This study was approved by the Ethics Committee of Weifang Second People's Hospital (Weifang, China; approval no. KY2024-023-01; approval date: March 15, 2024) and conducted in strict accordance with the principles of the Declaration of Helsinki. Written informed consent was obtained from all patients prior to BALF sample collection and study inclusion; for patients unable to provide consent due to critical illness (e.g., severe respiratory failure requiring mechanical ventilation or altered consciousness), consent was obtained from their legal representatives. All participants were informed about the purpose of the study, the procedures involved, potential risks and benefits, and their right to withdraw at any time without affecting their clinical care. Patient data were anonymized and de-identified prior to analysis to ensure confidentiality.

PJ infection was defined based on the following composite criteria: i) Clinical manifestations consistent with PCP, including progressive dyspnea, nonproductive cough, fever and hypoxemia; ii) characteristic chest imaging findings of diffuse bilateral ground-glass opacities, interstitial infiltrates or cystic lesions on computed tomography (CT); and iii) laboratory confirmation of PJ nucleic acid detection by tNGS from BALF samples. Patients meeting all three criteria were classified as confirmed PJ infection cases. Patients with negative tNGS results and no clinical or radiological evidence of PCP were classified as non-PJ controls.

Inclusion criteria were as follows: i) Patients aged ≥18 years who underwent bronchoscopy and BALF collection for suspected lower respiratory tract infection at Weifang Second People's Hospital between March 18, 2024 and December 31, 2024; ii) BALF samples with sufficient volume (≥3 ml) and quality for nucleic acid extraction and molecular testing; iii) complete clinical, laboratory and imaging data available for review; iv) for PJ-positive cases, confirmation by tNGS detection of PJ with consistent clinical and radiological findings. Exclusion criteria were as follows: i) BALF samples with insufficient volume (<3 ml) or visible blood contamination exceeding 50% of the sample volume; ii) patients with incomplete medical records or missing tNGS results; iii) samples stored for >72 h at 4˚C or subjected to repeated freeze-thaw cycles (>2 cycles) prior to testing; iv) patients receiving empirical anti-PCP therapy (trimethoprim-sulfamethoxazole, pentamidine or atovaquone) for >72 h before BALF collection; and v) concurrent diagnosis of other fungal infections (e.g., invasive aspergillosis, mucormycosis) that could confound the clinical presentation. The participant selection process is summarized in Fig. 1.

A total of 44 BALF samples positive for PJ by tNGS and 17 BALF samples negative for PJ by tNGS were collected. All samples were processed within 2 h of collection.

Pathogen controls. Genomic DNA of common clinically relevant pathogens, including Acinetobacter baumannii, Klebsiella pneumoniae, Pseudomonas aeruginosa, Proteus mirabilis, Stenotrophomonas maltophilia, Enterobacter aerogenes, Enterobacter cloacae, Serratia marcescens, Candida albicans and Candida krusei, was obtained from the Clinical Microbiology Laboratory of Weifang No. 2 People's Hospital (Weifang, China), having been isolated from the patient's clinical specimen during diagnostic examination.

Kits and instruments. The RAA Nucleic Acid Amplification kit (Conventional Method; cat. no. B00000A-96; lot no. BH7OA0) and RAA Nucleic Acid Amplification kit (Fluorescent Method; cat. no. F00001A-96; lot no. FH9KD0) were purchased from QT Biotech Co., Ltd. Thermo Scientific™ PCR Premix (2X) (cat. no. K0172) was purchased from Thermo Fisher Scientific, Inc. The MGIEasy Nucleic Acid Extraction Kit (lot no. 111942507003; cat. no. RM0766) was purchased from BGI Genomics Co., Ltd. The Equalbit 1X dsDNA HS Assay Kit (cat. no. EQ121-02; lot no. 7E1352F5) was purchased from Vazyme Biotech Co., Ltd. DL1500 Ladder (100 bp to 1,500 bp; cat. no. MD1019) was purchased from Simgen Biotechnology Co., Ltd. Instruments included a Qubit 4 Fluorometer (Thermo Fisher Scientific, Inc.), CFX96™ Real-Time PCR System (Bio-Rad Laboratories, Inc.), T100™ Thermal Cycler (Bio-Rad Laboratories, Inc.) and ChemiDoc™ Imaging System (Bio-Rad Laboratories, Inc.).

Nucleic acid extraction. Nucleic acid extraction was performed using the MGIEasy Nucleic Acid Extraction Kit, with steps as follows: i) Reagent and sample preparation: Lysis buffer: Antifoaming agent was added to lysis buffer; Wash Buffer A/B: Anhydrous ethanol was added to Wash Buffer A/B, respectively. For BALF samples: Transferred 350 µl of BALF to a clean 1.5 ml centrifuge tube. ii) Preparation and addition of extraction reaction mix for each sample. iii) Magnetic bead extraction. Viscous residues were homogenized and the mixture was then incubated at 56˚C for 10 min. Isopropanol and magnetic beads were added and after vortexing, the mixture was left to stand at room temperature for 10 min (invert every 3 min). Samples were centrifuged, placed on a magnetic rack and the supernatant was discarded. Beads were washed with Wash Buffer A/B sequentially and then air-dried at room temperature, followed by addition of nuclease-free water and elution of nucleic acid, which was stored at -20˚C. iv) Quality control. The DNA concentration was measured with the Equalbit 1X dsDNA HS Assay Kit (Vazyme Biotech Co., Ltd.) and Qubit 4 Fluorometer (Thermo Fisher Scientific, Inc.).

Samples were considered to meet the requirements for subsequent library preparation if their DNA concentration, as determined by NanoDrop DS-11, was ≥1 ng/µl (total amount ≥10 ng), and if the A260/A280 ratio and A260/A230 ratio, as measured by spectrophotometry (NanoDrop DS-11; Thermo Fisher Scientific, Inc.), were 1.8-2.0 and ≥1.8, respectively.

Construction of recombinant plasmid standard. A 149 bp conserved region of the PJ major surface glycoprotein gene (GenBank ID: KP168716.1) was synthesized and cloned into the pUC57 vector by Sangon Biotech, generating a recombinant plasmid. The plasmid was serially diluted to concentrations ranging from 10⁴ to 10 copies/µl with nuclease-free water. The number of copies/µl was calculated using the following formula:

Primer and probe design. The 149 bp conserved region of the PJ major surface glycoprotein gene was selected as the target for RAA development. Specific primers and an exo probe for RAA were designed using Oligo 7 software (Molecular Biology Insights, Inc.). The probe was labeled with a FAM fluorophore at the 5' end and a BHQ-1 quencher at the 3' end, and a tetrahydrofuran spacer was included to facilitate cleavage by the exonuclease. All oligonucleotides were synthesized by Sangon Biotech (Shanghai, China). Sequences are listed in Table I.

Specificity of primers and RAA assay. To evaluate the specificity of PJ-RAA-forward(F)/reverse (R) primers and confirm the amplification capability of the fluorescent RAA assay for PJ, conventional PCR was performed. Conventional PCR used genomic DNA templates including 8 common respiratory bacteria, 2 opportunistic fungi and a PJ-positive BALF sample. Nuclease-free water served as the negative control (NC) for both assays.

Workflow of tNGS. tNGS was performed using the PTseq Respiratory Pathogen Targeted High-Throughput Genetic Detection Kit (lot no. 513392407006; cat. no. RM1261; MGI Tech Co., Ltd.) on the MGI DNBSEQ-G99 platform according to the manufacturer's instructions. The tNGS workflow encompasses sample collection, nucleic acid extraction, multiplex targeted enrichment, library construction and bioinformatics analysis, covering a total of 274 common respiratory pathogens, including 72 bacteria, 95 fungi, 50 viruses and 15 other pathogens. Bioinformatics analysis was performed using the MGI platform to align reads against the reference genome. A positive result was defined as detection of 13 or more unique reads with >95% coverage of the target region.

The total volume of the PCR reaction system was 25 µl. Each reaction contained 12.5 µl of Thermo Scientific™ PCR Premix (2X) (containing Taq DNA polymerase, dNTPs and reaction buffer), 2 µl of 10 µM PJ-F primer, 2 µl of 10 µM PJ-R primer, 2 µl of template DNA (genomic DNA extracted from PJ-positive samples, non-PJ pathogens or NC) and 6.5 µl of nuclease-free water to bring the volume to 25 µl.

PCR amplification program. Thermal cycling was carried out on a PCR instrument with the following parameters: First, pre-denaturation was performed at 94˚C for 10 min to fully denature genomic DNA. This was followed by 30 amplification cycles, each consisting of denaturation at 94˚C for 30 sec (to separate double-stranded DNA), annealing at 60˚C for 30 sec (to enable specific binding of primers to target sequences) and extension at 72˚C for 30 sec (to synthesize new DNA strands via Taq polymerase). After the cycling process, a final extension step was conducted at 72˚C for 6 min to ensure complete extension of all amplified products, and the reaction was then held at 4˚C indefinitely to preserve the amplified products.

Agarose gel electrophoresis. PCR products were analyzed by agarose gel electrophoresis to confirm amplification specificity. A 1% agarose gel was prepared using 1X Tris-acetate-EDTA (TAE) buffer, with ethidium bromide added for subsequent nucleic acid visualization. Electrophoresis was run at a constant voltage of 90 V; the running time was adjusted based on gel separation efficiency, typically 20-30 min, until DNA bands were clearly distinguishable. The PCR products were mixed with 6X DNA loading buffer. For sample loading, 1.3 µl of DNA Marker (GeneRuler 100 bp DNA Ladder; cat. no. SM0241; lot no. 3498729; Thermo Fisher Scientific, Inc.) and 4 µl of PCR product from each reaction were loaded into separate wells of the gel. After electrophoresis, gel images were captured using the ChemiDoc™ Imaging System (cat. no. 12003153; Bio-Rad Laboratories, Inc.), and the presence or absence of target bands was used to evaluate the specificity of the PJ primers.

The RAA assay was performed in a 50 µl reaction volume using an RAA Nucleic Acid Amplification kit according to the manufacturer's instructions. For fluorescent RAA amplification, the system included 25 µl of RAA reaction buffer, 2.6 µl of the forward primer PJ-RAA-F (10 µM) and 2.6 µl of the reverse primer PJ-RAA-R (10 µM), 0.8 µl of PJ-RAA-Probe (10 µM), 12 µl of nuclease-free water, 5 µl magnesium acetate and 2 µl of template DNA (25 ng/µl). For conventional (electrophoresis-based) RAA amplification, the system excluded the 0.8 µl of PJ-RAA-Probe (10 µM), and the volume of nuclease-free water was adjusted to 12.8 µl to maintain a total reaction volume of 50 µl. Fluorescent reactions were incubated at 40˚C for 20 min on the CFX96™ Real-Time PCR System, with fluorescence signals recorded every 20 sec (corresponding to 1 cycle per 20 sec).

Optimization of reaction conditions. Key parameters were optimized using DNA extracted from clinical samples as the template, with tests performed across a series of conditions: Temperatures of 37, 38, 39, 40, 41 and 42˚C (each with a 20 min incubation). For fluorescent RAA, samples that gave rise to an exponential amplification curve above the NC threshold within 20 min were judged as positive. Optimal reaction conditions were defined as the combination that yielded the earliest amplification initiation and the fastest plateau phase arrival.

For sensitivity evaluation, the recombinant pUC57-PJ plasmid was serially diluted to generate a concentration gradient ranging from 10⁴ to 10¹ copies/µl (10⁴, 10³, 10², 10) for fluorescent RAA assays. Fluorescent RAA was performed under the optimized reaction conditions. All dilutions were tested in triplicate to ensure reproducibility.

Clinical validation. The optimized fluorescent RAA assay was used to detect 44 PJ-positive and 17 PJ-negative BALF samples. Diagnostic performance indicators, including sensitivity, specificity, positive predictive value, negative predictive value, total agreement and κ value (κ: 0.61-0.80=substantial agreement; 0.41-0.60=moderate agreement) were calculated using SPSS 26.0 (IBM Corp.).

Data were analyzed using ImageJ (version 1.52a; National Institutes of Health) and SPSS 26.0. The clinical diagnostic performance of the RAA assays was evaluated, with tNGS used as the reference standard. Analyses were based on a 2x2 contingency matrix using the chi-square test. A two-tailed P<0.05 was considered to indicate statistical significance.

Conventional PCR using PJ-RAA-F/R primers amplified a 149 bp fragment from PJ-positive BALF DNA (Fig. 2A), confirming the validity of the target region within the PJ major surface glycoprotein gene. A specific amplification band was only visible in the PJ-positive sample, while no bands were detected in the NC and all non-PJ pathogens. This result confirmed the high specificity of the primers for PJ detection. NGS of pUC57-PJ showed 100% homology with the reference PJ major surface glycoprotein sequence gene (GenBank ID: KP168716.1), confirming successful plasmid construction (Fig. 2B).

Fluorescence amplification was successfully detected across all tested temperatures (37-42˚C) in the detection of 3 PJ-positive BALF samples. At temperatures <40˚C (37, 38 and 39˚C), amplification initiation was relatively delayed, and the arrival time to the plateau phase was prolonged. By contrast, the system exhibited the earliest amplification initiation (mean: 20 cycles) and the fastest arrival at the plateau phase at 40˚C (Fig. 3), demonstrating the most robust and efficient amplification kinetics. This superior performance could be attributed to the highest and most stable recombinase activity at 40˚C in the fluorescent RAA system, which ensured consistent and rapid nucleic acid amplification. When the temperature exceeded 40˚C (41 and 42˚C), the amplification curves showed suboptimal morphological characteristics, with slower overall kinetics compared to those at 40˚C. Collectively, these results indicated that 40˚C was the optimal amplification temperature, as it balanced both the speed of amplification initiation and the efficiency of plateau phase arrival.

To evaluate the sensitivity of the fluorescent RAA assay for PJ detection, serial dilutions of the PJ-positive plasmid (10⁴, 10³, 10² and 10 copies/µl) were used as templates for amplification under the optimized reaction conditions, and the corresponding curves are shown in Fig. 4. The amplification curves exhibited a concentration-dependent trend: The template with the highest concentration (10⁴ copies/µl) initiated amplification earliest (within 5-10 cycles) and produced the strongest final relative fluorescence unit (RFU) signal. As the plasmid concentration decreased (10³, 10² and 10 copies/µl), the initiation cycle of amplification was gradually delayed, and the final RFU signal was sequentially reduced. Notably, even the lowest concentration template (10 copies/µl) still generated a detectable amplification signal (crossing the baseline within 25 cycles), indicating that the fluorescent RAA assay has good sensitivity and can effectively amplify PJ target sequences at low copy numbers.

Clinical validation. To evaluate the diagnostic performance of fluorescent RAA, 44 BALF samples from patients suspected of PCP were tested with tNGS as the reference standard. Among 44 PJ-positive samples confirmed by tNGS, 36 were detected by fluorescent RAA, and no false-positive results were observed in 17 PJ-negative samples (100% specificity for all assays). Diagnostic efficacy metrics (Table II) showed that fluorescent RAA achieved 81.82% sensitivity, 86.89% overall agreement and a κ value of 0.715 (P<0.001). The fluorescent RAA assay exhibited good rule-out diagnostic value, coupled with excellent clinical diagnostic performance. Among the 61 patients, 48 (78.7%) were male and 13 (21.3%) were female, with a mean age of 70.2±9.8 years. In the PJ-positive group, 36 (81.8%) were male and 8 (18.2%) were female, with a mean age of 70.4±10.8 years; in the PJ-negative group, 12 (70.6%) were male and 5 (29.4%) were female, with a mean age of 69.5±6.9 years. The characteristics of the cohort are summarized in Table III. There were no significant differences between the two groups in terms of gender or age (P>0.05).