Update: Biomarkers for Guiding Treatment in Early Stage Melanoma

Melanoma incidence has been dramatically increasing over the last 20 years in the US, and it is projected that more than 100,000 new cases of melanoma will be diagnosed in this country by the end of 2020 compared to 65,647 in 2011.^1,2 Excluding nonmelanoma skin cancer (NMSC), melanoma is the fifth most common cancer in this country. Although it accounts for only one percent of all skin cancers, melanoma causes the majority of deaths, and it is projected that nearly 7,000 people will die from melanoma by the end of this year.³

The American Joint Committee on Cancer (AJCC) staging system classifies melanoma into stage I through IV disease. Survival in early stage melanoma is fairly high, but there is a significant decrease in survival with more advanced disease. The prognosis of stage 0 to II (early stage) melanoma is related to the depth of invasion. Ten-year survival rates range between 98 percent in stage IA disease and 75 percent in stage IIC disease. Notably, however, a subset of patients with thicker primary melanomas (T3 or T4), ulceration, or more extensive lymph node involvement have been identified who have a higher baseline risk of recurrence and death from disease.^4-6 In stage III disease, melanoma has spread to the lymph nodes, which is associated with a significantly worse outcome, with 10-year survival of 88 percent, 77 percent, 60 percent, and 24 percent for stage IIIA, IIIB, IIIC, and IIID disease, respectively. In metastatic stage IV disease, there is a significant drop in survival with a five-year survival rate of 23 percent.^7-9

The differences in survival according to stage of disease influences the choice of therapy. Surgery with wide local excision is generally curative for earlier stage I and II melanoma, adjuvant therapy is recommended for stage III disease, and stage IV disease often requires aggressive treatment with immunotherapy, targeted therapy, radiation, or a combination of these modalities. Advances in therapy have been made for more advanced (stage III and IV) disease, but earlier stages of disease have received less attention, perhaps due to their overall more favorable outcomes. Currently, work is being done to improve diagnosis, biomarker development for risk stratification, staging, and treatment for early stage 0, I, and II melanoma. Here, we discuss the current tools for risk stratification of melanoma, controversy surrounding the sentinel lymph node biopsy (SLNB) for staging, and development of novel adjuvant immunotherapies for early stage disease.

Risk stratification of melanoma

Primary melanomas are assessed through the AJCC staging system, guiding both surgical and medical therapy and predicting outcomes such as likelihood of recurrence and death. This staging system includes primary tumor characteristics, such as thickness, ulceration, and mitotic rate, lymph node assessment, and search for distant metastases. Breslow thickness and ulceration are each independent predictors of melanoma specific survival (MSS) and recurrence-free survival (RFS).^10-12

Sentinel lymph node biopsy. Lymph node staging is often used for prognostication and is currently recommended for staging of thicker melanomas. Its use in thin melanoma (<1mm) is controversial, especially in the head and neck areas where lymphatic drainage is ambiguous.

The American Academy of Dermatology (AAD) recommends that SLNB be considered in patients with melanomas >1mm thick or <0.8mm thick with ulceration or 0.8 to 1mm thick with or without ulceration. The multicenter selective lymphadenectomy trial-1 (MSLT-1) trial of SLNB versus nodal observation found that in patients with intermediate to thick melanoma (1.2 to 3.5mm), positive SLN metastasis was associated with poorer outcomes than in patients with a negative SLN as evaluated by 10-year MSS (62 percent with positive SLN vs. 85 percent with negative SLN), highlighting the prognostic value of the SLNB. In addition, the MSLT-1 trial found that biopsy-based therapy improved 10-year disease free survival and melanoma specific survival in patients with intermediate to thick melanoma and with positive sentinel lymph nodes. The role of SLNB in thin melanoma (<1.2mm) could not be determined in the study as the sample size was too small to achieve statistical significance.¹³ A further meta-analysis evaluating the prognostic value of the SLNB found that negative SLN status was not associated with survival benefit compared to positive SLN status in thin melanoma but it was associated with worse outcomes in thick melanoma (>4mm).¹⁴ Yet, these studies did not take into account characteristics of the primary tumor, which would increase the likelihood of regional node micro-metastasis including ulceration, high mitotic rate, and Clark level IV/V invasion.¹⁵ The current AAD guidelines take these factors into account, recommending SLNB in thin melanomas (<0.8mm) with ulceration.

An additional concern surrounding the SLNB is its use in areas with ambiguous lymphatic draining such as the head and neck. Some studies have shown that there are a greater number of false negative SLNBs in these areas (1.5 percent for head and neck versus 0.5 percent for trunk and extremities).^16-19 However, more recent studies have shown similar SLNB identification rates in head and neck and non-head and neck sites (99.7 percent and 97.5 percent identification, respectively).²⁰

The follow-up MSLT-2 trial examined the use of a complete lymph node dissection for patients with positive SLNB and found that there was no significant improvement in three-year melanoma specific survival. This suggests that the survival advantage conferred after SLNB in MSLT-1 was due to disease that was limited to the sentinel node. Additionally, significant complications occurred after SLN dissection, and thus it is not recommended as the standard of care.²¹

Further contributing to the controversy surrounding the SLNB, it has been shown that patients with stage IIC disease (SLN negative, ulcerated primary >4mm thick) have worse survival than patients with stage IIIA disease (any Breslow thickness, non-ulcerated, SLN positive). The estimated five-year survival rate for IIC vs. IIIA disease was 56 [95 percent confidence interval: 30 to 70] percent versus 79 [95 percent confidence interval: 67 to 87] percent. Additionally, patients with matched stage IIIB disease had similar 5-year survival to patients with IIC disease. The authors suggested that these results call into question the value of the SLNB as a staging tool in the population.^6,22

Biomarkers. As the value of the SLNB status is controversial, there is a need for other prognostic biomarkers to risk stratify patients with melanoma and to guide treatment decisions. While tumor thickness, ulceration, mitotic index, and presence of an immune infiltrate allow for appropriate general staging of tumors, they cannot distinguish high- from low-risk for relapse within the same stage. Several studies conducted to identify prognostic biomarkers in early stage melanoma have identified Ki67 expression, driver mutations such as BRAF, and gene expression profiles.²³

One commercially available test is DecisionDx-Melanoma (Castle Biosciences) which classifies patients into low-risk (class 1) or high-risk (class 2) groups based on a 31-gene panel.²⁴ DecisionDx-Melanoma identifies up to 90 percent of patients with high-risk stage I and II disease. Although the test has high analytic validity, there is no perfect correlation between gene expression profiling and AJCC staging. When used in conjunction with the AJCC staging system, DecisionDx-Melanoma has the potential to improve identification of the high risk patients with stage I and II disease who are at high risk of recurrence and metastasis.²⁵

Extensive evidence shows that the tumor microenvironment (TME) modulates the progression of melanoma. A 53-gene melanoma immune profile (MIP) was identified that predicts both distant metastatic recurrence and disease-specific survival (MSS).^26-29 Additionally, quantitative multiplex immunofluorescence (qmIF) found that a low cytotoxic T lymphocyte (CTL) to macrophage ratio in the stroma was associated with shortened overall survival, highlighting its potential use as a prognostic biomarker. Development of these tissue-based biomarkers in primary melanoma tumors has been limited by the small size of tumors which necessitates that all tissue be formalin fixed and paraffin embedded for pathology review. Additionally, samples often need to be sent between institutions for analysis. The use of artificial intelligence (AI) to analyze hematoxylin and eosin (H&E) slides of primary melanoma samples would provide a method to risk stratify patients based on scanned images of primary melanoma samples, overcoming these limitations.⁷ This promising biomarker for risk stratification would allow for inter-institutional collaboration and low patient costs. Analysis of images from H&E slides of the original melanoma biopsy has shown potential in predicting melanoma recurrence.⁷

Immunotherapy for Early Stage Melanoma

Development of biomarkers is crucial for selecting patients with early stage melanoma who would benefit from treatment with immunotherapy. The Society for Immunotherapy of Cancer developed consensus guidelines using risk stratification as the basis for decision-making, and they recommend that the use of adjuvant therapy is driven by the risk of disease recurrence.

Several adjuvant therapies have shown significant benefit in stage III and IV melanoma.³⁰ The first adjuvant treatment available for high-risk melanoma was interferon-alpha (IFN-a). The ECOG 1684 and Intergroup E1694 trials demonstrated prolonged relapse-free survival and overall survival in patients with T4 or N1 melanoma.^31,32 Yet, IFN-a is no longer used in the adjuvant setting for the treatment of melanoma, due to severe side effects (constitutional symptoms, chronic fatigue, myelosuppression, hepatotoxicity, neurologic and psychiatric effects), and has been replaced by newer, less toxic agents.³²

The next advancement in the treatment of metastatic melanoma was development of checkpoint inhibitors. Ipilimumab, an anti-cytotoxic lymphocyte associated protein (CTLA-4) antibody, was the first checkpoint inhibitor to be approved for adjuvant therapy. The EORTC 18071 Phase 3 clinical trial demonstrated increased metastasis-free survival and overall survival in the adjuvant setting for stage III melanoma.³³ However, ipilimumab is also associated with severe adverse events and has thus been replaced by the more effective and less toxic checkpoint inhibitors, nivolumab and pembrolizumab.

Nivolumab and pembrolizumab (anti-PD1 antibodies) have greater efficacy in metastatic disease. They work by blocking the interactions between PD-1 and PD-L1 allowing for the activation of a latent immune response against the malignancy. The mechanism of modifying the immune response rather than acting directly on the cancer cells is especially important in the adjuvant setting where tumor burden is at a microscopic level. These newer agents lead to prolonged disease-free survival when used as an adjuvant treatment for resected stage IIIB, IIIC, or IV melanoma or in unresected stage IV melanoma.^34-37 Nivolumab has been shown to be superior to ipilimumab as measured by increased recurrence-free survival with fewer adverse events (CheckMate 238 trial, 14.4 percent vs. 45.9 percent grade 3-4 adverse events, respectively). Additionally, combination nivolumab and ipilimumab therapy has greater benefit than nivolumab alone, but its use clinically is limited by the side effect profile of this regimen.³⁵ Targeted agents have also been developed for melanoma with specific genetic mutations. Adjuvant treatment with dabrafenib (BRAF inhibitor) plus trametinib (MEK inhibitor) in stage III melanoma improves outcomes in tumors with the BRAFV600E mutation.³⁸

Given the success of immunotherapies in advanced disease, there has been interest in their role in earlier stage disease to prevent recurrence. Yet, the majority of clinical trials conducted in the adjuvant setting have enrolled patients at higher risk for relapse, defined as those with AJCC stage III disease (Table 1). Since the majority of patients with stage I and II melanoma are cured with wide local excision alone, none of the trials testing immunotherapy-based regimens include this patient subset. However, there is a high-risk subset of stage II melanoma patients for which there is no FDA-approved therapy. Clinical trials are currently ongoing to examine the role of immunotherapy in earlier stage disease. One of the ongoing clinical trials (Keynote 716) is comparing the safety and efficacy of pembrolizumab to placebo as adjuvant therapy in stage II melanoma.³⁹ Another ongoing clinical trial is evaluating the use of adjuvant nivolumab in patients with stage IIB and C melanoma.⁴⁰ These trials could have significant impact on patients with high risk stage II disease for which no adjuvant therapy is currently approved.

Because the recommendations for adjuvant therapy are based on risk stratification, several studies have examined whether biomarkers can predict response to treatment with nivolumab. Cell-based and serum biomarkers have been shown to predict anti-PD-1 response in metastatic melanoma. Weber and collaborators analyzed patients with stage IV melanoma on adjuvant treatment with nivolumab and found that mass spectrometry was able to differentiate a group of patients with survival over 50 percent from those with less than 20 percent at three years follow up.³⁴ Further protein set enrichment analysis showed that the subjects with higher levels of acute phase reactants, complement, and wound healing biomarkers were more resistant to treatment, in line with literature showing that complement activation may inhibit the efficacy of adaptive antitumor immunity independent of the PD-1/PD-L1 immune checkpoint pathway.⁴¹

Additionally, analysis of peripheral blood mononuclear cells (PBMCs) from patients with stage IV melanoma before and after three months of treatment with nivolumab showed that frequency of CD14+CD16-CD33+HLA-DRhi monocytes, high levels of activation markers, and better migration predicted response to the immunotherapy. This suggests that activated monocytes may heighten the development of an effective anti-tumor response that is induced by the anti-PD-1 immunotherapy.⁴² Similar studies conducted in stage II melanoma may identify novel biomarkers that predict response to anti-PD-1 immunotherapy, allowing for patient-directed treatment.

Conclusions

Melanoma is continuing to rise in incidence in the US, and although significant advances have been made for treatment of patients with stage III and IV disease, FDA-approved treatments for high-risk stage II patients are needed. Additionally, there are few well-established biomarkers that can differentiate high risk from low risk stage II melanoma patients. The SLNB in conjunction with the AJCC staging may provide additional prognostic information in thick melanoma, but its use in thin melanoma is controversial. Artificial intelligence algorithms show promise in determining risk of recurrence of primary melanoma samples. These biomarkers would allow for targeted treatment of high-risk patients with adjuvant immunotherapy, reducing unnecessary cost and toxicity for patients who would be less likely to benefit from this treatment.

Disclosures: None relevant to this publication

1. Netscher DT, et al. “Cutaneous malignancies: melanoma and nonmelanoma types,” Plastic and Reconstructive Surgery. 2011;127(3):37e-56e.

2. Guy GP, Jr. et al. “Vital Signs: Melanoma Incidence and Mortality Trends and Projections - United States, 1982-2030,” Morbidity and Mortality Weekly Report, 2015.

3. American Cancer Society. “Key Statistics for Melanoma Skin Cancer.”https://www.cancer.org/cancer/melanoma-skin-cancer/about/key-statistics.html. Accessed February 22, 2020.

4. Dickson PV, Gershenwald JE. “Staging and prognosis of cutaneous melanoma,” Surgical Oncology Clinics of North America. 2011;20(1):1-17.

5. Whiteman DC, Baade PD, Olsen CM. “More People Die from Thin Melanomas (≤1 mm) than from Thick Melanomas (&gt;4 mm) in Queensland, Australia,” Journal of Investigative Dermatology. 2015;135(4):1190-1193.

6. Yushak M, et al. “Approaches to High-Risk Resected Stage II and III Melanoma,” American Society of Clinical Oncology Annual Meeting. 2019;39:e207-e211.

7. Gershenwald JE, et al. “Melanoma staging: Evidence-based changes in the American Joint Committee on Cancer eighth edition cancer staging manual,” CA. 2017;67(6):472-492.

8. Essner R, et al. “Contemporary surgical treatment of advanced-stage melanoma,” Archives of Surgery. 2004;139(9):961-966; discussion 966-967.

9. Sandru A, et al. “Survival rates of patients with metastatic malignant melanoma,” Journal of Medicine and Life. 2014;7(4):572-576.

10. Thomas NE, et al. “Tumor-infiltrating lymphocyte grade in primary melanomas is independently associated with melanoma-specific survival in the population-based genes, environment and melanoma study,” J Clin Oncol. 2013;31(33):4252-4259.

11. Lee N, et al. “Tumour-infiltrating lymphocytes in melanoma prognosis and cancer immunotherapy,” Pathology. 2016;48(2):177-187.

12. Trinidad CM,et al. “Update on eighth edition American Joint Committee on Cancer classification for cutaneous melanoma and overview of potential pitfalls in histological examination of staging parameters,” J Clin Pathol. 2019;72(4):265-270.

13. Gershenwald JE, Scolyer RA, Hess KR, et al. “Melanoma staging: Evidence-based changes in the American Joint Committee on Cancer eighth edition cancer staging manual,” CA. 2017;67(6):472-492.

14. Freeman SR, et al. “Prognostic value of sentinel lymph node biopsy compared with that of Breslow thickness: implications for informed consent in patients with invasive melanoma,” Dermatologic Surgery. 2013;39(12):1800-1812.

15. Balch CM, Gershenwald JE. “Clinical value of the sentinel-node biopsy in primary cutaneous melanoma,” The New England Journal of Medicine. 2014;370(7):663-664.

16. O’Brien CJ, et al. “Prediction of potential metastatic sites in cutaneous head and neck melanoma using lymphoscintigraphy,” AmericanJjournal of Surgery. 1995;170(5):461-466.

17. Willis AI, Ridge JA. “Discordant lymphatic drainage patterns revealed by serial lymphoscintigraphy in cutaneous head and neck malignancies,” Head & Neck. 2007;29(11):979-985.

18. Chao C, et al. “Sentinel lymph node biopsy for head and neck melanomas,”Annals of Surgical Oncology. 2003;10(1):21-26.

19. Carlson GW, et al. “Regional recurrence after negative sentinel lymph node biopsy for melanoma,” Annals of Surgery. 2008;248(3):378-386.

20. Erman AB, et al. “Sentinel lymph node biopsy is accurate and prognostic in head and neck melanoma,” Cancer. 2012;118(4):1040-1047.

21. Faries MB, et al.” Completion Dissection or Observation for Sentinel-Node Metastasis in Melanoma,” The New England Journal of Medicine. 2017;376(23):2211-2222.

22. Kim CC, Najita JS, Tan SY-M, et al. “Factors associated with worse outcome for patients with AJCC stage IIC relative to stage IIIA melanoma,” Journal of Clinical Oncology. 2015;33(15_suppl):9078-9078.

23. Rizk EM, Seffens AM, Trager MH, et al. “Biomarkers Predictive of Survival and Response to Immune Checkpoint Inhibitors in Melanoma,” American Journal of Clinical Dermatology. 2019.

24. Cook RW, et al. “Analytic validity of DecisionDx-Melanoma, a gene expression profile test for determining metastatic risk in melanoma patients,” Diagn Pathol. 2018;13(1):13-13.

25. Ferris LK, 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,” Journal of the American Academy of Dermatology. 2017;76(5):818-825.e813.

26. Gartrell RD, et al. “Quantitative Analysis of Immune Infiltrates in Primary Melanoma,” Cancer Immunol Res. 2018;6(4):481-493.

27. Gartrell RD, et al. “Validation of Melanoma Immune Profile (MIP), a Prognostic Immune Gene Prediction Score for Stage II-III Melanoma,” Clin Cancer Res. 2019;25(8):2494-2502.

28. Sivendran S, Chang R, Pham L, et al. “Dissection of immune gene networks in primary melanoma tumors critical for antitumor surveillance of patients with stage II-III resectable disease,” J Invest Dermatol. 2014;134(8):2202-2211.

29. Gartrell-Corrado RD, et al. “Linking transcriptomic and imaging data defines features of a favorable tumor immune microenvironment and identifies a combination biomarker for primary melanoma, Cancer Res. 2020.

30. Geskin L, Brown CR, Kirkwood JM. “Adjuvant therapy of melanoma. Seminars in Cutaneous Medicine and Surgery,” 2003;22(1):55-67.

31. Tarhini AA, Gogas H, Kirkwood JM. “IFN-α in the treatment of melanoma. Journal of Immunology,” 2012;189(8):3789-3793.

32. Kirkwood JM, et al. “Interferon alfa-2b adjuvant therapy of high-risk resected cutaneous melanoma: the Eastern Cooperative Oncology Group Trial EST 1684,” Journal of Clinical Oncology: 1996;14(1):7-17.

33. Eggermont AMM, et al. “Prolonged Survival in Stage III Melanoma with Ipilimumab Adjuvant Therapy,” New England Journal of Medicine. 2016;375(19):1845-1855.

34. Weber J, et al. “Adjuvant Nivolumab versus Ipilimumab in Resected Stage III or IV Melanoma,” The New England Journal of Medicine. 2017;377(19):1824-1835.

35. Larkin J,et al. “Combined Nivolumab and Ipilimumab or Monotherapy in Untreated Melanoma, The New England Journal of Medicine. 2015;373(13):1270-1271.

36. Schachter J, et al. “Pembrolizumab versus ipilimumab for advanced melanoma: final overall survival results of a multicentre, randomised, open-label phase 3 study (KEYNOTE-006),” Lancet. 2017;390(10105):1853-1862.

37. Eggermont AMM, et al. “Adjuvant Pembrolizumab versus Placebo in Resected Stage III Melanoma,” New England Journal of Medicine. 2018;378(19):1789-1801.

38. Long GV, et al. “Adjuvant Dabrafenib plus Trametinib in Stage III BRAF-Mutated Melanoma,” The New England Journal of Medicine. 2017;377(19):1813-1823.

39. National Library of Medicine. Identifier NCT03553836. “Safety and Efficacy of Pembrolizumab Compared to Placebo in Resected High-risk Stage II Melanoma (MK-3475-716/KEYNOTE-716),” Available from: https://clinicaltrials.gov/ct2/show/NCT03553836

40. CNational Library of Medicine. Identifier NCT03405155. “Nivolumab in Treating Patients with Stage IIB-IIC Melanoma That Can Be Removed By Surgery,” Available from: https://clinicaltrials.gov/ct2/show/NCT03405155

41. Wang Y, et al. “Autocrine Complement Inhibits IL10-Dependent T-cell-Mediated Antitumor Immunity to Promote Tumor Progression,” Cancer Discovery. 2016;6(9):1022-1035.

42. Krieg C, et al. “High-dimensional single-cell analysis predicts response to anti-PD-1 immunotherapy,” Nature Medicine. 2018;24(2):144-153.

43. Trager MH, Queen, D., Geskin, L.J. “Emerging Therapies for Early Stage Melanoma,” Jacobs Journal of Experimental Dermatology. 2019;6(2).