Transmural Remission Characterized by High Biologic Concentrations Demonstrates Better Prognosis in Crohn’s Disease

Abstract

Background

The importance and pathophysiology of transmural healing in patients with Crohn’s disease [CD] remains to be verified. We aimed to examine the association between serum concentrations of biologics and transmural remission evaluated via magnetic resonance enterography [MRE].

Methods

We enrolled patients with CD who received maintenance biologics 1 year after induction and prospectively followed up for at least 1 year after baseline laboratory, endoscopic and MRE examination. We evaluated the relationship between baseline factors including the presence of transmural remission and patient prognosis, as well as between serum concentrations and transmural remission.

Results

We included 134 patients, of whom 65, 31, 27 and 11 received infliximab, adalimumab, ustekinumab and vedolizumab, respectively. Those who achieved transmural remission showed a lower risk of hospitalization and surgery than those who did not achieve remission [p < 0.01]. Adjusted hazard ratios of transmural remission for predicting hospitalization and surgery were 0.11 and 0.02, respectively, which were lower than those of clinical remission, biochemical remission and endoscopic remission. Regarding serum concentrations, the median concentration was higher in patients with transmural remission than in patients with transmural activity for all agents [p < 0.01 for infliximab, p = 0.04 for adalimumab, p < 0.01 for ustekinumab, p = 0.08 for vedolizumab].

Conclusions

Transmural remission was the best predictor for prognosis in CD patients who received maintenance biologic therapy. High drug concentration levels were associated with transmural remission confirmed via MRE.

Graphical Abstract

MRE, treat to target, drug monitoring

1. Introduction

Crohn’s disease [CD] is a chronic inflammatory bowel disease [IBD] characterized by relapse and remission. The treatment of CD has progressed greatly with the advent of biologics, which began with tumour necrosis factor alpha blockers [e.g. infliximab and adalimumab], followed by new biologics including p-40 interleukin-12/23 blockers [ustekinumab] and adhesion-molecule blockers [vedolizumab].^1–4 Unfortunately, some patients either do not respond to treatment with these agents or have only a transient response that later requires dose escalation or switching to another therapy.

An important and emerging treatment strategy for patients with CD is the treat-to-target strategy,⁵ which aims primarily to further improve gastrointestinal inflammation that could not be achieved by symptom-based treatment alone. Although the main outcome for improvement in inflammation has classically been endoscopic remission, CD is also characterized by bowel wall inflammation, and the importance of transmural remission has also been highlighted.⁶ Despite the number of studies on the relevance of transmural inflammation in patient prognosis, the majority of evidence comes from retrospective analyses, where the timing of imaging was not controlled.^7–9 Furthermore, serum drug concentration correlates positively with favourable therapeutic outcomes, such as clinical and endoscopic outcomes.¹⁰ However, data on transmural outcomes are limited.

In this prospective study, we aimed to examine the importance of transmural remission evaluated via magnetic resonance enterography [MRE] in patients with CD receiving maintenance biologics and to verify the association between drug concentrations of biologics and transmural remission confirmed via MRE.

2. Methods

2.1. Patients

We recruited patients with CD at Tokyo Medical and Dental University Hospital [a tertiary referral hospital, Tokyo, Japan] who met the following inclusion criteria from June 2018 to June 2021: [i] patients with endoscopically or radiologically proven active bowel disease before starting biologic therapy [infliximab, adalimumab, ustekinumab, vedolizumab] and [ii] patients who received maintenance therapy for 1 year after induction. Biologic therapy was commenced at the discretion of the multidisciplinary team based on clinical, endoscopic, radiological and biochemical evidence of disease activity, as well as individual patient characteristics.

The exclusion criteria were as follows: [i] patients who underwent intestinal surgery after treatment began, [ii] those who experienced primary and secondary failure and were switched to other biologics within 1 year, [iii] those who had prior colostomy/ileostomy, [iv] those who underwent treatment modification within 3 months of treatment induction, and [v] those with contraindications for endoscopic or MRE procedures.

The study was approved by the ethics committee of the Tokyo Medical and Dental University, and informed consent was obtained from all patients.

2.2. Endoscopic procedure and evaluation

Endoscopy [ileocolonoscopy or balloon-assisted enteroscopy; Olympus Medical Systems] was performed by IBD endoscopy experts. Balloon-assisted enteroscopy was performed for patients who had proximal ileal lesions. The procedures were recorded on video, and the recordings were assessed centrally by an expert endoscopist [with 15 years of experience in IBD endoscopy] who was blinded to the patients’ conditions.

Endoscopic evaluation was conducted using the modified simple endoscopic score for CD [SES-CD]; the item on strictures was excluded from the original scoring.¹¹ The scores regarding the size of ulcers, ulcerated surface and affected surface in each segment were evaluated. Endoscopic remission was defined as having a modified SES-CD score <4 [which correlates with the absence of mucosal defects ≥0.5 cm].¹²

2.3. MRE procedure and evaluation

We performed MRE using a 3.0-T scanner (FUJIFILM Healthcare Corporation [formerly HITACHI, Ltd]) with oral and intravenous contrast. Patients received 50 g of magnesium citrate [Horii] the day before MRE and 1000 mL of polyethylene glycol [EA Pharma Co., Ltd] within 60 min before MRE. All magnetic resonance images were taken while the patient was in the supine position. The MRE sequence [Supplementary Table S1] included cine magnetic resonance imaging [MRI], dynamic T1-weighted imaging, T2-weighted imaging and diffusion-weighted imaging [DWI].

Radiologists, who were blinded to the clinical presentation, determined the simplified magnetic resonance index of activity [sMaRIA].¹³ The sMaRIA was calculated as 1 × thickness > 3 mm + 1 × edema + 1 × fat stranding + 2 × ulcers. Radiological remission was defined as the sMaRIA score of the most affected segment <2. MRE scores were determined via consensus between two radiologists.

2.4. Blood collection and measurements

Routine blood tests were taken, after which serum was stored at −80°C for a maximum of 1 month. For patients receiving infliximab, ustekinumab and vedolizumab, serum was collected just before drug administration to measure trough levels. For patients receiving adalimumab, which is self-injected and exhibits only small fluctuations in its concentration,¹⁴ serum was collected at the time of hospital visit which was not synchronized with timing of injection. Coded blood samples were sent to Funakoshi Co., Ltd for analysis, who were blinded to the clinical status. The serum concentrations of biologics and antidrug antibodies were measured using commercially available IDKmonitor enzyme-linked immunosorbent assay kits specific for each biologic, performed as per the kit protocol [Immundiagnostik AG]. Other routine blood tests were performed within the hospital.

2.5. Study design and endpoints

We performed endoscopy and MRE and recorded clinical information including baseline laboratory data at 1 year after induction. All examinations were performed within 2 months and no changes were made to treatment from the time of endoscopy or MRE to blood collection.

Clinical remission was defined as Crohn’s Disease Activity Index [CDAI] < 150. Biochemical remission was defined as C-reactive protein [CRP]/calprotectin negative [≤3 mg/L and ≤250 μg/g, respectively]. Transmural remission was defined as endoscopic and radiological remission. Disease location and behaviour were classified according to the Montreal classification at study baseline: ileal [L1] or ileocolonic [L3] with inflammatory [B1], stricturing [B2] or penetrating [B3] behaviour.¹⁵

After baseline assessment, patients were regularly seen at the clinic every 1–2 months and were followed up for at least 1 year. The duration to hospitalization, surgery and clinical relapse were evaluated as endpoints. Hospitalization was defined as the development of severe symptoms [CDAI > 450] and/or CD-related complications that required hospitalization and/or surgery. Clinical relapse was defined as an increase in the CDAI score by 75 in patients in clinical remission [CDAI ≤ 150] at baseline. The relationship between baseline factors and endpoints was defined as the primary outcome. Sub-analysis was performed for patients with clinical, biochemical and endoscopic remission; and with B2/3 behaviour. We also evaluated the relationship between the serum drug concentrations and baseline factors.

2.6. Statistical analysis

We aimed to recruit 116 patients using sample size calculations that assumed a 75% relapse-free rate in the remission group and a 50% relapse-free rate in the active disease group based on our retrospective evaluation [presented at the Japan Digestive Disease Week 2014 international session], using 80% power and 5% significance.

Continuous variables were reported as means with standard deviation or medians with interquartile range [IQR], as appropriate. The Kaplan–Meier method and the log-rank test were used for calculation of cumulative endpoint-free rates. Cox’s proportional hazards model was used for single/multiple regression analysis. The predictor and confounders with p < 0.1 on univariate analysis were subjected to a final multivariable analysis. Independent variables, such as age, sex, smoking status, prior biologic exposure, prior surgery, concomitant immunomodulator use, CRP level, disease duration, disease behaviour and presence of antidrug antibodies, were considered confounders. Drug concentrations were compared using the Mann–Whitney U test. A receiver operating curve was used to determine the cutoff values of drug concentrations based on the Youden index. Contingency tables were prepared to determine the accuracy of the cutoff when assessed against transmural remission. Multiple logistic regression analysis was used to evaluate the relationship between drug concentrations and each baseline target. The level of statistical significance was set at a probability [p] value of <0.05. SPSS version 21.0 [IBM] and R version 3.3.0 [R Foundation for Statistical Computing] were used. The data underlying this article will be shared on reasonable request to the corresponding author.

3. Results

3.1. Baseline endoscopic and MRE findings

This prospective study included 134 patients [Figure 1]. Their clinical characteristics are summarized in Table 1. Median age was 32 years and median disease duration was 7 years. Of all included patients, 35 [26%], 61 [46%] and 38 [28%] showed inflammatory, stricturing and penetrating behaviour, respectively. Clinical and biochemical remission rates were 85% and 75%, respectively.

Table 1.

Clinical characteristics of all patients [N = 134]

Variable	All patients
Male, n [%]	99 [74%]
Median age [years] [IQR]	32 [26–44]
Median disease duration [years] [IQR]	7 [4–13]
Location, n [%]
L1: Ileal	36 [27%]
L2: Colonic	3 [2%]
L3: Ileocolonic	95 [71%]
Behaviour, n [%]
B1: Inflammatory	35 [26%]
B2: Stricturing	61 [46%]
B3: Penetrating	38 [28%]
Perianal involvement, n [%]
Active	27 [20%]
Remission	48 [36%]
None	59 [44%]
Previous intestinal resection, n [%]	47 [35%]
Biologic therapy, n [%]
Infliximab	65 [49%]
Adalimumab	31 [23%]
Ustekinumab	27 [20%]
Vedolizumab	11 [8%]
Biologic experienced, n [%]
Naïve	97 [72%]
Experienced [one]	27 [20%]
Experienced [two]	10 [8%]
Concomitant treatment, n [%]
Steroids	9 [7%]
Immunomodulators	67 [50%]
Clinical remission, n [%]	114 [85%]
Biochemical remission, n [%]	101 [75%]
Endoscopic remission, n [%]	94 [70%]
Transmural remission, n [%]	54 [40%]

IQR, interquartile range.

Figure 1.

Flowchart for included patients and their outcomes. Of the 134 patients included in the study, 64, 31, 27 and 11 received infliximab, adalimumab, ustekinumab and vedolizumab maintenance therapy, respectively. After a median follow-up period of 30 months, 43 patients had experienced hospitalization and 18 had undergone surgery.

Regarding treatment, 65, 31, 27 and 11 patients were administered infliximab, adalimumab, ustekinumab and vedolizumab, respectively. The doses of infliximab, adalimumab and ustekinumab were increased in five, four and two patients within 1 year after induction, respectively. Nine [7%] and 67 [50%] patients were administered concomitant treatment with steroids or immunomodulators. Among the four different agents [Supplementary Table S2], the infliximab group tended to have long disease duration and an ileocolonic type. In total, 83% and 81% of patients administered infliximab and adalimumab respectively were biologically naïve, and concomitant immunomodulators were frequently used in the same groups.

Endoscopic and transmural remissions were achieved by 94 [70%] and 54 [40%] patients, respectively. High signal intensities on DWI were confirmed in 93% [50/54] of patients with transmural activity and 59% [47/80] of patients with transmural remission, respectively. The infliximab, adalimumab and ustekinumab group showed a higher endoscopic remission rate than transmural remission rate; in contrast, the vedolizumab group showed a higher transmural remission rate [Supplementary Table S2].

3.2. Relationship between endpoints and target

Treatment was changed in 12 patients after endoscopy and MRE. After the median follow-up duration of 30 months, the endpoints of hospitalization and surgery were reached by 43 [32%] and 18 [13%] patients, respectively. The reasons for hospitalization were need for surgical intervention [n = 18], intestinal obstruction [n = 16], worsening symptoms [n = 8] and gastrointestinal bleeding [n = 1]. Indications for surgery were bowel strictures [n = 12], intestinal perforation [n = 2], abdominal abscess [n = 2], intestinal fistula [n = 1] and anastomotic stricture [n = 1].

Figure 2 shows the Kaplan–Meier curve analysis for endpoints stratified with each baseline target. Those patients who achieved clinical, biochemical, endoscopic or transmural remission showed a lower risk of hospitalization than those who did not achieve remission [p < 0.01]. Patients with transmural remission showed the highest hospitalization-free rates; 96, 94 and 91% at 12, 24 and 36 months, respectively. Similarly, those patients who achieved biochemical, endoscopic or transmural remission showed a lower risk of surgery than those who did not achieve remission [p < 0.01]. Patients with transmural remission showed the highest surgery-free rates: 100, 100 and 100% at 12, 24 and 36 months, respectively.

Figure 2.

Kaplan–Meier curve analysis for [a] hospitalization and [b] surgery stratified by their target at baseline. Patients who achieved clinical [p < 0.01], biochemical [p < 0.01], endoscopic [p < 0.01] or transmural remission [p < 0.01] showed lower risks of hospitalization compared with those not in remission. Patients who achieved clinical [p = 0.09], biochemical [p < 0.01], endoscopic [p < 0.01] or transmural remission [p < 0.01] showed lower risks of surgery compared with those not in remission.

Using Cox’s proportional hazards model [Figure 3], the adjusted hazard ratio [HR] of transmural remission for predicting hospitalization was 0.11 (p < 0.01; 95% confidence interval [CI]: 0.04–0.32), which was lower than that of clinical remission [0.36], biochemical remission [0.40] or endoscopic remission [0.39]. Similarly, the adjusted HR of transmural remission for predicting surgery was 0.02 [p = 0.04; 95% CI: 0.00–0.92], which was lower than that of clinical remission [0.42], biochemical remission [0.27] or endoscopic remission [0.19].

Figure 3.

[a] Adjusted hazards ratio [HR] of transmural remission for predicting hospitalization was 0.11 [95% CI, 0.04–0.32], which was lower than that of clinical remission [0.36], biochemical remission [0.40] and endoscopic remission [0.39]. [b] Adjusted HR of transmural remission for predicting surgery was 0.02 [95% CI, 0.00–0.92], which was lower than that of clinical remission [0.42], biochemical remission [0.27] and endoscopic remission [0.19].

Clinical relapse was seen in 52 [46%] of 113 patients with clinical remission at baseline. Kaplan–Meier curve analysis showed that patients who achieved biochemical, endoscopic or transmural remission had a lower risk of clinical relapse than those who did not achieve remission [Supplementary Figure S1]. Those patients with transmural remission showed the highest relapse-free rates: 96, 96 and 87% at 12, 24 and 36 months, respectively.

Sub-group analysis showed that transmural remission was associated with a lower risk both for hospitalization and for surgery in patients with clinical remission, biochemical remission or endoscopic remission [Supplementary Figures S2–S4; p < 0.01 for all].

3.3. Relationship between serum biologic concentrations and transmural remission

The median concentrations of infliximab, adalimumab, ustekinumab and vedolizumab were 3.4 [IQR, 1.1–6.1], 13.4 [IQR, 10.9–20.2], 1.4 [IQR, 0.5–3.4] and 18.5 [IQR, 8.6–22.3] μg/mL, respectively. Anti-drug antibodies were detected in 8% [5/65], 6% [2/31], 4% [1/27] and 9% [1/11] of patients, respectively. Drug concentrations were higher in patients with transmural remission than in patients with transmural activity [Figure 4; p < 0.01 for infliximab; p = 0.04 for adalimumab; p < 0.01 for ustekinumab; p = 0.08 for vedolizumab]. Quartile analyses showed that rates of transmural and endoscopic remission increased with increasing drug concentrations [Supplementary Figure S5].

Figure 4.

Box plot of [a] infliximab, [b] adalimumab, [c] ustekinumab and [d] vedolizumab serum concentrations stratified according to transmural findings. The median concentration was higher in patients with transmural remission than in patients with transmural activity [p < 0.01 for infliximab; p = 0.04 for adalimumab; p < 0.01 for ustekinumab; p = 0.08 for vedolizumab].

Cutoff levels of 4, 14, 2.5 and 19 μg/mL best predicted transmural remission for infliximab, adalimumab, ustekinumab and vedolizumab, respectively [Table 2]. Multiple logistic regression analysis showed that high drug levels and B2/B3 behaviour were independently associated with transmural remission [p < 0.01 for both]. High drug levels were also independently associated with endoscopic remission [p = 0.01], but not with clinical [p = 0.357] or biochemical [p = 0.15] remission.

Table 2.

Cutoff of serum concentration and diagnostic accuracy for transmural remission

Biologics	AUC [95% CI]	Cutoff [μg/mL]	Sensitivity [%]	Specificity [%]	PPV [%]	NPV [%]
Infliximab	0.75 [0.64–0.87]	4	53	77	67	66
Adalimumab	0.74 [0.56–0.92]	14	78	64	47	88
Ustekinumab	0.82 [0.65–0.98]	2.5	56	89	71	80
Vedolizumab	0.83 [0.57–0.99]	19	67	80	80	67

AUC, area under the curve; NPV, negative predictive value; PPV, positive predictive value.

3.4. Sub-analysis of patients with structuring or penetrating behaviour

We then conducted sub-group analysis on the patients with B2/B3 behaviour [n = 99]. Endoscopic and transmural remissions were achieved by 65 [67%] and 30 [30%] patients, respectively. The rate of transmural remission was relatively lower than that of all patients [41%]. Kaplan–Meier curve analysis [Supplementary Figure S6] showed that patients who achieved transmural remission were at a lower risk for hospitalization and surgery [p < 0.01 for both]. Regarding treatment, 48, 20, 22 and nine patients were administered infliximab, adalimumab, ustekinumab and vedolizumab, respectively. Median infliximab, adalimumab, ustekinumab and vedolizumab concentrations were 3.3, 14.2, 1.4 and 19.3 μg/mL, respectively.

4. Discussion

Our findings highlight the importance of transmural remission in patients with CD. We were able to evaluate whether it could be an ultimate target in patients with CD who are treated with biologics. Our prospective study showed that transmural remission was the best predictor for patient prognosis and that high drug concentration levels were associated with transmural remission.

In the present study, the achievement of transmural remission at 1 year after the induction of biologics was associated with a favourable prognosis. A treat-to-target treatment strategy [STRIDE-II] was recently proposed,⁵ in which clinical/biochemical/endoscopic remission were considered important targets. Transmural remission in CD has been newly recognized as an important adjunct but has not been endorsed as a new target. One reason for this was that past studies confirmed the usefulness of cross-sectional imaging modalities in detecting therapy-related changes, but there was lack of evidence for patient prognoses.^7–9 Our results showed that patients with transmural remission [defined by the combination of sMaRIA and SES-CD scores] achieved the best clinical course [hospitalization-free and surgery-free]. This scoring system does not consider the influence of intestinal complications such as structuring/penetrating disease, although transmural remission itself was an independent factor for patient prognosis even after adjusting for baseline factors including disease behaviour. Figure 3 shows transmural remission was the better predictor for patient prognoses than clinical/biochemical/endoscopic remission. However, we should be careful in interpreting the findings as the 95% CIs for transmural remission overlapped with those of other targets. Additionally, although transmural remission could be an ultimate outcome in patients with CD, the ability of currently available treatments is limited in achieving transmural remission, which is pointed out in STRIDE-II.

Next, we evaluated the relationship between serum drug concentration and transmural remission, which may contribute to further identification of the pathophysiology of CD. Although several studies have reported on the relationship between drug concentrations and CD-related outcomes, the present study focused on transmural inflammation. Figure 4 shows that the serum concentration in patients who achieved transmural remission was higher than that detected in patients with transmural activity, which was a trend that was observed for all agents. The calculated cutoff was an accurate predictor for transmural remission, and high drug levels were associated with transmural remission on multiple logistic regression analysis. Transmural remission is considered an adjunctive target in patients with CD and our results suggest that higher drug concentrations may be needed to control transmural inflammation in the gut.

Therapeutic drug monitoring has emerged as a tool to manage patients with CD receiving biologic therapy.¹⁶ For patients on infliximab or adalimumab therapy, although maintenance trough levels of 3–5 μg/mL are associated with clinical remission, higher concentrations are associated with endoscopic remission.^17,18 In our study, adalimumab concentrations were not necessarily taken as trough levels, and this may be why the rather high threshold of 14 µg/mL was associated with transmural remission. For patients administered ustekinumab, maintenance trough concentrations above 4.5 µg/mL were associated with endoscopic response.¹⁹ For patients administered vedolizumab, serum concentrations above 10 µg/mL at week 22 were associated with endoscopic remission.²⁰ Based on these results, we assume that a direct comparison of our results with past studies is difficult because the therapeutic window appears to vary based on the outcome of interest, measurement assay and disease phenotype. Currently, studies that compare proactive and reactive drug monitoring are also scarce. A prospective interventional study is needed to provide further clarification of the results.

Mucosal assessments may not be feasible in certain scenarios, such as in proximal small bowel disease, and a mismatch between endoscopic and transmural remission is not uncommon. Given the complementary nature of cross-sectional imaging, it has been increasingly used in addition to endoscopic assessments. It allows frequent assessments and has the advantage of assessing the entire gastrointestinal tract, including transmural remission. We evaluated both the rate of endoscopic and transmural remissions using validated scoring systems and provided important information for the physicians. However, since this was an observational study, comparison between each biologic was not possible.

The current study has some limitations. First, physicians were not blinded to the clinical evaluation and endoscopic/MRE findings, and this may have influenced the results pertaining to endpoints. The same limitation was discussed in a previous study that showed the efficacy of objective evaluation for the prediction of patient prognosis.²¹ Second, this study was not evaluated for proactive drug monitoring; therefore, there is a possibility of reverse causation [patients being in remission may result in a higher drug concentration]. Third, because the number of patients in each biologic group was relatively small and we did not show the significant differences, caution is warranted in interpreting the results and a larger size study is needed to validate the results in the present study. Finally, we only included patients who were able to continue maintenance therapy for 1 year without surgery, and who do not develop primary or secondary loss of response to treatment. This is a highly selected population, which may lead to bias in interpretation of the results and therefore our results are not generalizable to all patients who start biologic therapy.

In conclusion, transmural remission was the best predictor for patient prognosis in patients with CD who received maintenance biologic therapy for 1 year after induction. High serum drug concentration levels were associated with transmural remission evaluated via MRE. Our findings need to be validated in future prospective intervention studies with a larger sample size.

Funding

None. There was no assistance in drafting the manuscript

Conflict of Interest

None.

Author Contributions

KT conceived and designed the study; KT, AK, YK, TF, YU, HS, SH, MN and KO acquired the data; KT and YK analysed and interpreted the data; KT and YK drafted the manuscript; AK critically revised the manuscript; and MW and RO supervised the study.

Ethics

The study was approved by the ethics committee of the Tokyo Medical and Dental University. Written informed consent was obtained from all patients.

Data Availability Statement

The data underlying this article will be shared on reasonable request to the corresponding author.