Tumor‐infiltrating lymphocyte therapy and the mucosal melanoma treatment landscape

The systemic treatment of advanced mucosal melanoma (MM) is largely extrapolated from approaches within landmark melanoma clinical trials in which non-cutaneous subtypes are poorly represented. Relevant data are often derived from underpowered subgroup analyses that fail to capture the genetic, microenvironmental and immunological differences of this distinct entity, which, alongside later presentation, likely drive its poorer outcomes relative to cutaneous disease. Arising from melanocytes within sun-sheltered (and typically clinically occult) mucosal epithelia, MM genomes are characterized by structural rearrangements and usually lack the ultraviolet light mutational signature associated with high neoantigen burden and sensitivity to immune checkpoint inhibition (ICI) [1]. In a pooled analysis of patients with advanced MM treated in trials of nivolumab with or without ipilimumab, overall response rates (ORR) were 37.1% and 23.3% respectively, compared to 60.4% and 40.9% in cutaneous disease [2]. Canonical BRAF V600 point mutations are rare in MM, and whilst other potentially druggable molecular drivers (most frequently cKIT) have been identified, small trials of targeted agents show limited and short-lived efficacy [3, 4]. Our institution's practice is to offer first-line combination ICI, reserving KIT inhibitors for cKIT-mutated disease in the second line, and best supportive care or cytotoxic chemotherapy thereafter. Given current limitations, clinical trials are preferred at any line where available, and novel data linking meaningful responses with translational insights into MM biology are encouraged. Lifileucel, a tumor-infiltrating lymphocyte (TIL) therapy (Figure 1), has received accelerated US Food and Drug Administration approval for advanced melanoma resistant or refractory to anti-programmed cell death protein 1 (PD-1) therapy, based on the single-arm, phase II C-144-01 trial [5, 6]. Kluger et al. [7] recently describe outcomes for the 12 patients with MM within the trial who received lifileucel (amongst 153 treated patients in total), and reported an ORR of 50% with one complete response. This study described a small and highly selected group of patients who not only remained fit for later-line therapy after progression on ICI, but were well enough to undergo surgery followed by intensive non-myeloablative lymphodepleting chemotherapy, with a disease tempo compatible with the required TIL manufacturing process. This selection likely enriched for a subpopulation with less aggressive disease biology than typical for MM. Nevertheless, disease burden at baseline was high, with a median sum of diameters of target lesions being 118.9 mm in the MM subpopulation (101.1 mm across all 153 patients). Despite this, clinically meaningful and durable responses were observed. Median duration of response was not reached at median follow-up of 35.7 months, with a 2-year overall survival rate of 45.5% amongst treated patients. Numerically, responses exceeded the 31.4% ORR in the overall anti-PD-1-refractory melanoma population. It is biologically plausible that TIL therapy could be relatively more effective than ICI in MM compared with cutaneous disease. The largest translational dataset to date exploring the immune microenvironment and drivers of ICI resistance in MM confirmed lower immune cell infiltration at baseline compared with cutaneous disease [8]. Patients whose MM microenvironment bore a “more-infiltrated” transcriptional signature responded more frequently to ICI (albeit not reaching statistical significance), with the inference that pre-existing anti-tumor immunity predisposes to ICI response as in cutaneous disease. Those who did not respond despite a “more-infiltrated” microenvironment were more likely than responders to have a high signature of immunosuppressive cancer-associated fibroblasts and tumor-associated macrophages. The rationale for TIL therapy is that ex vivo expansion and in vivo interleukin-2-driven activation of otherwise scarce tumor-reactive lymphocytes may counterbalance immunosuppressive local features (including cytokines, cell populations and metabolites) to produce an effective anti-tumor immune response [5]. Preconditioning chemotherapy may also play a role in depleting negative regulatory immune cells. The benefit of this approach compared to ICI (alone) may be greater where the neoantigen burden is lower and microenvironment more immunosuppressive at baseline. These hypothesis-generating, preliminary data provide some support for trials bringing TILs into the first line in MM, where the likelihood of benefit from ICI is lower. The ongoing randomized phase III “TILVANCE” trial (NCT05727904) of lifileucel with pembrolizumab (versus pembrolizumab alone) for previously untreated advanced melanoma includes those with MM, and outcomes in this subgroup will be of particular interest. Other strategies to recruit/retain and activate immune effector cells within the tumor microenvironment are also appealing in MM, but efficacy data remain limited by its low incidence and under-representation in clinical trials The cancer testis antigen PRAME is frequently expressed in MM [9], but trials of a PRAME-directed bispecific T cell engager that have shown promising efficacy alone and in combination with anti-PD-1 ICI in cutaneous melanoma have excluded MM (NCT06112314, NCT04262466). One of two patients with MM treated within a phase I study of an autologous PRAME-directed T-cell receptor T cell therapy experienced a deep and durable response, meriting further exploration [10]. Outside immunotherapy, rational approaches with promise include targeting pathways implicated in tumorigenesis by recurrent structural rearrangements. In addition to activating cKIT mutations (which may occur outside the canonical exon 11 L576 locus [4]), whole-genome sequencing of 67 MM identified structural variants affecting other components of the mitogen-activated protein kinase and WNT signalling pathways as well as regulation of telomere length [1]. Notably, 70% of the cohort had mutations linked to response to cyclin-dependent kinase (CDK)4/6 inhibition. Dalpiciclib suppressed the growth of patient-derived xenograft models of CDK4-amplified head and neck MM in preclinical studies but was associated with a limited ORR of 6% in a small (n = 17), non-randomized, open-label, phase II trial in this population [11]. Nonetheless, the reported disease control rate of 81%, wide availability, and established tolerability make further evaluation of CDK4/6 inhibitors in this space is reasonable. Prospective clinical data are evidently required, but this will always remain a challenge in a rare, clinically and biologically heterogeneous cancer such as MM, particularly when appetite to include this group of patients (with historically poor prognosis and low response rates) in industry-sponsored studies may be low. An ongoing collective international translational effort is urged to better understand the biology and vulnerabilities of this entity and thus prioritise and direct avenues of clinical research. Dr Jennifer Thomas and Dr Max Julve wrote and edited the manuscript. Dr Andrew Furness and Professor James Larkin provided expert opinion and review. N/A J.T. and M.J. declare no conflicts of interest. J.L. declares the following: honorariums (Eisai, Novartis, Incyte, Merck, touchIME, touchEXPERTS, Pfizer, Royal College of Physicians, Cambridge Healthcare Research, Royal College of General Practitioners, VJOncology, Agence Unik, BMS, Immatics, Insighter, and GCO); consultancies (iOnctura, Apple Tree, Merck, BMS, Eisai, Debipharm, Incyte, Pfizer, Novartis, MSD. Speaker fee: Pierre Fabre, BMS, Ipsen, Roche, EUSA Pharma, Novartis, Aptitude, AstraZeneca, GSK, Eisai, Calithera, Ultimovacs, Seagen, Merck, eCancer, Inselgruppe, Pfizer, Goldman Sachs, MSD, Regional British Society of Gastroenterology, and Agence Unik); institutional research support (BMS, MSD, Novartis, Pfizer, Achilles Therapeutics, Roche, Nektar Therapeutics, Covance, Immunocore, Pharmacyclics, and Aveo); and grants (Achilles, BMS, MSD, Nektar, Novartis, Pfizer, Roche, Immunocore, Aveo, and Pharmacyclics). A.J.S.F. declares the following: speakers bureau and educational events (BMS, Eisai, Ipsen, Merck, and Pfizer); support for attending meetings and/or travel (Achilles Therapeutics, Adaptimmune, ESMO, Iovance Biotherapeutics. Advisory Boards: Achilles Therapeutics, GSK, Immunocore, and Neogene; and patents planned, issued, or pending: WO2019008375A1 (co-inventor of methods of identifying responders to cancer treatment). N/A N/A N/A