Renal mass biopsy using Raman spectroscopy identifies malignant and benign renal tumors: potential for pre-operative diagnosis.
Renal mass biopsy using Raman spectroscopy identifies malignant and benign renal tumors: potential for pre-operative diagnosis.
复制标题
DOI：
10.18632/oncotarget.16419
复制
发表时间：
2017-05-30
期刊：
影响因子：
--
通讯作者：
Jiang H
中科院分区：
文献类型：
--
作者：
Liu Y;Du Z;Zhang J;Jiang H
肾肿块活检诊断恶性肿瘤的准确性受多种因素的影响。在这里，我们研究了拉曼光谱技术用于区分肾肿瘤活检标本良恶性的可行性。样本来自63例接受根治性或部分肾切除术的患者，疑似癌症的肿块和切除的肾脏远端实质用18号活检针采集。每个样品获得四个拉曼光谱，并应用判别分析进行数据分析。最终共收集到383个拉曼光谱，每种类型的肿瘤都有自己的特征光谱。拉曼光谱对肿瘤和正常组织的识别准确率为82.53%，对良、恶性肿瘤的识别灵敏度为91.79%，特异度为71.15%。对低、高级别肿瘤的分类准确率为86.98%。透明细胞癌与嗜酸细胞瘤、血管平滑肌脂肪瘤的鉴别准确率分别为100%和89.25%。组织学亚型判别准确率为93.48%。与最终的病理和活检相比，拉曼光谱能够正确识别11例“漏诊”的活检诊断中的7例。这些结果表明，拉曼光谱可能成为一种有前途的无创性的术前诊断方法。
The accuracy of renal mass biopsy to diagnose malignancy can be affected by multiple factors. Here, we investigated the feasibility of Raman spectroscopy to distinguish malignant and benign renal tumors using biopsy specimens. Samples were collected from 63 patients who received radical or partial nephrectomy, mass suspicious of cancer and distal parenchyma were obtained from resected kidney using an 18-gauge biopsy needle. Four Raman spectra were obtained for each sample, and Discriminant Analysis was applied for data analysis. A total of 383 Raman spectra were eventually gathered and each type of tumor had its characteristic spectrum. Raman could separate tumoral and normal tissues with an accuracy of 82.53%, and distinguish malignant and benign tumors with a sensitivity of 91.79% and specificity of 71.15%. It could classify low-grade and high-grade tumors with an accuracy of 86.98%. Besides, clear cell renal carcinoma was differentiated with oncocytoma and angiomyolipoma with accuracy of 100% and 89.25%, respectively. And histological subtypes of cell carcinoma were distinguished with an accuracy of 93.48%. When compared with final pathology and biopsy, Raman spectroscopy was able to correctly identify 7 of 11 “missed” biopsy diagnoses. These results suggested that Raman may serve as a promising non-invasive approach in the future for pre-operative diagnosis.