The Disease Detective Work of Lab Diagnostics

Core Concepts

In this article, you will explore the biochemical underpinnings of lab diagnostics including rapid antigen tests and PCR. You’ll learn about their advantages and challenges, as well as the role of COVID-19 diagnostics in curbing a burgeoning pandemic.

This is the sixth article in a special ChemTalk mini-series about the intersections between chemistry and public health, using COVID-19 as a case study. Across this series, you can expect to learn about the drug discovery and development processes, chemistry’s central role in diagnosing and preventing diseases, and careers that are on the front line of chemistry and public health.

Previous Articles in This Mini-Series

> The Chemistry Behind Coronaviruses
> The Drug Discovery Process
> The Drug Development Process
> Chemistry Careers in the Public Health Field
> The Biochemistry Behind COVID-19 Medications

How do we diagnose diseases?

You know the feeling: you went to bed feeling fine, only to wake up under the weather. You’re certain something’s not right. What do you do next?

Maybe you’re the type of person who calls out sick from school, then seeks support straight away from their doctor. Other folks pick up over-the-counter medications from their local pharmacy and power through the rest of their day. No matter which category you find yourself in, how do you know where to start?

First and foremost, you need to understand what’s causing your illness. Without that key piece of information, you can’t decide how to treat it. This is where medical diagnostic testing shines. Diagnostic testing is a standardized, consistent strategy to determine whether a person has a specific type of infection. It’s one handy technique in the medical arsenal that helps health care professionals (HCPs) and public health experts track disease. The tests are tailor-made to detect that specific type of infection, while excluding others. In doing so, this helps narrow down the symptoms’ possible causes.

Strictly speaking, diagnostic testing isn’t always necessary. Some infections clear up on their own thanks to basic remedies like rest and fluids. Better yet, sometimes your symptoms resolve themselves before you even manage to obtain medicine in the first place. But in other circumstances, like a local outbreak or a widespread pandemic, it’s important to know for sure what’s causing your illness. (Though it’s worth noting that, in some cases, diagnosis happens without a formal diagnostic test. For example, an HCP at an urgent care clinic might prescribe you medication based on your symptoms alone, without waiting for official diagnostic test results.)

Repeatedly, we’ve witnessed this firsthand over the past several years. COVID-19 still circulated during flu season, compounded by the co-presence of RSV, with all three viruses boasting similar symptoms. Taking COVID-19, flu, or RSV tests helped people ascertain which virus they’d contracted or rule out those they hadn’t. Diagnostic testing helps health experts monitor cases and track disease spread, both among individual patients and across entire populations. To do that to the best of their ability, for everyone’s benefit, they must have an accurate measure of when, where, and how many cases arise. On a much smaller scale, your HCP needs to know what’s causing your infection, so they can design a successful treatment plan for you. Either way, there’s a mystery afoot — and we need to solve it.

Jinkies, Let’s Split up and Look for Clues!

So you woke up with a sore throat, nasal congestion, and a cough. During the pause in your pity party, you ask yourself, “Is this a cold, the flu, COVID-19, or something else?”

This type of inquiry characterizes the process of differential diagnosis. Differential diagnosis is the task of distinguishing between multiple conditions that present similar signs or symptoms. It’s a complex journey through which your HCP strives to identify the precise pathogen that’s making you feel this way.

This might lead you to wonder next about the purpose of differential diagnosis. After all, why bother identifying the cause of your illness, when you could instead be focusing on relieving your symptoms? Ideally, your course of treatment would address both the cause and the symptoms of your illness. But if you try to alleviate your symptoms without addressing their underlying cause, your relief might only be brief. Symptomatic treatment often serves as a temporary, surface-level solution, and can’t cure what’s ailing you. An example might be drinking herbal tea to soothe your ailing throat, or taking over-the-counter acetaminophen for fast headache relief. Contrarily, if you seek out and stamp out the underlying cause of your illness, you can manage it much better. You’d likely even see your symptoms naturally subside due to you (or your HCP) identifying and directly addressing the cause.

How does diagnostic testing impact patients?

This whole process of addressing the cause of your ailment begins with a diagnosis. Diagnostic testing can contribute to a more comprehensive, accurate diagnosis. A diagnostic test may present a positive test result that definitively concludes you’re infected with a specific disease. Or, it may present a negative test result that rules out a disease you don’t have. Either outcome is a great starting point for understanding what your immediate next steps should be.

HCPs frequently use diagnostic testing alongside their other medical mystery-solving skills. They pair the test results with what they already know about your signs, symptoms, medical history, lifestyle, risk factors, and even details like your travel history. In doing so, they paint a well-rounded picture of your health and the circumstances leading to your infection. Combining all of this information together improves their chances of giving you an accurate diagnosis.

That completely transforms the landscape of health care, and the benefits of diagnostic testing are twofold. It gives patients priceless peace of mind, and HCPs the full confidence to move forward with a suitable treatment plan. Test results, whatever they are, provide HCPs with concrete answers that help determine the patient’s prognosis and next steps. With this in mind, we clearly see that it’s crucial to make an accurate diagnosis from the start. After all, this diagnosis will directly influence the course of treatment. Therefore, an incorrect diagnosis can have serious consequences for the patient.

Why does differential diagnosis matter?

Let’s return to our example. As of this morning, you’re feeling unwell with those bothersome upper respiratory symptoms we described above. It could certainly be the common cold, which usually doesn’t require any specialized treatment and doesn’t have any drug treatments available. Or maybe it’s the flu, which is a bit more incapacitating, but at least flu-specific antiviral medications exist for you.

You visit your HCP to seek treatment. Based on your description of those (rather nonspecific) symptoms, they diagnose you with a bacterial sinus infection and prescribe you antibiotics. However, if your HCP’s diagnosis was incorrect and you actually have, let’s say, the flu, those antibiotics won’t help. Not even a little bit! Antibiotics are only effective against bacteria, not viruses. (Although some treatments might ease the symptoms of multiple diseases, antibiotics operate based on the unique biochemistry underlying a particular bacterial infection. Likewise, antiviral medications are tailor-made to fight specific viruses, so their utility against bacterial infections is next to none.) When all is said and done, that delay in proper treatment leaves you suffering from your bacterial sinus infection for longer.

Differential diagnosis is critical in scenarios like this one. Without distinctive symptoms, your HCP must find some way to identify your illness with as much certainty as possible. Diagnostic tests can lend a very helpful hand here, giving your HCP information that your symptoms can’t: specific, scientific proof. If your HCP had, say, performed a flu test and it was negative, they could at least rule out the possibility that you had the flu. They also could have sampled a culture from your sinuses, driving the conclusion that you have a bacterial sinus infection. Diagnostic testing may be a multi-step process, and sometimes ruling out non-causes is more productive than identifying actual causes.

Luckily, biomedical researchers are improving contemporary diagnostic tests all the time to be more sensitive, detailed, and precise. As we’ll see shortly, the development of these fine-tuned antigen tests and PCR tests represent modern innovations in diagnostic science. Although their applications extend elsewhere, these two methods formed the basis of COVID-19 testing as the pandemic emerged and proliferated. Next, we’ll investigate the science of how these tests deliver vital, reliable, urgent information when it’s needed most.

COVID-19 Diagnostics

What makes the difference between a “good” diagnostic test and a “bad” one? Good diagnostic tests deliver conclusive, unambiguous, understandable results that the patient or their HCP can use productively.

To achieve this, a test needs to have some key qualities that set it up for success. First and foremost, as with any medical apparatus, it must be safe and effective. Put differently, the act of testing shouldn’t present any unnecessary health risks to the patient. Even better, non-invasive tests pose virtually no such risks whatsoever. Invasive procedures involve entering the body, either through its natural orifices or by breaking the skin, like in surgery. Meanwhile, non-invasive procedures can be performed without needing to enter the inside of the body. Very soon, we’ll learn about COVID-19 tests that technically fall into the “invasive” category, because they involve inserting a swab deep into the nasal cavity through the nostrils.

What about effectiveness? No test is perfect, and it’s inevitable that false positive or false negative results occasionally happen. (More on that later!) In general, though, a sound test will present accurate results with reliable consistency. If a test can’t be trusted to deliver the right diagnosis, what’s the point of taking it at all?

Finally, from a population-level scope, diagnostic testing is most effective when it’s easily accessible and widely available. That means managing factors that impede test manufacturing and distribution, and financial barriers that prevent patients from obtaining it. In a perfect world, we’d have all types of diagnostic tests for all kinds of diseases accessible at all times. Any physical discomfort, flu-like symptom, or itch in the throat would be diagnosed right then and there. No waiting, no pre-authorization, and certainly no need to negotiate with your health insurance company!

Unfortunately, it’s not realistic to expect this scenario to always be the case. In actuality, these barriers do exist for many patients, and ultimately, we’re all at the mercy of the supply chain. Later in this article, we’ll discuss specific challenges that diagnostic testing entails. For now, let’s see how these qualities play out in a pandemic, when test availability and ease of access are at their most imperative, and how diagnostic testing helped control COVID-19.

How do COVID-19 diagnostics work?

Sometimes we can diagnose a single disease using multiple different kinds of diagnostic tests. Although this may seem redundant, it’s actually quite useful in many instances. If a patient can’t access or afford a specific test, it’s beneficial to have other, inexpensive test types available too. Along the same lines, insurance companies might not cover the cost of a particular test, so over-the-counter test options or public testing clinics might offer more convenience. Finally, as we’ll explore in greater depth soon, it is possible for diagnostic tests to yield false results. If there’s any doubt about the test’s accuracy, a patient may want to take a different, more reliable type of test to compare it to the original result. This way, having multiple test options available helps illustrate a more detailed image of the patient’s current health.

No matter the circumstances, having multiple diagnostic tests on the market for a single disease is a strong strategy. We observed this during the COVID-19 pandemic, as biomedical experts developed various diagnostic tests to identify the SARS-CoV-2 virus. In this section, we’ll cover a few of the most common ones, antigen tests and PCR tests. We’ll also investigate the science behind how they helped us make big strides in managing COVID-19.

Antigen Tests: When You Need to Know Now

The first type of COVID-19 diagnostic test that we’ll cover is an antigen test, also called a rapid test. True to their name, rapid tests are a swift way to determine if you have a particular disease. Soon we’ll discuss how they work in deeper detail, but for now, you should know that their job is to give patients a straightforward, yes-or-no answer as soon as possible. Results are typically ready in less than 30 minutes, although this varies depending on the test. These convenient, easy-to-use tests minimize many of those barriers we mentioned above. Patients don’t need to visit a lab for the test, often purchasing them over-the-counter and performing them at home instead.

The beauty of antigen tests is that their speedy results are thanks to a simple antigen-antibody binding reaction. Antigens, substances that elicit the body’s immune response, are a feature of infectious agents like viruses and bacteria. Each viral or bacterial disease presents unique antigens that are characteristic to that disease. Antibodies are proteins that detect the presence of antigens. Like a lock and key, a specific type of antibody only binds to a specific type of antigen. Whether or not any binding occurs translates into either a positive or negative result on the antigen test. This small reaction goes a long way in moving public health forward — let’s see how.

How do antigen tests work?

Now we understand that viruses like SARS-CoV-2 produce antigens, which specific antibodies can detect and bind to. This binding leads to a positive result on an antigen test. Since a reaction between immune agents is the basis of this test, we can describe antigen tests as an immunoassay.

There are many kinds of immunoassays. Among the most widely known is the enzyme-linked immunosorbent assay. In an ELISA, antibodies are linked to enzymes that, in turn, bind to a molecule of interest. The amount of binding indicates to scientists how much of the molecule of interest is present in a sample. Even if you haven’t performed one yourself, you’ve likely seen the colorful 96-well plates that carry out an ELISA. ELISAs are abundant in the research industry, with applications in pharmaceuticals, biotechnology, forensic science, toxicology, and lots more.

A less complex, less sophisticated version of the ELISA reaction is at the heart of antigen tests. Any given antigen test can only detect a particular kind of antigen, and thus can only detect a specific disease. In other words, an accurate COVID-19 antigen test can’t detect if you have strep throat or malaria, only COVID-19. This specificity exists because, before use, the antigen test is already equipped with antibodies selective for that antigen of interest. These antibodies will uniquely and precisely bind to those antigens only. (We’ll go into more detail about specificity later.)

This approach to antigen testing is known as a lateral flow immunoassay. The “lateral flow” portion of the name means it operates via capillary action, moving a liquid through the test strip. For COVID-19 tests, this liquid is a body fluid sample, usually a nasopharyngeal sample mixed with the test’s buffer. These tests have everything you need (except for the sample, which you provide) to conduct them in the comfort of your own home.

Journey to the End of a COVID-19 Antigen Test Strip

Let’s walk through the process of taking a lateral flow antigen test. You or your HCP carefully insert a swab into your nose — the deeper, the better to obtain a high-quality sample, despite the physical discomfort — and collect a sample. The sample is placed on the test strip, along with some buffer. If there’s any SARS-CoV-2 in the sample, this buffer suspends the virus and extracts its proteins. Proteins from the virus serve as the antigens in the test.

Now, the buffer-sample mixture (the analyte) moves through the test strip. The test strip already contains adntibodies, unique to those SARS-CoV-2 antigens. These antibodies are pre-labeled with a tag such as colored nanoparticles. As the analyte migrates along the test strip, the strip’s antibodies bind to viral antigens if present in the analyte. Then, the bound antibodies move with the antigens down the strip.

Waiting farther down the test strip is a test line, which is a row of SARS-CoV-2–specific antibodies. Beyond that is a control line, whose antibodies (usually immunoglobulins from a rabbit or similar animal) can bind to the antibody-nanoparticle complex.

The analyte reaches the test line first. If the analyte contains SARS-CoV-2 antigens, those antigens will bind to the test line’s antibodies. The test line will then appear as a visible, colored band due to the colored nanoparticles.

Meanwhile, the analyte continues moving down the test strip and reaches the control line. If any SARS-CoV-2 antigens remain in the analyte at this point, the control line’s immunoglobulins anchor them now via binding. Again a colored band forms, this time at the control line. If the sample didn’t contain any SARS-CoV-2 antigens in the first place, the control line’s immunoglobulins bind only the antibody-nanoparticle (A.K.A. conjugate-antibody) complex.

The outcome is either one or two colored bands, according to whether the analyte lacks or contains SARS-CoV-2 antigens. In either scenario, a colored band will form at the control line because it binds to antibodies no matter what. (It binds to the test’s antibody-nanoparticle complex if the analyte lacks antigens, or it binds to the antibody-antigen complex if the analyte contains antigens.) If the analyte contains antigens, a second colored band will form at the test line, whose SARS-CoV-2–specific antibodies bind them. This second band won’t form if the analyte lacks antigens.

To phrase it differently, the formation of two bands (at the test line and the control line) reflects a COVID-19–positive result. However, formation of only one band (at the control line) doesn’t necessarily mean that the patient is COVID-19–negative. Instead, it means that no SARS-CoV-2 antigens were detectable in the patient’s sample. Due to the risk of a false negative result, it’s best to follow up a negative antigen test result with another confirmatory antigen test or a more accurate diagnostic test like PCR. But what goes into making the result what it is?

What factors impact an antigen test result?

Prior to starting the antigen test, there’s no visible band at the test line. During the course of the test, if a band forms at the test line, that band may appear quite faint. Remember how, in the image of the 96-well ELISA plate, darker colors indicate more antibody-antigen binding? The same notion applies to antigen tests, too. The test line’s band appears darker if there are more SARS-CoV-2 antigens in the analyte, or lighter if fewer antigens. But the appearance of any band at all, even a light one, implies that there are at least some antigens present in the analyte. For that reason, if the test yields two bands, no matter how light, patients should consider this a positive test result.

So, antigen tests are easiest to interpret when the patient (and therefore their sample too) have a lot of virus. High viral load will generate a clearer, darker colored band at the test line. It takes time for the virus to replicate inside the patient’s body, so viral load is low during a COVID-19 infection’s early stages. We must remember this because it means antigen tests might inadvertently miss infections if patients take the test too early. To circumvent this, if you’ve had a known COVID-19 exposure, the FDA recommends taking two antigen tests at least 48 hours apart. If you’re infected, the virus will replicate a lot in those 48 hours and increase the viral load. This increases the likelihood of a positive second test, even if your first test was negative.

Head, Shoulders, Nose and Throat, Nose and Throat!

Nasopharyngeal swabs are the go-to samples for COVID-19 tests, but antigen tests for other diseases may take other formats. For example, antigen tests for strep throat prefer a throat swab, while those for malaria use blood samples (via fingerstick). It all depends on where each disease’s antigens are most concentrated in an infected person’s body. The bacteria that cause strep throat, and their antigens, are abundant in the throat, while malarial antigens are present in the bloodstream. SARS-CoV-2 antigens, in turn, are readily found in the nasal passages. That’s because cells in the respiratory tract often have ACE2 receptors, the receptor that the virus uses to infect its host.

Once the sample is collected, the antigen test proceeds quickly. The only factor limiting its speed is how long it takes for the analyte to fully traverse the test strip. It’s usually only a matter of minutes, and this capillary action’s fairly short duration is what makes them “rapid” tests. The benefit of these fast results is offset by the fact that antigen test results may be less accurate. If the test is done incorrectly, as sometimes happens when the patient performs the test instead of an HCP, it won’t reliably detect SARS-CoV-2 antigens. For example, the user might not sufficiently rehydrate their sample in the buffer before placing it on the test strip. And as we mentioned earlier, if the test happens too soon before the virus has adequately reproduced, the test might not pick up on the analyte’s antigens.

From an epidemiological perspective, COVID-19–specific antigen tests are a fantastic first line of defense in preventing viral spread. By the end of the pandemic, antigen tests’ wide availability let patients easily access them without a prescription or pre-authorization at pharmacies, supermarkets, and similar retailers. Antigen tests’ safe, minimally invasive technique made them an appealing and approachable way for anyone to get answers fast. In situations like a pandemic, when infections are spreading fast, the ability to get test results faster is immensely valuable. Nevertheless, antigen tests can be helpful in other circumstances too — not just during fast-moving pandemics or for infectious diseases. Sometimes, even routine applications of antigen tests can have life-changing results.

Additional Applications of Antigen Tests

Most of us likely have personal experience with COVID-19 antigen tests, but their technology precedes the pandemic. It’s long been applied to detect other diseases like strep throat, malaria, influenza, and more. Did you know that even the familiar, over-the-counter pregnancy tests are a form of antigen test? In this context, the “antigen” involved is human chorionic gonadotropin (hCG), a hormone present in the urine during pregnancy. The pregnancy test is outfitted with antibodies that yield a positive test result if hCG binds to them. So, after the pregnancy test interacts with a urine sample, the test result will be positive (“pregnant”) if the urine contains hCG, or negative (“not pregnant”) if the urine lacks hCG.

This is a great example of how, interestingly, antigen tests can take either the form of invasive or non-invasive tests. Earlier, we discussed how COVID-19 antigen tests collect the patient’s sample via a swab that’s inserted through the nostrils. That’s an invasive procedure, since it requires entering the inside of the body. However, pregnancy antigen tests as described above are a non-invasive type of antigen test. Entering the body isn’t necessary to generate a test result; that happens outside of the body, when the test comes into contact with the patient’s urine sample. Invasive pregnancy tests exist too, where an HCP collects a blood sample and the blood’s hCG levels are evaluated. The difference is, this highly accurate procedure can’t be performed from the convenient comfort of the patient’s own home.

But this type of at-home pregnancy test, just like any antigen test, may present inaccurate results. (False negatives are more common than false positives, since hCG is almost always only produced during pregnancy. Timing also matters: if a pregnant person takes an antigen test too early in their pregnancy, the test might not be able to detect hCG yet.) Besides, there’s always a chance that the individual will conduct the test wrong due to user error or not following the test’s instructions properly. With this in mind, it might take a few repeats of an antigen test to confirm the result. Since people often take them alone or without an HCP present, it’s important to design tests such that the average, non-HCP can easily interpret its results.

The best antigen tests provide results consistently and unambiguously. Note how, in the image above, there are only two possible outcomes: “pregnant” (two lines) or “not pregnant” (one line). It would be a problem if the test showed one obvious line and one fainter line — how is a user supposed to interpret that? Is it possible for a person to be half-pregnant, or only slightly pregnant? (We’ll save you an Internet search: The answer is no.) The antigen test’s accuracy no longer matters if its results are impossible to interpret anyway. Having a binary result, like pregnancy tests do, helps the user easily and correctly interpret the result the first time.

We can expand the same concept of binary results to infectious diseases, too. When you use an antigen test to diagnose your illness, there are only two possible outcomes. You’re either “infected” or “not infected” with whatever disease the test is designed to detect. And antigen tests have utility in diagnosing a huge range of conditions, COVID-19 and pregnancy being only a few. Across all of these applications, the same immunological principles play out to rapidly give patients meaningful information about their health. But although their results are available quickly, antigen tests run a relatively high risk of being less accurate, and therefore less reliable, than PCR tests. Next, we’ll dive into the scientific strategies that distinguish PCR, its mechanisms, and its level of accuracy from those of antigen tests.

PCR: Worth the Wait

If you’ve got time to spare or need to confirm a questionable antigen test result, you might turn to PCR. Polymerase chain reaction is a modern molecular marvel that helps scientists make countless copies of a nucleic acid. But why does this matter? And what does it have to do with diagnosing COVID-19?

Our cells naturally replicate their nucleic acids, like DNA, during mitosis. This happens on a frequent basis, and without fanfare, so what’s the big deal about a scientist doing it? Well, an individual nucleic acid is hard to study. It’s microscopic, vulnerable to degradation, and fragile and unstable under the wrong environmental conditions. By making many copies of it, scientists can accumulate enough of it to study in a productive way.

Recall how, if you take an antigen test too early in your infection, it might not successfully detect viral antigens. That’s not the case for PCR, which can detect even a single copy of DNA. This quality is known as sensitivity. More sensitive tests are more effective at analyzing the analyte of interest (here, SARS-CoV-2 antigens). PCR is significantly more sensitive than antigen tests, so PCR is better at picking up on the antigen’s presence.

Sensitivity: Making Something from (Nearly) Nothing

As we’ll soon see, PCR tests involve a solution of primers, DNA polymerase, NTPs, the DNA sample you’re analyzing, and usually a buffer that maintains the right pH value for the reaction. This solution’s total volume is pretty small: about 50 μL (microliters), for example. Within it, the concentration of DNA sample might be super small! It’s not uncommon to start PCR with a very small sample of DNA, around 1 ng per μL. This makes PCR a great option for detecting viral particles early on in an infection, when their numbers are still scarce. Later on, we’ll see how this quality also makes PCR a forensic MVP and crime-solving hero!

But to reach that level of sensitivity, we have to use biochemistry to adjust that DNA sample. First of all, the DNA sample includes the DNA segment of interest, plus a whole lot of extra DNA. Working with the DNA’s biochemistry in mind, we can ignore the extra, and focus on only the segment we want. Also, through PCR’s three-stage cycle, we exponentially increase the amount of DNA present in the sample. After the final cycle, we have an abundance of DNA — millions of, if not over a billion, copies’ worth. Now, it’s much easier for scientists to analyze test results, compared to trying to crack a single, measly DNA molecule.

Together, these two principles of PCR allow scientists to isolate and then replicate a specific segment of DNA that interests them. In the context of diagnostic tests, that DNA segment is an identifying antigen within a virus’s or bacteria’s DNA sample. Out of the serpentine mazework of the full DNA molecule, we’ve managed to make many copies of a particular DNA segment of interest. That’s no easy feat; the patient’s sample, after all, isn’t pure. In addition to the DNA of interest, it may contain other viruses and bacteria (and their genetic material!), proteins, lipids, electrolytes, and more. Normally, these extra components muddle the picture and hinder scientists’ ability to focus on just the DNA sample. So, the fact that we can use PCR to see beyond those components is pretty phenomenal.

Remember, if a diagnostic test fails to detect an antigen when it’s actually present, that’s a false negative result. Highly sensitive tests, like PCR, have the big advantage of minimizing the frequency of false negative results. For the patient receiving a PCR test, that makes their results more reliable than if they took an antigen test. This is true among diagnostic tests for other diseases too, not just those for COVID-19. PCR is the heavyweight champion of lab diagnostics, and its applications don’t stop there. Let’s read on to learn more about how it works its molecular magic to replicate nucleic acids, and how that elevates the accuracy of a diagnosis.

How do PCR tests work?

PCR comes in multiple formats, depending on the test’s purpose and what kind of nucleic acid is under study. We’ll go into detail about those applications soon. For now, we’ll talk about PCR in the context of studying DNA.

First, let’s unpack what “polymerase chain reaction” actually means on a scientific level. When DNA replicates, it uses the DNA polymerase enzyme to add new nucleotides in the correct order when synthesizing a new DNA strand. Without the helping hand of this polymerase, the nucleic acid wouldn’t be able to make a copy of itself. “Polymerase chain reaction,” then, refers to the process of a DNA sample repeatedly undergoing reactions with polymerase. The outcome is many copies of the original DNA sample.

This chain reaction doesn’t occur naturally. It’s only possible within the hot confines of a piece of lab equipment called a thermal cycler. A thermal cycler generates high temperatures over and over again, in a repeating cycle, for a predetermined number of cycles. When inside the thermal cycler, the DNA sample’s structure changes in response to the heat. All the while, polymerase is hard at work replicating it.

It sounds like a simple recipe, right? Pre-heat thermal cycler to 200°F, combine one sample of DNA with one unit of polymerase, mix thoroughly, and bake for 30 cycles. Yields approximately one billion copies of DNA. PCR is actually a very intricate procedure, developed through decades of refining molecular cloning technology. Luckily, we can break down this complex process into three stages universal among all PCR tests, no matter their purpose.

Stage 1: Denaturation

Before we can make more copies of a DNA sample, we need to know the base pairs it consists of. That involves separating, or denaturing, its strands to expose its nucleotide sequence.

With its precise temperature control, the thermal cycler creates the perfect environment for DNA denaturation. Maintained at temperatures of about 95°C (203°F), the double-stranded DNA sample breaks apart into two single strands. Now, each strand of template DNA is easier to access, which is central to the success of the next stage.

Stage 2: Annealing

In the second phase of a PCR cycle, the temperature lowers to roughly 60°C (140°F), though the exact temperature may vary. By reducing the temperature, we make it possible for primers to bind to the single-stranded template DNA. This part is referred to as annealing.

Like the template DNA, primers are also single-stranded DNA, a type of oligonucleotide. As short sequences of DNA, their job is to physically tag the region of template DNA that will be copied. To accomplish this, scientists must customize the pair of primers to suit the template DNA. Each primer (one “forward” and one “reverse”) will locate the nucleotide sequence on either single-stranded template DNA that’s complementary to itself, and bind there. The annealing stage results in two single-stranded pieces of template DNA (still), now with a primer attached to each strand. Between the locations where each primer binds is the template DNA sequence to be copied.

PCR doesn’t make copies of the entire DNA strand, only a certain portion of it. Without primers, the incoming polymerase has no way of knowing which portion of the strand to replicate. Once the primers have bound to the template DNA, it becomes clear where replication should begin and end. Equipped with this information, we can now proceed to the next stage.

Stage 3: Extension

Finally, we can actually build the new copies of DNA. The thermal cycler raises the temperature again, this time to about 70°C (158°F). This slight increase in temperature will facilitate the construction of a new DNA sequence.

In saunters our main character, Taq polymerase, the polymerase enzyme that’s most common in PCR. Its quirky name alludes to the fact that it was originally isolated from the bacterium Thermus aquaticus. In nature, T. aquaticus is a happy inhabitant of hot springs and hydrothermal vents, so Taq polymerase is stable in high-heat conditions — making it ideal for use in PCR. This enzyme remains unbothered and structurally intact while the thermal cycler’s hot temperatures maintain the DNA as two separate strands. For comparison, human polymerase functions best at our body temperature of 37°C (98.6°F), so it wouldn’t perform as well in PCR as Taq polymerase does.

During the extension stage of PCR, Taq polymerase wastes no time constructing the copies of DNA. It begins by locating the primers, which have already bound to the template DNA. Starting at the primers’ binding site, it extends each primer by adding free-floating nucleotide triphosphates (NTPs) to the primer’s own DNA sequence. These NTPs are free-floating in the PCR solution, eagerly awaiting their chance to hop into place on the daughter strand. Extension continues until the entire DNA sequence of interest has been copied, forming a daughter strand.

To recap, DNA derived from the patient’s sample provides the template that instructs how the daughter strands should look (what their nucleotide sequences should be). Also present in the same solution are NTPs, which become the building blocks of the daughter strands, and Taq polymerase. Taq polymerase is what launches this reaction into action, turning these free-floating molecules into a cohesive DNA daughter strand.

After extension, the original one sample of template DNA has duplicated into two copies of itself (two daughter strands). If you begin with multiple samples of template DNA, which is sometimes the case, you’ll end up with many more copies of DNA after extension. In PCR, the act of copying DNA is amplification. Since a PCR test can succeed with even one double-stranded template DNA sample as the starting material, we can say that PCR is a very sensitive test: able to detect and work with even the smallest quantities of nucleic acid.

One cycle of PCR consists of these three stages: denaturation, annealing, and extension. This cycle gets repeated, usually 25 to 40 times, until enough copies of DNA have been made. By this point, from even one sample of double-stranded DNA, we might achieve millions or billions of copies of DNA. That’s because each individual cycle of PCR generates two copies of DNA from however many copies existed during the previous cycle. In other terms, the copies produced in each cycle of PCR are recycled as the template DNA in the next cycle. This leads to exponential doubling, creating innumerable copies of DNA by the final cycle — enough for a scientist to analyze, manipulate, or quantify.

Completing all 25 to 40 cycles of PCR requires up to a few hours. But realistically, once a patient submits their nasopharyngeal sample to the lab for a COVID-19 PCR test, it might take days to receive their test results. That’s because it takes additional time to ship and process the sample, prepare the assay, and compile results. It’s clear now why PCR is a less convenient diagnostic test than antigen tests, which deliver ready-to-interpret results in thirty minutes or less. Also, PCR always requires specialized lab equipment, unlike antigen tests, which are easy to perform at home. However, PCR makes up for this with its high sensitivity, which directly fuels its high accuracy as a diagnostic test. This makes all the difference during a pandemic, when you need to trust that your diagnostic test result is reliable so you can make appropriate decisions about your next steps.

How does PCR help diagnose COVID-19?

Our overview of PCR demonstrated how it detects, replicates, and analyzes DNA in a patient’s sample. That comes in handy when diagnosing a condition that involves DNA. For example, biomedical experts can use PCR to study mutations in oncogenes (genes that may contribute to cancer), or to assess the likelihood of parents passing down genetic diseases to their child.

A key prerequisite to any PCR diagnostic test is the nucleic acid segment of interest. Any given sample might contain loads of DNA, for instance, but we don’t want to amplify all of it. We only care about, and want to work with, a particular segment of DNA within that sample. To cherry-pick a specific nucleotide sequence out of billions of base pairs, scientists use primers that target that particular sequence. The primers align perfectly with that DNA segment, thanks to the primers’ and DNA’s nucleotide sequences being complementary to each other. Once the primers have located the nucleotide sequence of interest along the DNA strand, a polymerase can swoop in and begin building a daughter strand from it. This is why PCR uses so many cycles: this many “extension” processes are necessary to create enough DNA material in order to generate a diagnostic result.

If the primers are off by even one incorrect base pair, this alignment won’t work. The primers won’t be able to locate the correct DNA segment. This could have two disastrous outcomes on your PCR test. First, the primers might instead align with some other segment in the DNA sample, and that other segment will end up getting amplified. As a result, the nucleotide sequence that gets analyzed is completely irrelevant to the DNA segment of interest. You’d be unable to identify the virus’s presence based on this irrelevant DNA, so you can’t definitively say whether the test results are positive or negative.

Second, if there’s absolutely no DNA in the sample that happens to align with these incorrect primers, then the PCR test will simply have no meaningful diagnostic result. The primers were unable to align with any complementary DNA sequence, so the extension stage of the PCR cycle never happened. In other words, the primers never successfully gave the polymerase instructions about where on the DNA to start building a daughter strand. Therefore, no DNA (viral or otherwise) got replicated — that is, amplified — at all during the PCR test.

So, disease detection depends on amplifying DNA, and more importantly, the right segment of DNA from a larger sample. PCR can detect pathogens that use DNA, like herpesviruses and adenoviruses, but this is where we encounter a dilemma. SARS-CoV-2, the culprit behind COVID-19, doesn’t use DNA as its genetic material. Instead, it’s an RNA virus, with its viral genome consisting of RNA. While the general principles of PCR still hold true when diagnosing COVID-19 — we must perform repeated, three-stage cycles to determine if any viral material is present in a patient’s sample — we have to tweak the PCR process slightly to account for this.

Back to Basics: DNA, RNA, and Everything in Between

Rather than using traditional PCR to diagnose COVID-19, we use its cooler, younger sibling: reverse transcription polymerase chain reaction (RT-PCR). Most of the steps of PCR and RT-PCR are identical, but there are a couple important distinctions. First of all, these methods use different starting materials. PCR uses a DNA sample (template DNA) as its starting material, while RT-PCR uses RNA as its starting material. This makes RT-PCR a great choice in situations when we don’t even have one unit of DNA available to analyze. RT-PCR allows us to utilize the basic framework of PCR to analyze a nucleic acid, even if that nucleic acid isn’t DNA.

Second, before we carry out the three-stage cycle in RT-PCR, we must convert our RNA starting material to DNA. What makes this possible is the enzyme we call reverse transcriptase. According to the central dogma of biology, DNA converts into RNA through the transcription process. That won’t help us here, since we’re already starting with RNA. In order to convert RNA to DNA, so we can use that DNA as our template DNA in PCR, we have to make the RNA undergo reverse transcription.

During reverse transcription, the reverse transcriptase enzyme synthesizes a DNA molecule from the RNA starting material. This DNA molecule’s base pair sequence is complementary to that of the RNA, so we call it complementary DNA (cDNA). Now we have DNA! From here on out, the rest of the RT-PCR process closely resembles traditional PCR, except this cDNA molecule serves as the template DNA for our RT-PCR test.

Using RT-PCR as a COVID-19 Diagnostic Test

Since SARS-CoV-2 is an RNA virus, whose genome lacks DNA, we need to approach a COVID-19 diagnostic test using RT-PCR. Before beginning the test, we have a patient sample, but don’t know whether or not it contains any SARS-CoV-2 virus. We’ll treat the sample in such a way that will make SARS-CoV-2 material detectable to the RT-PCR test, assuming it’s present in the sample. If there’s no viral material in the sample, then no harm done; the RT-PCR test simply won’t detect any.

We begin by giving our sample the reagents it needs in order to undergo reverse transcription. This way, if there is SARS-CoV-2 RNA in the sample, it will convert into cDNA that we can use during RT-PCR. First, we have to isolate and extract viral RNA from the sample. We want RNA to be as pure as possible, free of substances like proteins that could interfere with PCR. Then, just like in traditional PCR, we introduce primers (made of DNA) that will bind to the viral RNA strand at a specific location. We also provide reverse transcriptase, which will meet the primers at their binding locations and synthesize a cDNA strand that’s complementary to the viral RNA. Remember, reverse transcription and cDNA synthesis are interchangeable terms for the same process: by reverse-transcribing an RNA sequence, we’re creating a new cDNA strand.

Ideally, we’ll select a reverse transcriptase that has enough thermal stability to withstand the repeated high-temperature stages of many RT-PCR cycles. In the lab, reverse transcription generally happens around 50°C (122°F). After reverse transcription, our cDNA molecule is initially single-stranded. However, single-stranded DNA is more unstable than double-stranded DNA because double-stranded DNA has hydrogen bonds between the base pairs of its two strands, which stabilize the entire molecule. During PCR, it’s easier to work with double-stranded DNA since it’s more stable. So, before we initiate our first RT-PCR cycle, we’ll use a DNA polymerase to turn our single-stranded cDNA into double-stranded DNA.

To recap, we started with a patient sample that theoretically contains viral RNA. We treated the sample so as to convert that viral RNA into single-stranded cDNA, then double-stranded DNA, which will be our template DNA. Now, we can amplify this template DNA via PCR’s characteristic stages of denaturation, annealing, and extension. If the sample did indeed contain viral RNA (a positive test result), we end up with many, detectable copies of the virus’s genetic material. If it didn’t contain viral RNA, the RT-PCR test won’t detect any genetic material, leading to a negative test result.

PCR and RT-PCR work well as diagnostic tests because their strengths compensate for the shortcomings of an antigen test. As we’ve described, PCR and all of its variations show much higher sensitivity than antigen tests do. Due to this sensitivity, PCR tests require very little DNA or RNA starting material. As such, a PCR test can be accurate early on in a patient’s infection. This early on in an infection, an antigen test might yield a false negative result because the viral load has not yet replicated to a detectable extent. By contrast, PCR and RT-PCR are the gold standard of diagnostic testing because their results are accurate the vast majority of the time. We’ll talk more about sensitivity and false negatives. For now, let’s explore how we can utilize PCR’s benefits in many areas of science and society, not just for disease diagnosis.

Additional Applications of PCR Tests

Within health care settings, PCR and RT-PCR are well-established as diagnostic tools. That being said, there are other contexts, even beyond the medical realm, where making many copies of genetic material can be very practical.

In forensic science, for example, DNA evidence found at a crime scene is almost irrefutable. But there might not be much DNA sample available at a given scene. Even if a sample is present, it may only exist in trace amounts. This is where PCR can solve an otherwise-unsolvable whodunit mystery! We can use PCR to amplify the DNA from a crime scene. After PCR, we now have enough material to easily compare it to a suspect’s DNA. Trace amounts of DNA wouldn’t let us do this process of pinpointing the crime’s perpetrator. Likewise, if there’s a need to identify human remains, we can amplify any DNA sources found with the remains, like blood or tissue. PCR enables us to draw evidence-based forensic conclusions, increase credibility in our criminal justice system, reduce the risk of wrongly convicting innocent suspects, and bring closure to a victim’s surviving loved ones.

Bridging Modern and Ancient Science

Through a similar strategy, we can conduct paternity testing using PCR. All it takes is amplifying the DNA of a child, their mother, and one or more men who could potentially be the child’s father. DNA samples, especially those found at a forensic scene, might be scarce. To make working with the DNA easier, a scientist must use PCR to amplify the DNA content in that sample. Then, when there are enough DNA fragments to tinker with, a scientist loads the fragments onto a gel. The gel spans an electric field: one end has a negative electrode and the other has a positive electrode. The scientist activates the electrical circuit, thereby subjecting the DNA fragments to an electric field.

A DNA molecule’s phosphate groups give it an overall negative charge, and like charges repel while opposite charges attract. As a result, the negatively-charged DNA fragments migrate through the gel, from the negatively-charged side to the positively-charged side. Built into the gel are pores that “capture” migrating fragments, depending on each fragment’s size. Larger DNA fragments get “captured” in the pores earlier in their migration, whereas smaller fragments face less resistance and can migrate farther toward the positive electrode. In other words, the distance a band migrates down the gel is inversely related to that DNA fragment’s size. Once captured, each fragment stops migrating and appears as a horizontal band. The end result is a series of horizontal bands at different points along the gel. Therefore, each band tells us something about the size of that particular DNA fragment, and how its size compares to that of the other bands.

This test, gel electrophoresis, lets a scientist visualize the DNA of each person (child, mother, and prospective fathers) as bands on the gel. The scientist can visually compare how far the child’s and mother’s bands migrated along the gel, compared to the father’s bands. Since a child receives DNA from both of their parents, we expect the child’s bands to resemble their mother’s and father’s bands equally. Ultimately, the man whose DNA banding pattern most closely resembles the child’s is most likely the father.

PCR can also unlock long-buried stories about the past and reconstruct historical family trees. Royal Egyptian mummies, for example, are so well-preserved that their DNA samples are often still intact even millennia later. When modern archaeologists uncover a tomb but don’t know the identity of the mummy, they can count on PCR results to help. At this point, we’ve analyzed the genetic profiles of enough royal mummies that, nowadays, predicting a newly-discovered mummy’s identity is relatively straightforward. Think of it as an extension of PCR-based paternity testing!

By now, you’ve learned a great deal about why PCR matters, both within and beyond disease diagnostics. You also understand PCR forwards and backwards! Let’s shift gears and look closer at how we translate diagnostic test results into meaningful, usable information for patients and HCPs.

Interpreting Diagnostic Test Results

For the sake of safeguarding public health, we need to have diagnostic tests whose accuracy and reliability we can trust. This was especially true during the COVID-19 pandemic, when new variants frequently arose, and the disease spread extremely fast during its most severe waves. In early January, 2022, as the Omicron variant swelled amid the winter holiday season, over one million Americans tested positive for COVID-19 in a single day. With cases surging higher than ever, diagnostic testing became more urgent than ever.

From an epidemiological point of view, the news that more than a million new cases were detected overnight was deeply concerning. Even more alarming, though, was the realization that there were almost certainly additional cases that had gone diagnosed. Unfortunately, not everyone exposed to COVID-19 will seek a diagnostic test, even if tests are readily available. Because of that, public health experts rely heavily on what data they do have: the results of tests that people did take.

When you (or your HCP, or a professional lab scientist) analyzes your sample during a diagnostic test, it’s important that the test presents your results in a way that’s easy to understand. To build people’s trust in public health, and in the science that sustains it, we have to know how to interpret it. If you can’t make sense of your test result, it’s impossible to know if the test is reliable at all, or to plan your next move.

Three core points guide how we use diagnostic test results to make sense of, and make decisions about, public health. First, when an infected person takes a diagnostic test, the test’s accuracy increases over time as their infection progresses. This is because viral load, and therefore the detectable substances the virus produces, increase as symptoms arise and intensify. We especially see this pattern of increasing accuracy for antigen tests, because PCR tests are sensitive enough to detect much smaller quantities of viral substances, so PCR tests easily detect an incipient infection — no need to wait several days for the viral load to increase.

Going along with that, if you have any doubt about the accuracy of your test result, you might choose to take another test a few days later. The idea is that, by this point, your viral load will have become more detectable (assuming you’re infected). Although the spread of COVID-19 can move fast, determining if you’re infected is, in some ways, a waiting game.

Finally, no matter which stage of infection you’re in — early, late, miserable, or anywhere in between — you can take action to get a result that’s as accurate as possible. If your test involves inserting a swab into your nasopharyngeal cavity, that’s an undeniably uncomfortable but immensely important step. Inserting the swab deep enough into your nose, for example, ensures that you’ll collect a sufficient amount of sample. If your test has other instructions, like immersing your sample in a buffer or waiting a certain amount of time between steps, follow those instructions carefully. Strong attention to detail here minimizes user error, a common cause of false results, and makes your test’s outcome more accurate.

In this section, we’ll see what happens scientifically to produce a positive or negative result in antigen tests and PCR tests. By gaining a clear understanding of the science involved, you’ll be able to translate that science into a practical plan the next time you take a diagnostic test.

Positive Test Results

A positive test result is anything but; it means that some amount of viral antigen is present in the user’s sample. How much, exactly, doesn’t necessarily matter. If you have any viral material in your sample, you’re infected. Though it’s possible to quantitate viral material via PCR’s other cooler, younger sibling quantitative PCR (qPCR), traditional PCR and antigen tests are qualitative. In other words, a positive COVID-19 test result will look very different depending on whether it’s an antigen test or an RT-PCR test.

Antigen tests, as we’ve witnessed, manifest a positive result in the form of two colored bands. As the sample moves along the test strip, antibodies on the strip bind any SARS-CoV-2 antigens that exist in the sample. This binding phenomenon generates a band at the test line in addition to the band at the control line. This is the most common format of a COVID-19 antigen test, where two visible bands implies a positive test result.

Meanwhile, an RT-PCR test is positive if any SARS-CoV-2 genetic material is detectable after many cycles. That material started as RNA, but has transformed into double-stranded DNA by the time RT-PCR begins, and remains in that format through all the cycles. Since PCR tests have a very high sensitivity, it doesn’t take much genetic material to prompt a positive result. In spite of this, RT-PCR doesn’t blindly detect any and all genetic material and interpret it as an infection. Remember that custom-made primers keep RT-PCR on track throughout all of those dizzying cycles by ensuring that only a specific portion of cDNA gets amplified. Only the cDNA that has any implication on a COVID-19 diagnosis will be taken into consideration for the test result. This leads us into an important discussion on specificity.

False Positives

Specificity is a test’s ability to correctly identify negative results. If a test is specific, it will ignore confounding variables that might muddle an otherwise negative result.

Of course, any diagnostic test should ideally be highly specific to the disease that it tests for. If you take a COVID-19 test, you want it to interpret only the presence of SARS-CoV-2 genetic material as an antigen, causing a positive result. Since your own genetic material is also present in the sample, you don’t want the test to accidentally interpret the presence of your own genetic material as an antigen. If it did misinterpret your own genetic material as an antigen, then the test wouldn’t be specific to SARS-CoV-2 antigens, now would it? You’d find yourself with a false positive result.

Put differently, highly specific diagnostic tests have a low rate of false positives. When they present a positive result, you can be pretty certain that it’s accurate. There’s still a small chance that your positive result is wrong, since no diagnostic test is 100% perfect, but this is unlikely. If you have any doubt, you can always take a second test to confirm.

False Positive Antigen Test Results

Both antigen tests and PCR tests are highly specific, although PCR might have the edge. For lateral flow antigen tests like those used to diagnose COVID-19, their specificity comes from the antibodies on the test line. Test manufacturers use antibodies that will bind with high affinity to, and only to, a SARS-CoV-2 antigen of interest. If your sample happens to contain antigens from any other infection you currently have, those antigens will simply skate on by the test line, without binding at all to its antibodies.

The only antigens that a COVID-19 antigen test is capable of detecting are SARS-CoV-2 antigens, so it’s highly specific to COVID-19. False positive COVID-19 antigen test results make up less than two percent of antigen test results, roughly. When they happen, they’re usually due to the user not following the test instructions properly. They might handle their sample wrong, use a test beyond its expiration date, or use a test after having stored it under the wrong environmental conditions.

The false positive rate of a diagnostic test is inversely related to its specificity. We can infer the specificity by subtracting the false positive rate from the number one, or 100%. For example, if we assume that this COVID-19 antigen test’s false positive rate is two percent, we can conclude that its specificity is 98%. An antigen test can be followed up with a more accurate PCR test, to confirm that the positive result is indeed true. If the subsequent PCR test result is negative, chances are that your initial antigen test was a false positive result.

False Positive PCR Test Results

RT-PCR doesn’t operate based on antibody-antigen binding, but this doesn’t mean that it’s not specific. In fact, PCR takes multiple precautions, built directly into its assay framework, to maximize its specificity and prevent false positives. First, as previously mentioned, the primers have a big role in maintaining specificity. The primers are specially designed to bind to specific locations on the cDNA strand. If they’re designed incorrectly, they might bind elsewhere, which would amplify the wrong DNA sequence.

Another factor influencing PCR’s specificity is the temperature during the annealing stage. Recall that the annealing stage happens at a lower temperature than the denaturation and extension stages. Scientists carefully select the exact temperature during annealing to be conducive to the primer’s binding with the template DNA. If the annealing temperature is off by even a few degrees, the primers might not bind to the template DNA as well as they should. This could lead to nonspecific binding of the primers to other sites on the DNA strand instead. That causes other, unrelated DNA sequences to get amplified instead of the target sequence.

As long as the scientist optimizes the reagents and conditions to suit the DNA sequence of interest, which is characteristic of the disease being detected, the PCR test will be specific to that disease. This is an element that antigen tests don’t need to take into consideration, but one that helps make PCR as specific as it can be. In turn, this means that PCR test results are as accurate as possible. Their false positive rate is similar to, if not a bit lower than, that of COVID-19 antigen tests. This also implies that its specificity is comparable to, or a bit higher than, that of an antigen test.

When it comes to negative and false negative results, we lean on the metric of test sensitivity. We’ve covered sensitivity in some detail already, but next we’ll see how it works in tandem with specificity to make a diagnostic test the best it can be.

Negative Test Results

When people get a negative test result, it’s usually an invitation for relief. Maybe they’ve been anxiously fretting over their health ever since being exposed days earlier. Maybe they live with an immunocompromised roommate and want to avoid bringing any contagious cooties and critters home to that person. Or, maybe proof of a negative test result is the very last thing standing between them and their dream vacation.

While a positive antigen test depicts two colored bands on the test strip, a negative result displays only one. This band appears at the control line, whereas there should be nothing at the test line — not even the faintest trace of color. A band occurs at the control line because the control line’s immunoglobulins bind to the antibody-nanoparticle complex, which developed as the analyte progressed down the test strip. If the analyte has no detectable SARS-CoV-2 antigens, then there are none to interact with the test line’s SARS-CoV-2–specific antibodies. No colored band will form at the test line, only at the control line, yielding a negative result. With just a glance, you know whether or not the test detected COVID-19 in your sample.

For RT-PCR tests, viral RNA is detectable only if its cDNA is present in the sample and gets amplified sufficiently. There are a few possible causes of a false RT-PCR result. First, and hopefully this is the case, there’s no viral RNA in the sample that you provided. Even if you treated this sample with the chemical reagents and heating conditions that would normally amplify cDNA, you’d end up with no detectable cDNA because you started with no viral RNA. A second possibility is non-specific amplification. Your selected primers must precisely match the cDNA segment that complements the viral RNA of interest. If the primers don’t match it, then they might accidentally bind to another cDNA segment instead. Then, the cDNA that ends up getting amplified is actually unrelated to the viral RNA of interest.

For the most part, if you get a negative test result, you can breathe that sigh of relief. But it’s important to stay vigilant and seek a confirmatory test if appropriate. Though false negative results are unlikely, a negative result isn’t always an all-clear, as we’ll discuss next.

False Negatives

At face value, a negative test result sounds reassuring. But to comprehend this result, we have to read a bit deeper. A negative test result doesn’t necessarily mean that you’re not infected with the disease in question. All it means, in the strictest sense, is that the test didn’t detect the disease’s antigens in your sample.

False negatives happen when a test fails to detect antigens in a person who’s actually infected. Overlooking a patient’s positive case is never a good thing, as it can harm the timeliness and quality of treatment. In the dire circumstances of a pandemic, like COVID-19, failing to detect a real case can have extra dire circumstances. It’s easiest to manage, control, and contain the spread of an infection when you know that the infection exists.

Thanks to good old science, missing a positive case is unlikely, because most antigen and PCR tests exhibit strong sensitivity. The higher a test’s sensitivity, the more accurate results it can yield even with few antigens in the sample. PCR has consistently high sensitivity levels, able to make use of even a single copy of DNA. On the other hand, the sensitivity of an antigen test varies. Let’s look at why.

False Negative Antigen Test Results

How can a test fail to detect something that’s actually there? When it comes to antigen tests, sometimes it’s a simple reason, like the user isn’t swabbing their nasopharynx correctly. The most common cause behind a false negative result, though, is an infected person taking the test too soon.

Viral load (the amount of virus in your system) increases over the first few days after the onset of a COVID-19 infection. As viral load increases, so does the amount of viral antigens. Antigen tests look for these SARS-CoV-2 antigens, so they become more abundant and easier to detect a few days after the infection actually starts. When symptoms are in full force, that’s a sign that viral load is higher than before symptoms developed, and it’s a good time to take an antigen test.

Why does this matter? Folks are most infectious when their viral load is high — that is, when they’re actively symptomatic. It’s especially important to catch a COVID-19 infection when the patient is infectious, so action can be taken to isolate them, connect them with treatment, and start contact tracing. Infectiousness is a huge factor in disease transmission, so intervening when patients are most infectious is a way to prevent the spread.

Since antigen tests become more adept at picking up COVID-19 mid-infection, their false negative rate lowers with time. Very shortly after infection, the false negative rate may be as high as 60%, perhaps because antigen levels might not be high enough for the test to identify. As the days pass, symptoms gradually arise and evolve as viral load reaches a detectable level. Now, positive results become more likely, even if the patient’s previous tests were (presumably) false. Maybe we ought to call false negatives “delayed positives” instead.

This is all because antigen tests have a relatively low sensitivity (compared to PCR). PCR doesn’t experience the issue of navigating viral load quandaries; it can effortlessly detect virus-derived cDNA, even with very little available in the sample. High sensitivity reduces the risk of false negatives because a sensitive test readily detects small amounts of the virus. Conversely, high specificity reduces the risk of false positives by lowering the likelihood that the test will misinterpret a totally unrelated antigen as the virus’s antigen.

Antigen tests and PCR tests are both high-specificity, but sensitivity is where antigen tests fall somewhat short. Nonetheless, antigen tests are pretty good at finding positive cases when they exist. If you find yourself holding a double-lined antigen test, you can take that as a strong indication of infection. However, if you doubt your antigen test’s negative result, and you have the means to obtain a PCR test, you may want to do so. All things considered, PCR tests are more reliable overall than antigen tests due to their superior sensitivity. However, as we’ll see shortly, they aren’t infallible.

False Negative PCR Test Results

Ideally, we want our lab diagnostics to be both highly sensitive and highly specific. A test that’s only sensitive, and not specific, risks detecting the wrong antigens and diagnosing the wrong disease. On the flip side, a specific-but-not-sensitive test fails to pick up on the pathogen it’s so carefully crafted to detect.

Both qualities combined, then, contribute to a test’s accuracy and reliability. Antigen tests work fine in general, but with PCR, we get the best of both. One caveat: the best isn’t perfect.

False negative test results occur when an RT-PCR test doesn’t detect cDNA that’s been reverse-transcribed from viral RNA. Like with antigen tests, this is often a game of timing, but for different reasons. As we just covered, antigen tests might yield false negative results when people take them too soon. The same holds true for PCR tests, though it’s less of a concern because PCR can detect much lower viral content. Instead, timing matters in PCR tests because it often takes over a day to receive results. Consequently, when the patient finally receives their results, they might not truly reflect the infection as it is today.

That’s particularly dangerous if the result is a false negative. Let’s say a person is feeling fine now, but as a precaution, they submit their sample for a PCR test directly after having a COVID-19 exposure. It takes a few days for a scientist to analyze the sample and summarize the test result. During that time, the patient’s viral load is increasing. They finally receive their test result, and it’s negative. But by now, their infection has taken hold, and they might have transmitted it to others while awaiting the test result.

During the COVID-19 incubation period, the span of days between a patient getting infected and first experiencing symptoms, their increasing viral load influences the false negative result rate. Positive test results are most likely to be accurate once symptoms are in full swing (and that goes for antigen tests too!). During the incubation period preceding symptom onset, false negative results might happen if the viral load hasn’t reached detectable levels yet. Viral load increases over several days, making a true positive result more likely and a false negative less likely.

Special Considerations in PCR Diagnostic Testing

Obviously, we want our sample to be a solid indication of what’s actually happening in our body. The slow turnaround time of PCR testing is just one factor that can hinder that. Collecting the patient’s sample is another potential source of inaccuracy. HCPs, not patients, are usually the ones who collect the patient’s PCR sample. That’s unlike the situation for antigen tests, which are user-friendly for the typical nonprofessional and can be self-administered. When an HCP is responsible for sample collection, human error is less likely. (Partly because HCPs are better trained in sample collection technique than patients are, and partly because it’s tricky to perform a nasopharyngeal swab on yourself!) If the sample was collected improperly in the first place, the sample might lack viral antigens even if those antigens are actually present in the body.

Cross-reaction between different viruses (and their antigens) is rare, but when it happens, it confounds the test result. For instance, it might simultaneously generate a false negative reading for SARS-CoV-2, and a false-positive reading for another virus. We want to trust our RT-PCR test to amplify the right cDNA segment, from the right RNA segment, from the right virus. By its very nature, PCR makes good use of its specificity by keeping false negative results in check.

Although we’ve talked primarily about viruses, remember that antigens (like those detected by lab diagnostics) can come in other forms. Bacterial components, toxins, and allergens can also be antigens in these tests; theoretically, PCR can detect anything possessing nucleic acids. These all-encompassing test tactics have advanced modern medicine to new heights, but they’re not all-powerful. In the next section, we’ll discuss how diagnostics fit into the bigger picture of public health. We’ll explore how they’ve reshaped our health care approaches, and how our health care industry is transforming to make them stronger.

Challenges in Lab Diagnostics

Diagnostic tests have reshaped the ways HCPs approach patient care and public health personnel approach disease. Unfortunately, as we’ve briefly touched upon, this doesn’t come without its share of challenges. Now, we’ll revisit them in the context of the COVID-19 pandemic.

Result Underreporting and Health Inequity

In addition to pitfalls in their design that could generate false positives or false negatives, diagnostic tests’ results depend partly on how people take them. For one thing, we’ve already noted that taking an antigen test too early after COVID-19 exposure could make a misleading negative result. But an even bigger issue than how someone takes a test is whether someone takes a test at all.

Throughout the pandemic, diagnostic tools have helped epidemiologists and HCPs gauge how many cases there are at a given time, but this was only ever an estimate. Despite our best public health efforts, the reality is that not everybody who’s been exposed to or infected with COVID-19 will get tested. Some individuals simply don’t want to take a test, while others simply can’t. Especially on a global scale, access to diagnostic resources differs widely. Across countries with varying income levels, the extent of available resources, pandemic-prepared HCPs, and laboratory infrastructure varies too.

This raises the bigger question of how to confront health inequity. Health equity is the pursuit of fair access to health resources and care, regardless of someone’s background or circumstances. It involves addressing health disparities among patients, closing preventable gaps in our health care systems, and giving everyone an equal opportunity to obtain high-quality care. Health inequity, then, refers to the forces that hinder this pursuit.

Due to underreporting, where not everyone who ought to take a test actually takes one, the number of diagnosed positive cases is never a true figure. Instead, public health experts use the number of diagnosed cases to extrapolate how many cases there probably are. The truth is, although we’ve heavily discussed antigen tests and PCR tests in this article, there are other methods that can indirectly serve as diagnostic tests.

One example of this is wastewater surveillance. Through wastewater surveillance, health authorities collect and test wastewater samples to determine the levels of a pathogen that are present. This happened during COVID-19, where authorities monitored SARS-CoV-2 viral levels in local wastewater. Infected people (whether they know they’re infected or not) may shed viral particles in their waste, and it ends up in wastewater. By conducting wastewater surveillance, health authorities can track the geographic spread of diseases like COVID-19 and draw conclusions about how many people in a community may be infected.

Let’s circle back now to that phrase, “whether they know they’re infected or not.” When it comes to COVID-19, specifically, this invites a big dilemma. In some people, COVID-19 infections can happen without manifesting any physical symptoms. If someone has no symptoms, they have less of a motive to seek a test. Even among people who have symptoms, they might not report those symptoms or seek medical care, leaving their HCP unaware of their infection. More concerning yet is that asymptomatic folks who aren’t aware that they’re infected likely aren’t taking precautions to protect others from catching their infection.

To account for this scenario, a diagnostic test should be performed after a suspected exposure, even if the exposed person hasn’t manifested any symptoms. Timing, and the type of test, matter here. Antigen tests, as we know, might miss an infection if taken too soon after exposure. The user (assuming they’re infected) needs to give the virus time to replicate in their body, increasing their viral load high enough to be detectable. PCR tests would be able to detect a positive case sooner, since they require very little antigens in order to yield accurate results. Perhaps that’s the best quality of our well-rounded lab diagnostics: PCR tests generally excel where antigen tests fall short, and vice versa.

Keeping up with the Koronaviruses

As our understanding of COVID-19’s symptoms and spread evolved, our understanding of the virus itself evolved too. We learned that, although SARS-CoV-2’s infamous spike proteins (S proteins) have mutated over time, leading to the advent of new variants, other parts of the virus have not. As waves of new variants washed over the world, so did the worry that diagnostic tests might struggle to detect these changed strains.

As long as a diagnostic test targets a viral feature (antigen) that stays relatively constant, the test will still be able to detect new variants. When it comes to SARS-CoV-2, the S protein would not be a good antigen to use for this purpose. That’s because, across individual variants, the S protein’s appearance can fluctuate wildly. To keep up with newborn variants, to help maintain the sensitivity of our diagnostic tests, and to prevent false negative results, our tests should instead look for a more stable antigen.

The good news is that SARS-CoV-2 particles have other proteins besides the S protein. Most COVID-19 diagnostic tests are designed to detect its nucleocapsid proteins (N proteins), which contain the virus’s RNA. N proteins tend to stay consistent, even as the virus mutates, because they’re crucial to the virus’s structural integrity. They protect its precious genetic material from damage, making it possible for the virus to infect the host with its viral genome. If the N protein were to change, the virus’s approach to infecting a host would also have to fundamentally change. Hence, the N protein’s design is fairly universal across different variants — making it an ideal antigen to target via diagnostic testing.

Meanwhile, the molecular mechanisms of a PCR test can be tailored to detect multiple regions of the viral genome. Remember, a PCR test will evaluate whatever cDNA segment gets tagged with primers. Amplifying, and therefore detecting, a different segment of cDNA begins with redesigning the primers you’re using. This helps mitigate the risk of the PCR test overlooking a positive case, because the test can be adjusted slightly to amplify a different part of the cDNA. Another way of looking at it is, even if the virus’s RNA (and therefore the cDNA that we use in PCR) changes due to mutations, the PCR test can still detect the virus based on other parts of its genome.

Interestingly, targeting a stable antigen like the N protein is not the goal of our COVID-19 vaccines. Vaccines we’ve developed mainly target the virus’s S protein, because it’s how viral particles attach to and penetrate host cells. When a new SARS-CoV-2 variant arises, its S protein is often one of its features that has changed. Think of the S protein as a disguise that the viral particles wear — new variants need to keep changing their disguise so they can evade antibodies from vaccines and from past infections. Each vaccine we’ve made suits a particular “edition” of the S protein more than others, so as the variants’ S proteins mutate, we need to update our vaccines to accommodate this rotating lineup.

In the next article, we’ll talk in more depth about COVID-19 vaccines. We’ll also discuss how their innovative mRNA technology, which pinpoints SARS-CoV-2’s spike protein with laser focus, flipped the pandemic completely upside down and brought it under control. In combination with each other, vaccines and diagnostic tests are two of our most powerful public health tools. Used together, they help us attack a disease outbreak from all angles: vaccines from a preventive perspective, and diagnostic tests from a treatment standpoint.

Conclusion

Diagnostic testing represents yet another special strategy that guides HCPs as they solve the puzzle of what’s afflicting their patient. Two common diagnostic test types, rapid antigen tests and PCR tests, served as reliable tools in managing the COVID-19 pandemic. Despite their undeniable utility, however, diagnostic tests have some consequential pitfalls: their inability to keep up with viral evolution, the risk of a false positive or false negative result, and the fact that they simply can’t quantify the exact number of infected people because not everyone who’s infected seeks a test. Once a patient knows their status, they can collaborate with their HCP to craft a treatment plan that addresses the root cause of their condition, meets their medical needs, manages their symptoms effectively, and prevents the infection from spreading to others.