The natural killer cell response and tumor debulking are associated with prolonged survival in recurrent glioblastoma patients receiving dendritic cells loaded with autologous tumor lysates

Abstract

Recurrent glioblastomas (GBs) are highly aggressive tumors associated with a 6-8 mo survival rate. In this study, we evaluated the possible benefits of an immunotherapeutic strategy based on mature dendritic cells (DCs) loaded with autologous tumor-cell lysates in 15 patients affected by recurrent GB. The median progression-free survival (PFS) of this patient cohort was 4.4 mo, and the median overall survival (OS) was 8.0 mo. Patients with small tumors at the time of the first vaccination (< 20 cm³; n = 8) had significantly longer PFS and OS than the other patients (6.0 vs. 3.0 mo, p = 0.01; and 16.5 vs. 7.0 mo, p = 0.003, respectively). CD8⁺ T cells, CD56⁺ natural killer (NK) cells and other immune parameters, such as the levels of transforming growth factor β, vascular endothelial growth factor, interleukin-12 and interferon γ (IFNγ), were measured in the peripheral blood and serum of patients before and after immunization, which enabled us to obtain a vaccination/baseline ratio (V/B ratio). An increased V/B ratio for NK cells, but not CD8⁺ T cells, was significantly associated with prolonged PFS and OS. Patients exhibiting NK-cell responses were characterized by high levels of circulating IFNγ and E4BP4, an NK-cell transcription factor. Furthermore, the NK cell V/B ratio was inversely correlated with the TGFβ2 and VEGF V/B ratios. These results suggest that tumor-loaded DCs may increase the survival rate of patients with recurrent GB after effective tumor debulking, and emphasize the role of the NK-cell response in this therapeutic setting.

Keywords: IFNγ; NK cells; dendritic cells; glioblastoma; immunotherapy.

Figure 1. Tumor volume and Karnofsky Performance Score affected patient survival. (A) Treatment schema for patients 1–6 (top) and for patients 7–15 (bottom). (B) Kaplan-Meier analysis correlating tumor volume with progression-free survival (PFS) and overall survival (OS). Median PFS of patients with small (< 20 cm³) vs. large (≥ 20 cm³) tumors: 6 mo vs. 3 mo (n = 8 vs. n = 7; p = 0.01); median OS: 16.5 mo vs. 7 mo (n = 8 vs. n = 7; p = 0.003). (C) Kaplan-Meier analysis of the correlation between Karnofsky Performance Score (KPS) with PFS and OS. Median PFS of patients with high KPS (≥ 70, n = 9) vs. low KPS (< 70, n = 6): 5 mo vs. 2.5 mo (p = 0.005); median OS: 12.5 (high, n = 9) vs. 7 mo (low, n = 6) (p = 0.4).

Figure 2. Patients with increased frequencies of natural killer (NK) cells showed an increased expression of E4BP4 and interferon γ. (A) The time course of NK-cell frequency evaluated by flow cytometry shows a significant increase in some patients (n = 6, blue line) but not in others (n = 8, red line). (B and C) The time course of E4BP4 and interferon γ (IFNγ) expression levels as evaluated by real-time PCR shows a significant increase in patients with increased NK-cell frequency (n = 3, blue line) but not in others (n = 6, red line) (*p < 0.01, **p < 0.001, ***p < 0.0001 vs. first vaccination). (D and E) Flow cytometry histograms of IFNγ production by CD3⁺CD8⁺ T cells and CD3^-CD56⁺ NK cells from patient 9 and patient 11.

Figure 3. Time course of the immunological parameters observed for patient 9 (prolonged survival) and patient 11 (short survival). (A–F) Data for patient 9 (A–C) and patient 11 (D–F) are shown. Each time point represents the natural killer (NK) and CD8⁺ T-cell frequency; Interferon γ (ΙFNγ), E4BP4 and granzyme B expression levels as well as interleukin-12 (IL-12) serum levels are expressed as means ± SD (*p < 0.01, **p < 0.001, ***p < 0.0001).

Figure 4. Magnetic resonance imaging of patient 9 (prolonged survival) and patient 11 (short survival). (A and B) MRI findings for patients 11 (disease progression) and 9 (stable disease) after 6 mo of follow-up. Axial T2W and post-contrast T1W images (right panels and left panels for each patient, respectively) are shown before surgery (upper panels) and 6 mo after the subtotal resection of the lesion and vaccination therapy (lower panels).

Figure 5. Patient survival correlated positively with the vaccination/baseline (V/B) ratio of natural killer-cell frequency and negatively with the V/B ratio of transforming growth factor β2 and vascular endothelial growth factor. (A–C) Kaplan-Meier analyses show the correlations with progression-free survival (PFS) and overall survival (OS) for the natural killer (NK)-cell V/B ratio (A) [median PFS: 8 mo (high, n = 7) vs. 3 mo (low, n = 7), p = 0.004; median OS: 16.5 mo (high, n = 7) vs. 7 mo (low, n = 7), p = 0.02] and transforming growth factor β2 (TGFβ2) and vascular endothelial growth factor (VEGF) V/B ratio (B and C) [median PFS 8.5 mo (low, n = 4) vs. 2 mo (high, n = 5), p = 0.035; median OS 16.5 mo (low, n = 4) vs. 7 mo (high, n = 5) p = 0.02].