Patients selection
Under conditions established by the General Health Service, stem cells were allocated for patients with anal fistulas related to CD after failure of good surgical drainage and adequate treatment with biologics. Before deciding injection, previous drainage with a seton of all collections and a complete endoscopic response or moderate activity (CDAI < 220) were mandatory. Biologics and immunosuppressants were reviewed and adapted according to last ECCO guidelines and available staff. All the patients had pre-operative MRI and pre-operative endoscopy to guarantee the marketing authorization. Concerning the pelvic MRI, the acquisition protocol used high-resolution, T2-weighted, fat-saturation, fast-spin echo pulse sequences and T1-weighted, fat-saturation, fast-spin echo or three-dimensional gradient echo sequences before and after intravenous injection of gadolinium. MRI data collected were inflammation of the fistula tract, its total length and maximal depth, and the presence of collection. All MRIs were reviewed in each center by specialized radiologists. In Saint-Joseph Hospital, patient consent was obtained before data collection, which was performed prospectively in a REDCAP database (Research Electronic Data Capture, https://projectredcap.org/resources/citations/). At Rennes University Hospital, a prospective database for monitoring all patients with anoperineal Crohn's disease has been in place since January 2005. The extraction of the specific data for this study from this database and patient consent were performed retrospectively. The data recorded were as follows: sex, height, weight, body mass index (BMI), age, luminal CD phenotype at diagnosis according to the Montreal classification [15], extra-intestinal manifestations, smoking habits, past and actual treatments (steroids, immunosuppressants, and anti-TNFα) and past surgical history (anal fistula surgery, ileal and/or colonic surgery, and stoma). The characteristics of the fistula were assessed based on the data of the exploration under anesthesia performed for all the patients as recommended by the ADMIRE study 2 to 3 weeks before injection and of the last MRI. Anatomical classification of perianal CD according to the Cardiff-Hughes classification was used to classify the tracts (F0, F1, F2) [16]. Before injection, all patients had an F2 fistula according to this classification and a complex one according to the classification of the AGA with of course no abscess [17]. Fistulas were considered multibranched when there was a main tract and at least one secondary tract such as horseshoe or supralevator extension. In addition, CD activity was assessed with the Harvey-Bradshaw Index (HBI) [18] and perianal CD activity was assessed with the Perianal Disease Activity Index (PDAI) [19].
Procedure & Follow-up
Stem cell injections were prepared and the cells injected as previously described (10). Surgical procedures were performed under anesthesia after antibiotic prophylaxis (metronidazole). Once the setons were removed and the fistula tract(s) curetted, the internal openings were closed by direct suturing with absorbable sutures. In addition, patients with a large internal opening underwent a minimal flap advancement. According to the manufacturer`s recommendations, darvadstrocel was injected around the internal openings and along the fistula tract(s) (24 ml). The procedure was completed by gentle massaging of the tract(s) [20]. Follow-up and events were recorded over at least one year (3, 6, and 12 months and last visit) complemented by an MRI evaluation at least six months after the procedure. Therapeutic procedures after stem cell injection were recorded, including medical treatments modification (optimization, switch of immunosuppressant/biological therapy) and surgeries (fistula drainage, intestinal resection).
Endpoints
The primary endpoints were clinical and radiological response.
The clinical response was defined as follows:
- Complete: complete closure with invisible external openings with no discharge at pressure
- Partial: complete closure of ≥ 50% of the external openings with no discharge at pressure from any of the external openings
- Failure: closure of < 50% of the external openings ± the persistence of discharge at pressure from one or all of the external openings ± the occurrence of a new tract or abscess.
The radiological response was defined as follows:
- Complete: fibrous sequelae with no T2 hypersignal or enhancement after the injection of gadolinium into the tract(s), and no abscesses
- Partial: decrease of the inflammatory signal (T2 hypersignal and enhancement after the injection of gadolinium) ± the length ± the largest diameter of the tracts, with no abscesses
- Failure: absence of radiological modification or occurrence of a new fistula tract ± abscesses
A complete combined clinical-radiological response (deep response) was defined as the association of a complete clinical response with a complete radiological response.
The secondary endpoints were:
- Symptomatic efficacy, assessed with the PDAI
- Outcomes
- Factors associated with deep response
Statistical analysis
Data are expressed as medians and percentiles (interquartile range: 25% and 75%) or as numbers and percentages of the cohort. The nonparametric Wilcoxon signed-rank test was used for continuous variables and nonparametric Pearson or Fisher exact tests for categorical variables. The judgement criterion was deep response that combined an asymptomatic status plus clinical closure of the external openings combined with MRI healing of the tract (no inflammation and/or disappearance of the tract). Asymptomatic response was defined by the complete relief of discomfort and discharge and a stepwise regression analyses was performed to identify factors associated with deep response. The results are expressed as odds ratios (ORs) with 95% confidence intervals [CIs]. Statistical analyses were performed using JMP Pro 13.0.0 software (SAS institute).
Ethical considerations
The study protocol was approved by the “Comité de Protection des Personnes” GH Paris Saint-Joseph (IRB 00012157) and registered in clinicaltrial.gov (NCT05177003). The database Fondamentum is secured (nominative access code) with a CNIL (Commission nationale de l'informatique et des libertés) declaration (CNIL n°1412467).