PD-L1 Testing in Gastric Cancer by the Combined Positive Score of the 22C3 PharmDx and SP263 Assay with Clinically Relevant Cut-offs

Yujun Park; Jiwon Koh; Hee Young Na; Yoonjin Kwak; Keun-Wook Lee; Sang-Hoon Ahn; Do Joong Park; Hyung-Ho Kim; Hye Seung Lee

doi:10.4143/crt.2019.718

Articles

Page Path: HOME > Cancer Res Treat > Volume 52(3); 2020 > Article

Original Article PD-L1 Testing in Gastric Cancer by the Combined Positive Score of the 22C3 PharmDx and SP263 Assay with Clinically Relevant Cut-offs: Yujun Park, MD¹, Jiwon Koh, MD, PhD², Hee Young Na, MD¹, Yoonjin Kwak, MD, PhD², Keun-Wook Lee, MD, PhD³, Sang-Hoon Ahn, MD, PhD⁴, Do Joong Park, MD, PhD⁴, Hyung-Ho Kim, MD, PhD⁴, Hye Seung Lee, MD, PhD^1,; Cancer Research and Treatment : Official Journal of Korean Cancer Association 2020;52(3):661-670.
DOI: https://doi.org/10.4143/crt.2019.718
Published online: January 10, 2020

¹Department of Pathology, Seoul National University Bundang Hospital, Seoul National University College of Medicine, Seongnam, Korea

²Department of Pathology, Seoul National University Hospital, Seoul National University College of Medicine, Seoul, Korea

³Department of Internal Medicine, Seoul National University Bundang Hospital, Seoul National University College of Medicine, Seongnam, Korea

⁴Department of Surgery, Seoul National University Bundang Hospital, Seoul National University College of Medicine, Seongnam, Korea

Correspondence: Hye Seung Lee, MD, PhD Department of Pathology, Seoul National University Bundang Hospital, Seoul National University College of Medicine, 82 Gumi-ro 173beon-gil, Bundang-gu, Seongnam 13620, Korea
Tel: 82-31-787-7714 Fax: 82-31-787-4012 E-mail: hye2@snu.ac.kr

• Received: November 25, 2019 • Accepted: January 9, 2020

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

18,030 Views
718 Download
78 Web of Science
79 Crossref
82 Scopus

prev next

Full Article

Download PDF

Abstract

Purpose
We provide a comparison between 22C3 pharmDx and SP263 assay, for evaluating programmed death ligand 1 (PD-L1) expression in advanced gastric cancer (GC) patients.
Materials and Methods
The PD-L1 immunohistochemistry by 22C3 pharmDx and SP263 assays was performed in the center of the tumor (CT) and invasive margin (IM) in 379 GC tissues using tissue microarrays and interpreted as combined positive score (CPS) and tumor proportion score (TPS). Of the total samples, 55 samples were independently reviewed by five pathologists.
Results
The two assays showed a high correlation in both the CPS and TPS. At a CPS ≥ 1 cut-off, 219 (57.8%) and 231 (60.9%) GCs were positive for PD-L1 with the 22C3 and SP263 assays, and at ≥ 10 cut-off, 37 (9.8%) and 36 (9.5%) GCs were positive, respectively. The overall percent agreement (OPA) was greater than 90% with CPS ≥ 1 and ≥ 10 cut-offs, and TPS ≥ 1% and ≥ 10% cut-offs. There was higher OPA between the two assays with a CPS cut-off ≥ 10 (99.2%) than ≥ 1 (94.7%). The percent agreement between the CT and IM was higher with a CPS cut-off ≥ 10 (92.9%) than ≥ 1 (77.6%). Patient with positive expression at CPS ≥ 5 cut-off had a significantly better outcomes in both assays. Interobserver variability among five pathologists was higher than the assay variability.
Conclusion
Two assays for PD-L1 expression in GC showed high agreement. These results provide guidance for selecting eligible patients with GC for pembrolizumab treatment.
Key words: Programmed cell death ligand 1, Immunohistochemistry, 22C3 pharmDx, SP263 assay, Gastric neoplasms

Introduction

Gastric cancer (GC) is the fifth most common cancer and the third leading cause of cancer-related death worldwide [1]. The 5-year relative survival rate is approximately 55% in patients with stage II or III GC [2]. The last few decades have witnessed great advances in the treatment of patients with advanced GC, including postoperative adjuvant chemotherapy [3] and molecular targeted therapeutics [4].

Recent studies have demonstrated favorable outcomes of immunotherapy for patients with advanced cancer treated with immune checkpoint inhibitors, including anti–programmed death 1 receptor (PD-1)/programmed death ligand 1 (PD-L1) inhibitor [5]. PD-1 binds to its ligands PD-L1 and PD-L2 on the tumor cells, allowing for immune escape [6]. PD-L1 protein expression in viable cancer cells determined by immunohistochemistry (IHC) is correlated with a therapeutic effect of immune checkpoint inhibitors, and is thus considered an important biomarker for the use of anti–PD-1/PD-L1 inhibitors in clinical trials. Based on these clinical trial results, the Food and Drug Administration (FDA) approved PD-L1 IHC as a companion diagnostic modality for some solid tumors, including GC [7].

Based on the phase II KEYNOTE 59 trial [8], in September 2017, pembrolizumab was approved by the FDA for patients with advanced or metastatic GC and gastroesophageal junction (GEJ) cancer who had undergone previous treatment with at least two lines of chemotherapy. The PD-L1 IHC 22C3 pharmDx was approved by the FDA as a companion diagnostic assay for the use of pembrolizumab. PD-L1 expression in patients with GC and GEJ cancer evaluated using a combined positive score (CPS) has been proposed, in which a cutoff CPS ≥ 1 would indicate positive PD-L1 expression [8]. The more recent phase III KEYNOTE-061 trial evaluated pembrolizumab monotherapy as a second-line chemotherapy for patients with advanced GC or GEJ cancer with CPS ≥ 1, who were previously treated with first-line chemotherapy of platinum-containing and fluoropyrimidine-containing drugs, which demonstrated no significant improvement of pembrolizumab for improving overall survival (OS) compared to paclitaxel as second-line therapy. However, advanced GC patients with higher levels of PD-L1 expression, such as CPS ≥ 10, did achieve a significant therapeutic benefit from pembrolizumab [9].

Thus, FDA-approved PD-L1 IHC companion diagnostic assays should be performed to assess whether anti–PD1/PD-L1 inhibitors are appropriate for a given patient. Each companion diagnostic assay includes a specific antibody clone and staining platform associated with a specific inhibitor. However, not all pathology laboratories have the available resources to perform the various companion diagnostic assays required for specific inhibitors. To address this challenge, some studies have investigated the possibility of harmonizing these various PD-L1 IHC assays for non-small cell lung cancer (NSCLC) and urothelial carcinoma (UC) [10,11]. The 22C3 pharmDx is currently the only companion diagnostic assay used in GC, and no comparative studies have been conducted on the interchangeability of different IHC assays such as the SP263 assay for GC patients, especially in terms of the CPS.

Therefore, in this study, the PD-L1 IHC 22C3 pharmDx and SP263 assay were performed on surgically resected specimens from patients with advanced GC, and the level of PDL1 expression was determined according to the CPS and tumor proportion score (TPS) with clinically relevant cut-off values. Additional independent assessments by four other pathologists were further compared to determine the interobserver variability for each assay.

Materials and Methods

1. Patients and samples

In total, 379 consecutive patients with stage II and III GC who underwent surgical resection at Seoul National University Bundang Hospital (Seongnam, Korea) between 2006 and 2013 were enrolled in this study. All patients were treated by fluoropyrimidine-based adjuvant chemotherapy after radical surgical resection. Clinicopathological characteristics were retrospectively collected from medical records and pathology reports. OS was defined as the time from the date of operation to death of any cause or censored observation.

Formalin-fixed paraffin-embedded tissue specimens were prepared from the surgically resected GC specimens. The 2-mm core tissue microarrays (TMAs) were constructed as described previously (SuperBioChips Laboratories, Seoul, Korea) [12]. In all 379 samples, we selected representative tissue cores at the center of the tumor (CT) and invasive margin (IM), respectively.

2. PD-L1 IHC and interpretation

Two PD-L1 IHC diagnostic assays were performed on each specimen according to the manufacturer’s instructions: 22C3 pharmDx (mouse monoclonal primary anti–PD-L1 antibody, prediluted, clone 22C3, Dako, Carpinteria, CA) on the Autostainer Link 48 with EnVision DAB Detection System (Agilent Technologies, Santa Clara, CA), and Ventana SP263 (rabbit monoclonal primary anti–PD-L1 antibody, prediluted, Ventana Medical Systems, Tucson, AZ) on the Benchmark XT staining systems and Ultra with OptiView Universal DAB Detection Kit (Ventana Medical Systems) [8,13].

Interpretation of the 22C3 pharmDx and SP263 assays was performed from stained slides by two of the authors (Y.P. and Y.K.) who received appropriate training. PD-L1 expression in the tumor cell membrane and membrane and/or cytoplasm of tumor-associated mononuclear inflammatory cells such as lymphocytes and macrophages was scored. The CPS was defined as the total number of tumor cells and immune cells (including lymphocytes and macrophages) stained with PD-L1 divided by the number of all viable tumor cells, then multiplied by 100 [9]. Each countable array core section contained at least 100 viable GC cells.

3. Interobserver agreement

We randomly selected one TMA block containing 55 GC samples. To assess the variabilities observed between two PD-L1 IHC assays and five observers, five pathologists (Y.P., J.K., H.Y.N., Y.K., and H.S.L.) independently evaluated PDL1 expression according to the CPS and TPS in the 55 samples. All pathologists were blinded to clinical information as well as the evaluation results of other pathologists.

4. p53 IHC

IHC for p53 (DO7, mouse monoclonal, Dako) was performed using an automated immunostainer (BenchMark XT, Ventana Medical Systems), according to the manufacturer’s instructions. To evaluate the overexpression of p53 in tumor cells, strong intensity of nuclear staining in ≥ 10% of tumor cells was defined as p53 overexpression/positive, while the cases with < 10% positive tumor cells including those expressing dispersed or partial positive cells were defined as negative [14].

5. Microsatellite instability analysis

To evaluate the microsatellite instability (MSI) status of tumor cells, polymerase chain reaction (PCR) was performed using five NCI markers (BAT-26, BAT-25, D5S346, D17S250, and S2S123) in both tumor cells and non-neoplastic normal tissues of the same specimens. An automated DNA sequence analyzer (ABI 3731 Genetic Analyzer, Applied Biosystems, Foster City, CA) was used to determine the MSI status in the PCR product. MSI status was assessed by alteration of the allele profiles in tumor cells compared with matched non-neoplastic normal tissues. Two or more NCI markers with unstable peaks were defined as MSI-high (MSI-H) and with one or no unstable NCI marker as MSI-low (MSI-L)/microsatellite stable (MSS) [15].

6. Epstein-Barr virus in situ hybridization

Epstein-Barr virus (EBV) in situ hybridization (ISH) via a fluorescein-conjugated EBV‐encoded small RNA (EBER) oligonucleotide probe (INFORM EBV-encoded RNA probe, Ventana Medical Systems) was used to determine the EBV status of tumor cells. EBV-positive GCs were defined as the presence of tumor cells with EBER expression in the nucleus [15].

7. Statistical analysis

SPSS ver. 25.0 (IBM Corp., Armonk, NY) and Stata 14 (StataCorp LLC, College Station, TX) were used for statistical analyses. For pairwise positive correlations between assays, Spearman correlation coefficients were calculated. For pairwise comparison of the 22C3 pharmDx and SP263 assay, the overall percent agreement (OPA), positive percent agreement (PPA), and negative percent agreement (NPA) were calculated with the following clinically relevant cut-off values: CPS ≥ 1, CPS ≥ 10, TPS ≥ 1%, and TPS ≥ 10% [9,16]. An OPA of at least 90% between the two assays was considered to indicate agreement between assays [17]. Interobserver agreement between the pathologists was evaluated according to intraclass correlation coefficients (ICC) for CPS and TPS, and by Fleiss’ kappa value for CPS cut-off values of ≥ 1 and ≥ 10, respectively. Each pathologist obtained OPA, PPA, NPA, and Cohen's kappa values between the two assays with CPS cut-offs ≥ 1 and ≥ 10. OS was calculated by KaplanMeier method and evaluated by the log-rank test. Cox proportional hazards model for both univariate and multivariate analyses of survival to identify the significance of independent prognostic factors were tested. p-values of < 0.05 were considered to indicate statistically significant difference.

8. Ethical statement

All tissue samples used in the study were obtained from Department of Pathology, Seoul National University Bundang Hospital, and the clinicopathological information and patients’ survival data in the hospital medical records were used. The Institutional Review Board (IRB) of Seoul National University Bundang Hospital approved our study (IRB number: B-1606/349-308). Since this study was the retrospective study, written informed consent were waived for all participants.

Results

1. PD-L1 staining by 22C3 pharmDx and SP263 assay

PD-L1 IHC was performed using the 22C3 pharmDx and SP263 assay and the CPS and TPS were calculated for each sample. Representative IHC images of PD-L1 determined with the 22C3 pharmDx and SP263 assay showed similar staining patterns for the same tissue core (Fig. 1). To compare the SP263 assay with the 22C3 pharmDx, the SP263 assay showed stronger staining intensity and prominent membrane staining than the 22C3 pharmDx. There was a strong positive correlation between the two assays for all specimens, including the CT and the IM, when comparing the CPS and TPS (Fig. 2). The Spearman correlation coefficient between the two assays was 0.943 for TPS, which was slightly higher than that for CPS (0.914). Correlation coefficients of the TPS between the two assays were 0.951 for CT and 0.935 for IM, and those of the CPS were 0.916 in the CT and 0.912 in the IM.

The distribution of CPS and TPS is shown in Fig. 3. In general, the distribution of CPS and TPS results was similar for the two assays, although the number of PD-L1–negative cases (CPS < 1 or TPS < 1) was slightly higher based on the 22C3 pharmDx results than the SP263 assay results. For both assays, the distribution of CPS results ranged from 0 to 10 in most cases, and the TPS results similarly ranged from 0 to 10% for the CT and IM.

2. Clinical diagnostic comparison between the two assays

Of the 379 cases, when positive expression of PD-L1 was defined as CPS ≥ 1, similar numbers of PD-L1–positive expression were determined by the two assays, although the positive rates for SP263 assay were slightly higher than for the 22C3 pharmDx for both the CT and IM (Table 1). At the CPS ≥ 10 cut-off, the 22C3 pharmDx detected slightly more PD-L1–positive cases than the SP263 assay for both the CT and IM. However, the difference in the positive cases between the two assays was greater for CPS ≥ 10 than for CPS ≥ 1.

The correlations between clinicopathologic characteristics and PD-L1 expression are summarized in S1 Table. In brief, at the CPS ≥ 10 cut-off, PD-L1 expression was significantly associated with less aggressive clinicopathologic features, including pTNM stage II (p=0.010), expanding tumor border (p < 0.001), and absence of perineural invasion (p < 0.001), except for the correlation with poorly differentiation (p < 0.001). At both CPS ≥ 1 and ≥ 10 cut-offs, PD-L1 expression were significantly associated with EBV positivity (p=0.001 at CPS ≥ 1, p < 0.001 at CPS ≥ 10) and MSI-H status (p=0.002 at CPS ≥ 1, p < 0.001 at CPS ≥ 10) in GCs.

Representative IHC images of the discordant results for PD-L1 expression in the CT and IM are shown in S2 Fig. The concordance rate and Cohen's κ values of PD-L1 expression between the CT and IM were higher for CPS ≥ 10 with both assays (> 90%) than with CPS ≥ 1 (Table 2).

OPA, PPA, and NPA were calculated at their clinically relevant PD-L1 CPS cut-offs (≥ 1 and ≥ 10) and TPS cut-offs (≥ 1% and ≥ 10%). As shown in Table 3, at all CPS and TPS cut-offs, the OPA was greater than 90% between the two assays and was greater for CPS ≥ 10 (CT, 99.2%; IM, 98.7%) than for CPS ≥ 1 (94.7%; 94.5%) at both the CT and the IM. The PPA and NPA also ranged from 90% to 100% for all cutoffs.

3. Survival analysis comparison of two assays

According to Kaplan-Meier survival analysis, PD-L1 positive expression tended to have a better prognosis at CPS both CPS ≥ 1 and ≥ 10 cut-offs, but there was no statistically significant difference (all p > 0.05) (S3A-D Fig.). On the other hand, patient with negative expression of PD-L1 at CPS ≥ 5 cut-off had a significantly poor outcome in both assays (22C3 pharmDx, p=0.021; SP263 assay, p=0.033) (S4A and S4B Fig.). Multivariate Cox regression analysis indicated that PD-L1 expression by both assays was an independent prognostic factor (22C3 pharmDx, p=0.010; SP263 assay, p=0.035) as well as tumor size and pathologic TNM stage (S5 Table). In addition, at all TPS cut-offs including ≥ 5 cut-off value, there were no significant OS differences according to PD-L1 expression (all p > 0.05, data not shown).

4. Interobserver agreement

To determine the interobserver agreement of PD-L1 interpretation, CPS and TPS results in 55 array cores were obtained from five pathologists, and the ICC for CPS and TPS and Fleiss’ kappa at two CPS cut-offs (≥ 1 and ≥ 10) were measured to assess interobserver variability between five pathologists for the two assays. Table 4 shows the ICCs for CPS among the five pathologists with 55 of the array cores, which was slightly higher for the 22C3 pharmDx (ICC, 0.387) than for the SP263 assay (0.349), while the ICCs for TPS was higher for the SP263 assay (0.710) than for the 22C3 pharmDx (0.596). The ICCs for TPS were higher for both assays than the ICCs for CPS. The concordance as measured by the Fleiss’ kappa analysis was higher for the 22C3 pharmDx (CPS ≥ 1, κ=0.389; CPS ≥ 10, κ=0.256) than for the SP263 assay (κ=0.224 and κ=0.140, respectively) at both the CPS ≥ 1 and CPS ≥ 10 cut-offs.

OPA, PPA, NPA, and Cohen’s κ values between the two assays were assessed by each of the five pathologists at multiple CPS cut-offs (≥ 1 and ≥ 10) (Table 5). At the CPS ≥ 1 cutoff, the OPA was greater than 90% by pathologists A, C, and E, and the PPA was greater than 90% by all pathologists. In contrast, only one pathologist (pathologist E) found an NPA greater than 90%. Moderate concordance, as measured by the Cohen’s kappa value, was observed by pathologists A (κ=0.757), D (κ=0.505), and E (κ=0.963). At the CPS ≥ 10 cutoff, the OPA was greater than 90% by pathologists A, D, and E. These same three pathologists also found PPA and NPA values greater than 80% and reported substantial concordance, as measured by the Cohen’s kappa value (A, κ=0.819; D, κ=0.707; E, κ=1.000) (Table 5).

Discussion

Since pembrolizumab was first approved by the FDA in September 2014 for the treatment of advanced melanoma, immune checkpoint inhibitors, including various anti–PD-1/PD-L1 inhibitors, have shown remarkable outcomes in clinical trials of some advanced solid tumors [7]. The companion diagnostic test that predicts the therapeutic response of patients with advanced cancer to an anti–PD-1/PD-L1 inhibitor has also been approved by the FDA. A PD-L1 IHC assay is considered to be the most effective and widely used diagnostic test for anti–PD-1/PD-L1 inhibitors [7]. However, some studies have reported that PD-L1 expression has some limitations as a biomarker because of the relatively high heterogeneity and interobserver variability [18,19].

Each anti–PD-1/PD-L1 inhibitor is associated with its own specific diagnostic assay, staining platform, and PD-L1 expression evaluation cut-off value set for selecting patients for treatment. As a representative example, pembrolizumab was shown to be applicable for patients with PD-L1–positive expression when using only the FDA-approved PD-L1 IHC 22C3 pharmDx and Agilent Autostainer Link 48 platform. This rigid application has complicated use of the biomarker test for anti–PD-1/PD-L1 inhibitors [20]. To resolve this complexity, the interchangeability of various PD-L1 IHC assays has been evaluated [21]. A comparative study of the Dako 22C3 pharmDx, Dako 28-8 pharmDx, and Ventana SP263 assay in NSCLC patients by AstraZeneca showed that suitable alternatives for each other could be used to select patients with NSCLC who were suitable for anti–PD-1/PD-L1 inhibitors [22]. Based on this study, Ventana SP263 assay performed on the Ventana Benchmark Ultra platform obtained CE-marked to determine that pembrolizumab could be used for patients with NSCLC. Although the results of comparative studies of different PD-L1 IHC diagnostic assays for patients with NSCLC and UC have shown some good agreement considering interchangeability [10,11,23], recent study has raised concerns about this interchangeability [24]. Furthermore, no comparative study of these assays has been conducted for patients with GC to date; thus, we here provide the first comparison between the 22C3 pharmDx and SP263 assay in GC.

In a previous study comparing the results of the 22C3 pharmDx and SP263 assay in patients with NSCLC, the staining of the SP263 assay in tumor cells was more intense and membrane staining was particularly strong, suggesting that this assay could result in more PD-L1 positivity [24]. Consistently, we found that the number of PD-L1–positive cells was higher at all CPS and TPS cut-off values for the SP263 assay.

Our data further showed that the OPA between the 22C3 pharmDx and SP263 assays was greater than 90% at all CPS and TPS cut-off values. Compared with the CPS method, the OPA was higher based on the TPS, and higher OPA values were also observed with higher CPS and TPS cut-off values. In addition, the ICC of the TPS was higher than that of the CPS to confirm the interobserver variability among the five pathologists. Previous studies have shown that PD-L1 expression in immune cells in patients with NSCLC was more discordant than that in tumor cells in comparing different PD-L1 IHC assays, including the 22C3 pharmDx and SP263 assay, and high interobserver variability was observed [10,25]. In contrast with the TPS method, the CPS method evaluates immune cells such as lymphocytes and macrophages as well as tumor cells. The clinical utility of the CPS method was confirmed in patients with GC in the KEYNOTE-059 trial and has the advantage of being a predictive biomarker capable of comprehensively evaluating PD-L1 expression in both tumor and immune cells [26]. Despite the apparent reliability and reproducibility of the CPS method in this trial [26], there are still limited studies on the use of CPS for the evaluation of immune cells in GC, and thus further investigation is needed in this regard as well as in other solid tumors such as NSCLC.

Previous studies demonstrated that PD-L1 expression heterogeneity and interobserver variability were higher than the assay variability in PD-L1 IHC assays [18,19]. Consistently, we found that the PD-L1 expression heterogeneity and interobserver variability were higher than the assay variability between the 22C3 pharmDx and SP263 assay. With respect to expression heterogeneity, the concordance of the results for the CT and IM with CPS ≥ 1 was 77.6% (Cohen’s κvalue= 0.559) with the 22C3 pharmDx assay and was 77.8% (Cohen’s κvalue=0.564) with the SP263 assay, which were lower than the values determining the overall agreement between the two assays. Previous studies have shown higher discrepancies with higher TPS cut-offs in biopsy and resection specimens [18,27], whereas we found higher agreement between the CT and IM with CPS ≥ 10 than with the lower cut-off of CPS ≥ 1. When comparing interobserver variability with assay variability, three pathologists (A, D, and E) obtained higher κ values between the two assays than those obtained among the five pathologists for both CPS ≥ 1 and ≥ 10 cutoffs. Two pathologists (A and D) went through a formal training program to evaluate CPS by the 22C3 pharmDx assay, and pathologist E learned the PD-L1 IHC interpretation method from pathologist D. The other two untrained pathologists (B and C) showed higher assay variability compared to that of the trained participants for both the CPS ≥ 1 and CPS ≥ 10 cut-offs. Three pathologists (A, D, and E) also showed higher agreement between the two assays at CPS ≥ 10 than at CPS ≥ 1. The concordance, as measured by the Cohen’s kappa value, was also the highest between pathologists D and E at the CPS ≥ 1 (22C3 pharmDx, κ=0.741; SP263 assay, κ=0.712) and ≥ 10 (κ=0.867 and κ=0.824, respectively) cut-offs for both assays. The lowest interobserver variability between pathologists D and E may be considered to reflect the influence of their better training. Since there are limited studies on the expression and interobserver variability in GC, further data are needed to validate these results.

In our study, significant differences in OS were identified in the CPS method, which evaluated tumor cells and tumorassociated immune cells, not the TPS method, which evaluated tumor cells alone. Previous studies, including recent randomized controlled trial, reported that patient with positive expression of PD-L1 in stromal immune cells had a significantly good outcome, but no statically significant difference in positive expression of PD-L1 in tumor cells [28]. However, prognostic significance of PD-L1 expression is still uncertain because PD-L1 expression did not have prognostic significance in some studies, but positive expression of PDL1 in tumor cells was a poor prognostic factor in other studies [29]. This diversity might be due to difference in PD-L1 antibody, difference in the clinical stage of the patient cohort and difference in subject and cut-off value to assess PD-L1 expression. We applied clinically relevant PD-L1 antibodies and interpretation guidelines, and PD-L1 positive expression by both 22C3 pharmDx and SP263 assays was an independent better prognostic factor at the CPS ≥ 5 cut-off. Further studies are needed for confirming these prognostic differences.

There are some limitations to this study that should be mentioned. Because this was a retrospective study design conducted in a single institution, specimen bias is possible. In addition, effects of different staining platforms could influence the results of PD-L1 IHC in clinical practice, which was not considered in the study. However, this study was conducted in a large homogenous cohort of patients with advanced GC. Since the PD-L1 IHC was performed on the TMA slide, it was possible to reduce the difference in staining conditions between the cases. In the first place, this study represents the first comparative study to use clinically relevant PD-L1 IHC cut-offs in GC. Therefore, this study may provide a basis for a better diagnostic process for the use of anti–PD-1/PD-L1 inhibitors in GC patients.

In conclusion, the 22C3 pharmDx and SP263 assay showed high agreement for the same GC specimens, but expression heterogeneity and interobserver variability were also found to be higher than assay variability. In addition, the higher cut-off value with the CPS method resulted in greater interchangeability between the 22C3 pharmDx and SP263 assay. More studies are needed in this regard with consideration of the best ways to evaluate PD-L1 expression, such as those addressing issues of expression heterogeneity, interobserver variability, and assay variability in GC when compared to NSCLC. Therefore, our study can provide insights for improving the selection of patients eligible for pembrolizumab treatment.

Electronic Supplementary Material

Supplementary materials are available at Cancer Research and Treatment website (https://www.e-crt.org).

S1 Table.

Clinicopathologic correlation of PD-L1 expression in patients with gastric cancers using CPS

crt-2019-718-suppl1.pdf

S2 Fig.

Discordant programmed death ligand 1 expression in the center of the tumor (CT) and inva sive margin (IM) in the same gastric cancer sample based on the by 22C3 pharmDx (×50).

crt-2019-718-suppl2.pdf

S3 Fig.

Kaplan Meier analysis of overall survival in patients with programmed death ligand 1 expression for the 22C3 pharm Dx at the combined positive score (CPS) 1 (A) and CPS 10 (B) cut offs, and the SP263 assay at the CPS 1 (C) and CPS 10 (D) cut offs.

crt-2019-718-suppl3.pdf

S4 Fig.

Kaplan Meier analysis of overall survival in patients with programmed death ligand 1 expression for the 22C3 pharmDx (A) and the SP263 assay (B) at the combined positive score 5 cut offs.

crt-2019-718-suppl4.pdf

S5 Table.

Univariate and multivariate analysis of overall survival using Cox proportional hazards model

crt-2019-718-suppl5.pdf

NOTES

Conflict of interest relevant to this article was not reported.

Acknowledgements

This work was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (NRF-2016R1D1A1B03931744).

Fig. 1.

Representative tumor microarray core stained with the 22C3 pharmDx and SP 263 assay according to the combined positive score (CPS) distribution (core magnification, ×50; inlet, ×200).

Fig. 2.

Correlation between the 22C3 pharmDx and SP263 assay for the combined positive score (CPS) (A) and tumor proportion score (TPS) (B) at the center of the tumor and invasive margin in gastric cancer samples.

Fig. 3.

Programmed death ligand 1 expression distribution of the 22C3 pharmDx and SP263 assay for the combined positive score (CPS) (A, B) and tumor proportion score (TPS) (C, D) at the center of the tumor (CT) and invasive margin (IM) by intervals.

Table 1.

The number of PD-L1–positive gastric cancer cases

	22C3 (n=379)				SP263 (n=379)
	CT		IM		CT		IM
	Positive	Negative	Positive	Negative	Positive	Negative	Positive	Negative
CPS ≥ 1	219 (57.8)	160 (42.2)	170 (44.9)	209 (55.1)	231 (60.9)	148 (39.1)	175 (46.2)	204 (53.8)
CPS ≥ 10	37 (9.8)	342 (90.2)	50 (13.2)	329 (86.8)	36 (9.5)	343 (90.5)	49 (12.9)	330 (87.1)
TPS ≥ 1%	118 (31.1)	261 (68.9)	104 (27.4)	275 (72.6)	125 (33.0)	254 (67.0)	115 (30.3)	264 (69.7)
TPS ≥ 10%	28 (7.4)	351 (92.6)	30 (7.9)	349 (92.1)	29 (7.7)	350 (92.3)	34 (9.0)	345 (91.0)

Values are presented as number (%). PD-L1, programmed death ligand 1; CT, center of tumor; IM, invasive margin; CPS, combined positive score; TPS, tumor proportion score.

Table 2.

Agreement and Cohen’s kappa values between center of tumor and invasive margin

	22C3		SP263
	Agreement (%)	Cohen's kappa (lower 95% CI)	Agreement (%)	Cohen's kappa (lower 95% CI)
CPS ≥ 1	77.6	0.559 (47.8)	77.8	0.564 (48.5)
CPS ≥ 10	92.9	0.650 (52.9)	91.8	0.590 (46.0)
TPS ≥ 1%	84.7	0.631 (54.6)	85.8	0.671 (59.0)
TPS ≥ 10%	94.7	0.623 (47.5)	94.5	0.637 (49.3)

CI, confidence interval; CPS, combined positive score; TPS, tumor proportion score.

Table 3.

OPA, PPA, and NPA between 22C3 pharmDx and SP263 assay

	OPA (lower 95% CI, %)		PPA (lower 95% CI, %)		NPA (lower 95% CI, %)
	CT	IM	CT	IM	CT	IM
CPS ≥ 1	94.7 (92.0)	94.5 (91.7)	98.2 (95.4)	95.3 (90.9)	90.0 (84.3)	93.8 (89.6)
CPS ≥ 10	99.2 (97.7)	98.7 (97.0)	94.6 (81.8)	94.0 (83.5)	99.7 (98.4)	99.4 (97.8)
TPS ≥ 1%	97.1 (94.9)	96.6 (94.2)	98.3 (94.0)	99.0 (94.8)	96.6 (93.6)	95.6 (92.5)
TPS ≥ 10%	99.7 (98.5)	98.4 (96.6)	100 (87.7)	96.7 (82.8)	99.7 (98.4)	98.6 (96.7)

OPA, overall percent agreement; PPA, positive percent agreement; NPA, negative percent agreement; CI, confidence interval; CT, center of tumor; IM, invasive margin; CPS, combined positive score; TPS, tumor proportion score.

Table 4.

Interobserver variation between five pathologists by 22C3 pharmDx and SP263 assay

	Intraclass correlation coefficient (lower 95% CI)		Fleiss' Kappa (lower 95% CI)
	CPS	TPS	CPS ≥ 1	CPS ≥ 10
22C3	0.387 (20.9)	0.596 (40.5)	0.389 (26.4)	0.224 (8.0)
SP263	0.349 (13.5)	0.710 (57.2)	0.256 (15.6)	0.140 (2.4)

CI, confidence interval; CPS, combined positive score; TPS, tumor proportion score.

Table 5.

OPA, PPA, NPA, and Cohen’s kappa values between 22C3 pharmDx and SP263 assay (55 samples)

	OPA (lower 95% CI, %)	PPA (lower 95% CI, %)	NPA (lower 95% CI, %)	Cohen's kappa (lower 95% CI)
CPS ≥ 1
A	90.9 (80.1)	100 (91.0)	68.8 (41.3)	0.757 (55.4)
B	87.3 (75.5)	100 (92.5)	12.5 (0.3)	0.196 (–14.5)
C	98.2 (90.3)	100 (93.4)	0	-^a)
D	76.4 (63.0)	96.7 (82.8)	52.0 (31.3)	0.505 (28.5)
E	98.2 (90.3)	100 (89.4)	95.5 (77.2)	0.962 (88.5)
CPS ≥ 10
A	90.9 (80.1)	100 (86.8)	82.8 (64.2)	0.819 (66.6)
B	87.3 (75.5)	100 (92.5)	12.5 (0.3)	0.196 (–14.5)
C	80.0 (67.0)	88.6 (73.3)	65.0 (40.8)	0.554 (31.5)
D	90.9 (80.1)	88.9 (51.8)	91.3 (79.2)	0.707 (46.1)
E	100 (93.5)	100 (66.4)	100 (92.3)	100 (100)

OPA, overall percent agreement; PPA, positive percent agreement; NPA, negative percent agreement; CI, confidence interval; CPS, combined positive score.

^a) In all samples, SP263 CPS 1 is positive.

REFERENCES

1. Ferlay J, Soerjomataram I, Dikshit R, Eser S, Mathers C, Rebelo M, et al. Cancer incidence and mortality worldwide: sources, methods and major patterns in GLOBOCAN 2012. Int J Cancer. 2015;136:E359–86. Article PubMed
2. Jung KW, Won YJ, Kong HJ, Oh CM, Shin A, Lee JS. Survival of korean adult cancer patients by stage at diagnosis, 20062010: national cancer registry study. Cancer Res Treat. 2013;45:162–71. Article PubMed PMC PDF
3. Sasako M, Sakuramoto S, Katai H, Kinoshita T, Furukawa H, Yamaguchi T, et al. Five-year outcomes of a randomized phase III trial comparing adjuvant chemotherapy with S-1 versus surgery alone in stage II or III gastric cancer. J Clin Oncol. 2011;29:4387–93. Article PubMed
4. Bang YJ, Van Cutsem E, Feyereislova A, Chung HC, Shen L, Sawaki A, et al. Trastuzumab in combination with chemotherapy versus chemotherapy alone for treatment of HER2-positive advanced gastric or gastro-oesophageal junction cancer (ToGA): a phase 3, open-label, randomised controlled trial. Lancet. 2010;376:687–97. Article PubMed
5. Sharma P, Allison JP. Immune checkpoint targeting in cancer therapy: toward combination strategies with curative potential. Cell. 2015;161:205–14. Article PubMed PMC
6. Ribas A. Tumor immunotherapy directed at PD-1. N Engl J Med. 2012;366:2517–9. Article PubMed
7. Gong J, Chehrazi-Raffle A, Reddi S, Salgia R. Development of PD-1 and PD-L1 inhibitors as a form of cancer immunotherapy: a comprehensive review of registration trials and future considerations. J Immunother Cancer. 2018;6:8.Article PubMed PMC PDF
8. Fuchs CS, Doi T, Jang RW, Muro K, Satoh T, Machado M, et al. Safety and efficacy of pembrolizumab monotherapy in patients with previously treated advanced gastric and gastroesophageal junction cancer: phase 2 clinical KEYNOTE-059 trial. JAMA Oncol. 2018;4:e180013Article PubMed PMC PDF
9. Shitara K, Ozguroglu M, Bang YJ, Di Bartolomeo M, Mandala M, Ryu MH, et al. Pembrolizumab versus paclitaxel for previously treated, advanced gastric or gastro-oesophageal junction cancer (KEYNOTE-061): a randomised, open-label, controlled, phase 3 trial. Lancet. 2018;392:123–33. Article PubMed
10. Hirsch FR, McElhinny A, Stanforth D, Ranger-Moore J, Jansson M, Kulangara K, et al. PD-L1 immunohistochemistry assays for lung cancer: results from phase 1 of the Blueprint PD-L1 IHC Assay Comparison Project. J Thorac Oncol. 2017;12:208–22. Article PubMed
11. Hodgson A, Slodkowska E, Jungbluth A, Liu SK, Vesprini D, Enepekides D, et al. PD-L1 immunohistochemistry assay concordance in urothelial carcinoma of the bladder and hypopharyngeal squamous cell carcinoma. Am J Surg Pathol. 2018;42:1059–66. Article PubMed PMC
12. Yun S, Koh J, Nam SK, Park JO, Lee SM, Lee K, et al. Clinical significance of overexpression of NRG1 and its receptors, HER3 and HER4, in gastric cancer patients. Gastric Cancer. 2018;21:225–36. Article PubMed PDF
13. Planchard D, Yokoi T, McCleod MJ, Fischer JR, Kim YC, Ballas M, et al. A phase III study of durvalumab (MEDI4736) with or without tremelimumab for previously treated patients with advanced NSCLC: rationale and protocol design of the ARCTIC study. Clin Lung Cancer. 2016;17:232–6. Article PubMed
14. Chang MS, Lee JH, Kim JP, Kim HS, Lee HS, Kim CW, et al. Microsatellite instability and Epstein-Barr virus infection in gastric remnant cancers. Pathol Int. 2000;50:486–92. Article PubMed
15. Lee HS, Kim WH, Kwak Y, Koh J, Bae JM, Kim KM, et al. Molecular testing for gastrointestinal cancer. J Pathol Transl Med. 2017;51:103–21. Article PubMed PMC PDF
16. Statistical guidance on reporting results from studies evaluating diagnostic tests: guidance for industry and FDA staff [Internet]. Silver Spring, MD: U.S. Food and Drug Administration; 2019. [cited 2019 Apr 28]. Available from: https://www.fda.gov/regulatory-information/search-fda-guidancedocuments/statistical-guidance-reporting-results-studiesevaluating-diagnostic-tests-guidance-industry-and-fda
17. Fitzgibbons PL, Bradley LA, Fatheree LA, Alsabeh R, Fulton RS, Goldsmith JD, et al. Principles of analytic validation of immunohistochemical assays: guideline from the College of American Pathologists Pathology and Laboratory Quality Center. Arch Pathol Lab Med. 2014;138:1432–43. Article PubMed
18. Munari E, Zamboni G, Marconi M, Sommaggio M, Brunelli M, Martignoni G, et al. PD-L1 expression heterogeneity in nonsmall cell lung cancer: evaluation of small biopsies reliability. Oncotarget. 2017;8:90123–31. Article PubMed PMC
19. Brunnstrom H, Johansson A, Westbom-Fremer S, Backman M, Djureinovic D, Patthey A, et al. PD-L1 immunohistochemistry in clinical diagnostics of lung cancer: inter-pathologist variability is higher than assay variability. Mod Pathol. 2017;30:1411–21. Article PubMed PDF
20. Hansen AR, Siu LL. PD-L1 Testing in cancer: challenges in companion diagnostic development. JAMA Oncol. 2016;2:15–6. Article PubMed
21. Buttner R, Gosney JR, Skov BG, Adam J, Motoi N, Bloom KJ, et al. Programmed death-ligand 1 immunohistochemistry testing: a review of analytical assays and clinical implementation in non-small-cell lung cancer. J Clin Oncol. 2017;35:3867–76. Article PubMed
22. Ratcliffe MJ, Sharpe A, Midha A, Barker C, Scott M, Scorer P, et al. Agreement between programmed cell death ligand-1 diagnostic assays across multiple protein expression cutoffs in non-small cell lung cancer. Clin Cancer Res. 2017;23:3585–91. Article PubMed
23. Chan AW, Tong JH, Kwan JS, Chow C, Chung LY, Chau SL, et al. Assessment of programmed cell death ligand-1 expression by 4 diagnostic assays and its clinicopathological correlation in a large cohort of surgical resected non-small cell lung carcinoma. Mod Pathol. 2018;31:1381–90. Article PubMed PDF
24. Munari E, Rossi G, Zamboni G, Lunardi G, Marconi M, Sommaggio M, et al. PD-L1 assays 22C3 and SP263 are not interchangeable in non-small cell lung cancer when considering clinically relevant cutoffs: an interclone evaluation by differently trained pathologists. Am J Surg Pathol. 2018;42:1384–9. Article PubMed
25. Scheel AH, Dietel M, Heukamp LC, Johrens K, Kirchner T, Reu S, et al. Harmonized PD-L1 immunohistochemistry for pulmonary squamous-cell and adenocarcinomas. Mod Pathol. 2016;29:1165–72. Article PubMed PDF
26. Kulangara K, Zhang N, Corigliano E, Guerrero L, Waldroup S, Jaiswal D, et al. Clinical utility of the combined positive score for programmed death ligand-1 expression and the approval of pembrolizumab for treatment of gastric cancer. Arch Pathol Lab Med. 2019;143:330–7. Article PubMed
27. Kitazono S, Fujiwara Y, Tsuta K, Utsumi H, Kanda S, Horinouchi H, et al. Reliability of small biopsy samples compared with resected specimens for the determination of programmed death-ligand 1 expression in non-small-cell lung cancer. Clin Lung Cancer. 2015;16:385–90. Article PubMed
28. Choi YY, Kim H, Shin SJ, Kim HY, Lee J, Yang HK, et al. Microsatellite instability and programmed cell death-ligand 1 expression in stage II/III gastric cancer: post hoc analysis of the CLASSIC randomized controlled study. Ann Surg. 2019;270:309–16. Article PubMed
29. Kawazoe A, Kuwata T, Kuboki Y, Shitara K, Nagatsuma AK, Aizawa M, et al. Clinicopathological features of programmed death ligand 1 expression with tumor-infiltrating lymphocyte, mismatch repair, and Epstein-Barr virus status in a large cohort of gastric cancer patients. Gastric Cancer. 2017;20:40715.Article PDF

Figure & Data

REFERENCES

Citations

Citations to this article as recorded by

PD-L1 as a Biomarker in Gastric Cancer Immunotherapy
Yunjoo Cho, Soomin Ahn, Kyoung-Mee Kim
Journal of Gastric Cancer.2025; 25(1): 177. CrossRef
CT radiomics-based intratumoral and intertumoral heterogeneity indicators for prognosis prediction in gastric cancer patients receiving neoadjuvant chemotherapy
Jiazheng Li, Zhenhui Li, Yinkui Wang, Yuzhuo Li, Jing Zhang, Ziyu Li, Lei Tang
European Radiology.2025;[Epub] CrossRef
Biomarkers in gastroesophageal cancer 2025: an updated consensus statement by the Spanish Society of Medical Oncology (SEOM) and the Spanish Society of Pathology (SEAP)
Maria Alsina Maqueda, Ana Teijo Quintáns, Miriam Cuatrecasas, Maria Jesús Fernández Aceñero, Ana Fernández Montes, Carlos Gómez Martín, Paula Jiménez Fonseca, Carolina Martínez Ciarpaglini, Fernando Rivera Herrero, Mar Iglesias Coma
Clinical and Translational Oncology.2025;[Epub] CrossRef
Neoadjuvant Immunotherapy in Gastroesophageal Cancer: A Promising Early Signal?
Mojun Zhu, Harry H. Yoon
Journal of Clinical Oncology.2024; 42(4): 373. CrossRef
Programmed death ligand-1 (PD-L1) clone 22C3 expression in resected colorectal cancer as companion diagnostics for immune checkpoint inhibitor therapy: A comparison study and inter-rater agreement evaluation across proposed cut-offs and predictive (TPS, C
Dordi Lea, Claudia Zaharia, Kjetil Søreide
Cancer Treatment and Research Communications.2024; 38: 100788. CrossRef
The relationship between high ratios of CD4/FOXP3 and CD8/CD163 and the improved survivability of metastatic triple-negative breast cancer patients: a multicenter cohort study
Jeffry Beta Tenggara, Andhika Rachman, Joedo Prihartono, Lisnawati Rachmadi, Sonar Soni Panigoro, Didik Setyo Heriyanto, Noorwati Sutandyo, Intan Russianna Nasution, Familia Bella Rahadiati, Ricci Steven, Rachelle Betsy, Samuel Juanputra, Aru Wisaksono Su
BMC Research Notes.2024;[Epub] CrossRef
Who Should Receive Immunotherapy for Advanced Gastroesophageal Cancer?
Suhaib Khateeb, Ludimila Cavalcante, Noor Alnairat, Meghana Singh, Ibrahim Halil Sahin, Azhar Saeed, Anwaar Saeed
Current Treatment Options in Oncology.2024; 25(4): 496. CrossRef
Evaluation of PD-L1 Expression in Colorectal Carcinomas by Comparing Scoring Methods and Their Significance in Relation to Clinicopathologic Parameters
Mirela Frančina, Mislav Mikuš, Marin Mamić, Tihomir Jovanović, Mario Ćorić, Božica Lovrić, Ivan Vukoja, Goran Zukanović, Kristijan Matković, Jasmina Rajc, Ferdinand Slišurić, Mateja Jurić-Marelja, Goran Augustin, Ilijan Tomaš
Diagnostics.2024; 14(10): 1007. CrossRef
The tumor immune composition of mismatch repair deficient and Epstein-Barr virus-positive gastric cancer: A systematic review
J. Bos, T.S. Groen-van Schooten, C.P. Brugman, F.S. Jamaludin, H.W.M. van Laarhoven, S. Derks
Cancer Treatment Reviews.2024; 127: 102737. CrossRef
Discordant PD-L1 results between 28–8 and 22C3 assays are associated with outcomes of gastric cancer patients treated with nivolumab plus chemotherapy
Hyung-Don Kim, Jinho Shin, In Hye Song, Jaewon Hyung, Hyungeun Lee, Min-Hee Ryu, Young Soo Park
Gastric Cancer.2024; 27(4): 819. CrossRef
Claudin-18 status and its correlation with HER2 and PD-L1 expression in gastric cancer with peritoneal dissemination
Haruki Ogawa, Hiroyuki Abe, Koichi Yagi, Yasuyuki Seto, Tetsuo Ushiku
Gastric Cancer.2024; 27(4): 802. CrossRef
Interpretation of PD-L1 expression in gastric cancer: summary of a consensus meeting of Korean gastrointestinal pathologists
Soomin Ahn, Yoonjin Kwak, Gui Young Kwon, Kyoung-Mee Kim, Moonsik Kim, Hyunki Kim, Young Soo Park, Hyeon Jeong Oh, Kyoungyul Lee, Sung Hak Lee, Hye Seung Lee
Journal of Pathology and Translational Medicine.2024; 58(3): 103. CrossRef
PD-L1 Immunohistochemistry in Gastric Cancer: Comparison of Combined Positive Score and Tumor Area Positivity Across 28-8, 22C3, and SP263 Assays
Samuel J. Klempner, Eric S. Cowden, Samuel L. Cytryn, Matteo Fassan, Hisato Kawakami, Hideaki Shimada, Laura H. Tang, Daniel-Christoph Wagner, Yasushi Yatabe, Alexander Savchenko, Jennifer Salcius, Dorhyun Johng, Jing Chen, Giuliana Montenegro, Markus Moe
JCO Precision Oncology.2024;[Epub] CrossRef
PD-L1 expression in squamous cervical carcinomas of Mozambican women living with or without HIV
Lucília Lovane, Satish Tulsidás, Carla Carrilho, Christina Karlsson
Scientific Reports.2024;[Epub] CrossRef
First-line immune checkpoint inhibitors in low programmed death-ligand 1-expressing population
Feiyang Zhang, Guoming Chen, Yixin Yin, Xiaojiang Chen, Runcong Nie, Yingbo Chen
Frontiers in Pharmacology.2024;[Epub] CrossRef
Concordance of PD-L1 status in primary gastroesophageal adenocarcinoma and matched peritoneal metastases: a single institution study
V. Massa, F. Signorini, F. Salani, M.E. Filice, G. Grelli, P. Lippolis, P. Faviana, V. Genovesi, S. Santi, C. Vivaldi, S. Cesario, A. Bertolucci, C. Cremolini, V. Nardini, G. Masi, C. Ugolini, L. Fornaro
ESMO Gastrointestinal Oncology.2024; 5: 100089. CrossRef
A Phase 1a/1b Dose Escalation/Expansion Study of the Anti-PD-1 Monoclonal Antibody Nofazinlimab in Chinese Patients with Solid Tumors or Lymphoma
Jifang Gong, Ye Guo, Yanqiao Zhang, Yi Ba, Tong Chen, Wei Li, Caicun Zhou, Mengzhao Wang, Haiyan Yang, Yuhong Zhou, Qiqing Cai, Ziping Wang, Gang Huang, Wei Zhang, Rila Su, Zhongheng Cai, Zenglian Yue, Jinzhou Dou, Peiqi Li, Rachel Wu, Archie N. Tse, Lin
Targeted Oncology.2024; 19(5): 723. CrossRef
Immune checkpoint inhibitors in advanced gastroesophageal adenocarcinoma: a series of patient-level meta-analyses in different programmed death-ligand 1 subgroups
A.G. Leone, A.S. Mai, K.Y. Fong, D.W.T. Yap, K. Kato, E. Smyth, M. Moehler, J.T.C. Seong, R. Sundar, J.J. Zhao, F. Pietrantonio
ESMO Open.2024; 9(11): 103962. CrossRef
HIPEC for Metastatic Gastric Cancer: Moving the Needle Towards 3-Year Survival
Neal Bhutiani, Y. David Seo, Kristen A. Robinson, Michael G. White, Naruhiko Ikoma, Paul F. Mansfield, Jenny J. Li, Mariela Blum Murphy, Jaffer A. Ajani, Brian D. Badgwell
European Journal of Surgical Oncology.2024; : 108790. CrossRef
Biomarkers for Predicting Response to Personalized Immunotherapy in Gastric Cancer
Moonsik Kim, Ji Yun Jeong, An Na Seo
Diagnostics.2023; 13(17): 2782. CrossRef
Development of an automated combined positive score prediction pipeline using artificial intelligence on multiplexed immunofluorescence images
Abhishek Vahadane, Shreya Sharma, Devraj Mandal, Madan Dabbeeru, Josephine Jakthong, Miguel Garcia-Guzman, Shantanu Majumdar, Chung-Wein Lee
Computers in Biology and Medicine.2023; 152: 106337. CrossRef
CT-based delta radiomics in predicting the prognosis of stage IV gastric cancer to immune checkpoint inhibitors
Jiazheng Li, Zifan Chen, Yang Chen, Jie Zhao, Meng He, Xiaoting Li, Li Zhang, Bin Dong, Xiaotian Zhang, Lei Tang, Lin Shen
Frontiers in Oncology.2023;[Epub] CrossRef
Efficacy of PD-1/PD-L1 inhibitors in gastric or gastro-oesophageal junction cancer based on clinical characteristics: a meta-analysis
Gengwei Huo, Wenjie Liu, Peng Chen
BMC Cancer.2023;[Epub] CrossRef
Clinical relevance of PD-1 positive CD8 T-cells in gastric cancer
Joan Choo, Ley Fang Kua, Mu Yar Soe, Bernadette Reyna Asuncion, Benjamin Kye Jyn Tan, Chong Boon Teo, Ryan Yong Kiat Tay, Jimmy So, Asim Shabbir, Kim Guowei, Hon Lyn Tan, Gloria Chan, Haoran Ma, Gokula Krishnan Ramachandran, Jeffrey H. Y. Lum, Cheng Ean C
Gastric Cancer.2023; 26(3): 393. CrossRef
Genetic and immune microenvironment characterization of HER2‐positive gastric cancer: Their association with response to trastuzumab‐based treatment
Hyun Jung Kwon, Yujun Park, Soo Kyung Nam, Enoch Kang, Ka‐Kyung Kim, Inhae Jeong, Yoonjin Kwak, Jeesun Yoon, Tae‐Yong Kim, Keun‐Wook Lee, Do‐Youn Oh, Seock‐Ah Im, Seong‐Ho Kong, Do Joong Park, Hyuk‐Joon Lee, Hyung‐Ho Kim, Han‐Kwang Yang, Hye Seung Lee
Cancer Medicine.2023; 12(9): 10371. CrossRef
High Interobserver Variability Among Pathologists Using Combined Positive Score to Evaluate PD-L1 Expression in Gastric, Gastroesophageal Junction, and Esophageal Adenocarcinoma
Marie E. Robert, Josef Rüschoff, Bharat Jasani, Rondell P. Graham, Sunil S. Badve, Manuel Rodriguez-Justo, Liudmila L. Kodach, Amitabh Srivastava, Hanlin L. Wang, Laura H. Tang, Giancarlo Troncone, Federico Rojo, Benjamin J. Van Treeck, James Pratt, Iryna
Modern Pathology.2023; 36(5): 100154. CrossRef
Heterogeneous distribution pattern of CD3+ tumor-infiltrated lymphocytes (TILs) and high combined positive score (CPS) favored the prognosis of resected early stage small-cell lung cancer
Liang Zhu, Guoping Cheng, Meijuan Wu, Ming Chen, Ying Jin
Translational Oncology.2023; 34: 101697. CrossRef
The Role of Immunotherapy in Esophageal and Gastric Cancer
Hans Dedecker, Laure-Anne Teuwen, Timon Vandamme, Andreas Domen, Hans Prenen
Clinical Colorectal Cancer.2023; 22(2): 175. CrossRef
Tissue- and liquid biopsy-based biomarkers for immunotherapy in breast cancer
Luca Licata, Marco Mariani, Federico Rossari, Giulia Viale, Giulia Notini, Matteo Maria Naldini, Carlo Bosi, Marta Piras, Matteo Dugo, Giampaolo Bianchini
The Breast.2023; 69: 330. CrossRef
PD‐L1andHLA‐class I expression status and their therapeutic implication in oesophageal small‐cell carcinoma
Satoshi Yamashita, Hiroyuki Abe, Hiroharu Yamashita, Koichi Yagi, Yasuyuki Seto, Tetsuo Ushiku
Histopathology.2023; 83(2): 264. CrossRef
Programmed death-ligand 1 expression and overall survival in Thai patients with gastric cancer
Taned Chitapanarux, Pawut Gumrai, Sarawut Kongkarnka, Komson Wannasai, Nirush Lertprasertsuke
Scientific Reports.2023;[Epub] CrossRef
Efficacy of Pd-1/Pd-L1 Inhibitors in Advanced Gastroesophageal Cancer Based on Characteristics: A Meta-Analysis
Wenjie Liu, Gengwei Huo, Peng Chen
Immunotherapy.2023; 15(10): 751. CrossRef
Immunology and immunotherapy in gastric cancer
Xiaqing Xu, Jiaxing Chen, Wenxing Li, Chenlu Feng, Qian Liu, Wenfang Gao, Meng He
Clinical and Experimental Medicine.2023; 23(7): 3189. CrossRef
Targeted Therapies and Developing Precision Medicine in Gastric Cancer
Rille Pihlak, Caroline Fong, Naureen Starling
Cancers.2023; 15(12): 3248. CrossRef
Programmed death ligand 1 (PD-L1) expression in gastric cancer: literature review
T. N. Sotnikova, N. V. Danilova, P. G. Malkov, T. V. Polushkina
Advances in Molecular Oncology.2023; 10(2): 70. CrossRef
Novel Biomarkers of Gastric Cancer: Current Research and Future Perspectives
Yasushi Sato, Koichi Okamoto, Yutaka Kawano, Akinari Kasai, Tomoyuki Kawaguchi, Tamotsu Sagawa, Masahiro Sogabe, Hiroshi Miyamoto, Tetsuji Takayama
Journal of Clinical Medicine.2023; 12(14): 4646. CrossRef
Programmed death-ligand 1 (PD-L1) expression in primary gastric adenocarcinoma and matched metastases
Drolaiz H. W. Liu, Heike I. Grabsch, Beat Gloor, Rupert Langer, Bastian Dislich
Journal of Cancer Research and Clinical Oncology.2023; 149(14): 13345. CrossRef
Chemoimmunotherapy for esophageal squamous cell carcinoma—Summary and discussion of recent clinical trials
Zhen Zhang, Jiaqian Huang, Yuhong Xu, Huiyan Luo
MedComm – Future Medicine.2023;[Epub] CrossRef
Next generation immuno-oncology biomarkers in gastrointestinal cancer: what does the future hold?
Hassan Abushukair, Obada Ababneh, Ayah Al-Bzour, Ibrahim Halil Sahin, Anwaar Saeed
Expert Review of Molecular Diagnostics.2023; 23(10): 863. CrossRef
Impact of programmed death‐ligand 1 (PD‐L1) positivity on clinical and molecular features of patients with metastatic gastric cancer
Minkyue Shin, Soomin Ahn, Jaeyun Jung, Sujin Hyung, Kyoung‐Mee Kim, Seung Tae Kim, Won Ki Kang, Jeeyun Lee
Cancer Medicine.2023; 12(18): 18633. CrossRef
Regulatory T cells in gastric cancer: Key controllers from pathogenesis to therapy
Ion Negura, Mariana Pavel-Tanasa, Mihai Danciu
Cancer Treatment Reviews.2023; 120: 102629. CrossRef
The prognostic value of a combined immune score in tumor and immune cells assessed by immunohistochemistry in triple-negative breast cancer
Ji Eun Choi, Jae Seok Lee, Min-Sun Jin, Ilias P. Nikas, Kwangsoo Kim, Sunah Yang, Soo Young Park, Jiwon Koh, Sohyeon Yang, Seock-Ah Im, Han Suk Ryu
Breast Cancer Research.2023;[Epub] CrossRef
Comparison of SP142 and 22C3 Immunohistochemistry PD-L1 Assays for Clinical Efficacy of Atezolizumab in Non–Small Cell Lung Cancer: Results From the Randomized OAK Trial
Shirish Gadgeel, Fred R. Hirsch, Keith Kerr, Fabrice Barlesi, Keunchil Park, Achim Rittmeyer, Wei Zou, Namrata Bhatia, Hartmut Koeppen, Sarah M. Paul, David Shames, Jing Yi, Christina Matheny, Marcus Ballinger, Mark McCleland, David R. Gandara
Clinical Lung Cancer.2022; 23(1): 21. CrossRef
Adrenal Nodules Detected at Staging CT in Patients with Resectable Gastric Cancers Have a Low Incidence of Malignancy
Hae Young Kim, Won Chang, Yoon Jin Lee, Ji Hoon Park, Jungheum Cho, Hee Young Na, Hyungwoo Ahn, Sung Il Hwang, Hak Jong Lee, Young Hoon Kim, Kyoung Ho Lee
Radiology.2022; 302(1): 129. CrossRef
Immunohistochemistry as predictive and prognostic markers for gastrointestinal malignancies
Matthew W. Rosenbaum, Raul S. Gonzalez
Seminars in Diagnostic Pathology.2022; 39(1): 48. CrossRef
Histopathology of Gastrointestinal Immune-related Adverse Events
M. Lisa Zhang, Vikram Deshpande
American Journal of Surgical Pathology.2022; 46(1): e15. CrossRef
Safety and efficacy of HER2 blockade by trastuzumab-based chemotherapy-containing combination strategies in HER2+ gastroesophageal adenocarcinoma
D.V.T. Catenacci, H.C. Chung, L. Shen, M. Moehler, H.H. Yoon, M.K. Rosales, Y.-K. Kang
ESMO Open.2022; 7(1): 100360. CrossRef
PD-L1 and HER2 expression in gastric adenocarcinoma and their prognostic significance
Jie Lian, Guanjun Zhang, Yun Zhang, Heng Liu, Jiaojiao Zhang, Pengfei Nan, Wei Tian
Digestive and Liver Disease.2022; 54(10): 1419. CrossRef
Tissue miR-200c-3p and circulating miR-1290 as potential prognostic biomarkers for colorectal cancer
Enoch Kang, Sung Cheol Jung, Soo Kyung Nam, Yujun Park, Soo Hyun Seo, Kyoung Un Park, Heung-Kwon Oh, Duck-Woo Kim, Sung-Bum Kang, Hye Seung Lee
Scientific Reports.2022;[Epub] CrossRef
Gastric Cancer: Mechanisms, Biomarkers, and Therapeutic Approaches
Sangjoon Choi, Sujin Park, Hyunjin Kim, So Young Kang, Soomin Ahn, Kyoung-Mee Kim
Biomedicines.2022; 10(3): 543. CrossRef
Immunotherapy use in oesophagogastric cancers—a review of the literature
Annabel Smith, Amitesh Roy, Christos S. Karapetis, Vy Broadbridge, Timothy Price
British Journal of Cancer.2022; 127(1): 21. CrossRef
Interspatial Distribution of Tumor and Immune Cells in Correlation with PD-L1 in Molecular Subtypes of Gastric Cancers
Bastian Dislich, Kirsten D. Mertz, Beat Gloor, Rupert Langer
Cancers.2022; 14(7): 1736. CrossRef
Immunotherapy in Gastro-Oesophageal Cancer: Current Practice and the Future of Personalised Therapy
Mary E. Booth, Elizabeth C. Smyth
BioDrugs.2022; 36(4): 473. CrossRef
Anlotinib Combined with Toripalimab as Second-Line Therapy for Advanced, Relapsed Gastric or Gastroesophageal Junction Carcinoma
Man Jiang, Chuantao Zhang, Yabin Hu, Tianjun Li, Guangjie Yang, Guanqun Wang, Jingjuan Zhu, Changfeng Shao, Helei Hou, Na Zhou, Kewei Liu, Xiaochun Zhang
The Oncologist.2022; 27(11): e856. CrossRef
Abdominal Computed Tomography Enhanced Image Features under an Automatic Segmentation Algorithm in Identification of Gastric Cancer and Gastric Lymphoma
Lihua Zhou, Hao Hu, Lei Zhou, Yi Zhou, Ahmed Faeq Hussein
Computational and Mathematical Methods in Medicine.2022; 2022: 1. CrossRef
Margetuximab with retifanlimab as first-line therapy in HER2+/PD-L1+ unresectable or metastatic gastroesophageal adenocarcinoma: MAHOGANY cohort A
D.V.T. Catenacci, Y.-K. Kang, H.H. Yoon, B.Y. Shim, S.T. Kim, D.-Y. Oh, A.I. Spira, S.V. Ulahannan, E.J. Avery, P.M. Boland, J. Chao, H.C. Chung, F. Gardner, S.J. Klempner, K.-W. Lee, S.C. Oh, J. Peguero, M.B. Sonbol, L. Shen, M. Moehler, J. Sun, D. Li, M
ESMO Open.2022; 7(5): 100563. CrossRef
Advances in molecular biomarkers research and clinical application progress for gastric cancer immunotherapy
Hongzhen Cai, Man Li, Ruiyi Deng, Mopei Wang, Yanyan Shi
Biomarker Research.2022;[Epub] CrossRef
Immune Checkpoint Inhibitors for Gastrointestinal Malignancies: An Update
Kathryn DeCarli, Jonathan Strosberg, Khaldoun Almhanna
Cancers.2022; 14(17): 4201. CrossRef
Immunotherapy for Gastroesophageal Tumors: Is there still Hope for Efficacy?
Hannah Christina Puhr, Aysegul Ilhan-Mutlu
Current Cancer Drug Targets.2022; 22(8): 651. CrossRef
The seen and the unseen: Molecular classification and image based-analysis of gastrointestinal cancers
Corina-Elena Minciuna, Mihai Tanase, Teodora Ecaterina Manuc, Stefan Tudor, Vlad Herlea, Mihnea P. Dragomir, George A. Calin, Catalin Vasilescu
Computational and Structural Biotechnology Journal.2022; 20: 5065. CrossRef
Ornithine aminotransferase and carbamoyl phosphate synthetase 1 involved in ammonia metabolism serve as novel targets for early stages of gastric cancer
Zhen Jiang, Chen Wei, Yaomin Luo, Yang Xiao, Li Wang, Wubin Guo, Xiaoxia Yuan
Journal of Clinical Laboratory Analysis.2022;[Epub] CrossRef
Indirect Clinical Validation of a Programmed Death-Ligand 1 Laboratory-Developed Test for Gastric/Gastroesophageal Junction Adenocarcinoma with 22C3 Antibody Concentrate
Ji Min Kim, Binnari Kim, Eunji Kim, Minsun Jang, Jun Hun Cho, Hye Seung Lee, Yoonjin Kwak, Lingkang Huang, Radha Krishnan, Sally Y. Bai, Mounia Mounawar, Kyoung-Mee Kim
Molecular Diagnosis & Therapy.2022; 26(6): 679. CrossRef
High interobserver and intraobserver reproducibility among pathologists assessing PD‐L1 CPS across multiple indications
Shanthy Nuti, Yiwei Zhang, Nabila Zerrouki, Charlotte Roach, Gudrun Bänfer, George L Kumar, Edward Manna, Rolf Diezko, Kristopher Kersch, Josef Rüschoff, Bharat Jasani
Histopathology.2022; 81(6): 732. CrossRef
Validation of E1L3N antibody for PD-L1 detection and prediction of pembrolizumab response in non-small-cell lung cancer
Lianxi Song, Liang Zeng, Huan Yan, Qinqin Xu, Qing Xia, Jian Lei, Xiaoyan Chen, Xiaoping Hu, Zhan Wang, Hong Liu, Nong Yang, Yongchang Zhang
Communications Medicine.2022;[Epub] CrossRef
Loss of SATB2 expression correlates with cytokeratin 7 and PD-L1 tumor cell positivity and aggressiveness in colorectal cancer
Jan Hrudka, Radoslav Matěj, Andrej Nikov, Igor Tomyak, Hana Fišerová, Karolína Jelínková, Petr Waldauf
Scientific Reports.2022;[Epub] CrossRef
Programmed cell death protein 1/programmed death ligand 1 but not HER2 is a potential therapeutic target in gastric neuroendocrine carcinoma
Satoshi Yamashita, Hiroyuki Abe, Akiko Kunita, Hiroharu Yamashita, Yasuyuki Seto, Tetsuo Ushiku
Histopathology.2021; 78(3): 381. CrossRef
MAHOGANY: Margetuximab Combination in HER2+ Unresectable/metastatic Gastric/gastroesophageal Junction Adenocarcinoma
Daniel V T Catenacci, Minori Rosales, Hyun Cheol Chung, Harry H Yoon, Lin Shen, Markus Moehler, Yoon-Koo Kang
Future Oncology.2021; 17(10): 1155. CrossRef
PD-L1 expression in paired biopsies and surgical specimens in gastric adenocarcinoma: A digital image analysis study
You Jeong Heo, Binnari Kim, Hyunjin Kim, Soi Kim, Min Sun Jang, Kyoung-Mee Kim
Pathology - Research and Practice.2021; 218: 153338. CrossRef
Targeting Oncoimmune Drivers of Cancer Metastasis
Chie Kudo-Saito, Yukinori Ozaki, Hiroshi Imazeki, Hideyuki Hayashi, Jun Masuda, Hiroki Ozawa, Yamato Ogiwara
Cancers.2021; 13(3): 554. CrossRef
PD-L1 as a biomarker of response to immune-checkpoint inhibitors
Deborah Blythe Doroshow, Sheena Bhalla, Mary Beth Beasley, Lynette M. Sholl, Keith M. Kerr, Sacha Gnjatic, Ignacio I. Wistuba, David L. Rimm, Ming Sound Tsao, Fred R. Hirsch
Nature Reviews Clinical Oncology.2021; 18(6): 345. CrossRef
Cancer Immunotherapy Update: FDA-Approved Checkpoint Inhibitors and Companion Diagnostics
Julianne D. Twomey, Baolin Zhang
The AAPS Journal.2021;[Epub] CrossRef
High Expression of PD-L1 Is Associated with Better Survival in Pancreatic/Periampullary Cancers and Correlates with Epithelial to Mesenchymal Transition
Nishant Thakur, Kwang Yeol Paik, Gyoyeon Hwang, Yosep Chong
Diagnostics.2021; 11(4): 597. CrossRef
Prädiktive Diagnostik für Checkpoint-Inhibitoren
Hans-Ulrich Schildhaus, Wilko Weichert
Der Pathologe.2021; 42(4): 380. CrossRef
PD-1/PD-L1 in Cancer: Pathophysiological, Diagnostic and Therapeutic Aspects
Enrico Munari, Francesca R. Mariotti, Linda Quatrini, Pietro Bertoglio, Nicola Tumino, Paola Vacca, Albino Eccher, Francesco Ciompi, Matteo Brunelli, Guido Martignoni, Giuseppe Bogina, Lorenzo Moretta
International Journal of Molecular Sciences.2021; 22(10): 5123. CrossRef
PD-L1 expression in gastric cancer: interchangeability of 22C3 and 28-8 pharmDx assays for responses to immunotherapy
Soomin Ahn, Kyoung-Mee Kim
Modern Pathology.2021; 34(9): 1719. CrossRef
PD-L1 Expression Harmonization in Gastric Cancer Using 22C3 PharmDx and SP263 Assays
Tamara Z. Dabbagh, Maher A. Sughayer
Applied Immunohistochemistry & Molecular Morphology.2021; 29(6): 462. CrossRef
Expression of the immune checkpoint receptors PD-1, LAG3, and TIM3 in the immune context of stage II and III gastric cancer by using single and chromogenic multiplex immunohistochemistry
Yujun Park, An Na Seo, Jiwon Koh, Soo Kyoung Nam, Yoonjin Kwak, Sang-Hoon Ahn, Do Joong Park, Hyung-Ho Kim, Hye Seung Lee
OncoImmunology.2021;[Epub] CrossRef
Appropriate PD-L1 Cutoff Value for Gastric Cancer Immunotherapy: A Systematic Review and Meta-Analysis
Tong Xie, Zhening Zhang, Xiaotian Zhang, Changsong Qi, Lin Shen, Zhi Peng
Frontiers in Oncology.2021;[Epub] CrossRef
Prediction of TP53 mutations by p53 immunohistochemistry and their prognostic significance in gastric cancer
Hye Jung Hwang, Soo Kyung Nam, Hyunjin Park, Yujun Park, Jiwon Koh, Hee Young Na, Yoonjin Kwak, Woo Ho Kim, Hye Seung Lee
Journal of Pathology and Translational Medicine.2020; 54(5): 378. CrossRef

PubReader
ePub Link

Cite

CITE: export Copy Download Format; Close

Download Citation

Download a citation file in RIS format that can be imported by all major citation management software, including EndNote, ProCite, RefWorks, and Reference Manager.

Format:

RIS — For EndNote, ProCite, RefWorks, and most other reference management software
BibTeX — For JabRef, BibDesk, and other BibTeX-specific software

Include:

Citation for the content below
Citation and abstract for the content below

PD-L1 Testing in Gastric Cancer by the Combined Positive Score of the 22C3 PharmDx and SP263 Assay with Clinically Relevant Cut-offs

Cancer Res Treat. 2020;52(3):661-670. Published online January 10, 2020

DOI: https://doi.org/10.4143/crt.2019.718

XML Download

Abstract
Introduction
Materials and Methods
Results
Discussion
Electronic Supplementary Material
NOTES
REFERENCES

PD-L1 Testing in Gastric Cancer by the Combined Positive Score of the 22C3 PharmDx and SP263 Assay with Clinically Relevant Cut-offs

Fig. 1. Representative tumor microarray core stained with the 22C3 pharmDx and SP 263 assay according to the combined positive score (CPS) distribution (core magnification, ×50; inlet, ×200).

Fig. 2. Correlation between the 22C3 pharmDx and SP263 assay for the combined positive score (CPS) (A) and tumor proportion score (TPS) (B) at the center of the tumor and invasive margin in gastric cancer samples.

Fig. 3. Programmed death ligand 1 expression distribution of the 22C3 pharmDx and SP263 assay for the combined positive score (CPS) (A, B) and tumor proportion score (TPS) (C, D) at the center of the tumor (CT) and invasive margin (IM) by intervals.

Fig. 1.

Fig. 2.

Fig. 3.

PD-L1 Testing in Gastric Cancer by the Combined Positive Score of the 22C3 PharmDx and SP263 Assay with Clinically Relevant Cut-offs

	22C3 (n=379)				SP263 (n=379)
	CT		IM		CT		IM
	Positive	Negative	Positive	Negative	Positive	Negative	Positive	Negative
CPS ≥ 1	219 (57.8)	160 (42.2)	170 (44.9)	209 (55.1)	231 (60.9)	148 (39.1)	175 (46.2)	204 (53.8)
CPS ≥ 10	37 (9.8)	342 (90.2)	50 (13.2)	329 (86.8)	36 (9.5)	343 (90.5)	49 (12.9)	330 (87.1)
TPS ≥ 1%	118 (31.1)	261 (68.9)	104 (27.4)	275 (72.6)	125 (33.0)	254 (67.0)	115 (30.3)	264 (69.7)
TPS ≥ 10%	28 (7.4)	351 (92.6)	30 (7.9)	349 (92.1)	29 (7.7)	350 (92.3)	34 (9.0)	345 (91.0)

	22C3		SP263
	Agreement (%)	Cohen's kappa (lower 95% CI)	Agreement (%)	Cohen's kappa (lower 95% CI)
CPS ≥ 1	77.6	0.559 (47.8)	77.8	0.564 (48.5)
CPS ≥ 10	92.9	0.650 (52.9)	91.8	0.590 (46.0)
TPS ≥ 1%	84.7	0.631 (54.6)	85.8	0.671 (59.0)
TPS ≥ 10%	94.7	0.623 (47.5)	94.5	0.637 (49.3)

	OPA (lower 95% CI, %)		PPA (lower 95% CI, %)		NPA (lower 95% CI, %)
	CT	IM	CT	IM	CT	IM
CPS ≥ 1	94.7 (92.0)	94.5 (91.7)	98.2 (95.4)	95.3 (90.9)	90.0 (84.3)	93.8 (89.6)
CPS ≥ 10	99.2 (97.7)	98.7 (97.0)	94.6 (81.8)	94.0 (83.5)	99.7 (98.4)	99.4 (97.8)
TPS ≥ 1%	97.1 (94.9)	96.6 (94.2)	98.3 (94.0)	99.0 (94.8)	96.6 (93.6)	95.6 (92.5)
TPS ≥ 10%	99.7 (98.5)	98.4 (96.6)	100 (87.7)	96.7 (82.8)	99.7 (98.4)	98.6 (96.7)

	Intraclass correlation coefficient (lower 95% CI)		Fleiss' Kappa (lower 95% CI)
	CPS	TPS	CPS ≥ 1	CPS ≥ 10
22C3	0.387 (20.9)	0.596 (40.5)	0.389 (26.4)	0.224 (8.0)
SP263	0.349 (13.5)	0.710 (57.2)	0.256 (15.6)	0.140 (2.4)

	OPA (lower 95% CI, %)	PPA (lower 95% CI, %)	NPA (lower 95% CI, %)	Cohen's kappa (lower 95% CI)
CPS ≥ 1
A	90.9 (80.1)	100 (91.0)	68.8 (41.3)	0.757 (55.4)
B	87.3 (75.5)	100 (92.5)	12.5 (0.3)	0.196 (–14.5)
C	98.2 (90.3)	100 (93.4)	0	-^a)
D	76.4 (63.0)	96.7 (82.8)	52.0 (31.3)	0.505 (28.5)
E	98.2 (90.3)	100 (89.4)	95.5 (77.2)	0.962 (88.5)
CPS ≥ 10
A	90.9 (80.1)	100 (86.8)	82.8 (64.2)	0.819 (66.6)
B	87.3 (75.5)	100 (92.5)	12.5 (0.3)	0.196 (–14.5)
C	80.0 (67.0)	88.6 (73.3)	65.0 (40.8)	0.554 (31.5)
D	90.9 (80.1)	88.9 (51.8)	91.3 (79.2)	0.707 (46.1)
E	100 (93.5)	100 (66.4)	100 (92.3)	100 (100)

Table 1. The number of PD-L1–positive gastric cancer cases

Values are presented as number (%). PD-L1, programmed death ligand 1; CT, center of tumor; IM, invasive margin; CPS, combined positive score; TPS, tumor proportion score.

Table 2. Agreement and Cohen’s kappa values between center of tumor and invasive margin

CI, confidence interval; CPS, combined positive score; TPS, tumor proportion score.

Table 3. OPA, PPA, and NPA between 22C3 pharmDx and SP263 assay

Table 4. Interobserver variation between five pathologists by 22C3 pharmDx and SP263 assay

CI, confidence interval; CPS, combined positive score; TPS, tumor proportion score.

Table 5. OPA, PPA, NPA, and Cohen’s kappa values between 22C3 pharmDx and SP263 assay (55 samples)

OPA, overall percent agreement; PPA, positive percent agreement; NPA, negative percent agreement; CI, confidence interval; CPS, combined positive score.