Artificial Intelligence in Dermatology: A Systematic Review of Its Applications in Melanoma and Keratinocyte Carcinoma Diagnosis.

BACKGROUND: Limited access to dermatologic care may pose an obstacle to the early detection and intervention of cutaneous malignancies. The role of artificial intelligence (AI) in skin cancer diagnosis may alleviate potential care gaps. OBJECTIVE: The aim of this systematic review was to offer an in-depth exploration of published AI algorithms trained on dermoscopic and macroscopic clinical images for the diagnosis of melanoma, basal cell carcinoma, and cutaneous squamous cell carcinoma (cSCC). METHODS: Adhering to the Preferred Reporting Items for Systematic Reviews and Meta-Analysis guidelines, a systematic review was conducted on peer-reviewed articles published between January 1, 2000, and January 26, 2023. RESULTS AND DISCUSSION: Among the 232 studies in this review, the overall accuracy, sensitivity, and specificity of AI for tumor detection averaged 90%, 87%, and 91%, respectively. Model performance improved with time. Despite seemingly impressive performance, the paucity of external validation and limited representation of cSCC and skin of color in the data sets limits the generalizability of the current models. In addition, dermatologists coauthored only 12.9% of all studies included in the review. Moving forward, it is imperative to prioritize robustness in data reporting, inclusivity in data collection, and interdisciplinary collaboration to ensure the development of equitable and effective AI tools.

High-volume facilities are significantly more likely to use guideline-adherent systemic immunotherapy for metastatic Merkel cell carcinoma: implications for cancer care regionalization.

Merkel cell carcinoma (MCC) is a neuroendocrine skin cancer with a high rate of mortality. While still relatively rare, the incidence of MCC has been rapidly rising in the US and around the world. Since 2017, two immunotherapeutic drugs, avelumab and pembrolizumab, have been FDA-approved for the treatment of metastatic MCC and have revolutionized outcomes for MCC. However, real-world outcomes can differ from clinical trial data, and the adoption of novel therapeutics can be gradual. We aimed to characterize the treatment practices and outcomes of patients with metastatic MCC across the US. A retrospective cohort study of adult cases of MCC in the National Cancer Database diagnosed from 2004 to 2019 was performed. Multivariable logistic regressions to determine the association of a variety of patient, tumor, and system factors with likelihood of receipt of systemic therapies were performed. Univariate Kaplan-Meier and multivariable Cox survival regressions were performed. We identified 1017 cases of metastatic MCC. From 2017 to 2019, 54.2% of patients received immunotherapy. This increased from 45.1% in 2017 to 63.0% in 2019. High-volume centers were significantly more likely to use immunotherapy (odds ratio 3.235, p = 0.002). On univariate analysis, patients receiving systemic immunotherapy had significantly improved overall survival (p < 0.001). One-, 3-, and 5-year survival was 47.2% (standard error [SE] 1.8%), 21.8% (SE 1.5%), and 16.5% (SE 1.4%), respectively, for patients who did not receive immunotherapy versus 62.7% (SE 3.5%), 34.4% (SE 3.9%), and 23.6% (SE 4.4%), respectively, for those who did (Fig. 1). In our multivariable survival regression, receipt of immunotherapy was associated with an approximately 35% reduction in hazard of death (hazard ratio 0.665, p < 0.001; 95% CI 0.548-0.808). Our results demonstrate that the real-world survival advantage of immunotherapy for metastatic MCC is similar to clinical trial data. However, many patients with metastatic disease did not receive this guideline-recommended therapy in our most recent study year, and use of immunotherapy is higher at high-volume centers. This suggests that regionalization of care to high-volume centers or dissemination of their practices, may ultimately improve patient survival.

Evaluating Rates of Positive Margins After Standard Excision of Cutaneous Adnexal Malignancies.

BACKGROUND: It is recommended to excise adnexal neoplasms with standard local excision or Mohs micrographic surgery (MMS), although many occur on high-risk sites such as the head and neck (H&N) and exhibit subclinical extension. Minimal evidence exists on the efficacy of standard excisions for these tumors. OBJECTIVE: To evaluate the rate of positive surgical margins after standard excision of adnexal tumors. METHODS: Retrospective cohort study of cutaneous adnexal malignancies from the National Cancer Database diagnosed from 2004 to 2019. RESULTS: The authors identified a total of 4,402 cases treated with standard excision. Tumors on the H&N were approximately twice as likely as those on the trunk and extremities (T&E) to be excised with positive margins (odds ratio 2.146, p < .001), with the highest estimated rate for eccrine adenocarcinoma (12.1%, SE: 2.3%). The subtype with the highest positive margin rate on the T&E was microcystic adnexal carcinoma (8.0%, SE: 2.9). Positive margins were associated with poorer overall survival on multivariable survival analysis (hazard ratio 1.299, p = .015). CONCLUSION: The authors present subtype- and site-specific positive margin rates for adnexal tumors treated with standard excision, which suggest that tumors on the H&N and some T&E subtypes, should be considered for MMS.

Overall Survival After Mohs Surgery for Early-Stage Merkel Cell Carcinoma.

Importance: Merkel cell carcinoma (MCC) is a rare cutaneous malignant neoplasm with increasing incidence and high mortality. Although it is accepted that the optimal treatment for localized tumors is surgical, the data surrounding the optimal surgical approach are mixed, and current National Comprehensive Cancer Network guidelines state that Mohs micrographic surgery (MMS) and wide local excision (WLE) can both be used. The current National Comprehensive Cancer Network guidelines do not advocate a preference for MMS or WLE and suggest that they can be used interchangeably. Objective: To evaluate the association of surgical approach with overall survival after excision of localized T1/T2 MCC. Design, Setting, and Participants: This retrospective cohort study used the National Cancer Database to assess adults with T1/T2 MCC who were diagnosed between January 1, 2004, and December 31, 2018, with pathologically confirmed, negative regional lymph nodes and treated with surgery. The National Cancer Database includes all reportable cases from Commission on Cancer-accredited facilities. Data analysis was performed from October 2022 to May 2023. Exposure: Surgical approach. Main Outcomes and Measures: Overall survival. Results: A total of 2313 patients (mean [SD] age, 71 [10.6] years; 1340 [57.9%] male) were included in the study. Excision with MMS had the best unadjusted survival, with mean (SE) survival rates of 87.4% (3.4%) at 3 years, 84.5% (3.9%) at 5 years, and 81.8% (4.6%) at 10 years vs 86.1% (0.9%) at 3 years, 76.9% (1.2%) at 5 years, and 60.9% (2.0%) at 10 years for patients treated with WLE. Patients treated with narrow-margin excision had similar survival as those treated with WLE, with mean (SE) survival rates of 84.8% (1.4%) at 3 years, 78.3% (1.7%) at 5 years, and 60.8% (3.6%) at 10 years. On multivariable survival analysis, excision with MMS was associated with significantly improved survival compared with WLE (hazard ratio, 0.59; 95% CI, 0.36-0.97; P = .04). High-volume MCC centers were significantly more likely to use MMS over WLE compared with other centers (odds ratio, 1.99; 95% CI, 1.63-2.44; P < .001). Conclusions and Relevance: In this cohort study, the use of MMS (compared with WLE) was associated with significantly improved survival for patients with localized MCC with pathologically confirmed negative lymph nodes treated with surgery. These data suggest that Mohs surgery may provide a more effective treatment for MCC primary tumors than conventional WLE, although the lack of randomization and potential for selection bias in this study highlight the need for future prospective work evaluating this issue.

Defining and quantifying histopathologic risk factors for regional and distant metastases in a large cohort of vulvar squamous cell carcinomas.

BACKGROUND: Vulvar squamous cell carcinoma (vSCC) is a rare tumor with a good prognosis when treated at a localized stage. However, once regional/distant metastasis occurs, vSCC can be rapidly fatal. Thus, it is important to identify tumor prognostic features so that high-risk cases can be prioritized for further diagnostic workup and treatment. OBJECTIVE: To estimate the risk of regional/distant metastasis at presentation and sentinel lymph node status for vSCC based on histopathologic characteristics. METHODS: A retrospective cohort study of 15,188 adult vSCC cases from the National Cancer Database diagnosed from 2012 to 2019. RESULTS: We provide specific estimates of the risk of clinically positive nodes and metastatic disease at presentation and sentinel lymph node positivity according to tumor size, moderate/poor tumor differentiation, and lymph-vascular invasion. These histopathologic factors were all significantly associated with the tested clinical outcomes in a multivariable analysis. Moderate (hazard ratio, 1.190; P < .001) and poor differentiation (hazard ratio, 1.204; P < .001) and lymph-vascular invasion (hazard ratio, 1.465; P < .001) were also associated with significantly poorer overall survival. LIMITATIONS: Data on disease-specific survival not available in the data set. CONCLUSIONS: We demonstrate the association of the histopathologic characteristics of vSCC with clinically important outcomes. These data may provide individualized information when discussing diagnostic/treatment recommendations, particularly regarding sentinel lymph node biopsy. These data may also guide future staging and risk stratification efforts for vSCC.

Assessing Rates of Positive Surgical Margins After Standard Excision of Vulvar Melanomas.

BACKGROUND: Vulvar melanoma is a rare malignancy with frequent recurrence and poor prognosis. National guidelines recommend wide local excision of these tumors with allowances for narrower margins for anatomic and functional limitations, which are common on specialty sites. There is presently a lack of data of margin positivity after standard excision of vulvar melanomas. OBJECTIVE: We aim to evaluate the rate of positive margins after standard excision of vulvar melanomas. MATERIALS AND METHODS: Retrospective cohort study of surgically excised vulvar melanomas from the NCDB diagnosed from 2004 to 2019. RESULTS: We identified a total of 2,226 cases. Across surgical approaches and tumor stages, 17.2% (Standard Error [SE]: 0.8%) of cases had positive surgical margins. Among tumor stages, T4 tumors were most commonly excised with positive margins (22.9%, SE: 1.5%). On multivariable survival analysis, excision with positive margins was associated with significantly poorer survival (Hazard Ratio 1.299, p = .015). CONCLUSION: We find that positive margin rates after standard excision of vulvar malignancies are higher than for other specialty site melanomas. Our data suggest that use of surgical approaches with complete margin assessment may improve local control and functional outcomes for patients with vulvar melanoma as they have for patients with other specialty site melanomas.

Understanding the squamous cell carcinoma immune microenvironment.

Primary cutaneous squamous cell carcinoma (cSCC) is the second most common human cancer with a rising incidence of about 1.8 million in the United States annually. Primary cSCC is usually curable by surgery; however, in some cases, cSCC eventuates in nodal metastasis and death from disease specific death. cSCC results in up to 15,000 deaths each year in the United States. Until recently, non-surgical options for treatment of locally advanced or metastatic cSCC were largely ineffective. With the advent of checkpoint inhibitor immunotherapy, including cemiplimab and pembrolizumab, response rates climbed to 50%, representing a vast improvement over chemotherapeutic agents used previously. Herein, we discuss the phenotype and function of SCC associated Langerhans cells, dendritic cells, macrophages, myeloid derived suppressor cells and T cells as well as SCC-associated lymphatics and blood vessels. Possible role(s) of SCC-associated cytokines in progression and invasion are reviewed. We also discuss the SCC immune microenvironment in the context of currently available and pipeline therapeutics.

Staged melanoma excision requires larger margins for tumor clearance and results in low rates of recurrence.

There is controversy regarding the optimal surgical modality and ideal recommended margins for treating melanoma in situ (MIS) and invasive melanoma (IM). Although wide local excision is recommended, staged excision offers excellent margin control and low recurrence rates. In this manuscript, we reviewed a 10-year experience of staged excisions for the treatment of MIS and IM. A retrospective review was performed of 130 MIS and 32 IM cases treated with staged excision from April 2012 to April 2022. Staged excision was performed on the head and neck in 102 (79%) MIS and 23 (72%) IM cases. Approximately 10% of cases required surgical margins above the current recommendations (11 (9%) MIS and 6 (19%) IM). Twenty-three (19%) MIS and 7 (22%) IM cases required more than one excision to obtain clearance. Recurrence rates among MIS and IM were 0.0% and 0.6%, respectively. Upstaging occurred in 5 (4%) MIS and 7 (22%) IM cases. Complex repairs were performed on 82 (63%) MIS and 17 (53%) IM cases. Our findings revealed that staged excision provides effective margin control and low recurrence rates. Approximately 10% of patients required margins greater than the current recommendations, leading to larger defects and more complex repairs.

Immunosuppressed patients are at increased risk of local recurrence, metastasis, and disease specific death from cutaneous squamous cell carcinoma.

It is well established that immunosuppressed patients are at increased risk for poor outcomes (PO) from cutaneous squamous cell carcinoma (cSCC), including local recurrence (LR), nodal metastasis (NM), distant metastasis (DM), and disease-specific death (DSD). Defining PO risk is challenging but may be beneficial in guiding management. We aimed to define PO risk factors and evaluated their importance in immunosuppressed versus immunocompetent patients. We conducted a 4-year single-center retrospective review of patients with cSCC. Patient and tumor characteristics were evaluated in those that experienced PO. Immunosuppressed patients were ~ 11-fold more likely than immunocompetent patients to develop PO (10/85 vs. 15/1332, p < 0.0001). Among those with PO, immunosuppressed patients had diminished relapse free (p = 0.026) and progression free (p < 0.001) survival compared to immunocompetent. Immunosuppression was significantly associated with LR (p < 0.00001). Immunosuppressed patients were also more likely to develop NM, DM and experience DSD (p = 0.027). Mohs Appropriate Use Criteria was associated with NM, DM and DSD (p = 0.029), with area H tumors more likely to result in metastasis and death. In conclusion, immunosuppressed patients are more likely to develop LR, metastasis, and DSD from cSCC compared to immunocompetent patients. Immunosuppressed status was an independent risk factor for PO in this cohort and further considered for its inclusion in prognostication schema is warranted.