Levels of peripheral CD4(+)FoxP3(+) regulatory T cells are negatively associated with clinical response to adoptive immunotherapy of human cancer

Abstract

CD4(+)FoxP3(+) regulatory T cells (Tregs) have been shown to suppress T cell-mediated host immune responses against self- and nonself-antigens; however, the impact of CD4(+) Tregs on human antitumor immune responses and their influence on cancer treatment are unknown. In the present study, we explored the factors that influence CD4(+) Treg reconstitution in patients receiving adoptive immunotherapy following conditioning regimens designed to enhance T-cell function and evaluated potential associations between CD4(+) Treg levels and clinical responses to therapy. The analysis of 4 trials employing nonmyeloablative chemotherapy with or without total body irradiation (TBI) before adoptive T-cell transfer revealed that the percentage and number of reconstituting CD4(+)FoxP3(+) Tregs observed in the peripheral blood was higher in nonresponders than in responders. The addition of TBI resulted in a further depletion of CD4(+) Tregs, and the degree of depletion was dependent on the TBI dose. The number of administered doses of IL-2 was found to be positively associated with peripheral Treg reconstitution. These observations provide strong evidence that endogenous CD4(+) Tregs have a negative impact on cancer therapy, and suggest that strategies reducing Treg levels may provide clinical benefit to cancer patients. All 5 clinical trials are registered at www.clinicaltrials.gov as NCT00001832, NCT00096382, NCT00335127, NCT00509496, and NCT00513604.

Figure 1

Widely varying frequencies of CD4⁺FoxP3⁺ T cells in patient PBMCs after adoptive immunotherapy. (A) Samples of patient PBMCs from 5 NMA clinical trials (detailed in “Methods”) were analyzed for the presence of CD4⁺FoxP3⁺ T cells. Numbers in parentheses indicate the total number of patients enrolled in these trials. (B) The percentages (% of CD3⁺ T cells) and (C) absolute cell count per microliter of CD4⁺FoxP3⁺ T cells in patient peripheral blood were analyzed before (PRE), approximately 1 week (week 1), and 4 weeks (week 4) after adoptive transfer. The maximum, 75th percentile, median, 25th percentile, and minimum values are shown. The number of CD4⁺FoxP3⁺ T cells represents the product of the absolute lymphocyte count, the fraction of the absolute lymphocyte count corresponding to CD3⁺ T cells, and the fraction of CD3⁺ T cells expressing FoxP3⁺ T cells.

Figure 2

Phenotype of CD4⁺FoxP3⁺ T cells in the peripheral blood of ACT patients. (A) Cell-surface expression of CD25, CD27, and CD127, as well as intracellular CTLA-4 and FoxP3 expression, were evaluated on CD4-gated T cells from 1 of 29 patient PBMC samples analyzed before, 1 week, and 4 weeks after treatment. Numbers in the top left quadrants indicate the mean fluorescence intensity of CD25/FoxP3 or CTLA-4/FoxP3 expression. (B) Expression of CD45RO on gated CD4⁺FoxP3⁺ T cells from 1 of 8 representative patient PBMC samples obtained before, 1 week, and 4 weeks after treatment. (C) Expression of Ki-67 on gated CD4⁺FoxP3⁺ T cells from 1 of 29 representative patients before, 1 week, and 4 weeks after treatment. (D) CFSE dilution of labeled CD4⁺CD25⁻ effector T cells stimulated with anti-CD3 Ab was assessed after a 96-hour coculture with CD4⁺CD25^high T cells containing > 60% FoxP3⁺ T cells at varying effector T cell-to-Treg ratios. The percentages of T cells undergoing at least 1 division are shown, and data are representative of 2 experiments carried out with week-4 PBMC samples from 2 patients in the CD8⁺ young tumor-infiltrating lymphocyte (TIL) trial.

Figure 3

Infused T cells lack characteristic features of CD4⁺ Tregs. (A-B) Intracellular cytokine expression was examined on CD4-gated PBMCs after 6 hours of stimulation with P/I. IFN-γ and IL-2 production were evaluated on pretreatment, week-1, and week-4 samples of PBMCs. Numbers indicate the percentage of cells in each quadrant. Representative data from 1 of 15 patient samples analyzed are shown. (C) Intracellular IFN-γ expression was examined after a 6-hour stimulation of infusion TILs with P/I. Representative data from 1 of 15 patient samples analyzed are shown. (D) DNA methylation assays of week-4 PBMC sample from 1 patient in the CD8⁺ young TIL trial. Cells were sorted into CD4⁺CD25H (high), CD4⁺CD25M (medium), and CD4⁺CD25L (low), and the frequency of FoxP3⁺ cells in each population is shown. The methylation status of 11 CpG dinucleotides within intron 1 of the FOXP3 gene shown previously to be undermethylated in CD4⁺ Tregs were analyzed. The percentages of the 11 CpGs that were methylated, evaluated as described in “Methods,” are shown in heat maps, and the average percentages of the 11 CpGs that were methylated re indicated below the heat maps. (E) DNA methylation analysis of a sample of infusion TILs. DNA isolated from CD4⁺CD25⁺, CD4⁺CD25⁻, and CD8⁺CD25⁺ T cells that were sorted using magnetic beads as described in “Methods” was subjected to methylation analysis.

Figure 4

CD4⁺ Treg reconstitution in peripheral blood of NMA, NMA + 2-Gy TBI, and NMA + 12-Gy TBI TIL patients. (A) The total numbers of CD3⁺, CD8⁺, and CD4⁺ TILs that were infused into patients in the NMA-alone, NMA + 2-Gy TBI, and NMA + 12-Gy TBI TIL trials are shown. (B) The absolute CD3, CD8, and CD4 T-cell counts of week-1 PBMCs in the NMA alone, NMA + 2-Gy TBI, and NMA + 12-Gy TBI trials are indicated. Horizontal bars represent median values. (C-D) Percentages of Treg in CD3⁺ T cells and Treg absolute cell counts in the 3 NMA TIL trials were evaluated at week 1. (E) Percentages of CD4⁺ Treg as a percentage of total CD4⁺ T cells in the NMA-alone, NMA + 2-Gy TBI, and NMA + 12-Gy TBI TIL trials were compared at week 1. (F) The ratios of CD4⁺ Tregs to total CD8⁺ T cells in 3 NMA TIL trials were compared at week 1.

Figure 5

Association between frequencies of CD4⁺ Tregs and patient response to therapy. (A) Nonparametric analysis of the relationship between doses of IL-2 administrated and percentages or absolute cell count of Tregs from week-1 PBMCs in 3 NMA trials. Spearman r and P values are shown. (B) The percentages of Tregs in CD3⁺ T cells and absolute Treg counts of week-1 samples from nonresponders and responders in 3 NMA TIL trials are shown. Horizontal bars represent the median values. (C) The percentages of CD4⁺ Tregs as a percentage of total CD4⁺ T cells in week-1 samples and absolute CD4⁺ T-cell counts in the peripheral blood of nonresponders (NR) and responders (R) in the 3 NMA TIL trials are shown. (D) Ratio of CD4⁺ Tregs to total CD8⁺ T cells in week-1 samples and absolute CD8⁺ T-cell counts in the peripheral blood of nonresponders and responders in the 3 NMA TIL trials are shown.

Figure 6

Treg reconstitution in CD8⁺ young TIL trial. (A-B) Percentages and cell counts of Tregs in PBMCs from CD8⁺ young patients are shown. (C) Nonparametric correlation between doses of IL-2 administrated and percentages or absolute cell count of Treg from week-1 and week-4 PBMCs in CD8⁺ young TIL trial. Spearman r and P values are shown. (D) Percentages of Tregs in CD3⁺ T cells and Treg counts of nonresponders (NR) and responders (R) in the CD8⁺ young TIL trial are shown. (E-F) Nonparametric analysis was used to compare the percentages and absolute cell counts of CD4⁺ Tregs at week 1 in the peripheral blood of nonresponders and responders from the 4 NMA TIL trials. Horizontal bars represent the median values.

Figure 7

Evidence for lack of conversion of CD4⁺ T cells in TILs into CD4⁺ Tregs after adoptive transfer. (A) Cell-surface expression of the mouse TCRβ chain was analyzed on PMBCs from a representative patient (1 week to 15 months) from the gp100 TCR trial. Numbers indicate the percentages in each quadrant. The week-1 figure is a representative sample of 1 of 16 week-1 samples analyzed. Nonparametric correlation between the percentages of Tregs and persistent (mTCRβ⁺) or nonpersistent (mTCRβ⁻) CD4⁺ T cells from week-1 PBMCs in the gp100 trial (B), the percentages of Treg and CD4⁺ T cells from week-1 PBMCs in the CD8⁺ young TIL trial (C), and the percentages of Treg and persistent or nonpersistent CD8⁺ T cells from week-1 PBMCs in the gp100 TCR trial (D). Spearman r and P values are shown.