What question did this study set out to answer?

This research aims to evaluate the correlation between PD-L1, desmocollin-3, and stromal tumor infiltrating lymphocytes in recurrent head and neck squamous cell carcinomas.

April 5, 2026

Abstract 7721: Correlation between PD-L1, desmocollin 3, and stromal lymphocytes in recurrent head and neck squamous cell carcinomas

Key Points

This research aims to evaluate the correlation between PD-L1, desmocollin-3, and stromal tumor infiltrating lymphocytes in recurrent head and neck squamous cell carcinomas.
Evaluated PD-L1 expression using combined proportion score (CPS) and tumor proportion score (TPS) standards.
Examined DSC3 expression as TPS with immunohistochemistry on patient tissue samples.
Assessed stromal tumor infiltrating lymphocytes (TIL) via hematoxylin and eosin staining, expressed as a percentage of total cells.
Samples were collected from patients who provided informed consent.
77% of samples exhibited PD-L1 expression as CPS, while 51% showed PD-L1 expression as TPS.
93% of samples expressed DSC3, with levels ranging from 0-80%.
No correlation was found between PD-L1 (CPS) and DSC3 expression; a moderate negative correlation was observed with PD-L1 (TPS).
40% of evaluated samples exhibited TIL ≥ 10%, with TIL being absent in five samples, all of which expressed DSC3.

Abstract

Abstract Introduction: Immune checkpoint inhibitors has improved outcome of recurrent /metastatic head and cancer (rmHNSCC) . Benefit is seen in around 20% pf patients and is proportionate to PD-L1 expression. We evaluated correlation between PD-L1, Desmocollin-3 (DSC3) and stromal tumor infiltrating lymphocytes (TIL) to explore possibility of using DSC3 specific therapy along rmHNSCC . DSC3 is a predictive biomarker for an investigational product CADI-05. Methods: Samples from patients with rmHNSCC were evaluated PD-L1 expression using standard protocol and expressed as combined proportion score (CPS) and tumor proportion score ( TPS) . DSC3 expression was evaluated as tumor proportion score (TPS) by immunohistochemistry, using mouse anti-DSC3 monoclonal antibody (Progen, Heidelberg, Germany Cat#61093 dilution: 1:50) following removal of paffinization of formalin-fixed, paraffin-embedded tissue. TIL was evaluated after staining with hematoxylin and eosin and expressed as a percentage of total cells. All evaluated samples were received after informed consent of patients. Results: Total of seventy (70) samples were evaluated between September 2023 to May 2025 for PD-L1, DSC3 and TIL. PD-L1 expression as CPS and TPS was observed in 77% and 51% of samples. PD-L1 values ranged from 0-95% ( CPS) and 0-95%(TPS). DSC3 expression was observed in 93% of samples. DSC3 (TPS) values ranged from 0-80%. No correlation was found between PD-L1(CPS) and DSC3 expression (Spearman correlation coefficient r = -0.1432, p = 0.3846; 95% Confidence interval = -0.4389 to 0.1805). There was a moderate negative correlation (r= -0.3) between PD-L1(TPS) and DSC3. All PD-L1 negative samples expressed DSC3.Mean and median DSC3 expression values were not different between PD-L1 positive and negative samples. Based on expression level TIL were grouped as samples having TIL ≥ 10% and 10%. TIL ≥ 10% were observed in 40% of evaluated samples. All samples expressing PD-L1(CPS and TPS) as well as DSC3 had TIL ≥ 10% . TIL were absent in five samples and all expressed DSC3 from 10-40%. Conclusion: DSC3 is expressed in 93% of rmHNSCC , irrespective of PD-L1 and TIL expression suggesting DSC3 specific therapy may be useful in management of patients with rmHNSCC with absent/low PD-L1 and or TIL . Citation Format: Vedanti Newaskar, Shivani Sharma, Bakulesh M. Khamar. Correlation between PD-L1, desmocollin 3, and stromal lymphocytes in recurrent head and neck squamous cell carcinomas abstract. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 1 (Regular Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(7 Suppl):Abstract nr 7721.

Mark Helpful

Bookmark

Relay