CD4⁺ T Cell Fate in Multiple Myeloma

The average incidence of multiple myeloma (MM) is in the 7^th decade that coincides with the development of immunosenescence and thymic atrophy, meaning that lymphocyte recovery after lymphopenia-inducing therapies (most notably autologous stem cell transplant, ASCT) is largely reliant on homeostatic proliferation of peripheral T cells rather than replenishing the T cell pool with new thymic emigrants. We have previously shown that there is a significant reduction in circulating naïve T cells with a reciprocal expansion of antigen-experienced cells from newly diagnosed MM (NDMM) to relapsed/refractory disease (RRMM). This results in a reduced TCR repertoire and the accumulation of senescence-associated secretory phenotype cytotoxic T cells, which maintain the ability to produce IFNγ but lose proliferative potential. A reduction in CD4:8 ratio is also a characteristic finding in MM with disease progression, which can be explained by high IL-15 levels in lymphopenic states that preferentially drive expansion of CD8⁺ memory T cells.

We wanted to further evaluate what changes were occurring in the CD4⁺ T cell population with disease progression in MM. We analyzed paired peripheral blood (PB) samples from patients with NDMM and RRMM, and compared with age-matched normal donors (ND). In the NDMM cohort, we examined T cells from PB samples at baseline, after 4 cycles of lenalidomide and dexamethasone (len/dex), and after ASCT; and in the RRMM cohort samples from baseline and after 6 cycles of len/dex.

We firstly confirmed in flow cytometric analysis of T cells at serial intervals in NDMM patients that the reduction in circulating naïve T cells and in CD4:8 ratio occurs post ASCT and does not recover by time of last follow-up.

We next utilised RNA-seq to analyse differences in CD4⁺ T cells from NDMM, RRMM and ND. CD4⁺ T cells from RRMM showed downregulation of cytosolic ribosomal activity but maintenance of mitochondrial ribosomal activity and significant upregulation of pathways involved with calcium signalling. To this end, we evaluated mitochondrial biogenesis and metabolic pathways involved with mitochondrial respiration. Flow cytometric analysis of mitochondrial mass showed a marked increase in RRMM compared with ND, in keeping with a shift towards memory phenotype. Key rate-limiting enzymes in fatty acid β-oxidation (CPT1-A, ACAA2 and ACADVL) were all significantly increased in RRMM compared with ND.

To analyse whether these cells were metabolically active, we also measured mitochondrial membrane potential and reactive oxygen species (ROS), gating on cells with high mitochondrial mass. Mitochondrial membrane potential was significantly increased in RRMM compared with ND, although ROS was reduced. The significance of this is not clear, as ROS are not only implicated in cell senescence and activation-induced cell death, but are also positively involved in tyrosine kinase and PI3K-signalling pathways.

PD-1 has been shown to play a role in transitioning activated CD4⁺ T cells from glycolysis to FAO metabolism, and elevating ROS in activated CD8⁺ T cells. We analysed PD-1 expression on T cells in RRMM and at treatment intervals in NDMM (as described earlier). The proportion of CD4⁺ and CD8⁺ T cells expressing PD-1 was increased 4-6 months post-ASCT and remained elevated in CD4⁺ T cells 9-12 months post-ASCT, but normalised to baseline levels in CD8⁺ T cells. Increased PD-1 expressing CD4⁺ T cells was also evident in RRMM patient samples. This may suggest that in the lymphopenic state, PD-1 expression enhances longevity in a subset of CD4⁺ T cells by promoting reliance on mitochondrial respiration; however, their ability to undergo homeostatic proliferation is impaired. In CD8⁺ T cells, high PD-1 expression may lead to cell death via ROS accumulation, and these cells do not persist.

ASCT remains a backbone of myeloma treatment in medically fit patients. However, this leads to significant permanent defects in the T cell repertoire, which may have unintended adverse outcomes. Additionally, T cells post-ASCT may not be metabolically adequate for the production of CAR-T cells, nor respond to checkpoint blockade therapies.