Development of a prognostic genetic signature to predict the metastatic risk associated with cutaneous melanoma

31-GEP (DecisionDx): a review of clinical utility and performance in a Mayo Clinic cohort

BACKGROUND

Cutaneous melanoma (CM) is a significant health concern because of its high metastatic potential. Gene Expression Profile (GEP) testing, particularly the 31-GEP test (DecisionDx-Melanoma), has been increasingly used for risk stratification in CM patients. This study aimed to evaluate the clinical utility and performance of the 31-GEP test in a real-world setting.

METHODS

Patients with CM who underwent 31-GEP testing from August 2014 to August 2022 at our institution were identified through searches of electronic health records. The study analyzed the influence of 31-GEP testing on clinical decision-making related to sentinel lymph node biopsy (SLNB), medical oncology referral, and postdiagnosis surveillance. Kaplan-Meier curves and Cox proportional hazard models were used to elucidate the test's performance, focusing on relapse-free survival (RFS) and melanoma-specific survival (MSS).

RESULTS

The study included 65 CM patients. Dermatologists ordered more than 80% of 31-GEP tests. In 81.5% of cases, 31-GEP results did not alter standard clinical management. SLNB decisions were unaffected in 92% of patients with pre-SLNB 31-GEP results. Among patients with stage I-IIA melanoma, 25% of those with high-risk 31-GEP results were referred to medical oncology. Contrary to expectations, the rate of nodal metastasis was higher in low-risk than in high-risk 31-GEP cases. Survival analysis showed overlapping RFS and MSS curves between different 31-GEP classes, suggesting limited prognostic value.

CONCLUSIONS

The 31-GEP test has a limited impact on clinical management decisions and shows limited prognostic value.

Reporting tumor genomic test results to SEER registries via linkages

BACKGROUND

Precision medicine has become a mainstay of cancer care in recent years. The National Cancer Institute (NCI) Surveillance, Epidemiology, and End Results (SEER) Program has been an authoritative source of cancer statistics and data since 1973. However, tumor genomic information has not been adequately captured in the cancer surveillance data, which impedes population-based research on molecular subtypes. To address this, the SEER Program has developed and implemented a centralized process to link SEER registries' tumor cases with genomic test results that are provided by molecular laboratories to the registries.

METHODS

Data linkages were carried out following operating procedures for centralized linkages established by the SEER Program. The linkages used Match*Pro, a probabilistic linkage software, and were facilitated by the registries' trusted third party (an honest broker). The SEER registries provide to NCI limited datasets that undergo preliminary evaluation prior to their release to the research community.

RESULTS

Recently conducted genomic linkages included OncotypeDX Breast Recurrence Score, OncotypeDX Breast Ductal Carcinoma in Situ, OncotypeDX Genomic Prostate Score, Decipher Prostate Genomic Classifier, DecisionDX Uveal Melanoma, DecisionDX Preferentially Expressed Antigen in Melanoma, DecisionDX Melanoma, and germline tests results in Georgia and California SEER registries.

CONCLUSIONS

The linkages of cancer cases from SEER registries with genomic test results obtained from molecular laboratories offer an effective approach for data collection in cancer surveillance. By providing de-identified data to the research community, the NCI's SEER Program enables scientists to investigate numerous research inquiries.

Sentinel lymph node risk prognostication in primary cutaneous melanoma through tissue-based profiling, potentially redefining the need for sentinel lymph node biopsy

PURPOSE OF REVIEW

The role of Sentinel Lymph Node Biopsy (SLNB) is pivotal in the contemporary staging of cutaneous melanoma. In this review, we examine advanced molecular testing platforms like gene expression profiling (GEP) and immunohistochemistry (IHC) as tools for predicting the prognosis of sentinel lymph nodes. We compare these innovative approaches with traditional staging assessments. Additionally, we delve into the shared genetic and protein markers between GEP and IHC tests and their relevance to melanoma biology, exploring their prognostic and predictive characteristics. Finally, we assess alternative methods to potentially obviate the need for SLNB altogether.

RECENT FINDINGS

Progress in adjuvant melanoma therapy has diminished the necessity of Sentinel Lymph Node Biopsy (SLNB) while underscoring the importance of accurately identifying high-risk stage I and II melanoma patients who may benefit from additional anti-tumor interventions. The clinical application of testing through gene expression profiling (GEP) or immunohistochemistry (IHC) is gaining traction, with platforms such as DecisionDx, Merlin Assay (CP-GEP), MelaGenix GEP, and Immunoprint coming into play. Currently, extensive validation studies are in progress to incorporate routine molecular testing into clinical practice. However, due to significant methodological limitations, widespread clinical adoption of tissue-based molecular testing remains elusive at present.

SUMMARY

While various tissue-based molecular testing platforms have the potential to stratify the risk of sentinel lymph node positivity (SLNP), most suffer from significant methodological deficiencies, including limited sample size, lack of prospective validation, and limited correlation with established clinicopathological variables. Furthermore, the genes and proteins identified by individual gene expression profiling (GEP) or immunohistochemistry (IHC) tests exhibit minimal overlap, even when considering the most well-established melanoma mutations. However, there is hope that the ongoing prospective trial for the Merlin Assay may safely reduce the necessity for SLNB procedures if successful. Additionally, the MelaGenix GEP and Immunoprint tests could prove valuable in identifying high-risk stage I-II melanoma patients and potentially guiding their selection for adjuvant therapy, thus potentially reducing the need for SLNB. Due to the diverse study designs employed, effective comparisons between GEP or IHC tests are challenging, and to date, there is no study directly comparing the clinical utility of these respective GEP or IHC tests.

Inconsistent Associations Between Risk Factor Profiles and Adjuvant Radiation Therapy (ART) Treatment in Patients with Cutaneous Squamous Cell Carcinoma and Utility of the 40-Gene Expression Profile to Refine ART Guidance

INTRODUCTION

National Comprehensive Cancer Network (NCCN) recommendations for adjuvant radiation therapy (ART) use are similar for High Risk and Very High Risk cutaneous squamous cell carcinoma (cSCC) with negative post-surgical margins. Although studies report reductions in disease progression following ART treatment, ART use is likely inconsistent when guided by available risk factors. This study evaluated the association of ART with clinical risk factors in ART-treated and untreated patients and showed the clinical utility of the 40-gene expression profile (40-GEP) for guiding ART.

METHODS

A multicenter study of 954 patients was conducted with institutional review board (IRB) approval. The 40-GEP test was performed using primary tumor tissue from patients with either a minimum of 3 years of follow-up or a documented regional or distant metastasis. Unsupervised hierarchical cluster analysis identified patterns of clinical risk factors for ART-treated patients, then identified untreated patients with matching risk factor profiles. Results were cross-referenced to 40-GEP test results to determine utility of the test to guide ART.

RESULTS

Analysis demonstrated inconsistent implementation of ART for eligible patients. Cluster analysis identified four patient profiles based on clusters of risk factors and, notably, matching profiles in ART-treated and untreated patients. Further, the analysis identified patients who received but could have deferred ART on the basis of 40-GEP test result and biologically low risk of metastasis, and untreated patients who likely would have benefitted from ART on the basis of their 40-GEP test result.

CONCLUSIONS

ART guidance is not determined by the presence of specific clinicopathologic factors, with treated and untreated patients sharing the same risk factor profiles. cSCC risk determination based on NCCN recommendations for clinical factor assessment results in inconsistent use of ART. Including tumor biology-based prognostic information from the 40-GEP refines risk and identifies patients who are most appropriate and likely to benefit from ART, and those that can consider deferring ART.

Factors Affecting Recurrence and Survival for Patients with High-Risk Stage II Melanoma

BACKGROUND

In the current era of effective adjuvant therapies and de-escalation of surgery, distinguishing which patients with high-risk stage II melanoma are at increased risk of recurrence after excision of the primary lesion is essential to determining appropriate treatment and surveillance plans.

METHODS

A single-center retrospective study analyzed patients with stage IIB or IIC melanoma. Demographic and tumor data were collected, and genomic analysis of formalin-fixed, paraffin-embedded tissue samples was performed via an internal next-generation sequencing (NGS) platform (SNaPshot). The end points examined were relapse-free survival (RFS), distant metastasis-free survival (DMFS), overall survival (OS), and melanoma-specific survival (MSS). Uni- and multivariable Cox regressions were performed to calculate the hazard ratios.

RESULTS

The study included 92 patients with a median age of 69 years and a male/female ratio of 2:1. A Breslow depth greater than 4 mm, a higher mitotic rate, an advanced T stage, and a KIT mutation had a negative impact on RFS. A primary lesion in the head and neck, a mitotic rate exceeding 10 mitoses per mm2, a CDH1 mutation, or a KIT mutation was significantly associated with a shorter DMFS. Overall survival was significantly lower with older age at diagnosis and a higher mitotic rate. An older age at diagnosis also had a negative impact on MSS.

CONCLUSION

Traditional histopathologic factors and specific tumor mutations displayed a significant correlation with disease recurrence and survival for patients with high-risk stage II melanoma. This study supported the use of genomic testing of high-risk stage II melanomas for prognostic prediction and risk stratification.

Prognostic and predictive biomarkers in melanoma

Biomarkers help to inform the clinical management of patients with melanoma. For patients with clinically localised primary melanoma, biomarkers can help to predict post-surgical outcome (including via the use of risk prediction tools), better select patients for sentinel lymph node biopsy, and tailor catch-all follow-up protocols to the individual. Systemic drug treatments, including immune checkpoint inhibitor (ICI) therapies and BRAF-targeted therapies, have radically improved the prognosis of metastatic (stage III and IV) cutaneous melanoma patients, and also shown benefit in the earlier setting of stage IIB/C primary melanoma. Unfortunately, a response is far from guaranteed. Here, we review clinically relevant, established, and emerging, prognostic, and predictive pathological biomarkers that refine clinical decision-making in primary and metastatic melanoma patients. Gene expression profile assays and nomograms are emerging tools for prognostication and sentinel lymph node risk prediction in primary melanoma patients. Biomarkers incorporated into clinical practice guidelines include BRAF V600 mutations for the use of targeted therapies in metastatic cutaneous melanoma, and the HLA-A∗02:01 allele for the use of a bispecific fusion protein in metastatic uveal melanoma. Several predictive biomarkers have been proposed for ICI therapies but have not been incorporated into Australian clinical practice guidelines. Further research, validation, and assessment of clinical utility is required before more prognostic and predictive biomarkers are fluidly integrated into routine care.

The Use of Gene Expression Profiling and Biomarkers in Melanoma Diagnosis and Predicting Recurrence: Implications for Surveillance and Treatment

Cutaneous melanoma is becoming more prevalent in the United States and has the highest mortality among cutaneous malignancies. The majority of melanomas are diagnosed at an early stage and, as such, survival is generally favorable. However, there remains prognostic uncertainty among subsets of early- and intermediate-stage melanoma patients, some of whom go on to develop advanced disease while others remain disease-free. Melanoma gene expression profiling (GEP) has evolved with the notion to help bridge this gap and identify higher- or lower-risk patients to better tailor treatment and surveillance protocols. These tests seek to prognosticate melanomas independently of established AJCC 8 cancer staging and clinicopathologic features (sex, age, primary tumor location, thickness, ulceration, mitotic rate, lymphovascular invasion, microsatellites, and/or SLNB status). While there is a significant opportunity to improve the accuracy of melanoma prognostication and diagnosis, it is equally important to understand the current landscape of molecular profiling for melanoma treatment. Society guidelines currently do not recommend molecular testing outside of clinical trials for melanoma clinical decision making, citing insufficient high-quality evidence guiding indications for the testing and interpretation of results. The goal of this chapter is to review the available literature for GEP testing for melanoma diagnosis and prognostication and understand their place in current treatment paradigms.

The 31-Gene Expression Profile Test Outperforms AJCC in Stratifying Risk of Recurrence in Patients with Stage I Cutaneous Melanoma

BACKGROUND

Patients with stage I cutaneous melanoma (CM) are considered at low risk for metastasis or melanoma specific death; however, because the majority of patients are diagnosed with stage I disease, they represent the largest number of melanoma deaths annually. The 31-gene expression profile (31-GEP) test has been prospectively validated to provide prognostic information independent of staging, classifying patients as low (Class 1A), intermediate (Class 1B/2A), or high (Class 2B) risk of poor outcomes.

METHODS

Patients enrolled in previous studies of the 31-GEP were combined and evaluated for recurrence-free (RFS) and melanoma-specific survival (MSS) (n = 1261, "combined"). A second large, unselected real-world cohort (n = 5651) comprising clinically tested patients diagnosed 2013-2018 who were linked to outcomes data from the NCI Surveillance, Epidemiology, and End Results (SEER) Program registries was evaluated for MSS.

RESULTS

Combined cohort Class 1A patients had significantly higher RFS than Class 1B/2A or Class 2B patients (97.3%, 88.6%, 77.3%, p < 0.001)-better risk stratification than AJCC8 stage IA (97.5%) versus IB (89.3%). The SEER cohort showed better MSS stratification by the 31-GEP (Class 1A = 98.0%, Class 1B/2A = 97.5%, Class 2B = 92.3%; p < 0.001) than by AJCC8 staging (stage IA = 97.6%, stage IB = 97.9%; p < 0.001).

CONCLUSIONS

The 31-GEP test significantly improved patient risk stratification, independent of AJCC8 staging in patients with stage I CM. The 31-GEP provided greater separation between high- (Class 2B) and low-risk (Class 1A) groups than seen between AJCC stage IA and IB. These data support integrating the 31-GEP into clinical decision making for more risk-aligned management plans.

Relationship of Histopathologic Parameters and Gene Expression Profiling in Malignant Melanoma

BACKGROUND

Histopathologic characteristics (HC) are a mainstay in melanoma prognosis; gene expression profiling (GEP) has emerged as a potential additional independent value.

OBJECTIVE

To elucidate HC predictive of groups obtained via GEP of malignant melanoma.

METHODS

A retrospective study analyzing HC of 265 melanomas submitted for GEP over the course of 8 years. GEP was conducted as a part of regular clinicopathologic workup through Castle Biosciences Decision Dx®.

RESULTS

Of the 265 cases, the major HC found to have an association with reported gene expression profiles were melanoma histology subtype, depth of invasion, and presence of ulcer.

LIMITATIONS

This study is limited by its cross-sectional nature. Causation and long-term related outcomes of the use of GEP versus American Joint Committee on Cancer histopathologic staging cannot be ascertained by this design.

CONCLUSIONS

An association, but no definitive prediction, exists between histopathologic categories of depth of invasion, melanoma subtype, and presence or absence of ulcer and gene expression profiles. GEP adds valuable data to the evaluation of malignant melanomas that cannot be definitively predicted by conventional models. The findings add to needed groundwork for comparison of traditional markers and molecular genotyping and begins to build a robust predictive model for better outcomes in patients with malignant melanoma.

Sentinel lymph node assessment in melanoma: current state and future directions

Assessment of sentinel lymph node status is an important step in the evaluation of patients with melanoma for both prognosis and therapeutic management. Pathologists have an important role in this evaluation. The methodologies have varied over time, from the evaluation of dimensions of metastatic burden to determination of the location of the tumour deposits within the lymph node to precise cell counting. However, no single method of sentinel lymph node tumour burden measurement can currently be used as a sole independent predictor of prognosis. The management approach to sentinel lymph node-positive patients has also evolved over time, with a more conservative approach recently recognised for selected cases. This review gives an overview of past and current status in the field with a glimpse into future directions based on prior experiences and clinical trials.

Routine imaging guided by a 31-gene expression profile assay results in earlier detection of melanoma with decreased metastatic tumor burden compared to patients without surveillance imaging studies

Patients with early-stage disease typically have a good prognosis, but still have a risk of recurrence, even with negative sentinel lymph node biopsy (SLNB). This study explores the utility of routine imaging to detect metastases in patients with negative SLNB but high-risk 31 gene expression profile (31-GEP) scores. We retrospectively identified melanoma patients with negative SLNBs. Patients with high-risk GEP results were placed in the experimental group and patients without GEP testing were placed in the control group. Among both cohorts, recurrent melanoma groups were identified. The tumor burden at the time of recurrence and the time to recurrence were compared between experimental group patients with routine imaging and control group patients without imaging schedules. We identified 327 control patients and 307 experimental patients, of which 14.1% versus 20.5% had melanoma recurrence, respectively. Of the patients with recurrent melanoma, those in the experimental group were older (65.75 versus 59.20), had higher Breslow depths (3.72 mm versus 3.31 mm), and had advanced tumor staging (89.5% versus 71.4% of patients presenting clinical stage ≥ II) compared to the control group at primary diagnosis. However, melanoma recurrence was detected earlier (25.50 months versus 35.35 months) in the experimental group at a lower overall tumor burden (73.10 mm versus 27.60 mm). A higher percentage of experimental patients started immunotherapy when offered (76.3% and 67.9%). Patients who received routine imaging after high-risk GEP test scores had an earlier recurrence diagnosis with lower tumor burden, leading to better clinical outcomes.

Robust dimethyl-based multiplex-DIA doubles single-cell proteome depth via a reference channel

Single-cell proteomics aims to characterize biological function and heterogeneity at the level of proteins in an unbiased manner. It is currently limited in proteomic depth, throughput, and robustness, which we address here by a streamlined multiplexed workflow using data-independent acquisition (mDIA). We demonstrate automated and complete dimethyl labeling of bulk or single-cell samples, without losing proteomic depth. Lys-N digestion enables five-plex quantification at MS1 and MS2 level. Because the multiplexed channels are quantitatively isolated from each other, mDIA accommodates a reference channel that does not interfere with the target channels. Our algorithm RefQuant takes advantage of this and confidently quantifies twice as many proteins per single cell compared to our previous work (Brunner et al, PMID 35226415), while our workflow currently allows routine analysis of 80 single cells per day. Finally, we combined mDIA with spatial proteomics to increase the throughput of Deep Visual Proteomics seven-fold for microdissection and four-fold for MS analysis. Applying this to primary cutaneous melanoma, we discovered proteomic signatures of cells within distinct tumor microenvironments, showcasing its potential for precision oncology.

31-Gene Expression Profile Testing in Cutaneous Melanoma and Survival Outcomes in a Population-Based Analysis: A SEER Collaboration

PURPOSE

The DecisionDx-Melanoma 31-gene expression profile (31-GEP) test is validated to classify cutaneous malignant melanoma (CM) patient risk of recurrence, metastasis, or death as low (class 1A), intermediate (class 1B/2A), or high (class 2B). This study aimed to examine the effect of 31-GEP testing on survival outcomes and confirm the prognostic ability of the 31-GEP at the population level.

METHODS

Patients with stage I-III CM with a clinical 31-GEP result between 2016 and 2018 were linked to data from 17 SEER registries (n = 4,687) following registries' operation procedures for linkages. Melanoma-specific survival (MSS) and overall survival (OS) differences by 31-GEP risk category were examined using Kaplan-Meier analysis and the log-rank test. Crude and adjusted hazard ratios (HRs) were calculated using Cox regression model to evaluate variables associated with survival. 31-GEP tested patients were propensity score-matched to a cohort of non-31-GEP tested patients from the SEER database. Robustness of the effect of 31-GEP testing was assessed using resampling.

RESULTS

Patients with a 31-GEP class 1A result had higher 3-year MSS and OS than patients with a class 1B/2A or class 2B result (MSS: 99.7% v 97.1% v 89.6%, P < .001; OS: 96.6% v 90.2% v 79.4%, P < .001). A class 2B result was an independent predictor of MSS (HR, 7.00; 95% CI, 2.70 to 18.00) and OS (HR, 2.39; 95% CI, 1.54 to 3.70). 31-GEP testing was associated with a 29% lower MSS mortality (HR, 0.71; 95% CI, 0.53 to 0.94) and 17% lower overall mortality (HR, 0.83; 95% CI, 0.70 to 0.99) relative to untested patients.

CONCLUSION

In a population-based, clinically tested melanoma cohort, the 31-GEP stratified patients by their risk of dying from melanoma.

The 31-gene expression profile test informs sentinel lymph node biopsy decisions in patients with cutaneous melanoma: results of a prospective, multicenter study

BACKGROUND

The 31-gene expression profile test (Class 1A: low-risk; 1B/2A: intermediate-risk; 2B: high-risk) is validated to identify patients with cutaneous melanoma who can safely forego sentinel lymph node biopsy (SLNB). The objective of the current study is to quantify SLNB reduction by clinicians using 31-GEP.

METHODS

Patients with T1-T2 tumors eligible for SLNB were seen by surgical oncologists (89.1%), dermatologists (7.8%), and medical oncologists (3.1%). After receiving 31-GEP results but before SLNB, clinicians were asked which clinical and pathological features influenced SLNB decisions (n = 191). The Exact binomial test was used to compare SLNB procedure rates to a contemporary study (78% SLNB baseline rate). Logistic regression modeling (odds ratio [OR], 95% CI) was used to identify features associated with SLNB procedure rates.

RESULTS

One hundred clinical decisions (52.4%) were influenced by the 31-GEP to forego SLNB and 70% (70/100) were not performed. Of the 30 performed, 0% (0/30) were positive. The 31-GEP influenced sixty-three clinical decisions (33.0%) to perform SLNB, and 92.1% (58/63) were performed. There was a clinically meaningful 29.4% reduction of SLNBs performed in patients with a Class 1A result relative to the baseline rate of 78.0% (p < .01). In patients ≥55 or ≥65-year-old, SLNB reduction was 32.3% (p < .01), 28.3% (p < .01), respectively. Overall, 85.3% of decisions relating to SLNB were influenced by 31-GEP results.

CONCLUSION

In this prospective, multicenter study, clinicians demonstrated clinically meaningful use of the 31-GEP test to forego or pursue SLNB in patients with T1-T2 tumors resulting in a significant, risk appropriate decrease in SLNBs.

Identification of stage I/II melanoma patients at high risk for recurrence using a model combining clinicopathologic factors with gene expression profiling (CP-GEP)

PURPOSE

Patients with cutaneous melanoma stage I/IIA disease are currently not eligible for adjuvant therapy, despite their risk for relapses and death. This study validates the ability of a model combining clinicopathologic factors with gene expression profiling (CP-GEP) to identify patients at high risk for disease recurrence in stage I/II and subgroup stage I/IIA.

PATIENTS AND METHODS

543 patients with stage I/II primary cutaneous melanoma from the University of Tuebingen diagnosed between 2000 and 2017 were analysed. All patients received sentinel lymph node biopsy (SLNB). Analysis was conducted for a separate group of 80 patients who did not undergo SLNB.

RESULTS

CP-GEP stratified 424 stage I/IIA patients (78% of the cohort) according to their risk for recurrence, with five-year relapse-free survival (RFS) rates of 77.8% and 93% for CP-GEP high risk (195 patients) and low risk (229 patients), respectively, and hazard ratio of 3.53 (p-value <0.001). In patients who did not receive SLNB biopsy, CP-GEP captured 6 out of 7 relapses.

CONCLUSION

CP-GEP can be used to identify primary cutaneous melanoma patients with a high risk for disease recurrence - especially for stage I/IIA, who are considered low risk by AJCC 8th. These patients may benefit from adjuvant therapy. Also, in the future, when SLNB may become irrelevant, CP-GEP may serve as a risk stratification tool.

Molecular testing in melanoma for the surgical pathologist

The diagnostic work-up of melanocytic tumours has undergone significant changes in the last years following the exponential growth of molecular assays. For the practising pathologist it is often difficult to sort through the multitude of different tests that are currently available for clinical use. The molecular tests used in melanocytic pathology can be broadly divided into four categories: (1) tests that predict response to systemic therapy in melanoma; (2) tests that predict prognosis in melanoma; (3) tests useful in determining the type or class of melanocytic tumour; and (4) tests useful in the differential diagnosis of naevus versus melanoma (primarily used as an aid in the diagnosis of histologically ambiguous melanocytic lesions). This review will present an updated synopsis of major molecular ancillary tests used in clinical practice.

CDCA3 is a prognostic biomarker for cutaneous melanoma and is connected with immune infiltration

Introduction

Dysregulation of cell cycle progression (CCP) is a trait that distinguishes cancer from other diseases. In several cancer types, CCP-related genes serve as the primary risk factor for prognosis, but their role in cutaneous melanoma remains unclear.

Methods

Data from cutaneous melanoma patients were acquired from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO). Using a Wilcoxon test, the level of CCP-related gene expression in cutaneous melanoma patient tissues was compared to that in normal skin tissues. Logistic analysis was then utilized to calculate the connection between the CCP-related genes and clinicopathological variables. The important functions of the CCP-related genes were further investigated using Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis, and single-sample Gene Set Enrichment Analysis (ssGSEA). Univariate and multivariate Cox analyses and Kaplan-Meier analysis were used to estimate the association between CCP-related genes and prognosis. In addition, using Cox multivariate analysis, a nomogram was constructed to forecast the influence of CCP-related genes on survival rates.

Results

High expression of CCP-related genes was associated with TNM stage, age, pathological grade, and Breslow depth (P < 0.05). Multivariate analysis demonstrated that CCP-related genes were an independent factor in overall survival and disease-specific survival. High levels of gene expression originating from CCP were shown by GSEA to trigger DNA replication, the G1-S specific transcription factor, the mitotic spindle checkpoint, and the cell cycle. There was a negative association between CCP-related genes and the abundance of innate immune cells. Finally, we revealed that knockdown of cell division cycle-associated gene 3 (CDCA3) significantly suppressed the proliferation and migration ability of cutaneous melanoma cells.

Conclusion

According to this study, CCP-related genes could serve as potential biomarkers to assess the prognosis of cutaneous melanoma patients and are crucial immune response regulators.

Molecular Skin Cancer Diagnosis: Promise and Limitations

Skin cancer is a significant and increasing global health burden. Although the current diagnostic workflow is robust and able to provide clinically actionable results, it is subject to notable limitations. The training and expertise required for accurate diagnoses using conventional skin cancer diagnostics are significant, and patient access to this workflow can be limited by geographic location or unforeseen events, such as coronavirus disease 2019 (COVID-19). Molecular biomarkers have transformed diagnostics and treatment delivery in oncology. With rapid advancements in molecular biology techniques, understanding of the underlying molecular mechanism of cancer pathologies has deepened, yielding biomarkers that can be used to monitor the course of malignant diseases. Herein, commercially available, clinically validated, and emerging skin cancer molecular biomarkers are reviewed. The qualities of an ideal molecular biomarker are defined. The potential benefits and limitations of applying molecular biomarker testing over the course of skin cancer from susceptibility to treatment are explored, with a view to outlining a future model of molecular biomarker skin cancer diagnostics.

Insights and Strategies of Melanoma Immunotherapy: Predictive Biomarkers of Response and Resistance and Strategies to Improve Response Rates

Despite the recent successes and durable responses with immune checkpoint inhibitors (ICI), many cancer patients, including those with melanoma, do not derive long-term benefits from ICI therapies. The lack of predictive biomarkers to stratify patients to targeted treatments has been the driver of primary treatment failure and represents an unmet medical need in melanoma and other cancers. Understanding genomic correlations with response and resistance to ICI will enhance cancer patients' benefits. Building on insights into interplay with the complex tumor microenvironment (TME), the ultimate goal should be assessing how the tumor 'instructs' the local immune system to create its privileged niche with a focus on genomic reprogramming within the TME. It is hypothesized that this genomic reprogramming determines the response to ICI. Furthermore, emerging genomic signatures of ICI response, including those related to neoantigens, antigen presentation, DNA repair, and oncogenic pathways, are gaining momentum. In addition, emerging data suggest a role for checkpoint regulators, T cell functionality, chromatin modifiers, and copy-number alterations in mediating the selective response to ICI. As such, efforts to contextualize genomic correlations with response into a more insightful understanding of tumor immune biology will help the development of novel biomarkers and therapeutic strategies to overcome ICI resistance.

Optimizing treatment approaches for patients with cutaneous melanoma by integrating clinical and pathologic features with the 31-gene expression profile test

BACKGROUND

Many patients with low-stage cutaneous melanoma will experience tumor recurrence, metastasis, or death, and many higher staged patients will not.

OBJECTIVE

To develop an algorithm by integrating the 31-gene expression profile test with clinicopathologic data for an optimized, personalized risk of recurrence (integrated 31 risk of recurrence [i31-ROR]) or death and use i31-ROR in conjunction with a previously validated algorithm for precise sentinel lymph node positivity risk estimates (i31-SLNB) for optimized treatment plan decisions.

METHODS

Cox regression models for ROR were developed (n = 1581) and independently validated (n = 523) on a cohort with stage I-III melanoma. Using National Comprehensive Cancer Network cut points, i31-ROR performance was evaluated using the midpoint survival rates between patients with stage IIA and stage IIB disease as a risk threshold.

RESULTS

Patients with a low-risk i31-ROR result had significantly higher 5-year recurrence-free survival (91% vs 45%, P < .001), distant metastasis-free survival (95% vs 53%, P < .001), and melanoma-specific survival (98% vs 73%, P < .001) than patients with a high-risk i31-ROR result. A combined i31-SLNB/ROR analysis identified 44% of patients who could forego sentinel lymph node biopsy while maintaining high survival rates (>98%) or were restratified as being at a higher or lower risk of recurrence or death.

LIMITATIONS

Multicenter, retrospective study.

CONCLUSION

Integrating clinicopathologic features with the 31-GEP optimizes patient risk stratification compared to clinicopathologic features alone.

Not Your Mother's Melanoma: Causes and Effects of Early Melanoma Diagnosis

The year 2022 will herald approximately 100,000 new cases of cutaneous melanoma (CM), and over 7000 deaths from CM. Over the past 40 years, CM incidence has increased nearly six-fold; however, annual mortality has remained relatively constant. These trends encapsulate the phenomenon of overdiagnosis. Increased recognition of indolent lesions that appear histologically malignant may be leading to a melanoma epidemic. Enhanced melanoma awareness, screening efforts, physician uncertainty, medical-legal pressures, and diagnostic scrutiny using tools like immunohistochemical staining, mole mapping, dermoscopy, confocal microscopy, and molecular diagnostics contribute to increased CM diagnosis. As a result, current melanoma staging and treatment guidelines are being challenged. Existing standards fail to accurately identify histologically benign lesions that are lethal or, conversely, histologically malignant lesions that are innocuous. Healthcare systems and, more importantly, patients suffer from this diagnostic ambiguity that leads to the over-treatment of innocuous melanomas and under-treatment of aggressive melanomas. As dermatology continues to experience a shift towards earlier diagnosis of melanoma, management strategies must adapt. Herein, we review factors that may contribute to the increased incidence of melanoma, emphasize deficiencies in current staging systems, and provide insights into the future of melanoma management via precision medicine.

Discovery of pathway-independent protein signatures associated with clinical outcome in human cancer cohorts

Proteomic data provide a direct readout of protein function, thus constituting an information-rich resource for prognostic and predictive modeling. However, protein array data may not fully capture pathway activity due to the limited number of molecules and incomplete pathway coverage compared to other high-throughput technologies. For the present study, our aim was to improve clinical outcome prediction compared to published pathway-dependent prognostic signatures for The Cancer Genome Atlas (TCGA) cohorts using the least absolute shrinkage and selection operator (LASSO). RPPA data is particularly well-suited to the LASSO due to the relatively low number of predictors compared to larger genomic data matrices. Our approach selected predictors regardless of their pathway membership and optimally combined their RPPA measurements into a weighted risk score. Performance was assessed and compared to that of the published signatures using two unbiased approaches: 1) 10 iterations of threefold cross-validation for unbiased estimation of hazard ratio and difference in 5-year survival (by Kaplan-Meier method) between predictor-defined high and low risk groups; and 2) a permutation test to evaluate the statistical significance of the cross-validated log-rank statistic. Here, we demonstrate strong stratification of 445 renal clear cell carcinoma tumors from The Cancer Genome Atlas (TCGA) into high and low risk groups using LASSO regression on RPPA data. Median cross-validated difference in 5-year overall survival was 32.8%, compared to 25.2% using a published receptor tyrosine kinase (RTK) prognostic signature (median hazard ratios of 3.3 and 2.4, respectively). Applicability and performance of our approach was demonstrated in three additional TCGA cohorts: ovarian serous cystadenocarcinoma (OVCA), sarcoma (SARC), and cutaneous melanoma (SKCM). The data-driven LASSO-based approach is versatile and well-suited for discovery of new protein/disease associations.

SRC-RAC1 signaling drives drug resistance to BRAF inhibition in de-differentiated cutaneous melanomas

Rare gain-of-function mutations in RAC1 drive drug resistance to targeted BRAF inhibition in cutaneous melanoma. Here, we show that wildtype RAC1 is a critical driver of growth and drug resistance, but only in a subset of melanomas with elevated markers of de-differentiation. Similarly, SRC inhibition also selectively sensitized de-differentiated melanomas to BRAF inhibition. One possible mechanism may be the suppression of the de-differentiated state, as SRC and RAC1 maintained markers of de-differentiation in human melanoma cells. The functional differences between melanoma subtypes suggest that the clinical management of cutaneous melanoma can be enhanced by the knowledge of differentiation status. To simplify the task of classification, we developed a binary classification strategy based on a small set of ten genes. Using this gene set, we reliably determined the differentiation status previously defined by hundreds of genes. Overall, our study informs strategies that enhance the precision of BRAFi by discovering unique vulnerabilities of the de-differentiated cutaneous melanoma subtype and creating a practical method to resolve differentiation status.

Dermatologic Disease-Directed Targeted Therapy (D3T2): The Application of Biomarker-Based Precision Medicine for the Personalized Treatment of Skin Conditions-Precision Dermatology

Precision dermatology uses individualized dermatologic disease-directed targeted therapy (D³T²) for the management of dermatoses and for the evaluation and therapy of cutaneous malignancies. Personalized/precision strategies are based on biomarkers that are most frequently derived from tissue transcriptomic expression or genomic sequencing or from circulating cytokines. For instance, the pathologic diagnosis of a pigmented lesion and determining the prognosis of a malignant melanocytic neoplasm can be enhanced by genomic/transcriptomic analysis. In addition to biopsy, innovative techniques have been developed for obtaining transcriptomes in skin conditions; as an example, patches can be applied to a psoriasis plaque for a few minutes to capture the epidermis/upper dermis transcriptome. Atopic dermatitis and prurigo nodularis may also be candidate conditions for precision dermatology. Precision dermatology has a role in managing melanoma and nonmelanoma skin cancers and rare cutaneous tumors-such as perivascular epithelioid cell tumor (PEComa)-that can originate in or metastasize to the skin. For instance, advanced/metastatic basal cell carcinomas can be treated with Hedgehog inhibitors (vismodegib and sonidegib) targeting the smoothened (SMO) or patched 1 (PTCH1) gene alterations that are a hallmark of these cancers and activate the Hedgehog pathway. Advanced/metastatic basal and cutaneous squamous cell cancers often have a high tumor mutational burden (which predicts immunotherapy response); immune checkpoint blockade with cemiplimab, a programmed cell death protein 1 (PD1) inhibitor, is now approved for these malignancies. Gene expression profiling of primary cutaneous squamous cell carcinoma can identify those individuals at high risk for subsequent metastases. In the realm of rare neoplasms, PEComas-which can originate in the skin, albeit uncommonly-have tuberous sclerosis complex 1 (TSC1)/tuberous sclerosis complex 2 (TSC2) gene alterations, which activate mammalian target of rapamycin (mTOR) signaling, and can be suppressed by nab-sirolimus, now approved for this condition. In summary, precision dermatologic techniques/strategies are an important emerging approach for evaluation and management of skin disorders and cutaneous neoplasms, and may serve as a paradigm for the application of precision medicine beyond dermatology.

Attitudes of patients with cutaneous melanoma toward prognostic testing using the 31-gene expression profile test

OBJECTIVE

Although most patients diagnosed with early-stage cutaneous melanoma (CM) have excellent outcomes, because of the large number diagnosed each year, many will experience recurrence or death. Prognostic testing for CM using the 31-gene expression profile (31-GEP) test can benefit patients by helping guide risk-appropriate treatment and surveillance plans. We sought to evaluate patients' attitudes toward prognostic testing with the 31-GEP and assess whether patients experience decision regret about having 31-GEP testing.

METHODS

A 43-question survey was distributed by the Melanoma Research Foundation in June-August 2021 to CM patients enrolled in their database. Patients were asked questions regarding their decision to undergo 31-GEP testing and the extent to which they experienced decision regret using a validated set of Decision Regret Scale questions.

RESULTS

We analyzed responses from patients diagnosed in 2014 or later (n  = 120). Of these, 28 had received 31-GEP testing. Most respondents (n  = 108, 90%) desired prognostic information when diagnosed. Of those who received 31-GEP testing, most felt the results were useful (n  = 22 out of 24) and had regret scores significantly less than neutral regret, regardless of their test results (Class 1: p  < 0.001; Class 2: p  = 0.036). Further, decision regret scores were not significantly different between patients who received a Class 1 31-GEP result and those who received a Class 2 result (mean Class 1 = 1.39 and mean Class 2 = 1.90, p  = 0.058).

CONCLUSIONS

Most newly diagnosed CM patients desired prognostic information about their tumors. Patients who received 31-GEP testing felt it was useful and did not regret their decision to undergo 31-GEP testing.

Expanded evidence that the 31-gene expression profile test provides clinical utility for melanoma management in a multicenter study

OBJECTIVE

National Comprehensive Cancer Network (NCCN) guidelines for cutaneous melanoma (CM) recommend physicians consider increased surveillance for patients who typically have lower melanoma survival rates (stages IIB-IV as determined by the American Joint Committee on Cancer (AJCC), 8th edition). However, up to 15% of patients identified as having a low recurrence risk (stages I-IIA) experience disease recurrence, and some patients identified as having a high recurrence risk will not experience any recurrence. The 31-gene expression profile test (31-GEP) stratifies patient recurrence risk into low (Class 1) and high (Class 2) and has demonstrated risk-appropriate impact on disease management and clinical decisions.

METHODS

Five-year plans for lab work, frequency of clinical visits, and imaging pre- and post-31-GEP test results were assessed for a cohort of 509 stage I-III patients following an interim subset analysis of 247 patients.

RESULTS

After receiving 31-GEP results, 50.6% of patients had a change in management plans in at least one of the following categories-clinical visits, lab work, or surveillance imaging. The changes aligned with the risk predicted by the 31-GEP for 76.1% of patients with a Class 1 result and 78.7% of patients with a Class 2 result. A Class 1 31-GEP result was associated with changes toward low-intensity management recommendations, while a Class 2 result was associated with changes toward high-intensity management recommendations.

CONCLUSION

The 31-GEP can stratify patient recurrence risk in patients with CM, and clinicians understand and apply the prognostic ability of the 31-GEP test to alter patient management in risk-appropriate directions.

Characterization of the CpG Island Hypermethylated Phenotype Subclass in Primary Melanomas

Cutaneous melanoma can be lethal even if detected at an early stage. Epigenetic profiling may facilitate the identification of aggressive primary melanomas with unfavorable outcomes. We performed clustering of whole-genome methylation data to identify subclasses that were then assessed for survival, clinical features, methylation patterns, and biological pathways. Among 89 cutaneous primary invasive melanomas, we identified three methylation subclasses exhibiting low methylation, intermediate methylation, or hypermethylation of CpG islands, known as the CpG island methylator phenotype (CIMP). CIMP melanomas occurred as early as tumor stage 1b and, compared with low-methylation melanomas, were associated with age at diagnosis ≥65 years, lentigo maligna melanoma histologic subtype, presence of ulceration, higher American Joint Committee on Cancer stage and tumor stage, and lower tumor-infiltrating lymphocyte grade (all P < 0.05). Patients with CIMP melanomas had worse melanoma-specific survival (hazard ratio = 11.84; confidence interval = 4.65‒30.20) than those with low-methylation melanomas, adjusted for age, sex, American Joint Committee on Cancer stage, and tumor-infiltrating lymphocyte grade. Genes hypermethylated in CIMP compared with those in low-methylation melanomas included PTEN, VDR, PD-L1, TET2, and gene sets related to development/differentiation, the extracellular matrix, and immunity. CIMP melanomas exhibited hypermethylation of genes important in melanoma progression and tumor immunity, and although present in some early melanomas, CIMP was associated with worse survival independent of known prognostic factors.

Analysis of Melanoma Gene Expression Signatures at the Single-Cell Level Uncovers 45-Gene Signature Related to Prognosis

Since the current melanoma clinicopathological staging system remains restricted to predicting survival outcomes, establishing precise prognostic targets is needed. Here, we used gene expression signature (GES) classification and Cox regression analyses to biologically characterize melanoma cells at the single-cell level and construct a prognosis-related gene signature for melanoma. By analyzing publicly available scRNA-seq data, we identified six distinct GESs (named: “Anti-apoptosis”, “Immune cell interactions”, “Melanogenesis”, “Ribosomal biogenesis”, “Extracellular structure organization”, and “Epithelial-Mesenchymal Transition (EMT)”). We verified these GESs in the bulk RNA-seq data of patients with skin cutaneous melanoma (SKCM) from The Cancer Genome Atlas (TCGA). Four GESs (“Immune cell interactions”, “Melanogenesis”, “Ribosomal biogenesis”, and “Extracellular structure organization”) were significantly correlated with prognosis (p = 1.08 × 10−5, p = 0.042, p = 0.001, and p = 0.031, respectively). We identified a prognostic signature of melanoma composed of 45 genes (MPS_45). MPS_45 was validated in TCGA-SKCM (HR = 1.82, p = 9.08 × 10−6) and three other melanoma datasets (GSE65904: HR = 1.73, p = 0.006; GSE19234: HR = 3.83, p = 0.002; and GSE53118: HR = 1.85, p = 0.037). MPS_45 was independently associated with survival (p = 0.002) and was proved to have a high potential for predicting prognosis in melanoma patients.

Using a Clinicopathologic and Gene Expression (CP-GEP) Model to Identify Stage I-II Melanoma Patients at Risk of Disease Relapse

Background: The current standard of care for patients without sentinel node (SN) metastasis (i.e., stage I−II melanoma) is watchful waiting, while >40% of patients with stage IB−IIC will eventually present with disease recurrence or die as a result of melanoma. With the prospect of adjuvant therapeutic options for patients with a negative SN, we assessed the performance of a clinicopathologic and gene expression (CP-GEP) model, a model originally developed to predict SN metastasis, to identify patients with stage I−II melanoma at risk of disease relapse. Methods: This study included patients with cutaneous melanoma ≥18 years of age with a negative SN between October 2006 and December 2017 at the Sahlgrenska University Hospital (Sweden) and Erasmus MC Cancer Institute (The Netherlands). According to the CP-GEP model, which can be applied to the primary melanoma tissue, the patients were stratified into high or low risk of recurrence. The primary aim was to assess the 5-year recurrence-free survival (RFS) of low- and high-risk CP-GEP. A secondary aim was to compare the CP-GEP model with the EORTC nomogram, a model based on clinicopathological variables only. Results: In total, 535 patients (stage I−II) were included. CP-GEP stratification among these patients resulted in a 5-year RFS of 92.9% (95% confidence interval (CI): 86.4−96.4) in CP-GEP low-risk patients (n = 122) versus 80.7% (95%CI: 76.3−84.3) in CP-GEP high-risk patients (n = 413; hazard ratio 2.93 (95%CI: 1.41−6.09), p < 0.004). According to the EORTC nomogram, 25% of the patients were classified as having a ‘low risk’ of recurrence (96.8% 5-year RFS (95%CI 91.6−98.8), n = 130), 49% as ‘intermediate risk’ (88.4% 5-year RFS (95%CI 83.6−91.8), n = 261), and 26% as ‘high risk’ (61.1% 5-year RFS (95%CI 51.9−69.1), n = 137). Conclusion: In these two independent European cohorts, the CP-GEP model was able to stratify patients with stage I−II melanoma into two groups differentiated by RFS.

Predictors of Sentinel Lymph Node Metastasis in Patients with Thin Melanoma: An International Multi-institutional Collaboration

BACKGROUND

Consideration of sentinel lymph node biopsy (SLNB) is recommended for patients with T1b melanomas and T1a melanomas with high-risk features; however, the proportion of patients with actionable results is low. We aimed to identify factors predicting SLNB positivity in T1 melanomas by examining a multi-institutional international population.

METHODS

Data were extracted on patients with T1 cutaneous melanoma who underwent SLNB between 2005 and 2018 at five tertiary centers in Europe and Canada. Univariable and multivariable logistic regression analyses were performed to identify predictors of SLNB positivity.

RESULTS

Overall, 676 patients were analyzed. Most patients had one or more high-risk features: Breslow thickness 0.8-1 mm in 78.1% of patients, ulceration in 8.3%, mitotic rate > 1/mm² in 42.5%, Clark's level ≥ 4 in 34.3%, lymphovascular invasion in 1.4%, nodular histology in 2.9%, and absence of tumor-infiltrating lymphocytes in 14.4%. Fifty-three patients (7.8%) had a positive SLNB. Breslow thickness and mitotic rate independently predicted SLNB positivity. The odds of positive SLNB increased by 50% for each 0.1 mm increase in thickness past 0.7 mm (95% confidence interval [CI] 1.05-2.13) and by 22% for each mitosis per mm² (95% CI 1.06-1.41). Patients who had one excised node (vs. two or more) were three times less likely to have a positive SLNB (3.6% vs. 9.6%; odds ratio 2.9 [1.3-7.7]).

CONCLUSIONS

Our international multi-institutional data confirm that Breslow thickness and mitotic rate independently predict SLNB positivity in patients with T1 melanoma. Even within this highly selected population, the number needed to diagnose is 13:1 (7.8%), indicating that more work is required to identify additional predictors of sentinel node positivity.

Human liver single nucleus and single cell RNA sequencing identify a hepatocellular carcinoma-associated cell-type affecting survival

Background

Hepatocellular carcinoma (HCC) is a common primary liver cancer with poor overall survival. We hypothesized that there are HCC-associated cell-types that impact patient survival.

Methods

We combined liver single nucleus (snRNA-seq), single cell (scRNA-seq), and bulk RNA-sequencing (RNA-seq) data to search for cell-type differences in HCC. To first identify cell-types in HCC, adjacent non-tumor tissue, and normal liver, we integrated single-cell level data from a healthy liver cohort (n = 9 non-HCC samples) collected in the Strasbourg University Hospital; an HCC cohort (n = 1 non-HCC, n = 14 HCC-tumor, and n = 14 adjacent non-tumor samples) collected in the Singapore General Hospital and National University; and another HCC cohort (n = 3 HCC-tumor and n = 3 adjacent non-tumor samples) collected in the Dumont-UCLA Liver Cancer Center. We then leveraged these single cell level data to decompose the cell-types in liver bulk RNA-seq data from HCC patients’ tumor (n = 361) and adjacent non-tumor tissue (n = 49) from the Cancer Genome Atlas (TCGA) multi-center cohort. For replication, we decomposed 221 HCC and 209 adjacent non-tumor liver microarray samples from the Liver Cancer Institute (LCI) cohort collected by the Liver Cancer Institute and Zhongshan Hospital of Fudan University.

Results

We discovered a tumor-associated proliferative cell-type, Prol (80.4% tumor cells), enriched for cell cycle and mitosis genes. In the liver bulk tissue from the TCGA cohort, the proportion of the Prol cell-type is significantly increased in HCC and associates with a worse overall survival. Independently from our decomposition analysis, we reciprocally show that Prol nuclei/cells significantly over-express both tumor-elevated and survival-decreasing genes obtained from the bulk tissue. Our replication analysis in the LCI cohort confirmed that an increased estimated proportion of the Prol cell-type in HCC is a significant marker for a shorter overall survival. Finally, we show that somatic mutations in the tumor suppressor genes TP53 and RB1 are linked to an increase of the Prol cell-type in HCC.

Conclusions

By integrating liver single cell, single nucleus, and bulk expression data from multiple cohorts we identified a proliferating cell-type (Prol) enriched in HCC tumors, associated with a decreased overall survival, and linked to TP53 and RB1 somatic mutations.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13073-022-01055-5.

High-Plex Spatial RNA Profiling Reveals Cell Type‒Specific Biomarker Expression during Melanoma Development

Early diagnosis of melanoma is critical for improved survival. However, the biomarkers of early melanoma evolution and their origin within the tumor and its microenvironment, including the keratinocytes, are poorly defined. To address this, we used spatial transcript profiling that maintains the morphological tumor context to measure the expression of >1,000 RNAs in situ in patient-derived formalin-fixed, paraffin-embedded tissue sections in primary melanoma and melanocytic nevi. We profiled 134 regions of interest (each 200 μm in diameter) enriched in melanocytes, neighboring keratinocytes, or immune cells. This approach captured distinct expression patterns across cell types and tumor types during melanoma development. Unexpectedly, we discovered that S100A8 is expressed by keratinocytes within the tumor microenvironment during melanoma growth. Immunohistochemistry of 252 tumors showed prominent keratinocyte-derived S100A8 expression in melanoma but not in benign tumors and confirmed the same pattern for S100A8's binding partner S100A9, suggesting that injury to the epidermis may be an early and readily detectable indicator of melanoma development. Together, our results establish a framework for high-plex, spatial, and cell type‒specific resolution of gene expression in archival tissue applicable to the development of biomarkers and characterization of tumor microenvironment interactions in tumor evolution.

Development of an Image Analysis-Based Prognosis Score Using Google's Teachable Machine in Melanoma

BACKGROUND

The increasing number of melanoma patients makes it necessary to establish new strategies for prognosis assessment to ensure follow-up care. Deep-learning-based image analysis of primary melanoma could be a future component of risk stratification.

OBJECTIVES

To develop a risk score for overall survival based on image analysis through artificial intelligence (AI) and validate it in a test cohort.

METHODS

Hematoxylin and eosin (H&E) stained sections of 831 melanomas, diagnosed from 2012-2015 were photographed and used to perform deep-learning-based group classification. For this purpose, the freely available software of Google's teachable machine was used. Five hundred patient sections were used as the training cohort, and 331 sections served as the test cohort.

RESULTS

Using Google's Teachable Machine, a prognosis score for overall survival could be developed that achieved a statistically significant prognosis estimate with an AUC of 0.694 in a ROC analysis based solely on image sections of approximately 250 × 250 µm. The prognosis group "low-risk" (n = 230) showed an overall survival rate of 93%, whereas the prognosis group "high-risk" (n = 101) showed an overall survival rate of 77.2%.

CONCLUSIONS

The study supports the possibility of using deep learning-based classification systems for risk stratification in melanoma. The AI assessment used in this study provides a significant risk estimate in melanoma, but it does not considerably improve the existing risk classification based on the TNM classification.

Development and validation of a nomogram incorporating gene expression profiling and clinical factors for accurate prediction of metastasis in patients with cutaneous melanoma following Mohs micrographic surgery

BACKGROUND

There is a need to improve prognostic accuracy for patients with cutaneous melanoma. A 31-gene expression profile (31-GEP) test uses the molecular biology of primary tumors to identify individual patient metastatic risk.

OBJECTIVE

Develop a nomogram incorporating 31-GEP with relevant clinical factors to improve prognostic accuracy.

METHODS

In an IRB-approved study, 1124 patients from 9 Mohs micrographic surgery centers were prospectively enrolled, treated with Mohs micrographic surgery, and underwent 31-GEP testing. Data from 684 of those patients with at least 1-year follow-up or a metastatic event were included in nomogram development to predict metastatic risk.

RESULTS

Logistic regression modeling of 31-GEP results and T stage provided the simplest nomogram with the lowest Bayesian information criteria score. Validation in an archival cohort (n = 901) demonstrated a significant linear correlation between observed and nomogram-predicted risk of metastasis. The resulting nomogram more accurately predicts the risk for cutaneous melanoma metastasis than T stage or 31-GEP alone.

LIMITATIONS

The patient population is representative of Mohs micrographic surgery centers. Sentinel lymph node biopsy was not performed for most patients and could not be used in the nomogram.

CONCLUSIONS

Integration of 31-GEP and T stage can gain clinically useful prognostic information from data obtained noninvasively.

Improved cutaneous melanoma survival stratification through integration of 31-gene expression profile testing with the American Joint Committee on Cancer 8th Edition Staging

Cutaneous melanoma (CM) survival is assessed using averaged data from the American Joint Committee on Cancer 8th edition (AJCC8). However, subsets of AJCC8 stages I-III have better or worse survival than the predicted average value. The objective of this study was to determine if the 31-gene expression profile (31-GEP) test for CM can further risk-stratify melanoma-specific mortality within each AJCC8 stage. This retrospective multicenter study of 901 archival CM samples obtained from patients with stages I-III CM assessed 31-GEP test predictions of 5-year melanoma-specific survival (MSS) using Kaplan-Meier and Cox proportional hazards. In stage I-III CM population, patients with a Class 2B result had a lower 5-year MSS (77.8%) than patients with a Class 1A result (98.7%) and log-rank testing demonstrated significant stratification of MSS [χ2 (2df, n = 901) = 99.7, P < 0.001). Within each stage, 31-GEP data provided additional risk stratification, including in stage I [χ2 (2df, n = 415) = 11.3, P = 0.004]. Cox regression multivariable analysis showed that the 31-GEP test was a significant predictor of melanoma-specific mortality (MSM) in patients with stage I-III CM [hazard ratio: 6.44 (95% confidence interval: 2.61-15.85), P < 0.001]. This retrospective study focuses on Class 1A versus Class 2B results. Intermediate results (Class 1B/2A) comprised 21.6% of cases with survival rates between Class 1A and 2B, and similar to 5-year MSS AJCC stage values. Data from the 31-GEP test significantly differentiates MSM into lower (Class 1A) and higher risk (Class 2B) groups within each AJCC8 stage. Incorporating 31-GEP results into AJCC8 survival calculations has the potential to more precisely assess survival and enhance management guidance.

Expert Consensus on the Use of Prognostic Gene Expression Profiling Tests for the Management of Cutaneous Melanoma: Consensus from the Skin Cancer Prevention Working Group

BACKGROUND

Prognostic assessment of cutaneous melanoma relies on historical, clinicopathological, and phenotypic risk factors according to American Joint Committee on Cancer(AJCC) and National Comprehensive Cancer Network (NCCN) guidelines but may not account for a patient's individual additional genetic risk factors.

OBJECTIVE

To review the available literature regarding commercially available gene expression profile (GEP) tests and their use in the management of cutaneous melanoma.

METHODS

A literature search was conducted for original, English-language studies or meta-analyses published between 2010 and 2021 on commercially available GEP tests in cutaneous melanoma prognosis, clinical decision-making regarding sentinel lymph node biopsy, and real-world efficacy. After the literature review, the Skin Cancer Prevention Working Group, an expert panel of dermatologists with specialized training in melanoma and non-melanoma skin cancer diagnosis and management, utilized a modified Delphi technique to develop consensus statements regarding prognostic gene expression profile tests. Statements were only adopted with a supermajority vote of > 80%.

RESULTS

The initial search identified 1064 studies/meta-analyses that met the search criteria. Of these, we included 21 original articles and meta-analyses that studied the 31-GEP test (DecisionDx-Melanoma; Castle Biosciences, Inc.), five original articles that studied the 11-GEP test (Melagenix; NeraCare GmbH), and four original articles that studied the 8-GEP test with clinicopathological factors (Merlin; 8-GEP + CP; SkylineDx B.V.) in this review. Six statements received supermajority approval and were adopted by the panel.

CONCLUSION

GEP tests provide additional, reproducible information for dermatologists to consider within the larger framework of the eighth edition of the AJCC and NCCN cutaneous melanoma guidelines when counseling regarding prognosis and when considering a sentinel lymph node biopsy.

Using a 31-Gene Expression Profile Test to Stratify Patients with Stage I-II Cutaneous Melanoma According to Recurrence Risk: Update to a Prospective, Multicenter Study

Simple Summary

Many people with skin cancer will have their cancer come back. The 31-gene expression profile (31-GEP) test can help predict if a cancer has a low (Class 1) or high (Class 2) chance of returning. This study looked at 86 patients with early skin cancer to see how well the 31-GEP test predicted if their cancer would return. None of the patients with a Class 1 GEP result had their cancer return within 3 years, but one-fourth of patients with a Class 2 result did. This study showed that the 31-GEP test can help predict if a patient’s skin cancer will return. Accurate risk prediction can help doctors make better treatment plans for patients with skin cancer.

Abstract

Background: Fifteen to forty percent of patients with localized cutaneous melanoma (CM) (stages I–II) will experience disease relapse. The 31-gene expression profile (31-GEP) uses gene expression data from the primary tumor in conjunction with clinicopathologic features to refine patient prognosis. The study’s objective was to evaluate 31-GEP risk stratification for disease-free survival (DFS) in a previously published cohort with longer follow-up. Methods: Patients with stage IB–II CM (n = 86) were prospectively tested with the 31-GEP. Follow-up time increased from 2.2 to 3.9 years. Patient outcomes were compared using Kaplan-Meier and Cox regression analysis. Results: A Class 2B result was a significant predictor of 3-year DFS (hazard ratio (HR) 8.4, p = 0.008) in univariate analysis. The 31-GEP significantly stratified patients by risk of relapse (p = 0.005). A Class 2B result was associated with a lower 3-year DFS (75.0%) than a Class 1A result (100%). The 31-GEP had a high sensitivity (77.8%) and negative predictive value (95.0%). Conclusions: The 31-GEP is a significant predictor of disease relapse in patients with stage IB–II melanoma and accurately stratified patients by risk of relapse.

Revisiting miRNA Association with Melanoma Recurrence and Metastasis from a Machine Learning Point of View

The diagnostic and prognostic value of miRNAs in cutaneous melanoma (CM) has been broadly studied and supported by advanced bioinformatics tools. From early studies using miRNA arrays with several limitations, to the recent NGS-derived miRNA expression profiles, an accurate diagnostic panel of a comprehensive pre-specified set of miRNAs that could aid timely identification of specific cancer stages is still elusive, mainly because of the heterogeneity of the approaches and the samples. Herein, we summarize the existing studies that report several miRNAs as important diagnostic and prognostic biomarkers in CM. Using publicly available NGS data, we analyzed the correlation of specific miRNA expression profiles with the expression signatures of known gene targets. Combining network analytics with machine learning, we developed specific non-linear classification models that could successfully predict CM recurrence and metastasis, based on two newly identified miRNA signatures. Subsequent unbiased analyses and independent test sets (i.e., a dataset not used for training, as a validation cohort) using our prediction models resulted in 73.85% and 82.09% accuracy in predicting CM recurrence and metastasis, respectively. Overall, our approach combines detailed analysis of miRNA profiles with heuristic optimization and machine learning, which facilitates dimensionality reduction and optimization of the prediction models. Our approach provides an improved prediction strategy that could serve as an auxiliary tool towards precision treatment.

Role of Biomarkers in the Integrated Management of Melanoma

Melanoma, which is an aggressive skin cancer, is currently the fifth and seventh most common cancer in men and women, respectively. The American Cancer Society reported that approximately 106,110 new cases of melanoma were diagnosed in the United States in 2021, with 7,180 people dying from the disease. This information could facilitate the early detection of possible metastatic lesions and the development of novel therapeutic techniques for melanoma. Additionally, early detection of malignant melanoma remains an objective of melanoma research. Recently, melanoma treatment has substantially improved, given the availability of targeted treatments and immunotherapy. These developments have highlighted the significance of identifying biomarkers for prognosis and predicting therapy response. Biomarkers included tissue protein expression, circulating DNA detection, and genetic alterations in cancer cells. Improved diagnostic and prognostic biomarkers are becoming increasingly relevant in melanoma treatment, with the development of newer and more targeted treatments. Here, the author discusses the aspects of biomarkers in the real-time management of patients with melanoma.

Breslow thickness 2.0: Why gene expression profiling is a step toward better patient selection for sentinel lymph node biopsies

Risk-stratification of cutaneous melanoma is important. Patients want to know what to expect after diagnosis, and physicians need to decide on a treatment plan. Historically, melanoma that had spread beyond the skin and regional lymph nodes was largely incurable, and the only approach to preventing a bad outcome was surgery. Through the seminal work of Alexander Breslow and Donald Morton, a system was devised to carefully escalate surgery based on primary tumor thickness and sentinel lymph node status. Today, we know that prophylactic lymph node dissections do not improve survival, but we continue to appreciate the prognostic implications of a positive sentinel node and the benefits of removing nodal metastases, which facilitates locoregional disease control. However, the question arises whether we can better select patients for sentinel lymph node biopsies (SLNB) as, currently, 85% of these procedures are negative and non-therapeutic. Here, we argue that gene expression profiling (GEP) of the diagnostic biopsy is a valuable step toward better patient selection when combined with reliable clinicopathologic (CP) information such as patient age and Breslow thickness. Recently, a CP-GEP-based classifier of nodal metastasis risk, the Merlin Assay, has become commercially available. While CP-GEP is still being validated in prospective studies, preliminary data suggest that it is an independent predictor of nodal metastasis, outperforming clinicopathological variables. The hunt is on for Breslow thickness 2.0.

Sentinel Lymph Node Biopsy Positivity in Patients With Acral Lentiginous and Other Subtypes of Cutaneous Melanoma

IMPORTANCE

Acral lentiginous melanoma (ALM) is a rare subtype of malignant melanoma typically occurring on the palmar and plantar surfaces. Although it has distinctive genetic, prognostic, and behavioral characteristics relative to cutaneous melanomas overall, owing to its rarity, treatment is largely guided by data extrapolated from more common subtypes. Although sentinel lymph node (SLN) status has been shown to be a significant prognostic factor for ALM, the independent effect of ALM-subtype disease on the likelihood of SLN positivity and the stage-specific positivity rates for ALM are not well characterized.

OBJECTIVE

To evaluate the association of ALM with SLN status as well as to characterize the clinical stage-specific rates of SLN positivity for ALM based on the AJCC Cancer Staging Manual, 8th edition (AJCC-8).

DESIGN, SETTING, AND PARTICIPANTS

The National Cancer Database (NCDB) includes all reportable cases from Commission on Cancer accredited facilities and represents approximately 50% of all newly diagnosed melanoma cases in the US. This retrospective cohort study included cases of AJCC-8 clinical stage I to II melanomas from the NCDB diagnosed from 2012 to 2015. The analysis took place between April 2021 and September 2021.

EXPOSURES

Melanoma histopathologic subtype.

MAIN OUTCOMES AND MEASURES

Sentinel lymph node status.

RESULTS

We identified 60 148 patients with malignant melanomas, 959 of whom had ALM-subtype disease. Among patients in the cohort, 25 550 (42.5%) were women and the mean (SD) age was 64 (16) years. Multivariable logistic regression controlling for demographic and histopathologic characteristics revealed that ALM was independently associated with the highest risk for SLN positivity among included subtypes (vs superficial spreading melanoma: odds ratio, 1.91; 95% CI, 1.59-2.28). Subgroup analysis by AJCC clinical stage demonstrated that ALM was independently associated with the highest risk for SLN positivity for both stage IB and II disease. The rate of SLN positivity for patients with stage IB and II ALM was 18.39% (95% CI, 13.82%-24.03%) and 39.53% (34.98%-44.26%), respectively.

CONCLUSIONS AND RELEVANCE

In this cohort study ALM was independently associated with SLN positivity and had relatively high positivity rates at clinical stage IB and II. This suggests that SLNB should be encouraged for all patients with clinical stage IB and II ALM, and such patients should receive appropriate counseling about the higher regional metastatic risk of their cancers. Future work with a larger cohort is required to elucidate the risk of SLN positivity for stage IA ALM.

Clinical Utility of the 40-Gene Expression Profile (40-GEP) Test for Improved Patient Management Decisions and Disease-Related Outcomes when Combined with Current Clinicopathological Risk Factors for Cutaneous Squamous Cell Carcinoma (cSCC): Case Series

Introduction

While improvements have been made to risk assessment of cutaneous squamous cell carcinoma (cSCC) patients, there is a critical need for a uniform and more precise stratification system of their care. To address this unmet clinical need, a prognostic 40-gene expression profile (40-GEP) test has recently been developed and independently validated to show improved stratification of metastatic risk in high-risk cSCC patients compared with current staging systems.

Methods

Two cSCC cases, both male with similar patient profiles and the same staging status across two different staging systems, yet with opposing outcomes, were chosen for retrospective review of their primary biopsy using the 40-GEP test.

Results

Case 1 declined further treatment, even when presented with evidence of a small focus of cSCC found in the last layer of nonmarginal tissue obtained from Mohs micrographic surgery (MMS). Case 1 remained recurrence free, and retrospective analysis of the initial biopsy with the 40-GEP test provided a Class 1 result (low likelihood of metastasis). Case 2, even with subsequent clearing of the primary cSCC with MMS, noted another metastatic cSCC 3 months later. Case 2, after multimodal adjuvant treatments, died due to disease progression. Retrospective analysis of the initial biopsy with the 40-GEP test provided a Class 2B result (high likelihood of metastasis).

Conclusions

The cases discussed highlight the utility in 40-GEP to provide additional information to guide treatment decisions and improve outcomes. Integrating novel molecular prognostication with traditional clinicopathological risk factors can improve stratification of high-risk cSCC patients and may inform selection of risk-appropriate treatment and surveillance strategies.

Impact of Next-generation Sequencing on Interobserver Agreement and Diagnosis of Spitzoid Neoplasms

Atypical Spitzoid melanocytic tumors are diagnostically challenging. Many studies have suggested various genomic markers to improve classification and prognostication. We aimed to assess whether next-generation sequencing studies using the Tempus xO assay assessing mutations in 1711 cancer-related genes and performing whole transcriptome mRNA sequencing for structural alterations could improve diagnostic agreement and accuracy in assessing neoplasms with Spitzoid histologic features. Twenty expert pathologists were asked to review 70 consultation level cases with Spitzoid features, once with limited clinical information and again with additional genomic information. There was an improvement in overall agreement with additional genomic information. Most significantly, there was increase in agreement of the diagnosis of conventional melanoma from moderate (κ=0.470, SE=0.0105) to substantial (κ=0.645, SE=0.0143) as measured by an average Cohen κ. Clinical follow-up was available in all 70 cases which substantiated that the improved agreement was clinically significant. Among 3 patients with distant metastatic disease, there was a highly significant increase in diagnostic recognition of the cases as conventional melanoma with genomics (P<0.005). In one case, none of 20 pathologists recognized a tumor with BRAF and TERT promoter mutations associated with fatal outcome as a conventional melanoma when only limited clinical information was provided, whereas 60% of pathologists correctly diagnosed this case when genomic information was also available. There was also a significant improvement in agreement of which lesions should be classified in the Spitz category/WHO Pathway from an average Cohen κ of 0.360 (SE=0.00921) to 0.607 (SE=0.0232) with genomics.

Perturbation of BRMS1 interactome reveals pathways that impact metastasis

Breast Cancer Metastasis Suppressor 1 (BRMS1) expression is associated with longer patient survival in multiple cancer types. Understanding BRMS1 functionality will provide insights into both mechanism of action and will enhance potential therapeutic development. In this study, we confirmed that the C-terminus of BRMS1 is critical for metastasis suppression and hypothesized that critical protein interactions in this region would explain its function. Phosphorylation status at S237 regulates BRMS1 protein interactions related to a variety of biological processes, phenotypes [cell cycle (e.g., CDKN2A), DNA repair (e.g., BRCA1)], and metastasis [(e.g., TCF2 and POLE2)]. Presence of S237 also directly decreased MDA-MB-231 breast carcinoma migration in vitro and metastases in vivo. The results add significantly to our understanding of how BRMS1 interactions with Sin3/HDAC complexes regulate metastasis and expand insights into BRMS1's molecular role, as they demonstrate BRMS1 C-terminus involvement in distinct protein-protein interactions.

Bioinformatic and Machine Learning Applications in Melanoma Risk Assessment and Prognosis: A Literature Review

Over 100,000 people are diagnosed with cutaneous melanoma each year in the United States. Despite recent advancements in metastatic melanoma treatment, such as immunotherapy, there are still over 7000 melanoma-related deaths each year. Melanoma is a highly heterogenous disease, and many underlying genetic drivers have been identified since the introduction of next-generation sequencing. Despite clinical staging guidelines, the prognosis of metastatic melanoma is variable and difficult to predict. Bioinformatic and machine learning analyses relying on genetic, clinical, and histopathologic inputs have been increasingly used to risk stratify melanoma patients with high accuracy. This literature review summarizes the key genetic drivers of melanoma and recent applications of bioinformatic and machine learning models in the risk stratification of melanoma patients. A robustly validated risk stratification tool can potentially guide the physician management of melanoma patients and ultimately improve patient outcomes.

More sentinel lymph node biopsies for thin melanomas after transition to AJCC 8th edition do not increase positivity rate: A Danish population-based study of 7148 patients

BACKGROUND

We evaluated the outcome of sentinel lymph node biopsies (SLNB) in patients with thin melanoma before and after the implementation of AJCC 8th edition (AJCC8) and identified predictors of positive sentinel lymph nodes (+SLN).

METHODS

Patients diagnosed with T1 melanomas (Breslow thickness ≤1 mm) during 2016-2017 as per AJCC 7th edition (AJCC7) (n = 3414) and 2018-2019 as per AJCC8 (n = 3734) were identified in the Danish Melanoma Database.

RESULTS

More SLNBs were performed in the AJCC8 cohort compared to the AJCC7 (22.2% vs. 16.2%, p < 0.001), with no significant difference in +SLN rates (4.7% vs. 6.7%, p = 0.118). In the AJCC7 + SLN subgroup, no melanomas were ulcerated, 94.6% had mitotic rate (MR) ≥ 1, 67.6% were ≥0.8 mm and 32.4% would be T1a according to AJCC8. In the AJCC8 + SLN subgroup, 10.3% were ulcerated, 74.4% had MR≥ 1, 97.4% were ≥0.8 mm and 23.1% would be T1a according to AJCC7. On multivariable analysis younger age and MR ≥ 1 were significant predictors of +SLN.

CONCLUSION

More SLNBs were performed in T1 melanomas after transition to AJCC8 without an increase in +SLN rate. None of the AJCC8 T1b criteria were significant predictors of +SLN. We suggest that mitosis and younger age should be considered as indications for SLNB in thin melanoma.

Low expression of endoplasmic reticulum stress-related gene SERP1 is associated with poor prognosis and immune infiltration in skin cutaneous melanoma

Stress-associated endoplasmic reticulum protein 1 (SERP1) is a gene induced by endoplasmic reticulum (ER) stress and a major contributor to multiple tumor types. Skin cutaneous melanoma (SKCM) is a highly aggressive and fatal cancer with poor treatment outcomes after progression. In this study, we evaluated SERP1's role in tumorigenesis, prognosis, and immune infiltration in SKCM. Patients with SKCM had low SERP1 expression. We identified differentially expressed genes between high- and low-SERP1 expression groups and conducted functional, pathway, and gene enrichment analyses. Protein-protein (PPI) and gene-gene interaction (GGI) networks were constructed via STRING and GeneMANIA, respectively. SERP1 mutation information was obtained through cBioPortal; location in the skin was identified through the Human Protein Atlas. Kaplan-Meier analysis revealed an association between low SERP1 expression and overall survival (OS), disease-specific survival (DSS), progress-free interval (PFI) rates, and worse prognosis in patients with multiple clinicopathological features. Cox regression analysis and nomograms further presented SERP1 level as an independent prognostic factor for patients with SKCM. Furthermore, there were significant correlations between SERP1 expression and immune infiltrates; thus, low SERP1 expression is associated with immune cell infiltration and can be considered a poor prognostic biomarker in patients with SKCM. Stress-associated endoplasmic reticulum protein 1 (SERP1) is a gene induced by endoplasmic reticulum (ER) stress and a major contributor to multiple tumor types. Skin cutaneous melanoma (SKCM) is a highly aggressive and fatal cancer with poor treatment outcomes after progression. In this study, we evaluated SERP1's role in tumorigenesis, prognosis, and immune infiltration in SKCM. Patients with SKCM had low SERP1 expression. We identified differentially expressed genes between high- and low-SERP1 expression groups and conducted functional, pathway, and gene enrichment analyses. Protein-protein (PPI) and gene-gene interaction (GGI) networks were constructed via STRING and GeneMANIA, respectively. SERP1 mutation information was obtained through cBioPortal; location in the skin were identified through the Human Protein Atlas. Kaplan-Meier analysis revealed an association between low SERP1 expression and overall survival (OS), disease-specific survival (DSS), progress-free interval (PFI) rates, and worse prognosis in patients with multiple clinicopathological features. Cox regression analysis and nomograms further presented SERP1 level as an independent prognostic factor for patients with SKCM. Furthermore, there were significant correlations between SERP1 expression and immune infiltrates; thus, low SERP1 expression is associated with immune cell infiltration and can be considered a poor prognostic biomarker in patients with SKCM.