Head and Neck Pathology, King's College London, Guy's Hospital, London, SE1 9RT, United Kingdom.
Department of Oral and Maxillofacial Clinical Sciences, Faculty of Dentistry, University of Malaya, 50603, Kuala Lumpur, Malaysia.
Sci Rep. 2018 Oct 26;8(1):15874. doi: 10.1038/s41598-018-34165-5.
The value of image cytometry DNA ploidy analysis and dysplasia grading to predict malignant transformation has been determined in oral lesions considered to be at 'high' risk on the basis of clinical information and biopsy result. 10-year follow up data for 259 sequential patients with oral lesions clinically at 'high' risk of malignant transformation were matched to cancer registry and local pathology database records of malignant outcomes, ploidy result and histological dysplasia grade. In multivariate analysis (n = 228 patients), 24 developed carcinoma and of these, 14 prior biopsy samples were aneuploid. Aneuploidy was a significant predictor (hazard ratio 7.92; 95% CI 3.45, 18.17) compared with diploidy (p < 0.001). The positive predictive value (PPV) for severe dysplasia was 50% (95% CI 31.5, 68.5) and for aneuploid lesions, 33.3% (95% CI 19.0, 47.6). Combined DNA aneuploidy and severe dysplasia increased PPV to 56.3% (95% CI 31.9, 80.6). Diploid-tetraploid and non-dysplastic status had high negative predictive values (NPV) of 94.6% (95% CI 91.4, 97.8) and 99.17% (95% CI 97.4, 100.8) respectively. DNA ploidy predicts malignant transformation well and combining it with dysplasia grading gave the highest predictive value. The predictive values reported here exceed those from other investigations to date.
基于临床信息和活检结果,将被认为具有“高”恶变风险的口腔病变进行图像细胞 DNA 倍体分析和不典型增生分级,以确定其对恶性转化的预测价值。对 259 例临床具有“高”恶变风险的口腔病变患者进行了 10 年的随访,将其与癌症登记处和局部病理学数据库中的恶性结果、倍体结果和组织学不典型增生分级记录进行了匹配。在多变量分析(n=228 例患者)中,24 例发展为癌,其中 14 例先前的活检样本为非整倍体。非整倍体是一个显著的预测因子(风险比 7.92;95%可信区间 3.45, 18.17),与二倍体相比(p<0.001)。重度不典型增生的阳性预测值(PPV)为 50%(95%可信区间 31.5, 68.5),非整倍体病变的 PPV 为 33.3%(95%可信区间 19.0, 47.6)。DNA 非整倍体和重度不典型增生联合后,PPV 增加至 56.3%(95%可信区间 31.9, 80.6)。二倍体-四倍体和非不典型增生状态的阴性预测值(NPV)分别为 94.6%(95%可信区间 91.4, 97.8)和 99.17%(95%可信区间 97.4, 100.8)。DNA 倍体很好地预测了恶性转化,将其与不典型增生分级相结合可获得最高的预测价值。本研究报告的预测值高于迄今为止其他研究的预测值。