Brazilian Journal of Medical and Biological Research (2005) 38: 683-694
`FTY720 in kidney transplant recipients
`ISSN 0100-879X
`
`683
`
`S.I. Park1, C.R. Felipe1,
`P.G. Machado1, R. Garcia1,
`A. Skerjanec2,
`R. Schmouder2,
`H. Tedesco-Silva Jr.1 and
`J.O. Medina-Pestana1
`
`Correspondence
`H. Tedesco-Silva Jr.
`Divisão de Nefrologia
`Hospital do Rim e Hipertensão
`EPM, UNIFESP
`Rua Borges Lagoa, 960, 11º andar
`04038-002 São Paulo, SP
`Brasil
`Fax: +55-11-5087-8008
`E-mail: heliotedesco@hrim.com.br or
`parksung@hotmail.com
`
`Part of the data were presented
`at the American Transplant Congress,
`May 30-June 4, 2003, Washington,
`DC, USA.
`
`Publication supported by FAPESP.
`
`Received March 22, 2004
`Accepted February 3, 2005
`
`Pharmacokinetic/pharmacodynamic
`relationships of FTY720 in kidney
`transplant recipients
`
`1Divisão de Nefrologia, Hospital do Rim e Hipertensão, Escola Paulista de Medicina,
`Universidade Federal de São Paulo, São Paulo, SP, Brasil
`2Novartis Pharmaceuticals, East Hanover, NJ, USA
`
`Key words
`• FTY720
`• Lymphopenia
`• Pharmacokinetics
`• Pharmacodynamics
`• Immunosuppression
`• Renal transplants
`
`Abstract
`
`FTY720 is a new and effective immunosuppressive agent, which
`produces peripheral blood lymphopenia through a lymphocyte
`homing effect. We investigated the relationship between the dose
`of FTY720 or blood concentration (pharmacokinetics, PK) and
`peripheral lymphopenia (pharmacodynamics, PD) in 23 kidney trans-
`plant recipients randomized to receive FTY720 (0.25-2.5 mg/day) or
`mofetil mycophenolate (2 mg/day) in combination with cyclosporine
`and steroids. FTY720 dose, blood concentrations and lymphocyte
`counts were determined weekly before and 4 to 12 weeks after
`transplantation. The effect of PD was calculated as the absolute
`lymphocyte count or its reductions. PK/PD modeling was used to
`find the best-fit model. Mean FTY720 concentrations were 0.36 ±
`0.05 (0.25 mg), 0.73 ± 0.12 (0.5 mg), 3.26 ± 0.51 (1 mg), and 7.15 ± 1.41
`ng/ml (2.5 mg) between 4 and 12 weeks after transplantation. FTY720
`PK was linear with dose (r2 = 0.98) and showed low inter- and intra-
`individual variability. FTY720 produced a dose-dependent increase
`in mean percent reduction of peripheral lymphocyte counts (38 vs
`42 vs 56 vs 77, P < 0.01, respectively). The simple Emax model
`[E = (Emax * C)/(C + EC50)] was the best-fit PK/PD modeling for
`FTY720 dose (Emax = 87.8 ± 5.3% and ED50 = 0.48 ± 0.08 mg, r2 = 0.94)
`or concentration (Emax = 78.3 ± 2.9% and EC50 = 0.59 ± 0.09 ng/ml,
`r2 = 0.89) vs effect (% reduction in peripheral lymphocytes). FTY720
`PK/PD is dose dependent and follows an Emax model (EC50 = 0.5 mg
`or 0.6 ng/ml). Using lymphopenia as an FTY720 PD surrogate
`marker, high % reductions (~80%) in peripheral lymphocytes are
`required to achieve best efficacy to prevent acute allograft rejection.
`
`Introduction
`
`FTY720 (2-amino-2-[2-(4-octylphenyl)
`ethyl] propane-1,3 diol hydrochloride) is a
`novel immunomodulator agent developed by
`a chemical modification of myriocine (1-3),
`
`a metabolite isolated from Iscaria sinclairii
`broth (4). This immunomodulator is effec-
`tive in protecting solid organ grafts from
`acute rejection in experimental transplant
`models (5). FTY720 is effective even in the
`absence of cyclosporine, and also shows
`
`Braz J Med Biol Res 38(5) 2005
`
`APOTEX - EXHIBIT 1019
`
`

`

`684
`
`S.I. Park et al.
`
`synergistic interactions with cyclosporine,
`tacrolimus, sirolimus, or everolimus (6-14).
`FTY720 is currently undergoing phase III
`clinical trials in kidney transplant recipients
`after FTY720 doses and combination thera-
`py were identified in phase II studies.
`FTY720 produces peripheral lymphope-
`nia in both animals and humans as a conse-
`quence of altered T-cell trafficking and redi-
`rection to secondary lymphatic organs (15,
`16). At therapeutic concentrations, FTY720
`does not interfere with cytokine synthesis or
`binding, does not inhibit cell proliferation and
`does not cause cell death by apoptosis (17,18).
`Lymphopenia is fully reversible upon drug
`discontinuation (19). FTY720 needs to be
`phosphorylated by sphingosine phosphatase
`and the phosphorylated form (FTY720-P)
`appears to mediate its biological effect through
`the binding to sphingosine-1-phosphate, G-
`protein-coupled receptors expressed in the
`membrane of lymphocytes and endothelial
`cell of lymph nodes (20-22). The proposed
`molecular mechanism of action is that lym-
`phocyte homing is altered by FTY720 as a
`sphingosine-1-phosphate agonist targeting
`G-protein-coupled receptors expressed in
`the membrane of lymphocytes and endothe-
`lial cell of lymph nodes (23). This binding
`results in drug-receptor internalization, in-
`creasing intrinsic lymphocyte mobility and
`chemotactic response, accelerating lympho-
`cyte homing, and trapping in secondary lym-
`phoid tissues (21,24).
`FTY720 has unique pharmacokinetic
`properties. Its absorption is slow, reaching
`peak concentrations 8 to 36 h after oral
`administration (19). FTY720 has a large
`volume of distribution (1116-1737 L) and a
`clearance of about 123 to 383 ml/min, result-
`ing in a long elimination half-life of about 108
`h (4.5 days) (19). Since it is given once a day,
`presumably it takes about 4 weeks to reach
`steady-state concentration, with 11- to 19-
`fold accumulation in tissues compared to
`first dose administration (19).
`Pharmacokinetic and pharmacodynamic
`
`(PK/PD) modeling has been used increas-
`ingly in clinical pharmacology and drug de-
`velopment, not only to speed up the develop-
`ment process but, more importantly, to de-
`termine the optimal dosage of new drugs,
`which will deliver an appropriate effect. The
`relationship between FTY720 pharmacoki-
`netics and pharmacodynamics has been stud-
`ied in various experimental transplant models
`(13). The aim of the present study was to
`determine the relationship among FTY720
`dose, blood concentration and surrogate bio-
`logical effect in the peripheral blood com-
`partment (lymphopenia). This information
`will help to choose and possibly individualize
`drug dose regimens, which will result in best
`efficacy/toxicity relationships.
`
`Material and Methods
`
`Population
`
`Between June 1st and September 30,
`2000, 23 kidney transplant recipients were
`enrolled at our center as part of an interna-
`tional multicenter prospective, open-label,
`randomized, dose-finding, and exploratory
`trial, which included 208 patients. The pri-
`mary objective was to evaluate the safety,
`tolerability and preliminary efficacy of in-
`creasing doses of FTY720 versus mofetil
`mycophenolate (MMF, CellCept®, Roche
`Laboratories, New Jersey, NJ, USA) in com-
`bination with a cyclosporine microemulsion
`and prednisone.
`The local Medical Ethics Committee ap-
`proved the protocol and the study was per-
`formed in accordance with the Declaration of
`Helsinki and US Food and Drug Administration
`guidelines for good clinical practice. All pa-
`tients signed an informed consent term after
`being informed of the details of this study, and
`were enrolled according to study-specific in-
`clusion and exclusion criteria. The present
`report describes a retrospective analysis of
`data obtained from 23 patients enrolled at our
`transplant center during the clinical trial.
`
`Braz J Med Biol Res 38(5) 2005
`
`

`

`FTY720 in kidney transplant recipients
`
`Study design
`
`The study involved 24 weeks of observa-
`tion divided into 12 weeks of treatment and
`12 weeks of follow-up. Patients were ran-
`domized 24 h after renal transplantation to
`one of four FTY720 groups (0.25 mg (N =
`4), 0.5 mg (N = 4), 1 mg (N = 5), 2.5 mg (N
`= 5)) or to one MMF group (N = 5). To be
`enrolled in the study patients had to be
`between 18 and 65 years old and to be
`receiving their first living or cadaver donor
`allografts. All patients received initial doses
`of cyclosporine of 8 to 10 mg kg-1 day-1
`twice a day adjusted to achieve therapeutic
`concentrations of 200-400 ng/ml during the
`first month, 150-250 ng/ml during the sec-
`ond and third months, and 100-200 ng/ml
`thereafter. Prednisone was introduced two
`days after the transplant, with initial doses of
`0.5 mg kg-1 day-1 followed by a gradual
`reduction to 0.2 mg kg-1 day-1 at 12 weeks
`post-transplant. Within 24 h after transplant
`surgery, a loading dose of 1, 2, 4, and 4 mg
`was administered to patients randomized to
`receive fixed maintenance doses of 0.25,
`0.5, 1, or 2.5 mg/day of FTY720. Patients
`randomized to the MMF group received 2 g
`twice a day. FTY720 treatment lasted 12
`weeks with the patients being converted to
`azathioprine, 2 mg kg-1 day-1, or MMF, 2 g/
`day thereafter at the discretion of the inves-
`tigator. Drug doses were also adjusted for
`safety and tolerance.
`
`Pharmacokinetics/pharmacodynamics
`
`There were 14 study visits, namely pre-
`transplant and at 1, 2, 4 to 12 weeks during
`the treatment phase and 16, 20 and 24 weeks
`during the follow-up phase. On the occasion
`of each study visit, blood samples were
`obtained to measure blood concentrations of
`cyclosporine and FTY720 and to perform
`lymphocyte counts.
`Pharmacokinetic analysis. Whole blood
`cyclosporine concentrations were measured
`
`daily during the first week and doses were
`adjusted until therapeutic concentrations
`(>200 ng/ml) were achieved. Subsequent
`monitoring was performed twice a week
`during the first month, once a week during
`the second month, and every other week
`thereafter. Whole blood cyclosporine con-
`centrations were measured by a fluores-
`cence polarization immunoassay (Abbott
`Laboratories, Chicago, IL, USA) according
`to the manufacturer instructions. Since
`FTY720 has a long terminal half-live of about
`100 h, steady-state concentrations are
`achieved only after 4 weeks of treatment.
`Therefore, blood samples were collected to
`measure whole blood FTY720 concentra-
`tions between 4 and 12 weeks. Blood con-
`centrations of FTY720 were determined us-
`ing a validated HPLC/mass spectrometry/
`mass spectrometry method from Novartis
`(East Hanover, NJ, USA).
`Pharmacodynamics study. The surrogate
`marker of the pharmacodynamic effect of
`FTY720 used in this study was the peripheral
`lymphocyte count in the blood compart-
`ment. Lymphocyte counts were performed
`on the occasion of each study visit (pre-
`transplant and 1, 2, 4, 5, 6, 7, 8, 9, 10, 11, 12,
`16, 20, and 24 weeks after transplant) in all
`5 groups using the CELL-DYN 3200 auto-
`matic counter (Abbott Park, Chicago, IL,
`USA) whose methods of determinations are
`based on flow cytometry. Since MMF has no
`effect on peripheral lymphocytes (25), lym-
`phocyte counts obtained from patients re-
`ceiving MMF were used to calculate the
`basal pharmacodynamic effect when FTY720
`doses or exposures were equal to zero. The
`pharmacodynamic effect of FTY720 was
`determined either as absolute reduction in
`peripheral lymphocyte counts or as percent
`reduction compared to the lymphocyte count
`obtained pre-transplant and before the ad-
`ministration of the first dose of FTY720 or
`MMF.
`PK/PD correlations. Peripheral lympho-
`cyte counts were correlated with FTY720
`
`685
`
`Braz J Med Biol Res 38(5) 2005
`
`

`

`686
`
`S.I. Park et al.
`
`doses and blood concentrations. PK/PD
`modeling was used to find the best-fit model
`of the correlation between absolute or per-
`cent reduction in peripheral lymphocyte count
`and increasing doses or blood concentra-
`tions of FTY720. For these analysis we used
`the scientific software WinNonlin (SCI Soft-
`ware) to find the best fit PK/PD model that
`explains the relationship between FTY720
`dose or blood concentration and the chosen
`pharmacodynamic surrogate marker.
`
`Statistical analysis
`
`Demographic characteristics were ana-
`lyzed by analysis of variance (ANOVA) for
`continuous variables and by the chi-square
`test for categorical variables. Summary sta-
`tistics were expressed as means ± SD, or
`frequencies or median and range, respective-
`ly. The observed and calculated PK/PD pa-
`rameters were presented as means ± SD and
`as dose/response and exposure/response re-
`lationships. Linear regression analyses were
`used to correlate FTY720 dose and blood
`concentration. Nonlinear regressions were
`used to correlate the dose and drug concen-
`tration in blood with biological response. All
`statistical analyses were performed using the
`SPSS 7.5 software (SPSS Inc., Chicago, IL,
`USA), with the level of significance set at P
`< 0.05.
`
`Results
`
`Demographics and baseline characteristics
`
`Demographic characteristics of the al-
`lograft recipients were similar across treat-
`ment groups (Table 1). Mean age was 40.1
`± 11.0 years, weight 62.2 ± 8.8 kg, and mean
`body mass index was 22.8 ± 2.6 kg/m2. Most
`patients were males (61%) and white (61%),
`with 4 (17%) blacks and 5 (22%) with
`various degrees of miscegenation. The pri-
`mary etiology leading to renal failure was
`chronic glomerulonephritis (22%), followed
`
`by diabetes (17%) and nephrosclerosis
`(13%). Mean number of HLA mismatches
`was 2.9 ± 0.5 and 100% showed no pre-
`formed anti-HLA antibodies. All recipients
`tested negative for hepatitis B and C viruses
`and 50% of the recipients were positive for
`cytomegalovirus IgG. No significant differ-
`ences in demographic characteristics were
`observed between the 5 groups.
`
`Pharmacokinetics
`
`Cyclosporine doses and whole blood con-
`centrations did not differ statistically be-
`tween the 5 groups during the 24 weeks of
`treatment. Mean doses/weight were 9.7 ±
`2.5 in the first week, 6.5 ± 1.9 at week 4, 4.7
`± 1.5 at week 12, and 3.8 ± 1.5 mg kg-1
`day-1 at week 24, with no significant differ-
`ences between groups (Figure 1A). Corre-
`sponding mean whole blood cyclosporine
`concentrations were 271.0 ± 104.1, 351.1 ±
`121.4, 213.5 ± 71.3, and 149.1 ± 72.7 ng/ml,
`respectively (Figure 1B). Also dose-normal-
`ized cyclosporine concentrations did not dif-
`fer significantly between the 5 groups, rang-
`ing from 0.50 ± 0.35 at week 1 to 0.63 ± 0.30
`ng ml-1 mg-1 at week 24. Therapeutic con-
`centrations of cyclosporine were achieved in
`all groups during the study period. Mean
`prednisone doses were 30.4 ± 3.7 (week 1),
`27.3 ± 3.7 (week 4), 13.3 ± 3.7 (week 12),
`and 9.9 ± 0.5 mg/day (week 24), with no
`significant differences between the 5 groups.
`Mean FTY720 blood concentrations ac-
`cording to dose level and study visit are
`shown in Figure 2. There were no significant
`differences in mean blood concentration be-
`tween study visits (General Linear Model for
`repeated measurements) within each dose
`level, confirming that at week 4 steady-state
`FTY720 concentrations had been achieved
`(Table 2). Interindividual variability (% coef-
`ficients of variation, %CV) increased from
`24 to 41% with increasing doses of FTY720
`(P = 0.05) but no significant differences
`were observed between study visits. Aver-
`
`Braz J Med Biol Res 38(5) 2005
`
`

`

`FTY720 in kidney transplant recipients
`
`687
`
`Table 1. Demographic characteristics of the transplant population.
`
`N
`
`Age (years)
`
`Total
`
`23
`
`FTY720
`(0.25 mg)
`
`FTY720
`(0.5 mg)
`
`FTY720
`(1 mg)
`
`FTY720
`(2.5 mg)
`
`4
`
`4
`
`5
`
`5
`
`MMF
`(2 g)
`
`5
`
`40.1 ± 11.0
`(23.0-59.0)
`
`45.5 ± 5.0
`(40.0-52.0)
`
`38.3 ± 14.0
`(23.0-52.0)
`
`40.4 ± 14.3
`(24.0-59.0)
`
`38.6 ± 11.8
`(24.0-57.0)
`
`38.6 ± 10.7
`(25.0-52.0)
`
`Gender (male/female)
`
`14/9
`
`3/1
`
`3/1
`
`0/5
`
`4/1
`
`4/1
`
`Weight (kg)
`
`BMI (kg/m2)
`
`62.2 ± 8.8
`(47.4-83.6)
`
`22.8 ± 2.6
`(19.2-27.0)
`
`68.6 ± 10.1
`(62.3-83.6)
`
`24.7 ± 2.2
`(22.3-27.0)
`
`60.4 ± 5.6
`(53.1-65.9)
`
`22.1 ± 1.7
`(20.5-24.2)
`
`55.6 ± 7.5
`(47.4-61.8)
`
`23.9 ± 3.8
`(19.2-27.0)
`
`61.8 ± 10.4
`(53.8-79.7)
`
`65.4 ± 7.2
`(57.6-76.0)
`
`21.5 ± 2.0
`(20.1-24.9)
`
`22.0 ± 2.5
`(19.7-26.0)
`
`Ethnicity (white/black/mulatto/others)
`
`14/4/2/3
`
`3/0/0/1
`
`4/0/0/0
`
`2/1/1/1
`
`3/1/0/1
`
`2/2/1/0
`
`HLA mismatches
`
`2.9 ± 0.5
`
`3.3 ± 0.5
`
`3.0 ± 0.0
`
`2.6 ± 0.5
`
`2.8 ± 0.4
`
`2.9 ± 0.5
`
`Cause of ESRD
`Chronic glomerulonephritis
`Nephrosclerosis
`Polycystic kidney disease
`Diabetic nephropathy
`Unknown
`
`Time on dialysis (months)
`
`5 (21.7%)
`3 (13.0%)
`2 (8.9%)
`4 (17.4%)
`9 (39.1%)
`
`2
`0
`0
`1
`1
`
`1
`0
`0
`1
`2
`
`0
`1
`0
`1
`3
`
`1
`1
`1
`0
`2
`
`1
`1
`1
`1
`1
`
`17.7 ± 10.2
`(2.0-36.0)
`
`22.3 ± 12.9
`(6.0-36.0)
`
`18.5 ± 4.9
`(12.0-24.0)
`
`20.0 ± 13.6
`(2.0-36.0)
`
`16.2 ± 11.3
`(7.0-36.0)
`
`12.6 ± 6.8
`(5.0-23.0)
`
`Panel reactive antibody <5/>5 (%)
`
`23/0
`
`Viral status
`Anti-HCV positive (IgG)
`HbsAg positive
`CMV positive (IgG)
`
`0
`0
`12
`
`4/0
`
`0
`0
`3
`
`4/0
`
`0
`0
`1
`
`5/0
`
`0
`0
`2
`
`5/0
`
`5/0
`
`0
`0
`3
`
`0
`0
`3
`
`Data are reported as means ± SD (range). MMF = mofetil mycophenolate; BMI = body mass index; HLA = human leukocyte antigen; ESRD
`= end-stage renal disease; HCV = hepatitis C virus; CMV = cytomegalovirus. Panel reactive antibody between donor and recipients greater or
`less than 5%.
`
`Figure 1. Time course of mean cyclosporine doses (A) and blood concentrations (B). There were no significant differences for mean cyclosporine
`doses or blood concentrations among study groups at each visit (independent Student t-test). MMF = mofetil mycophenolate.
`
`Braz J Med Biol Res 38(5) 2005
`
`

`

`688
`
`S.I. Park et al.
`
`age intra-individual variability (%CV) ranged
`from 25 to 28 and was not dose dependent.
`In the 10-fold dose range studied in this trial
`(0.25 to 2.5 mg/day), FTY720 pharmacoki-
`netics was linear with dose (r2 = 0.679 and
`0.982), as can be observed in Figure 3.
`
`Pharmacodynamics
`
`The time course of the FTY720-induced
`reduction of peripheral lymphocyte count is
`shown in Figure 4. There were no significant
`differences in pre-transplant values between
`groups and lymphopenia was not observed in
`the MMF group throughout the study. The
`
`Figure 2. Time course of mean FTY720 blood concentrations between 4 and 12 weeks
`post-transplant. P < 0.01 for 1.0 and 2.5 mg FTY720 compared to 0.25 and 0.5 mg
`FTY720 during administration time (ANOVA).
`
`Table 2. FTY720 blood concentrations and percent coefficient of variation (%CV) between weeks 4 and 12 according to FTY720 dose level.
`
`Visit (week)
`
`FTY720 (0.25 mg)
`N = 4
`
`FTY720 (0.50 mg)
`N = 4
`
`FTY720 (1.00 mg)
`N = 5
`
`FTY720 (2.50 mg)
`N = 5
`
`4a
`5a
`6a
`7a
`8a
`9a
`10a
`11a
`12a
`Total a
`Average interindividual %CVb,c (range)
`Average intra-individual %CVd (range)
`
`0.41 ± 0.13 (32)
`0.37 ± 0.12 (32)
`0.26 ± 0.05 (14)
`0.36 ± 0.16 (44)
`0.37 ± 0.01 (3)
`0.43 ± 0.11 (26)
`0.33 ± 0.07 (21)
`0.37 ± 0.09 (24)
`0.35 ± 0.08 (23)
`0.36 ± 0.05 (24)
`24 ± 12 (14-44)
`25 ± 14 (7-40)
`
`0.81 ± 0.13 (16)
`0.78 ± 0.16 (21)
`0.72 ± 0.26 (26)
`0.81 ± 0.25 (31)
`0.80 ± 0.36 (45)
`0.79 ± 0.25 (32)
`0.76 ± 0.31 (41)
`0.66 ± 0.27 (41)
`0.44 ± 0.30 (68)
`0.73 ± 0.12 (36)
`36 ± 15 (16-68)
`26 ± 16 (12-42)
`
`2.59 ± 1.08 (42)
`3.18 ± 1.35 (42)
`3.74 ± 2.51 (67)
`3.77 ± 1.39 (37)
`3.31 ± 1.05 (32)
`3.44 ± 1.53 (44)
`3.37 ± 0.66 (20)
`3.65 ± 1.09 (30)
`2.28 ± 1.22 (54)
`3.26 ± 0.51 (41)
`41 ± 14 (20-67)
`25 ± 8 (17-35)
`
`4.26 ± 1.65 (39)
`6.25 ± 2.80 (45)
`6.60 ± 1.82 (29)
`6.61 ± 1.81 (27)
`7.36 ± 2.58 (36)
`9.02 ± 2.82 (32)
`8.00 ± 2.56 (32)
`8.24 ± 4.17 (51)
`7.97 ± 4.05 (51)
`7.15 ± 1.41 (38)
`38 ± 9 (27-51)
`28 ± 13 (14-49)
`
`Data are reported as means ± SD in ng/ml and %CV within parentheses.
`aFTY720 blood level, P < 0.05 between groups (ANOVA); binterindividual %CV according to FTY720 dose, P = 0.05 (ANOVA); cinterindividual
`%CV according to visit, P > 0.05 (general linear model for repeated measures); dintra-individual %CV, P > 0.05 between groups (ANOVA).
`
`Figure 3. Linear regression analysis correlating individual (A) or mean (B) FTY720 blood concentrations and dose levels.
`
`Braz J Med Biol Res 38(5) 2005
`
`

`

`FTY720 in kidney transplant recipients
`
`689
`
`reduction in the number of lymphocytes was
`observed as early as at week 1, reached its
`peak at about week 4, and was fully reversed
`4 to 8 weeks after treatment interruption.
`The extent of lymphopenia and the time to
`return to baseline values appear to depend on
`the magnitude of the FTY720 dose. Mean
`percent reduction in lymphocyte counts af-
`ter week 4 was dose dependent (37.8 ± 19.2
`(0.25 mg) vs 42.4 ± 19.9 (0.50 mg) vs 56.0
`± 17.8 (1.00 mg) vs 76.5 ± 10.8% (2.50 mg),
`P < 0.01). Mean lymphocyte counts varied
`from 624 to 1320 (0.25 mg), 675 to 1085
`(0.5 mg), 567 to 853 (1 mg), and 327 to 696/
`mm3 (2.5 mg; Table 3). Inter- (42-55%) and
`intra-individual variability (25-55%) ranges
`
`Figure 4. Time course of FTY720-induced lymphopenia during the 12-week treatment
`phase and return to baseline values upon drug discontinuation. MMF = mofetil myco-
`phenolate.
`
`Table 3. Mean peripheral blood lymphocyte counts and percent lymphocyte reduction between weeks 4 and 12 in the 5 groups of patients.
`
`MMF (2 g)
` (N = 5)
`
`FTY720 (0.25 mg)
`(N = 4)
`
`FTY720 (0.5 mg)
`(N = 4)
`
`FTY720 (1 mg)
`(N = 5)
`
`FTY720 (2.5 mg)
`(N = 5)
`
`Lymphocytes
`(N/mm3)
`
`%
`Reduction
`
`Lymphocytes
`(N/mm3)
`
`%
`Reduction
`
`Lymphocytes
`(N/mm3)
`
`%
`Reduction
`
`Lymphocytes
`(N/mm3)
`
`%
`Reduction
`
`Lymphocytes
`(N/mm3)
`
`%
` Reduction
`
`1746 (35)
`
`2488 (8)
`2059 (45)
`2217 (34)
`2344 (28)
`2556 (26)
`2518 (25)
`2209 (27)
`2474 (24)
`2472 (46)
`2364 (28)
`
`29 ± 11
`[8-46]
`
`22 ± 7
`[16-32]
`
`1740 (31)
`1744 (19)
`1853 (19)
`
`-
`
`-
`-
`-
`-
`-
`-
`-
`-
`-
`-
`
`-
`
`-
`
`-
`-
`-
`
`1677 (21)
`
`-
`
`1372 (26)
`
`-
`
`1567 (38)
`
`-
`
`1657 (29)
`
`-
`
`872 (44)
`624 (59)
`1559 (61)
`900 (75)
`1067 (81)
`1023 (98)
`1284 (12)
`1320 (20)
`924 (12)
`823 (42)
`
`51 ± 31
`[12-98]
`
`55 ± 27
`[21-86]
`
`39.3 (49)
`46.6 (66)
`39.2 (23)
`43.8 (49)
`38.5 (32)
`49.2 (40)
`19.5 (94)
`15.1 (42)
`26.1 (4)
`37.8 (56)
`
`44 ± 26
` [4-94]
`
`36 ± 8
`[27-45]
`
`675 (45)
`805 (69)
`852 (30)
`890 (35)
`981 (75)
`837 (35)
`1085 (28)
`934 (36)
`1271 (45)
`906 (49)
`
`44 ± 17
`[28-75]
`
`28 ± 14
`[12-47]
`
`58.2 (3)
`52.6 (4)
`46.9 (48)
`44.1 (61)
`45.9 (62)
`46.8 (41)
`23.4 (84)
`45.8 (42)
`18.6 (44)
`42.4 (47)
`
`43 ± 26
`[3-84]
`
`42 ± 9
`[31-53]
`
`567 (49)
`674 (61)
`953 (33)
`588 (50)
`853 (41)
`845 (46)
`800 (42)
`724 (17)
`650 (41)
`731 (41)
`
`42 ± 12
`[17-61]
`
`25 ± 14
`[13-46]
`
`65.1 (32)
`59.0 (43)
`43.6 (47)
`62.9 (34)
`52.6 (8)
`48.8 (45)
`52.3 (33)
`53.4 (45)
`62.6 (14)
`56.0 (32)
`
`33 ± 14
`[8-47]
`
`18 ± 12
`[2-28]
`
`511 (53)
`459 (68)
`327 (21)
`663 (33)
`562 (60)
`645 (88)
`696 (70)
`545 (52)
`623 (53)
`429 (37)
`
`55 ± 20
`[21-88]
`
`35 ± 15
`[21-59]
`
`79.6 (6)
`82.6 (12)
`79.5 (11)
`69.9 (10)
`75.5 (19)
`78.8 (12)
`74.0 (13)
`76.9 (11)
`70.9 (29)
`76.5 (14)
`
`17 ± 7
`[6-29]
`
`10 ± 7
`[2-53]
`
`1717 (39)
`1848 (31)
`1720 (43)
`
`-
`-
`-
`
`2520 (52)
`1770 (32)
`2080 (41)
`
`-
`-
`-
`
`1968 (46)
`1945 (48)
`2128 (33)
`
`-
`-
`-
`
`1220 (28)
`1730 (23)
`1620 (25)
`
`-
`-
`-
`
`Visit (week)
`
`Baseline
`0
`
`FTY720 treatment
`4a
`5a
`6a
`7a
`8a
`9a
`10a
`11a
`12a
`4-12a
`
`Average
`interindividual
`%CVb,c [range]
`Average
` intra-individual
`%CVd [range]
`
`Recovery
`16
`20
`24
`
`Percent coefficient of variation (%CV) is reported within parentheses. aLymphocyte count and reduction, P < 0.05 between groups (ANOVA); binterindividual %CV
`according to group, P > 0.05 for lymphocyte count, and P < 0.01 for reduction (ANOVA); cinterindividual %CV according to visit, P > 0.05 (general linear model for
`repeated measures); dintra-individual %CV according to group, P > 0.05 for lymphocyte count and P < 0.01 for reduction (ANOVA).
`
`Braz J Med Biol Res 38(5) 2005
`
`

`

`690
`
`S.I. Park et al.
`
`Figure 5. Linear regression analysis correlating mean FTY720-induced reduction of peripheral lymphocyte counts and drug dose (A) or
`concentrations (B).
`
`Figure 6. Pharmacokinetic and pharmacodynamic inhibitory Emax model correlating FTY720 dose (A) or blood concentration (B) with absolute
`lymphocyte counts in peripheral blood.
`
`Figure 7. Pharmacokinetic and pharmacodynamic simple Emax model correlating FTY720 dose (A) or blood concentration (B) with percent reduction
`in lymphocyte counts in peripheral blood.
`
`Braz J Med Biol Res 38(5) 2005
`
`

`

`FTY720 in kidney transplant recipients
`
`691
`
`of absolute lymphocyte counts did not differ
`according to FTY720 dose level. However, a
`decrease in both inter- (44-17%) and intra-
`individual variability (42-10%) ranges of per-
`cent reduction in lymphocyte counts ac-
`cording to FTY720 dose level was observed
`(P < 0.01). There was no significant differ-
`ence in interindividual %CV according to
`study visit.
`
`PK/PD analysis
`
`The observed linear correlation between
`FTY720 dose and steady-state blood con-
`centrations did not translate into linear corre-
`lations between dose or blood concentra-
`tions and lymphopenia. In the 10-fold dose
`range studied in this trial (0.25 to 2.5 mg/day)
`FTY720 pharmacodynamics was not dose-
`linear (r2 = 0.57 and 0.53), as can be ob-
`served in Figure 5A and B. Searching for a
`best-fit model we identified the inhibitory
`Emax model as the one with best prediction of
`the relationship between both FTY720 dose
`or blood concentrations and absolute periph-
`eral lymphocyte count. This model applies
`the following formula: E = Emax - (Emax - E0)
`* [C/(C + EC50)], where E is the pharmaco-
`dynamic effect at a given drug dose or
`concentration (C), Emax is the maximum phar-
`macodynamic effect attributed to the drug,
`E0 is the effect when drug dose or concen-
`tration is equal to zero, and EC50 is the drug
`dose or concentration which produces 50%
`of the maximum pharmacodynamic effect.
`The last parameter determines the drug po-
`tency. Using this model, the correlation be-
`tween individual FTY720 doses or steady-
`state blood concentrations and absolute pe-
`ripheral lymphocyte counts produced coef-
`ficients of determination of r2 = 0.96 and r2
`= 0.87, respectively. For FTY720 dose and
`concentration correlations, E0 and Emax were
`very similar, with EC50 of 0.08 mg or 0.13
`ng/ml, respectively (Figure 6A,B). Con-
`versely, the simple Emax model was the one
`that best predicted the relationship between
`
`either FTY720 dose or blood concentration
`and percent reduction in peripheral lympho-
`cyte count. This model applies the following
`formula: E = (Emax * C)/(C + EC50), where E
`is the pharmacodynamic effect at a given
`drug dose or concentration (C), Emax is the
`maximum pharmacodynamic effect attrib-
`uted to the drug, and EC50 is the drug dose or
`concentration which produces 50% of the
`maximum pharmacodynamic effect. Again,
`using this model, the correlation between
`individual FTY720 doses or steady-state blood
`concentrations and percent reduction in pe-
`ripheral lymphocyte counts produced coef-
`ficients of determination of r2 = 0.94 and r2
`= 0.89, respectively. For FTY720 dose and
`concentration correlations, Emax was 87.8%
`and 78.3%, with EC50 of 0.48 mg or 0.59 ng/
`ml, respectively (Figure 7A,B).
`
`Discussion
`
`FTY720 is a novel immunomodulating
`drug with a unique mechanism of action,
`which is different from the currently used
`immunosuppressive agents. Results from
`phase II clinical trials have demonstrated that
`the combination of FTY720 with cyclo-
`sporine produces a very low incidence of
`acute rejection (26). Furthermore, except
`for a first dose-associated mild negative
`chronotropic effect, FTY720 is not nephro-
`toxic, does not potentiate cyclosporine neph-
`rotoxicity and lacks other specific drug-
`related negative side effects (27).
`In the present study, we set out to deter-
`mine FTY720 PK/PD relationships in a co-
`hort of patients undergoing a dose-finding
`phase II clinical trial (28). The demographic
`characteristics revealed a low risk transplant
`population since this was an exploratory
`study, but no significant differences were
`observed among the study groups (Table 1).
`All patients received comparable initial cy-
`closporine doses and underwent similar dose
`reductions resulting in virtually equal cyclo-
`sporine concentrations throughout the first
`
`Braz J Med Biol Res 38(5) 2005
`
`

`

`692
`
`S.I. Park et al.
`
`24 post-transplant weeks. Moreover FTY720
`did not produce any significant drug interac-
`tion with cyclosporine since blood cyclo-
`sporine concentrations were similar for all
`FTY720 dose groups and for the MMF
`group (29).
`The pharmacokinetic properties of
`FTY720 have been described previously (30).
`The long terminal half-life required a loading
`dose to achieve steady-state concentrations
`more rapidly. Except for the higher dose
`level, whereby mean blood concentrations
`were still increasing after week 4 (Table 2,
`Figure 2), it appears that steady-state con-
`centrations were achieved after 4 weeks of
`daily administration. In this cohort of pa-
`tients, we observed a relatively low inter- and
`intra-individual variability in FTY720 trough
`blood concentrations, regardless of FTY720
`dose or time post-transplant. Moreover,
`FTY720 showed dose-linear pharmacoki-
`netics (13). Taken together, these character-
`istics may result in less need for dose adjust-
`ments, and fast and more predictable
`achievement of target concentrations when
`dose adjustments are needed.
`As shown in Table 3, MMF did not
`produce lymphopenia during the study pe-
`riod. FTY720-induced lymphopenia began
`early and could be detected as early as 1 week
`after transplantation (Figure 4). Actually, in
`more detailed studies, lymphopenia was al-
`ready noticed after the loading dose and
`paralleled the decrease in heartbeat (28). This
`has been attributed to the physiological mech-
`anism of action of FTY720, which, after oral
`administration, is phosphorylated (FTY720-
`P) and binds to sphingosine-1-phosphate
`receptors present in the plasma membrane of
`lymphocytes and cardiac myocytes, respec-
`tively (31,32). Nadir lymphopenia is reached
`between 2 and 4 weeks and after drug
`discontinuation lymphocyte counts return to
`baseline values within 4 to 8 weeks depend-
`ing on the magnitude of the FTY720 dose.
`This indicates that FTY720-induced lym-
`phopenia is reversible and therefore safer,
`
`since it does not produce prolonged apopto-
`sis-related lymphopenia and does not pro-
`duce long-term effects like those observed
`using depleting monoclonal or polyclonal
`antibodies (33). Inter- and intra-individual
`variability of the peripheral absolute lympho-
`cyte count is apparently higher than that seen
`for the FTY720 pharmacokinetics. Inter-
`and intra-individual variability of percent re-
`duction of lymphocyte counts decreases
`with increasing FTY720 doses, perhaps due
`to achievement of maximal pharmacody-
`namic effect.
`Visual observation of the relationship be-
`tween FTY720 doses or blood concentration
`absolute lymphocyte count or percent re-
`duction in lymphocyte count revealed a non-
`linear pattern (Figure 3), supported by the
`low coefficient of determination (r2) of the
`linear regression analysis (Figure 4). Search-
`ing for a best-fit model that would explain
`this relationship we found two PK/PD mod-
`els. For the absolute lymphocyte count, the
`inhibitory Emax model was selected since
`there was a decrease (inhibition) in the num-
`ber of lymphocytes in peripheral blood with
`increasing drug dose or concentration, show-
`ing high coefficients of determinations (Fig-
`ure 6). The E0 term was included to describe
`the effect when drug dose or concentrations
`were equal to zero (in this study, lymphocyte
`counts from patients receiving MMF). For
`the percent reduction in lymphocyte counts,
`the simple Emax model was selected since
`there was an increase in percent reduction of
`the number of lymphocytes in peripheral
`blood with increasing drug dose or concen-
`tration, also showing high coefficients of
`determination (Figure 7). There was an evi-
`dent difference in EC50 between the two PK/
`PD models for both FTY720 dose (0.08 vs
`0.48 mg) and blood concentration (0.13 vs
`0.59 ng/ml). A simple explanation for this
`finding is the large but frequently observed
`variability of the peripheral lymphocyte counts
`when drug dose or concentrations were
`equal to zero (E0 term), suggesting that the
`
`Braz J Med Biol Res 38(5) 2005
`
`

`

`FTY720 in kidney transplant recipients
`
`693
`
`use of percent reduction, which naturally
`corrects for baseline differences, is more
`appropriate as a surrogate marker of the
`pharmacodynamic effect of FTY720 in the
`blood compartment.
`Finally, relationships between pharmaco-
`kinetics and pharmacodynamics are useful
`to find ideal doses to produce the desired
`effect. However, it is still important to deter-
`mine the relationship between the pharmaco-
`dynamic effect and efficacy and safety,
`which is the ultimate goal of pharmacology.
`In a phase II clinical trial, the 2.5-mg dose of
`FTY720 showed good efficacy for the pre-
`vention of acute rejection after kidney trans-
`plantation. This dose produced a reduction
`of about 80% in lymphocyte count, close to
`the maximum effect (Emax). Since FTY720
`accumulates 11- to 19-fold in various tis-
`sues, perhaps in the secondary lymphoid
`organs, which are rich in lymphocytes, and
`therefore in sphingosine-1-phosphate recep-
`tors, the EC50 required for drug-receptor
`complex internalization might be higher.
`Moreover, data from the same phase II
`clinical trial demonstrated that FTY720 at
`doses of 5 mg/day, which probably produce
`
`maximum percent reduction of lymphocyte
`count, is still effective and safe. Recently, it
`has been demonstrated that the active form
`of FTY720 is the phosphorylated form
`(FTY720-P) and studies are needed to assess
`the correlation between FTY720-P and the
`degree of lymphopenia. Of course, the lacks
`of similar dose-response curves for toxicity
`impair the establishment of a therapeutic
`window.
`In conclusion, the PK/PD relationship of
`FTY720 follows an Emax model. Using pe-
`ripheral lymphopenia as an FTY720 pharma-
`codynamic surrogate marker, high pharma-
`codynamic