Summary: difficulties and key points of PD-L1 immunohistochemistry

< br / >PD-L1 is a transmembrane protein expressed on T cells, B cells and tumor cellsThe study shows that when PD-L1 on tumor cell membrane combines with PD-1 on T cells and other immune cells, tumor cells send inhibitory signals, and then T cells can not recognize tumor cells and have killing effect on tumor cells, so the immune function of the body is inhibitedHow to select patients who may benefit from PD-1 / PD-L1 inhibitor therapy is the most concerned problem in clinical practice< br / > PD-L1 immunohistochemistry is a simple and effective method to predict the efficacy of PD-1 / PD-L1 inhibitors< br / > at present, the indications, PD-L1 detection platform and interpretation methods of PD-1 / PD-L1 inhibitors approved by FDA / nmpa are shown in the table below< br / > PD-L1 has been approved by FDA as a concomitant or supplementary diagnosis of immunotherapyAt present, there are five kinds of PD-L1 immunohistochemistry detection kits / antibodies: 22c3, 28-8, sp263, sp142 and 73-10, which are detected on two immunohistochemistry platforms Dako and Ventana respectively< br / > < br / > FDA only approved the PD-L1 detection of Dako 22c3 pharmdx as the concomitant diagnosis of drug K, while Dako 28-8 and Ventana sp142 were the supplementary diagnosis of drug O and drug t (atezolizumab), respectivelyVentana sp263 is certified by the European Union as a complementary diagnosis of three immunosuppressants< br / >2 different interpretation standards of antibody detection resultsat present, the main interpretation methods used for PD-L1 antibody detection are TPS, CPS, TC and IC respectively< br / >CPS is also recommended in cervical cancer and urothelial cancerTherefore, PD-L1 interpretation needs a lot of training from a professional pathologist to ensure the accuracy of interpretation< br / > < br / > in addition, different types of samples (fresh or paraffin section), different sources of samples (primary or metastatic sites), different sampling locations of tumor samples will lead to certain differences in the detection results < br / > therefore, different immunotherapy drugs need to be tested with different antibodies Meanwhile, the detection platform, quality control system and personnel's interpretation experience of PD-L1 are highly required In recent years, there have been many reports about the consistency of different detection kits / antibodies in the world < br / > < br / > the expression of PD-L1 in tumor cells and immune cells was detected by 22c3, 28-8, sp142 and e1l3n respectively in 90 NSCLC resected specimens The percentage of staining was evaluated by 13 pathologists The results showed that PD-L1 detected by sp142 antibody in tumor cells and immune cells was significantly reduced, and the consistency of tumor cell score detected by different antibodies was 0.813 (95% CI: 0.815-0.839), while the consistency of immune cell score was 0.277 (95% CI: 0.222-0.334) < br / > < br / > a comparative study of staining patterns of 22c3, 28-8, sp142 and sp263 in adenocarcinoma and squamous cell carcinoma of the lung was conducted in Germany in 2016 In 15 specimens of surgical resection, the proportion of tumor cells stained with 28-8 and 22c3 monoclonal antibodies was similar; in contrast, sp142 showed less staining in 4 specimens, and sp263 showed more staining proportion in 9 specimens < br / > < br / > in 2017, the blue print project was also carried out to evaluate the consistency of these four antibodies The experimental results of 39 NSCLC samples showed that when 22c3, 28-8 and sp263 were used for analysis, the percentage of tumor cells stained was comparable, while the detection results of sp142 showed that the total number of tumor cells stained was less < br / > < br / > blue print plan is to find out whether there is a high consistent rate of PD-L1 expression in tumor cells detected by different immunohistochemistry antibodies The consistency of four antibodies (28-8, 22c3, sp263 and sp142) was evaluated in the study of blueprint phase 1 (BP1) The results showed that the three antibodies, 28-8, 22c3 and sp263, had a high consistency in the expression of PD-L1 in tumor cells, while the sensitivity of sp142 was low and the consistency with the other three antibodies was poor < br / > < br / > of course, the blue print phase I study has certain limitations: first, all cases are surgical resection specimens, while most of the samples to be tested in the real world are small biopsy specimens, which are not involved in phase I study; second, only three pathologists participated in the interpretation of PD-L1 immunohistochemistry in phase I study, which is not statistically reliable < br / > < br / > < br / > < br / > < br / > bp2a was evaluated by 25 experienced pathologists In addition to surgical resection specimens, puncture specimens and cytological specimens were added The above results of bp2a phase I were verified in a larger cohort of clinical practice specimens < br / > Thirty one cases were included, all of which contained tissue wax block, hollow core needle or clamp biopsy, and fine needle puncture cell block embedding < br / > < br / > there are some differences between German consistency study and blue print plan in the comparability conclusion of sp263 and 22c3, 28-8 McAb, which may be caused by too few samples In the same year, Ratcliffe used 22c3, 28-8 and sp263 to detect 500 NSCLC archived samples The percentage of tumor cell membrane positive staining including 1%, 10%, 25% and 50% cutoff values was set as the PD-L1 positive standard The overall coincidence rate between different monoclonal antibodies was more than 90% < br / > < br / > based on the above results, it can be found that the detection results of 22c3, 28-8 and sp263 have similar positive percentage of tumor cells with high consistency, while the tumor cells with sp142 staining are less < br / > < br / > The expression of PD-L1 (TPS ≥ 1%) in tumor cells was approved by FDA, and there was no EGFR and ALK gene variation, Patients with stage III or metastatic non-small cell lung cancer who are not suitable for surgical resection or radiotherapy and chemotherapy < br / > < br / > < br / > < br / > < br / > < br / > cervical cancer < br / > < br / > < br / > < br / > PD-L1 < br / > < br / > 22c3 CPS ≥ 1 < br / > < br / > < br / > FDA approved the use of pabolizumab in the treatment of PD-L1 (CPS ≥ 1) expressed by tumor cells in patients with previously treated disease progression or intractable or metastatic cervical cancer after chemotherapy < br / > < br / > < br / > < br / > < br / > < br / > < br / > gastric cancer / gastric esophageal junction adenocarcinoma < br / > < br / > < br / > PD-L1 < br / > < br / > 22c3 CPS ≥ 1 < br / > < br / > < br / > FDA approved the use of pabolizumab in the treatment of tumor cells expressing PD-L1 (CPS ≥ 1) after two or more treatment regimens (including fluorouracil, platinum containing chemotherapy), recurrent locally advanced or metastatic adenocarcinoma of the gastric or gastroesophageal junction < br / > < br / > < br / > < br / > < br / > < br / > < br / > squamous cell carcinoma of the head and neck < br / > < br / > < br / > PD-L1 < br / > < br / > 22c3 CPS ≥ 1 < br / > < br / > < br / > < br / > FDA approved the use of pabolizumab in the treatment of tumor cells with PD-L1 (CPS ≥ 10) which were not suitable for cisplatin chemotherapy, Or it is not suitable for patients with locally advanced or metastatic urothelial carcinoma treated with platinum containing drugs < br / > < br / > < br / > < br / > < br / > < br / > esophageal carcinoma < br / > < br / > < br / > PD-L1 < br / > < br / > 22c3 CPS ≥ 10 < br / > < br / > < br / > < br / > FDA approved pabolizumab for the treatment of patients with recurrent locally advanced or metastatic esophageal squamous cell carcinoma, In these patients, PD-L1 expression was positive (CPS ≥ 10) After one-line or multiple line systemic treatment, the disease still progressed < br / > < br / > < br / > < br / > < br / > < br / > * * TPS: the percentage of tumor cells with partial or complete membrane staining under any intensity accounted for all tumor cells **CPS:PD-L1 the number of staining cells (tumor cells, lymphocytes, macrophages) divided by the number of all tumor cells and multiplied by 100 Ventana PD-L1 test interpretation standard interpretation Ventana SP263 as an example, in non-small cell lung cancer (NSCLC), SP263 staining of tumor cells (
) can present different patterns, including basal and lateral membrane staining, basement membrane staining and cytoplasmic peripheral staining (Fig 3) The latter is not actually stained in the cell membrane, and should be excluded from positive masculine < br / > < br / > < br / > < br / > sp263 showed diffuse characteristics in the staining of immune cells (IC) The evaluation should include IC in the following tumor areas: ① the surrounding matrix continuous with the tumor, ② the matrix in the tumor cell group, but the necrotic area should be excluded (Figure 4); the types of IC include lymphocytes, plasma cells, macrophages, dendritic cells and granulocytes < br / > < br / > < br / > Figure 4: evaluation range of tumor-related IC < br / > < br / > the scoring method of PD-L1 expression in immune cells is: the percentage of IC stained with any intensity PD-L1 in total IC, and only the decile and quartile are scored However, different from TC evaluation, if the total IC is 1%, the score report result can only be 0%, < 100% or 100% < br / > < br / > take Ventana sp142 as an example < br / > < br / > sp142 needs to use a set of Amplification Reagents for detection, which results in its staining characteristics different from other clones, that is, after amplification, it will show dot or granular staining Sp142 for n