Use of an empirically derived cyclophosphamide dosing nomogram for treatment of vasculitis

Treatment of systemic necrotizing vasculitides in patients aged sixty-five years or older: results of a multicenter, open-label, randomized controlled trial of corticosteroid and cyclophosphamide-based induction therapy

OBJECTIVE

To investigate a new therapeutic strategy, with rapid corticosteroid dose tapering and limited cyclophosphamide (CYC) exposure, for older patients with systemic necrotizing vasculitides (SNVs; polyarteritis nodosa [PAN], granulomatosis with polyangiitis [Wegnener's] [GPA], microscopic polyangiitis [MPA], or eosinophilic GPA [Churg-Strauss] [EGPA]).

METHODS

A multicenter, open-label, randomized controlled trial comprising patients ≥65 years old and newly diagnosed as having SNV was conducted. The experimental treatment consisted of corticosteroids for ∼9 months and a maximum of six 500-mg fixed-dose intravenous (IV) CYC pulses, every 2-3 weeks, then maintenance azathioprine or methotrexate. The control treatment included ∼26 months of corticosteroids for all patients, combined with 500 mg/m(2) IV CYC pulses, every 2-3 weeks until remission, then maintenance for all patients with GPA or MPA and for those with EGPA or PAN with a Five-Factors Score (FFS) of ≥1. Randomization used a 1:1 ratio computer-generated list and was performed centrally with sealed opaque envelopes. The primary outcome measure was ≥1 serious adverse event (SAE) occurring within 3 years of followup. Secondary outcome measures included remission and relapse rates.

RESULTS

Among the 108 patients randomized, 4 were excluded (early consent withdrawal or protocol violation). Mean ± SD age at diagnosis was 75.2 ± 6.3 years. Analysis at 3 years included 53 patients (21 GPA, 21 MPA, 8 EGPA, and 3 PAN) in the experimental arm and 51 patients (15 GPA, 23 MPA, 6 EGPA, and 7 PAN) in the conventional arm. In total, 32 (60%) versus 40 (78%) had ≥1 SAE (P = 0.04), most frequently infections; 6 (11%) versus 7 (14%) failed to achieve remission (P = 0.71); 9 (17%) versus 12 (24%) died (P = 0.41); and 20 (44%) of 45 versus 12 (29%) of 41 survivors in remission experienced a relapse (P = 0.15).

CONCLUSION

For older SNV patients, an induction regimen limiting corticosteroid exposure and with fixed low-dose IV CYC pulses reduces SAEs in comparison to conventional therapy, and does not affect the remission rate. Three-year relapse rates remain high for both arms.