Background/Purpose: Two randomized controlled trials showed modest but significant effects for oral (po) CYC ( superior to placebo and  equal to mycophenolate mofetil). A number of European centres seem to prefer intravenous (IV) CYC based on a presumed better safety profile. However, no direct comparisons of these two administration methods have been published. In this context, our goal was to compare the relative efficacy and safety of po versus IV CYC(IV) for treating ILD and/or skin involvement in SSc.

Methods: Patients were derived from the EUSTAR database and Scleroderma Lung Studies I and II, receiving >= 6 months of po or IV CYC ,all with 12 months follow-up. Serious (SAEs) and non-serious adverse events (AEs) and efficacy (change in ppFVC, ppDLCO, mRSS) were analyzed at end of treatment (EoT) and at follow-up (FU).  Analysis included descriptive statistics and linear regressions

Results: Between the 149 po versus 153 IV CYC patients there were baseline statistical differences in ethnicity, previous DMARD exposure, previous and concomitant steroid exposure and dosage, current/previous smoking, digital ulcers and arterial hypertension (Table 1).

Efficacy: after adjusting for significant baseline ppFVC, ppDLCO and mRSS, at EOT vs baseline and at FU vs EOT, changes in ppFVC, ppDLCO and mRSS were similar for both groups (p=NS, Table 2). In a multivariate analysis, route of administration had no impact on efficacy.

Safety: there was more leukopenia (22.1% vs 1.3%, p<0.001), haemorrhagic cystitis (5.5% vs 0%, p=0.011) and alopecia (19.5% vs 1.3%, p<0.001) at EOT visit after po than IV CYC. In contrast, there were more SAEs and need for oxygen supplementation at FU in the IV vs po group (19.3% vs 9.4%, p=0.025, and 7.2% vs 2%, p=0.049). The median cumulative CYC dosage was significantly higher in the po group [39.4 mg vs 9.2 mg, p<0.001). No significant differences were found for deaths, anemia, thrombocytopaenia, gastro-intestinal bleeding, serious infections, malignancies, need for total parenteral nutrition, new onset of cardiomyopathy or amenorrhea.

Conclusion: This hypothesis generating study is limited by the different  source data, patients’ selection (i.e. presence of active ILD and absence pf PAH for SLSI and SLS 2), post-treatment medications (higher dosage corticosteroids, more csDMARD and O2 in IV-CYC) and short follow-up duration. It showed similar efficacy of one year of oral versus IV CYC. AEs were more frequent with po CYC but more SAEs were noted for IV-CYC. Well-controlled studies are warranted to confirm and extend our data.