Background: Therapeutic augmentation of the intestinal microbiome to improve immunotherapy outcomes is an active area of investigation. Microbial Ecosystem Therapeutics (METs) are consortia of human-derived bacteria designed to be reproducible, scalable and safe alternatives to fecal transplant. MET4 is a first-in-class consortium of taxa associated with immune checkpoint inhibitor (ICI)-responsiveness. Here we describe preliminary results of MET4-IO, an interventional trial assessing the safety and ecological effects of MET4 in ICI recipients. Methods: MET4-IO is a randomized investigator-initiated trial, evaluating MET4 in solid cancer patients treated with ICI. MET4-IO involves 3 cohorts of 65 total patients: Group A, a safety cohort of 5 patients already on ICI; Group B, patients starting ICI, randomized 3:1 to receive MET4 or not; Group C, patients on ICI who experience radiological progression but not clinical deterioration, randomized 1:1 to receive MET4 or not. Stool and blood samples are collected at baseline and 4-5 additional time-points. For this interim analysis, 16S rRNA gene sequencing was performed on fecal specimens. Shannon diversity, relative abundance (RA), number and fold-change of MET4 taxa > RA 0.01 were assessed and compared to controls. Results: As of January 26, 2020, 21 patients were enrolled (A = 5,B = 12,C = 4), and 15 (71%) received MET4. The mean age was 65.9 years, 40% were females, 52% had head and neck cancer and 19% melanoma. Sixteen patients (76%) were treated with an anti-PD1 agent as monotherapy and 5 with a combination of anti-PD1 and anti-CTLA4 antibodies. G3-4 toxicities (CTCAEv5.0) attributed to ICI were observed in 13% vs. 17% of MET4 exposed and control patients, respectively. Three patients (20%) experienced toxicities attributed to MET4, all grade 1 except G2 dyspepsia in 1 patient. A greater number of MET4-associated taxa were detectable in MET4 recipients than controls (p < 0.01), with a trend towards higher cumulative RA (p = 0.10). No significant change in Shannon diversity after MET4 was observed, however controls were more likely to lose diversity overtime than MET4 recipients (p = 0.05). Colonization with MET4 varied by recipient and by taxon. Bifidobacterium, Collinsella and Enterococcus were significantly more common and abundant in MET4 recipients than controls. Conclusions: In this cohort, MET4 treatment was safe and associated with higher MET4-associated taxa in recipients than controls. Further analyses including peripheral blood immunophenotyping are ongoing. Clinical trial information: NCT03686202.