Pharmacokinetic Modeling of Fludarabine to Control Exposure and Improve Outcomes after Reduced Intensity Conditioning Transplantation

Fludarabine (FLU) is a chemotherapeutic and immunosuppressive agent used in reduced intensity conditioning (RIC) regimens prior to hematopoietic stem cell transplantation. High exposure to F-ara-A, the active metabolite of FLU, has been associated with more treatment related mortality (TRM). No standard dosing models allow estimation of exposure and individualized dosing. We developed a population pharmacokinetic model for adults that estimates F-ara-A clearance (Cl) from which area under the curve (AUC_0-∞) is calculated.(Clin Pharmacol Ther, Vol95, S87, March 2014) We retrospectively tested the performance of the model by associating individual predicted AUC _0-∞to clinical outcomes.

TV F-ara-A CL (L/hr) = (7.04 + 3.9 * (CrCl/85) * (70/IBW)) * (IBW/70)^0.75

AUC _0-∞ (μg-hr/ml)= Administered dose in F-ara-A equivalents (mg)/Estimated TV F-ara-A Cl (L/hr)

Adult HCT patients (n=301, 2008 to 2014) who received IV FLU in their conditioning regimen were included. FLU doses, actual body weight, height, serum creatinine, along with TRM, engraftment and acute GVHD were reviewed. TRM was defined as death without relapse or disease progression. GVHD was staged and graded according to the standard GVHD criteria. The median age (range) was 58 yrs. (18-75). 131(43.5%) received PBSCT, 31(10.3%) bone marrow and cord blood in 139(46.2%). FLU doses were 25-40 mg/m2/day; nearly all x 5 days. The median (range) daily dose was 67 mg (38 mg-100 mg). The pharmacokinetic model was used to estimate the F-ara-A Cl and then predict their AUC_0-∞ from the first dose and the total cumulative AUC_0-∞. Recursive partitioning regression analysis was used to determine the optimal cut points for the first dose AUC_0-∞ and cumulative AUC_0-∞towards clinical outcomes. The cumulative incidence of engraftment, TRM and acute GVHD (grades II-IV and III-IV) was calculated using death prior to event as a competing risk. The proportional hazards model of Fine and Gray was used to assess the association of F-ara-A exposures towards TRM, acute GVHD and engraftment.

The median (range) F-ara-A Cl, AUC_0-∞ and cumulative AUC_0-∞ predicted from the model were 10.92 L/hr (7.51-15.37), 4.79 μg-hr/ml (2.40 -7.52) and 23.93 μg-hr/ml (11.20-37.62), respectively. Patients with a higher first dose AUC_0-∞ ≥6 μg-hr/ml had a higher incidence [95%CI] of day 100 TRM (20% [7-33%] vs 6% [4-9%], p<0.01) compared to those with <6 μg-hr/ml. Similarly, higher cumulative AUC_0-∞ ≥30 μg-hr/ml was also associated with higher risk of day 100 TRM (21% [7-34%] vs 6% [3-9%], p<0.01) compared to those <30 μg-hr/ml. Both PK measures [first dose AUC_0-∞ ≥6 μg-hr/ml and cumulative AUC_0-∞≥30 μg-hr/ml] had higher risks of 12 month TRM (32% [16-48%] vs 15% [11-20%], p=0.02) and (34% [17-51%] vs 15% [11-20%], p=0.02), respectively.

An AUC_0-∞ ≥4 μg-hr/mL led only to a marginally higher risk of 6 month GVHD grades II-IV and III-IV (42% [35-48%] vs 28% [14-41%] p=0.07 and 43% [37-50] vs. 30%[16-44%], p=0.06, respectively) compared to <4 μg-hr/ml. However a cumulative AUC_0-∞≥20 μg-hr/ml was associated with a significantly greater risk of acute GVHD grades II-IV (43% [36-49%] vs. 26% [14-38%], p= 0.02) and grades III-IV (44%, [38-51%] vs, 28% [15-41%], p=0.03).

Lower F-ara-A clearance and higher AUC _0-∞ were associated with greater risks of TRM and acute GVHD. These data support personalizing FLU dose rather than using empirical body surface area based dosing. Targeting FLU therapy may improve outcomes after HCT.