Purpose: To determine the impact of high-dose cytarabine (ARA-C) (HDAC) dose modification, based on renal function, on the incidence of neurotoxicity (NT).
Patients and methods: We retrospectively analyzed the records of 256 patients treated with HDAC (> or = 2.0 g/m2 per dose) for acute myelogenous leukemia (AML) at the University of California, San Francisco (UCSF). From 1985 to 1994, a total of 358 cycles of HDAC were administered, using either a twice-daily schedule (n = 208) or a once-daily regimen (n = 48). In 1989, a dose-modification algorithm was initiated at our institution, which reduced ARA-C doses in the setting of renal insufficiency (RI). For patients with a serum creatinine (Cr) level of 1.5 to 1.9 mg/dL during treatment, or an increase in Cr during treatment (deltaCr) of 0.5 to 1.2 mg/dL, ARA-C was decreased to 1 g/m2 per dose. For patients with a Cr > or = 2.0 mg/dL or a deltaCr greater 1.2 mg/dL, the dose was reduced to 0.1 g/m2/d.
Results: Overall, the incidence of NT was 16% (34 of 208) for patients treated with twice-daily HDAC and 0% (none of 48) for patients treated with daily HDAC (P = .003). NT occurred more often in patients treated on a twice-daily schedule with 3 g/m2 per dose compared with 2 g/m2 per dose (25% v 8%; P = .009). NT occurred in 55% of the twice-daily-treated patients with RI, compared with 7% of those with normal renal function (P = .00001). In patients with RI, NT occurred in none of 11 dose-modified cycles versus five of 11 (45%) total unmodified cycles (P = .01). None of 14 patients treated with once-daily HDAC given during RI developed NT, compared to 55% of patients (23 of 42) receiving twice-daily HDAC during RI (P = .009). By univariate analysis, NT was not associated with patient age or serum alkaline phosphatase, but NT was significantly increased in patients treated with twice-daily HDAC when the serum bilirubin was > or = 2.0 mg/dL compared with twice-daily HDAC given when the total bilirubin was less than 2.0 mg/dL (33% v 14%; P = .017). Multivariate analysis confirmed that RI was the most significant risk factor associated with the development of NT.
Conclusion: HDAC NT is strongly associated with RI. The risk of HDAC NT can be reduced by the following: (1) routinely reducing the ARA-C dose from 3 to 2 g/m2 per dose; (2) modifying the ARA-C dose based on daily Cr values; and (3) administering HDAC on a once-daily rather than twice-daily schedule.