Dual-targeted therapy for the management of refractory Crohn’s disease: a retrospective cohort study

Abstract

Crohn’s disease (CD) is a chronic relapsing autoimmune disease. Dual-targeted therapy (DTT) has been proven effective and safe in refractory inflammatory bowel disease (IBD), especially the combination of ustekinumab with vedolizumab (UST + VDZ). However, the efficacy and safety of DTT on refractory CD are still limited or insufficient. We aim to evaluate the efficacy and safety of DTT including ustekinumab combined with adalimumab (UST + ADA) and UST + VDZ in refractory CD patients. Refractory CD patients who failed at least two biologics from Xijing IBD Center were included. Clinical evaluation was performed by Crohn’s Disease Activity Index (CDAI). Endoscopic data were assessed using Simple Endoscopic Score-Crohn’s Disease (SES-CD). Other clinical indicators were also collected. Totally, 23 patients were enrolled, and 20 completed 4–27 months of follow-up. Among these patients, there were significant post-treatment reductions in median SES-CD [14.5–5 (p < 0.005)] and CDAI [326.25–94.75 (p < 0.005)]. For UST + ADA (n = 16), 68.75% reached endoscopic response, 56.25% achieved endoscopic remission, 75% clinical response occurred, and 68.75% of patients achieved clinical remission. For UST + VDZ (n = 4), 25% had endoscopic response, 50% had clinical response, and one achieved clinical remission. The regimen of UST + ADA exhibited a more favorable performance during follow-up. 7 patients discontinued DTT due to no significant improvement or exacerbation, and no severe adverse events were observed. The study supports the effectiveness and safety of DTT for refractory CD and especially suggests the potential efficacy and safety of UST + ADA. Larger sample and randomized controlled trial (RCT) studies are needed for confirmation.

Introduction

Crohn’s disease (CD), which belongs to inflammatory bowel disease (IBD), is a chronic inflammatory condition of the gastrointestinal system resulting from a dysregulated immune response and has seen an increasing global prevalence [1, 2]. It can affect any part of the gastrointestinal tract, with the terminal ileum and proximal colon being the most commonly involved areas [1]. Additionally, 50% of patients will have stenosis, fistulas, and abscesses, and up to 50% may have extra-intestinal symptoms, most commonly in the joints, skin, and eyes [1, 3]. Traditional treatment options include aminosalicylates, corticosteroids, and immunomodulators [4]. Effective biological agents include TNF-α antagonists (Infliximab, Adalimumab, Certolizumab pegol) and antibodies inhibiting IL-12/23 p40 subunit (Ustekinumab) and targeting leukocyte α4β7 integrins (Vedolizumab) [5]. Currently, biological therapies are widely used in patients with moderate-to-severe early stage CD; however, the remission rates are only approximately 40% after one year [6, 7]. Additionally, a retrospective study indicated that patients who required surgery had a recurrence rate of nearly 50% within 10 years [8]. With the emergence of newer biological agents targeting different molecular pathways and small-molecule drugs, the medical treatment options for CD have further expanded [9]. Nevertheless, some patients still have persistent symptoms and disease activity. They are defined as refractory CD according to the International Organization for the Study of Inflammatory Bowel Disease (IOIBD) [10]. The utilization of at least two biologic agents, or the concurrent use of one biologic agent and one small-molecule drug, might be a promising strategy for patients with refractory CD [11]. A 2007 randomized controlled trial first observed that combining natalizumab with infliximab in patients with refractory Crohn’s disease did not result in more adverse events than Infliximab alone [12]. Since then, the efficacy and safety of dual-targeted therapy (DTT) have been progressively explored. In a meta-analysis of 30 cohort studies, Ahmed et al. summarized pooled clinical and endoscopic remission rates of 58.8% (95% CI 42–74%) and 34.3% (95% CI 23–46%) on 288 IBD patients who received dual biologic or small molecule therapy, and they observed 31% (95% CI 13–54%) of adverse events, but only 7% (95% CI 2–13%) were severe event [13]. Although limited data, ustekinumab plus vedolizumab (UST + VDZ) seems to be the most promising in all DTT trials reported [14]. Studies have demonstrated that the inhibition of IL-23 is especially efficacious among patients who show no response to anti-TNF-α, and the most prevalent combination is Ustekinumab combined with anti-TNF-α antibodies [15, 16]. Adalimumab, being a widely utilized anti-TNF-α agent, exhibits superior efficacy in patients with moderate-to-severe CD in comparison with infliximab [17, 18]. It is possible that the combination of ustekinumab and adalimumab (UST + ADA) yields a better outcome. However, in the prior exploration of the application of dual biologic agents in patients with IBD, only four cases were treated with UST + ADA, and all patients achieved a clinical response without any adverse effects [15]. Its efficacy and safety in treating refractory CD also need to be further explored. This study aims to further assess the efficacy and safety of DTT, including UST + ADA and UST + VDZ, after treatment of patients with refractory CD.

Materials and methods

Study design and patient population

This was a single-center retrospective study with the following inclusion criteria: (i) age ≥ 18 years; (ii) diagnosed with CD according to the Chinese guidelines for the diagnosis and treatment of CD [19]; (iii) patients met with refractory CD criteria according to the consensus of IOIBD. Refractory CD was defined per IOIBD criteria as persistent symptoms or inflammation despite adequate treatment with  ≥ 2 biologics (dose optimization confirmed) [10]; (iv) patients had previously experienced treatment failures with at least two biological agents; and (v) patients were proactively willing to undergo treatment for long-term observation. Exclusion criteria: (i) patients who failed to satisfy refractory CD criteria; (ii) refractory CD patients were receiving surgery; (iii) refractory CD patients were receiving treatment with small-molecule drugs; and (iv) patients who were not willing to be engaged in this study.

Outcome assessment

Baseline demographic data, including sex, age, Montreal classification, duration of disease, prior therapies, surgical history, EIM (Extra-Intestinal Manifestations), and comorbidity, were collected for all the patients. Follow-up intervals were determined by clinical need, with endoscopic/clinical assessments performed at the treating physician’s discretion. Adverse events, including infections, immunogenicity (anti-drug antibodies), and injection-site reactions, were systematically monitored via patient-reported outcomes and laboratory tests.

The primary endpoint of this study is endoscopic response, which is defined as a  ≥ 50% decrease in the Simple Endoscopic Score-Crohn’s Disease (SES-CD) [20]. The secondary endpoint is clinical response, characterized by either a 70-point decrease in the Crohn’s Disease Activity Index (CDAI) or a decrease of less than 70 points but with a total score of  ≤ 150 [21]. Other outcome measures include endoscopic remission (SES-CD ≤ 2) and clinical remission (CDAI ≤ 150). SES-CD scores were assessed independently by two blinded gastroenterologists (κ = 0.85). CDAI was calculated retrospectively using clinical records (stool frequency, pain, well-being, hematocrit, weight). During the treatment process, ESR (erythrocyte sedimentation rate), CRP (C-reactive protein), BMI (body mass index), PLT( Platelet), and stool frequency were recorded.

Statistical analysis

Percentages were used for categorical variables. Between-group comparisons were performed using Wilcoxon rank. Continuous variables are indicated as the median (interquartile range). P values  ≤ 0.05 were considered significant. All results should be considered as exploratory. Data were explored using IBM SPSS Statistics version 27.

Results

A total of 23 patients were enrolled in the study, three patients discontinued follow-up due to loss to contact (n = 2) and personal reasons unrelated to treatment (n = 1), while 20 patients finished 4-27 months follow-up and then were evaluated within this study, and the median time to endoscopic and clinical assessment was 7 months (5,12) (Fig. 1). Of the 20 patients, 75% (n = 15) of patients were male. The median medical history was 4.5 years (2.55, 10.38). 18 patients were diagnosed before the age of 40 (90%); 12 patients were involved with the ileocolonic (60%). 9 patients (45%) had a structuring phenotype, and 1 patient had a penetrating phenotype. Two patients had EIM of oral ulcers, and one had a comorbidity of rheumatoid arthritis (Table 1).

Fig. 1.

Patient inclusion flowchart

Table 1.

Baseline characteristics of 20 patients

Case	Sex	Age	Montreal classification	Duration (Years)	Previous therapy	Combination	EIM/ Comorbidity	Surgical history
1	M	26	A2L3B1	3.4	5-ASA, ADA, UST	UST + ADA	–	–
2	M	21	A1L2B2	5	5-ASA, IFX, UST	UST + ADA	–	–
3	F	19	A2L3B1	2.4	5-ASA, GC, IM, IFX, ADA	UST + ADA	–	–
4	M	30	A2L2B1	2.2	5-ASA, GC, ADA, UST	UST + ADA	Oral ulcer	–
5	M	35	A2L3B2	15	5-ASA, IFX, ADA	UST + ADA	–	Colectomy
6	F	22	A2L3B1	5	5-ASA, IM, ADA、UST	UST + ADA	–	–
7	F	26	A2L3B2	2	5-ASA, IFX, UST	UST + ADA	–	–
8	F	56	A3L3B1	0.5	5-ASA, IFX, UST	UST + ADA	Rheumatoid arthritis	–
9	M	32	A2L1B2	10	5-ASA, ADA, UST	UST + ADA	–	Colectomy
10	M	51	A2L1B3	12.4	5-ASA, GC, IM, IFX, ADA, UST	UST + ADA	Oral ulcer	Colectomy
11	M	37	A2L3B2	11.2	5-ASA, GC, IFX, ADA	UST + ADA	–	–
12	M	24	A2L3B1	3	5-ASA, ADA, UST	UST + ADA	–	–
13	M	37	A2L1BI	10	5-ASA, GC, IFX, UST	UST + ADA	–	–
14	M	26	A2L3B1	9.2	5-ASA, ADA, UST	UST + ADA	–	–
15	M	44	A2L2B2	3	5-ASA, IM, IFX, UST	UST + ADA	–	–
16	M	42	A3L2B2	0.5	5-ASA, GC, IFX, UST	UST + ADA	–	–
17	F	24	A2L3B2	10.8	5-ASA, IM, IFX, VDZ, UST	UST + VDZ	–	–
18	M	39	A2L1B1	4	5-ASA, GC, IM, IFX, UST	UST + VDZ	–	–
19	M	25	A2L3B1	3.4	5-ASA, ADA, UST	UST + VDZ	–	–
20	M	32	A2L1B2	10.5	5-ASA, ADA, UST	UST + VDZ	–	Colectomy

5-ASA, 5-aminosalicylic acid; GC, Corticosteroid; IM, immunomodulator; IFX, infliximab; ADA, Adalimumab; UST, Ustekinumab; VDZ, Vedolizumab; EIM, extra-intestinal manifestation

All 20 patients used 5-aminosalicylic acid (5-ASA), 6 patients used immunomodulators, 6 patients used corticosteroids before, and moreover, all patients received at least 2 biological monotherapies. 20% (4/20) of patients had a history of colectomy. In this study, the DTT includes UST + ADA (n = 16) and UST + VDZ (n = 4) (Table 1). During DTT, two patients received enteral nutrition. No immunomodulators and corticosteroids were used concomitantly.

All patients were followed up for 4–27 months, the median SES-CD score reduced from 14.5 (12, 18.75) to 5 (0.5, 11.5) (p < 0.005), and the median CDAI score decreased from 326.25 (273.3, 385.77) to 94.75 (37.8, 241.5) (p < 0.005) (Fig. 2). Endoscopic response occurred in 60% of treatments, and endoscopic remission was achieved in 45% of patients. Clinical response occurred in 70% of treatments, and clinical remission was achieved in 60% of patients. UST + ADA showed superior endoscopic response vs UST + VDZ (68.75% vs 25%) and clinical remission (68.75% vs 25%), though statistical significance was limited by sample size (Table 2).

Fig. 2.

A, B Changes in CDAI and SES-CD in all patients, C, D Changes in CDAI and SES-CD in patients who received UST + ADA. CDAI: Crohn’s Disease Activity Index; SES-CD: Simplified Endoscopic Score–Crohn’s Disease. Median values with IQR are shown

Table 2.

Response and remission rates on clinical and endoscopic of two DTT regimens

	Endoscopic response	Endoscopic remission	Clinical response	Clinical remission
UST + ADA	68.75% (11/16)	56.25% (9/16)	75% (12/16)	68.75% (11/16)
UST + VDZ	25% (1/4)	0	50% (2/4)	25% (1/4)
All DTT	60% (12/20)	45% (9/20)	70% (14/20)	60% (12/20)

UST, Ustekinumab; ADA, adalimumab; VDZ, vedolizumab; DTT, dual-targeted therapy

In the regimen of UST + ADA, clinical scores and endoscopic scores were significantly reduced (Fig. 2). Endoscopic response occurred in 68.75% (11/16) of trials, 56.25% (9/16) achieved endoscopic remission, and the median SES-CD score reduced from 14.5 (12, 18.75) to 2 (0, 11.25). Clinical response occurred in 75% (12/16) of treatments, 68.75% (11/16) of patients achieved clinical remission, and the median CDAI score decreased from 319.5 (241.35, 380.5) to 83 (36.73, 225.5) (Table 2). Patients’ clinical indicators were also significantly improved. The median stool frequency decreased from 4.5 times to 1.5 times, the median ESR level decreased from 23.65 to 7.17 mm/h, and the median CRP concentration declined from 27.65 to 3.99 mg/L. Overall patients’ BMI level also increased, with the median BMI level increasing from 20.9 to 22.35 kg/m2 (Table 3). Two patients initially combined with enteral nutrition therapy and ultimately achieved endoscopic remission. For the patients with concomitant oral ulcers, the symptoms also disappeared after treatment.

Table 3.

Paired sample Wilcoxon signed-rank test for the median changes in in the regimen of ustekinumab combined with adalimumab

	Pre-treatment	Post-treatment	Z value	P value
SES-CD score	14.5 (12, 18.75)	2 (0, 11.25)	− 3.099	0.002
CDAI score	319.5 (241.35, 380.5)	83 (36.73, 225.5)	− 3.361	< 0.001
ESR	23.65 (6.40, 43.35)	7.17 (4.00, 11.90)	− 2.999	0.003
C-reactive protein	27.65 (5.99, 69.55)	3.99 (1.29, 15.04)	− 2.982	0.003
Stool frequency	4.5 (3.75, 8)	1.5 (1, 2.75)	− 3.159	0.002
BMI	20.9 (19.30, 22.78)	22.35 (21.32, 23.48)	− 2.379	0.017
Platelet	299.5 (222.25, 397.75)	211 (194.25, 236.25)	− 2.637	0.008

In the regimen of UST + VDZ, during the follow-up, 25% (1/4) of patients reached endoscopic response with no endoscopic remission occurring, and the median SES-CD score decreased from 14 (9, 22.75) to 8.5 (6.25, 11.5). 50% (2/4) of patients reached clinical response, one of them achieved clinical remission, and the median CDAI score reduced from 373.65 (301.25, 401.2) to 220.5 (117, 32.25). (Table 2). Compared with the regimen of UST + ADA, the improvement of the four patients’ other clinical indicators’ median values is not significant (Table 4).

Table 4.

Paired sample Wilcoxon signed-rank test for the mean changes in the regimen of ustekinumab combined with vedolizumab

	Pre-treatment	Post-treatment	Z value	P value
SES-CD score	14 (9, 22.75)	8.5 (6.25, 11.5)	− 1.826	0.068
CDAI score	373.65 (301.25, 401.2)	220.5 (117, 320.25)	− 1.826	0.068
ESR	30.5 (10.25, 83)	19.35 (6.83, 22.35)	− 1.826	0.068
C-reactive protein	21.55 (5.48, 44.28)	11.95 (3.89, 20.63)	− 2.599	0.068
Stool frequency	5.5 (3.25, 7.75)	3 (2.25, 3.75)	− 1.604	0.109
BMI	20.65 (16.87, 22.18)	20.9 (18.43, 21.73)	− 0.816	0.414
Platelet	400 (252, 783.5)	233.5 (214, 298.75)	− 1.461	0.144

In total, 7 patients discontinued the treatment. Four patients were UST + ADA; among them, three patients showed exacerbated diarrhea after 5 months of treatment and one patient showed a slight improvement in endoscopy after 4 months of initial treatment. The other three were treated with UST + VDZ; they chose to discontinue the treatment due to no significant benefit in the endoscopic examination during the 5 months’ treatment. No major adverse events such as serious infection, malignancy, major adverse cardiovascular events, or venous thromboembolism were recorded. Mild adverse events (e.g., transient fatigue, headache) occurred in 4 patients but did not require treatment discontinuation.

Discussion

Ustekinumab, a monoclonal antibody targeting the p40 subunit of interleukin 12 and 23, has shown efficacy and long-term safety in patients with moderate-to-severe active CD [22, 23]. Adalimumab, a widely used traditional anti-TNF-α agent, is more effective for CD than Infliximab and has similar efficacy to Ustekinumab in treating moderate–severe CD [18, 24]. Vedolizumab, an integrin antagonist, binds to α4β7 integrins to prevent the migration from the bloodstream to the intestinal mucosa, thus reducing the influx of leukocytes into inflamed tissue, and exhibited good effectiveness and safety for CD patients in clinical trials and the real-world settings [25, 26]. In this study, 16 patients were administered UST + ADA, while 4 patients received UST + VDZ.

In a retrospective study, Edward Yang, MD, et al. reported that endoscopic and clinical responses were observed in 43% and 50% of refractory CD patients treated with dual-biological therapy, respectively [27]. In our study, endoscopic response and clinical response rates reached 60% (12/20) and 70% (14/20), respectively, which is better than previous studies. Among the patients who received UST + VDZ in Yang et al.’s study, 63% (5/8) of the patients achieved endoscopic improvement and 71% (5/7) achieved clinical improvement. However, 25% (1/4) of the patients who received UST + VDZ in our study had an endoscopic response, and 50% (2/4) achieved clinical response. It is possible that bias occurred due to the small volume of patients in this type of combination. Nonetheless, 68.75% (11/16) of patients with UST + ADA achieved endoscopic response, and 75% (12/16) of the patients achieved clinical response and clinical indicators significantly reduced, indicating that UST + ADA is more effective than UST + VDZ in this study.

According to Privitera et al., adverse effects of dual-biological therapy involving UST + VDZ were identified in 13% to 30% of patients, with infections being the most prevalent [28]. Yang et al. noted adverse events in 13% of the trials (3/22) [27]. Among patients who were reported to have received UST + ADA, no adverse events were reported¹⁵. In our study, after careful screening and timely monitoring, no adverse events were observed in either of the DTT regimens.

For the combinations of different drugs, besides efficacy and safety of a single drug, the mechanisms of action (MOAs) are also needed to be considered [11]. Based on the MOAs of various effective molecules, it is theoretically believed the combination of anti-TNF or anti-12/13 or anti-23 with anti-integrin, having some overlap in the MOAs, may possess good efficacy in enhancing activity over a single agent and low additional risks; however, the combination of anti-TNF and IL-12/13 or anti-23, which have similar MOAs, might exhibit low efficacy in enhancing activity over a single agent and exhibit an increase in additional risks [29]. The findings of our study do not fully align with the theory; the results of our study and the previous four cases all demonstrated the efficacy and safety of UST + ADA, which indicates that there may be other underlying mechanisms. However, the overall sample size of this combination is limited, and it is necessary to further enlarge the sample size to explore its latent mechanisms.

The strengths of this study are presented as follows: (1). Firstly, this study emphasized UST + ADA as well as that of UST + VDZ. (2). Secondly, all patients were assessed with objective endoscopic and clinical endpoints. (3). Thirdly, our study encompassed the largest number of patients receiving UST + ADA and meticulously analyzed its efficacy and safety. However, this study still had the following limitations. (1). Firstly, it is a single-center respective study with a small number of patients. (2). Secondly, since there are significant differences in follow-up time, the long-term efficacy of both DTT regimens cannot be established. (3). Thirdly, 4 patients have not finished the follow-up and were not included in the evaluation.

In conclusion, our study suggests potential efficacy and safety of DTT in refractory CD, especially UST + ADA. However, more extended investigation needs to be carried out on long-term effectiveness and safety.

Authors’ contributions

J.H. and J.Z. were responsible for the design and conception of the overall project, data curation, statistical analyses, and writing the original draft. Y.S. carried out data curation, formal analysis, interpretation of results, and writing—review and editing. J.L. was involved in conceptualization, funding acquisition, supervision, and writing—review and editing. All authors reviewed and approved the final draft of the article before submission.

Funding

This work was supported by the National Natural Science Foundation of China (82370588); Natural Science Foundation of Shaanxi Province, Key Industrial Innovation Project Fund (2023-ZDLSF-44); Medical Personnel Training and Promotion Program of Xijing Hospital (XJZT24QN41); National Natural Science Foundation of China Major Research Program Integration Project (92259302); Independent Funds of the Key Laboratory (CBSKL2022ZZ34).

Data availability

The original contributions presented in the study are included in the article material. Further inquiries can be directed to the corresponding authors.

Declarations

Conflict of interest

All the authors confirmed no conflicts of interest statement.

Ethical approval

This study was conducted following the guidelines of the Declaration of Helsinki and proved by the Medical Ethics Committee of the First Affiliated Hospital of the Air Force Medical University (reference number: KY20222333-C-1). Informed consent was obtained from all patients.