Ruling out Melanoma: A Practical Guide to Improving Performance Through Non-Invasive Gene Expression Testing (Family Practice)

Jason K. Rivers, MD, FRCPC, FAAD^1,2; Darrell S. Rigel, MD³

¹Department of Dermatology and Skin Science, University of British Columbia, Vancouver, BC, Canada
²Pacific Derm, Vancouver, BC, Canada
³Department of Dermatology, New York University Medical Center, New York, NY, USA

Conflict of interest:
Jason Rivers and Darrell Rigel are consultants for DermTech.

Introduction

The task of correctly assessing pigmented skin lesions to rule out melanoma via the current standard of care (visual assessment plus histopathology) remains a challenge even for experienced melanocytic lesion experts because of the inherent limitations of image recognition. Tools such as dermoscopy or computer-aided image analysis of skin lesions can reduce but not overcome some of these inherent limitations. However, therapeutic challenges continue even after a decision has been made to biopsy a pigmented lesion suspicious for melanoma because pathology also focuses on pattern recognition, and histologic criteria to distinguish between benign and malignant skin lesions may overlap in borderline cases. Although immunohistochemistry can offer additional information, a growing number of investigators have demonstrated that molecular analysis techniques such as fluorescence in situ hybridization, comparative genomic hybridization and messenger RNA expression profiling of surgically obtained specimens can help to improve the prediction about the behavior of melanocytic neoplasms, including melanoma. However, even these current molecular techniques fall short because the tests depend on tissue samples from surgical biopsies. A simple yet accurate non-invasive tool (similar to Cologuard for colon cancer) to guide biopsy decisions and rule out melanoma would be attractive and desirable to health care providers and patients alike. In this article, we summarize a large body of literature on such a tool (a non-invasive adhesive patch based gene expression test for LINC and PRAME) and provide a guide to primary care physicians on how to use this tool to improve pigmented lesion management by biopsying less while at the same time reducing the chance of missing melanomas.

Background¹

• Melanoma is potentially the deadliest form of skin cancer and develops from melanocytes.
• In the past decade, the number of new melanoma cases diagnosed annually has increased by 53%.
• An estimated 178,560 cases of melanoma will be diagnosed in the US in 2018; Canadian numbers will be about one-tenth of this number.
• An estimated 9,320 people will die of melanoma in the US in 2018.
• Compared with stage I invasive melanoma patients treated within 30 days of being biopsied, those treated 30-59 days after biopsy have a 5% higher chance of dying from the disease and those treated more than 119 days after biopsy have a 41% higher risk.
• The estimated five-year survival rate for patients whose melanoma is detected early is about 99% in the US and Canada. The survival rate falls to 63% when the disease reaches the lymph nodes and 20% when the disease metastasizes to distant organs.
• Only 20-30% of melanomas are found in existing moles, while 70-80% arise on apparently normal skin.

The Current Care Standard of Diagnosing Melanoma

• Management of atypical pigmented lesions involves ruling out melanoma via visual and/or dermoscopic assessment followed by surgical biopsy and histopathologic examination (Figure 1A).²
• Ideally, when melanomas are identified, they are found at the earliest stages (melanoma in situ (MIS)/Stage 1a) when a high cure rate is possible by wide excision.^3,4
• While the purpose of the visual assessment/surgical biopsy paradigm is to rule out melanoma, this approach has relatively poor performance metrics with an estimated 3-10% specificity for visual examination alone.⁵
• This, coupled with the low sensitivity of 65-85% for histologic assessment^6,7 and the estimated in-office prevalence of around 5-10% for melanoma, leads to a low negative predictive value (NPV) for early stage disease (83%, Figure 1).
• Thus, during histopathologic assessment, only a small number of melanomas are identified from a large pool of biopsied pigmented lesions.
• Perhaps even more concerning is the risk of false negative histologic diagnoses resulting from a significant overlap in the histopathologic criteria between atypical nevi and early stage melanoma.^6-8
• Elmore et al. concluded the diagnosis of early stage melanoma was not accurate after finding 187 pathologists misinterpreted 35% of slide interpretations for MIS/Stage 1a melanomas.⁶
• Given the prevalence of early stage melanoma in biopsied lesions is approximately 5-10%, the NPV (i.e. there is no melanoma when the pathologist states there is no melanoma) can be approximated as between 75-89%.^5-7,9,10
• Under the current approach to pigmented lesions, the number of surgical biopsies needed to identify one melanoma (NNB, number needed to biopsy) averages around 20-25 and ranges from 8 to >30 depending on the clinical setting.^5,9,11-13
• Further complicating the issue is that the histopathologic assessment of routinely biopsied lesions, without serial sectioning, is limited. With routine step sectioning of the tissue block providing less than 2% of the material for evaluation, there remains uncertainty as to what is present in the rest of the specimen.
• With the current diagnostic approach, it is estimated that, in the United States, 3 million pigmented lesion surgical biopsies were performed in the year 2017 alone, yielding <200,000 melanoma diagnoses.^14,15 Canadian numbers are about a tenth of these numbers.

Pigmented Lesion Assay (PLA) Overview

• The PLA is a gene expression test that helps clinicians rule out melanoma and avoid the need for a surgical biopsy of concerning pigmented lesions (Figure 1B).^16-22
• The PLA is based on a new technology for non-invasive skin testing that permits gene expression analysis of skin samples collected with adhesive patches.19 In order to retrieve enough genetic material, the lesion is sampled consecutively four times, each time with a different patch. For each patch, the clinical margin of the lesion is delineated in pen and then the outlined tissue is dissected from the surrounding sample at the processing lab.
• The recovered RNA is extracted and analyzed for two indicator genes. The indicator genes used are PRAME (Preferentially Expressed Antigen in Melanoma) and LINC00518 (Long Intergenic Non-Coding RNA 518), both of which are overexpressed in melanoma. These genes were categorized as the key factors driving test performance in a microarray-based gene expression screen that identified a group of 312 genes differentially expressed in melanoma versus non-melanoma pigmented lesion samples.^16,21
• The recovered RNA is extracted and analyzed for two indicator genes. The indicator genes used are PRAME (Preferentially Expressed Antigen in Melanoma) and LINC00518 (Long Intergenic Non-Coding RNA 518), both of which are overexpressed in melanoma. These genes were categorized as the key factors driving test performance in a microarray-based gene expression screen that identified a group of 312 genes differentially expressed in melanoma versus non-melanoma pigmented lesion samples.^16,21
• Sampling of the most superficial skin layers contains information from deeper epidermal cells as a result of normal skin physiology in which basal cells migrate to the surface of the skin as they differentiate into squamous cells.
• In contrast to histopathologic sectioning, this method of genetic tissue sampling allows for the collection of material from the entire lesion.
• The PLA is intended for use in patients 18 years of age or older with pigmented lesions measuring 5 mm or larger and suspicious for melanoma.
• It is not intended for use on non-melanocytic lesions (e.g., seborrheic keratoses), non-melanoma skin cancers (e.g., basal cell carcinomas) and bleeding or ulcerated lesions. Further, at present, the PLA cannot be used on palms, soles, nails, or mucous membranes.
• Of importance, the PLA is intended to aid clinicians in surgical biopsy decisions but not to be used as a diagnostic test for melanoma.
• Positive PLA tests should be followed with a surgical biopsy, while patients with a negative test can have the lesion monitored per standard of care.
• The time for clinicians to receive the molecular pathology test reports is generally 48-72 hours.
• The test is available in all 50 US states and increasingly also in Canada following Health Canada approval of the platform.
• Adhesive patches to collect PLA samples are made available free of charge.
• The PLA significantly reduces cost (please see Table 1) based on a recently published cost effectiveness analysis.²²

PLA Versus Current Standard of Care

• Table 1 compares the key performance metrics of the PLA against the current standard of care (visual assessment and surgical biopsy/histopathology) for pigmented lesion management.
• In contrast to the current standard of care, the PLA has a very high NPV (>99%) coupled with a high sensitivity (91-95%), ensuring a very low probability of missing a melanoma.^16,17
• The relatively high specificity of the PLA (69-91%) also helps to effectively reduce the number of lesions that would require subsequent histopathology review.16 Consequently, using the PLA, the number of lesions needed to be biopsied to find one melanoma is reduced from 20-25 to 2.7 (Table 1).^5,9,11-13,17
• By utilizing the PLA, unnecessary surgical procedures may be reduced by as much as 88%.¹⁸
• The findings of this internal data set is consistent with a recently published review of 18,715 biopsied pigmented lesions where 83% of those lesions were either benign or mildly atypical nevi and did not require additional treatment.¹⁰
• Thus, using the PLA, about 90% of surgical biopsies performed on pigmented lesions in the general community may be avoided.

Conclusion

At present, the relatively low specificity of the clinical examination has resulted in a large number of biopsies of pigmented lesions being taken to ensure the detection of melanoma. The addition of PLA to this diagnostic pathway can lead to fewer surgical procedures and would provide significant benefits to patients such as reduced pain, infections and scarring. In addition, significant benefits accrue to the healthcare system because the PLA can reduce the costs associated with unnecessary surgical procedures.²² Most important, however, is the lower probability of missing a melanoma compared with the current standard of care. The PLA provides a unique and disruptive technology for the assessment of pigmented lesions, and it may soon transform our current approach to one that is less often invasive, highly reproducible, and a cost saving to the health care system.^16-22