Back

A probiotic bacterium modulates antitumor γδ T-cell responses in lung cancer

Goto, Y.; Dolton, G.; Thomas, H.; Morin, T.; Tajima, Y.; Imamura, K.; Sakata, S.; Oka, K.; Hayashi, A.; Takahashi, M.; Ueno, T.; Sakagami, T.; Tomita, Y.; Sewell, A. K.; Motozono, C.

2025-07-19 immunology
10.1101/2025.07.16.664923 bioRxiv
Show abstract

The link between the intestinal microflora and cancer outcomes has been recognized for over a decade. Several recent studies have demonstrated that the gut microbiome is associated with the efficiency of T-cell checkpoint blockade therapy for cancer raising interest in strategies to harness this effect via consumption of live microorganisms (probiotics). The probiotic Clostridium butyricum strain MIYAIRI 588 (CBM588) enhances both response rates and overall survival in patients receiving immune checkpoint blockade therapy for non-small cell lung cancer and metastatic renal cell carcinoma but the mechanism underpinning this benefit remains unclear. Here, we show that CBM588 spores induce a population of V{gamma}9V{delta}2 T-cells from the peripheral blood of healthy donors and lung cancer patients. A subset of these T-cells responded to, and directly lysed, cancer cell lines via a butyrophilin 3A-dependant mechanism. In patients taking CBM588 alongside checkpoint blockade, peripheral V{delta}2 T-cells expressed the activation marker CD69 more frequently than those receiving checkpoint blockade alone and the frequency of V{delta}2CD69 cells increased following initiation of CBM588 treatment (p = 0.0041). Our findings provide a potential mechanism by which the manipulation of the intestinal microflora might directly contribute to cancer prognosis by activating immune effector cells with intrinsic anticancer properties.

Matching journals

The top 9 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.