Clinical outcomes of upadacitinib dose escalation in inflammatory bowel disease: A single-center retrospective cohort study

Abstract

BACKGROUND

Upadacitinib is a selective Janus kinase-1 (JAK1) inhibitor approved for the treatment of moderate-to-severe ulcerative colitis (UC) and Crohn’s disease (CD). Standard therapy consists of induction dosing at 45 mg daily followed by maintenance dosing of 15-30 mg daily. Although pivotal trials demonstrated sustained efficacy with maintenance therapy, secondary loss of response occurs in clinical practice. Data regarding the effectiveness of dose re-escalation remain limited.

AIM

To evaluate clinical outcomes following upadacitinib dose escalation in inflammatory bowel disease (IBD) with loss of response to maintenance dosing.

METHODS

We conducted a single-center retrospective cohort study of adults with IBD who underwent re-escalation to upadacitinib 45 mg daily beyond the labeled induction period. Data were obtained through electronic medical record review. The primary endpoint was clinical response, defined as provider-documented symptom improvement and/or objective improvement in biomarkers or endoscopic assessment. Secondary outcomes included indications for escalation, corticosteroid exposure, IBD-related healthcare utilization, and treatment durability.

RESULTS

A total of 56 patients were included [38 CD, 18 UC; median age 34 years (interquartile range: 26-46), 43% female] with a mean of 2.49 prior biologic exposures. Among 49 patients with paired symptom data, abdominal pain decreased from 57% to 29% (P = 0.003), hematochezia from 49% to 20% (P = 0.004), urgency from 45% to 16% (P = 0.003), and diarrhea from 55% to 37% (P = 0.049). Corticosteroid use declined from 50% to 32% (P = 0.031). The proportion with ≥ 1 IBD-related hospitalization or emergency department visit decreased from 32% to 20%. Biomarkers and endoscopic outcomes improved numerically but were limited by small paired samples. At most recent follow-up, 89% of patients remained on escalated dosing.

CONCLUSION

In this single-center cohort, upadacitinib dose escalation improves symptoms and reduces corticosteroid use, supporting dose intensification as a pragmatic strategy for secondary loss of response.

Key Words: Upadacitinib; Janus kinase 1 inhibitor; Inflammatory bowel disease; Crohn’s disease; Ulcerative colitis; Dose intensification; Secondary loss of response; Real-world evidence; Treatment durability; Corticosteroid reduction

Core Tip: Real-world data guiding dose re-escalation of upadacitinib in inflammatory bowel disease (IBD) are limited. In this single-center cohort, off-label intensification in patients with secondary loss of response was associated with significant symptomatic improvement and reduced corticosteroid exposure, with favorable trends in objective measures. Most patients remained on the escalated dose at follow-up. No new laboratory safety signals were observed. These findings suggest dose intensification may represent a pragmatic management strategy in selected patients with IBD.

INTRODUCTION

Upadacitinib is an oral small-molecule Janus kinase (JAK) inhibitor that selectively targets JAK1, with additional inhibitory activity against JAK2, JAK3, and tyrosine kinase 2[1]. It is approved for the treatment of multiple autoimmune diseases, including atopic dermatitis, rheumatoid arthritis, and ankylosing spondylitis. Upadacitinib was approved by the United States Food and Drug Administration (FDA) in March 2022 for ulcerative colitis (UC) and in May 2023 for moderate-to-severe Crohn’s disease (CD)[2,3]. As an oral JAK1 inhibitor, it represents the first targeted oral therapy available for CD and one of the few advanced oral options for UC, providing an alternative to infusion-based and injectable therapies[2,3].

The phase 3 clinical trial program established its efficacy in both induction and maintenance settings. In the U-ACHIEVE (UC1) and U-ACCOMPLISH (UC2) induction trials for UC, 45 mg daily achieved clinical remission in 26% and 33% of patients, respectively, compared to 4%-5% with placebo[1,4]. In the U-ACHIEVE maintenance study, both 15 mg and 30 mg daily doses were superior to placebo at maintaining remission at 52 weeks[1,4]. Similar findings were observed in phase 3 trials for CD, where the U-EXCEL and U-EXCEED induction trials and the U-ENDURE maintenance study demonstrated superiority over placebo in both induction and maintenance outcomes[5].

Current FDA labeling restricts induction dosing of 45 mg to 8 weeks for UC and 12 weeks for CD, after which patients transition to maintenance dosing of either 15 mg or 30 mg daily. In real-world practice, however, a subset of patients experience a secondary loss of response on maintenance dosing. Dose escalation is an established strategy for other advanced therapies in inflammatory bowel disease (IBD), including anti-tumor necrosis factor agents, vedolizumab, and ustekinumab. Tofacitinib, another JAK inhibitor approved for UC, is also associated with secondary loss of response in some patients. After an 8-16 weeks induction phase, dosing is typically reduced; however, real-world data suggest that de-escalation may lead to disease worsening. In one cohort, more than one-half of patients experienced increased disease activity after dose reduction, and nearly 40% did not regain clinical response despite re-escalation to induction dosing[6]. These observations underscore the uncertainty surrounding optimal JAK inhibitor dosing.

Only one other study has examined high-dose upadacitinib in patients with IBD who experienced a loss of response to standard maintenance dosing. This study specifically investigated the effectiveness and safety of reinduction with upadacitinib 45 mg daily in patients with IBD. Findings indicate that reinduction successfully recaptured clinical response in more than 80% of patients and remission was best sustained by continuing the 45 mg dose indefinitely. The authors concluded that the intensified dosing strategy was effective and appeared safe for patients who had relapsed on the lower maintenance dose[7].

To build upon this limited evidence, we conducted a retrospective, single-center cohort study to characterize patient demographics, indications for dose escalation, gastrointestinal symptom profiles, and clinical outcomes among individuals receiving off-label escalated upadacitinib therapy.

MATERIALS AND METHODS

We conducted a retrospective single-center cohort study within the Atrium Health System at Carolinas Medical Center, affiliated with Atrium Health Wake Forest Baptist, that included patients with IBD who received off-label escalated dosing of upadacitinib between February 2023 and March 2025 at the time of data extraction. The study was approved by the Atrium Health Wake Forest Baptist Institutional Review Board (No. 00125815), with a waiver of informed consent due to the retrospective nature of data collection.

Patient selection

Eligible patients were adults between 18 and 80 years of age with a diagnosis of UC or CD for at least 3 months and endoscopic confirmation of disease who underwent at least one reinduction cycle with upadacitinib 45 mg daily beyond the FDA-approved induction period. Patients were included if they had a documented loss of response on maintenance dosing requiring re-escalation or were continued on induction dosing beyond 8 weeks (UC) or 12 weeks (CD) at provider discretion. Patients were excluded if they did not complete the full 8- or 12-week reinduction cycle or discontinued therapy for non-medical reasons. The study size was determined by the number of eligible patients during the study period. All consecutive eligible patients meeting inclusion criteria were included.

Data collection

A list of patients prescribed the escalated dose of upadacitinib was generated through the institutional specialty pharmacy database. Clinical data were extracted through electronic medical record review and entered into a standardized REDCap data collection form. Most recent follow up was August 2025, when data were entered into the REDCap database. Collected variables included the following.

Demographics: Age, sex, race, IBD subtype, prior biologic or JAK inhibitor exposure.

Treatment characteristics: Indication for escalation, duration of escalated therapy, concurrent IBD medications, subsequent treatment modifications (discontinuation or additional therapy).

Clinical outcomes: Provider-documented gastrointestinal symptoms (abdominal pain, nausea, vomiting, diarrhea, mucus in stool, hematochezia, urgency, tenesmus, constipation, weight loss, anorexia, fever, fatigue, joint pain, mouth ulcerations, skin rash, dizziness, other), systemic corticosteroid use, IBD-related hospitalizations within 12 months before and during follow-up after dose escalation.

Biomarker outcomes pre- and post-escalation: Complete blood count, complete metabolic panel, C-reactive protein (CRP), erythrocyte sedimentation rate (ESR), fecal calprotectin.

Endoscopic outcomes: Findings from colonoscopy or flexible sigmoidoscopy performed within three months before and at least three months after escalation including Mayo endoscopic subscore (UC) or Simple Endoscopic Score for CD (SES-CD) when reported, pathology reports, and descriptive assessment of improvement.

Radiographic imaging of the abdomen: XR, computed tomography, and magnetic resonance imaging studies from before and after dose escalation were reviewed when available.

Outcomes

The primary outcome was clinical response after dose escalation. Response was defined as improvement in symptoms documented by the treating provider and/or objective improvement in inflammatory biomarkers or endoscopic findings. Because this was a retrospective study conducted in routine clinical practice, fully validated disease activity indices (such as the Mayo endoscopic subscore for UC or the SES-CD) were not consistently documented at standardized intervals. In real-world settings, clinical decision-making is frequently guided by provider-documented symptom assessment, laboratory trends, and endoscopic impressions rather than systematically recorded composite scores. Therefore, provider-documented symptom improvement and objective biomarker or endoscopic changes were used to reflect pragmatic clinical response.

Secondary outcomes included the reason for dose escalation, changes in corticosteroid exposure, IBD-related hospitalizations and emergency department visits before and after escalation, and treatment durability.

Statistical analysis

Continuous variables were presented as medians with interquartile ranges (IQR), and categorical variables were reported as n (%). Paired categorical comparisons were performed using McNemar’s test, and paired continuous variables were analyzed using the Wilcoxon signed-rank test. All tests were two-sided, with statistical significance defined as P < 0.05. Statistical analyses were performed using Microsoft Excel (Microsoft Corporation, Redmond, WA, United States). Analyses were conducted using available paired data for each outcome; no imputation was performed for missing values. A predefined subgroup analysis was performed among patients escalated for uncontrolled symptoms. Other subgroup analyses were not conducted due to limited sample sizes.

RESULTS

Study population

Fifty-six patients met inclusion criteria, including 38 (68%) with CD and 18 (32%) with UC. The median age was 34 years (IQR: 26-46), and 24 (43%) were female. Most patients were White (75%), with smaller proportions of Black (14%), Asian (5%), and other (5%). Before initiating upadacitinib, patients had been exposed to a mean of 2.49 prior biologic therapies. Baseline characteristics are shown in Table 1.

Table 1 Baseline characteristics of the study population, n (%)/mean ± SD/median (interquartile range).

Characteristic	Value
Age (years)	34 (26-46)
Female sex	24 (43)
Race/ethnicity
White	42 (75)
Black	8 (14)
Asian	3 (5)
Other	3 (5)
IBD subtype
Crohn’s disease	38 (68)
Ulcerative colitis	18 (32)
Prior biologic exposure	2.49 ± 1.25

IBD: Inflammatory bowel disease.

Indications for escalation

Fifty-two patients underwent re-escalation after loss of response on maintenance dosing, while four continued induction dosing beyond the labeled 8- or 12-week periods. When indications for escalation were analyzed inclusively, worsening gastrointestinal symptoms were present in 31 patients, endoscopic activity in 27, and laboratory abnormalities in 10; categories in Table 2 represent mutually exclusive combinations of these indications.

Table 2 Indications for off-label escalated dosing of upadacitinib.

Reason for off-label escalated dosing	Number of patients
Symptoms	22
Endoscopic activity	19
Laboratory findings	3
Symptoms and endoscopic activity	5
Symptoms and laboratory findings	3
Endoscopic activity and laboratory findings	2
Symptoms, endoscopic activity, and laboratory findings	1
Imaging and laboratory findings	1

Categories represent mutually exclusive indications (n = 56 total patients); each patient is counted once based on the combination of clinical symptoms, endoscopic activity, imaging findings, and laboratory abnormalities prompting dose escalation.

Clinical outcomes

Provider-documented symptoms before and after escalation were available for 49 patients. Forty-four of the patients had symptoms initially, which decreased to 35 patients after escalation (P = 0.022). The most frequent baseline symptoms included abdominal pain (n = 28), diarrhea (n = 27), hematochezia (n = 24), and urgency (n = 22). After escalation, significant improvement was noted in abdominal pain (28 to 14 patients, P = 0.003), diarrhea (27 to 18 patients, P = 0.049), hematochezia (24 to 10 patients, P = 0.004), and urgency (22 to 8 patients, P = 0.003). Improvement was observed across nearly all symptom domains, with no change in weight loss, ocular symptoms, or oral ulcerations. Symptom data for all patients are shown in Table 3.

Table 3 Symptom improvement in entire cohort, n (%).

Symptom	Before, n = 49	After, n = 49	P value
Any symptoms	44 (89.8)	35 (71.4)	0.022
Abdominal pain	28 (57.1)	14 (28.6)	0.003
Urgency	22 (45.0)	8 (16.3)	0.003
Hematochezia	24 (49.0)	10 (20.4)	0.004
Diarrhea	27 (55.1)	18 (36.7)	0.049
Extra-intestinal manifestations	20 (35.7)	13 (23.2)	0.089
Constipation	5 (10.2)	2 (4.1)	0.375
Mucus	7 (14.3)	4 (8.2)	0.450
Nausea	8 (16.3)	7 (14.3)	1.000
Vomiting	2 (4.1)	1 (2.0)	1.000
Tenesmus	5 (10.2)	4 (8.2)	1.000

Gastrointestinal and extra-intestinal symptoms were assessed in all patients. P values were calculated using McNemar’s test. Seven patients were excluded due to lack of paired data.

Subset analysis was performed among the 31 patients who were escalated for uncontrolled symptoms, 30 of whom had paired data. In this group, the most frequent baseline complaints were abdominal pain (n = 19), diarrhea (n = 19), hematochezia (n = 16), and urgency (n = 15). After escalation, significant improvements were noted in abdominal pain (19 to 8 patients, P = 0.003), diarrhea (19 to 12 patients, P = 0.039), and hematochezia (16 to 5 patients, P = 0.007). There was a trend toward improvement in urgency (15 to 7 patients, P = 0.077), but this was not statistically significant. P values for symptoms were calculated using McNemar’s test. Symptom data for this subset of patients are shown in Table 4.

Table 4 Symptom improvement in patients escalated for uncontrolled symptoms, n (%).

Symptom	Before, n = 30	After, n = 30	P value
Any symptoms	30 (100)	25 (83.3)	0.063
Abdominal pain	19 (63.3)	8 (26.7)	0.003
Urgency	15 (50.0)	7 (23.3)	0.077
Hematochezia	16 (53.3)	5 (16.7)	0.007
Diarrhea	19 (63.3)	12 (40.0)	0.039
Extra-intestinal manifestations	14 (46.7)	10 (33.3)	0.160
Mucus	4 (13.3)	2 (6.7)	0.625
Tenesmus	1 (3.3)	3 (10.0)	0.625
Constipation	3 (10.0)	2 (6.7)	1.000
Nausea	5 (16.7)	5 (16.7)	1.000
Vomiting	2 (6.7)	1 (3.3)	1.000

Gastrointestinal and extra-intestinal symptoms were documented among patients escalated for uncontrolled symptoms. P values were calculated using McNemar’s test.

Systemic corticosteroid use significantly decreased from 28 (50%) in the year before escalation to 18 (32%) afterward (McNemar’s P = 0.031). The number of patients with at least one IBD-related hospitalization or emergency department visit declined from 18 (32%) in the year prior to dose escalation to 11 (20%) after escalation (McNemar’s P = 0.118). The total number of IBD-related hospitalizations or emergency department visits decreased from 44 in the year prior to escalation to 29 after escalation. These findings are summarized in Figure 1. Among 56 patients, 14 experienced fewer hospitalizations or emergency department visits after dose escalation, whereas 6 experienced more. Change in number of events per patient was analyzed using the Wilcoxon signed-rank test and did not reach statistical significance (P = 0.094).

Figure 1 The changes in the number of patients before and after the drug dosage was increased. A: Corticosteroid use before and after dose escalation. The number of patients receiving systemic corticosteroids decreased from 28 to 18 (McNemar’s P = 0.031). Four patients initiated corticosteroids after escalation, whereas 14 discontinued therapy; B: Patients with ≥ 1 inflammatory bowel disease-related hospitalization or emergency department visit before and after dose escalation. The number of patients with at least one event decreased from 18 to 11 (McNemar’s P = 0.118). Eleven patients with a prior event had no subsequent events after escalation, whereas four experienced a new event. IBD: Inflammatory bowel disease.

Biomarkers

CRP values before and after escalation were available for 40 patients. Median CRP (reference < 5 mg/L) decreased from 0.25 mg/L (IQR: 0-1.75) to 0 mg/L (IQR: 0-0.96), though this did not reach statistical significance (Wilcoxon signed-rank P = 0.470). ESR values were available for 12 patients; median ESR (reference 0-20 mm/hour) decreased from 27 mm/hour (IQR: 7.25-43.25) to 17.5 mm/hour (IQR: 7.75-59.5), also without significant change (Wilcoxon signed-rank P = 0.630). Fecal calprotectin (reference < 50 μg/g) was available for 21 patients and decreased from a median of 276 μg/g (IQR: 136.5-1030.5) to 146 μg/g (IQR: 46-497.5), which was not statistically significant (Wilcoxon signed-rank P = 0.390). Biomarker outcomes are shown in Table 5.

Table 5 Biomarker outcomes before and after dose escalation (paired observations), median (interquartile range).

Outcome	Before	After	P value
CRP (mg/L)	0.25 (0-1.75)	0 (0-0.96)	0.470
ESR (mm/hour)	27 (7.25-43.25)	17.5 (7.75-59.5)	0.630
Fecal calprotectin (μg/g)	276 (136.5-1030.5)	146 (46-497.5)	0.390

P values were calculated using the Wilcoxon signed-rank test (two-sided). Paired data were available for C-reactive protein (n = 40), erythrocyte sedimentation rate (n = 12), and fecal calprotectin (n = 21). CRP: C-reactive protein; ESR: Erythrocyte sedimentation rate.

There were insufficient paired data to perform subgroup analyses limited to those escalated for abnormal laboratory markers. Among those patients, only five had paired CRP values, three had paired ESR values, and four had paired fecal calprotectin values. Complete blood count and metabolic panel parameters remained stable, and no significant laboratory safety signals were identified.

Endoscopic outcomes

Among UC patients, Mayo endoscopic subscores (n = 3) improved from a median of 3 (IQR: 2-5) before escalation to 0 (IQR: 0-0.5) after escalation; however, this did not reach statistical significance (Wilcoxon signed-rank P = 0.250). Among CD patients, SES-CD scores (n = 7) remained stable, from a median of 7 (IQR: 5.5-9) to 7 (IQR: 4-9.5), without statistical significance (Wilcoxon signed-rank P = 0.440). Given the small number of patients with paired endoscopic data, these findings should be interpreted with caution. Endoscopic outcomes are shown in Table 6. There were insufficient concordant data points to perform subgroup analyses limited to those escalated for endoscopic findings.

Table 6 Endoscopic outcomes before and after dose escalation (paired observations), median (interquartile range).

Endoscopic outcome	Before	After	P value
Mayo endoscopic subscore	3 (2-5)	0 (0-0.5)	0.250
SES-CD	7 (5.5-9)	7 (4-9.5)	0.440

Mayo endoscopic subscore (ulcerative colitis); n = 3 paired observations. Simple endoscopic score for Crohn’s disease; n = 7 paired observations (6 non-zero paired differences included in Wilcoxon analysis). Values are presented as median (interquartile range). P values were calculated using the two-sided Wilcoxon signed-rank test. Given the small sample sizes, results should be interpreted with caution. SES-CD: Simple endoscopic score for Crohn’s disease.

Treatment durability and modifications

At the most recent follow-up, 50 of 56 patients (89.3%) remained on escalated upadacitinib dosing. Six patients (10.7%) discontinued therapy, most commonly due to persistent disease (n = 3), followed by colorectal cancer (n = 1), skin infection (n = 1), and allergic reaction (n = 1). Additional therapeutic modifications included initiation of concurrent dual-biologic/small molecule therapy in 14 patients and addition of 5-aminosalicylate therapy in three patients.

DISCUSSION

In this single-center retrospective cohort study, off-label escalation of upadacitinib to induction dosing was associated with clinically meaningful benefits in patients with IBD who had lost response to maintenance therapy or who had severe disease necessitating prolonged induction dosing. Most patients underwent dose escalation due to uncontrolled symptoms, followed by endoscopic activity. Symptom improvements were most pronounced in abdominal pain, diarrhea, and hematochezia in both the total cohort and the subset analysis. Urgency significantly improved in the total cohort and trended toward improvement in the subset. Corticosteroid exposure also significantly decreased following dose escalation.

No new laboratory safety signals were identified during the observation period. One patient discontinued therapy following a diagnosis of colorectal cancer during follow-up; causality cannot be established in this retrospective analysis. Upadacitinib, as a JAK inhibitor, carries boxed warnings including risks of serious infection, malignancy, major adverse cardiovascular events, and venous thromboembolism. These warnings are primarily derived from broader JAK inhibitor safety data and post-marketing surveillance. Given the small sample size and retrospective design, this study was not powered to evaluate rare adverse events. Careful patient selection, individualized risk assessment, and ongoing monitoring remain essential when considering dose intensification strategies in clinical practice. Larger prospective studies are warranted to better define the safety profile of dose escalation in IBD.

Both the total number of IBD-related hospitalizations and emergency department visits and the number of patients with at least one event declined after dose escalation, although these differences did not reach statistical significance. Inflammatory biomarkers, including ESR, CRP, and fecal calprotectin, also showed numerical improvement without achieving statistical significance. The limited number of paired measurements and variability in baseline disease activity likely contributed to this finding. It is also possible that symptomatic improvement may precede or occur independently of measurable biochemical changes. In routine clinical practice, where treatment decisions are often driven by symptom recurrence rather than protocolized scoring systems, these findings provide practical guidance for clinicians considering dose re-escalation in patients with secondary loss of response.

Endoscopic scores moved in a favorable direction as well, but interpretation is constrained by the small number of patients with paired examinations. While these results should be viewed cautiously, the consistency of directional improvement across multiple domains suggests a possible treatment effect that warrants further study in larger cohorts.

Dose intensification is a well-recognized strategy for biologic therapies in IBD, including anti-tumor necrosis factor agents, vedolizumab, and ustekinumab. Experience with tofacitinib has shown that dose reduction can lead to relapse and that remission is not always regained after re-escalation. In this context, our findings suggest that upadacitinib re-escalation may offer symptomatic benefit and potentially reduce healthcare utilization in selected patients, even if objective markers demonstrate more modest changes.

Our results align with the only other published real-world study examining upadacitinib reinduction and high-dose maintenance. In that multicenter cohort, Ho et al[7] reported recapture of response in the majority of patients restarted on 45 mg daily, with sustained remission in most responders and no safety concerns identified. Although that analysis focused primarily on clinical response and treatment durability, our study adds more granular symptom-level data, as well as information on corticosteroid use and hospitalization trends. Together, these observations support dose escalation as a reasonable management approach in carefully selected patients.

This study has several limitations. First, its retrospective design introduces the potential for selection and information bias, including variability in symptom documentation and in the timing of laboratory and endoscopic assessments. In the absence of a comparator group, observed improvements may reflect regression to the mean or natural fluctuation of disease activity rather than a true causal treatment effect. Second, as a single-center study with a relatively small sample size, the results may not be generalizable. Third, incomplete biomarker and endoscopic data, including very small paired endoscopic sample sizes, limited the ability to detect statistically significant differences in objective outcomes. Additionally, validated composite disease activity indices were not uniformly available due to the retrospective and pragmatic nature of the study design as well as variability in physician documentation. While standardized scoring systems improve comparability in clinical trials, real-world clinical practice often relies on physician assessment, symptom documentation, and selective objective testing to guide management decisions. As such, the outcomes measured in this study reflect typical clinical decision-making patterns rather than protocolized trial assessments. Finally, patients were escalated for differing clinical indications including symptoms, laboratory abnormalities, or endoscopic activity, introducing heterogeneity in clinical context.

CONCLUSION

Despite these limitations, this study provides real-world data on upadacitinib dose escalation in patients with IBD who experience secondary loss of response to maintenance therapy. The observed improvements in key symptoms and reduction in corticosteroid exposure suggest that dose intensification may represent a pragmatic strategy in selected patients. Prospective, multicenter studies with standardized protocols are warranted to confirm these findings, particularly with regards to objective outcomes, determine optimal patient selection criteria and duration of escalation, and evaluate long-term safety and cost-effectiveness.