Description of invasive aspergillosis in Mexican patients with chronic granulomatous disease

Chronic granulomatous disease (CGD) is a congenital immune disorder of phagocytic function caused by a defect in nicotinamide adenine dinucleotide phosphate (NADPH) oxidase (1). Patients with CGD lack reactive oxygen species production, which predisposes them to severe bacterial and fungal infections along with various inflammatory manifestations. Genes associated with classical CGD include NADPH oxidase 2 (CYBB), cytochrome B-245 α chain (CYBA), neutrophil cytosolic factor 1 (NCF1), and NCF2. CYBB has an X-linked (XL) inheritance pattern, whereas the other genes are inherited in an autosomal recessive pattern (1). Inherited p40^phox (NCF4) and chaperone EROS (CYBC1) deficiencies indicate an underlying distinctive condition resembling a mild, atypical form of CGD (2, 3).

Among the most common pathogens that infect patients with classical CGD include Staphylococcus aureus, Aspergillus spp., and Burkholderia spp. (4). Aspergillus infections have only been described in patients with pathogenic CYBC1 variants and not in those with NCF4 variants (2, 5, 6). Invasive Aspergillus infections are the most common fungal infections associated with CGD (4, 7, 8, 9, 10). Moreover, Aspergillus fumigatus is the most common pathogenic species, followed by Aspergillus nidulans (11). The clinical presentation of invasive aspergillosis (IA) in CGD is often indolent with minimal symptoms. Therefore, it is important to maintain a high index of clinical suspicion (8). The primary sites of IA in most patients are the lungs, followed by bones (8). These patients require prolonged and even multiple antifungal treatments (12). Stem cell transplantation (SCT) is a successful treatment for CDG and is an alternative for refractory IA (13). However, information on aspergillosis infections in patients with CGD is limited in Latin America (14, 15). Therefore, we aimed to describe the clinical, microbiological, and therapeutic characteristics of patients with CGD and IA in a Latin American country.

The current study included 45 patients with a genetic diagnosis of CGD recruited between 2005 and 2024 from seven public hospitals in four Mexican cities (Mexico City, Monterrey, Tlaxcala, and Puebla). Tirthy (40%) of the 75 patients in this cohort were not included because they did not have aspergillosis at the time of the study. All subjects in this study were born in Mexico and continued to live there. The subjects were recruited from 44 unrelated families hailing from 15 of the 32 states in Mexico. All patients had infections other than IA and inflammatory manifestations (granulomas in different organs, autoimmunity, macrophage activation syndrome, inflammatory bowel disease, or Kawasaki disease). Three patients exhibited additional risk factors, including exposure to mulch (P7), visits to caves (P7), and residence with his mother in a jail cell with unsanitary conditions (P31). Thirthy-nine patients (87%) included in this study were male. The median age at the first manifestation of CGD was 3 mo (1–6.5); only one patient (P31) presented with IA as the first manifestation of CGD. The median age at CGD genetic diagnosis was 25 mo (9–66.5). The median time between the first manifestation of CGD and the first episode of IA was 60 mo (26.5 days–126 mo). The median age of all patients at last follow-up was 133 mo (73–200.5), and the median age of patients alive at the time of the study was 154.5 mo (111–238.2). The genotypes identified were CYBB (XL) in 33 cases (73%) and recessive genes in 12 cases (27%), with the following distribution: NCF2 in six, CYBA in three, and NCF1 in three patients. Regarding prophylaxis, 44 patients (97%) received itraconazole and trimethoprim with sulfamethoxazole, whereas 39 patients (86%) received additionally recombinant interferon gamma. Adherence to these treatments remains unknown (Table 1). The data on 45 patients have been previously published (4, 16).

Of the 44 families of the 45 patients, one family had two affected members with IA. The median age at the first IA event in all 45 patients was 65 mo (30.5–140). A total of 79 IA events were recorded in 45 patients. The difference in the median age at the first IA event between the 33 patients with XL inheritance (64 mo [30.5–140]) and the 12 patients with AR inheritance (85.5 mo [25.2–166.7]) was not significant (p = 0.586). Thirteen patients (29%) had at least one IA event before CGD diagnosis, median period time between the first IA event and the CGD diagnosis was 1 mo (1, 2, 3), and 32 patients (71%) had one or more IA events after CGD diagnosis, median period time between the CGD diagnosis and the first IA event was 46.5 mo (13.2–85.5). Of the 45 patients, 26 (58%) had one IA event, nine (20%) had two, five (11%) had three, and five (11%) had four IA events. The median time between the first and second events was 35 mo (15–65), that between the second and third events was 26 mo (17.5–32.5), and that between the third and fourth events was 22 mo (15.5–43.5) (Fig. 1).

Based on the European Organization for Research and Treatment of Cancer/the Mycoses Study Group Education and Research Consortium (EORTC/MSGERC) criteria (17), from the 79 events, 13 (17%) IA events were proven, 35 (44%) were probable, and 31 (39%) were possible (Table 1). The mortality rate was 57.7% for proven events, 23.1% for probable events, and 19.2% for possible events (p= 0.038). The median time between the onset of symptoms and the isolation of Aspergillus spp. was 35 days (19.5–55.5). Aspergillus spp. was detected in 30 events, 13 of which were Aspergillus fumigatus and the other 17 included Aspergillus versicolor (n = 7), Aspergillus terreus (n = 2), Aspergillus nidulans (n = 2), Aspergillus flavus (n = 4), Aspergillus ustus (n = 1), and Aspergillus niger (n = 1) (Tables 2 and 3). The proportion of mortality was not significantly different between patients infected with one Aspergillus species (48%) and those infected with ≥2 Aspergillus species (60%) (p = 0.5). The median time from symptom onset to Aspergillus antigen (galactomannan) positivity was 21 days (9.7–59.7) among the 35 probable events. The 79 IA events affected the following organs: lungs (n = 59); lungs and bones (n = 3); lungs and spleen (n = 3); lungs and central nervous system (CNS; n = 3); lungs, skin, and CNS (n = 2); bones (n = 1); skin (n = 1); lungs, and skin (n = 1); lungs and liver (n = 1); lungs and pericardium (n = 1); lungs, livers, and spleen (n = 1); lungs, bones, and skin (n = 1); lungs, liver, spleen, and kidneys (n = 1); and lungs, liver, spleen, and CNS (n = 1) (Fig. 2). The proportion of mortality was not significantly different between patients with only pulmonary IA (32.2%) and IA pulmonary and one or more IA organs affected (35%) (p = 0.818). Co-infection with bacteria and fungi was not uncommon; the Table 1 shows the different microorganisms that co-infected each patient in each of the IA events (Table 1).

Fourty-four (97%) patients with 78 events received antifungal therapy. The median time between hospital admission and initiation of antifungal treatment was 13.5 days (7–30). One patient (P32) did not receive treatment because IA was diagnosed postmortem. All 44 patients received one or more of the following antifungal agents: voriconazole, amphotericin B, caspofungin, posaconazole, itraconazole, or anidulafungin. Of the 78 IA events, 43 (55%) were treated with monotherapy, 16 (20.5%) with dual therapy, 15 (19.2%) with triple therapy, three (3.8%) with quadruple therapy, and one (1.5%) with quintuple therapy (Table 1). Patients who received more than two antifungal agents per event had a higher rate of mortality (48.6%) compared with patients who received only one antifungal agent (20.5%) (p = 0.008). We compared the number of days of antifungal treatment in the group of patients who died and those who survived; we found no statistically significant differences. Receiver-operating characteristic curve (ROC) curve analysis was performed to relate the number of treatment days to mortality, resulting in a cutoff point of <75 treatment days. Comparing mortality with this cutoff point, 36% of those receiving <75 days of treatment died versus 16.7% of those receiving >75 days (p = 0.05). Among patients coinfected with tuberculosis and aspergillosis there was no statistically significant difference (p = 0.516) in the proportion of deaths in patients receiving voriconazole plus antimycobacterial drugs, including rifampicin (27.3%) versus voriconazole plus other antimycobacterial drugs without rifampicin (40%). In 51 (65%) of the 78 events, parallel therapy with corticosteroids (methylprednisolone, prednisone, dexamethasone, or hydrocortisone) was administered; however, information on the exact indication is unavailable.

Among the described patients, 12 (27%) underwent at least one SCT, and the median age at the time of transplantation was 57.5 mo (19.2–158.2). In the cohort under investigation, nine patient received reduced intensity and three myeloablative conditioning. Among the 12 transplanted patients, those receiving complete myeloablative conditioning reported no deaths and those receiving reduced intensity six deaths (66.7%) (p = 0.091). Four out of 12 patients experienced an IA event prior to undergoing SCT. Of these patients, three underwent successful grafting, while one patient died owing to septic shock (microorganism undetected) (Fig. 3). The remaining eight patients were not evaluated purposely for aspergillosis before the procedure. Five out of the eight cases died; one as a result of septic shock (microorganism undetected) and four as a result of IA during the procedure, while the remaining three had a successful graft. In the four patients with IA after SCT, the median time between the procedure and the diagnosis of IA was 21 days (7–145).

Twenty-six of the 45 patients died, 18 (69%) from IA and the remaining from other causes (Table 4). Sixteen (89%) of the deceased patients had XL inheritance, while the rest had AR inheritance (p = 0.054). The overall survival of the 45 patients was 80.8% at 64 mo and 67.9% at 120 mo, with a median survival of 195 mo. The median survival time for patients with XL-CGD was 167 mo, with 20 deaths, and for patients with AR-CGD it was more than 313 mo, with 6 deaths (p = 0.07). When survival analysis of the AR-CGD subgroups was performed, no significant differences were observed (p = 0.254), median survival for the p22^phox group was 201 mo with three deaths, for the p47^phox group 358 mo with one death, and finally, for the p67^phox group 184 mo with two deaths (Fig. 4). The median age at death was 107.5 mo (44.7–196), and the median time between the first IA event and death was 6 mo (1–15). Aspergillus spp. was isolated from 10 (55%) of the 18 patients who died of IA; the species are listed in Table 1.

This is the first study on patients with CGD and IA in a Latin American country. The 45 patients included in this study were treated in public hospitals, with the majority coming from low socioeconomic backgrounds. All patients in the present study had pathogenic variants of the four genes involved in classical CGD (CYBB, CYBA, NCF1, and NCF2) (18). In a cohort of 24 North American patients with CGD receiving prophylactic itraconazole, the first event of invasive fungal infection occurred at the age of 10 years compared with 4 years in patients that did not receive itraconazole (12). Most of the patients described here experienced their first IA event before the age of 5 years, and although they were prescribed prophylactic itraconazole, we have no knowledge of their treatment adherence and compliance. Routine monitoring of itraconazole plasma levels (target level of >250 μg/ml for prophylaxis) is mandatory as a method to assess adherence and compliance; levels must be monitored every 3 mo in patients with CGD that are not transplanted (19).

Each event was classified according to the EORTC/MSGERC criteria, with 13 events classified as proven, 35 as probable, and 31 as possible. The proven events had a higher mortality; a possible explanation is that the other two groups had a lower infectious burden or were actually free of aspergillosis. Similar to other studies, we found that in few patients with proven IA events, the previous results for serum galactomannan antigen were negative (12, 13). Blumental et al. reported that the lack of sensitivity can be attributed to the absence of angioinvasion by fungal hyphae observed during IA in patients with CGD, in contrast to neutropenic subjects (12). In addition, fungal elements have been identified in lung biopsies with no growth in culture (20). Therefore, a negative test result in patients with CGD does not rule out the possibility of aspergillosis. Similar to in other patient groups, A. fumigatus was the most common species to infect the 45 Mexican patients (7, 21, 22, 23, 24, 25, 26). Other species identified in our study included A. nidulans, A. terreus, A. versicolor, A. flavus, A. ustus, and A. niger. Interestingly, there was no increased mortality in patients infected with two species compared with those infected with one species. To the best of our knowledge, A. ustus (P14) has not been previously described in patients with CGD. A. ustus has been found on the surfaces of the walls of caves, in the indoor air of buildings, including hospitals, and in soils and bat dung (27). A comparison of our results with other descriptions of CGD and IA confirmed that Aspergillus primarily affected the lungs, followed by other anatomical sites, such as the bones, CNS, skin, and spleen (7, 10, 21, 25, 28). In this study, we found no increased mortality in patients with pulmonary IA versus those with pulmonary and other organ involvement. Pericardial involvement of Aspergillus was found in one patient (P43). A similar case was reported by Abd Elaziz et al. (25). Furthermore, one patient (P3) had IA involving the kidneys, which is the first reported case in literature involving this organ. We found that after the first IA event, subsequent aspergillosis events occurred, and the time interval between the following IA events decreased (Fig. 1).

According to Kline et al., single-agent therapy for IA often provides unsatisfactory results; however, the benefits of combined therapy are debated (13). In the 78 patients, voriconazole monotherapy was the first-line treatment for most patients. However, several patients experienced subsequent IA events. In this study, patients who received more than two antifungals per event had a higher mortality rate than those who received a single antifungal; one explanation could be that by the time an additional antifungal was administered, the aspergillosis had spread and was refractory despite polytherapy. Combination antifungal therapy starting from the first episode of aspergillosis may be effective in preventing subsequent episodes. Among the 45 patients who received <75 days of treatment versus those who received >75 days had higher mortality. The neonatal patient (P31) was administered voriconazole monotherapy for 28 days with apparent improvement; however, a few months later, he presented with a second fatal event. Prolonged administration of multiple antifungals is necessary to achieve remission of infection (12). Blumental et al. suggested ∼446 days of treatment for cure (12).

Patients with inborn errors of immunity are at risk of exposure to endemic microorganisms depending on their geographical region of residence. Mycobacterial infections are widespread among patients with CGD in Latin America (13). These patients are often co-infected with fungi and mycobacteria in tuberculosis endemic areas, which require treatment using antifungals in addition to antituberculosis drugs (Table 1) (29). In such cases, it is crucial to note that combining rifampicin with voriconazole treatment leads to a significant reduction in the plasma concentration of the antifungal drug (30), thereby compromising treatment efficacy. Alternative medications should be considered in such cases. Aspergillus infection triggers an abnormal immune response in patients with CGD. This response is characterized by hyperinflammation leading to tissue damage and ineffective fungal control (31). Deffert et al. showed that when a preparation containing the A. fumigatus cell wall is introduced, a severe and persistent inflammatory response is induced in a murine knockout model that lacks NADPH oxidase (32). Studies in humans and mice have conclusively demonstrated that infection with A. nidulans and A. fumigatus induces a hyper-inflammatory response driven by the excessive production of interleukin-1 (31). The treatment for inflammatory complications in patients with CGD and IA includes corticosteroids and other immunosuppressants (31, 33). The 35% of the patients described in our study received steroids; some were part of an anti-inflammatory treatment in conjunction with antimicrobials.

The access to transplants is only available to a minority of patients in Latin America (15). Apart from cost, other limiting factors include lack of expertise in performing transplants on patients with CGD. Among the patients in the present study, 12 underwent SCT, and 50% died during the procedure, five secondary to IA. Countries such as Mexico should take into account the experience gained from other groups: transplantation in CGD should be strongly considered at a younger age and especially if a well-matched donor is available (34, 35). In high-risk patients with CGD, reduced intensity conditioning SCT regimen is safe and effective (36, 37). All patients with CGD should be screened for aspergillosis prior to SCT. Since serum ß-d-glucan and serum galactomannan provide unreliable results, a negative result does not mean that Aspergillus spp. infection does not exist; radiological imaging is a relatively more effective indicator of the disease extent (8, 13, 37, 38, 39). Additionally, it is crucial to consider antifungal therapy before and after transplantation regardless of whether the patient is infected by Aspergillus (38, 40). Multidrug antifungal therapy is essential for successful transplant outcomes in patients with CGD and fungal infections (13). IA was responsible for an overwhelming majority (69%) of all deaths in this study group of 45 patients. This rate is higher than those reported in other international cohorts (7, 9, 10, 22, 26, 28, 41).

Finally, the early treatment of IA is crucial, which includes empirical combined and aggressive treatments, introduction of new antifungal drugs, measurement of therapeutic levels, and implementation of SCT. These measures will reduce morbidity and mortality in countries with a situation similar to Mexico.

Patients with CGD (genetic diagnosis) and IA in Mexico (from seven different hospitals) between January 2005 and September 2024 were included. The study protocol was approved by the Research Committee of the National Institute of Pediatrics (019/2011). Since this was a retrospective study of chart information, informed consent was not required. Attending physicians were contacted and invited to participate by completing a questionnaire on demographic, clinical, microbiological, and therapeutic information related to the IA event(s) of each patient with CGD. Each physician reviewed the medical records of the patients and recorded the requested information. Questionnaires were received electronically. Data were entered into an Excel spreadsheet.

Descriptive statistics were used for the statistical analysis. Data are expressed according to distribution, categorical variables with proportions, and numerical variables as medians (interquartile ranges are in parentheses). Differences between groups were compared using Fisher’s exact test and Mann–Whitney U test. Survival was analyzed by the Kaplan–Meier method, with log-rank test performed for comparisons values of p < 0.05 were considered significant. All statistical analyses were performed using Statistical Package for Social Science (SPSS) software version 25.0 (SPSS, Inc.).

Each IA event was classified as possible, probable, or proven according to the criteria established by the EORTC/MSGERC (17). Patients were considered to have a proven IA event if they satisfied at least one of the following mycological criteria: (1) histopathological, cytopathological, or direct microscopic examination of a specimen obtained by needle aspiration or biopsy in which hyphae specific to Aspergillus spp. were observed accompanied by evidence of associated tissue damage (with necessary confirmation by means of culture or polymerase chain reaction). (2) Recovery of Aspergillus spp. by culture of a specimen obtained using a sterile procedure from a normally sterile site and a clinically or radiologically abnormal site consistent with an infectious disease process, while excluding bronchoalveolar lavage fluid, cranial sinus cavity specimen, and urine. Probable IA requires the presence of a host factor, a clinical criterion, and a mycological criterion. Cases that met the criteria for host factors and clinical criteria, but not for mycological criteria, were considered as possible IA (17).

The first IA event was defined as the first Aspergillus infection. Subsequent events were defined based on the presence of new clinical/radiological data with or without new mycological evidence (direct or indirect) after the completion of antifungal treatment and evidence of clinical/radiological improvement for the first episode (11, 17, 42).

Data available upon request.

We would like to thank FUMENI A.C. for the support provided for the professional writing services in this project.

Author contributions: T. Cova-Guzmán: data curation, investigation, methodology, software, and writing—original draft. D. Palacios Reyes: data curation, investigation, and writing—original draft. C.M. Román-Montes: data curation and writing—original draft. M.A. Yamazaki-Nakashimada: data curation, resources, and visualization. A.T. Staines Boone: conceptualization, investigation, methodology, supervision, validation. L. Silva-Goytia: resources. M.L. García-Cruz: investigation. H. Gómez-Tello: investigation. U. Pérez-Blanco: methodology. N. Jiménez-Polvo: investigation. E. Mamani Velasquez: investigation. N. Ramírez Uribe: resources and visualization. I. Medina Vera: methodology. S. Espinosa Padilla: resources, supervision, validation, visualization, and writing—original draft, review, and editing. L. Blancas-Galicia: conceptualization, methodology, supervision, and writing—review and editing.