炎症性肠病患者孕期安全使用硫唑嘌呤、TNF抑制剂单药治疗

芝加哥——旧金山加州大学的Uma Mahadevan博士在2014年美国消化疾病周上报告的一项对1,289妇女和1,039活产婴儿进行的研究发现，炎症性肠病（IBD）女性患者怀孕期间能够继续安全使用硫唑嘌呤/6-巯基嘌呤或肿瘤坏死因子-α（TNF-α）抑制剂单药治疗以持续缓解症状。而暴露于硫唑嘌呤/6-巯基嘌呤和TNF-α抑制剂的婴幼儿发育里程碑评分达标比例与未暴露的婴儿相比类似；在出生的所有婴儿中，共有55例婴儿被诊断为先天性异常。与无药物暴露的胎儿相比（12例），单独暴露于硫唑嘌呤、TNF抑制剂或者同时暴露于两种药物的胎儿先天畸形发生率并没有更高（分别为17例、7例和19例）。

Uma Mahadevan博士

该分析是基于PIANO的注册数据进行的一项前瞻性队列研究，研究者在患者孕期的每三个月、生产时、生产后第4、9、12个月通过电话或线下对IBD妊娠患者进行问卷随访调查，同时每年收集他们的孩子相关数据，直到4岁。共7例妊娠妇女在孕期被诊断为IBD，59.4％被诊断为克罗恩病，38.3％为溃疡性结肠炎，还有2.3％诊断未明确。暴露定义为在妊娠前3个月直至妊娠结束期间任意时间使用过硫唑嘌呤/6-巯基嘌呤或TNF-α抑制剂，包括英利昔单抗（Remicade，强生），阿达木单抗（Humira，艾伯维），赛妥珠单抗（Cimzia，优时比）。

有三分之二的患者选择母乳喂养，与药物暴露的女性相比，未暴露女性更倾向母乳喂养（P<0.001）。在控制药物暴露后，母乳喂养与婴儿感染或身高、体重降低的风险增加无关。

而同时暴露在两种药物中的婴幼儿早产率较高（OR，2.6，P<0.05）。溃疡性结肠炎的母亲接受联合治疗则增加了婴幼儿的早产率（OR ，4.9）、低出生体重率（OR ，6.1）、以及新生儿重症监护病房的留观率（OR，3.9）。Mahadevan博士认为，出现该现象的原因之一可能是患溃疡性结肠炎患者在孕期的疾病活动更加活跃导致，另外一个可能则是溃疡性结肠炎母亲存在低水平的未处理炎症。

PIANO研究者曾对161女性进行的分析发现，同时暴露英夫利昔单抗​​或阿达木单抗和硫唑嘌呤药物中的婴幼儿在12月龄时感染几率显著增加。大部分TNF抑制剂在第二和第三孕期能够透过胎盘，这引起了人们对免疫系统发育和继发性感染风险的担忧。

而在近期的分析中，研究者们并没有发现TNF抑制剂增加婴儿感染率的证据。 “但是，当我们剔除赛妥珠单抗再进行分析时，这种显著性则与2年前的结果一致，”Mahadevan博士在接受采访时说。 “赛妥珠单抗并不能主动透过胎盘。幸运的是，大部分感染都是较轻微的，包括中耳炎和上呼吸道感染。”

发育里程碑

暴露于硫唑嘌呤/6-巯基嘌呤和TNF-α抑制剂的婴幼儿发育里程碑【基于丹佛儿童发育筛查测验（DDST）和年龄与发育进程问卷（ASQ-3）】评分达标比例与未暴露的婴儿类似。

在控制了早产因素后，未暴露的4月龄婴儿平均发展里程碑得分为0.92分，暴露于硫唑嘌呤、TNF抑制剂和两种药物中婴儿的平均​​增量差异分别为0.01、0.01和0.00分，无显著性差异。

Mahadevan博士说，出乎意料的是，单独暴露于TNF抑制剂的婴儿12月龄时的评分显著优于未暴露于药物中的婴儿。“我不会认为这个数据代表着药物使你的宝宝更聪明，关键是它并没有使情况变得更糟，”她指出。

ASQ-3问卷显示，暴露组在36个月时的粗大运动技能平均分显著优于未暴露组（55.94 vs. 47.53，P = 0.01），而48个月的粗大运动技能评分（60与51.20；P = 0.02），48个月精细动作技能评分（53.50 vs. 42.11，P = 0.01），以及48个月个人/社会互动能力评分（58 vs. 51.93，P=0.04）也都稍高于未暴露组。

Mahadevan博士说，硫唑嘌呤暴露组的评分总是优于未暴露组，包括24个月的个人/社会互动能力评分（50.75 vs. 47.34，P=0.04），36个月的解决问题能力评分（平均52.04 vs. 48.66，P= 0.05）以及48个月的解决问题能力评分（平均59.92 vs. 57.66，P = 0.02），均具有统计学显著性。

同时，最近的一份报告发现IBD母亲的孩子出现注意力及多动症障碍（ADHD）和粗大运动异常的比例较高（J. Crohns Colitis 2013;7:542-50），另一个病例系列则报道，连续入组的25例孩子在子宫内暴露于TNF-α抑制剂中的神经精神发育正常（Inflamm. Bowel Dis. 2014;20:495-501）。另一项对58文章或摘进行的系统综述则发现，TNF-α抑制剂的使用与妊娠合并IBD的不良妊娠转归、先天性异常、第1年感染均无相关性（BMC Medicine 2013;11:174）。

该研究由美国克罗恩病和结肠炎基金会支持，Mahadevan博士报告担任杨森、AbbVie，武田和UCB制药公司的顾问，并接受普罗米修斯实验室和葛兰素史克提供的研究经费。

爱思唯尔版权所有  未经授权请勿转载

By: PATRICE WENDLING, Ob.Gyn. News Digital Network

AT DDW 2014

VITALS

Key clinical point: Monotherapy with azathioprine and anti-TNF agents can be continued in pregnancy with IBD to maintain remission.

Major finding: Congenital anomalies were not higher in children with exposure to azathioprine, anti-TNF agents, or combination therapy, compared with no drug exposure (12 vs. 17 vs. 7 vs. 19 anomalies).

Data source: Analysis of 1,289 women with IBD and 1,039 live births.

Disclosures: The study was supported by the Crohn’s and Colitis Foundation of America. Dr. Mahadevan reported serving as an advisor for Janssen-Cilag, AbbVie, Takeda, and UCB Pharma, and receiving research grants from Prometheus Laboratories and GlaxoSmithKline.

CHICAGO – Women with inflammatory bowel disease can safely continue monotherapy with azathioprine/6-mercaptopurine or antitumor necrosis–alpha agents during pregnancy to maintain remission, according to an analysis of 1,289 women and 1,039 live births.

Infants exposed to azathioprine/6 MP and anti-TNF-alpha agents achieve developmental milestone scores at rates similar to unexposed infants and do not have higher rates of congenital anomalies, lead author Dr. Uma Mahadevan said at the annual Digestive Disease Week.

Overall, there were 55 congenital anomalies, with 21 diagnosed at birth. Events were similar among infants with in utero exposure to azathioprine, anti-TNF agents, combination therapy, and those with no drug exposure (12 vs. 17 vs. 7 vs. 19 events).

Exposure was defined as any use of azathioprine/6 MP or the anti-TNF-alpha agents infliximab (Remicade), adalimumab (Humira), or certolizumab pegol (Cimzia) at any time from 3 months prior to conception to the end of the pregnancy. Nine women had exposure to the recombinant monoclonal antibody natalizumab (Tysabri) and were included in the anti-TNF agents group.

The analysis was based on the PIANO registry, a prospective cohort of pregnant women with IBD who were followed by telephone or in-person questionnaire at every trimester, at delivery, at 4, 9, and 12 months after delivery, and annually for the first 4 years of their child’s life. Seven women received their diagnosis of IBD during pregnancy, 59.4% had Crohn’s disease, 38.3% ulcerative colitis, and 2.3% were indeterminate.

Two-thirds of women reported breastfeeding, with unexposed women significantly more likely to breastfeed than women with exposure to azathioprine, anti-TNF agents, or combination therapy (85% vs. 65% vs. 71% vs. 61%; P less than .0001).

After controlling for drug exposure, breastfeeding was not associated with an increased risk of infant infection or reduced height or weight, reported Dr. Mahadevan, with the University of California San Francisco.

Infants exposed to combination therapy, however, had higher rates of preterm birth, after adjustment for none/mild vs. moderate/severe IBD activity (odds ratio, 2.6; P less than .05).

Infants exposed to combination therapy who were born to mothers with ulcerative colitis also had increased rates of preterm birth (odds ratio, 4.9), low birth weight (OR, 6.1), and NICU stay (OR, 3.9).

One possible reason for this signal is that women with ulcerative colitis in the PIANO registry, as well as two other studies, have more disease activity during pregnancy, Dr. Mahadevan suggested. Another possibility is that mothers with ulcerative colitis may have low levels of untreated inflammation.

With just 161 women available for analysis, the PIANO investigators previously reported a significant increase in infant infections at 12 months of age in those exposed to combination infliximab or adalimumab plus azathioprine. Most anti-TNF agents cross the placenta in the second and third trimester, which has raised concerns about immune system development and risk of subsequent infections.

In the current analysis, there was no increase in infant infections when the investigators looked at all anti-TNF drugs. "But when we removed certolizumab from the analysis, it was significant just as it was 2 years ago," Dr. Mahadevan said in an interview. "Certolizumab is not actively transported across the placenta. Fortunately, the majority of the infections were minor, including otitis media and upper respiratory infections."

Developmental milestones

Overall, infants exposed to azathioprine or anti-TNF agents were found to achieve developmental milestones at rates similar to unexposed infants of mothers with IBD based on Denver Childhood Developmental Test and Ages and Stages Questionnaire scores, she said.

After controlling for preterm birth, the mean developmental milestone score at 4 months was 0.92 for infants with no drug exposure, with a nonsignificant mean delta difference of 0.01, 0.01, 0.00 points for infants exposed to azathioprine, anti-TNF agents, and combination therapy, respectively.

In a surprising twist, infants exposed to anti-TNF agents alone actually had significantly better scores at 12 months than unexposed infants, bettering the average score of 0.81 in the unexposed group by a mean delta of 0.02 points (P = .01), Dr. Mahadevan said.

"I wouldn’t interpret this data [as indicating] that biologics make your baby smarter. I think the key thing is that it doesn’t make it worse," she remarked. "When you see the deltas, they are pretty small."

On the Ages and Stages Questionnaire, the biologic group had significantly better mean scores than the unexposed group for gross motor skills at 36 months (55.94 vs. 47.53; P = .01), gross motor skills at 48 months (60 vs. 51.20; P = .02), fine motor skills at 48 months (53.50 vs. 42.11; P = .01), and personal/social interaction at 48 months (58 vs. 51.93; P = .04).

The azathioprine-exposed group always had better scores than the unexposed group, Dr. Mahadevan said. Where it was statistically significant was for personal/social interaction at 24 months (50.75 vs. 47.34; P = .04), problem solving at 36 months (mean 52.04 vs. 48.66; P = .05), and problem solving at 48 months (mean 59.92 vs. 57.66; P = .02).

Dr. Mahadevan observed that the data may change because not all infants have reached each time point and may not reflect outcomes in lower socioeconomic groups since more than 90% of women were of similar socioeconomic and educational status.

Still, the findings are reassuring in the absence of any randomized controlled trials. A recent report found children born to mothers with IBD had higher rates of attention-deficit hyperactivity disorder and gross motor abnormalities (J. Crohns Colitis 2013;7:542-50), while a case series reported normal neuropsychological development in 25 consecutive children exposed in utero to TNF-alpha inhibitors (Inflamm. Bowel Dis. 2014;20:495-501), she noted.

A recent systematic review of 58 articles or abstracts also found no association between TNF-alpha inhibitor use during pregnancy with IBD and adverse pregnancy outcomes, congenital abnormalities, or infections in the first year of life (BMC Medicine 2013;11:174).

The study was supported by the Crohn’s and Colitis Foundation of America. Dr. Mahadevan reported serving as an advisor for Janssen-Cilag, AbbVie, Takeda, and UCB Pharma, and receiving research grants from Prometheus Laboratories and GlaxoSmithKline.