RESUMEN
Background: Skin cancers are the most common malignancies diagnosed worldwide. While the early detection and treatment of pre-cancerous and cancerous skin lesions can dramatically improve outcomes, factors such as a global shortage of pathologists, increased workloads, and high rates of diagnostic discordance underscore the need for techniques that improve pathology workflows. Although AI models are now being used to classify lesions from whole slide images (WSIs), diagnostic performance rarely surpasses that of expert pathologists. Objectives: The objective of the present study was to create an AI model to detect and classify skin lesions with a higher degree of sensitivity than previously demonstrated, with potential to match and eventually surpass expert pathologists to improve clinical workflows. Methods: We combined supervised learning (SL) with semi-supervised learning (SSL) to produce an end-to-end multi-level skin detection system that not only detects 5 main types of skin lesions with high sensitivity and specificity, but also subtypes, localizes, and provides margin status to evaluate the proximity of the lesion to non-epidermal margins. The Supervised Training Subset consisted of 2188 random WSIs collected by the PathologyWatch (PW) laboratory between 2013 and 2018, while the Weakly Supervised Subset consisted of 5161 WSIs from daily case specimens. The Validation Set consisted of 250 curated daily case WSIs obtained from the PW tissue archives and included 50 "mimickers". The Testing Set (3821 WSIs) was composed of non-curated daily case specimens collected from July 20, 2021 to August 20, 2021 from PW laboratories. Results: The performance characteristics of our AI model (i.e., Mihm) were assessed retrospectively by running the Testing Set through the Mihm Evaluation Pipeline. Our results show that the sensitivity of Mihm in classifying melanocytic lesions, basal cell carcinoma, and atypical squamous lesions, verruca vulgaris, and seborrheic keratosis was 98.91% (95% CI: 98.27%, 99.55%), 97.24% (95% CI: 96.15%, 98.33%), 95.26% (95% CI: 93.79%, 96.73%), 93.50% (95% CI: 89.14%, 97.86%), and 86.91% (95% CI: 82.13%, 91.69%), respectively. Additionally, our multi-level (i.e., patch-level, ROI-level, and WSI-level) detection algorithm includes a qualitative feature that subtypes lesions, an AI overlay in the front-end digital display that localizes diagnostic ROIs, and reports on margin status by detecting overlap between lesions and non-epidermal tissue margins. Conclusions: Our AI model, developed in collaboration with dermatopathologists, detects 5 skin lesion types with higher sensitivity than previously published AI models, and provides end users with information such as subtyping, localization, and margin status in a front-end digital display. Our end-to-end system has the potential to improve pathology workflows by increasing diagnostic accuracy, expediting the course of patient care, and ultimately improving patient outcomes.
RESUMEN
BACKGROUND: Resveratrol (RESV) is a naturally occurring compound that may possess anticancer capabilities in both prostate carcinoma and melanoma. METHODS: The in vitro and in vivo cytotoxic activity of RESV and 3,5-dihydroxy-4'-acetoxy-trans-stilbene (4-ACE) was tested using cellular assays and a xenograft model. Five prostate carcinoma cell lines were used for in vitro evaluation. A melanoma cell line (Duke melanoma 738 [DM738]) and the prostate carcinoma line CWR22 were used for in vivo experiments. Mice were randomized to osmotic mini pumps with 200 µL of RESV (250 mg/mL), 4-ACE (335 mg/mL), or vehicle (50% dimethyl sulfoxide, 50% polyethylene glycol). Serum drug and metabolite levels were calculated by high-performance liquid chromatography with diode-array detection. Western blots were performed on treated tumors. Results were analyzed using a student's t-test, analysis of variance, and the Mann-Whitney rank sum test. RESULTS: RESV and 4-ACE were cytotoxic in a time- and dose-dependent manner in all prostate carcinoma cell lines tested. Enhanced growth compared with controls was seen at the 24 h time point in four lines treated with RESV and two lines treated with 4-ACE (Ps < 0.048). In vivo, no difference in either tumor growth or postmortem tumor weight was detected in either DM738 (P = 0.555, P = 0.562) or CWR22 (P = 0.166, P = 0.811) xenografts treated with either drug. Serum drug levels did not correlate with tumor growth rates for any treatment group (all Ps > 0.11). Treated tumors demonstrated protein changes by western blot. CONCLUSION: Although in vitro data were promising, RESV and 4-ACE have limited potential as single agents in the treatment of prostate carcinoma and melanoma.
Asunto(s)
Antineoplásicos/uso terapéutico , Melanoma/tratamiento farmacológico , Neoplasias de la Próstata/tratamiento farmacológico , Neoplasias Cutáneas/tratamiento farmacológico , Estilbenos/uso terapéutico , Ensayos Antitumor por Modelo de Xenoinjerto , Animales , Línea Celular Tumoral , Proliferación Celular , Modelos Animales de Enfermedad , Relación Dosis-Respuesta a Droga , Humanos , Técnicas In Vitro , Masculino , Melanoma/patología , Ratones , Ratones Desnudos , Neoplasias de la Próstata/patología , Resveratrol , Neoplasias Cutáneas/patología , Factores de Tiempo , Resultado del TratamientoRESUMEN
CONTEXT: Struma ovarii exhibiting malignant histology are uncommon, and an aggressive clinical course in the form of initial extraovarian spread or recurrence is even more exceptional for these tumors. OBJECTIVE: To determine whether specific histologic features have predictive value in distinguishing clinically benign from clinically malignant struma ovarii. DESIGN: Blinded analysis of 19 histologic characteristics of thyroid tumors was performed in 60 clinically benign and 26 clinically malignant struma ovarii cases, with long-term follow-up. RESULTS: Except for lack of fibrosis and macrofollicular pattern, which were more common in biologically malignant tumors (P â=â .04 and P â=â .008, respectively), and trabecular pattern, which was associated with a benign clinical course (P â=â .03), none of the other histologic features was found to be correlated with clinical behavior. The presence of the following features was similar in the biologically benign and malignant tumors: papillae, pseudo-papillae, psammoma bodies, nuclear grooves, nuclear overlap, "orphan Annie" nuclei, nuclear pseudo-inclusions, prominent nucleoli, hypercellularity, colloid scalloping, eosinophilic cytoplasm, mitoses, vascular invasion, cytologic atypia, nuclear pleomorphism, and cell size and type. Trabecular pattern and absence of fibrosis were uncommon, and there was considerable overlap of macrofollicular pattern ratio between benign and malignant cases. Thus, their practical usefulness is uncertain. CONCLUSIONS: The clinical outcome of struma ovarii cannot be predicted based on the microscopic diagnosis of the thyroid tissue or on specific histologic features. The lack of correlation between morphology and outcome in proliferative and histologically malignant struma ovarii is striking, making the behavior of these tumors particularly unpredictable.
Asunto(s)
Neoplasias Ováricas/patología , Ovario/patología , Estruma Ovárico/patología , Glándula Tiroides/patología , Neoplasias de la Tiroides/patología , Adulto , Femenino , Fibrosis , Estudios de Seguimiento , Humanos , Persona de Mediana Edad , Recurrencia Local de Neoplasia/patología , Valor Predictivo de las PruebasRESUMEN
BACKGROUND: Resveratrol (RESV) is a naturally occurring compound that possesses anti-cancer capabilities. The goal of this study was to evaluate the potential of RESV as an adjunct to chemotherapy in melanoma treatment. METHODS: The in vitro and in vivo cytotoxic activity of RESV with or without chemotherapy was tested using cellular assays and a xenograft model. Two Duke melanoma cell lines (DM738, DM443) were used for both in vivo and in vitro experiments, and two nonmalignant human fibroblast lines (NHDF, HS68) were used for in vitro cellular assays. Xenografts were randomized to treatment arms and tumors measured to evaluate response. Results were analyzed using a Student's t-test and ANOVA. Western blots were performed on in vivo tissue. RESULTS: In vitro RESV significantly decreased melanoma cell viability in all lines tested (all P < 0.0001). Treatment of fibroblast cell lines revealed that RESV selectively spared NHDF and HS68 cells compared with its cytotoxic effects on melanoma cells (P < 0.0001). Treatment of malignant cells with 50 µM RESV and temozolomide (TMZ) for 72 h significantly enhanced cytotoxicity compared with treatment with TMZ alone (P < 0.0001). In vivo, however, there was no significant difference between any treatment arms (P = 0.65). CONCLUSION: RESV shows promise as a novel therapeutic in the management of melanoma for its selective anti-tumor activity in vitro. Translating in vitro results to in vivo models has proven difficult. Barriers thought to prevent such translation are identified, and a rationale for overcoming them is discussed.
Asunto(s)
Antineoplásicos/uso terapéutico , Melanoma/tratamiento farmacológico , Neoplasias Cutáneas/tratamiento farmacológico , Estilbenos/uso terapéutico , Animales , Línea Celular , Línea Celular Tumoral , Quimioterapia Adyuvante , Dacarbazina/análogos & derivados , Dacarbazina/uso terapéutico , Modelos Animales de Enfermedad , Quimioterapia Combinada , Humanos , Técnicas In Vitro , Melanoma/patología , Melfalán/uso terapéutico , Ratones , Ratones Desnudos , Ratones SCID , Resveratrol , Neoplasias Cutáneas/patología , Temozolomida , Resultado del Tratamiento , Ensayos Antitumor por Modelo de XenoinjertoRESUMEN
A 57-year-old man status post several myocardial infarcts and heart transplantation presented with a slowly growing violaceous plaque on his lateral left knee at the site of prior minor trauma. A biopsy revealed a suppurative dermatitis with associated pseudocarcinomatous epithelial hyperplasia. There were multiple non-pigmented eosinophilic organisms with clear cytoplasmic halos within the infiltrate. A methenamine silver stain showed round to ovoid organisms of slightly variable size. Rare uni-polar budding, some of which was broad based, was apparent. A few short hyphae with indeterminate septa were also noted. Fontana-Masson, mucicarmine, Alcian blue and Fite stains were all negative. These findings suggested a diagnosis of blastomycosis. However, a fungal culture grew colonies of Alternaria species. Alternariosis has been previously shown to possess morphologic characteristics that can simulate other fungal infections. To our knowledge, a striking similarity to blastomycosis, as seen in our case, has not been previously reported. Dermatopathologists should be aware that alternariosis may mimic blastomycosis, especially when hyphal forms are rare or absent in tissue specimens. Culture is necessary for definitive classification.
Asunto(s)
Alternaria/aislamiento & purificación , Alternariosis/diagnóstico , Blastomicosis/diagnóstico , Alternariosis/tratamiento farmacológico , Alternariosis/microbiología , Antifúngicos/uso terapéutico , Terapia Combinada , Procedimientos Quirúrgicos Dermatologicos , Diagnóstico Diferencial , Humanos , Huésped Inmunocomprometido , Inmunosupresores/efectos adversos , Rodilla , Masculino , Persona de Mediana Edad , Pirimidinas/uso terapéutico , Piel/microbiología , Piel/patología , Triazoles/uso terapéutico , VoriconazolRESUMEN
4'-Ester analogs of the disease preventative agent resveratrol were synthesized and evaluated for their potential as anti-melanoma and pancreatic cancer agents. A decarbonylative Heck coupling was used to assemble the protected stilbene core structure. The 4'-acetate and the palmitoate analogs demonstrated selective activity with DM443 and DM738 cells over normal NHDF cells.