Can Combination MEK and Akt Inhibition Slay the Giant Congenital Nevus?

Abstract

The clinical management of large and giant congenital melanocytic nevi (lgCMN) relies heavily upon iterative surgical procedures. In this issue Rouille et al. (2019) use lgCMN explants and a newly developed patient-derived xenograft model to show that the local administration of MEK and Akt inhibitors limits the lgCMN proliferative potential. These findings, along with emerging reports, support continued investigation of targeted therapies in lgCMN.

COMMENTARY

Congenital melanocytic nevi (CMN) are benign, pigmented tumors resulting from the hyperproliferation of cutaneous melanocytes in utero or shortly after birth. CMN are classified into four major groups based upon the largest predicted diameter of the nevus at adulthood: small (<1.5 cm), medium(1.5–20 cm), large (20–40 cm), and giant (>40 cm) (Krengel et al., 2013). Large and giant CMN (lgCMN) have the lowest incidence rates, collectively affecting approximately 1 in 20,000 individuals (Kinsler et al., 2017a). Nevertheless, these CMN are also the most clinically concerning and difficult to manage. Individuals with lgCMN are at increased risk for melanoma development, and a handful of these patients will be diagnosed with life-threatening neurocutaneous melanocytosis (Kinsler et al., 2017a). Other common complications of lgCMN include pruritus, pain, and psychosocial trauma due to cosmetic disfigurement (Krengel et al., 2013).

Surgical resection remains the mainstay of lgCMN management and aims to mitigate melanoma risk, reduce disease complications, and improve cosmetic outcomes (Bauer and Corcoran, 2005). Iterative surgical approaches employing tissue expanders and skin grafts are frequently required to appropriately manage lgCMN lesions (Bauer and Corcoran, 2005). However, even with these techniques, the complete removal of lgCMN can be precluded by lesion size, anatomical location, or deep tissue invasion. Beyond the surgical challenges associated with lgCMN management are the difficulties surrounding melanoma diagnosis in these patients. Proliferative nodules arising within lgCMN often resemble melanoma pathologically, with >80%S of lgCMN harboring the same oncogenic NRAS mutations observed in cutaneous melanoma (Kinsler et al., 2017a). Due to this similarity, a differential diagnosis cannot be based upon the presence of an NRAS mutation. Thus, the management of symptoms and cancer risk is a persistent challenge for these patients.

A new preclinical model of lgCMN

Preclinical models play fundamental roles in the development of new therapeutic strategies. To this end, several reported genetically engineered mouse models exhibit phenotypes analogous to human CMN. Although the majority of these genetically engineered mouse models drive the formation of large nevocytic lesions, only two replicate both the genetic driver and central nervous system involvement observed in advanced lgCMN patients. In the first model, the melanocytic expression of NRas^12D leads to leptomeningeal hyperpigmentation and subsequent melanoma formation (Pedersen et al., 2013). In the second model, the expression of NRas^61R, in combination with activated WNT signaling (Apc loss), causes the mice to develop leptomeningeal melanosis (Pawlikowski et al., 2015). Still, whether the cutaneous and central nervous system lesions that develop in these models fully recapitulate the genetic and pathological heterogeneity of human CMN remains unclear. Furthermore, the structural differences between murine and human skin could affect drug availability and efficacy in preclinical trials.

Rouille et al. (2019) describe a new patient-derived CMN xenograft model in which surgically resected human lgCMN tissue is implanted on the dorsal side of Rag2^−/− mice. Once engrafted, these explants retain the classic structure and signaling profiles of human CMN. Using this model, the authors were able to test injectable therapies under clinically relevant conditions. Although the use of Rag2^−/− mice, which lack mature B- and T-cells, precluded their assessment of immunotherapeutic regimens, the authors were able to provide proof- of-principal for a new, targeted inhibitor-based approach to lgCMN management.

Identification of MEK and Akt inhibition as a potential lgCMN management strategy

The increased availability of pharmaceutical kinase inhibitors has provided new opportunities for small molecule interventions in lgCMN. Due to the high frequencies of oncogenic NRAS mutations in lgCMN, Rouille et al. (2019) hypothesized that inhibitors of NRAS signaling would aid in the management of lgCMN. Using protein kinase arrays, they discovered that MAPK and Akt signaling was elevated in CMN-derived nevocytes when compared with normal human melanocytes. This initial observation led the authors to treat NRAS-mutant nevocytes with MEK (binimetinib) and Akt (GSK690693) inhibitors. Treatment with these drugs, either alone or in combination, reduced the viability and clonogenic potential of cultured CMN nevocytes. This prompted further explant and xenograft studies, which revealed a promising reduction in nevocyte proliferation and abundance in CMN biopsies treated with binimetinib and GSK690693.

The promise of targeted therapy in lgCMN

Toxicity due to the blockade of proliferation in healthy tissues has been a barrier to the clinical implementation of MEK and Akt inhibitors. However, the use of an injectable drug delivery system by Rouille et al. (2019) showed that local therapeutic administration might be both feasible and efficacious in the setting of lgCMN. Based on these data, they proposed a new lgCMN management strategy involving the use of injection-based targeted therapies to treat unresectable lesions and reduce lgCMN burden prior to surgery.

In support of the therapeutic strategy put forth by Rouille et al. (2019), Pawlikowski et al. showed that the MEK inhibitor, selumetinib, can limit the extent of cutaneous and central nervous system disease in an NRAS- mutant, Apc-null CMN mouse model (Pawlikowski et al., 2015). Early clinical data also suggest a role for MEK inhibitors in controlling human lgCMN. For example, the MEK inhibitor, trametinib, was shown to decrease the size and thickness of a lgCMN driven by a BRAF gene fusion (Mir et al., 2019). Trametinib has also been prescribed to four lgCMN patients with NRAS-mutant central nervous system melanoma on a compassionate basis (Kinsler et al., 2017b). In these cases, the drug provided temporary symptomatic relief with no apparent side effects. Notably, such responses may not be limited to congenital nevi as the involution of acquired nevi is also reported in melanoma patients treated with a combination of BRAF and MEK inhibitors (McClenahan et al., 2014).

Although these data support the use of MEK inhibitors in the treatment of CMN, the effectiveness of Akt inhibitors alone, or in combination with MEK inhibition, has yet to be determined. Clinical trials using a combination of MEK and Akt inhibitors show that Akt inhibition provides no additional benefit to patients with cutaneous melanoma (Algazi et al., 2018). How- ever, the local administration of MEK and Akt inhibitors may limit drug toxicity, permitting the use of higher drug doses that could improve therapeutic response. It is also possible that lgCMN may be more sensitive to Akt inhibitors than fully transformed, cutaneous melanomas. Thus, the findings of Rouille et al. (2019) provide the impetus to explore the role of a growing resource of Akt inhibitors in lgCMN management.

Next steps in the evolution of lgCMN management

The work of Rouille et al. (2019) bolsters the accumulating evidence that supports the use of targeted inhibitors in lgCMN patients; however, a number of questions remain. For instance, how will the heterogeneity and mosaicism of lgCMN affect patient responses? As recently highlighted by Martins da Silva et al., distinct genetic drivers and secondary gene alterations can evolve within the same CMN lesion (Martins da Silva et al., 2019). Therefore, patient responses are expected to vary unless lgCMN are universally dependent on MEK or Akt signaling for survival.

Another question that arises is how well locally injected inhibitors will distribute throughout lgCMN lesions. If the distribution is limited, the number of injections required to achieve lesional control may negate reductions in drug toxicity that are achieved by the use of a local, rather than a systemic, approach. One potential advantage of systemic therapy, however, would be the ability to manage and reduce the risk of neurocutaneous melanocytosis. Notably, Rouille et al. (2019) were unable to conduct any pharmacodynamic or dose escalation studies in their models, presumably because of the limited availability of lgCMN biopsies and the considerable amount of time required to generate patient-derived xenografts. The potential for acquired therapeutic resistance after repeated dosing was also unexplored. Therefore, additional clinical and preclinical trials will be needed to characterize and optimize local drug delivery.

Not every lgCMN is driven by an NRAS mutation; therefore, the efficacy of other targeted therapies should be explored. Oncogenic BRAF mutations are the second most common genetic drivers of CMN (Martins da Silva et al., 2019). In this study, combination dabrafenib (BRAF inhibitor) and trametinib therapy, as used currently in melanoma, may be a viable option for managing these lesions. As mentioned above, such therapy is already known to cause the involution of acquired, BRAF-mutant nevi in melanoma patients (McClenahan et al., 2014). Therefore, extension of the proposed therapeutic approach of Rouille et al. (2019) to other lgCMN genetic subtypes may not be far down the line. The promising results of Rouille et al. (2019), along with early clinical insights, are signs that targeted therapies may soon make their way into standard of care practices for lgCMN management.

Clinical Implications.

• The generation of patient-derived xenograft models of large and giant congenital melanocytic nevi (lgCMN) enables therapeutic screening in a native tissue environment.

• Local injection of MEK and Akt inhibitors could limit complications from lgCMN by reducing the lesion size and surgical burden.

• The work of Rouille et al. (2019), along with emerging case reports and preclinical data, supports continued investigation of targeted therapies in lgCMN management.