PDF(Pubmed)
Abstract：
BACKGROUND: Frozen section analysis of oral cancer specimens is ideal for assessing margin distances and depth of invasion (DOI); the latter impacts intraoperative decisions regarding elective neck dissection (END). Here, we show that intraoperative determination of worst pattern of invasion (WPOI), specifically WPOI-5, has a high level of accuracy. This relates to our demonstration herein that WPOI-5 predicts occult cervical metastases (OCM) for pT1 oral squamous carcinoma (OSC).
METHODS: The presence of OCM was correlated with WPOI in 228 patients with primary T1/T2/cN0 OSC undergoing resection and END. Concordance between intraoperative and final pathology WPOI determination was assessed on 51 cases of OSC.
RESULTS: WPOI-5 predicts OCM in pT1 patients, compared with WPOI-4/WPOI-3 (p < 0.0001). Most pT1 WPOI-5 tumors had DOI of 4-5 mm (24/59 or 40.7%). Only two pT1 WPOI-5 tumors had DOI < 4 mm (3.0 and 3.5 mm). If END were performed in this pT1 cohort for all WPOI-5 OSC patients regardless of DOI, OR all OSC patients with DOI ≥ 4 mm regardless of WPOI, then no OCM would be missed (p = 0.017, 100% sensitivity, 29% specificity, 77% positive predictive value, 23% negative predictive value). With respect to intraoperative WPOI-5 determination, the accuracy, sensitivity, and specificity was 92.16, 73.33, and 100.0%, respectively.
CONCLUSIONS: DOI ≥ 4 mm is the dominant predictor of OCM. For the rare WPOI-5 OSC with DOI < 4 mm, it is reasonable to suggest that surgeons perform END. WPOI-5 may be accurately determined intraoperatively. As microscopic instruction is needed to accurately assess WPOI-5, a teaching link is included in this manuscript.
METHODS: The presence of OCM was correlated with WPOI in 228 patients with primary T1/T2/cN0 OSC undergoing resection and END. Concordance between intraoperative and final pathology WPOI determination was assessed on 51 cases of OSC.
RESULTS: WPOI-5 predicts OCM in pT1 patients, compared with WPOI-4/WPOI-3 (p < 0.0001). Most pT1 WPOI-5 tumors had DOI of 4-5 mm (24/59 or 40.7%). Only two pT1 WPOI-5 tumors had DOI < 4 mm (3.0 and 3.5 mm). If END were performed in this pT1 cohort for all WPOI-5 OSC patients regardless of DOI, OR all OSC patients with DOI ≥ 4 mm regardless of WPOI, then no OCM would be missed (p = 0.017, 100% sensitivity, 29% specificity, 77% positive predictive value, 23% negative predictive value). With respect to intraoperative WPOI-5 determination, the accuracy, sensitivity, and specificity was 92.16, 73.33, and 100.0%, respectively.
CONCLUSIONS: DOI ≥ 4 mm is the dominant predictor of OCM. For the rare WPOI-5 OSC with DOI < 4 mm, it is reasonable to suggest that surgeons perform END. WPOI-5 may be accurately determined intraoperatively. As microscopic instruction is needed to accurately assess WPOI-5, a teaching link is included in this manuscript.
摘要：
背景：口腔癌标本的冰冻切片分析是评估边缘距离和浸润深度（DOI）的理想选择；后者影响选择性颈清扫术（END）的术中决策。这里,我们表明，术中确定最坏的侵袭模式（WPOI），特别是WPOI-5，具有较高的精度。这与我们在本文中的证明有关，即WPOI-5预测pT1口腔鳞状细胞癌（OSC）的隐匿性宫颈转移（OCM）。
方法：在228例原发性T1/T2/cN0OSC切除和END患者中，OCM的存在与WPOI相关。对51例OSC进行了术中和最终病理WPOI测定之间的一致性评估。
结果：WPOI-5预测pT1患者的OCM，与WPOI-4/WPOI-3相比（p<0.0001）。大多数pTlWPOI-5肿瘤的DOI为4-5mm(24/59或40.7%)。只有两个pTlWPOI-5肿瘤的DOI<4mm(3.0和3.5mm)。如果在此pT1队列中对所有WPOI-5OSC患者进行END，而不考虑DOI，或所有DOI≥4mm的OSC患者，无论WPOI如何，则不会错过OCM(p=0.017，100%灵敏度，29%的特异性，77%的阳性预测值，23%的阴性预测值)。关于术中WPOI-5的测定，准确性,灵敏度,特异性分别为92.16、73.33和100.0%，分别。
结论：DOI≥4mm是OCM的主要预测因子。对于DOI<4mm的稀有WPOI-5OSC，建议外科医生进行END是合理的。WPOI-5可以在手术中准确地确定。由于需要微观指导才能准确评估WPOI-5，因此本手稿中包含了教学链接。
方法：在228例原发性T1/T2/cN0OSC切除和END患者中，OCM的存在与WPOI相关。
结果：WPOI-5预测pT1患者的OCM，
结论：DOI≥4mm是OCM的主要预测因子。
