Cancer immunotherapy is revolutionizing cancer care, moving steadily into earlier lines of treatment and previously unresponsive cancer types. But while it can be dramatically effective, not all patients respond and immunotherapy can cause significant toxicity. Practical and reliable methods to select patients for safe and effective immunotherapy remain lacking. The influence of commensal microbiota on systemic and anti-tumour immunity is an under-recognised modulator of cancer development and treatment outcomes with relevance to clinical practice. From allogeneic stem cell transplant recipients to advanced solid cancer patients, we now know that gut microbiota influence immunotherapy response rates, survival outcomes, and toxicities [1]. Recent data supports a clear association of gut microbes with anti-cancer immunotherapy responses at both the community level and for specific taxa (e.g. Faecalibacterium, Akkermansia, and Bifidobacterium – anti-PD-(L)1 favourable; and Bacteroidales – anti-PD-1 unfavourable but some species anti-CTLA-4 ± anti-PD-1 favourable) [2-5]. Gut microbiota are also closely linked to the development of significant treatment-related toxicities such as intestinal/colonic inflammation, being particularly associated with abundance of Bacteroides intestinalis in the faecal microbiome [6]. Ongoing efforts to understand the immunological mechanisms linking microbes in the gut or other organ sites – including tumours – will need to integrate tumour genomic factors, immune state, and microbial functionality into more comprehensive biomarker-driven models of tumour immunology. Therapeutically, microbial manipulation via administration of defined bacterial products or metabolites, microbial transplantation, or dietary modification, are all under investigation as adjuncts to immunotherapy. Microbial-based manipulation thus holds promise as the next layer of personalised cancer therapy.