Virtual Library
Start Your Search
Y. Leung
Author of
-
+
P1.07 - Immunology and Immunotherapy (ID 693)
- Event: WCLC 2017
- Type: Poster Session with Presenters Present
- Track: Immunology and Immunotherapy
- Presentations: 1
- Moderators:
- Coordinates: 10/16/2017, 09:30 - 16:00, Exhibit Hall (Hall B + C)
-
+
P1.07-022 - Routine PD-L1 Immunohistochemistry Testing by 22C3 in a Canadian Reference Testing Centre (ID 9487)
09:30 - 09:30 | Author(s): Y. Leung
- Abstract
Background:
Immunohistochemical testing for PD-L1 expression is increasingly used as a predictive biomarker for anti PD-1/PD-L1 immunotherapies in non-small cell lung cancer (NSCLC). We report here the experience of population-based PD-L1 testing using the 22C3 antibody in the routine clinical practice of a large regional reference pathology laboratory.
Method:
PD-L1 testing was performed by the PD-L1 immunohistochemistry (IHC) 22C3 PharmDx assay (Agilent) using a Dako Link48 autostainer, according to the manufacturer’s instructions. Testing was performed on (1) all biopsies and resections for NSCLC performed at University Health Network (UHN) and partner hospitals between August 1, 2016 and March 31, 2017; (2) all cases of NSCLC referred from external sources for EGFR/ALK testing; and (3) cases of pulmonary squamous cell carcinoma referred in specifically for PD-L1 testing. PD-L1 IHC was also performed retroactively on archived UHN cases upon request by oncologists, dating from January 1, 2013. Tumour Proportion Scores (TPS) were reported in three categories (no expression, <1%; low expression, 1-49%; and high expression, ≥50%).
Result:
A total of 2027 PD-L1 IHC were performed on specimens from 1814 patients, of which 110 (5.4%) were reported as indeterminate, mostly due to insufficient tumor cellularity. Indeterminate results were more frequent in biopsy (6.4%) vs. resection (2.7%) specimens. For the remaining 1917 evaluable tests, proportions in each TPS category were: <1% (42.3%); 1-49% (28.5%); ≥50% (29.3%). In 1482 tests with known EGFR mutation status, EGFR-mutated tumors (n=296) demonstrated lower rates of PD-L1 expression [TPS <1% (51.7%); 1-49% (29.4%); ≥50% (18.9%); P<0.01]. No statistically significant difference in PD-L1 expression was detected in ALK-rearranged tumors (n=29). Of 100 patients with successful PD-L1 staining in both a biopsy and paired resection specimen, 57/100 (57%) demonstrated concordant TPS categories in both specimens. Only 22/49 (44.9%) biopsies with TPS<1% showed TPS<1% in the resection, while 26/49 (53.1%) and 1/49 (2.0%) showed TPS 1-49% and ≥50%, respectively. Of biopsies with TPS 1-49%, 17/29 (58.6%) were concordant in the resection, while 5/29 (17.2%) and 7/29 (24.1%) showed TPS <1% and ≥50%, respectively. Among biopsies with TPS≥50%, 18/22 (81.8%) were concordant in the resection, while 4/22 (18.2%) showed TPS 1-49% in the resection.
Conclusion:
In our population-based study, PD-L1 expression in NSCLC using the 22C3 antibody demonstrated similar prevalence as reported in clinical trials. EGFR mutated but not ALK rearranged tumors were associated with lower PD-L1 expression. Intra-tumoral heterogeneity of PD-L1 expression may result in its under-estimation in biopsy specimens compared to paired resection specimens.
