在II-III期肿瘤中对局部晚期直肠癌患者进行术前放化疗(RCT)已被证明在很高比例的病例中是有效的。尽管如此,20-30%的患者没有反应甚至疾病进展。目前,术前反应通过影像学和肿瘤组织病理学回归的组合来评估,但是最近的研究表明,各种基因异常可能与直肠癌肿瘤细胞对新辅助治疗的敏感性或耐药性有关。在本研究中,我们调查了高密度单核苷酸多态性(SNP)阵列6.0检测到的遗传病变与对新辅助RCT的反应之间的关系,治疗前根据Dworak标准对39例直肠癌肿瘤进行了评估。检测到的拷贝数(CN)丢失的最高频率对应于染色体18q(n=27;69%),1p(n=22;56%),15q(n=19;49%),8p(n=18;48%),4q(n=17;46%),和22q(n=17;46%);反过来,CN增益更频繁地涉及染色体20p(n=22;56%),8p(n=20;51%),和15q(n=16;41%)。在1p的改变之间存在显着关联,3q,7q,12p,17q,20p,和22q染色体区域以及手术前对治疗的反应程度。然而,4q,15q11.1和15q14染色体区域改变被五种预测算法确定为重要的,即,那些对预测术前RCT治疗的肿瘤反应影响最大的患者。预后因素的多因素分析显示,诊断时血清15q11.1和癌胚抗原(CEA)水平的增加是预测无病生存(DFS)的唯一独立变量。在多变量分析中,淋巴结受累也显示出对总体生存率(OS)的预后影响。基于深度学习的算法在诊断疾病后60个月预测DFS和OS的成功率为100%。总之,我们的结果表明在诊断时肿瘤遗传异常之间存在关联,对新辅助治疗的反应,以及局部晚期直肠癌患者的生存率。除了局部晚期直肠癌患者的临床和生物学特征外,这些可以在未来用作治疗和预后生物标志物,为了识别对术前治疗敏感或耐药的患者,帮助指导治疗决策。需要在更大系列患者中进行其他前瞻性研究,以确认新鉴定的生物标志物的临床实用性。
Administering preoperative radiochemotherapy (RCT) in stage II-III tumors to locally advanced rectal carcinoma patients has proved to be effective in a high percentage of cases. Despite this, 20-30% of patients show no response or even disease progression. At present, preoperative response is assessed by a combination of imaging and tumor regression on histopathology, but recent studies suggest that various genetic abnormalities may be associated with the sensitivity or resistance of rectal cancer tumor cells to neoadjuvant therapy. In the present study we investigated the relationship between genetic lesions detected by high-density single-nucleotide polymorphisms (SNP) arrays 6.0 and response to neoadjuvant RCT, evaluated according to Dworak criteria in 39 rectal cancer tumors before treatment. The highest frequency of copy-number (CN) losses detected corresponded to chromosomes 18q (n = 27; 69%), 1p (n = 22; 56%), 15q (n = 19; 49%), 8p (n = 18; 48%), 4q (n = 17; 46%), and 22q (n = 17; 46%); in turn, CN gains more frequently involved chromosomes 20p (n = 22; 56%), 8p (n = 20; 51%), and 15q (n = 16; 41%). There was a significant association between alterations in the 1p, 3q, 7q, 12p, 17q, 20p, and 22q chromosomal regions and the degree of response to therapy prior to surgery. However, 4q, 15q11.1, and 15q14 chromosomal region alterations were identified as important by five prediction algorithms, i.e., those with the greatest influence on predicting the tumor response to treatment with preoperative RCT. Multivariate analysis of prognostic factors showed that gains on 15q11.1 and carcinoembryonic antigen (CEA) levels serum at diagnosis were the only independent variables predicting disease-free survival (DFS). Lymph node involvement also showed a prognostic impact on overall survival (OS) in the multivariate analysis. A deep-learning-based algorithm showed a 100% success rate in predicting both DFS and OS at 60 months after diagnosis of the disease. In summary, our results indicate the existence of an association between tumor genetic abnormalities at diagnosis, response to neoadjuvant therapy, and survival of patients with locally advanced rectal cancer. In addition to the clinical and biological characteristics of locally advanced rectal cancer patients, these could be used in the future as therapeutic and prognostic biomarkers, to identify patients sensitive or resistant to preoperative treatment, helping guide therapeutic decision-making. Additional prospective studies in larger series of patients are required to confirm the clinical utility of the newly identified biomarkers.