Therapeutic granulocyte transfusions in 1993 [editorial] [see comments]

Abstract

Bacterial, yeast, and fungal infections are an undeniable consequence of both severe neutropenia (<0.5 X lo9 PMNs/ L blood) and disorders of abnormal PMN function. In many studies, benefit has been shown for GTX as treatment for infections in certain clinical settings in animals and in humans.* However, enthusiasm has waxed and waned. In the 1970s, zealous advocates of GTX stated that it was unnecessary, unwarranted, and perhaps even unethical to conduct controlled, clinical trials to define the seemingly obvious efficacy of therapeutic GTX.3-5 Despite such enthusiasm and the plethora of published reports supporting the transfusion of granulocyte concentrates, the use of GTX has diminished strikingly. Although this can be explained, in part, by the development of more effective antibiotics and alternative therapies such as intravenous (IV) gammaglobulin and recombinant myeloid growth factors, many physicians hold strong negative opinions about the value of GTX, for reasons that are not readily apparent. At present, many physicians believe that there is little, if any, role for GTX in the management of infected neutropenic patients! In my view, this excessively negative attitude is unjustified scientifically, and it is reasonable to critically assess both the role of therapeutic GTX in the light of current standards of practice and the feasibility of using G-CSF, per the method of Bensinger et al,’ to stimulate PMN donors. ROLE OF THERAPEUTIC GTX IN NEUTROPENIC PATIENTS A detailed analysis of 30 papers reporting use of GTX plus antibiotics to treat infections in severely neutropenic patients (usually <0.5 PMN X 109/L blood) has been presented previously.’ In summary, data from all 30 studies are displayed in Table 1. Patients were tabulated according to the index infection that prompted GTX therapy. Patients were counted only once, and those with septicemia were listed in the sep ticemia section, whether or not they exhibited another infection such as pneumonia, abscess, etc. All patients administered GTX for a designated type of infection were tabulated in the column labeled “treated.” Patients for whom the actual course and mortality of the infection could be clearly documented were then listed again in the “evaluable” column. GTX were considered to be successful, if so stated by the investigator. Many of the 30 reports were of uncontrolled studies consisting of small numbers of patients with a diversity of underlying diseases, types of infections, antimicrobial therapy, and GTX management (ie, variable dose and quality of PMNs). Because of these complex issues, combining data from multiple reports is of limited value. It was done here primarily to document the breadth of experience reported. It is clear that information regarding the efficacy of GTX for specific types of infection is fairly limited, with the possible exception of bacterial septicemia. To obtain more definitive information, the seven controlled studies were analyzed in more detail.’-14 In the seven controlled studies, the response of infected, neutropenic patients to treatment with GTX plus antibiotics (study group) was compared with that of comparable patients, evaluated concurrently, who were administered antibiotics alone (control group). The results of these seven studies are presented in Table 2. In three of the seven studies a significant overall benefit for GTX was found.”-13 In two additional overall success was not demonstrated for GTX, but certain subgroups of patients were found to benefit significantly. One study with partial success was the first controlled study of GTX reported.’ In this early study, many patients received an inadequate dose of GTX by current standards, and overall success could not be demonstrated. However, 100% of patients who received at least four GTX, and 80% of those receiving at least three GTX survived, as compared with only 30% survival of controls. In the other study with partial benefit,” no advantage for GTX could be demonstrated when all patients were analyzed. However, when the subgroup of patients with persistent bone marrow (BM) failure was analyzed separately, 75% of those receiving GTX responded favorably compared with only 20% of controls. This indicated that transfused PMN were likely to benefit patients with persistently severe neutropenia, but not those with recovering marrow function. Thus, some measure of success for GTX was evident in five of the seven controlled studies. However, success was not uniform, as evidenced by two negative studie~.~.’~ One explanation for the inconsistent success of the controlled studies is evident on critical analysis of the adequacy ofGTX support (Table 3). Patients in the three trials reporting overall success received relatively high doses of PMNs, collected from donors who were selected on the basis of leukocyte