Deciphering dermatological distinctions: Cornulin as a discriminant biomarker between Basal Cell Carcinoma and Squamous Cell Carcinoma detected through E-biopsy and Machine Learning

Background: Clinical misdiagnosis between cutaneous squamous cell carcinoma (cSCC) and basal cell carcinoma (BCC) poses treatment challenges and carries risks of recurrence, metastases, and increased morbidity and mortality. Objective: We aimed to identify discriminant proteins markers for cSCC and BCC using a minimally invasive proteome sampling method called e-biopsy, employing electroporation for non-thermal cell permeabilization and machine learning. Methods: E-biopsy facilitated ex vivo proteome extraction from 21 cSCC and 21 BCC pathologically validated human cancers. LC/MS/MS profiling of 126 proteomes was followed by Machine Learning analysis to identify proteins distinguishing cSCC from BCC. For identified panel validation, we used proteomes sampled by e-biopsy from unrelated 20 cSCC and 46 BCC human cancers, and differential expression analysis of published transcriptomics. The most commonly chosen discriminant biomarker by machine learning models, cornulin, was also validated using fluorescent immunohistochemistry. Results: 192 proteomes sampled from 108 patients were analyzed. Machine Learning-based approaches resulted in a set of 11 potential biomarker proteins that can be used to construct a model with 95.2% average cross-validation accuracy, BCC precision of 93.6±14.5%, cSCC precision of 98.4±7.2%, specificity of 97.7±11.8%, and per-patient sensitivity 92.7±15.3%. Protein-protein interaction analysis revealed a novel interaction network connecting 10 of the 11 resulted proteins. Histological and transcriptomic validation confirmed cornulin as a discriminant marker significantly lower in cSCC than in BCC. Conclusions: E-biopsy combined with machine learning provides a novel approach to molecular biomarkers sampling from skin for biomarker detection and differential expression analysis between cSCC and BCC

背景：临床中皮肤鳞状细胞癌（cutaneous squamous cell carcinoma, cSCC）与基底细胞癌（basal cell carcinoma, BCC）的误诊会给治疗带来困难，并存在复发、转移风险，同时会增加患者的发病率与死亡率。
目的：本研究旨在通过一种名为e-biopsy的微创蛋白质组采样方法，结合非热细胞透化电穿孔技术与机器学习，筛选出可区分cSCC与BCC的鉴别性蛋白标志物。
方法：研究人员利用e-biopsy对经病理学验证的21例cSCC和21例BCC人类癌组织进行离体蛋白质组提取。对126份蛋白质组开展液相色谱-串联质谱（LC/MS/MS）分析后，通过机器学习分析筛选出可区分cSCC与BCC的蛋白。为验证所筛选的标志物组合，研究人员使用了从另外20例cSCC和46例BCC人类癌组织中通过e-biopsy采集的蛋白质组，并对已发表的转录组学数据进行差异表达分析。机器学习模型最常筛选出的鉴别性生物标志物——丝聚蛋白原（cornulin），还通过荧光免疫组化完成了验证。
结果：本研究共分析了来自108例患者的192份蛋白质组样本。基于机器学习的方法筛选出11种潜在生物标志物蛋白，可构建出平均交叉验证准确率达95.2%的模型，其中BCC的精准度为93.6±14.5%，cSCC的精准度为98.4±7.2%，特异性为97.7±11.8%，患者水平灵敏度为92.7±15.3%。蛋白质相互作用分析揭示了连接该11种蛋白中10种的全新相互作用网络。组织学与转录组学验证证实，丝聚蛋白原（cornulin）在cSCC中的表达量显著低于BCC，可作为鉴别标志物。
结论：e-biopsy联合机器学习为从皮肤组织中采集分子生物标志物，并开展cSCC与BCC之间的差异表达分析与生物标志物检测提供了一种全新的方法。