Next-generation sequencing of the BALF in the diagnosis of community-acquired pneumonia in immunocompromised patients.

Dear Editor,

We read with great interest the recent publication by Xie et al., addressing the issue of next-generation sequencing (NGS) for diagnosis of severe pneumonia in China. They suggested that NGS might lead to more rapid and accurate diagnosis with better clinical prognosis than conventional detection methods in severe pneumonia in ICU. NGS is a novel method to DNA/RNA sequencing. Due to its high-throughput capacity and fast turnover time, NGS technology has been widely applied to the field of medical microbiology in recent years. NGS has been reported in the application for pathogen detection from blood, cerebrospinal fluid, tissue and intraoperative samples.2, 3, 4 However, no studies have been reported from bronchoalveolar lavage fluid (BALF) samples in community-acquired pneumonia (CAP) patients with immunosuppression. Compared with the immunocompetent host, CAP can be potentially fatal and present significant challenges in immunocompromised host (ICH). In addition to common pathogens, some unusual organisms can cause severe infection in ICH. Besides, multiple concurrent infections are also common in ICH. These characteristics pose a severe challenge to establish a specific diagnosis to guide therapy. A diagnostic tool should be pursued aggressively for favorable outcomes in this population. We recently explored the utility of NGS of bronchoalveolar lavage on the diagnosis of pneumonia in immunocompromised patients.

Thirteen immunocompromised patients with a diagnosis of communty-acquired pneumonia (CAP) were recruited from critical care department of Shanghai Ruijin Hospital. The diagnosis of CAP is based on the presence of select clinical features (e.g., cough, fever, sputum production, and pleuritic chest pain) and is supported by imaging of the lung, usually by chest radiography. Immunosuppression, defined as chemotherapy or neutropenia < 1000 µL during the past 28 days; treatment ≥ 20 mg corticosteroids daily for ≥ 14 days; human immunodeficiency virus infection; immunosuppressive therapy after organ or bone marrow transplantation; and active tuberculosis. Fiberoptic bronchoscopy was performed under local anesthesia with topical lidocaine for patients. BALF specimens using three aliquots of 20 mL of 0.9% saline were obtained. The first was discarded as recommended. The others were pooled and separated into two aliquots. One aliquot was sent to the microbiology lab for direct examination with Gram stain and culture. The other one was sent for DNA extraction and sequencing.

Standard procedures (SP) of BALF microbiologic testing were performed in all patients, which included quantitative cultures for bacteria, mycobacteria and fungi. In addition, multiplex PCR for influenza A/B, parainfluenza 1/2/3/4, coronavirus OC43, coronavirus 229E/NL63, respiratory syncytial virus A/B, adenovirus, human metapneumovirus, human rhinovirus, enterovirus and Human bocavirus was performed on BALF and nasopharyngeal. CMV, EBV, HSV-I, and HSV-II serology for IgG and IgM antibodies were measured.

Clinical characteristics including sex, age, APACHE II score at the time of diagnosis of CAP were demonstrated in Table 1 . All patients were considered to be immunocompromised due to corticosteroids therapy or other immunosuppressive drugs.

Table 1

Clinical characteristics of patients included.

Characteristics	13 patients
Age (years)	52 (46.5–71)
Female/male (n)	5/8
APACHEII score	23 (19–30.5)
Laboratory findings
White blood cells (109 cells/L)	8.7 (5.3–12.1)
Lymphocytes (109 cells/L)	0.4 (0.2–0.99)
CRP (mg/L)	46.6 (19.5–89)
PCT (ng/ml)	1.03 (0.22–4.75)
CD3+T cell count (cells/ul)	215 (141.5–564)
CD3+T %	63.6 (51.75–81.65)
CD3+4+ T cell count (cells/ul)	97 (59.5–181.5)
CD3+4+ T %	31.4 (17.1–39.45)
CD3+8+ T cell count (cells/ul)	89 (63.5–201.5)
CD3+8+ T %	24.9 (21–42.9)
Radiologic findings
Consolidation	7
Interstitial/patchy involvement	8
Immunodeficiency
Nephrotic syndrome	3
Renal transplantation	2
Sjogren Syndrome	2
Adult onset Still's disease	1
Dermatomyositis	1
Rheumatic arthritis	1
Systemic vasculitis	1
Psoriasis	1

SP identified pathogens in 6 patients (6/13), four of which were bacteria, one fungi and one virus detected by clinical respiratory viral PCR panels (Table 2 ). It is worth noting that two samples positive for Acinetobacter baumannii by SP were considered co-infected with PJP or CMV respectively by NGS. NGS detected pathogens in 12 patients (12/13) and one sample with non-infectious etiologies was confirmed by NGS (Table 2). Importantly, five PJP infections and one Aspergillus fumigatus infection were recognized by NGS. Moreover, among the six fungi-positive samples, CMV was also identified in three of them (two PJP and one Aspergillus fumigatus). Being relatively uncommon pathogens among the immunocompromised, PJP were identified in 5 cases and CMV in 4 cases by NGS. However, CMV serology for IgM were negative in these patients. No methenamine silver staining was performed in these cases, for the detecting item has not been carried out in our hospital. Overall, four bacteria, one fungus and one virus were detected by SP. By comparison, four bacteria, seven fungi and five viruses were identified by NGS (Table 2).

Table 2

Next-generation sequencing of BALF for the patients with pneumonia.

Case number	Pathogens identified by SP	Pathogens identified by NGS	Raw Reads	SMRN	Genomic coverage (%)
1	None	Pneumocystis Jiroveci	1883	1810	2.63
		Cytomegalovirus	4	4	6.09
2	None	Pneumocystis Jiroveci	2183	2060	2.85
		Cytomegalovirus	16	16	0.84
3	Human Bocavirus	Human Bocavirus	25	24	43
4	None	Aspergillus Fumigatus	848	608	0.3
		Cytomegalovirus	25834	24833	95
5	Stenotrophomonas Maltophilia	Stenotrophomonas Maltophilia	8641	6153	18
6	Acinetobacter Baumannii	Acinetobacter Baumannii	3322	2415	8.43
		Cytomegalovirus	4	4	0.20
7	Candida Albicans	Candida Albicans	30	28	0.03
8	None	Pneumocystis Jiroveci	96	95	0.15
9	None	Human Simple Virus I	26192	25472	95
10	Acinetobacter Baumannii	Acinetobacter Baumannii	1181	747	1.43
		Pneumocystis Jiroveci	3975	3791	2.32
11	None	Pneumocystis Jiroveci	105	103	0.06
12	Acinetobacter Baumannii	Acinetobacter Baumannii	186860	139486	85.44
13	None	None

SMRN: strictly map reads number.

In terms of bacteria detection, NGS showed no significant advantages in this study. It is noteworthy that our study demonstrated NGS has a distinct advantage in the field of fungi and viral testing, especially PJP and CMV detection.

The definitive PJP diagnosis usually requires demonstration of fungus in tissue, BALF, or induced sputum samples. Methenamine silver staining and PCR in BALF samples have good sensitivity and specificity and is an emerging modality for diagnosis in the correct clinical setting.7, 8 There are a few laboratory tests to detect CMV infection. Histopathologic analysis or viral culture from tissue, blood, urine, or respiratory specimens is less utilized because of poor sensitivity and longer time-consuming. CMV serology for IgG and IgM antibodies is another diagnostic tool but may be negative and inconclusive in immunocompromised hosts. The other methods such as PCR and pp65 antigenemia assays have not been carried out widely. , NGS technology showed its remarkable advantages in terms of opportunistic pathogens in patients with immunocompromised states.

In recent years, clinicians have been increasingly aware of the problem of mixed infection in immunocompromised hosts. However, to diagnosis the co-occurrence of two commonly seen opportunistic infections is difficult in immunocompromised patients with lung infiltrates. It is worth noting the distinct advantage of NGS in concurrent detection of bacteria, fungi and viruses. In our study, standard procedures performed poorly in detecting mixed infections, whereas five cases with coinfection of bacteria, fungi or viruses were identified by NGS. This advantage may help clinicians more comprehensive evaluation of patients and make effective treatment.

In conclusion, our study explored the application of NGS in the microbiologic diagnosis in CAP patients with immunosuppression. NGS technology showed its remarkable advantages in detecting opportunistic pathogens and mixed infection in those patients.

Ethical approval

The study protocol was approved by the Ruijin Hospital Ethics Committee, Shanghai Jiaotong University School of Medicine, China. Formal consent was obtained from the patient or the next of kin.

Funding

This study was supported by the National Key R&D Program of China (2017YFC1309700, 2017YFC1309705), National Natural Science Foundation of China (81772040, 81770005, 81801885) and Shanghai Sailing Program (18YF1413800).

Conflicts of interest

The authors declare no conflicts of interest.