Imaging Memory T-Cells Stratifies Response to Adjuvant Metformin Combined with αPD-1 Therapy

Abstract

The low response rates associated with immune checkpoint inhibitor (ICI) use has led to a surge in research investigating adjuvant combination strategies in an attempt to enhance efficacy. Repurposing existing drugs as adjuvants accelerates the pace of cancer immune therapy research; however, many combinations exacerbate the immunogenic response elicited by ICIs and can lead to adverse immune-related events. Metformin, a widely used type 2 diabetes drug is an ideal candidate to repurpose as it has a good safety profile and studies suggest that metformin can modulate the tumour microenvironment, promoting a favourable environment for T cell activation but has no direct action on T cell activation on its own. In the current study we used PET imaging with [¹⁸F]AlF-NOTA-KCNA3P, a radiopharmaceutical specifically targeting K_V1.3 the potassium channel over-expressed on active effector memory T-cells, to determine whether combining PD1 with metformin leads to an enhanced immunological memory response in a preclinical colorectal cancer model. Flow cytometry was used to assess which immune cell populations infiltrate the tumours in response to the treatment combination. Imaging with [¹⁸F]AlF-NOTA-KCNA3P demonstrated that adjuvant metformin significantly improved anti-PD1 efficacy and led to a robust anti-tumour immunological memory response in a syngeneic colon cancer model through changes in tumour infiltrating effector memory T-cells.

Keywords: immune checkpoint inhibitors (ICI); metformin; positron emission tomography (PET); potassium channels.

Figure 1

(A) Schematic showing the treatment, tumour volume assessment and imaging regimen. (B) Individual tumour volumes for each animals highlighting the variability in response. (C) Tumour volumes in each treatment cohort after therapy response stratification. (D) Average tumour volume after tumour re-challenge. Data are represented as mean ± S.D. (TR, responding tumours; TNR, non-responding tumours).

Figure 2

(A) Representative MIP images indicating tumour retention of [¹⁸F]AlF-NOTA-KCNA3P across the treatment cohorts. Tumour borders are shown as yellow dotted lines. (B) Bar graph indicating tumour retention of [¹⁸F]AlF-NOTA-KCNA3P in each treatment cohort (Control, αPD1, metformin, combined αPD1 + metformin and TNRs; n = 5–10 mice/group; ** p < 0.01, *** p < 0.001 compared to TNR, ^$ p < 0.05 compared to αPD1; data shown as the mean %ID/g ± S.E.M.). (C) Retention of [¹⁸F]AlF-NOTA-KCNA3P in individual tumours from TR and TNRs (**** p < 0.0001).

Figure 3

Tumour infiltrating immune cell populations determined using FACS across each treatment cohort (control, αPD1, metformin, combined αPD1 + metformin and TNRs). Data shown as (A) CD3⁺ cells as a % of total CD45⁺ cells, (B) CD8⁺ T_EM cells as a % of total CD8⁺ cells, (C) CD4⁺ T_EM cells as a % of total CD4⁺ cells, (D) K_V1.3⁺ T_EM cells as a % of total CD3⁺ cells, and (E) F4/80⁺ cells as a % of total CD45⁺ cells. Data indicated are individual values with mean ± S.D. representative of n = 5–10 mice/cohort. * p < 0.05; ** p < 0.01 compared to TNR.