Highlights from the Literature

Gene Expression Profile Testing in Cutaneous Melanoma

By William B. Henghold II, MD, FACMS

In preparing this report, I thought about the sheer number of patients (as well as the individual ones I’ll never forget) with cutaneous melanoma (CM) that I have had the privilege to treat since I started my career in dermatology in 1993. I don’t remember that many during residency, but there were certainly enough to gain clinical competence in diagnosis and standard surgical management, as well as to learn how to talk about prognosis, which often dominated the discussion. “Doctor, will that little black spot kill me?” As time went on the numbers grew, as did my experience, but I still remember the frustration and inadequacy I felt in trying to allay fears or justify concerns, particularly in those patients with Stage I and II disease. I now treat several CM patients a week, most of them with Mohs surgery with the benefit of MART-1 immunohistochemistry. I see every one of them in consultation prior to surgery. Prognosis still dominates the discussion.

The majority of patients in the United States diagnosed with CM each year are initially grouped as Stage I or II according to current American Joint Committee on Cancer (AJCC) guidelines. (1) And the majority of patients who will ultimately die of their disease are these same patients, due to the larger total number first included in these purportedly less risky categories, compared to Stage III and IV. So clearly the “horse got out of the barn” sooner than we expected, as would be initially predicted by tumor thickness, presence or absence of ulceration, mitotic rate, and sentinel lymph node biopsy (SLNB) status.

The overarching question is: How can we do better at predicting those patients who are at higher risk for metastatic disease than would be predicted by current AJCC staging guidelines alone, at the time patients first receive their melanoma diagnosis? Is there additional information we can add to more appropriately stage our patients so that our prospective efforts on their behalf will be more meaningful?

There is a growing body of evidence that definitively supports the routine use of a gene expression profile (GEP) test to help us identify sooner those patients who will develop metastatic disease. Due to its relatively recent appearance on the CM scene, there is nothing in the current AJCC guidelines nor the American Academy of Dermatology Clinical Guidelines for the management of CM that addresses the use of GEP. Although the National Comprehensive Care Network Guidelines makes a brief mention of GEP, it is not recommended outside of a clinical study. (1,2,3) However, the guidelines are unanimous in their recommendation that patient management should be customized to address the individual’s overall risk for disease progression.

The DecisionDx-Melanoma assay (Castle Biosciences, Inc., Friendswood, TX) is a GEP test performed on primary tumor tissue (the formalin-fixed, paraffin-embedded tissue obtained at the time of the patient’s initial biopsy or excisional specimen) using reverse transcriptase-polymerase chain reaction technology to determine the expression of 31 genes (28 prognostic and 3 controls). The results are then compared to a set of patients with known outcomes. Twenty-four of the gene targets are known to be differentially expressed in tumors that are metastatic. The likelihood of metastasis developing within a five-year period is then reported as Class IA/B or 2A/B. Class 1A (lowest risk) and Class 2B (highest risk) have the highest predictive value for metastasis. Class IB (low risk) and 2A (high risk) have reduced, although still valid, predictive confidence for five-year metastatic risk. Overall, patients with Class 1 tumors were found to have a five-year disease free survival (DFS) rate of 97%, while those in Class 2 have a five-year DFS rate of 31%. (4)

SLNB status is currently recognized as the single most powerful prognostic marker for disease-specific survival for patients with CM with a tumor thickness > 1 mm, although the vast majority of patients who eventually develop metastatic disease are SLNB negative. (5) A recent study addressed the utility of GEP to improve the risk stratification of SLNB negative patients to help identify those patients who are indeed at higher risk of metastasis. In addition, on multivariate analysis this same study found GEP class to be independent of SLNB for disease-free survival, distant metastasis-free survival, and overall survival. (6) GEP appears to be superior to SLNB at identifying patients who are at high risk of developing metastasis. From a cost standpoint, GEP runs in the hundreds of dollars while SLNB the tens of thousands. And there is no procedural risk to the patient with GEP. Taking into account the recent findings in the Second Multicenter Selective Lymphadenectomy Trial (MSLT-II), SLNB is looking less attractive all the time.

Another recent study combining AJCC criteria with GEP noted a 21% discordance rate between the two, with GEP able to predict 85% of the deaths in that group as high risk, while by AJCC criteria it was categorized as low risk. (7) This study suggests that GEP is a superior prognostic tool to AJCC staging. In any event, the test itself has value as a stand-alone option as well as utility in combining it with other staging modalities. A number of other studies, both retrospective and prospective in nature, all point to the validity of the GEP test, especially with respect to accurately identifying patients at high risk for metastasis. Interestingly, in the majority of the studies performed to date, the median time for a recurrence or metastatic event is 1.1 years for class 2 patients. This provides valuable information when deciding on how closely to monitor a given patient, either with frequency of clinical follow-up and/or imaging studies, and referral to medical oncology for consideration of adjuvant therapy. (8,9,10,11)

In the nearly 25 years since I’ve been caring for patients with CM, I’m more optimistic and encouraged than ever by the progress that has been made in understanding this disease. Gone (for the most part) are the mutilating surgeries. Sentinel lymph node biopsy is clearly just a prognostic staging tool. And there are now some promising therapies we can offer patients with stage III and IV disease, whereas before there were none. With the advent of advances in molecular biology and genetics, we’re getting much better at offering CM patients a clearer picture of what they can expect when first faced with their diagnosis. Giving our patients all the best and current evidenced-based information we possibly can at the time of their initial diagnosis is critical to their informed decision making.

From my review of the available literature, it seems clear that the GEP test provides valid and timely information, especially for Class IA and Class IIB cases. At a minimum, it provides piece of mind for the former, and an opportunity to face the cold reality of the situation for the latter. What remains to be seen is if the information will affect clinical outcomes by allowing for more aggressive management of those patients at higher risk while they are asymptomatic and their tumor burden is low. In any event, “It is what it is.” Those words were uttered by a patient I’ll never forget.

References

1. Balch CM, Gershenwald JE, Soong SJ, et al. Final version of 2009 AJCC melanoma staging and classification. J Clin Oncol 2009;27:6199–6206.
2. Bichakjian CK, Halpern AC, Johnson TM, et al. Guidelines of care for the management of primary cutaneous melanoma. J Am Acad Dermatol 2011;65(5):1032-1047.
3. Coit DG, Thompson JA, Algazi A, et al. Melanoma, Version 2.2016, NCCN Clinical Practice Guidelines in Oncology. J Natl Compr Canc Netw 2016;14:450-73.
4. Gerami P, Cook RW, Wilkinson J, et al. Development of a prognostic genetic signature to predict the metastatic risk associated with cutaneous melanoma. Clin Cancer Res 2015;21:175–83.
5. Morton, D.L., Thompson, J.F., Cochran, A.J. et al. Sentinel-node biopsy or nodal observation in melanoma. N Engl J Med 2006;355:1307–1317.
6. Gerami P, Cook RW, Russell MC, et al. Gene expression profiling for molecular staging of cutaneous melanoma in patients undergoing sentinel lymph node biopsy. J Am Acad Dermatol 2015;72:780-5.
7. Ferris LK, Farberg AS, Middlebrook B, et al. Identification of high-risk cutaneous melanoma tumors is improved when combining the online American Joint Committee on Cancer Individualized Melanoma Patient Outcome Prediction Tool with a 31-gene expression profile-based classification. J Am Acad Dermatol 2017;76;818-25.
8. Hsueh EC, DeBloom JR, Lee J, et al. Interim analysis of survival in a prospective, multi-center registry cohort of cutaneous melanoma tested with a prognostic 31-gene expression profile test. J Hematol Oncol 2017;10(152):1-8.
9. Fleming MD, Middlebrook B, Covington KR, et al. Performance of a prognostic 31-gene expression profile test in Stage III cutaneous melanoma subjects. Poster presented at ASCO 2017 (abstract #9578).
10. Vetto JT, Leachman S, Middlebrook B, et al. Performance of a 31-gene expression profile (GEP) test for metastatic risk prediction in cutaneous melanomas (CM) of the head & neck. Poster presented at ASCO 2017 (abstract #9576).
11. Greenhaw BN, Brodland DG, Zitelli JA. Estimation of prognosis in invasive melanoma using a gene expression profile test. Presented at: 48th Annual Meeting of the American College of Mohs Surgery: April 29, 2016: Orlando FL.