Effectiveness of granulocyte and monocyte adsorptive apheresis in patients with ulcerative colitis and 5-aminosalicylate intolerance

Abstract

5-Aminosalicylate (5-ASA) intolerance complicates ulcerative colitis (UC) management, often necessitating alternative treatment approaches. Granulocyte and monocyte adsorptive apheresis (GMA) has been recognized as a safe treatment option for patients with UC who are refractory or intolerant to conventional drugs, including steroids; however, its effectiveness and safety in patients with 5-ASA intolerance remain unclear. This study aimed to investigate the effectiveness and safety of GMA in patients with UC and 5-ASA intolerance. We conducted a retrospective observational study to evaluate the effectiveness and safety of GMA in patients with UC and 5-ASA intolerance. Eighty-five patients with UC who underwent GMA were assessed, including 21 with 5-ASA intolerance and 64 with 5-ASA tolerance. This study compared the patient characteristics, treatment outcomes, concomitant drugs, and adverse events between the 2 groups. The remission rate was 28.6% (6/21) in the 5-ASA-intolerant group and 31.3% (20/64) in the 5-ASA-tolerant group, with no significant difference between the groups (P = .967). Both groups showed significant reductions in the dose of oral prednisolone and Lichtiger clinical activity index scores following the GMA. Our study suggests that GMA is equally effective and safe in 5-ASA-intolerant and 5-ASA-tolerant patients, offering an alternative treatment option in this challenging clinical scenario.

1. Introduction

Ulcerative colitis (UC) is a chronic inflammatory bowel disease that primarily affects the colonic mucosa and is characterized by recurring episodes of inflammation and remission; however, its etiology remains elusive.^[1] Recently, various biologics and small molecules have been developed to improve treatment outcomes.^[2–4] Despite these advances, a substantial proportion of patients with mild-to-moderate active UC can achieve and maintain remission using 5-aminosalicylate (5-ASA) formulations alone, and owing to abundant evidence, these formulations are the first-line treatment in the guidelines.^[5,6]

Although 5-ASA formulations rarely result in severe adverse events, 5-ASA intolerance is known to manifest,^[7–20] with documented incidence rates ranging from 5% to 15%.^[7,12–20] The most common symptoms of 5-ASA intolerance are fever, abdominal pain, diarrhea, and bloody stools.^[7–20] Given that 5-ASA formulations are pivotal anchor drugs for inducing and maintaining remission in UC, choosing an alternative treatment is difficult once 5-ASA intolerance is diagnosed.

5-ASA intolerance is associated with a significant increase in hospitalization and steroid administration, early discontinuation of anti-TNF antibody therapy, and an increased risk of surgery.^[16,20] In addition, it is difficult to distinguish between 5-ASA intolerance and an exacerbation of active ulcerative colitis. If treatment for active ulcerative colitis is intensified despite the presence of 5-ASA intolerance, it may lead to unnecessary additions of drug therapies such as steroids, biologics, and small molecule compounds, potentially resulting in various complications such as adverse effects and financial burden. Therefore, careful attention must be paid to the diagnosis and treatment of 5-ASA intolerance.

Granulocyte and monocyte adsorptive apheresis (GMA), which removes granulocytes and monocytes from the peripheral blood, has been used as a non-pharmacological therapy that has anti-inflammatory effects in inflammatory disorders such as UC, Crohn disease (CD), pustular psoriasis, and psoriatic arthritis. Numerous reports have indicated that GMA is a safe and effective treatment option.^[21–31]

In our department, GMA is primarily used as an induction therapy for remission in cases resistant to or dependent on prednisolone (PSL), based on Japanese guidelines.^[6] Additionally, for patients with low disease activity, those with concomitant diseases such as diabetes, or those who wish to avoid PSL administration, GMA is administered in combination with 5-ASA preparations without using PSL. Since the approval of infliximab in 2010 in Japan, various biologics and small molecules have become available, and their prescription has increased in our department, particularly for PSL-resistant or dependent cases. However, the remission induction rate for these drugs ranges from 13% to 50%, indicating a limitation in their therapeutic efficacy.^[4]

GMA is considered a therapeutic option for certain patients such as the elderly and those with comorbidities, due to its lower incidence of adverse events compared to the aforementioned drugs, and it has demonstrated effectiveness and safety.^[30,31] Motoya et al investigated the safety and effectiveness of GMA in patients with IBD who have special situations, such as the elderly and those with comorbidities, and no adverse events or side effects were reported in 8 patients (7 UC cases, 1 CD case) with 5-ASA intolerance.^[30] However, the effectiveness was not reported due to the small number of cases of 5-ASA intolerance in this report. Additionally, only one case report exists on the use of GMA in pediatric patients with UC and 5-ASA intolerance.^[32]

Although GMA is expected to be a therapeutic option for patients with UC and 5-ASA intolerance, evidence supporting the use of GMA in such patients is currently insufficient, indicating a need for further investigation. Clarifying GMA effectiveness and safety could provide a new treatment option for this population. This retrospective observational study aimed to assess the effectiveness and safety of GMA in patients with UC and 5-ASA intolerance by comparing outcomes between those with and without 5-ASA intolerance.

2. Methods

2.1. Patients and therapy protocol

This retrospective observational study reviewed the clinical records of all patients with active UC who underwent GMA at St. Marianna University Hospital (Kawasaki City, Japan) between January 2010 and January 2024. Patients with a Lichtiger clinical activity index (CAI) score of 3 or lower before GMA were excluded. The patients were separated into 2 groups based on whether or not they had 5-ASA intolerance.

The Lichtiger CAI score is a disease activity score for UC that includes parameters such as diarrhea (number of bowel movements per day), nocturnal bowel movements, presence of bloody stools, fecal incontinence, abdominal pain/cramping, general well-being, abdominal tenderness, and need for antidiarrheal drug.^[33] Physicians extracted data from medical records and calculated their scores.

GMA was performed using Adacolumn® (JIMRO Co., Ltd., Takasaki, Japan) as previously reported.^[21] The duration of one GMA session was 60 minutes at a blood flow rate of 30 mL/min. Blood access was via the antecubital vein in one arm and from the Adacolumn® outflow, the blood returned to the patient via a venipuncture in the antecubital vein in the contralateral arm. During GMA, heparin (including low-molecular weight heparin) or nafamostat mesylate was used as the anti-coagulant for continuous infusion during the 60 minutes GMA session. A typical dose for continuous infusion is 2000 U for heparin and 30 mg for nafamostat mesylate. According to the reimbursement scheme in Japan, GMA is administered once or twice per week for 10 sessions (up to 11 sessions in fulminant cases). There are no restrictions on the intervals for GMA, but it is usually administered once or twice a week.^[6,28,34]

2.2. Definition of 5-ASA intolerance

Since the definition of 5-ASA intolerance is not yet clearly established, this study defined 5-ASA intolerance, according to previous reports, as patients in whom 2 or more symptoms, such as those indicating an exacerbation of UC (fever, abdominal pain, diarrhea, bloody stools), extraintestinal complications (skin, joint pain), or an increase in C-reactive protein (CRP) levels following oral 5-ASA administration, with improvement in these symptoms upon discontinuation of the drug.^[7–20]

While a fraction of the patients underwent drug lymphocyte stimulation tests (DLST), those who met the aforementioned definition were classified as intolerant to 5-ASA, regardless of negative results.

The DLST is a drug allergy diagnostic test that measures the level of 3H-thymidine uptake in a combination of sensitized lymphocytes collected from the patient in response to the suspected drug to determine the degree of lymphocyte proliferation.^[35]

2.3. Data collection

Patient data were collected at 2 time points: before the administration of GMA and after the final GMA session.

Before administering GMA, the following patient data were collected: gender, age, disease type (pancolitis, left-sided colitis, and proctitis), concomitant disease, 5-ASA intolerance, Lichtiger CAI score, CRP level, dose of oral PSL, PSL-dependent (relapse during tapering of appropriate steroids or frequent relapse after discontinuation) or resistant (no effect after appropriate steroid administration for 1 to 2 weeks), thiopurine intolerance, and concomitant drugs. After GMA administration, Lichtiger CAI score, CRP level, dose of oral PSL, usage of GMA (number of sessions, frequency), and GMA-induced adverse events requiring its discontinuation were collected.

In patients diagnosed with 5-ASA intolerance, the following data were also collected: the type of 5-ASA formulation (time-dependent, pH-dependent, multi-matrix system [MMX]), dosage, and the results of DLST when 5-ASA intolerance symptoms occurred.

2.4. Outcomes

Clinical remission was defined as a decrease in the Lichtiger CAI score to ≤3 after the final GMA session. Clinical response was defined as a decrease in the Lichtiger CAI score to <10 after the final GMA session and a decrease of ≥3 points from before GMA, or a decrease to ≤3.

2.5. Statistical analysis

Quantitative variables are expressed as medians (interquartile range [IQR]), whereas qualitative variables are expressed as absolute numbers (%). The Mann–Whitney U or Fisher exact test was used to compare the 5-ASA-intolerant and -tolerant groups. We used the Wilcoxon t-test to compare the Lichtiger CAI scores, dose of oral PSL, and CRP levels before and after GMA. Furthermore, the predictive factor for GMA effectiveness was examined by multiple logistic regression. Odds ratios (ORs) with 95% confidence intervals (CIs) were calculated for the selected variables. The statistical software JMP Pro, Ver. 14.0.0 (SAS Inc., Cary) was used.

3. Results

During the study period, a total of 87 patients with UC underwent GMA at our hospital. We excluded 2 cases: one with a Lichtiger CAI score of 3 before GMA, and one initially diagnosed with UC but later diagnosed with familial Mediterranean fever. Thus, 85 patients with UC who underwent GMA were assessed in this study. Of these, 21 cases were 5-ASA-intolerant (including 2 cases of GMA re-administration owing to relapse), and 64 cases were 5-ASA-tolerant (including 14 cases of GMA re-administration owing to relapse), (Fig. S1, Supplemental Digital Content, https://links.lww.com/MD/P494). There were no significant differences between the 2 groups in terms of gender, age, disease type, concomitant disease, Lichtiger CAI score, CRP level, dose of oral PSL, PSL-resistant, and thiopurine intolerance (Table 1). However, the 5-ASA-intolerant group had a considerably higher proportion of PSL-dependent patients than the 5-ASA-tolerant group (47.6% [10/21] vs 21.9% [14/64]; P = .023). Furthermore, the 5-ASA-intolerant group had substantially lower baseline oral and topical 5-ASA administration rates than the 5-ASA-tolerant group (28.6% [6/21] vs 92.2% [59/64], P < .001, and 0% [0/21] vs 23.4% [15/64], P = .017, respectively). Meanwhile, no significant difference was found between the 2 groups in terms of GMA usage (frequency and number of sessions) (Table 2).

Table 1.

Baseline characteristics of patients enrolled in this study.

	5-ASA intolerance (n = 21)	5-ASA tolerance (n = 64)	P
Gender, male/female	9 (42.9)/12 (57.1)	37 (57.8)/27 (42.2)	.314
Age, yr	41 (23–62.5)	43 (30.3–55)	.771
Disease type, pan colitis/left side colitis/rectum	18 (85.7)/2 (9.5)/1 (4.8)	50 (78.1)/14 (21.9)/0	.109
Concomitant disease	0	7 (11.0)	.000
Lichtiger CAI score	11 (7–12.5)	10.5 (8–13)	.623
CRP, mg/dL	0.4 (0.1–1.8)	1.1 (0.2–4.4)	.118
Dose of oral PSL, mg/day	40 (27.5–50)	40 (32.5–50)	.667
PSL-dependent	10 (47.6)	14 (21.9)	.023
PSL-resistant	3 (14.3)	4 (6.3)	.357
Thiopurine intolerance	1 (4.8)	3 (4.7)	1.000

Data are presented as n (%) or median (IQR).

5-ASA = 5-aminosalicylic acid, CAI = clinical activity index, CRP = C-reactive protein, IQR = interquartile range, PSL = prednisolone.

Table 2.

Concomitant medications at baseline and usage of GMA.

		5-ASA intolerance (n = 21)	5-ASA tolerance (n = 64)	P
Concomitant medication	5-ASA or SASP	6 (28.6)	59 (92.2)	<.001
	Topical 5-ASA	0	15 (23.4)	.017
	Oral PSL	10 (47.6)	20 (31.2)	.196
	Topical PSL	2 (9.5)	3 (4.7)	.593
	Thiopurine	5 (23.8)	8 (12.5)	.293
	Calcineurin inhibitor	0	1 (1.6)	1.000
	Anti-TNF	1 (4.8)	4 (6.3)	1.000
	Vedolizumab	3 (14.3)	2 (3.1)	.094
	Ustekinumab	1 (4.8)	2 (1.6)	.435
	Janus kinase inhibitor	0	5 (7.8)	.327
Usage of GMA	GMA sessions per week, 1/2	5 (23.8)/16 (76.2)	17 (26.6)/47 (73.4)	1.000
	Number of GMA sessions	10 (4–10)	10 (5–10)	.238

Data are presented as n (%) or median (IQR).

5-ASA = 5-aminosalicylic acid, GMA = granulocyte monocyte adsorptive apheresis, IQR = interquartile range, PSL = prednisolone, SASP = salazosulfapyridine, TNF = tumor necrosis factor.

In 19 cases of 5-ASA intolerance (excluding 2 cases of GMA re-administration), the following 5-ASA formulations were used when 5-ASA intolerance symptoms were observed: pH-dependent in 10 cases, MMX in 6, time-dependent in 2, and a combination of time-dependent and salazosulfapyridine (SASP) in one (Table 3). The most common intolerance symptom was diarrhea in 14 cases, followed by bloody stools in 12 and abdominal pain in 9 (Fig. S2, Supplemental Digital Content, https://links.lww.com/MD/P494), with many cases exhibiting more than 2 symptoms (Table 3). Of the 19 cases, 15 underwent DLST, and 5 cases tested positive, yielding a positivity rate of 33.3% (5/15) (Fig. S3, Supplemental Digital Content, https://links.lww.com/MD/P494). In terms of the primary outcomes of this study concerning the effectiveness of GMA, there was no significant difference in remission rates between the 5-ASA-intolerant and 5-ASA-tolerant groups (28.6% [6/21] vs 31.3% [20/64], P = .967). Similarly, there was no significant difference in response rates between the 2 groups (66.7% [14/21] vs 71.9% [46/64], P = .783) (Fig. 1). Table 4 shows the median values [IQR] of the Lichtiger CAI scores, dose of oral PSL, and CRP levels before and after GMA for both the 5-ASA-intolerant and 5-ASA-tolerant groups. While the Lichtiger CAI scores and dose of oral PSL were significantly lower in both groups after GMA, there was no significant difference in CRP levels in the 5-ASA-intolerant group (0.4 [0.1–1.8] vs 0.2 [0.1–3.3], P = .700), whereas a significant decrease was observed in the 5-ASA-tolerant group (1.1 [0.2–4.4] vs 0.2 [0.0–0.8], P < .001).

Table 3.

5-ASA regimen at diagnosis of 5-ASA intolerance, and during GMA.

Case	Age group/gender	Disease type	Type and dose (g) of mesalamine at diagnosis	DLST	Symptoms of 5-ASA intolerance	During GMA
1	60s/F	Pan colitis	Time-dependent, 2	N.A.	CRP elevation, bloody stool.	–
2	70s/M	Pan colitis	Multi-matrix, 3.6	Multi-matrix; negative	Abdominal pain, diarrhea	–
3	50s/F	Pan colitis	pH-dependent, 3.6	pH-dependent; negative	Abdominal pain, diarrhea	SASP, 3
				SASP; negative
4	70s/F	Rectum	pH-dependent, 3.6	pH-dependent; negative	Diarrhea, bloody stool, respiratory difficulty	–
5	10s/M	Pan colitis	pH-dependent, 3.6	pH-dependent; negative	Abdominal pain, diarrhea, bloody stool	pH-dependent, 3.6
6	20s/M	Pan colitis	pH-dependent, 3.6	Time-dependent; positive	Increased frequency of bowel movements, abdominal pain, diarrhea, bloody stool, fever	–
				pH-dependent; positive
7	50s/M	Pan colitis	Multi-matrix, 4.8	Multi-matrix; negative	Bloody stool, diarrhea, abdominal pain	–
8	50s/M	Pan colitis	–	–	–	–
9	20s/M	Pan colitis	pH-dependent, 3.6	Time-dependent; positive	Increased frequency of bowel movements, abdominal pain	–
				pH-dependent; positive
10	40s/F	Pan colitis	Multi-matrix, 1.2	N.A.	Abdominal pain, diarrhea, bloody stool	–
11	50s/F	Pan colitis	pH-dependent, 3.6	pH-dependent; negative	Abdominal pain, diarrhea, bloody stool	–
12	20s/F	Pan colitis	Multi-matrix. 4.8	pH-dependent, positive	Abdominal pain, diarrhea, bloody stool	–
				Multi-matrix; positive
13	10s/M	Pan colitis	pH-dependent, 2.4	N.A.	Abdominal pain, diarrhea, bloody stool	SASP, 3
14	60s/M	Pan colitis	pH-dependent, 2.4	pH-dependent; negative	Diarrhea, bloody stool	–
15	40s/F	Left side colitis	Multi-matrix, 4.8	Time-dependent; positive	Diarrhea, bloody stool	Multi-matrix, 4.8
				multi-matrix; negative
16	30s/F	Pan colitis	Multi-matrix, 4.8	N.A.	Vomiting, appetite loss	–
17	20s/M	Pan colitis	pH-dependent, 3.6	pH-dependent; negative	Diarrhea, fever	–
18	50s/F	Pan colitis	pH-dependent, 1.2	N.A.	Increased frequency of bowel movements, bloody stool, fever, headache	–
19	60s/F	Pan colitis	Time-dependent, 4	Time-dependent; positive	Abdominal pain, diarrhea, increased frequency of bloody stool	SASP, 4
				SASP; negative
20	10s/F	Pan colitis	Time-dependent; 1.5	Time-dependent; negative	Fever, diarrhea, bloody stool, respiratory difficulty	–
			SASP, 8	SASP; negative
21	10s/F	Pan colitis	–	–	–	SASP, 3

No. 8 is a retreatment of No. 7, and No. 21 is a retreatment of No. 20.

5-ASA = 5-aminosalicylate, CRP = C-reactive protein, DLST = drug-induced lymphocyte stimulation test, GMA = granulocyte monocyte adsorptive apheresis, N.A. = not applicable, SASP = salazosulfapyridine.

Figure 1.

Remission rate and response rate.

Table 4.

Lichtiger CAI score, dose of oral PSL, and CRP levels before and after GMA.

		5-ASA intolerance	5-ASA tolerance
Lichtiger CAI score	Before GMA	11 (7–12.5), n = 21	10.5 (8–13), n = 64
	After GMA	6 (3–11), n = 21	6 (3–8), n = 64
	P	<.001	<.001
Dose of oral PSL, mg/day	Before GMA	40 (27.5–50), n = 10	40 (32.5–50), n = 20
	After GMA	25 (8.8–42.5), n = 10	20 (11.3–40), n = 20
	P	.016	<.001
CRP, mg/dL	Before GMA	0.4 (0.1–1.8), n = 21	1.1 (0.2–4.4), n = 64
	After GMA	0.2 (0.1–3.3), n = 21	0.2 (0.0–0.8), n = 63
	P	.700	<.001

Data are presented as n (%) or median (IQR).

5-ASA = 5-aminosalicylic acid, CAI = clinical activity index, CRP = C-reactive protein, GMA = granulocyte monocyte adsorptive apheresis, IQR = interquartile range, PSL = prednisolone.

In the univariate analysis of factors contributing to remission among all 85 patients who underwent GMA, the proportion of patients with 5-ASA-intolerance was the same in the remission and non-remission groups (23.1% [6/26] vs 25.4% [15/59], P = 1.000). However, the Lichtiger CAI score before GMA was significantly higher in the non-remission group (11 [9–13] vs 8 [6–13.3], P = .004) (Table 5). Similarly, in the results of the multivariate analysis, the Lichtiger CAI score before GMA (odds ratio [OR]: 0.807, 95% confidence interval [CI]: 0.676–0.945, P = .007), rather than 5-ASA intolerance (OR: 0.719, 95% CI: 0.225–2.302, P = .575), was found to be an independent predictive factor of induction of remission with GMA (Table 6).

Table 5.

Univariate analysis of factors contributing to the remission induction of GMA.

	Remission (n = 26)	Non-remission (n = 59)	P
Gender, male/female	13 (50)/13 (50)	33 (55.9)/26 (44.1)	.644
Age, yr	36.5 (29.3–55.3)	45 (30–58)	.333
Disease type, pan colitis/left side colitis/rectum	20 (76.9)/6 (23.1)/0	48 (81.4)/10 (17.0)/1 (1.7)	.655
CRP, mg/dL	0.5 (0.1–2.5)	1 (0.1–4.2)	.564
Concomitant disease	0	7 (11.0)	1.000
Dose of oral PSL, mg/day	40 (27.5–50)	40 (40–50)	.357
Lichtiger CAI score	8 (6–13.3)	11 (9–13)	.004
5-ASA intolerance	6 (23.1)	15 (25.4)	1.000
Thiopurine intolerance	0	4 (6.8)	.308
PSL-dependent	7 (26.9)	19 (32.2)	1000
PSL-resistant	2 (7.7)	5 (8.5)	1.000
Concomitant 5-ASA or SASP	6 (23.1)	12 (20.3)	.779
Concomitant PSL	13 (50)	17 (28.8)	.085

Data are presented as n (%) or median (IQR).

5-ASA = 5-aminosalicylic acid, CAI = clinical activity index, CRP = C-reactive protein, GMA = granulocyte monocyte adsorptive apheresis, IQR = interquartile range, PSL = prednisolone, SASP = salazosulfapyridine.

Table 6.

Multivariate analysis of factors contributing to the remission induction of GMA.

Variables	Odds ratio	95% CI		P
		Lower	Upper
Age, yr	1.011	0.982	1.040	.462
Lichtiger CAI score	0.807	0.676	0.945	.007
5-ASA intolerance	0.719	0.225	2.302	.575
Concomitant PSL	2.100	0.764	5.770	.151

5-ASA = 5-aminosalicylic acid, CAI = clinical activity index, CI = confidence interval, GMA = granulocyte monocyte adsorptive apheresis, PSL = prednisolone.

In terms of GMA safety, one patient in the 5-ASA-intolerant group discontinued GMA due to anxiety, while 2 patients in the 5-ASA-tolerant group discontinued GMA due to needle puncture discomfort and an allergic reaction to the apheresis system. The incidence rates were 4.8% (1/21) in the 5-ASA-intolerant group and 3.1% (2/64) in the 5-ASA-tolerant group, with no significant difference observed between the 2 groups (P = .742).

4. Discussion

In this study, we evaluated the effectiveness and safety of GMA in patients with UC and 5-ASA intolerance, with data collected from both intolerant and tolerant groups. We found that the effectiveness and safety of GMA in patients with 5-ASA intolerance were comparable to those in patients with 5-ASA tolerance.

To date, the definition of 5-ASA intolerance in UC has not been established. The definition of 5-ASA intolerance in this study was based on previous reports,^[7–20] characterized by the manifestation of symptoms indicating an exacerbation of UC (fever, abdominal pain, diarrhea, bloody stools) or extraintestinal complications (skin, joint pain) following oral 5-ASA administration, with improvement in these symptoms upon discontinuation of the medication. In this study, the confirmed symptoms of 5-ASA intolerance, as illustrated in Figure S1, Supplemental Digital Content, https://links.lww.com/MD/P494, were predominantly diarrhea, bloody stools, abdominal pain, and fever, consistent with previous reports. However, while cases presenting extraintestinal complications (such as skin lesions and joint pain) similar to those reported previously were also defined as 5-ASA intolerance in this study, only one case of respiratory difficulty was observed as an extraintestinal symptom and no clear occurrence of extraintestinal complications was noted. Of the 13 previous reports, extraintestinal complications were reported in 7,^{[13–16,18–20]} while the remaining, similar to our report, did not mention extraintestinal complications. As mentioned earlier, the primary symptoms of 5-ASA intolerance are diarrhea, bloody stools, abdominal pain, and fever, and extraintestinal complications may not always be expected. Furthermore, it was confirmed that intolerance symptoms occurred regardless of the type of 5-ASA formulation, consistent with previous reports.^{[7,10,13–15,17,19]} Additionally, the positive rate of DLST in the 15 cases tested was 33.3%, which is in line with previous reports (Saito et al: 25%, Hiraoka et al: 26.1%, Minagawa et al: 57.1%).^[14,15,19] Based on the comparison with previous reports, the diagnosis of 5-ASA intolerance in this study was considered appropriate.

In this study, the baseline characteristics of the patients before the GMA showed no significant differences in gender, age, disease type, CRP level, concomitant disease, Lichtiger CAI score, dose of oral PSL, thiopurine intolerance, and PSL resistance; however, the proportion of PSL-dependent patients was significantly higher in the 5-ASA-intolerant group. The significantly lower concurrent use of oral 5-ASA formulations and higher use of PSL in the 5-ASA-intolerant group, although not statistically significant, might have led to an increased number of PSL-dependent patients due to the inability to use 5-ASA formulations and the subsequent administration of PSL as alternatives. Mizuno et al reported a significant increase in hospitalization and steroid administration rates in cases of 5-ASA intolerance and significantly higher rates of additional induction of remission with anti-TNF therapy.^[16] Similarly, Hibiya et al reported a significantly higher proportion of patients progressing to total colectomy in patients intolerant to 5-ASA,^[20] suggesting that the inability to use 5-ASA formulations as anchor drugs due to 5-ASA intolerance may pose difficulties in drug therapy and contribute to poor prognosis. These reports were considered supportive of our findings.

In terms of GMA effectiveness, this study reported remission rates in the 5-ASA-intolerant and tolerant groups were 28.6% and 31.3%, respectively, with no significant differences between the 2 groups. Motoya et al particular study, which explored the effectiveness and safety of GMA in patients with UC in special situations, reported remission induction rates of 46.4%,^[30] which was slightly higher than those in our study. Compared with the particular study, the lower remission rate in our study could be attributed to the higher severity of the patients, as reflected by the median Lichtiger CAI scores of 11 indicated a predominance of severe cases,^[33] while in the aforementioned particular study, the median partial Mayo score of 6 indicated a predominant of moderate cases.^[36] This discrepancy in severity levels, with a higher proportion of severe cases in our study, may have contributed to the lower remission rate compared with the particular study. In addition, univariate and multivariate analyses of the factors contributing to remission showed that higher baseline Lichtiger CAI scores had a negative impact on remission induction, indicating that the severity of UC disease activity, rather than 5-ASA intolerance, influences the effectiveness of GMA.

Furthermore, the changes in each efficacy parameter, as shown in Table 4, both Lichtiger CAI scores and dose of oral PSL were significantly reduced in both groups after GMA. Given that steroids can cause a range of adverse effects, including osteoporosis,^[37,38] it is important to reduce its dosage promptly. The PSL-sparing effect of GMA has been reported,^{[21,25,27,29]} and in the present study, we also confirmed the PSL-sparing effect of GMA regardless of 5-ASA intolerance. In contrast, CRP levels decreased substantially only in the 5-ASA-tolerant group and not in the 5-ASA-intolerant group. Although the baseline CRP levels were numerically higher in the 5-ASA-tolerant group compared to the 5-ASA-intolerant group, this difference was not statistically significant. Notably, the 5-ASA-tolerant group included 7 cases with extraintestinal complications, all with high CRP levels (data not shown). Furthermore, posttreatment CRP levels decreased to 0.2, which is considered negative, in both groups. This suggests that GMA is effective in reducing CRP levels to a certain extent, regardless of 5-ASA intolerance. However, CRP levels do not necessarily reflect the disease activity of UC,^[39,40] indicating limitations in evaluating effectiveness based solely on CRP levels.

Regarding safety, the incidence rates of adverse events that resulted in the discontinuation of GMA were sufficiently low in both the 5-ASA-intolerant group and the 5-ASA-tolerant group, with no significant difference between the 2. Furthermore, the abovementioned particular study reported no adverse events in 7 patients with 5-ASA intolerance treated with GMA.^[30] Furthermore, Yamamoto et al reported a discontinuation rate of 5% due to disease exacerbation or adverse events in a retrospective observational study involving 894 UC patients.^[27] In our study, the incidence of adverse events necessitating treatment discontinuation (4.8%) was comparable to the reported rate of serious adverse events and discontinuation.

In recent years, with the launch of various biologics and small molecules, the treatment options for UC have expanded. In this context, our reported remission rate of 28.6% for GMA in 5-ASA-intolerant patients falls within the range of the 6-week remission rates for these drugs (13–50%),^[4] demonstrating that GMA is an effective treatment option in the current therapeutic environment. Additionally, the confirmed PSL-sparing effect of GMA highlights its potential to mitigate steroid dependence. The incidence of adverse events in GMA showed no significant difference between the 5-ASA-intolerant and 5-ASA-tolerant groups, and it was comparable to the reported rates of adverse events and treatment discontinuation. This confirms that GMA can be safely administered to 5-ASA-intolerant patients. These results demonstrate the effectiveness and safety of GMA for 5-ASA-intolerant patients, providing valuable insights for treatment decisions for these patients.

This study is a retrospective observational analysis of a single trial; hence, it has certain limitations. The absence of restrictions on GMA and drug selection may result in bias in their selection. Endoscopic data, histological findings, and biomarkers such as calprotectin were not included in the treatment efficacy evaluation items, which may have resulted in an inadequate evaluation of the efficacy of GMA. The relatively small sample size may have resulted in a loss of statistical power. Prospective studies with larger numbers of patients are required to confirm the results of this study. Furthermore, because GMA has been covered by insurance as maintenance therapy since 2022, it is vital to evaluate the long-term prognosis of patients with 5-ASA intolerance treated with GMA maintenance therapy. Nevertheless, the strength of this study lies in its evaluation of the efficacy of GMA in patients with 5-ASA intolerance for the first time.

In conclusion, in this study, we demonstrated that the effectiveness and safety of GMA in 5-ASA-intolerant patients are comparable to those in 5-ASA-tolerant patients. Consequently, GMA is a viable treatment option for UC patients intolerant to 5-ASA in the current therapeutic environment. However, further studies with larger patient populations are needed to validate our findings.

Acknowledgments

The authors would also like to thank Editage (www.editage.com) for English language editing.

Author contributions

Conceptualization: Masaki Kato.

Data curation: Masaki Kato, Masaki Yamashita, Kenichi Araki, Jun Ishida, Hiroshi Yasuda.

Formal analysis: Masaki Kato.

Investigation: Masaki Yamashita, Mitsuto Tsukui, Yoshihiko Iijima, Kenichi Araki, Jun Ishida, Takumi Komatsu, Yusuke Nakamoto, Akiyo Kawashima, Maki Konno, Hirofumi Kiyokawa.

Supervision: Yoshinori Sato, Tsutomu Sakurada, Tadateru Maehata, Hiroshi Yasuda, Keisuke Tateishi.

Writing – original draft: Masaki Kato.

Writing – review & editing: Masaki Yamashita, Shigeki Kojima, Mitsuto Tsukui, Yoshihiko Iijima, Kenichi Araki, Jun Ishida, Takumi Komatsu, Yusuke Nakamoto, Akiyo Kawashima, Maki Konno, Hirofumi Kiyokawa, Yoshinori Sato, Tsutomu Sakurada, Tadateru Maehata, Hiroshi Yasuda, Keisuke Tateishi.

Supplementary Material

medi-104-e43389-s001.pptx ^{(192.3KB, pptx)}