [19]. The first trial compared ipilimumab 3 mg/Kg, 4 cycles at 3-weeks interval, single-agent therapy or in combination with glycoprotein (gp-100) peptide vaccine, with gp-100 vaccine monotherapy [18]. The second trial compared ipilimumab 10 mg/Kg, 4 cycles at 3-weeks interval, in combination with dacarbazine, with dacarbazine alone until week 22; the responders (patients with stable disease or patients with an objective response, and no unresolved adverse events) received ipilimumab or placebo every 12 weeks thereafter as maintenance therapy [19]. The results of both trials showed, for the patients who received ipilimumab, an improved response rate and an increase in the duration of the response, in addition to better results for PFS (progression-free survival) and OS (overall survival) for both previously treated [18] or untreated advanced melanoma patients [19]. The CA 184–169 clinical trial compared the standard and high doses of ipilimumab; the survival results were not significantly different [20].

Pooled data from several phases II and phase III trials demonstrate a median survival time of 11.4 months for ipilimumab monotherapy [21]. The survival curves reached a plateau after 3 years and appeared stable even after 10 years [21]. In CA 184–024 trial, approximately 20% of the patients treated with ipilimumab showed longer overall survival compared with chemotherapy (18.2% 5-year OS for ipilimumab in combination with dacarbazine versus 8.8% for dacarbazine alone) [22].

The real-world data from the Expanded Access Program for Ipilimumab confirmed the efficacity of this therapy for previously treated metastatic melanoma patients [23, 24, 25, 26, 27, 28, 29, 30]. More than 1600 patients were treated with single-agent ipilimumab 3 mg/Kg, 4 cycles at 3-weeks interval (induction phase). The median PFS and median OS were similar between 6 European countries and South Africa (Table 1).

Country	Patients number	median PFS (months)	median OS (months)	1-year OS (%)	2-year OS (%)
Czech Republic (23)	196	—	7.5	—	—
Italy (24)	855	3.7	7.2	—	—
Netherlands (25)	31	—	7	45.2	28.8
Poland (26)	50	3	8	—	—
Romania (27)	89	4.13	6.3	—	—
Spain (28)	144	—	6.5	32.9	—
South Africa (29)	108	3.44	—	36	20
UK (30)	193	2.8	6.1	31	14.8
Total	1616	(2.8–4.13)	(6.1–8)	(31–45.2%)	14.8–28.8%

Table 1.

The efficacity of ipilimumab monotherapy; results from expanded access program in 6 European countries and South-Africa (23–30).

Safety results showed a high risk, 10–15%, of severe (grade 3 and 4) immune-mediated adverse events (irAEs) for standard dose ipilimumab monotherapy [18], 30% for high-dose ipilimumab monotherapy [20], and 38% risk of severe irAEs for ipilimumab combined with dacarbazine [19]. The study CA 184–002 reported seven deaths caused by immune-mediated AEs [18].

As a result, clinical guidelines do not recommend the association of ipilimumab with dacarbazine due to high risk for severe adverse events, and the FDA-recommended dose of ipilimumab is now 3 mg/Kg instead of 10 mg/Kg, 4 cycles at 3-weeks interval (induction therapy) [31].

The second anti-CTLA-4 antibody, tremelimumab, also generated promising anticancer responses in early clinical trials [32]. Unfortunately, a phase III clinical trial of tremelimumab versus standard-of-care chemotherapy in advanced melanoma was stopped early due to a lack of survival benefits [33].

3. PD-1 inhibitors

The anti-PD-1 monoclonal antibodies, pembrolizumab, and nivolumab, humanized immunoglobulins (IgG4), were both approved as single-agent therapies by FDA in 2014 for unresectable advanced or metastatic melanoma.

Pembrolizumab is administered intravenously at 2 mg/Kg body weight, or 200 mg fixed dose every 3 weeks until progression of the disease or until a severe toxicity develops. The treatment can be administered continuously, over a period of 1–2 years, depending on the response of the disease and the tolerance of the treatment. However, the optimal treatment duration has not been established until now [34].

The initial results from the phase I KEYNOTE-001 clinical trial showed a response rate of 34% and a median OS of 25.9 months for ipilimumab refractory metastatic melanoma [35]. The KEYNOTE-002 clinical trial compared two pembrolizumab doses (2 mg/Kg and 10 mg/Kg every 3 weeks) with chemotherapy for the same population as the previous study [36]. Long-term follow-up showed that both doses of pembrolizumab provide higher response rates (22–28%) and longer duration of response along with improvements in progression-free survival (16–22% PFS 2-year rate), compared with chemotherapy (4% response rate and < 1% PFS 2-year rate) [37]. Furthermore, pembrolizumab therapy was better tolerated than chemotherapy [38].

In the end, the results of phase III KEYNOTE-006 clinical trial support the recommendation of American (NCCN) and European (ESMO) guidelines that pembrolizumab should be considered as first-line therapy in patients with unresectable or metastatic melanoma [39, 40]. The clinical trial compared two pembrolizumab regimens (10 mg/Kg every 2 or every 3 weeks) with ipilimumab for the patients with metastatic melanoma previously untreated with ICI [41, 42]. All the study endpoints aligned: 36–37% response rate for pembrolizumab compared with 13% for ipilimumab (statistically significant), 28–31% PFS 2-year rate versus 14% for ipilimumab (statistically significant), and a trend to improve the OS 2-year rate for pembrolizumab [42].

The kinetics of the response to pembrolizumab reflects the response to immunotherapy. Long-term follow-up during clinical trials showed a late response to pembrolizumab therapy, more than a year after the start of the treatment; in addition, some partial responders may become complete responders over time [37, 41, 43].

Nivolumab is administered intravenously at 3 mg/Kg body weight or 240 mg fixed dose every 2 weeks, or 480 mg fixed dose every 4 weeks until progression of the disease or until a severe toxicity develops.

The phase III study CheckMate 037 compared nivolumab with chemotherapy for the patients with ipilimumab-refractory metastatic melanoma (BRAF wild-type) and for the patients with ipilimumab and BRAF inhibitors refractory metastatic melanoma (BRAF mutated) [44]. Immunotherapy improved the response rate (27% versus 10%) and the duration of the response compared with chemotherapy, but after 2 years, it did not improve neither median PFS (3.1 versus 3.7 months) nor median OS (15.7 versus 14.7 months) [44, 45].

The subsequent phase III CheckMate 066 and 067 clinical trials demonstrated nivolumab efficacy in unresectable stage III and metastatic stage IV melanoma. In CheckMate 066, nivolumab monotherapy was compared with chemotherapy [46, 47]. The response rate (40% versus 13.9%), median PFS (5.1 versus 2.2 months), and median OS (37.5 versus 11.2 months) were statistically significant in favor of immunotherapy [46, 47]. Nivolumab therapy led to long-term survival in up to 40% of patients, as the survival curves suggest [47].

In the CheckMate 067 clinical trial, the dual immune combination of CTLA-4 and PD-1 inhibitors was compared with nivolumab (monotherapy) and with ipilimumab (monotherapy) as first-line treatments for metastatic melanoma; the results demonstrated the superiority of dual immune combination and also of single-agent PD-1 inhibitor over ipilimumab monotherapy [48, 49, 50]. In monotherapy, nivolumab was superior to ipilimumab in terms of response rate (45% versus 19%), median PFS (6.9 versus 2.9 months), and median OS (36.9 versus 19.9 months) [48, 49, 50].

The kinetics of the response to nivolumab, ipilimumab, and pembrolizumab was almost identical, with late complete response seen more than a year after the start of the treatment [45, 48, 50]. Across clinical trials, response to nivolumab tends to persist after the discontinuation of the drug [48, 50].

4. Dual immune blockade (CTLA-4 and PD-1 Inhibitors)

Preclinical studies demonstrated that dual immune blockade with anti-CTLA-4 combined with anti-PD-1 was more effective than with either alone [51]. A phase I study of immune combination therapies found that the maximum tolerated dose of concurrent administration is 3 mg/Kg q3w for ipilimumab and 1 mg/Kg for nivolumab q3w; in this study, the overall response rate was 40% and the grade 3–4 AEs rate was 53% [52].

The nivolumab and ipilimumab combination arms from CheckMate 067 and CheckMate 069 clinical trials showed higher response rates (58% vs. 19%, p < 0.001 for CheckMate 067 and 59% vs. 11% for CheckMate 069), prolonged response durations, longer time to subsequent therapies, prolonged median PFS (11.5 vs. 2.9 months, p < 0.001 for CheckMate 067 and not reached vs. 3.0 months in CheckMate 069), and larger median OS compared with single-agent ipilimumab [50, 53]. These effects persisted during long-term follow-up, with 4-year survival rates of 53% for the combination arm compared with 46% for single-agent nivolumab and with 30% for single-agent ipilimumab in the CheckMate 067 study [48]. For a subgroup of patients with high levels of PD-L1 expression, the median OS and median PFS were similar for single-agent nivolumab compared with the ipilimumab and nivolumab combination, but the number of toxicities was smaller for monotherapy [48].

Long-term follow-up (6.5 years) in the CheckMate 067 study showed a longer median OS of 72.1, 36.9, and 19.9 months in the combination arm compared with nivolumab and ipilimumab monotherapy [54].

CheckMate 067 and 069 showed significantly increased toxicity of dual immune blockade versus monotherapy [50, 53]. The rate of grade 3–4 related adverse events (AEs) in CheckMate 067 was 59% for the ipilimumab and nivolumab arm compared with 21% for nivolumab alone and with 28% for ipilimumab monotherapy [50]. In CheckMate 069 the rate of AEs for the combination was 54%, compared with 20% for ipilimumab monotherapy [53].

A pooled analysis of the immune combination trials found that response rates, PFS, and OS of the patients who discontinued the treatment in the induction phase due to the AE, were similar to those of the patients who completed the treatment [55].

The kinetics of the response to combination therapy includes a late complete response (CR) that was seen more than a year after the start of treatment, with a double rate of CR, and increased response duration [48, 49].

Subgroup analysis, in both CheckMate clinical studies, demonstrated improved efficacy with nivolumab and ipilimumab combination therapy, regardless of BRAF mutation status [48, 49, 50, 53].

In order to identify a possible biomarker that could predict the response to immunotherapy, the researchers assessed PD-L1 expression in tumor samples from the patients included in CheckMate and KEYNOTE trials [45, 47, 48, 49, 50, 56]. In these randomized clinical studies, the improved response rate, PFS, and OS for anti-PD-1 therapy had a statistically significant correlation with increased PD-L1 expression, [45, 46, 48, 49, 50, 56].

However, it was not possible to identify an expression level cutoff for PD-L1 with a cert prognostic value. Furthermore, in these clinical trials, there were patients who experienced durable responses to anti-PD-1 inhibitors, regardless of the PD-L1 expression in biopsy specimens [56].

At the present time, we know the following [39]:

1. Anti-PD-1 therapy (nivolumab) and dual immune therapy (ipilimumab in combination with nivolumab) efficacies appear to improve with increasing PD-L1 expression; however, this biomarker is not the only one that predicts the response to ICI [48].

2. For high PD-L1 tumor expression, improvements in outcome with dual immune therapy or with nivolumab monotherapy were similar; instead, for low PD-L1 tumor expression, the outcome was better with dual immune therapy [48].

3. Unlike CTLA-4 inhibitor monotherapy, the dual immune therapy led to good responses even in patients with very low PD-L1 tumor expression [48].

4. PD-L1 tumor expression cannot be used in order to exclude patients from anti-PD-1 monotherapy [39]; however, the use of combination therapy for patients with low PD-L1 tumor expression, in order to increase efficacy, and the use of PD-L1 monotherapy for patients with a high level of PD-L1 tumor expression, in order to decrease the toxicity, prove effective and are consequently preferred [39].

Treatment for stage III In-transit melanoma represents a real challenge for medical oncologists, dermatologists, and surgeons. Local therapy (e.g., intralesional injections) can be combined with regional therapy (e.g., Isolated Limb Perfusion and Infusion) and systemic therapy [39].

Talimogen laherparepvec (T-VEC), an agent that uses a modified herpes simplex virus to induce tumor cell lysis and deliver a localized expression of GM-CSF is the main intralesional agent approved for this indication, according to the results of a phase 3 clinical study [57]. T-VEC produced local durable response rates (16.3% versus 2.1% for injection of GM-CSF) and remission of oligometastatic disease (bystander effect). The overall response rate was superior for intralesional T-VEC compared with intralesional GM-CSF (26.4% vs. 5.7%, p < 0.001) with higher rates of complete response (11% vs. 1%) [57].

The AEs rate produced by T-VEC injection was 20%, with 11% serious-AEs (grade 3–4). The most frequent AEs were local, e.g., injection-site reactions (cellulitis, pain, and peripheral edema), but also systemic toxicities appeared (fatigue, chills, pyrexia, and other flu-like symptoms) [57].

Immune Checkpoint Inhibitors combined with T-VEC intralesional injections represent a new approach in clinical ongoing trials. At first, the combination ipilimumab with T-VEC was tested, with a spectacular reduction of tumor burden for the injected lesions and also for some distant lesions [58]. However, the good clinical response did not engender longer PFS, and the rate of AEs was higher for the combination, compared with both agents in monotherapy. The phase 3 MASTERKEY-265 trial combined pembrolizumab with T-VEC in order to improve previous results by reducing toxicities [59]. The anti-PD-1 and T-VEC combination demonstrated a 43% CR with 4-year PFS and OS rates of 55.9 and 71.4%, respectively [59].

5. Immune Checkpoint Inhibitors (ICI) and Brain Metastasis (BM)

Treatment of melanoma BM is a real challenge for oncologists and radiotherapists. Clinical studies confirmed that immune therapy can be used safely and efficiently, especially in asymptomatic patients with BM. The CA 184–042 study demonstrates the superiority of HD-ipilimumab in asymptomatic patients (compared with symptomatic patients) in terms of response rate (16% vs. 5%), median PFS (2.6 vs. 1.3 months), and median OS (7.0 months vs. 3.7 months). Interestingly, good response rates were obtained for both intracranial and extracranial disease [60]. The patients with asymptomatic BM from the CA 184–169 trial had the same median OS for HD-ipilimumab and for standard ipilimumab doses [20].

For PD-1 inhibitors, used in the asymptomatic BM population, clinical studies showed good response rates, 30% for pembrolizumab [61] and 29% for nivolumab [62], and also high median OS (17 months for pembrolizumab and 18.5 months for nivolumab) [61, 62]. In the subset of patients with symptomatic BM and leptomeningeal disease, usually with bad prognostic, the CA 209–170 study finds a comparable response rate (25%), to the response rate (29%) for asymptomatic BM, but with much lower median OS (5.1 vs. 18. months) [62].

The real impact on asymptomatic BM patients was seen in the CA 209–170 clinical trial arm treated with a dual combination of ipilimumab and nivolumab, with 57% response rate for extracranial disease and 46% for intracranial disease; the median OS was not reached for immune combination, compared with 18.5 months for nivolumab monotherapy [62]. The good results for dual combination were confirmed by the CheckMate 204 clinical trial, with a more than 50% response rate for both extra and intracranial disease and with median OS not reached [63].

NCCN and ESMO guidelines concluded that ipilimumab and nivolumab combination is superior to anti-PD-1 monotherapy and that anti-PD-1 therapy provides higher response rates and better median OS compared with ipilimumab monotherapy, especially for asymptomatic BM melanoma patients [39, 40].

Accordingly, whole-brain radiotherapy (WBRT) is now reserved, only with palliative intent, for symptomatic BM patients [40]. Stereotactic radiosurgery (SRS) has replaced WBRT for non-bulky (< 3 cm), <5–10 asymptomatic BM, as upfront therapy. For more advanced disease, guidelines recommend first-line systemic therapy, mainly combination immune therapy; in this case, SRS will be used as salvage therapy for disease progression [40]. SRS and immune therapy can be administered simultaneously, but with close MRI evaluation, as a result of the increased risk for asymptomatic radio-necrosis (15% of patients) [64].

6. The algorithm for BRAF wild-type (wt) melanoma treatment

The current first-line standard therapies for inoperable stage III and IV BRAF wt melanoma are the PD-1 blockade (nivolumab or pembrolizumab) and the dual blockade CTLA-4 and PD-1 (ipilimumab and nivolumab) (Table 2) [39, 40]. Different regimens and doses from clinical guidelines are underlined in Table 3.

Inoperable stage III/IV melanoma
	Therapy	Melanoma subtype
First-line	Immune therapy; Anti-PD-1/Anti-CTLA-4 combination Anti-PD-1 monotherapy T-VEC	BRAF wt; BRAF mutated
	Targeted therapy; BRAFi + MEKi	BRAF mutated
Second-line	Clinical Trial (after IT)	BRAF wt
	IT rechallenge: ipilimumab after PD-1 monotherapy PD-1 therapy after bridging treatment (e.g., chemo) anti-PD-1 and anti-CTLA-4 combination if it is not given previously	BRAF wt
	Switched therapy (TT after IT or IT after TT)	BRAF mutated
Subsequent lines	pembrolizumab / low dose ipilimumab combination for tumors that have progressed after prior anti-PD-1 therapy HD-IL-2* Ipilimumab and intralesional T-VEC combination* pembrolizumab and Lenvatinib combination* BSC	BRAF wt
	rechallenge with both TT and IT chemotherapy** BSC	BRAF mutated

Table 2.

Algorithm for inoperable stage III/IV melanoma treatment (IT – Immune therapy; TT – Targeted therapy; BSC – Best supportive care).

^*

Only for NCCN guideline [39].

^**

Only for ESMO guideline [40].

Treatment	Dosing	Treatment duration
Nivolumab	240 mg q2w or; 480 mg q4w	until disease progression or unacceptable toxicity; most common regimens in daily-practice
	3 mg/Kg q2w	until disease progression or unacceptable toxicity; it is allowed to continue the treatment for clinical benefit even in the case of progression of disease.
Pembrolizumab	200 mg q3w	until disease progression or unacceptable toxicity; most common regimens in daily-practice
	2 mg/Kg q3w; 10 mg/Kg q2w or q3w	until disease progression or unacceptable toxicity; the treatment can be stopped after 24 months for complete responders
Ipilimumab / Nivolumab combination	1 mg/Kg nivo + 3 mg/kgc ipi q3w for 4 doses, followed by nivo; 240 mg q2w or; 480 mg q4w	until disease progression or unacceptable toxicity;
	1 mg/kg nivo + 3 mg/kgc ipi q3w for 4 doses, followed by 3 mg/kg nivo monotherapy q2w	until disease progression or unacceptable toxicity; it is allowed to continue the treatment for clinical benefit even in the case of progression of disease.

Table 3.

Immune checkpoint inhibitor treatment regimens.

Adapted from NCCN guideline [40].

T-VEC is also an option for in-transit unresectable melanoma.

For second-line treatment and beyond, ESMO guidelines [40] recommend clinical trials if available, or ICI rechallenge.

Immune therapy rechallenge includes at least 3 options [40]:

• Ipilimumab after PD-1 monotherapy (nivolumab or pembrolizumab) [65]

• Nivolumab or pembrolizumab if another line of treatment was given after ICI failure (e.g., chemotherapy)

• Ipilimumab and nivolumab combination if not given previously [66]

Two clinical trials demonstrated that immune therapy with ipilimumab or with the dual combination ipilimumab and anti-PD-1 should be considered a viable treatment option after failure of anterior PD-1 therapy [65, 66]. The combination appeared to be highly effective in terms of response rate, duration of the response, and median OS compared with ipilimumab monotherapy (20.4 vs. 8.8 months for median OS) [66]. The grade 3–5 toxicities for both groups were the same [66].

The NCCN guidelines have additional recommendations [39]:

• Pembrolizumab + low-dose ipilimumab combination for tumors that progressed after prior anti-PD-1 therapy [66]

• HD-IL-2

• Ipilimumab + intralesional T-VEC combination

• Pembrolizumab + lenvatinib combination

The combination of anti-PD-1 (pembrolizumab) with VEGF inhibitor (lenvatinib) produces a higher overall response rate of 48% compared with pembrolizumab alone, in a small phase I/II trial [67].

In particular cases, other options can be considered, such as imatinib for tumors with activating mutations of Kit, larotrectinib, and entrectinib for NTRK gene fusion-positive tumors [68] and cytotoxic agents.

7. The algorithm for BRAF mutated melanoma treatment

The current first-line treatment for inoperable stage III and IV BRAF-mutated melanoma is also immune therapy (IT), or the dual combination of BRAF inhibitors with MEK inhibitors (TT, Targeted Therapy) (Table 2). The best sequence of IT and TT is currently unknown [69]. No direct randomized comparison exists between IT and TT, but one meta-analysis suggests a better outcome after 1 year in favor of IT [70, 71], despite a very good response rate to TT in the first 12 months [72]. The main advantage of immune therapy as a first option is long-term/durable disease control even after treatment is ended [73].

There are several ongoing trials that study the optimal sequence for the first-line treatment, TT-IT or IT-TT (SECOMBIT, DREAMseq). Randomized three-arm phase 2 study (SECOMBIT / NCT02631447) revealed a better trend for OS and total PFS at 2 and 3 years for the arm with upfront ipilimumab and nivolumab combination and for the arm with short targeted therapy followed by immune combination therapy, compared with upfront targeted therapy with BRAFi + MEKi [74]. A randomized DREAMseq trial was designed to compare the efficacy and toxicity of the sequence IT-TT (Ipilimumab + nivolumab – dabrafenib + trametinib) with the sequence TT-IT. OS and duration of overall response (DOR) were better for upfront immune combination therapy (2-year OS of 72% vs. 52% p = 0.0095 and median DOR not reached for upfront immune therapy, and 12.7 months for targeted therapy) [75].

ESMO recommendation [76]:

• Elevated LDH level: ipilimumab and nivolumab combination preferred.

• LDH >1x and ≤ 2x ULN – anti-PD-1 monotherapy preferred.

• The tumor burden is not clearly defined yet.

• Switching TT to IT after short therapy should not be considered outside clinical trials.

First-line therapy selection for BRAF-mutated melanoma should be based on treatment goals (short-term benefit or long-term benefit), on the clinical characteristics of the disease (LDH level, organs involved, number of metastases or tumor burden, disease progression kinetics), on co-morbidities and performance status of the patient, and on the patient’s preference and compliance for oral or iv agents [76]. However, it seems prudent to start with immune therapy for the cases with tumors that do not progress very quickly and do not immediately threaten an important organ or function [40].

ESMO recommendation [76]:

• Patients treated by TT-IT sequence can be rechallenged with targeted therapy.

• Patients treated by IT-TT sequence can be rechallenged with anti-PD-1 therapy (no data exist for ipilimumab and nivolumab combination).

• Patients treated with first-line anti-PD1 monotherapy and second-line TT might benefit from ipilimumab-based treatment.

• Finally, after using all options, rechallenge with the drugs that showed the best response should be considered.

As a second-line treatment, NCCN and ESMO guidelines recommend the switch from one treatment to another, depending on the previously used first-line therapy (Table 2) [39, 40].

Subsequent lines are not well established; as an option, clinical trials or rechallenge with both TT and IT can be considered. Another option can be chemotherapy with single-agent DTIC or Temozolomide, and Paclitaxel + Carboplatin combination, with palliative intent or as “bridging therapy” (Table 2).

8. Stopping immunotherapy in metastatic melanoma treatment

ESMO recommendation [76]:

• Stopping anti-PD-1 therapy for patients with CR that persist on radiological evaluation and who received treatment for at least 6 months should be considered.

• Stopping anti-PD-1 therapy for patients with PR and SD after 2 years of treatment should be considered.

• Stopping targeted therapy outside clinical trials is not recommended.

Sixty-seven patients from KEYNOTE-001 trial stopped the pembrolizumab therapy after complete response (CR) was confirmed by radiological evaluation and after completing minimum 6 months of treatment [43]. The 2-years DFS from the time of CR was ~90% [43]. In KEYNOTE-006 the patients stopped the treatment after 2 years and 85.4% did not suffer a relapse after 5 years of follow-up [77].

Both CheckMate 067 and KEYNOTE-006 trials revealed a good Hazard Ratio (HR) for progressive disease (PD) after 2 years with anti-PD-1 monotherapy (nivolumab, respectively pembrolizumab) for responders (partial response – PR and stable disease – SD) [43, 48].

9. Treatment for patients with disease progression after adjuvant therapy

ESMO recommendation [76]:

• The patients with primary resistance should be treated with another option.

• The patients with acquired resistance can be treated with the same treatment option or with another agent.

• The decision should be taken in accordance with BRAF status

Primary resistance (disease progression during the 12-months adjuvant therapy or < 6 months from the treatment ending): it is unlikely to have a clinical benefit from using the same agent [76, 78].

Acquired resistance (disease progression >6 months from the treatment ending): it is possible to use the same agent or an alternative agent from the same class [76, 79].

A multicenter randomized clinical study with 300 metastatic melanoma patients (56% BRAF wt and 44% BRAF mutated) evaluated the treatment for the patients who stopped responding after the initial response (acquired resistance) [78]. The most commonly used agent after the first progression was anti-PD-1 (51% of patients from the cohort study), followed by targeted therapy (19%), dual immune combination CTLA-4 and PD-1 (12%), investigational drugs (11%), and ipilimumab monotherapy (6%). The ORR was 46% for anti-PD-1 monotherapy, 67% for TT, 56% for immune combination therapy, 20% for the investigational agent, and 0% for CTLA-4 monotherapy, but no difference in OS after about 2 years of follow up was observed [78]. Another clinical trial demonstrated a higher response rate for the patients treated with the immune dual combination, compared with ipilimumab monotherapy for patients with progressive metastatic melanoma after first-line anti-PD-1 therapy (31% vs. 13%) [66].

A multicenter randomized clinical trial, that focused on the treatment of recurrence after adjuvant therapy, demonstrated a very good response to immune checkpoint inhibitors, similar to the response rate for the patients treated by first-line immune therapy; the three-year OS was 79% for anti-PD-1 based therapy (monotherapy or dual immune combination), 55% for targeted therapy rechallenge, and 25% for ipilimumab monotherapy [80].

10. Immune adverse events iAEs

The current treatments for melanoma produce high-grade toxicity rates, with 55–59% for ipilimumab and nivolumab combination, 20% for nivolumab alone, and 27% for ipilimumab alone [78]. The most common AEs associated with immune checkpoint inhibitors are autoimmune (iAEs). The most frequent immune toxicities to ICI, across all options (anti-PD-1 monotherapy, anti-CTLA-4 monotherapy, or dual immune combination), were cutaneous (pruritus, maculopapular rash, and vitiligo), gastrointestinal (diarrhea/colitis) and fatigue [39]. The most common high-grade, potentially life-threatening iAEs were endocrinopathies (hypophysitis, adrenal insufficiency, and hypo- or hyperthyroidism), pancreatitis, and hepatic AEs (elevated ALT, AST, hepatitis) [39]. Other potentially lethal iAEs were nephritis, pneumonitis, and myocarditis.

A retrospective study from the WHO pharmacovigilance database identified 613 fatal ICI toxic events, reported from 2009 to 2018. The most death-related AEs were pneumonitis, hepatitis, and neurotoxic effects, for the dual immune combination and colitis for anti-CTLA-4 treatment [81]. A meta-analysis of 112 trials showed higher toxicity-related fatality rates for CTLA-4 and PD-1 combination (1.23%) and for anti-CTLA-4 monotherapy (1.08%), compared with single-agent anti-PD-1 (0.36%) [81].

The treatment with ICI requires a routine monitoring for immune toxicities, with physical examination, anamnesis for autoimmune or infectious diseases (screening for HIV, hepatitis A, B, and C), complete blood count, comprehensive metabolic panel, cardiac evaluation with ECG and measurements of oxygen saturation, and endocrine evaluation (TSH, FT4 and serum cortisone), at baseline and periodically, for the entire treatment duration [82]. The NCCN elaborated a comprehensive guideline for the management of immunotherapy-related toxicities [82].

The kinetics of iAEs is different for different types of immune-related toxicities. The first toxicities that become evident are skin-related AEs (median time to onset 3 weeks), but the risk persists throughout treatment. Later, gastrointestinal (median time to onset 7 weeks) and hepatic toxicities appear, and finally pulmonary, endocrine, and renal AEs may develop [83, 84]. The patients who experienced AEs of any grade had a significantly higher objective response rate [83]. Most treatment-related AEs resolve completely after specific treatment, with the exception of endocrinopathies, which require long-term hormone replacement therapy [84]. Median time to resolution for grade 3–4 iAEs was under 5 weeks, apart from endocrinopathies excepted [84].

11. Immune checkpoint inhibitors and their impact on intestinal flora

ESMO recommendation [76]:

• Restrictive use of empirical antibiotics in melanoma patients treated by immune checkpoint inhibitors

Specific species of gut microbiome or microbiota can influence antitumoral responses, either through innate or adaptive immune pathways. In severely immunocompromised patients, the modification of intestinal flora through diet or fecal microbiota transplants could improve the response to ICI [85].

On the other hand, the excessive use of antibiotics decreases the diversity of gut microbiome and eliminates the most immunogenic bacteria, having thus a negative impact on patients treated with ICI [86].

12. New directions for immune therapy

12.1 Anti-LAG-3 and Anti-PD-1 immune combination

Lymphocyte-activation gene 3 (LAG-3) is a cell-surface receptor on activated CD4+ T cells and represents an alternate immune checkpoint [87]. The anti-LAG-3 agent relatlimab and anti-PD-1 agent nivolumab were combined in phase II/III RELATIVITY-047 clinical study [88]. The study compared the dual immune combination relatlimab and nivolumab with nivolumab monotherapy, favoring the immune combination in terms of median PFS (10.1 months for combination vs. 4.6 months for monotherapy, with HR for progression or death of 0.75). The obtained results were better for a subgroup of patients with positive LAG-3 expression (≥ 1%). The rate of AEs was 18.9% for the combination and 9.7% for the monotherapy group [88].

In the CheckMate 067 clinical trial, PFS for the nivolumab and ipilimumab combination, the current first-line indication for stage III and IV inoperable melanoma, was 11.5 months, with a 59% rate of AEs [48]. If the first results of the RELATIVITY-047 clinical study will be supported also by better overall survival rates, it would give good grounds for the expectation that the relatlimab and nivolumab combination will replace the ipilimumab and nivolumab combination in the first-line treatment of metastatic melanoma [89].

12.2 Other Novel ICIs

Anti-VISTA small molecule ICI (CA 170) combined with nivolumab and anti-Tim-3 antibody combined with spartalizumab (anti-PD-1) are among the most promising immune combinations for the treatment of advanced melanoma [90, 91]. V-domain immunoglobulin suppressor of T-cell activation (VISTA) is a negative regulator of T-cell function; anti-VISTA agents show synergistic effects with anti-PD-1 agents [90]. T-cell immunoglobulin and mucin domain 3 (Tim-3) is a cell surface molecule expressed on lymphocytes, dendritic cells, and tumor cells (including melanoma cells) that breaks off T-cell activation and diminishes antitumor immunity [91]. Anti-Tim-3 monoclonal antibody stops T-cell inhibition and amplifies tumor cell disintegration.

Other potential new immune combinations are the associations between anti-PD-1 inhibitors with agonists of IL-2 described in the PIVOT-02 phase II clinical trial [92], or between anti-PD-1 agents and different oncolytic viruses (e.g., polio, coxsackie, herpes simplex or poxvirus) [93].

12.3 Adoptive cellular therapy (ACT)

ACT with the use of tumor-infiltrating lymphocytes (TILs) can be a future option for solid tumors, including metastatic melanoma [69, 94]. The clinical development of TILs started more than 40 years ago, but it was not approved by FDA for melanoma treatment, despite the good response rates [93]. The combination of TILs with chemotherapy and IL-2 was associated with a 24% CR and with 55% ORR among patients with disease recurrence after previous systemic treatment [94]. Because of high rates of potentially lethal AEs, this therapy can be safely administered only in a high-facility oncological center, trained for IL-2 administration.

One of the pivotal multicenter clinical trials was designed to evaluate TILs administration (lifileucel, an autologous, centrally manufactured TILs) in conjunction with IL-2, followed by sequential ICI, in patients with solid tumors, including melanoma (NCT 02360579) [95]. The ORR was 36%, with 2 from 66 patients with CR and 22 from 66 patients with PR. Median duration of response was not reached after 18.7 months of median follow-up. This treatment could be used as salvage therapy for metastatic melanoma patients, refractory to anti-PD-1 and targeted therapy [95].

Systemic treatment for metastatic melanoma improved dramatically in the last 10 years with enhanced long-term survival of these patients. Immune therapy is part of the change of the treatment paradigm in melanoma; shifting from direct cytotoxic tumor destruction to increasing the immune system activity in order to destroy the cancer cells. Undoubtedly, the near future will be the time of different dual immune combinations, with or without new targeted therapy approaches.