Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
Brain Metastases from Cutaneous Melanoma: Biology and its Implications for More Rational Therapeutic Approaches
Introduction
Cutaneous melanoma is the fifth and seventh most frequently diagnosed cancer in the United States in men and women, respectively, for 2014 ( ). In other countries, such as Australia, cutaneous melanoma has reached epidemic proportions. Although it is rare among men and women less than 30 years old, it is the fourth and second most frequently diagnosed cancer in this age group, respectively. Surgery is curative for early, localized, nonmetastatic cutaneous melanomas with a 5-year relative survival rate of 98%. However, the corresponding 5-year rates for patients with regional and distant metastatic melanoma (MM) are 62% and 16%, respectively. For regional lymph node positive and distant MM, multimodality treatment is frequently used with variable success ( ). Over the past 15 years, tremendous progress has been made in identifying histopathologic prognostic factors for patients with cutaneous melanoma. Most of these factors are included in the latest 2009 revision of the American Joint Committee on Cancer (AJCC) staging system, including important histopathologic features associated with primary melanoma, such as Breslow depth of invasion, ulceration, mitotic rate, and tumor infiltrating immune cells as well as features identified in the regional metastases, such as number and size of metastatic lesions within regional lymph nodes ( ). Nevertheless, the current AJCC staging system frequently fails to predict patients with adverse prognosis, in particular patients with regional or distant MM ( ). To this end, better strategies to quantify and characterize the role of host immune response are expected to further refine existing prognostic factors ( ). Objective assessment of immune response in tumor tissues is an emerging need given the unique role of host immune response and the therapeutic benefit of several immunotherapies in a subset of patients with regional and distant MM ( ). Even less progress has been achieved toward understanding existing or identifying new prognostic factors for patients with established distant metastases, with the exception of the role of metastatic organ site of involvement (lymph nodes and subcutaneous lesions vs lung vs other visceral organ involvement) and serum LDH levels ( ).
In parallel with the advances in identifying prognostic histopathologic features, considerable progress has been made in understanding the biology of cutaneous melanoma. The discovery of hotspot mutations in the V600 codon of BRAF (35–50% of melanomas) and the G12 and Q61 codons of NRAS (10–25%) in the 1990s followed by the development of highly selective kinase inhibitors targeting the MAPK pathway in the late 2000s resulted in proof-of-principle clinical trials that led to the FDA approval of three such inhibitors in MM in 2011 and 2013 ( ). Recent applications of next generation DNA sequencing technologies in melanoma have shown additional genetic aberrations that not only add further layers of heterogeneity to the biology of this disease, but have also important prognostic and treatment implications ( ). Research efforts based on large melanoma tissue cohorts, with multiple molecular data platforms and complex integrative analysis, such as the Cancer Genome Atlas Project (TCGA) in cutaneous melanoma, are expected to unravel more robust associations that could better define subgroups with the highest risk for relapse and more personalized systemic treatments ( ).
Challenges of Melanoma Brain Metastases in 2014
Understand Biology, Predict Future Development, Manage Established Brain Metastases More Effectively and with Less Toxicity.
Overview and the “Old Wisdom”
Frequency of brain metastasis from melanoma (MBM) follows that of brain metastases from lung, breast, gastrointestinal, and cancers of unknown primary ( ). Adjusting for incidence, however, melanoma has the highest propensity for central nervous system (CNS) spread among other metastatic solid tumors (up to 70%) and is associated with a worse prognosis than that of distant MM without brain metastases ( ). In particular, MBM can be the original presentation in 20–30% patients with newly diagnosed distant MM and their incidence may rise to 60% over the next two years in patients enduring distant MM ( ). The grim prognosis of patients with established MBM can be attributed to an approximately 50% incidence of intracranial hemorrhage, the highest among any other tumor type ( ). It is important to emphasize, however, that extracranial disease remains the leading cause of death in patients with distant MM, whereas deaths from active MBM are presumably responsible for up to 40% of all deaths, most frequently as a result of development of new intracranial lesions and other catastrophic events (hemorrhage, seizures) rather than progression of existing intracranial lesions ( ). These clinical observations suggest three fundamental aspects about MM: its inherent high propensity for development of CNS metastases, its high incidence of spontaneous bleeding, and the importance of some organ site-nonspecific factor, presumably from the host, that may play an important role in overall survival (OS).
Treatment of MBM has not significantly changed for decades, with the exception of targeted therapies and immunotherapies directed towards distinct patient subsets that will be discussed later in this chapter ( ). External beam irradiation therapy, either in the form of whole brain irradiation or stereotactic radiosurgery, has been the cornerstone therapy for the majority of patients enduring MBM, whereas craniotomy is used for a much more restricted patient subgroup ( ). With respect to systemic treatments, corticosteroids have been the standard “bridge” therapy for most patients, whereas the clinical benefit from any chemotherapy, including agents that adequately penetrate the blood–brain barrier (BBB), has been minimal ( ). Finally, immunotherapies, including the FDA-approved high dose bolus IL-2, had long been avoided for patients with brain metastases due to the increased, at least theoretical, concern of worsening intracranial edema ( ) for a site that had been, for the most part, considered immune-privileged, at least under physiologic conditions ( ). Lack of clinical studies dedicated for patients with active MBM, limited understanding of the biology of MBM, and even more limited translational research on this tumor site have plagued the field for decades, which may explain the empiricism and paucity of systemic treatment options. Recent and ongoing advances on all three fronts have begun to change the approach to management of MBM, an important priority as patients with distant MM now live longer than ever, and therefore may more frequently endure brain metastases at some point in the natural history of their disease.
How Do We Predict Melanoma Brain Metastases?
As mentioned above, the increased propensity of melanomas to metastasize to the brain has been a long-standing observation. At the moment, our ability to predict patients at risk for the development of MBM has been limited to retrospective analyses of large single-institution melanoma patient cohorts for various clinicopathologic parameters, including the incidence of future development of brain metastases. Results from these investigations have shown inconsistent associations between several histopathologic features of the primary melanoma (nodular histology, Breslow thickness of invasion, ulceration), clinical factors (age, primary melanomas from the head and neck), and future development of MBM. Although these data suggest that features from the primary lesion may regulate the CNS tropism of this disease, they do not substantially improve the prognostic value of the existing AJCC staging system ( ). Therefore, it is reasonable to postulate that factors linked with the biology of the primary or even premalignant conditions, but not regional lymph node MM ( ) may be the culprit. To this end, patients with MM bearing oncogenic mutations in the BRAF and NRAS oncogene were associated with higher frequency of MBM development compared to melanomas that did not bear any hotspot BRAF/NRAS mutations ( ). The higher frequency in development of brain metastases for oncogene-driven melanomas can in part explain the worse OS seen in those patients. It also underlies the powerful role of various oncogenes to foster steps leading up to cancer development and progression, such as migration, invasion, survival during hematogenous dissemination, and establishment of brain metastases ( ).
The molecular mechanisms, however, that account for this fascinating metastatic organ “tropism” remain speculative. Nevertheless, most melanoma investigators believe that at least two unique, incompletely understood, but established properties of melanoma cells may account for the propensity of melanoma cells for CNS metastases. First, melanoma cells originate from transformed melanocytes, the embryologic derivatives of neural crest cells . This is a transient, highly migratory embryonal cell population that gives rise to several different cell types, including melanocytes. There is growing evidence that genes important for neural crest or embryologic development are overexpressed in a subset of melanomas ( ). To this list of neural crest-related genes, it is important to include Rac1, a member of the Ras-related Rho family of GTPases that is important for the maintenance of postmigratory neural crest stem cells. Activating hotspot mutations for Rac1 have been observed in 5–6% of melanomas ( ) and were associated with a worse prognosis ( ). Second, melanomas can transdifferentiate along the neural pathway . A handful of non-neural crest-associated neuronal proteins have been identified in melanoma cells ( ). Apart from these two major properties of melanoma cells, the role of factors other than oncogenic mutations or more specific chemokines in the CNS tropism of melanoma cells is incompletely understood ( ). Future translational research that focuses on prospective tumor imaging analysis of primary melanomas for particular neural crest- or neural-associated molecules may help predict metastatic events to the CNS.
You're Reading a Preview
Become a Clinical Tree membership for Full access and enjoy Unlimited articles
If you are a member. Log in here
