Background CD155 (poliovirus receptor, PVR) is an immunosuppressive molecule overexpressed in lung adenocarcinoma (LUAD) and breast cancers (BRCA). However, no mutation has been identified that could be linked to such overexpression, and therefore it is likely regulated on the transcriptional level. Previously we identified interleukin-22 (IL-22) signaling as one of the pathways that upregulate CD155 expression in mouse models of lung and breast cancer. However, it is difficult to assess the activity of the IL-22 axis in the publicly available datasets since IL-22 signaling involves several components that must be considered: IL22, IL22RA1 and IL10RB, which encode heterodimeric IL-22 receptors found on tumor cells, and IL22RA2, which encodes soluble IL-22 binding protein (IL-22BP), an antagonist of IL-22 secreted by myeloid cells. The expression of IL22 itself is often missing in the available data due to the insufficient depth of sequencing, which prompts scientists to utilize one of the available components of the axis as a surrogate.

Materials and Methods Here we used agglomerative clustering, a bottom-up method of hierarchical clustering, to stratify the dataset by gene expression patterns in an unsupervised way. For this, we used LUAD and BRCA sequencing datasets from The Cancer Genome Atlas (TCGA). In the current analysis, we focused on HER2+ samples of the BRCA dataset. For clustering, we utilized PVR, IL22RA1, IL22RA2, and IL10RB gene expression. In identified clusters, we compared overall (OS) and restricted mean survival time (RMST) for the first five-year follow-up.

Results In both cohorts, we identified three clusters that are characterized by the following patterns of gene expression: cluster 0 (IL22RA1 ^high, IL22RA2 ^low, IL10RB ^med, PVR ^high), cluster 1 (IL22RA1 ^low, IL22RA2 ^high, IL10RB ^high, PVR ^low), and cluster 2 (IL22RA1 ^low, IL22RA2 ^low, IL10RB ^low, PVR ^medium). Here, cluster 0, identified by a high expression of IL-22 receptor and CD155, and low expression of IL-22BP, was characterized by the poor OS in both cohorts. Moreover, the average difference in RMST between clusters 1 and 0 constituted 361 days in lung and 93 days in HER2+ breast cancer. This difference could be explained by the prevalence of advanced-stage patients in the lung cancer but not in the breast cancer cohort. Moreover, we identified that this difference in survival between clusters stems from differences in early (I and II), but not late-stage (III and IV) patient entries.

Conclusions Here we identified that early-stage lung and HER2+ breast cancer patients could be stratified according to their IL22RA1, IL22RA2, IL10RB, and PVR expression with cluster 0 predicting lower OS and shorter RMST. Mechanistically, the activity of such a pathway defines the immunosuppressive axis we identified previously.

Disclosure Information D. Briukhovetska: A. Employment (full or part-time); Significant; ♣ University Hospital, LMU Munich. B. Research Grant (principal investigator, collaborator or consultant and pending grants as well as grants already received); Significant; Novartis. J. Jobst: None. S. Endres: A. Employment (full or part-time); Significant; University Hospital, LMU Munich. B. Research Grant (principal investigator, collaborator or consultant and pending grants as well as grants already received); Significant; Deutsche Forschungsgemeinschaft, Elitenetzwerk Bayern, Bio-M, Munich, Germany, TCR2, Cambridge, MA, USA. D. Speakers Bureau/Honoraria (speakers bureau, symposia, and expert witness); Significant; Paul-Martini-Stiftung, Else Kröner-Fresenius Stiftung. E. Ownership Interest (stock, stock options, patent or other intellectual property); Modest; Carina Biotech Ltd, Mawson Lakes, Australia. E. Ownership Interest (stock, stock options, patent or other intellectual property); Significant; TCR2, Cambridge, MA, USA. F. Consultant/Advisory Board; Significant; Gilde Healthcare, Utrecht, Netherlands. S. Kobold: A. Employment (full or part-time); Significant; University Hospital, LMU Munich. B. Research Grant (principal investigator, collaborator or consultant and pending grants as well as grants already received); Significant; European Union, Hector Foundation, Elite Network of Bavaria, Melanoma Research Alliance, EKFS, German Cancer Aid, Ernst-Jung-Stiftung, German Excellence Initiative, BMBF, Research Council, DFG, SFB-TRR 338/1, Fritz-Bender Foundation, José-Carreras Foundation, Bio-M, TCR2 Inc. D. Speakers Bureau/Honoraria (speakers bureau, symposia, and expert witness); Modest; BMS, GSK. D. Speakers Bureau/Honoraria (speakers bureau, symposia, and expert witness); Significant; Novartis. E. Ownership Interest (stock, stock options, patent or other intellectual property); Modest; Carina Biotech Ltd, Mawson Lakes, Australia. E. Ownership Interest (stock, stock options, patent or other intellectual property); Significant; TCR2, Cambridge, MA, USA. F. Consultant/Advisory Board; Modest; Tabby therapeutic ltd. F. Consultant/Advisory Board; Significant; TCR2 Inc, Novartis.