(P 129) MATURE DENDRITIC CELLS ON TISSUE MICROARRAY CORES CORRELATE WITH THE PRESENCE OF ADJACENT TERTIARY LYMPHOID STRUCTURES IN EPITHELIOID SARCOMA

Objective: Immunotherapy has emerged as a new treatment modality in some sarcomas, particularly for tumors with tertiary lymphoid structures (TLS). TLS are functional lymphoid aggregates that form in sites of chronic inflammation and are indicative of an active anticancer immune response when present in the tumor microenvironment. While TLS are emerging as a predictive biomarker for immunotherapy benefit, their assessment at scale on clinical specimens remains challenging. Accurately identifying TLS requires multiplex techniques that highlight the diverse immune cells forming these structures, something that is not easy to perform on large series of full faced tissue slides. Tissue microarrays facilitate assessment of the clinical value of biomarkers on large numbers of specimens in parallel, at low cost. However, it is not clear whether small tissue microarray cores are capable of representing associated TLS responses. The objective of this project is to evaluate the sufficiency of tissue microarray cores for TLS representation, in epithelioid sarcoma, as an example of a sarcoma with variable degrees of TLS formation.

Methods: We have developed and optimized a multiplex immunohistochemistry panel of CD3, CD20, and CD208 to identify T cells, B cells, and mature dendritic cells, respectively. This multiplex panel was applied to 19 duplicated cases on an epithelioid sarcoma tissue microarray, and to their respective full faced sections from their source blocks. The tissue microarray cores were scored by a trained pathologist for the number of CD3, CD20, and CD208 positive cells. Cores with at least one cell expressing CD3, CD20, or CD208 were considered positive for each immune biomarker. Cognate full faced slides were separately assessed for the presence of TLSs that were defined as specific lymphoid-like aggregates of B cells, T cells, and dendritic cells, blinded to the tissue microarray result.

Results: As shown in Table 1, of 19 cases included in the study, 13 cases had at least one TLS on their full faced sections. CD20 (B cell) positivity was observed to be always accompanied by CD3+ T cells. All the tissue microarray cases with CD208 were also positive for CD20 and CD3. Interestingly, all the CD208 positive tissue microarray cases had at least one TLS on their cognate full faced sections. On the other hand, none of the cases lacking TLS showed CD208+ dendritic cells on their tissue microarray cores, even when there was a high degree of non-TLS immune infiltration. Of all three biomarkers, CD208 had the highest specificity and CD3 showed the highest sensitivity on tissue microarrays for detection of an associated TLS response.

Conclusion: Results of our study show that even though CD3 and CD20 may have more sensitivity as TLS markers, CD208 is much more specific for the presence of TLS on cognate full faced sections. More soft tissue sarcoma types will be added to this ongoing project to assess the value of CD208 as a specific immunohistochemistry biomarker for TLS that can be applied to tissue microarrays. Results from this project may facilitate the evaluation of TLS as a predictive biomarker for immunotherapy benefit in larger series of sarcomas.

CTOS abstract-table.pdf