Life-threatening infections with bacteria, yeast and fungus continue to be a consequence of severe neutropenia (fewer than 500/μl blood neutrophils) and disorders of neutrophil (PMN) dysfunction. The most frequent situation is neutropenic fever and infection with yeast or fungus during either hematopoietic progenitor cell transplantation or intense chemotherapy for hematologic malignant disease. Neutropenic infections cause considerable morbidity, occasionally are fatal and add considerable cost to the treatment of these patients.
Previous attempts either to prevent or treat infections in severely neutropenic patients by transfusing PMN concentrates − commonly called granulocyte transfusions (GTXs) − achieved only questionable success, despite many reports documenting benefit [1,2][1,2]. This lack of enthusiasm for GTX can be explained, primarily, because historical GTX collected for transfusion from donors stimulated only with corticosteroids contained woefully inadequate numbers of PMNs. Currently, much larger number of PMNs can be collected from healthy donors by means of granulocyte colony-stimulating factor and corticosteroid (dexamethasone) marrow stimulation followed by large-volume leukapheresis . This article discusses the current status of ‘high-dose’ therapeutic GTX.
HISTORICAL OVERVIEW OF THERAPEUTIC GRANULOCYTE TRANSFUSION
Regarding the efficacy of therapeutic GTX collected without donor G-CSF stimulation, seven historical controlled studies will be summarized [3–9][3–9][3–9][3–9][3–9][3–9][3–9]. The response of infected neutropenic patients to treatment with GTX and antibiotics was compared with that of comparable patients given antibiotics alone. Three of the seven studies reported a significant overall benefit for GTX [6–8][6–8][6–8]. In two additional studies [3,5][3,5], overall success was not demonstrated for GTX, but subgroups of patients benefited significantly. Thus, some measure of success for GTX was evident in five of the seven controlled studies. However, this success was counterbalanced by four studies reporting negative results − two with negative overall results [4,9][4,9] and two with negative results for some types of patients [3,5][3,5].
In explanation, patients in the three overall successful trials received relatively high doses of PMNs (≥1.7 × 1010/day). By contrast, two of the four studies with negative findings used PMNs collected by filtration leukapheresis [3,5][3,5] − PMNs now known to be defective. In the negative studies using PMNs collected by centrifugation leukapheresis [3,5,9][3,5,9][3,5,9], the dose was extremely low (0.41−0.56 × 1010 per GTX). The seven historical controlled GTX trials were analyzed by meta-analysis , and conclusions were that the low doses of PMNs transfused and the relatively high survival rate of the nontransfused control individuals were primarily responsible for the differing success rates of these studies − prompting the suggestion that severely neutropenic patients with life-threatening infections should be considered for GTX given in high doses .
ASSESSMENT OF MODERN THERAPEUTIC GRANULOCYTE TRANSFUSION
A modern therapeutic GTX is defined as a PMN collected from donors stimulated with G-CSF, with or without corticosteroids, by means of centrifugation leukapheresis using an erythrocyte-sedimenting agent while processing of large volumes of donor blood. An ideal PMN collection should include: 300–480 μg G-CSF subcutaneously and 8 mg dexamethasone given orally to the donor approximately 12 h before leukapheresis begins; hetastarch + concentrated citrate infused throughout the leukapheresis procedure at a ratio of 1 part starch to 12−14 parts donor blood and processing of 8−10 l of donor blood. The goal should be to transfuse 6−8 × 1010 neutrophils per GTX, with a lower limit of 4 × 1010.
As previously reviewed , many case reports and small series of patients suggested success for modern GTX − with the caveat that neutropenic patients with invasive/tissue fungal infections may not respond even to ‘high-dose’ GTX. Unfortunately, at this time, no definitive randomized clinical trials of modern therapeutic GTX collected after G-CSF and dexamethasone donor stimulation exist to establish either their efficacy or potential toxicity. Two randomized clinical trials have been reported or presented, but both have shortcomings/flaws that preclude firm guidelines for clinical practice.
The first published randomized clinical trial of ‘modern’ GTX provided no definitive guidelines for transfusion practices  because of problems including: never completed because of poor enrolment; donors stimulated only with G-CSF and no dexamethasone, resulting in relatively low PMN doses; GTXs were given every other day, not daily, again resulting in low PMN doses. Thus, the trial failed to test efficacy of modern ‘high-dose’ GTX .
The second, and most recent, randomized clinical trial (RING) was presented at the 2014 annual meeting of the American Society for Hematology with publication as an abstract [13▪]. Like the first trial, it stopped before completion and, accordingly, failed to provide definitive guidelines for clinical practice.
RING is a multicenter (14 clinical sites in the United States) randomized clinical trial in which 114 infected neutropenic (<500 neutrophils/microliter blood) patients were randomly allocated to treatment either with daily GTX collected from donors stimulated with G-CSF + dexamethasone and antibiotics (N = 56) or with antibiotics alone (N = 58). The primary endpoint was a composite of survival at 42 days after randomization and microbial response determined by a blinded adjudication panel. Results by intention-to-treat analysis (i.e. all individuals entered into the two arms, regardless of whether or not they received a GTX) were nearly identical (P > 0.99) with 42% in the GTX arm versus 43% of controls having a favorable response. Similarly, by per protocol analysis (i.e. only patients actually treated as intended), results were not different (P = 0.64) 49% favorable in the GTX arm versus 41% of controls. No differences in success rates were noted when analyzed per specific types of infection (i.e. bacteremia, fungemia, invasive/tissue bacterial infections or invasive/tissue fungal infections).
Although results appeared to be ‘negative’ (i.e. no advantage for GTX), 32% of patients in the GTX arm received less than the intended dose of PMNs/granulocytes. When patients actually given the intended ‘high-dose’ GTX were compared with those given the unintended lower doses of GTX, the ‘high-dose’ GTX patients had significantly higher success rates (P = 0.01). Specifically, the success rate was 58% for 26 patients given at least three GTXs with an average dose per transfusion of at least 5 × 1010 compared with 11% success for nine patients receiving lower leukocyte doses.
The reason(s) for the poor PMN/granulocyte doses were not determined, but it is clear that the potential benefits of high-dose GTX still remain a possibility − provided that high-dose GTXs (at least 4 × 1010 per transfusion) are actually given [13▪]. It is also clear that the collection of quality neutrophil/granulocyte concentrates requires expertise and careful quality assessment to consistently provide a product with any hope of benefit.
As a case-in-point, in a retrospective review of patients with invasive Fusarium infections, 11 patients with severe neutropenia and failure to respond to antifungal drugs received a median of seven GTXs containing a mean of 6.84×1010 granulocytes per GTX [14▪]. Ten of 11 patients (91%) had a favorable response, compared with an expected response of less than 50% when antifungal drugs are given without GTX. Over 98% of donors were unrelated community donors who were given 480 μg G-CSF subcutaneously 12–18 h + 8 mg dexamethasone orally 12 h before leukapheresis. Seven liters of donor blood were processed by continuous-flow leukapheresis using 500 ml 6% hetastarch anticoagulated with 30 ml of 46.7% trisodium citrate at a donor blood:hetastarch ratio of 12 : 1 throughout the procedure.
The use of G-CSF + dexamethasone to stimulate PMN donors has brought GTX therapy into a new era, as it is now possible to collect a relatively large number of PMNs/granulocytes (>4 × 1010). However, definitive/quality data to provide firm guidelines for clinical practice remain elusive. Randomized clinical trials have proven extremely difficult to complete because of poor/slow patient enrolment and the lack of dependable neutrophil/granulocyte leukapheresis products. There are strong hints that transfusing more than 5 × 1010 neutrophils/granulocytes daily to infected severely neutropenic patients may be beneficial.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
REFERENCES AND RECOMMENDED READING
Papers of particular interest, published within the annual period of review, have been highlighted as:
- ▪ of special interest
- ▪▪ of outstanding interest
1. Strauss RG. Therapeutic granulocyte transfusions in 1993. Blood 1993; 81:1675–1678.
2. Strauss RG. McLeod BC, Szczepiorkowski ZM, Weinstein R, Winters JL. Granulocyte (neutrophil) transfusions. Apheresis: principles and practice 3rd ed.Bethesda, MD: AABB Press; 2010. 215–218.
3. Alavi JB, Root RK, Djerassi I, et al. A randomized clinical trial of granulocyte transfusions for infection in acute leukemia. N Engl J Med 1977; 296:706–711.
4. Fortuny IE, Bloomfield CD, Hadlock DC, et al. Granulocyte transfusion: a controlled study in patients with acute nonlymphocytic leukemia. Transfusion 1975; 15:548–558.
5. Graw RG Jr, Herzig G, Perry S, Henderson IS. Normal granulocyte transfusion therapy. N Engl J Med 1972; 287:367–371.
6. Herzig RH, Herzig GP, Graw RG, et al. Successful granulocyte transfusion therapy for gram-negative septicemia. A prospectively randomized controlled study. N Engl J Med 1977; 296:701–705.
7. Higby DJ, Yates JW, Henderson ES, Holland JF. Filtration leukapheresis for granulocytic transfusion therapy. N Engl J Med 1975; 292:761–766.
8. Vogler WR, Winton EF. A controlled study of the efficacy of granulocyte transfusions in patients with neutropenia. Am J Med 1977; 63:548–555.
9. Winston DJ, Ho WG, Gale RP. Therapeutic granulocyte transfusions for documented infections: a controlled trial in 95 infectious granulocytopenic episodes. Ann Intern Med 1982; 97:509–515.
10. Vamvakas EC, Pineda AA. Meta-analysis of clinical studies of the efficacy of granulocyte transfusions in the treatment of bacterial sepsis. J Clin Apher 1996; 11:1–9.
11. Strauss RG. State of the art: role of granulocyte/neutrophil transfusions for hematology/oncology patients in the modern era. Br J Haematol 2012; 158:299–306.
12. Seidel MG, Peters C, Wacker A, et al. Randomized phase III study of granulocyte transfusions in neutropenic patients. Bone Marrow Transplantation 2008; 42:679–684.
13▪. Price T, McCullough J, Ness P, et al. A randomized controlled trial on the efficacy of high-dose granulocyte transfusion therapy in neutropenic patients with infection (RING). Blood First Edition paper, September 2, 2015; DOI 10.1182/blood-2015-05-645986.
Unable to document the efficacy of granulocyte transfusions overall, but offers a strong hint that when high doses of neutrophils/granulocytes are actually given daily, some infected neutropenic patients benefit.
14▪. Kadri SS, Remy KE, Strich JR, et al. Role of granulocyte transfusions in invasive fusariosis: systematic review and single center experience. Transfusion 2015; in press. doi:10.1111/trf.13099. [Epub ahead of print].