Increased cellular hypoxia and reduced proliferation of both normal and leukaemic cells during progression of acute myeloid leukaemia in rats.

The microenvironmental changes in the bone marrow, spleen and liver during progression of the transplantable promyelocytic leukaemia in the Brown Norwegian rat (BNML) have been studied. We used flow cytometry to estimate cellular hypoxia and proliferation based on in vivo pulse-labelling with a mixture of 2-nitroimidazole linked to theophylline (NITP) and bromodeoxyuridine (BrdUrd). The leukaemic cells were identified with the RM124 antibody. In rats inoculated with leukaemic cells the fraction of RM124+ cells was significantly increased from day 20 onwards in the spleen and from day 27 in the bone marrow and liver, reaching a level of 65-87% in these organs at day 32. At day 32, the NITP+ fraction of RM124+ cells had increased significantly in the bone marrow and spleen to 88% and 90%, respectively. The corresponding fractions of NITP+ normal cells reached 63% and 65%, respectively. From day 13 to day 32, the DNA-synthesizing (BrdUrd+) fraction of RM124+ cells in the bone marrow decreased significantly from 52% to 25%, and of normal cells from about 20% to 6%. In the bone marrow and spleen at day 27 and 32, the S-phase and G2/M-phase fractions according to DNA content were higher for the NITP+ than for the NITP- cells. This could partly be explained by an impaired cell cycle progression due to hypoxia. Nevertheless, we found indications of leukaemic cells that were simultaneously labelled with NITP and BrdUrd, in the bone marrow and spleen. These latter findings suggest that in contrast to normal cells some of the leukaemic cells can proliferate even during hypoxia, and this subpopulation may consequently renew and expand the leukaemic cell load.

S-phase induction by interleukin-6 followed by chemotherapy in patients with chronic lymphocytic leukemia and non-Hodgkin's lymphoma.

Interleukin-6 (IL-6) has in vitro demonstrated growth regulatory effects on tumor cells from patients with chronic lymphocytic leukemia (CLL) and lymphoma. The proliferation rate of these cells is usually very low and this is thought to be one of the reasons for the lack of a curative potential of cytostatic chemotherapy in CLL and low grade NHL. Recombinant human (rh) IL-6 might increase the in vivo proliferation rate leading to a higher sensitivity for chemotherapy. We tested this hypothesis by administering rhIL-6 to 9 CLL patients and 3 NHL patients in doses of 2.5 micrograms/kg, 5 micrograms/kg and 10 micrograms/kg s.c. daily for 5 days followed by CHOP chemotherapy on the last day of rhIL-6 injection. Six patients had two treatment cycles. The proportion of cells in S-phase was determined by the bromodeoxyuridine labeling index (LI). Three patients achieved a partial remission, one patient had progressive disease and the remaining patients demonstrated no change. Two patients, who received 10 micrograms/kg/day rhIL-6, demonstrated a significant increase in LI, one of these was first observed in the second treatment cycle. A significant decrease was seen in two patients receiving 2.5 micrograms/kg and 5 micrograms/kg respectively. Immunophenotypic assessment demonstrated that rhIL-6 increased the expression of CD20 in all CLL patients with a reversal after cessation of rhIL-6. We conclude that rhIL-6, in the dosage and schedule used in this study, did not increase the proportion of the cells in S-phase and that the growth stimulatory effects of rhIL-6 in CLL in vivo probably are insignificant. However, the role of rhIL-6 in CLL as inducer of increased CD20 expression prior to anti-CD20 antibody treatment remains to be determined.

Linear reduction of clonal cells in stem cell enriched grafts in transplanted multiple myeloma.

In 30 patients with multiple myeloma who were scheduled for peripheral blood stem-cell transplantation, a quantitative analysis of the stem cells following enrichment by anti-CD34 was carried out. To detect the cells of the specific myeloma clone, polymerase chain reaction (PCR) was performed using unique allele-specific oligo primers for the immunoglobulin heavy chain rearrangement. The clonogenic cells before and after stem-cell enrichment, were quantified by a limiting dilution assay and a highly sensitive semi-nested PCR combined with a real-time quantitative PCR. In order to accomplish a statistically adequate end-point analysis, a large number of PCR analyses (40 per sample) were performed. By this technique the lowest detection limit observed was one myeloma cell per 106 cells. Myeloma cells were detected in 29/30 samples from the CD34-enriched fraction. The CD34 selection procedure resulted in a median 28-fold enrichment of CD34+ haemopoietic precursor cells. The stem-cell selection reduced the median concentration of clonal cells per million total cells by half, with a highly significant linear relationship between the number of myeloma cells before and after stem cell enrichment. The median depletion of clonal cells by the overall procedure was 2.15 log units, corresponding to a reduction of the total quantity of clonal cells reinfused into the patients by at least 99.3%. We conclude that CD34+ cell enrichment led to a reliable tumour cell depletion of the order of 2 log, which may not be sufficient since the total number of tumour cells in the leukapheresis product was 7.2 log (median).

S-phase induction by interleukin-6 followed by chemotherapy in patients with refractory multiple myeloma.

The plasma cell labeling index (PCLI) in patients with multiple myeloma (MM) is relatively low and this has been associated with the low rate of remission following chemotherapy. Interleukin-6 (IL-6) has been demonstrated to be a major growth factor of myeloma cells. In order to increase the S-phase proportion of myeloma cells, which might increase the sensitivity to chemotherapy, we gave rhIL-6 followed by chemotherapy to 15 myeloma patients with refractory disease. A total of 25 treatment cycles were administered since ten patients had two cycles. The rhIL-6 dose was 2.5 (n = 3), 5.0 (n = 6) and 10.0 microg/kg (n = 6) by subcutaneous injection once daily for 5 days and chemotherapy was administered on the last day of rhIL-6 injection. The effect of rhIL-6 treatment on labeling index (LI) was heterogeneous, but no statistically significant change was noted for this particular group as a whole. In two patients an increase (mean 7.7%) in LI of mononuclear bone marrow cells during the rhIL-6 treatment was demonstrated and in one patient a decrease of 2.8% was seen. Assessment of PCLI demonstrated an increase of 2.9% in one out of six patients and a decrease of 1.9% in one out of six patients. None of the 15 patients achieved remission according to standard criteria. During the rhIL-6 treatment, 14 of the 15 patients developed mild constitutional adverse events (AE) well known in patients treated with IL-6, and none of the AE in the subsequent chemotherapy phase were related to IL-6. In conclusion, our study demonstrated that rhIL-6 can be administered safely to patients with refractory MM, but the cell cycle recruitment approach was not sufficiently effective to be of clinical value.

BEAM+autologous stem cell transplantation in malignant lymphoma: 100 consecutive transplants in a single centre. Efficacy, toxicity and engraftment in relation to stem-cell source and previous treatment.

One hundred consecutive patients with malignant lymphoma treated with high-dose chemotherapy and autologous stem cell transplantation, followed at least 1 yr post-transplant, are reported, 68 with non-Hodgkin's lymphoma and 32 with Hodgkin's disease. At transplant, 23 patients were in first remission, 69 in later chemosensitive disease and 8 were chemotherapy resistant. Based on previous treatment and stem-cell source, the patients were subdivided into 3 cohorts: BMT1: bone-marrow harvest and transplant after > or =3 treatment regimens (38 patients); BMT2: bone marrow harvest and transplant after less than 3 treatment regimens (24 patients); PBSCT: peripheral-blood stem cell transplant (38 patients, 5 of these with CD34+ cell selected PBSC). The 4-yr survival and progression-free survival of all patients was 45 and 40%, respectively. Forty-one patients have died, 27 of lymphoma, evenly distributed in the cohorts. Fourteen treatment-related deaths occurred, 13 of these in the BMT1 cohort, significantly more than in the other cohorts (p=0.001). In univariate survival analysis cohort, age, disease status at transplant and number of previous treatment regimens were significant. In multivariate survival analysis cohort, age and sex were independently significant, women having a shorter survival. The patients transplanted with unselected PBSC had significantly shorter duration of pancytopenia and hospital stay than the otherwise comparable BMT2 patients, but their progression-free survival was identical. We confirm that high-dose therapy with autologous stem cell transplant from blood or bone marrow in not-too-heavily pretreated patients is a safe procedure but will cure only half the patients.

Changing bone marrow micro-environment during development of acute myeloid leukaemia in rats.

The Brown Norwegian rat transplanted with promyelocytic leukaemic cells (BNML) has been used as a model for human acute myeloid leukaemia. We have previously shown that both the blood supply to the bone marrow and the metabolic rate decrease in relation to the leukaemic development in these rats. Here we have investigated how the development and progression of this leukaemia affect oxygenation, pH and proliferation of normal and leukaemic cells in vivo. Bone marrow pH was measured by a needle electrode. Nitroimidazol-theophylline (NITP) was used to identify hypoxic cells, and we applied bromodeoxyuridine (BrdUrd) to identify DNA replicating cells. The leukaemia progressed slowly until day 27 after which a rapid deterioration could be observed leading to severe changes over the following 5 d. In whole blood there was evidence of progressing metabolic acidosis. In bone marrow the fraction of leukaemic cells increased to > 90% and the pH dropped to about 6.5. The fraction of NITP+ cells increased to > 80% in bone marrow and to about 40% in blood. The fraction of BrdUrd+ cells was unchanged in blood, but decreased in bone marrow both for normal cells (from about 20% to 5%), and for leukaemic cells (from about 45% to 25%), evidently as a result of the severely changed microenvironment. In this study we have demonstrated in vivo the development of an acidic and hypoxic bone marrow hampering normal haemopoiesis during leukaemic growth. Our data support the notion of BNML as a valuable tool for studying leukaemogenesis.

Immediate bromodeoxyuridine labelling of unseparated human bone marrow cells ex vivo is superior to labelling after routine laboratory processing.

It is important to evaluate the proliferation of bone marrow cells in several disease conditions and during treatment of patients with for example cytokines. Labelling with bromodeoxyuridine (BrdUrd), immunocytochemical staining with anti-BrdUrd antibody and analysis by flow cytometry provides a reliable and reproducible technique for estimation of the fraction of cells that incorporated BrdUrd into DNA during S-phase. We have compared immediate BrdUrd labelling of unseparated bone marrow cells with the previously used labelling in the laboratory after routine separation of the mononuclear cells. Bone marrow aspirates from seven lymphoma patients without bone marrow involvement were studied with these two methods. We found higher BrdUrd labelling indices (LI) in the mononuclear cells, when cells were labelled immediately. A large variation in LI was found between patients. Our results suggest that ex vivo BrdUrd labelling of bone marrow cells should be performed immediately after aspiration and before separation, because these data are closer to values reported from in vivo labelling with BrdUrd.

Retroviral-mediated marker gene transfer in hematopoiesis-supportive marrow stromal cells.

A Moloney-derived retrovirus containing both LacZ and NeoR genes (G1BgSVNa from Genetic Therapy, Inc.), was used to transduce human and murine bone marrow stromal cells. Different kinds of stromal cells that were able to support hematopoiesis were transduced by incubation for 24 h in the presence of virus-containing supernatant. Semiconfluent layers of MRC-5 (human, myofibroblastic, fetal, pulmonary) and MS-5 (murine, myofibroblastic, medullary) cells were successfully transduced after one 24-h incubation, as demonstrated by G418 resistance and Escherichia coli beta-galactosidase staining. In contrast, human stromal cells, purified from primary confluent layers grown for 3-4 weeks, could not be transduced. However, stromal cells generated after 10-12 days in culture from Stro-1+ and 1B10+ stromal precursors were successfully transduced in the presence of basic fibroblast growth factor. Transduced stromal cells maintained a myofibroblastic phenotype, although with a decreased number of alpha-SM actin-positive microfilaments in MS-5 cells. The ability to support the generation of stroma-adherent colony-forming cells from cocultured cord blood CD34+ cells after 4 weeks in culture was similar before and after transduction and G418 selection. In conclusion, human primary stromal precursors can be efficiently transduced, and the stromal cell phenotype and function are not significantly altered after retroviral-mediated transfer of marker genes.

Cytokines active on granulomonopoiesis: release and consumption by human marrow myoid [corrected] stromal cells.

Haemopoiesis is sustained and preferentially committed to granulomonopoiesis by myoid [corrected] stromal cells generated by colony-derived cell lines (CDCL). Using ELISA and RIA, we studied, in the supernatant of cells from CDCL, the time course of interleukins 3 and 6 (IL-3, IL-6), stem cell factor (SCF), granulocyte-macrophage, granulocyte and macrophage colony stimulating factors (GM-CSF, G-CSF and M-CSF), macrophage-inflammatory protein-1alpha (MIP-1alpha) and transforming growth factor beta1 (TGF beta1). IL-6, GM-CSF, M-CSF and MIP-1alpha were released into the supernatant after medium renewal and, except for M-CSF, addition of IL-1beta. G-CSF was detected only after addition of IL-1beta. SCF, contained in medium, first declined and then increased 24 h after medium renewal. Release of TGF beta1 started 24 h after medium renewal and lasted until day 7. IL-3, provided by horse serum, declined throughout the 7d of observation. In conclusion, stromal cells from CDCL synthesized and released into the supernatant. IL-6, GM-CSF, G-CSF, M-CSF and MIP-1alpha after stimulation by seric factor(s) and/or IL-1beta. TGF beta1 was synthesized and released without any obvious extraneous stimuli. There is no definite argument for synthesis of soluble SCF and IL-3. These data support a model where growth factors increase shortly after medium renewal, and negative regulators take over at a later time.

[Autologous stem cell transplantation. From bone marrow to selected blood stem cells: 100 consecutive procedures at a single center]. / Autolog stamcelletransplantation. Fra knoglemarv til oprensede blodstamceller: 100 konsekutive procedurer i et enkelt center.

One hundred consecutive autologous stem cell transplants are reported: Non-Hodgkin's lymphoma 51 cases, Hodgkin's disease 27 cases, acute leukaemia 14 cases, multiple myeloma seven cases and chronic myeloid leukaemia one case. Most patients were in their second or later remission. The overall three-year survival for all patients was 60% and the three-year disease-free survival was 50% for lymphoma patients and 30% for acute leukaemia patients. The dominant source of stem cells was bone marrow during 1993, but from 1994 it has been peripheral blood, now totalling 33 cases. There were 12 toxic deaths, all among patients who were heavily treated before bone marrow harvest and transplantation. The patients transplanted with blood stem cells had significantly shorter duration of pancytopenia, and hospital stay, but their disease-free survival was not longer than that of a comparable group of bone marrow transplanted patients. Six patients were transplanted with purified CD34+ cells (selected by avidity column (Ceprate (R)), and had duration of thrombocytopenia and hospital stay similar to the patients transplanted with unmanipulated blood stem cells, but slightly longer duration of neutropenia. We conclude that high-dose therapy with autologous stem cell transplantation in not too heavily pretreated patients is a safe procedure irrespective of the source of stem cells.

Constitutive and modulated cytokine expression in two permanent human bone marrow stromal cell lines.

We present a detailed analysis of cytokine expression patterns of the two permanent human bone marrow stromal cell lines, L87/4 and L88/5. These cell lines, previously established in our laboratory, are highly radiotolerant without cell detachment and support long-term cultures of CD(34+)-enriched human cord blood cells. RT-PCR analysis of 22 different cytokines or cytokine receptor mRNAs showed an almost identical expression pattern in the two stromal cell lines compared to primary human Dexter-type stroma. Since stromal feeder lines employed in long-term cultures usually are irradiated and grown in media containing corticosteroids, we analyzed the impact of irradiation and dexamethasone on cytokine production in the two cell lines by RT-PCR, Northern blot analysis, bioassays, and RIAs. By RT-PCR analysis, constitutive mRNA expression of c-kit, G-CSF, GM-CSF, IL-1 beta, IL-6, IL-7, IL-8, IL-11, Kit ligand (KL), LIF, M-CSF, MIP-1 alpha, TGF-beta, and TNF-alpha was demonstrated in both cell lines, with L87/4 a more potent cytokine producer than L88/5. Northern blot data showed an increase in mRNA levels for GM-CSF, IL-1 beta, and LIF by irradiation and IL-1 alpha treatment in both cell lines. IL-1 alpha-induced GM-CSF, IL-1 beta, IL-6, IL-11, and LIF mRNA levels were reduced by the addition of dexamethasone, whereas dexamethasone had no influence on the amounts of IL-1 alpha-induced G-CSF mRNA. L87/4 and, to a lower extent, L88/5 cells showed dexamethasone-dependent increases in KL mRNA, while KL mRNA levels were not stimulated by IL-1 alpha.

Improved vaccination response during ranitidine treatment, and increased plasma histamine concentrations, in patients with B cell chronic lymphocytic leukemia.

Patients with B cell chronic lymphocytic leukemia (B-CLL) have decreased capacity to mount relevant antibody responses upon immunization, and development of hypogammaglobulinemia is part of the natural history of the disease. We investigated the influence of histamine type-2 (H2) receptor blockade by ranitidine on the in vivo antibody production in B-CLL patients following vaccination. Anti-polysaccharide antibodies in B-CLL patients, vaccinated with a tetanus-toxoid conjugated vaccine against Haemophilus influenzae type-B (Hib), reached long-term protective levels in more than 90% of B-CLL patients randomized to ranitidine treatment, as compared to 43% of the untreated patients (P = 0.024). No difference in the response to vaccination against influenza virus types A and B protein could be detected between the two groups. Plasma histamine levels were 2-fold to 20-fold higher in 23 out of 31 B-CLL patients, compared to normal controls, and these levels showed a significant positive correlation to disease duration. These findings indicate the possibility of improving in vivo antibody production against a highly relevant pathogen in B-CLL patients by histamine type-2 receptor blockade, and the combined finding of an immune-stimulatory effect of ranitidine and increased plasma histamine levels, strongly suggests the involvement of histamine in the pathogenesis of B-CLL immunodeficiency.

Pharmacokinetic and pharmacodynamic studies of subcutaneously administered recombinant human interleukin-3 following chemotherapy for non-Hodgkin's lymphoma.

The pharmacokinetics and the pharmacodynamic profile of subcutaneously administered recombinant human non-glycosylated interleukin-3 (rhIL-3) was studied in lymphoma patients after standard CHOP chemotherapy. 30 patients received 0.5, 1.0, 5.0, 7.5 and 10 micrograms/kg (six patients at each dose level) of rhIL-3 for 14 d. Serum rhIL-3 samples were obtained regularly, during the treatment and serially over a 24 h period on the first (cycle day 2) and the last (cycle day 15) day of rhIL-3 treatment for pharmacokinetic evaluation. Following s.c. injection on cycle day 2. the maximum rhIL-3 serum concentration ranged from 289 pg/ml (0.5 micrograms/kg) to 4690 pg/ml (10 micrograms/kg). Both the maximum serum concentration (R = 0.90. P < 0.0001) and the area under the serum concentration-time curve (R = 0.95, P < 0.0001) were related to dose. The elimination half-life T1/2 beta was 160 min for 0.5 micrograms/kg and 134 min for 10 micrograms/kg, with no apparent dose relationship. The systemic clearance of 3.0-6.0 ml/min/kg was comparable at all dose levels. No significant difference was noted between pharmacokinetic parameters on the first day of rhIL-3 and the last day of treatment, and no accumulation of the drug was noted throughout the study. The pharmacokinetic parameters correlated poorly to the clinical response of the growth factor. where dose in micrograms/kg seemed to be the most important single factor.

Recombinant human interleukin-3: pharmacokinetics after intravenous and subcutaneous bolus injection and effects on granulocyte kinetics.

The pharmacokinetics of E. coli derived recombinant human interleukin-3 (rhIL-3) was studied following intravenous (i.v.) and subcutaneous (s.c.) bolus injection of rhIL-3. After i.v. bolus injection in eight patients, serum peak levels of 34.5-135.0 ng/ml were reached, followed by a rapid decline with a t1/2 alpha of 17 +/- 2 min and a t1/2 beta of 59 +/- 7 min. After s.c. bolus injection in five patients, the absorption was more prolonged with peak serum levels reached at 2.8 +/- 0.4 h. Elimination was also more protracted, and serum base-line levels were reached at 14-24 h. The immediate effect of rhIL-3 on peripheral white blood cells was less pronounced and more variable than previously found for G- or GM-CSF. Following i.v. administration, neutrophils showed a moderate drop to median 64% of initial values (range 42-85%) at median 30 min after injection (range 15-60 min) followed by an increase at 24 h to 69-288% of initial values. Eosinophils dropped to a median nadir of 34% and then gradually increased to maximum values in the range 135-720% at 18-24 h. The effect of rhIL-3 was further examined following i.v. injection of autologous 111Indium-labelled granulocytes in six patients. In steady state, i.v. injection of rhIL-3 caused a moderate drop in 111Indium activity of peripheral blood within 20 min without tendency to subsequent recovery. No change occurred in the activity recorded over the lungs and liver. The activity over the spleen decreased moderately in two patients. These results are strikingly different from those previously obtained after i.v. injection of rhGM-CSF.

Development and application of a sensitive radioimmunoassay for human granulocyte-macrophage colony-stimulating factor able to measure normal concentrations in blood.

A radioimmunoassay (RIA) for human granulocyte-macrophage colony-stimulating factor (GM-CSf) was developed based on antibodies from rabbits immunized with glycosylated recombinant human (rh) GM-CSF. The antibodies are specific for human GM-CSF and do not crossreact with other human hematopoietic growth factors or mouse GM-CSF. The antibodies also react with nonglycosylated rhGM-CSF, so E. coli-derived rhGM-CSF can be assayed as well. The RIA has a measuring range of about 10 to 200 pg/mL. Normal blood was found to contain 13 to 24 pg/mL (95% limits) with a mean of 18.5 pg/mL (n = 34). Monoclonal antibodies against GM-CSF could remove GM-CSF from normal human serum, thus ensuring that the GM-CSF measured in serum is real and does not represent nonspecific reactivity with our polyclonal rabbit antibodies. While previously published methods have been unable to measure GM-CSF in human serum under normal conditions, our more sensitive RIA does confirm the presence of small amounts of GM-CSF in serum or plasma and can therefore be used to detect fluctuations of GM-CSF in health and in disease.

Anomalous plasma concentrations and impaired secretion of growth factors in Fanconi's anemia.

In Fanconi's anemia, which is known to be an autosomal recessive Mendelian trait with four complementary groups. In addition to stunning phenotypic variation at clinical and cellular levels, aplastic pancytopenia is a common feature. Since either an early block of differentiation in stem cells or their insufficient support by stromal functions could be an underlying factor, levels of stem cell factor (SCF) and cytokines have been measured in blood and in supernatants of monocytes after stimulation with granulocyte-macrophage colony stimulating factor (GM-CSF). In two of three FA patients, no GM-CSF was detectable, and simultaneously SCF was decreased to 8% and 15% of normal values. The combination of low SCF and GM-CSF may be implied in the pathogenesis of marrow aplasia, since comparison with W/Sl mice shows that impairment of the SCF/c-kit function alone has different effects. Also, this explains that treatment with GM-CSF in a recent study enhanced only leukogenesis and not all three lineages. In the third patient, both factors were normal, and here a different mechanism may act. In all three FA patients, interleukin 6 (IL-6) production in stimulated monocytes was decreased, which may hamper immune defense of infections in a nonspecific way.

Comparative pharmacokinetics of single-dose administration of mammalian and bacterially-derived recombinant human granulocyte-macrophage colony-stimulating factor.

Pharmacokinetics of recombinant human non-glycosylated bacterially-synthesized (E. coli) granulocyte-macrophage colony-stimulating factor (GM-CSF) were studied following single intravenous (i.v.) and subcutaneous (s.c.) bolus injection, and compared to equivalent doses of glycosylated mammalian-derived CHO-GM-CSF. Each route of administration gave a different GM-CSF concentration-time profile. The highest peak serum concentrations (Cmax) were observed following i.v. bolus injection. After i.v. administration, a two-phase decline in concentration was noted for both types of GM-CSF with a significantly shorter t1/2 alpha of 7.8 minutes for the E. coli GM-CSF versus 20.0 min for the CHO-GM-CSF, while no significant difference was observed for the terminal phase. Following s.c. administration of equivalent doses, a higher peak serum concentration was observed in the E. coli-treated patients and, again, a faster elimination where pretreatment serum levels were reached after 16-20 h, versus more than 48 h after administration of CHO-GM-CSF. Although the non-glycosylated E. coli GM-CSF thus seems to undergo a faster elimination that the glycosylated CHO-GM-CSF no significant difference could be demonstrated in the in vivo effect of corresponding doses of the two compounds with respect to stimulation of granulopoiesis--with reservation for small patient numbers and a large individual variations in response.

In vivo kinetics of 111indium-labelled autologous granulocytes following i.v. administration of granulocyte-macrophage colony-stimulating factor (GM-CSF).

Administration of both glycosylated and non-glycosylated recombinant human granulocyte-macrophage colony-stimulating factor (rhGM-CSF) induces an immediate transient granulocytopenia of 1-3 hours' duration. In order to explore this phenomenon, granulocytes were labelled with 111Indium and the effect on the kinetics of granulocytes after administration of rhGM-CSF was studied in 10 previously untreated patients with malignant lymphoma. For both types and doses of rhGM-CSF, a significant and dramatic accumulation of the 111Indium-labelled granulocytes was observed in the lung within a few minutes after i.v. injection of rhGM-CSF. The accumulation of radioactivity coincided with the pronounced and transient granulocytopenia in peripheral blood. The 111Indium-labelled granulocytes later reappeared in the peripheral blood, indicating reversible pulmonary vascular margination of the granulocytes. Half-life of labelled granulocytes after reappearance was comparable to half-life values under normal conditions. The transient accumulation of granulocytes in the pulmonary vessels seems not to be of clinical importance in the management of patients, but it may to some degree explain previously described side-effects, such as transient hypoxemia ("first-dose" reaction) following administration of rhGM-CSF.

Clinical pharmacokinetic studies of a human haemopoietic growth factor, GM-CSF.

The pharmacokinetics of glycosylated recombinant human granulocyte-macrophage colony-stimulating factor (rhGM-CSF) was studied following intravenous (i.v.) and subcutaneous (s.c.) bolus injection of rhGM-CSF, 8 micrograms kg-1 employing a sensitive radioimmunoassay. After a single i.v. bolus injection, an initial high serum level of rhGM-CSF was observed, followed by a rapid decrease that occurred in two phases with a half-life (t1/2) alpha of 20.0 +/- 5 min and a t1/2 beta of 68.3 +/- 8 min. Following s.c. bolus injection the absorption was more prolonged. Peak serum concentrations did not occur until about 15-20 h, and were followed by a more protracted elimination than by the i.v. route. In all patients the single rhGM-CSF injection led to an increase in peripheral white blood cells (WBC), after a temporary drop of 2-5 h duration. The increase in WBC was of longer duration after s.c. than after i.v. bolus treatment. Since the subcutaneous administration leads to prolonged serum concentration of rhGM-CSF and prolonged increase in peripheral WBC, it seems preferable to i.v. bolus injection, and as effective as continuous i.v. infusion.