医学
胰腺导管腺癌
免疫组织化学
腺癌
病理
免疫染色
普通外科
内科学
癌症
胰腺癌
作者
Michael Muggilli,Donna Russell,Zhongren Zhou
出处
期刊:CytoJournal
[Medknow]
日期:2021-11-03
卷期号:18: 28-28
被引量:7
标识
DOI:10.25259/cytojournal_78_2020
摘要
Objectives: Pancreatic ductal adenocarcinoma (PDAC) has a poor prognosis with surgery or chemotherapy. Programmed death ligand 1 expression (PD-L1) immunotherapy has been successful for treating lung and other cancers with PD-L1 expression. However, in many unresectable PDAC cases, cytological samples are the only available tissues for PD-L1 testing. The aim of this study is to retrospectively compare the expression of PD-L1 using cytological and surgical samples. Material and Methods: Paired formalin-fixed cell blocks and surgical samples from the same patients with confirmed diagnoses of PDAC ( n = 28) were sectioned for PD-L1 immunohistochemistry. Using tumor proportion score (TPS) and combined positive score (CPS) to evaluate paired cell blocks and surgical samples, we counted and analyzed the data. Results: With TPS, the PD-L1 was expressed in 9/28 (32%) of PDAC surgical samples and in 9/28 (32%) of paired cytological samples. Overall, the PD-L1 expression had a correlation of 26/28 (93%). With CPS, the PD-L1 was expressed in 20/28 (71%) of PDAC surgical samples and in 16/28 (57%) of paired cytological samples. The PD-L1 expression had a correlation of 20/28 (71%) and a discrepancy of 8/28 (29%). The PD-L1 expression was significantly higher in moderately-differentiated PDAC than in well-differentiated with TPS. Conclusion: Cytological samples are useful for evaluating PD-L1 expression with TPS because the concordant rate was 93%. With CPS, cytological samples are limited due to the scant inflammatory cells with the concordant rate of 71%. Extensive sampling of the pancreatic tumor may improve the detection of immune cells expressing PD-L1 in cytological samples. With TPS, PD-L1 expression was significantly higher in moderate-differentiation of PDAC than in poor- and well-differentiation.
科研通智能强力驱动
Strongly Powered by AbleSci AI