Programmed Death-ligand 1 Immunohistochemistry Research Articles

High Interobserver Variability Among Pathologists Using Combined Positive Score to Evaluate PD-L1 Expression in Gastric, Gastroesophageal Junction, and Esophageal Adenocarcinoma

Reliable, reproducible methods to interpret programmed death ligand-1 (PD-L1) expression on tumor cells (TC) and immune cells (IC) are needed for pathologists to inform decisions associated with checkpoint inhibitor therapies. Our international study compared interpathologist agreement of PD-L1 expression using the combined positive score (CPS) under standardized conditions on samples from patients with gastric/gastroesophageal junction/esophageal adenocarcinoma. Tissue sections from 100 adenocarcinoma pretreatment biopsies were stained in a single laboratory using the PD-L1 immunohistochemistry 28-8 and 22C3 (Agilent) pharmDx immunohistochemical assays. PD-L1 CPS was evaluated by 12 pathologists on scanned whole slide images of these biopsies before and after a 2-hour CPS training session by Agilent. Additionally, pathologists determined PD-L1–positive TC, IC, and total viable TC on a single tissue fragment from 35 of 100 biopsy samples. Scoring agreement among pathologists was assessed using the intraclass correlation coefficient (ICC). Interobserver variability for CPS for 100 biopsies was high, with only fair agreement among pathologists both pre- (range, 0.45-0.55) and posttraining (range, 0.56-0.57) for both assays. For the 35 single biopsy samples, poor/fair agreement was also observed for the total number of viable TC (ICC, 0.09), number of PD-L1–positive IC (ICC, 0.19), number of PD-L1–positive TC (ICC, 0.54), and calculated CPS (ICC, 0.14), whereas calculated TC score (positive TC/total TC) showed excellent agreement (ICC, 0.82). Retrospective histologic review of samples with the poorest interpathologist agreement revealed the following as possible confounding factors: (1) ambiguous identification of positively staining stromal cells, (2) faint or variable intensity of staining, (3) difficulty in distinguishing membranous from cytoplasmic tumor staining, and (4) cautery and crush artifacts. These results emphasize the need for objective techniques to standardize the interpretation of PD-L1 expression when using the CPS methodology on gastric/gastroesophageal junction cancer biopsies to accurately identify patients most likely to benefit from immune checkpoint inhibitor therapy.

Loss of CDKN2A/B is a Molecular Marker of High-grade Histology and is Associated with Aggressive Behavior in Acinic Cell Carcinoma

Acinic cell carcinoma (AciCC) is a rare salivary gland cancer with excellent prognosis in most cases. However, a subset of patients will develop distant metastasis and die of disease. Recently, a 2-tiered grading scheme in AciCC was proposed to recognize patients at risk of poor outcome. We performed a genetic analysis of AciCC to explore the underlying molecular correlates of the tumor grade and examined programmed death ligand 1 (PD-L1) expression to identify potential candidates for immunotherapy. A retrospective cohort of 55 patients included 34 high-grade (HG) and 21 low-grade AciCCs. Forty-three cases were subjected to targeted exome sequencing by Memorial Sloan Kettering-Integrated Mutation Profiling of Actionable Cancer Targets. PD-L1 immunohistochemistry was performed in 33 cases. Tumor mutation burden was low with a median of 1 and 2 mutations in low-grade and HG AciCCs, respectively. CDKN2A/B was the most frequently altered gene, and loss-of-function mutations were found only in HG but not in low-grade AciCCs (18/31 [58.1%] vs 0/12 [0%], P < .001). CDKN2A/B alterations were significantly associated with distant metastasis, which occurred in 16/18 (88.9%) CDKN2A/B mutants versus 11/25 (44%) wild-type cases (P = .004, Fisher exact test). Sequential profiling of multiple temporally distant samples from the same patient demonstrated intratumor heterogeneity, including the detection of CDKN2A/B deletion in the second, in HG metastasis only. ATM and PTEN mutations were detected in 6/31 (19.4%) and 5/31 (16.1%); ARID2, BIRC3, and FBXW7 mutations each in 4/31 (12.9%); and TP53, MTAP, and FAT1 each in 3/31 (9.7%) HG AciCC. PD-L1–positive labeling was more common in HG AciCC (9/17, 52.9% vs 3/16, 18.9%, P = .071). CDKN2A/B mutations in AciCC represent a molecular marker of HG histology and disease progression, providing a rationale for further studies to determine their prognostic and therapeutic significance in this salivary gland cancer. AciCC with ATM mutations may be amenable to targeted therapy. Immunotherapy can be considered to be a treatment option for a subset of patients with AciCC.

Programmed death-ligand1 is a determinant of recurrence in alveolar echinococcosis

ObjectivesAlveolar echinococcosis (AE) recurrence is one of the major stakes in patients undergoing surgery, the main curative treatment. Preliminary data demonstrated an effect of programmed death-ligand1 (PD-L1) inhibitors on AE proliferation in animals. The current study aimed to analyze the prognostic value of PD-L1 expression in tissue samples of patients with AE undergoing surgery. MethodsA cross-sectional study of patients operated for AE between 2002 and 2017 was performed. Patients with recurrence were matched 1: 2 with patients without recurrence. The matching criteria were PNM staging (P = hepatic localization of the parasite, N = extra-hepatic involvement of neighboring organs, and M = absence or presence of metastasis), resection status, preoperative albendazole treatment, and lesion size. PD-L1 immunohistochemistry staining was performed in surgical liver specimens. The expression of PD-L1 was assessed in immune cells. Disease-free survival was calculated using the Kaplan-Meier method. ResultsAmong 68 consecutive patients, eight patients with recurrence were matched to 16 patients without recurrence. PD-L1 was overexpressed in patients with recurrence (recurrence: PD-L1 <1%: one, PD-L1 ≥1%: seven; no recurrence: PD-L1 <1%: nine, PD-L1 ≥1%: seven, P = 0.040). Moreover, patients with lower PD-L1 expression (<1%) showed better median disease-free survival (120 months, 95% confidence interval 104-135 vs 74, 95% confidence interval 44-104, P = 0.050). ConclusionThese findings highlight the proof of concept of PD-L1 in AE, but further data on its prognostic importance and the role of immune checkpoint blockade as a promising therapeutical strategy are needed.

Programmed death ligand-1 expression and its association with the degree of differentiation and the presence of necrosis in non-small cell lung carcinoma

The mechanisms underlying the expression of programmed death ligand-1 (PD-L1) in non-small cell lung carcinoma (NSCLC) are not yet fully clarified. In this study, surgical resections of 730 lung cancer patients with diagnosed NSCLC were analyzed. Results of PD-L1 immunohistochemistry (using clone 22C3) were correlated with clinicopathological variables including the degree of tumor differentiation and the presence of confluent areas of coagulative necrosis. PD-L1 immunohistochemistry was analyzed in tumor cells, whereas PD-L1 positivity was defined as membranous staining in ≥1 of tumor cells. A significantly higher proportion of PD-L1 positive cases was noted in poorly differentiated (grade 3) adenocarcinomas compared to better differentiated (grade 1 and grade 2) subtypes (63.8% vs. 28.7%; p<0.001). Contrary to this, better differentiated (keratinizing) and less differentiated (non-keratinizing) squamous cell carcinoma subtypes were found to have a similar proportion of PD-L1 positive cases (51.4% vs. 55.8%; p=0.570). High levels of PD-L1 expression significantly correlated with the presence of necrosis in NSCLC: seventy-nine of 109 NSCLC cases with the presence of necrosis were PD-L1 positive compared to 256 out of 621 NSCLC without necrosis (72.5% vs. 41.2%; p<0.001). High PD-L1 expression was not positively correlated with age, gender, and advanced T stage but a significant association between PD-L1 positivity and higher N stage was observed (p<0.001) in NSCLC patients. In conclusion, the proportion of PD-L1 positive cases is higher only in poorly differentiated NSCLC of the adenocarcinoma type. A significantly higher overall rate of PD-L1 positive cases was noted in NSCLC with the presence of necrosis. Further investigation is suggested to elucidate the intricated interconnections between the plethora of hypoxic biomarkers and immunological factors in different types and subtypes of NSCLC.

Durable responders in advanced NSCLC with elevated TMB and treated with 1L immune checkpoint inhibitor: a real-world outcomes analysis

BackgroundFor patients with advanced non-small cell lung carcinoma (NSCLC), immune checkpoint inhibitor (ICPI) and chemotherapy (chemo) ICPI represent two distinct first-line standard-of-care regimens without clear and established biomarkers to inform...

Feasibility of shape-sensing robotic-assisted bronchoscopy for biomarker identification in patients with thoracic malignancies

ObjectiveMolecular diagnostic assays require samples with high nucleic acid content to generate reliable data. Similarly, programmed death-ligand 1 (PD-L1) immunohistochemistry (IHC) requires samples with adequate tumor content. We investigated whether shape-sensing robotic-assisted bronchoscopy (ssRAB) provides adequate samples for molecular and predictive testing. MethodsWe retrospectively identified diagnostic samples from a prospectively collected database. Pathologic reports were reviewed to assess adequacy of samples for molecular testing and feasibility of PD-L1 IHC. Tumor cellularity was quantified by an independent pathologist using paraffin-embedded sections. Univariable and multivariable linear regression models were constructed to assess associations between lesion- and procedure-related variables and tumor cellularity. ResultsIn total, 128 samples were analyzed: 104 primary lung cancers and 24 metastatic lesions. On initial pathologic assessment, ssRAB samples were deemed to be adequate for molecular testing in 84% of cases; on independent review of cellular blocks, median tumor cellularity was 60% (interquartile range, 25%-80%). Hybrid capture-based next-generation sequencing was successful for 25 of 26 samples (96%), polymerase chain reaction-based molecular testing (Idylla; Biocartis) was successful for 49 of 52 samples (94%), and PD-L1 IHC was successful for 61 of 67 samples (91%). Carcinoid and small cell carcinoma histologic subtype and adequacy on rapid on-site evaluation were associated with higher tumor cellularity. ConclusionsThe ssRAB platform provided adequate tissue for next-generation sequencing, polymerase chain reaction-based molecular testing, and PD-L1 IHC in >80% of cases. Tumor histology and adequacy on intraoperative cytologic assessment might be associated with sample quality and suitability for downstream assays.

Validation of E1L3N antibody for PD-L1 detection and prediction of pembrolizumab response in non-small-cell lung cancer

BackgroundThe programmed death-ligand 1 (PD-L1) 22C3 assay is one of the approved companion diagnostic assays for receiving anti-programmed cell death ligand 1 (PD-L1) therapy. Our study evaluated the performance of E1L3N and 22C3 antibodies in estimating PD-L1 expression in non-small cell lung cancer (NSCLC).MethodsOur retrospective study included 46 patients diagnosed with unresectable EGFR/ALK/ROS1-negative NSCLC who received first-line pembrolizumab therapy between 2018 and 2021. PD-L1 immunohistochemistry of baseline tissue biopsy samples was performed using PDL1-E1L3N and PDL1-22C3 antibodies. The concordance between the PD-L1 assays and the treatment outcomes was assessed.ResultsUsing a tumor proportion score (TPS) cutoff of ≥1%, 67.4% of patients are evaluated to be positive using PDL1-E1L3N and 73.9% using PDL1-22C3. Using a TPS of ≥50% as the cutoff, 26.1% of patients are positive using PDL1-E1L3N and 30.4% using PDL1-22C3. The PDL1-22C3 and PDL1-E1L3N assays are highly concordant and reveal a correlation coefficient of 0.925 (p < 0.0001). Patients with PDL1-E1L3N TPS > 50% have a significantly higher objective response rate than patients with PDL1-E1L3N TPS < 1% (p = 0.047), with a similar trend observed for PDL1-22C3 (p = 0.051). Consistent with PDL1-22C3, patients with higher PDL1-E1L3N expression (≥50%, 1–49%) have longer progression-free survival than those with PDL1-E1L3N TPS < 1%.ConclusionOur study provides clinical evidence on the concordance of PD-L1 TPS scores between clones E1L3N and 22C3. Moreover, the treatment responses to pembrolizumab are also comparable between the PDL1-E1L3N and PDL1-22C3. These findings indicate that E1L3N is a reliable and cost-effective assay and may serve as an alternative to 22C3.

Programmed Death Ligand 1 (PD-L1) as a Therapeutic Target in Gestational Trophoblastic Neoplasia (GTN) (098)

<b>Objectives:</b> Despite early efficacy signals for anti-PD-L1/PD-1 immunotherapy in chemotherapy-resistant GTN, response rates are variable. Lack of consistent evaluation and reporting of trophoblast PD-L1 expression and other candidate predictive biomarkers makes developing a precise strategy for PD-L1/PD-1 blockade in the treatment of GTN challenging. We characterize trophoblast PD-L1 expression using the clinically applicable PD-L1 IHC 22C3 assay, tumor-infiltrating immune cell (TIIC) density and clinical outcomes in one of the largest GTN cohorts to date. <b>Methods:</b> Cases of paraffin-embedded primary tumors in our GTN library were used for this investigation (<i>n</i>=60). Immunohistochemistry (IHC) staining for PD-L1, CD68 (macrophages), CD8, CD4, and CD56 (NK cells) was performed on fresh-cut 4-mm tissue sections. Tumor proportion scores (TPS) were independently assigned by two gynecologic pathologists. Digital image analysis using QuPath v0.2.3 to average TIIC density in 3 40x hotspots per case was performed. Complete clinical data was possible for 18 patients. Descriptive statistics, Wilcoxon rank-sum test, and simple linear regression were used. <b>Results:</b> PD-L1 expression and TIIC density were quantified for all 60 cases of choriocarcinoma (CC, <i>n</i>=32), placental site trophoblastic tumor (PSTT, <i>n</i>=9), and epithelioid trophoblastic tumor (ETT, <i>n</i>=19). CC and PSTT have higher trophoblast PD-L1 expression than ETT; the median TPS for CC and PSTT was 85% (range: 0-98%) and 90% (range: 0-90%), compared to 35% (range: 0-100%) for ETT (p=0.02). CD68 monocytes were more abundant in CC and PSTT than ETT. CD8 cytotoxic T (p<0.01) and CD4 regulatory T (p=0.04) cell densities were higher in ETT. CD56 expression was associated with PD-L1 expression (slope=0.36, p=0.03). The Table summarizes the clinically annotated subset of patients. Similar patterns of PD-L1 expression/ TPS and TIICs were observed for CC and ETT; complete data were available for only two PSTT cases. Of the 15 patients who did not see anti-PD-L1/PD-1 treatment, seven (47%) would have been eligible based on a TPS of >0 and might have been offered it as an alternative to multi-agent chemotherapy or fertility loss. Two (PSTT case 16 and ETT case 17) of the three patients who received salvage anti-PD-1 therapy had durable complete responses. The third case (ETT case 18) had a complete response but then progressed after two years. All three had above-median TPS scores for their histologic subtypes and had moderate to high levels of CD56 expression. The two patients without evidence of disease had above median levels of CD8 and CD68 for their subtype, while the patient who recurred had below-median levels of both immune markers. <b>Conclusions:</b> Contrary to the published literature, ubiquitously high trophoblastic PD-L1 expression was not observed using the clinically applicable PD-L1 IHC 22C3 assay. Rather, there is a wide range of PD-L1 expression, including some cases with a TPS score of 0%. We added three cases of non-CC treated with salvage anti-PD-L1/PD-1 therapy to the four reported to date. In these cases, tumor-infiltrating CD8, CD68, and CD56 cell density may influence anti-PD-1 efficacy. There may be an opportunity to spare GTN patients from toxicity and/or adverse outcomes by prescribing anti-PD-L1/PD-1 therapy earlier in their disease course.

Programmed Death Ligand-1 Combined Positive Score Concordance and Interrater Reliability in Primary Tumors and Synchronous Lymph Node Metastases in Resected Cases of p16+ Oropharyngeal Squamous Cell Carcinoma.

Pembrolizumab is used in patients with metastatic head and neck squamous cell carcinoma contingent upon the programmed death ligand-1 (PD-L1) combined positive score (CPS). To compare PD-L1 CPS scores derived from paired resected primary tumors (PTs) and lymph node metastases (LMs) in patients with p16+ oropharyngeal squamous cell carcinoma (OPSCC). We identified 38 resected p16+ OPSCCs for which paired PTs and LMs were available. PD-L1 immunohistochemistry using the SP263 antibody clone was done on both the PT and the LM. CPS scoring was performed by 4 observers, and data were analyzed at the CPS cut points of greater than or equal to 1 and 20 in regard to interobserver and interspecimen agreement. Overall agreement between consensus CPS scoring of PT and LM was seen in 76% of paired specimens (κ = 0.53). No specimen received a negative consensus score. Interobserver agreement for both PT and LM was fair to substantial (κ = 0.54 and 0.51, respectively) and was inferior to that seen in a prospective series of unselected head and neck squamous carcinoma cases evaluated at our institution (κ = 0.84). Given the high rates of interobserver and interspecimen variability, evaluation of additional material or by additional observers may be of value in performing CPS scoring in cases of p16+ OPSCC. This is particularly the case when a negative or low-positive result is being evaluated in a patient who is otherwise a good candidate for immunotherapy.

Significance of the effects of chemotherapy on programmed death-ligand 1 expression in triple-negative breast cancer.

Atezolizumab has been approved as an antibody against programmed death-ligand 1 (PD-L1)-positive immune cells in patients with advanced or recurrent triple-negative breast cancer. However, the optimal timing to examine PD-L1 expression remains controversial. We retrospectively researched PD-L1 positivity rates in biopsy, surgical and recurrent specimens from patients with triple-negative breast cancer treated with neoadjuvant chemotherapy. We also examined alterations in PD-L1 and their meaning. In total, 35 triple-negative breast cancer biopsy specimens obtained before neoadjuvant chemotherapy, 20 corresponding specimens obtained after neoadjuvant chemotherapy and 5 corresponding recurrent specimens were obtained. We examined PD-L1 immunohistochemistry on tumor cells and tumor-infiltrating immune cells using SP142 antibody. In comparison with specimens obtained before neoadjuvant chemotherapy, PD-L1 expression randomly changed in immune cells after neoadjuvant chemotherapy, but PD-L1 expression was significantly reduced in tumor cells. Pre-neoadjuvant chemotherapy specimens with low PD-L1 expression (PD-L1 scores of ≤1 for both immune cells and tumor cells) were linked to better disease-free survival (P<0.001) and overall survival (P<0.001) than the other specimens. This is the first study to evaluate PD-L1 expression both before and after chemotherapy in breast cancer and examine its relationship with prognosis. The results suggest that the PD-L1 level may be useful for predicting the prognosis of patients with triple-negative breast cancer who do not have pathological complete responses to neoadjuvant chemotherapy.

Choice of PD-L1 immunohistochemistry assay influences clinical eligibility for gastric cancer immunotherapy

BackgroundImmune checkpoint inhibitors (ICI) are now standard-of-care treatment for patients with metastatic gastric cancer (GC). To guide patient selection for ICI therapy, programmed death ligand-1 (PD-L1) biomarker expression is routinely assessed via immunohistochemistry (IHC). However, with an increasing number of approved ICIs, each paired with a different PD-L1 antibody IHC assay used in their respective landmark trials, there is an unmet clinical and logistical need for harmonization. We investigated the interchangeability between the Dako 22C3, Dako 28–8 and Ventana SP-142 assays in GC PD-L1 IHC.MethodsIn this cross-sectional study, we scored 362 GC samples for PD-L1 combined positive score (CPS), tumor proportion score (TPS) and immune cells (IC) using a multiplex immunohistochemistry/immunofluorescence technique. Samples were obtained via biopsy or resection of gastric cancer.ResultsThe percentage of PD-L1-positive samples at clinically relevant CPS ≥ 1, ≥ 5 and ≥ 10 cut-offs for the 28–8 assay were approximately two-fold higher than that of the 22C3 (CPS ≥ 1: 70.3 vs 49.4%, p < 0.001; CPS ≥ 5: 29.1 vs 13.4%, p < 0.001; CPS ≥ 10: 13.7 vs 7.0%, p = 0.004). The mean CPS score on 28–8 assay was nearly double that of the 22C3 (6.39 ± 14.5 vs 3.46 ± 8.98, p < 0.001). At the clinically important CPS ≥ 5 cut-off, there was only moderate concordance between the 22C3 and 28–8 assays.ConclusionOur findings suggest that scoring PD-L1 CPS with the 28–8 assay may result in higher PD-L1 scores and higher proportion of PD-L1 positivity compared to 22C3 and other assays. Until stronger evidence of inter-assay concordance is found, we urge caution in treating the assays as equivalent.

Assessing PD-L1 without a biopsy and through PD-L1 PET imaging with 18F-BMS-986229.

2578 Background: Programmed death ligand-1 (PD-L1) is usually determined by immunohistochemistry (IHC). Determining PD-L1 status by whole body, non-invasive PD-L1 PET imaging with 18F-BMS-986229 tracer infusion may be a way to circumvent limitations to PD-L1 IHC such as poor tissue availability; discrepant PD-L1 IHC results among different tumors within the same patient; and inability to measure PD-L1 longitudinally given challenges with repeated biopsies. Methods: Within the prospective ADAPT-IT trial (NCT03122522) testing an abbreviated course of nivolumab (nivo) + ipilimumab (ipi) in patients with unresectable stage III or IV melanoma, we investigated PD-L1 PET imaging in a 10-patient expansion cohort at baseline and after 6 weeks of treatment. Maximum standard uptake value (SUV max) and mean of all SUV max for lesions for PD-L1 PET were calculated on a whole patient level for each target/ non target lesion. Standard treatment response was determined by RECIST v1.1 at weeks 6 and 12. PD-L1 IHC was scored using the E1L3N antibody by tumor proportion score (TPS), Immune Cell Score (ICS), and a Combined Positive Score (CPS). Comparisons between PD-L1 PET parameters and response were determined by complete or partial response (CR/PR); stable disease (SD), and progressive disease (PD) at respective imaging timepoints. Absolute changes in PD-L1 SUV mean and max were calculated between scans. The correlation between PD-L1 PET and IHC was estimated using Spearman’s coefficient correlation. Results: Five patients (50%) had a PR to treatment at week 6; one of these PR became a CR at week 12 and one PR was not assessable at week 12. All RECIST responders at week 6 (n = 5) had baseline PD-L1 SUV mean ≥3.00, and all progressors at week 6 (n = 3) had baseline PD-L1 SUV mean ≤2.60 (Table). A similar trend was observed when assessing response at week 12 and when considering baseline PD-L1 SUV max. Patients had changes in their PD-L1 SUV levels between baseline and week 6. Baseline PD-L1 SUV mean (inclusive of all lesions within a patient) correlated with PD-L1 IHC by TPS, ICS, and CPS (r = 0.64, 0.5, 0.47, respectively). No patients had side effects from 18F-BMS-986229 tracer infusion. Conclusions: The signal of PD-L1 positivity by PET imaging with 18F-BMS-986229 at baseline appears associated with early efficacy from nivo + ipi in this small cohort and may offer additional information than PD-L1 IHC. The ability to assess PD-L1 on a whole patient level at multiple timepoints on treatment may have future implications in how best to sequence and combine immunotherapies but further study in larger patient cohorts is needed. Clinical trial information: NCT03122522. [Table: see text]

Observer performance study to examine the feasibility of the AI-powered PD-L1 analyzer to assist pathologists’ assessment of PD-L1 expression using tumor proportion score in non–small cell lung cancer.

8529 Background: Programmed death ligand 1 (PD-L1) expression is the standard biomarker for PD-L1 inhibitors in advanced non-small cell lung cancer (NSCLC). However, evaluation of PD-L1 tumor proportion score (TPS) by pathologists causes inter-observer variation and demands time to interpret. This study aimed to evaluate the benefit of the artificial intelligence (AI) algorithm in assisting pathologists to determine TPS on PD-L1 immunohistochemistry (IHC) whole-slide images (WSIs) in NSCLC. Methods: Lunit SCOPE PD-L1, an AI-powered PD-L1 TPS analyzer, was developed from 393,565 tumor cells annotated by board-certified pathologists for PD-L1 expression in 802 WSIs stained by 22C3 pharmDx IHC. The AI model was developed based on a region-based convolutional neural network, and the model can detect and count PD-L1 positive or negative tumor cells from WSIs to calculate TPS. Seven independent board-certified pathologists scored ground truth (GT) of PD-L1 TPS from 199 WSI of NSCLC stained by 22C3 pharmDx IHC. TPS from each GT reader was grouped as negative (&lt; 1%), low (1% to 49%), or high (≥ 50%). The GT of each slide was determined by the consensus of GT readers. Another twelve independent board-certified pathologists scored PD-L1 TPS from the same WSIs as observer performance testers (OPT). They scored TPS twice with a washout interval of 4 weeks, with or without AI assistance. TPS accuracy change and reading time of OPT reader according to the presence or absence of AI assistance were analyzed. Results: The standalone accuracy of the AI model was 0.809 (95% CI: 0.690–0.941). With AI assistance, the overall accuracy of TPS had been changed from 0.799 (95% confidence interval [CI]: 0.764–0.836) to 0.832 (95% CI: 0.796–0.869) (P = 0.004). AI assistance increased the accuracy rate in 11 out of 12 OPT readers. The result of the generalized linear mixed model revealed that AI assistance and specimen type affected the probability of correct answer, while the order of reading did not (Table). The mean time to read with AI was 195.4±506.5 (mean±standard deviation) seconds, which was significantly shorter than the mean time to read without AI (285.1±1578.4, P &lt;0.001). Conclusions: This study demonstrates that an AI-powered PD-L1 TPS analyzer can assist board-certified pathologists in evaluating TPS of NSCLC by improving the accuracy of TPS group evaluation and reducing the time to read slides.[Table: see text]

Choice of PD-L1 immunohistochemistry assay influences clinical eligibility for gastric cancer immunotherapy.

4026 Background: Immune checkpoint inhibitors (ICI) are now standard-of-care treatment for patients with metastatic gastric cancer (GC). To guide patient selection for ICI therapy, programmed death ligand-1 (PD-L1) biomarker expression is routinely assessed via immunohistochemistry (IHC). Regulatory approval for ICIs is granted based on PD-L1 expression status, scored using metrics such as the combined positive score (CPS). However, with an increasing number of approved ICIs, each paired with a different PD-L1 antibody IHC assay used in their respective landmark trials, there is an unmet clinical and logistical need for harmonization. We thus investigated the interchangeability between the Dako 22C3, Dako 28-8 and Ventana SP-142 assays in GC PD-L1 IHC. Methods: In this cross-sectional study, samples were obtained via biopsy or resection of gastric cancer at the National University Hospital, Singapore. We scored 362 GC samples for PD-L1 CPS, tumor proportion score (TPS) and immune cells (IC) using a multiplex immunohistochemistry/immunofluorescence technique. 344 samples were developed into a tissue microarray (TMA), while 18 samples were used as whole slides for orthogonal validation. The samples selected for whole slide analysis were obtained from GC patients treated with ICI therapy. Results: The percentage of PD-L1 positive samples at clinically relevant CPS ≥1, ≥5 and ≥10 cut-offs (Table) for the 28-8 assay were approximately two-fold higher than that of the 22C3 (CPS≥1: 70.3% vs 49.4%, p&lt;0.001; CPS≥5: 29.1% vs 13.4%, p&lt;0.001; CPS≥10: 13.7% vs 7.0%, p=0.004). The mean CPS score on 28-8 assay was nearly double that of the 22C3 (6.39 ±14.5 vs 3.46±8.98, p&lt;0.001). At the clinically important CPS≥5 cut-off, there was only moderate concordance between the 22C3 and 28-8 assays. Conclusions: Our findings suggest that scoring PD-L1 CPS with the 28-8 assay may result in higher proportion of PD-L1 positivity and higher PD-L1 scores compared to assessment with the 22C3 and other assays. Clinically, this could lead to a larger number of patients eligible and approved for ICI therapy. If assays are viewed and used interchangeability, a substantial number of patients may be inaccurately denied or granted treatment with ICIs based on the assay chosen. As such, until stronger evidence of inter-assay concordance is found, we urge caution in treating the assays as equivalent.[Table: see text]

The utilization of artificial intelligence to produce clinically relevant scores on PD-L1 immunostained non-small cell lung cancer biopsies.

e14576 Background: Immunotherapies targeting complex programmed death-ligand 1 (PD-L1) interactions have been groundbreaking in the treatment of non-small cell lung cancer (NSCLC), offering new strategies for patients who are likely to respond. PD-L1 binds to the checkpoint PD-1 to regulate T cells, which has an important role in boosting the immune response. Identifying patients who may benefit from PD-L1 focused care is dependent on the interpretation of the drug’s associated companion diagnostic (CDx) immunohistochemistry (IHC) assay. PD-L1 IHC assays stratify NSCLC patients using a Tumor Proportion Score (TPS), a manual scoring paradigm which must be applied by a clinical pathologist. Application of the score is complex as it requires the reader to visually exclude stromal and tumor-infiltrating immune cells across a full sample, retaining PD-L1 expression information for true tumor cells only. Methods: In this study, we utilized the PD-L1 [28-8] immunohistochemistry (IHC) assay to stain 10 NCSLC specimens. Stained slides were then imaged using the AT2 scanner (Leica Biosystems, Buffalo Grove, IL) scanner and analyzed using the Visiopharm Image Analysis platform. Results: The intricacy of TPS application is time consuming for the pathologist and introduces opportunity for inter- and intra-reader variability. Our novel image analysis approach utilizes artificial intelligence (AI) to automatically denote tumor nests and exclude tumor-infiltrating immune cells. The remaining true tumor cells may then be quantified for PD-L1 expression across the full biopsy, producing a TPS for each sample. Conclusions: This method allows for great accuracy and time-efficiency in providing a TPS than traditional methods, which may result in improved patient care.

Harmonization of programmed death-ligand 1 immunohistochemistry and mRNA expression scoring in metastatic melanoma: a multicentre analysis.

The aim of this multicentre study was to harmonize programmed death-ligand 1 (PD-L1) immunohistochemistry (IHC) and melanoma scoring. To provide a reference for PD-L1 expression independently of the IHC protocol, PD-L1 mRNA expression was compared with IHC. Standardized PD-L1 assays (22C3, 28-8, SP142, SP263) and laboratory-developed tests (QR1, 22C3) were evaluated on three IHC platforms with a training set (seven cases). mRNA expression was determined by RNAscope (CD274/PD-L1 probe) and analysed with image analysis. PD-L1 IHC findings were scored by seven blinded pathologists using the tumour proportion score (TPS), the combined positive score (CPS), and the MELscore. This method was validated by three blinded pathologists on 40 metastatic melanomas. Concordances among various antibody/platforms were high across antibodies [intraclass correlation coefficient (ICC) >0.80 for the CPS], except for SP142. Two levels of immunostaining intensity were observed: high (QR1 and SP263) and low (28-8, 22C3, and SP142). Reproducibilities across pathologists were higher for QR1 and SP263 (ICC ≥0.87 and ICC ≥0.85 for the TPS and the CPS, respectively). QR1, SP263 and 28-8 showed the highest concordance with mRNA expression. We developed a standardized method for PD-L1 immunodetection and scoring, tested on 40 metastatic melanomas. Concordances among antibodies were excellent for all criteria, and concordances among pathologists were better for the MELscore than for other scores. Harmonization of PD-L1 staining and scoring in melanomas with good concordance is achievable with the PD-L1 IHC protocols applied to other cancers; this reproducible approach can simplify daily practice.

Expression of CD274 mRNA Measured by qRT-PCR Correlates With PD-L1 Immunohistochemistry in Gastric and Urothelial Carcinoma.

Programmed death-ligand 1 (PD-L1) immunohistochemistry (IHC) is widely used to predict the clinical responses to immune checkpoint inhibitors (ICIs). However, PD-L1 IHC suffers from the complexity of multiple testing platforms and different cutoff values caused by the current one drug-one diagnostic test co-development approach for ICIs. We aimed to test whether PD-L1 (CD274) mRNA expression levels measured using quantitative reverse transcription-polymerase chain reaction (qRT-PCR) can represent PD-L1 IHC and predict responses to ICI. The FDA-approved PD-L1 IHC results with 22C3 pharmDx (gastric cancer) and SP142 (urothelial carcinoma) were compared with CD274 mRNA expression levels via qRT-PCR using the same formalin-fixed, paraffin-embedded tissue blocks from 59 gastric cancer and 41 urothelial carcinoma samples. CD274 mRNA expression was identified using three independent sets of primers and TaqMan® probes targeting exon 1–2, exon 3–4, and exon 5–6. CD274 mRNA levels in spanning exon 1–2, exon 3–4, and exon 5–6 junctions of CD274 correlated well with PD-L1 expression (r2=0.81, 0.65, and 0.59, respectively). The area under the curve of exon 1–2 was the highest (0.783), followed by exon 3–4 (0.701), and exon 5–6 (0.671) of the CD274 gene against the PD-L1 combined positive score cutoff of 10. When CD274 mRNA expression was matched for response to immunotherapy, the overall response rate was higher in patients with high CD274 mRNA levels with a cutoff of 0.0722 (gastric cancer) and 0.0480 (urothelial carcinoma) than in those with low CD274 mRNA expression (P < 0.001 and P = 0.018, respectively). These results show that CD274 mRNA levels predicted ICI responses in patients with gastric or urothelial carcinomas and could be used as alternatives for PD-L1 IHC.

Companion Diagnostics: Lessons Learned and the Path Forward From the Programmed Death Ligand-1 Rollout.

Programmed death ligand-1 (PD-L1) immunohistochemistry companion diagnostic assays play a crucial role as predictive markers in patients being considered for immune checkpoint inhibitor therapy. However, because of a convergence of several factors, including recognition of increased types of cancers susceptible to immunotherapy, increasing numbers of immune checkpoint inhibitors, and release of multiple PD-L1 immunohistochemistry antibodies with differing reporting systems, this complex testing environment has led to significant levels of confusion for pathologists and medical oncologists. To identify which processes and procedures have contributed to the current challenges surrounding programmed death receptor-1 (PD-1)/PD-L1 companion diagnostics and to propose potential remedies to this issue. This is based upon input from key industrial stakeholders in conjunction with the College of American Pathologists Personalized Health Care Committee. A meeting of representatives of pharmaceutical and in vitro diagnostic companies along with the Personalized Health Care Committee reviewed the process of release of the PD-L1 companion diagnostic assays using a modified root cause analysis format. The modified root cause analysis envisioned an ideal circumstance of development and implementation of a companion diagnostic to identify shortcomings in the rollout of the PD-L1 assay and to suggest actions to improve future companion diagnostic assay releases. The group recommended improvements to key principles in companion diagnostics implementation related to multi-stakeholder communication, increased regulatory flexibility to incorporate postapproval medical knowledge, improved cross-disciplinary information exchange between medical oncology and pathology societies, and enhanced postmarket training programs. The rapidly changing nature of and increasing complexity associated with companion diagnostics require a fundamental review of processes related to their design, implementation, and oversight.

Analysis of Immunological Characteristics and Genomic Alterations in HPV-Positive Oropharyngeal Squamous Cell Carcinoma Based on PD-L1 Expression.

Programmed death-ligand 1 (PD-L1) expression has been approved as an immune checkpoint inhibitor (ICI) response predictive biomarker; however, the clinicopathological and molecular features of HPV-positive oropharyngeal squamous cell carcinoma [HPV(+)OPSCC] based on PD-L1 expression are not well studied. We aimed to characterize clinicopathological, tumor immune microenvironmental, and molecular features of HPV(+)OPSCC with different PD-L1 expression scored by combined positive score (CPS). A total of 112 cases were collected from 2008-2021 and received PD-L1 and CD8 immunohistochemistry (IHC) staining. 71 samples received DNA sequencing out of which 32 samples received RNA sequencing for immune-related gene alterations or expression analysis. The 32 samples were also subjected to analysis of CD20, CD4, CD8, CD68, Foxp3 and P16 by multiplex immunofluorescence (mIF) staining, and the immune markers were evaluated in the tumor body (TB), tumor margin (TM) and normal stroma (NS) regions separately. Our results showed that of 112 HPV(+)OPSCC tumors, high(CPS≥20), intermediate(1≤CPS<20), and low(CPS<1) PD-L1 expression was seen in 29.5%, 43.8% and 26.8% cases respectively. Non-smoking patients and patients with tumors occurring at the tonsils or having rich lymphocytes infiltration had significantly higher PD-L1 expression. Patients with CPS≥20 had significantly higher tumor mutation burden (TMB, p=0.0058), and PD-L1 expression correlated significantly with CD8+ T cells infiltration, which were ample in tumor regions than in NS in mIF. CD20+, CD4+, CD68+, Foxp3+CD4+ cells were demonstrated to infiltrate higher in TM while CD20+ and CD68+ cells were also enriched in NS and TB regions respectively. However, none of them showed correlations with PD-L1 expression. ARID1A, STK11 alterations were enriched in the low PD-L1 group significantly, while anti-viral immune associated APOBEC mutation signature and immune-related genes expression such as XCL1 and IL11 were positively associated with PD-L1 expression (p<0.05). This is a comprehensive investigation revealing immune and molecular features of HPV(+)OPSCC based on PD-L1 expression. Our study suggested that 73.2% of HPV(+)OPSCC patients may benefit from immunotherapy, and high PD-L1 expression reflects immune-active status of HPV(+)OPSCC accompanied by higher immune effect factors such as TMB, CD8+ cytotoxic T cells and immune-related genomic alterations. Our study offers valuable information for understanding the immune features of HPV(+)OPSCC.

Determining PD-L1 expression in head and neck squamous cell carcinoma using immunohistochemistry.

Advanced head and neck squamous cell carcinoma (HNSCC) has limited treatment options. Programmed death-ligand1 (PD-L1) expressed by tumor cells interacts with PD-1 receptor on T lymphocytes leading to immune evasive response and survival advantage. Therapy with immune check-point inhibitors target PD-1/PD-L1 blockade inducing tumor regression. Immunohistochemistry (IHC) for PD-L1 expression enables patient selection for immunotherapy and may be considered a potential predictor of clinical response. A retrospective analysis of IHC for PD-L1 expression using manual laboratory developed technique (LDT) with antibody clone 22C3 (Dako) in 93 cases of HNSCC. PD-L1 expression was correlated with age, gender, tumor site, grade and stage. PD-L1 IHC was performed in 93 cases and immunopositivity was noted in 59 (63.4%) cases. High expression with combined proportion score (CPS) ≥50 was seen in 15 (16.1%) cases and low expression with CPS ≥1 expression was seen in 44 (47.3%) cases. An almost-perfect interobserver agreement was noted by two pathologists for PD-L1 IHC expression (Cohen's kappa coefficient = 0.910). No statistically significant correlation was noted between PD-L1 score and patient demographics, tumor site, grade or stage. Detection of PD-L1 status by IHC enables identification of HNSCC patients eligible for future targeted immunotherapy.