Immune presence and tumor recognition
An immune-rich tissue sample tells us immune cells are present. Whether those cells can recognize and kill the malignant cells is a separate question.
Antigens and presentation
Three separate questions about the same tissue. Immune-cell presence does not by itself establish presentation, recognition, or killing.
An antigen is something the immune system can recognize. A neoantigen is a new antigen arising from a tumor-specific change. For many CD8 T-cell targets, the relevant object is a short protein fragment, or peptide, displayed by an HLA class I molecule.
HLA is the human version of the major histocompatibility complex, MHC. HLA-I includes HLA-A, HLA-B, and HLA-C. B2M, beta-2-microglobulin, helps form the class I complex. Antigen-processing machinery (APM) includes proteins that generate, transport, and load peptides. These steps connect protein production to surface display. Antigen processing and presentation
The evidence ladder is:
Authentic tumor change → expressed RNA → protein/peptide generation → HLA presentation → T-cell recognition → tumor killing → clinical benefit.
Evidence at one step does not establish the later steps. Immunopeptidomics measures recovered HLA-bound peptides; it does not itself prove a useful T-cell response. Immunogenicity means capacity to provoke an immune response, which differs from expression or predicted binding.
What a gene copy or RNA score can tell us
HLA haplotype loss removes one inherited set of HLA alleles, potentially changing the available peptide display. Single-copy B2M loss leaves one copy and does not automatically abolish protein or presentation. Expression, assembly, loading, and localization still matter. Primary lung-cancer research illustrates allele-specific HLA loss as an escape mechanism; it does not establish escape in every tumor. McGranahan et al.
Interferons (IFNs) are signaling proteins. Type I includes IFN-alpha/beta; IFN-gamma is type II. An interferon-response module measures a related RNA program. A low baseline score does not prove the cell cannot respond to stimulation. IFN-gamma can induce antigen-presentation components, but transcription and assembled surface complexes remain separate measurements. Primary interferon/assembly experiment
The surrounding immune vocabulary
| Term | Plain-language meaning | Interpretation limit |
|---|---|---|
| sTILs | Stromal tumor-infiltrating lymphocytes; pathology scores lymphocytes in the relevant stromal area | A percentage of stromal area under scoring rules, not the fraction of all cells that are T cells |
| Plasma cells | Antibody-secreting cells derived from B cells | Their presence does not identify what the antibodies recognize |
| IgA / IgG | Different antibody classes | An antibody-class RNA signal does not prove tumor-specific activity |
| Chemokine | A signal affecting cell movement and recruitment | A ligand's presence does not establish where responding cells moved |
| CXCL12 | A chemokine involved in cell positioning and interactions | A stromal RNA signal is not a direct measurement of T-cell exclusion |
International TILs Working Group scoring recommendations define the stromal area used for sTIL assessment.
The October 4 review distinguishes immune-rich surrounding tissue from lower tumor-cell APM modules and withdraws its computational sTIL estimate. The pathology measurements from different specimens remain separate. Report qualifications
Say it in your own words
“There are immune cells in the sample. To say they recognize the cancer, we also need evidence that the tumor displays the relevant target and that T cells respond to it.”
Check yourself: Does an expressed mutation prove an immunogenic neoantigen? No. Processing, presentation, and recognition still need evidence. Continue to comparisons and the evidence ladder, or the HLA lesson for depth.