Developing profiles for early pancreatic cancer through bioinformatics analysis

胰腺癌是英国癌症相关死亡的第五大原因。统计数据显示，如果肿瘤可以在早期阶段被发现，五年生存率将从3%显著增加到39%（癌症事实与数据，2021）。然而，由于缺乏针对通常无症状的早期胰腺癌患者的有效和特异性筛查测试，超过80%的胰腺癌病例在晚期被诊断，突出了识别用于早期癌症检测的敏感和特异性标志物的需求。研究表明，循环肿瘤细胞在胰腺癌的所有阶段都是可检测的，但根据用于富集的标志物，检测率不确定为11%-92%（Martini等人，2019年）;相反，对早期胰腺癌进展有反应的免疫细胞在外周血中表现出不同的特征，并可用于以非侵入性方式早期诊断胰腺癌（佩雷拉et al.，2020).该项目的重点是对早期胰腺癌病例产生的外周血RNA-seq数据进行生物信息学分析。我们将我们的技术与其他技术区分开来，其独特的卖点如下所示。(1)与检测肿瘤DNA相比，具有高灵敏度和特异性（49.5%的准确度; Klein et al.，（2）使用大型训练数据集以避免年龄、性别、疾病状态（例如，糖尿病，胰腺炎）;（3）从不同类型的样品（例如，血液，活组织检查）和数据（例如，RNA、DNA、蛋白质等）以满足独特的诊断需求。我们有以下目标。(1)升级我们的生物信息学软件包，可用于自动分类和预测早期癌症病例（2）验证同意对我们产品进行beta测试的早期采用者的样本中产生的生物标志物。(3)申请专利以保护具有诊断潜力的新生物标志物。项目完成后，我们可以将经验证的生物标志物授权给我们的生物技术和临床合作伙伴，用于诊断产品的商业化。这将产生一个明确的临床路径，从血液样本，通过数据分析，诊断结果和个性化治疗，以满足胰腺癌患者的个体需求。

Pancreatic cancer is the 5th leading cause of cancer-related death in the UK. Statistics have shown that the five-year survival rate will be markedly increased from 3% to 39% if tumours can be detected at early stages (Cancer Facts & Figures, 2021). However, due to the lack of validated and specific screening tests for early-stage pancreatic cancer patients who are usually asymptomatic, more than 80% of the pancreatic cancer cases are diagnosed at advanced stage, highlighting the needs to identify sensitive and specific markers for detection of the cancer at early stages. Studies have shown that circulating tumour cells remain detectable across all stages of pancreatic cancer, but with an uncertain detection rate of 11%-92% depending on the markers used for enrichment (Martini et al., 2019); in contrast, immune cells that respond to early pancreatic cancer progression exhibit distinct characteristics in peripheral blood and can be exploited for early diagnosis of pancreatic cancer in a non-invasive manner (Pereira et al., 2020).This project focuses on bioinformatics analysis of the peripheral blood RNA-seq data generated from early pancreatic cancer cases. We differentiate our technology from others with the unique selling points listed as follows. (1) High sensitivity and specificity compared to detecting tumour DNA (49.5% accuracy; Klein et al., 2021); (2) Using large training data sets to avoid demographic differences regarding age, sex, disease status (e.g., diabetes, pancreatitis); (3) Ability of extracting novel biomarkers from different types of samples (e.g., blood, biopsy) and data (e.g., RNA, DNA, protein, etc.) to meet unique diagnostic needs.We have the following objectives. (1) Upgrading our bioinformatics software package that can be used to automatically classify and predict early cancer cases (2) Validating the resulting biomarkers in samples from our early adopters who agree to beta test our product. (3) Filing patents to protect the new biomarkers with diagnostic potential.Upon completion of the project, we are able to license the validated biomarkers to our biotech and clinical partners for commercialisation of diagnostic products. This will result in a clear clinical pathway from the blood sample, through data analysis, to diagnostic outcome and personalised therapy that meets the needs of pancreatic cancer patients on an individual basis.