Faculty Bibliography

SIGNIFICANCE/CONCLUSIONS:Melanoma is a deadly cancer that physicians struggle to diagnose early because they lack the knowledge to differentiate benign from malignant lesions. Deep machine learning approaches to image analysis offer promise but lack the transparency to be widely adopted as stand-alone diagnostics. AIM/OBJECTIVE:We aimed to create a transparent machine learning technology (i.e., not deep learning) to discriminate melanomas from nevi in dermoscopy images and an interface for sensory cue integration. APPROACH/METHODS:Imaging biomarker cues (IBCs) fed ensemble machine learning classifier (Eclass) training while raw images fed deep learning classifier training. We compared the areas under the diagnostic receiver operator curves. RESULTS:Our interpretable machine learning algorithm outperformed the leading deep-learning approach 75% of the time. The user interface displayed only the diagnostic imaging biomarkers as IBCs. CONCLUSIONS:From a translational perspective, Eclass is better than convolutional machine learning diagnosis in that physicians can embrace it faster than black box outputs. Imaging biomarkers cues may be used during sensory cue integration in clinical screening. Our method may be applied to other image-based diagnostic analyses, including pathology and radiology.

Cutaneous squamous cell carcinoma (cSCC) causes approximately 10,000 deaths annually in the U. S. Current therapies are largely ineffective against metastatic and locally advanced cSCC. There is a need to identify novel, effective, and less toxic small molecule cSCC therapeutics. We developed a 3-dimensional bioprinted skin (3DBPS) model of cSCC tumors together with a microscopy assay to test chemotherapeutic effects in tissue. The full thickness SCC tissue model was validated using hematoxylin and eosin (H&E) and immunohistochemical histological staining, confocal microscopy, and cDNA microarray analysis. A nondestructive, 3D fluorescence confocal imaging assay with tdTomato-labeled A431 SCC and ZsGreen-labeled keratinocytes was developed to test efficacy and general toxicity of chemotherapeutics. Fluorescence-derived imaging biomarkers indicated that 50% of cancer cells were killed in the tissue after 1μM 5-Fluorouracil 48-hour treatment, compared to a baseline of 12% for untreated controls. The imaging biomarkers also showed that normal keratinocytes were less affected by treatment (11% killed) than the untreated tissue, which had no significant killing effect. Data showed that 5-Fluorouracil selectively killed cSCC cells more than keratinocytes. Our 3DBPS assay platform provides cellular-level measurement of cell viability and can be adapted to achieve nondestructive high-throughput screening (HTS) in bio-fabricated tissues.

There have been several significant advances in cancer treatment in the last decade that are applicable to the treatment of melanoma and advanced nonmelanoma skin cancers. Among these are the development of immune checkpoint inhibitors targeting the programmed death protein-1 (PD-1)/programmed death legand-1 (PDL-1) axis, as well as targeted inhibitors of the BRAF/MEK signaling cascade in melanoma, and the hedgehog signaling pathway in basal cell carcinoma (BCC). These immune-based and targeted therapies have dramatically changed the treatment options for locally advanced and metastatic melanoma, Merkel's cell carcinoma, cutaneous squamous cell carcinoma (cSCC), and BCC. In this article, we will briefly review the currently approved targeted and immunotherapy-based treatments for locally advanced and metastatic melanoma, Merkel's cell carcinoma, and cSCC and discuss various combinations of approved therapies, as well as emerging therapeutic candidates that are currently in clinical trials, including novel checkpoint inhibitors in development, intratumoral oncolytic agents (viral and nonviral), and various immune-based therapies such as toll-like receptor (TLR) agonists, adoptive T-cell therapy, T-cell costimulation, and innate immune cell therapy. For advanced BCC, we will discuss trials investigating the currently approved smoothened (SMO) inhibitors for neoadjuvant use, emerging SMO inhibitors in development, topical SMO inhibitors, alternative targets in the hedgehog signaling pathway, and the use of anti-PD-1 agents for advanced BCC both alone and in combination with SMO inhibitors.

BACKGROUND:Opioid overprescribing is a major contributor to the opioid crisis. The lack of procedure-specific guidelines contributes to the vast differences in prescribing practices. OBJECTIVE:Create opioid-prescribing consensus guidelines for common dermatologic procedures. METHODS:We utilized a four-step modified Delphi method to conduct a systematic discussion among a panel of providers in the fields of general dermatology, dermatologic surgery, and cosmetics/phlebology to develop opioid-prescribing guidelines for some of the most common dermatologic procedural scenarios. Guidelines were developed for opioid-naïve patients undergoing routine procedures. Opioid tablets were defined as oxycodone 5-mg oral equivalents. RESULTS:Postoperative pain after most uncomplicated procedures (76%) can be adequately managed with acetaminophen and/or ibuprofen. Group consensus identified no specific dermatological scenario that routinely requires more than 15 oxycodone 5-mg oral equivalents to manage postoperative pain. Group consensus found that 23 percent of the procedural scenarios routinely require 1-10 opioid tablets, while only one routinely requires 1-15 opioid tablets. LIMITATIONS/CONCLUSIONS:These recommendations are based on expert consensus in lieu of quality evidence-based outcomes research. These recommendations must be individualized to accommodate patients' comorbidities. CONCLUSIONS:Procedure-specific opioid-prescribing guidelines may serve as a foundation to produce effective and responsible postoperative pain management strategies after dermatologic interventions.

Perineural invasion is a pathologic process of neoplastic dissemination along and invading into the nerves. Perineural invasion is associated with aggressive disease and a greater likelihood of poor outcomes. In this study, 3 of 9 patients with cutaneous squamous cell carcinoma and perineural invasion exhibited poor clinical outcomes. Tumors from these patients expressed high levels of MAGE-A3, a cancer testis antigen that may contribute to key processes of tumor development. In addition to perineural invasion, the tumors exhibited poor differentiation and deep invasion and were subsequently classified as Brigham and Women's Hospital tumor stage 3. Cyclin E, A and B mRNA levels were increased in these tumors compared with normal skin tissues (102.93±15.03 vs. 27.15±4.59, 36.83±19.41 vs. 11.59±5.83, 343.77±86.49 vs. 95.65±29.25, respectively; p<0.05). A431 cutaneous squamous cell carcinoma cells pretreated with MAGE-A3 antibody exhibited a decreased percentage S-phase cells (14.13±2.8% vs. 33.97±1.1%; p<0.05) and reduced closure in scratch assays (43.88±5.49% vs. 61.17±3.97%; p = 0.0058). In a syngeneic animal model of squamous cell carcinoma, immunoblots revealed overexpression of MAGE-A3 and cyclin E, A, and B protein in tumors at 6 weeks. However, knockout of MAGE-A3 expression caused a reduction in tumor growth (mean tumor volume 155.3 mm3 vs. 3.2 mm3) compared with parental cells. These results suggest that MAGE-A3 is a key mediator in cancer progression. Moreover, elevated collagen XI and matrix metalloproteases 3, 10, 11, and 13 mRNA levels were observed in poorly differentiated cutaneous squamous cell carcinoma with perineural invasion compared with normal skin tissue (1132.56±882.7 vs. 107.62±183.62, 1118.15±1109.49 vs. 9.5±5, 2603.87±2385.26 vs. 5.29±3, 957.95±627.14 vs. 400.42±967.66, 1149.13±832.18 vs. 19.41±35.62, respectively; p<0.05). In summary, this study highlights the potential prognostic value of MAGE-A3 in clinical outcomes of cutaneous squamous cell carcinoma patients.

T-cell landscape differences between cutaneous squamous cell carcinoma (cSCC) tumors in immune competent (SCC in IC) and immunocompromised organ transplant recipients (TSCC in OTR) are unclear. We developed an analytical method to define tumor infiltrating lymphocyte (TIL) phenotype in cSCC from immune competent and immune suppressed patients using single-cell TCR sequencing and gene expression data. TSCC exhibits reduced proportions of cytotoxic and naïve TILs and similar numbers of regulatory TILs. Fewer, more heterogeneous TCR clonotypes are observed in TIL from OTR. Most TCR sequences for top ten clonotypes correspond to known antigens, while 24% correspond to putative neoantigens. OTR show increased cSCC events over 12 months possibly due to reduced cytotoxic T-cells. Our novel method of barcoding CD8+ T-cells is the first providing gene expression and TCR sequences in cSCC. Knowledge regarding putative antigens recognized by TCRs with phenotypic function of T-cells bearing those TCRs could facilitate personalized cSCC treatments.