Chronic granulomatous disease (CGD) is an inborn error of immunity (IEI) characterized by a defective respiratory burst in phagocytes, resulting in defective clearance of phagocytosed microorganisms (1). It is caused by mutations in genes encoding different protein subunits of nicotinamide adenine dinucleotide phosphate (NADPH) oxidase enzyme complex (2). Clinical manifestations vary from an immunodeficient phenotype with repeated infections (3, 4) to those characterized by uncontrolled hyper-inflammation (5) and other autoimmune manifestations. Common patterns of infections in CGD include pneumonia, lymphadenitis, hepatosplenomegaly, and abscesses (6).

Laboratory diagnosis of CGD can be made by the nitroblue tetrazolium dye reduction test (NBT) (7, 8) or by flow-cytometry based dihydrorhodamine (DHR) assay (9, 10). The DHR assay is now considered to be the preferred screening test due to its higher reproducibility, sensitivity, rapidity, and ability to detect X linked carriers (11, 12). While several centers in India have published their individual experiences on CGD (13–16), countrywide data have never been collated before. Such collaborative efforts are the need of the hour and are especially important for uncommon conditions like CGD, for which there is paucity of data on disease burden in the country. The present work is the first multi-centric study in India to provide data on clinical, immunological, and molecular features of CGD.

We contacted all centers that are recognized as Foundation for Primary Immunodeficiency Diseases (FPID) regional centers for diagnosis or treatment for primary immunodeficiencies in India, and also other Indian institutions involved in care of patients with IEI. All centers were requested to provide clinical and laboratory details of patients with CGD on a pre-designed Excel sheet. Data collection was completed by July 2020. Details included demographic information, family history including consanguinity, clinical manifestations, immunological investigations, genetic diagnosis, treatment, and follow-up.

Participating institutes included—Postgraduate Institute of Medical Education and Research (PGIMER), Chandigarh, North India (80 patients); Bai Jerbai Wadia Hospital for Children, Mumbai, West India (57 patients); Aster CMI Hospital, Bengaluru, South India (22 patients); National Institute of Immunohematology (NIIH), Mumbai, West India (22 patients); Kanchi Kamakoti Childs Trust Hospital, Chennai, South India (21 patients); Apollo Hospitals, Chennai, South India (17 patients); Christian Medical College, Vellore, South India (five patients); Kasturba Medical College, Mangalore, South-West, India (five patients); Sir Ganga Ram Hospital (SGRH), New Delhi, North India (four patients) and one patient each from R G Kar Medical College, Kolkata, East India; Indraprastha Apollo Hospital, New Delhi, North India; and Medens Hospital, Haryana, North India. Data were collated and subsequently translated on to a database and analyzed ( Figure 1 ). Before the analysis, all patient identity details were anonymized. Three (3) female patients had a probable skewed X-linked inactivation. These patients have been described separately.

CGD was first described in four male children from Minnesota in the year 1954 (34) and was later referred as ‘fatal granulomatosus of childhood’ by Berendes et al. in 1957 (35). Subsequently, molecular mechanism of disease pathogenesis (36) and genetic defect (CYBB) was discovered in the years 1967 and 1986 (37, 38), respectively. CGD is now one of the commonly recognized IEI in children. Prognosis of affected children has improved significantly over the last few decades with the introduction of long-term antimicrobial prophylaxis and initiation of curative therapies such as HSCT. Incidence of CGD differs between populations, and it ranges from an estimated one per 200,000 lakh live births in the US and Europe (1, 6), to 1.5 per 100,000 live births among Arabs in Israel where consanguinity rates are high (39).

Initial reports of CGD from India date back to late 1990s (40). Subsequently, the two Indian Council of Medical Research Centres for Advanced Research in Primary Immunodeficiency Diseases (PGIMER, Chandigarh and NIIH, Mumbai) published large case series on CGD from their respective centres (13, 15, 16). Several case reports and case series from other centers have also emanated in the last few years. The present multi-centric study is an effort to collate the data of patients with CGD diagnosed at multiple centers across the country to understand epidemiology, profile of infections, and molecular spectrum of CGD in India. For this study, we obtained data on 236 patients with CGD diagnosed at 12 different centers in India.

We observed a higher proportion of AR forms of CGD as compared to X-linked forms. This is consistent with the previously published data on CGD from India (15). Higher proportion of AR-CGD may be a reflection of high rates of consanguinity and endogamous marriages in the country. Median age at diagnosis in our cohort was 24 months (IQR: 8–60 months), which is comparable with data from other large multi-centric studies from the USA, Europe, and China (1, 6, 41, 42) ( Supplementary Table 3 ). Patients with X-linked CGD have an earlier age of presentation and are reported to have a more severe illness as compared to patients with AR-CGD. We also observed that age at onset of infections and diagnosis of AR-CGD was significantly higher than XL-CGD. However, unlike reports from several other countries (1, 43), there was no difference in survival and mortality rates between AR and XL types of CGD. We attribute this difference to three factors—delays in diagnosis of patients with AR-CGD, a probable missed diagnosis of milder forms of AR-CGD in many patients probably due to lack of awareness especially among the adult physicians, and lack of appropriate laboratory facilities for diagnosis of CGD at many hospitals in India. At present, NBT test can be performed at most centers involved in care of IEI and DHR is currently performed only at six centers. However, many tertiary care hospitals in the country still do not have the facility to perform either NBT or DHR. Median follow-up duration of our cohort is 1 year suggesting that majority of patients are only diagnosed only in the recent years. Increase in awareness of IEI and upscaling of laboratory facilities for carrying out basic immunological investigations are the needs of the hour.

Among the breakthrough infections, pneumonia was most common followed by abscesses (skin, subcutaneous, and deep abscesses, inclusive of liver abscesses), lymphadenitis, and BSI. India, being a country endemic for tuberculosis, it is not surprising that many children with persistent pneumonia (54%) in our cohort had received empirical ATT in our cohort. This has led to significant delays in diagnosis of CGD. Pediatricians and internists practising in developing countries such as India need to keep in mind that persistent pneumonia may be a presentation of CGD. Empirical ATT may not be warranted in such cases.

Aspergillus sp. is the most common etiology for pneumonia in our cohort followed by gram-negative bacteria, Mycobacterium sp., and Staphylococcus sp. Microbiological profile of pneumonia is similar to cohorts from China (44), Iran (45), and Turkey (43). Pneumonia spreading to contiguous tissues such as ribs and vertebra has been documented only with pneumonia due to fungal or mycobacterial etiology (Aspergillus sp.—four; Mucorales sp. —one; Mycobacterium sp. —two). Pulmonary infection due to Mucor sp. was documented in two patients who had not been exposed to corticosteroids. Delays in diagnosis, malnutrition, and prolonged antibiotic therapy for pneumonia could be the reasons for the development of Mucorales infection in our cohort. Nocardia sp. has been isolated in two patients—one with a lung abscess and the other with persistent pneumonia. Unusual bacteria documented with pneumonia include C. gleum, Citrobacter sp., and Francisella noatuensis. A patient with persistent pneumonia had grown C. lusitaniae from the lung tissue which was resistant to amphotericin B.

A wide spectrum of microorganisms that have been isolated in CGD patients with pneumonia suggests the importance of microbiological isolation of organism and appropriate targeting of anti-microbials. We suggest that patients with severe pneumonia in CGD must be managed in tertiary-care centers with facilities for advanced microbiological testing such as MALDI-TOF, 16S rRNA PCR, fungal cultures, and PCR.

Apart from signature organisms such as Aspergillus sp. and Staphylococcus sp., we also documented a high incidence of mycobacterial infections (18.5%) in our cohort. Most common presentation of mycobacterial infection was pneumonia followed by disseminated forms and lymphadenitis ( Table 4 ). Localized or disseminated infections due to BCG have been documented in 8% of patients in our cohort. Abnormal BCG response has been documented at higher rates in other cohorts such as Iran (55.9%) (45), Turkey (22.5%) (43), Latin America (29.6%) (46), and China (64%) (41) that also administer BCG to all children. We do not investigate all patients with BCG adenitis for CGD. It is possible that milder forms of CGD may present only as BCG adenitis and that such patients may have been missed in our cohort. Incidence of Serratia infection in our cohort is very low (2/236, 0.8%) when compared to studies from North America and Europe (47). Possible reasons for this observation include tropical climate patterns and low microbiological isolation rates. Lee et al. have previously reported that infection due to Burkholderia pseudomallei and Chromobacterium violaceum in CGD have been predominantly noted in tropical countries (48). Melioidosis is common in Southern India, and we have also documented melioidosis in a patient from South India who had X-linked CGD (49). Septicemia due to C. violaceum has also been documented in a child with NCF1 defect who hailed from Central India. It is possible that many patients with melioidosis are not screened for CGD in the country and diagnosis of CGD may be missed in them.

Staphylococcus aureus was the commonest organism isolated in suppurative lymphadenitis and skin abscess. Other bacteria isolated in such infections included gram-negative organisms like Burkholderia sp., Klebsiella sp., and Pseudomonas sp. Fungi such as Candida sp. and Basidiobolus sp. were only occasionally isolated in such infections. The microbiological spectrum in these infections is similar to reports from other countries (1, 6, 43, 50). Among the BSI, the microbiological spectrum was almost similar—Staphylococcus sp. was the commonest, followed by Burkholderia sp. and other gram-negative bacteria. However, non-typhoidal Salmonella (8%) was isolated more frequently from blood as compared to other sites. This suggests the importance of frequent handwashing and avoidance of ingestion of raw food items in patients with CGD, as these factors increase the risk of acquiring non-typhoidal Salmonella. S. aureus is the only organism documented in patients with liver abscess. Oral prednisolone was successfully used in three children with liver abscess for enhanced resolution, and none of them required surgical intervention. Our experience is in line with recent evidence concerning use of corticosteroids in liver abscess in CGD (51).

Common non-infective manifestations documented in our cohort include inflammatory bowel disease-like colitis, visceral granuloma, and secondary HLH. We document a lower rate of colitis (5%) in our cohort compared to studies from North America, Europe and China. However, our rates are comparable to cohorts from Turkey, Iran, and Israel, where autosomal recessive forms of CGD are predominant compared to X-linked CGD ( Supplementary Table 3 ) (43, 45). Two patients who had colitis since early infancy were managed as inflammatory bowel disease elsewhere, and were diagnosed to have CGD at 5 and 7 years, respectively. One of the children also developed fulminant pneumonia following infliximab therapy for IBD-like colitis, similar to the report by Uzel et al. (52). This suggests that work-up for CGD must be considered in children with early-onset colitis – pediatricians and gastroenterologists need to be aware of this presentation. Unlike reports from other studies (1, 53), children with AR-CGD had a higher incidence of colitis in our cohort compared to X-linked forms. The difference is probably due to predominance of AR forms of CGD in the present series. While four patients achieved remission with only oral prednisolone and mesalamine, three achieved partial control with azathioprine. Complete remission of colitis following HSCT was documented in a child with X-linked CGD. This shows that immunological reconstitution following HSCT is also beneficial for the inflammatory component of CGD.

Flow cytometry-based evaluation of NADPH oxidase components is a surrogate marker for identifying the genetic defect in CGD. However, this assay can be currently performed in only in two centers in India—PGIMER, Chandigarh and NIIH, Mumbai. We identified regional heterogeneity in molecular spectrum of CGD within India. Overall, NCF1 is the commonest molecular defect identified in patients with AR-CGD, similar to the cohorts from other parts of world. However, CYBA is the second common type of AR-CGD in patients from Southern and Western India, whereas, NCF2 defect is commoner in patients from North India. Moreover, a common dinucleotide deletion in NCF2 (c.835_836delAC) was observed in 10 patients from North India, suggesting a Founder effect in this population. Observed regional differences in molecular spectrum could be due to heterogeneity in genetic background of the population in India. Genescan is considered to be the preferred mode for identification of common dinucleotide deletion in Exon 2 of NCF1 (c.75_76delGT). This modality is, however, currently available only at Chandigarh and Mumbai. The preferred molecular approach for work-up of patients with CGD at Mumbai has been detailed in a previous publication. At Chandigarh, we follow a similar approach, except that Sanger sequencing for NCF2 c.835_836delAC variant is carried out first for patients who have decreased p67^phox expression by flow cytometry, as this variant is commonly and exclusively observed in North Indian population. This is followed by NGS whenever required. We have not identified p40^phox (NCF4) defect in our cohort. However, patients with NCF4 defect can have near normal neutrophil stimulation in the conventional DHR assay done using phorbol myristate acetate (PMA) as the stimulant (54). It is possible that this defect has been missed in many patients in India, as all patients suspected to have CGD were screened with DHR assay using stimulation with PMA. Recently, EROS (CYBC1) defect resulting in CGD has also been described (55). We have not screened for this defect in our cohort.

Most of the patients in our cohort were kept on antimicrobial prophylaxis—cotrimoxazole and itraconazole. Interferon-gamma was not used in any of our patients as it is not available in India. Number and severity of infections, especially those due to Staphylococcus sp., significantly came down after initiation of antimicrobial prophylaxis ( Table 8 ). However, many patients had documented breakthrough infections even while being continued on prophylactic medications. Overall mortality in our cohort was 38.5% which is similar to reports from other countries such as Mexico and China (41, 50, 56, 57). Our mortality rates are, however, much higher than those reported from the USA, Iran, and European countries. Delays in diagnosis, severe infection at the presentation, and lack of widespread availability of pediatric HSCT services in the country accounted for high mortality rates in our cohort. However, we also document significant decrease in number of infections following initiation of antibiotic prophylaxis ( Table 8 ). This suggests that patients with CGD in developing countries, especially AR forms (NCF1 defect), can have a reasonable quality of life provided they are diagnosed early and continued on long-term antimicrobial prophylaxis ( Figure 5 ).

HSCT could be carried out in only 23 patients in our cohort. Most of the HSCTs have been carried out in Apollo Children’s Hospitals, Chennai. Only five other centers (BJ Wadia Hospitals for Children, Mumbai; PGIMER, Chandigarh; SGRH, New Delhi; Aster CMI Hospitals, Bangalore; Apollo Indraprastha Hospitals, New Delhi) have carried out HSCT for CGD until date. Successful engraftment was documented in 15 patients who received either bone marrow or peripheral blood-derived stem cells. A child who received cord-blood transplantation failed to develop early myeloid engraftment and succumbed to severe infection. This is similar to the experience reported by Morio et al. (58). Four out of six children who received haploidentical transplantation have had a successful engraftment, indicating that haploidentical transplantation is a potential life-saving option in children when completely matched donors are not available. We reiterate that establishment of dedicated pediatric HSCT centres, and government support for patients undergoing HSCT are essential for state-of-the-art management of CGD in developing countries.

To conclude, we document the first multicentric study on clinical and molecular features of 236 patients with CGD from India. To the best of our knowledge, ours is the third largest cohort of patients with CGD documented till date after the multicentric reports from the USA (n = 368) and Europe (n = 429). We have documented a wide spectrum of bacterial and fungal infections and a high incidence of mycobacterial infections in our cohort. Though AR-CGD predominates in our cohort, significant regional differences in molecular spectrum have been observed. Reasons for a high mortality rate (30%) include lack of awareness of CGD among internists and pediatricians, lack of easy access to immunological tests, and paucity of pediatric HSCT services in our country. Establishment of a nation-wide registry for a rare disease like CGD will be needed in future to decipher precise estimate of disease burden in the country and for better identification of barriers in the existing diagnostic and management strategies.

The data sets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found in the article/ Supplementary Material .

Ethical review and approval was not required for the study on human participants in accordance with the local legislation and institutional requirements. Written informed consent to participate in this study was provided by the participants’ legal guardian/next of kin. Written informed consent was obtained from the individuals for the publication of any potentially identifiable images or data included in this article.

PV, MSu, DS, AJi, AnG, SL, PP, AK, PT, AP, VG, MD, SB, CG, HK, DM, MSi, RR, RU, FN, BG, HL, MKa, AS, SSe, TS, AmG, and SSi—clinical management of patients and follow-up and contribution of clinical data. AR, MSh, JS, GK, SC, BS, RM, VA, MKu, GH, UB, PK, MM, AJa, FN, KV, MD, KC, KI, OO, SN, and YL—laboratory work-up of patients and contribution of laboratory data. PV, MSu, MSh, and SL—preparation of first draft and literature review. AR, PV, and SSi—critical review and editing of manuscript. AR and PV—final approval of manuscript. All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.