← Back to List

Program Number: B046
Session Name: Poster Session

The immune landscape of radiated versus non-radiated head and neck squamous cell carcinomas

Katherine C Wai, MD; Zachary Stensland, BA; Sophia Guldberg, PhD; Minna Apostolova, BA; Maha Rahim, PhD; Trine Line Okholm, PhD; Kyle Jones, DDS, PhD; Stanley Tamaki, PhD; Patrick K Ha, MD; Alain P Algazi, MD; Matthew H Spitzer, PhD; UCSF

Background: Preclinical models of head and neck squamous cell carcinoma (HNSCC) have demonstrated that radiation can negatively impact the systemic antitumor immune response, but these changes are not well studied in humans. We thus sought to characterize the systemic immune changes among patients with and without a history of head and neck radiation.

Methods: Matched tumor and blood samples were collected from patients undergoing curative intent surgery. Specimens were analyzed using mass cytometry by time-of-flight (CyTOF). Immune cell populations were clustered using the FlowSOM algorithm. CD8+ T-cells were classified and further subclustered. Population frequencies were compared between radiated and non-radiated patients, corrected for multiple hypothesis testing. For lymph node (LN) analysis, we utilized single-cell data from a pre-existing database. Briefly, a tissue microarray was created from surgically resected LN specimens from HNSCC patients and analyzed using multiplexed ion beam imaging (MIBI). This allowed for spatial immunophenotyping. Individual cells were identified through segmentation algorithm Mesmer and phenotyped using protein expression. A spatial enrichment analysis was then conducted to determine which immune cell pairs may be more likely to interact. For example, for each cell type pair of cell type X and cell type Y, the number of times cell X was within a 20um radius of cell Y was counted. These counts were compared between radiated and non-radiated patients.

Results: Overall, there were 65 patients who had either tumor and/or blood data generated by CyTOF and 14/65 (22%) had a history of radiation (18/65). In the MIBI cohort, 4/23 (17%) had a history of radiation. In the blood, there was overall lymphopenia among radiated patients, with decreased proportion of CD4+, CD8+ and γδT-cells (p<0.01). This was similarly demonstrated in tumor, with a decreased number of CD4+ and γδT-cells (p<0.05), and a trend toward a decreased number of CD8+ T-cells (p=0.11). When CD8+ T-cells were further subclustered in tumor, we found a decreased proportion of progenitor exhausted T-cells (Tpex) and activated CD103+CD8+ T-cells (p<0.05) in radiated tumors. Within the LN, among radiated patients, there was a trend towards decreased frequency of cells important in mediating the antitumor response including Tpex, intermediate exhausted cells (Tex-int), and dendritic cells, with an increased frequency of immunosuppressive T regulatory cells (Tregs) [Figure 1]. Furthermore, CD163+ macrophages were more likely to be neighboring Tpex (p=0.02), and a similar finding was observed with Tregs to Tpex (p=0.25) [Figure 2].

Conclusion: HNSCC patients with a history of radiation demonstrate a less robust antitumor response compared to patients without prior radiation. These differences are systemic and are within the tumor microenvironment, LN and blood. This impaired response may be multifactorial, both in the overall decreased frequency of cells essential to the antitumor response and the increased number of immunosuppressive niches within the lymph node, impacting the ability of key T-cells to coordinate the antitumor response. Further understanding of the short-term and long-term effects of radiation may guide future studies for treatment strategies and timing.

 

 

← Back to List