Pharmacodynamics of Single Doses of the Novel Immunosuppressant FTY720 in Stable Renal Transplant Patients


  • This work was funded by Novartis Pharmaceuticals Corp., Clinical Pharmacology, East Hanover, NJ.

* Corresponding author: Dr Klemens Budde,


FTY720, a new and potent immunosuppressant, causes in animal models a rapid, reversible reduction of all subsets of peripheral blood lymphocytes, inducing their migration to secondary lymphoid organs. In this human phase I trial, the pharmacodynamics of single oral doses of FTY720 were evaluated. A randomized, double-blind, placebo-controlled, time-lagged study of six different single ascending oral doses of FTY720 ranging from 0.25 to 3.5 mg was conducted in stable renal transplant patients receiving a cyclosporine-based regimen. Absolute and subset lymphocyte counts, as well as absolute differential leukocyte counts, were determined by differential blood counts and flow cytometry at screening and multiple intervals thereafter. A pharmacodynamic model was established. Twenty-four single doses of FTY720 that were administered caused a transient, reversible pan-lymphopenia within 4 h. Lymphocyte subgroup analysis revealed that almost all subsets declined, with CD4- and CD45RA-positive cells being affected the most. Natural killer cells, granulocytes and monocytes were not influenced by FTY720. The lymphocyte count returned to baseline within 72 h in all dosing cohorts except the highest. Pharmacokinetik/pharmacodynamic modelling revealed a nonlinear dose effect and resulted in a good fit with observed values. These data show that FTY720 is highly effective in humans, with single oral doses of FTY720 ranging from 0.25 to 3.5 mg causing a reversible selective panlymphopenia.


Despite recent progress, immunosuppression for the prevention and treatment of rejection after organ transplantation remains unsatisfactory. Although 1-year graft survival rates are high and acute rejection episodes continue to be reduced, there is a well defined need for an immunosuppressive agent free of the side-effects associated with currently used agents. FTY720 (FTY) is a new immunosuppressive compound that was discovered by structural modification of myriocin, a product of an ascomycete, Isaria sinclairii (1,2). FTY alone prolongs allograft survival in different animal transplantation models with remarkable potency (3–10). When used alone, FTY significantly prolongs allograft survival in animal studies of hepatic, renal, and cardiac transplantation. A synergistic effect between FTY and subtherapeutic or therapeutic doses of cyclosporine (CsA) was observed in several different allograft models (5–11). A similar synergistic effect was seen when FTY was added to subtherapeutic doses of sirolimus in a rat cardiac transplantation model (11). A recent study that combined 0.01 mg/kg/day FTY with low-dose everolimus (0.3–1.0 mg/kg/day) resulted in prolonged graft survival (12). Importantly, long-term use of FTY has not been associated with discernible side-effects or toxicities in these animal models (3–12).

The mechanism of action of FTY is not completely understood but is different from all other immunosuppressants. In experimental animals, the administration of FTY results in a rapid, transient depletion of peripheral blood lymphocytes, effecting T cells more than B cells, the CD3+, CD4+, and CD8+ T-cell lines appear to decrease the most (13–15). It was initially thought that FTY reduced circulating lymphocytes via induction of lymphocyte apoptosis (16,17), but more recent evidence suggests that lymphocyte sequestration into lymph nodes and Peyer's patches, mediated by an enhanced migratory response to homing chemokines, is the primary mode of action of FTY (18–20).

A consistent feature of all animal models treated at pharmacological doses of FTY is a decrease in lymphocyte count, which returns to baseline with cessation of drug dosing. As the pharmacodynamic effect of FTY is lymphopenia (12–15,18–20) a dose range that would be expected to result in lymphopenia was calculated using a previously described allometric scaling model (21). Allometric scaling models comparing animals with humans indicated that the single, initial doses of FTY used in this study, e.g. 0.25 mg or 0.50 mg, may result in lymphopenia. In all animal models tested, lymphopenia occurred at a dose 3–10-fold lower than the dose required for toxicity in the most sensitive animal, the dog. The design of this first in-human study exploited this feature of FTY by using the onset of significant, consistent lymphopenia as a stop point for further dose escalation. In a previous report (22) we described the safety, pharmacokinetics and reversible lymphopenia after single oral doses of FTY in stable renal transplant patients. The objective of the present study was to further evaluate the pharmacodynamic effect of single oral doses of FTY in renal transplant recipients. Here we analyze the kinetics of lymphocyte subsets and thoroughly explore the pharmacokinetic/pharmacodynamic relationship in this first human trial of FTY.

Materials and Methods

Study design and subjects

This was a randomized, double-blind, placebo-controlled, time-lagged, two-center, ascending single-oral dose study. The study was approved by the Ethics Committee at both centers. All subjects gave witnessed, informed, written consent before the onset of study procedures. The study protocol has been described in greater detail in a separate report (22). In brief, this entry-into-human study was designed to measure the pharmacodynamic effects of single doses of FTY (0.25, 0.5, 0.75, 1.0, 2.0 and 3.5 mg) in up to 32 stable renal allograft recipients, with four subjects (three randomized to FTY and one to placebo) in each treatment group. The pharmacodynamic measure of this study is the lymphocyte count. Based upon the results of an evaluation of safety and tolerability at each dose, the next group of subjects would proceed to a higher dose, repeat the same dose, revert to a lower dose, or stop dose escalation. Dose escalation was terminated if at least two subjects in a group developed a nadir lymphopenia ≤ 20% of the average baseline lymphocyte count, or if significant toxicity was detected. Subjects participating in a lower dose group could re-enter a higher-dose group 1 month later if they still fulfilled the inclusion and exclusion criteria, and had no lymphopenia.

Subjects were domiciled for 24 h before dosing with FTY and then for 96 h following dosing. Subjects fasted overnight before oral administration of study medication and consumed an average of 120 mL of water per hour before dosing and 24 h after dosing. Unless performing a study assessment, subjects had to rest quietly in the upright position for the next 4 h after administration of the drug. Strenuous physical exercise was prohibited for 7 days before dosing and until after study completion.

Subjects included into the trial were 18–65-year-old medically stable renal transplant recipients who had received their transplant at least 12 months earlier. All subjects had to be on CsA-based (Neoral™, Novartis Pharma, Basel, Switzerland) immunosuppression for at least 3 months, with a stable serum creatinine of ≤ 3.0 mg/dL. Concomitant therapy with mycophenolate, azathioprine or cyclophosphamide was prohibited. Subjects were excluded if they had received antilymphocytic therapy (OKT3, antithymocyte globulin, etc.) within the last year, suffered an episode of graft rejection within the last 6 months, or had persistent abnormally low lymphocyte counts (1.5 × 109/L).

Pharmacodynamic assessments

The lymphocyte count was the primary pharmacodynamic measure of this study. Whole blood was drawn by venipuncture from a peripheral forearm vein or in-dwelling venous cannula at screening (day-21 to day-2), baseline (day-1), 1 h and immediately predose, and at 0.5, 1, 2, 6, 12, 24, 48, 72 and 96 h postdosing into EDTA-containing tubes for differential blood counts. Absolute counts were analyzed with a Micro-Diff-II cell counter (Coulter-Immunotech Diagnostics, Hamburg, Germany) in a central laboratory. Complete leucocyte differential blood counts were performed at baseline, 24, 48, 72, 96 h after dosing. The absolute lymphocyte count data collected over time was used to derive the following variables:

I. Average baseline lymphocyte count defined as the mean of the baseline lymphocyte count taken the day before dosing and the lymphocyte count at time − 1 h and 0 h (predose)

II. Ratio of treated to placebo lymphocyte count defined as (average lymphocyte count of a particular dose at a particular time) divided by (average baseline lymphocyte count for that dose) divided by (average lymphocyte count for placebo at that time point) divided by (average baseline lymphocyte count for placebo).

This constructed variable corrects for any circadian variation in lymphocyte count observed in the placebo group, and isolates the effect of FTY alone on the lymphocyte count.

Lymphocyte subset analysis

Analysis of lymphocyte subsets was performed at screening, baseline, and at 12, 24, 48, 96 h after intake. For immunophenotyping two monoclonal antibodies and the appropriate isotype controls were used for direct immunofluorescence staining. In brief, whole blood was incubated for 10 min at room temperature with the monoclonal antibodies. One antibody was labelled with fluorescein isothiocyanate (FITC), and the other antibody was labelled with phycoerythrin (PE). In order to eliminate erythrocytes, the samples were incubated with lysis solution (ImmunoPrep-reagent, Coulter-Immunotech Diagnostics, Hamburg, Germany) using an automated, standardized system (Q-Prep, Coulter-Immunotech). The following lymphocyte subsets were measured using a two-color flow cytometer (Coulter EpicsXL/MCL; Coulter-Immunotech): CD45 (leucocyte common antigen) together with CD14 (monocyte), CD3 (T cell) in combination with CD4 (T-helper), CD8 (T-suppressor), CD16 (natural killer), and CD20 (B cell). CD4-positive cells were further characterized using specific antibodies against CD45RO (T memory) and CD45RA (T naïve). All antibodies were purchased from Coulter-Immunotech.

In order to better define the initial lymphopenia, all the patients from one center participated in a substudy. Blood was drawn at predose and 4, 8, 24 and 96 h after intake when CD3 (Tcells), CD4 (T-helper), CD8 (T-suppressor), CD16 (natural killer) and CD19 (B cells)-positive cells were investigated. Absolute lymphocyte numbers for the time points at 4 and 8 h were calculated as the individual mean value of the value after 2 and 6 h, or after 6 and 8 h, respectively. Direct immunofluorescence was performed using the whole-blood technique as described earlier using a two-color flow cytometer (Becton Dickinson, Heidelberg, Germany). In this set of experiments all monoclonal antibodies and the appropriate isotype controls were purchased from Becton Dickinson.

Statistics and pharmacodynamic modeling methodology

Lymphocyte counts were expressed as a percentage of their baseline value, expressed as percentage of the baseline value (PCTBL). The following model was fitted to the data of PCTBL vs. time:


The parameters _1 and _2 describe the rates of disappearance of lymphocytes from the systemic circulation and their reappearance, respectively. The dose dependence of the response was modeled by allowing _1 and _2 to vary across doses. The parameters were also allowed to vary across subjects by adding random subject effects to each parameter and fitting the resulting mixed-effects model using the SAS macro NLINMIX.

The area under the effect curve (AUE) was calculated by the linear trapezoidal rule using absolute lymphocyte counts. Descriptive statistics were computed for lymphocyte variables. Results from placebo subjects were pooled across groups. Paired t-tests were performed within groups to assess changes from baseline with a p-value of 0.05 considered to be significant. Kruskal–Wallis one-way analysis of variance on ranks was employed to detect differences in the median AUE-values among the treatment groups. To isolate the treatment group that differs from the others, a multiple comparison vs. placebo as a control group test was performed (Dunn's method).


Pharmacodynamics and pharmacodynamic modeling results

A total of 20 subjects were enrolled, with 12 re-enrolling in later dose groups, for a total of 32 pharmacodynamic profiles (24 FTY, 8 placebo). According to the protocol, the 0.25- and 0.5-mg dose groups were replicated in order to define more clearly the degree of lymphopenia observed in the initial test groups. Details of the patients' characteristics are the subject of a separate report (22). In brief, only one woman participated in this phase I trial, all the patients received a cyclosporine-based therapy, all but one in combination with low-dose steroids (prednisolone n = 14; methylprednisolone n = 5). All the patients were in a stable medical condition, with a creatinine clearance of 70.3 ± 19.0 mL/min (range: 42.3–128.4). Renal transplantation had been performed 8.7 ± 3.9 years (range: 3.3–16.2) ago. Concomitant therapy remained unchanged throughout the dosing interval. Except for clinically insignificant abnormalities, the safety laboratory data were unremarkable in pretreatment evaluations with normal leucocyte (7.92 ± 1.79 × 109/L) counts at screening.

As described in our initial report (22), all treatment groups exhibited a temporal reversible lymphopenia. The mean lymphocyte counts for each treatment group are plotted in Figure 1. Note that even after placebo, lymphocyte counts initially dropped, evincing an underlying circadian pattern under steroid therapy. All treated groups, 0.25–3.5 mg of FTY, consistently manifested a more pronounced decrease in lymphocyte counts compared with the placebo group.

Figure 1.

Lymphocytes as percent of baseline (mean ± standard error for different dosing cohorts) vs. hours post dose, with pharmacokinetic/pharmacodynamic model fit superimposed.

Exploratory examination of the data suggested that the doses 0.25–1.0 mg were not markedly different in terms of lymphocyte response. Hence, the first model to be fitted to the data posited that _1 and _2 each depended on whether the dose was in one of four categories: 0, 0.25–1, 2, or 3.5 mg. It was found that _1 did not vary significantly across dose categories, but _2 did. In the final model, _1 was assumed to have the same value for all four dose categories, while _2 was still allowed to vary across dose categories. The fitted curves from that model are superimposed on the data in Figure 1. Parameter estimates are displayed in Table 1. The estimated values of _1 and _2 in Table 1 should be considered as average values across patients. The intersubject coefficients of variation were estimated to be 35% and 37% for _1 and _2, respectively.

Table 1.  Parameter estimates of pharmacokinetic–pharmacodynamic model
Dose (mg)Estimated _1 ± SE (h−1)Estimated _2 ± SE (h−1)
00.291 ± 0.0270.148 ± 0.028
0.25–1 0.067 ± 0.007
2 0.061 ± 0.016
3.5 0.024 ± 0.007

In order to further explore the pharmacodynamic response to the drug and because the placebo-treated subjects exhibited a significant degree of diurnal variation of lymphocyte numbers we calculated a constructed variable. The constructed variable is the mean ratio of the relative treatment to relative placebo lymphocyte counts. This ratio corrects for the earlier-described variation in lymphocyte count observed in the placebo group, and instead isolates the effect of FTY alone. The time course of lymphopenia, stratified by dosage group, is presented graphically in Figure 2. Compared with placebo, all treatment groups manifested a temporal pattern of relative lymphopenia with a sharp decline of lymphocytes numbers, which was detectable at the latest by the 6-h time point. The lower dosing cohorts (0.25–1.0 mg FTY) had interchangeable curves with lowest values between 6 and 12 h and a slow recovery over time. The nadir of these groups was very similar and approximated 60% of placebo (range 57–60%). After 24 h, relative lymphopenia had recovered to 81 ± 16% of placebo, reaching baseline after 72 h (96 ± 12% of placebo) in these dose groups. The more rapid onset, the higher degree and the sustained effect of 3.5 mg FTY on lymphocyte numbers is clearly documented in the Figure 2. The curve for the 2.0-mg dosing cohort is between the highest and the lower dosing cohort with an exceptionally high value after 12 h as a result of an outlier.

Figure 2.

Effect of FTY on the ratio of treated to placebo lymphocyte count as a function of time post FTY dose.

Lymphocyte subset results

In order to better characterize the observed changes in lymphocyte numbers, we analyzed the lymphocyte subsets at several time points before and after the administration of FTY and placebo. Placebo-treated patients exhibited no significant changes in their lymphocyte subsets over the time points investigated (Table 2) with approximately 82 ± 6% T cells, 6 ± 2% B cells and 8 ± 5% NK cells. Stable expression of CD4 (49 ± 7% of lymphocytes) and CD8 (33 ± 5% of lymphocytes) with a CD4/CD8 ratio of 1.51 ± 0.41 was observed in the placebo-treated renal allograft recipients. Twenty-nine ± 4% of lymphocytes expressed CD45RO, and 22 ± 5% CD45RA.

Table 2A.  Mean number of cells by leucocyte subset, time post dose and treatment, and mean CD4/CD8 ratio by time post dose and treatment
Treatment Baseline12244896
PlaceboB cells135 ± 41157 ± 89149 ± 64148 ± 58155 ± 57
 T cells2082 ± 4672057 ± 2662246 ± 7292421 ± 8322498 ± 678
 CD41272 ± 3981234 ± 2241381 ± 5391421 ± 5311481 ± 485
 CD8825 ± 166820 ± 156913 ± 277966 ± 341959 ± 288
 CD4/CD81.54 ± 0.431.57 ± 0.471.52 ± 0.381.49 ± 0.391.59 ± 0.51
0.25 mgB cells111 ± 40102 ± 53104 ± 34122 ± 41147 ± 76
 T cells1700 ± 4951210 ± 5391518 ± 2971875 ± 5021971 ± 558
 CD41021 ± 297666 ± 355971 ± 3171152 ± 3721190 ± 458
 CD8612 ± 231488 ± 180523 ± 35666 ± 187701 ± 173
 CD4/CD81.80 ± 0.741.34 ± 0.341.86 ± 0.611.76 ± 0.571.74 ± 0.71
0.5 mgB cells142 ± 4591 ± 32110 ± 48101 ± 42129 ± 40
 T cells1871 ± 4271051 ± 3981474 ± 5951491 ± 5161729 ± 478
 CD41190 ± 308608 ± 166927 ± 410932 ± 3351106 ± 340
 CD8623 ± 304387 ± 287495 ± 280503 ± 287577 ± 321
 CD4/CD82.49 ± 1.572.29 ± 1.522.33 ± 1.312.34 ± 1.222.56 ± 1.52
0.75 mgB cells103 ± 3163 ± 2496 ± 21107 ± 24108 ± 19
 T cells2015 ± 3441038 ± 2971709 ± 3611920 ± 3041993 ± 404
 CD41157 ± 422425 ± 146911 ± 3721005 ± 3771043 ± 458
 CD8814 ± 148549 ± 225765 ± 208887 ± 243897 ± 272
 CD4/CD81.50 ± 0.720.81 ± 0.191.31 ± 0.791.24 ± 0.641.30 ± 0.77
1.0 mgB cells164 ± 31100 ± 7119 ± 36148 ± 13192 ± 99
 T cells2163 ± 5851144 ± 1241780 ± 162060 ± 2282123 ± 539
 CD41225 ± 448543 ± 63839 ± 1031011 ± 1321075 ± 249
 CD8837 ± 295556 ± 156825 ± 209969 ± 294892 ± 286
 CD4/CD81.58 ± 0.651.07 ± 0.481.09 ± 0.411.14 ± 0.461.28 ± 0.41
2.0 mgB cells143 ± 39105 ± 49106 ± 795 ± 11104 ± 21
 T cells2011 ± 8871387 ± 6871690 ± 4191694 ± 3811938 ± 522
 CD41200 ± 610628 ± 276951 ± 161984 ± 1581151 ± 423
 CD8747 ± 256700 ± 410702 ± 232745 ± 163749 ± 169
 CD4/CD81.54 ± 0.301.06 ± 0.441.45 ± 0.511.36 ± 0.351.56 ± 0.53
3.5 mgB cells 235 (n = 1)54 ± 1568 ± 2361 ± 2157 ± 31
 T cells3384 (n = 1)742 ± 101765 ± 114872 ± 239789 ± 200
 CD42726 (n = 1)361 ± 105423 ± 111473 ± 158486 ± 239
 CD81645 (n = 1)277 ± 103288 ± 95286 ± 126254 ± 49
 CD4/CD8 1.66 (n = 1)1.38 ± 0.391.50 ± 0.181.76 ± 0.521.91 ± 0.85
Table 2B.  Mean number of cells by leukocyte subset, time post dose and treatment
Treatment Baseline12244896
PlaceboCD45RO742 ± 244773 ± 133838 ± 344877 ± 325894 ± 289
 CD45RA644 ± 212523 ± 131615 ± 259645 ± 245640 ± 183
 NK cells174 ± 75213 ± 78266 ± 125203 ± 57192 ± 69
 Monocytes526 ± 139533 ± 139576 ± 251604 ± 171429 ± 178
 Granulocytes5106 ± 17815059 ± 18694795 ± 15584324 ± 12674774 ± 1951
0.25 mgCD45RO512 ± 130442 ± 253478 ± 124581 ± 178683 ± 200
 CD45RA561 ± 206298 ± 172495 ± 224592 ± 240580 ± 226
 NK cells169 ± 93165 ± 134133 ± 59147 ± 45139 ± 26
 Monocytes464 ± 202531 ± 129431 ± 224480 ± 261416 ± 141
 Granulocytes4100 ± 15854248 ± 5082813 ± 13863510 ± 8013420 ± 1170
0.5 mgCD45RO737 ± 189460 ± 149597 ± 255587 ± 197672 ± 198
 CD45RA506 ± 185169 ± 99348 ± 188370 ± 168443 ± 171
 NK cells229 ± 221138 ± 41198 ± 81165 ± 72166 ± 53
 Monocytes493 ± 120399 ± 84406 ± 90390 ± 93353 ± 111
 Granulocytes3752 ± 11514135 ± 19103977 ± 13043517 ± 12913760 ± 963
0.75 mgCD45RO687 ± 117312 ± 109520 ± 204571 ± 147579 ± 89
 CD45RA559 ± 387138 ± 69429 ± 270496 ± 303433 ± 261
 NK cells116 ± 8109 ± 44168 ± 4143 ± 9169 ± 16
 Monocytes603 ± 85400 ± 87477 ± 163520 ± 161503 ± 237
 Granulocytes5233 ± 7267257 ± 4835003 ± 3965457 ± 955067 ± 397
1.0 mgCD45RO716 ± 270449 ± 76599 ± 108654 ± 49661 ± 136
 CD45RA684 ± 262230 ± 68434 ± 35434 ± 35604 ± 128
 NK cells202 ± 88129 ± 78280 ± 224294 ± 172252 ± 47
 Monocytes327 ± 67407 ± 117600 ± 193670 ± 215390 ± 96
 Granulocytes3957 ± 12275633 ± 7174283 ± 14823813 ± 13603767 ± 2159
2.0 mgCD45RO703 ± 364550 ± 244648 ± 129620 ± 133723 ± 254
 CD45RA554 ± 122109 ± 18292 ± 53371 ± 78408 ± 104
 NK cells140 ± 61241 ± 209215 ± 58231 ± 50197 ± 84
 Monocytes413 ± 119367 ± 67400 ± 202623 ± 193497 ± 232
 Granulocytes5653 ± 29196613 ± 9284697 ± 17775113 ± 22084640 ± 1942
3.5 mgCD45RO1504 (n = 1)255 ± 52282 ± 43333 ± 87347 ± 156
 CD45RA1551 (n = 1)122 ± 53149 ± 72188 ± 35199 ± 117
 NK cells376 (n = 1)132 ± 59136 ± 50309 ± 168219 ± 109
 Monocytes780 (n = 1)270 ± 26297 ± 78397 ± 174357 ± 338
 Granulocytes5710 (n = 1)4250 ± 15063653 ± 20813440 ± 13683933 ± 1648

Flow cytometric analysis of lymphocyte subsets revealed that almost all lymphocyte subsets, including B cells, T cells, T-helper cells, T-suppressor cells, memory T cells and naïve T cells, were affected by FTY (Table 2). No changes occurred in the CD16-positive cells (NK cells) after FTY dosing. The highest dose group (n = 3) had the most intense and sustained effect on lymphocyte subsets. However, in general we did not observe a clear dose–response relationship: some patients in the lowest dose group developed a greater extent of relative lymphopenia compared with individuals who had received a higher dose of FTY. Because of the missing dose-relationship, the overall variability and the small numbers in each group the data for all FTY treatment groups were pooled (Figure 3A). Overall we observed a 41% decrease in lymphocyte numbers 12 h after dosing. At this time point the T cells had fallen by 37% (p < 0.001) compared with baseline, which was more pronounced in the CD4+ cells (45% decrease compared with baseline; p < 0.001) compared with the CD8+ cells (26% decrease compared with baseline; p < 0.003). Interestingly, the relative amount of CD4+ cells dropped from 49% at baseline to 39% (12 h after FTY; p < 0.001), whereas the relative amount of CD8+ cells increased from 29.6% to 32.3% (p < 0.004). As a consequence the ratio of CD4/CD8 declined from 1.88 at baseline to 1.45 (12 h after FTY; p < 0.004). Further analysis revealed that mainly the CD45RA+ naïve cells were affected by FTY. At baseline 0.61 ± 0.30 × 109/L CD45RA+ naïve cells (24 ± 8% of lymphocytes) were present. Twelve hours after dosing the absolute (0.19 ± 0.12 × 109/L; 41% of baseline; p < 0.001) and relative (13 ± 6% of lymphocytes; p < 0.001) amounts of CD45RA+ naïve cells had decreased significantly. Memory cells (CD45RO; 30% decease; p < 0.001) and B cells (27% decrease; p < 0.001) were affected to a lesser extent, without any changes in their relative amount. Twenty-four hours after dosing T-cell (84% of baseline; p < 0.01) and B-cell (86% of baseline; p < 0.01) numbers were still significantly lower compared with baseline. Again, CD4+ cells (82% of baseline; p < 0.02) and CD45RA+ naïve cells (73% of baseline; p < 0.005) were mainly affected, CD8+ cells (88% of baseline; n.s.) and CD45RO+ memory cells (86% of baseline; p < 0.05) to a lesser extent. Forty-eight hours after dosing all cell populations had recovered to baseline values. Except lymphopenia, FTY exhibited no further effect on the white blood cells, and the granulocyte and monocyte counts remained stable and within normal limits in all groups. The data for monocytes and granulocytes are included in Table 2B.

Figure 3.

Effect of FTY on lymphocytes and lymphocyte subsets: (A) 24-h time course of the mean of 24 profiles after the administration of different doses of FTY (0.25–3.5 mg). Twelve hours after a single dose of FTY all lymphocyte subpopulations, except NK cells (CD16+), were significantly (p < 0.01) lower compared with baseline. After 24 h all values, except for CD8 and CD16 cells, differed significantly (p < 0.05) from baseline, while after 48 h and 96 h no significant differences were detected. (B) Twenty-four-hour time course of the mean of 13 profiles after the administration of different doses of FTY (0.25–3.5 mg). After dosing all subpopulations decreased significantly (p < 0.05), except for B cells at the 24-h timepoint.

In additional experiments all 10 patients from one center (16 pharmacodynamic profiles: placebo: n = 3; 0.25 mg FTY: n = 3; 0.5 mg FTY: n = 4; 0.75 mg FTY: n = 1; 1.0 mg FTY: n = 2; 2.0 mg FTY: n = 1; 3.5 mg FTY: n = 2) participated in a substudy to further characterize the initial time course of lymphopenia. Again, in this set of experiments, placebo-treated subjects (n = 3) exhibited only minor changes in their lymphocyte subsets. Because of the small numbers, all patients from FTY treated cohorts were pooled for further analysis (Figure 3B). Except CD16+ cells, all lymphocyte subsets decreased significantly within 4 h after administration of FTY, reached a nadir 8 h after dosing, slowly recovered after 24 h, and reached baseline after 96 h. Four hours after intake lymphocyte numbers had decreased by 32 ± 14% (p < 0.001), while T cells had decreased by 38 ± 18% (p < 0.001), B cells by 22 ± 45% (p < 0.05). Already 4 h after dosing, CD4+ cells were affected to a greater extent (53 ± 18% decrease, p < 0.001) compared with CD8+ cells (27 ± 20% decrease, p < 0.001). Similar results were obtained after 8 h: the nadir being 64% (p < 0.001) of predose lymphocyte numbers, 56% (p < 0.001) of predose T-cell numbers, and 75% (p < 0.05) of predose B-cell numbers. Again, we observed the predominant effect upon CD4 cells (61% decrease, p < 0.001), and CD8 cells decreased to a lesser extent (41% decrease, p < 0.01). As reported earlier, lymphocyte (83%, p < 0.05), T cell (78%, p < 0.01), and CD8 numbers (86%, p < 0.05) in this set of experiments reached approximately 80% of predose after 24 h, while CD4+ cells had reached only 66% of predose values (p < 0.005). At this time point B cells had already reached predose values. Ninety-six hours after dosing, lymphocyte subsets had again reached predose values.

Pharmacokinetic–pharmacodynamic relationship

Further pharmacokinetic and pharmacodynamic analysis of FTY was performed to better understand the relationship of FTY plasma concentration and its presumed effect on lymphocyte count. Standard pharmacokinetic parameters (FTY dose, FTY blood concentrations at different time points, Cmax, AUC) exhibited only poor correlations (correlation coefficients less than 0.6) with all pharmacodynamic parameters (lymphocyte counts at different time points, nadir lymphocyte count, constructed variables, AUE) tested.

A measure of the 96-h AUE is given for all dose groups in Figure 4. Using Kruskal–Wallis one-way analysis of variance on ranks, there were statistically significant differences in the median values among the treatment groups (p = 0.030). However, we observed considerable overlap between treatments, and given the small sample size the difference in AUE values between lower doses of FTY and placebo did not reach statistical significance. The only treatment that showed statistically clear separation from the placebo effect was the highest, the 3.5-mg dosing group. However, a clear trend in increased effect with FTY relative to placebo is seen from the box plots in Figure 4. When analyzing the 24-h AUE the difference between the 0.25-mg and 0.5-mg FTY dose groups and placebo reached statistical significance (p < 0.05). Again, there was no separation of AUE values between different FTY doses ranging from 0.25 to 2.0 mg.

Figure 4.

Boxplots of the 96-h area under the effect curve vs. FTY dose. There were significant (p < 0.03) differences in the median values among the treatment groups (Kruskal–Wallis analysis), and the 3.5-mg dose group differed significantly from placebo (p < 0.05; Dunn's method).


There remains a need for a new immunosuppressant that can be used alone or in combination with existing therapies. Immunosuppressive drugs with distinct mechanisms of action but without overlapping toxicity offer the best options for synergistic antirejection therapy. FTY has a novel mechanism of action, is highly effective in animal models of transplantation without added toxicities (3–12), and is synergistic in combination with CsA (3–11), sirolimus (11) or everolimus (12,23). Thus, FTY has significant potential as a new immunosuppressive treatment. In this first human study of the use of FTY, single oral doses of FTY ranging from 0.25 to 3.5 mg were administered to stable renal transplant subjects on a CsA-based immunosuppression regimen. As FTY induces a dose-dependent lymphopenia in animals (12–15,18–20), the anticipated pharmacodynamic response in this Phase I study was a reduction in lymphocyte counts. In fact, all FTY-treated groups manifested lymphopenia that was more profound and followed a different time course compared with placebo. It is suggestive that the well-described phenomena of a circadian rhythm in combination with steroid-induced lymphopenia are responsible for the observed lymphopenia under placebo treatment (22). In order to isolate the effect of FTY alone we calculated a constructed variable, clearly showing the profound effect of FTY on lymphocyte numbers, which is clearly independent of any circadian rhythm. Similar to all animal models, lymphocyte numbers reflect the pharmacodynamic response to the drug. In future clinical trials the monitoring of lymphocyte numbers may be a valuable tool to assess the pharmacodynamic response in the individual.

Extending our previous report (22), the FTY effect upon lymphocyte counts was almost pan-lymphocytic, affecting all lymphocyte subsets except natural killer cells. As early as 4 h after dosing profound effects on the different lymphocyte subsets were documented. Interestingly CD4 and naïve T cells decreased to a greater extent. The predominant effect on CD4-positive cells has been described for nonhuman primates. Although the mechanism of action of FTY is unclear at the present time, there is increasing evidence that FTY modulates the sequestration into lymph nodes and Peyer's patches (18–20). Recently, accelerated homing of naïve lymphocytes to secondary lymphoid tissues under FTY treatment has been described (24). It is tempting to speculate that depletion of naïve cells reflects the sequestration into lymph nodes mediated by FTY. The mechanism by which FTY modulates lymphocyte homing to peripheral lesions remains to be investigated. While this lymphopenic effect of FTY is seen with multiple lymphocyte subsets, monocyte and granulocyte counts are unaffected by single doses of FTY. However it is of note that repeated administration of FTY may have different effects, as seen in the present single-dose study.

The interaction of single-dose FTY pharmacokinetics and pharmacodynamics is complex. A poor correlation between blood levels and the extent of lymphodepletion has been observed in nonhuman primates (14) similar to the present study. The rapid onset of lymphopenia with measurable effects on lymphocyte subsets, occurred at time points when drug levels started to rise. At maximal drug levels however (after 12–36 h), the lymphocyte counts already returned slowly to baseline values. Although FTY exhibits very predictable pharmacokinetics with a linear dose–response relationship (22), the pharmacodynamic effect was not linear over the dose range tested in this study. Further efforts in developing a pharmacokinetic/pharmacodynamic model are being explored. The current model describing the time course of lymphocyte response should be viewed as an exploratory effort to quantify the relationship between the pharmacodynamic response and FTY dose. There is clearly a need to improve the model by including a more complex, mechanistic component of FTY action on lymphocyte trafficking. As we continue to gather new experimental data, further efforts to refine the model and to extend the model into multiple-dose simulations will be ongoing.

For nonhuman primates, the lymphopenic effect of FTY has no linear dose–response relationship (14). In baboons, the response onset was faster and the duration was longer at higher doses, but the maximal lymphodepletion was similar within the large dose range (0.03–0.3 mg/kg) administered. The results of the present study support this observation: higher doses caused a more rapid and more sustained lymphopenia; however, the degree of lymphopenia showed only minor differences. Quesniaux et al. (14) observed decreasing interindividual variability of the pharmacodynamic response with increasing dosage. Especially during the recovery phase, interindividual variability was noted. Similar observations were made in the present study, although because of small numbers this can not be statistically proven.

Because drug combinations, rather than monotherapy, will continue to be used in the future after organ transplantation, the availability of drugs of different mechanisms of action that can be safely combined will be essential for further improvement of outcome post-transplant. It is foreseen that immunosuppression of the future will be ‘customized’ according to the stage after transplantation, the type of organ transplanted, and the different recipient populations, which will further increase the need for a larger choice of immunosuppressive agents. The unique mechanism of action, the absence of additive toxicities with cyclosporine or tacrolimus, and the unique pharmacodynamic profile of FTY offer important potential advantages in this setting. In summary, single oral doses of FTY in doses ranging from 0.5 mg to 3.5 mg caused a dose-dependent, reversible lymphopenia. FTY exerts its effect predominately on CD4 and naïve cells. Our data indicate that pharmacokinetics may be a poor predictor for the pharmacodynamic response to the drug. Pharmacodynamic monitoring by measuring the lymphocyte numbers may be more advisable for this new immunomodulator. Together, these data support further investigation of FTY to add potency and synergy, without toxicity, to an immunosuppressive regimen.


The authors gratefully acknowledge the assistance of Renate Schötschel, Markus Hiss, Manuela Schütz, and Denise Barilla.