Community-acquired pneumonia (CAP) is one of the most common and serious infections in children worldwide, potentially leading to severe complications such as sepsis, acute respiratory distress syndrome, or death (Vaughn et al., 2024). The etiological spectrum of CAP is broad, encompassing a variety of bacteria and viruses (Liu et al., 2023). Among bacterial pathogens, Haemophilus influenzae and Streptococcus pneumoniae are the two most frequently implicated species. Furthermore, these two pathogens, along with Neisseria meningitidis, are leading causes of invasive diseases, including bacteremia, pneumonia, and meningitis. They are also common agents of secondary infections following viral respiratory disease and frequent co-pathogens with Mycoplasma pneumoniae (Brueggemann et al., 2021).

H. influenzae is a Gram-negative, non-motile bacterium first isolated from the nasopharynx of patients during an influenza epidemic (Wen et al., 2020). S. pneumoniae is a Gram-positive, encapsulated diplococcus pathogen first identified from the sputum of a pneumonia patient in 1881 (Zivich et al., 2018). Both organisms are transmitted via respiratory droplets from infected or colonized individuals and subsequently colonize the mucosal surfaces of the nasopharynx and upper airway. This colonization can lead to a range of infections, including otitis media, epiglottitis, sinusitis, pneumonia, meningitis, and bacteremia (Revai et al., 2008). The identification of S. pneumoniae and H. influenzae is also critical for guiding appropriate antibiotic therapy. A recent study demonstrated that antibiotics were of little benefit for children with acute sinusitis if these bacterial pathogens were absent in the nasopharynx, suggesting that accurate testing for these two bacteria could help reduce unnecessary antibiotic use (Shaikh et al., 2023). Given the clinical significance of these pathogens, the development of accurate and rapid molecular detection methods is crucial for timely diagnosis and treatment.

Despite being the traditional gold standard for diagnosis, bacterial culture often yields a lower positive rate for these two fastidious organisms compared to other pathogens, which is a major constraint in many clinical laboratories (Farajzadeh Sheikh et al., 2021; Sharma et al., 2021). Recombinase-aided amplification (RAA) assay is an isothermal amplification technology known for its high specificity, sensitivity, and portability. The RAA system is composed of three key proteins: a recombinase (which anneals primers to the template DNA), a single-strand DNA binding protein (SSB), and a DNA polymerase (for amplification and extension). The coordinated action of these components with specific primers and fluorescent probes enables a rapid and specific reaction, making the technology highly suitable for point-of-care testing (POCT) and the detection of fastidious bacteria. Recently, RAA has been successfully used to detect various microbial pathogens, such as Mycoplasma pneumoniae, Monkeypox virus, and Candida auris (Xue et al., 2020; Cui et al., 2023; Feng et al., 2024).

In this study, we developed a duplex RAA assay for the simultaneous detection of S. pneumoniae and H. influenzae in a single tube. We then validated the application of this method using invasive clinical samples and compared its performance against bacterial culture and a previously published real-time PCR method.

Streptococcus pneumoniae and Haemophilus influenzae are common inhabitants of the human nasopharyngeal microflora but can migrate to sterile sites to cause invasive disease (Eichner et al., 2021). Recent metagenomic studies of BALF from children have identified S. pneumoniae and H. influenzae as dominant pathogens (Yang et al., 2022). Similarly, advanced sequencing of cerebrospinal fluid in resource-limited settings has shown S. pneumoniae to be a primary cause of bacterial meningitis (Pallerla et al., 2022). Given the clinical importance of these bacteria, a rapid and accurate detection method applicable directly to clinical samples is essential for early diagnosis. Any diagnostic delay can increase mortality, prolong hospital stays, and elevate healthcare expenditures (Pallerla et al., 2022).

Since both pathogens can asymptomatically colonize the nasopharynx, a positive result from noninvasive samples (e.g., pharyngeal swabs) can make it difficult to distinguish between pathogenic infection and carriage. To address this diagnostic ambiguity, our study focused on developing a duplex recombinase-aided amplification (RAA) assay and evaluating it on invasive clinical samples (BALF and CSF). This approach provides a more conclusive diagnostic basis for clinicians. The RAA assay achieves amplification in approximately 10 min, significantly reducing the turnaround time compared with real-time PCR and enhancing its utility in clinical settings.

For developing robust diagnostic methods, selecting highly conserved target sequences is crucial to ensure comprehensive detection across numerous serotypes. Given the genomic variability among different genotypes and serotypes of these pathogens, this initial step is critical for accuracy. Previous research on S. pneumoniae has utilized various gene targets, including those encoding pneumolysin (ply), autolysin (lytA), Spn9802, capsular polysaccharide biosynthesis proteins (cps), and pneumococcal surface antigen A (psaA) (Blaschke, 2011; Wessels et al., 2012; Chang et al., 2021). Notably, the lytA gene, which encodes a cell wall hydrolase, is a recognized virulence factor due to its role in releasing highly inflammatory cell wall components and pneumolysin. Consequently, it has been repeatedly validated as an effective and reliable detection target (Seki et al., 2005). For H. influenzae, specific genes such as the outer membrane protein P6 gene (omp6) and 16S rRNA have been successfully targeted (Maleki et al., 2020; Wang et al., 2022). For instance, Cao et al. established an isothermal amplification technology based on the omp6 gene, achieving sensitive detection by coupling multiple cross displacement amplification with a nanoparticle-based lateral flow biosensor (Cao et al., 2021).

Informed by these previous studies, we selected the lytA and omp6 genes as target markers for our assay. Sequence alignments confirmed that these genes exhibit high sequence homology across diverse serotypes. Consequently, primers and probes were designed based on these conserved regions. The resulting single-tube duplex assay demonstrated high specificity for various strains of H. influenzae and S. pneumoniae, with no cross-reactivity observed against other common respiratory pathogens. The analytical sensitivity was determined to be 72 copies per reaction for S. pneumoniae (lytA) and 35 copies per reaction for H. influenzae (omp6). This performance is comparable to the sensitivity of previously reported real-time PCR assays, which typically ranges from 10 to 100 copies per reaction (Abdeldaim et al., 2009; Meyler et al., 2012; Hajia et al., 2014), indicating that our RAA assay exhibits competitive detection sensitivity.

To evaluate the clinical utility of this method, we analyzed 168 invasive clinical samples. The results from both the RAA and real-time PCR assays demonstrated 100% concordance, with positive detection rates of 10.71% for S. pneumoniae and 12.5% for H. influenzae. In contrast, the positive rates for bacterial culture were only 5.36% and 4.17%, respectively. Notably, for samples that were culture-negative but RAA-positive, the amplification dynamics suggested a low bacterial load, likely below the detection threshold of conventional culture. This finding highlights the superior sensitivity of the RAA assay. Furthermore, we observed a high rate of co-infection: 33.3% (6/18) of the S. pneumoniae-positive samples were also co-infected with H. influenzae. As previous studies have indicated, co-infection with these two pathogens can exacerbate pneumonia and increase mortality in pediatric patients (Pneumonia Etiology Research for Child Health (PERCH) Study Group, 2019). These findings underscore the clinical importance of simultaneous detection.

However, this study has limitations. The primary limitation is the qualitative nature of the RAA assay. Although the amplification curve can provide a semi-quantitative indication of bacterial load, it does not offer the precise quantification achievable with methods such as real-time PCR.

In summary, we have developed a duplex RAA assay for the simultaneous detection of H. influenzae and S. pneumoniae in clinical samples, demonstrating high specificity and sensitivity. This assay offers significant advantages, including simplified reaction conditions, a reduced turnaround time, and lower cost compared with traditional culture methods and real-time PCR. Therefore, it holds substantial potential for clinical application, particularly for the rapid analysis of invasive specimens.