In December, we launched our first ancillary diagnostic test, MelaPro Dx™, that is intended to help dermatopathologists differentiate ambiguous melanocytic lesions. MelaPro Dx is a molecular profiling test that evaluates the proteomic contents of melanocytic cells to distinguish between benign and malignant lesions. In this post, I would like to give you a brief overview of the different ways that molecular profiling tests are employed in the diagnosis and treatment of melanocytic lesions.
Melanocytic lesions can be difficult to diagnose based on histopathology alone, as they can exhibit a high degree of ambiguity in cellular features that leads to discordance in diagnosis, even among experienced dermatopathologists. To help with these challenging cases, clinicians rely on a variety of molecular tests, ranging from those that target only one or a few molecules to those that provide a more global analysis.
An initial approach in melanoma diagnosis may be to use immunohistochemical staining for targeted antigens in the biopsies. These targets may be general cancer markers, such as ki67 or cytokeratins, or more specific makers for melanoma, such as a decrease in BRCA-associated protein 1 (BAP-1) or an increase in histone 3 methylation, both of which have been correlated with malignancy.
Comparative genomic hybridization (CGH) has been used, although mostly in academic settings, to evaluate variations in chromosomal copy numbers relative to ploidy level, where gains and losses, particularly in chromosomes 1, 6, 8, 9, and 11, have been linked to a melanoma diagnosis. This analysis is achieved through direct comparison of a questionable sample to a reference, or normal control, sample. Fluorescence in situ hybridization (FISH) takes a more targeted approach of evaluating copy number variations at specific genomic foci that have been implicated in malignant melanoma. FISH is offered through several academic institutions as well as commercially on a limited basis.
More recently, gene expression profiling and protein profiling tests have entered the market. The myPath® Melanoma test uses a 23-gene expression signature to differentiate benign from malignant melanocytic lesions. The test targets 14 genes that have been implicated in malignant melanoma as well as 9 housekeeping genes. In contrast, our MelaPro Dx test differentiates benign nevi from malignant melanomas by applying a machine learning algorithm to a proteomic fingerprint consisting of 1,075 peptide peaks arising from in situ enzymatic digestions of proteins in the specimen. Proteins, and their posttranslational modifications, change their expression levels much more rapidly than DNA or RNA, allowing for a better evaluation of the active state of the targeted cells.
Ancillary diagnostic tests have also found their way into the dermatologist’s office via the Pigmented Lesion Assay. This test allows a dermatologist to apply an adhesive patch to a suspicious lesion to extract biomolecules from the skin surface. The extracted RNA is then evaluated for expression levels of LINC and PRAME that can help a dermatologist decide whether or not to biopsy the lesion for further evaluation.
Once a definitive diagnosis of melanoma is rendered, the next question is how best to manage it. Early stage, thin melanomas, generally have a good prognosis. However, some of these lesions behave more aggressively and ultimately metastasize. The DecisionDx-Melanoma gene expression profiling test uses a 28-gene signature to stratify lesions into low and high risk of metastasis. This information can be used to determine an appropriate surveillance or treatment plan for the patient.
Determining an appropriate treatment strategy can be challenging, and a variety of approaches are being used to more quickly identify a drug or immunotherapy that is most likely to be effective against a patient’s cancer. This approach to cancer treatment is collectively known as precision medicine.
One technique that has shown promise is through the use of patient-derived xenografts (PDX). A small piece of the patient’s tumor is implanted into the flank of a mouse or rat, and once it is established in the animal, different drug treatments can be evaluated to determine to which one(s) the tumor shows a response. By determining the best course of treatment in a PDX model, unnecessary cost and side effects to the patient can be avoided.
Molecular testing is also being used to determine specific mutations in melanocytic tumors that can be indicative of response to a particular type of treatment. For example, BRAF mutations may indicate that the tumor will respond to BRAF or MEK inhibitor drugs, including vemurafenib, dabrafenib, trametinib, and cobimetinib; while mutations to KIT may indicate that imatinib mesylate may be an effective treatment. Expression of programmed death-ligand 1 (PD-L1) by melanoma cells has been indicative that treatment with anti-PD-1 therapy may be effective; however, the predictive value of PD-L1 expression has not been as high for melanoma as for other cancers.
Recently, a study was carried out that examined a 209-protein signature from serum of patients with metastatic melanoma that better correlated with progression-free survival and overall survival for patients treated with anti-PD-1 therapy than did melanoma tumor expression of PD-L1. Clinical validation studies showed that the signature can differentiate patients whose cancer is likely to be sensitive to anti-PD-1 therapy from those who are likely to be resistant to treatment.
This post is intended to serve as an introduction to the types of molecular tests that are currently applied to melanoma diagnosis and treatment selection. For more in-depth information, the reader is directed to the following references:
Lee JJ and Lian CG. “Molecular Testing for Cutaneous Melanoma An Update and Review.” Arch Pathol Lab Med. doi: 10.5858/arpa.2018-0038-RA
Helgadottir H, et al. “Personalized Medicine in Malignant Melanoma: Towards Patient Tailored Treatment.” Front Oncol. 2019, 8, 1-6.