Temporary Glucose − 6 phosphate dehydrogenase enzyme deficiency, an additional risk factor for invasive fungal infection in immune-compromised AML patients

Abstract

Invasive fungal infections (IFI) represent a major threat to acute myeloid leukemia (AML) patients, especially after induction chemotherapy. Intense chemotherapeutic drugs further weaken the immune function of AML patients which increases their susceptibility to life-threatening fungal infection, including invasive aspergillosis (IA). At present, there is no single reliable serological marker or method to predict accurately the risk of IA after induction chemotherapy. We hypothesized that some anti-cancer medicines and antibiotics may trigger temporary Glucose-6-Phosphate Dehydrogenase (G6PD) level which may increase the risk of fungal infection. In this study, we found that the serum galactomannan (GM) level was high in AML patients during IA and decreased in patients who recovered from IA. Along with the serum GM level, the level of G6PD was measured during IA and recovery. We found that AML patients had reduced serum G6PD levels during IA and experienced restoration of G6PD levels upon recovery in paired samples. Interleukin-6 (IL-6) was found to be higher in serum during fungal infection and positively correlated with the level of lactate dehydrogenase (LDH). Overall, monitoring G6PD levels along with IL-6 and GM in serum may be helpful in predicting IFI in AML patients.

Clinical trial number

Not applicable.

Introduction

Invasive fungal infections (IFIs) are a serious threat for acute myeloid leukemia (AML) patients, accounts for 24–30% of morbidity and mortality in patients with AML [1]. Patients with haematological malignancies are particularly susceptible to IFI due to multiple factors, including the underlying disease, treatment related immune-suppression during induction therapy, prolonged neutropenia and severe gastrointestinal mucosal damage because of intensive cytotoxic therapy [2]. In addition, prolonged exposure to broad-spectrum antibiotics is a well-known risk factor for fungal colonization and subsequent invasive infections. Historically, Candida and Aspergillus species were the predominant causes of IFIs in patients with haematological malignancies [3]. However, in recent years, both the incidence and clinical significance of invasive aspergillosis have increased worldwide [4]. Aspergillus species have now emerged as the most common mold pathogens, accounting for nearly 90% of mold infections and causing severe disease in approximately 2–28% of patients with AML [5]. Furthermore, chemotherapy-induced dysregulation of key signalling pathways involved in antifungal immunity may further increase susceptibility to IFIs [6].

Glucose-6-phosphate dehydrogenase (G6PD) plays a central role in several cellular immune responses [7]. Sanna et al. reported that G6PD may be an important determinant of susceptibility to opportunistic infection and host immune response in neutropenic patients [8]. G6PD, a critical cytoplasmic enzyme that catalyses the first step of the pentose phosphate pathway (PPP), generating nicotinamide adenine dinucleotide phosphate (NADPH) from NADP⁺ [9]. NADPH serves a dual function, acting both as an essential substrate for the generation of reactive oxygen species (ROS) and as a signalling molecule in infection-related pathways [10].

Antifungal phagocytic activity involves both oxidative and non-oxidative mechanisms. During the oxidative burst, phagocytic leukocytes utilize NADPH to generate toxic ROS, which are essential for the intracellular killing of pathogens. G6PD activity is therefore crucial for maintaining redox homeostasis by regulating cytotoxic ROS levels [11]. Previous studies have demonstrated that neutrophil killing activity is enhanced by NADPH oxidases [12]. Consequently, G6PD deficiency, which impairs NADPH oxidase function in phagocytes, may increase susceptibility to recurrent infections caused by catalase-positive organisms, including Aspergillus species [13]. Though G6PD expression is up regulated in most of the cancer but has been found to be down regulated in AML.

Low levels of G6PD may result from reduced enzyme synthesis, accelerated degradation, impaired dimerization, loss of catalytic activity, or a combination of all these mechanisms. Since reduced G6PD synthesis has not been widely reported, decreased enzyme activity resulting from impaired dimerization or accelerated degradation is considered the most common cause of reduced G6PD levels and may be responsible for altered functional properties of the enzyme [14]. In addition, lower levels of G6PD may be associated with both anaemia and neutrophil abnormalities through increased cellular susceptibility to oxidative stress. Although G6PD deficiency is not typically considered a direct cause of neutropenia, but lower G6PD levels in leukemia patients may be associated with neutrophil abnormalities due to factors such as oxidative stress, bone marrow suppression, chemotherapy-related toxicity and overall disease severity. Moreover, the stability and enzymatic activity of G6PD depend on its binding to NADP+, making intracellular NADP+ levels critical for maintaining its function [15].

Neutrophils play a pivotal role in host defence against fungal infections. In response to infection, stimulated neutrophils release chromosomal DNA to form neutrophil extracellular trap (NET) for trapping and killing invading microbes. The formation of NETs requires a metabolic shift toward the pentose phosphate pathway (PPP), a process that may be disrupted by G6PD inactivation [16].

A number of cytokines play a crucial role in the immune response against fungal pathogens, particularly Interleukin-6 (IL-6). In response to infections and tissue injuries, IL-6 produced immediately to contribute in host defence. Although its expression is tightly regulated under physiological conditions, but sustained dysregulated IL-6 production may contribute to chronic inflammation and pathological immune responses [17]. Since IL-6 is a soluble mediator with a pleiotropic effect on inflammation and immune response, measurement of IL-6 may have prognostic and diagnostic value in patients with the fungal infections [18].

Lactate dehydrogenase (LDH) is an enzyme that catalyses the reversible conversion of lactate to pyruvate with the interconversion of NAD + to NADH [19]. Elevated serum LDH levels have been observed in a various lymphoproliferative disorders and inflammatory conditions, including infections [20]. We hypothesized that serum LDH may serve as a useful inflammatory marker, particularly in patients with invasive pulmonary aspergillosis (IPA).

Early diagnosis of IFIs remains challenging, especially during the initial stages of infection when antifungal therapy is most effective [5]. Currently, enzyme-linked immunosorbent assays (ELISAs) for galactomannan and β-D-glucan are the only commercially available biomarkers commonly used to detect IFIs in patients with haematological malignancies. However, sometimes these assays are limited by false-positive results associated with conditions such as bacteremia caused by Gram-positive organisms, concomitant use of amoxicillin-clavulanate, Streptococcus pneumoniae infection and administration of albumin products or coagulation factors [21]. In this context, the identification of additional complementary biomarkers may improve the accuracy of predicting invasive fungal infections.

The present study aimed to evaluate the association of G6PD and IL-6, in combination with galactomannan and LDH levels, with invasive fungal disease in high-risk AML patients undergoing induction therapy. The utility of these markers as predictive biomarkers warrants further prospective validation.

Materials and methods

Patient selection

33 adult Acute Myeloid Leukemia patients, treated in the Department of Medical Oncology, All India Institute of Medical sciences (AIIMS), New-Delhi, India were recruited, out of which 30 patients were included in this study. As controls, 7 healthy, age matched adults, without any prior exposure to medication were also included. Peripheral blood samples were collected from patients at baseline i.e. at the time of diagnosis, and during febrile neutropenic condition. Routinely serum sample were analysed for galactomannan antigen twice per week during induction therapy to correlate positivity for GM with the diagnosis of invasive aspergillus infection until the high-risk condition for developing IFI had subsided [22]. No patients with G6PD deficiency were enrolled in this study. The availability of serial serum samples together with complete clinical records were useful to review patient’s characteristics retrospectively which are summarized in Table 1.

Table 1.

Demographic characteristics of AML patients enrolled in this study

Patient Characteristics (n = 30)	Median (n)
Age (Range: 20–60 years)	≤ 36 (15) > 36 (15)
Gender (%) Male Female	40% (12) 60% (18)
Infection Status	Fungal (Probable) Infection (+): 23 Fungal Infection (-): 7
Induction Therapy	3 + 7 (20) 3 + 5 (3) Decitabine (4) Low Dose Ara-C (3)
Pre-emptive anti-fungal therapy	Voriconazole (7) Amphotericin B (8) None (15)

Collection and storage of serum samples

Five millilitres (5ml) of whole peripheral blood samples were collected in BD Vacutainer™. Blood samples were allowed to clot for 2 h at room temperature before centrifugation for 15 min at 1000×g. Serum (supernatant) was separated from the blood and tested for galactomannan. The serum samples were stored frozen at -80 °C until tested for G6PD and LDH etc. For Interleukin-6 evaluation, plasma separation was done from two ml of peripheral blood collected in an EDTA tube. Patient’s blood samples were centrifuged at 1600–2000×g for 15 min. The separated plasma removed carefully and stored into plain eppendorf at -80 degree centigrade for further use.

Management of patients for fungal infection

All patients were nursed in rooms with HEPA filtration. 50% of AML patients included in this study were treated with antifungal drugs. In our setting, voriconazole remains the standard of care for patients with proven or probable invasive aspergillosis, while amphotericin B is reserved for those with suspected azole-resistant infections or who develop toxicity or intolerance to azole therapy. Voriconazole (400 mg twice daily, orally) was given to 7(23.3%) of 30 patients whereas 8 patient (26.6%) received liposomal amphotericin B (1–3 mg/kg/day IV).

Definition of invasive aspergillosis

Episodes of invasive aspergillosis (IA) were classified according to the European Organization for Research and Treatment of Cancer/Invasive Fungal Infections Cooperative Group and National Institute of Allergy and Infectious Diseases Mycoses Study Group (EORTC/MSG) consensus criteria. The diagnosis of probable IA was based on a combination of host factors, clinical features, microbiological evidence, and radiological findings. Serum galactomannan (GM) positivity was included as part of the diagnostic evaluation for all probable IA cases in this study [23].

Galactomannan detection (GM)

To diagnose fungal infection, blood serum was tested for galactomannan antigen by using the commercially available sandwich ELISA kit. For the detection of galactomannan antigen in serum, the Patelia™ Aspergillus ELISA kits (Cat no# 62794, Bio-Rad) was used as recommended by manufacturer. The results were expressed as optical density (OD), which is obtained after dividing the OD of patient’s serum sample with the average OD of cut-off controls provided in the kit [24]. Optical density more than 0.8 in two consecutive samples for a patient was considered true positive for aspergillus infection. Results in-between 0.5 and 0.8 OD (gray zone) were considered borderline positive whereas OD less than 0.5 was considered as negative.

Diagnostic work-up of IA

In cases of clinical suspicion of fungal infection, patients having GM index more than 1 in two consecutive ELISA tests was further tested by lateral flow (IMMY, cat no # AF2003) following recommended protocol. As a confirmation of Aspergillus infection, bronchoalveolar lavage (BAL) fluid was also tested for GM antigen, followed by pulmonary high-resolution computed tomography (HRCT) scan.

G6PD determination

G6PD concentration in serum was detected by using ELISA kit (Cat no#E-EL-H1816) sandwich-ELISA method. The assay was performed according to the recommended protocol. The OD was measured by using spectrophotometry at a wavelength of 450 nm. The concentration of G6PD present in patient’s serum was calculated by comparing the OD of the standard curve. The OD value is proportional to the concentration of G6PD.

Detection of interleukin-6

To determine the IL-6 concentration in plasma of patients and in healthy control, sandwich-ELISA has been performed by using human IL-6 ELISA kit (Cat. no# E-EL-H0102). The ELISA plate provided in this kit was pre-coated with an antibody specific to human IL-6. The plasma separated from whole blood samples were added in the wells of ELISA plate and followed the recommended protocol. The OD value will be proportional to the concentration of IL-6 present in the sample. The concentration of IL-6 in the sample was calculated by comparing with the standard curve based on the OD value of standards with known concentration.

Estimation of lactate dehydrogenase

To estimate the lactate dehydrogenase in blood serum, kit (Clinical System, cat. no# 1102160025) approved by International Federation of Clinical Chemistry and laboratory medicine (IFCC) were used to run LDH assay according to manufacturer’s protocol. This protocol is based on an enzymatic coupling reaction and results were represented as International Units per Litre (IU/L).

Statistical Analysis: Differences between independent groups were assessed using the Mann-Whitney U test, due to the non-normal distribution of data as confirmed by the Shapiro-Wilk test. The Wilcoxon signed-rank test was used to compare paired samples from the same patients at two time points. Spearman’s rank correlation coefficient (p) was used to assess association between continuous variables. Survival distributions were compared using the Kaplan-Meier method and the log-rank (Mantel-Haenszel) test. All tests were two-tailed and p < 0.05 was considered statistically significant. Data were analysed using GraphPad Prism version 8.0.0 (GraphPad Software, San Diego, CA, USA) and Stata version 11.2.

Result

Demographic characteristics and clinical management

Overall baseline characteristics for all enrolled Acute Myeloid Leukemia (AML) patients involved in this study are summarized in Table 1. Out of 30 total patients, 12 were male (40%), whereas 18 patients were female (60%). The median age of the patients was 36, ranging between 20 and 60 years old. Two thirds of AML patients (20 out of 30) received the standard ‘3 + 7’ induction therapy regimen, which comprised of daunomycin (60 mg/ m²/day), administered on days 1 to 3 and cytosine arabinoside (100–200 mg/m²/day) given from days 1 to 7. Three patients received ‘3 + 5’ regimen, consisting of daunorubicin (60 mg/ m²/day) or another anthracycline administered from day 1 to 3, along with cytarabine (100–200 mg/m²/day) given as a continuous infusion from days 1 to 5. Decitabine was administered to 4 patients at a dose of 20mg/m²/day for 5 consecutive days. In addition, low-dose cytarabine (LD-Ara-C) was given to 3 elderly patients at a dose of 20mg/m²/day as treatment for AML.

Out of the 30 patients, 23 at least had single episode of (probable) fungal infection, among them 10 patients had more than one episode of fungal infection during induction chemotherapy. Seven patients remained free from fungal infection during the therapy. Fifteen AML patients were treated with either Amphotericin B or voriconazole as antifungal therapy to treat (probable) fungal infection, whereas 8 patients did not receive therapy for presumed IFI. Antifungal treatment allocation was not randomized and was influenced by multiple clinical considerations, including physician judgment, evolving clinical condition, drug tolerance, and resource-related factors. Pre-emptive therapy with voriconazole was given to 7 (23.3%) out of 30 patients. Eight patients (26.6%) received liposomal amphotericin B. Criteria for initiating pre-emptive therapy included (i) patients having persistent fever after 5 days of intravenous antibiotic treatment, (ii) identification of fungal mycelia in the respiratory tract and (iii) recurrence of fever after an afebrile interval of at least 48 h in neutropenic patients still receiving broad spectrum antibiotics. Patients age less than 18 years or having secondary AML or pre-exposed to any chemotherapeutic drugs due to having other malignancy were not included in this study.

Presence of galactomannan antigen in blood serum during fungal infection

Galactomannan, a component of fungal cell wall generally released into the blood due to invasive fungal infection by Aspergillus species. To check the presence of invasive fungal infection by Aspergillus, galactomannan antigen was measured in the serum of all the patients twice per week during induction chemotherapy. A higher galactomannan antigen level was found in serum of the AML patients with (probable) invasive fungal infection in comparison to AML patients without invasive fungal infection and IFI-negative patients had comparable GM values to healthy individuals (p = 0.46) (Fig. 1a). To explore further, galactomannan was also measured in the same patients in presence or absence of fungal infection as a paired sample. GM levels were significantly elevated during infection (median OD: 1.64) compared to the non-infected state (median OD: 0.28; p < 0.0001), demonstrating a consistent increase following infection onset (Fig. 1b; Table 2). We have seen that galactomannan level was less than 0.5 optical densities in AML patients without infection but increased significantly once the same patient experienced a probable infection.

Fig. 1.

Increased Galactomannan (GM) level in the serum of AML patients with Invasive Fungal Infection (IFI). (a) Serum GM antigen levels were compared among IFI-positive AML patients, IFI-negative AML patients and healthy controls using the Mann-Whitney U test. GM levels were significantly higher in IFI-positive patients than in IFI-negative patients. IFI-negative patients and healthy controls had comparable GM levels (p = 0.46). (b) Paired serum GM levels from the same AML patients during invasive fungal infection and at their pre-infection baseline. GM levels were significantly elevated during the infected state compared to the non-infected state (median OD: 1.64 vs. 0.28; p < 0.0001; Wilcoxon signed-rank test), reflecting a consistent within-patient increase upon infection onset

Table 2.

Comparison of galactomannan, G6PD, IL-6, and LDH levels in AML patients with and without fungal infection and healthy controls

	Galactomannan (OD Value)/Median (Range)	G6PD (pg/mL) /Median (Range)	IL-6 (pg/mL) /Median (Range)	LDH (U/L) /Median (Range)
AML patients during fungal infection	1.64 (0.83–4.66)	11.6 (2.25–16.7)	463.8 (156–1166)	462 (159-2244.5)
AML patients without fungal infection	0.28 (0.15–0.58)	17.25 (10.55-22)	200 (40.62-777.37)	238.5 (58.5-1141.25)
p-value (Infected vs. Non-Infected)	< 0.0001	0.001	< 0.0001	0.0005
Healthy Individual	0.29 (0.22–2.8)	-	2.47 (1.01–68.08)	376 (265.5-424.5)
p-value (Patient vs. Healthy)	0.46	-	< 0.0001	0.92

Comparisons between groups were performed using the Mann–Whitney U test; statistical significance was defined as p < 0.05

Correlation of Glucose 6-phosphate dehydrogenase level with invasive fungal infection

Glucose-6-phosphate dehydrogenase is a cytoplasmic enzyme which affects many cellular functions including immune responses. Since it catalyses the production of NADPH, which is required to destroy microorganisms by white blood cells in the respiratory burst reaction (phagocytic leukocytes), we assessed G6PD levels in all AML patients with a probable invasive fungal infection and in those free from fungal infection. Interestingly, G6PD levels were low in patients during a probable fungal infection (median: 11.6 ng/ml) whereas significantly higher levels of G6PD were observed at baseline, prior to development of infection (median: 17.25 ng/ml; Fig. 2a; Table 2), consistent with an association between reduced serum G6PD levels and active fungal infection in this cohort. To further analyse whether lower level of G6PD is associated with the severity of fungal burden, we assessed the correlation between serum G6PD and galactomannan levels in AML patients during invasive fungal infection. A significant negative correlation was observed between G6PD and galactomannan values, demonstrating that lower G6PD levels were associated with higher serum Aspergillus antigen levels (Fig. 2b).

Fig. 2.

Reduced Serum G6PD levels in AML patients during Invasive Fungal Infection. (a). Paired serum G6PD levels from the same AML patients at pre-infection baseline and during invasive fungal infection. G6PD levels were significantly reduced during the infected state compared to the non-infected state (median: 11.6 vs. 17.25 ng/mL; p = 0.0001; Wilcoxon signed-rank test). (b). Spearman rank correlation between serum G6PD levels and galactomannan (GM) values in IFI-positive AML patients. An inverse association was observed, whereby lower G6PD levels corresponded with higher GM values (ρ = -0.4303; p = 0.040)

Impact of interleukin-6 in fungal infection

Interleukin-6 plays crucial role in the defence mechanism against fungal infections. To assess the impact of IL-6 in relation to probable fungal infection, we have evaluated the level of IL-6 in all the AML patients with infection versus no fungal infection. Samples from healthy individuals were also checked for IL-6 levels and compared with AML patients. The overall levels of interleukin-6 were elevated significantly in serum of haematological malignancy patients in comparison to healthy individuals (Fig. 3a). Moreover, the IL-6 level was higher in the same patients with AML and probable fungal infection (median IL-6: 463.8 pg/ml) compared to when they were not infected (median IL-6: 200 pg/ml) (Fig. 3b; Table 2). However, despite the elevation, IL-6 levels did not correlate with galactomannan levels in infected patients, suggesting that IL-6 elevation may reflect general immune activation rather than a measure of fungal antigen burden (Fig. 3c).

Fig. 3.

Elevated IL-6 levels in AML patients having invasive fungal infection. (a) Serum IL-6 levels compared among healthy individuals, IFI-negative AML patients, and IFI-positive AML patients using the Mann-Whitney U test. IL-6 levels were significantly higher in IFI-negative AML patients than in healthy individuals (p = 0.0024), and were further significantly elevated in IFI-positive AML patients compared to both IFI-negative AML patients (p = 0.01) and healthy individuals (p < 0.0001). (b) Paired plasma IL-6 levels from the same AML patients at pre-infection baseline and during invasive fungal infection. IL-6 levels were significantly elevated during the infected state compared to the non-infected state (median: 200 vs. 463.8 pg/mL; p < 0.0001; Wilcoxon signed-rank test). (c) Spearman rank correlation between plasma IL-6 levels and galactomannan values in IFI-positive AML patients. No significant association was observed (ρ = −0.02; p = 0.9)

Connection between lactate dehydrogenase (LDH) and fungal infection

As high LDH level is often associated with several disease conditions including inflammation and infection related cellular damage. To determine the association between serum lactate dehydrogenase levels with opportunistic invasive fungal infection in AML patients, we checked the overall serum LDH levels in same set of AML patients with probable fungal infection or in AML patients with no evidence of fungal infection. Serum LDH levels were markedly elevated in presence of probable fungal infection among AML patients (Fig. 4a). Serum LDH levels were analysed in paired samples collected from the same AML patients before and after the onset of fungal infection. Elevated LDH levels were observed in infected patients with AML (median LDH: 462 U/L) compared to uninfected patients with AML (median LDH: 238.5U/L) (Fig. 4b; Table 2) (p = 0.0005). LDH levels also showed a significant positive correlation with galactomannan values during infection (p = 0.003) (Fig. 4c), suggesting that a higher Aspergillus antigen burden may be associated with an increased serum LDH release.

Fig. 4.

Increased LDH levels in AML patients with invasive fungal infection. (a) Serum LDH levels compared among healthy individuals, IFI-negative AML patients, and IFI-positive AML patients using the Mann-Whitney U test. LDH levels were significantly higher in IFI-positive AML patients than in healthy individuals (p = 0.0009), while no significant difference was observed between IFI-negative AML patients and healthy individuals, or between IFI-negative and IFI-positive AML patients. (b) Paired serum LDH levels from the same AML patients at pre-infection baseline and during invasive fungal infection. LDH levels were significantly elevated during the infected state compared to the non-infected state (median: 238.5 vs. 462 U/L; p = 0.0005; Wilcoxon signed-rank test). (c) Spearman rank correlation between serum LDH levels and galactomannan values in IFI-positive AML patients. A significant positive association was observed, whereby higher LDH levels corresponded with higher GM values (ρ = 0.5773; p = 0.003)

The impact of fungal infection and management on the probability of patient’s survival

As invasive fungal infection contributes substantially to morbidity and mortality in AML, the association between infection status and survival was examined using unadjusted Kaplan–Meier analysis. Patients with IFI did not have a statistically significant difference on survival over the follow-up period compared to those without infection (Fig. 5a). The relationship between antifungal treatment and survival was similarly explored using unadjusted Kaplan-Meier curves. Similarly, there was no significant difference in survival between patients treated with voriconazole compared to patients treated with liposomal amphotericin B (Fig. 5b).

Fig. 5.

Impact of Infection and pre-emptive therapy on patient’s survival. (a) Kaplan-Meier survival curves comparing AML patients with invasive fungal infection (+ IFI) and without invasive fungal infection (-IFI), compared using the log-rank (Mantel-Haenszel) test. Patients with IFI had no significant difference in overall survival over the follow-up period (p = 0.6045). (b) Kaplan-Meier survival curves comparing AML patients receiving voriconazole pre-emptive versus liposomal amphotericin B, compared using the log-rank (Mantel-Haenszel) test. The Kaplan-Meier curve demonstrating no statistically significant difference in overall survival between the two treatment groups (p = 0.15)

Discussion

Despite advances in prophylaxis and antimicrobial therapies, acute leukemia patients remain at risk for life threating infections. Invasive fungal infections contribute a significant number of mortalities in patients worldwide especially in India with haematological malignancies during treatment with myelosuppressive chemotherapeutic drugs. For better management of the fungal infection, accurate diagnosis at the initial stage and immediate effective treatment is needed but early and correct diagnosis of fungal infections is difficult even till date. The detection of galactomannan antigen in blood or BAL fluid is the only option to diagnose invasive aspergillosis in AML patients with having fever. But due to various conditions, false positive results of galactomannan sometimes create confusion which results into excessive medical intervention and unnecessary expensive treatment [25]. To combat morbidity and mortality, early diagnosis is a smart option for AML patients especially during induction therapy when they are more vulnerable to severe infection. In this study, we initially checked the GM antigen level as a gold standard in the enrolled patient’s serum at base line and during therapy. We have found in most of the time while patients having infection, had GM level (OD) was more than 0.5 while less than 0.5 in patients without having any infection. But due to false positivity, one non-infected AML patient and one healthy (control) individual had higher GM value.

An association between G6PD deficiency and an increased risk of fungal infections has been observed in patients with acute myeloid leukemia, summarized as ‘less enzyme, more fungus [26], though direct correlation was not been explored. The prominent role of G6PD in generating ROS via NADPH oxidase is well established. In immune responses, NADPH oxidases are crucial, play a vital role in fighting infections with the help of reactive oxygen species. These ROS helps to kill pathogens and regulate inflammation [27]. According to previous study, the expression and activity of G6PD depend on multiple aspects from the transcriptional level to the post-translational modification [28]. The mechanism underlying the G6PD deficiency may include alterations to protein assembly, dimer formation and protein turnover [14, 29]. Interactions with other genes and proteins could also impact on the functions of G6PD. Studies by Mele et al., reported that oncogenes such as PI3k/AKT, Ras, Src, mTORC1, as well as tumor suppressors such as p53 and TAp73, can alter G6PD activity [30, 31]. It has also been reported that the tumor suppressor protein p53 also can inactivate G6PD by binding to G6PD, which prevents the formation of the active dimer and NADPH production [32]. G6PD deficiency affects many cellular immune responses such as impaired inflammasome activation [33]. In addition, G6PD deficiency has been shown to enhance viral infections [34]. Even a high frequency of erythrocyte G6PD deficiency has been reported in patients with viral hepatitis and among patients with leprosy [35]. An increased frequency of G6PD deficiency was also noted among males with pneumococcal and tuberculosis infections [36]. Since the interference with dimerization result in severe G6PD deficiency, the level of NADP⁺ also can alter G6PD enzyme activity [37].

In this study as well, we have seen that during infection, AML patients had lower G6PD level but became normal in the paired sample upon recovery from infection, further suggest that patients with temporary reduced level of G6PD were at high risk of IA. However, whether G6PD reduction precedes infection onset or reflects a consequence of the infectious and inflammatory state requires further investigation. Since more fungal infection found in patients with less G6PD enzyme, the group of patients with lower G6PD level might need a more intensive surveillance strategy with some other reliable markers which will allow early or immediate detection and administration of appropriate medicine to get rid-off fungal infection.

Since G6PD is known to be associated with variable drug response, chemotherapy drugs also can affect G6PD levels in various ways. Specific chemotherapy drugs like ibandronic acid, oxaliplatin, and carboplatin have been shown to inhibit G6PD activity in vitro [38]. The probable cause of temporary reduction of G6PD level in AML patients may be due to chemotherapeutic drugs used in disease management. The fluctuating levels of G6PD were shown in this study could be related with insufficient production of NADPH for achieving an effective respiratory burst or hamper fungicidal activity.

The production of cytokines that exerts regulatory activity in the inflammatory response has a role in the development of IA [39]. On the other hand, the species of Aspergillus have been shown to induce different subsets of T-helper cells which elevate several cytokines [40]. IL-6 is one of them, plays an important role in defending against the infection. In response to fungal infection, IL-6 activates and recruits different sets of immune cells [41]. Since IL-6 gets produced in response to aspergillus infection, a systemic evaluation of IL-6 in the plasma sample of infected vs. non-infected AML patients was assessed where higher IL-6 level was found to be associated with the fungal infection status of AML patients. In comparison to healthy control, IL-6 level was more in the plasma of AML patients and there was a significant difference in the IL-6 level in paired samples before and after infection further confirmed its link with fungal infection. These findings suggest that elevated IL-6 is associated with active fungal infection in AML patients; whether IL-6 has independent predictive value prior to infection onset requires prospective evaluation.

Lactate dehydrogenase enzyme often gets elevated and released into the blood stream upon tissue damage due to disease or infection. It has been reported that the patients with various infections such as disseminated histoplasmosis or tuberculosis had markedly elevated LDH [42]. To correlate the LDH level with infection, we have checked the LDH level along with G6PD and IL6 in AML patients with and without fungal infection. It has been found that overall high LDH levels were associated with AML disease, moreover further elevated LDH has been reported in those patients especially while having Aspergillus infection. Though LDH is not only definitive marker for fungal infection, but still can be helpful in diagnosing fungal infection in combination with other biomarkers like G6PD and IL-6.

In AML patients, IL 6 and LDH cannot distinguish bacterial from fungal or viral infections, as both may be elevated in response to systemic inflammation, including sepsis, pneumonia, and bloodstream infections, and thus are more useful for risk stratification and prognostication than for inadequate to independently discriminate infectious etiology [43].

Generally, infection is a significant contributor towards morbidity and mortality especially in malignancy. AML patients with severe infection may have a higher risk of early mortality in comparison with those who have mild infections or no infectious episodes. Moreover, untreated or delayed treatment of fungal infection can weaken the host immune system, makes individuals more susceptible to other infection. As a preventive measure, prophylactic treatments are sometimes recommended for better management of infection to combat mortality. The reduction of colonization is also important in preventing systemic infections [44].

In AML patients, antifungal prophylaxis during induction varies widely [45]. Voriconazole has been reported to be highly effective in reducing both fungal colonization and superficial fungal infections [46].

Furthermore, in a randomized clinical trial, voriconazole demonstrated a significantly higher treatment success rate than amphotericin B deoxycholate in patients with invasive aspergillosis (52.8% vs. 31.6%) [47]. We have also evaluated survival probability in voriconazole prophylaxis vs. amphotericin β group to see their impact on survival. Our observations suggest that pre-emptive voriconazole may offer an initial advantage over empirical amphotericin B therapy; however, no significant impact on long-term overall survival was observed may be due to smaller sample size.

In this study, several potential sources of confounding and bias should be considered. Antifungal therapy was not randomly assigned; patients treated with voriconazole or amphotericin B differed from untreated patients with respect to disease severity and clinical indication. Furthermore, G6PD, IL-6, and LDH are non-specific biomarkers, may be influenced by multiple factors common in patients with AML undergoing induction chemotherapy, including malignancy burden, haemolysis, hypoxia, and concomitant bacterial or viral infections. Since adjustment for these covariates were not performed, observed associations may be confounded by disease severity along with fungal infection. In addition, the retrospective single-centre design and small sample size (n = 30) limit statistical power. With these considerations in mind, the findings provide a rationale for further investigation and validation in larger, prospective, multicentre cohorts with appropriate covariate adjustment prior to clinical application.

Conclusions

Invasive fungal infections are a serious concern for AML patients, contributes significantly to their mortality due to compromised immune system as an effect of chemotherapy. Initially, the symptoms for IA is nonspecific and clinical signs appear late in the course of infection. Moreover, in absence of single effective markers to diagnose invasive fungal infection at initial stage, G6PD can be used as an indicator for the risk of fungal infection. Along with G6PD, monitoring IL-6 and LDH level can be a useful predictive non-invasive biomarker in conjugation with galactomannan. Prospective studies are required to determine their independent predictive value. Correct prediction and reliable detection of circulating markers for IA development would be helpful in timely incorporation of appropriate antifungal therapy in the vulnerable group to improve survival outcomes.

Limitations

Our study has several limitations. First, this is a retrospective, single-centre study with a relatively small sample size, limiting the ability to establish G6PD as an independent risk factor for invasive fungal infection. In addition, the mechanistic role of G6PD inactivation during fungal infection could not be explored in patient samples and requires further investigation through in vitro studies.

Furthermore, reduced G6PD levels and elevated LDH and IL-6 concentrations are not disease-specific and may be influenced by factors such as malignancy-associated inflammation, tissue damage, hypoxia, haemolysis, chemotherapy, and concurrent infections. Therefore, these parameters cannot serve as standalone predictive biomarkers when assessed individually. However, when interpreted collectively alongside clinical findings, imaging, and established fungal biomarkers, they may provide supportive predictive value for invasive fungal infections.

Finally, due to limited sample availability, additional cytokines associated with infection and immune response could not be evaluated. Further studies investigating the cytokine network involved in opportunistic fungal infections may help improve early prediction and diagnosis of invasive aspergillosis in high-risk patients.

Abbreviations

AML

Acute myeloid leukemia

BAL

Broncho alveolar lavage

ELISA

Enzyme-linked Immunosorbent Assay

GM

Galactomannan

G6PD

Glucose-6-Phosphate Dehydrogenase

HRCT

High-resolution computed tomography

IL-6

Interlukin-6

IFCC

International Federation of Clinical Chemistry and laboratory medicine

IA

Invasive aspergillosis

IFI

Invasive fungal infections

IPA

Invasive pulmonary aspergillosis

LDH

Lactate dehydrogenase

NET

Neutrophil extracellular trap

NADPH

Nicotinamide adenine dinucleotide phosphate

OD

Optical Density

Author contributions

SG designed and monitored experimental protocol, data analysis and drafted the manuscript. PS performed the experiments, data collection and conducted data analysis. CW supported experimental data analyses and edited the manuscript. RKS provided valuable input regarding patient’s management and clinical data correlation. SKS, SB and RKS contributed to data interpretation and manuscript editing. All the authors reviewed, edited and approved the final version of the manuscript.

Funding

This study was funded by intramural project from host institute and Department of Biotechnology (BT/PR41087/Swdn/135/5/2020), India.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethical approval

The Institute Ethics Committee of the All India Institute of Medical Sciences, New Delhi, granted prospective ethical approval for this study (Ref. No.: IEC-163/05.03.2021, RP-28/2021), effective from 30th March 2021. The collection of five millilitres (ml) of peripheral blood from AML patients was also approved. The AIIMS Ethics Committees adhere to international standards, including the ICH-GCP guidelines and the Declaration of Helsinki, as well as national guidelines issued by the Indian Council of Medical Research (ICMR).

Consent to participate

All patients were provided with a participant information Sheet in both English and Hindi (the local language). The study details—including the expected duration, potential benefits, and assurance of confidentiality were thoroughly explained to them. As the study involved no intervention and posed no associated risks, written informed consent was obtained from all willing participants prior to collecting peripheral blood samples (5 ml).

Competing interests

The authors declare no competing interests.