They say:

> Outperforming chemotherapy and immunotherapy

And then later:

> As a facultative anaerobe, it preferentially accumulates within the hypoxic tumor microenvironment, where it rapidly proliferates and exerts direct cytotoxic effects while simultaneously activating a broad immune response. Within hours, tumors become infiltrated with T cells, B cells, and neutrophils, accompanied by surges in key inflammatory cytokines like TNF-α and IFN-γ.

So this is immunotherapy. Although it is clever immunotherapy. Gut bacteria doesn't usually survive long in the bloodstream because there's too much oxygen present (that's part of why it's gut bacteria, its unlikely to go all Leeroy Jenkins on the rest of the organism).

The TME is often so densely packed with growth that its less oxidative than surrounding tissue. So the bacteria that don't find a tumor don't last long enough to cause problems and the ones that do find one see it as a bit of a refuge from the problematic environment and colonize it specifically, throwing off whatever subterfuge the tumor was using to keep the immune system from getting involved.

It's a bit like throwing a brick through the window of a bank that was being quietly robbed by somebody else. The cops show up and realize they've been overlooking a separate problem.