Artificial intelligence-assisted metastasis and prognosis model for patients with nodular melanoma

OBJECTIVE

The objective of this study was to identify the risk factors that influence metastasis and prognosis in patients with nodular melanoma (NM), as well as to develop and validate a prognostic model using artificial intelligence (AI) algorithms.

METHODS

The Surveillance, Epidemiology, and End Results (SEER) database was queried for 4,727 patients with NM based on the inclusion/exclusion criteria. Their clinicopathological characteristics were retrospectively reviewed, and logistic regression analysis was utilized to identify risk factors for metastasis. This was followed by employing Multilayer Perceptron (MLP), Adaptive Boosting (AB), Bagging (BAG), logistic regression (LR), Gradient Boosting Machine (GBM), and eXtreme Gradient Boosting (XGB) algorithms to develop metastasis models. The performance of the six models was evaluated and compared, leading to the selection and visualization of the optimal model. Through integrating the prognostic factors of Cox regression analysis with the optimal models, the prognostic prediction model was constructed, validated, and assessed.

RESULTS

Logistic regression analyses identified that marital status, gender, primary site, surgery, radiation, chemotherapy, system management, and N stage were all independent risk factors for NM metastasis. MLP emerged as the optimal model among the six models (AUC = 0.932, F1 = 0.855, Accuracy = 0.856, Sensitivity = 0.878), and the corresponding network calculator (https://shimunana-nm-distant-m-nm-m-distant-8z8k54.streamlit.app/) was developed. The following were examined as independent prognostic factors: MLP, age, marital status, sequence number, laterality, surgery, radiation, chemotherapy, system management, T stage, and N stage. System management and surgery emerged as protective factors (HR < 1). To predict 1-, 3-, and 5-year overall survival (OS), a nomogram was created. The validation results demonstrated that the model exhibited good discrimination and consistency, as well as high clinical usefulness.

CONCLUSION

The developed prediction model more effectively reflects the prognosis of patients with NM and differentiates between the risk level of patients, serving as a useful supplement to the classical American Joint Committee on Cancer (AJCC) staging system and offering a reference for clinically stratified individualized treatment and prognosis prediction. Furthermore, the model enables clinicians to quantify the risk of metastasis in NM patients, assess patient survival, and administer precise treatments.

Development and validation of a novel model to predict recurrence-free survival and melanoma-specific survival after sentinel lymph node biopsy in patients with melanoma: an international, retrospective, multicentre analysis

BACKGROUND

The introduction of adjuvant systemic treatment for patients with high-risk melanomas necessitates accurate staging of disease. However, inconsistencies in outcomes exist between disease stages as defined by the American Joint Committee on Cancer (8th edition). We aimed to develop a tool to predict patient-specific outcomes in people with melanoma rather than grouping patients according to disease stage.

METHODS

Patients older than 13 years with confirmed primary melanoma who underwent sentinel lymph node biopsy (SLNB) between Oct 29, 1997, and Nov 11, 2013, at four European melanoma centres (based in Berlin, Germany; Amsterdam and Rotterdam, the Netherlands; and Warsaw, Poland) were included in the development cohort. Potential predictors of recurrence-free and melanoma-specific survival assessed were sex, age, presence of ulceration, primary tumour location, histological subtype, Breslow thickness, sentinel node status, number of sentinel nodes removed, maximum diameter of the largest sentinel node metastasis, and Dewar classification. A prognostic model and nomogram were developed to predict 5-year recurrence-free survival on a continuous scale in patients with stage pT1b or higher melanomas. This model was also calibrated to predict melanoma-specific survival. Model performance was assessed by discrimination (area under the time-dependent receiver operating characteristics curve [AUC]) and calibration. External validation was done in a cohort of patients with primary melanomas who underwent SLNB between Jan 30, 1997, and Dec 12, 2013, at the Melanoma Institute Australia (Sydney, NSW, Australia).

FINDINGS

The development cohort consisted of 4071 patients, of whom 2075 (51%) were female and 1996 (49%) were male. 889 (22%) had sentinel node-positive disease and 3182 (78%) had sentinel node-negative disease. The validation cohort comprised 4822 patients, of whom 1965 (41%) were female and 2857 (59%) were male. 891 (18%) had sentinel node-positive disease and 3931 (82%) had sentinel node-negative disease. Median follow-up was 4·8 years (IQR 2·3-7·8) in the development cohort and 5·0 years (2·2-8·9) in the validation cohort. In the development cohort, 5-year recurrence-free survival was 73·5% (95% CI 72·0-75·1) and 5-year melanoma-specific survival was 86·5% (85·3-87·8). In the validation cohort, the corresponding estimates were 66·1% (64·6-67·7) and 83·3% (82·0-84·6), respectively. The final model contained six prognostic factors: sentinel node status, Breslow thickness, presence of ulceration, age at SLNB, primary tumour location, and maximum diameter of the largest sentinel node metastasis. In the development cohort, for the model's prediction of recurrence-free survival, the AUC was 0·80 (95% CI 0·78-0·81); for prediction of melanoma-specific survival, the AUC was 0·81 (0·79-0·84). External validation showed good calibration for both outcomes, with AUCs of 0·73 (0·71-0·75) and 0·76 (0·74-0·78), respectively.

INTERPRETATION

Our prediction model and nomogram accurately predicted patient-specific risk probabilities for 5-year recurrence-free and melanoma-specific survival. These tools could have important implications for clinical decision making when considering adjuvant treatments in patients with high-risk melanomas.

FUNDING

Erasmus Medical Centre Cancer Institute.

Population-based prognostic instrument (SweMR 2.0) for melanoma-specific survival - An ideal tool for individualised treatment decisions for Swedish patients

INTRODUCTION

The prognosis for patients with melanoma has improved due to better treatments in recent years and updated tools to accurately predict an individual's risk are warranted. This study aims to describe a prognostic instrument for patients with cutaneous melanoma and its potential as a clinical device for treatment decisions.

METHODS

Patients with localised invasive cutaneous melanoma diagnosed in 1990-2021 with data on tumour thickness were identified from the population-based Swedish Melanoma Registry. The parametric Royston-Parmar (RP) method was used to estimate melanoma-specific survival (MSS) probabilities. Separate models were constructed for patients (≤1 mm) and (>1 mm) and prognostic groups were created based on all combinations of age, sex, tumour site, tumour thickness, absence/presence of ulceration, histopathologic type, Clark's level of invasion, mitoses and sentinel lymph node (SLN) status.

RESULTS

In total, 72 616 patients were identified, 41 764 with melanoma ≤1 mm and 30 852 with melanoma >1 mm. The most important variable was tumour thickness for both (≤1 mm) and (>1 mm), that explained more than 50% of the survival. The second most important variables were mitoses (≤1 mm) and SLN status (>1 mm). The prognostic instrument successfully created probabilities for >30 000 prognostic groups.

CONCLUSIONS

The Swedish updated population-based prognostic instrument, predicts MSS survival up to 10 years after diagnosis. The prognostic instrument gives more representative and up-to-date prognostic information for Swedish patients with primary melanoma than the present AJCC staging. Additional to clinical use and the adjuvant setting, the information retrieved could be used to plan future studies.

Predictive accuracy of risk prediction models for recurrence, metastasis and survival for early-stage cutaneous melanoma: a systematic review

OBJECTIVES

To identify prognostic models for melanoma survival, recurrence and metastasis among American Joint Committee on Cancer stage I and II patients postsurgery; and evaluate model performance, including overall survival (OS) prediction.

DESIGN

Systematic review and narrative synthesis.

DATA SOURCES

Searched MEDLINE, Embase, CINAHL, Cochrane Library, Science Citation Index and grey literature sources including cancer and guideline websites from 2000 to September 2021.

ELIGIBILITY CRITERIA

Included studies on risk prediction models for stage I and II melanoma in adults ≥18 years. Outcomes included OS, recurrence, metastases and model performance. No language or country of publication restrictions were applied.

DATA EXTRACTION AND SYNTHESIS

Two pairs of reviewers independently screened studies, extracted data and assessed the risk of bias using the CHecklist for critical Appraisal and data extraction for systematic Reviews of prediction Modelling Studies checklist and the Prediction study Risk of Bias Assessment Tool. Heterogeneous predictors prevented statistical synthesis.

RESULTS

From 28 967 records, 15 studies reporting 20 models were included; 8 (stage I), 2 (stage II), 7 (stages I-II) and 7 (stages not reported), but were clearly applicable to early stages. Clinicopathological predictors per model ranged from 3-10. The most common were: ulceration, Breslow thickness/depth, sociodemographic status and site. Where reported, discriminatory values were ≥0.7. Calibration measures showed good matches between predicted and observed rates. None of the studies assessed clinical usefulness of the models. Risk of bias was high in eight models, unclear in nine and low in three. Seven models were internally and externally cross-validated, six models were externally validated and eight models were internally validated.

CONCLUSIONS

All models are effective in their predictive performance, however the low quality of the evidence raises concern as to whether current follow-up recommendations following surgical treatment is adequate. Future models should incorporate biomarkers for improved accuracy.

PROSPERO REGISTRATION NUMBER

CRD42018086784.

Identification of a Seven-Differentially Expressed Gene-Based Recurrence-Free Survival Model for Melanoma Patients

Melanoma is a malignant tumor that originates in melanocytes of the skin or mucous membrane, which has a high mortality rate and worse prognosis. Therefore, perspective prognosis evaluation seems more important for patients' treatment. Gene expression profiles of melanoma were downloaded from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) databases, respectively. 130 consistent differentially expressed genes (DEGs) were identified between melanoma and nevus tissues from two GEO cohorts. Prognostic genes were identified by univariate analysis, and 20 of them were regarded to be associated with the recurrence-free survival (RFS) of melanoma patients. Then, the LASSO Cox regression analysis chose seven of them to establish a seven-DEG-based RFS predicting signature. We demonstrated that this model was more powerful to predict RFS risk than other individual clinical features and was able to independently predict the RFS outcomes in different subsets of patients. We attempted to search for the underlying mechanisms by analyzing the coexpression genes of the seven candidates, and the pathway enrichment analyses indicated that immune response-related pathways might play a critical role in melanoma progression. Finally, we establish a robust seven-DEG-based RFS predicting signature, which will facilitate the personalized treatment of melanoma patients.

Detection of cancer metastasis: past, present and future

The clinical importance of metastatic spread of cancer has been recognized for centuries, and melanoma has loomed large in historical descriptions of metastases, as well as the numerous mechanistic theories espoused. The "fatal black tumor" described by Hippocrates in 5000 BC that was later termed "melanose" by Rene Laennec in 1804 was recognized to have the propensity to metastasize by William Norris in 1820. And while the prognosis of melanoma was uniformly acknowledged to be dire, Samuel Cooper described surgical removal as having the potential to improve prognosis. Subsequent to this, in 1898 Herbert Snow was the first to recognize the potential clinical benefit of removing clinically normal lymph nodes at the time of initial cancer surgery. In describing "anticipatory gland excision," he noted that "it is essential to remove, whenever possible, those lymph glands which first receive the infective protoplasm, and bar its entrance into the blood, before they have undergone increase in bulk". This revolutionary concept marked the beginning of a debate that rages today: are regional lymph nodes the first stop for metastases ("incubator" hypothesis) or does their involvement serve as an indicator of aggressive disease with inherent metastatic potential ("marker" hypothesis). Is there a better way to improve prediction of disease outcome? This article attempts to address some of the resultant questions that were the subject of the session "Novel Frontiers in the Diagnosis of Cancer" at the 8th International Congress on Cancer Metastases, held in San Francisco, CA in October 2019. Some of these questions addressed include the significance of sentinel node metastasis in melanoma, and the optimal method for their pathologic analysis. The finding of circulating tumor cells in the blood may potentially supplant surgical techniques for detection of metastatic disease, and we are beginning to perfect techniques for their detection, understand how to apply the findings clinically, and develop clinical followup treatment algorithms based on these results. Finally, we will discuss the revolutionary field of machine learning and its applications in cancer diagnosis. Computer-based learning algorithms have the potential to improve efficiency and diagnostic accuracy of pathology, and can be used to develop novel predictors of prognosis, but significant challenges remain. This review will thus encompass latest concepts in the detection of cancer metastasis via the lymphatic system, the circulatory system, and the role of computers in enhancing our knowledge in this field.

Stage IIIa Melanoma and Impact of Multiple Positive Lymph Nodes on Survival

BACKGROUND

For patients with cutaneous melanoma, having >1 positive lymph node (LN) is associated with worse survival. We hypothesized that for stage IIIA patients, N2a disease (2 to 3 positive LN) would be associated with a worse prognosis compared to those with N1a disease (1 positive LN).

STUDY DESIGN

Stage IIIA melanoma patients in the NCDB Participant User File from 2010 to 2016 were analyzed. Overall survival (OS) between N1a and N2a patients was compared. Subgroup analyses were made between patients undergoing sentinel lymph node (SLN) biopsy alone and those undergoing subsequent completion lymph node dissection (CLND). A separate post hoc analysis of T2a patients undergoing SLN biopsy and CLND from a prospective multicenter randomized clinical trial was performed to validate the findings.

RESULTS

Records of 2,305 IIIA patients were evaluated. In an adjusted survival model, N2a disease was an independent risk factor for worse OS (hazard ratio [HR] 1.56, p = 0.0052). In the subgroup analysis, there was no difference in OS between N1a and N2a disease for patients who underwent SLN biopsy without CLND (p = 0.59), but there was a significant difference in OS for patients who underwent SLN biopsy plus CLND (p = 0.0009). The separate clinical trial database confirmed that for patients with SLN-only disease, there was no difference in OS between N1a and N2a disease.

CONCLUSIONS

For stage IIIA melanoma patients, the distribution of micrometastatic lymph node disease (SLN or non-SLN), rather than the absolute number of SLNs, should be considered when individualizing adjuvant therapy recommendations.

Circulating tumour DNA and melanoma survival: A systematic literature review and meta-analysis

We reviewed and meta-analysed the available evidence (until December 2019) about circulating tumour DNA (ctDNA) levels and melanoma patients survival. We included twenty-six studies (>2000 patients overall), which included mostly stage III-IV cutaneous melanoma patients and differed widely in terms of systemic therapy received and somatic mutations that were searched. Patients with detectable ctDNA before treatment had worse progression-free survival (PFS) (summary hazard ratio (SHR) 2.47, 95 % confidence intervals (CI) 1.85-3.29) and overall survival (OS) (SHR 2.98, 95 % CI 2.26-3.92), with no difference by tumour stage. ctDNA detectability during follow-up was associated with poorer PFS (SHR 4.27, 95 %CI 2.75-6.63) and OS (SHR 3.91, 95 %CI 1.97-7.78); in the latter case, the association was stronger (p = 0.01) for stage IV vs. III melanomas. Between-estimates heterogeneity was low for all pooled estimates. ctDNA is a strong prognostic biomarker for advanced-stage melanoma patients, robust across tumour (e.g. genomic profile) and patients (e.g. systemic therapy) characteristics.

Identification of stage I/IIA melanoma patients at high risk for disease relapse using a clinicopathologic and gene expression model

PURPOSE

Patients with stage I/IIA cutaneous melanoma (CM) are currently not eligible for adjuvant therapies despite uncertainty in relapse risk. Here, we studied the ability of a recently developed model which combines clinicopathologic and gene expression variables (CP-GEP) to identify stage I/IIA melanoma patients who have a high risk for disease relapse.

PATIENTS AND METHODS

Archival specimens from a cohort of 837 consecutive primary CMs were used for assessing the prognostic performance of CP-GEP. The CP-GEP model combines Breslow thickness and patient age, with the expression of eight genes in the primary tumour. Our specific patient group, represented by 580 stage I/IIA patients, was stratified based on their risk of relapse: CP-GEP High Risk and CP-GEP Low Risk. The main clinical end-point of this study was five-year relapse-free survival (RFS).

RESULTS

Within the stage I/IIA melanoma group, CP-GEP identified a high-risk patient group (47% of total stage I/IIA patients) which had a considerably worse five-year RFS than the low-risk patient group; 74% (95% confidence interval [CI]: 67%-80%) versus 89% (95% CI: 84%-93%); hazard ratio [HR] = 2.98 (95% CI: 1.78-4.98); P < 0.0001. Of patients in the high-risk group, those who relapsed were most likely to do so within the first 3 years.

CONCLUSION

The CP-GEP model can be used to identify stage I/IIA patients who have a high risk for disease relapse. These patients may benefit from adjuvant therapy.

Risk Prediction Models for Melanoma: A Systematic Review on the Heterogeneity in Model Development and Validation

The rising incidence of cutaneous melanoma over the past few decades has prompted substantial efforts to develop risk prediction models identifying people at high risk of developing melanoma to facilitate targeted screening programs. We review these models, regarding study characteristics, differences in risk factor selection and assessment, evaluation, and validation methods. Our systematic literature search revealed 40 studies comprising 46 different risk prediction models eligible for the review. Altogether, 35 different risk factors were part of the models with nevi being the most common one (n = 35, 78%); little consistency in other risk factors was observed. Results of an internal validation were reported for less than half of the studies (n = 18, 45%), and only 6 performed external validation. In terms of model performance, 29 studies assessed the discriminative ability of their models; other performance measures, e.g., regarding calibration or clinical usefulness, were rarely reported. Due to the substantial heterogeneity in risk factor selection and assessment as well as methodologic aspects of model development, direct comparisons between models are hardly possible. Uniform methodologic standards for the development and validation of risk prediction models for melanoma and reporting standards for the accompanying publications are necessary and need to be obligatory for that reason.

Survival Differences Between Pediatric Head and Neck Versus Body Melanoma in the Surveillance, Epidemiology, and End Results Program

OBJECTIVES/HYPOTHESIS

To review the demographics, treatment, and survival of pediatric melanoma of the head and neck and to determine if melanoma of the head and neck has worse survival than melanoma of other body sites.

STUDY DESIGN

Retrospective database review.

METHODS

Pediatric patients from 0 to 21 years in the Surveillance, Epidemiology, and End Results 18 registries database were included from 1975 to 2016 based on a diagnosis of melanoma of the skin using the primary site International Classification of Diseases for Oncology, Third Edition codes from C44.0-C44.9.skin of lip, C44.1-eyelid, C44.2-external ear, C44.3-skin other/unspecified parts of face, C44.4-skin of scalp and neck, C44.5-skin of trunk, C44.6-skin of upper limb and shoulder, C44.7-skin of lower limb and hip, C44.8-overlapping lesion of skin, and C44.9-skin, NOS (not otherwise specified).

RESULTS

A total of 4,561 pediatric melanomas of the skin were identified. There were 854 (18.7%) cases of melanoma of the head and neck (MHN) and 3,707 (81.3%) cases of melanoma of the body (MOB). The hazard ratio for MHN versus MOB was 1.6 (95% confidence interval: 1.3-2.1) after accounting for sex, race, and age. Of MHN sites, the hazard ratio for melanoma of the scalp and neck was 2.2 (1.1-4.7). The 2- and 5-year Kaplan-Meier overall survival for MHN were 94.6% and 90.7%, respectively, compared with 96.6% and 94.7%, respectively, for MOB (P < .01).

CONCLUSIONS

Survival outcomes of pediatric melanoma are notably related to anatomic site. Children with melanoma of the scalp and neck have the worst survival of all sites. Additionally, children who are older/white/male are at greater risk for worse survival outcomes.

LEVEL OF EVIDENCE

3 Laryngoscope, 131:E635-E641, 2021.

Model Combining Tumor Molecular and Clinicopathologic Risk Factors Predicts Sentinel Lymph Node Metastasis in Primary Cutaneous Melanoma

PURPOSE

More than 80% of patients who undergo sentinel lymph node (SLN) biopsy have no nodal metastasis. Here we describe a model that combines clinicopathologic and molecular variables to identify patients with thin and intermediate thickness melanomas who may forgo the SLN biopsy procedure due to their low risk of nodal metastasis.

PATIENTS AND METHODS

Genes with functional roles in melanoma metastasis were discovered by analysis of next generation sequencing data and case control studies. We then used PCR to quantify gene expression in diagnostic biopsy tissue across a prospectively designed archival cohort of 754 consecutive thin and intermediate thickness primary cutaneous melanomas. Outcome of interest was SLN biopsy metastasis within 90 days of melanoma diagnosis. A penalized maximum likelihood estimation algorithm was used to train logistic regression models in a repeated cross validation scheme to predict the presence of SLN metastasis from molecular, clinical and histologic variables.

RESULTS

Expression of genes with roles in epithelial-to-mesenchymal transition (glia derived nexin, growth differentiation factor 15, integrin β3, interleukin 8, lysyl oxidase homolog 4, TGFβ receptor type 1 and tissue-type plasminogen activator) and melanosome function (melanoma antigen recognized by T cells 1) were associated with SLN metastasis. The predictive ability of a model that only considered clinicopathologic or gene expression variables was outperformed by a model which included molecular variables in combination with the clinicopathologic predictors Breslow thickness and patient age; AUC, 0.82; 95% CI, 0.78-0.86; SLN biopsy reduction rate of 42% at a negative predictive value of 96%.

CONCLUSION

A combined model including clinicopathologic and gene expression variables improved the identification of melanoma patients who may forgo the SLN biopsy procedure due to their low risk of nodal metastasis.

A Population-Based Comparison of the AJCC 7th and AJCC 8th Editions for Patients Diagnosed with Stage III Cutaneous Malignant Melanoma in Sweden

BACKGROUND

Cutaneous melanoma is steadily increasing worldwide. The new AJCC 8th edition was recently launched and introduced several changes in melanoma staging, particularly for stage III. We conducted a population-based registry study with the purpose to evaluate the impact and prognostic accuracy of the new classification in Sweden.

METHODS

Consecutive patients diagnosed with stage III melanoma between January 2005 and September 2017 were identified by the Swedish Melanoma Registry (SMR) and included for analyses. Patients with multiple primary melanomas were excluded. Patients were classified according to the AJCC 7th as well as the 8th edition. Melanoma-specific survival (MSS) was retrieved from the Swedish Cause of Death Registry.

RESULTS

A total of 2067 eligible patients were identified from the SMR; 1150 patients (57%) changed stage III subgroup when reclassified according to the AJCC 8th edition. The median 5- and 10-year MSS for the whole cohort of stage III melanoma patients was 59% and 51% respectively. The MSS for substage IIIA, B, and C were all improved when patients were reclassified by using to the AJCC 8th edition. The newly defined substage IIID had the worst prognosis with a 10-year MSS of 16%.

CONCLUSIONS

A high proportion of patients diagnosed with stage III melanoma in Sweden between 2005 and 2017 was restaged to another subgroup, when they were reclassified according to the AJCC 8th of staging manual. We established an improved MSS for all substages compared with the former AJCC 7th edition. This may have implications on decisions about adjuvant treatment.

FOXD2-AS1 correlates with the malignant status and regulates cell proliferation, migration, and invasion in cutaneous melanoma

Long noncoding RNA (lncRNA) FOXD2 adjacent opposite strand RNA 1 (FOXD2-AS1) has been shown to be dysregulated in several types of human cancer. However, the role of FOXD2-AS1 in cutaneous melanoma was still unclear. In our study, FOXD2-AS1 expression has been found to be upregulated in cutaneous melanoma tissue specimens and cell lines compared with that in normal tissue specimens and normal human epidermal melanocyte, respectively. Furthermore, high expression of FOXD2-AS1 was obviously correlated with deep Breslow thickness, present ulceration, high Clark level and distant metastasis in cutaneous melanoma patients. However, there were no statistical associations between FOXD2-AS1 expression and cutaneous melanoma patients' disease-free survival and overall survival. The results of loss-of-function study showed that inhibition of FOXD2-AS1 suppresses cutaneous melanoma cell proliferation, migration and invasion through regulating phospho-Akt expression. In conclusion, FOXD2-AS1 is associated with clinical progression in cutaneous melanoma patients, and functions as oncogenic lncRNA in cutaneous melanoma cells.

Melanoma-Induced Reprogramming of Schwann Cell Signaling Aids Tumor Growth

The tumor microenvironment has been compared with a nonhealing wound involving a complex interaction between multiple cell types. Schwann cells, the key regulators of peripheral nerve repair, have recently been shown to directly affect nonneural wound healing. Their role in cancer progression, however, has been largely limited to neuropathic pain and perineural invasion. In this study, we showed that melanoma activated otherwise dormant functions of Schwann cells aimed at nerve regeneration and wound healing. Such reprogramming of Schwann cells into repair-like cells occurred during the destruction and displacement of neurons as the tumor expanded and via direct signaling from melanoma cells to Schwann cells, resulting in activation of the nerve injury response. Melanoma-activated Schwann cells significantly altered the microenvironment through their modulation of the immune system and the extracellular matrix in a way that promoted melanoma growth in vitro and in vivo. Local inhibition of Schwann cell activity following cutaneous sensory nerve transection in melanoma orthotopic models significantly decreased the rate of tumor growth. Tumor-associated Schwann cells, therefore, can have a significant protumorigenic effect and may present a novel target for cancer therapy. SIGNIFICANCE: These findings reveal a role of the nerve injury response, particularly through functions of activated Schwann cells, in promoting melanoma growth.

Smartphone applications for triaging adults with skin lesions that are suspicious for melanoma

BACKGROUND

Melanoma accounts for a small proportion of all skin cancer cases but is responsible for most skin cancer-related deaths. Early detection and treatment can improve survival. Smartphone applications are readily accessible and potentially offer an instant risk assessment of the likelihood of malignancy so that the right people seek further medical attention from a clinician for more detailed assessment of the lesion. There is, however, a risk that melanomas will be missed and treatment delayed if the application reassures the user that their lesion is low risk.

OBJECTIVES

To assess the diagnostic accuracy of smartphone applications to rule out cutaneous invasive melanoma and atypical intraepidermal melanocytic variants in adults with concerns about suspicious skin lesions.

SEARCH METHODS

We undertook a comprehensive search of the following databases from inception to August 2016: Cochrane Central Register of Controlled Trials; MEDLINE; Embase; CINAHL; CPCI; Zetoc; Science Citation Index; US National Institutes of Health Ongoing Trials Register; NIHR Clinical Research Network Portfolio Database; and the World Health Organization International Clinical Trials Registry Platform. We studied reference lists and published systematic review articles.

SELECTION CRITERIA

Studies of any design evaluating smartphone applications intended for use by individuals in a community setting who have lesions that might be suspicious for melanoma or atypical intraepidermal melanocytic variants versus a reference standard of histological confirmation or clinical follow-up and expert opinion.

DATA COLLECTION AND ANALYSIS

Two review authors independently extracted all data using a standardised data extraction and quality assessment form (based on QUADAS-2). Due to scarcity of data and poor quality of studies, we did not perform a meta-analysis for this review. For illustrative purposes, we plotted estimates of sensitivity and specificity on coupled forest plots for each application under consideration.

MAIN RESULTS

This review reports on two cohorts of lesions published in two studies. Both studies were at high risk of bias from selective participant recruitment and high rates of non-evaluable images. Concerns about applicability of findings were high due to inclusion only of lesions already selected for excision in a dermatology clinic setting, and image acquisition by clinicians rather than by smartphone app users.We report data for five mobile phone applications and 332 suspicious skin lesions with 86 melanomas across the two studies. Across the four artificial intelligence-based applications that classified lesion images (photographs) as melanomas (one application) or as high risk or 'problematic' lesions (three applications) using a pre-programmed algorithm, sensitivities ranged from 7% (95% CI 2% to 16%) to 73% (95% CI 52% to 88%) and specificities from 37% (95% CI 29% to 46%) to 94% (95% CI 87% to 97%). The single application using store-and-forward review of lesion images by a dermatologist had a sensitivity of 98% (95% CI 90% to 100%) and specificity of 30% (95% CI 22% to 40%).The number of test failures (lesion images analysed by the applications but classed as 'unevaluable' and excluded by the study authors) ranged from 3 to 31 (or 2% to 18% of lesions analysed). The store-and-forward application had one of the highest rates of test failure (15%). At least one melanoma was classed as unevaluable in three of the four application evaluations.

AUTHORS' CONCLUSIONS

Smartphone applications using artificial intelligence-based analysis have not yet demonstrated sufficient promise in terms of accuracy, and they are associated with a high likelihood of missing melanomas. Applications based on store-and-forward images could have a potential role in the timely presentation of people with potentially malignant lesions by facilitating active self-management health practices and early engagement of those with suspicious skin lesions; however, they may incur a significant increase in resource and workload. Given the paucity of evidence and low methodological quality of existing studies, it is not possible to draw any implications for practice. Nevertheless, this is a rapidly advancing field, and new and better applications with robust reporting of studies could change these conclusions substantially.

Conditional recurrence-free survival in patients with primary stage I-II cutaneous malignant melanoma - a population-based study

Conditional survival in patients with localized primary cutaneous malignant melanoma (CMM) is well described. However, conditional recurrence-free survival (RFS) has not been investigated before. The aim of this study was to determine conditional RFS and test for time dependency in prognostic factors in patients with localized stage I-II CMM. This study included 1437 CMM patients registered in one region of Sweden during 1999-2012 followed up through 31 December 2012. To identify first recurrence of CMM disease, data from a care data warehouse, the pathology and radiology department registries were used. Patients were also followed through a Census Register and the National Cause of Death Register. The time-dependent risk of recurrence was analysed in a Cox's proportional hazard regression. The 5-year conditional RFS increased from 86% (95% confidence interval: 84-88) at diagnosis to 96% (95% confidence interval: 94-98) at 5 years after diagnosis. Women showed a 60% lower risk of recurrence than men and this effect was stable over time (P=0.39). Patients aged greater than or equal to 65 years had a 40% higher risk of recurrence than patients aged less than 65 years, and this effect was stable over time (P=0.65). Patients with tumour ulceration showed a 70% higher risk of recurrence than nonulcerated patients, but this effect disappeared after 2 years (P=0.04). For patients with T3-T4 CMM, the hazard ratios decreased over time and were similar to hazard ratio of patients with T2 CMM after 2 years and later. The decreasing impact of tumour thickness and ulceration over time could have important implications for CMM patients in terms of counselling and follow-up.

Development and validation of a nomogram to predict recurrence and melanoma-specific mortality in patients with negative sentinel lymph nodes

Background

Patients with melanoma and negative sentinel nodes (SNs) have varying outcomes, dependent on several prognostic factors. Considering all these factors in a prediction model might aid in identifying patients who could benefit from a personalized treatment strategy. The objective was to construct and validate a nomogram for recurrence and melanoma‐specific mortality (MSM) in patients with melanoma and negative SNs.

Methods

A total of 3220 patients with negative SNs were identified from a cohort of 4124 patients from four EORTC Melanoma Group centres who underwent sentinel lymph node biopsy. Prognostic factors for recurrence and MSM were studied with Cox regression analysis. Significant factors were incorporated in the models. Performance was assessed by discrimination (c‐index) and calibration in cross‐validation across the four centres. A nomogram was developed for graphical presentation.

Results

There were 3180 eligible patients. The final prediction model for recurrence and the calibrated model for MSM included three independent prognostic factors: ulceration, anatomical location and Breslow thickness. The c‐index was 0·74 for recurrence and 0·76 for the calibrated MSM model. Cross‐validation across the four centres showed reasonable model performance. A nomogram was developed based on these models. One‐third of the patients had a 5‐year recurrence probability of 8·2 per cent or less, and one‐third had a recurrence probability of 23·0 per cent or more.

Conclusion

A nomogram for predicting recurrence and MSM in patients with melanoma and negative SNs was constructed and validated. It could provide personalized estimates useful for tailoring surveillance strategies (reduce or increase intensity), and selection of patients for adjuvant therapy or clinical trials.

Could personalize care

Melanoma staging: Varying precision and terminal digit clustering in Breslow thickness data is evident in a population-based study

BACKGROUND

Errors in Breslow thickness reporting can give misclassification of T category, an important classifier in melanoma staging.

OBJECTIVE

We sought to investigate precision (number of digits) and terminal digit clustering in Breslow thickness and potential consequences for T category.

METHODS

All first primary and morphologically verified invasive melanomas in Norway between 2008 and 2015 were included. A smoothing model was fitted to estimate the underlying Breslow thickness distribution without digit clustering.

RESULTS

Thickness was reported for 13,057 (97.5%) patients; the median was 1.0 mm (range, 0.09-85). It was reported as whole numbers (15.6%), to 1 decimal (78.2%) and 2 decimal places (6.2%)-thin tumors with more precision than thick tumors. Terminal digit clustering was found with marked peaks in the observed frequency distribution for terminal digits 0 and 5, and with drops around these peaks. Terminal digit clustering increased proportions of patients classified with T1 and T4 tumors and decreased proportions classified with T2 and T3.

LIMITATIONS

Breslow thickness was not reported in 2.5% of cases.

CONCLUSIONS

The Norwegian recommendation of measurement to the nearest 0.1 mm was not followed. Terminal digit clustering was marked, with consequences for T category. Pathologists, clinicians, and epidemiologists should know that clustering of thickness data around T category cut points can impact melanoma staging with consequent effect on patient management and prognosis.

Review of diagnostic, prognostic, and predictive biomarkers in melanoma

Melanoma is an aggressive cutaneous malignancy with rapidly rising incidence. Diagnosis of controversial melanocytic lesions, correct prognostication of patients, selection of appropriate adjuvant and systemic therapies, and prediction of response to a given therapy remain very real challenges. Despite these challenges, multiple high throughput, nucleic-acid based biomarkers have been developed that can be assayed from histologic tissue specimens. FISH, CGH, Decision-Dx, and other multi-marker assays have been combined to improve overall predictability. This review discusses some of the most promising nucleic acid based assays that can be obtained from tissue specimens to assist with diagnosis, prognostication, and prediction of treatment response.

Survival prediction tools for esophageal and gastroesophageal junction cancer: A systematic review

BACKGROUND

Clinical, pathological, and molecular information combined with cancer stage in prognostication algorithms can offer more personalized estimates of survival, which might guide treatment choices. Our aim was to evaluate the quality of prognostication tools in esophageal cancer.

METHODS

We systematically searched MedLine and Embase from 2005 to 2017 for studies reporting development or validation of models predicting long-term survival in esophageal cancer. We evaluated tools using the Critical Appraisal and Data Extraction for Systematic Reviews of Prediction Modelling Studies guidelines and the American Joint Committee on Cancer acceptance criteria for risk models.

RESULTS

We identified 16 prognostication tools for patients treated with curative intent and 1 for patients with metastatic disease. These tools frequently excluded adenocarcinoma, contained outdated data, and were developed with a limited sample size. Nine tools were developed in China for squamous cell cancer, and 11 used data on patients diagnosed before 2010. Most tools excluded key prognostic factors such as age and sex. Tumor stage and grade were the most commonly, but not universally, included factors. Twelve tools were designed to predict overall survival; 5 predicted cancer-specific survival. Bootstrap internal validation was performed for most tools; c-statistics ranged from 0.63 to 0.77 and graphically evaluated calibration was "good." Five tools were externally validated; c-statistics ranged from 0.70 to 0.77.

CONCLUSIONS

Existing tools cannot be confidently used for esophageal cancer prognostication in current clinical practice. Better-quality tools might help to more individually and accurately estimate disease course, select further treatments, and risk-stratify for future clinical trials.

The Limitations of Standard Clinicopathologic Features to Accurately Risk-Stratify Prognosis after Resection of Intrahepatic Cholangiocarcinoma

BACKGROUND

The ability to provide accurate prognostic data after hepatectomy for intrahepatic cholangiocarcinoma (ICC) remains poor. We sought to develop and validate a nomogram to predict survival, as well as investigate the clinical implications of underestimating patients' risk of recurrence.

METHODS

Patients undergoing curative-intent resection of ICC between 1990 and 2015 at 14 major hepatobiliary centers were included. Variables significant on multivariable analysis were used to construct a nomogram to predict disease-free survival (DFS). The nomogram assigned a score to each variable included in the model and calculated the risk of recurrence.

RESULTS

Eight hundred ninety-seven patients are included in the analytic cohort. On multivariable Cox regression analysis, tumor size > 5 cm (HR 1.98, 95% CI 1.44-2.13; p < 0.001), multifocal ICC (HR 1.64, 95% CI 1.32-2.03; p < 0.001), lymph node metastasis (HR 1.63, 95% CI 1.25-2.11; p < 0.001), poorly differentiated tumor grade (HR 1.50, 95% CI 1.21-1.89; p < 0.001), and periductal infiltrating type (PI) morphology (HR 1.42, 95% CI 1.09-1.83; p = 0.008) were independent adverse risk factors associated with decreased DFS. The Harrell's c-index for the nomogram was 0.633 (with n = 5000 bootstrapping resamples) and the plot comparing predicted and actuarial DFS demonstrated a good calibration of the model. A subset of patients (n = 282) had a DFS worse than predicted (ΔPredicted DFS - Actuarial DFS > 6 months). Moreover, underestimation of a recurrence risk was more common among patients with clinicopathologic features traditionally considered "favorable."

CONCLUSION

A nomogram based on standard clinicopathologic characteristics was suboptimal in its ability to predict accurately risk of recurrence among patients with ICC after curative-intent liver resection. Particularly, the risk of underestimating patient risk of recurrence was highest among patients with historically favorable characteristics. Over one third of patients recurred > 6 months earlier than the DFS predicted by the nomogram.

Variability in Predictions from Online Tools: A Demonstration Using Internet-Based Melanoma Predictors

BACKGROUND

Prognostic models are increasingly being made available online, where they can be publicly accessed by both patients and clinicians. These online tools are an important resource for patients to better understand their prognosis and for clinicians to make informed decisions about treatment and follow-up. The goal of this analysis was to highlight the possible variability in multiple online prognostic tools in a single disease.

METHODS

To demonstrate the variability in survival predictions across online prognostic tools, we applied a single validation dataset to three online melanoma prognostic tools. Data on melanoma patients treated at Memorial Sloan Kettering Cancer Center between 2000 and 2014 were retrospectively collected. Calibration was assessed using calibration plots and discrimination was assessed using the C-index.

RESULTS

In this demonstration project, we found important differences across the three models that led to variability in individual patients' predicted survival across the tools, especially in the lower range of predictions. In a validation test using a single-institution data set, calibration and discrimination varied across the three models.

CONCLUSIONS

This study underscores the potential variability both within and across online tools, and highlights the importance of using methodological rigor when developing a prognostic model that will be made publicly available online. The results also reinforce that careful development and thoughtful interpretation, including understanding a given tool's limitations, are required in order for online prognostic tools that provide survival predictions to be a useful resource for both patients and clinicians.

Personalizing prognosis in colorectal cancer: A systematic review of the quality and nature of clinical prognostic tools for survival outcomes

Integrating diverse types of prognostic information into accurate, individualized estimates of outcome in colorectal cancer is challenging. Significant heterogeneity in colorectal cancer prognostication tool quality exists. Methodology is incompletely or inadequately reported. Evaluations of the internal or external validity of the prognostic model are rarely performed. Prognostication tools are important devices for patient management, but tool reliability is compromised by poor quality. Guidance for future development of prognostication tools in colorectal cancer is needed.

Perioperative BRAF inhibitors in locally advanced stage III melanoma

BACKGROUND AND OBJECTIVES

Stage III malignant melanoma is a heterogeneous disease where those cases deemed marginally resectable or irresecatble are frequently incurable by surgery alone. Targeted therapy takes advantage of the high incidence of BRAF mutations in melanomas, most notably the V600E mutation. These agents have rarely been used in a neoadjuvant setting prior to surgery.

METHODS

Thirteen consecutive patients with confirmed BRAF^V600E regionally advanced melanoma deemed marginally resectable or irrresectable, were treated with BRAF inhibiting agents, prior to undergoing surgery. The primary outcome measures were a successful resection and pathological response. Disease-free survival was a secondary outcome measure.

RESULTS

Overall, 12/13 patients showed a marked clinical responsiveness to medical treatment, enabling a macroscopically successful resection in all cases. Four patients had a complete pathological response with no viable tumor evident in the resected specimens and eight patients showed evidence of minimally residual tumor with extensive tumoral necrosis and fibrosis. One patient progressed and died before surgery. At a median follow up of 20 months, 10 patients remain free of disease.

CONCLUSIONS

Perioperative treatment with BRAF inhibiting agents in BRAFV600E mutated Stage III melanoma patients facilitates surgical resection and affords satisfactory disease free survival.

Underexpression of Specific Interferon Genes Is Associated with Poor Prognosis of Melanoma

Because the prognosis of melanoma is challenging and inaccurate when using current clinical approaches, clinicians are seeking more accurate molecular markers to improve risk models. Accordingly, we performed a survival analysis on 404 samples from The Cancer Genome Atlas (TCGA) cohort of skin cutaneous melanoma. Using our recently developed gene network model, we identified biological signatures that confidently predict the prognosis of melanoma (p-value < 10-5). Our model predicted 38 cases as low-risk and 54 cases as high-risk. The probability of surviving at least 5 years was 64% for low-risk and 14% for high-risk cases. In particular, we found that the overexpression of specific genes in the mitotic cell cycle pathway and the underexpression of specific genes in the interferon pathway are both associated with poor prognosis. We show that our predictive model assesses the risk more accurately than the traditional Clark staging method. Therefore, our model can help clinicians design treatment strategies more effectively. Furthermore, our findings shed light on the biology of melanoma and its prognosis. This is the first in vivo study that demonstrates the association between the interferon pathway and the prognosis of melanoma.