CAR T-Cell Therapy: Targets, Response Markers and Escape

Choosing a Target

A good CAR target is expressed uniformly on the cancer cells and absent, or dispensable, on essential normal tissue. CD19 and CD22 on B-cell cancers and BCMA on myeloma fit this because loss of normal B cells or plasma cells is survivable.

Because the CAR recognises surface protein directly, it bypasses the MHC class I and antigen-processing steps, so tumours cannot escape CAR T cells by downregulating antigen presentation the way they escape natural T cells.

What Predicts Response

The degree and durability of CAR T-cell expansion in the blood after infusion correlates strongly with response; poor expansion predicts failure. Lower disease burden before infusion, adequate T-cell fitness in the starting material, and the specific costimulatory domain in the construct also matter.

Persistence of CAR T cells, tracked by loss of normal B cells (B-cell aplasia) as a surrogate, is associated with durable remission in acute lymphoblastic leukaemia.

Toxicity and Antigen Escape

Cytokine release syndrome and neurotoxicity are the characteristic acute toxicities, both related to the intensity of CAR T-cell activation and manageable with the interleukin-6 blocker tocilizumab and corticosteroids.

Relapse after CAR T cells is often driven by antigen loss: the tumour stops expressing the target, for example through CD19 mutation, splicing changes or lineage switch. This has motivated dual-target CARs.

Monitoring After Infusion

The first weeks are dominated by watching for cytokine release syndrome and neurotoxicity, graded by standardised scales and treated with tocilizumab, corticosteroids and supportive care. Ferritin, C-reactive protein and cytokine trends help track severity but do not replace clinical grading.

Longer term, response is assessed by standard disease measures — marrow and imaging, and increasingly circulating tumour DNA or flow-based minimal residual disease — while B-cell aplasia and CAR transgene levels serve as surrogates for persistence. Prolonged cytopenias and low immunoglobulins are common and may need growth-factor or immunoglobulin support; loss of the target antigen is checked for at relapse.

Key Takeaways

• ·CAR T cells recognise a surface protein directly, bypassing MHC-restricted antigen presentation.
• ·Robust in-vivo CAR T-cell expansion and persistence predict durable response.
• ·Antigen loss, such as CD19 loss, is a major mechanism of relapse after initial response.