Benign adnexal tumours associated with naevus sebaceous are dermoscopic mimickers of basal cell carcinomas. A pigmented nodule arising in a naevus sebaceous with a total blue large ovoid nest on dermoscopy could be a trichoblastoma, basal cell carcinoma, hidrocystoma or hidradenoma. Dermoscopy can be a useful diagnostic tool which generates a more accurate preoperative diagnosis.

<b><i>Background:</i></b> Preoperative diagnosis of malignant collision tumors (MCT) is extremely difficult. The value of dermoscopy to improve the correct detection of these tumors has not been previously studied. This study aims to evaluate the diagnostic accuracy of MCT with and without dermoscopy and to describe the dermoscopic features of a large series of MCT. <b><i>Methods:</i></b> Dermoscopic images of 161 MCT were evaluated. Clinical and dermoscopic images of histopathologically proven MCT intermingled with other tumors were randomly presented to clinicians with different levels of experience, blinded to the diagnosis and objective of the study. The clinical and dermoscopic diagnostic accuracies were measured separately. <b><i>Results:</i></b> A total of 161 histopathologically proven cases of MCT were collected. The most frequent MCT was basal cell carcinoma-seborrheic keratosis collision tumor (CT; 37.9%), followed by basal cell carcinoma-melanocytic nevus CT (19.9%), and melanoma-seborrheic keratosis CT (6.8%). Diagnostic accuracy among experts on dermoscopy was 71.4%. The study included 119 participants. The percentage of correct diagnoses was 8% by naked eye examination and 36.4% by dermoscopy (<i>p</i> &#x3c; 0.001). The presence of the malignant component in the cases of MCT was not recognizable in 19.1% of cases by naked eye examination and in 11.8% of cases by dermoscopy (<i>p</i> &#x3c; 0.001). <b><i>Conclusions:</i></b> The diagnosis of MCT can be assisted and clarified by dermoscopy. However, many of these lesions manifest complex morphologies and continue to be challenging, even for experts on dermoscopy. Atypical, uncertain, or non-classifiable lesions still need a complete excision for the final diagnosis.

Objective: The purpose of this study was to examine the efficacy and toxicity of treating small arteriovenous malformations (AVMs) (≤3 cm in diameter) with a median marginal applied dose of 14 Gy. Methods: Two hundred and thirteen patients diagnosed with AVMs were treated between January 1991 and December 2005. Seventy-three percent of the patients had hemorrhaged prior to treatment, 13% had had previous surgery and 19.2% had had previous embolization. The median follow-up duration was 48.1 months. Results: The Kaplan-Meier analysis estimated that the 36-month obliteration rate was 65.5% for patients undergoing their first stereotactic radiosurgery (SRS) and 68.3% for those undergoing repeated SRS. The Kaplan-Meier analysis estimated the 60-month AVMs obliteration rate for the entire cohort to be 82.4%. The median time to AVM obliteration was 40 ± 2.8 months. We found a statistically significant relationship between the time of obliteration and the following factors: site of the AVMs (sites other than brainstem), a higher prescribed dose and a positive history of previous hemorrhage. Thirteen patients (7.6%) experienced toxicities. Conclusions: SRS was an effective and safe treatment for AVMs ≤3 cm in diameter, with acceptable toxicity.

e14524 Background: Immunogenic cell death (ICD) is known for the release of DAMPS from tumor cells. We aimed to find signals of ICD by assessing the variation of plasma DAMPS (HMGB1 and S100A8) after vs. before standard of care (SoC) systemic treatment in patients with advanced solid tumors. Methods: Patients scheduled to start a new line of systemic treatment were included. Plasma concentrations of HMGB1 and S100A8 were measured (ng/mL) before and after three months of treatment. CD44 immunohistochemical (IHC) expression was determined in tumor tissue. After vs. before variation of paired median concentrations was analyzed with the Wilcoxon signed-rank test for the whole population and selected subgroups according to RECIST response, baseline plasmatic iron levels, CD44 expression, and platinum-based treatment. Results: Fifty-two patients were included. The most frequent tumor sites were colorectal (35%) and lung (25%). Forty-two patients (81%) received this treatment as first-line. Thirty-six patients (69%) received chemotherapy (CT) alone, ten (19%) CT plus targeted therapy (7 FOLFOX or XELOX plus bevacizumab, and one each FOLFOX plus cetuximab, FOLFOX plus panitumumab, docetaxel-trastuzumab-pertuzumab), two (3.8%) carboplatin-pemetrexed-pembrolizumab, three (5.8%) pembrolizumab alone and one (1.9%) cetuximab alone. Overall response rate (RECIST) was 42%, rate of patients with low baseline plasmatic iron levels was 53%, and CD44 expression was positive in 35% of the patients. Results on plasma concentration of DAMPS are shown in the table. Conclusions: Signals of ICD were not observed in these patients. HMGB1 variation was not significant while plasma concentration of S100A8 significantly decreased after treatment, more markedly in those patients who experienced tumor response.[Table: see text]