ESMO @ ECC 2015: Novel Assay Developed to Determine PD-L2 Expression in Tumour Samples

PD-L2 is expressed in various human tumour types and may explain the response seen to anti-PD-1 therapy in patients with tumours lacking PD-L1 expression

Expression of programmed death ligand 2 (PDL-2), a known ligand of PD-1, has been demonstrated in seven different human tumour types, according to findings reported at the European Cancer Congress (ECC 2015), convening in Vienna, Austria, 25 - 29 September, 2015.

Agents targeting the PD-1 pathway have transformed cancer treatment by reactivating the patient’s immune system, thus enabling it to aid in eradicating the tumour. The study presented at ECC 2015 investigated the expression of PD-L2 in different tumour types and evaluated the potential role it may play regarding clinical responsiveness to anti-PD-1 therapies. 

Investigators showed a significant association between the extent and distribution of PD-L2, determined using a newly-developed assay, in human archival tumour tissue of different types of cancer compared to standard immunohistochemistry (IHC) labelling of the same sample (range: p  = 0.0012 to p <0.0001). PDL-2 expression was also shown to significantly correlate with levels of PD-L2 mRNA in the tumour samples, which were quantitated using Nanostring (range: p = 0.0037 to p < 0.0001). 

On Sunday 27 September 2015, during the Proffered Paper Session on Immunotherapy in Cancer, Jennifer Yearley, Department of Biologics Operations, Profiling and Expression, Merck & Company Inc., Kenilworth, USA discussed the significance of PD-L2 expression and explained how it may allow broader use of PD-1 targeting agents in the treatment of an expanded range of human tumour types.

Some PD-L1-negative patients also respond to PD-1 axis targeted therapies

According to Dr. Yearley, biomarker screening by IHC is generally done prior to using agents targeting the PD-1 pathway to determine whether the tumour type is a candidate for treatment with these agents. Biomarker screening has been shown to enrich responder populations and has largely focused on the extent of expression of PD-L1, one of the two known ligands of PD-1, in the tumour. 

However, some PD-L1-negative patients have also responded to PD-1 axis targeted therapies, possibly via tumour expression of the other known ligand, PD-L2.

Dr. Yearley presented findings on behalf of a team from Merck from a study utilising a novel IHC assay for PD-L2. The assay was applied to groups of archival tissue from seven human tumour types, including renal cell carcinoma, bladder carcinoma, melanoma, non-small-cell lung cancer (NSCLC), head and neck squamous carcinoma (HNSC), triple-negative breast cancer (TNBC), and gastric carcinoma. Sample sizes of the groups evaluated ranged from 22 tissue samples in TNBC to 94 samples in NSCLC, with a median sample size of 71 samples per tumour type.

The investigators also compared the findings of PD-L2 IHC staining to PD-L2 mRNA levels, as determined using the Nanostring platform, and to results obtained using PD-L1 IHC staining done with Merck clone 22C3. 

This study demonstrated that PD-L2 expression within the tumour microenvironment can be significant in these seven tumour types and that the expression levels associated with RNA levels of PD-L2 in the tumour cells.

The expression of PD-L1 may be discordant with PD-L2 expression

Moreover, the investigators observed that individual tumour samples showed PD-L1 expression in the absence of PD-L2, while other tumour samples showed PD-L2 expression but not PD-L1 expression.

The investigators observed that PD-L2 was expressed in individual tumour samples in the absence of PD-L1. Therefore, screening for both biomarkers may guide the selection of which PD-1 axis targeting agent will provide a stronger clinical response. For example, in a patient with a tumour lacking PD-L1 expression, treatment with anti-PD-1 agents may be more likely to result in clinical benefit than treatment with anti-PD-L1 focused agents.


The investigators summarised that PD-L2 may be expressed within human tumours in the absence of PD-L1, thus providing an alternate target for agents directed to the PD-1 pathway. PD-L2 expression may at least partially account for PD-L1-negative patients having a positive clinical response to anti-PD-1 targeted therapies. 

These findings also suggests that there may be patient populations that may benefit from anti-PD-1 targeted therapies that block PD-1 interaction with both PD-L1 and PD-L2, who may not experience benefit from therapies that only target PD-L1. 


18LBA PD-L2 expression in human tumors: relevance to anti-PD-1 therapy in cancer

The study was funded by Merck & Company Inc.