gms | German Medical Science

The aim of the study presented in this contribution is to give an overview of the presently applied methodologies for nucleic acid tests (NATs) for infectious diseases in clinical and analytical laboratories and to discuss emerging technologies for NATs potentially suited to support rapid and point-of-care (POC) diagnostics. Near patient instruments will directly influence clinical outcomes for patients and allow to manage patients in one consultation rather than require further follow-ups. This can allow treatment of the patient which should be appropriate and timely. The most relevant POC tests for monitoring infectious diseases involve rapid polymerase chain reaction (PCR) based approaches and novel isothermal amplification methods. The development of these nucleic acid based methods has improved the analytical performance, i.e. sensitivity and specificity, of methods in the area of infectious diseases. These methods have the potential to complement microbiological methods for pathogen detection, which are time consuming and may lack sensitivity.

Important for POC applications are low costs and an acceptable turnaround time [1], [2], [3]. The turnaround time is correlated to the term “rapid diagnostics”, which has been subject to dramatic changes with the availability of PCR and isothermal nucleic amplification technologies. Whereas MRSA detection based on chromogenic agar technology typically requires 2 days to 3 days, application of PCR allows the reduction of the time to obtain the result in a few hours [1], [2]. The transport time from sample collection to the laboratory contributes significantly to turn around times in PCR based examination [2], which could be avoided using POC tests. Using isothermal nucleic amplification techniques, thermocycling necessary for PCR is avoided and the time to result can be reduced even further to less than 60 min or even to 20 min [3]. Isothermal amplification technology has the potential for NATs integrated in disposable microchips [4], [5], [6], [7] and instrumentation for isothermal amplification and detection of the reaction is simpler compared to PCR. Hence, besides rapid diagnostics in industrial countries, application of POC nucleic acid detection based on isothermal nucleic amplification is of high interest in developing countries, where easy handling shall be taken into account.

The literature study revealed that a variety of methods for the detection of nucleic acids are developed, which are not based on thermo-cycling PCR (Table 1 [Tab. 1]). Commercially available are instruments and related kits based on branched chain DNA signal amplification (bDNA) [8], loop-mediated isothermal amplification (LAMP) [9], [10] and recombinase polymerase amplification (RPA) [7].

The bDNA method was first demonstrated to provide reliable results in 1995 [11] by quantitative measurements of HIV-1. Later, the performance of bDNA was evaluated by comparing different NAT methods using the VERSANT HCV RNA Assay and the Siemens System 340 bDNA analyser (Siemens Healthcare Diagnostics) [12], [13].

The LAMP method is used in the real time turbidimeter developed by Eiken Chemical Co., Ltd. (Japan) and applied e.g. for the diagnosis of the avian influenza viruses (H5N1) [14]. Lumora Ltd. (United Kingdom) also implemented the LAMP method in a portable small instrument (PDQ) and a high throughput development system. In contrast to the turbidimetric approach, the amplification of the nucleic acid is monitored by fluorescence measurements, called bioluminescent assay in real-time (BART) [15]. The instruments provided by OptiGene also rely on the LAMP methodology and isothermal master mix kits (licensed for LAMP by Eiken Chemical Company) are provided. The recombinase polymerase amplification (RPA) method [7] is used by TwistDx, Ltd (United Kingdom) in their Twista^® portable real-time fluorometer and reagent kits are provided for food safety.

The development of improved variants for LAMP and RPA is still in progress. Instead of using displacement primers “Stem primers” are used for LAMP [16] and a chemical heating approach was applied to detect HIV-1 [17]. An advantage of RPA is that no initial heating (e.g. to 95°C to obtain single stranded DNA) is required and only 37°C are needed compared to 65°C for LAMP; this was exploited in a microfluidic lab-on-a-foil device [18]. Furthermore, an active field of research of particular interest for rapid and POC diagnostics is the miniaturisation of nucleic acid amplification systems [19] including the adaption of NATs in microfluidic chips [20].

Two questionnaires (see Attachment 1 [Attach. 1] and Attachment 2 [Attach. 2]) were created comprising manufacturer specific or end-user specific inquiries. In addition, questions for both, companies and analytical laboratories were included in each of the questionnaires. The end-user specific survey was distributed by INSTAND to 1,000 analytical laboratories in Germany regularly participating in ring trials for external quality assessment addressing detection of infectious pathogens by NATs. In addition, INSTAND contacted 336 laboratories from other European and Non-European countries. These laboratories perform infectious disease diagnostics in the fields of virology, bacteriology, mycobacteriology, parasitology and mycology. Many of these laboratories offer diagnostic services in several of the above fields.

We received replies from 48 laboratories in 14 European countries. All except 2 laboratories offer analytical test in infectious disease medicine in several fields. Three quarters of the laboratories having replied perform more than 1,000 NAT based analyses per month. The numbers of laboratories from the respective countries that participated in the survey are shown in Figure 1 [Fig. 1]. Most participants came from Germany and Scandinavian countries, but we received also answers from Baltic and Eastern European counties. In total, 36 manufacturers were contacted, of which 12 responded. Only 8 forms returned could be included in the analysis, since 4 manufacturers focus on the development of new instrumentation or kits and are not directly in contact with analytical laboratories. The results of the survey, summarised in Tables 2, 3, 4 and 5 are presented in the following paragraphs. In these tables, the most frequent answers are marked in grey.

The answers for manufacturer specific questions are summarised in Table 2 [Tab. 2]. The majority of manufacturers offer both, instrumentation and specific kits (Table 2 [Tab. 2], row 1). About a quarter of the answers are from companies producing instrumentation only and few firms (12%) focus on the development of kits for nucleic acid amplification. With respect to the number of kits supplied (Table 2 [Tab. 2], row 2) for the detection of human cytomegalovirus (CMV), human immunodeficiency virus (HIV), hepatitis B and C viruses (HBV/HCV) most of the manufacturers (89%) provide more the 1,000 tests per month and the remaining companies (11%) supply 100 to 1,000 tests per month.

Half of the fluid handling systems are based on conventional techniques, whereas 38% of the companies claim that they apply microsystem technology (Table 2 [Tab. 2], row 3). Interpretation of these answers is difficult since the term “microsystem technology”, i.e. application of ultra precision milling or lithography for production of embossing tools of injection molds yielding surface roughness in the 10 nm region, was not defined in the questionnaire. More significant is the volume of the cuvette or chamber used for the amplification since less material is required and shorter amplification times are expected for smaller volumes, in particular if PCR is used. For cuvette volumes between 10 µL to 20 µL and 20 µL to 50 µL the same numbers (38%) are given and in few cases (13%) volumes above 50 µL are required (Table 2 [Tab. 2], row 4). However, in the context of nucleic acid amplification, it follows from the answers concerning the amplification time that no substantial reduction is reached, all responses state times above 25 min (Table 2 [Tab. 2], row 5). Hence, the contribution of amplification times to the turnaround time cannot be neglected and its reduction is highly desirable.

The market shares of kits and instruments according to the answers of end-users are depicted in Figure 2 [Fig. 2]. According to Figure 2a [Fig. 2] the amplification is based on in house tests (blue colour, 14 laboratories) and to an equal amount kits provided by Roche (red, 14 nominations). Other suppliers were quoted less frequently. For each vendor, the same colour was chosen in Figure 2a and 2b [Fig. 2] to visualise the correlation between kits applied for nucleic amplification tests and instrumentation. In Figure 2b [Fig. 2], equipment for nucleic acid amplification as well as apparatuses for DNA/RNA extraction and purification is accounted for. It follows from Figure 2b [Fig. 2] that – besides supply of kits – Roche (red, 24 laboratories) is also the market leader for equipment. The use of “no standard equipment” is accordingly interpreted as “in house tests” and indicated in blue. Other companies providing kits and instrumentation are Cepheid (8/11), Abbott (5/8), Qiagen (4/10), Applied Biosystems (1/4) and Becton Dickinson (1/4). The other companies referred to as either manufacturer of kits (Altona Diagnostics, Amplex Biosystems, Fisher Scientific, GeneProof) or manufacturer of instruments (BioRad, bioMérieux, Hologic, SensoQuest) are marked in different grey values.

In Table 3 [Tab. 3] and Table 4 [Tab. 4] the answers of the end-users (Table 3 [Tab. 3]) and the manufacturers’ responses (Table 4 [Tab. 4]) with respect to technologies applied for nucleic acid amplification, pathogens, sample type, workflow, turnaround time, sensitivity and price are listed.

PCR is the leading technology applied by almost all end-users (98%) and offered by most companies (75%). Isothermal amplification of nucleic acid is used in 8% of the analytical laboratories and 25% of the companies produce corresponding instruments and/or kits. None of the participants, neither end-users nor manufacturers indicated the application of branched DNA methodology. Microarray hybridisation was also listed (Table 4 [Tab. 4], row 1), but it is less relevant in the context of analysis of pathogens due to low potential for quantification.

In Figure 3 [Fig. 3], we summarise responses of analytical laboratories (blue bars) with respect to the target bacteria and viruses, which were explicitly listed in the questionnaire as multiple choice. The red bars in Figure 3 [Fig. 3] indicate the answers given by the manufacturers concerning the supply of corresponding kits. The end-users (Table 3 [Tab. 3], row 2) reported that in more than 50% of the laboratories methicillin-resistant Staphylococcus aureus (MRSA), influenza A and B viruses and hepatitis B and C viruses (HBV/HCV) are diagnosed. M. tuberculosis (27%), human cytomegalovirus (CMV) (40%), adenovirus (35%), human immunodeficiency virus (HIV) (48%) and Epstein-Barr virus (EBV) (40%) are also frequently analysed. On the other hand, only few laboratories (6%) provide molecular diagnostics for SARS-associated coronavirus (SARS-CoV) causative for severe acute respiratory syndrome.

For all these NATs commercial kits are available and provided by some of the manufacturers. In particular, according to Table 4 [Tab. 4], row 2, kits for M. tuberculosis are offered by 25% of the responding companies, 38% supply kits for CMV, 50% for influenza A and B viruses and 38% for EBV. A smaller fraction of the companies provide kits for MRSA (38%), SARS-CoV (13%), adenovirus (13%), HIV (25%) and HBV/HCV (25%).

M. tuberculosis, tested in 27% of the analytical laboratories (Table 3 [Tab. 3], row 2), is reported to reveal a much higher notification rate in WHO Member States in the east than in the west [21]. This tuberculosis report 2014 of the WHO indicates a worldwide increase of 3.5% for the multidrug-resistant tuberculosis (MDR-TB) in the year 2013. In some countries, however, i.e. the Russian Federation, Uzbekistan, Republic of Moldova, Kazakhstan, Kyrgyzstan and Belarus, the incidence of new cases ranged between 19%–35% in the years 2011–2013. Particularly distressing is the high percentage of extensively drug-resistant tuberculosis (XDR-TB) and the continuing decline of the success rate of medical treatment of tuberculosis.

The relative use of diagnostic tests for the detection of pathogens by the replying laboratories as well as the supply of commercial diagnostic tests by the responding manufacturers seem to reflect the diagnostics needs of the corresponding methods for public health in the countries from where replies were received (part of northern, eastern and mid Europe). This correlation may be true for MRSA, as well as HBV/HCV and influenza A and B viruses. Interestingly, replying laboratories and manufactures hang on to diagnostic tests for the detection of SARS-CoV, although there have not been any known cases of SARS-CoV infection reported anywhere in the world since 2004 (http://www.cdc.gov/coronavirus/about/index.html).

Apart from the pathogens explicitly listed in the questionnaire, the participants were also asked to note other important target bacteria and viruses. The most frequent answers are summarised in row 3 of Table 3 [Tab. 3]. Besides Chlamydia trachomatis, detected in 44% of the laboratories, norovirus (27%), HSV-1/HSV-2 (27%), varicella zoster virus (23%), Neisseria gonorrhoeae (23%) and Mycoplasma genitalium (17%) are analysed.

The distribution of samples for NATs is depicted in Figure 4 [Fig. 4]. It follows from the response of the end-users (Figure 4 [Fig. 4], blue bars; Table 3 [Tab. 3], row 4) that most often swab (83%) and blood/blood plasma (60%) are used. Besides these samples, urine (44%), tissue from biopsies (38%), faeces (33%), bronchoalveolar lavage (BAL) (25%), cerebrospinal fluid (CSF) (21%) and sputum (13%) serve for nucleic acid tests. The percentage of manufactures providing kits for these materials (Table 4 [Tab. 4], row 3) is shown as red bars in Figure 4 [Fig. 4]. Most nominations concern faeces (50%), swab (46%) and blood/blood plasma (36%), followed by bioptic tissue (22%). These samples are among the five materials most frequently reported by the analytical laboratories.

In the questionnaire, we offered three options for the workflow of nucleic acid tests. The manual preparation requires DNA/RNA extraction, purification and pipetting of reagents, complete tests include automated preparation and amplification in one or different instruments and the third selection is for complete, cartridge or cassette based tests which contain all reagents needed and only the sample has to be fed in. As presented in Figure 5 [Fig. 5] and Table 3 [Tab. 3], row 5, in analytical laboratories, manual preparation is done in 33% of cases. At present, tests which include automated preparation (63%) of samples are being preferred. Fully automated tests, based on cartridge or cassette systems are applied in 21% of the laboratories. The sum exceeds 100% since in some laboratories various modalities are being used. The most common indication of manufacturers (Table 4 [Tab. 4], row 4) concerns the manual preparation (63%), automated preparation is reported with 25% and cartridge based workflows in 13% of cases. Compared to the end-users, the order is reversed with respect to manual and automated preparation, possibly because most of the laboratories have a high throughput which requires the highest possible degree of automation. On the other hand, both groups are ranking complete, cartridge based tests on the third rank, the relative market share is estimated to be in the range of 10% to 20%.

The turnaround time for pathogen detection using NATs is generally above 2 h (77%) according to the analytical laboratories (Table 3 [Tab. 3], row 6), 23% and 2% claimed turnaround times between 1 h to 2 h or even below 60 min, respectively. The analytical sensitivity or limit of detection in pure samples or calibrators requested by most of the end-users for reliable pathogen detection is about the same for <10 copies mL^–1 (29%), 10 copies mL^–1 to 50 copies mL^–1 (26%) and 50 copies mL^–1 to 100 copies mL^–1 (31%) (Table 3 [Tab. 3], row 7). The broad range given by the end-users are probably due to the fact that pathogen and sample specific sensitivity and limit of detections are required for reliable diagnostics. The manufacturers’ responses to these questions are in close concordance with the end-users requirements and show a trend to lower turnaround times (38% quote the range 1 h to 2 h) (Table 4 [Tab. 4], row 5) and to higher sensitivity (50% state a sensitivity of 10 copies mL^–1 to 50 copies mL^–1) (Table 4 [Tab. 4], row 6). However, the costs per tests requested by the analytical laboratories are lower compared to the manufacturers’ announcements. The range up to 4 € is mentioned by 66% of the end-users (Table 3 [Tab. 3], row 8) and 42% of the companies (Table 4 [Tab. 4], row 7). The upper limit stated by the laboratories is 6 €, but the manufacturers indicate that the costs for several kits (57%) exceed 8 €. Again, the broad spread in acceptable costs and cost-covering prices indicate that for different pathogens and samples the complexity of test kits and the corresponding developments are crucial.

An overview on the response to the questions specific for analytical laboratories is given in Table 5 [Tab. 5]. As already mentioned when discussing the high percentage of laboratories using automated preparation and cartridge based tests (see Table 3 [Tab. 3], row 5: sum 84%), most of the participating analytical laboratories (75%) investigate more than 1,000 samples with respect to infectious pathogens (Table 5 [Tab. 5], row 1). For high throughput applications automation is necessary to provide analyses in short times at reasonable costs. About 21% of the users perform 100 tests per month to 1,000 tests per month. The number of analyses is less than 100 tests per month in 4% of the analytical laboratories. The relative share of NATs at all analyses covers the complete range from <25% to >75% with the most common indication for <25%, given by 54% of the laboratories (Table 5 [Tab. 5], row 2). The distribution reveals a second (relative) maximum, 23% of the laboratories perform in more than 75% of the analyses nucleic acid tests. This survey result demonstrates the increasing importance of nucleic acid tests, used in more than half of the laboratories (54%) in addition to other diagnostic methods and applied as leading procedure in specialised facilities (23%).

About 40% (Table 5 [Tab. 5], row 3) of the end-users report the application of “rapid” PCR/NATs when answering the corresponding question of the survey sheet. However, this statement is not supported by the answers concerning the turnaround times, since only 2% reach times below 60 min (Table 3 [Tab. 3], row 6). These 2% of the laboratories seem to better represent the current situation than the 38% given in Table 5 [Tab. 5], row 3 with respect to “rapid” diagnostic. We interpret these discrepancies as non-conform definitions and suggest the use of the term “rapid” PCR/NAT as indicative for turnaround times less than 30 min to better account for the present state of the art of NATs.

The bacteria and viruses for which the end-users declare an urgent need of rapid nucleic acid tests are shown in Figure 6 [Fig. 6] (Table 5 [Tab. 5], row 4). The percentage of laboratories requesting a rapid test is plotted on a logarithmic scale to account for the large variations. Apart from SARS, for all pathogens explicitly listed in the questionnaire there seems to be an interest for significantly shorter turnaround times, specified to be <15 min by the majority of users (Table 5 [Tab. 5], row 5). However, the number of laboratories requesting such rapid NATs ranges from only 2% for M. tuberculosis, adenovirus, HIV and EBV to 54% for MRSA. Besides MRSA, a large proportion of the answers indicate the demand for rapid tests with respect to norovirus (27%) and influenza A and B viruses (21%). About 10% identify the need for rapid CMV and HBV/HCV nucleic acid tests. In Figure 6 [Fig. 6], we include the expected number of rapid tests (Table 5 [Tab. 5], row 6) per analytical laboratory – based on the most frequent statements – as colour code, violet represents <10 tests per month, light-brown indicates 10 tests per month to 50 tests per month and green stands for >50 tests per month. Alternatively, instead of using the most frequent answers for each bacterium or virus, we summarized the expected numbers of tests stated by each laboratories for all pathogens which are analysed. From Figure 7 [Fig. 7] one can conclude that most of the end-users, who intend to apply rapid NATs expect more than 50 tests per month. Please note that the total number of nominations (60) exceeds the number of participating laboratories (48), since generally more pathogens are analysed in each analytical laboratory. Compared to the typical throughput of >1,000 analyses per month the relative contribution of rapid test would be in the order of 5%.

The question whether qualitative tests or quantitative tests are needed yielded different opinions. According to Table 5 [Tab. 5], row 7, most of the users (44%) prefer to choose between a qualitative or quantitative tests dependent on the pathogen to be detected, 31% would be satisfied with a qualitative result.

The answers concerning the purchase of equipment (Table 5 [Tab. 5], row 8) and the establishment of rapid nucleic acid analyses in analytical laboratories (50% do not intend to introduce rapid tests and 19% responded with “later than the next three years”) reflect that the end-users are currently waiting, possibly because the technologies for rapid NATs are still being developed and improved by corresponding research activities.

The results of this survey refer to the answers of 48 analytical laboratories, most located in Germany. About a quarter of the responses were from Scandinavian countries. With respect to the current state of the art of NATs reasonable consensus was observed except for the costs between end-users and manufacturers, albeit only 8 companies responded.

The pathogens most frequently analysed in more than half of the laboratories are MRSA, influenza A and B viruses and HBV/HCV, followed by HIV, EBV, CMV and adenovirus detected in about 40% of the laboratories. M. tuberculosis was mentioned as target by 27% of the end-users. Apart from these pathogens, which were explicitly listed in the questionnaire as multiple choice option, Chlamydia trachomatis, norovirus and Neisseria gonorrhoeae are also frequently (in 20% to 40% of the laboratories) analysed.

The majority of the participating analytical laboratories perform more than 1000 analyses per month to detect infectious diseases. Hence, a high degree of automation is required to achieve high sample throughput applying complete conventional or cartridge based tests. The relative share of in-house tests for nucleic acid extraction/purification and amplification is about 27% while 73% utilize kits provided by various manufacturers. In this context, internal and external quality assurance is highly relevant to ensure that results from different laboratories are in agreement within defined limits of equivalence, regulated for selected pathogens in the corresponding guidelines [22] of the German medical association.

The questionnaire revealed a non-consistent use of the term “rapid” analysis, due to the dramatic reduction of turnaround times from several days required by cultures to typically 2 h – 4 h after the introduction of PCR. It follows from the questionnaire that the current state of the art for the turnaround time of PCR based nucleic acid tests is still >2 h. The next step towards rapid point of care tests is expected by the implementation of complete tests involving isothermal methodology for NATs. Such a setting might allow turnaround times below 30 min and possibly between 10 min – 15 min. Hence we suggest the definition of a rapid NAT as analysis with turnaround times <30 min. For rapid NATs complete cartridge based tests with integrated microfluidic chips are the most promising approach. Of particular interest is the RPA method [7], [23], because the amplification can be carried out at about 37°C.

The majority of the laboratories (75%) points out the need for rapid nucleic tests for certain pathogens. Of particular interest is MRSA for more than 50% of the end-users, 27% and 21% demand for rapid NATs to detect Norovirus and influenza A/B, respectively (Table 5 [Tab. 5], row 4). According to about 10% of the laboratories, further pathogens for which rapid tests are needed are HBV, HCV and CMV. The desired turnaround time is <15 min (Table 5 [Tab. 5], row 5) for rapid nucleic acid tests of these disease-causing agents. Slightly more laboratories (44% compared to 31%, Table 5 [Tab. 5], row 7) indicated that – dependent on the pathogen – quantitative test would be preferable to qualitative analyses. The results of the questionnaire allow the estimation that the relative share of rapid NATs is expected to be around 5% compared to the total number of analysis per month. It should be noted that the bacteria and viruses for which rapid tests are requested are being frequently examined at present in the participating laboratories utilising PCR.

Besides conventional nucleic amplification technique by PCR, only few laboratories (8%) apply isothermal amplification. This demonstrates that research and development is still necessary to overcome the drawbacks of isothermal methodologies for nucleic acid amplification. In particular, the possibility to obtain quantitative results would certainly accelerate the application of isothermal tests in routine laboratories and for point of care applications. In addition, integration of isothermal tests in disposable cartridges is essential to improve handling, reproducibility and to reduce the turnaround time to the requested range of 15 min. Because of the ongoing development most of the end-users (85%) are still waiting before purchasing corresponding equipment to utilise rapid (isothermal) nucleic analysis. However, we conclude from the questionnaire that for certain applications rapid NATs are needed to improve the measurement support for diagnostic and therapeutic decisions.

Heinz Zeichhardt and Hans-Peter Grunert are shareholders of GBD (Gesellschaft fuer Biotechnologische Diagnostik) mbH, Berlin (Germany), which is a manufacturer of materials for external quality control. The other authors declare that they have no competing interests.

The questionnaire was disseminated among analytical laboratories participating in round robin tests for nucleic acid tests organised by INSTAND e.V. (Society for the Promoting Quality Assurance in Medical Laboratories, Düsseldorf, Germany) for external quality assurance. We gratefully acknowledge the support of INSTAND e.V., in particular Michael Spannagl, chairman of INSTAND and Ingo Schellenberg, Vice-Chairman of INSTAND. The work was funded by the European Union within the European Metrology Research Programme (EMRP) HLT-08, 2011 ‘INFECT-MET’.