Endoscopic Healing Overall
The relative ability of infliximab, vedolizumab, adalimumab, and ustekinumab to achieve overall EH, defined using a definition of total SES-CD < 3, was assessed. Results from this analysis are presented in Table 2.
Patients treated with adalimumab [17/61 (27.9%)] and infliximab [39/141 (27.7%)] had the highest rate of one-year EH, followed by ustekinumab [7/41 (17.1%)] and vedolizumab [4/56 (7.1%)]. Compared to vedolizumab-treated patients, adalimumab-treated patients had a greater than fivefold likelihood of achieving one-year EH, which remained significant after adjustment for covariates [aOR: 5.79 (95% CI: 1.77-18.95), p=0.004] (Table 4). Similarly, after adjustment, patients treated with infliximab had a greater likelihood of achieving one-year EH compared to those treated with vedolizumab [aOR: 4.59 (95% CI: 1.48-14.22, p=0.008].
No significant differences were observed between ustekinumab and vedolizumab. Of the 299 patients included in this analysis, a total of 242 patients had large ulcers at baseline. Vedolizumab [2/47 (4.3%)] and ustekinumab [5/34 (14.7%)] demonstrated the lowest rates of one-year EH in patients with large ulcers, while adalimumab [11/51 (21.6%)] and infliximab [27/110 (24.6%)] had higher rates of one-year EH. Compared to vedolizumab, adalimumab demonstrated a significantly greater ability to achieve one-year EH in those with large ulcers at baseline after adjustment for covariates [aOR: 6.45 (95% CI: 1.32-31.38), p=0.021]. Similarly, this was observed among patients treated with infliximab [aOR: 6.04 (95% CI: 1.31-27.82), p=0.021]. To evaluate whether a higher SES-CD enrollment threshold would impact the relative positioning of therapies, sensitivity analyses were performed on the study population restricted to 250 patients with baseline SES-CD ≥ 7, of which 108 were treated with infliximab, 32 with ustekinumab, 54 with adalimumab, and 56 with vedolizumab.
Similar rates of one-year EH were observed in this sensitivity analysis, with infliximab-treated patients achieving the highest rate of one year EH [28/108 (25.9%)],followed closely by adalimumab [12/54 (22.2%)], while ustekinumab [5/32 (15.6%)] and vedolizumab [4/56 (7.1%)] were the worst performing (p=0.035). No difference was observed between ustekinumab and vedolizumab for achieving one-year EH in patients with large ulcers at baseline. Similar trends were observed for those with very large ulcers at baseline (n=90) and also presented in Table 2.
Endoscopic Healing Overall Among Biologic Naïve Patients
Among the 299 participants included in this analysis, a total of 240 participants were biologic naïve.Table 3 demonstrates the one-year endoscopic outcomes of these participants.
Adalimumab demonstrated the highest rate of one-year EH [12/36 (33.3%)], followed by infliximab [39/141 (27.7%)],ustekinumab [5/22 (22.7%)], and vedolizumab [4/41 (9.8%)]. Compared to vedolizumab, again, adalimumab demonstrated a significantly greater ability to achieve one-year EH after adjustment for covariates [aOR: 4.68 (95% CI: 1.34-16.30), p=0.015] (Supplementary Table 1,
http://links.lww.com/AJG/C508). Similar findings were observed for infliximab [aOR: 3.99 (95% CI: 1.27-12.51), p=0.018].
No significant differences were observed between vedolizumab and ustekinumab. Similar trends were observed among biologic naïve patients with very large or large ulcers.
Endoscopic Healing of the Ileum Supplementary Table 2,
http://links.lww.com/AJG/C508 demonstrates the number of patients with ileal involvement at baseline and the proportion achieving endoscopic outcomes at one-year. In total, 184 patients had a SES-CD ≥ 3 in the ileum.
Vedolizumab demonstrated the lowest rate of ileal EH at one-year [8/43 (18.6%)]. Compared to vedolizumab, infliximab demonstrated the highest rate of ileal EH at one-year [29/79 (36.7%)], followed by adalimumab [12/40 (30%)], and ustekinumab [5/22 (22.7%)]. After adjustment for covariates, no significant difference in the rate of ileal EH was observed between vedolizumab and infliximab [aOR: 2.25 (95% CI: 0.85-5.96), p=0.102], adalimumab [aOR: 2.02 (95% CI: 0.69-5.97), p=0.202], or ustekinumab [aOR: 1.81 (95% CI: 0.47-6.91), p=0.389] (Table 4). Sensitivity analysis using the outcome of absence of ileal ulcers were performed on 119 patients with large ileal ulcers at baseline.
Vedolizumab had the lowest rate of absent ileal ulcers at one-year [2/23 (8.7%)].Compared to vedolizumab, infliximab demonstrated the highest rate of absence of ileal ulcers at one-year [20/49 (40.8%), aOR: 5.39 (95% CI: 1.03-28.05), p=0.045]. This was followed by adalimumab [9/30 (30%)] and ustekinumab [3/17 (17.7%)], which were not significant after adjustment for covariates (p=0.077 and p=0.259, respectively). Among 24 participants with very large ulcers at baseline, none of the 7 participants who were treated with vedolizumab achieved absence of ileal ulcers at one-year. In contrast, 3/5 (60%) of infliximab-treated patients and 1/6 (16.7%) of adalimumab-treated patients with very large ileal ulcers had clearance of ileal ulcers.
Endoscopic Healing of the Ileum Among Biologic Naïve Patients
Outcomes of the 145 patients with ileal disease at baseline who were biologic naïve are presented in Supplementary Table 3,http://links.lww.com/AJG/C508. Vedolizumab demonstrated the lowest rate of ileal EH at one-year [7/32 (21.9%)]. Ustekinumab demonstrated the highest rate of ileal EH at one-year[4/10 (40%), aOR: 2.13 (95% CI: 0.43-10.46), p=0.354] followed by adalimumab [9/24 (37.5%), aOR: 1.88 (95% CI: 0.55-6.38), p=0.313], and infliximab [29/79 (36.7%), aOR: 1.96 (95% CI: 0.59-6.45), p=0.270].When biologics among biologic naïve patients with large ileal ulcers were compared, again, vedolizumab had the lowest rate of absence of ileal ulcers at one-year [2/17 (11.8%)]. Adalimumab demonstrated the highest rate of absence of ileal ulcers at one-year [7/17 (42.3%), aOR: 3.75 (95% CI: 0.60-23.30), p=0.156] compared to infliximab [20/49 (40.8%), aOR: 3.10 (95% CI: 0.48-20.04), p=0.234], and ustekinumab [2/7 (28.6%), aOR: 1.92 (95% CI: 0.18-20.24), p=0.589]. Similar trends were observed among the 14 biologic naïve participants with very large ileal ulcers. Once adjusted for covariates, no significant differences were observed between biologics for any of the ileal outcomes above among biologic naïve patients.
Endoscopic Healing of the Colon
Supplementary Table 4,
http://links.lww.com/AJG/C508 demonstrates the outcomes of patients with disease in any of the colonic segments at baseline and the proportion achieving endoscopic outcomes atone-year. In total, 232 patients had at least one colonic segment with a SES-CD ≥ 3.
Ustekinumab had the lowest rate of colonic EH [9/31 (29.0%)]. Compared to ustekinumab, the highest proportion of patients achieving segment EH at one-year were treated with adalimumab [30/48, (62.5%), aOR: 3.97 (95% CI: 1.45-10.90), p=0.007] followed by infliximab [55/105 (52.4%), aOR: 2.08 (95% CI: 0.82-5.27), p=0.121]. There was no significant difference in colonic EH between ustekinumab and vedolizumab [aOR: 1.00 (95% CI: 0.35-2.80), p=0.995]. Among the 176 patients with large colonic ulcers at baseline, ustekinumab again had the lowest rate of absence of colonic ulcers at one-year [8/27 (29.6%)].
Infliximab demonstrated the highest rate of absence of colonic ulcers at one-year [55/78 (70.5%), aOR:3.82 (95% CI: 1.35-10.81), p=0.012] followed by adalimumab (24/37 (64.9%), aOR: 4.27 (95% CI: 1.37-13.31), p=0.012]. There were no significant differences between vedolizumab and ustekinumab [aOR: 1.42 (95% CI: 0.46-4.41), p=0.540] (Table 4). Similar results were obtained when evaluating biologics and absence of colonic ulcers at one-year among patients with very large colonic ulcers at baseline.
Endoscopic Healing of the Colon Among Biologic Naïve Patients
Among the 184 biologic naïve patients with colonic ulcers at baseline,
ustekinumab had the lowest rate of segment EH at one-year [5/17 (29.4%)]. Adalimumab demonstrated the highest rate of segment EH at one-year [18/27 (66.7%), aOR: 4.96 (95% CI: 1.29-19.03), p=0.020], followed by infliximab [55/105 (52.4%), aOR: 2.71 (95% CI: 0.89-8.27), p=0.081], and vedolizumab [13/35 (37.1%), aOR: 1.48 (95% CI: 0.41-5.38), p=0.548] (Supplementary Table 1/5,
http://links.lww.com/AJG/C508). Sensitivity analyses among those with large colonic ulcers at baseline (n=133) using the outcome of absence of colonic ulcers at one-year demonstrated that again ustekinumab had the lowest rate of absence of colonic ulcers at one-year [5/14 (35.7%)].
Infliximab had the highest rate of absence of colonic ulcers at one-year [55/78 (70.5%), aOR: 4.73 (95% CI: 1.40-15.94), p=0.012], followed by adalimumab [13/19 (68.4%), aOR: 4.33 (95% CI: 0.96-19.55), p=0.057], and vedolizumab [12/22 (54.6%), aOR: 2.47 (95% CI: 0.59-10.27), p=0.213]. Among the 51 patients with very large colonic ulcers at baseline, compared to ustekinumab [2/9 (22.2%)], adalimumab [7/9 (77.8%)] demonstrated the highest rate of absence of colonic ulcers at one-year [aOR: 18.53 (95% CI: 1.68-204.51), p=0.017].
After adjustment for known confounders, infliximab [19/25 (76.0%)] also demonstrated a significantly higher rate of absence of colonic ulcers at one-year compared to ustekinumab [aOR: 15.71 (95% CI: 2.26-109.12), p=0.005]. No significant difference was observed when ustekinumab and vedolizumab were compared.
Endoscopic Healing of the Individual Colonic Segments
Endoscopic outcomes across individual colonic segments and among patients who were biologic naïve were also evaluated.
Overall, adalimumab demonstrated the highest rate of segment EH in the ascending colon, transverse colon, and descending colon, while infliximab had the highest rate of segment EH in the rectum. Among patients who were biologic-naïve, adalimumab demonstrated the greatest ability to achieve segment EH across all colonic segments. Among patients with large ulcers, adalimumab continued to demonstrate the highest rate of absence of ulcers in the ascending colon [5/11 (45.5%)], transverse colon [6/14 (42.9%)], and descending colon [14/25 (56.0%)], although no significant differences were observed after adjustment for covariates. Similarly, infliximab continued to demonstrate the highest rate of absence of ulcers in the rectum among patients with large rectal ulcers at baseline, although this was not statistically significant compared to the worst performing biologic, ustekinumab [12/28 (42.9%) vs. 0/18 (0%)].