P10. Babi et al.

 Double IHC assay for PD-L1 and SOX10 in melanoma

Cigdem Babi, Freja Stubkjær Bukalo, Anne Vishoff Petersen and Søren Nielsen. Institute of Pathology, Aalborg University Hospital, Denmark

Background: A novel treatment for many cancer types including malignant melanoma is based on immune therapy blocking the signal between the programmed death receptor 1 (PD-1) and its ligand 1 (PD-L1), (1).  PD-1 is mainly expressed on activated T cells, whereas PD-L1 mostly is expressed on several types of tumor cells and macrophages/dendritic cells. Clinical studies have shown that inhibition of the interaction between PD-1 and PD-L1 enhances the T-cell response and mediates antitumor activity, (2). The stratification of patients for the treatment can be guided by immunohistochemistry (IHC) for PD-L1. There may be challenges in the interpretation of PD-L1 IHC as it is difficult to distinguish PD-L1 positive tumor cells from PD-L1 positive macrophages and T-cells as these are among each other in the tissue (3),(4).
This project aims to develop an immunohistochemical double staining method for PD-L1 and SOX10 and examine whether it can help to distinguish the PD-L1 expression in macrophages, T-cells and tumor cells in malignant melanoma.

Methods: 28 FFPE malignant melanoma samples were included in the study and all used in a tissue micro array. IHC assays for PD-L1 (mAb clone 22C3) and SOX10 (mAb BC34) were performed. Validated single IHC assays for PD-L1 and SOX10 were used as references for the development and calibration of the double staining for PD-L1 and SOX10. In the double staining, PD-L1 was visualized by OptiView DAB which stain the membranes brown and SOX10 by UltraView RED which stain the nuclei of melanocytes red. All IHC assays were performed on BenchMark Ultra, Ventana. All samples were scored using NordiQC criteria and for each sample a Tumor Proportion Score (TPS) was assigned. All samples were scored by two groups of observers using virtual slide scans. The TPS for each sample was divided into three cutoff values; <1%, >1-5% and >5%.

Results: Using PD-L1 as single IHC assay with a positive cut-off of ≥1%, 54% (n= 15 of 28) of the melanomas were identified as PD-L1 positive. Using same cut-off, the PD-L1/SOX10 double IHC assay identified 18% (n= 5 of 28) of the melanomas as positive.
Using PD-L1 as single IHC assay with a positive cut-off of >5%, 21% (n= 6 of 28) of the melanomas were identified as PD-L1 positive. Using same cut-off, the PD-L1/SOX10 double IHC assay identified 11% (n= 3 of 28) of the melanomas as positive.

PD-L1 + SOX10 Total
PD-L1 Cut-off <1% >1-5% >5%  
<1% 13 0 0 13
>1-5% 8 1 0 9
>5% 2 1 3 6
Total   23 2 3 28

  Conclusion: The results generated showed a significantly reduced proportion of PD-L1 positive melanomas using the double IHC assay compared to the single IHC assay.The difference might be explained by many factors. First of all, immune cells encountered and interpreted as melanoma cells by single IHC assay will increase the TPS, whereas these cells will be identified as non-melanoma cells by negative SOX10 reaction by the double IHC assay.
However, also a reduced analytical sensitivity for PD-L1 in the double PD-L1/SOX10 IHC assay seemed to cause the difference. This was in particular indicated by a reduced proportion of positive neoplastic cells in the cases with a TPS >5%.


  1. Clinical utility of nivolumab in the treatment of advanced melanoma. Asmar R et al. Ther Clin Risk Manag. 2016 Feb 26;12:313-25.

  2. PD-1/PD-L1 blockade in cancer treatment: perspectives and issues. Hamanishi J et al. Int J Clin Oncol. 2016 Jun;21(3):462-73.

  3. Agilent PD-L1 Interpretation Manual – PD-L1 IHC 22C3 pharmDx is CE-IVD-marked – For in Vitro Diagnostic Use

  4. Agilent PD-L1 IHC 28-8 pharmDx – Interpretation Manual – Melanoma – 29142_pd-l1-ihc-28-8-pharmdx-melanoma-interpretation-manual.pdf