The Journey to a COVID-19 Diagnostic Test

The Journey to a COVID-19 Diagnostic Test

by Dr. Zvi Loewy

[Editor’s Note: In light of the recent COVID-19 outbreak, we are including articles by those responding to the challenge of testing and treating the disease. The first two companies to receive Emergency Use Authorization from the FDA to roll out assays to identify infected individuals were Thermo Fisher Scientific and Roche Molecular Systems. A Q&A with Thermo Fisher Scientific is included in a separate piece in this issue; in this article we highlight the Roche Cobas Amplicor System currently being used for screening large numbers of samples.

We contacted Zvi Loewy to ask him to explain how the system works. Dr Loewy was the Group Leader at Hoffmann-La Roche from 1990 – 1995 when he directed the Cobas Amplicor development program from inception up to commercial launch, including initiating and supporting international evaluations as well as managing international clinical trials for FDA and international regulatory approval.

The current version of the Roche Cobas Sars-CoV-2 test runs on the Cobas 6800 and 8800 with a capability of processing samples in standard multi-well plates in about 3.5 hours. Depending on the number of systems and staffing, a laboratory can run thousands of samples per day.]

Introduction: Nucleic Acid Amplification and the Evolution of Molecular Diagnostics

The Polymerase Chain Reaction (PCR), a method to replicate DNA and generate multiple copies of a specific nucleic acid, was invented by Kary Mullis a research scientist at Cetus Corporation in 1983. The initial reporting of PCR was presented at  the  American Society for Human Genetics annual meeting  in 1985 and subsequently published.1 The key steps in the PCR process are denaturation of target nucleic acid, annealing of primers and extension. To amplify the region of a specific target sequence, PCR requires a series of temperature cycles. The temperature cycles ensure that a specific sequence of DNA is doubled every cycle, leading to exponential amplification of the targeted segment. In theory, the number of copies made after 20 cycles is about one million, one billion copies after 30 cycles; in practice, the number of copies per cycle may not exactly double but close enough that a rough estimate of amplification is sufficient for most purposes An important point here is that PCR needs DNA as the template; for viruses with an RNA-based genome, an additional reverse transcription step needs to be performed at the start of  the amplification process.

Three key technology enablers have been fundamental in transforming PCR from a theoretical process to routine diagnostic practice:

  1. The identification and integration of thermophilic DNA polymerases;
  2. Development of high throughput detection formats; and
  3. Process automation.

In 1988, a Cetus research team isolated a thermophilic DNA polymerase from the bacterium Thermus aquaticus (Taq DNA polymerase) found in hot springs.2 Because Taq polymerase will maintain its functionality at temperatures up to 95°C (even after multiple cycles), the incorporation of Taq polymerase into PCR obviated significant human intervention during the cycling process and enabled the automation of PCR through the development  and deployment of a thermal cycler by PerkinElmer in 1988. In 1991, Roche purchased the rights to PCR from Cetus with the intent to build a molecular diagnostics business.

Format Development

1. Microwell Plate Detection

For PCR to be used in routine clinical diagnostics it was necessary to identify detection format technologies that were familiar to the clinical diagnostics community, enabled high throughput and integrated well with the thermal cycler footprints. To address these requirements, Roche developed and commercialized the Amplicor product line. The Amplicor platform was predicated upon use of microwell plates  that would enable high-throughput processing. The Amplicon PCR assay was performed  with biotin labeled primers with the resulting amplicons (PCR products) subsequently captured on microwell plates coated with a specific capture probe and detected by colorimetric detection.3,4 This ELISA-based approach was very successful in introducing PCR to the clinical diagnostics industry. In 1993, the AMPLICOR® test for Chlamydia trachomatis received 510(k) clearance by the FDA making it the first FDA approved diagnostic IVD test predicated on PCR technology.

“The cobas® master mix contains detection probes specific for the target, each probe labeled with unique fluorescent dyes acts as reporter”


Historically, automated systems have been the workhorse in clinical diagnostics. Although advances in chemistry allowed PCR technology to progress initially at a rapid rate, systems integration and process automation, with

the exception of thermal cyclers, lagged  behind. The first fully automated system for performing routine diagnostic PCR, the COBAS AMPLICORTM system, was described in 1996.5 The COBAS AMPLICORTM is a fully integrated system that amplifies and detects nucleic acids; input targets include both DNA, as well as RNA that was subjected to reverse transcription.

The COBAS AMPLICORTM system includes an amplification subsystem comprised of two independently controlled thermal cyclers and a detection subsystem predicated on ELISA-like technology that uses oligonucleotide probe- coated paramagnetic microparticles. A broad portfolio of infectious disease tests has been developed for the COBAS AMPLICORTM system.

3. cobas 6800/8800 Systems

Since 2014, the cobas 6800 and cobas 8800 systems have provided fully integrated, automated solutions for viral load monitoring, donor screening, sexual health and microbiology. The cobas®  6800/8800  instrument  is composed of four system modules:

  1. Sample supply;
  2. Sample Transfer;
  3. Sample Processing; and
  4. Analysis.

Samples are aliquoted into wells of standard assay plates, either individually or as multiple samples pooled into a single well. Sample preparation in each plate well is initiated by release of the target  nucleic acid through the addition of proteinase and lysis reagent to the sample, then capturing the nucleic acid onto the silica surface of magnetic glass particles added after cell lysis. These particles are immobilized in the well with an array of magnets positioned  under the plate. Unbound substances are removed by successive wash steps; purified nucleic acids bound to the complement sequence on the magnetic glass particles held on the well surface are eluted from the particles with elution buffer at elevated temperature. Selective amplification

of target nucleic acid is enabled by the use of target-specific primers. The cobas® master mix contains detection probes specific for the target, each probe labeled with unique fluorescent dyes acts as reporter. Reporter probes are measured at defined wave lengths tuned to the dyes specific to the assay. Each probe also has a dye which acts as a quencher; when the probe is not bound to the target sequence, the fluorescent signal of the probe is suppressed by the quencher dye. During PCR amplification hybridization of the probes to the specific DNA template results in cleavage by the 5’ to 3’ nuclease activity of the DNA polymerase; the separation of the reporter and quencher dyes allows the generation of a fluorescent signal. With each PCR cycle, increasing amounts of the cleaved probes are generated and correspondingly the cumulative signal of the reporter dye is increased. Results on the cobas® 6800/8800 system can be viewed directly on  the system screen, printed as a report, or sent to a Laboratory Information System (LIMS).


Coronaviruses are members of the subfamily Coronavirinae in the family Coronaviridae and the order Nidovirales. The subfamily Coronavirinae includes four genera: Alphacoronavirus, Betacoronavirus, Gammacoronavirus and Deltacoronavirus. The Alphacoronavirus and Betacoronavirus genera infect mammals exclusively, whereas the Gammacoronavirus and Deltacoronavirus genera infect birds primarily. The Alphacoronaviruses and Betacoronavirsues cause respiratory illness in humans and gastrointestinal disease in animals. Based on sequence databases, all human coronaviruses have animal origins.

Although many coronaviruses are pathogenic to humans, the majority corelate with mild clinical symptoms.6  Two exceptions are the severe acute respiratory syndrome coronavirus (SARS-CoV) that originated in southern China in 2002 and resulted in greater than 8000 human infections7 and the Middle-East respiratory syndrome coronavirus (MERS-CoV) originally detected in Saudi Arabia in 2012 and accounted for approximately 2500 infections.8

In December 2019, several pneumonia cases caused by a newly identified b-coronavirus were identified in Wuhan, China. The disease outbreak started in a seafood market. The new coronavirus was named COVID-19 (coronavirus disease 2019) by the World Health Organization (WHO) in February 2020. The Coronavirus Study Group (CSG) of the International Committee proposed the name SARS-CoV-2.

Based on virus sequencing and evolutionary analysis, the bat has been proposed as the natural host. The COVID-19 genome sequence and Bat CoV RaTG13 sequence have been shown to be 96.2% identical; suggestive of a common ancestor.9 It is likely that SARS-CoV-2 was originally transmitted from bats via an intermediate host culminating in human infection.


The genomes of the viruses belonging to the Coronaviridae family are single-strand, positive-sense RNA, 26 to 32 kilobases in length. The sequence of SARS-CoV-2 was obtained from analysis of eight complete genome sequences generated by next generation sequencing from bronchoalveolar lavage fluid samples isolated from eight patients. The eight complete genomes were found to be nearly identical across the entire genome.10 The single-stranded RNA genome of SARS-CoV-2 is 29891 bases in size and encodes for 9860 amino acids. The GC content of SARS-CoV-2 is 38%. The organization of the  SARS-CoV-2  genome  corresponds  to a 5’ replicase and structural proteins that include the spike, envelope, membrane and nucleocapsid proteins.

In addition, there are 12 putative open reading frames.11 The SARS-CoV-2 sequence was found to be most closely related to the bat-SLCoVZC45 virus sequence and the bat-SL-CoVZXC21 sequence. The highest level of sequence identity was determined to be in the E gene with 98.7% homology. The lowest degree of sequence identity was in the S gene with 75% homology. The majority of encoded proteins showed high sequence identity between SARS-CoV-2 and the bat coronaviruses. The exceptions were the spike protein with 80% sequence identity and protein 13 with 73.2% sequence homology. In addition to genome sequencing and analysis, SARS- CoV-2 cell membrane proteins were evaluated for possible clues to cell entry and subsequent infection. In humans, SARS-CoV-2 entry into cells is likely mediated by the human cellular receptor angiotensin-converting enzyme 2 (ACE2) and the serine protease TMPRSS2 for S protein priming.12

cobas SARS-CoV-2 Test

The Roche cobas 6800/8800 system is used to perform the cobas SARS-CoV-2 test.13 The system throughput is up to 96 results in about three and a half hours and a total of 384 results are obtained with the cobas 6800 system and 960 results with the cobas 8800 system in eight hours. The cobas 6800 / cobas 8800 COVID-19 / SARS-CoV-2 test is a single-well  dual target qualitative assay for the detection of nucleic acid from SARS-CoV-2 in nasopharyngeal and oropharyngeal swab samples. The SARS- CoV-2 requires the addition of an RNA internal control consisting of non-Sarbecovirus related RNA that is added to each specimen. The internal control is used as an extraction control and

tested in concert with each specimen. As such, the internal control is used to monitor the sample preparation and PCR amplification process. Performance of the assay entails the simultaneous extraction and purification of nucleic acid from patient samples and the internal control. Amplification of SARS-CoV-2 target nucleic acid from the sample is enabled by primers directed to the SARS-CoV-2 specific ORF 1 reg (Target 1) ion and a conserved region of the envelope E-gene (Target 2). The two targets are detected by sequence-specific probes labeled with different fluorophores.

In silico analysis demonstrated that the Cobas SARS-CoV-2 assay will pick up all analyzed SARS-CoV-2 sequences in the NCBI and  GISAID databases. Cobas SARS-CoV-2 had an approximate 100% match with the exception of all but one Target 1 sequence. When analyzing Target 2, the Cobas SARS-CoV-2 had a 100% match with the exception of three sequences.

In silico analysis for cross reactivity with greater than 35 infectious agent target sequences did not result in any evidence of cross reactivity. Cross reactivity was also laboratory tested. Testing a panel of multiple unique sub-species microorganisms did not result in any false positive results; Roche’s fact sheet states that the test is designed to minimize false positives with a proviso about the consequences of a false positive.14

Prior to launch of the assay, a clinical evaluation was performed with nasopharyngeal swab samples. A set of 100 negative samples and 50 “simulated” positive samples were tested. Low positive and moderate positive “simulated” clinical samples were prepared by spiking cultured USA-WA1/2020 strain  virus into negative clinical samples to either 1.5x LoD or 4x LoD. All low positive and moderate positive samples were positive; all negative samples were negative.13,14 Roche received Emergency Use Authorization (EUA) for the cobas® SARS-CoV-2 test on March 13, 202014; the first shipment was in mid-March for large-scale testing.


The field of molecular diagnostics has developed at a rapid pace over the past twenty- five years. Assays and fully integrated systems are now readily available for many pathogens and genetic disorders. The emergence of a pandemic during the first quarter of 2020 necessitated, and correspondingly, resulted in the accelerated development and approval of a diagnostic for SARS-CoV-2. The Roche cobas SARS-CoV-2 was implemented in the clinical laboratory in March 2020. Since then, the system has been used to diagnose and track the disease through large-scale laboratory testing and analysis.

Dr. Zvi LoewyDr. Zvi Loewy is Professor and Associate Dean for Research at Touro College of Pharmacy, New York, NY and Professor of Microbiology and Immunology at New York Medical College, Valhalla, NY. Professor Loewy is a senior academic leader and an experienced global pharmaceutical – biotechnology executive who leverages a diversified background in big-pharma senior management, biotech startup creation and academia. Dr. Loewy is a true visionary, who has been the key inventor on many patents including human DNA identity assays, fully integrated molecular diagnostic platforms, innovative medical devices and consumer healthcare products. Of note, Dr. Loewy led the development of the Cobas Amplicor™ System. Dr. Loewy’s international experience has included leading international research teams; championing the penetration and commercial launch of consumer healthcare products in China and developing markets; and leading open innovation in the Mid-East. Professor Loewy received his education at Rensselaer Polytechnic Institute and at the Albert Einstein College of Medicine. Dr. Loewy is on the boards of the Jerusalem College of Technology and the New Jersey Technology Incubator; and is an Editor of the Journal of Prosthodontics. Professor Loewy has published broadly and has over 25 issued patents.


  1. Saiki, K., Scharf, S., Faloona, F., Mullis, K.B., Horn, G.T., Erlich, H.A. and Arnheim, N. (1985) Enzymatic amplification of globin genomic sequences and restriction site analysis for diagnosis of sickle cell anemia. Science 230: 1350-1354.
  2. Saiki, K., Gelfand, D.H., Stoffel, S., Scharf, S.J., Higuchi, R., Horn, G.T., Mullis, K.B. and Erlich, H.A. (1988) Primer-directed enzymatic amplification of DNA. Science 239: 487-491.
  3. Pottathil, R and Loewy, G. (1993) Aids diagnostics, p. 411-423. In B. Ram, P. Singh and P. Tyle (ed.). Diagnostics in the year 2000. Van Nostrand Reinhold, New York.
  4. Loewy, G., Mecca, J. and Diaco, R. (1994) Enhancement of Borrelia burgorferi PCR by Uracil N-Glycosylase.
  5. Clin. Microbio. 32: 135-138. DiDomenico, , Link, H., Knobel, R., Caratsch, T. Weschler, W., Loewy, Z.G. and Rosenstraus, M. (1996) COBAS AMPLICORTM: fully automated RNA and DNA amplification and detection system for routine diagnostic PCR. Clin. Chem. 42: 1915-1923.
  6. Su, , Wong, G., Shi, W., Liu, J., Lai, ACK., Zhou, J., Liu, W., Bi, Y. and Gao, G.F. (2016) Epidemiology, genetic recombination, and pathogenesis of coronaviruses. Trends Microbiol. 24: 490-502.
  7. Chan-Yeung, and Xu, R.H. (2003) SARS: epidemiology. Respirology 8: S9-14.
  8. Zaki, M., van Boheemen, S., Bestebroer, T.M., Osterhaus, A.D. and Fouchier, R.A. (2012) Isolation of a novel coronavirus from man with pneumonia in Saudi Arabia. N. Engl. J. Med. 367: 1814-1820.
  9. Zhou, , Yang, X-L, Wang, X-G, Hu, B., Zhang, L., Zhang, W., Si, H-R, Zhu, Y., Li, B., Huang, C-L., Chen, H-D., Chen, J., Luo, Y., Guo, H., Jiang, R-D., Liu, M-Q., Chen, Y., Shen, X-R., Wang, X., Zheng, X-S., Zhao, K., Chen, Q-J., Deng, F., Liu, L-L., Yan, B., Zhan, F-X., Wang, Y-Y., Xiao, G-F. and Shi, Z.L. (2020) A pneumonia outbreak associated with a new coronavirus of probable bat origin. Nature 579: 270-273.
  10. Lu, , Zhao, X., Li, J., Niu, P., Yang, B., Wu, H., Wang, W., Song, H., Huang, B., Zhu, N., Bi, Y., Ma, X., Zhan, F., Wang, L., Hu, T., Zhou, H., Hu, Z., Zhou, W., Zhao, L., Chen, J., Meng, Y., Wang, J., Lin, Y., Yuan, J., Xie, Z., Ma, J., Liu, W.J., Wang, D., Xu, W., Holmes, E.C., Gao, G.F., Wu, G., Chen, W., Shi, W. and Tan, T. (2020) Genomic characterization and epidemiology of 2019 novel coronavirus: implications for virus origins and receptor binding. Lancet 395: 565-574.
  11. Chan, F-K., Kok, K-H., Zhu, Z., Chu, H., To, K. K-W., Yuan, S. and Yuen, K-Y. (2020) Genomic characterization of the 2019 novel human-pathogenic coronavirus isolated from a patient with atypical pneumonia after visiting Wuhan. Emerging Microbes and infections 9: 221-236.
  12. Hoffmann, , et al., SARS-CoV-2 Cell Entry Depends on ACE2 and TMPRSS2 and Is Blocked by a Clinically Proven Protease Inhibitor, Cell 181, 1–10, April 16, 2020,
  13. cobas®SARS-CoV-2 Test,
  14. Roche submission to FDA; see:
    1. Roche submission letter to FDA,
    2. Updated instructions,
    3. Fact sheet for cobas®SARS-CoV-2 – Molecular Systems, I,
    4. Roche begins shipments of fi rst 400,000 COVID-19 tests to laboratories across US to begin patient testing
      under FDA Emergency Use Authorization