Development of a Novel Fluorescent-Based Lateral Flow Assay for the Detection of Neisseria gonorrhoeae at the Point of Care

Development of a low-cost lateral flow assay for the detection of Neisseria gonorrhoeae Background Neisseria gonorrhoeae (NG) has acquired significant resistance, primarily due to extensive and unwarranted antibiotic utilization over several decades. This resistance has largely been associated with the syndromic management of sexually transmitted infections, particularly in low- and middle-income countries where affordable point of care tests are unavailable. To address this diagnostic gap, FIND has developed a low-cost lateral flow assay for the detection of NG at the point of care. Methods The early performance of the lateral flow assay was evaluated using frozen clinical samples. Limit of detection, inclusivity, and exclusivity studies were performed using well-characterized NG strains, common commensal genital microorganisms, and other Neisseria bacteria. Subsequently, clinical performance was evaluated at 2 sexual health clinics in Birmingham, Alabama. Results The observed limit of detection with reference NG strains was 5 × 103 CFU/mL. Inclusivity was demonstrated for 31 NG strains. Exclusivity testing showed no cross-reactivity with 28 non-Neisseria and nongonococcal Neisseria species; cross-reactivity was observed with Neisseria meningitidis, Neisseria lactamica, and Neisseria polysaccharea. The lateral flow assay demonstrated clinical sensitivity and specificity of 78.6% and 100% in female vaginal swabs and 100% and 89.7% in male urine, respectively. Conclusions FIND has developed a lateral flow assay that aligns with the majority of the World Health Organization Target Product Profile criteria for confirming or excluding NG infection at the point of care. The NG lateral flow assay has now achieved design freeze (final device optimization) and is ready for technology transfer to a manufacturing partner. This test has the potential to support the shift in patient management from a syndromic to an etiological approach.

S exually transmitted infections (STIs) remain one of the most critical global health challenges of the 21st century. 1,2In 2006 and again in 2019, the World Health Organization (WHO) highlighted the importance of a comprehensive STI control strategy, which includes the promotion and provision of prevention strategies, targeted community-based interventions, reliable data to guide the response, and effective clinical services for STI patients with a particular focus on primary health care, sexual and reproductive health, and HIV prevention and care services.Moreover, the WHO emphasizes the value of awareness of STI prevalence within a population, as this promotes a drop in the rate of STI incidence and prevalence in the long run. 3,4mong the most common and manageable STIs are Chlamydia trachomatis (CT) and Neisseria gonorrhoeae (NG). 5,6The WHO estimates that in 2020, 82.4 million new cases of NG infection occurred among adolescents and adults aged 15 to 49 years worldwide, with a global incidence rate of 19 per 1000 female and 24 per 1000 male individuals, with the highest magnitude in WHO Western Pacific and African Regions. 7The WHO has identified NG as a high-priority pathogen because of widespread antimicrobial resistance, including emergent resistance to the "last line" extended-spectrum cephalosporins, cefixime, and ceftriaxone.Patients with gonorrhea-like symptoms (vaginal and urethral discharge) in low-and middle-income countries (LMICs) are mostly seen in primary health care settings, which do not have accessible and affordable diagnostics to confirm NG infection or provide guidance on effective antibiotic treatment regimens.Many STI treatment guidelines still recommend a syndromic-based treatment strategy for NG infection that entails a combination of 2 antibiotics, resulting in many patients being treated for NG without necessarily being infected. 80][11] Depending on disease prevalence and the specific syndromic approach used, for example, the estimates of the proportion of patients with vaginal discharge treated against gonorrhea/chlamydia without actually being infected through the syndromic approach range between 18% and 62%, and the proportion of missed cases between 0.5% and 22% in an analysis of pooled data. 8To improve case detection rates, highly sensitive and specific tests are required to enable effective detection of NG in key populations to (i) diagnose which STI(s) the individual has; (ii) reduce transmission through timely detection and management, particularly for asymptomatic individuals; (iii) provide appropriate antibiotic treatment; and (iv) provide prevalence data to public health agencies.If syndromic evaluation remains the primary approach to guide the treatment of STIs, there is a significant risk of misdiagnosis and antibiotic overuse, leading to further antimicrobial resistance. 12,13ccurate diagnostic tests such as molecular tests for NG are widely used in high-income countries but are too expensive, require substantial infrastructure, or are unavailable in many LMICs.Accessible and affordable diagnostics are needed in LMICs to help guide diagnosis and treatment decisions to foster antibiotic stewardship of existing and new antibiotics.Even in countries where available, testing is often expensive and not widely accessible, and the long turnaround time for results can impede follow-up for care or treatment.Furthermore, the stewardship strategy must fit within a redefined, WHO-supported clinical algorithm that includes the use of diagnostics for patients presenting to primary health care settings.In 2021, the WHO revised its guidelines for the management of vaginal discharge and has recommended that, in settings where same-day treatment is not feasible or available with molecular testing, treatment should be based on a qualityassured rapid test with a minimum sensitivity of 80% and specificity of 90%, if available, to confirm or exclude infection with NG. 9 To address the diagnostic gap in STI stewardship, FIND collaborated with the WHO and the Global Antibiotic Research & Development Partnership to define the user needs and instrument characteristics of diagnostic tests for NG and CT in the form of a Target Product Profile (TPP) for both men and women in LMICs to drive the development of point-of-care (POC) tests for NG and CT intended for low-resource settings. 14To this end, FIND has developed, in collaboration with the contract research organization, DCN Dx, a rapid (<30 minutes), easy-to-use and low-cost (< $3) lateral flow assay (LFA) for detecting NG (NG LFA).This is the first publication to describe the NG LFA design and development with analytical and clinical performance.

Lateral Flow Test Design
Between 2019 and 2022, FIND funded the development of an LFA to detect NG at the POC, which included NG-specific antibody screening, antibody pairing, test optimization, and evaluation.The NG LFA consists of a nitrocellulose membrane, conjugate pad, sample pad, wicking pad, and backing card housed in a custom cassette (Supplemental Digital Content, http://links.lww.com/OLQ/B30).The nitrocellulose membrane contains a test line composed of NG-specific surface antigen capture antibodies and a control line composed of anti-chicken antibodies.To ensure the validity of the result, it is essential for the control line to be detected in all tests.The test line, on the other hand, should only be detected if NG antigens are present.Any result lacking a detectable control line is deemed invalid.The conjugate pad contains a mixture of the NG-specific antibody and chicken antibodies conjugated to latex particle reporters.After adding the premixed sample and buffer to the sample pad, the reagents are allowed to migrate through the conjugate pad for 20 minutes until reaching the absorbent wicking pad at the end of the LFA.The NG LFA test line and control line signal cutoffs are empirically determined for each device lot to differentiate positive and negative samples.A negative result consists of a control line with intensity above the cutoff and a test line below the cutoff, whereas a positive result has both control and test line intensities above the cutoff.
To meet the sensitivity requirements set out in the TPP, the final NG LFA design incorporated a fluorescent europium reporter, which requires a fluorescence reader to interpret the results.A low-cost (<US $50) and easy-to-use portable POC reader was developed in parallel with the NG LFA device development and preliminary testing; however, in the studies presented here, a commercial LFA cassette reader (Axxin AX-2X-S; Axxin, Fairfield, Victoria, Australia) was used at the time of this laboratory evaluation.a

Limit of Detection, Inclusivity, Exclusivity, and Stability
7][18][19] The limit of detection for the NG LFA was empirically determined through a dilution series with 2 reference NG strains (ATCC 49226, Zeptometrix Z017) spiked into negative urine.The analyte concentration that produced a positive result at a statistic frequency of 95% for a minimum of 20 replicates was designated as the limit of detection.
Inclusivity studies were performed using well-characterized NG strains (n = 31) tested 1 Â 10 5 CFU/mL and spiked into a negative urine matrix in duplicate.Each NG strain was diluted serially down to its lowest detectable dilution, as described above.Exclusivity studies were performed on 21 common commensal microorganisms and 10 nongonococcal Neisseria species (n = 31), based on microorganisms tested on recently Food and Drug Administrationcleared assays as well as microorganisms commonly found in urogenital clinical samples (source: Zeptometrix/ATCC).
Clinical & Laboratory Standards Institute EP25-A guidance 20 was used to evaluate the stability of the NG LFA test device, extraction buffer solution in the dropper bottle, and lyophilized controls to predict shelf life and storage conditions for the NG LFA kit.NG LFA devices were assembled and sealed in a pouch with desiccant, and the external controls were lyophilized in vials and sealed with tops.Sample panels consisted of low and medium positive quality control and one negative quality control, prepared and aliquoted for each testing event.Test devices, extraction buffer, and controls were placed at 40°C/70% relative humidity for real-time stability for up to 575 days and 2°C to 8°C and 55°C at 70 days for accelerated stability.The formula used for accelerated stability uses the Arrhenius equation. 20

Laboratory Evaluation
Preliminary NG LFA device performance was characterized with frozen clinical specimens (n = 80).The sample set consists of de-identified biobanked urine samples from Zimbabwe (Biomedical Research and Training Institute), Australia (Victorian Infectious Disease Reference Laboratory), and the United States a Field evaluations with the POC reader currently in progress. 15vel Lateral Flow Assay for Neisseria gonorrhoeae Sexually Transmitted Diseases • Volume 51, Number

Clinical Evaluation
Study Design, Setting, and Participants A cross-sectional pilot study was conducted between January and April 2022 at 2 sites in Birmingham, Alabama.Cisgender female (n = 41) and cisgender male individuals (n = 51) were recruited from treatment-seeking population at the Jefferson County Department of Health STD Clinic and the UAB Sexual Health Research Clinic; the age ranged from 18 to 63 years (mean, 26 years) with 91% Black/African American and 9% White.Symptomatic participants (62%) described abnormal discharge, genital itching or pain, or discomfort or pain during intercourse.This pilot study was designed to accommodate potential changes in device and assay during the final development process and was not powered for formal clinical sensitivity and specificity.
Cisgender female or male individuals, 18 years or older, sexually active, presenting with STI symptoms, or who have been recalled to the clinic with a NG diagnosis were included in the study.Symptoms could include burning sensation or pain upon urination, genital itching or pain, abnormal genital discharge, pelvic pain, or discomfort or bleeding during intercourse.Participants were treated for STIs after a diagnosis based on the validated reference test results only (cobas CT/NG assay on the cobas6800, Roche Diagnostics, Indianapolis, IN).The study was reviewed and approved by the Research Ethics Committee of the University of Alabama at Birmingham.Written informed consent was obtained from all study participants.

Study Procedures
After completing a health questionnaire, 3 vaginal swabs were collected by clinical staff, and male individuals provided 20 mL of first-catch urine.Vaginal swabs were used for female testing because of the significantly better performance of vaginal swabs compared with female urine for molecular tests, 21 which we assumed would be amplified when using an antigen-based assay.The first vaginal swab/urine sample was used for reference testing using the Food and Drug Administration-approved cobas CT/NG assay on the cobas 6800 platform as per the manufacturer's instructions.One vaginal swab/urine sample was used for NG LFA testing, as per Supplementary Figure 1, http://links.lww.com/OLQ/B30.The remaining vaginal swab and urine were stored at −80°C for further testing if required.
Each NG strain was evaluated for inclusivity at the lowest dilution factor, in which both duplicates could still produce a positive NG LFA result.All 31 NG strains produced a positive result for all NG strains at 1 Â 10 5 CFU/mL, with some species still detectable at 1 CFU/mL (Table 2).No cross-reactivity was observed for the 21 microorganisms that most frequently colonize or infect the genital tract or for ~70% (7 of 10) of nongonococcal Neisseria species tested when tested at 1 Â 10^6 CFU/mL (Table, Supplemental Digital Content, http:// links.lww.com/OLQ/B30).However, the NG LFA did cross-react with Neisseria meningitidis, Neisseria lactamica, and Neisseria polysaccharea at 1 Â 10 6 CFU/mL; each titered down to determine the lowest concentration that produced a positive result (Table 3).
The NG LFA stability was evaluated with negative, low positive, and medium positive quality controls.The NG LFA test signal remained stable at 40°C/70% relative humidity at 575 days (19 months) of real-time stability and at 2°C to 8°C and 55°C at 70 days for accelerated stability (Supplemental Digital Content, http://links.lww.com/OLQ/B30).

Laboratory Evaluation
The NG LFA demonstrated promising preliminary performance for frozen male urine clinical specimens (n = 40 positive; 40 negative), where the NG LFA was positive in 39 of 40 (97.5% sensitivity) and negative in 40 of 40 (100% specificity) of the samples tested (Table 4).

Clinical Evaluation
Preliminary clinical sensitivity was calculated from study participants with positive reference test results, and preliminary clinical specificity was calculated from participants with negative reference test results (Table 5).Study participants with an invalid reference test result were excluded from the sensitivity analysis.For discordant results where the reference test is positive and the NG LFA is negative, the NG LFA result was recorded as false negative.For discordant results where the reference test is negative and the NG LFA is positive, the NG LFA test result was recorded as false positive.The NG LFA was positive in 11 of 14 female participants (78.6% sensitivity) with a positive reference test with Ct < 42, and in 11 of 12 (91.7%sensitivity) with Ct < 35.The NG LFAwas positive in 12 of 12 male participants (100% sensitivity) with a positive reference test.The NG LFA was negative in 27 of 27 female participants (100% specificity) and 35 of 39 male participants (89.7% specificity) with a negative reference test result, respectively.

DISCUSSION
This study describes the analytical and preliminary clinical evaluation of a novel POC rapid test for the detection of NG within 30 minutes.Until now, commercial lateral flow tests for NG have shown poor sensitivity and specificity, 18,19,22 whereas NG molecular tests remain more expensive, require more infrastructure, and have a minimal time to result of 90 minutes. 23The NG LFA presents an affordable and easy-to-use POC test for detecting NG that will enable clinicians to identify patients infected with NG during the patient visit.The NG LFA is specifically designed to be performed in nonlaboratory settings, eliminating the need for technical personnel to conduct or interpret the test.A recent study conducted in South Africa has confirmed that the NG LFA is acceptable, convenient, and easy to use in primary health care facilities, particularly among field workers and nurses. 24The ability to test and treat patients during a medical consultation holds great potential for significantly improving patient care. 25romisingly, the NG LFA limit of detection was significantly more sensitive than the predefined analytical specifications.Although the NG LFA limit of detection of 10 3 CFU/mL is less sensitive than NG molecular assays (Xpert CT/NG; Cepheid, Sunnyvale, CA: 1.5 CFU/mL for vaginal swabs and 2.7 CFU/mL for urine), 26 the NG LFA detected the majority of NG-positive cases in this pilot study (79% of swabs, 100% of urine), and the accuracy increased to 92% for female participants with higher bacterial load.The only study to date to quantify bacterial loads in male urine and female vaginal swabs demonstrated a mean bacterial load 100-fold higher (approx.1.14 Â 10 5 CFU/mL) than the presented NG LFA limit of detection.Of note, symptomatic infections were associated with a higher NG bacterial load in urine samples but not in vaginal swabs. 27As a caveat, it should be noted that the NG LFA detects NG surface antigen, which does not necessarily correlate to CFU/mL, which measures viable bacteria.Therefore, the limit of detection as determined by this method does not necessarily represent the true clinical performance for a given strain.In addition, it is unclear whether a cultured NG strain expresses the same amount of antigen as the same strain in human infection.
Because of the high degree of similarity between Neisseria species, one risk associated with testing for NG within the general population is the possibility of cross-reaction with nongonococcal Neisseria species.The current analytical studies showed no cross-reactivity with most known genital commensal bacteria;  however, the NG LFA does cross-react with other Neisseria species such as N. meningitidis.This cross-reactivity may be less clinically relevant for genital or rectal specimens where colonization with commensal nongonococcal Neisseria species is less common; furthermore, the clinical treatment of N. meningitidis and NG are the same in many LMICs. 28However, this may be a risk for oropharyngeal specimens where cross-reactivity can occur with commensal Neisseria species present in the pharynx.The current iteration of the NG LFA was designed for urine and vaginal swabs only; however, given that 70% of NG infections are single-site infections, of which 66% are extragenital, future work will need to confirm the inclusion of other samples, including rectal and oropharyngeal matrices. 27Future evaluation studies could include more geographical locations and a variety of clinical settings, and also be inclusive of trans and intersex populations.
Although the observed specificity for vaginal swabs was 100%, the NG LFA sensitivity of 78.6% fell just below the minimal WHO TPP requirement of >80%.In the present study, the 3 false-negative NG LFA vaginal swab results have PCR reference test Ct > 30, with 2 samples at Ct 41.1 and Ct 38.9 near the upper limit of the reference test detection; these false negatives may be attributed to low bacterial load in vaginal specimens.It bears noting that LFA-PCR concordance is higher for low cycle threshold times (Ct < 30 or Ct < 35, depending on the reference test). 29,30xcluding the results for the 2 samples with Ct > 35, the observed NG LFA sensitivity for vaginal swabs increases to 91.7%.
Although the observed sensitivity for male urine was 100%, the NG LFA specificity of 87.9% did not meet the minimal WHO TPP requirement of >95%.Specificity can be affected by interfering substances or the presence of microorganisms expressing the antigen detected by the lateral flow antibodies. 31The subset of commensal urogenital bacteria tested indicated no crossreactivity (Supplemental Digital Content, http://links.lww.com/OLQ/B30); however, potential interference of exogenous and endogenous substances has not been evaluated.
One significant limitation of this study was the small sample size to allow for the findings to be accurately extrapolated.A recent similar clinical evaluation of the NG LFA in South Africa demonstrated superior performance with sensitivities of 96.1% and 91.7% and specificities of 97.2% and 96.3% in 200 symptomatic males and 200 symptomatic female individuals, respectively (in press). 32he reason for variations in estimated performance between the 2 sites is unknown but may be related to differences in access to services in the 2 settings (more routine diagnostic testing is available in the US setting than in South Africa), which may suggest that infections in the South African population are at a different stage regarding organism load.This reflects the need for additional studies in diverse populations as mentioned previously.
Based on the preliminary performance assessed by these studies, future work should advance to in-country field evaluations in LMICs for asymptomatic populations and other specimen types as noted previously.Ongoing field evaluation is a critical step in supporting the development of these technologies to meet the needs of LMICs.
Currently, POC testing for NG is unavailable in primary health care centers in LMICs.The development of low-cost and easy-to-use technologies would be a significant advancement to decentralize diagnostic testing for STIs-particularly where molecular systems are not available, not affordable, or where infrastructure limitations would inhibit their use.The NG LFA presented here has the potential to confirm or exclude NG infection at the POC in virtually any setting and support the shift in STI patient management guidelines from a syndromic to an etiological approach.The NG LFA developed by FIND is now ready for technology transfer to a commercial manufacturing partner with the aim of having a product on the market by 2024.
3, March 2024(SLR Research in Carlsbad, CA).All samples were frozen raw without preservatives or transport media.Clinical samples were characterized by NG status, including the cycle threshold (Ct) of real-time polymerase chain reaction (PCR) results, sample age, sample site origin, and patient symptom status if available.NG LFA testing included 2 replicates per sample, using one lot of NG LFA devices.The testing procedures are outlined in Supplementary Figure1, http://links.lww.com/OLQ/B30, and are the same procedures used for the clinical performance evaluation.Measurement and interpretation of the test result were performed using the Axxin AX-2X-S commercial lateral flow test reader.

TABLE 1 .
NG Lateral Flow Assay Limit of Detection

TABLE 2 .
Inclusivity Evaluation of 31 N. gonorrhoeae Stains and Their Limit of Detection *Strains provided by Peter Rice, UMass.

TABLE 4 .
Performance of N. gonorrhoeae Lateral Flow Assay When Tested With Archived Frozen Urine Clinical Specimens

TABLE 5 .
Diagnostic Performance of the NG LFA When Compared With cobas CT/NG Assay in Female Vaginal Swabs and Male Urine *Ct values: 33.8, 38.9, 41.1.Novel Lateral Flow Assay for Neisseria gonorrhoeaeSexually Transmitted Diseases • Volume 51,Number 3, March 2024