Combinations of antibiotics have shown promise in off-label treatment of inflammatory bowel disease (IBD) flares in adult and pediatric patients (1–6). In a small pediatric cohort of moderate-severe medically refractory ulcerative colitis (UC), half achieved remission with a regimen of 3 to 4 antibiotics (6). The rationale is to decrease mucosal inflammation by altering intestinal microbiota; however, to date no study has documented these changes. We report our experience with combination oral antibiotics in symptomatic pediatric patients with steroid-dependent or steroid-resistant UC, and demonstrate significant alterations in gut microbiota during and after antibiotic therapy with reductions in inflammatory markers.
We describe a retrospective review of pediatric patients with histology-confirmed UC, steroid dependence or resistance, and incomplete response to biologics treated with combination oral antibiotics at Children's Hospital Los Angeles (CHLA) between July 2016 and June 2017 (institutional review board [IRB] CHLA-17–00269). Antibiotics were selected by physician discretion based on Turner et al (6) for intended minimum duration of 2 to 3 weeks depending on Pediatric Ulcerative Colitis Activity Index (PUCAI) response. The oral antibiotics included Amoxicillin (50 mg/kg divided tid, max 500 mg tid), Metronidazole (5 mg/kg tid, max 250 mg tid), Doxycycline (2 mg/kg bid, max 100 mg bid) for ages over 7, and Vancomycin (250 mg QID) if hospitalized. Fluconazole (200 mg qd) was added for a few patients per physician discretion. Oral Gentamicin (2.5 mg/kg tid) replaced any allergenic drug. Standard of care medications continued concomitantly.
For some patients who enrolled in the institution's biorepository and provided biological samples during clinical visits, targeted 16S rRNA sequencing of the V4 region using Illumina MiSeq was used to characterize oral and intestinal microbiota (IRB CHLA-15–00395). Available PUCAI and clinical labs were evaluated at baseline, during and after antibiotics up to 6 months. Clinically significant response of PUCAI (change ≥ 20), inflammatory markers (C-reactive protein [CRP], erythrocyte sedimentation rate [ESR], white blood cells [WBC], calprotectin), hemoglobin, albumin, and remission (PUCAI ≤ 10) were assessed. Mixed effects regression was applied to outcome parameters as a function of time compared to pre-treatment baseline while controlling for the individual as a random effect. All P values were adjusted for multiple comparisons using the Benjamini-Hochberg method. Alpha diversity was calculated using the Shannon Diversity Index, and pairwise comparisons were performed using Wilcoxon rank-sum test in R. The log ratio of fecal Bacteroidetes/Firmicutes abundances was compared using Kruskal-Wallis test in R.
Eight steroid-dependent or refractory UC patients ages 8 to 20 years (mean 14.6), 38% female, most with extensive colitis (n = 6) with mean pretreatment PUCAI of 60 (25–80), and history of poor response to multiple immunosuppressants including biologics were included in this analysis (Table 1). Duration of antibiotics ranged from 6 to 84 days (median 14). Two patients who had complete response received an extended duration since they had been medically refractory to other therapies and physicians and parents believed a prolonged course would maintain remission. In 88% (N = 7), the PUCAI significantly declined at all timepoints (P < 0.001) achieving PUCAI drop ≥ 20 (mean change 40) with 3 (38%) of these patients achieving PUCAI ≤ 10 at end of antibiotics (Fig. 1A). One patient with severe refractory disease had a negligible response after 7 days (PUCAI 75 from 80), and subsequently underwent elective colectomy.
Two of the 8 patients had complete enduring response (PUCAI 0) and remained steroid-free 6 months post-antibiotics. One of these patients de-escalated from Infliximab. A third patient later developed a flare 5 months after responding to antibiotics (PUCAI 60 to 30), and again improved after a second round of combination antibiotics (PUCAI 70 to 15). Two other patients had a relatively short follow-up period but demonstrated significant response (PUCAI of 20 to 25 from 60) at 2 months after antibiotics. Two patients relapsed quickly after antibiotics completion. Specifically, one received 6 days of antibiotics and achieved remission (PUCAI 75 to 10) before a scheduled research fecal transplant (FMT) for UC (IRB CHLA-16–00050, IND 17083). This patient developed moderate disease 7 days after antibiotics/FMT, and 6 months later is awaiting colectomy. Another steroid-dependent, Infliximab primary non-responder relapsed within 2 days post-antibiotics after responding to 21 days of therapy (PUCAI 65 to 35) with concomitant maintenance steroid wean. Three months later, this patient underwent colectomy after not responding to Vedolizumab. Overall antibiotics were well tolerated, except for one episode of metronidazole-responsive Clostridium difficile (CDI) infection 3.5 weeks after antibiotics in a patient with history of recurrent CDI.
Markers of inflammation decreased in most subjects (Fig. 1). Albumin was significantly increased at 6 months (P = 0.02) (Fig. 1B). CRP, ESR, WBC, and calprotectin levels decreased, but were not significant (Fig. 1C). Microbial alpha diversity significantly increased (P = 0.0037) post-antibiotics compared to during therapy in which a monolithic microbial profile emerged (Fig. 2A). Oral microbiome has higher alpha diversity with different bacterial families than corresponding fecal communities during therapy. The shifts in microbial populations led to differences in Bacteroidetes/Firmicutes log ratio along the course of therapy (Fig. 2B). Bacteroidaceae, Enterobacteriaceae, and Pasteurellaceae initially reappeared in the first 2 weeks post-antibiotics. Over time, Corynbacteriaceae, Erysipelotrichaceae, and Prevotellaceae appeared 6 months later (Fig. 2C).
Intestinal dysbiosis contributing to disease activity may be responsive to broad-spectrum antibiotics in select patients with UC. Overall, 7/8 steroid dependent or resistant UC patients with active disease treated with oral antibiotics had a reduction of PUCAI scores greater than 20 points (P < 0.001). PUCAI ≤10 was achieved in 3/8 (38%) at the end of antibiotics. Two of these patients had sustained steroid-free remission (PUCAI 0) 6 months post-antibiotics, one of whom was able de-escalate from Infliximab. In all, 4/7 responders discontinued steroids in the months post-antibiotics. Two initial responders relapsed within a week post-antibiotics, though one received FMT shortly after discontinuation and another was concomitantly weaning maintenance steroids. Antibiotics were well tolerated without allergic or other reactions, but one patient with recurrent CDI developed C difficile post-antibiotics.
Markers of inflammation WBC, CRP, ESR and calprotectin decreased, while hemoglobin and albumin increased. Due to small sample size and sparse sampling, only changes in albumin achieved statistical significance (P = 0.02) possibly secondary to improved nutrition and decreased gut loss. These clinical findings align with prior reports utilizing a similar antibiotic regimen (6).
Microbial sequencing reveals dramatic monolithic microbiota change during broad-spectrum antibiotics as well as a significant increase in alpha diversity post-therapy. Gut microbial patterns months after antibiotics, however, lead to expansion of potentially immunogenic bacteria such as Prevotellaceae typically seen in UC and implicated in Th17-mediated mucosal inflammation (7). Additionally, Erysipelotrichaceae is a colitogenic bacteria in IBD (1) and positively correlates with tumor necrosis factor levels (8).
In summary, broad-spectrum antibiotics may provide a safe and relatively fast-acting, steroid-sparing therapeutic option for short-term control of active disease in refractory pediatric UC patients. Our data support the hypothesis that antibiotic alteration of microbial communities result in decreased inflammation, however, potentially immunogenic bacteria repopulate in some individuals post-antibiotics. These observations merit an adequately powered randomized clinical trial to define the value of this approach and identify patients who would derive the most benefit.
The authors would like to thank the CHLA Division of Pediatric Gastroenterology and most importantly the families and patients.
1. Sato K, Chiba T, Ohkusa T. Serial changes of cytokines in active ulcerative colitis: effects of antibiotic combination therapy. Hepatogastroenterology
2. Terao S, Yamashiro K, Tamura I, et al. Antibiotic combination therapy for steroid withdrawal in steroid-dependent ulcerative colitis. Digestion
3. Gionchetti P, Rizzello F, Ferrieri A, et al. Rifaximin in patients with moderate or severe ulcerative colitis refractory to steroid-treatment: a double-blind, placebo-controlled trial. Dig Dis Sci
4. Turunen UM, Farkkila MA, Hakala K, et al. Long-term treatment of ulcerative colitis with ciprofloxacin: a prospective, double-blind, placebo-controlled study. Gastroenterology
5. Dickinson RJ, O’Connor HJ, Pinder I, et al. Double blind controlled trial of oral vancomycin as adjunctive treatment in acute exacerbations of idiopathic colitis. Gut
6. Turner D, Levine A, Kolho KL, et al. Combination of oral antibiotics may be effective in severe pediatric ulcerative colitis: a preliminary report. J Crohns Colitis
7. Larsen JM. The immune response to Prevotella bacteria in chronic inflammatory disease. Immunology
8. Dinh DM, Volpe GE, Duffalo C, et al. Intestinal microbiota, microbial translocation, and systemic inflammation in chronic HIV infection. J Infect Dis