- Email: [email protected]
In-vivo effect of interleukin 3 and erythropoietin, either alone or in combination, on the hematopoietic toxicity associated with zidovudine
IN-VIVO EFFECT OF INTERLEUKIN 3 AND ERYTHROPOIETIN, EITHER ALONE OR IN COMBINATION, ON THE HEMATOPOIETIC TOXICITY ASSOCIATED WITH ZIDOVUDINE Vincent
S. Gallicchio,1-5
Nedda K. Hughes,
‘-5 Kam-Fai
Tsel74
We studied the effect of erythropoietin (EPO) and interleukin 3 (IL-3), either alone or in combination, on the hematopoietic toxicity associated with zidovudine in vivo, as determined by peripheral blood indices, and assay of hematopoietic progenitors, i.e. erythroid (CFU-E/BFU-E), myeloid (CFU-GM) and megakaryocyte (CFU-Meg) from hone marrow and spleen. Previous studies from this laboratory have established that dose escalation of zidovudine to normal mice induced a dose-dependent decrease in hematocrit, white blood cells and platelets with altered populations of marrow and splenic erythroid, myeloid and megakaryocyte progenitors. Daily administration of EPO (50 U/animal, i.p.) and/or IL-3 (5 U/animal, i.p.) was associated with altered peripheral blood indices and progenitor cells. In general, use of EPO and IL-3 alone reduced zidovudine-induced toxicity, notably in erythropoiesis; however, combination EPO/IL3 was associated with enhanced toxicity with an observed rebound only with the use of c2.5 mg/ml drug; 2.5 mg/ml drug in the presence of combination EPO/IL3 accelerated zidovudine-erythroid toxicity. A similar response was noted with circulating platelets and megakaryocyte progenitors. Use of EPO or IL-3, either alone or in combination, failed to reverse zidovudine-induced neutropenia. These studies demonstrate that use of EPO or IL-3, either alone or in combination may serve as an effective adjuvant therapy to modulate the erythroid toxicity associated with lower doses of zidovudine; however, this cytokine therapy was ineffective modulating zidovudine-induced myelosuppression when used in vivo. A reversal in zidovudine-induced myeloid toxicity, therefore may require the use of a myelopoiesis inducing cytokine.
The drug zidovudine, 3’-azido-3’-deoxythymidine (AZT), a synthetic thymidine analog, has been used clinically in the management of acquired immune deficiency syndrome, A1DS.l Studies have demonstrated that zidovudine reduces morbidity and mortality late in the course of infection,2 and
1HematologyiOncology Division, Departments of From the 2Microbiology and Immunology, 3Clinical Sciences, Medicine, and “Toxicology, Lucille P. Markey Cancer Center, University of Kentucky Medical Center and 5Department of Veterans Affairs, Lexington, KY 405360084, USA. Address correspondence to: Dr Vincent S. Callicchio, Hematology/Oncology Division, CC-406, Lucille P. Markey Cancer Center, 800 Rose Street, Lexington, KY 4053GOO84, USA. Received 5 June 1992; revised and accepted for publication 9 October 1992 0 1993 Academic Press Limited 1043-4666/93/010062+ 10 $08.00/O KEY WORDS: erythropoietiniinterleukin cells/hematopoiesislzidovudine/hematopoietic
62
3ihematopoietic toxicity
stem
reduces morbidity when administered to symptomatic or asymptomatic patients earlier in the course of their infection.4 The drug is an effective inhibitor of viral reverse transcriptase activity due to its altered chemical structure having an azido group in place of hydroxyl group at the 3’-carbon position in the basic sugar molecule.5 This alteration results in the inability of newly synthesized nucleotides, e.g. zidovudine triphosphate, to be effective in attachment to continue the 5’- to 3’-phosphodiester linkage necessary to construct DNA chain synthesis. Although zidovudine has been demonstrated clinically to induce immunologic improvement, to decrease the incidence of opportunistic infections and to reduce AIDS mortality, an unfortunate side-effect has been the induction of hematopoietic suppression manifested by anemia, neutropenia and overall bone marrow failure.s-1” The hematopoietic toxicity associated with zidovudine has been identified, however, its mechanism has remained poorly understood. Several in-vitro co-culture studies have identified zidovudineCYTOKINE,
Vol.
5, No. 1 (January),
1993: pp 62-71
IL-3,
induced marrow toxicity as the result of inhibition of hematopoietic progenitors, e.g. human colonyforming unit (CFU)-granulocyte/macrophage (CFUGM) ,11.12 CFU-erythroid (E) and burst-forming unit (BFU-E), and multipotential CFU-GEMM,l* and murine CFU-GM, BFU-E and CFU-megakaryocyte (Meg) ,I3 in addition to an inhibition of the number of marrow- and spleen-derived CFU-S, CFU-GM and BFU-E following in-vivo administration to mice.14,L5 Several hematopoietic growth factors/cytokines such as granulocyte-macrophage colony-stimulating factor (GM-CSF) and erythropoietin (EPO) are currently under investigation in order to evaluate their ability to modulate the hematopoietic toxicity associated with zidovudine therapy in the treatment of human-AIDS. Recently this laboratory has demonstrated that the cytokines IL-l, IL-616 and IL-317 reduce zidovudine toxicity on hematopoietic progenitors when combined in vitro. We report here the results of studies performed to evaluate whether EPO and IL-3, either alone or in combination, can modulate zidovudine toxicity when administered in vivo to normal mice.
erythropoietin
toxicity
i 63
Zidovudine-treated animals receiving EPO (1.0 and 2.5 mg/ml) failed to reverse zidovudine-induced neutropenia when evaluated over the 5-week treatment period. Low dose zidovudine, i.e. 0.1 mgiml produced only moderate neutropenia. Normal control animals receiving zidovudine by 3-weeks treatment produced a 50% reduction in the WBC from the 2.5 mg/ml treatment group as previously reported.18
0 A a 0 n
0
1
2
3 Week
RESULTS
4
5
Control EPO 0.1 AZT/EPO 1 .O AZTlEPO 2.5 AZT/EPO
6
0 . a 0
Effect of IL-3 and EPO Either Alone or in Combination on Zidovudine-Induced Hematopoietic Toxicity The effect on hematocrit (Fig. lA-C) demonstrates that EPO alone, during the course of zidovudine treatment, was effective in ameliorating hematocrit reduction in the groups receiving 0.1 mgiml and 1.0 mg/ml (P < O.OS), but failed to reverse zidovudine anemia from the 2.5 mgiml group. Zidovudine control animals demonstrated a hematocrit value of 2.5 k 2 (by week 4 post-zidovudine treatment) as reported from this laboratory.18 Zidovudine-treated animals receiving IL-3 produced a reduced hematocrit value 2 weeks following initiation of therapy in all zidovudine groups that was dose-dependent. By week 4, zidovudinetreated mice (0.1 and 1.0 mgiml) had hematocrit values that were within the normal control range (P < 0.05); however, 2.5 mg/ml was still significantly below normal controls. With IL-3 and EPO, hematocrit values were within control range from groups receiving 0.1 and 1.0 mg/ml zidovudine (P < 0.05); however, IL-3/EPO together failed to reverse reduced hematocrit values when administered to the 2.5 mg/ml group. IL-3/EPO administered to normal control mice induced hematocrit values in the 70% range through the entire study (Fig. 1C).
and zidovudine
n
Control IL-3 0.1 AZT 1 .O AZT 2.5 AZT
0 A A 0 .
Control IL-3 + EPO 0.1 IL-3 + EPO 1.0 IL-3 + EPO 2.5 IL-3 + EPO
IL-3 IL-3 IL-3
Week Figure 1. (A) Effect of erythropoietin on hematocrit (%) from mice administered dose-escalation of zidovudine (0.1, 1.0 and 2.5 mg/ml). f SEM from triplicate studies. (B) Effect of IL-3 on hematocrit (%‘a) from mice administered dose-escalation of zidovudine (0.1, 1.0 and 2.5 mg/ml). f SEM from triplicate studies. (C) Effect of erythropoietin/lL-3 on hematocrit (%) from mice administered dose-escalation of zidovudine (0.1, 1.0 and 2.5 mgiml). f SEM from triplicate studies.
64 I Gallicchio et al.
Zidovudine-treated animals receiving IL-3 alone had reduced WBC; however, their reduction was not as severe when compared with zidovudine-treated controls,ls but were still below normal non-drug treated controls. IL-3 administered to normal animals produced a significant leukocytosis during weeks 2 and 3 (P < 0.05). Administration of IL-3 plus EPO to mice receiving dose-escalation of zidovudine is given in Fig. 2C. Combination of IL-3 and EPO failed to ameliorate zidovudine-induced neutropenia. By 3 weeks post-treatment mice receiving 1.0 and 2.5 mg/ml zidovudine had WBC values far below normal controls and were below zidovudine-treated controls.18 These results indicate IL-3 and EPO had a marginal effect and in combination enhanced zidovudine-induced myelosuppression when administered in vivo. The effect of IL-3 and EPO on circulating platelets in animals treated with zidovudine is given in Fig. 3A-C). Zidovudine-treated animals receiving EPO had reduced platelet values compared with normal controls, however, when compared with zidovudinetreated controls,18 administration of EPO reduced platelet reduction observed following 0.1 and 1.0 mg/ml zidovudine (P < 0.05). This was not observed in the 2.5 mg/ml zidovudine group. Administration of IL-3 to reverse reduced platelets was only observed in mice receiving low doses, i.e. 0.1 mg/ml and 1.0 mg/ml; however, in mice receiving 2.5 mgiml zidovudine, IL-3 failed to reverse zidovudine-induced thrombocytopenia (P < 0.05). IL-3 was capable of producing thrombocytosis in normal mice. Administration of IL-3 and EPO to zidovudine-treated mice (1.0 and 2.5 mg/ml) failed to reverse zidovudine induced thrombocytopenia.rs Animals treated with low dose, i.e. 0.1 mgiml responded to IL-3/EPO treatment with platelet values within the normal range. These results indicate, as was observed when monitoring WBC, that treatment with IL-3/EPO marginally influenced circulating platelet values in mice receiving zidovudine.
CYTOKINE,
Vol.
5, No.
1
(January1993:
62-71)
zidovudine treated. However, the increase was not sustained and gradually decreased during the remaining 3-week evaluation period with the lowest level observed from the 2.5 mg/ml treatment group. Splenic CFU-E were also increased from all treatment groups, control and zidovudine; however, unlike the marrow response, the splenic response was more sustained through 4 weeks. By week 5 CFU-E were at the
0 . A 0 n
I
I
1
2
,
,
3 Week
I
/
,I,
!,
3
S
6
0 . A 0 n
Control EPO 0.1 AZT/EPO 1 .O AZT/EPO 2.5 AZTIEPO
Control IL-3 0.1 AZT 1.0 AZT 2.5 AZT
IL-3 IL-3 IL-3
Week
0 Control
Effect of IL-3 and EPO, Either Alone or in Combination, on Zidovadine-Induced Toxicity Assessment of Hematopoietic Progenitors The ability of EPO and IL-3 to influence several classes of hematopoietic progenitors, i.e. erythroid (CFU-E and BFU-E), myeloid (CFU-GM) and megakaryocyte (CFU-Meg) were evaluated from bone marrow and spleen from mice receiving zidovudine in vivo . The effect of EPO on the late erythroid progenitor cell CFU-E, following zidovudine treatment in vivo is given in Fig. 4A, bone marrow (top) and spleen (bottom). Marrow CFU-E were increased following two weeks of treatment from all groups, control and
A a 0 .
IL-3 + EPO 0.1 IL-3 + EPO 1.0 IL-3 + EPO 2.5 IL-3 + EPO
Figure 2. (A) Effect of erythropoietin on WBC (X 103) from mice admininistered dose-escalation of zidovudine (0.1, 1.0 and 2.5 mg/ml). + SEM from triplicate studies. (B) Effect of IL-3 on WBC (X 10s) from mice administered dose-escalation of sidevudine (0.1, 1.0 and 2.5 mg/ml). f SEM from triplicate studies. (C) Effect of erythropoietin/IL-3 on WBC (X 103) from mice administered dose-escalation of zidovudine (0.1, 1.0 and 2.5 mg/ml). f SEM from triplicate studies.
IL-3, erythropoietin
level of the normal control. The effect of IL-3 on CFU-E is given in Fig. 4B, bone marrow (top) and spleen (bottom). Marrow CFU-E were below controls during the first 2-week period followed by an increase above control only from the 1.0 mg/ml zidovudine group at week 5 post-treatment. From IL-3-treated normal controls, marrow CFU-E were increased above the normal control by week 2 post-treatment. Splenic-
10
A
0 Control A EPO A 0.1 AZTiEPO 0 1.O AZTiEPO n 2.5 AZTIEPO
‘* 1 B
f
0 Control . IL-3 A 0.1 AZT IL-3 o I .O AZT IL-3 n 2.5 AZT IL-3
‘0
1
2
3
4
5
6
Week
Control IL-3 + EPO 0.1 IL-3 + EPO 1.O IL-3 + EPO 2.5 IL-3 + EPO
Figure 3. (A) Effect of erythropoietin on platelets (X 105) from mice administered dose-escalation of zidovudine (0.1, 1.0 and 2.5 mg/ml). + SEM from triplicate studies. (B) Effect of IL-3 on platelets (X 10s) from mice administered dose-escalation of zidovudine (0.1, 1.0 and 2.5 mgiml). f SEM from triplicate studies. (C) Effect of erythropoietin/IL-3 on platelets (X 10s) from mice administered dose-escalation of zidovudine (0.1, 1.0 and 2.5 mg/ml). f SEM from triplicate studies.
and zidovudine toxicity / 65
derived CFU-E were increased above controls with the greatest increase observed 3 weeks post-treatment and significantly from the 1 .O mg/ml zidovudine group (P < 0.01). The effect of combination EPO/IL-3 on CFU-E is given in Fig. 4C; marrow (top) and spleen (bottom). Combination treatment was increased in marrow CFU-E at 1 week post-treatment from the 1.0 mg/ml group with the most significant response observed at 4 weeks post-treatment (P < 0.05); however, from the 2.5 mg/ml group CFU-E was below controls. From the spleen (bottom) CFU-E were elevated over the normal control but generally not as significant compared with IL-3IEPO treated controls. In summary, compared with CFU-E from zidovudine-treated controls,rs treatment with EPO and IL-3, alone or in combination, increased CFU-E early which was not sustained during the course of treatment in all groups. The effect of EPO on the early erythroid progenitor cell BFU-E, following zidovudine treatment in vivo is given in Fig. 5A, bone marrow (top) and spleen (bottom). Marrow BFU-E in response to EPO were increased above normal control from zidovudinetreated mice only at 5 weeks post-treatment from all groups (P < 0.05). Splenic BFU-E were increased from all groups at several time points examined except at week 1 and from post-treatment and only from the EPO control. The response of BFU-E to IL-3 was more dramatic (Fig. 5B). Marrow BFU-E were significantly increased from controls, 0.1 and 1 .O mg/ml treated groups (P < O.Ol), while BFU-E from 2.5 mgiml were at the level of the normal control only at week 2 post-treatment. Splenic BFU-E in contrast were below normal control from all treatment groups except at week 2 from the 1 .O mgiml treatment group. BFU-E from animals treated with combined IL-3/EPO (Fig. 5C) demonstrated marrow increase at 4 weeks post-treatment from all groups except the 2.5 mg/ml group (P < 0.01). Splenic BFU-E were evaluated from all treatment groups except the 2.5 mg/ml group, which differed from the time of appearance, i.e. week 3 for the 0.1 mgiml treatment group, 1.0 mg/ml at week 3 and the IL-3/EPO control at week 4. Compared with zidovudine controls,ls BFU-E were increased following treatment with EPO, IL-3 or IL-3/EPO from marrow but only IL-3/EPO from spleen. The effect of EPO on the myeloid progenitor cell CFU-GM, following zidovudine treatment in vivo is given in Fig. 6A, bone marrow (top) and spleen (bottom). Marrow CFU-GM were reduced below normal controls at virtually all time points examined, with the values from the 2.5 mg/ml the most significant (P < 0.01). From the spleen, CFUGM were increased above normal controls during the early phase of treatment that peaked by week 1, which then gradually declined for the remaining 3 weeks.
64 I Gallicchio
0
CYTOKINE,
et al.
1
2
0 -.* Q -m-
Control EPO 0.1 AZT 1.0 AZT 2.5 AZT
IF c * -0 -m-
Control EPO 0.1 AZT 1.0 AZT 2.5 AZT
3 Week
+ EPO + EPO + EPO
4
5
6
0
I
3
3
5
5, No.
1 (January
1993: 62-71)
6
Week 0 Control A- IL-3 + EPO A 0.1 IL-3 + EPO -0 1.0 IL-3 + EPO -* 2.S IL-3 + EPO
Week
+ EPO + EPO + EPO
2
Vol.
b- 0.1 AZT 0 I.0 AZT -m- 2.5 AZT
+ IL-3 + IL-3 + IL-3
Week 4%
Control
-.ir o -=-
EPO + IL-3 0.1 IL-3 + EPO 1.0 IL-3 + EPO 2.5 IL-3 + EPO
-c-.* 0 -m-
Control IL-3 0.1 AZT 1.0 AZT 2.5 AZT
+ IL-3 + IL-.? + IL-.?
Figure 4. (A) Effect of erythropoietin on bone marrow (top), spleen (bottom) CFU-E (X 103) from mice administered dose-escalation of zidovudine (0.1, 1.0 and 2.5 mg/ml). f SEM from triplicate studies. (B) Effect of IL-3 on bone marrow (top), spleen (bottom) CFU-E (X 103) from mice administered dose-escalation of zidovudine (0.1, 1.0 and 2.5 mg/ml). k SEM from triplicate studies. (C) Effect of erythropoietin/IL-3 on bone marrow (top), spleen (bottom) CFU-E (X 103) from mice administered dose-escalation of zidovudine (0.1, 1.0 and 2.5 mg/ml). _+ SEM from triplicate studies.
Compared with zidovudine controls,is EPO, over the 5-week time-course failed to increase CFU-GM from marrow and spleen. IL-3 increased marrow CFU-GM only at the last time point examined (week 5) from all groups except the 2.5 mg/ml group (Fig. 6B). Splenic-derived CFU-GM following IL-3 treatment were increased from weeks 3 to 5 from all the zidovudine treatment groups. IL-3 failed to increase splenic-derived CFU-GM administered to normal animals. IL-3/EPO treatment failed to increase CFUGM from zidovudine-treated marrow significantly at any time point (Fig. 6C). This differed significantly from the splenic response, when compared with normal controls, all zidovudine-treated groups
expressed increased CFU-GM (P < 0.05). Compared with zidovudine controls,is only treatment with IL-3 alone was effective in reversing zidovudine-induced suppression of myeloid precursors CFU-GM. The effect of EPO on the megakaryocyte progenitor cell CFU-Meg following zidovudine treatment in vivo is given in Fig. 7A, bone marrow (top) and spleen (bottom). Marrow CFU-Meg were elevated above normal controls from zidovudine-treated mice with 1.0 mg/ml at week 4 and 2.5 mgiml at weeks 1, 3, 4 and 5 post-treatment. Splenic CFU-Meg were increased from all groups above normal control (P < 0.05). The effect of IL-3 on CFU-Meg is given in Fig. 7B. Marrow CFU-Meg were increased only at
IL-3.
erythropoietin
and zidovudine
toxicity
i 67
DISCUSSION
weeks 4 and 5 from zidovudine-treated 1.0 and 2.5 mgiml. Splenic CFU-Meg were increased over normal controls from 0.1 mgiml only at week 5, 1.0 mgiml at weeks 2-4, and 2.5 mgiml at weeks 4 and 5 (P < 0.05). IL-3IEPO in combination improved marrow production of CFU-Meg which was most significant at week 4 post-treatment (Fig. 7C) (P < 0.01). Splenic CFU-Meg showed a similar response: 0.1 mgiml that peaked at week 2 and 1.0 mgiml that peaked at week 4. Compared with zidovudine controls,r* EPO, IL-3 and IL-3IEPO produced an increase in CFU-Meg from spleen compared with bone marrow.
Although zidovudine has been shown to be an effective agent in prolonging life in HIV-infected patients,‘-4 zidovudine is not without toxic effects, particularly in relation to the bone marrow.9 The extent of marrow suppression is often the doselimiting factor in further therapy, and results in either dose reduction or discontinuation of drug therapy. The nature of current therapeutic strategies warrants continued zidovudine treatment, potentially for long durations, possibly for the remainder of a patient’s life, and a thorough analysis of the longterm consequences of zidovudine exposure and its
1 2
0
I
cl A A 0 n
I
t
2
3 Week
Control EPO 0.1 AZT+ 1 .O AZT+ 2.5 AZT+
I
I
EPO EPO EPO
I
4
I
1 I
5
I
t Week 0 i A 0 n
Control IL-3 0.1 AZT I .O AZT 2.5 AZT
Week D i A 0
+ IL-3 + IL-3 + IL-3
n
0 Week 0 A A 0 .
Control EPO 0.1 AZT+ 1.0 AZT+ 2.5 AZT+
EPO EPO EPO
1
Week 0 Control A IL-3 A 0.1 AZT 0 1.0 AZT n 2.5 AZT
+ IL-3 + IL-3 + IL-3
Control IL-3 + EPO 0.1 IL-3 + EPO 1.0 IL-3 + EPO 2.5 IL-3 + EPO
2
5 Week
0 . A 0 n
Control IL-3 + EPO 0.1 IL-3 + EPO 1.0 IL-3 + EPO 2.5 IL-3 + EPO
Figure 5. (A) Effect of erythropoietin on bone marrow (top), spleen (bottom) BFU-E (X 103) from mice administered dose-escalation of zidovudine (0.1,l.O and 2.5 mg/ml). 2 SEM from triplicate studies. (B) Effect of IL-3 on bone marrow (top), spleen (bottom) BFU-E (X loj) from mice administered dose-escalation of zidovudine (0.1, 1.0 and 2.5 mgiml). t SEM from triplicate studies. (C) Effect of erythropoietin/IL-3 on bone marrow (top), spleen (bottom) BFU-E (X 103) from mice administered dose-escalation of zidovudine (0.1, 1.0 and 2.5 mg/ml). 5 SEM from triplicate studies.
6
68 i Gallicchio
CYTOKINE,
et al.
140
Vol. 5, No. 1 (January
1993: 62-71)
60
B A
0
0 Week
Week 0 A /, 0 m
0 Control A EPO A 0.1 AZT + EPO Ol.OAZT+EPO l 2.5 AZT + EPO 30
Control IL-3 0.1 AZT 1.0 AZT 2.5 AZT
123456 Week Control A IL-3 + EPO A 0.1 IL-3 + EPO 0 1.0 IL-3 + EPO n 2.5 IL-3 + EPO q
+ IL-3 + IL-3 + IL-3
60 C
60 SO 40 30 20 10
0
123456
0
123456
Week 0 . A 0 n
Control EPO 0.1 AZT 1.0 AZT 2.5 AZT
+ EPO + EPO + EPO
1 Week
0 A A 0 n
Control IL-3 0.1 AZT 1.0 AZT 2.5 AZT
+ IL-3 + IL-3 + IL-3
0 . A 0 n
2
3 Week
4
5
6
Control IL-3 + EPO 0.1 IL-3 + EPO 1.0 IL-3 + EPO 2.5 IL-3 + EPO
Figure 6. (A) Effect of erythropoietin on bone marrow (top), spleen (bottom) CFU-GM (X 103) from mice administered dose-escalation of zidovudine (0.1, 1.0 and 2.5 mg/ml). + SEM from triplicate studies. (B) Effect of IL-3 on bone marrow (top), spleen (bottom) CFU-GM (X 103) from mice administered dose-escalation of zidovudine (0.1, 1.0 and 2.5 mg/ml). + SEM from triplicate studies. (C) Effect of erytbropoietin/IL-3 on bone marrow (top), spleen (bottom) CFU-GM (X 103) from mice administered dose-escalation of zidovudine (0.1, 1.0 and 2.5 mg/ml). i SEM from triplicate studies.
effects on hematopoiesis has recently been established.‘s Dose-escalation of zidovudine (0.1, 1.0 and 2.5 mg/ml) induced a dose-dependent decrease in hematocrit, WBC and platelets. High dose drug, i.e. > 1.0 mg/ml, reduced marrow CFU-E, while splenic CFU-E increased after 1 week then declined. BFU-E increased at weeks 1 and 2, then declined to control levels. Splenic BFU-E rose during the examination period in a pattern that was dose-dependent. Femoral CFU-GM was cyclic, i.e. with low-dose drug, 0.1 mg/ml, it increased gradually, then declined; with the higher dose (2.5 mg/ml) it increased initially, then declined below controls (week 6). Femoral CFU-Meg increased after the low-dose drug and was inhibited after high dose (2.5 mg/ml). Splenic CFU-Meg was
reduced initially, followed by an increase at week 4. The use of hematopoietic growth factors such as GM-CSF and erythropoietin (EPO) have been evaluated clinically in the treatment of the complications associated with human AIDS such as patients receiving zidovudine,s mitigation of the neutropenia of combined interferon-a and zidovudine treatment associated with HIV-Kaposi’s sarcoma,19 and in supplement to HIV-patients receiving chemotherapy for AIDS-related nonHodgkin’s lymphoma. 2o Erythropoietin continues to be evaluated for treatment of the anemia and overall inadequate erythopoiesis in patients with AIDS and Kaposi’s sarcoma.2,21 Although the clinical use of erythropoietin to reverse the anemia associated
IL-3,
/,I’
1
Week Control EPO O.lAZT+EPO l.OAZT+EPO 2.5 AZT + EPO
0 A A 0 n
Control IL-3 0.1 AZT 1.0 AZT 2.5 AZT
/I,,
2
3Week
5
6
0 A A 0
+ IL-3 + IL-3 + IL-3
n
Control IL-3 + EPO 0.1 IL-3 + EPO 1.0 IL-3 + EPO 2.5 IL-3 + EPO
i
/ 1
2
3
4
5
6
0
123456
Week
0 Control A EPO A O.lAZT+EPO 0 l.OAZT+EPO n 2.5 AZT + EPO
0 . A o n
/ 69
1
!,
4
6000
0
toxicity
-
T
0 A A 0 m
and zidovudine
and was inversely related to the hematocrit level.18 In response to long-term zidovudine treatment reported in this study, administered erythropoietin, either alone or in combination with IL-3, reversed zidovudineinduced anemia that was present following low and medium drug levels, but failed to reverse zidovudine anemia following use of the drug in high dose. As a consequence of zidovudine treatment administered to normal mice,l* erythroid progenitor cells, in particular BFU-E, were increased, notably in the spleen. These data suggested that erythropoietin was not the factor required to induce the differentiation of these progenitors and that induction of differentiation of this population would require the use of IL-3. This laboratory has recently demonstrated that use of
with zidovudine toxicity in AIDS patients has been reported,Q23 serum erythropoietin levels in the vast majority of AIDS patients not receiving zidovudine therapy have been found to be significantly elevated, and in some cases increased lOOO-fold, although not all AIDS-patients on zidovudine therapy have increased erythropoietin titers.21 For the vast majority of AIDS patients, the erythropoietin level is inversely related to the hemoglobin concentration, thereby indicating that the normal feedback loop mechanism that controls erythropoietin production in response to anemia is impaired in AIDS patients. In response to zidovudine treatment administered to normal mice, the erythropoietin titer was evaluated and was found to increase in response to dose-escalation of the drug
0
erythropoietin
Control IL-3 0.1 AZT 1.0 AZT 2.5 AZT
+ IL-3 + IL-3 + IL-3
0 A A 0 n
Week Control IL-3 + EPO 0.1 IL-3 + EPO 1.0 IL-3 + EPO 2.5 IL-3 + EPO
Figure 7. (A) Effect of erythropoietin on bone marrow (top), spleen (bottom) CFU-Meg (X 103) from mice administered dose escalation of zidovudine (0.1, 1.0 and 2.5 mg/ml). + SEM from triplicate studies. (B) Effect of IL-3 on bone marrow (top), spleen (bottom) CFU-Meg (X 103) from mice administered dose escalation of zidovudine (0.1, 1.0 and 2.5 m&ml). + SEM from triplicate studies. (C) Effect of erythropoietin/IL-3 on bone marrow (top), spleen (bottom) CFU-Meg (X 103) from mice administered dose-escalation of zidovudine (0.1, 1.0 and 2.5 mg/ml). f SEM from triplicate studies.
70 I Gallicchio
IL-3 is effective in ameliorating zidovudine toxicity on human-derived erythroid progenitor cells when combined in vitro.17 As described in these studies, use of IL-3 in vivo was effective in ameliorating zidovudine-induced suppression of erythropoiesis only when combined with erythropoietin. IL-3 alone induced an earlier observed reduction in erythropoiesis that was present following 0.1 and 1.0 mg/ml drug; the higher dose, 2.5 mg/ml, IL3 failed to reverse reduced erythropoiesis. These studies suggest that a combination of IL-3 and EPO may be efficacious in reversing zidovudine-induced hematopoietic toxicity, as measured by anemia and possibly thrombocytopenia; however, this treatment combination failed to reverse zidovudine-induced myeloid suppression. Reversal of myeloid toxicity may require inclusion of a myeloid stimulating growth factor such as GM-CSF or G-CSF. Such studies are currently in progress.
MATERIALS
CYTOKINE,
et al.
AND METHODS
Vol. 5, No. 1 (January 1993: 62-71)
daily basis); erythropoietin (EPO) 50 U per animal; sp. act. > 10 000 U/mg protein, Upjohn, Kalamazoo, MI, i.p. on a daily basis) or both growth factors administered daily for a 5-week period. Control groups (125 animals) received either IL-3, EPO alone or in combination with normal mice receiving sterile water. Dose-escalation of zidovudine, i.e. 0.1, 1.0 and 2.5 mg/ml and the effect on normal hematopoiesis in vivo has been recently reported from this laboratory.18
Assessment of Hematology
Parameters
Prior to sacrifice, blood was collected via cardiac puncture with 3-ml syringes containing a total of 1 mg of NazEDTA. Femurs and spleens were removed and processed as described above. Hematocrits were measured after microcapillary centrifugation. White blood count values were determined with a Coulter counter (model ZN), with differentials determined following staining of peripheral smears with Wright-Giemsa stain. Platelets were determined via use of the ammonium oxalate procedure. Femoral marrow and spleen cells were evaluated for their progenitor cell content of erythroid (CFU-E and BFU-E), granulocyte (CFU-GM) and megakaryocyte (CFU-Meg) using semisolid clonal assays in vitro as performed routinely in the laboratory.i”rs
Mice C57BU6 female mice, l&12 weeks of age, were purchased from Harlan Laboratories, Indianapolis, IN. Animals were housed in plastic cages and fed Purina Lab Chow and water ad libitum. After sacrifice by cervical dislocation, femurs and spleens were removed and singlecell suspensions were prepared using McCoy’s 5A medium (GIBCO, Grand Island, NY) following repeated passage through a 25gauge needle. After centrifugation (400 g, 10 min, 4”C), the cells were resuspended in McCoy’s medium for routine cell counting using a Coulter counter (model ZN; Coulter Electronics, Hialeah, FL). Cell viability (> 95%) was confirmed via use of the trypan blue exclusion test before plating.
Treatment Interleukin
of Animals and Zidovudine, 3 and Erythropoietin
The effect of sustained long-term zidovudine treatment and the ability of continuous interleukin 3 (IL-3) and erythropoietin (EPO), either alone or in combination, to modulate zidovudine toxicity was evaluated in the following studies. Normal mice were to receive zidovudine treatment in a dose-escalation fashion, i.e. 0.1 mg/ml, 1.0 mgiml and 2.5 mg/ml placed in sterile drinking water. Zidovudine in the form of the pure powder was kindly supplied by Dr Philip Furman, Burroughs-Wellcome, Research Triangle Park, NC. These zidovudine drug concentrations corresponded to a suboptimal, optimal and pharmacological dose of drug with respect to its anti-viral potency. Zidovudine levels were confirmed using high-performance liquid chromatography analysis. Normal control animals received only sterile water. Selected groups (125 animals) of zidovudine-treated animals received either IL-3 (5 U per animal; 0.5 pgikglbw; sp. act. 1 x 107 U/ml, Genzyme, Cambridge, MA, i.p. on a
Statistical Analysis Mean -t SE for values comparing control with experimental utilizing triplicate studies were determined and analyzed using the two sample ranks test of Wilcoxon-White (P < 0.05) to determine significance.
Acknowledgements
These studies were supported in part by a Merit Review Grant awarded from the Department of Veterans Affairs. The authors are grateful to Mrs Brenda Frazier for typing this manuscript.
REFERENCES 1. Yarchoan R, Klecker RW, Weinhold KJ, Markham PD, Lyerly HK, Durack DT, Galmann E, Nursinoff-Lehrman S, Blum RM, Barry DW, Shearer GM, Fischl MA, Mitsuya H, Gallo RC, Collins JM, Bolognesi DP, Myers S (1986) Administration of 3’-azido-3’-deoxythymidine, an inhibitor of HTLV-IIIILAV replication, to patients with AIDS or AIDS-related complex. Lancet i:575-577. 2. Fischl MA, Richman DD, Grieco MH, Gottlieb MS, Volverding PA, Laskin GL, Leedom JM, Groopman JE, Mildvan D, Schooley RT, Jackson GG, Durack DT, King D (1987) The efficacy of 3’-azido-3’.deoxythymidine in the treatment of AIDS and AIDS-related complex: a double-blind, placebo-controlled trial. N Engl J Med 317:185-191. 3. Fischl MA, Richman DD, Hansen N. Collier AC, Carey JT, Para MF, Hardy D, Donin R, Powderly WA, Allen JD, Wong B, Mergian TC, McAuliffe VG, Hyslop NE, Rhame FS, Balfour HH, Spector SA, Volberding P, Petinelli C, Anderson J, The AIDS Clinical Trials Group (1990) The safety and efficacy of zidovudine (AZT) in the treatment of mildly symptomatic human
IL-3,
immunodeficiency placebo-controlled
virus type I (HIV) infection: a double-blind trial. Ann Intern Med 112:727-737.
4. Volberding PA, Lagakos SW, Koch MA, Pettinelli CA, Myers MW, Booth DK, Balfour HH, Reichman RC, Bartlett JA, Hirsch MS, Murphy RL, Hardy WD, Soeiro R, Fischl MA, Bartlett JG, Merigan TC. Hyslop NE, Richman DD, Valentine FT, Corey L (1990) Safety and -efficacy of zidovudine in asymptomatic human immundeficiencv virus infected individuals with less than 500 CD4+ cells/mm’. N Engl J Med 322:941-949. 5. Forester G (1987) Profound azidothymidine. N Engl J Med 317:772.
cytopenia
secondary
to
6. Gill PS, Raick M. Brynes RH, Causey D, Loureiro C, Levine AM (1987) Azidothymidine associated with bone marrow failure in the acquired immunodeficiency syndrome (AIDS). Ann Int Med 107:502-505. 7. Gottlieb MB, Wolfe PR, Chafey S (1987) of AIDS-related thrombocytopenia to intravenous azidothymidine (3’.azido-3’.deoxythymidine). AIDS Retroviruses 3: 109: 112.
Response and oral Res Hum
8. Groopman JE, Mitsuyasu RT, deLeo MJ, Oette DH, Golde DW (1987) Effect of recombinant human granulocyte macrophage colony stimulating factor on myelopoiesis in the acquired immunodeficiency syndrome. N Engl J Med 317:593-598. 9. Richman DD, Fischl MA, Grieco MH. Gottlieb MS, Volberding PA, Laskin OL, Leedom JM, Groopman JE, Mildran D. Hirsch MS, Jackson GG, Durach DT, Nusinoff-Lehrman S, AZT Collaborative Working Group (1987) The toxicity of azidothymidine (AZT) in the treatment of AIDS and AIDS-related complex: A double-blind. placebo-controlled trial. N Engl J Mcd 317:192-197. 10. Dournon E, Rogenbaum W. Michon C, Penonne C, DeTruchis P. Bouret E. Leevachen M. Matherson S, Gharakhanian S, Girard PM, Salmon D, Ceport C, Dazza MC, Regnier B (1988) Effects of zidnvudine in 365 consecutive patients with AIDS or AIDS-related complex. Lancet 2: 1297-1300. 11. Sommadossi JP, Zhu Z, Carisle R (1987) Mechanism of toxicity of 3’.azido-3’.deoxythymidine in normal hematopoietic progenitor cells: evidence of inhibition of the de novo pyrimidine pathway. Proc Am Assoc Cancer Res 28:329A. 12. Dainiak N, Goldman L (1988) proliferation in vitro Br J Hacmatol69:299-304.
Worthington M, Riordon 3’-azido-3’-deoxythymidine of human haematopoietic
MA, Kreckzo S, (AZT) inhibits progenitor cells.
13. Gallicchio VS, Doukas MA, Hulette BC, Hughes NK. Gass C (1989) Protection of 3’-Azido-3’-deoxythymidine induced toxicity to murine hematopoietic progenitors (CFU-GM, BFU-E, and CFUMeg) with interleukin-1. Proc Sot Exp Biol Med 192:201-204.
erythropoietin
and zidovudine
toxicity
/ 71
14. Boglioio G, Lerza R, Mercibani M, Saviane A, Pannacciulli I (1989) Azidothymidine-induced depression of murine hemopoietic progenitor cells. Exp Hematol 16:938-940. 15. Cronkite EP, Bullis J (1990) In vivo toxicity of 3’.azido3’.deoxvthvmidine (AZT) on CBAiCa mice. Int J Cell Cloning 8:332-3;15.’ ’ ’ 16. Gallicchio VS, Hughes NK, Hulette BC, Noblitt LN (1991) Effect of IL-l, IL-6, GM-CSF and erythropoietin on the in vitro toxicity associated with AZT on human bone marrow haematopoietic progenitor stem cells: CFU-GM and BFU-E. Antiviral Chem Chemother 2:76-85. 17. Gallicchio VS, Hughes NK (1992) Influence of interleukin3 on zidovudine (AZT)-induced in vitro toxicity to human hematopoietic progenitors. Int J Cell Cloning 10:99-104. 18. Gallicchio VS, Hughes NK (1992) Suppression of murine hematopoiesis in vivo after chronic administration of zidovudine: evidence that zidovudine-induced anemia is the result of decreased bone marrow-derived. erythropoietin-responsive progenitor cells. Proc Sot Exp Biol Med 199:459-465. 19. Scadden DT, Bering HA, Levine JD, Bresnahan J, Evans L, Epstein C, Groopman JE (1991) Granulocyte-macrophage colony stimulating factor mitigates the neutropenia of combined interferon-alpha and zidovudine treatment of acquired immune deficiency syndrome-associated Kaposi‘s sarcoma. J Clin Oncol 9:802-808. 20. Kaplan LD, Kahn JO. Crowe S. Northfeet D, Neville P, Grossberg H, Abrams DI, Tracey J, Mills J, Volberding PA (1991) Clinical and virologic effects of recombinant human granulocyte-macrophage colony-stimulating factor in patients receiving chemotherapy for human immunodeficiency virus associated non-Hodgkin’s lymphoma: results of a randomized trial. J Clin Oncol 9:929-940. 21. Walker RE, Parker RI, Kovacs JA, Masur H, Lane HC, Carleton S. Kirk LE. Granick HR, Fauci AS (1988) Anemia and erythropoiesis in patients with the acquired immunodeficiency syndrome (AIDS) and Kaposi sarcoma treated with zidovudine. Ann Intern Med 108:372-376. 22. Fischl MA, Galpin JE, Levine JD, Groopman JE. Henry DH. Kennedy P. Miles S, Robbins W, Stanett B, ialusky R, Abeis RI, Tsai HC, Rusnick (1990) Recombinant human erythropoietin for patients with AIDS treated with zidovudine. N Engl J Med 322: 1488-1493.
S. Gallicchio,1-5
Nedda K. Hughes,
‘-5 Kam-Fai
Tsel74
We studied the effect of erythropoietin (EPO) and interleukin 3 (IL-3), either alone or in combination, on the hematopoietic toxicity associated with zidovudine in vivo, as determined by peripheral blood indices, and assay of hematopoietic progenitors, i.e. erythroid (CFU-E/BFU-E), myeloid (CFU-GM) and megakaryocyte (CFU-Meg) from hone marrow and spleen. Previous studies from this laboratory have established that dose escalation of zidovudine to normal mice induced a dose-dependent decrease in hematocrit, white blood cells and platelets with altered populations of marrow and splenic erythroid, myeloid and megakaryocyte progenitors. Daily administration of EPO (50 U/animal, i.p.) and/or IL-3 (5 U/animal, i.p.) was associated with altered peripheral blood indices and progenitor cells. In general, use of EPO and IL-3 alone reduced zidovudine-induced toxicity, notably in erythropoiesis; however, combination EPO/IL3 was associated with enhanced toxicity with an observed rebound only with the use of c2.5 mg/ml drug; 2.5 mg/ml drug in the presence of combination EPO/IL3 accelerated zidovudine-erythroid toxicity. A similar response was noted with circulating platelets and megakaryocyte progenitors. Use of EPO or IL-3, either alone or in combination, failed to reverse zidovudine-induced neutropenia. These studies demonstrate that use of EPO or IL-3, either alone or in combination may serve as an effective adjuvant therapy to modulate the erythroid toxicity associated with lower doses of zidovudine; however, this cytokine therapy was ineffective modulating zidovudine-induced myelosuppression when used in vivo. A reversal in zidovudine-induced myeloid toxicity, therefore may require the use of a myelopoiesis inducing cytokine.
The drug zidovudine, 3’-azido-3’-deoxythymidine (AZT), a synthetic thymidine analog, has been used clinically in the management of acquired immune deficiency syndrome, A1DS.l Studies have demonstrated that zidovudine reduces morbidity and mortality late in the course of infection,2 and
1HematologyiOncology Division, Departments of From the 2Microbiology and Immunology, 3Clinical Sciences, Medicine, and “Toxicology, Lucille P. Markey Cancer Center, University of Kentucky Medical Center and 5Department of Veterans Affairs, Lexington, KY 405360084, USA. Address correspondence to: Dr Vincent S. Callicchio, Hematology/Oncology Division, CC-406, Lucille P. Markey Cancer Center, 800 Rose Street, Lexington, KY 4053GOO84, USA. Received 5 June 1992; revised and accepted for publication 9 October 1992 0 1993 Academic Press Limited 1043-4666/93/010062+ 10 $08.00/O KEY WORDS: erythropoietiniinterleukin cells/hematopoiesislzidovudine/hematopoietic
62
3ihematopoietic toxicity
stem
reduces morbidity when administered to symptomatic or asymptomatic patients earlier in the course of their infection.4 The drug is an effective inhibitor of viral reverse transcriptase activity due to its altered chemical structure having an azido group in place of hydroxyl group at the 3’-carbon position in the basic sugar molecule.5 This alteration results in the inability of newly synthesized nucleotides, e.g. zidovudine triphosphate, to be effective in attachment to continue the 5’- to 3’-phosphodiester linkage necessary to construct DNA chain synthesis. Although zidovudine has been demonstrated clinically to induce immunologic improvement, to decrease the incidence of opportunistic infections and to reduce AIDS mortality, an unfortunate side-effect has been the induction of hematopoietic suppression manifested by anemia, neutropenia and overall bone marrow failure.s-1” The hematopoietic toxicity associated with zidovudine has been identified, however, its mechanism has remained poorly understood. Several in-vitro co-culture studies have identified zidovudineCYTOKINE,
Vol.
5, No. 1 (January),
1993: pp 62-71
IL-3,
induced marrow toxicity as the result of inhibition of hematopoietic progenitors, e.g. human colonyforming unit (CFU)-granulocyte/macrophage (CFUGM) ,11.12 CFU-erythroid (E) and burst-forming unit (BFU-E), and multipotential CFU-GEMM,l* and murine CFU-GM, BFU-E and CFU-megakaryocyte (Meg) ,I3 in addition to an inhibition of the number of marrow- and spleen-derived CFU-S, CFU-GM and BFU-E following in-vivo administration to mice.14,L5 Several hematopoietic growth factors/cytokines such as granulocyte-macrophage colony-stimulating factor (GM-CSF) and erythropoietin (EPO) are currently under investigation in order to evaluate their ability to modulate the hematopoietic toxicity associated with zidovudine therapy in the treatment of human-AIDS. Recently this laboratory has demonstrated that the cytokines IL-l, IL-616 and IL-317 reduce zidovudine toxicity on hematopoietic progenitors when combined in vitro. We report here the results of studies performed to evaluate whether EPO and IL-3, either alone or in combination, can modulate zidovudine toxicity when administered in vivo to normal mice.
erythropoietin
toxicity
i 63
Zidovudine-treated animals receiving EPO (1.0 and 2.5 mg/ml) failed to reverse zidovudine-induced neutropenia when evaluated over the 5-week treatment period. Low dose zidovudine, i.e. 0.1 mgiml produced only moderate neutropenia. Normal control animals receiving zidovudine by 3-weeks treatment produced a 50% reduction in the WBC from the 2.5 mg/ml treatment group as previously reported.18
0 A a 0 n
0
1
2
3 Week
RESULTS
4
5
Control EPO 0.1 AZT/EPO 1 .O AZTlEPO 2.5 AZT/EPO
6
0 . a 0
Effect of IL-3 and EPO Either Alone or in Combination on Zidovudine-Induced Hematopoietic Toxicity The effect on hematocrit (Fig. lA-C) demonstrates that EPO alone, during the course of zidovudine treatment, was effective in ameliorating hematocrit reduction in the groups receiving 0.1 mgiml and 1.0 mg/ml (P < O.OS), but failed to reverse zidovudine anemia from the 2.5 mgiml group. Zidovudine control animals demonstrated a hematocrit value of 2.5 k 2 (by week 4 post-zidovudine treatment) as reported from this laboratory.18 Zidovudine-treated animals receiving IL-3 produced a reduced hematocrit value 2 weeks following initiation of therapy in all zidovudine groups that was dose-dependent. By week 4, zidovudinetreated mice (0.1 and 1.0 mgiml) had hematocrit values that were within the normal control range (P < 0.05); however, 2.5 mg/ml was still significantly below normal controls. With IL-3 and EPO, hematocrit values were within control range from groups receiving 0.1 and 1.0 mg/ml zidovudine (P < 0.05); however, IL-3/EPO together failed to reverse reduced hematocrit values when administered to the 2.5 mg/ml group. IL-3/EPO administered to normal control mice induced hematocrit values in the 70% range through the entire study (Fig. 1C).
and zidovudine
n
Control IL-3 0.1 AZT 1 .O AZT 2.5 AZT
0 A A 0 .
Control IL-3 + EPO 0.1 IL-3 + EPO 1.0 IL-3 + EPO 2.5 IL-3 + EPO
IL-3 IL-3 IL-3
Week Figure 1. (A) Effect of erythropoietin on hematocrit (%) from mice administered dose-escalation of zidovudine (0.1, 1.0 and 2.5 mg/ml). f SEM from triplicate studies. (B) Effect of IL-3 on hematocrit (%‘a) from mice administered dose-escalation of zidovudine (0.1, 1.0 and 2.5 mg/ml). f SEM from triplicate studies. (C) Effect of erythropoietin/lL-3 on hematocrit (%) from mice administered dose-escalation of zidovudine (0.1, 1.0 and 2.5 mgiml). f SEM from triplicate studies.
64 I Gallicchio et al.
Zidovudine-treated animals receiving IL-3 alone had reduced WBC; however, their reduction was not as severe when compared with zidovudine-treated controls,ls but were still below normal non-drug treated controls. IL-3 administered to normal animals produced a significant leukocytosis during weeks 2 and 3 (P < 0.05). Administration of IL-3 plus EPO to mice receiving dose-escalation of zidovudine is given in Fig. 2C. Combination of IL-3 and EPO failed to ameliorate zidovudine-induced neutropenia. By 3 weeks post-treatment mice receiving 1.0 and 2.5 mg/ml zidovudine had WBC values far below normal controls and were below zidovudine-treated controls.18 These results indicate IL-3 and EPO had a marginal effect and in combination enhanced zidovudine-induced myelosuppression when administered in vivo. The effect of IL-3 and EPO on circulating platelets in animals treated with zidovudine is given in Fig. 3A-C). Zidovudine-treated animals receiving EPO had reduced platelet values compared with normal controls, however, when compared with zidovudinetreated controls,18 administration of EPO reduced platelet reduction observed following 0.1 and 1.0 mg/ml zidovudine (P < 0.05). This was not observed in the 2.5 mg/ml zidovudine group. Administration of IL-3 to reverse reduced platelets was only observed in mice receiving low doses, i.e. 0.1 mg/ml and 1.0 mg/ml; however, in mice receiving 2.5 mgiml zidovudine, IL-3 failed to reverse zidovudine-induced thrombocytopenia (P < 0.05). IL-3 was capable of producing thrombocytosis in normal mice. Administration of IL-3 and EPO to zidovudine-treated mice (1.0 and 2.5 mg/ml) failed to reverse zidovudine induced thrombocytopenia.rs Animals treated with low dose, i.e. 0.1 mgiml responded to IL-3/EPO treatment with platelet values within the normal range. These results indicate, as was observed when monitoring WBC, that treatment with IL-3/EPO marginally influenced circulating platelet values in mice receiving zidovudine.
CYTOKINE,
Vol.
5, No.
1
(January1993:
62-71)
zidovudine treated. However, the increase was not sustained and gradually decreased during the remaining 3-week evaluation period with the lowest level observed from the 2.5 mg/ml treatment group. Splenic CFU-E were also increased from all treatment groups, control and zidovudine; however, unlike the marrow response, the splenic response was more sustained through 4 weeks. By week 5 CFU-E were at the
0 . A 0 n
I
I
1
2
,
,
3 Week
I
/
,I,
!,
3
S
6
0 . A 0 n
Control EPO 0.1 AZT/EPO 1 .O AZT/EPO 2.5 AZTIEPO
Control IL-3 0.1 AZT 1.0 AZT 2.5 AZT
IL-3 IL-3 IL-3
Week
0 Control
Effect of IL-3 and EPO, Either Alone or in Combination, on Zidovadine-Induced Toxicity Assessment of Hematopoietic Progenitors The ability of EPO and IL-3 to influence several classes of hematopoietic progenitors, i.e. erythroid (CFU-E and BFU-E), myeloid (CFU-GM) and megakaryocyte (CFU-Meg) were evaluated from bone marrow and spleen from mice receiving zidovudine in vivo . The effect of EPO on the late erythroid progenitor cell CFU-E, following zidovudine treatment in vivo is given in Fig. 4A, bone marrow (top) and spleen (bottom). Marrow CFU-E were increased following two weeks of treatment from all groups, control and
A a 0 .
IL-3 + EPO 0.1 IL-3 + EPO 1.0 IL-3 + EPO 2.5 IL-3 + EPO
Figure 2. (A) Effect of erythropoietin on WBC (X 103) from mice admininistered dose-escalation of zidovudine (0.1, 1.0 and 2.5 mg/ml). + SEM from triplicate studies. (B) Effect of IL-3 on WBC (X 10s) from mice administered dose-escalation of sidevudine (0.1, 1.0 and 2.5 mg/ml). f SEM from triplicate studies. (C) Effect of erythropoietin/IL-3 on WBC (X 103) from mice administered dose-escalation of zidovudine (0.1, 1.0 and 2.5 mg/ml). f SEM from triplicate studies.
IL-3, erythropoietin
level of the normal control. The effect of IL-3 on CFU-E is given in Fig. 4B, bone marrow (top) and spleen (bottom). Marrow CFU-E were below controls during the first 2-week period followed by an increase above control only from the 1.0 mg/ml zidovudine group at week 5 post-treatment. From IL-3-treated normal controls, marrow CFU-E were increased above the normal control by week 2 post-treatment. Splenic-
10
A
0 Control A EPO A 0.1 AZTiEPO 0 1.O AZTiEPO n 2.5 AZTIEPO
‘* 1 B
f
0 Control . IL-3 A 0.1 AZT IL-3 o I .O AZT IL-3 n 2.5 AZT IL-3
‘0
1
2
3
4
5
6
Week
Control IL-3 + EPO 0.1 IL-3 + EPO 1.O IL-3 + EPO 2.5 IL-3 + EPO
Figure 3. (A) Effect of erythropoietin on platelets (X 105) from mice administered dose-escalation of zidovudine (0.1, 1.0 and 2.5 mg/ml). + SEM from triplicate studies. (B) Effect of IL-3 on platelets (X 10s) from mice administered dose-escalation of zidovudine (0.1, 1.0 and 2.5 mgiml). f SEM from triplicate studies. (C) Effect of erythropoietin/IL-3 on platelets (X 10s) from mice administered dose-escalation of zidovudine (0.1, 1.0 and 2.5 mg/ml). f SEM from triplicate studies.
and zidovudine toxicity / 65
derived CFU-E were increased above controls with the greatest increase observed 3 weeks post-treatment and significantly from the 1 .O mg/ml zidovudine group (P < 0.01). The effect of combination EPO/IL-3 on CFU-E is given in Fig. 4C; marrow (top) and spleen (bottom). Combination treatment was increased in marrow CFU-E at 1 week post-treatment from the 1.0 mg/ml group with the most significant response observed at 4 weeks post-treatment (P < 0.05); however, from the 2.5 mg/ml group CFU-E was below controls. From the spleen (bottom) CFU-E were elevated over the normal control but generally not as significant compared with IL-3IEPO treated controls. In summary, compared with CFU-E from zidovudine-treated controls,rs treatment with EPO and IL-3, alone or in combination, increased CFU-E early which was not sustained during the course of treatment in all groups. The effect of EPO on the early erythroid progenitor cell BFU-E, following zidovudine treatment in vivo is given in Fig. 5A, bone marrow (top) and spleen (bottom). Marrow BFU-E in response to EPO were increased above normal control from zidovudinetreated mice only at 5 weeks post-treatment from all groups (P < 0.05). Splenic BFU-E were increased from all groups at several time points examined except at week 1 and from post-treatment and only from the EPO control. The response of BFU-E to IL-3 was more dramatic (Fig. 5B). Marrow BFU-E were significantly increased from controls, 0.1 and 1 .O mg/ml treated groups (P < O.Ol), while BFU-E from 2.5 mgiml were at the level of the normal control only at week 2 post-treatment. Splenic BFU-E in contrast were below normal control from all treatment groups except at week 2 from the 1 .O mgiml treatment group. BFU-E from animals treated with combined IL-3/EPO (Fig. 5C) demonstrated marrow increase at 4 weeks post-treatment from all groups except the 2.5 mg/ml group (P < 0.01). Splenic BFU-E were evaluated from all treatment groups except the 2.5 mg/ml group, which differed from the time of appearance, i.e. week 3 for the 0.1 mgiml treatment group, 1.0 mg/ml at week 3 and the IL-3/EPO control at week 4. Compared with zidovudine controls,ls BFU-E were increased following treatment with EPO, IL-3 or IL-3/EPO from marrow but only IL-3/EPO from spleen. The effect of EPO on the myeloid progenitor cell CFU-GM, following zidovudine treatment in vivo is given in Fig. 6A, bone marrow (top) and spleen (bottom). Marrow CFU-GM were reduced below normal controls at virtually all time points examined, with the values from the 2.5 mg/ml the most significant (P < 0.01). From the spleen, CFUGM were increased above normal controls during the early phase of treatment that peaked by week 1, which then gradually declined for the remaining 3 weeks.
64 I Gallicchio
0
CYTOKINE,
et al.
1
2
0 -.* Q -m-
Control EPO 0.1 AZT 1.0 AZT 2.5 AZT
IF c * -0 -m-
Control EPO 0.1 AZT 1.0 AZT 2.5 AZT
3 Week
+ EPO + EPO + EPO
4
5
6
0
I
3
3
5
5, No.
1 (January
1993: 62-71)
6
Week 0 Control A- IL-3 + EPO A 0.1 IL-3 + EPO -0 1.0 IL-3 + EPO -* 2.S IL-3 + EPO
Week
+ EPO + EPO + EPO
2
Vol.
b- 0.1 AZT 0 I.0 AZT -m- 2.5 AZT
+ IL-3 + IL-3 + IL-3
Week 4%
Control
-.ir o -=-
EPO + IL-3 0.1 IL-3 + EPO 1.0 IL-3 + EPO 2.5 IL-3 + EPO
-c-.* 0 -m-
Control IL-3 0.1 AZT 1.0 AZT 2.5 AZT
+ IL-3 + IL-.? + IL-.?
Figure 4. (A) Effect of erythropoietin on bone marrow (top), spleen (bottom) CFU-E (X 103) from mice administered dose-escalation of zidovudine (0.1, 1.0 and 2.5 mg/ml). f SEM from triplicate studies. (B) Effect of IL-3 on bone marrow (top), spleen (bottom) CFU-E (X 103) from mice administered dose-escalation of zidovudine (0.1, 1.0 and 2.5 mg/ml). k SEM from triplicate studies. (C) Effect of erythropoietin/IL-3 on bone marrow (top), spleen (bottom) CFU-E (X 103) from mice administered dose-escalation of zidovudine (0.1, 1.0 and 2.5 mg/ml). _+ SEM from triplicate studies.
Compared with zidovudine controls,is EPO, over the 5-week time-course failed to increase CFU-GM from marrow and spleen. IL-3 increased marrow CFU-GM only at the last time point examined (week 5) from all groups except the 2.5 mg/ml group (Fig. 6B). Splenic-derived CFU-GM following IL-3 treatment were increased from weeks 3 to 5 from all the zidovudine treatment groups. IL-3 failed to increase splenic-derived CFU-GM administered to normal animals. IL-3/EPO treatment failed to increase CFUGM from zidovudine-treated marrow significantly at any time point (Fig. 6C). This differed significantly from the splenic response, when compared with normal controls, all zidovudine-treated groups
expressed increased CFU-GM (P < 0.05). Compared with zidovudine controls,is only treatment with IL-3 alone was effective in reversing zidovudine-induced suppression of myeloid precursors CFU-GM. The effect of EPO on the megakaryocyte progenitor cell CFU-Meg following zidovudine treatment in vivo is given in Fig. 7A, bone marrow (top) and spleen (bottom). Marrow CFU-Meg were elevated above normal controls from zidovudine-treated mice with 1.0 mg/ml at week 4 and 2.5 mgiml at weeks 1, 3, 4 and 5 post-treatment. Splenic CFU-Meg were increased from all groups above normal control (P < 0.05). The effect of IL-3 on CFU-Meg is given in Fig. 7B. Marrow CFU-Meg were increased only at
IL-3.
erythropoietin
and zidovudine
toxicity
i 67
DISCUSSION
weeks 4 and 5 from zidovudine-treated 1.0 and 2.5 mgiml. Splenic CFU-Meg were increased over normal controls from 0.1 mgiml only at week 5, 1.0 mgiml at weeks 2-4, and 2.5 mgiml at weeks 4 and 5 (P < 0.05). IL-3IEPO in combination improved marrow production of CFU-Meg which was most significant at week 4 post-treatment (Fig. 7C) (P < 0.01). Splenic CFU-Meg showed a similar response: 0.1 mgiml that peaked at week 2 and 1.0 mgiml that peaked at week 4. Compared with zidovudine controls,r* EPO, IL-3 and IL-3IEPO produced an increase in CFU-Meg from spleen compared with bone marrow.
Although zidovudine has been shown to be an effective agent in prolonging life in HIV-infected patients,‘-4 zidovudine is not without toxic effects, particularly in relation to the bone marrow.9 The extent of marrow suppression is often the doselimiting factor in further therapy, and results in either dose reduction or discontinuation of drug therapy. The nature of current therapeutic strategies warrants continued zidovudine treatment, potentially for long durations, possibly for the remainder of a patient’s life, and a thorough analysis of the longterm consequences of zidovudine exposure and its
1 2
0
I
cl A A 0 n
I
t
2
3 Week
Control EPO 0.1 AZT+ 1 .O AZT+ 2.5 AZT+
I
I
EPO EPO EPO
I
4
I
1 I
5
I
t Week 0 i A 0 n
Control IL-3 0.1 AZT I .O AZT 2.5 AZT
Week D i A 0
+ IL-3 + IL-3 + IL-3
n
0 Week 0 A A 0 .
Control EPO 0.1 AZT+ 1.0 AZT+ 2.5 AZT+
EPO EPO EPO
1
Week 0 Control A IL-3 A 0.1 AZT 0 1.0 AZT n 2.5 AZT
+ IL-3 + IL-3 + IL-3
Control IL-3 + EPO 0.1 IL-3 + EPO 1.0 IL-3 + EPO 2.5 IL-3 + EPO
2
5 Week
0 . A 0 n
Control IL-3 + EPO 0.1 IL-3 + EPO 1.0 IL-3 + EPO 2.5 IL-3 + EPO
Figure 5. (A) Effect of erythropoietin on bone marrow (top), spleen (bottom) BFU-E (X 103) from mice administered dose-escalation of zidovudine (0.1,l.O and 2.5 mg/ml). 2 SEM from triplicate studies. (B) Effect of IL-3 on bone marrow (top), spleen (bottom) BFU-E (X loj) from mice administered dose-escalation of zidovudine (0.1, 1.0 and 2.5 mgiml). t SEM from triplicate studies. (C) Effect of erythropoietin/IL-3 on bone marrow (top), spleen (bottom) BFU-E (X 103) from mice administered dose-escalation of zidovudine (0.1, 1.0 and 2.5 mg/ml). 5 SEM from triplicate studies.
6
68 i Gallicchio
CYTOKINE,
et al.
140
Vol. 5, No. 1 (January
1993: 62-71)
60
B A
0
0 Week
Week 0 A /, 0 m
0 Control A EPO A 0.1 AZT + EPO Ol.OAZT+EPO l 2.5 AZT + EPO 30
Control IL-3 0.1 AZT 1.0 AZT 2.5 AZT
123456 Week Control A IL-3 + EPO A 0.1 IL-3 + EPO 0 1.0 IL-3 + EPO n 2.5 IL-3 + EPO q
+ IL-3 + IL-3 + IL-3
60 C
60 SO 40 30 20 10
0
123456
0
123456
Week 0 . A 0 n
Control EPO 0.1 AZT 1.0 AZT 2.5 AZT
+ EPO + EPO + EPO
1 Week
0 A A 0 n
Control IL-3 0.1 AZT 1.0 AZT 2.5 AZT
+ IL-3 + IL-3 + IL-3
0 . A 0 n
2
3 Week
4
5
6
Control IL-3 + EPO 0.1 IL-3 + EPO 1.0 IL-3 + EPO 2.5 IL-3 + EPO
Figure 6. (A) Effect of erythropoietin on bone marrow (top), spleen (bottom) CFU-GM (X 103) from mice administered dose-escalation of zidovudine (0.1, 1.0 and 2.5 mg/ml). + SEM from triplicate studies. (B) Effect of IL-3 on bone marrow (top), spleen (bottom) CFU-GM (X 103) from mice administered dose-escalation of zidovudine (0.1, 1.0 and 2.5 mg/ml). + SEM from triplicate studies. (C) Effect of erytbropoietin/IL-3 on bone marrow (top), spleen (bottom) CFU-GM (X 103) from mice administered dose-escalation of zidovudine (0.1, 1.0 and 2.5 mg/ml). i SEM from triplicate studies.
effects on hematopoiesis has recently been established.‘s Dose-escalation of zidovudine (0.1, 1.0 and 2.5 mg/ml) induced a dose-dependent decrease in hematocrit, WBC and platelets. High dose drug, i.e. > 1.0 mg/ml, reduced marrow CFU-E, while splenic CFU-E increased after 1 week then declined. BFU-E increased at weeks 1 and 2, then declined to control levels. Splenic BFU-E rose during the examination period in a pattern that was dose-dependent. Femoral CFU-GM was cyclic, i.e. with low-dose drug, 0.1 mg/ml, it increased gradually, then declined; with the higher dose (2.5 mg/ml) it increased initially, then declined below controls (week 6). Femoral CFU-Meg increased after the low-dose drug and was inhibited after high dose (2.5 mg/ml). Splenic CFU-Meg was
reduced initially, followed by an increase at week 4. The use of hematopoietic growth factors such as GM-CSF and erythropoietin (EPO) have been evaluated clinically in the treatment of the complications associated with human AIDS such as patients receiving zidovudine,s mitigation of the neutropenia of combined interferon-a and zidovudine treatment associated with HIV-Kaposi’s sarcoma,19 and in supplement to HIV-patients receiving chemotherapy for AIDS-related nonHodgkin’s lymphoma. 2o Erythropoietin continues to be evaluated for treatment of the anemia and overall inadequate erythopoiesis in patients with AIDS and Kaposi’s sarcoma.2,21 Although the clinical use of erythropoietin to reverse the anemia associated
IL-3,
/,I’
1
Week Control EPO O.lAZT+EPO l.OAZT+EPO 2.5 AZT + EPO
0 A A 0 n
Control IL-3 0.1 AZT 1.0 AZT 2.5 AZT
/I,,
2
3Week
5
6
0 A A 0
+ IL-3 + IL-3 + IL-3
n
Control IL-3 + EPO 0.1 IL-3 + EPO 1.0 IL-3 + EPO 2.5 IL-3 + EPO
i
/ 1
2
3
4
5
6
0
123456
Week
0 Control A EPO A O.lAZT+EPO 0 l.OAZT+EPO n 2.5 AZT + EPO
0 . A o n
/ 69
1
!,
4
6000
0
toxicity
-
T
0 A A 0 m
and zidovudine
and was inversely related to the hematocrit level.18 In response to long-term zidovudine treatment reported in this study, administered erythropoietin, either alone or in combination with IL-3, reversed zidovudineinduced anemia that was present following low and medium drug levels, but failed to reverse zidovudine anemia following use of the drug in high dose. As a consequence of zidovudine treatment administered to normal mice,l* erythroid progenitor cells, in particular BFU-E, were increased, notably in the spleen. These data suggested that erythropoietin was not the factor required to induce the differentiation of these progenitors and that induction of differentiation of this population would require the use of IL-3. This laboratory has recently demonstrated that use of
with zidovudine toxicity in AIDS patients has been reported,Q23 serum erythropoietin levels in the vast majority of AIDS patients not receiving zidovudine therapy have been found to be significantly elevated, and in some cases increased lOOO-fold, although not all AIDS-patients on zidovudine therapy have increased erythropoietin titers.21 For the vast majority of AIDS patients, the erythropoietin level is inversely related to the hemoglobin concentration, thereby indicating that the normal feedback loop mechanism that controls erythropoietin production in response to anemia is impaired in AIDS patients. In response to zidovudine treatment administered to normal mice, the erythropoietin titer was evaluated and was found to increase in response to dose-escalation of the drug
0
erythropoietin
Control IL-3 0.1 AZT 1.0 AZT 2.5 AZT
+ IL-3 + IL-3 + IL-3
0 A A 0 n
Week Control IL-3 + EPO 0.1 IL-3 + EPO 1.0 IL-3 + EPO 2.5 IL-3 + EPO
Figure 7. (A) Effect of erythropoietin on bone marrow (top), spleen (bottom) CFU-Meg (X 103) from mice administered dose escalation of zidovudine (0.1, 1.0 and 2.5 mg/ml). + SEM from triplicate studies. (B) Effect of IL-3 on bone marrow (top), spleen (bottom) CFU-Meg (X 103) from mice administered dose escalation of zidovudine (0.1, 1.0 and 2.5 m&ml). + SEM from triplicate studies. (C) Effect of erythropoietin/IL-3 on bone marrow (top), spleen (bottom) CFU-Meg (X 103) from mice administered dose-escalation of zidovudine (0.1, 1.0 and 2.5 mg/ml). f SEM from triplicate studies.
70 I Gallicchio
IL-3 is effective in ameliorating zidovudine toxicity on human-derived erythroid progenitor cells when combined in vitro.17 As described in these studies, use of IL-3 in vivo was effective in ameliorating zidovudine-induced suppression of erythropoiesis only when combined with erythropoietin. IL-3 alone induced an earlier observed reduction in erythropoiesis that was present following 0.1 and 1.0 mg/ml drug; the higher dose, 2.5 mg/ml, IL3 failed to reverse reduced erythropoiesis. These studies suggest that a combination of IL-3 and EPO may be efficacious in reversing zidovudine-induced hematopoietic toxicity, as measured by anemia and possibly thrombocytopenia; however, this treatment combination failed to reverse zidovudine-induced myeloid suppression. Reversal of myeloid toxicity may require inclusion of a myeloid stimulating growth factor such as GM-CSF or G-CSF. Such studies are currently in progress.
MATERIALS
CYTOKINE,
et al.
AND METHODS
Vol. 5, No. 1 (January 1993: 62-71)
daily basis); erythropoietin (EPO) 50 U per animal; sp. act. > 10 000 U/mg protein, Upjohn, Kalamazoo, MI, i.p. on a daily basis) or both growth factors administered daily for a 5-week period. Control groups (125 animals) received either IL-3, EPO alone or in combination with normal mice receiving sterile water. Dose-escalation of zidovudine, i.e. 0.1, 1.0 and 2.5 mg/ml and the effect on normal hematopoiesis in vivo has been recently reported from this laboratory.18
Assessment of Hematology
Parameters
Prior to sacrifice, blood was collected via cardiac puncture with 3-ml syringes containing a total of 1 mg of NazEDTA. Femurs and spleens were removed and processed as described above. Hematocrits were measured after microcapillary centrifugation. White blood count values were determined with a Coulter counter (model ZN), with differentials determined following staining of peripheral smears with Wright-Giemsa stain. Platelets were determined via use of the ammonium oxalate procedure. Femoral marrow and spleen cells were evaluated for their progenitor cell content of erythroid (CFU-E and BFU-E), granulocyte (CFU-GM) and megakaryocyte (CFU-Meg) using semisolid clonal assays in vitro as performed routinely in the laboratory.i”rs
Mice C57BU6 female mice, l&12 weeks of age, were purchased from Harlan Laboratories, Indianapolis, IN. Animals were housed in plastic cages and fed Purina Lab Chow and water ad libitum. After sacrifice by cervical dislocation, femurs and spleens were removed and singlecell suspensions were prepared using McCoy’s 5A medium (GIBCO, Grand Island, NY) following repeated passage through a 25gauge needle. After centrifugation (400 g, 10 min, 4”C), the cells were resuspended in McCoy’s medium for routine cell counting using a Coulter counter (model ZN; Coulter Electronics, Hialeah, FL). Cell viability (> 95%) was confirmed via use of the trypan blue exclusion test before plating.
Treatment Interleukin
of Animals and Zidovudine, 3 and Erythropoietin
The effect of sustained long-term zidovudine treatment and the ability of continuous interleukin 3 (IL-3) and erythropoietin (EPO), either alone or in combination, to modulate zidovudine toxicity was evaluated in the following studies. Normal mice were to receive zidovudine treatment in a dose-escalation fashion, i.e. 0.1 mg/ml, 1.0 mgiml and 2.5 mg/ml placed in sterile drinking water. Zidovudine in the form of the pure powder was kindly supplied by Dr Philip Furman, Burroughs-Wellcome, Research Triangle Park, NC. These zidovudine drug concentrations corresponded to a suboptimal, optimal and pharmacological dose of drug with respect to its anti-viral potency. Zidovudine levels were confirmed using high-performance liquid chromatography analysis. Normal control animals received only sterile water. Selected groups (125 animals) of zidovudine-treated animals received either IL-3 (5 U per animal; 0.5 pgikglbw; sp. act. 1 x 107 U/ml, Genzyme, Cambridge, MA, i.p. on a
Statistical Analysis Mean -t SE for values comparing control with experimental utilizing triplicate studies were determined and analyzed using the two sample ranks test of Wilcoxon-White (P < 0.05) to determine significance.
Acknowledgements
These studies were supported in part by a Merit Review Grant awarded from the Department of Veterans Affairs. The authors are grateful to Mrs Brenda Frazier for typing this manuscript.
REFERENCES 1. Yarchoan R, Klecker RW, Weinhold KJ, Markham PD, Lyerly HK, Durack DT, Galmann E, Nursinoff-Lehrman S, Blum RM, Barry DW, Shearer GM, Fischl MA, Mitsuya H, Gallo RC, Collins JM, Bolognesi DP, Myers S (1986) Administration of 3’-azido-3’-deoxythymidine, an inhibitor of HTLV-IIIILAV replication, to patients with AIDS or AIDS-related complex. Lancet i:575-577. 2. Fischl MA, Richman DD, Grieco MH, Gottlieb MS, Volverding PA, Laskin GL, Leedom JM, Groopman JE, Mildvan D, Schooley RT, Jackson GG, Durack DT, King D (1987) The efficacy of 3’-azido-3’.deoxythymidine in the treatment of AIDS and AIDS-related complex: a double-blind, placebo-controlled trial. N Engl J Med 317:185-191. 3. Fischl MA, Richman DD, Hansen N. Collier AC, Carey JT, Para MF, Hardy D, Donin R, Powderly WA, Allen JD, Wong B, Mergian TC, McAuliffe VG, Hyslop NE, Rhame FS, Balfour HH, Spector SA, Volberding P, Petinelli C, Anderson J, The AIDS Clinical Trials Group (1990) The safety and efficacy of zidovudine (AZT) in the treatment of mildly symptomatic human
IL-3,
immunodeficiency placebo-controlled
virus type I (HIV) infection: a double-blind trial. Ann Intern Med 112:727-737.
4. Volberding PA, Lagakos SW, Koch MA, Pettinelli CA, Myers MW, Booth DK, Balfour HH, Reichman RC, Bartlett JA, Hirsch MS, Murphy RL, Hardy WD, Soeiro R, Fischl MA, Bartlett JG, Merigan TC. Hyslop NE, Richman DD, Valentine FT, Corey L (1990) Safety and -efficacy of zidovudine in asymptomatic human immundeficiencv virus infected individuals with less than 500 CD4+ cells/mm’. N Engl J Med 322:941-949. 5. Forester G (1987) Profound azidothymidine. N Engl J Med 317:772.
cytopenia
secondary
to
6. Gill PS, Raick M. Brynes RH, Causey D, Loureiro C, Levine AM (1987) Azidothymidine associated with bone marrow failure in the acquired immunodeficiency syndrome (AIDS). Ann Int Med 107:502-505. 7. Gottlieb MB, Wolfe PR, Chafey S (1987) of AIDS-related thrombocytopenia to intravenous azidothymidine (3’.azido-3’.deoxythymidine). AIDS Retroviruses 3: 109: 112.
Response and oral Res Hum
8. Groopman JE, Mitsuyasu RT, deLeo MJ, Oette DH, Golde DW (1987) Effect of recombinant human granulocyte macrophage colony stimulating factor on myelopoiesis in the acquired immunodeficiency syndrome. N Engl J Med 317:593-598. 9. Richman DD, Fischl MA, Grieco MH. Gottlieb MS, Volberding PA, Laskin OL, Leedom JM, Groopman JE, Mildran D. Hirsch MS, Jackson GG, Durach DT, Nusinoff-Lehrman S, AZT Collaborative Working Group (1987) The toxicity of azidothymidine (AZT) in the treatment of AIDS and AIDS-related complex: A double-blind. placebo-controlled trial. N Engl J Mcd 317:192-197. 10. Dournon E, Rogenbaum W. Michon C, Penonne C, DeTruchis P. Bouret E. Leevachen M. Matherson S, Gharakhanian S, Girard PM, Salmon D, Ceport C, Dazza MC, Regnier B (1988) Effects of zidnvudine in 365 consecutive patients with AIDS or AIDS-related complex. Lancet 2: 1297-1300. 11. Sommadossi JP, Zhu Z, Carisle R (1987) Mechanism of toxicity of 3’.azido-3’.deoxythymidine in normal hematopoietic progenitor cells: evidence of inhibition of the de novo pyrimidine pathway. Proc Am Assoc Cancer Res 28:329A. 12. Dainiak N, Goldman L (1988) proliferation in vitro Br J Hacmatol69:299-304.
Worthington M, Riordon 3’-azido-3’-deoxythymidine of human haematopoietic
MA, Kreckzo S, (AZT) inhibits progenitor cells.
13. Gallicchio VS, Doukas MA, Hulette BC, Hughes NK. Gass C (1989) Protection of 3’-Azido-3’-deoxythymidine induced toxicity to murine hematopoietic progenitors (CFU-GM, BFU-E, and CFUMeg) with interleukin-1. Proc Sot Exp Biol Med 192:201-204.
erythropoietin
and zidovudine
toxicity
/ 71
14. Boglioio G, Lerza R, Mercibani M, Saviane A, Pannacciulli I (1989) Azidothymidine-induced depression of murine hemopoietic progenitor cells. Exp Hematol 16:938-940. 15. Cronkite EP, Bullis J (1990) In vivo toxicity of 3’.azido3’.deoxvthvmidine (AZT) on CBAiCa mice. Int J Cell Cloning 8:332-3;15.’ ’ ’ 16. Gallicchio VS, Hughes NK, Hulette BC, Noblitt LN (1991) Effect of IL-l, IL-6, GM-CSF and erythropoietin on the in vitro toxicity associated with AZT on human bone marrow haematopoietic progenitor stem cells: CFU-GM and BFU-E. Antiviral Chem Chemother 2:76-85. 17. Gallicchio VS, Hughes NK (1992) Influence of interleukin3 on zidovudine (AZT)-induced in vitro toxicity to human hematopoietic progenitors. Int J Cell Cloning 10:99-104. 18. Gallicchio VS, Hughes NK (1992) Suppression of murine hematopoiesis in vivo after chronic administration of zidovudine: evidence that zidovudine-induced anemia is the result of decreased bone marrow-derived. erythropoietin-responsive progenitor cells. Proc Sot Exp Biol Med 199:459-465. 19. Scadden DT, Bering HA, Levine JD, Bresnahan J, Evans L, Epstein C, Groopman JE (1991) Granulocyte-macrophage colony stimulating factor mitigates the neutropenia of combined interferon-alpha and zidovudine treatment of acquired immune deficiency syndrome-associated Kaposi‘s sarcoma. J Clin Oncol 9:802-808. 20. Kaplan LD, Kahn JO. Crowe S. Northfeet D, Neville P, Grossberg H, Abrams DI, Tracey J, Mills J, Volberding PA (1991) Clinical and virologic effects of recombinant human granulocyte-macrophage colony-stimulating factor in patients receiving chemotherapy for human immunodeficiency virus associated non-Hodgkin’s lymphoma: results of a randomized trial. J Clin Oncol 9:929-940. 21. Walker RE, Parker RI, Kovacs JA, Masur H, Lane HC, Carleton S. Kirk LE. Granick HR, Fauci AS (1988) Anemia and erythropoiesis in patients with the acquired immunodeficiency syndrome (AIDS) and Kaposi sarcoma treated with zidovudine. Ann Intern Med 108:372-376. 22. Fischl MA, Galpin JE, Levine JD, Groopman JE. Henry DH. Kennedy P. Miles S, Robbins W, Stanett B, ialusky R, Abeis RI, Tsai HC, Rusnick (1990) Recombinant human erythropoietin for patients with AIDS treated with zidovudine. N Engl J Med 322: 1488-1493.