Graft-versus-host disease can be separated from graft-versus-lymphoma effects by control of lymphocyte trafficking with FTY720 (original) (raw)
Lack of toxicity of FTY in syngeneic BMT recipients. There was no clinical evidence of toxicity in syngeneic BMT recipients receiving FTY at doses of 6, 3, 1, or 0.1 mg/kg/d for 29 or 100 days. In mice receiving 3 mg/kg/d for 29 or 100 days, spleen, liver, lung, skin, and large and small bowel did not show any histological evidence of tissue damage (data not shown).
FTY decreases the mortality of GvHD and reduces clinical GvHD. To determine whether FTY could inhibit GvHD, we lethally irradiated male B6D2F1 (H2bxd) mice (9.75 Gy) and reconstituted them with either syngeneic bone marrow (5 × 106 cells i.v.) for hematopoietic rescue, or an inoculum of haplotype-mismatched GvHD-inducing allogeneic marrow cells (5 × 106) and spleen cells (1 × 106) from male C3D2F1 (H2kxd) donors. Three different doses of FTY (3 mg/kg, 1 mg/kg, or 0.1 mg/kg) or sterile H2O as control were administered daily by oral gavage from day 0 until day 29 post-BMT. Mortality of the H2O-treated GvHD control mice began on day 7 post-BMT, and five of ten mice succumbed by day 13 (Figure 1a). The remaining mice in this group survived more than 29 days. In contrast, all recipients of FTY at a dose of 3 mg/kg (n = 10) or 1 mg/kg (n = 10) survived more than 29 days (P < 0.05 vs. allogeneic BMT with H2O). Three of ten recipients of the lowest dose of FTY (0.1 mg/kg) died by day 10 post-BMT, and the remainder of the group survived more than 29 days. The syngeneic control groups receiving either H2O or different doses of FTY (3 mg/kg/d, n = 5; 1 mg/kg/d, n = 5; 0.1 mg/kg/d, n = 5) all survived more than 29 days. Therefore, treatment with FTY at a daily dose of 3 or 1 mg/kg/d led to a decreased rate of mortality associated with haplotype-mismatched allogeneic marrow and spleen cell transplantation. The lowest dose of FTY (0.1 mg/kg/d) showed only weak protective activity.
FTY decreases GvHD mortality and inhibits GvHD-associated weight loss. (a) GvHD mortality. (b and c) Weight curves of the syngeneic groups (b) and allogeneic groups (c). Lethally irradiated B6D2F1 mice received haplotype-mismatched GvHD-inducing C3D2F1 BMCs (5 × 106) and spleen cells (1 × 106), as well as daily treatment, by oral gavage on days 0–29 post-BMT, with either H2O (filled inverted triangles; n = 10) or FTY at a dose of 3 mg/kg (filled circles; n = 10), 1 mg/kg (filled squares; n = 10), or 0.1 mg/kg (filled diamonds; n = 10). Syngeneic control groups received 5 × 106 B6D2F1 BMCs and either H2O (open inverted triangles; n = 5) or different doses of FTY (3 mg/kg, open circles, n = 5; 1 mg/kg, open squares, n = 5; or 0.1 mg/kg, open diamonds, n = 5). *Statistically significant difference compared with H2O-treated control group (P < 0.05).
We monitored all transplanted animals for clinical signs of GvHD. Figure 1, b and c, shows the average weight changes of the groups compared with those on day –1 pre-BMT. All transplanted animals, including syngeneic controls, showed initial weight loss. All syngeneic control groups later recovered their weights (Figure 1b). Control animals receiving allogeneic BMT without FTY rapidly lost more weight than syngeneic controls until day 9 post-BMT. The survivors began to regain weight on day 11 post-BMT (P < 0.05 on day 10 post-BMT vs. syngeneic BMT recipients treated with H2O) (Figure 1c). The maximum weight loss of the allogeneic control animals receiving only H2O was significantly greater than that in the group receiving allogeneic BMT and FTY at a daily dose of 3 mg/kg (P < 0.05 on days 7–13 post-BMT), 1 mg/kg (P < 0.05 on days 7–14 post-BMT), or 0.1 mg/kg (P < 0.05 on days 10–12 post-BMT), and the group receiving 3 mg/kg FTY showed less weight loss than did the groups treated with 1 or 0.1 mg/kg/d (P < 0.05) (Figure 1c). Thus, FTY at a dose of 3, 1, or 0.1 mg/kg/d alleviated GvHD-associated weight loss in a dose-related fashion.
Examination for generalized signs of GvHD (fur texture, posture, and inflammation of the eyes) revealed higher GvHD scores for the GvHD control group receiving allogeneic bone marrow and spleen cells (BMCs/SPCs) with H2O compared with the FTY-treated allogeneic groups. By day 8, the maximum mean GvHD score of 6.58 ± 0.42 was reached in the allogeneic GvHD control group, and all animals had diarrhea (Table 1). The group receiving allogeneic BMCs/SPCs and the lowest dose of FTY (0.1 mg/kg/d) showed a maximum mean GvHD score of 5 ± 0 on day 7, with diarrhea in only two of ten animals. The recipients of allogeneic BMCs/SPCs and FTY at a dose of 3 mg/kg/d or 1 mg/kg/d showed further reductions in GvHD. The group receiving 3 mg/kg/d FTY had a maximum mean GvHD score of 3 ± 0 on days 9, 12, and 13, with no diarrhea in any of the animals (n = 10). The group receiving FTY at a dose of 1 mg/kg/d had a maximum mean GvHD score of 2 on days 8 and 9, and two of ten animals had diarrhea on day 9. All clinical signs of GvHD resolved by day 18 post-BMT in all groups. Thus, FTY at a dose of 3 mg/kg/d and 1 mg/kg/d led to a marked reduction in the severity of GvHD-associated clinical signs.
Clinically reduced GvHD after FTY at a dose of 3 or 1 mg/kg/d
In order to explore the potential of FTY to inhibit a more rapidly lethal form of GvHD, we lethally irradiated B6D2F1 (H2bxd) mice (9.75 Gy) and reconstituted them with either syngeneic bone marrow (5 × 106 cells i.v.) for hematopoietic rescue, or haplotype-mismatched allogeneic marrow (5 × 106 cells) with an increased spleen cell inoculum (3 × 106 cells) from C3D2F1 (H2kxd) mice. To determine whether the presence of high levels of FTY at the time of transplant could better ameliorate GvHD, we included groups in which the administration of FTY was initiated on day –2 pre-BMT. To determine whether an augmented dose would further impede GvHD, we also evaluated a dose of 6 mg/kg/d of FTY. In groups receiving these doses starting on day –2 or day 0, FTY was continued until day 29 post-BMT. In order to determine whether continuation of treatment with FTY was required to maintain freedom from GvHD in the group receiving FTY at a dose of 3 mg/kg/d beginning on the day of BMT, we discontinued FTY administration in about half of the mice (n = 3) on day 29 and continued it in the remaining animals until day 100 post-BMT (n = 4).
In these recipients of higher spleen cell doses, mortality in the control group receiving only sterile H2O began on day 6 post-BMT (Figure 2a), and ten of ten mice succumbed by day 7. In contrast, recipients of 3 mg/kg/d FTY beginning on day 0, with discontinuation on day 29 or day 100, were almost completely protected from GvHD mortality (P < 0.05 vs. allogeneic BMT with H2O). Recipients of 3 or 6 mg/kg/d FTY from day –2 until day 29 showed similar protection from GvHD mortality to that observed in recipients of 3 mg/kg/d FTY beginning on day 0.
FTY inhibits GvHD induced by a rapidly lethal dose of allogeneic spleen cells. (a) GvHD mortality. (b and c) Weight curves of the syngeneic (b) and allogeneic (c) groups. Lethally irradiated B6D2F1 mice received haplotype-mismatched C3D2F1 BMCs (5 × 106) and an increased dose of spleen cells (3 × 106), as well as H2O (filled inverted triangles; n = 10) or FTY at a dose of 3 mg/kg (filled triangles; n = 10) or 6 mg/kg (×’s; n = 7), from day –2 pre-BMT until day 29 post-BMT. FTY was given at a dose of 3 mg/kg beginning on day 0 to an additional allogeneic group until day 29 post-BMT (filled circles; n = 3), or until day 100 post-BMT (+’s; n = 4). Syngeneic control groups received 5 × 106 B6D2F1 BMCs, as well as H2O (open inverted triangles; n = 5), different doses of FTY from day –2 until day 29 (3 mg/kg, open triangles, n = 5; 6 mg/kg, asterisks, n = 5), or 3 mg/kg FTY from day 0 until day 29 (open circles; n = 5).
The corresponding average weight changes compared with those on day 0 are shown in Figure 2, b and c. All experimental animals showed initial weight loss, but the control animals receiving only allogeneic BMCs/SPCs showed the greatest weight loss before they died. In contrast, all FTY-treated animals, regardless of the dose or time of initiation of administration, showed similar weight loss until day 7 post-BMT, then began to recover their weights on day 10. Beginning FTY on day –2, with or without an increase in dose to 6 mg/kg/d, did not further attenuate the weight loss compared with that observed in recipients of 3 mg/kg/d FTY beginning on day 0 (Figure 2c). Continuing the administration of FTY beyond day 29 did not result in any measurable difference in weight by day 100 compared with that of FTY-treated allogeneic recipients in which FTY was discontinued on day 29 (Figure 2c). The syngeneic controls treated with FTY (3 or 6 mg/kg/d) or water showed similar, milder weight loss compared with the allogeneic BMT recipients and showed recovery and continuous weight increases regardless of FTY dose (Figure 2b).
In the group receiving allogeneic BMT with water alone, the mean GvHD score on day 6 was 2.5 ± 0.24 of a possible score of 11, and all animals died the following day. The GvHD scores of the recipients of allogeneic BMCs/SPCs and FTY at a daily dose of 3 mg/kg starting on day –2 (1 ± 0) or day 0 (2 ± 0) or 6 mg/kg beginning on day –2 (1 ± 0) were reduced on day 6, and clinical signs of GvHD resolved by day 15 post-BMT in all surviving groups (data not shown). Only one animal (receiving the highest dose of 6 mg/kg/d from day –2 until day 29) showed any subsequent evidence of GvHD. Neither starting administration prior to the day of BMT nor increasing the dose of FTY led to further protection above the potent inhibition of GvHD observed with FTY administered at a dose of 3 mg/kg/d from days 0–29.
FTY reduces GvHD-associated tissue infiltration. To evaluate our hypothesis that FTY could reduce GvHD-associated tissue infiltration, we lethally irradiated B6D2F1 (H2bxd) mice (9.75 Gy) and reconstituted them with either syngeneic BMCs (5 × 106) and syngeneic spleen cells (1 × 106) (n = 2), or haplotype-mismatched GvHD-inducing allogeneic BMCs (5 × 106) and spleen cells (1 × 106) from C3D2F1 (H2kxd) mice with (n = 7) or without (n = 7) FTY at a dose of 3 mg/kg/d beginning on day 0. Animals were sacrificed on day 11 or 13, and tissues (skin, lung, liver, and small and large bowel) were preserved for histological analysis. Tissue infiltration consisting of lymphocytes, macrophages, and neutrophils was assessed as mild, moderate, or severe. In the untreated allogeneic group, two of three mice sacrificed on day 11 showed severe large-intestinal infiltration consisting of lymphocytes, macrophages, and neutrophils, thickening of the mucosa, and loss of crypts, and the third animal showed moderate infiltration. In contrast, in the allogeneic group receiving FTY, three of three mice showed only mild focal infiltration, with preserved crypts, less thickening of mucosa, and a few localized areas with loss of crypts and tissue infiltration, demonstrating a marked decrease in tissue pathology with FTY treatment. Among the untreated allogeneic control BMC/SPC recipients sacrificed on day 13, three of four animals showed severe, and the fourth animal showed moderate, inflammatory cell infiltration of the large bowel. In marked contrast, only one of four FTY-treated allogeneic recipients showed moderate infiltration, two of four showed mild focal infiltration, and one animal showed no pathological changes at all. Figure 3 depicts large-bowel tissues of representative animals from the syngeneic control group (Figure 3a), the untreated allogeneic group (Figure 3b), and the FTY-treated allogeneic group (Figure 3c), sacrificed on day 13. Syngeneic BMC/SPC control recipients showed no pathological changes on days 11 and 13, indicating recovery of the large intestine from the conditioning treatment. Liver, lung, and skin were normal in syngeneic and allogeneic groups on days 11 and 13, and the small intestine was largely unaffected in both allogeneic groups. Weight loss was more severe in the untreated than in the FTY-treated allogeneic group (P < 0.05 on day 10). These analyses demonstrate that FTY attenuates tissue infiltration associated with GvHD.
FTY reduces GvHD-associated tissue infiltration. Large-bowel tissues are shown (×50) from one representative animal in each group. Lethally irradiated B6D2F1 mice (9.75 Gy) were reconstituted with either syngeneic BMCs (5 × 106) and spleen cells (1 × 106) for hematopoietic rescue (n = 2), or an inoculum of haplotype-mismatched GvHD-inducing allogeneic BMCs (5 × 106) and spleen cells (1 × 106) from C3D2F1 mice with (n = 7) or without (n = 7) 3 mg/kg/d FTY (days 0–11 or 0–13). Tissues are from representative animals of the syngeneic control group (a), the untreated allogeneic group (b), and the FTY-treated allogeneic group (c) that were sacrificed on day 13 post-BMT.
GvL effects proceed unhindered, while FTY inhibits GvHD. To determine whether inhibition of GvHD by FTY would be associated with impaired GvL effects, we used a previously established T cell leukemia/lymphoma, EL4 (52). This H2b tumor is matched to the unshared host MHC haplotype. We lethally irradiated B6D2F1 (H2bxd) mice (9.75 Gy) and administered either syngeneic marrow (5 × 106 cells i.v.) for hematopoietic rescue, or an inoculum of haplotype-mismatched GvHD-inducing allogeneic marrow cells (5 × 106) and spleen cells (3 × 106) from C3D2F1 (H2kxd) mice; both groups also received 5,000 EL4 cells on the day of BMT. GvHD control groups received similar inocula without EL4.
Mortality induced by EL4 cells in the syngeneic control group receiving only H2O began on day 19 post-BMT. Median survival time (MST) was 19 days, and five of five mice died by day 21 post-BMT (Figure 4b). Similarly, the syngeneic BMT group receiving EL4 tumor and FTY treatment (3 mg/kg/d) showed mortality beginning on day 18 post-BMT, and six of six mice succumbed by day 21. Therefore, FTY had no measurable antitumor effect against EL4 in mice receiving syngeneic BMT. Macroscopic examination of carcasses in these groups showed clearly enlarged lymph nodes and spleen, and nodular infiltration of spleen, liver, and kidney (data not shown), consistent with EL4-induced death (48).
FTY inhibits GvHD mortality while permitting GvL effects to be observed. Survival curves of nonleukemic control groups (a) and leukemic groups (b) are shown. (a) Nonleukemic groups received syngeneic B6D2F1 BMCs (5 × 106) with either H2O (open inverted triangles; n = 7) or FTY (3 mg/kg/d; open circles; n = 4); or they received allogeneic haplotype-mismatched C3D2F1 BMCs (5 × 106) and spleen cells (3 × 106) with either H2O (filled squares; n = 10) or 3 mg/kg/d FTY from day 0 until day 29 post-BMT (filled triangles; n = 10). (b) Leukemic groups received 5,000 EL4 cells on day 0. Mice received 5 × 106 syngeneic B6D2F1 BMCs alone (×’s; n = 6) or with FTY (3 mg/kg/d; open diamonds; n = 6). Allogeneic groups received haplotype-mismatched C3D2F1 BMCs (5 × 106) and spleen cells (1 × 106) with either H2O (filled inverted triangles; n = 10) or FTY treatment (*; n = 10).
The GvHD control group receiving allogeneic BMCs and 3 × 106 donor spleen cells began showing mortality on day 7 post-BMT, and six of seven mice died by 8 days post-BMT (Figure 4a). In contrast, all FTY-treated mice (3 mg/kg from day 0 to day 29 post-BMT) that received similar inocula survived until day 100 post-BMT (n = 9; P < 0.05 vs. allogeneic group without FTY treatment) (Figure 4a). The allogeneic group receiving tumor (without FTY) began showing mortality on day 6 post-BMT, apparently due to GvHD, and eight of ten mice died by day 9 post-BMT (Figure 4b). Thus, a GvL effect could not be measured in this group, presumably because of rapid GvHD mortality. While several mice in the allogeneic tumor group receiving 3 mg/kg/d FTY treatment (days 0–29 post-BMT) died between day 5 and day 7 post-BMT, MST for the group was 32 days, and three of ten mice survived more than 100 days (P < 0.05 vs. allogeneic tumor group without FTY treatment, and P < 0.05 vs. syngeneic groups receiving EL4 with or without FTY) (Figure 4b). Thus, simultaneous protection from lymphoma and from GvHD mortality was observed in FTY-treated recipients of allogeneic BMCs/SPCs.
All animals showed initial weight loss (data not shown). The greatest weight loss occurred, as seen previously, in the allogeneic BMT groups that did not receive FTY treatment (P < 0.05 vs. allogeneic group with FTY). Clinical signs of GvHD were also most severe in the group receiving allogeneic BMT with water alone: on day 7 the mean GvHD score was 6.57 ± 0, and all animals died by day 10. In contrast, the average GvHD score of the recipients of allogeneic BMCs/SPCs and FTY at a dose of 3 mg/kg/d (days 0–29) was lower on day 7 (1.25 ± 0), and all surviving animals were free of signs of GvHD by day 11.
Thus, FTY treatment markedly inhibits GvHD mortality while allowing a significant antitumor effect to occur. Since allogeneic control mice that received tumor but not FTY died early from GvHD, an antitumor effect could only be seen in the FTY-treated mice that were protected from GvHD.
To compare the magnitude of GvL effects in allogeneic BMC/SPC recipients treated or not treated with FTY, we repeated the experiment with the allogeneic spleen cell dose (1 × 106) that did not previously cause lethal GvHD in control mice that did not receive FTY. We lethally irradiated B6D2F1 (H2bxd) mice (9.75 Gy) and administered either syngeneic marrow (5 × 106 cells i.v.) for hematopoietic rescue, or an inoculum of haplotype-mismatched allogeneic marrow cells (5 × 106) and spleen cells (1 × 106) from C3D2F1 (H2kxd) mice; both groups also received 5,000 EL4 cells on the day of BMT. Control groups received similar inocula without EL4. As in other experiments, allogeneic recipients of 3 mg/kg/d FTY with and without EL4 showed significantly less weight loss than the corresponding groups receiving similar inocula without FTY (P < 0.05 on days 7 and 8 post-BMT without EL4, P < 0.05 on days 7, 9, and 10 post-BMT with EL4). Almost all nonleukemic allogeneic control animals, with or without FTY treatment, survived until the end of the experiment, permitting comparison of GvL effects in untreated and FTY-treated allogeneic recipients. EL4-induced mortality in the syngeneic control group began on day 19 post-BMT, and seven of seven animals succumbed by day 21 post-BMT (Figure 5a). In contrast, in the allogeneic group receiving EL4 cells and H2O, leukemic mortality began 21 days post-BMT, MST was 50 days, and six of ten animals died by day 50 post-BMT (P < 0.05 vs. syngeneic BMT with EL4). Thus, allogeneic BMC/SPC administration significantly delayed and reduced the mortality associated with EL4, as we have previously reported (48). In this experiment, mortality in the FTY-treated allogeneic recipients was similar to that in the allogeneic group receiving EL4 cells and sterile H2O only. Leukemic mortality in the FTY-treated group began on day 25 post-BMT, MST was 32 days, and seven of ten animals died by day 42 post-BMT (P < 0.05 vs. syngeneic BMT with EL4; no significant difference vs. allogeneic BMT with EL4 and H2O) (Figure 5a). Thus, results of this experiment indicated that the magnitude of the GvL effect in mice receiving FTY was similar to that in the group receiving allogeneic BMT with no GvHD prophylaxis. This observation was confirmed in another experiment involving administration of 2,000 or 5,000 EL4 cells (data not shown).
FTY reduces clinical GvHD while preserving GvL. (a) Mortality. (b and c) Weight curves of the syngeneic (b) and allogeneic (c) groups. Lethally irradiated B6D2F1 mice received haplotype-mismatched allogeneic C3D2F1 BMCs (5 × 106) and spleen cells (1 × 106), with either H2O (filled squares; n = 10) or 3 mg/kg FTY treatment from day 0 until day 29 post-BMT (filled triangles; n = 10). Five thousand EL4 cells were administered to allogeneic groups on the day of BMT with H2O (filled inverted triangles; n = 10) or with FTY treatment (open diamonds; n = 10). Syngeneic control groups received 5 × 106 B6D2F1 BMCs with H2O (open inverted triangles; n = 5), with FTY (3 mg/kg; open circles; n = 5), or with 5,000 EL4 cells (×’s; n = 5).
To determine whether the tumor had been completely eradicated in the surviving allogeneic BMC/SPC recipients of EL4, we performed an adoptive transfer of spleen cells from individual surviving animals that had received allogeneic BMCs, spleen cells, and EL4 with (n = 3) or without (n = 4) 3 mg/kg FTY, into lethally irradiated host-type B6D2F1 secondary recipients. To eliminate any donor T cells that might have mediated an ongoing GvL effect at the time of sacrifice, these T cells were depleted ex vivo from both spleen cell and BMC inocula before transfer to secondary recipients. To ensure that residual donor T cells did not survive in the secondary recipients, these mice were treated with T cell–depleting mAb’s (anti-CD4 mAb GK1.5, anti-Ly2.2 mAb 2.43, and anti Ly2.1 mAb 116-13.1) on day –1 with respect to the cell transfer. EL4 does not express CD4 or CD8 (data not shown). All of the T cell–depleted BMCs and spleen cells from each experimental mouse were administered to a single B6D2F1 secondary recipient. As a control, one recipient was given T cell–depleted BMCs (5 × 106) and spleen cells (1 × 106) from a normal C3D2F1 donor. All recipients survived and remained healthy until day 100 post-BMT, when the experiment was terminated. These results suggest that EL4 had been completely eradicated in the original full-haplotype MHC-disparate BMT recipients treated or not treated with FTY (data not shown).
A therapeutic dose of FTY does not directly reduce EL4 proliferation or migration in vivo. We addressed the possibility that the antitumor effects observed in FTY-treated mice reflected, in part, a direct effect of FTY on EL4 tumor proliferation or migration into tissues. Thirty thousand EL4 cells were administered to normal B6 mice, which were sacrificed 5, 10, 17, or 24 days later (n = 3 per group per time point). Treatment with 3 mg/kg/d FTY for the duration of the experiment did not diminish the number of Vβ12+Thy1+ EL4 cells in the lymphoid tissues and did not delay or decrease the level of tumor infiltration in other tissues (kidney, liver, and lungs) compared with that in controls receiving EL4 with H2O treatment alone (data not shown). Thus, FTY had no direct antitumor effect on EL4 and did not decrease its migration into nonlymphoid organs.
FTY does not directly sensitize EL4 to alloreactive cytotoxic T lymphocytes. FTY has been reported to have a direct toxic effect on human and mouse tumor cell lines in vitro (53–56). Although the above studies in normal B6 mice and syngeneic BMT recipients rule out a direct antitumor effect of FTY in vivo, we performed an in vitro experiment to address the possibility that the drug might enhance the sensitivity of EL4 cells to killing by alloreactive cytotoxic T lymphocytes (CTLs), which we have previously shown to be the major effectors of GvL in the EL4 model (51). BALB/c anti-B6 CTLs were generated using standard techniques (57), and cytolytic activity was assessed using EL4 targets that were cultured in medium alone, or with 0.5, 1, or 2 μM FTY (concentrations similar to those used in the studies cited above; ref. 56) for the last 4 hours before 51Cr labeling. In two similar experiments, culture with FTY did not affect the lysis of EL4 by alloreactive CTLs compared with that of EL4 cultured in medium alone (data not shown).
A GvHD-inhibitory dose of FTY does not prevent donor-marrow engraftment. To determine whether full donor chimerism developed in animals protected from GvHD by FTY, chimerism was evaluated by FACS analysis of the peripheral WBCs on day 63 post-BMT in the FTY-treated animals presented in Figure 2. The results confirmed that the animals that received allogeneic BMCs/SPCs and that were protected from GvHD by FTY, regardless of dose or mode of administration, reconstituted fully with donor-type WBCs in all lineages (data not shown). Thus, FTY inhibits GvHD induced by allogeneic BMCs/SPCs while permitting full donor-type hematopoietic reconstitution.