Oops, you're using an old version of your browser so some of the features on this page may not be displaying properly.

MINIMAL Requirements: Google Chrome 24+Mozilla Firefox 20+Internet Explorer 11Opera 15–18Apple Safari 7SeaMonkey 2.15-2.23

Poster Display session 1

4910 - Expression of PD-L1 in Chinese Patients with Common Cancers


28 Sep 2019


Poster Display session 1


Pathology/Molecular Biology

Tumour Site


Min Zheng


Annals of Oncology (2019) 30 (suppl_5): v797-v815. 10.1093/annonc/mdz269


M. Zheng1, S. Wang2, X. Chen3, N. Mao2, H. Zhong4, J. Guo5, X. Pan6, Y. Dai6, D. Chen6, K. Wang7, X. Dong6

Author affiliations

  • 1 The Department Of Thoracic Surgery, The Affiliated Hospital of Guilin Medical College, 541001 - Guilin/CN
  • 2 Thoracic Surgery, Affiliated Tumor Hospital of Guangxi Medical University, 530021 - Nanning/CN
  • 3 Medical Oncology, The First Affiliated Hospital of Guangxi Medical University, 530021 - Nanning/CN
  • 4 Thoracic And Cardiovascular Surgery, The 9th Affiliated Hospital of Guangxi Medical University, 536000 - Beihai/CN
  • 5 Thoracic Surgery, First Affiliated Hospital of Guangxi Medical University, 530021 - Nanning/CN
  • 6 Clinical Pathology, OrigiMed, 201100 - Shanghai/CN
  • 7 Medical Informatics, OrigiMed, Shanghai, 201100 - Shanghai/CN


Login to get immediate access to this content.

If you do not have an ESMO account, please create one for free.

Abstract 4910


PD-L1 immunohistochemistry (IHC) data in China was less reported. We systematically investigated PD-L1 IHC data of Chinese patients generated by Origimed.


PD-L1 was stained using IHC on cancer samples from 2060 Chinese patients in Origimed since 2017. Written informed consent was obtained from each patient. IHC staining was performed on FFPE tissue sections using anti-PD-L1 antibody 28-8. Tumor Proportion Score (TPS) was applied on all the samples. We investigated PD-L1 TPS in lung adenocarcinoma (n = 893), lung squamous carcinoma (n = 172), liver cancer (n = 404), esophageal cancer (n = 186), colorectal cancer (n = 166), pancreatic cancer (n = 136) and gastric cancer (n = 103). We then further applied PD-L1 combined positive score (CPS) on the 103 gastric cancer samples. All the slides were reviewed by the same senior pathologist. 95% confidence interval (CI) was obtained by bootstrap. Information entropy (Shannon’s formula) was measured in natural units.


The highest proportion of PD-L1 TPS > = 1% was observed in lung squamous carcinoma (49%; 95% CI: [43%, 59%]), followed by lung adenocarcinoma (25%; 95% CI: [22%, 28%]), liver cancer (14%; 95% CI: [11%, 18%]), esophageal cancer (12%; 95% CI: [8.1%, 17%]), pancreatic cancer (11%; 95% CI: [5.9%, 16%]), gastric cancer (7%; 95% CI: [1.9%, 12%]) and colorectal cancer (4%; 95% CI: [1.2%, 6.6%]). The proportion of PD-L1 CPS > = 1 for gastric cancer was 17% (95% CI: [9.7%, 24%]). In gastric cancer, information entropy of TPS and CPS was 0.40 and 0.78, respectively.


The proportion of PD-L1 TPS > = 1% in Chinese lung squamous cell carcinoma (49%) was much higher than that in lung adenocarcinoma (25%). PD-L1 CPS contained more information than TPS in gastric cancer. Thus, we recommend applying CPS to gastric cancer in China.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

The authors.


Has not received any funding.


X. Pan: Full / Part-time employment: Origimed. Y. Dai: Full / Part-time employment: Origimed. D. Chen: Full / Part-time employment: Origimed. K. Wang: Full / Part-time employment: Origimed. X. Dong: Full / Part-time employment: Origimed. All other authors have declared no conflicts of interest.

This site uses cookies. Some of these cookies are essential, while others help us improve your experience by providing insights into how the site is being used.

For more detailed information on the cookies we use, please check our Privacy Policy.

Customise settings
  • Necessary cookies enable core functionality. The website cannot function properly without these cookies, and you can only disable them by changing your browser preferences.