PF757 REDUCED INTENSITY CONDITIONING AND AVOIDING USE OF BROAD SPECTRUM ANTIBIOTHERAPY ACTIVE AGAINST GRAM‐NEGATIVE BACTERIA MIGHT HAVE A CUMULATIVE IMPACT IN REDUCING INTESTINAL MICROBIOTA MISBALANCE

Background: Intestinal microbiota misbalance of allo‐HCT recipients is characterized by loss of alfa‐diversity and domination of microbial communities by single organisms. These injury patterns have been associated with poor survival and might be attributable to regimen‐related mucosal injury, antibiotic exposure and nutritional alterations, between others. We hypothesized that potential sources of microbiome insult may vary with time and that conditioning related toxicity and specific antibiotic pressure may have a sequential and possibly cumulative impact, leading to poor survival trough increase rate and severity of acute graft versus host disease. Aims: to perform an exploratory preliminary analysis aiming at identifying possible trends in microbiota misbalance after 1) myeloablative vs. reduced conditioning and 2) broad‐spectrum antibiotherapy active against Gram‐negative bacteria vs. not receiving it. Methods: We prospectively recruited a multicenter cohort of 24 consecutive recipients of allo‐HCT after signing informed consent form from January through September 2018. We performed 16S rRNA gene profiling of 87 stool samples collected at the following time points: pre‐conditioning (sample1), day0 (sple2), day+7 (sple3), day of fever whenever it occurred (sample 4/inconstant), day of antibiotic discontinuation (sple5). Most frequent diagnosis was AML (41%) and variables of interest are summarized in table 1. Phylogenetic diversity analyses were performed using Qiime2 and in‐house R scripts. No statistical comparison between groups was performed due to small sample size and operational taxonomic units (OUT) in each sample were obtained by using Silva database. Secondary analyses were performed for calculating Shannon's and inverse Simpsońs diversity indexes for alpha and beta diversity. Patients received antibiotherapy per local policy along with other usual standards of care. Results: Median follow‐up 9 months (4–13). Six patients died (3 from relapse, 1 from TRM, 2 from other causes). Five patients had grade III‐IV aGVHD. Seven patients received broad‐spectrum antibiotherapy against Gram‐negative bacteria. Following trends were identified: a) patients receiving myeloablative conditioning (MAC) had greater reduction in microbiota diversity between sample 1 (pre‐conditioning) and sample 2 (day 0) vs. patients receiving reduced intensity conditioning (RIC) (figure, higher panel). b) patients receiving broad‐spectrum antibiotherapy against Gram‐negative bacteria had greater reduction in microbiota diversity between sample 4 (day of starting antibiotherapy) and sample 5 (day of antibiotherapy withdrawal) vs. patients that received reduced spectrum antibiotherapy or no antibiotherapy (figure, medium panel). c) patients developing grade III‐IV aGVHD had greater reduction in microbiota diversity for each collection time‐point vs. patients having grade I‐II aGVHD or no aGVHD. Summary/Conclusion: 1) This exploratory preliminary analysis showed in a small prospective multicenter cohort of adult recipients of allo‐HCT that MAC and use of broad spectrum antibiotherapy active against Gram‐negative bacteria might have a sequential cumulative impact in reducing the diversity of intestinal microbiota. image 2) This hypothesis merits further testing in a bigger cohort of patients. Our clinical research group is carrying out this study and data will be updated. 3) If the hypothesis is confirmed this could be used for designing co‐ordinated strategies targeting at the same time conditioning and antibiotherapy in selected individual patients.