Clinical features, genotypes, and geographic distribution of 238 Latin American CGD patients

Chronic granulomatous disease (CGD) is a heterogeneous inborn error of immunity (IEI) caused by defects in the nicotinamide adenine dinucleotide phosphate (NADPH) oxidase complex of phagocytes, leading to varying levels of impairment in the oxidative burst in response to stimuli, a relevant mechanism for effective microbial killing (1). Due to this defect, CGD is characterized by excessive inflammation and recurrent and severe infections caused mainly by extracellular pyogenic bacteria, intracellular bacteria and fungi (particularly catalase producers), challenging inflammatory manifestations, and autoimmunity (2).

Phagocyte NADPH oxidase is composed of five main subunits: Two proteins anchored to the plasma membrane or phagosomal/phagolysosomal membrane that form the cytochrome b₅₅₈, and three cytoplasmic proteins that migrate and anchor to cytochrome b₅₅₈ to form the NADPH oxidase complex upon stimulation (2). CGD can occur in the X-linked (XL) form when pathogenics variants affect the CYBB gene (OMIM# 306400), (Xp21.1), which encodes the 91-kDa β subunit of cytochrome b₅₅₈, gp91^phox, and in the autosomal recessive (AR) form when mutations occur in the CYBA (16q24), NCF1 (7q11.23), NCF2 (1q25), and NCF4 (22q13.1) genes (the latter associated with p40^phox deficiency, a CGD-like but distinct phagocyte disorder), and the more recently identified CYBC1 (17q25.3) gene (1, 2), (OMIM# 233690, 233700, 233710, 613960, 618935, respectively), which encode p22^phox (α subunit of cytochrome b₅₅₈), p47^phox, p67^phox, p40^phox, and EROS, respectively (3, 4, 5, 6, 7).

The main manifestations of CGD include pneumonia (PNM), lymphadenitis, skin infections with or without abscesses, deep abscesses (especially hepatic), hepatosplenomegaly, osteomyelitis, and infectious or inflammatory gastroenteropathies. The latter, in a state of hyperinflammation, can also present as chronic colitis with the formation of obstructive granulomas (also observed in the genitourinary tract), making the gastrointestinal tract the most affected by inflammatory processes (8, 9, 10, 11, 12, 13, 14, 15, 16). The most commonly isolated pathogens in CGD include Staphylococcus species, especially Staphylococcus aureus, Burkholderia spp., Serratia spp., Nocardia (more common in temperate countries), Aspergillus spp., Mycobacterium tuberculosis, and Bacillus Calmette-Guérin (BCG) infections, in countries where tuberculosis (TB) is endemic and BCG vaccination is mandatory (8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20). Treatment involves using antibiotics and antifungals for active infections and corticosteroids for challenging inflammatory processes. However, prophylaxis with cotrimoxazole plus azole antifungals provides a better quality of life and survival for these patients, sometimes also benefiting from the use of interferon-gamma (IFN-γ) (21, 22). The only curative therapy currently is hematopoietic stem cell transplantation (HSCT) (23, 24, 25).

The estimated incidence of CGD is around 1:200,000–1:250,000 live births in the United States (9). Still, it varies considerably across different geographic regions: 1:450,000 in Sweden, 1:300,000 in Japan, 1:218,000 among Israeli Jews, and the highest estimated incidence among Israeli Arabs at 1:100,000 (11, 26). Generally, XL-CGD accounts for about 60–70% of cases, followed by p47^phox defects in up to 30% of cases. However, the AR form may occur at a frequency similar to or even higher than the XL form in countries with high consanguinity rates (11, 26). The true incidence of CGD in Latin American countries remains largely unknown.

Studies on CGD in Latin America are mostly confined to case reports and clinical studies with few patients, rarely exceeding a dozen (27), with few more representative clinical-epidemiological studies (8, 13, 28). As the literature shows, the clinical characteristics of CGD patients in developing countries differ from those classically reported in patients from developed countries. Therefore, clinical-epidemiological studies are relevant to better understanding CGD patients in this region.

This study summarizes the clinical, epidemiological, and genetic characteristics of 238 patients with CGD from eight Latin American countries, making it the most extensive clinical-epidemiological study on CGD in Latin America to date.

A total of 238 patients from 210 unrelated families (109 familial cases, 22 consanguineous families) at 53 pediatric healthcare centers in Mexico (n = 118; 50%), Brazil (n = 96; 40%), Chile (n = 6; 2%), Costa Rica (n = 5; 2%), Argentina (n = 4; 2%), Paraguay (n = 4; 2%), Peru (n = 3; 1%), and Uruguay (n = 2; 1%) were included in the study. Among the 238 patients, 196 were male and 42 were female. Of the male patients, 107 had confirmed XL-CGD and 12 had AR-CGD. Among the females, 20 had AR-CGD and two were likely XL-CGD carriers with clinical manifestations, possibly due to skewed X-inactivation. The remaining patients had undefined CGD subtype (unknown genotype [UG]-CGD). The patients were diagnosed between 1976 and 2021, with 229 diagnosed by dihydrorhodamine-1,2,3 (DHR) (including confirmations by nitroblue tetrazolium [NBT]), eight by NBT, and one with a genetic diagnosis of XL-CGD and a carrier mother (P170 and P145, respectively). Table 1 summarizes the demographic characteristics of the patients, and more detailed individual descriptions are available in Table 2 and Table S1. The median age at presentation of the first clinical manifestation of CGD was 5 mo (range: 0.1–336 mo). The median age at diagnosis was 2 years and 2 mo (range: 0.3–366 mo), with a median diagnosis delay of 1 year and 2 mo (range: 0.1–232 mo). 173 patients (73%) were diagnosed before the age of 6 years. For statistical purposes, female patients with UGs, all with AR-CGD phenotype, were included in the AR-CGD group, forming the AR-CGD + UG female (UGfem) group with 52 individuals, allowing the following analyses. The median age at onset of clinical manifestations for patients with XL-CGD was 3 mo (range, 0.1–336 mo), which was earlier than the median age for patients with AR-CGD, at 8.5 mo (range: 0.25–204 mo), and for the AR-CGD + UGfem group, at 9 mo (range: 0.25–204 mo) (P < 0.001). The date or age of the first clinical manifestation was obtained for 225 patients. A similar pattern was observed in the median age at diagnosis: patients with XL-CGD were diagnosed at a median age of 23 mo (range: 0.6–366 mo), compared to 60 mo (range: 1–235 mo) for AR-CGD and 67 mo (range: 1–250 mo) for the AR-CGD + UGfem group (P < 0.001). Date or age at diagnosis was available for 229 patients. Again, the median diagnosis delay for XL-CGD patients was shorter, at 13 mo (range: 0.1–186 mo), compared to 37 mo (range: 0.1–227 mo) for AR-CGD and 40 mo (range: 0.1–232 mo) for the AR-CGD + UGfem group (P < 0.001) (Fig. 1).

Brazil and Mexico account for 90% of the patients described here, and the regional distribution of the patients is broad, unlike in other countries, where patients are concentrated in more developed regions. In Mexico, the regional origins were as follows: Northwest (n = 8), Northeast (n = 8), West (n = 11), East (n = 9), Central-North (n = 7), Central-South (n = 63), Southwest (n = 3), Southeast (n = 7), and unknown (n = 2), while in Brazil, the distribution was as follows: North (n = 3), Northeast (n = 19), Central-West (n = 5), Federal District (n = 3), Southeast (n = 45), South (n = 12), and unknown (n = 9). The regions with the most patients are also where the main medical and research centers for IEI are located. In Mexico (Central-South, n = 63), this is the National Institute of Pediatrics. In contrast, in Brazil (Southeast, n = 45), it is the Federal Universities of São Paulo, Rio de Janeiro, and the University of São Paulo (USP) (Fig. 2).

The genetic diagnosis was obtained for 141 patients (59%) from 122 families, with the majority having XL-CGD (n = 109, 77%). 11 novel pathogenic variants were identified, including nine in the CYBB gene and two in NCF2, all predicted to be deleterious according to Combined Annotation Dependent Depletion (CADD), SIFT, MutationTaster, and POLYPHEN-2 (Table 1). Patients from Mexico comprised 64% of those with genetic analysis (n = 91, 77% of all Mexican patients), while Brazil had only 35 patients genetically diagnosed (36.5%). Argentina, Peru, Chile, Costa Rica, Uruguay, and Paraguay had, respectively, four (100%), three (100%), six (67%), five (60%), one (50%), and no patients genetically diagnosed. For several reasons, including lack of genetic material or technical issues, genetic diagnosis was not possible for 97 patients from 88 families. Overall, 109 cases (77%) had XL-CGD, and 32 (22%) had AR-CGD, with 92 different mutations identified. No gene hot spots were found, with exonic regions most frequently affected (n = 122), followed by splicing sites (intron and exon) (n = 13), large deletions (≥1 exon) (n = 8), and a single promoter deletion case with no gp91^phox expression (P206) (Table 2 and Table S1). Pathogenic variants in CYBB, CYBA, and NCF2 were heterogeneous (Fig. 3). Nonsense mutations (n = 43) were most frequent in CYBB, followed by missense mutations (n = 26), deletions (n = 18), splice sites (n = 11), copy number variation with large deletions covering the CYBB region (≥1 exon) (n = 8), insertions, and indels (n = 3). Patient P150 was the only one with McLeod syndrome.

In CYBA and NCF2, deletions were most common (n = 6), followed by missense (n = 3), nonsense, splicing sites, deletion/nonsense, deletion/missense, and splicing site/missense (each in 1 patient). There were four compound heterozygous cases: three in CYBA (P162, P175, and P192) and one in NCF2 (P60) (Fig. 3). All 18 patients with p47^phox deficiency from 15 families had the same homozygous mutation, ΔGT (c.75_76delGT), in NCF1. No pathogenic variants in NCF4 (not typical CGD), CYBC1, or RAC were identified in this study. Data on neutrophil NADPH oxidase expression were obtained through mutation databases, scientific articles, or in silico analysis: 63 patients were X91⁰ (undetectable), 9 were X91⁻ (low), three were X91⁺, and 28 were X91^? (the level of protein expression was not determined). Among the X91⁰ patients, one had a de novo mutation (P146), and one had McLeod syndrome (P150). Two symptomatic carrier mothers (P145 and P205), unrelated, had the same pathogenic variant in exon 7 of CYBB, c.676C>T, leading to abolished gp91^phox expression—likely cases of skewed X-inactivation (Table 2 and Table S1). Among the AR-CGD cases, 27 patients were homozygous: five were A22⁰, three were A67⁰, 18 were A47⁰, and one was A67^? (P167). In cases of compound heterozygosity, mutations c.55_63delAAGAAGGAC (P162 and P175) and c.366+1G>A (P192) led to an absence of p67^phox (A67⁰), though the respective allele mutations could not determine protein expression (A67^?). For compound heterozygous CYBA (P60: c.472_484del/c.399delC), protein expression could not be determined (A22^?) (Table 2 and Table S1).

The first clinical manifestation of CGD occurred within the first 2 years of life in 83% (n = 188) of our patients, with some cases presenting multiple simultaneous manifestations. Infectious manifestations were predominant, affecting 196 patients (88%), followed by gastroenteropathy, typically expressed as diarrhea (n = 16, 7%), with no clear distinction between infectious or inflammatory causes, nonspecific persistent fever (n = 9), seizures (n = 1), and failure to thrive (n = 1). The relative frequency of the first clinical manifestations, categorized by genotype/phenotype, can be seen in Fig. 4 A. PNM was the most common initial clinical manifestation (n = 57), followed by adverse reactions to the BCG vaccine (n = 48), skin infections (n = 32), lymphadenopathy (n = 17), gastroenteropathy (n = 16), and sepsis (n = 9). Analyzing infections by anatomical site, we observed that the lungs (27%) were the most frequently affected, followed by the skin (17%), lymph nodes (8%), and intestines (7%). The sites of BCG infection (21%) were multiple and are described in detail in the section “BCG infectious disease and TB in CGD patients” and in Fig. 6. Although there was no statistical difference, PNM was more frequent among patients with XL-CGD, as were liver abscesses. Conversely, skin infections were more frequent in patients with AR-CGD and in the AR-CGD + UGfem group, manifesting as pyoderma, pustulosis, impetigo, dermatitis, cellulitis, omphalitis, furunculosis, and other abscesses.

Infectious manifestations occurred in all patients, with PNM being the most frequent, affecting 80% of patients (n = 191), with an average of 3.8 episodes per patient (range, 1–25 episodes), showing no significant difference between groups (XL, AR, UG, and AR + UGfem), as shown in Fig. 4 B. Recurrent upper respiratory tract infections (URTIs) were less common than PNM (n = 86, 36.1%), with an average of nearly four episodes per patient (range, 1–20). Otitis media was the most common (n = 32), followed by tonsillitis/pharyngitis (n = 24) and sinusitis (n = 12). Among the pathogens identified in patients with PNM, the M. tuberculosis complex was the most frequent (17%): Mycobacterium spp. (n = 14), M. tuberculosis (n = 8), and Mycobacterium bovis/BCG (n = 11), followed by Klebsiella pneumoniae (n = 8), Burkholderia cepacia (n = 8), Serratia marcescens (n = 7), S. aureus (n = 6), and Pseudomonas aeruginosa (n = 5), with others identified in one to four patients. Regarding fungal infections, Aspergillus species were predominant, confirmed in 47 patients (25% of patients with PNM and 76% of patients with fungal PNM); however, species identification, such as Aspergillus fumigatus, was made in only five patients. Fig. 5 describes the main microorganisms isolated by infection site. In URTIs, Streptococcus pneumoniae and Streptococcus milleri were isolated from one patient each, as well as Staphylococcus spp., S. aureus, and Pseudomonas fluorescens. Among fungi, A. fumigatus, Candida spp., and Pneumocystis jirovecii were isolated from one patient each.

The second most frequent clinical manifestation was lymphadenopathy, affecting 63% of patients (n = 149), with an average of 3.3 episodes per patient (range, 1–16 episodes). The genus Staphylococcus was the most common (16.5%), except in patients with lymphadenopathy due to adverse reactions to the BCG vaccine (n = 42, 46% of bacterial lymphadenitis and 93% of patients with lymphadenopathy caused by mycobacteria) (Fig. 4 B and Fig. 5). Skin infections occurred in 132 patients (55.5%), with an average of five episodes per patient (range, 1–37 episodes), with subcutaneous abscesses being the most common manifestation (50%), primarily due to S. aureus (n = 16, 12%). Patients with AR-CGD and the AR + UGfem group experienced more skin infections than those with XL-CGD (P = 0.032/P = 0.01), although the infections were more frequent in the XL group with 233 episodes compared to 158 in the AR form. Other manifestations occurred in 2.3–8.3% of patients, including pyoderma (caused by S. aureus and K. pneumoniae), furunculosis (caused by Staphylococcus spp. and K. pneumoniae), fistulas, cellulitis (caused by Staphylococcus spp., S. marcescens, and Chromobacterium violaceum), and impetigo (caused by Staphylococcus spp., Serratia spp., and S. aureus) (Fig. 4 B and Fig. 5).

One hundred and twenty-three patients (52%) had gastroenteropathy, with an average of 5 episodes per patient (range, 1–25 episodes). Most cases were confirmed or inferred to be infectious; however, only 28% of patients (n = 35) had confirmation of the etiological agent in at least one of the episodes. Of these, the majority were bacterial (80%), with the genus Salmonella being the most common (n = 16, 13%), affecting 46% of patients with isolated microorganisms. Chronic diarrhea (lasting >4 wk) was present in 31 patients, followed by hematochezia (n = 8), stomatitis/esophagitis/duodenitis/appendicitis (n = 7), and intestinal obstruction due to granulomatous inflammation (n = 3). Other events were associated with allergies (n = 3) or food intolerance (n = 1). Inflammatory bowel disease (IBD) was diagnosed in 31 patients (13%). It was more common among Brazilians (n = 21), especially those with XL-CGD (n = 7), accounting for 29% of XL-CGD patients in the country (Fig. 4 B and Fig. 5). Perianal lesions were identified in 26 patients (11%), with an average of 3.1 episodes per patient (range, 1–24), the most frequent being perianal abscesses (n = 25), followed by fistulas (n = 3) and granulomas (n = 1). Only four patients (15%) had isolated microorganisms, all bacterial: K. pneumoniae (n = 2), Citrobacter freundii (n = 1), and S. aureus (n = 1). Nearly half of the patients experienced sepsis (n = 119), with an average of 1.5 episodes per patient (range, 1–8 episodes), being more recurrent in patients with XL-CGD, 1.6 times more than those with AR-CGD or the AR + UGfem group (P < 0.02). Salmonella (n = 18) was the most frequent pathogen associated with sepsis, followed by Staphylococcus spp. (n = 14), primarily S. aureus (n = 10), Serratia spp. (n = 7), Klebsiella spp. (n = 5), and Burkholderia spp. (n = 6). Aspergillus spp. was the principal fungus associated with sepsis (n = 16), followed by Candida spp. (n = 6), and one case associated with Mucor spp. infection (Fig. 5). Septic shock was the leading cause of death in this study (n = 36), accounting for 45.5% of total deaths.

Deep abscesses occurred in at least five different organs, affecting 61 patients (26%), with an average of 1.7 episodes per patient (range: 1–8 episodes). Hepatic abscesses were the most frequent (n = 45, 74%), with two patients experiencing recurrent hepatic abscesses (P124, n = 6 and P196, n = 8), followed by pulmonary abscesses (n = 11), cerebral abscesses (n = 10), splenic abscesses (n = 3), and intestinal abscesses (n = 2). S. aureus (n = 11) and M. bovis (n = 9) were the primary etiological agents, followed by S. marcescens (n = 3) and Aspergillus spp. (n = 3). Hepatosplenomegaly was clinically diagnosed in 30 patients (16%), with an average of 1.5 episodes per patient (range: 1–9 episodes), and was recurrent in only two patients (P63, n = 9 and P196, n = 8). Eight patients had only hepatomegaly (one episode each), and three had only splenomegaly. 10 patients had hepatosplenomegaly concomitant with a hepatic abscess, one with a splenic abscess (P192), and one with both hepatic and splenic abscesses (P196). Urinary tract infections (UTIs) occurred in 58 patients (24%), with an average of two episodes per patient (range, 1–12 episodes), and were more frequent in non-Brazilian patients (P = 0.049), particularly among Mexican patients (P = 0.008). K. pneumoniae and Klebsiella oxytoca (n = 5), Escherichia coli (n = 4), and Salmonella enterica (n = 2) were the main microorganisms isolated.

Osteomyelitis was diagnosed in 45 patients (19%), with an average of 1.3 episodes per patient (range, 1–4 episodes), and was more frequent in non-Brazilian patients (P = 0.038). Serratia spp. (n = 7), M. bovis (BCG dissemination), Aspergillus spp. (n = 4), and A. fumigatus (n = 3) were the most common, with only one case of osteomyelitis caused by Scedosporium spp. (P227) (Fig. 5). Infections of the meninges affected 20 patients (8.4%), with an average of 1.25 episodes per patient (range, 1–2 episodes). Mycobacterium spp. were the most common (n = 6), including at least two cases of TB, including one case of neurotuberculosis (P11). Granulomas were identified in 46 patients (19.3%), with an average of 3 episodes per patient (range 1–12). Granulomas were most commonly found in lymph nodes (43.5%), lungs (34.8%), skin (21.7%), and intestines/mesentery (13%). Despite the low isolation rate, the genus Mycobacterium was the most frequently isolated group (n = 8, 17% of the total), with M. bovis identified in five of these cases (reaction to BCG). The genus Aspergillus caused pulmonary granulomas in two patients (P33 and P136).

In all participating countries, the BCG vaccine is mandatory at birth (36). 208 patients (87%) were vaccinated with BCG. Among the 30 patients without vaccination confirmation, six did not receive the vaccine due to a family history of CGD, and P156 did not receive the vaccine as it was not administered in the United States of America (USA) where they were born. 94 patients (45%) experienced an adverse reaction, with an average age of 5 mo (range, 0.25–51 mo), and this was the first infectious manifestation of CGD in 48 individuals (51%). 15 patients (16%) had a local reaction (fistulization at the vaccination site, subcutaneous abscess, or delayed healing), and 62 (66%) had a locoregional reaction (regional adenitis), both forms of BCG-itis. 48 patients (51%) had disseminated BCG infection (BCG-osis) (Fig. 6). Pharmacological treatment was administered to 73 patients (77.6%) with BCG-itis, including rifampicin, isoniazid, ethambutol, and pyrazinamide, as well as surgical resection of axillary/cervical lymph nodes in some cases. Although pyrazinamide is ineffective against BCG strains due to intrinsic resistance, it was initially included in empirical TB treatment in a few patients before the diagnosis of BCG infection was established. In all such cases, pyrazinamide was discontinued once BCG was confirmed. The average age of presentation of BCG-osis was 5.2 mo (range: 0.25–36 mo) and affected various anatomical sites: systemic lymph nodes (n = 36); liver (n = 14); lungs (n = 13); bone (n = 7); spleen (n = 6); regional lymph node and lung (n = 4); regional lymph node, lung, liver, and bone (n = 2); intestine (n = 2); regional lymph node, lung, and bone (n = 1); and regional lymph node, lung, liver, and spleen (n = 1) (Fig. 6). Anti-TB treatment was administered to 42 patients with BCG-osis (87.5%), with the combination of rifampicin and isoniazid being the most commonly used (n = 38). Three of the seven patients with BCG-osis who did not receive anti-TB treatment died shortly after diagnosis, and two due to septic shock (P166 and P181). Generally, adverse reactions to BCG were more frequent in patients with XL-CGD (P = 0.002), as was BCG-itis (P < 0.001), but not BCG-osis, for which no significant association was identified. Less frequent than BCG reactions, TB was diagnosed in 37 patients (15%), with a median age of 1 mo (range: 1–5 mo), and M. tuberculosis confirmation in only 35% of cases (n = 13). Pulmonary TB was the most common (n = 24), followed by disseminated forms affecting lymph nodes (n = 10), meninges (n = 5), intestine (n = 3), bones (n = 2), spleen (n = 2), skin (n = 1), and brain (n = 1). BCG infection and TB did not manifest simultaneously, and no cases of environmental mycobacterial infection were confirmed.

Of the total patients, 227 (95%) used cotrimoxazole (sulfamethoxazole [SMX] + trimethoprim [TMP]) for bacterial prophylaxis. One patient, who was intolerant to SMX, substituted it with doxycycline + TMP (P57), while the remaining 10 patients died shortly after diagnosis, which was made in a hospital setting. Regarding antifungal prophylaxis, 207 patients (87%) used azole antifungals, with itraconazole being the most commonly used (97%), and the combination of cotrimoxazole and itraconazole was employed in 187 patients (79%) (37). IFN-γ prophylaxis was adopted by 93 patients (39%) from only three countries (Mexico, Brazil, and Argentina), predominantly in XL-CGD patients (63%) and consistently in conjunction with TMP-SMX, with Mexico being the primary country, having 83 patients (89%). The main reason for discontinuing of IFN-γ prophylaxis was adverse effects, including persistent fever, headache, myalgia, and skin manifestations. Corticosteroids were used by 94 patients (39%) to manage inflammatory processes, ranging from the nasal application of fluticasone for URTI episodes to oral or intravenous administration for more intense inflammation in the intestines (acute or chronic colitis, IBD, and Crohn-like IBD), lungs, granulomas, and hepatic abscesses that were poorly responsive to antibiotics.

Currently the only curative therapy for CGD, HSTC was performed in 53 patients (22%), totaling 68 transplants. Among these, 41 patients underwent a single transplant (60%), and 12 required a second transplant due to primary or secondary graft failure or graft-versus-host disease (GVHD). Of these, three patients needed a third transplant—one for primary failure (P3), another for secondary failure (P149), and the third for reactivation of cytomegalovirus (CMV) (P15). For the first HSCT, bone marrow was the primary stem cell source for 35 patients (66%), followed by peripheral blood (11, 20%) and umbilical cord blood (6, 11%), with data not available for patient P235. The median age at the first HSCT was 4 years (range, 10–236 mo). For the second HSCT, bone marrow was the primary cell source, except for patient P19, who received umbilical cord cells. For the third HSCT, bone marrow (67%) and peripheral blood (33%) were the cell sources. Notably, more Brazilian patients underwent HSCT (32 patients, 60%), with the majority lacking a defined genetic diagnosis (26 patients, 81%). 26 patients with UG-CGD underwent HSCT (49%), 22 with XL-CGD (42%), and only 5 with AR-CGD (9%): 2 patients with p47^phox deficiency (P16 and P157) and 3 with p67^phox deficiency (P162, P163, and P192). The success rate for the first HSCT was 56%, for the second HSCT was 66.6%, and for the third HSCT was 66.7%, with a total success rate of 70% among all transplants. Post-HSCT reactions were observed in 40 patients (75%) following the first transplant, with 18 achieving successful recovery: 10 cases of GVHD (P82, P83, P119, P148, P151, P156, P162, P204, P216, and P235), 4 cases of CMV reactivation (P17, P38, P87, and P91), 3 cases of secondary graft failure and infection (P81, P155, and P217), and 1 case of primary graft failure (P173). Among those who received a second transplant, four were due to primary graft failure, five were due to secondary failure, one was due to CMV reactivation, and two cases were unspecified. Two cases of primary graft failure (P3 and P149) and one case of CMV reactivation (P15) required a third transplant. Primary graft failure occurred again in patient P3, who deteriorated with worsening pulmonary conditions (Aspergillus spp. and K. pneumoniae in blood cultures) and died from septic shock.

225 patients (95%) required at least one hospitalization during the follow-up period, with 40 needing intensive care at least once. When evaluating the number of hospitalizations per year (total hospitalizations/follow-up time in years), patients with XL-CGD had a higher frequency compared to those in the AR-CGD + UGfem group (P < 0.05) but not compared to patients in the CGD-UG male (UGm) group (P = 0.4051). The CGD-UGm group, in turn, showed a difference from the AR-CGD + UGfem group (P < 0.05), as shown in Fig. 7 A. PNM and sepsis were the leading causes of hospitalization, as well as the primary causes of death among patients in this study. PNM was the cause of hospitalization for 175 patients (range, 1–12 hospitalizations), while sepsis caused hospitalization in 106 patients (range, 1–8 hospitalizations). Both conditions were more frequent among patients with XL-CGD (P < 0.05), as illustrated in Fig. 7 B.

80 died (one-third of the total), predominantly male (n = 69, 86%). Seven deaths occurred during data collection (six males, including three with XL-CGD and one female with AR-CGD). The overall survival analysis yielded a median of 300 mo (range: 1 mo–35 years), excluding two patients: one due to missing birth date (P83) and the other due to missing date of death (P169). The overall survival rate for the 236 patients (total) was 84.3% at 5 years and 76.1% at 10 years (Fig. 8 A). The median survival for patients with XL-CGD was 200 mo, lower than for patients in the AR-CGD + UGfem group and very similar to the CGD-UGm group, as shown in Fig. 8, B and C (P = 0.0247). Brazilian patients exhibited better overall survival than those from other countries (P = 0.0013). Analysis of overall survival between transplanted (HSCT) and non-transplanted patients did not show a significant difference between groups, as depicted in Fig. 8 D (P = 0.728). The median age at death for patients with CGD was 66 mo (5 years 6 mo; 1 mo–31 years), with XL-CGD at 60 mo (range: 5 mo–31 years), AR-CGD + UGfem at 88 mo (range: 17 mo–23 years), and CGD-UGm at 75 mo (range: 1 mo–25 years). Mortality among patients with XL-CGD was higher (n = 43), accounting for 54% of the total deaths, compared to only four deaths in patients with AR-CGD, 12 in the AR-CGD + UGfem group, and 25 deaths in the CGD-UGm group.

Infections were the primary causes of death in patients with CGD, both with and without genetic diagnosis (82.5%), with septic shock (n = 36) and PNM (n = 21) accounting for a total of 70% of deaths. Aspergillus spp. infection was the leading cause of septic death (40% of cases with confirmed etiology), all originating from pulmonary sources. Among the deaths due to PNM, only 13 had isolated microorganisms, including eight cases of Aspergillus spp. (one with co-infection of Aspergillus flavus and Aspergillus versicolor—P109, and another with A. fumigatus and B. cepacia—P110), four of which involved dissemination of the infection and septicemia, and one case of disseminated aspergillosis after HSCT (P157). Candida species were present in five patients: four with septic shock (P129 and P132 with Candida spp., P138 with Candida albicans, and P121 with co-infection of Salmonella sp. and C. albicans) and one with PNM associated with Candida sp. (P134), although it is not possible to say that there was no cross-contamination by colonization in catheters or probes during hospitalization in at least one or more cases. Deaths due to PNM related to other fungal infections included Mucor sp. (P9), Histoplasma sp. (P205), and one case with fungal structures in bronchoalveolar lavage (P126). Different bacteria were also isolated from deceased patients: B. cepacia (P105, P109, and P139), S. enterica (P113 and P171), Salmonella sp. and C. albicans (P121), S. marcescens (P131), S. aureus (P231), and Pseudomonas spp. and M. tuberculosis complex (P11) in cases of septic shock. Other infectious causes of death included phagocyte activation syndrome/hemophagocytic lymphohistiocytosis (P99, P100, P188, and P228), pulmonary TB (P54), infectious myocarditis (P170), severe gastroenteropathy (P193), and two deaths due to BCG dissemination: one from PNM and respiratory failure (P236) and another from septic shock originating from the pulmonary focus (P147), both with XL-CGD, along with one case of disseminated fungal infection after HSCT (P19). Non-infectious causes included autoimmunity: autoimmune thrombocytopenia and hemorrhage in the central nervous system following gene therapy (P158) and lupus nephropathy (P215), neoplasia (craniopharyngioma—P108), and complications after HSCT (GVHD—P59, P86, P89, and P94; thrombotic microangiopathy—P65).

The crude mortality rate did not significantly differ between groups (P = 0.117). However, the specific mortality rate due to PNM among patients with XL-CGD (12.8%) was statistically significant when compared to the AR-CGD + UGfem group (1.9%) (P = 0.038), although no significant difference was observed for septic death (P = 0.272). When examining the case-fatality rate for PNM and sepsis among the groups, a higher case-fatality rate for PNM was noted within the XL-CGD group (14.7%), with 66.7% of deaths due to PNM, compared to the AR-CGD + UGfem group (2.4%), with only 5% of deaths due to PNM (P = 0.035), no significant differences were observed for deaths due to sepsis (P >0.3). PNM was the cause of death exclusively in male patients (66.7% were XL-CGD patients), except for one female carrying a mutation in the CYBB gene (P205), considered as having XL-CGD, and another female with a UG (P212). When assessing association measures, the prevalence ratio for patients with XL-CGD dying from PNM is 6.68 times higher than that of patients in the AR-CGD + UGfem group (1.19–39.4–95% confidence interval [CI]), as is the odds ratio, which is 7.52 times higher (1.15–81.2–95% CI), without confirming this association in cases of death due to sepsis.

This study, the largest in Latin America, highlights the genetic and clinical diversity of CGD patients across the region and provides valuable insights into the disease’s epidemiology, treatment outcomes, and challenges. The high rate of XL-CGD observed is consistent with global patterns (8, 10, 11, 12, 13); however, the relatively high frequency of AR forms reflects the social context of certain Latin American regions, where consanguinity is relatively more common, with 46% consanguinity among patients with AR-CGD versus 20% overall. This is considerably lower than the consanguinity rates observed among Arabs and Israeli Jews, where consanguineous marriage is frequent, with high consanguinity rates and AR-CGD cases similar or even higher than XL-CGD (11, 38). The genetic diversity identified, particularly the novel pathogenic variants in CYBB and NCF2, highlights the necessity for region-specific genetic resources and diagnostic tools to improve early detection and management, especially concerning Brazilian patients.

The results highlight considerable delays in diagnosis, with most diagnoses made more than a year after symptom onset. This delay is likely due to limited awareness, diagnostic resources, and healthcare disparities across Latin America. Our work demonstrated the lowest median/mean age of definitive diagnosis when compared to large-scale studies: USA (64.8 mo) (39), Europe (63.6 mo) (10), Latin American studies, such as in Mexico (30 mo) (13), and the last and largest study on CGD in the region (52.7 mo) (8). The same pattern can be observed when we evaluated patients with XL-CGD and AR-CGD, demonstrating that efforts for early diagnosis in recent years have shown positive results. Early diagnosis is crucial in CGD, as prompt treatment and monitoring can reduce infection-related morbidity and mortality, as well as increase the chances of successful HSCT. Recent results from Europe (the largest series of transplants in patients with CGD reported to date) have shown that even with haploidentical transplants there was a substantial improvement in overall survival and event-free survival in patients with CGD, with even better results with compatible donors and younger patients (24). Among the 53 transplanted patients, 2 did not have age of the first transplant. Of the remaining patients, 10 were transplanted up to 12 mo after diagnosis, and the others after 1 year of definitive diagnosis (76.7%), ranging from 13 mo to 13 years 11 mo. Despite clinical stability at the time of transplantation, many patients exhibited chronic infections, potentially affecting therapeutic outcomes. Since there is insufficient information on the conditioning regimen, it is not possible to discuss the relationship between conditioning and outcome.

All transplants were performed in referral hospitals for pediatrics and HSCT. 34 patients were transplanted when ≤8 years old (64%) and 17 >8 years (32%). When we evaluated the survival of the two groups (eight deaths in each group), we found more favorable survival in patients who underwent HSCT earlier (P = 0.01), as observed in other series (24, 40). Thus, a relative benefit in the success rate of HSCT and survival of Latin American patients may be obtained with the earliest possible transplant, since most transplants in this region are performed when previous therapies or prophylaxis have not demonstrated efficacy in the long term. This suggestion is partly based on the two important declines in the transplant curve observed in Fig. 8 D, referring to Brazilian patients who died a few months after HSCT (P3, P59, P79, and P86—first decline, and P18 and P94—second decline), even with new transplant attempts in some of them (P3 [3×]; P59 and P79 [2×]). Patient P94, a supposed success, presented GVHD that severely affected the lung, leading to acute chronic respiratory failure—PNM with bronchiectasis—dying 5 years after HSCT.

The age of onset of symptoms and the diagnostic delay varies between studies, but the trends are consistent across developing countries (14). The difference in age of diagnosis is due to the considerable variation in manifestations among patients, even in developed countries (9, 10, 16, 41), such as the early age of first clinical manifestations and more severe events in XL-CGD and generally later onset of the disease with less severe and frequent events in AR-CGD. Implementing more widespread newborn screening for CGD in Latin America could mitigate these delays, facilitating timely intervention and improving outcomes.

Infectious complications, particularly with pathogens such as Aspergillus and Staphylococcus, remain the primary challenge for CGD patients, where invasive Aspergillus spp. infections remain a significant challenge despite the implementation of azole antifungals, the main cause of severe pulmonary infections and death in patients with CGD, as observed in this study (9, 10, 12, 13, 41, 42). The diversity of microorganisms isolated in this study, despite the technical, operational, and logistical challenges of pathogen isolation, demonstrates that health systems should pay more attention to the socio-environmental context of patients with CGD in Latin America since the infectious profile has important implications for morbidity and mortality, as well as on consequences for future HSCT. The observed high prevalence of Aspergillus infections could be linked to environmental factors, while the occurrence of severe BCG vaccine reactions suggests a need to re-evaluate vaccination policies for CGD patients in endemic regions (8, 10, 12, 13, 16, 20).

The gastrointestinal involvement, particularly in cases of granulomatous colitis, highlights the need for multidisciplinary approaches to manage both infectious and inflammatory complications. IBD in CGD can affect up to 50% of patients and is a significant cause of weight and height deficit in these patients (21, 43). Overall, 31 patients presented IBD (13%) with or without granulomas, being more common among Brazilian patients with XL-CGD, who were mostly treated with corticosteroids, as they help reduce inflammation and also assist in the treatment of hepatic abscesses and granulomas, which occur in approximately one-third of CGD patients (2, 44). In cases of poor response to corticosteroids or hyperinflammation, immunobiologicals are especially effective (2, 43). Infliximab, a chimeric monoclonal antibody against tumor necrosis factor (TNF)-α, has shown promising results in persistent colitis associated with CGD and fistula closure; however, it can lead to an increase in the frequency of serious infections with typical CGD pathogens that can quickly progress to death (45, 46). Other anti-TNF antibodies, such as adalimumab and golimumab and the antagonist of the p40 subunit of interleukin (IL)-12 and IL-23 (ustekinumab), have been used in inflammatory bowel processes with good results (7, 47). Finally, HSCT leads to complete and stable colitis remission in patients with CGD (21, 23).

All patients received standard prophylaxis (cotrimoxazole and itraconazole). At the same time, IFN-γ was almost exclusively administered to patients from Mexico (89% of those with access). Therefore, it was not possible to compare it with other countries that made this drug available (Brazil and Argentina). IFN-γ has been shown to reduce the number and severity of infections and the relative risk of serious infections without significant adverse effects in prolonged use with a decrease in hospitalization time (21, 22). IFN-γ is currently accepted and administered in several countries to treat CGD (8, 9, 10, 13). Of note, in Brazil, this drug remains unregistered with the National Health Surveillance Agency, which limits access and dissemination of its use (only nine patients had access) (8).

HSCT outcomes varied, with a higher success rate in Brazilian patients, likely due to better access to specialized centers. Despite the curative potential of HSCT, the risk of complications remains high, highlighting the importance of careful patient selection and optimized transplantation protocols since this therapeutic option is frequently utilized when standard prophylaxis fails to control infections, resulting in patients being transplanted with significant sequelae from previous infections (2, 40, 45, 48). Expanding access to HSCT, the early transplantation and enhancing supportive care could improve outcomes for more patients throughout Latin America.

This study provides a comprehensive analysis of CGD in Latin America, detailing clinical and genetic characteristics and highlighting the disparities in healthcare access and outcomes across the region. Our findings emphasize the relevance of early diagnosis, access to genetic testing, and improved therapeutic options to enhance patient quality of life and survival rates. These needs become evident when we compare the survival curve of patients in Latin America with the curve of the study in the USA, which evaluated the survival of 227 patients with CGD based on residual O₂⁻ production, showing a delay of more than 10 years for the first deaths, while in Latin America survival reached almost 70% in the same interval, reaching almost 50% at 25 years (49). The high frequency of novel pathogenic variants observed underscores the need for a Latin America-specific genetic database to aid in early diagnosis and personalized treatment planning.

Future efforts to develop region-specific guidelines, expand HSCT access, and implement newborn screening programs for CGD are essential steps for reduce the disease burden. This study provides the basis for future research and policy development to support CGD patients in Latin America and improve healthcare equity within this population. Continued collaborative efforts across the region will be essential to advance CGD treatment and ensure timely and effective interventions for affected patients.

Patient data were collected from the diagnostic and research service registry of the Human Immunology Laboratory at the Institute of Biomedical Sciences (ICB) at the USP and subsequently from the Latin American Society for Immunodeficiencies (LASID) registry after obtaining consent from the participating physicians and informed consent signed by the patients or their parents. Additional information was obtained through a detailed online questionnaire completed by the physicians, which included demographic, clinical, microbiological, laboratory, genetic, and family data (history of CGD, early death due to infection, and consanguinity), as well as information on treatment (prophylaxis and therapy, including HSTC), hospitalization, reactions to the BCG vaccine, TB, chronic inflammatory manifestations, and death. Data collection occurred between 2020 and 2022, using records from patients diagnosed between 1981 and 2021 from 53 pediatric hospitals in Brazil, Mexico, Chile, Costa Rica, Argentina, Paraguay, Peru, and Uruguay (30 of which were registered in LASID). After correcting divergent or duplicate data, the study included 238 patients. The LASID registry and the use of recorded information were approved by the National Health Council of the Brazilian Ministry of Health for international studies (CONEP 25000.040727/2008-16, CAAE 0034.1.146.000-08). Ethical approvals were obtained from all research ethics committees of the participating countries’ institutions, in accordance with the Helsinki Convention, as well as from the Research Ethics Committee of ICB-USP (CAAE: 39743920.5.0000.5467) and the Ethics Committee of the Hospital das Clínicas da Faculdade de Medicina da USP (CAAE: 39743920.5.3001.0068).

The present study included patients of both sexes diagnosed with CGD, based on the identification of laboratory and clinical findings: (1) reduced or absent production of reactive oxygen species (ROS) or defective expression of components of the phagocyte NADPH oxidase complex (gp91^phox, p22^phox, p47^phox, p67^phox, p40^phox, and EROS) (33, 50, 51, 52, 53); (2) identification of a pathogenic variant in any of the involved genes (CYBB, CYBA, NCF1, NCF2, NCF4, and CYBC1), either in homozygosity, compound heterozygosity (AR form), or hemizygosity (XL form—or heterozygosity in the case of symptomatic carrier females) (3, 4, 5, 7); and (3) individual or family history of CGD with frequent infections or early death from severe infection or granulomatous disease (brother, maternal uncle, or maternal cousin with XL-CGD) or mother with dual populations regarding ROS production—suggestive of being a carrier (52, 54).

Laboratory diagnosis of CGD was performed for all patients (except for one with a prior genetic diagnosis). It was based on the abnormal oxidative burst in granulocytes, using either the NBT assay (50) or DHR assay, or both (51). The NBT test was conducted in some patients, mainly those diagnosed before the 1990s, with subsequent confirmation of diagnosis in many of them using the DHR assay. In total, 229 patients were diagnosed using DHR, eight using NBT, and one patient had only a genetic diagnosis. The NBT test assesses the reduction of NBT to formazan (a visible blue pigment) in neutrophils after stimulation. Cells from CGD patients do not reduce NBT, except in cases with residual ROS production (50). DHR, which has replaced NBT and is the current gold standard for CGD diagnosis, detects by flow cytometry the oxidation of DHR to rhodamine-123 (fluorescent) by ROS in stimulated neutrophils. The median fluorescence intensity of activated cells directly correlates with superoxide production (typically <5% of normal control), allowing not only for the diagnosis of the deficiency but also the inference of inheritance, carrier status in females, and the association of ROS levels with disease severity (33, 51, 52, 53).

Each healthcare center provided information on the results of genetic and molecular diagnoses. Only Mexico provided information on the expression of NADPH oxidase subunits from a previous study that also included additional genetic analysis of the Mexican patients described here (13). For genetic investigation of Brazilian patients at ICB-USP, genomic deoxyribonucleic acid (gDNA) was extracted from peripheral blood using the Wizard Genomic DNA Purification Extraction Kit (Promega Corporation), after erythrocyte lysis according to the manufacturer’s instructions. The concentration and quality of the purified gDNA were assessed using a NanoDrop 2000 (Thermo Fisher Scientific), and DNA integrity was confirmed by electrophoresis on a 1.5% agarose gel (Sigma-Aldrich) stained with SYBR Green-II (Thermo Fisher Scientific). The exons and flanking intron sequences of the genes CYBB, CYBA, NCF2, and NCF4 were amplified by polymerase chain reaction (primers and conditions used are available upon request) using Taq DNA polymerase (Life Technologies), and the amplicons were sent to the Human Genome Study Center (ICB-USP, São Paulo) for Sanger sequencing and whole-exome sequencing.

The onset of symptoms was estimated based on data regarding the first relevant symptoms presented before diagnosing the oxidative burst deficiency by DHR or NBT. Diagnostic delay was calculated by subtracting the age at diagnosis from the age at onset of symptoms. Infectious and inflammatory clinical manifestations occurred over their lifetime, with etiological agents not always isolated but clinically inferred, with emphasis on pyogenic and mycobacterial infections, including reactions to the BCG vaccine, which were categorized into BCG-osis, BCG-itis, and local reaction. Confirmed cases of infection were quantified and evaluated regarding the main infectious sites. Information on treatment was limited to standard prophylaxis against infections, IFN-γ, corticosteroids, and HSCT. Genetic variants not found in specialized databases (e.g., HGMD, GnomAD, 1000 Genomes Project, and ClinVar) or in scientific articles were considered “novel” variants and assessed using pathogenicity predictors (CADD, SIFT, MutationTaster, and POLYPHEN-2). Hospitalization per year was calculated by dividing the total number of hospitalizations, regardless of cause, by the follow-up time in years and then compared between groups of patients with XL-CGD, AR-CGD + UGfem (unknown genetic females), and UG-CGD (unknown genetic CGD) using the Kruskal–Wallis test (P < 0.05). A similar analysis was performed, considering the cause of hospitalization, between the XL-CGD and AR-CGD + UGfem groups, using the Mann–Whitney test (P < 0.05). For survival analysis, it was necessary to group AR-CGD patients with female patients with UG, clinically diagnosed as AR-CGD, to obtain a statistically valid “n” (two patients [P145 and P205] carrying of a pathogenic variant in CYBB, were excluded from this group). The analysis period started from the date of birth and continued until the last follow-up update or the analyzed outcome (death). Two patients were not included in the survival analysis: one due to missing date of birth and another due to missing exact date of death. Mortality was assessed by crude rate (number of deaths in the group/number of individuals in the group × 100), specific rate (number of deaths from a specific cause/number of patients in the group × 100), and case fatality rate (deaths from specific infection/total number of patients with the same infection by group × 100). Comparison between groups was assessed using the χ² test or Fisher’s exact test (95% CI) and the prevalence ratio and odds ratio (95% CI).

For the geospatial analysis, maps were created using Geographic Information System (GIS) technology and spatial statistics. In spatial statistics, the distribution of patient origins and diagnostic and follow-up centers in Brazil was explored and described using Quantum Geographic Information System (QGIS) software. The maps were referenced using the cartographic base obtained from the Instituto Brasileiro de Geografia e Estatística and equivalent institutions in other countries: Instituto Geográfico Nacional of Argentina, Instituto Nacional de Estadística (INE) of Paraguay, Instituto Nacional de Estadística e Informática of Peru, INE of Uruguay, INE of Chile, and Comisión Nacional para el Conocimiento y Uso de la Biodiversidad of Mexico. The geographic distribution of the IEIs network centers in Brazil was obtained from the LASID Registry and the detailed questionnaire. In contrast, for the other countries, the analysis was performed by geographic region of the patient’s origin due to the lack of precise identification of diagnostic and follow-up locations (geographic coordinates). Shapefiles (.shp), an Environmental System Research Institute vector data format for storing geographic coordinates, shape, and attributes of geographic features in GIS, were obtained from each of the described institutions, and, in QGIS software, the geographic location spreadsheet of each center (Brazil only) and patients was incorporated into the shapefile of each country (13, 55, 56). To represent the data, geometric shapes (triangles for medical centers and circles for patients) were used, with size determined by class.

Table S1 shows the genetic characterization and clinical outcomes of all patients with CGD in this study (n = 238).

This study was performed in line with the principles of the Declaration of Helsinki. Approval was granted by the Ethics Committee of the ICB, USP, and by the Ethics Committee of Hospital das Clínicas da Faculdade de Medicina da Universidade de São Paulo - Children’s Institute (CAAE: 39743920.5.0000.5467 and 39743920.5.3001.0068).

Informed consent was obtained from all participants included in the study according to local ethics guidelines.

All authors agree to the publication of this manuscript.

The data are available from the corresponding author upon reasonable request.

The authors would like to thank all patients and their families for participating in this study. We would like to thank Dr. Lizbeth Blancas-Galicia for her efforts in providing most of the data from Mexican patients and for mediating contact with more professionals to participate in the study. We would also like to thank Eliel Gonçalves, administrator of the LASID Registry, for providing important information to contact physicians throughout Latin America. We would also like to thank Rosana Duarte Prisco for her assistance with the statistical analyses.

Conselho Nacional de Desenvolvimento Científico e Tecnológico supported Oliveira Tiago Santos de Oliveira (Grant #133246/2019-1) and Fundação de Amparo à Pesquisa do Estado de São Paulo supported Antonio Condino-Neto (Grant # 2016/22158-3). The Jeffrey Modell Foundation supported Antonio Condino-Neto.

Author contributions: Tiago Santos Santos de Oliveira: conceptualization, data curation, formal analysis, investigation, methodology, project administration, resources, validation, visualization, and writing—original draft, review, and editing. Ranieri Coelho Salgado: conceptualization, supervision, validation, visualization, and writing—original draft, review, and editing. Lillian Nunes Gomes: conceptualization, investigation, and writing—original draft. Carolina Sanchez Aranda: data curation, formal analysis, resources, and supervision. Janáira Fernandes Severo Ferreira: investigation. Ekaterini Simões Goudouris: investigation, visualization, and writing—original draft, review, and editing. Fernanda Pinto Mariz: resources. Maria Luiza Oliva Alonso: resources and writing—review and editing. Solange Oliveira Rodrigues Valle: resources and writing—review and editing. Luiz Fernando Bacarini Leite: resources. Ana Carolina da Matta Ain: resources. Pérsio Roxo-Junior: resources. Marília Montanaro: resources. Flávia Alice Timburibá de Medeiros Guimãraes: data curation, investigation, and writing—review and editing. Simone de Castro Resende Franco: conceptualization. Herberto José Chong-Neto: conceptualization, investigation, methodology, resources, and writing—review and editing. Nelson Augusto Rosario Filho: investigation and writing—original draft, review, and editing. Daniélli Christinni Bichuetti-Silva: resources. Vera Maria Dantas: resources. Jackeline Motta Franco: resources. Viviane Guimaro Mendes Barreto Resende: resources. Simone Soares Lima: resources. Dorcas Lamounier Costa: investigation, validation, and visualization. Mayra de Barros Dorna: resources. Antônio Carlos Pastorino: data curation and supervision. Beni Morgenstern: investigation. Vera Esteves Vagnozzi Rullo: resources. Constantino Giovanni Braga Cartaxo: investigation. Naiade R. de Sá: resources. Irma Cecília Douglas Paes Barreto: resources. Nancy Viegas Chen Lobo: resources. Flavia Amendola Anisio de Carvalho: resources. Monica Soares de Souza: conceptualization, data curation, investigation, project administration, and resources. Gesmar Rodrigues Silva Segundo: conceptualization, resources, visualization, and writing—original draft, review, and editing. Dewton de Moraes Vasconcelos: investigation and resources. Juliana Folloni Fernandes: resources. Gislene Santana Tusani: investigation. Carolina Cardoso de Mello Prando: data curation, investigation, validation, and writing—review and editing. Ana Paula Zaninelli Diniz Iwamura: resources and writing—review and editing. Arturo López-Larios: resources. Ana Jocelyn Carmona Vargas: resources. Ana Paola Macias-Robles: resources. Maria Edith Gonzalez-Serrano: conceptualization, data curation, formal analysis, investigation, resources, and writing—original draft, review, and editing. Sara Elva Espinosa Padilla: investigation, validation, visualization, and writing—review and editing. Ana Karen Peñafiel Vicuna: data curation. Aidé Tamara Staines-Boone: data curation and investigation. Omar J. Saucedo-Ramirez: writing—review and editing. Miguel García-Domínguez: data curation, investigation, resources, supervision, validation, and visualization. Maria R. Canseco-Raymundo: investigation. Federico Saracho-Weber: resources. Oscar Zavaleta Martínez: investigation, resources, and validation. Susana Garcia-Pavón-Osorio: resources. Rogelio Guzmán Cotaya: investigation. Francisco J. Espinosa-Rosales: conceptualization, resources, supervision, and writing—review and editing. Roselia Ramirez-Rivera: investigation. María de Luz García-Cruz: investigation. Carmen Zarate-Hernández: investigation. Lizbeth Blancas-Galicia: investigation. Alejandra King: investigation, resources, visualization, and writing—review and editing. Oscar Porras: investigation, validation, and visualization. Lorena Regairaz: resources. Alejandra Lampugnani: resources. Celia Martínez de Cuellar: resources. Juan Carlos Aldave Becerra: resources. Liz Eliana Veramendi Espinoza: investigation, validation, and writing—original draft, review, and editing. Cecilia Montenegro: resources, validation, and writing—review and editing. Magdalena Schelotto: investigation. Virginia Patiño: resources. Mariane Cunha de Oliveira: formal analysis and writing—review and editing. Antonio Condino-Neto: conceptualization, data curation, formal analysis, funding acquisition, investigation, methodology, project administration, resources, supervision, validation, visualization, and writing—review and editing.