Society: AGA

Background: The impact of pregnancy on vedolizumab (VDZ) and ustekinumab (UST) pharmacokinetics is not well defined. Similarly, the time to infant clearance and outcomes following in-utero exposure require elucidation. We aim to define the stability of UST and VDZ levels in pregnancy, placental drug transfer, and the time to and factors influencing infant drug clearance.
Methods: This multicentre prospective cohort study recruited women with IBD who were pregnant or planning pregnancy and receiving VDZ or UST. Trough drug levels, and clinical and biochemical data were documented pre-conception and in each trimester (T) of pregnancy. Maternal and infant drug levels were obtained at delivery and repeated in infants until clearance was acheived. Infant outcomes were assessed to 2 years of age. Drug levels were measured by ELISA (Theradiag) with a lower limit of detection of 0.25 mg/mL (VDZ) and 0.4 mg/mL (UST).
Results: 97 women aged median 31 (IQR 29-34) y with 54 receiving VDZ (41,17 with Crohn’s disease respectively) were included. Clinical and biochemical disease remission was maintained in the majority (Table 1). Median VDZ levels decreased by 42% from T1 to T3 (p<0.001; Skilling’s Mack) without a corresponding increase in median faecal calprotectin (T1 21.5ug/g (14-110) vs T3 59 (17-131) p=0.65; Wilcoxon signed-rank). UST levels were stable over the course of pregnancy (p=0.14) (Fig 1). 82% (VDZ) and 78% (UST) of participants continued drug into T3. Favourable delivery and infant outcomes, comparable to previously reported cohorts, were observed (Table 1). Infant:maternal drug level ratio was 1.74 (1.24-3.5, n=35) for UST and 0.71 (0.52-0.91, n=44) for VDZ. Delivery levels in infants correlated positively with concomitant maternal levels and gestation at time of final drug dose (Table 1). 67% UST and 90% VDZ exposed infants cleared drug by 15 weeks (p=0.06; two sided test of proportions), with a time to clearance of 13 (9-22) weeks for UST and 11 (8-13) weeks for VDZ (p=0.11; non-parametric comparison of medians). Time to clearance correlated positively with infant delivery level (Table 1). There was no difference in the time to infant clearance between those receiving their last dose in T2 vs T3 (UST p=0.45 or VDZ p=0.44)
Conclusion: UST levels are stable over the course of pregnancy. VDZ levels fall but without a parallel increase biochemical disease activity. Proactive dose adjustment and level monitoring during pregnancy is therefore not necessary. Differences in placental transfer of UST and VDZ were confirmed, with the infant:maternal drug level ratio at delivery lower for VDZ than UST. Both are cleared prior to 15 weeks in the majority of infants, but the higher placental transfer of UST results in numerically fewer exposed infants clearing by this timepoint. No concerning infant outcomes were identified in this cohort.

Trough drug levels across pregnancy

Introduction: The peak age of onset for inflammatory bowel disease (IBD) overlaps with women’s reproductive age. There has also been an increase in IBD in the pregnant population in recent years. Studies suggest women with IBD may have worse maternal and fetal outcomes and increased hospital resource utilization during pregnancy. We hypothesize that insurance status as a social determinant of health (SDOH) may play a role in these outcomes.

Methods: The National Inpatient Sample (NIS) from 2016-2017 was analyzed to identify all pregnant women and stratified by IBD diagnosis based on ICD 10 codes. Maternal outcomes of interest were hypertensive disorders, gestational diabetes, post-partum hemorrhage, venous thromboembolism, postpartum endometritis, spontaneous abortion, missed abortion, post-abortion complications, ectopic pregnancy, polyhydramnios, hyperemesis, mental disorder during pregnancy, prolonged pregnancy, and length of stay. Fetal outcomes of interest were preterm delivery, fetal growth restriction, fetal death, fetal distress, cord complications and large for gestational age. Univariate analyses were conducted to determine the effect of insurance status (public, private, self-pay) on maternal and fetal outcomes. Outcomes were adjusted for age, race, hospital bed size, location/teaching status of hospital, hospital region, median household income, and Charlson Comorbidity index.

Results:
There was a total of 17,144 records for hospitalized pregnant women with IBD, of which 4,970 had public health insurance, 12,024 had private health insurance and 150 were self-pay (Table 1). After adjusting for potential confounders, pregnant women with IBD with public health insurance had increased odds of hospitalization for pregnancy-related mental disorders, hyperemesis, fetal growth restriction, and spontaneous abortion and lower odds of hospitalization for prolonged pregnancy than those with private insurance. The mean length of stay (LOS) for pregnant IBD patients was higher than those without IBD. Patients with public insurance had a higher mean LOS than other insurance types regardless of IBD status (Table 2).

Conclusion: NIS data suggest an association with increased adverse maternal/fetal outcomes for pregnant IBD patients with public compared to private health insurance. While public health insurance appears to have a protective factor for prolonged pregnancy, the increased odds of the other adverse outcomes described are likely of higher threat to maternal and fetal health. Other SDOH factors may also have a role in these outcomes among pregnant women with IBD and further studies are warranted.

Background:
The risk of anal squamous cell carcinoma (SCC) is elevated in Crohn’s disease (CD). Other risk factors for anal cancer include HIV, MSM (men who have sex with men) status, smoking, and age. Few studies have investigated IBD as a risk factor for anal cancer. None have investigated the long-term prevalence of precursor high grade squamous intraepithelial lesions (HSIL) and anal cancer in patients with IBD who identify as MSM. This study aims to compare the prevalence of HSIL and anal cancer in patients with IBD who identify as MSM to those without IBD through a retrospective cohort study design. Secondly, we aim to evaluate how often MSM patients with IBD seen at a large, tertiary IBD clinic had at least 1 visit to the associated anal SIL (aSIL) screening clinic.
Methods:
We analyzed data from the electronic health record (EHR) to identify all MSM patients receiving care at a referral clinic for screening and treatment of aSIL, and all patients at a tertiary IBD clinic who identify as MSM. MSM status was determined by natural language processing of clinic notes and social history information. We extracted demographic data, history of HIV, tobacco status, and IBD medication use from the EHR. HSIL and anal cancer history was determined by extraction of pathology and cytology reports augmented by manual chart review. IBD history was obtained via diagnosis codes and confirmed with review of clinical notes.
Results:
Of the 4675 patients who identify as MSM seen at the aSIL clinic, 57 had IBD (27 Ulcerative Colitis, 30 Crohn’s Disease). Of the patients with IBD, 28 (49.1%) were living with HIV (LWH) vs. 3222 (69.8%) of the patients without IBD. 23 (43.4%) patients with IBD had a positive tobacco use history vs. 1849 (45.2%) patients without IBD. IBD medication exposure is listed in table 2. HSIL was found in 25 (43.9%) patients with IBD and 2417 (52.3%) without IBD. Anal cancer was found in 2 (3.6%) patients with IBD and 139 (3.0%) without. Of the 3203 men seen at the IBD referral clinic, 145 identified as MSM and of these patients, 20 (13.8%) had been seen in the aSIL clinic.
Discussion:
In this group of patients identifying as MSM evaluated at a referral clinic for aSIL, history of IBD was not associated with increased prevalence of HSIL or anal cancer. Among all risk factors, HIV status had the strongest association with HSIL and anal cancer prevalence. HIV was less prevalent in the IBD group, but when controlling for HIV in a multivariable analysis, we did not find an association between IBD history and HSIL or anal cancer. Further analysis will identify any referral pattern differences between groups that may have affected the results. Future studies on the role of biologics and immunosuppressants in a larger population are needed to guide screening recommendations.

Table 1: Prevalence of HSIL or Anal Cancer compared to no history of HSIL or Anal Cancer among patients with and without a history of Inflammatory Bowel Disease, with analysis adjusted for HIV prevalence.

Table 1: Prevalence of HSIL or Anal Cancer compared to no history of HSIL or Anal Cancer among patients with and without a history of Inflammatory Bowel Disease, with analysis adjusted for HIV prevalence.

Table 2: Characteristics of MSM patients with Inflammatory Bowel Disease.

Table 2: Characteristics of MSM patients with Inflammatory Bowel Disease.

Background and Aims: Dietary fibers can modulate gut dysbiosis and could have benefits in inflammatory bowel disease (IBD), but data is limited. This prospective longitudinal study aimed to investigate the effect of dietary fiber intake on clinical relapse rate, changes in fecal calprotectin levels, and gut microbiota in clinically quiescent IBD patients.
Methods: Sixty-one IBD patients (23 Crohn's disease and 38 ulcerative colitis) with clinical remission were enrolled. All were required to have unchanged medical treatment in the last 2 months. Baseline clinical parameters, dietary intake, fecal calprotectin, and fecal microbiota were obtained. The diet was assessed by a nutritionist using food photographs 3 days a week. Fecal microbiota was analyzed using 16S rRNA analysis. At enrollment and every visit, the nutritionist advised all patients to increase their fiber intake. All patients had clinical and diet assessments at week 6 and every 12 weeks until the development of clinical flare requiring medical treatment escalation or completing the study at 1 year. During follow-up, fecal calprotectin was measured at weeks 6, 30, 54, and at the time of clinical flares. Fecal microbiota was re-analyzed at the end of the study. Patients were categorized into 3 groups based on a significant change in dietary fiber consumption, which was defined as a relative change in mean fiber intake of >25% from baseline.
Results: At baseline, the average dietary fiber consumption was 10.7 g a day. During follow-up, 19 (31.1%) patients significantly increased their fiber intake, 30 (49.2%) patients did not change their fiber intake, and 12 (19.7%) patients significantly decreased their fiber intake. The overall cumulative rate of the clinical flare was 22.6%. The cumulative clinical flare rate was 12.6%, 20.5%, and 46.5% in the group with increased, unchanged, and decreased fiber intake, respectively, with a significant difference observed between the group with increased and decreased fiber intake (p=0.03) (Figure 1). Fecal calprotectin had a trend of decreasing in the group with increased and unchanged fiber intake, while the trend was increasing in the group with decreased fiber intake, but not statistically significant. For microbiota analysis, as shown in Figure 2, there was a significant increase in the whole Clostridia class in the increased fiber intake group, but not in the other groups. The other changes, including the increase in genus Blautia, Doria, Coprococcus, and Ruminococcus and the decrease of several members of the order Lactobacillales were observed in all groups.
Conclusions: Increasing dietary fiber consumption may improve microbial dysbiosis and prevent clinical flares in some clinically quiescent IBD patients.

<b>Figure 1</b>. Cumulative rate of clinical flare in patients with increased, unchanged, and decreased fiber intake

Figure 1. Cumulative rate of clinical flare in patients with increased, unchanged, and decreased fiber intake

<b>Figure 2</b>. LEfSe cladograms comparing before and after the intervention of decreased, unchanged, and increased fiber intake groups

Figure 2. LEfSe cladograms comparing before and after the intervention of decreased, unchanged, and increased fiber intake groups

Background
Expeditious initiation of biologic or novel small molecule therapy is important in inflammatory bowel disease (IBD), namely Crohn’s disease (CD) and ulcerative colitis (UC). However, starting therapy in the outpatient setting may be challenged by various clinical, social, and financial variables which can significantly delay treatment initiation. In addition, the COVID-19 pandemic, which negatively impacted healthcare utilization, may have slowed approval or commencement of therapy. We aimed to identify factors that may contribute to delays in initiation of biologics or small molecules for treating patients with IBD.
Methods
This was a multi-center retrospective study. Data was collected from three large tertiary academic medical centers - University of Vermont Medical Center (UVM), Beth Israel Deaconess Medical Center (BIDMC), and University of Mississippi Medical Center (UM). Patient who received first-time outpatient prescriptions of biologics or novel small molecule therapies from 3/1/2019 to 9/30/20 were included in the study. Univariate and multivariate linear regression analysis was performed to identify variables associated with a delay in biologic or small molecule treatment initiation. Variables included demographic, clinical, insurance, prior authorization, and facility information as well as pre-pandemic (3/1/2019 - 2/29/2020) vs. post-pandemic (3/1/2020 - 9/30/2020) period.
Results
In total 411 patients (CD, n=276; UC, n=129) were included in the analysis. (UVM, n=123; BIDMC, n=165; and UM, n=123). Of these, 68% were covered by commercial insurance, 31% were Medicare or Medicaid, and 1% were self-pay. The median [interquartile range-(IQR)] delay was 20 [12-37] days (infliximab, 19 [13-33] days; adalimumab, 10 [5-26] days; vedolizumab, 21 [14-42] days and ustekinumab, 21 [14-42] days). Variables associated with a delay in biologic or small molecule treatment initiation are described in Table 1. Blacks had the longest delay with a median [IQR] of 34 [17-53] days and were the only racial group with a median delay >20 days (Table 2). UM had the longest delay compared to BIDMC and UVM (Table 2). Distance impacted delay, with those living ≥49.5 miles from the medical center having a median delay of 30 days compared to 18 for those closer. Patients who were declined initial approval by insurance had a median of 51 (IQR 27-84) days delay compared to initial approval of 18 days (IQR 11-33). Finally, the COVID-19 pandemic in our population saw an improvement of a median delay from 22 days to 18 days (Table 2).
Conclusion
This large retrospective study showed that there may be a significant delay of a median 20 days in biologic or small molecule treatment initiation. The most important variables associated with this delay included patient’s race, the distance to the medical center, and the initial coverage.

Table 1. Variables associated with delay in biologic or small molecule treatment initiation.

Table 1. Variables associated with delay in biologic or small molecule treatment initiation.

Table 2. Delay in biologic or small molecule treatment initiation according to variables identified from multivariate linear regression analysis.

Table 2. Delay in biologic or small molecule treatment initiation according to variables identified from multivariate linear regression analysis.

Anna Owings

Appearances