Cancer Diagnosis Program | Biorepositories and Biospecimen Research Branch

Search BRD Papers

Comprehensive Comparison of 22C3 and SP263 PD-L1 Expression in Non-Small-Cell Lung Cancer Using Routine Clinical and Conditioned Archives.

Author(s): Kim SY, Kim TE, Park CK, Yoon HK, Sa YJ, Kim HR, Woo IS, Kim TJ

Publication: Cancers (Basel), 2022, Vol. 14, Page

PubMed ID: 35804910 PubMed Review Paper? No

Purpose of Paper

Conclusion of Paper

Studies

1. Study Purpose

   This study compared PD-L1 immunohistochemical staining results from two different assays using NSCLC specimens that were procured by different methods and were stored as FFPE blocks for <6 months or 6-12 months prior to analysis. A total of 431 NSCLC specimens were analyzed, which included adenocarcinoma (282 specimens), squamous cell carcinoma (133 specimens), adenosquamous carcinoma (7 specimens), and large cell neuroendocrine carcinoma and sarcomatoid carcinoma specimens (9 specimens total).  NSCLC specimens included those that were collected by bronchoscopy biopsy (83 specimens), percutaneous needle aspiration biopsy (PCNB, 256 specimens), biopsy of a metastatic site (26 specimens), endobronchial ultrasound biopsy (EBUS, 15 specimens), an unspecified cell aspiration method (6 specimens), and surgical resections that included both lobectomy (41 specimens) and wedge resection (4 specimens). Specimens were fixed in 10% buffered formalin immediately (biopsies) or after receipt in the pathology department (surgical resections) and stored as FFPE blocks for <6 months or 6-12 months.  Additional details of tissue fixation, processing, and storage were not provided. Case-matched FFPE sections (4 µm thick) were stained for PD-L1 immunostaining with the PD-L1 22C3 pharmDx Kit and the SP263 assay; slides were semi-quantitatively scored for tumor proportion score (TPS; the number of immunopositive tumor cells divided by the number of viable tumor cells) by two pathologists. Any perceptible staining in the membrane of viable tumor cells was considered positive.  The authors defined three categories based on perceived clinical relevance of PD-L1 results: negative (1% TPS), low expression (1-49% TPS), high expression (≥50% TPS).

   Summary of Findings:

   When all FFPE NSCLC specimens were considered (regardless of method of acquisition or FFPE storage duration), overall concordance between the 22C3 and SP263 PD-L1 immunohistochemistry assays was high when results were classified as positive (TPS≥1%) or negative (0%) (89.9-92.3%), as well as across TPS categories of protein expression (high expression: 79.6-100.0%; low expression: 89.9-100.0%). However, concordance between the two PD-L1 assays was significantly lower in cytology (66.7%) and biopsy (metastatic site biopsy= 69.2%, EBUS=80.0%, PCNB=81.3%, bronchoscopy biopsy=81.9%) specimens than surgically resected specimens (wedge resection=100%, lobectomy= 90.2%)(p<0.001). Concordance between PD-L1 assays was also significantly higher among FFPE blocks stored for less than 6 months compared to those stored for 6-12 months before analysis (p=0.003).  Concordance between PD-L1 assays was lower in biopsy specimens collected from a metastatic site than those sampled from lung (R²=0.68 versus 0.85, respectively).

   Biospecimens

   • Tissue - Lung

   Preservative Types

   • Formalin

   Diagnoses:

   • Neoplastic - Carcinoma

   Platform:

   Analyte	Technology Platform
   Protein	Immunohistochemistry

   Pre-analytical Factors:

   Classification	Pre-analytical Factor	Value(s)
   Biospecimen Acquisition	Method of tissue acquisition	Ultrasound-guided biopsy Fine needle aspiration Lobectomy Partial surgical resection Endoscopic biopsy
   Storage	Storage duration	<6 months ≥6 months
   Immunohistochemistry Specific	Type of antibody	22C3 SP263
   Biospecimen Acquisition	Biospecimen location	Primary tumor Metastatic site

   View more

2. Study Purpose

   This study evaluated the concordance of the tumor proportion score (TPS) for PD-L1 positive immunohistochemical staining using two different assays that employ different antibody clones (PD-L1 22C3 and SP263) in non-small-cell lung cancer (NSCL) specimens (histological subtype was not specified) using 431 specimens stored for < 1 y and in 314 archival specimens stored as FFPE blocks for 1 y (45 specimens), 1-3 y (56 specimens), 3-6 y (104 specimens), and longer than 6 y (109 specimens). The stability of antigenicity in FFPE paraffin sections was evaluated in sections from archival blocks that were analyzed immediately or after storage at room temperature or 4°C for 1 week, 2 weeks, or 1 month. Specimens were obtained through biopsy or surgical resection and were fixed in 10% buffered formalin immediately or after receipt in the pathology department, respectively. Additional details on tissue processing were not provided.  Matched FFPE sections (4 µm thick) were stained for PD-L1 immunostaining with the PD-L1 22C3 pharmDx Kit and the SP263 assay; slides were semi-quantitatively scored for tumor proportion score (TPS; the number of immunopositive tumor cells divided by the number of viable tumor cells) by two pathologists. Any perceptible staining in the membrane of viable tumor cells was considered positive. The authors defined three categories based on perceived clinical relevance of PD-L1 results: negative (<1% TPS), low expression (1-49% TPS), high expression (≥50% TPS).

   Summary of Findings:

   While FFPE blocks were not matched for case, procurement method, or histological subtype, FFPE blocks stored for longer periods of time (≥ 3 y) had lower PD-L1 expression than those stored for < 3 y.  FFPE blocks stored for <3 y had a higher proportion of specimens with high and low PD-L1 expression than those with negative expression regardless of whether the SP263 (p=0.009, both) or 22C3 (p=0.007, p=0.022, respectively) assay was used.  FFPE blocks stored for <3 y also had a higher proportion of specimens with low PD-L1 expression relative to those that were negative for both SP263 (p=0.009) or 22C3 (p=0.022) assays. Conversely, the predominant TPS for FFPE blocks stored longer than 3 y was <1% (PD-L1 negative) for both assays.

   The duration and temperature of FFPE slide storage prior to analysis also affected PD-L1 expression in an antibody-specific manner. Based on a multivariate model, the authors conclude that FFPE sections stained immediately after sectioning had a higher probability of low and high PD-L1 staining when the 22C3 antibody was used relative to when staining was with the SP263 antibody (p=0.006 and p=0.002, respectively), but the distribution of specimens in PD-L1 categories differed when FFPE sections were stored for > 1 week.  However, PD-L1 immunostaining with the SP263 was better preserved during section storage since there were no significant differences in the distribution of PD-L1 staining among FFPE sections stored for up to 1 month at room temperature or 4°C and those that were stained immediately after sectioning when the SP263 assay was used. Linear regression analysis revealed that the strength of the correlation between PD-L1 assays (22C3 and SP263) decreased with the duration of FFPE slide storage when storage was at room temperature (R²= 0.86-0.73) but not 4°C (R²=0.85-0.86).

   Biospecimens

   • Tissue - Lung

   Preservative Types

   • Formalin

   Diagnoses:

   • Neoplastic - Carcinoma

   Platform:

   Analyte	Technology Platform
   Protein	Immunohistochemistry

   Pre-analytical Factors:

   Classification	Pre-analytical Factor	Value(s)
   Storage	Storage duration	<1 y (FFPE block) 1-3 y (FFPE block) 3-6 y (FFPE block) >6 y (FFPE block) Fresh (FFPE slide) 1 week (FFPE slide) 2 weeks (FFPE slide) 1 month (FFPE slide)
   Immunohistochemistry Specific	Type of antibody	22C3 SP263
   Storage	Storage temperature	Room temperature (FFPE slide storage) 4°C (FFPE slide storage)

   View more