Preoperative immune checkpoint inhibition and cryoablation in early-stage breast cancer

Abstract

Local cryoablation can engender systemic immune activation/anticancer responses in tumors otherwise resistant to immune checkpoint blockade (ICB). We evaluated the safety/tolerability of preoperative cryoablation plus ipilimumab and nivolumab in 5 early-stage/resectable breast cancers. The primary endpoint was met when all 5 patients underwent standard-of-care primary breast surgery undelayedly. Three patients developed transient hyperthyroidism; one developed grade 4 liver toxicity (resolved with supportive management). We compared this strategy with cryoablation and/or ipilimumab. Dual ICB plus cryoablation induced higher expression of T cell activation markers and serum Th1 cytokines and reduced immunosuppressive serum CD4⁺PD-1^hi T cells, improving effector-to-suppressor T cell ratio. After dual ICB and before cryoablation, T cell receptor sequencing of 4 patients showed increased T cell clonality. In this small subset of patients, we provide preliminary evidence that preoperative cryoablation plus ipilimumab and nivolumab is feasible, inducing systemic adaptive immune activation potentially more robust than cryoablation with/without ipilimumab.

Keywords: Health sciences; Immunology; Oncology.

Graphical abstract

Figure 1

Study schema Combined immune checkpoint blockade (immunotherapy) was administered 1–5 days prior to, and cryoablation (Cryo) was performed 7–10 days prior to standard-of-care surgery. Toxicity evaluation continued for 12 weeks after drug administration. Blood for immune correlates was obtained at baseline, cryoablation, surgery, and 2–4 weeks thereafter (see Table S1 for individual patient timelines). Tumor samples were obtained at cryoablation and surgery.

Figure 2

The combination of ipilimumab, nivolumab, and cryoablation induces T cell activation in the periphery (A) Representative bivariate plots of PD-1 vs. Foxp3 surface expression on CD3⁺CD4⁺ T cells at baseline and 2 weeks post-treatment from a single patient from each treatment arm to identify CD4⁺PD-1^hi T cells (4PD-1^hi). (B) Quantitation of CD4⁺PD-1^hi T cells (top) and the ratio of CD8⁺ to 4PD-1^hi cells (bottom) in each treatment arm. Pre = baseline (pre-treatment), PI = post-immunotherapy, PC = post-cryoablation, PS = post-surgery. Cohort numbers are: cryoablation, n = 7; ipilimumab, n = 6; cryoablation plus ipilimumab, n = 6; cryoablation plus ipilimumab plus nivolumab, n = 5. (C) Heatmaps of expression of T cell activation markers in CD4⁺ T effector (Teff) cells and CD8⁺ T cells in each treatment arm. Data are represented as the average log fold-change (log fold change [FC]) relative to baseline (t = 0) for each time point. (D) Comparison of TNFα and IFNγ serum concentration between patients in each treatment arm. Statistics were calculated using two-way ANOVA (∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001). (E) Serum concentrations of CRP, SAA, TARC, and MCP-4 pooled from 4 patients at baseline and 1, 2, and 6 weeks post-treatment with ipilimumab, nivolumab, and cryoablation. Statistics were calculated using paired t tests: ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.05. For all cytokine data, dotted lines represent lower limit of detection for each cytokine.

Figure 3

T cell receptor (TCR) sequencing analysis of blood samples pre- and post-treatment (A) Frequencies of T cell clones ranked and color-coded, with the most abundant clones in color. (B) Simpson index for each time point. (C) Volcano plots of log₂ fold-change (Fc) versus -log₁₀ p value vs. pre-treatment. Lines through the y axis indicate a change in p value scale. (D) Frequency over time of the clones in the blood that expanded at 2 weeks compared with pre-treatment.