BRIDGING THE GAP BETWEEN DEVELOPMENT OF POINT-OF-CARE NUCLEIC ACID TESTING AND PATIENT CARE FOR SEXUALLY TRANSMITTED INFECTIONS

Abstract

The rate of sexually transmitted infections (STIs), including the four major curable STIs – chlamydia, gonorrhea, trichomoniasis and, syphilis – continue to increase globally, causing medical cost burden and morbidity especially in low and middle-income countries (LMIC). There have seen significant advances in diagnostic testing, but commercial antigen-based point-of-care tests (POCTs) are often insufficiently sensitive and specific, while near-point-of-care (POC) instruments that can perform sensitive and specific nucleic acid amplification tests (NAATs) are technically complex and expensive, especially for LMIC. Thus, there remains a critical need for NAAT-based STI POCTs that can improve diagnosis and curb the ongoing epidemic. Unfortunately, the development of such POCTs has been challenging due to the gap between researchers developing new technologies and healthcare providers using these technologies. This review aims to bridge this gap. We first present a short introduction on the four major STIs, followed by a discussion on the current landscape of commercial near-POC instruments for the detection of these STIs. We present relevant research toward addressing the gaps in developing NAAT-based STI POCT technologies and supplement this discussion with technologies for HIV and other infectious diseases, which may be adapted for STIs. Additionally, as case studies, we highlight the developmental trajectory of two different POCT technologies, including one approved by the United States Food and Drug Administration (FDA). Finally, we offer our perspectives on future development of NAAT-based STI POCT technologies.

Graphical Abstract

Toward tackling ongoing sexually transmitted infections epidemic, this review aims to bridge the gap between researchers who develop nucleic acid amplification-based point-of-care tests and end-users who need such tests.

1. INTRODUCTION

The 21st century has seen a global resurgence of sexually transmitted infections (STIs),¹ which refer to a variety of clinical syndromes and infections caused by pathogens that can be acquired and transmitted through sexual activity. The incidence of STIs, including the four major curable STIs – chlamydia, gonorrhea, trichomoniasis, and syphilis is increasing by over 1 million new cases daily and represents a global public health crisis.² Although many STIs (especially bacterial STIs) are curable, they result in a substantial burden of disease. The complications of STIs include pelvic inflammatory disease, ectopic pregnancy, infertility, chronic pelvic pain, seronegative arthropathy, and neurological and cardiovascular diseases. STIs in pregnancy can cause fetal or neonatal death, premature delivery, and neonatal encephalitis, eye infections, and pneumonia. STIs can also increase the risk of acquisition and transmission of HIV.³ STIs disproportionally affect low- and middle-income countries (LMIC), developing countries, and marginalized groups such as commercial sex workers and men who have sex with men (MSM).⁴ Stigma and disparaging moral attitudes toward STIs have resulted in reluctance (or even neglect) to prioritize STI control policies.⁵

In response to this crisis, the World Health Organization (WHO) has established the Global STI Strategy⁶, aimed at ending STI epidemic as a major public health threat by 2030⁷. The development of rapid, accurate, easy to use, and low-cost STI point-of-care tests (POCTs) represents a critical component of this strategy. Such STI POCTs that can rapidly and accurately diagnose STIs could facilitate the etiological management (i.e., based on causal and treatable pathogens) of STIs over syndromic management (i.e., based on symptoms),⁸ which lacks the specificity and sensitivity to reduce disease burden. Furthermore, syndromic management of STIs has also been associated with overtreatment of non-infected patients, as well as with the emergence of antimicrobial resistance (AMR), resulting from the treatment with unnecessary or ineffective antibiotics.⁹ As such, STI POCTs could help to determine antimicrobial susceptibility, permitting precision treatment, and help reduce the emergence of AMR.

Previously, STI POCTs generally refer to antigen- or antibody-based “rapid diagnostic tests” (RDTs). Though low cost, fast, and easy to use for the detection of syphilis,¹⁰ chlamydia,¹¹ or gonorrhea,¹² RDTs are less sensitive than nucleic acid amplification tests (NAATs). However, given their cost, sophistication, and infrastructure requirements, NAATs are typically performed in centralized laboratories and out of the reach for LMIC. Moreover, the test turnaround time (TAT) of most NAATs can take 2 – 4 days, requiring patients to return for test results on a subsequent clinic visit, which often leads to significant loss to follow-up. Lately, commercial instruments such as the Cepheid GeneXpert® platform have brought NAATs closer to patients with STIs. For example, the Cepheid GeneXpert® platform can detect chlamydia and gonorrhea with acceptable sensitivity and specificity. Unfortunately, it requires several steps, takes 90 minutes to obtain result, and requires a stable source of electricity, thus limiting its utility as a practical STI POCT. Furthermore, its cost presents a significant barrier for uptake in LMIC.¹³ Thus, there remains significant needs for new NAAT-based STI POCT technologies, especially those that can be practically deployed in LMIC.

Despite the clear and urgent needs, the development of NAAT-based STI POCTs has met significant challenges – evident by the scarcity of successful demonstration and use of such diagnostics. We recognize that the challenges arise from the gap between researchers developing new technologies and healthcare providers using these technologies. The development of STI POCT technologies that are affordable, sensitive, specific, easy to perform, and deliverable to those who need them is indeed challenging. But equally challenging are barriers at the patient, provider, and health system levels, which can prevent both implementation and uptake of STI POCTs at POC settings such as primary care clinics.⁸ Bridging the two sides – researchers and providers – would be immensely helpful for accelerating the development of NAAT-based STI POCTs and facilitating their uptake and widescale use at POC settings.

This review aims to bridge the gap between development of NAAT-based STI POCT technologies and patient care for STIs, with a special emphasis on LMIC. To do so, we begin with a concise primer on major STIs, followed by an introduction on the current landscape of commercial NAAT-based near-POC instruments. We then briefly discuss the research gaps foci, and priorities viewpoints from both the POCT technology development community and the STI research community. Next, we highlight current research toward developing NAAT-based STI POCTs and provide in-depth case studies of two NAAT-based STI POCTs – one from industry and one from academia. Finally, we offer our perspectives on the development of POC tests with the goal of facilitating future developments. Our article therefore complements current reviews that emphasize on either technology development¹⁴ or STIs^15–19, and together offer a holistic view on bringing NAAT-based STI POCT technologies to patient care.

2. PRIMER FOR MAJOR CURABLE STIs

We begin by briefly introducing the four major curable STIs, which are discussed throughout this review. This introduction focuses on the causative pathogen, the global incidence (i.e., new case) of the disease, the global prevalence (i.e., total current case) of the disease, and the standard diagnostic methods (Figure 1). For a more in-depth introduction of these STIs, the work by Workowski provides an easy-to-read primer for these STIs.²⁰ In addition to the four major curable STIs, we include HIV in this review because significant research efforts toward developing NAAT-based HIV POCTs have resulted in considerable technological advances, which can serve as a roadmap for the development of NAAT-based POCTs for these four STIs.

Figure 1. Overview of the 4 major curable STIs and HIV.

The 4 major STIs – chlamydia, gonorrhea, trichomoniasis, and syphilis – and HIV impact millions of new patients per year. Here, global incidence refers to the number of new cases in 2020 while global prevalence refers to the number of total current cases as of 2020, both estimated by the World Health Organization.²² These STIs are currently diagnosed via commercial benchtop (or robotic) nucleic acid amplification tests (NAATs) or serological tests (syphilis and HIV).

2.1. Chlamydia

Chlamydia is caused by infection with Chlamydia trachomatis (CT), a Gram-negative obligate (i.e., requires oxygen for growth) intracellular bacterium whose only natural host is humans.²¹ The WHO estimated that in 2020 128 million new cases of chlamydia were diagnosed worldwide,²² making it the most prevalent bacterial STI. The WHO also estimated that as of 2020, there are 128 million prevalent cases of chlamydia worldwide.²² The majority of persons with genital CT infections are asymptomatic²¹ and, as a result, detection of infection often relies on screening.²³ NAATs are considered the gold standard for detection of CT, which are commercially available from Abbott Laboratories, BD Biosciences, Hologic, Roche Diagnostics, and others.

2.2. Gonorrhea

Gonorrhea is caused by infection with Neisseria gonorrhoeae (NG), a Gram-negative diplococcus bacterium. The WHO estimated that in 2020 82 million new cases of gonorrhea were diagnosed worldwide,²² making it the second most prevalent bacterial STI globally. The WHO also estimated that as of 2020, there are 29 million prevalent cases of gonorrhea worldwide.²² NG infections can also be asymptomatic, especially for women.^{24, 25} Co-infections with CT and NG are common. NAATs are recommended for detection of infections caused by NG with and without symptoms.²⁶

AMR in NG is particularly concerning,^{27, 28} as NG has developed resistance to all previously prescribed antibiotics. Consequently, gonorrhea might even be evolving into an untreatable infection in certain circumstances,²⁹ which is a major global public health concern. For example, NG isolates displaying resistance or decreased susceptibility to the currently-recommended treatment option, cephalosporins, have been reported globally.²⁷ In response to this concern, the WHO published a global action plan in 2012³⁰ which considers NG a high priority antibiotic-resistant bacteria for research and development of new antibiotics.³¹ Recent reviews on NG also emphasize the importance of incorporating AMR detection in POCTs.^32–35 For example, in reviewing recent progress in gonorrhea POC tests, Gaydos and Melendez point out that while POCTs for NG are currently available, they are not widely used, especially in LMIC. Further development of POC tests with AMR testing capacity is needed to help guide antimicrobial stewardship.³²

2.3. Trichomoniasis

Trichomoniasis is caused by infection with Trichomonas vaginalis (TV), a protozoan parasite. The WHO estimated that there were 156 million new cases of trichomoniasis worldwide in 2020,²² which is the highest incidence among the major STIs. Moreover, the WHO estimated that as of 2020, there are 105 million prevalent cases of trichomoniasis worldwide.²² In women, the most commonly infected part of the body is the lower genital tract (vulva, vagina, cervix, or urethra). In men, the most commonly infected body part is the urethra. Symptoms for TV infections range from mild irritation to severe inflammation, though asymptomatic infections are common. Trichomoniasis can increase the risk of contracting or transmitting other STIs and HIV. Pregnant women with trichomoniasis are more likely to have preterm delivery. Trichomoniasis is commonly treated with either metronidazole or tinidazole. NAAT-based testing is now more readily available for the diagnosis of TV in women and have replaced culture as the gold standard test due to their excellent sensitivity and specificity.²⁶ Like testing for CT and NG, there are a number of commercially available NAAT systems for TV testing. These include the APTIMA Trichomonas vaginalis Assay (Hologic), the BD ProbeTec Trichomonas Vaginalis Qx Amplified DNA Assay (BD), and the Affirm VPIII Microbial Identification Test (BD). Despite the increased use of NAATs, Gaydos and colleagues point out that diagnostic methods for TV remain suboptimal.³⁶

2.4. Syphilis

Syphilis is caused by the spirochete (a flexible spirally twisted bacterium) Treponema pallidum (TP). The WHO estimated that 7.1 million incident cases of syphilis occurred globally in 2020.²² The WHO also estimated that there are approximately 22.3 million prevalent syphilis cases worldwide as of 2020.²² Syphilis is clinically characterized by a variety of clinical presentations, depending on disease stage, ranging from ulcers to rash in the palms and soles to neurologic symptoms. If left untreated, syphilis can cause serious health complications and can travel to the nervous system (neurosyphilis) or to the eye (ocular syphilis).³⁷ Congenital syphilis occurs when an infected mother transmits the infection to the fetus during pregnancy and can result in adverse birth outcomes, including low-birth weight, miscarriage, stillbirth or death shortly after birth.

Unlike the other major STIs, syphilis is routinely diagnosed by serological testing, which detect antibodies specific to TP. Two types of tests – nontreponemal test and treponemal test – are used. The nontreponemal test is a rapid and simple but indirect method that detects biomarkers from the TP infection, whereas the treponemal test directly detects antibodies produced from the TP infection and is more sensitive and specific. Positivity of both type of tests are required to confirm active syphilis. Two different algorithms are currently used for syphilis diagnosis. In the traditional algorithm, a nontreponemal test, such as the rapid plasma reagin (RPR) or Venereal Disease Research Laboratory (VDRL) test, is initially performed and reactivity confirmed with a treponemal test. In the reverse algorithm, recommended by the CDC, the treponemal test is performed first and reactive-samples are then tested and tittered with the nontreponemal test for purposes of following treatment success.³⁸ One limitation of this approach is that because antibodies remain positive for life following infection, it is difficult to differentiate between prior and active infections.

In the U.S., the Syphilis Health Check (Trinity Biotech USA, Inc) is one of the serological POCT cleared by the US Food and Drug Administration (FDA) and is a Clinical Laboratory Improvement Amendments (CLIA)-waived test. This lateral flow assay detects the presence of treponemal antibodies directly from whole blood or plasma, it is easy to use by an untrained user, and results can be available in 10 minutes. The Syphilis Health Check has been shown to be highly sensitive and specific for detection of syphilis antibodies.¹⁰ The Dual Path Platform (DPP) HIV-Syphilis system is a rapid, multiplex immunoassay for the detection of antibodies to HIV and TP. It is FDA-approved for detection of HIV and syphilis directly from fingerstick whole blood, venous whole blood and plasma specimens.³⁹

Other POC tests, in use outside of the U.S., for detection of TP antibodies include the Alere Determine Syphilis TP (Abbott Laboratories, Inc), SD Bioline Syphilis 3.0 (Abbott Laboratories, Inc), Syphicheck (The Tulip Group/Qualpro), and Visitect Syphilis (Omega Diagnostics), which use plasma or whole blood specimens and return results in 5 to 30 minutes.¹⁵ In a recent meta-analysis of these tests, Jafari and colleagues, found that most of the tests performed reasonably well with either whole blood or serum specimens, with sensitivities ranging from 74% to 99% and specificities ranging from 94% to 99%. In that analysis, the Alere Determine Syphilis TP assay using serum was the best-performing test, with sensitivity of 90% and specificity of 94%.⁴⁰

NAATs for syphilis could play an important role in the diagnosis of syphilis and supplement current serological tests, especially by filling a diagnostic gap between transmission and seroconversion.⁴¹ However, there are currently no FDA-cleared NAATs for the detection of TP. An investigational transcription-mediated amplification (TMA) assay for detection of rRNA of TP (Hologic, San Diego, CA, USA) has shown to be a useful adjunct method for the diagnosis of syphilis. Getman and colleagues reported that the assay is highly sensitivity (1.4 organisms mL⁻¹) and specific for detection of TP directly from clinical samples; although the sensitivity is lower in serum specimens.⁴¹ Another study reported that this assay detected TP in some cases that were missed by classical serological testing and that it can be used to augment syphilis screening among MSM.⁴²

2.5. HIV

HIV is perhaps the most “notorious” STI. The WHO estimated that 1.5 million people worldwide acquired HIV in 2020, resulting in a total of 37.7 million people worldwide living with HIV as of 2020.²² HIV is caused by infection with the human immunodeficiency virus and weakens a person’s immune system by attacking CD4 cells. In the past two decades, due to revolutionary advances in antiviral therapies (ART), HIV/AIDS is now considered a manageable chronic condition. Indeed, in 2016, the Prevention Access Campaign, a health equity initiative with the goal of ending the HIV/AIDS pandemic as well as HIV-related stigma, launched the Undetectable = Untransmittable (U = U) initiative. U = U signifies that individuals with HIV who receive ART and have achieved and maintained an undetectable viral load cannot sexually transmit the virus to others. This concept, based on strong scientific evidence, has broad implications for the prevention and treatment of HIV infection from a scientific and public health standpoint.⁴³

The effectiveness of ART has led to optimism and the launch of aggressive campaigns to eradicate HIV. For example, the United Nation Program on HIV/AIDS (UNAIDS) 95–95-95 targets are that 95% of individuals with HIV are aware of their status, 95% of those aware of their status are on treatment, and 95% of those on treatment have virological suppression.⁴⁴ The Ending HIV Epidemic (EHE) campaign, introduced by the CDC, emphasizes the need for diagnosing all people with HIV as soon as possible by supporting accessible, high-quality HIV testing.⁴⁵ Toward meeting these goals, advances in diagnostic tools that facilitate early testing and viral load testing are important. The pivotal role of POC HIV viral load testing to sustainable global HIV/AIDS response is now widely recognized.⁴⁶ Although protein- and antibody-based tests have been the main diagnostic method used since the development of the first HIV diagnostic test, NAAT-based detection of HIV is now considered more ideal in the acute stage of infection than protein- and antibody-based tests.⁴⁴

2.6. Brief Remarks on Other Conditions

This review focuses only on the 4 major curable STIs and HIV, but other STIs also deserve attention. Mycoplasma genitalium (MG) is an emerging STI pathogen that has already developed AMR.^{29, 47, 48} Due to its fastidious growth requirements,⁴⁷ NAAT-based POCT with AMR detection is crucial for the diagnosis and treatment of MG infection.⁴⁹ To this end, ResistancePlus® MG from SpeeDX (Australia) is a commercial assay for detection of AMR in MG in clinical laboratory use in Australia and Europe. Bacterial vaginosis (BV) is a syndrome of vaginal dysbiosis – changes in the normal balance of bacteria in the vagina – that can be transmitted sexually and represents the most common vaginal condition in women ages 15 to 44. The diagnosis and treatment of BV would benefit significantly from new NAAT-based POCT technologies.^50–52 To this end, although some bacteria associated with BV have been identified (most notably Gardnerella vaginalis), searches for etiological agents via metagenomics would likely continue.

3. CURRENT LANDSCAPE OF COMMERCIAL NEAR-POC NAAT INSTRUMENTS

We briefly introduce notable commercial instruments that can perform near-POC NAAT-based detection of CT, NG, TV, TP, and HIV (Figure 2). These instruments are more advanced than the conventional benchtop assays introduced in the previous section. Yet several of these instruments are still laboratory-bound. Furthermore, all of them are too expensive as POCTs in LMICs. Nevertheless, an understanding of these instruments is useful for developing new NAAT-based POCT technologies. More comprehensive introduction to these instruments and additional instruments not covered in this review can be found in the literature.^{15, 53, 54} For example, Toskins and colleagues published a report on the landscape of STI molecular diagnostic instruments based on publicly available information, published and unpublished reports and prospectuses, and interviews with developers and manufacturers.⁵³ Lepej and Poljak reviewed the status of portable molecular diagnostic instruments in microbiology, which covered eight portable diagnostic molecular instruments that are commercially available or in the launching stage.⁴¹

Figure 2. Notable commercial NAAT-based STI near-POC platforms.

(A) Cepheid GeneXpert, (B) DRW SAMBA II, (C) binx io, (D) Visby Medical Sexual Health Test, and (E) Quidel Solana represent the most advanced NAAT-based STI near-POC platforms to date. These platforms have been evaluated in recent major clinical evaluation studies,^{55–57, 60, 62, 63, 65, 67} from which the sample type and the sensitivity are included herein. Among these platforms, Cepheid GeneXpert, binx io, Visby, Sexual Health Test, and Quidel Solana have received approvals from both FDA and CE, SAMBA II is CE approved, and binx io and Visby, Sexual Health Test are also CLIA waived.

3.1. Cepheid GeneXpert®

The Cepheid GeneXpert® System (Cepheid, Sunnyvale, CA, USA; Figure 2A) is one of the most popular instruments for performing near-POC NAATs. It is an automated and integrated system for PCR-based NAAT, which currently has 21 FDA-cleared and 27 CE-IVD marked assays, including a CT assay (Xpert® CT) a combined CT and NG assay (Xpert® CT/NG), and a test for TV (Xpert® TV). The GeneXpert® System integrates and automates sample preparation, amplification, and detection in a single-use, self-contained cartridge. Most liquids and dry reagents along with enzymes are prefilled so that pre-analytical steps are minimized, greatly reducing opportunities for sample mix-ups and operational errors. Each self-contained cartridge includes a sample adequacy control, which detects the presence of a single copy human gene, which monitors whether the sample contains human DNA for enhanced results integrity.

Cepheid GeneXpert instrument and assays have been evaluated in major clinical studies. In 2013, Gaydos and colleagues compared the Cepheid Xpert CT/NG assay to two nucleic acid amplification assays using samples from 1722 female and 1387 male volunteers.⁵⁵ The authors reported 97.4%, 98.7%, and 97.6% sensitivities for chlamydia from female endocervical, vaginal, and urine samples, respectively, and 97.5% sensitivity for chlamydia from male urine samples, with ≥ 99.4% specificities across all sample types. For gonorrhea, the sensitivities from female endocervical, vaginal, and urine samples were 100.0%, 100.0%, and 95.6%, respectively, and the sensitivity from male urine samples was 98.0%, with ≥ 99.8% specificities across all sample types. In 2018, Schwebke and colleagues reported the results of a multicenter study conducted using the Xpert TV assay to test specimens from 1867 women and 4791 men.⁵⁶ For female specimens (urine samples, self-collected vaginal swabs, and endocervical swabs), the sensitivities and specificities ranged from 99.5 to 100% and 99.4 to 99.9%, respectively. For male urine samples, the sensitivity and specificity were 97.2% and 99.9%, respectively.

Although the GeneXpert® systems are currently dependent on stable electricity, controlled temperature conditions, and calibration requirements and therefore considered more appropriate for use at the district hospital or above in LMIC, there are efforts in moving these systems closer to remote primary healthcare settings to reach the most patients. Ndlovu and colleagues concluded that it was feasible to use the GeneXpert® platform to test HIV (and tuberculosis) in rural Zimbabwe.⁵⁷ Similarly, Causer and colleagues examined the diagnostic performance of the GeneXpert CT/NG assay when used by clinicians in remote community health services in Australia with high prevalence of CT and NG infection and found strong concordance with conventional NAATs.⁵⁸ According to Cepheid, it is also developing a portable instrument named GeneXpert® Omni, which will be powered by a rechargeable battery, interfaced to mobile device application, and equipped with wireless and Bluetooth connectivity.

3.2. SAMBA II

SAMBA II (Figure 2B) was developed by a team at the Diagnostics Development Unit of the University of Cambridge and its spinoff company, Diagnostics for the Real World (DRW, CA, USA).⁵⁹ SAMBA II is a sample-in-result-out system that is comprised of an assay module and a tablet/display module linked via Bluetooth. However, it is not FDA-cleared for use in the U.S. Consistent electricity is required for operation. Each tablet/display module can control up to four assay modules, allowing some tailored scalability and throughput depending on the needs of healthcare facilities. SAMBA II uses isothermal NAAT. There are currently two assays: (i) HIV-1 Qualitative Whole Blood test (for early infant diagnosis and diagnosis of acute HIV infection, and (ii) HIV-1 Semi-Quantitative Plasma test (for HIV monitoring). Ritchie and colleagues evaluated the performance of the SAMBA II HIV-1 Semi-Quantitative Plasma test in London, Malawi, and Uganda and found overall 97.3% concordance with a standard commercial test (Roche TaqMan v2; n = 584).⁶⁰ A combined CT/NG assay is in the pipeline, but a release date has not been announced.

3.3. binx io

The binx io® platform (binx health, Trowbridge, UK and Boston, MA, formerly Atlas Genetics; Figure 2C) is a NAAT-based POCT for CT and NG that was CE marked in April 2019, FDA cleared in August 2019, and received CLIA-waived status in March 2021. The test is approved for use with vaginal swabs and urine samples for men. The io® CT/NG assay provides sample-to-result capability, including DNA extraction, rapid PCR, and electrochemical-based detection in 30 minutes, fitting the requirements for a short turn-around-time, so that results can be used to provide treatment to patients before they leave the clinical encounter.

Several evaluation studies using the io® platform have been conducted. A pilot study for detection of CT found high sensitivity and specificity, high patient acceptability, and that the majority of women (70%) preferred to self-collect vaginal swabs if a POC test was available.⁶¹ Another study evaluated the CT assay performance using samples collected from women attending four sexual health clinics in England and found that the assay’s sensitivity was similar to that of laboratory-based NAATs, and its specificity is at the lower end of the laboratory-based NAAT range.⁶² In a recently completed, multi-center clinical study with 1,523 participants, 96% of patient’s vaginal swab samples were processed on the binx io® platform by non-laboratorians in a POC setting,⁶³ reporting for women a sensitivity/specificity of 96.1%/99.1% for CT and 100%/99.9% for NG, and for male urine a sensitivity/specificity of 92.5%/99.3% for CT and 97.3%/100% for NG. Finally, the io® CT/NG assay and its performance characteristics were recently profiled by Van Der Pol and Gaydos.⁶⁴

3.4. Visby Medical Sexual Health Test

The Visby Medical Sexual Health Test (Visby Medical, San Jose, CA, USA; Figure 2D) is a new, single-use, rapid, NAAT-based diagnostic device for the detection of CT, NG, and TV. This device, which can be used at POC without complex instrumentation and provide results in < 30 min, received FDA approval and CLIA-waived status in August 2021. In a cross-sectional study, Morris and colleagues enrolled 1585 participants at ten clinical sites across seven US states.⁶⁵ Untrained operators tested the participant’s self-collected vaginal swabs using only the device manufacturer’s instructions and the results compared with other NAAT results (Hologic and BD). Of the 1555 participants with test ran on the device, 1532 (98.5%) had a valid result on either the first or repeat test. Among the patients with evaluable results (including a determinate patient-infected status), the device achieved a sensitivity/specificity of 97.6%/98.3% for CT (n = 1457), a sensitivity/specificity of 97.4%/99.4% for NG (n = 1468), and a sensitivity/specificity of 99.2%/96.9% for TV (n = 1449). As such, the Visby Medical Sexual Health Test could represent an important advance in the development of rapid diagnostics for STIs and other infectious diseases.^{19, 65}

3.5. Solana Trichomonas assay

The Solana Trichomonas Assay (Quidel, San Diego, CA, USA; Figure 2E) is a rapid (< 40 min), near-POC diagnostic that uses an isothermal NAAT method called helicase dependent amplification⁶⁶ (HDA, also see 4.2.3.3 for a brief introduction) and the Solana instrument to detect TV. This assay consists of two steps: (1) specimen preparation and (2) amplification and detection of a conserved repeat sequence in the TV genome by HDA in the presence of a target-specific fluorescence probe.³⁶ The assay is intended for use in clinical laboratories, but has the potential to be useful in a doctor’s office or a clinic that has a laboratory. It was recently FDA cleared as a moderately complex POC assay for the diagnosis of trichomoniasis. In the FDA clinical trial,⁶⁷ vaginal swabs and urines were obtained by clinicians from 1044 women (501 asymptomatic and 543 symptomatic). The prevalence of TV by swabs and/or urines was 11.5%. For the clinician-collected swabs, the Solana Trichomonas assay showed high sensitivity and specificity from both asymptomatic (100%/98.7%) and symptomatic (98.6%/98.5%) women compared with a FDA-composite reference standard of wet preparation and/or culture for detection of TV. Urines were similarly sensitive and specific from asymptomatic (98.0%/98.4%) and symptomatic (92.9%/97.9%) women when compared with the FDA-composite reference methods. Compared with a NAAT assay (Aptima-TV), the sensitivity/specificity performance was 89.7%/99.0% for swabs and 100%/98.9% for urines.⁶⁷

3.6. Other Assays and Instruments

There are also several near-POC assays and instruments for STIs currently being developed by various companies. Among them, the ResistancePlus® GC (beta) assay from SpeeDX is the first commercial, licensed NAAT-based test that utilizes PlexPrime® and PlexZyme® technologies⁶⁸ to simultaneously detect NG and susceptibility to ciprofloxacin. Though it is yet to be FDA cleared, there is a peer-reviewed, published evaluation of the assay, which reported > 96% sensitivity and specificity for detection of NG and > 97% sensitivity and specificity for detection of ciprofloxacin susceptibility.⁶⁹ Moreover, the development and clinical evaluations^{70, 71} of ResistancePlus® MG FleXible – ResistancePlus® MG assay adapted to a cartridge format using the Cepheid GeneXpert® platform – suggest that similar development for ResistancePlus® GC as a near-POC test is possible in the future. STD Direct Flow Chip Kit from Vitro (Spain) uses multiplex PCR and DNA probe hybridization to detect 9 species of STI pathogens including CT, NG, TV, and TP from clinical specimens (e.g., urine, endocervical swab, urethral swab) without DNA purification. It has been CE marked and evaluated, reporting 98.4% sensitivity and 99.9% specificity.⁷²

Although several near-POC instruments for STIs are currently under development by various companies, peer-reviewed, published evaluations of these instruments were not found in a literature search. These include the Truelab Real Time micro PCR system from Molbio Diagnostics (India), an instrument that integrates STI Multiplex Array from Randox Biosciences (UK) and the Vivalytic Analyzer from Bosch Healthcare Solutions (Germany), the HG Swift from Hibergene Diagnostics (Ireland), ID NOW™ (formerly Alere™-I) from Abbott (UK), and Q-POC™ from QuantuMDx (UK).

4. CURRENT GAPS

Despite advances of commercial near-POC NAAT instruments and tests, there is still significant gaps for developing NAAT-based STI POCTs, which must be addressed through research. In this section, we provide a brief overview of these gaps with viewpoints from both the POCT technology development community and the STI research community.

First, NAAT-based detection of AMR in NG is challenging. Due to complex mechanisms of AMR, it is critical to first identify the appropriate genetic targets that can indicate susceptibilities of NG to relevant antibiotics. The frequency of potential genetic targets for AMR testing in diverse NG populations must be investigated. Moreover, multiplexed assays that test for susceptibility to different antimicrobials must be developed⁷³. Subsequently, the new assays must be incorporated into a POCT. Such a POCT supplemented with AMR detection has the potential to provide patient-specific information in a time frame that allows clinicians to prescribe individualized treatment regimens. It could also provide a valuable tool for surveillance and epidemiological studies while fulfilling the need for a diagnostic-enabled antibiotic stewardship strategy.¹³

Second, despite significant progress in the development of new NAAT-based STI POCT technologies, considerable technical barriers remain. The most important requirements for POCTs are to ensure same-visit results for treatment and uptake in primary care facilities, especially in LMIC.⁸ These seemingly basic requirements have proven challenging due to the many stringent criteria that must be met. The most well-known are the ASSURED (Affordable, Sensitive, Specific, User-friendly, Rapid, Equipment-free, Delivered)^{74, 75} criteria. With rapid advances in digital technology and mobile health (m-health), future diagnostics should incorporate these elements to provide REASSURED (Real-time connectivity and Ease of specimen collection and environmental friendliness) diagnostic systems that can inform disease control strategies in real-time, strengthen the efficiency of health care systems and improve patient outcomes⁷⁶ and highlight additional conditions that should be considered in addressing POC testing needs. In addition to the RE/ASSURED criteria, target product profiles (TPPs)^{53, 77, 78} developed through landscape analysis, expert opinion, and end-user input can also serve as guidance for developing STI POCT technologies. For example, an international expert group (IEG), commissioned by the WHO, established after two technical consultations in 2014 and 2015,⁷⁹ have formulated TPPs for 5 STIs.⁵³ Organizations such as the Foundation for Innovative New Diagnostics (FIND) – a global non-profit organization driving innovation in the development and delivery of diagnostics to combat major diseases affecting the world’s poorest populations – also have TPPs (see https://www.finddx.org/tpps/). Finally, it is now well known in the Lab on a Chip research community that LMIC pose additional challenges such as extreme environments, poor infrastructure, few trained staff, and limited financial resources.⁵³

Third, among the many technical barriers, cost stands out as a particularly notable gap. Although the Cepheid GeneXpert® is one of the most mature commercial platforms, its high cost diminishes its potential for effective POCTs in LMIC. Companies who develop and market POCTs must ensure profitability, and therefore may prioritize affluent countries and neglect LMIC. Moreover, scientists and engineers who work on developing new STI POCT technologies are usually based in affluent countries and are often unaware of the cost barriers for the end-users in LMIC.⁸⁰ It is therefore imperative to bring technology developers together with end-users (i.e., patients and healthcare providers) in LMIC to understand these challenges and collaborate toward simple and cost-effective solutions.

Fourth, more broadly, technology developers and end-users, especially in LMIC, often lack synergy. As the development of NAAT-based STI POCT technologies remains a daunting challenge, researchers are increasingly recognizing that there could be some flexibility and tradeoffs in satisfying RE/ASSURED criteria and TPPs. As such, dialogues between technology developers and end-users about such tradeoffs are critical^{81, 82} (and would seem sensible) but have unfortunately been scarce, leaving technology developers based in affluent countries unaware of these settings and the real needs of the end-users in LMIC.⁸⁰ In truth, even the acceptance of STI POCTs should be assessed in their intended local settings instead of assumed. The aforementioned IEG for STI POCTs, for instance, even stressed the need for more studies to determine attitudes, perceived usefulness, confidence, and anticipated changes in the management of STIs from both patients and healthcare providers toward STI POCTs in different settings and populations.^{79, 83}

5. CURRENT RESEARCH AND POTENTIAL DIRECTIONS

We discuss in this section current research on and potential research directions for NAAT-based STI POCT technologies. The development of these technologies is multi-faceted and encompasses the following areas: discovering and identifying diagnostic targets (especially for AMR), streamlining sample preparation, establishing POC amenable NAAT, multiplexing, exploring detection modalities, exploring device materials, reducing power dependence, constructing integrated devices and instruments, pre-storing reagents, incorporating internet connectivity, and evaluating technologies (Figure 3). Within each of these areas, we introduce representative methods, notable works, and emerging trends. We supplement our discussion with relevant research geared toward infectious diseases, as it can inspire potential directions for STI research.

Figure 3. Overview of research toward NAAT-based STI POCT technologies.

Development of NAAT-based STI POCT technologies is multi-faceted. Broadly, the research encompasses (at least) 11 areas: discovering and identifying diagnostic targets (especially for antimicrobial resistance, AMR), streamlining sample preparation, establishing POC amenable NAAT, multiplexing, exploring detection modalities, exploring device materials, reducing power dependence, constructing integrated devices and instruments, pre-storing reagents, incorporating internet connectivity, and evaluating technologies. For several of these areas, several methods are notable and discussed in detail.

5.1. Discover and Identify Diagnostic Targets

NAAT-based detection of CT, NG, and TV through specific gene sequences is well established. On the other hand, NAAT-based detection of AMR in NG is still a challenge. Thus, for NG, the development of new POCT technologies would benefit from identifying the appropriate diagnostic targets that can indicate susceptibilities to relevant antibiotics. The diagnostic targets can be genetic determinants that are found to be associated with AMR for relevant antibiotics or NG nucleic acids that can be detected through a “pheno-molecular” approach, in which brief antibiotic exposure precede quantitative detection of the nucleic acids.

5.1.1. Genetic Determinants of AMR

The ability to directly detect genetic determinants of AMR through NAAT is ideal for incorporating AMR detection in NAAT-based POCTs for NG. This detection strategy would obviate antibiotic exposure and significantly simplify the development of POCTs. At present, several genetic determinants of AMR associated with resistance to key antibiotics for treating NG are known and characterized.^{84, 85} These include mutations in the gyrA gene for ciprofloxacin resistance,⁸⁶ mutations in the 23S rRNA gene for azithromycin resistance,^{87, 88} a mutation in the 16S rRNA gene for spectinomycin resistance,⁸⁹ and mutations in mosaic alleles of the penA gene for resistance to extended-spectrum cephalosporins^{90, 91} – a group of antibiotics that includes ceftriaxone (the current treatment option for patients infected with NG). Multiple research groups have reported NAAT-based assays to detect these genetic determinants of AMR. ^{73, 92–97} For example, Donà and colleagues developed a qPCR method with high-resolution melting (HRM) analysis to identify NG and detect AMR. This method uses one triplex and three duplex reactions to detect two sequences for NG identification and seven determinants of resistance to extended-spectrum cephalosporins (e.g., ceftriaxone and cefixime), azithromycin, ciprofloxacin, and spectinomycin.⁷³ The method was validated with 39 NG strains, 19 commensal Neisseria species strains, and 193 gonococcal isolates, reporting > 90% sensitivities and specificities. Low assay sensitivity (10³ to 10⁴ genomic DNA copies/reaction) and some cross-reactivity with non-gonococcal Neisseria species in this method suggest that clinical isolation would be required, which would prevent direct testing from clinical specimens, but this method would still offer faster AMR testing from gonococcal isolates than culture-based method. Thus, it could be an attractive approach to adapting such NAAT assays into POCT technologies.

Despite the potential advantages, direct NAAT-based detection of AMR via genetic determinants is suboptimal in sensitivity and specificity and cannot detect newly emerging AMR determinants^98–100. Discoveries of AMR determinants for other antimicrobials would be a significant advance,^{79, 98, 101} but such discoveries are reliant on sequencing technologies and can be resource-intensive still quite challenging. For example, Thomas and colleagues recently reported on whole genome sequences of 813 NG isolates collected through the Gonococcal Isolate Surveillance Project in the United States. Their phylogenomic analysis revealed a persisting lineage of NG that carries a mosaic allele in the penA gene and was present in the majority of isolates with elevated resistance to extended-spectrum cephalosporinss.¹⁰² Moreover, emerging nanopore-based sequencing technologies (e.g., MinION from Oxford Nanopore Technologies), which can be less resource-intensive, still must improve in accuracy to identify key genetic determinants for AMR in NG.¹⁰³ Nevertheless, sequencing technologies will continue to advance. And, as reviewed by Pai and colleagues,⁴⁴ sequencing technologies have already been used to achieve phylogenetic classification of HIV-1 and monitor antiviral resistance, we therefore foresee sequencing technologies playing an important role toward discovering and identifying genetic determinants of AMR in NG.

5.1.2. Pheno-molecular antimicrobial susceptibility testing (AST)

Pheno-molecular AST is an emerging strategy for AMR testing. In pheno-molecular AST, a brief antibiotic exposure precedes quantitative molecular detection of the bacteria, and the quantity of the detected molecules is used as a surrogate for bacterial growth and susceptibilities to antibiotics.¹⁰⁴ As a result, pheno-molecular AST can be more broadly applicable than NAAT-based detection of genetic determinants of AMR. To date, pheno-molecular AST has been demonstrated in a handful of studies for testing AMR in NG against several antibiotics. Antibiotic exposure is necessary, which not only lengthens the assay TAT and adds reagents and assay steps, which can complicate the development of POCTs.

Genomic DNA is a simple choice as the molecular surrogate in pheno-molecular AST. For example, Chen and colleagues developed a qPCR assay that quantitatively measures NG genomic DNA directly from a culture medium after antibiotic exposure without sample processing. The assay has an analytical sensitivity of 100 colony forming units (CFU) mL⁻¹ and is specific to NG in the presence of urogenital flora and clinical swab eluent. The authors tested several NG strains against penicillin, tetracycline, and ciprofloxacin, and achieved 95.2% categorical agreement with standard, culture-based E-test.¹⁰⁵ Though faster than the E-test method, which requires overnight culture, this assay still required 1 – 4 h for antibiotic exposure and an additional 1.5 h for qPCR. Recently, Savela and colleagues developed a new strategy for accelerating pheno-molecular AST by measuring DNA accessibility to exogenous nucleases after exposure to β-lactams (the largest antibiotic class used to treat NG).¹⁰⁶ The authors showed that genomic DNA in antibiotic-susceptible cells has increased accessibility upon exposure to β-lactams such as penicillin, cefixime, and ceftriaxone and that a judiciously chosen surfactant permeabilized the outer membrane and enhanced this effect. Moreover, the authors employed digital PCR and digital loop mediated isothermal amplification (LAMP) for precise quantification of DNA. In doing so, the authors could shorten exposure of NG clinical isolates to these antibiotics to 15 – 30 min and still achieved strong agreement with standard culture-based agar-dilution AST. Finally, as digital LAMP can be completed in ~35 min, the assay can be completed within 1 h even from clinical urine samples.¹⁰⁶ Despite its fast TAT, this assay has yet to be developed into a POCT and its manual and multi-step workflow may pose challenges toward this goal.

Messenger RNAs (mRNAs) that can rapidly increase or decrease in quantity (i.e., expression level) in response to antibiotic exposure have also been used as molecular surrogates in pheno-molecular AST. The appeal of mRNAs is that their quantities can change before bacterial replication, thus allowing for rapid antibiotic exposure time. The disadvantage, on the other hand, stems from that for each antibiotic, a panel of distinct mRNA markers must be first identified (through sequencing technologies) and then simultaneously detected to ensure accurate AST. Thus, the scalability of this method can be a concern. Khazaei and colleagues found that mRNA abundance changes in porB and rpmB, which were detected via digital RT-PCR, could determine the antibiotic susceptibility and resistance of 49 clinical isolates after 10 min exposure to ciprofloxacin.¹⁰⁷ Hashemi and colleagues identified two markers (arsR and rpsO) of AMR in NG upon 10 min of azithromycin exposure. They validated these markers using RT-qPCR against a total of 78 NG isolates.¹⁰⁸

Wadsworth and colleagues pointed out that development and testing of mRNA-based pheno-molecular AST should consider genetic diversity of the target pathogen.¹⁰⁹ As a demonstration, they tested phylogenetically diverse panel of NG previously reported diagnostic markers for ciprofloxacin resistance (porB and rpmB)¹⁰⁷ and found that porB and rpmB expression were not uniformly diagnostic of ciprofloxacin resistance. They also employed the same NG panel, RNA-seq, and RT-qPCR to identify azithromycin responsive mRNA markers and evaluate their potential diagnostic value. Their results confirmed evidence of genetic distance and population structure impacting transcriptional response to azithromycin and led to four candidate diagnostic transcripts (rpsO, rplN, omp3, and NGO1079) that categorically predicted resistance in 19/20 cases upon 60-min azithromycin exposure. These results point to the usefulness of employing a phylogenetically diverse panel of NG for the development of pheno-molecular AST assays, while also showing potentially suboptimal AST accuracy from rapid (e.g., 10 min) antibiotic exposure.

5.2. Streamline Sample Processing

The development of new NAAT-based STI POCT technologies begin with sample processing. To ensure access and uptake of new STI POCT technologies at the primary health care level without compromising the analytical performances, they must take into account the sample type, the collection method, and the sample volume.¹¹⁰ Currently, swabs collected by healthcare providers and stored in transport media are the most common samples for CT, NG, and TV testing. Whole blood collected by healthcare providers via venipuncture in standard collection tubes or plasma separated from the whole blood are the standard samples for syphilis and HIV. Self-sampling for the major curable STIs is an emerging trend,¹¹¹ and several studies have shown that self-collected samples (e.g., vaginal swab, penile swab, urine) offer comparable diagnostic outcomes as healthcare provider-collected swabs and can increase STI testing uptake and may even offer the opportunity for home-based screening.^112–115

After the sample type, the collection method, and the sample volume are determined, a suitable sample processing method that can extract target nucleic acids from the human matrix in clinical specimens while minimizing manual processing steps, maintaining low cost, and obviating bulky instruments can be developed accordingly. To date, membrane-based and magnetic bead-based sample processing methods have been most frequently employed for detecting STIs, while other methods (e.g., glass powder, silica beads, silicon micropillars) have yet to see heavy use for detecting STIs may still be explored. Several review articles that broadly discuss sample processing methods beyond STIs provide useful resources toward deciding on the sample processing method of choice. For example, Paul and colleagues reviewed POC nucleic acid extraction technologies for rapid diagnosis of human (and plant) diseases.¹¹⁶ Yin and colleagues recently reviewed advances in microfluidic systems with integrated sample preparation and nucleic acid amplification.¹¹⁷ Emaus and colleagues offered a comprehensive review that covered nucleic acid extraction methods including liquid-liquid extraction, solid phase extraction, microfluidic-based extraction, and sequence-specific extraction.¹¹⁸ Obino and colleagues provided an overview of microfluidic systems for nucleic acid extraction from human samples.¹¹⁹ As a potential template of assessing sample processing methods, Beall and colleagues assessed the workflow and performance of six nucleic acid extraction technologies for use in LMIC.¹²⁰ We briefly discuss these sample processing methods, with an emphasis on those that have been applied toward detecting STI pathogens.

5.2.1. Membrane-Based Sample Processing

Membrane filtration is a mature method for extracting nucleic acids from samples, which represents a key step in sample processing. Membrane filtration allows large sample volume to be processed, which can potentially concentrate the target nucleic acids and improve detection sensitivity. On the other hand, as the sample and reagents must be injected (either manually or by an instrument such as syringe pump) through the membrane, the workflow can complicate the development of POCT technologies. Nevertheless, membrane filtration has been employed toward the development of POC-amenable HIV detection. For example, Liu and colleagues integrated a flow-through, Flinders Technology Associates (Whatman FTA®) membrane in a microfluidic cassette to isolate HIV-1 RNA from oral fluids, and subsequently performed reverse transcription loop-mediated isothermal amplification (RT-LAMP) for detection.¹²¹ Jangam and colleagues developed fast isolation of nucleic acid (FINA), which used a separation membrane and an absorbent pad, to extract cellular HIV-1 proviral DNA from whole blood in 2 min (Figure 4A).¹²² The same team also developed a disposable assay card that includes on-board reagent storage and a battery-operated portable analyzer capable of automated PCR mix assembly using a novel reagent delivery system and performing qPCR-based HIV-1 proviral DNA detection.¹²³ More recently, they detected FINA-extracted DNA from the heel blood of 61 South African infants by benchtop qPCR, and demonstrated 100% sensitivity and 100% specificity for HIV-1 detection.¹²⁴

Figure 4. Advances in membrane-based sample processing toward developing NAAT-based STI POCTs.

(A) Jangam and colleagues developed fast isolation of nucleic acid (FINA), which used a separation membrane and an absorbent pad, to extract cellular HIV-1 proviral DNA from whole blood in 2 min.¹²² The same team subsequently leveraged FINA and developed a POC PCR test for HIV-1 detection in LMIC. In addition to FINA, they also developed a disposable assay card that includes on-board reagent storage, provisions for thermal cycling and fluorescence detection, and a battery-operated portable analyzer. The system was capable of automated PCR mix assembly using a novel reagent delivery system and performing qPCR.¹²³ (B) Rosenbohm and colleagues utilized a chitosan-modified Fusion 5 membrane to capture TV DNA from ~50 mL cell lysate loaded into a syringe and pumped through the membrane at 20 mL min⁻¹.¹²⁶ The chitosan functionalized membrane (and capture DNA) was transferred in an isothermal helicase dependent amplification mix, where TV DNA amplified with negligible inhibition. Finally, the amplification result was detected via lateral flow strip readout.

Chitosan, a linear polysaccharide that captures and concentrates nucleic acids via anion exchange chromatography, has been incorporated into membrane filtration. For example, Byrnes and colleagues developed a sample preparation system by coupling lateral flow in a porous membrane with chitosan.¹²⁵ In this method, charge-switchable chitosan can capture and release nucleic acid in a pH-dependent manner, which can then be used for downstream processes, including qPCR. Though the authors believe their method is simple-to-use, inexpensive, and disposable, the process was performed in a humidifier chamber to reduce evaporation and the recovery of nucleic acid required centrifugation, which significantly diminished the potential of deploying this system at the POC. In another example, Rosenbohm and colleagues utilized a chitosan-modified Fusion 5 membrane (a silica-based glass fiber membrane) to capture TV DNA from a large sample volume (~50 mL; Figure 4B).¹²⁶ To extract DNA, cell lysate was loaded into a syringe and pumped through the chitosan-modified filter using a syringe pump. After this extraction, the chitosan functionalized filter was transferred to a HDA reaction mix, where TV DNA amplified with negligible inhibition. Finally, the amplification result was detected via lateral flow strip readout. These results suggest that such direct amplification can simplify membrane filtration-based sample processing toward the development of POCT technologies.

An intriguing strategy for membrane-based sample processing is directly integrating the membrane into pipette tips. To this end, Lu and colleagues inserted FTA cards into pipette tips to extract DNA/RNA from crude samples with a few manual pipetting steps, and directly performed LAMP within the same pipette tips, thereby achieving sample-to-answer detection.¹²⁷ TruTip Nucleic Acid Purification Kit (Akonni Biosystems, USA) is a commercial product based on the same concept. Pipetting inevitably involves manual operation, which may introduce errors. However, the acceptability of this strategy depends on the end-users.

5.2.2. Magnetic-Based Sample Processing

Using nucleic-acid binding magnetic beads for sample processing is another mature method and has been routinely incorporated in commercial platforms. Magnetic-based sample processing offers several advantages, including target concentration and automation. For example, the commercially available Hologic Aptima assays use magnetic beads labeled with hybridization probes to capture ribosomal RNA from STI pathogens prior to detection by NAAT. Mulberry and colleagues developed a palm-sized, battery-operated sample preparation instrument with automated magnet movement to enable nucleic acid purification with simple contactless pouring for discarding supernatant.¹²⁸

Magnetic beads could also be used to transport purified nucleic acids and automate sample processing with the assistance of a moving external magnet – a concept often referred to as magnetic droplet microfluidics or droplet magnetofluidics.^129–131 Shin and colleagues leveraged this concept to purify DNA from swab samples for detection of CT in a custom-developed plastic cartridge and palm-sized instrument.¹³² Gaddes and colleagues reported on the development of a palm-sized droplet magnetofluidic-based nucleic acid extraction and purification instrument.¹³³ Xu and colleagues implemented this idea within a capillary with pre-filled lysis buffer and magnetic beads, and used a moving magnet to transports DNA released from lysed CT and NG and detection by LAMP.¹³⁴ Chen and colleagues reviewed advances in using magnetic particles actuated by magnetic fields for achieving integrated nucleic acid purification, amplification, and detection without manual operation or sophisticated microfluidic pumps and valves.¹³⁵

5.2.3. Other Sample Processing Methods

Unlike membrane-based and magnetic-based sample processing methods, other sample processing methods such as glass powder, silicon micropillars, and silica beads have mostly been used toward detecting non-STI bacteria. For example, Devi and colleagues reported using sterile glass powder to lyse microorganisms such as Bacillus subtilis (a Gram-positive bacteria), Escherichia coli (a Gram-negative bacteria), and Chlorella sorokiniana (a microalgae) followed by using powdered activated charcoal to purify their DNA.¹³⁶ Petralia and colleagues developed a microfluidic biofilter device composed of a silicon microchannel with silicon oxide-coated silicon micropillars distributed at the bottom surface to extract Hepatitis B Virus DNA.¹³⁷ Even using only Tris-EDTA buffer as the sample buffer, the authors found that DNA could be captured by the micropillars. The authors found that the DNA extraction efficiency increased with the surface area to volume ratio and managed to accomplish ~40% DNA extraction efficiency. Finally, Park and colleagues packed silica beads within an integrated rotary microfluidic system to extract DNA from food-borne pathogens such as Salmonella Typhimurium and Vibrio parahaemolyticus.¹³⁸ In the presence of chaotropic salts, DNA can adsorb to the surface of the silica beads, thus achieving DNA extraction. The authors also incorporated LAMP and colorimetric lateral flow strip-based detection in their system and demonstrated a limit of detection of 50 CFU in 80 min. This platform may be readily adapted to detect STI pathogens.

Finally, an alternative to sample processing is directly mixing the sample into the reaction mix and amplifying the target nucleic acids. Such direct NAAT assays can minimize or even obviate sample processing, which can be advantageous toward realizing NAAT-based STI POCTs. On the other hand, because a small sample volume is typically used to avoid inhibition of most current NAATs, the sensitivity of direct NAAT assays may be reduced. There are a few examples of STI detection via direct NAAT. Krolov and colleagues reported a RPA assay that can detect CT directly from urine samples.¹³⁹ In a follow-up study, Jevtuševskaja and colleagues reported that the addition of antimicrobial peptide lyses improves their LAMP assay for direct CT detection from urine samples.¹⁴⁰ Walker and Hsieh reviewed advances in direct amplification of nucleic acids from complex samples, including blood, dried blood spot, saliva, swabs, and urine,¹⁴¹ which may offer a starting point for developing new direct NAAT assays for detecting STI pathogens.

5.3. Establish POC Amenable NAAT

Because POCTs must be rapid, portable, and low-cost, standard NAAT methods – including PCR and isothermal NAAT methods (e.g., LAMP, HDA, RPA) – that are performed in benchtop instruments must be adapted accordingly, thereby becoming amenable for POC use. The typical approach is implementing these NAAT methods in microfluidic chips and portable instruments. Such implementation can significantly improve the speed and portability of PCR by directly addressing its limitation from thermal cycling. Meanwhile, as isothermal NAAT methods inherently obviate thermal cycling, they offer a more direct path for implementation within microfluidic chips and construction of portable instruments. Several reviews cover this research space. For example, Cantara and colleagues evaluated 8 NAAT methods for potential use to detect infectious agents in LMIC.¹⁴² Maffert and colleagues also reviewed NAAT-based technologies toward diagnostics for infectious diseases in LMIC.¹⁴³ Lee and colleagues reviewed emerging ultrafast NAAT methods, which include light-, radio frequency-, convective flow-, and resistive-assisted thermal cycling.¹⁴⁴ These reviews were not specific to STIs, but the conclusions can be applicable to STIs.

5.3.1. PCR

PCR remains a mainstay for developing NAAT-based STI diagnostics. For developing POCT technologies, PCR has been perceived as a disadvantage due to its requirement for thermal cycling, which can lengthen TAT and complicate miniaturization. However, emerging miniaturized and ultrafast PCR technologies are dispelling these perceptions. Petralia and colleagues reviewed the recent advances in genetic POC technologies, including the extraction and PCR amplification chemistry suitable for POC use and the new frontiers of research in this field.¹⁴⁵ Ullerich and colleagues also reviewed ultrafast PCR technologies for POCT.¹⁴⁶ We highlight herein some recent advances that emphasized research on STI.

The development of extreme PCR¹⁴⁷ and plasmonic PCR^{148, 149} – both of which can achieve sub-minute PCR – are challenging the conventional wisdom that PCR is slow. However, miniaturization of these ultrafast PCR into a low-cost setup that is amenable for POC and LMIC could still be challenging. To this end, Chan and colleagues used the extreme PCR concept to develop a “Thermos Thermal Cycler” that costs < US $200 (Figure 5A).¹⁵⁰ In this setup, 2 or 3 24-oz thermoses were filled with heated water to serve as water baths for performing PCR or RT-PCR. A pan-and-tilt servo, which was controlled via Arduino electronic controller and powered by a battery pack, was used to shuttle PCR vessels in and out of the thermoses. For demonstration, the authors performed PCR to detect extracted CT DNA in < 12 min (40 cycles), as well as RT-PCR to detect extracted HIV-1 RNA in ~16.4 min (45 cycles).¹⁵⁰ The authors believe that their setup can be coupled with a lateral-flow strip or a cell-phone-based fluorescence detector and bring NAAT to LMIC. However, the analytical sensitivities of these assays have yet to be fully characterized.

Figure 5. Advances in establishing point-of-care amenable PCR toward developing NAAT-based STI POCTs.

Emerging technologies that can reduce the transition time between PCR temperatures within in a compact footprint have led to ultrafast PCR, challenging the perception that PCR may be ill-suited for developing NAAT-based STI POCTs. (A) Chan and colleagues developed a “Thermos Thermal Cycler” that is rapid and low-cost.¹⁵⁰ In this setup, 2 or 3 24-oz thermoses were filled with heated water to serve as water baths for performing PCR or RT-PCR. A pan-and-tilt servo, which was controlled via Arduino electronic controller and powered by a battery pack, was used to shuttle PCR vessels in and out of the thermoses, thereby shortening the time of temperature transition. (B) (i) Trick and colleagues developed PROMPT (Portable, Rapid, On-cartridge, Magnetofluidic, Purification, and Testing) – a coffee mug-sized platform that incorporated (ii) a unique heat block with optimally minimized thermal mass to perform 40 PCR cycles in 12 min.¹⁵⁴ PROMPT also integrated sample processing, allowing it to detect NG and genotype one of its AMR genetic determinant in 15 min directly from clinical samples.

Miniaturization of thermal cyclers offers an effective strategy for accelerating PCR. Lee and colleagues used a high powered Peltier heater (20 mm × 20 mm × 5 mm) to drive a miniature thermal cycler that could perform 35 PCR cycles in 3 min and at a cost of < US $110.¹⁵¹ In their palm-sized real-time PCR devices,^{152, 153} Ahrberg and colleagues micromachined silicon heaters that could achieve 20 °C/s heating and −20 °C/s cooling. The 100 to 200 nL drops of PCR reactions surrounded by mineral oil provided small thermal masses, which facilitated such rapid thermal cycling. In a STI-relevant demonstration, Trick and colleagues miniaturized a unique heat block with optimally minimized thermal mass to achieve ultrafast PCR within an integrated and portable platform (Figure 5B).¹⁵⁴ This coffee mug-sized platform, coined PROMPT (Portable, Rapid, On-cartridge, Magnetofluidic, Purification, and Testing) could complete 40 PCR cycles in 12 min. Moreover, PROMPT integrated sample processing, and so could detect NG and genotype one of its AMR genetic determinant in 15 min directly from clinical samples. Elsewhere, Lim and colleagues developed a Peltier-based miniaturized thermal cycler that could detect CT, NG, TV, and 3 other STI pathogens from endocervical swab samples using a commercial assay (Seeplex STD6 ACE, Seegene, Korea).¹⁵⁵ Unfortunately, as this commercial assay required lengthy hot start, hold time at each temperature, and final extension, it is unclear if this miniaturized thermal cycler in fact accelerated PCR. Moreover, it is also unclear how the PCR products were detected.

Convective PCR¹⁵⁶ obviates thermal cycling while achieving fast TAT. Qiu and colleagues developed such a device that was integrated with a lateral-flow strip read-out for the detection of H1N1 RNA in 35 minutes.¹⁵⁷ To address potential susceptibilities to environmental temperature fluctuations, Chang and colleagues added a thermal baffle to stabilize their convective PCR such that it could remain functional even in a wind tunnel and detect 10 copies of plasmid target in 10 to 20 min.¹⁵⁸ Over the years, convective PCR has seen continuous refinement and portable instruments have been developed. For example, Shu and colleagues developed a palm-sized instrument that integrated magnetic-based sample processing, laser-based photothermal lysis, and closed-loop convective real-time PCR to detect 10 copies/μL of Staphylococcus aureus DNA in 28 min.¹⁵⁹ Zhuo and colleagues developed a portable instrument for performing their capillary convective PCR for Influenza A detection in 25 min.¹⁶⁰ Their instrument could automatically adjust the temperature to perform reverse transcription, though the necessity for adding this function somewhat undercuts the appeal of convective PCR, which is supposed to obviate such temperature control function. Carossino and colleagues evaluated a commercial “insulated isothermal PCR” platform, which performs convective PCR in a capillary vessel (R-tube™, GeneReach, USA) and a field-deployable device (POCKIT™ Nucleic Acid Analyzer, GeneReach, USA), for the detection of Zika virus. This platform could detect 10 to 100 in vitro transcribed RNA copies/μL in ~90 min. Finally, Khodakov and colleagues recently engineered a novel, toroidal convective PCR method that incorporated in situ microarray-based multiplexed detection (up to 48 microarray spots) and was implemented within a portable instrument.¹⁶¹ The authors demonstrated multiplexed pathogen detection in a version of the chip that houses a microarray with up to 15 pathogenic bacteria in 30 min. The TAT for convective PCR in the literature varied, but versions that are rapid, integrated with detection, and impervious to environmental temperature fluctuations could offer an attractive option for developing NAAT-based STI POCTs.

5.3.2. LAMP

LAMP,¹⁶² which typically operates between 60 and 65 °C with a TAT of ≤ 60 min, is one of the most established isothermal NAATs and has been used for detecting STI pathogens. Liu and colleagues reported a LAMP assay that targets and detects the porA pseudogene of NG.¹⁶³ Shimuta and colleagues developed two LAMP assays to detect, in 60 min, NG strains carrying the penA-60.001 allele which are commonly associated with ceftriaxone resistance.⁹⁴ The authors characterized their assays with 14 WHO NG reference strains with different ceftriaxone susceptibilities and penA alleles, as well as 204 NG strains and 95 strains of other Neisseria spp. Becherer and colleagues developed a real-time, multiplex LAMP assay based on mediator displacement probes with universal reporters to detect STI pathogens including HIV and TP.¹⁶⁴

LAMP can be monitored via a variety of detection methods.¹⁶⁵ Among them, colorimetric readout has become a fixture and enables naked-eye detection, thus offering an attractive strategy for developing POCT. Choopara and colleagues employed hydroxy naphthol blue-based readout, which changes from violet to sky blue to indicate a positive LAMP reaction¹⁶⁶ to detect CT via naked eye in 45 min.¹⁶⁷ The authors evaluated this assay with 284 endocervical swab samples and reported 90 to 100% sensitivity and 95% specificity after a thermal lysis protocol. The same research team later coupled LAMP to post-LAMP gold nanoparticle-based detection – another common strategy for achieving naked-eye readout.¹⁶⁸ This assay could detect 11.25 copies of CT DNA in 32 to 47 min and achieved 96%/99% sensitivity/specificity when tested against 130 clinical endocervical swab samples.

LAMP has also been implemented within microfluidic devices,¹⁶⁹ some of which are applied to detect STI pathogens. Xu and colleagues developed a microfluidic chip that uses surface acoustic waves to achieve the necessary reaction temperatures for performing a multiplexed LAMP assay and melt curve analysis to detect CT and NG.¹⁷⁰ Yu and colleagues developed a self-partitioning SlipChip to perform digital LAMP for quantifying human papillomavirus (HPV) viral load.¹⁷¹ Shin and colleagues have implemented LAMP in a thermoplastic cartridge and a portable instrument for detecting CT from swab samples. They also demonstrated the user-friendliness of the platform by allowing untrained personnel to use the platform in an emergency department setting upon a brief, video-based training¹³² – thereby demonstrating near-POC use.

5.3.3. HDA

Helicase-dependent isothermal amplification, or HDA,⁶⁶ which operates at ~65 °C with a TAT of ≤ 90 min, is another mature isothermal NAAT,^{172, 173} and has been employed for developing both commercial NAAT-based STI diagnostic assays and potential NAAT-based STI POCT technologies in academic research. For example, the AmpliVue™ Trichomonas Assay (Quidel) – a FDA-cleared test for qualitative detection of TV DNA in female vaginal specimens – employs HDA in conjunction with a disposable lateral-flow detection device to achieve a ~45 min TAT. Gaydos and colleagues evaluated this test and found comparable sensitivity and specificity to another NAAT for the detection of TV from 992 patients (342 symptomatic and 650 asymptomatic).¹⁷⁴ Horst and colleagues also employed HDA in their paper device that can detect NG from urethral and vaginal swab samples within 80 min. The authors reported an analytical sensitivity down to 10 NG cells and 95% sensitivity and 100% specificity from 40 urethral and vaginal swab samples.¹⁷⁵

5.3.4. RPA

Recombinase polymerase amplification, or RPA,¹⁷⁶ which operates at ~37 °C with a TAT ≤ 60 min, has become a solid alternative to PCR and LAMP for developing NAAT-based STI POCTs. Its ~37 °C reaction temperature lowers the heating constraint for performing the reaction and opens the door to non-conventional heating sources (e.g., body heat).¹⁷⁷ Li and colleagues presented an excellent review on RPA.¹⁷⁸ A number of STI detection studies employed RPA assays, such as assays for detecting HIV-1 proviral DNA for early infant diagnosis^{177, 179} and detecting CT directly from urine samples.¹³⁹ In particular, in a multicenter cross-sectional preclinical evaluation, Harding-Esch and colleagues collaborated with staff at TwistDx – the company that manufactures RPA assays – and evaluated a 15-min RPA assay for CT and NG. Single-blinded testing was performed by the staff at TwistDx, and comparable sensitivity and specificity to routine clinic NAATs were reported.¹⁸⁰

RPA has been miniaturized in microfluidic devices and used for detection of STIs and other pathogens. Ereku and colleagues implemented RPA in a multi-chamber microfluidic cartridge that was heated by a custom shoebox sized instrument to detect spiked-in extracted CT DNA down to 1 copy/μL in 25 min.¹⁸¹ Yeh and colleagues developed a self-powered integrated microfluidic POC low-cost enabling (SIMPLE) chip, which incorporated vacuum-based self-powered microfluidic pumping (i.e., without any external pumps, controllers, or power sources), blood plasma separation, and RPA, to detect 10 – 10⁵ copies/μL of methicillin-resistant S. aureus DNA in ~30 min.¹⁸²

A rapidly emerging trend is coupling RPA and CRISPR-based detection for developing novel diagnostic assays. This trend was pioneered by Gootenberg and colleagues – who developed the SHERLOCK assay¹⁸³ that couples RPA to CRISPR-Cas13a (an RNA activated RNase) – and Chen and colleagues – who developed the DETECTR assay¹⁸⁴ that couples RPA with CRISPR-Cas12a (a DNA activated DNase). This combination has been used to develop tests for STI viral pathogens. The DETECTR assay developed by Chen and colleagues was used to detect HPV, a sexually transmitted virus that can lead to cancer later in life. As these early assays employ at least two steps, Ding and colleagues successfully integrated RPA and CRISPR into a single-step assay called AIOD-CRISPR and demonstrated the detection of HIV¹⁸⁵ and SARS-CoV-2 (the causative virus for COVID-19)¹⁸⁶. This single-step assay represents a key advance that may truly propel assays that couple RPA and CRISPR-based detection in the development of practical diagnostic tools, including NAAT-based STI POCT technologies.

5.3.5. Other Isothermal NAAT

Besides LAMP, HDA, and RPA, several lesser-known isothermal NAAT techniques such as Cross Priming Amplification (CPA), Simultaneous Amplification and Testing (SAT), and Multiplex Strand Invasion Based Amplification (mSIBA) have also been reported in the literature. CPA is carried out with 5 primers – 1 of which being the cross primer – and a strand displacement DNA polymerase to detect DNA at 63°C in 60 to 90 min.¹⁸⁷ The CPA assays developed by Yu and colleagues could detect 45 and 65 copies of CT and NG DNA per reaction, respectively. They coupled the assay to disposable cartridges that allowed CPA amplicons in closed reaction tubes to be detected via lateral flow strips. They tested 80 cervical or vaginal swab specimens with both CPA and qPCR and found 98.8% and 97.5% agreement between the two methods for CT and NG, respectively.¹⁸⁸ Currently, Ustar Biotechnologies (China) manufactures and distributes CPA-based assays for detecting CT and NG, as well as the disposable cartridges. SAT is a RNA detection method and has been used to detect rRNA (3 to 4 orders of magnitude more abundant than genomic DNA) of STI pathogens. SAT uses a sense primer with T7 promotor sequence, an antisense primer, a reverse transcriptase, and T7 RNA polymerase to exponentially generate RNA amplicons in a 42 °C reaction in ~40 min.¹⁸⁹ Currently, Shanghai Rendu Biotechnology (China) manufactures and distributes SAT-based assays for detecting CT, NG, and 2 other STI pathogens. Liang and colleagues compared SAT and qPCR in detecting CT and NG from 123 urogenital swab samples and found that SAT outperformed qPCR in detecting samples < 1 × 10³ DNA copies mL⁻¹.¹⁹⁰ However, SAT currently requires the enzymes to be added after pre-incubation steps (60 °C for 10 min and then at 42 °C for 5 min); this workflow adds a wrinkle for the development of POCT technologies. Eboigbodin and Hoser developed mSIBA¹⁹¹ (an extension from SIBA¹⁹² reported previously by the same research team) that can simultaneously detect CT, NG, and an internal control in the same reaction tube at 40 °C in 60 min. In mSIBA, an invasion oligonucleotide is aided by a recombinase (UvsX) to invade the double-stranded target nucleic acid, hybridize to the complementary strand, and create room for target-specific primers and fluorescently-labeled probes that flank the invasion oligonucleotide to hybridize to the target strands, initiate polymerization, and produce fluorescence signals. The authors found comparable analytical sensitivities and specificities between mSIBA and a previously reported multiplex PCR assay, but mSIBA had shorter TAT.¹⁹¹ However, the POC amenability of mSIBA remains to be evaluated.

5.4. Multiplex

Multiplexing is critical for the development of NAAT-based STI POCT technologies. Ideally, a STI POCT must simultaneously screen for CT, NG, and TV from a single specimen because the same clinical symptoms can be caused by them and mixed infections are common. Inclusion of AMR detection for NG is also desirable. Multiplexing could also potentially increase throughput and data generation while simplifying formats and decreasing the time and cost required to perform several tests, which makes them more fit for use in LMIC than current POCT platforms. Of note, conventional multiplex PCR is unlikely to extend beyond 3- or 4-plex detection, and consequently may be insufficient for detecting STI pathogens and AMR. Such a deficiency therefore creates the need for other multiplexing strategies. Kim and colleagues recently reviewed multiplex molecular POCT for syndromic infectious diseases.¹⁹³ We highlight herein some multiplexing strategies that have been applied to detecting STI pathogens.

5.4.1. Parallelization via Microfluidic Devices

A straightforward multiplexing strategy is parallelizing single-plex assays in independent channels or wells in microfluidic devices. Turigan and colleagues used this strategy in a multi-lane biochip¹⁹⁴ to perform 14 single-plex microfluidic PCR reactions in parallel for rapid detection and strain typing of CT.¹⁹⁵ The 22-min microfluidic PCR reactions were coupled with 10-min DNA processing of clinical swabs and 28-min gel electrophoresis-based detection, resulting in an integrated platform with 60-min swab-in-result-out capability.¹⁹⁵ Kriesel and colleagues reported a research-use-only STI panel based on the FilmArray® system (BioFire Diagnostics, LLC, Salt Lake City, Utah), which employed multiple PCR primer sets for detecting CT, NG, TV, TP, and 5 other relevant STI pathogens.¹⁹⁶ The authors found 98% and 97% concordance between their FilmArray® STI panel and standard NAAT for CT and for NG, respectively, and suggested that their test has the potential to improve public health by providing rapid, sensitive, and reliable results in clinics or nearby laboratories.¹⁹⁶

5.4.2. Novel PCR Probe Design

The aforementioned ResistancePlus® assay uses a novel PCR probe technology called PlexZyme, which is based on MNAzyme qPCR¹⁹⁷ and is currently commercialized by SpeeDx (Australia), has been used to detect NG and its ciprofloxacin resistance gene (gyrA)^{198, 199} and M. genitalium and its azithromycin resistance gene (23S rRNA).^{68, 200} This technology uses a pair of partial nucleic acid enzymes to create a new class of qPCR probes with enhanced specificity and improved capacity for multiplexing when compared to qPCR using Taqman probes or molecular beacons, which have been empirically demonstrated.¹⁹⁷

5.4.3. Melt Curve Analysis

Post-amplification melt curve analysis offers a simple strategy for multiplexing and has been leveraged to develop multiplexed STI pathogen detection and AMR detection. Hu and colleagues designed and evaluated a multiplex real-time PCR melting curve assay that can detect CT, NG, TV, and 6 other STI pathogens after manual DNA extraction.²⁰¹ This assay appears to have both universal primers and species-specific primers in one-pot and is performed in a closed-tube and single-step format. Donà and colleagues employed melt curve analysis to identify NG and detect AMR. Their method, which included one triplex and three duplex reactions, could identify NG and detect seven genetic determinants of AMR to extended-spectrum cephalosporins (ESCs), azithromycin, ciprofloxacin, and spectinomycin.⁷³ Twin and colleagues developed a melt curve-based assay to detect genetic mutations from MG associated with resistance to azithromycin – a commonly used antibiotic for the treatment of MG infections.²⁰² Lengthened assay time can be a disadvantage for melt curve analysis. Also, as multiple targets may result in an unresolvable composite melt curve, melt curve analysis can have a limited capacity for detecting co-infections.^{203, 204} Compatibility with isothermal NAATs is another issue – although temperature control requirement for melt curve analysis would not be as stringent as PCR, the advantage brought forth by isothermal NAATs would be negated if melt curve analysis is performed.

5.4.4. Probe Hybridization

Hybridization between amplicons and specific probes is a well-known strategy for multiplexing, but it increases assay time, adds a step to the assay, and requires either a microarray chip or Luminex beads. Chung and colleagues developed multiplexed PCR reactions and a microarray chip for the diagnosis of genitourinary infections and identified 14 of the most prevalent pathogens, including CT, NG, TV, and MG among Koreans. They subsequently tested their chip with DNA extracted from clinical samples.²⁰⁵ Schmitt and colleagues developed and validated a bead-based multiplex sexually transmitted infection profiling (STIP) assay that detects 18 STI pathogens from urogenital samples using a multiplex PCR followed by Luminex bead-based hybridization. STIP demonstrated an overall concordance of 95–100% with commercially available quantitative PCR tests.²⁰⁶ Though broad, this method is unlikely to be used at the POC due to its reliance on Luminex beads, which can only be analyzed through dedicated instruments.

5.4.5. Ratiometric Fluorescence Detection

Zhang and colleagues developed a new multiplexed detection approach termed ratiometric fluorescence coding, in which each nucleic acid target is encoded with a specific ratio between two standard fluorophores. In the initial demonstration of this strategy, the authors employed the padlock probe chemistry and rolling circle amplification (RCA) to transform each nucleic acid target into a specific template with a specific number of binding sites for two probes labeled with different fluorophores, thereby achieving multiplexed detection. The authors detected DNA targets from six infectious diseases, including HIV, CT, NG, and TP.²⁰⁷ Despite the demonstrated multiplexing capacity, this ratiometric fluorescence coding assay still must be simplified into a single-step assay to facilitate POC use.

5.5. Explore Detection Modality

Portable detectors are essential to the development of NAAT-based STI POCT technologies. Current POCT technologies predominantly use fluorescence-based and lateral-flow-based detection, but other detection methods such as electrochemical detection may provide new opportunities. Several review articles can provide a starting point for such an exploration. Pashchenko and colleagues compared optical, electrochemical, magnetic, and colorimetric POC biosensors for infectious disease diagnosis.²⁰⁸ Yang and colleagues reviewed detection platforms for POC testing based on colorimetric, luminescent and magnetic assays. They designated a section on portable and handheld devices based on these detection modalities.²⁰⁹ Yang and colleagues and Ding and colleagues reviewed recent developments in mobile sensing based on the integration of microfluidic devices and smartphones.^{210, 211} The ubiquity and expanding capabilities of smartphones make them (in principle) useful detectors for NAAT-based STI POCT technologies, especially in academic research. However, standardization and regulatory approval for the wide variety of smartphones remain challenging.

5.5.1. Fluorescence Detection

Portable fluorescence detection can be realized in a variety of setups. Among them, low-cost fluorescence detection can be achieved by employing inexpensive UV light²¹² (US $2) or LEDs²¹³ as the excitation sources and relying on naked-eye visualization. Though in practice, a portable imaging device such as a digital camera²¹² or a smartphone²¹³ was still incorporated in such a setup. Smartphone-based fluorescence detection has become common and was incorporated in several NAAT-based STI POCT technologies.^{132, 213–216} Damhorst and colleagues reported a smartphone-imaged RT-LAMP on a chip, which was further coupled to a blood separation chip to detect HIV-1 from whole blood.²¹⁴ Alternatively, several research groups employed portable commercial fluorescence detectors^{121, 154, 217} or fluorescence microscopes.^{218, 219} These detectors and microscopes may incur additional cost to the instrument for executing POCTs, but they are already commercialized and standardized, which can be advantageous during technology development. A particularly interesting quantitative detection strategy that employed a portable fluorescence microscope was introduced by Liu and colleagues, who coined the strategy Nuclemeter.²¹⁹ Nuclemeter was based on reaction diffusion, in which the copy number of target nucleic acids undergoing enzymatic amplification was inferred from the position of the reaction-diffusion front, analogous to reading temperature in a mercury thermometer. Using RT-LAMP, the authors demonstrated comparable HIV-1 viral load quantification between Nuclemeter and conventional benchtop methods. Moreover, in contrast to conventional time-course monitoring of fluorescence signals, Nuclemeter needed only one time point to observe the signal and quantify target nucleic acids, which could offer a simple alternative for LMIC.

5.5.2. Naked-Eye Detection

Naked-eye detection is most often achieved through lateral flow strip-based readout and colorimetric readout. For example, lateral flow strips were employed to detect TV after HDA,¹²⁶ NG after HDA,¹⁷⁵ CT and NG after CPA,¹⁸⁸ and HIV after RT-LAMP.²²⁰ Moreover, the FDA-cleared AmpliVue™ Trichomonas Assay (Quidel) detects TV by employing HDA in conjunction with a disposable lateral-flow detection device. Similarly, colorimetric readouts were coupled to LAMP to detect CT.^{167, 168} Simplicity is the main advantage of naked-eye detection. On the other hand, as diagnostic results should be ideally acquired without user interpretation and recorded in a database without manual entry, a readout device (e.g., smartphone) that can accomplish both is ideally still needed to accompany naked-eye detection.

5.5.3. Electrochemical Detection

Electrochemical detection is a mature method for the detection of biomolecules including nucleic acids.²²¹ Electrochemical nucleic acid sensors typically employ electrodes and redox tags to produce an electronic signal (e.g., current) in the presence of the target. Besides the commercially available binx io® platform, electrochemical detection has only been employed sparingly in academic research and development of NAAT-based POCT technologies for the detection of STIs and HIV. For example, Singh and colleagues developed an electrochemical sensor for detecting NG DNA²²² and Wang and colleagues developed an exonuclease-amplified electrochemical sensor for detecting host HIV-1 DNA,²²³ but neither incorporated NAAT. Nevertheless, integration of NAAT, electrochemical detection, and microfluidics is well established for various non-STI pathogens,^224–227 which can serve as inspiration for developing electrochemical- and NAAT-based STI POCT technologies. Indeed, electrochemical detection has been coupled to a variety of NAAT techniques,^{228, 229} including PCR,^230–233 LAMP,^234–240 HDA,^{241, 242} and RPA,²⁴³ often with real-time measurements²⁴⁴ in microfluidic devices. Electrochemical melt curve analysis of DNA^{245, 246} and multiplex electrochemical DNA detection²⁴⁷ have also been demonstrated. Of note, many of these works were research prototypes that use microfabricated gold electrodes on glass substates, PDMS-based microfluidics, and benchtop potentiostats (i.e., the instruments for executing electrochemical detection), rendering them unsuitable as POCTs in LMIC. In response, there have been considerable research efforts toward developing inexpensive electrochemical sensors based on disposable, screen-printed electrodes^{248, 249} and paper electrodes,^250–252 constructing portable, low-cost, and even open-source potentiostats,^253–258 and coupling electrochemical detection with smartphones.^{259, 260} These developments suggest that there is considerable foundation but still ample opportunity for the development of electrochemical- and NAAT-based STI POCT technologies.

5.6. Explore Device Materials

The development of microfluidic NAAT-based POCT technologies has historically relied on PDMS and glass as the device materials. Unfortunately, devices made from PDMS and glass face significant challenges for POC use, especially in LMIC, because they can be fragile and pose difficulties for non-trained users. Furthermore, the fabrication processes of these devices from various research organizations often lack standardization, which can be an insurmountable challenge for the development of diagnostic tests. Therefore, technology developers should move past PDMS and glass and actively explore alternative device materials such as paper and plastic for the development of POCTs for STIs.

5.6.1. Paper Devices

Paper devices (often referred to as microfluidic paper-based analytical devices, or μPADs) have become a mainstay in microfluidics. Paper devices generally have low cost and can obviate external pumping system, making them attractive not only for the development of POCTs in LMIC, but potentially also for home-based testing. The state-of-the-art for paper devices was reviewed in 2013.²⁶¹ Since then, several others have reviewed the landscape of paper microfluidic devices,^{262, 263} advances toward broad diagnostic applications,^{264, 265} advances and challenges of fully integrated paper-based POC nucleic acid testing,²⁶⁶ and those that incorporated NAAT for detecting infectious diseases.²⁶⁷ These reviews also pointed out several areas that require further advances, including integration with sample collection tools (e.g., for precise sample volume), automation in fluidic delivery in paper, on-board reagent storage, reduction in manual operation, multiplexed detection, thorough examination on long-term stability, and incorporation of simple readout timer to track the assay endpoint for accurate readout or simple, but quantitative readout module (e.g., smartphone).^{265, 266}

Several paper devices for the detection of STI pathogens have been reported. Rohrman and colleagues developed a paper and plastic device that can perform RPA to amplify 10 copies of HIV DNA to detectable levels in 15 min. The device stores lyophilized enzymes, facilitates mixing of reaction components, performs RPA, and incorporates commercially available lateral flow strips for detection in five steps of operation. The authors believe that their device may serve as part of a POC HIV DNA test in LMIC,²⁶⁸ though the device was designed to operate after conventional DNA extraction from dried-blood spots, which must be integrated to reduce operational complexity in LMIC. Linnes and colleagues developed a paper device for CT detection by inserting chromatography paper into a micropipette tip.²⁶⁹ In this device, the bacterial cells were trapped on the paper, and then the captured cells were directly amplified and detected via in-situ HDA (i.e., in the pipette tip). As such, their diagnostic platform incorporates cell lysis, isothermal nucleic acid amplification, and lateral flow visual detection using only pressure and heat sources, eliminating the need for expensive laboratory equipment while achieving 100× more sensitive detection than current rapid chlamydia immunoassays in 1 h.²⁶⁹ Finally, Horst and colleagues developed a paperfluidic device to detect NG from urethral and vaginal swab samples within 80 min (Figure 6A).¹⁷⁵ The device integrates patient swab sample lysis, PES membrane-based nucleic acid extraction, HDA, an internal amplification control, and visual lateral flow detection, albeit at the expense of several manual steps. The authors achieved an analytical sensitivity of 10 NG cells per device and 95% sensitivity and 100% specificity from 40 urethral and vaginal swab samples.¹⁷⁵

Figure 6. Advances in exploring device materials, reducing power dependence, and pre-storing reagents toward developing NAAT-based STI POCTs.

(A) Horst and colleagues developed a paperfluidic device to detect NG from vaginal or urethral swab samples within 80 min. The device integrates patient swab sample lysis, PES membrane-based nucleic acid extraction, HDA, an internal amplification control, and visual lateral flow detection within an 80 min run time.¹⁷⁵ (B) Kong and colleagues developed a wearable microfluidic device combined with recombinase polymerase amplification (RPA) for simple and rapid amplification of HIV-1 DNA using human wrist body heat. With the aid of a cellphone-based fluorescence detection system, this device detected HIV-1 DNA in 24 min.²¹³ (C) (i) Song and colleagues developed a multifunctional NAAT reactor that combined membrane-based extraction, concentration, and purification of nucleic acids, refrigeration-free storage of LAMP reagents with just-in-time release, and on-chip amplification and detection of HPV16 DNA and HIV RNA.²¹⁸ (ii) The plastic reactor chip was made with poly(methyl methacrylate) (PMMA).

Inspiration for advancing paper- and NAAT-based STI POCT technologies can be found from paper devices that target non-STI pathogens. For example, Connelly and colleagues reported a multilayer paper device with a central patterned paper strip that can slide in and out of the fluidic path, which allows the introduction of sample, wash buffers, LAMP reaction mix, and detection reagents with minimal pipetting within the device.²⁷⁰ The testing result from the device is detected using a hand-held UV source and camera phone, which may be suitable for POC use in LMIC. Starting with human plasma spiked with E. coli, the authors reported a limit of detection of 5 cells in 1 h. Tang and colleagues developed a paper device that not only integrates nucleic acid extraction, HDA, and lateral flow assay detection, but also incorporates on-chip dried reagent storage and an integrated battery, which may facilitate POC use.²⁷¹ The authors demonstrated the detection of 10² CFU mL⁻¹ S. Typhimurium in wastewater and egg and 10³ CFU mL⁻¹ in milk and juice in ~1 h. Trinh and colleagues developed a foldable paper device with dried PCR reagents on paper and FTA cards to detect 3 major foodborne pathogens: Salmonella spp., S. aureus, and E. coli O157:H7.²⁷² Reboud and colleagues developed another foldable paper device that combines whole-blood sample preparation, LAMP, and a lateral flow-based visual readout to detect malaria from a finger prick of whole blood in < 50 min. The authors evaluated their technology in village schools in Uganda and detected malaria in 98% of infected individuals in a double-blind first-in-human study.²⁷³ Finally, an interesting technical advance was reported by Hamedi and colleagues, who integrated paper microfluidics and printed electronics in monolithic devices.²⁷⁴ This concept may be useful for advancing paper STI diagnostic tools in the future.

5.6.2. Plastic Devices

Plastic was once considered an attractive alternative material to silicon and glass for fabricating microfluidic devices due to its lower cost and greater compatibility with biology as well as manufacturing and commercialization.²⁷⁵ Although it had lost favor to PDMS over the past two decades,²⁷⁶ especially in the academic research community, its appeal for commercialization remains. Advances in fabrication methods such as laser-cutting, computer numeric control (CNC) machining, micromilling,²⁷⁵ and increasingly, 3D printing,^{277, 278} have invigorated the push for developing plastic microfluidic devices and even democratizing such developments to makerspaces.²⁷⁹ For the development of NAAT-based STI POCT technologies, some researchers – most notably Liu, Mauk, Bau, and colleagues – have remained focused on employing plastic materials such as poly(methyl methacrylate) (PMMA)^{121, 212, 218, 280} and polycarbonate (PC)^{281, 282} to fabricate microfluidic cartridges in their technologies.

In addition to conventional plastic cartridges with planar designs, plastic devices can be fabricated into different structures that can improve assay performance. For example, Shin and colleagues developed thermoplastic cartridges with extruded wells that were thermoformed from a 0.2-mm-thin PMMA film.²¹⁷ The same research team subsequently improved the fabrication process by employing a polypropylene film.¹⁵⁴ The extruded wells in these thermoplastic cartridges were compatible with miniaturized thermal cyclers that could perform ultrafast PCR.¹⁵⁴ Yang and colleagues employed Ecoflex to develop a bandage-like, flexible, and wearable microfluidic chip that uses body heat to perform RPA and detect as few as 10 copies/μL of a conserved nucleic acid fragment of the Zika virus within 10 min via a visual readout.²⁸³ The results suggest that this flexible plastic microfluidic chip could be readily adapted for detecting STI pathogens.

5.7. Reduce Power Dependence

Electricity may be unstable or even inaccessible in LMIC, which presents an additional consideration for developing NAAT-based STI POCT technologies. As such, technologies that are self-powered will be more suitable for meeting the need in such challenging environment. The most straightforward power source is battery. Alternative power sources include exothermic chemical reactions, body heat, and solar power. Choi reviewed advances in powering POCT technologies, with particular emphasis on use in LMIC.⁷⁹

5.7.1. Battery

Battery is perhaps the most straightforward strategy for powering POCT technologies in the absence of reliable electricity. As smartphone-integrated systems will be more generally used for POCTs, embedded batteries will be the primary power source for this type of application.²⁸⁴ For battery-powered devices, incorporation of heaters with low power consumption can lengthen the battery life. Byers and colleagues reported the development and application of resistive microheaters printed with nanosilver ink for improved heating in paper POC and in-field diagnostics.²⁸⁵ They pointed out the predictability and versatility of resistive heating and demonstrated both isothermal nucleic acid amplification at 65°C and 16-hour bacterial culture at 37 °C using their microheaters and low-power battery heating. The heaters cost between US $0.17 and $0.58, are reusable and stable over 6 months, and can be wetted without degradation or reduction in conductivity. These versatile printed microheaters enable thermal control for a variety of low power heating applications.

5.7.2. Exothermic Chemical Reaction

Exothermic chemical reactions (e.g., magnesium alloy reacting with water) present an option for performing isothermal NAAT assays such as LAMP. LaBarre and colleagues developed a non-instrumented nucleic acid amplification (NINA) heater based on exothermic chemical reactions and engineered phase change materials, and demonstrated that their NINA heater could perform LAMP assays comparably to commercially instruments in detecting genomic DNA from P. falciparum – the protozoan parasite that causes malaria.²⁸⁶ Curtis and colleagues developed a rapid, single-use nucleic acid amplification device (NINA-SUD) that used exothermic chemical reaction to perform RT-LAMP for HIV-1 from whole blood.²⁸⁷ Notably, the authors also lyophilized HIV-1 RT-LAMP reagents and demonstrated its stability over time.²⁸⁷ Liu and colleagues constructed a PMMA cartridge to perform water-triggered exothermic chemical reactions of a phase-change material for heating a LAMP assay that detects E. coli DNA.²¹² Liao and colleagues implemented exothermic chemical heat within an “smart cup” and integrated a smartphone to perform real-time fluorescence-based LAMP, which allowed them to detect 100 copies of HSV-2 DNA.²¹⁵ Singleton and colleagues implemented exothermic chemical reaction – whose material cost was estimated US $0.06 per test – within a thermos, thereby achieving an ambient temperature operating range of 16 °C to 30 °C and a thermal standard deviation at its operating temperature of < 0.5 °C.²²⁰ They employed their heater to perform RT-LAMP and a lateral flow strip for visual readout, and demonstrated sensitive and repeatable detection of HIV-1 with a ß-actin positive internal amplification control from processed sample to result in less than 80 minutes.²²⁰ It is foreseeable that these devices can be adapted to detect STIs.

5.7.3. Body Heat

Body heat may be harnessed to perform isothermal amplification methods that require low reaction temperatures, such as RPA. A few research groups demonstrated using body heat to incubate RPA reactions for detecting HIV-1 DNA.^{177, 213} For example, Kong and colleagues developed a wearable PDMS device that can be wrapped around the human wrist, where the body temperature varied from 33 – 34°C in the ambient environment, to perform RPA reactions (Figure 6B). With the aid of smartphone-based fluorescence detection, this device detected HIV-1 DNA quantitatively ranging from 10² to 10⁵ copies mL⁻¹ in 24 min. This wearable POC nucleic acid testing method offers advantages in time, portability, and independence from reliable electricity.²¹³ Despite these advantages, the PDMS device should be replaced in the future.

5.7.4. Solar Power

Solar power offers a good option in LMIC and has been employed to perform both PCR and LAMP. Jiang and colleagues developed solar thermal PCR for smartphone-assisted molecular diagnostics. They integrated solar heating with microfluidics to eliminate thermal cycling power requirements as well as create a simple device infrastructure for PCR that could be completed in < 30 min while consuming 80 mW power, which allowed a standard 5.5 Wh iPhone battery to provide 70 h of power to this system.²¹⁶ Snodgrass and colleagues described a portable device that can perform LAMP with power from electricity, sunlight or a flame, and that can store heat from intermittent energy sources for operation when electrical power is not available or reliable. They deployed the device in two Ugandan health clinics, where it successfully operated through multiple power outages, with equivalent performance when powered via sunlight or electricity.²⁸⁸ Although this portable and electricity independent system was not used to detect STIs, it is foreseeable that STI detection can be achieved.

5.8. Pre-store Reagents

For the development of POCTs, devices that pre-store all necessary reagents not only simplify operation for untrained users, but also obviate cold storage – a key benefit for POC use in LMIC. Several such devices have been reported for HIV detection. Song and colleagues developed a multifunctional NAAT reactor that combined membrane-based extraction, concentration, and purification of nucleic acids, refrigeration-free storage of LAMP reagents with just-in-time release, and on-chip amplification and detection of HPV16 DNA and HIV RNA.²¹⁸ The authors encapsulated lyophilized LAMP reagents in paraffin wax and pre-stored the wax pellet within the reactor (Figure 6C). When the temperature in the reactor increased to the amplification temperature, the paraffin encapsulating the reagents would melt and float away such that the reagents would be hydrated, just-in-time, for the LAMP reaction to proceed. A shelf life of 6 months was reported, but the authors also found that the assay reproducibility of their reaction was worse than that of the benchtop instrument.²¹⁸ In the NINA-SUD device developed by Curtis and colleagues, which used exothermic chemical reaction as the heat source, the HIV-1 targeting RT-LAMP reagents were lyophilized and found to be stable.²⁸⁷ Phillips and colleagues reported their microfluidic rapid and autonomous analytical device (microRAAD) to detect HIV from whole blood samples,²⁸⁹ in which they deposited and dried RT-LAMP reagents in polyethylene terephthalate (PET) film based on a published protocol.²⁹⁰ These examples indicate the feasibility of incorporating pre-stored NAAT reagents toward developing NAAT-based STI POCT technologies. In addition to these examples, Deng and Jiang reviewed recent advances in reagent storage and release in POC devices,²⁹¹ which could provide a helpful resource for developing STI POCTs with pre-stored reagents.

5.9. Construct Integrated Devices and Instruments

Despite seemingly significant progress in discrete components, devices that are fully integrated and POC-amenable are scarce, and POCT technologies that either can or have the potential to be deployed in LMIC are rare. Some researchers have presented prototypes, shown preliminary results, and discussed their insights,^292–295 while others have developed near-POC instruments^{280, 296, 297} rather than POCTs (as defined by this review). Among those, Bau and colleagues provided a particularly informative narrative of their quest for developing instrument-free NAAT-based diagnostic tools, which often focus on HIV.²⁹⁵ Examples of near-POC instruments include an all-in-one microfluidic nucleic acid diagnosis system developed by Ye and colleagues, which integrated centrifugal microfluidic chip with multiple units, membrane-based sample processing, and a new isothermal NAAT to simultaneously detect CT, NG, and 2 other STI pathogens from genitourinary secretions with minimal manual manipulations in ~50 min.²⁹⁶ Although this system demonstrates solid performances, preprocessing steps (e.g., external centrifugation) were still required. It is also unclear what the cost is or if the power consumption could be reduced. This instrument is therefore more like an alternative to GeneXpert, rather than a POCT. Similarly, Geng and colleagues developed a “3D extensible” microfluidic cassette technology with a tabletop control instrument that could directly accept contrived swab samples with spiked and dried CT, lyse cells, purify DNA, perform LAMP with pre-stored reaction mixture, and perform real-time fluorescence detection in ~82 min (Figure 7A).²⁸⁰ Mauk and colleagues recently reviewed miniaturized devices for NAAT-based HIV POCT technologies, in which they pointed out that there remain unmet needs for fully integrated, “sample-to-result” devices that are also low-cost.²⁹⁸ The same unmet needs apply to other STIs, especially given that the development of HIV POCTs already outpaces that of other STIs.

Figure 7. Advances in constructing integrated devices and instruments and incorporating internet connectivity toward developing NAAT-based STI POCTs.

(A) (i) Geng and colleagues developed a “3D extensible” microfluidic cassette technology that could pre-store reagents in valve compartments and directly accept contrived swab samples with spiked and dried CT, lyse cells, purify DNA, perform LAMP with pre-stored reaction mixture.²⁸⁰ (ii) Within the reaction chamber of the cassette (76.5 mm × 10 mm × 32 mm), a piece of chitosan-modified filter paper was embedded for capturing DNA and facilitating LAMP in situ. (iii) The cassette was inserted into a custom instrument (comparable to shoe box in size) to perform real-time, fluorescence-based LAMP in ~82 min. (B) (i) Phillips and colleagues developed microfluidic rapid and autonomous analytical device (microRAAD) to detect HIV from whole blood samples in 90 min.²⁸⁹ Inside microRAAD, the authors incorporated thermally-actuated valves for fluidic control, a temperature control circuit for low-power heating, dried RT-LAMP reagents, and a lateral flow strip for detection. (ii) microRAAD is compact and battery powered. (C) (i) Song and colleagues developed a hand-held, smartphone-empowered “smart-connected-cup” that can detect HIV virus from blood in 45 min and demonstrated spatiotemporal disease mapping function via GPS location of the smartphone.³⁰⁵ (ii) Smart-connected-cup incorporated exothermic chemical reaction for performing LAMP without electricity and used the smartphone to detect the bioluminescence signal from the LAMP reaction without an excitation source or optical filters, which reduced the cost. (iii) LAMP reactions were performed in custom plastic chip with four reactors and on-chip, membrane-based nucleic acid extraction.

Nevertheless, potential POCT technologies have been reported. An example is the aforementioned microRAAD that could detect HIV from whole blood samples (Figure 7B).²⁸⁹ Inside microRAAD, the authors incorporated thermally-actuated valves for fluidic control, a temperature control circuit for low-power heating, dried RT-LAMP reagents, and a lateral flow strip for detection. As a side note, this RT-LAMP employed Bst 3.0 polymerase, which includes reverse transcriptase capabilities. Even after three weeks of room-temperature reagent storage, microRAAD could automatically isolate the virus from whole blood, amplify HIV-1 RNA, and transport amplification products to the internal LFIA, detecting as few as 3 × 10⁵ HIV-1 viral particles, or 2.3 × 10⁷ virus copies per mL of whole blood, within 90 min. The current version of microRAAD, however, still must improve its analytical sensitivity and TAT.

5.10. Incorporate Internet Connectivity

The connected age has created a growing need for POCT technologies to be internet connected, thus facilitating so-called eHealth (i.e., application of information and communication technologies to healthcare)^{299, 300} and mHealth (i.e., application of mobile devices to healthcare)^{299, 301} across the globe. Indeed, such internet-connected POCT technologies offer new means to diagnose, track, and control infectious diseases and to improve the efficiency of the health system.³⁰¹ There are several review articles on this burgeoning research field.^{210, 300–303}

Integration of smartphones into the POCT technologies provides a natural pathway to establishing connectivity. Although smartphones have already become a significant part of new POCT technologies in academic research, they had been more often used for detection³⁰⁴ rather than internet connection. For NAAT-based STI POCT technologies, several researchers have incorporated smartphones,^{132, 213–216} but leveraging the smartphones for connectivity has been uncommon as well. A notable example was presented by Song and colleagues, who developed a hand-held “smart-connected-cup” that can detect HIV virus from blood in 45 min and demonstrated spatiotemporal disease mapping function (Figure 7C).³⁰⁵ Smart-connected-cup incorporated exothermic chemical reaction for performing LAMP without electricity and used the smartphone to detect the bioluminescence signal from the LAMP reaction without an excitation source or optical filters, which reduced the cost. For demonstrating its connectivity, the authors designed a website to map GPS locations of tests and test results performed by the smart-connected-cup. The authors believe that their smart-connected-cup represents a smart- and connected-diagnostic system that can obviate lab facilities and facilitate use at home, in the field, in the clinic, and particularly in LMIC, which exemplifies what the authors called “Internet of Medical Things” or IoMT.

5.11. Evaluate Technologies

Clinical evaluation of NAAT-based STI POCT technologies, which can be time-consuming and resource-intensive, should be thoroughly considered during technology development. A typical study entails measuring the sensitivity, specificity, and predictive values, and determining optimal operational characteristics of STI POCTs in sample sets from a variety of populations within laboratory and real-world settings for both screening for asymptomatic STIs and for case management. Additionally, the clinical utility should be comprehensively evaluated, by morbidity/mortality and adverse event profiles of the target infections. The aforementioned IEG for STI POCTs recognized the lack of independent evaluations of currently available POCTs, both laboratory-based and clinic-based, and utility evaluations. The IEG has called for the development of standardized evaluation protocols for novel STI POCT technologies across all stages of the evaluation pipeline,⁷⁹ as well as additional clinical and implementation research that can generate evidence for use of novel STI POCT technologies, which may address clinical and operational challenges in their global deployment.⁴⁴

On the other hand, it is worth noting that technologies that had been clinical evaluated varied substantially in their stages of development. For example, Smith and colleagues evaluated Atomo HIV Self Test – a device developed by a company called Atomo Diagnostics – and concluded that the high sensitivity, specificity, acceptability, usability, and fidelity of this device makes it an appropriate option for the enhancement of HIV testing strategies for harder to reach populations, such as young people and men.³⁰⁶ On the other hand, Melendez and colleagues evaluated the performance of prototype microwave-accelerated metal-enhanced fluorescence (MAMEF) assay that target either 16S rRNA or cryptic plasmid of CT with 257 dry vaginal swabs.³⁰⁷ Although capable of < 10 min detection, the MAMEF assays required manual operation from trained personnel. This example illustrates that clinical evaluation of potential POCT technologies could begin at an early stage without the need for a polished product.

6. CASE STUDIES ON DEVELOPMENT OF NAAT-BASED STI POCTS

As examples of POCT development, we describe the evolution of two STI POCTs – one in industry and one in academia. The industry case focuses on the evolution of binx io®, which has evolved significantly over the years to reach its present status of FDA clearance and CE approval. The academia case focuses on the droplet microfluidics technology, which the Wang group (the authors of this review) pioneered and have been using as the backbone technology for developing STI POCTs. Such case studies can provide insights into developing new STI POCT technologies and illustrate the importance of collaborating with clinical researchers and end-users.

6.1. Evolution of the binx io® for CT/NG – Perspective from a Small Biotech

Binx health (formerly Atlas Genetics) was founded in 2005 as a small biotechnology company with less than 15 employees and limited resources for the development of POCT. Using funding provided by the NIH through the Center for POC Technologies Research for STDs at Johns University, binx developed a rapid (< 25 min) POC assay for CT, which was later clinically validated in collaboration with Johns Hopkins Researchers.³⁰⁸ Their platform includes a novel technique for detection of electrochemical labels in a fully integrated fluidic card for sample processing and reagent handling. The CT assay includes extraction of chlamydial DNA, amplification of small sequence of the CT genome, and detection of the amplified DNA using the electrochemically labeled DNA probe. The assay is sensitive with an LOD of 25 chlamydial elementary bodies (EB) and free from cross-reactivity with other microorganisms commonly present in human genital swab samples.³⁰⁸ The clinical sensitivity and specificity, determined using 306 clinical samples, was 98.1% and 98.0%, respectively.

Following the development of the CT assay, Atlas Genetics continued to improve their platform. In 2013, they reported on the preliminary results of their TV test on the now-called io® platform.³⁰⁹ Though the platform was still at the prototype stage, it performed well in this study, which included 90 clinical samples, and achieved 95.5% sensitivity and 95.7% specificity. In 2014, Atlas Genetics completed the development of io® cartridge and set up manufacturing capability at Bespak (United Kingdom, a subsidiary of Consort Medical), which established a manufacturing line within a dedicated ISO Class 8 cleanroom. For the development of io® Reader, Atlas Genetics first partnered with TTP (United Kingdom) in 2013 to develop the prototype. Then in 2014, LRE (Germany, a subsidiary of Esterline) manufactured and delivered the instruments. Meanwhile, Atlas Genetics also supported academic research for multiplexing strategies.^{310, 311}

In 2018, clinical researchers from Johns Hopkins University and St. George’s University of London independently evaluated the performances of the new Atlas io® CT assay. The study conducted at Johns Hopkins University found high sensitivity and specificity for CT detection, high patient acceptability, and that 70% of women preferred to self-collect vaginal swabs if a POC test was available.⁶¹ The CT assay was also evaluated using 284 clinical samples collected at a teen Health Center and a sexually transmitted disease clinic, and participants were surveyed about their attitudes towards to POC testing.⁶¹ This study reported an adjudicated sensitivity and specificity of 92.9% and 98.8%, respectively. In terms of their attitudes towards POC testing, most (61%) were willing to wait up to 20 minutes, and 26% were willing to wait up to 40 minutes for results, if they could be treated before leaving clinic.⁶¹ The study conducted at St. George’s University of London found that the io® CT assay’s sensitivity was similar to that of laboratory-based NAATs, and its specificity was at the lower end of the laboratory-based NAAT range.⁶² Around the same time of these evaluations studies, the team at St. George’s University conducted a modeling-based evaluation of the costs, benefits and cost-effectiveness for platforms similar to the io® platform and concluded that such platforms could benefit symptomatic genitourinary medicine clinic attendees.³¹² Finally, during its rebranding phase from Atlas Genetics to binx health in 2018, the new io® CT/NG assay was evaluated through a non-interventional, cross-sectional clinical study that was conducted from September 18, 2018, through March 13, 2019. This clinical study enrolled 1,523 participants, 96% of patient’s vaginal swab samples were processed on the io® platform by non-laboratorians in a POC setting.⁶³ Clinical study performance for women demonstrated a 96.1% sensitivity and 99.1% specificity for chlamydia and 100% sensitivity and 99.9% specificity for gonorrhea. For male urine, sensitivity and specificity for chlamydia were 92.5% and 99.3%, respectively. For gonorrhea, the sensitivity and specificity were 97.3% and 100%, respectively.

More than a decade after the initial development of their CT assay, the io® CT/NG assay was CE marked in Europe in April 2019. It received FDA clearance in August 2019 and CLIA Waiver in March 2021. The evolution of this POC platform sheds light on the importance of collaboration between industry, clinical and academic researchers. Furthermore, it highlights the need for continued evaluation of end-users’ needs for POCT during the development process.

6.2. Evolution of Droplet Magnetofluidic (DM) Platforms in an Academic Setting

Our team began exploring DM as a potential technology for developing POCTs in 2011. In DM, magnetic beads are used as a mobile substrate for both nucleic acid capture and transport from discrete drops of reagents for nucleic acid extraction, purification, and amplification. This mechanism reduces fluid handling to particle translocation and streamlines sample preparation and NAAT by simply moving the nucleic acid-carrying magnetic beads with external magnetic forces, allowing detection of a variety of infectious agents from different sample matrices. Recent reviews from our team¹²⁹ and others^{130, 131} offer introductions and perspectives on DM technology.

Our initial efforts focused on demonstrating the capabilities of DM in performing integrated bioassays. The initial platform in 2011 employed silica superparamagnetic particles (SSP) for both solid phase extraction of DNA and magnetic actuation³¹³ (Figure 8). All bioassay buffers and reaction mixture – including cell lysis, DNA binding, wash, elution, and PCR – were pre-stored as discrete droplets within the microfluidic chip. In the microfluidic chip, the droplets were also separated by pairs of micro-elevations that resulted in slits. The unique surface topography facilitated efficient splitting of SSP from each droplet through each slit, thereby enabling magnetofluidic manipulation and achieving a sample-to-answer workflow without complex fluidic control. In addition, a compact sample handling stage including a magnet manipulator and a Peltier thermal cycler was built for performing bioassays in the microfluidic chip. The feasibility of the platform was demonstrated by detecting an ovarian cancer biomarker Rsf-1 via qPCR and detecting E. coli via real-time HDA.³¹³ Our team subsequently demonstrated in 2013 that DM can facilitate a comprehensive range of droplet manipulations toward performing sample processing steps, including droplet dispensing, transport, fusion, and particle extraction,³¹⁴ and that DM can be achieved through electromagnetic forces³¹⁵ (Figure 8).

Figure 8. Evolution of droplet magnetofluidic platforms toward NAAT-based STI POCTs.

Between 2011 and 2014, our efforts focused on demonstrating the capabilities of DM in performing integrated bioassays (Y. Zhang, et al.^{313, 314}) and exploring alternative actuation mechanisms such as electromagnetic forces (C.-H. Chiou, et al.³¹⁵). Since 2017, we have employed DM as the backbone technology toward developing STI POCTs. We have developed USB drive-sized plastic cartridges to replace conventional PDMS-glass chips, realized magnetic actuation via servo motors to manipulate magnetic particles, and engineered palm-sized instruments interfaced to mobile phones, tablets, or computers to automate the cartridge assay (D. J. Shin, et al.¹³² and D. J. Shin and A. Y. Trick, et al.²¹⁷). The most recent platform has been deployed in primary care clinics in Uganda to detect NG and determine its AMR (A. Y. Trick, et al.¹⁵⁴).

Having investigated the potential utility of DM, we began using it as the backbone technology toward developing STI POCTs. The first fully integrated platform, coined mobile nucleic acid amplification testing (mobiNAAT), was developed in 2017 to test for CT¹³² (Figure 8). In contrast to earlier DM platforms, our team employed a LAMP assay for CT detection, developed a USB drive-sized plastic cartridge to replace conventional PDMS-glass chips, implemented a magnet arm based on a servo motor to manipulate magnetic particles, and engineered a palm-sized instrument with a mobile phone user interface and Bluetooth connection to automate the cartridge assay. The costs of the plastic cartridge (including all reagents) and the instrument were < US $2 (comparable to lateral-flow tests) and < US $300, respectively. Toward realizing our long-term goal of developing a POCT, we deployed mobiNAAT in a hospital emergency department (ED) to test 30 potentially CT-infected patients and mobiNAAT achieved 100% concordance with laboratory-based NAAT.¹³² Moreover, when a non-laboratorian study coordinator performed the tests on the mobiNAAT platform, after receiving a brief training via an embedded mobile app module, concordant results were obtained with the traditional NAAT results and those obtained by the platform developer, demonstrating the ease of use and reproducibility of mobiNAAT.¹³²

We subsequently sought to determine end-user acceptability and to make technological improvements. Specifically, in 2018, we recruited 30 ED technicians to evaluate mobiNAAT and complete a survey of acceptability factors.³¹⁶ We found that 83.3% of the participants preferred POCTs to traditional laboratory tests, where 76.7% indicated that their preference was due to the fast turnaround time. The survey revealed that the primary value of a STI POCT to end-users in an acute care setting is the potential to save time, expedite patient care decisions, and reduce length of stay, and that testing cost appeared to be a secondary consideration. This survey illustrates our efforts in bridging the gap between developers and end-users. Meanwhile, we have begun improving our DM technology by introducing a new cartridge with extruded wells for immobilizing the reagent droplets within the cartridge, thereby improving the robustness of the cartridge (Figure 8). Manipulation of magnetic particles was achieved via a new biaxial magnetic arm. We demonstrated the utility of this platform with hepatitis C virus (HCV) RNA detection from clinical serum samples, which yielded results in ~1 h with no false positives.²¹⁷

Leveraging these experiences, we next developed PROMPT (Portable, Rapid, On-cartridge, Magnetofluidic, Purification, and Testing) that can detect NG and genotype one of its AMR genetic determinant in 15 min directly from clinical samples¹⁵⁴ (Figure 8). Such rapid turnaround time was achieved via ultrafast PCR in the PROMPT instrument, which was enabled by a unique heat block with optimally minimized thermal mass surrounding the extruded PCR well of the plastic cartridge. Built-in two-color fluorescence detection permitted multiplexed detection of the NG opa gene for pathogen identification, and the wildtype gyrA gene, which is highly predictive of ciprofloxacin susceptibility. We evaluated the clinical utility of the multiplex assay (NG and ciprofloxacin susceptibility testing) in two studies in Baltimore, Maryland, USA and Kampala, Uganda. Following the collecting of self-collected penile swabs, an aliquot of the swab eluate is loaded onto to the sample cartridge, followed by loading of the cartridge onto the PROMPT device for testing (Fig. 8). Using this approach, penile swab samples from sexual health clinics in Baltimore, MD (n = 66) and Kampala, Uganda (n = 151) were tested on the PROMPT device with an overall sensitivity and specificity of 97.7% (95% CI: 94.7%–100%) and 97.6% (95% CI: 94.1%–100%) respectively for NG detection and 100% concordance with culture results for ciprofloxacin susceptibility.¹⁵⁴

7. PERSPECTIVES FOR FUTURE DEVELOPMENT

We share in this section our perspectives on important topics regarding future development of NAAT-based STI POCT technologies. These topics include guidance for technology development, testing cost, emerging testing trends, and barriers beyond technology development. We hope our perspectives can offer a holistic view that can accelerate the development of NAAT-based STI POCT technologies and their deployment and uptake in primary care settings, especially in LMIC.

7.1. Development Guidelines and End-user Communication

During the development of new STI POCT technologies, developers should reference RE/ASSURED criteria and TPPs, but more importantly, should consult with end-users whenever possible. There have seen discussions about the necessity of the ASSURED criteria and the potential compromises that may have to be made.^{13, 80, 317, 318} Garcia and colleagues organized a study in which developers of diagnostic tests from affluent countries visited and collaborated with end-users in Peru.⁸⁰ Such open dialogue can produce a clear list of must-have test qualities for each setting, while attributes of less importance can be given lower priority. Additionally, faced with few resources, local health-care professionals have the ability to come up with simple, yet innovative solutions to challenges they encounter, and might be able to provide fresh insights into ways to solve challenges faced by developers.⁸⁰ In lieu of in-person meetings, teleconferences can be held or end-user surveys can be conducted.^{81, 154, 316, 319–322} Nayak and colleagues noted that social sciences present frameworks for analyzing user behavior, but that they have yet to be methodically applied to the development of POCTs.³²³ Using microfluidic POC HIV home testing as an example, they pointed out that engineering design of such a POCT can benefit from a structured evaluation of user behavior and experience, as guided by a social-psychological framework, which emphasizes user credibility, accessibility, acceptability, usability, and value.

7.2. Cost and Cost Analysis

A recurring theme of this review is that the cost of instruments and assays will be a critical factor in implementation and uptake of new STI POCTs. Expert consensus is that the ideal price point per test in LMIC would be about US $1, where higher costs could limit adoption of such tests. The reasonable price point per test in the United States would be about US $20.⁸ Notably, according to FIND (see https://www.finddx.org/pricing/genexpert/), the list prices for the Cepheid Xpert CT/NG and Xpert TV are US $16.20 and US $19.00 per test cartridge for eligible countries. While the cost of these tests may be acceptable in the United States, they are still too expensive for LMIC. Therefore, the cost should be actively incorporated into the design and the development process for new STI POCT technologies, while the use of inexpensive reagents, materials, and components must be embraced.

On the other hand, cost analysis of new STI POCT technologies prevent potentially unreasonable cost reduction, which may not only pose technical challenges, but also disincentivize commercialization. The cost savings obtained from the rapid delivery of results, reduction of patient follow-up, facility cost, decreased complications and onward transmission should be incorporated in the cost analysis.³²⁴ For example, based on a modelling study by Vickerman and colleagues, even at $4 per test, a moderately sensitive (> 60%) POCT for both CT and NG can still be a cost effective strategy for increasing the impact of STI treatment interventions in LMIC.³²⁵ The cost-effectiveness of multiplex POCTs (CT, NG, MG, TV) has been demonstrated in a separate modeling study.³¹² A cost-effectiveness analysis has shown that using a NG NAAT for screening women between 15 and 29 years of age can prevent 1247 cases of PID and save US $177 per patient compared with no screening, while using a potential POCT with about 75% sensitivity can prevent additional PID cases.³²⁶ Finally, a POCT with AMR detection would accrue an additional cost, but its use could reduce the cost from follow-up visits and could allow for the use of older and cheaper drugs, such as ciprofloxacin, and more importantly, conserve the current last-resort options of ceftriaxone.³²⁷

7.3. New Testing Trends and New Potential Issues

Self-testing can increase accessibility of testing, reduce the burden of STIs,⁸ and facilitate mHealth,³⁰¹ thus represent an emerging testing trends. Self-testing can be either facilitated by healthcare providers at settings like primary care clinics or pharmacies or unsupervised self-testing, where users understand the pre-test information, perform the self-test, and interpret results themselves, at private settings like home and office.³²⁸ If affordable and reliable, such testing options provide confidentiality that alleviates stigma, discrimination, and social visibility, convenience, and links to care with a shorter turnaround time than conventional testing. To determine the acceptability of such a test, Pearson and colleagues conducted surveys and found that people are willing to pay for and use STI self-tests for chlamydia and gonorrhea when they become available.³²⁹ Pai and colleagues pointed out the potential of mHealth and machine learning toward improving HIV detection, linkage to and retention in care, surveillance, prediction of hot spots, and overall quality of HIV management.⁴⁴ Such potential may be realized for the major STIs (permitted that data are accurately integrated within large-scale databases) and be expanded to potentially facilitate the capability of public health authorities to monitor outbreaks, implement response strategies and assess the impact of interventions across the world. Cao and colleagues conducted a global scoping review on digital health for STIs and HIV services; they noted that digital health, while still emerging, is a powerful and versatile tool that can be utilized toward high-quality, innovative strategies on STIs and HIV services.³³⁰

Although there are at-home tests for HIV, no NAAT-based STI self-tests have been approved by the FDA. There are technological challenges and other potential issues. Shrivastava and colleagues presented a comprehensive review on recent progress, challenges, and prospects of fully integrated mobile and wearable POCT technologies for self-testing.³³¹ Some of the challenges listed in this review, including sensitivity and robustness, overlap with the development of new NAAT-based STI POCT technologies. Peterman and colleagues are similarly concerned that self-tests for gonorrhea and chlamydia may have potential problems such as poor test accuracy and low predictive value.³³² For this testing format, they note that approaches that use small stand-alone testing units may still be impractical for self-tests; instead, more promising are test cartridges that can be attached to smartphones for both powering the test cartridges and detecting amplified products. Furthermore, a smartphone app might also directly report the positive test to the state or local health department and assist in linking patients to a provider for treatment though such a link might discourage testing among persons who want to remain anonymous.³³² Devices for home-use, sold over the counter, would ideally be non-instrumented (other than smartphone-based detection), and as easy to use as lateral flow strips currently sold in drug stores. In this setting, paper devices with a low-temperature isothermal NAAT method (e.g., RPA) may have an edge. Simple, manually-actuated microfluidic chips made in hard plastic with integrated pouches for fluid storage and actuation, and possibly incorporating chemical heating (as some described in the review article by Mauk and colleagues¹¹⁰) may offer another potential solution. Finally, as Wilson and colleagues pointed out, testing innovations must be accompanied by service innovations (e.g., remote consultation and postal treatment services) to improve treatment rates for those diagnosed by self-testing.³³³

7.4. Nontechnology Barriers and Need for STI Programs

The mere availability of rapid or simple POCTs does not automatically ensure their adoption or scale-up. Beyond the technology, there are a range of barriers at the patient, provider, and at the health system level, which could prevent the successful use of POCTs.^{8, 317, 334} At the patient level, barriers include costs for confirmatory tests, difficulties in performing the POCTs, distrust, lack of awareness, privacy issues, and/or a preference for conventional approaches. At the provider level, barriers include costs, a lack of awareness, complicated protocols, challenges in clinical workflow, or simply a resistance to adopting POC diagnostics. At the health system level, barriers include costs, integration challenges, quality assurance and quality control requirements, workflow challenges, required training, and technical support. Among the workflow challenges that could arise is the reporting of notifiable STIs.

The responsibilities for tackling these barriers rest on all stakeholders. For technology developers, there is an increasing need to consider beyond technology development and connect to as many stakeholders as possible in order for policy makers and companies to take the leap and bring POCTs to the patients.³¹⁸ To do so, understanding of the nontechnology barriers is crucial, as development of new technologies have to be designed in synergy with its target environment and implementation, instead of merely adapting to it later on. Nayak and colleagues discussed nontechnology components that developers must consider: clinical workflow, regulatory guidance, reimbursement, and legislation. They presented four different use cases for which the device metrics are tailored, as well as case studies on devices that had progressed from basic technologies to use-case demonstrations.³³⁵ Finally, they provided a list of companies that can manufacture microfluidic devices. Peeling and Boeras discussed the lessons they have learned from implementing POCTs. Although they mostly refer to antigen-based tests, their summaries on implementation and policy barriers present a useful primer for understanding these nontechnology challenges.³³⁶ Korte and colleagues urged better understanding for “value proposition” of new POCT technologies.³³⁷ Expanding the focus of research to address gaps in both the creation and evaluation of value propositions is imperative in order for value concepts to positively influence the adoption of POC testing. Malekjahani and colleagues shared their experiences and perspectives on the four critical steps for engineering POCTs: (1) Needs and Value Assessment, (2) Technology Development, (3) Preclinical Verification, and (4) Clinical Validation and Field Trials. During the first step, they recommend researchers focus on assessing specific diagnostic needs and the value, which a potential device would offer, rather than maximizing technical parameters such as sensitivity and time, which do not necessarily translate to increased clinical impact. During the second step, they recommend researchers develop equipment-free devices, which are agnostic to any mobile phone, rather than developing devices that may appear feasible at the POC, yet are dependent on laboratory resources. During the third step, they recommend that researchers simulate the field conditions during laboratory verification to achieve successful translation, as performances during field evaluations often decrease. In the last and fourth step, they recommend researchers perform proper field testing of devices in conditions that match with the final intended use.³³⁸

Review articles on commercialization of microfluidic POC diagnostic devices^{264, 339–341} can also be helpful, as there are significant overlaps between deployment within healthcare systems and commercialization. For example, Reyes and colleagues pointed out the importance of standardization of microfluidic-based medical devices and described the existing gaps in the standardization of flow control, interconnections, component integration, manufacturing, assembly, packaging, reliability, performance of microfluidic elements and safety testing of microfluidic devices throughout the entire product life cycle.³⁴¹ Gong and Sinton point out that a major barrier to commercialization is complying with regulatory policies. They encourage researchers to consult the relevant medical device regulations early and often during technology development to expedite commercialization, that is, to design for regulatory approval in addition to function and low cost.²⁶⁴

In the meantime, other stakeholders should work to facilitate the development, evaluation, implementation, and uptake of new STI POCT technologies. To this end, team science that integrates complementary expertise (e.g., cost analysis, technology development, evaluation) is crucial. In particular, implementation science needs to be at the forefront of developing, validating, and deploying state-of-the-art efforts to address STIs in real-world settings.³⁴² It is also helpful to include expertise on modeling and cost analysis to make an investment case to ensure buy-in among key stakeholders. Toskins and colleagues recommended setting up a network of evaluation sites for accelerating the evaluation of new STI POCT technologies.⁷⁹ Doing so can produce better STI data and facilitate the development of models to estimate cost-effectiveness, which may harmonize regulatory approval processes and accelerate policy development. Moreover, POCT implementation should consider integration within the STI management pathways, including patient flow, immediate treatment, partner management and retesting, and the existing healthcare systems.¹³ Toskins and colleagues suggested that demonstration and implementation research projects should be organized through a specific agenda to help roll-out new STI POCTs and integrate them into health care systems.⁵³

Furthermore, other stakeholders should also work to improve STI programs, which can accelerate the development of new NAAT-based STI POCT technologies through several fronts.⁷⁹ First, they can establish a pathway for technology developers to acquire additional funding and connect with evaluation sites. An example of existing efforts is the STI POC Center at Johns Hopkins University, which offers funding for technology development and opportunities for evaluating technologies. The COVID-19 XPrize³⁴³ also provides a model that can be emulated. Second, they can develop guidelines, norms, and standards. Presently, the case can be made that guidance tools and documents issued by regulatory agencies have not kept pace with the development of new NAAT-based STI POCT technologies from academia. Third, they can further support the introduction and implementation of new STI POCT technologies. For example, they can assist in the rollout and supply of the new technologies, as well as in organizing training at the target environment. Moreover, because different stakeholders can communicate through STI programs, tensions that may arise at the implementation level can be eased. Finally, they can promote multi-sectorial and even international research cooperation. For example, such programs can help create biobanks for collecting well-characterized specimens that can be accessed by different stakeholders. They can also provide a conduit to global health sector STI strategies and initiatives.

8. SUMMARY AND OUTLOOK

The incidence rates of the four major curable STIs – chlamydia, gonorrhea, trichomoniasis, and syphilis – continue to increase globally, resulting in epidemics with substantial medical cost burden and morbidity, especially in LMIC. The emergence of AMR in NG represents another worrisome trend. Commercial near-POC instruments that can perform NAATs for detecting STI pathogens (e.g., Cepheid GeneXpert) represent key advances, but can still be too complex and expensive as POCTs even in developed countries, while also lacking AMR testing. As such, there is still an unmet need for low-cost, simple, accurate, rapid, and self-contained STI POCTs that can support uptake and widescale use in primary care and community settings, especially in LMIC.⁸ There has been progress in academic research and industry development, and efforts directed to tackling HIV²⁹⁸ can be leveraged as foundation for tackling the major STIs. In academic research, isothermal NAATs and paper devices have received significant research attention, while droplet magnetofluidic technology coupled to miniaturized rapid PCR represent an emerging promising strategy for use in LMIC. Effective incorporation of AMR detection in NG into POCT technologies is beginning to receive attention. In industry, binx io® and Visby Medical Sexual Health Test have become promising POCTs, though their costs may still be high for LMIC. Thus, even with notable progress, significant opportunities for research and development remain for years to come, and viable paths have been charted through the development of potential POCT technologies for HIV and other infectious diseases – some of which are highlighted in this review. Finally, limited demand and challenges in the regulatory approval of such devices exist. Funders and developers will need vision and steady motivation to navigate through these nontechnology barriers and bring new POCT technologies to patient care.

As described in this review, the development of a new POCT is a multi-step process, which may involve platform and assay development, clinical validation, and lastly regulatory approval; a process that traditionally in total might take up to 10 years. However, the research community, including public health professionals, physicians, and technology developers, have independently and collectively amassed analytical tools and experiences for the development of new NAAT-based STI POCT technologies as highlighted in this review. These analytical tools and experiences can provide helpful guidance and accelerate the development of new technologies. Moreover, this increasingly connected world has created infrastructure and opportunities for collaborative, team-based research that is critical to the development of new NAAT-based STI POCT technologies. The rapid development of near-POC instruments and POCTs during the COVID-19 pandemic provides an excellent example of how needs-driven innovation can facilitate technological advances. STIs are an on-going epidemic and a global public health threat and as such should be prioritized. The analytical platforms and assays developed for detection of SARS-CoV-2, could and should be applied towards the detection of STIs to help mitigate and decrease the burden of STIs on a global scale.