Indian Journal of Pathology and Microbiology

: 2022  |  Volume : 65  |  Issue : 4  |  Page : 839--843

PD-L1 expression in muscle invasive urothelial carcinoma: Comparison of SP142 and SP263 assay

Gurudutt Gupta1, Sunil Pasricha1, Meenakshi Kamboj1, Anila Sharma1, NS Nayana1, Garima Durga1, Anurag Sharma2, Sudhir Rawal3, Anurag Meh1,  
1 Department of Pathology, Rajiv Gandhi Cancer Institute and Research Centre, Rohini, New Delhi, India
2 Department of Biostatistics, Rajiv Gandhi Cancer Institute and Research Centre, Rohini, New Delhi, India
3 Uro-Surgery, Rajiv Gandhi Cancer Institute and Research Centre, Rohini, New Delhi, India

Correspondence Address:
Meenakshi Kamboj
Department of Pathology, Rajiv Gandhi Cancer Institute and Research Centre, Sector 5, Rohini, New Delhi - 110 085


Introduction: High-grade urothelial carcinoma has a different molecular pathway than superficial low grade urothelial carcinoma, and is characterized by genomic instability. The high tumor mutation burden leads to neoantigen formation, evoking an immune response. The immune response has been keenly studied in last two decades and programmed death ligand-1 (PDL-1) has emerged as acceptable immunohistochemical marker for assessment of response to therapy, prognostication and patient selection for immunotherapy. The targeting of PD-1 and PDL-1 by checkpoint inhibitors (CPIs) is an attractive strategy to unblock the inhibitor and induce cytotoxic cell death. However, the presence of complementary and companion diagnostic testing with multiple PDL-1 assays and platforms for various CPIs make a diagnostic quagmire. Thus, it is the need of hour to harmonize these assays. In this undertaken study we evaluated the concordance in PD-L1 expression between the two PD-L1 clones: SP263 and SP142, in treatment naïve muscle invasive bladder cancer (MIBC). Methods: We evaluated Ventana PD-L1 “SP263 and SP142” qualitative immunohistochemical assay using rabbit monoclonal anti-PD-L1 clones in evaluation of PDL-1 immunoexpression on Ventana autostainer platform. The study includes 30 muscle invasive urothelial carcinomas, with 10 of 30 having nodal metastasis. Results: SP263 assay was statistically more sensitive than SP142 for tumor cell (TC) scoring (P = 0.0009), whereas SP142 was more sensitive for immune cell (IC) scoring (P = 0.0067). There was no statistical significant discordance for TC or IC scoring between primary tumor and metastatic lymph node. Conclusion: PD-L1 testing status can be done on both primary tumor and metastatic site, however in metachronous metastatic setting, testing on recent metastatic site should be preferred. The harmonization of immunoexpression between 2 PD-L1 clones could not be achieved.

How to cite this article:
Gupta G, Pasricha S, Kamboj M, Sharma A, Nayana N S, Durga G, Sharma A, Rawal S, Meh A. PD-L1 expression in muscle invasive urothelial carcinoma: Comparison of SP142 and SP263 assay.Indian J Pathol Microbiol 2022;65:839-843

How to cite this URL:
Gupta G, Pasricha S, Kamboj M, Sharma A, Nayana N S, Durga G, Sharma A, Rawal S, Meh A. PD-L1 expression in muscle invasive urothelial carcinoma: Comparison of SP142 and SP263 assay. Indian J Pathol Microbiol [serial online] 2022 [cited 2023 Feb 3 ];65:839-843
Available from:

Full Text


Invasive urothelial carcinoma poses a significant cause of morbidity and mortality, and has been treated through decades with surgery, chemotherapy and radiation, depending upon the stage of disease. The best outcomes are seen in patients amenable to surgical resection upfront. The checkpoint inhibitors (CPIs) offer a promising therapeutic option for managing advanced stage urothelial carcinoma requiring chemotherapy in cisplatin ineligible/chemoresistant patients.

The advent of CPI has led to laboratory testing of predictive markers like programmed death ligand-1 (PD-L1), mismatch repair (MMR), microsatellite instability (MSI), tumor mutation burden (TMB), polymerase-D and E mutations, cancer neoantigens, and tumor-infiltrating lymphocytes (TILs). The emerging biomarkers like transcriptional signatures of immune responsiveness, cancer immunotherapy resistance biomarkers, and the microbiome are still in research.[1] The precise role and indications of CPI in urothelial carcinoma is still evolving, and there is a persuasive evidence of utility of PD-L1 testing in complementary and companion settings for predicting the response to CPI in metastatic or chemo-resistant/chemo-ineligible cases.[2],[3]

During last few years, CPI like atezolizumab, pembrolizumab, nivolumab, avelumab and durvalumab have been approved after trials by FDA for metastatic/non-resectable urothelial carcinoma.[2],[3],[4],[5],[6] For cisplatin ineligible advanced urothelial carcinoma, PD-L1 testing with companion diagnostic tests: SP142 (Roche) and 22C3 (Dako) for atezolizumab and pembrolizumab, respectively, has been FDA approved for first line monotherapy, whereas SP263 assay has been approved in complementary setting.[7],[8],[9]

The comparison of different assays for urothelial carcinoma has been studied by various authors, however, with conflicting results for tumor cell (TC) staining versus immune cell (IC) staining.[3],[7],[10],[11]

The temporal and spatial heterogeneity of PD-L1 immunoexpression has been well described in the literature. The heterogeneity may manifest as intratumoral (within the tumor) or intertumoral {between primary and metastatic site (regional lymphnode or distant site)}. However, few studies have analyzed the heterogeneity in PD-L1 expression between primary and metastatic deposits.[12],[13],[14] Moreover, the TCs are believed to stain more uniformly at various sites, in contrast to the ICs.[2] The concordance between various clones/IHC platforms is also a limiting factor for PD-L1 testing for a harmonious standardized testing. With various clones and autostainer platforms as companion/complementary assay for respective malignancy and CPI, the testing protocol has become very customized and tedious. Hence, exchangeability of diagnostic PD-L1 antibody clones, algorithms and assays are need of the time and are being explored by many authors.[3],[10],[11],[15] The harmonization study among various PD-L1 clones and their respective assay has been done in blueprint studies[16],[17] for PD-L1 expression in lung cancer.

This undertaken study attempts to look for concordance in PD-L1 expression between the two popular anti-PD-L1 rabbit monoclonal clones, SP263 and SP142, both from Ventana, in treatment naïve muscle invasive bladder cancer (MIBC). We also evaluated the concordance of PD-L1 immunoexpression between the primary tumor (bladder) and metastatic nodal disease (paired cystectomy and lymph node dissection).

 Material and Methods

In this study, we evaluated and compared the two FDA approved clones of PD-L1 in terms of immunostaining on Ventana Benchmark platform. In the calendar year 2017, 250 cases of urothelial carcinoma presented to urology OPD. Only treatment naïve MIBC cases which underwent radical cystectomy/cystoprostatectomy, with or without nodal disease were included. Forty six cases of MIBC radical cystectomy cases based on this criterion were identified. Of these, 30 cases fulfilled the inclusion criterion of having at least one tissue section with muscle invasive tumor, adequately preserved tumor tissue (30% viable tumor with stroma) and no history of neo-adjuvant chemotherapy. The inclusion criterion ensured adequate, well preserved tumor area with stromal interface to evaluate ICs as well as TCs. Whole tissue block sections from selected cases were used for PDL-1 IHC for both clones, to take care of staining heterogeneity in TC & IC components. The protocol was as per the product insert, validated in our laboratory. The cases were scored only after the assurance of internal positive and negative on-slide controls. Ten out of 30 cases were metastatic to lymphnodes (regional nodes), and were scored for expression in both primary tumor and metastatic deposit in lymphnodes.

The TC and IC scoring was done according to the recommendations defined for urothelial carcinoma and for the clone used. The significant cut off scoring for both TC and IC has been mentioned in [Table 1].[8],[9] For all the cases, immunoscoring and results were recorded by three histopathologists (GG, AS and SP), and in case of any discrepant opinion a consensus result was obtained.{Table 1}

Statistical methods

The Chi-square test was used to determine the association between SP263 and SP142 in TC scoring and IC scoring. A 2×2 contingency table was used to calculate overall percent agreement (OPA), positive percent agreement (PPA), and negative percent agreement (NPA), for pair wise comparison between SP263 and SP142 assay. The values were calculated at SP142 and SP263 assay cutoffs.

Marginal Homogeneity test was used to test the significance of difference in TC and IC in primary tumor and lymph nodes with synchronous metastasis in 10 patients. All data entries and statistical analyses were performed by using SPSS® Version 23.0 software. All these statistical results are accompanied by 95% confidence intervals (CI). All the reported P values are two-sided, and P < 0.05 are considered to indicate statistical significance.

Ethical considerations

The study has been approved by a suitably constituted Ethics Committee of the institution within which the work was undertaken and it conforms to the provisions of the Declaration of Helsinki (as revised in Fortaleza, Brazil, October 2013).


In our analysis, for TC score, SP263 clone was statistically found to be more sensitive than SP142 clone [Table 2]a for detecting cases to be eligible for CPI (P = 0.0009). For IC scoring, there was significant difference in the detection of negative and positive cases. When scoring the primary tumor site only on the basis of IC at SP142 label cut off, 3 cases that were positive for SP142, were negative by SP263, and two cases were vice versa [Table 2]b. Hence, SP142 is more sensitive for detection of positive ICs (P = 0.0067). However, for lymph node IC scoring, the discordance was not statistically significant [Table 2]c between SP142 and SP263 (P = 0.133).{Table 2}

The scoring for both primary and synchronous metastasis (regional nodal metastasis in our cases) at SP263 and SP142 cutoff are shown in [Table 3]a. At SP263 label cutoff 45% & 40% of cases are positive for SP263 and SP142, respectively. In comparison, at SP142 label cutoff, only 17.5% & 27.5% of cases are positive. The OPA, PPA and NPA has been mentioned in [Table 3]a, [Table 3]b. The statistical analysis highlights the OPA of 77.5%, 85%and 90% for label cutoff, SP142 cut off and SP263 cut off, respectively. Thus the poor OPA of 77.5% at label cut off highlights the poor concordance between the two assays. The PPA was good (93.75%) for SP263 label cut off, whereas poor (54.5%) for SP142 label cutoff. The NPA, however, was best for SP142 label cutoff (96.55%) and intermediate for SP263 label cutoff at 87.5% [Table 3]a, [Table 3]b.{Table 3}

Marginal homogeneity showed that there was no statistically significant difference in PD-L1 TC score between primary tumor and metastatic lymph node by SP263 clone. Moreover there was no statistically significant difference in the PD-L1 [Table 4].{Table 4}

Our study is limited by small sample size in synchronous metastatic cases and no metachronous metastatic cases were included.


The PD-L1 status in case of urothelial carcinoma is assessed on TCs and/or ICs depending on the clone used for predicting response to intended CPI drug. IC scoring is done on lymphocytes, neutrophils, plasma cells and histiocytes.[7],[8],[9] The use of complementary or companion diagnostic test is desirable in assigning the PD-L1 status for particular drug. The use of multiple clones and platforms add to the confusion and dilemma for both pathologist and the oncologist. Moreover, for an individual laboratory, it is not feasible to have multiple IHC platforms, hence harmonization of these assays is practical need of the hour and this study attempts to answer some of these issues.

In our study, we observed discordant PD-L1 scoring in 11 of 40 (27.5%) cases (paired testing), at the SP 263 label cutoff [Table 3]a. These discordant results in our study [as shown in [Table 2] and [Table 3] are due to different scoring cutoffs used for SP142 and SP263. These findings were supported by many studies; one by Eckstein et al.[7] showed high discordance of 42% between patients which were positive for at least one algorithm (SP142 and SP22C3). The best OPA and PPA in our cases was seen with SP263 cut off; this can be explained by a cutoff that translates at > or equal to 1% of IC+ tumor area or TC score >25%. The best OPA was seen with SP263 label cutoff; this is due to the fact that, cases which fall short of threshold of IC+ score bySP263 (as SP142 has better sensitivity for staining IC), was compensated by increased TC scoring by SP263 clone. The PPA was worst with SP142 label cut off, again due to similar reason that IC+ staining is seen more with SP142 than SP263 (IC staining scores ≥5%, are less with SP263 in comparison for SP142). The NPA is best when SP142 cutoff is used as explained above. However, for other two cutoffs, it is suboptimal.

Other studies over years[10],[11],[15],[18] have evaluated harmonization among the various PDL-1 clones. Hodgson et al.[15] compared four clones SP142, SP263, 22C3, E1L3N and showed substantial agreement between all antibody clones (κ0.639-0.791). They showed excellent reliability among SP263, 22C3 and E1L3N antibodies with interclass correlation coefficient (ICC) of 0.929–0.949 for TC staining; whereas with SP142 staining was lower for TC, with moderate correlation (ICC 0.500-0.619). Reliability of IC staining was lower compared to TC (ICC 0.519–0.866). Zajac et al.[10] studied 335 cases of urothelial carcinoma (biopsies and radical resections) and compared SP142, SP263, 22C3 and 28-8 for analytical staining and classification concordance. They also showed good correlation among SP263, 22C3 and 28-8 for TC and IC, with spearman rank coefficient of 0.92–0.93 for TCs and 0.88–0.91 for ICs. SP142 was an outlier and showed significant difference with other assays for TC, whereas IC staining was similar. Hence, they showed interchangeability between three clones, but not for SP142.

Rijnder et al.[11] compared five assays, SP142, SP263, 22C3, E1L3N and 28-8, in muscle-invasive UC, for TC and IC scoring. They showed better concordance for companion/complementary diagnostics (SP142, SP263, 22C3, and 28-8), whereas research antibody E1L3N was an outlier. The pairwise agreement of PDL-1 staining in their study varied from 72 to 90% (κ 0.255–0.708). In their study also, PD-L1 expression by SP142 was less on TCs.

The SP142 stains IC more than TCs; however, SP263 outperforms SP142 for TC and IC scoring. SP142 was found to stain more IC in some studies, whereas some showed SP263 to be equivalent or superior. For TC staining, SP263 shows more sensitivity with higher percentage of TC positivity. The reason mentioned are two folds, though both SP142 and SP263 are raised against the cytoplasmic portion (intracellular domain of PDL-1 isoform-1), and hence stain IC better with more cytoplasmic staining. SP263 is said to be unaffected by changes in the amino acid sequence in PDL-1 isoforms and hence is more sensitive for TC staining.[2] The better and intense staining of IC by SP142 is also explained by cross reactivity of SP142 antibody with PD-L1 extra cytoplasmic domain.[3] The SP142 assay has been developed to score IC more easily by pathologist than other PDL-1 assays, that stains more TCs.[5] This fact is reflected in our results with better IC scoring and lower TC scoring with SP142 and a good NPA. Therefore, TC scoring or Combined positive scoring (TC + IC) is not recommended in urothelial carcinoma for SP142.

The clinical trial validated recommended scoring is important in-patient selection. Though some papers[2] recommend using 25% cut off for IC and TC scoring to better segregate responders from non-responders, the SP263 already uses 25% cut off for scoring of IC and TC. The blueprint studies[16],[17] in lung cancer (phase 1 and 2) showed good concordance for SP263, 22C3 and 28-8 for TC scoring, however SP142 was an outlier in term of TC staining and concordance for IC scoring was poor among pathologist (kappa score of 0.27-0.28). Schats et al.[18] studied four clones (SP142, SP263, 22C3 and 28-8) on other tumors (lung and melanoma) and showed that PDL-1 kits were not exchangeable and recommended comparison of assay with original kit for accessing clinical utility.

The paired primary and metastatic nodal disease have high degree of concordance in TC and IC scoring for both clones. The concordance of PD-L1 testing between paired nodal metastatic and primary disease and between metachronous metastatic disease and archival primary tumor tissue has been addressed by many large studies.[12],[13] The conclusion of all the studies was that for paired tumor tissue, the results were highly concordant and either of the available tissue can be used for PDL-1 testing by IHC.

Our results, though limited by small sample size, show high degree of correlation between paired primary and lymphnode metastasis for the both the clones. Tretiakova et al.[13] studied 79 matched primary and metastatic cases (most cases were paired primary carcinoma and lymph nodes metastasis), and found 90% concordance for PD-L1 expression using clones 22C3, 28.8, SP142, E1L3N

Zajac et al.,[14] analyzed 332 patients with paired primary and metastatic samples using combined TC/IC scoring, and showed good concordance (57.1 and 50.9%, respectively; P = 0.343, no significant difference). Our findings are in accord with these studies, wherein we found no statistical difference in primary and metastatic lymphnode tumor TC and IC staining. Thus, either of the paired samples may be tested.

However, when compared with metachronous metastatic disease, the studies show very poor concordance, hence emphasizing the spatial-temporal drift in immune response and PDL-1 expression of the tumor. Burgess et al.[13] in a study using SP142 concluded that IC PD-L1 expression is discordant between primary and metastatic lesions (metachronous disease) with simple agreement rate of 79.2% and for TC a better simple agreement of 89.6%, thus highlighting the dynamic PD-L1 expression. Moreover, there is persuasive evidence in the prevailing literature for the escalation of PD-L1 immunoexpression after chemotherapy in lung cancer, which can be significant cause here also. This guides the clinician and pathologist to test most recent biopsied tumor, for correct decision to predict effectiveness of CPI therapy.

To conclude, the testing for PD-L1 status can be done on both primary tumor and metastatic lymph node and in case of metachronous metastatic disease, the recently biopsied metastatic disease sample should be preferred. The harmonization of SP142 and SP263 for urothelial carcinoma is not achieved in this study and the exchangeability of these 2 clones could not be established. The two assays show poor OPA (77.5%), hence show poor concordance. Thus, it is prudent to use the recommended companion/complementary assay approved in a clinically validated trial for the intended drug.


The authors would like to acknowledge Ms. Sangeeta Arora and Mr. Arvind Bhuker for their technical assistance in performing immunohistochemical stains.

Financial support and sponsorship

The study was supported by intramural grant from RGCIRC (Res/SCM/23/2017/59).

Conflicts of interest

There are no conflicts of interest.


1Walk EE, Yohe SL, Beckman A, Schade A, Zutter MM, Pfeifer J, et al. The cancer immunotherapy biomarker testing landscape. Arch Pathol Lab Med 2020;144:706-24.
2Zhu J, Armstrong AJ, Friedlander TW, Kim W, Pal SK, George DJ, et al. Biomarkers of immunotherapy in urothelial and renal cell carcinoma: PD-L1, tumor mutational burden, and beyond. J Immunother Cancer 2018;6:4.
3Kintsler S, Cassataro MA, Drosch M, Holenya P, Knuechel R, Braunschweig T. Expression of programmed death ligand (PD-L1) in different tumors. Comparison of several current available antibody clones and antibody profiling. Ann Diag Pathol 2019;41:24-37.
4Available from:
5Vennapusa B, Baker B, Kowanetz M, Boone J, Menzl I, Bruey JM, et al. Development of a PD-L1 complementary diagnostic immunohistochemistry assay (SP142) for atezolizumab. Appl Immunohistochem Mol Morphol 2019;27:92-100.
6Udall M, Rizzo M, Kenny J, Doherty J, Dahm S, Robbins P, et al. PD-L1 diagnostic tests: A systematic literature review of scoring algorithms and test-validation metrics. Diagn Pathol 2018;13:12.
7Eckstein M, Cimadamore A, Hartmann A, Lopez-Beltran A, Cheng L, Scarpelli M, et al. PD-L1 assessment in urothelial carcinoma: A practical approach. Ann Transl Med 2019;7:690.
8SP263 Interpretation guide for urothelial carcinoma by ventana. Available from: [Last accessed on 2020 Sep 28].
9SP142 Interpretation guide for urothelial carcinoma by Ventana. Available from: [Last accessed on 2020 Sep 28].
10Zajac M, Scott M, Ratcliffe M, Scorer P, Barker C, Al-Masri H, et al. Concordance among four commercially available, validated programmed cell death ligand-1 assays in urothelial carcinoma. Diagn Pathol 2019;14:99.
11Rijnders M, van der Veldt AA, Zuiverloon TC, Grünberg K, Thunnissen E, de Wit R, et al. PD-L1 antibody comparison in urothelial carcinoma. Eur Urol 2019;75:538-40.
12Tretiakova M, Fulton R, Kocherginsky M, Long T, Ussakli C, Antic T, et al. Concordance study of PD-L1 expression in primary and metastatic bladder carcinomas: Comparison of four commonly used antibodies and RNA expression. Mod Pathol 2018;31:623-32.
13Burgess EF, Livasy C, Hartman A, Robinson MM, Symanowski J, Naso C, et al. Discordance of high PD-L1 expression in primary and metastatic urothelial carcinoma lesions. Urol Oncol 2019;37:299.e19-25.
14Zajac M, Boothman AM, Ben Y, Gupta A, Jin X, Antal J, et al. PD-L1 expression in primary lesions vs metastatic sites and by demographics in advanced urothelial carcinoma samples [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77 (13 Suppl):Abstract nr 656. doi: 10.1158/1538-7445.AM2017-656.
15Hodgson 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.
16Hirsch 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.
17Tsao MS, Kerr KM, Kockx M, Beasley MB, Borczuk AC, Botling J, et al. PD-L1 immunohistochemistry comparability study in real-life clinical samples: Results of blueprint phase 2 project. J Thorac Oncol 2018;13:1302-11.
18Schats KA, Van Vré EA, Boeckx C, De Bie M, Schrijvers DM, Neyns B, et al. Optimal evaluation of programmed death ligand-1 on tumor cells versus immune cells requires different detection methods. Arch Pathol Lab Med 2018;142:982-91.