Improving Prediction Marker Models with the Ratio of CD39+CD8+ to Total CD8+ T Cells: How Good is Good Enough?
Haitang Yang,Sean R. R. Hall,Feng Yao
DOI: https://doi.org/10.1016/j.jtho.2021.05.014
IF: 20.121
2021-01-01
Journal of Thoracic Oncology
Abstract:There is a paucity of biomarkers that can reliably identify patients who will respond to immunotherapy. We read with great interest the recent work by Yeong et al.1Yeong J. Suteja L. Simoni Y. et al.Intra-tumoral CD39+CD8+ T cells predict response to PD-1/PD-L1 blockade in patients with NSCLC [e-pub ahead of print]. J Thorac Oncol.https://doi.org/10.1016/j.jtho.2021.04.016Date accessed: May 10, 2021Google Scholar By implementing multiplex immunohistochemistry or immunofluorescence, the authors revealed that the proportion of CD39+CD8+ T cells predicted response to immune checkpoint inhibition (ICI) (area under the curve = 0.954), thus providing a potential biomarker to guide the selection of patients with NSCLC who will benefit from immunotherapy. Despite this, there are several critical issues with this study. First, whether CD39, a marker that identifies tumor antigen-specific T cells,2Simoni Y. Becht E. Fehlings M. et al.Bystander CD8+ T cells are abundant and phenotypically distinct in human tumour infiltrates.Nature. 2018; 557: 575-579Crossref PubMed Scopus (547) Google Scholar represents the best candidate to predict response to ICIs is not known, because intratumoral CD8 tumor-infiltrating lymphocytes (TILs) consist of multiple subpopulations that overexpress several additional tumor-specific markers (e.g., programmed cell death protein-1 [PD-1], LAG3, CD38, CD101).3Blank C.U. Haining W.N. Held W. et al.Defining ‘T cell exhaustion’.Nat Rev Immunol. 2019; 19: 665-674Crossref PubMed Scopus (385) Google Scholar Previous work revealed that PD-1high subset within the CD8+ TIL compartment strongly predicts both response and survival in patients with NSCLC treated with PD-1 blockade.4Thommen D.S. Koelzer V.H. Herzig P. et al.A transcriptionally and functionally distinct PD-1+ CD8+ T cell pool with predictive potential in non-small-cell lung cancer treated with PD-1 blockade.Nat Med. 2018; 24: 994-1004Crossref PubMed Scopus (453) Google Scholar We mined single-cell RNA sequencing analysis of two high-quality NSCLC data sets revealing CD39 (ENTPD1) is expressed not only by subpopulations of CD8+ TILs (particularly exhausted CD8+ T cells) but also by regulatory T cells , including subsets of monocytes/macrophages (Figs. 1 and 2). Unsurprisingly, exhausted CD8+ T cells coexpress CD39 and other exhaustion/chronic-activation markers, for example, PD-1 (PDCD1), CD38, and, in particular, LAG3. Together, whether CD39 represents the best candidate remains to be defined.Figure 2scRNA-seq analysis of immune cell types within human NSCLC. (A) Unsupervised clustering of viable immune cells from six patient samples with NSCLC, represented as a UMAP plot. Different immune cell type clusters were color coded. scRNA-seq data were downloaded from TISCH (http://tisch.comp-genomics.org/home/). (B) Distribution of the expression of indicated genes across different immune cell types in NSCLC, represented as a UMAP plot. CD8T, CD8+ T cell; CD8Tex, exhausted CD8+ T cell; scRNA-seq, single-cell RNA sequencing; TISCH, tumor immune single-cell hub; UMAP, Uniform Manifold Approximation and Projection.View Large Image Figure ViewerDownload Hi-res image Download (PPT) Second, response to ICI is predicated on the reinvigoration of pre-existing exhausted TILs. Nevertheless, emerging evidence challenges this long-held view revealing the T cell response to ICIs originates outside the tumor, evolving from a distinct repertoire of TILs recruited from the periphery.5Wu T.D. Madireddi S. de Almeida P.E. et al.Peripheral T cell expansion predicts tumour infiltration and clinical response.Nature. 2020; 579: 274-278Crossref PubMed Scopus (202) Google Scholar Notably, in treated patients, pre-existing exhausted T cell clones had little expansion. Third, a potential bias exists pertaining to the use of biopsy specimens to generate ratio or abundance scores. Spatial heterogeneity of TILs is missing in single sample biopsies of this study, thus potentially leading to bias when calculating the ratio of CD39+CD8+ to total CD8+ T cells. Moreover, spatial interaction between TILs and cancer cells is absent, which might influence not only baseline immune status but also response to treatment. Therefore, multiregional biopsies may partially reduce this bias. A final critical point relates to how the response evaluation was performed, which the authors did not clearly describe, given that conventional computed tomography scans cannot truly reflect the pathologic response to ICIs.6Lee H.Y. Lee H.J. Kim Y.T. et al.Value of combined interpretation of computed tomography response and positron emission tomography response for prediction of prognosis after neoadjuvant chemotherapy in non-small cell lung cancer.J Thorac Oncol. 2010; 5: 497-503Abstract Full Text Full Text PDF PubMed Scopus (28) Google Scholar Therefore, whether nonresponders defined in this study represent the tumors that were resistant to ICI remains to be defined, which could also result in potential bias. ICIs in a neoadjuvant setting can provide a precise evaluation of the therapeutic responses. Finally, the 35 patients receiving ICI had previous lines of treatment, which further confounds the use of CD39 expression on CD8+ TILs as a predictive biomarker in NSCLC. In summary, the work by Yeong et al.1Yeong J. Suteja L. Simoni Y. et al.Intra-tumoral CD39+CD8+ T cells predict response to PD-1/PD-L1 blockade in patients with NSCLC [e-pub ahead of print]. J Thorac Oncol.https://doi.org/10.1016/j.jtho.2021.04.016Date accessed: May 10, 2021Google Scholar provides an example of the importance of immunohistochemistry-based models for predicting the response to ICIs. Moving forward, it will be important to determine the origin of CD39+CD8+ TILs in responding patients and its role in baseline immune status, which remains key unresolved issues. Haitang Yang: Conceptualization, Methodology, Software, Data curation, Writing- original draft preparation, visualization, investigation. Sean R. R. Hall: Data curation, Writing- original draft preparation. Feng Yao: Supervision, Writing- reviewing and editing, Conceptualization. Improving Precision and Implementation of Immuno-Oncology BiomarkersJournal of Thoracic OncologyVol. 16Issue 10PreviewWe thank Yang et al.1 for showing interest in our recent work on the use of CD39+ CD8+ T cell as a potential marker for predicting responses to programmed cell death protein-1 or programmed death-ligand 1 inhibitors (immune checkpoint inhibitor [ICI]),2 specifically raising important points on the precision of CD39 as a biomarker candidate. Full-Text PDF