Randomised Clinical Trial

Adjunctive Induction Therapy With Oral Effervescent Budesonide in Newly Diagnosed Coeliac Disease

Evan D. Newnham; Daniel Clayton-Chubb; Meena Nagarethinam; Patrick Hosking; Peter R. Gibson


Aliment Pharmacol Ther. 2021;54(4):419-428. 

In This Article



A total of 37 participants were randomised to budesonide (n = 19) or placebo (n = 18). The study was closed prior to reaching the planned 40 patients due to loss of funding. The flow of patients is shown in Figure 1. Five patients did not attend for repeat duodenal biopsies at 8 weeks (two budesonide and three placebo) and those five plus an additional subject in the budesonide group did not undergo the 52-week biopsies. Participant groups were well balanced in terms of baseline demographic commitments and disease characteristics (Table 1).

Figure 1.

Patient flow

Study Compliance

Overall compliance to budesonide was 82% and to placebo 88%. Adherence to the gluten-free diet was 'excellent' in 22 (12 receiving budesonide), 'good' in 10 (5 budesonide) at weeks 4–8. In five patients, food diaries were not completed (two receiving budesonide). Participants had undertaken a diagnostic gastroscopy a mean (SD) of 5.2 (1.7) weeks and commenced a gluten-free diet 3.7 (1.2) weeks prior to commencing treatment.

Healing Rates

Mucosal response (Marsh 0 or 1) was observed in 12 patients (32%) at week 8 and 20 of 37 patients (54%) at week 52. Mucosal remission was noted in nine patients at week 8 (24%) and 14 (38%) at week 52. As shown in Figure 2, there were no significant differences when stratified by treatment arms in week-8 response (37% budesonide vs 28% placebo; P = 0.73) or remission (32% vs 17%; P = 0.45); similarly, there were no differences in week-52 rates of response (63% vs 44%; P = 0.33) or remission (42% vs 33%; P = 0.74). There were no statistically significant differences between mean villous-height : crypt-depth ratios in the two groups at baseline, at week 8 (budesonide: 2.26 ± 0.97 vs placebo: 1.93 ± 0.97) and at week 52 (2.59 ± 1.42 vs 2.29 ± 1.13). The change in villous-height : crypt-depth from baseline observed in those having duodenal biopsies at weeks 8 and 52 is illustrated in Figure 3.

Figure 2.

Marsh scores in patients at 8 and 52 weeks according to treatment group. The patients not having biopsies (but included in the intention-to-treat analysis) shown in light grey as the last-observation-carried forward (LOCF) cohort

Figure 3.

Change in villous height : crypt depth (Vh:Cd) ratio after 8 and 52 weeks relative to that at baseline according to the treatment group. Data are shown for patients who had histology at 8 and 52 weeks. The mean change from baseline at 8 weeks was 1.58 (95% confidence intervals 1.07, 2.09) for budesonide group compared with 1.64 (1.00, 2.27) for placebo, and at 52 weeks, 2.26 (1.75, 2.77) compared with 2.01 (1.44, 2.58), respectively

Other Clinical End-points

The results for other main clinical end-points are shown in Table 2. Overall, gastrointestinal symptoms tended to improve at 8 weeks but not 52 weeks, with no differences between the treatment groups. The proportion of patients who were satisfied with their symptoms were similar in the budesonide and placebo groups at 8 and 52 weeks. Fatigue Impact Scores improved over time in each group with no differences between treatment groups. Hospital Anxiety and Depression Scores also improved over time, but there were no differences between the groups after 8 weeks. However, as shown in Figure 4, total (P = 0.022), anxiety (P = 0.042) and depression scores (P = 0.013) were all lower in the budesonide-treated groups at 52 weeks, but none reached statistical significance when the p-value was corrected for multiple comparisons.

Figure 4.

Total, anxiety and depression scores for the Hospital Anxiety and Depression Scale at baseline and after 8 and 52 weeks in the budesonide and placebo treated groups. The bars shown represent the medians. Last observation was carried forward for missing data. At 52 weeks, scores were lower in the budesonide-treated than placebo-treated group for total score (P = 0.022), anxiety (P = 0.04) and depression (P = 0.0049). Only the score for depression was lower after Bonferroni correction

Safety and Adverse Events

As shown in Table 3, the safety profile of oral effervescent budesonide was acceptable, with no significant differences in either treatment-related adverse events or post-treatment adverse events compared with those in the placebo-treated patients. The only serious adverse event was a pregnancy in the post-treatment period. The majority of adverse events were judged as of mild intensity, and none were of severe intensity. All events were judged as recovered or recovering except for persisting fibromyalgia that emerged in a budesonide-treated patient in the post-treatment period. Causality of adverse events was judged as probable in one patient with vulvovaginal candidiasis while receiving budesonide.

No clinically relevant laboratory abnormalities were recorded (data not shown). Specifically, fasting serum cortisol levels and fasting blood sugar levels were within normal limits in all subjects at all time-points. There was no signal of deterioration of bone density on dual-energy X-ray absorptiometry. Thus, mean (SD) bone mineral density in the budesonide-treated group at the femoral neck was 0.98 (0.14) g/cm2 at baseline and 1.00 (0.12) g/cm2 at 52 weeks, and 1.17 (0.16) and 1.17 (0.14) g/cm2 at the lumbar spine, respectively. In the placebo-treated group, the respective bone mineral densities were 1.00 (0.16) and 1.04 (0.18) g/cm2 at the femoral neck and 1.23 (0.23) and 1.24 (0.14) g/cm2 at the lumbar spine.

Post hoc Analysis of Associations With Early Mucosal Remission

Given the lack of difference in healing rates between the treatment arms, the baseline data for patients who had mucosal remission at 8 weeks (n = 9) were pooled and compared with those without mucosal remission (n = 28) in a post hoc analysis. Dietary compliance was rated similarly between the two groups (67% vs 70% excellent, respectively). Baseline demographic and clinical measures of the two groups are shown in Table 4. Those achieving mucosal remission at 8 weeks had less severe histopathological lesions, whether rated via the Marsh score (Table 4) or the villous-height : crypt-depth ratio (Figure 5). Those not achieving remission tended to be more fatigued (P = 0.20) and depressed (P = 0.052) at baseline assessment.

Figure 5.

Post hoc analysis of villous height : crypt depth ratio (Vh:Cd), fatigue scored by the Fatigue Impact Scale, gastrointestinal (GI) symptoms on a 100-mm visual analogue scale and depression scored by the Hospital Anxiety and Depression Scale during the baseline assessment of patients according to whether they achieved mucosal remission ('healed') at 8 weeks or not ('unhealed'), irrespective of what experimental treatment they had. The villous height : crypt depth ratio was higher in patients who achieved mucosal remission at 8 weeks (P = 0.0145; t test)