SEARCH

SEARCH BY CITATION

Keywords:

  • prophylactic;
  • granulocyte transfusions;
  • allogeneic stem cell transplantation;
  • severe infection

Abstract

  1. Top of page
  2. Abstract
  3. Patients and methods
  4. Results
  5. Granulocyte donation efficacy
  6. Granulocyte donation safety
  7. Infections
  8. Nutrition and other outcomes
  9. Discussion
  10. Acknowledgments
  11. References

Summary. The predictable neutropenia that follows allogeneic stem cell transplantation (ASCT) may be associated with recurrence of previous life-threatening infection. We describe nine patients with either previous invasive aspergillosis (IA) or considered to be at high risk of developing IA who underwent ASCT with prophylactic granulocyte transfusions. The study group, when compared with a control group, had a significant reduction in the incidence and duration of fevers (P < 0·05) and maximum C-reactive protein (P < 0·05). There were significantly fewer days of neutropenia (P < 0·05). There was also radiological improvement of pulmonary infiltrates in four out of seven assessable patients. No serious toxicity was encountered in donors or recipients. We conclude that prophylactic granulocyte donations can be given safely, and that they significantly reduce the number of days of neutropenia. Further investigation is warranted to determine whether granulocyte donations can prevent the recurrence of IA in patients at risk of fungal infection.

Neutropenia is an inevitable consequence of the cytotoxic chemotherapy administered to patients before conventional allogeneic bone marrow transplantation (Bodey et al, 1966). Consequently, patients are at risk of serious bacterial and fungal infections that may be fatal, despite broad-spectrum antibacterial and antifungal agents (Williamson et al, 1998). A previous history of fungal infection and other pretransplant risk factors, such as prolonged neutropenia, places patients at an even higher risk of severe fungal infections and may preclude the patient from a bone marrow transplant procedure (Fukuda et al, 2002). In an attempt to counter the deleterious effects of neutropenia, granulocyte transfusions have been used in the treatment of established serious neutropenic sepsis, but with limited impact on outcome. There is, however, little experience of the prophylactic use of granulocyte transfusions to attempt to reduce the risk of invasive aspergillosis in patients considered to be at high risk from this complication. We therefore developed a protocol of prophylactic granulocyte transfusions with the aim of ablating the period of neutropenia in patients at high risk for invasive fungal infections and report here our preliminary experience in nine patients.

Patients.  Between October 2001 and December 2002, nine patients considered to be at high risk of invasive aspergillosis (IA) underwent allogeneic bone marrow transplant at the Bone Marrow Transplant Unit, Royal Hospital for Sick Children, United Bristol Healthcare Trust. Patients were considered to be at high risk of IA because of proven IA (n = 1), probable/possible IA (n = 6), according to the European Organization for Research and Treatment of Cancer Mycoses Study Group criteria (Ascioglu & Rex, 2002), or because of prolonged neutropenia pretransplant (n = 2). Because of the high risk of IA, these patients received prophylactic granulocyte transfusions as detailed below. Eighteen contemporaneous controls matched for conditioning regimen were identified. As conditioning protocols were uniform in patients with particular diseases and of certain ages, these factors were also controlled for. The conditioning regimen is the single most important factor that determines duration and severity of neutropenia and, hence, the likelihood of neutropenic sepsis. Patient disease and transplant details are given in Table I.

Table I.  Basic transplant characteristics.
  Cases (n = 9)Controls (n = 18)
  • AML, acute myeloid leukaemia; ALL, acute lymphoblastic leukaemia; JMML, juvenile myelomonocytic leukaemia.

  • *

    High CMV risk, donor or recipient were CMV seropositive; low CMV risk, both donor and recipient were CMV seronegative.

Median age (years)3223
Diagnosis
 AML64
 Kostmann's syndrome/AML10
 Granulocytic sarcoma10
 ALL15
 Lymphoma04
 JMML02
 Aplastic anaemia01
 Plasma cell dyscrasia01
 Congenital immune disorder01
Transplant type
 Sibling donor17
 Unrelated matched donor510
 Unrelated mismatched donor20
 Haploidentical11
CMV risk*
 High48
 Low510
Reduced-intensity conditioning48

Conditioning.  Patients were admitted to non-isolation beds on d −9 to commence conditioning therapy. They received myeloablative conditioning with Campath 1H (0·2 mg/kg) on d −9 to −5, cyclophosphamide (60 mg/kg) on d −6 and −5 and total body irradiation (TBI, 14·4 Gy) given in eight fractions on d −3 to d 0 or non-myeloablative conditioning with Campath 1H (0·2 mg/kg) on d −9 to −5, fludarabine (30 mg/m2) intravenously on d −7 to −3 and melphalan 140 mg/m2 intravenously on d −2. Fludarabine was added to the myeloablative conditioning for mismatched or haploidentical transplants.

Supportive care, infection prophylaxis and treatment of infection.  All patients were transplanted in positive-pressure single rooms with air filtration by EU8-grade filters and remained in protective isolation from d 0 until engraftment. Patients were free to leave the unit during conditioning. All patients received ‘clean’ diets, which excluded fresh fruit and vegetables and tap water from d −3 until 30 d after discharge from isolation. Patients who were readmitted to the unit for management of complications were housed in single rooms without air filtration. Intravenous access was achieved by double-lumen tunnelled central venous catheter. In the absence of line infections and provided lumens remained patent, lines were left in situ for 3–6 months, depending on venous access. Patients received oral ciprofloxacin (250 mg twice daily) from d 13 until intravenous antibiotics were commenced. Pneumocystis carinii prophylaxis was used routinely for 6–12 months, but continued longer in patients with chronic graft-versus-host disease (GVHD). Phenoxy-methyl penicillin (250 mg twice daily) was commenced at 6 months and continued indefinitely. Prophylactic and therapeutic antibacterial antibiotics were the same in the study and control groups.

Antifungal therapy.  The seven patients with possible/proven IA received a median of 27 d of treatment with liposomal amphotericin B before transplant. In addition, two patients received voriconazole, and one received caspofungin. During transplant, all seven patients received intravenous ambisome 1 mg/kg. In addition, two received voriconazole, and one received caspofungin during the transplant period. Control group patients received prophylactic itraconazole suspension or ambisome1 mg/kg if they were intolerant of itraconazole.

Granulocyte donations.  Potential donors were identified from recipients' friends and families. All potential donors filled in a standard risk assessment questionnaire and were tested for cytomegalovirus (CMV) complement fixation and latex agglutination status and blood group. All donors underwent full clinical assessment including medical examination, electrocardiogram, urinalysis, full blood count, coagulation and immunoglobulin assays. Blood was also tested for hepatitis B and C, syphilis and human immunodeficiency viruses 1 and 2. All donors were counselled by independent National Blood Service staff, and gave informed consent. Granulocyte mobilization was achieved with granulocyte colony-stimulating factor (G-CSF) 5 µg/kg subcutaneously and 8 mg of dexamethasone orally, both given 12 h before each granulocyte collection. Peripheral blood leukapheresis was performed on a Cobe Spectra with an interface of 7%, with 6% pentastarch as a sedimenting agent, aiming for a granulocyte collection of over 10 × 1011. Seven litres of blood were processed. Donors could donate a maximum of once per week.

Outcome measures.  Data were collected retrospectively using case notes and from pathology computer records. The time period for data collection was from the start of transplant conditioning until the day of engraftment, defined as an unsupported neutrophil count of above 1 × 109/l on three consecutive days. Fevers were defined as a temperature of > 38°C on two separate occasions or a single reading above 38·5°C. C-reactive protein (CRP) and neutrophil count were measured daily in all patients, and the highest single reading was recorded for cases and controls. The number of days of pyrexia was calculated from observation charts stored in the patients' notes. Information on the need for total parenteral nutrition (TPN) and parenteral opiates was extracted from daily progress charts. This information was used as a surrogate marker for the severity of mucositis. Radiological assessment was performed as clinically indicated. Computerized tomography (CT) scans were reviewed by two independent radiologists, who were blinded to the treatment received. Scans were examined for findings suggestive of IA (cavitating lesions, halo sign or air crescent sign). For patients with radiological evidence of IA, follow-up scans were performed to assess response.

Statistical analysis.  Data were analysed with winstat using Mann–Whitney U two-tailed test for continuous data and Fisher's exact test for analysis of two-way tables. A P-value < 0·05 was considered to be statistically significant.

Granulocyte donation efficacy

  1. Top of page
  2. Abstract
  3. Patients and methods
  4. Results
  5. Granulocyte donation efficacy
  6. Granulocyte donation safety
  7. Infections
  8. Nutrition and other outcomes
  9. Discussion
  10. Acknowledgments
  11. References

Nine patients received 40 granulocyte donations. The median granulocyte dose was 64·1 × 109 per donation (range 7·2 × 109−128·8 × 109). The median neutrophil increment at 12 h after donation was 1·7 × 109/l (range 0·1 × 109−7·4 × 109/l). One patient had known human leucocyte antigen antibodies (patient 6) and had minimal increments to granulocytes – this patient was not included in the statistical comparison with the control group, but was included for safety and tolerability data. The median number of days of neutrophils below 0·3 × 109/l and 0·5 × 109/l were 3·5 d and 5·5 d, respectively, for recipients of granulocyte transfusions compared with 9·5 d and 12·5 d for controls (P < 0·05) (see Table II).

Table II.  Basic granulocyte transfusion data.
Number of granulocyte transfusions40
Median dose of granulocytes64·1 × 109 per donation
Mean neutrophil increment at 12 h1·7 × 109/l
Median number of days of neutrophils < 0·5/l5·5

Granulocyte donation safety

  1. Top of page
  2. Abstract
  3. Patients and methods
  4. Results
  5. Granulocyte donation efficacy
  6. Granulocyte donation safety
  7. Infections
  8. Nutrition and other outcomes
  9. Discussion
  10. Acknowledgments
  11. References

Fourteen out of a total of 57 donors (24%) were unsuitable for various reasons including CMV risk, poor venous access and recent tattoos. No adverse effects of granulocyte donation occurred in any donor. One recipient had an episode of fever and bronchospasm after a granulocyte donation that settled with simple measures and did not recur. The day of engraftment was similar in patients and controls (Table III).

Table III.  Outcome measures.
 CasesControlsSignificance
  1. NS, not significant.

Number818 
Number of patients developing fever218P  < 0·001
Median days of fever2·57·1P  < 0·001
Median CRPmax4085P  = 0·006
D 20 mortality00NS
D 100 mortality11NS
Number of patients requiring TPN210NS
Number of patients requiring parenteral opiates38NS
Median number of platelet transfusions64NS
Median days neutrophils < 0·3 × 109/l3·59·5P  = 0·01
Median days neutrophils < 0·5 × 109/l5·512·5P  = 0·008
Days of antibiotics before discharge1413NS
Day of engraftment1616NS
Day of discharge3042NS
Acute GVHD grade I512NS
CMV reactivation23NS

Infections

  1. Top of page
  2. Abstract
  3. Patients and methods
  4. Results
  5. Granulocyte donation efficacy
  6. Granulocyte donation safety
  7. Infections
  8. Nutrition and other outcomes
  9. Discussion
  10. Acknowledgments
  11. References

Compared with controls, the recipients of granulocyte transfusions were significantly less likely to develop fever (P < 0·05), had fewer days of fever (P < 0·05) and had a lower maximum CRP (P < 0·05) (Table III). Of the seven patients with evidence of previous IA, four had persistent radiological abnormalities immediately before transplant (Table IV). All these four patients had objective radiological improvement within 60 d of transplant. Two of the seven patients had radiological evidence of progressive IA, one of whom subsequently improved and is still alive; the other died of leukaemic myocardial infiltration. One patient had no clinical indication for further imaging so could not be assessed. Neither of the two patients with prolonged neutropenia pretransplant required radiological investigation for IA during the first 100 d of transplant. There were no other proven bacterial, fungal or viral infections before engraftment in recipients of granulocyte transfusions.

Table IV.  Details of fungal infections.
PatientClinical features of invasive mycosisRadiographic features of invasive mycosisRadiography before transplantRadiography after transplant
  1. SVC, superior vena cava; BAL, broncho-alveolar lavage.

  2. NA, not applicable.

1None – heavily pretreated for refractory AML (six courses of consecutive chemotherapy) NANA
Fever chest pain, pleuritic chest pain, pleural rubCavity within an area of consolidation with positive halo signNear resolutionWorse at d +14
3Fever, pleuritic chest painRight basal consolidation with positive halo signNAImproved at d +28
4Fevers, chest pain, pleural rubRapidly enlarging mass encasing SVCAbnormalImproved at d +28
5Prolonged fevers after prolonged aplasiaRapidly progressive peribronchial inflammatory lesionsNormalNA
6Kostmanns/AMLNANANA
7Fevers with positive Aspergillus PCRConsolidation with positive halo signNAImproved at d +56
8FeversThree cavitating lesions on CTAbnormalSome improvement at d 12 and d +57
9Fevers; chest sounds hyphae on BALWidespread nodular lesions with positive halo signAbnormalWorse at d +58

Nutrition and other outcomes

  1. Top of page
  2. Abstract
  3. Patients and methods
  4. Results
  5. Granulocyte donation efficacy
  6. Granulocyte donation safety
  7. Infections
  8. Nutrition and other outcomes
  9. Discussion
  10. Acknowledgments
  11. References

One patient in each group had died by d 100. Two recipients of granulocyte donations reactivated CMV. Five recipients suffered grade 1 GVHD involving skin. There were no cases of grade II–IV acute GVHD. There was less use of TPN and opiates for mucositis in the treatment group, but this did not reach statistical significance. There was no difference in the number of platelet transfusions.

Discussion

  1. Top of page
  2. Abstract
  3. Patients and methods
  4. Results
  5. Granulocyte donation efficacy
  6. Granulocyte donation safety
  7. Infections
  8. Nutrition and other outcomes
  9. Discussion
  10. Acknowledgments
  11. References

In this study of nine allogeneic stem cell transplant recipients at high risk of IA, we have shown that a programme of prophylactic granulocyte transfusions was feasible and resulted in a significant reduction in the period of post-transplant neutropenia. Associated with this reduction in neutropenia was a reduction in the incidence of fever, the number of days of fever, the maximum CRP and the use of TPN and opiates for mucositis. The granulocyte transfusions were well tolerated, and there were no adverse events reported in donors. The apparently highly significant differences in fever and maximum CRP between the two groups require further explanation. Although antibacterial therapy was uniform in the two groups, antifungal prophylaxis differed. However, bacterial sepsis is the most common cause of neutropenic fever after transplant, and there were no proven fungal infections in the control group. An elevated CRP may also result from tissue damage caused by the conditioning regimen; we observed less mucositis in the study group (as shown by less use of TPN and parenteral opiates), although the difference was not statistically significant.

The correlation between leucopenia and infection was first noted over 35 years ago (Bodey et al, 1966), and the period of neutropenia that follows myelosuppressive chemotherapy is associated with an increased risk of both bacterial and fungal infection. Consequently, granulocyte infusions have been used in the treatment of severe neutropenic sepsis with evidence of benefit in some (Freireich et al, 1964) but not all studies (Freireich, 2000). A substantial problem in these early studies was the difficulty in collecting sufficient numbers of granulocytes, with doses of 109 proving ineffective (Freireich, 2000). Renewed interest in the use of granulocyte transfusions has occurred with the introduction of myeloid growth factors into routine clinical practice. The administration of both G-CSF and corticosteroids to healthy donors produces a marked increase in peripheral blood neutrophils and enhances their harvesting for clinical purposes (Heuft et al, 2002). Several reports have suggested that G-CSF/dexamethasone-mobilized granulocytes may be efficacious in severe infection (Grigull et al, 2002), although a phase I/II study showed no efficacy in invasive aspergillosis (Price et al, 2000). Patients undergoing allogeneic bone marrow transplantation are at risk of neutropenic sepsis, and certain subgroups of patients may be at additional risk for IA (Fukuda et al, 2002; Marr et al, 2002), which, once established, caries a high mortality (Williamson et al, 1998). We therefore developed a programme of prophylactic granulocyte transfusions for allogeneic transplant recipients who were deemed to be at a higher risk for IA.

We did not observe any significant acute toxicity in the donors, and the administration of a single dose of both dexamethasone and G-CSF is unlikely to be associated with significant long-term toxicity. There was, however, a substantial workload associated with the screening, counselling and harvesting of donors that had additional cost and staffing implications. It was also difficult to find sufficient donors to ensure the provision of CMV-seronegative donations to CMV-seronegative recipients, although this was achieved in our study population. In all nine patients, adequate numbers of granulocyte donations could be harvested to facilitate thrice-weekly granulocyte transfusions, which were well tolerated and produced a significant reduction in the number of days of neutropenia. However, neutropenia was not completely abolished in these patients, because of occasional low donation sizes, poor increments or delays in granulocyte collection resulting from difficulties in identifying adequate donor numbers. This therapy was also associated with a reduction in a number of markers of infection (incidence of fever, days of fever, CRP) and the indirect markers of mucositis (TPN and opiate usage). Approximately 30% of patients with previous IA will develop recurrent IA after transplant, and this carries a very high mortality (Williamson et al, 1998; Fukuda et al, 2002). Despite the high risk of IA in these patients, there was radiological evidence of regression in four patients, and progression in two, one of whom subsequently improved and remains well 9 months after the transplant; the other died at d +36 of leukaemic infiltration of the myocardium. There were no deaths attributable to IA. This finding should be interpreted with caution, as this was a small cohort of patients who also received pre-emptive treatment with antifungal drugs before and after transplantation, making it difficult to isolate the effect of prophylactic granulocyte transfusions on the incidence of progressive IA. Larger trials are required to confirm the benefit of prophylactic granulocyte transfusions in reducing the complications of neutropenia and specifically in reducing the risk of IA in patients at risk of this complication. Given the workload associated with this therapy, its cost effectiveness will require scrutiny in any such studies.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Patients and methods
  4. Results
  5. Granulocyte donation efficacy
  6. Granulocyte donation safety
  7. Infections
  8. Nutrition and other outcomes
  9. Discussion
  10. Acknowledgments
  11. References

Heather Hawkins helped with data collection. Joyce Wyatt helped to collate all the radiological findings. We thank the nursing and medical staff of the BMT unit for their excellent care of these patients.

References

  1. Top of page
  2. Abstract
  3. Patients and methods
  4. Results
  5. Granulocyte donation efficacy
  6. Granulocyte donation safety
  7. Infections
  8. Nutrition and other outcomes
  9. Discussion
  10. Acknowledgments
  11. References
  • Ascioglu, S. & Rex, J.H. (2002) Defining opportunistic invasive fungal infections in immunocompromised patients with cancer and haematopoietic stem cell transplants: an international consensus. Clinical Infectious Diseases, 34, 714.
  • Bodey, G.P., Buckley, M., Sathe, Y.S. & Freireich, E.J. (1966) Quantitative relationships between circulating leukocytes and infection in patients with acute leukemia. Annals of Internal Medicine, 64, 340.
  • Freireich, E.J. (2000) Supportive care for patients with blood disorders. British Journal of Haematology, 111, 6877.
  • Freireich, E.J., Levin, R.H., Whang, J., Carbone, P.P., Bronson, W. & Morse, E.E. (1964) The function and fate of transfused leukocytes from donors with chronic myelocytic leukemia in leukopenic recipients. Annals of the New York Academy of Science, 113, 10811089.
  • Fukuda, T., Owens, S., Boeckh, M., Guthrie, K., Mattson, D., Davis, C., Wauld, A., Corey, L., Storb, R. & Marr, K. (2002) Allogeneic hematopoietic stem cell transplantation in patients with a history of prior invasive aspergillosis: outcomes and risks for post-transplant invasive aspergillosis. ASH abstract.
  • Grigull, L., Schrauder, A., Schmitt-Thomssen, A., Sykora, K. & Welte, K. (2002) Efficacy and safety of G-CSF mobilized granulocyte transfusions in four neutropenic children with sepsis and invasive fungal infection. Infection, 30, 267271.
  • Heuft, H.G., Goudeva, L., Sel, S. & Blasczyk, R. (2002) Equivalent mobilization and collection of granulocytes for transfusion after administration of glycosylated G-CSF (3 µg/kg) plus dexamethasone versus glycosylated G-CSF (12 µg/kg) alone. Transfusion, 42, 928934.
  • Marr, K., Carter, R., Boeckh, M., Martin, P. & Corey, L. (2002) Invasive aspergillosis in allogeneic stem cell transplant recipients: changes in epidemiology and risk factors. Blood, 100, 43584366.
  • Price, T.H., Bowden, R.A., Boeckh, M., Bux, J., Nelson, K., Liles, W.C. & Dale, D.C. (2000) Phase I/II trial of neutrophil transfusions from donors stimulated with G-CSF and dexamethasone for treatment of patients with infections in hematopoietic stem cell transplantation. Blood, 95, 33023304.
  • Williamson, E., Millar, M., Steward, C., Cornish, J., Foot, A., Oakhill, A., Pamphilon, D., Reeves, B., Caul, E., Warnock, D. & Marks, D. (1998) Infections in adults undergoing unrelated donor bone marrow transplantation. British Journal of Haematology, 104, 560568.