Perianal fistulising Crohn's disease (PFCD) is a common yet challenging phenotype of Crohn's disease, due to its refractory nature and substantial impact on patients' health and quality of life [1]. PFCD is also associated with high direct health care costs that remain high for years beyond the initial presentation of perianal fistulas [2]. While studies have demonstrated the effectiveness of anti-tumour necrosis factor (TNF) therapy for this condition, a substantial proportion of patients remain symptomatic [3]. As a result, innovative treatment strategies are needed.

Bacsur and colleagues reported on the effectiveness and safety of locally injected allogenic, adipose-derived mesenchymal stem cells (MSC) (Darvadstrocel) for treatment of complex PFCD [4]. Like the ADMIRE trials, patients underwent fistula tract conditioning by “rigorous” curettage and seton placement before surgical closure of the internal fistula opening and local injection of Darvadstrocel [5, 6]. Among the 204 patients who underwent treatment in this real-world study, 62% achieved fistula closure at week 52. Importantly, closure was determined clinically without radiologic evaluation. Furthermore, the proportion of patients with a perianal disease activity score (PDAI) of > 4 decreased from 58% at baseline to 20% at week 52 (p < 0.001). Only 14% of patients experienced an adverse event requiring hospitalisation; most involved perianal abscess or pain.

The impressive rate of fistula closure reported in this study was higher than those reported in the ADMIRE trials (56% in ADMIRE I and 43% in ADMIRE II) [6, 7]. In addition to the open-label study design, the study population probably had a lower fistula disease burden overall given that 42% of patients had a PDAI score of 4 or lower at baseline. Furthermore, only 43% of patients had multiple fistula tracts, and 14% had branching tracts, both of which have been shown to be predictive of anti-TNF failure in PFCD [8].

While these results are promising, this study was uncontrolled. As a result, it is unclear if fistula closure was a result of MSC therapy itself, the surgical intervention that accompanied MSC treatment (curettage and closure of the internal fistula track opening), and/or the natural history of PFCD after seton removal. These are important considerations considering the results of the ADMIRE-II trial. This was a large phase III randomised controlled trial, which found that MSC therapy (Darvadstrocel) administered after fistula tract conditioning and surgical closure was no more effective in achieving combined clinical and radiologic remission at week 24 than fistula tract conditioning and surgical closure alone (49% vs. 46%, p = 0.57) [6]. In contrast, in ADMIRE-I, MSC therapy was significantly more effective than fistula tract conditioning and surgical closure alone (50% vs. 34%, p = 0.024) [5].

Given the tremendous cost of this therapy, future research should prioritise understanding the reasons for the discrepancy in the ADMIRE trials. Furthermore, comparative effectiveness studies to determine the optimal surgical approach to PFCD (fistula conditioning, epithelial ablation techniques and/or closing the internal fistula opening), and the natural history of PFCD after seton removal are required. Answers to these questions will help to determine if MSC therapy truly adds benefit beyond fistula tract conditioning and surgical closure.

肛周瘘是克罗恩病的一种常见但具有挑战性的表型，因为它具有难治性，并且对患者的健康和生活质量有重大影响 [1]。PFCD 还与高昂的直接医疗保健费用有关，在肛周瘘的初始表现后数年内，这些费用仍然很高 [2]。虽然研究已经证明了抗肿瘤坏死因子 （TNF） 治疗对这种情况的有效性，但很大一部分患者仍然存在症状 [3]。因此，需要创新的治疗策略。

Bacsur 及其同事报道了局部注射的同种异体脂肪来源的间充质干细胞 （MSC） （Darvadstrocel） 治疗复杂 PFCD 的有效性和安全性 [4]。与 ADMIRE 试验一样，患者在手术闭合内瘘口和局部注射 Darvadstrocel 之前，通过“严格”刮除术和挂线放置接受了瘘管预处理 [5， 6]。在这项真实世界研究中接受治疗的 204 名患者中，62% 的患者在第 52 周实现瘘管闭合。重要的是，临床确定闭合，无需放射学评估。此外，肛周疾病活动评分 （PDAI） 为 > 4 的患者比例从基线时的 58% 下降到第 52 周的 20% （p < 0.001）。只有 14% 的患者经历了需要住院治疗的不良事件;大多数涉及肛周脓肿或疼痛。

本研究中报告的瘘管闭合率令人印象深刻，高于 ADMIRE 试验中报告的瘘管闭合率（ADMIRE I 为 56%，ADMIRE II 为 43%）[6， 7]。除了开放标签研究设计外，鉴于 42% 的患者在基线时 PDAI 评分为 4 或更低，研究人群的瘘管疾病总体负担可能较低。此外，只有 43% 的患者有多条瘘管，14% 的患者有分支束，这两者都被证明可以预测 PFCD 中抗 TNF 失败 [8]。

虽然这些结果很有希望，但这项研究是不受控制的。因此，目前尚不清楚瘘管闭合是 MSC 治疗本身、伴随 MSC 治疗的手术干预（刮除和闭合内瘘管通道开口）和/或挂线去除后 PFCD 自然病程的结果。考虑到 ADMIRE-II 试验的结果，这些都是重要的考虑因素。这是一项大型 III 期随机对照试验，发现在瘘管预处理和手术闭合后给予 MSC 疗法 （Darvadstrocel） 在第 24 周实现临床和放射学联合缓解方面并不比单独使用瘘管预处理和手术闭合更有效（49% vs. 46%，p = 0.57）[6].相比之下，在 ADMIRE-I 中，MSC 治疗明显比单独的瘘管预处理和手术闭合更有效 （50% vs. 34%，p = 0.024） [5]。

鉴于这种疗法的巨大成本，未来的研究应优先了解ADMIRE试验中差异的原因。此外，还需要比较有效性研究，以确定 PFCD 的最佳手术方法（瘘管调节、上皮消融技术和/或关闭内瘘口）以及挂线去除后 PFCD 的自然病程。这些问题的答案将有助于确定 MSC 治疗是否真的在瘘管调节和手术闭合之外增加益处。