Markedly elevated serum level of presepsin in agranulocytosis with hematologic malignancy: A potential prognostic factor in a single-institution retrospective study after granulocyte transfusion

Abstract

Introduction

No established criteria exist for assessing the effectiveness of granulocyte transfusion (GTX) or biomarkers for predicting fatal infections in neutropenia. This study aimed to assess whether a novel sepsis marker, presepsin (P-SEP), is a useful prognostic indicator during GTX.

Methods

We collected frozen serum from 8 patients who had undergone GTX between September 2022 and October 2023 and measured their P-SEP levels. We compared these results with clinical records and assessed the alterations before and after GTX and their association with prognosis.

Results

The post-transfusion neutrophil count increased in all cases. In 5 of 8 patients (62.5%), P-SEP levels were reduced 1 day after GTX. Pretransfusion P-SEP levels were statistically significantly lower in the group of patients who survived and overcame infection after transfusion (GTX-survived) than in the group of patients who did not survive (GTX-nonsurvived) (1493 pg/mL vs 6658 pg/mL, P =.04). Transfused cell counts and changes in P-SEP levels 1 day after GTX were better in the GTX-survived group than in the GTX-nonsurvived group, although the difference was not statistically significant.

Discussion

Presepsin is a biomarker that can be assessed in patients undergoing GTX for agranulocytosis. A clinically significant increase in P-SEP levels before GTX may indicate ineffective GTX and an unfavorable prognosis.

Introduction

Neutrophils play a crucial role in host innate immunity; however, neutropenia is often prolonged in patients with hematologic malignancies who are undergoing chemotherapy or hematopoietic stem cell transplantation (HSCT).¹ Patients with neutropenia exhibit an increased risk of life-threatening bacterial and fungal infections that are proportional to the severity and duration of neutropenia.²

Pioneering studies over the past 5 decades have reported the efficacy of granulocyte transfusion (GTX) in the clinical management of sepsis in patients with neutropenia.^3,4 Subsequently, granulocyte apheresis was performed from a donor mobilized with granulocyte colony-stimulating factor (G-CSF), which enhanced granulocyte collection efficiency.⁵ In addition, combining G-CSF and dexamethasone in conditioning individuals for GTX harvest has advanced granulocyte mobilization.^6,7 An optimal granulocyte apheresis regimen ensures adequate granulocyte collection, and various centers have reported favorable infection control in patients with neutropenia.^8-10 Moreover, our institution has reported that GTX may serve as a safe bridging therapy for neutrophil engraftment following HSCT in patients with active infections.¹¹ When adequate granulocytes are available, GTX improved the outcomes of bacterial infection in patients with neutropenia.^12,13 However, there are limited indicators for determining the effectiveness of GTX, and the factors that affect post-GTX prognosis remain unclear.

In clinical practice, numerous immunologic biomarkers have been used for diagnosis and monitoring. Their continuous assessment aims to develop optimal indicators of infection.^14,15 Presepsin (P-SEP), a novel indicator of infection,^16,17 is an N-terminal fragment of the CD14 protein, which is a transmembrane protein expressed on neutrophils, monocytes, and macrophages.¹⁸ It is a soluble derivative of the lipopolysaccharide receptor, a member of the toll-like receptor family that recognizes pathogen-related molecular patterns and initiates innate immune responses. Presepsin is produced during infections and is strongly expressed during sepsis, with increased levels compared with those in patients with nonbacterial infections and healthy individuals.^17,19 Therefore, P-SEP has been used as a reliable sepsis marker in clinical practice. Despite the increase in biomarkers through phagocytosis, P-SEP is useful in patients with hematologic malignancies because P-SEP levels increase during infections, independent of the neutrophil count.^20-28 Although the majority of previous studies have focused on neutropenia, its association with agranulocytosis remains unclear. In contrast, GTX is administered to treat agranulocytosis, potentially inducing temporary immunologic alterations following GTX. To date, however, no studies have assessed the alterations in P-SEP levels. Therefore, this study aimed to assess whether alterations in P-SEP levels during GTX are useful indicators of its effectiveness.

Methods

Patient selection

This study included patients treated with GTX for agranulocytosis and lethal infection at Okayama University Hospital between September 2022 and October 2023. Patient data, including underlying disease diagnosis and treatment history as well as laboratory tests during GTX, were retrospectively collected from the medical records. Informed consent was obtained using an opt-out approach, and the protocol was approved by the Institutional Review Board of Okayama University Hospital. This study adhered to the ethical standards of the Institutional and National Research Committee and complied with the principles of 1964 Declaration of Helsinki and its subsequent amendments.

GTX recipients’ eligibility

The recipient eligibility criteria for GTX were as follows: (1) patients with agranulocytosis or near-agranulocytosis because of chemotherapy, conditioning regimen, or underlying hematologic diseases, (2) patients with an antimicrobial therapy-resistant infection before or after HSCT, and (3) patients at risk of lethal outcomes because of worsening infection before neutrophil recovery. Bacterial and fungal infections were defined through positive cultures from the infection site. In addition, patients with clinically evident pneumonia, enteritis, or sepsis but unknown pathogens were included. In conclusion, the attending physicians determined the GTX indications for each patient, considering donor availability, host factors, infectious disease severity, and tumor factors.

GTX donors’ eligibility

Donor eligibility criteria for GTX were as follows: (1) family members within the third degree of kinship to the recipient, (2) possessing an ABO blood type match or a minor mismatch with the recipient, (3) aged 18 to 65 years, and (4) no viral infection (HIV, hepatitis B virus, hepatitis C virus, or human T-cell leukemia virus type 1) at the time of apheresis. Informed consent was obtained from all donors before the procedure.

Granulocyte harvest procedure and infusion

Granulocytes were harvested from donors using the Spectra Optia system (Terumo BCT) and polymorphonuclear collection program protocol. Hydroxyethyl starch was used as a blood cell sedimentation agent to separate the granulocytes from the red blood cells. Granulocyte irradiation (15 Gy) was performed immediately after harvesting, followed by prompt injection of all granulocyte concentrates into the recipients. To prevent transfusion-related reactions, the patients received 100 mg hydrocortisone and 5 mg d-chlorpheniramine maleate before GTX.

Measurement of P-SEP

We collected frozen serum from 8 patients who had undergone GTX. We used existing specimens (residual specimens) from the completed clinical tests to measure P-SEP levels. The patients’ blood samples were collected on O-ringless screw cap tubes (1394-200-SS-C, WATSON) and centrifuged at 1930g for 6.5 minutes at 25 °C (S700FR, KUBOTA). The obtained blood serum was preserved at −30 °C, and P-SEP levels were measured collectively at a later date (median storage period, 4 months [range, 1-8 months]). Serum P-SEP levels were measured at LSI Medience Corporation (Tokyo, Japan) using STACIA CLEIA preassay kit (PHC Corporation). In brief, samples were first incubated with an alkaline-phosphatase–labeled antipresepsin rabbit monoclonal antibody (mAb) and subsequently with antipresepsin mouse mAb–coated magnetic particles. After washing, the magnetic particles were incubated with a chemiluminescent substrate solution (T2214, Thermo Fisher Scientific). The resulting luminescence was measured with the luminometer installed in the STACIA CLEIA system. The measurement range for this assay is 50 to 20 000 pg/mL.

Statistical analysis

Overall survival (OS) was defined as the time from the date of initial GTX to either the date of death or last follow-up. The Mann-Whitney U test was used to assess statistical significance between the group that survived and the group that did not survive. All statistical tests were 2 tailed, and statistical significance was set at P <.05. GraphPad Prism 9 software was used for the analysis.

Results

Patient characteristics

Table 1 presents the characteristics of the patients included in this study. Eight patients underwent GTX, with a median age of 42.5 years (range, 20-59 years). Among them, 3 patients underwent GTX during the postconditioning regimen and up to engraftment. The majority of the patients had acute leukemia. Table 2 outlines the clinical courses and patient outcomes. Patients 1, 2, and 6 died from infection, whereas patients 3 and 7 survived the infection but died of primary disease. The etiologies of the infections included sepsis (n = 3); sepsis and pneumonia (n = 2); pneumonia (n = 1); pneumonia and enterocolitis (n = 1); and pneumonia, sepsis, and septic pulmonary embolization (n = 1). Six patients developed sepsis. Of the 3 patients with pneumonia, 1 tested positive for Stenotrophomonas maltophilia and 1 tested positive for both S maltophilia and Staphylococcus haemolyticus; 1 patient with septic pulmonary embolization was the only case of fungal infection with Fusarium species in blood cultures. S haemolyticus was detected in the blood cultures of 2 of the 3 patients with sepsis only. No pathogens were identified in the remaining 3 cases. The median number of GTX per patient was 1 (range, 1-3). Three patients died of primary infection that contributed to GTX, and 2 of the remaining 5 patients died of their primary disease. The median follow-up period from GTX was 144 days (range, 2-323 days).

Table 1.

Patient Characteristics

Unique patient No.	Age, y	Sex	Body weight, kg	ECOG-ACRIN performance status	Hematopoietic cell transplantation–specific comorbidity index	Disease	Disease status at GTX	Most recent chemotherapy	GTX in peri-HSCT period
1	20	Male	50.6	2	1	Acute myeloid leukemia	Progressive disease	Venetoclax/azacitidine	No
2	25	Male	47.7	2	6	Acute myeloid leukemia	N/A	Fludarabine, melphalan, and total body irradiation	Yes (post-transplant cyclophosphamide haploidentical hematopoietic stem cell transplantation, pre-engraftment)
3	41	Female	70.8	1	0	Acute myeloid leukemia	Complete remission	Fludarabine, melphalan, and total body irradiation	Yes (cord blood transplantation, pre-engraftment)
4	59	Male	67.8	1	2	Myelodysplastic syndrome	Complete remission	Fludarabine, cytarabine, granulocyte colony-stimulating factor, and mitoxantrone	No
5	44	Female	75.4	4	1	Acute myeloid leukemia	Progressive disease	Fludarabine, melphalan, and total body irradiation	Yes (post-transplant cyclophosphamide haploidentical hematopoietic stem cell transplantation, pre-engraftment)
6	58	Female	60.4	3	8	Acute myeloid leukemia	N/A	Idarubicin/cytarabine	No
7	44	Female	50.5	1	3	Acute myeloid leukemia	Progressive disease	Fludarabine, cytarabine, and mitoxantrone	No
8	22	Female	52.8	1	3	Acute myeloid leukemia	Complete remission	Fludarabine, cytarabine, and mitoxantrone	No

GTX, granulocyte transfusion; HSCT, hematopoietic stem cell transplantation; N/A, not available.

Table 2.

Infectious Disease Treatment and Clinical Progress

Unique patient No.	Infection	Etiology	Type of antibiotics at GTX	Type of antifungal drugs at GTX	GTX Infusions, No.	Outcome of the primary infection	Outcome	Cause of death	OS after GTX, d
1	Pneumonia, sepsis, septic pulmonary embolization	Fusarium species	Meropenem, teicoplanin	Voriconazole	1	Not survived	Dead	Primary infection	2
2	Pneumonia, sepsis	S haemolyticus, S maltophilia	Meropenem, minocycline, sulfamethoxazole-trimethoprim, daptomycin	Micafungin, posaconazole	3	Not survived	Dead	Primary infection	24
3	Sepsis	Undetermined origin	Meropenem, levofloxacin, gentamicin, minocycline, teicoplanin	Liposomal amphotericin B	2	Survived	Dead	Primary disease	172
4	Pneumonia	Undetermined origin	Meropenem, minocycline, teicoplanin	Micafungin	1	Survived	Alive	N/A	323
5	Pneumonia, enterocolitis	Undetermined origin	Tazobactam/piperacillin, minocycline, teicoplanin	Micafungin, posaconazole	2	Survived	Alive	N/A	179
6	Pneumonia, sepsis	S maltophilia	Levofloxacin, minocycline, teicoplanin	Micafungin	1	Not survived	Dead	Primary infection	9
7	Sepsis	S haemolyticus	Meropenem, minocycline, vancomycin	Posaconazole	1	Survived	Dead	Primary disease	145
8	Sepsis	S haemolyticus	Tazobactam/piperacillin, minocycline, vancomycin	Posaconazole	1	Survived	Alive	N/A	143

GTX, granulocyte transfusion; N/A, not available; OS, overall survival; S haemolyticus, Staphylococcus haemolyticus; S maltophilia, Stenotrophomonas maltophilia.

Granulocyte recruitment and transfusion

Twelve GTX granulocyte products were obtained from 10 donors (Table 3). The donors had a median age of 43 years (range, 21-60 years) and a median body weight of 69.7 kg (range, 51.4-127.5 kg). Eight of the 10 harvests were performed 1 day before apheresis by administering 8 mg oral dexamethasone and 300 mg subcutaneous injection of filgrastim. In the other 2 harvests, apheresis was performed on the same day as that of the peripheral blood stem cell harvest following filgrastim mobilization on consecutive days or with pegfilgrastim. The median processed volume during granulocyte apheresis was 3.240 L (range, 2.949-3.419 L), with a median number of granulocytes in the product of 1.96 × 10¹⁰ cells (range, 1.074-4.145 × 10¹⁰). Eight patients underwent 12 GTXs, with a median number of granulocytes per patient’s body weight of 0.397 × 10⁹ cells/kg (range, 0.168-0.585 × 10⁹ cells/kg) (Table 3).

Table 3.

The Parameters of Granulocyte Harvest

Unique donor No.	Unique patient No.	Donor age, y	Donor sex	Donor body weight, kg	Conditioning for GTX harvest	Processed volume, L	Product neutrophil yields, × 1010	GTX dose, × 109/recipient body weight, kg
1	1	48	Male	76.4	G-CSF+dexamethasone	2.949	1.074	0.212
2	2	33	Male	77.8	G-CSF+dexamethasone	3.282	1.948	0.403
3		31	Female	58.5	G-CSF+dexamethasone	3.219	1.310	0.295
						3.244	1.378	0.318
4	3	39	Male	127.5	G-CSF+dexamethasone	3.235	4.145	0.585
						3.149	3.846	0.546
5	4	36	Female	63.1	Polyethylene glycol –G-CSF	3.181	2.826	0.416
6	5	48	Female	84.9	G-CSF	3.419	3.374	0.447
7		60	Male	57.1	G-CSF+dexamethasone	3.178	1.316	0.168
8	6	56	Female	57.4	G-CSF+dexamethasone	3.330	1.191	0.197
9	7	46	Female	51.4	G-CSF+dexamethasone	3.375	1.971	0.39
10	8	21	Male	76.2	G-CSF+dexamethasone	3.382	2.183	0.413

G-CSF, granulocyte colony-stimulating factor; GTX, granulocyte transfusion.

Alterations in P-SEP and outcomes following GTX

Residual serum samples were available at various time points: before GTX (pre-GTX), 1 hour after GTX, 1 day after GTX, and 2 days after GTX. Some patients received more than 1 GTX (patients 2, 3, and 5), but subsequent analyses were limited to the first round of transfusion (n = 8) to evaluate the association between trends in P-SEP levels before and after GTX and following clinical course.

Plotting alterations in P-SEP over time before and after the GTX revealed 2 groups based on the pre-GTX P-SEP levels: the GTX-survived group (red lines) and the GTX-nonsurvived group (blue lines) (Figure 1A). Pre-GTX P-SEP levels were statistically significantly lower in the GTX-survived group than in the GTX-nonsurvived group (1493 pg/mL [range, 491-2512 pg/mL] vs 6658 pg/mL [range, 6610-6717 pg/mL]; P =.04) (Figure 1B; Table S1). The P-SEP levels decreased 1 day after GTX in 5 of 8 patients (Figure 1A; Table S1). The median rate of change in P-SEP levels ([post-GTX day 1 − pre-GTX] / pre-GTX × 100) tended to be better in the GTX-survived group than in the GTX-nonsurvived group (−27.8% vs 7.3%, P =.25) (Figure 1C). The trends in creatinine and estimated glomerular filtration rate (eGFR) at the time P-SEP was measured are shown in Table S2. At most points, an eGFR of 60 mL/min/1.73 m² or higher (ie, grade 2 according to the Kidney Disease: Improving Global Outcomes 2024 guidelines) was achieved.²⁹ In this study, no cases were diagnosed with hemophagocytic lymphohistiocytosis (HLH) using bone marrow aspiration before or after GTX (data not shown). However, ferritin and triglycerides were not measured at the same time as P-SEP measurements; thus, HScores could not be evaluated.³⁰ In patients in the GTX-nonsurvived group, the prognosis was highly unfavorable, with a median OS of 9 days (range, 2-24 days) (Table 2). Patients who died of primary infections commonly had bacteria and fungi detected in blood cultures and presented with pneumonia and sepsis.

Figure 1.

P-SEP values were measured before and after GTX using residual serum from patients who had undergone GTX. A) The alteration of P-SEP after GTX. B) P-SEP before GTX in patients who survived (n = 5) and did not survive (n = 3) with a statistically significant difference (P =.04). C) The rate of change ([post 1 h − pre] / pre × 100) in P-SEP levels between the 2 groups before and 1 day after GTX (%) in patients who survived (n = 5) and did not survive (n = 3), not statistically significant (P =.25). The dotted line in (C) shows the baseline. Data are expressed as the mean (SEM). ns, not statistically significant; P-SEP, presepsin; GTX, granulocyte transplantation. *P <.05.

Neutrophil increase through GTX

The neutrophil counts following GTX treatment are shown in Figure 2. The neutrophil counts increased in all patients. The median neutrophil count increased from 0/μL (range, 0-444 µL) to 705/μL (range, 110-2250 µL) 1 hour after GTX and fell to 338/μL (range, 0-1211 µL) 1 day after GTX. Although there was no statistically significant difference between the 2 groups, the GTX-survived group (red line) exhibited higher neutrophil counts before GTX than did the GTX-nonsurvived group (blue line) (29/µL [range, 0-147/µL] vs 2/µL [range, 0-5 µL]). In addition, patients exhibited greater neutrophil increases 1 hour after GTX (634/µL [range, 330-920/µL] vs 347/µL [range, 110-750/µL]), 1 day after GTX (441/µL [range, 147-1191/µL] vs 188/µL [range, 21-502/µL]), and 2 days after GTX (183/µL [range, 75-360/µL] vs 153/µL [range, 20-410/µL]). The number of transfused granulocytes per patient’s body weight tended to be higher in the GTX-survived group; however, this difference was not statistically significant (0.45 × 10⁹/kg vs 0.27 × 10⁹/kg, P =.07) (Figure 2B).

Figure 2.

A) Change in neutrophil counts after GTX. Neutrophil counts in patients who survived (n = 5) and did not survive (n = 3). B) The number of transfusion cells of GTX in patients who survived (n = 5) and did not survive (n = 3). The red lines and symbols show patients who survived, and the blue lines and symbols show patients who did not survive, respectively. Data are expressed as mean (SEM). ns, not statistically significant (P = .71); GTX, granulocyte transplantation.

Clinical course in a case with 3 transfusions

In our case series, P-SEP levels were not measured in real time. In addition, serum from the period not involving GTX was unavailable because this study focused on assessing the effect of GTX and collected serum before and after GTX. Therefore, P-SEP levels were not measured at all stages of the clinical course. In contrast, C-reactive protein (CRP) levels in all patients and procalcitonin (PCT) levels in some patients could be extracted from the medical records (Table S1). Comparing these markers with the P-SEP levels regarding effectiveness is challenging because of insufficient data. Therefore, in 1 case (patient 2) for which we had complete data on these 3 inflammatory markers, we examined the data graphically to discuss the clinical course and outcome (Figure 3).

Figure 3.

Clinical course of patient 2, who had a fatal course despite multiple GTX cycles. CHDF, continuous hemodialysis and filtration; CRP, C-reactive protein; GTX, granulocyte transfusion; HSCT, hematopoietic stem cell transplantation; PCT, procalcitonin; S haemolyticus, Staphylococcus haemolyticus; S maltophilia, Stenotrophomonas maltophilia.

Because the patient developed sepsis from S haemolyticus on day 0 after HSCT, multiple antibiotic therapies were initiated, and GTX was scheduled because of anticipated delayed neutrophil recovery. Although P-SEP, CRP, and PCT levels were reduced following the initial GTX infusion on day 7, the patient required ventilatory support for pneumonia and continuous hemodialysis and filtration for acute kidney failure on day 9. In addition, the patient developed sepsis from S maltophilia on day 14, and additional GTX infusions were administered on days 15 and 22. An increase in neutrophil count was observed after GTX, and CRP and PCT levels gradually fell. In contrast, P-SEP levels remained consistently high and increased the subsequent day after the third GTX. The patient died of multiple organ failure on day 32, and an autopsy confirmed bacterial pneumonia. However, no other infections were identified, and the primary disease had infiltrated the liver and spleen.

Discussion

In this study, we assessed the association between clinical outcomes and serum P-SEP levels in patients who had received GTX for life-threatening infections because of agranulocytosis following chemotherapy or HSCT. Granulocyte transfusion was effective in overcoming infections in 5 of the 8 patients (Table 2). This study indicated that statistically significantly higher P-SEP levels before GTX were associated with an unfavorable prognosis, despite a transient increase in neutrophil counts in all patients. Patients with higher P-SEP levels on day 28 after HSCT (≥2000 pg/mL, as determined by the 75th percentile value) exhibited statistically significantly unfavorable OS (hazard ratio, 3.7 [95% CI, 1.8-7.7]; P <.01) compared with patients who had lower P-SEP levels, which is primarily associated with a higher incidence of therapy-related mortality (hazard ratio, 20.2 [95% CI, 4.7-87.6]; P <.01).²² They measured P-SEP levels using serum samples collected following neutrophil engraftment, which had been achieved at 28 days after HSCT, and reported an association with treatment-related mortality. In contrast, this study indicated an association between P-SEP levels before neutrophil recovery and infectious disease mortality.

A soluble CD14 subtype, P-SEP is a glycoprotein expressed on the membrane surfaces of phagocytes (neutrophils, monocytes, and macrophages). It serves as a pattern-recognition molecule in the innate immune response to microorganisms by activating an inflammatory signaling cascade upon contact with infectious agents.¹⁸ The fact that CD14 is expressed on neutrophils raises concerns regarding its reliability as an immunologic biomarker in patients with neutropenia. However, studies have demonstrated that in the absence of infection, P-SEP levels in patients with neutropenia remain at baseline compared with patients without neutropenia.^20,21 In our analysis of P-SEP levels in patients with neutropenia and an infection (Table 2), we observed that the levels were not lower than those observed in a study assessing P-SEP in sepsis that satisfied diagnostic criteria for systemic inflammatory response syndrome (mean (SD), 817.9 ± 572.7 pg/mL in sepsis and mean (SD), 1992.9 ± 1509.2 pg/mL in severe sepsis).¹⁷

To the best of our knowledge, there have been no previous reports of alterations in serum P-SEP levels among patients undergoing GTX. Although previous findings indicated no correlation between P-SEP levels and neutrophil or monocyte counts,^21,22 our data indicate that the rapid increase in neutrophil counts following GTX infusion did not increase P-SEP levels (Figure 1A). In contrast, the mechanism through which transfused granulocytes reduce inflammation in the human body following GTX and the reason for the reduction in P-SEP levels after GTX remain unclear. However, P-SEP has been reported to rapidly reflect the treatment effects.²¹ In addition, because all patients were receiving antimicrobial agents, it is possible that the influx of granulocytes affected the treatment effect, resulting in the reduction of P-SEP levels 1 hour after transfusion.

Various immunologic biomarkers are used to diagnose and monitor infections in daily clinical practice. For example, CRP and PCT are commonly explored as immunologic biomarkers to identify patients at risk of febrile neutropenia–related complications.^23,24 Recent studies have demonstrated the effectiveness of P-SEP and other immunologic biomarkers in patients with febrile neutropenia.^21,25-28 When the role of P-SEP is limited to febrile neutropenia, distinctive aspects can be noted. First, P-SEP levels increased earlier than levels of CRP or PCT within 2 hours and reached a peak 3 hours after infection.³¹ Similarly, P-SEP levels increase statistically significantly earlier in febrile neutropenia than in other immunologic biomarkers, facilitating early identification of infection.^25,26 Second, P-SEP exhibits high specificity for bacteremia.^26,27 Third, P-SEP was superior to PCT for daily monitoring of patients during infection and antibiotic treatment on consecutive days at the onset of neutropenia.²⁶

The efficacy of GTX is well documented. A review of previous studies on GTX in patients with hematologic disorders and postchemotherapy neutropenia or severe aplastic anemia reported that 17 studies have demonstrated recovery from infection in 36.7% to 82.6% of patients.¹³ Although P-SEP is useful in diagnosing and monitoring infections, its potential as a prognostic factor for life-threatening infections in patients with neutropenia has not been explored to our knowledge. The most statistically significant finding of this study was the association between higher pre-GTX P-SEP levels and an unfavorable prognosis in patients with neutropenia following GTX (Figure 1A, 1B). Studies on severe sepsis or septic shock not limited to neutropenia have demonstrated that baseline P-SEP levels independently predict mortality and organ failure, and are associated with the severity of sepsis.^32,33 In addition, it has been demonstrated that P-SEP levels do not reduce when inappropriate antibiotic therapy is administered for infection.³² Although P-SEP is a useful immunologic biomarker of infection, even in patients with neutropenia, the threshold for P-SEP levels before GTX that predict GTX efficacy remains unclear. However, it may be beneficial to measure P-SEP levels on consecutive days before and after GTX to predict its efficacy; assess therapeutic response; and adjust treatment strategies, such as preparing for subsequent GTX or altering antibiotics and antifungal drugs.

This study has several limitations. First, it should be regarded as preliminary because of its retrospective design, small and heterogeneous populations, short follow-up period, and reliance solely on descriptive data obtained through univariate analysis. Second, because there is no control group (ie, without GTX), it was difficult to compare directly the effects of P-SEP in patients with and without GTX. Finally, P-SEP levels are known to be elevated in patients with kidney dysfunction and in patients with HLH.^34,35 In this study, eGFR at the time of P-SEP measurement was below grade 2 at most points, which is expected to have little effect on the measured P-SEP levels,³⁴ and no patients revealed phagocytosis in the bone marrow aspirate specimens. However, the process was not able to completely rule out the possibility of falsely elevated P-SEP levels in patients with eGFR grade 3 (patient 2) or the presence of HLH in all patients. Notwithstanding these limitations, our study is the first to identify the variation in P-SEP before and after GTX in patients with fatal infections during chemotherapy and allo-HSCT for hematologic malignancies, and our findings may provide valuable insights for guiding and determining treatment.

In conclusion, GTX has demonstrated efficacy in addressing life-threatening infections in patients with neutropenia following chemotherapy or HSCT, although it is not effective in all cases. Pretransfusion P-SEP may serve as an appropriate biomarker for assessing the therapeutic efficacy and clinical outcomes in patients undergoing GTX for agranulocytosis. Further studies involving larger patient cohorts are required to assess these results and advance our understanding of GTX efficacy.

Supplementary Material

Supplementary material is available at Laboratory Medicine online.

Conflicts of interest

The authors have no conflicts of interest directly relevant to the content of this manuscript.

Funding

This work was supported by The Japan Society of Transfusion Medicine and Cell Therapy and PHC Corporation.