Update on eighth edition American Joint Committee on Cancer classification for cutaneous melanoma and overview of potential pitfalls in histological examination of staging parameters

Skin Cancer Associated With Chronic Leg Ulcers in the Population of Olmsted County, Minnesota

Malignant skin tumors in the setting of chronic leg ulcers (CLUs) are often underdiagnosed which may contribute to treatment delay and poor outcomes. The aims of our study were to determine the incidence and clinical characteristics of skin cancers in leg ulcers in the Olmsted County population from 1995 to 2020. We used the Rochester Epidemiology Project (a collaboration between health care providers) infrastructure to describe this epidemiology, allowing "population-based" research. Electronic medical records of adult patients with International Classification of Diseases diagnosis codes for leg ulcers and skin cancers on the legs were queried. Thirty-seven individuals with skin cancers in nonhealing ulcers were identified. The cumulative incidence of skin cancer over the 25-year period was 37:7864 (0.47%). The overall incidence rate was 470 per 100,000 patients. Eleven (29.7%) men and 26 (70.3%) women were identified with mean age of 77 years. History of venous insufficiency was present in 30 (81.1%) patients and diabetes in 13 (35.1%) patients. Clinical characteristics of CLU with skin cancer included abnormal granulation tissue in 36 (94.7%) and irregular borders in 35 (94.6%) cases. Skin cancers among CLUs included 17 (41.5%) basal cell carcinomas, 17 (41.5%) squamous cell carcinomas, 2 (4.9%) melanomas, 2 (4.9%) porocarcinomas, 1 (2.4%) basosquamous cell carcinoma, and 1 (2.4%) eccrine adenocarcinoma. The apparent association between chronic wounds and subsequent biopsy-proven skin cancer of the same site was primarily observed in elderly patients; malignant transformation of wounds favored basal cell carcinoma and squamous cell carcinoma. This retrospective cohort study further characterizes the association between skin cancers and chronic leg wounds.

A deep learning model based on whole slide images to predict disease-free survival in cutaneous melanoma patients

The application of deep learning on whole-slide histological images (WSIs) can reveal insights for clinical and basic tumor science investigations. Finding quantitative imaging biomarkers from WSIs directly for the prediction of disease-free survival (DFS) in stage I-III melanoma patients is crucial to optimize patient management. In this study, we designed a deep learning-based model with the aim of learning prognostic biomarkers from WSIs to predict 1-year DFS in cutaneous melanoma patients. First, WSIs referred to a cohort of 43 patients (31 DF cases, 12 non-DF cases) from the Clinical Proteomic Tumor Analysis Consortium Cutaneous Melanoma (CPTAC-CM) public database were firstly annotated by our expert pathologists and then automatically split into crops, which were later employed to train and validate the proposed model using a fivefold cross-validation scheme for 5 rounds. Then, the model was further validated on WSIs related to an independent test, i.e. a validation cohort of 11 melanoma patients (8 DF cases, 3 non-DF cases), whose data were collected from Istituto Tumori 'Giovanni Paolo II' in Bari, Italy. The quantitative imaging biomarkers extracted by the proposed model showed prognostic power, achieving a median AUC value of 69.5% and a median accuracy of 72.7% on the public cohort of patients. These results remained comparable on the validation cohort of patients with an AUC value of 66.7% and an accuracy value of 72.7%, respectively. This work is contributing to the recently undertaken investigation on how treat features extracted from raw WSIs to fulfil prognostic tasks involving melanoma patients. The promising results make this study as a valuable basis for future research investigation on wider cohorts of patients referred to our Institute.

Excision Margin and Survival in Patients with High-Risk, Primary Cutaneous Melanoma: A Retrospective Study Based on Surveillance Epidemiology and End Result (SEER) Database

OBJECTIVES

The optimal excision margin of primary cutaneous melanoma greater than 2 mm in thickness is still a controversial topic. The aim of the present study was to compare the long-term survival between narrow and wide excision margins in the surgical excision of patients with high-risk primary melanoma.

METHODS

We chose the patients with primary melanoma of the skin thicker than 2 mm in The Surveillance, Epidemiology, and End Results database. Patients were divided into a narrow margin group (1-2 cm) and a wide margin group (>2 cm) according to the resection margin information. The primary outcome was overall survival and disease-specific survival.

RESULTS

From 2004 to 2015, a total of 2,772 patients diagnosed as having melanoma of the skin were recruited into this study and were assigned to the narrow margin group (n = 1996) and the wide margin group (n = 776). A total of 1,098 patients died during the follow-up, and 681 of these were due to melanoma. There were 779 deaths in the narrow margin group and 319 deaths in the wide margin group (HR: 0.96, 95% CI: 0.84-1.10, p = 0.26). A total of 490 melanoma-specific deaths were reported in the narrow margin group and 191 were reported in the wide margin group (HR: 1.01, 95% CI: 0.85-1.19, p = 0.91).

CONCLUSIONS

Wider excision margin greater than 2 cm did not provide any additional therapeutic benefits compared to narrow excision margin between 1 and 2 cm. A 2-cm margin is adequate and safe for high-risk primary melanoma of the skin thicker than 2 mm.

Improving Skin Cancer Classification Using Heavy-Tailed Student T-Distribution in Generative Adversarial Networks (TED-GAN)

Deep learning has gained immense attention from researchers in medicine, especially in medical imaging. The main bottleneck is the unavailability of sufficiently large medical datasets required for the good performance of deep learning models. This paper proposes a new framework consisting of one variational autoencoder (VAE), two generative adversarial networks, and one auxiliary classifier to artificially generate realistic-looking skin lesion images and improve classification performance. We first train the encoder-decoder network to obtain the latent noise vector with the image manifold’s information and let the generative adversarial network sample the input from this informative noise vector in order to generate the skin lesion images. The use of informative noise allows the GAN to avoid mode collapse and creates faster convergence. To improve the diversity in the generated images, we use another GAN with an auxiliary classifier, which samples the noise vector from a heavy-tailed student t-distribution instead of a random noise Gaussian distribution. The proposed framework was named TED-GAN, with T from the t-distribution and ED from the encoder-decoder network which is part of the solution. The proposed framework could be used in a broad range of areas in medical imaging. We used it here to generate skin lesion images and have obtained an improved classification performance on the skin lesion classification task, rising from 66% average accuracy to 92.5%. The results show that TED-GAN has a better impact on the classification task because of its diverse range of generated images due to the use of a heavy-tailed t-distribution.

The Challenging Melanoma Landscape: From Early Drug Discovery to Clinical Approval

Melanoma is recognized as the most dangerous type of skin cancer, with high mortality and resistance to currently used treatments. To overcome the limitations of the available therapeutic options, the discovery and development of new, more effective, and safer therapies is required. In this review, the different research steps involved in the process of antimelanoma drug evaluation and selection are explored, including information regarding in silico, in vitro, and in vivo experiments, as well as clinical trial phases. Details are given about the most used cell lines and assays to perform both two- and three-dimensional in vitro screening of drug candidates towards melanoma. For in vivo studies, murine models are, undoubtedly, the most widely used for assessing the therapeutic potential of new compounds and to study the underlying mechanisms of action. Here, the main melanoma murine models are described as well as other animal species. A section is dedicated to ongoing clinical studies, demonstrating the wide interest and successful efforts devoted to melanoma therapy, in particular at advanced stages of the disease, and a final section includes some considerations regarding approval for marketing by regulatory agencies. Overall, considerable commitment is being directed to the continuous development of optimized experimental models, important for the understanding of melanoma biology and for the evaluation and validation of novel therapeutic strategies.

Mortality in cutaneous malignant melanoma and its association with Neutrophil-to-Lymphocyte ratio

INTRODUCTION

Cutaneous malignant melanoma (CMM) incidence has risen rapidly in the last 50 years. Poor progression and high mortality characterize CMM, making a thorough understanding of progression and associated factors essential for optimizing care.

AIMS

We assessed the association between the Neutrophil-to-Lymphocyte Ratio (NLR) and mortality in adults with CMM from an entirely mixed-race Hispanic population during 12 consecutive years of extensive follow-up.

MATERIAL & METHODS

We performed a retrospective cohort study in a tertiary hospital in Peru. NLR was categorized with a cutoff value higher or equal than 3. We collected demographic variables, laboratory results and treatments at baseline of follow-up. Cox regression analysis was performed, and we calculated crude and adjusted hazard ratios (HR) and their 95% confidence interval (95%CI).

RESULTS

The analysis was from 615 CMM cases, and there were 378 deaths. Most melanomas (63.6%) were acral lentiginous. The crude analysis showed that high NLR is a risk factor for mortality, HR = 2.52; 95%CI (2.03-3.14). High NRL ratio remains statistically significant after adjusting for confounding variables, aHR = 1.61; 95%CI (1.16-2.24). Other risk factors for mortality were clinical stages III and IV, older than 60 years, females and greater Breslow thickness.

CONCLUSIONS

We concluded that high NRL ratio is a risk factor for mortality and should be monitored in every patient who is diagnosed with malignant melanoma during their first blood count. It should then be carried out in follow-up controls for patients of clinical stage III and IV only, or in patients who present a relapse.

Recognition, Staging, and Management of Melanoma

Melanoma accounts for approximately 1% of all skin cancers but contributes to almost all skin cancer deaths. The developing picture suggests that melanoma phenotypes are driven by epigenetic mechanisms that reflect a complex interplay between genotype and environment. Furthermore, the growing consensus is that current classification standards, notwithstanding pertinent clinical history and appropriate biopsy, fall short of capturing the vast complexity of the disease. This article summarizes the current understanding of the clinical picture of melanoma, with a focus on the tremendous breakthroughs in molecular classification and therapeutics.

Histopathologic classification and prognostic factors of melanoma: a 2021 update

Despite the rapid recent advances in molecular analysis of tumors, which allow large-scale and high-resolution genomics, the "gold standard" for melanoma diagnosis continues to be histopathology, in conjunction with clinical characteristics and sometimes with important support of immunohistochemistry. Observations, where postulated that cutaneous melanomas may arise through two distinct pathways, discoveries such as that BRAF^V600E mutations were mostly common in melanomas on sun-exposed skin with little solar elastosis and seminal works for melanoma progression and evolution set the groundwork for the new WHO Classification of Melanoma: a classification of melanoma that not only encompasses histologic but also clinical, epidemiologic, and genetic characteristics. The melanomas were divided into those etiologically related to sun exposure and those that are not, based on their mutational signatures, anatomic site, and epidemiology. On the basis of degree of associated solar elastosis melanomas on the sun exposed skin were further divided by the histopathologic degree of cumulative solar damage (CSD) of the surrounding skin, into low and high CSD. On the low-CSD group of melanomas are included superficial spreading melanomas, while the high-CSD melanomas encompasses lentigo maligna and desmoplastic melanomas. The "non-CSD" classification includes acral melanomas, some melanomas in congenital nevi, melanomas in blue nevi, Spitz melanomas, mucosal melanomas, and uveal melanomas. Nodular and nevoid melanoma may occur in any pathway. A group of intermediate tumors termed melanocytoma is proposed for tumors that in addition to mutations that activate the MAPK pathway, harbor multiple driver mutations, and they are either low-grade or high-grade, to indicate that they may carry a higher risk of malignant transformation. In this review a summary of the most recent WHO classification of melanoma is provided. A short analysis of essential histopathologic prognostic parameters is also provided. The new classification of melanoma discriminates distinct types of melanoma based on their clinicopathologic, and genomic characteristics. Undoubtedly, melanoma research will continue to evolve as new clinical, pathological, molecular data accumulates. The challenge of the forthcoming years is to better characterize the intermediate category of melanocytic lesions.

Second Diagnostic Opinion by Experienced Dermatopathologists in the Setting of a Referral Regional Melanoma Unit Significantly Improves the Clinical Management of Patients With Cutaneous Melanoma

The diagnosis of cutaneous melanoma and melanocytic neoplasms in general is one of the most challenging fields in pathology, and the reported interobserver diagnostic agreement in the evaluation of melanocytic lesions is poor. Nevertheless, a correct histopathological diagnosis is crucial to ensure a good clinical management of the patients. The institution of multidisciplinary teams has recently modified the approach to the patients with cutaneous melanoma. Patients referred to a multidisciplinary melanoma unit after receiving a diagnosis of melanoma elsewhere are encouraged to have their histopathological diagnosis confirmed by a second opinion from the experienced pathologist of the team before any treatment is initiated. We performed a retrospective analysis on a series of 121 histopathological revisions required for melanocytic neoplasms in the context of a multidisciplinary team, in order to evaluate the effects of second diagnostic opinion (SDO) on the clinical management of the patients. We defined three types of diagnostic discrepancies between the first diagnosis and the second opinion, according to the greatness of their clinical impact. Overall, the incidence of diagnostic discrepancies of any type was quite high in our series (56%). Interestingly, the SDO determined relevant changes in the clinical management of the patients in 33 out of 121 (27.3%) cases. This study confirms that SDO by expert pathologists significantly affects the course of treatment of melanoma patients and helps improving the diagnostic accuracy and clinical outcome.

Automated digital TIL analysis (ADTA) adds prognostic value to standard assessment of depth and ulceration in primary melanoma

Accurate prognostic biomarkers in early-stage melanoma are urgently needed to stratify patients for clinical trials of adjuvant therapy. We applied a previously developed open source deep learning algorithm to detect tumor-infiltrating lymphocytes (TILs) in hematoxylin and eosin (H&E) images of early-stage melanomas. We tested whether automated digital (TIL) analysis (ADTA) improved accuracy of prediction of disease specific survival (DSS) based on current pathology standards. ADTA was applied to a training cohort (n = 80) and a cutoff value was defined based on a Receiver Operating Curve. ADTA was then applied to a validation cohort (n = 145) and the previously determined cutoff value was used to stratify high and low risk patients, as demonstrated by Kaplan–Meier analysis (p ≤ 0.001). Multivariable Cox proportional hazards analysis was performed using ADTA, depth, and ulceration as co-variables and showed that ADTA contributed to DSS prediction (HR: 4.18, CI 1.51–11.58, p = 0.006). ADTA provides an effective and attainable assessment of TILs and should be further evaluated in larger studies for inclusion in staging algorithms.

Differences in cutaneous melanoma survival between the 7th and 8th edition of the American Joint Committee on Cancer (AJCC). A multicentric population-based study

BACKGROUND

The 8th edition of the AJCC manual for melanoma includes many changes leading to major substage migrations, which could lead to important clinical reassessments.

OBJECTIVES

To evaluate the differences and prognostic value of the 8th AJCC classification in comparison with the 7th edition.

METHODS

Clinical and histopathological data were retrieved from five melanoma referral centers including 7815 melanoma patients diagnosed between January 1998 and December 2018. All patients were reclassified and compared using the 7th and 8th classifications of the AJCC. Sankey plots were used to evaluate the migration of patients between the different versions. The primary outcome was overall survival (OS), and curves based on the Kaplan-Meier method were used to investigate survival differences between the 7th and 8th editions.

RESULTS

The number of patients classified as stages IB, IIIA, and IIIB decreased while the patients classified as stages IA and IIIC increased notably. Migration analysis showed that many patients in group I were understaged whereas a significant percentage of patients in group III were upstaged. Indirect OS analysis showed a loss in the linearity in the AJCC 8th edition and the groups tended to overlap. Direct OS analysis between groups and versions of the AJCC showed a better prognosis within the new stage III patients, with no effect on those in stages I and II.

CONCLUSION

The 8th AJCC edition represents an important change in the classification of patients. We observe that the main migratory changes occur in stage I and III, that severity linearity is lost and groups overlap, and that a more advanced stage does not mean a worse prognosis.

SOX10 is as specific as S100 protein in detecting metastases of melanoma in lymph nodes and is recommended for sentinel lymph node assessment

BACKGROUND

Sentinel lymph node (SLN) biopsy remains crucial for melanoma staging. The European Organisation for Research and Treatment of Cancer Melanoma Group recommends performing immunohistochemical stainings for reproducible identification of melanoma metastases. S100 protein (pS100) is a commonly used melanocytic antigen because of its high sensitivity in spite of relatively low specificity. SRY-related HMG-box 10 protein (SOX10) is a transcription factor characterising neural crest-derived cells. It is uniformly expressed mostly in the nuclei of melanocytes, neural, and myoepithelial cells. Pathologists sometimes prefer SOX10 as a melanoma marker, but it has not yet been investigated on a large-scale to confirm that it is reliable and recommendable for routine SLN evaluation.

METHODS

Four hundred one treatment-naïve lymph node (LN) metastatic melanomas were included in high-density tissue microarrays and were assessed for the presence of SOX10 and pS100 by immunohistochemistry. The slides were digitalised, shared and evaluated by a panel of experienced melanoma pathologists.

RESULTS

The vast majority of melanomas were double-positive for pS100 and SOX10 (93.2%); a small percentage of the cases (3.9%) were double-negative melanomas. Discordance between the two markers was observed: 1.9% pS100(-)/SOX10(+) and 0.75% pS100(+)/SOX10(-). SOX10 was not expressed by immune cell types in the LN, resulting in a less controversial interpretation of the staining.

CONCLUSIONS

SOX10 is as equally specific as pS100 for the detection of melanoma metastases in LNs. The interpretation of SOX10 staining is highly reproducible among different centres and different pathologists because of the absence of staining of immune cells.

Deep Learning Based on Standard H&E Images of Primary Melanoma Tumors Identifies Patients at Risk for Visceral Recurrence and Death

PURPOSE

Biomarkers for disease-specific survival (DSS) in early-stage melanoma are needed to select patients for adjuvant immunotherapy and accelerate clinical trial design. We present a pathology-based computational method using a deep neural network architecture for DSS prediction.

EXPERIMENTAL DESIGN

The model was trained on 108 patients from four institutions and tested on 104 patients from Yale School of Medicine (YSM, New Haven, CT). A receiver operating characteristic (ROC) curve was generated on the basis of vote aggregation of individual image sequences, an optimized cutoff was selected, and the computational model was tested on a third independent population of 51 patients from Geisinger Health Systems (GHS).

RESULTS

Area under the curve (AUC) in the YSM patients was 0.905 (P < 0.0001). AUC in the GHS patients was 0.880 (P < 0.0001). Using the cutoff selected in the YSM cohort, the computational model predicted DSS in the GHS cohort based on Kaplan-Meier (KM) analysis (P < 0.0001).

CONCLUSIONS

The novel method presented is applicable to digital images, obviating the need for sample shipment and manipulation and representing a practical advance over current genetic and IHC-based methods.

Rsf‑1 regulates malignant melanoma cell viability and chemoresistance via NF‑κB/Bcl‑2 signaling

Remodeling and spacing factor 1 (Rsf-1) has been reported as overexpressed in numerous cancers; however, its expression, biological functions and mechanisms in malignant melanoma remain unknown. In the present study, the expression of Rsf-1 was investigated in 50 cases of malignant melanoma samples using immunohistochemistry. The results revealed that Rsf-1 expression was elevated in 38% of specimens. MTT, colony formation, Transwell and flow cytometry assays were performed to investigate the functions of Rsf-1. Knockdown of Rsf-1 in the MV3 and A375 melanoma cell lines decreased the viability, invasion and cell cycle transition of cells. Conversely, overexpression of Rsf-1 in M14 cells with low endogenous Rsf-1 expression induced opposing effects. Further analysis revealed that Rsf-1 knockdown decreased matrix metalloproteinase-2, cyclin E and phosphorylated-IκB expression. Additionally, Rsf-1 depletion reduced cisplatin resistance and significantly increased the cisplatin-associated apoptotic rate, whereas Rsf-1 overexpression exhibited opposing effects. Rsf-1 also maintained the mitochondrial membrane potential following cisplatin treatment. Analysis of apoptosis-associated proteins revealed that Rsf-1 positively regulated B-cell lymphoma 2 (Bcl-2), cellular inhibitor of apoptosis 1 (cIAP1) and cIAP2, and downregulated Bcl-2-associated X protein expression. Nuclear factor κ-light-chain-enhancer of activated B-cells (NF-κB) inhibition reversed the effects of Rsf-1 on Bcl-2. In conclusion, Rsf-1 was overexpressed in malignant melanoma and may contribute to the malignant behaviors of melanoma cells, possibly via the regulation of NF-κB signaling. Therefore, Rsf-1 may be a potential therapeutic target in the treatment of malignant melanoma.

Pathological review of primary cutaneous malignant melanoma by a specialist skin cancer multidisciplinary team improves patient care in the UK

AIMS

The National Institute for Health and Care Excellence advocated the development of specialist skin cancer multidisciplinary teams (SSMDTs) for the management of higher risk invasive skin cancers in the UK. The interobserver variability in the histopathological assessment of primary cutaneous malignant melanoma (PCMM) is well recognised.

METHODS

We evaluated the discordance rates in the assessment of the histopathological criteria of PCMM based on the eighth American Joint Committee on Cancer (AJCC) melanoma staging system and subsequent change in prognosis and management following pathology review by an SSMDT.

RESULTS

353 cases of PCMM were referred to our SSMDT between April 2015 and May 2016. Cases in which there was a discrepancy in one or more histological parameters following expert review were collected retrospectively. Of 341 eligible cases, there were 94 (27.6%) in which there was an alteration in any parameter. There was interobserver agreement in final diagnosis in 96.8%, Breslow thickness in 86.8%, ulceration in 98.2%, microsatellites in 98.5%, tumour mitotic rate in 88.9%, histological subtype in 92.4%, growth phase in 98.5%, angiolymphatic invasion in 97.7%, perineural invasion in 98.8%, regression in 95.3% and tumour-infiltrating lymphocytes in 95.0%. A corresponding change in AJCC stage occurred in 23 cases (6.7%), with a resulting change in clinical management in 10 cases (2.9%).

CONCLUSIONS

Disagreements in the pathological assessment of PCMM can have significant clinical implications for a small number of patients. Our findings highlight the value of the SSMDT for high-quality care of patients with melanoma in the UK.