Eosinophils are critical for response to immune checkpoint blockade in breast cancer and other cancer types


Year:

Session type:

Theme:

Hannah Garner, Olga Blomberg, Lorenzo Spagnuolo, Leonie Voorwerk, Ewald van Dyk, Myriam Chalabi, Olga Isaeva, Noor Bakker, Kelly Kersten, Chris Klaver, Maxime Duijst, Dorien Pastoors, Cheei-Sing Hau, Kim Vrijland, Elisabeth Raeven, Louisa Hoes, Willemijn Theelen, Paul Baas, Emile Voest, Lodewyk Wessels, Karin de Visser, Marleen Kok

Abstract

Background

Immune checkpoint blockade (ICB) fundamentally changed the treatment landscape of cancer. However, only a minority of advanced breast cancer patients benefit from ICB. There is an urgent need to improve our understanding of the immune mechanisms that underlie ICB response. By combining comprehensive profiling of circulating immune cells from ICB-treated patients with mechanistic mouse studies, we aimed to study immune parameters associated with response to ICB.

Method

Blood from metastatic triple negative breast cancer (TNBC) patients in the phase 2 TONIC-trial (NCT02499367, n=111) was analysed before and during anti-PD1 treatment by multiparameter flow cytometry. In parallel, we used genetically engineered breast cancer mouse models for mechanistic studies. We validated our findings in other patient cohorts treated with ICB: metastatic non-small cell lung cancer (NSCLC; n=55), early-stage colon cancer (CC; mismatch repair proficient (pMMR) n=17 and MMR-deficient (dMMR) n=21) and metastatic dMMR cancer (n=11). RNA-sequencing analysis was performed on biopsies from the TNBC and CC cohorts.


Results

A significant increase in circulating eosinophils was observed in TNBC patients responding to ICB (p=0.002), but not in non-responders. In line with this, circulating and intratumoral eosinophils in breast cancer mouse models responsive to ICB with cisplatin, were increased during ICB treatment. Importantly, ICB response in these preclinical models was abrogated upon depletion of eosinophils. Moreover, increased systemic eosinophil counts was associated with ICB-response in NSCLC (p=0.03) and in early-stage pMMR CC (p=0.04) but not in early-stage or metastatic dMMR tumours. Additionally, expression of eosinophil-related genes increased during ICB within tumours from metastatic TNBC and early-stage CC.

Conclusion

An increase in systemic eosinophils is associated with ICB response in multiple tumour types. Mouse models reveal an unexpected causal role for eosinophils during ICB response, indicating that eosinophils are mechanistically involved in response to ICB and exposing novel possibilities for designing therapeutic strategies that increase ICB efficiency by engaging eosinophils.

Impact statement

By revealing a causal role for eosinophils during response to ICB we uncover unexplored avenues for future research into treatment strategies that activate and mobilise eosinophils in combination with ICB to enhance response to ICB and induce durable responses in a larger fraction of patients.