Abstract
Despite profound T cell immunodeficiency, most patients treated with chemotherapy (CT) do not succumb to infection. The basis for residual protective immunity in lymphopenic patients is not known. Understanding the mechanisms underlying persistent immunity in some cancer patients might suggest treatment strategies to enhance immune competence in all. We therefore prospectively studied T cell homeostasis in 73 children (median age 4y; range 1–17) with ALL receiving a protracted, 2-year chemotherapy regimen. T cell frequency and phenotype and TREC levels were measured at diagnosis and every 5 months on therapy, and compared to an age-matched cohort of 805 healthy children. Twenty-three patients (32%) had high-risk features. All patients received a 5-drug induction, consolidation, and a continuation phase with pulses of vincristine and prednisone. High risk patients also received doxorubicin during consolidation. Patients received childhood vaccinations prior to diagnosis and standard PCP prophylaxis during therapy. Most patients had profound defects in CD4 and CD8 T cell compartments at diagnosis that failed to recover during the 2 years of therapy. Absolute counts of CD4 and CD8 T cells in patients remained below 10th %ile for age in 77% and 86% respectively at all timepoints. We evaluated whether this T cell defect affected both naive and memory compartments. Compared to healthy children, the fraction of CD4 cells with a naive (CD45RA/CD62L) phenotype was markedly reduced (77% of patients with <10th %ile values). Consistent with the reduction in naive T cells, thymopoiesis (measured by TREC levels) was significantly lower in ALL patients than in normal controls (p<0.0001). In contrast, the proportion of CD4 cells with a memory phenotype was elevated, and 60% of children had CD45RO% greater than 90th %ile for age. To confirm that this represented preservation of bona fide T cell memory, we studied functional T cell memory in vitro and in vivo. T cell responses to vaccine Ags administered prior to ALL therapy were measured in a sub-set of 10 patients using a novel CFSE-based assay. Proliferation to the vaccine Ags Tetanus and Varicella Zoster Virus was significantly higher in patients than in pediatric controls (p<0.05), suggesting that memory T cells specific for previously-encountered Ags were enriched in the T cell pool. As an in vivo measure of immunity we recorded the frequency of serious infections (positive blood culture or PCP infection). Despite profound T cell lymphopenia, the rate of infection following induction was low at 0.14 infections/patient-year, suggesting that pathogen-specific immunity was largely maintained. We demonstrate that naive and memory T cells show differential sensitivity to CT. Naive T cells and/or thymocytes are most profoundly affected, while the memory T cell pool is relatively spared. Although persistence of T cell memory offers protection from previously encountered pathogens, without replenishment of the T cell pool with naive cells, response to new antigens is likely to be limited. Interventions that specifically protect thymopoiesis and/or naive T cells such as IL7, KGF or androgen inhibitors may lessen the impact of CT on the peripheral T cell pool and improve functional immunity in cancer patients.
N-ras couples antigen receptor signaling to Eomesodermin and to functional CD8+ T cell memory but not to effector differentiation