Uncontrolled spread, p.15
Uncontrolled Spread, page 15
The CDC’s single biggest breakdown in its response to COVID was the abortive rollout of its COVID test. In their defense, officials at the CDC pointed out to me that developing tests for commercial and academic labs was never their job. It wasn’t what they had ever done before. Their experience was in working directly with state public health labs. They had done it successfully in 2009 in response to H1N1 swine flu. But once the agency ran into challenges with their development of that initial test for SARS-CoV-2, for weeks after, there still wasn’t another test being pursued in the US other than the effort that was under way at the CDC. Seeing the CDC’s challenges, the leadership at HHS needed to mount a concerted effort to develop an alternative, pulling together other public health agencies with relevant tools such as FDA, NIH, and the Biomedical Advanced Research and Development Authority. There was no plan B. It was almost as if once the CDC tried—and failed—to develop a test for COVID, the focus shifted exclusively to remedying the CDC test even though the agency was never meant to, and never would be able to, supply the entire market with testing. When Secretary Azar asked the CDC why the agency wasn’t shipping its test kits to private hospitals, CDC officials had to tell him that the agency never provided test kits to the private sector, only to state labs. But in this case, the CDC was expected, for the first time, to supply test kits directly to commercial and academic labs.
Into the late spring, CDC officials still found themselves explaining to the leadership at HHS that it wasn’t the CDC’s historical role to work with commercial and academic labs, or to help industry develop test kits. Every year the agency develops a test kit for flu, but it doesn’t send this kit to hospitals or commercial labs—only to state public health labs. Instead, the needs of hospitals and private labs for flu tests are met by commercial manufacturers.
We had repeated openings on testing; chances to start turning things around before the crisis spiraled out of control. One such opportunity came on March 4, when the entire lab industry gathered in Washington, DC, for the annual meeting of the American Clinical Laboratory Association. The trade association conference straddled a turning point when the pandemic was becoming firmly rooted, and there was a palpable sense that the US would be next to fall to COVID. Now, in the ballroom of the Grand Hyatt Washington were the chief executives of all of the major commercial labs. It was probably the last conference that these CEOs would attend in person for more than a year. Standing behind a glass podium and flanked by four potted trees, Dr. Stephen Hahn of the FDA would deliver a twenty-minute speech of which he would devote a mere five hundred words to the pandemic. He reminded the lab executives of their regulatory obligations if they wanted to develop tests for COVID and mentioned some accommodations that the FDA had mapped out to ease that process. It was largely a standard recitation of what was already on the FDA website. It was a missed opportunity. By that point, we needed a stronger call to action.59
I saw the annual meeting of the lab industry as a critical turning point. It was an opportunity to rally the commercial labs and test makers around the effort. The lack of a clear call to action was emblematic of a larger organizational problem. No single agency or person really owned the problem. And HHS leaders never convened the department’s operating divisions, including the FDA, the NIH, and the CDC, into an organized effort to tackle the key challenges we faced.
There was plenty of precedent for the FDA commissioner stepping in to quarterback an industrywide response in a moment of public health crisis. One came in 2004, when I was working as a senior advisor to FDA commissioner Dr. Mark McClellan.
There was pressure on the US government to allow the intellectual property protecting drugs for the treatment of HIV/AIDS, which had been developed by western manufacturers, to be appropriated by generic drug companies, mostly located in India. These generic firms had promised to expand treatment options in Africa by manufacturing less costly versions of the pills.
The entire public health community was committed to increasing the number of HIV patients in Africa who could be offered treatment. But some of the knockoff drugs were of suspect quality, and those of us working at the FDA at that time worried about the prospect of using American funds to buy millions of doses of what, we feared, could be inferior drugs.60 We believed that all patients deserved the same high-quality treatment options available to Americans.
So we led an effort to bring the drug industry together with the US government, whereby drug makers would allow their patented medicines to be developed into cheaper combination pills that would be easier to distribute and use in the more austere settings found in many public health clinics in Africa. The FDA, in turn, would put these new products through a process where the combination medicines could be reviewed; if they met the agency’s standards for approval, they would receive “tentative” FDA approval. Since the drugs still had patents blocking their commercialization in the US, they couldn’t win full FDA approval. But this provisional status would be sufficient to ensure the high quality of these copies and would in turn pave the way for the US government to legally purchase and distribute the drugs in Africa.
As part of this effort, the FDA agreed to publish a guidance that would outline all the circumstances where the FDA believed there was already sufficient clinical data to support the use of different drugs in combination. Any drug maker who could successfully combine different medicines to make one of these combination drugs could apply for tentative FDA approval just by demonstrating that the new combination pill was chemically the same as each of its individual components and would deliver the same amount of medicine. Drug makers wouldn’t need to also undertake the time and cost of developing clinical data to prove that these combinations were effective, since the FDA had already specified that in advance.
There were other aspects to our overall plan to promote greater access to HIV/AIDS drugs in Africa. As another part of the effort, I asked three drug makers to join a hastily arranged meeting at the FDA, where we sought the cooperation of Merck, Gilead, and Bristol Meyers Squibb to pool the intellectual property they each owned for the components of what we believed could be an ideal once-daily pill for the treatment of HIV. We promised the companies an efficient development and review process if they would join together to develop the pill, and they agreed. The drug they would make was Atripla, one of the most important HIV medications to be launched in the last twenty years.
Taken together, the process we set in motion became a key part of the President’s Emergency Plan for AIDS Relief or PEPFAR, and I believe that combined FDA effort helped save millions of lives. PEPFAR was a singular public health achievement.
Sixteen years later, with COVID spreading across America, I urged White House officials to meet with the lab industry executives while they were in town. Dr. Deborah Birx, who served as the White House coronavirus response coordinator, was urging the president and vice president to do the same thing, and Vice President Pence did meet with the chief executives of the major lab companies, though Hahn did not attend.61 The very morning of Hahn’s speech, I did what I could to encourage more concerted, urgent action, especially with the entire lab industry assembled in Washington at the moment that the epidemic was about to explode. So, following a television appearance on CNBC, where I had already raised the issue of getting more COVID tests into the marketplace, I sent a message on Twitter from the backseat of a car on the West Side Highway in New York. I wanted to make the stakes abundantly clear:
“On #coronavirus testing: The nation’s big clinical labs meet in Washington today for a convention and their CEOs are in town meeting with federal elected leaders. This is their moment. They ought to step up. They may be judged by what they now do.”62
Chapter 8
Not Enough Tests and Not Enough Labs
There were two main technologies available to test for SARS-CoV-2: antigen-based tests (which measure whether you have whole particles of live virus in your body, called a virion) and molecular tests that rely on PCR (and measure the presence of the virus’s RNA). Each has pros and cons. Besides their lower cost and greater convenience, antigen-based tests can offer some clinical advantages over molecular tests, which are otherwise considered the gold standard for diagnosing infection.
Once you’ve been infected for eight or ten days, you might not have live virus in your body anymore. The immune system has already attacked and destroyed whatever virus you had. What’s left, at that point, is often dead virus, RNA fragments, that can no longer grow in viral cultures, or infect someone, even though you continue to shed these viral particles.1
Sometimes the PCR tests can still detect genetic fragments, and return a positive result, even though you no longer have active virus in your system, and you’re therefore no longer contagious. By contrast, antigen-based tests are likely to return a positive result only when there’s whole, live virus still in your respiratory secretions. The general criticism of antigen tests is that they may not have the same sensitivity as PCR, meaning they can miss positive cases. But a good antigen test can have high specificity, meaning that when it detects virus particles it’s more likely to be picking up live SARS-CoV-2, and you’re more likely to have an active infection.
The PCR tests sometimes try to correct for this challenge by quantifying the amount of RNA found in a sample. When the RNA level is very low, it may be an indication that the person is recovering from the infection and is no longer shedding live virus. Therefore, the patient is no longer contagious.2 Or it may be because their infection is early, and the virus hasn’t had time to replicate in high numbers and reach its full saturation. The antigen tests can take longer to develop and may not be available at the outset of a crisis involving a novel pathogen. But the tests can often be produced more cheaply and in larger quantities than molecular tests like PCR tests. The trade-off is that antigen tests are generally less sensitive. They may be more likely to miss positive cases. The key is fitting the right test to the right public health purpose.
What you need is an all-of-the-above approach that sets in motion a deliberate effort to build the four layers of testing needed in a crisis. First are laboratory developed tests that can be deployed quickly to allow you to get initial tests into the field. Next are manufactured test kits that are based on PCR and can be run on high-throughput testing instruments and can allow high-volume testing at clinical laboratories. Third are point of care molecular tests that many doctors already have in their offices and can be updated to test for SARS-CoV-2. Finally, inexpensive antigen tests can also be used at the point of care or deployed to schools and workplaces, and eventually can be used as at-home tests. These are the four layers of a resilient system. A plan to develop each of these testing platforms should have been under way from the start. The part of the system that got short shrift for a long time was the non-PCR tests. We rely far too much on lab testing in this country. Government payers like Medicare favor lab-based tests over point-of-care diagnostics because they’re accustomed to paying for services. This has created a healthcare system that is biased against tests that can be put into the hands of consumers, which is precisely what’s needed in a pandemic.
The first antigen test directed against COVID was a lateral flow test developed by Quidel and authorized by the FDA on May 9.3 Similar to the way a home pregnancy test works, a lateral flow test for a virus uses antibodies bound to a nitrocellulose (paper) membrane. The antibodies are designed to bind to specific regions on a particular virus.4 When a sample is added (in this case, the swab from someone’s nose) the sample will flow along the test paper and then onto an absorbent pad where the antibodies are bound. If the virus is present in the sample, the antibodies will bind to the viral target, and together, the antibodies and the bound virus particles will continue to move along the test strip until the bound complex reaches a readable line. Other lateral flow tests would later be authorized by the FDA, including one by Abbott called BinaxNOW that’s about the size of a credit card and initially sold for five dollars.5 Upon its approval on August 26, 2020, it was made available in massive quantities owing to investments Abbott had made in manufacturing. A patient can swab himself or herself, add drops of some reagent fluid on the paper from a small bottle, and then fold the special paper over the nasal swab. A readable result appears in about fifteen minutes.6 This test was adopted widely in the fall of 2020 and was also being used to screen people entering the White House. By April 2021, it was finally sold directly to consumers, without a doctor’s prescription, in a pack of two tests for about twenty-five dollars.7
Each type of diagnostic test has an important role when we take the time to understand the context in which each can be optimally used. A comprehensive approach to pandemic planning will take into consideration the different kinds of tests, how they should be best used, and when they can be made available after a new pathogen emerges. We didn’t have a deliberate strategy to deploy these different layers of testing. Every company stepped into this market at different times, with little meaningful coordination. A real pandemic playbook would have mapped out these decisions in advance and seeded the development of these different testing layers.
Some government entity or official should have been charged with advancing these different options in a coordinated fashion. We should have prioritized the development of the point-of-care tests early on, and supported their production with US government funding, and guaranteed purchasing and reimbursement. Such an effort would have needed to be led by the secretary of Health and Human Services and would have required collaboration across the CDC, the FDA, and other federal agencies like BARDA and the Centers for Medicare & Medicaid Services. No one agency, or public health official, owned this challenge. We needed an all-of-theabove approach to get the four layers of testing into the market. We needed a coordinated response that deployed the right test to the right patient at the right time.
When antigen tests finally reached the market in large quantities, the federal government initially cornered the entire supply, buying more than 150 million of the BinaxNOW tests by September (spending $760 million) with no detailed plan on how to use them.8 They shipped most of these tests into nursing homes, following a strategy to focus resources on preventing infection in the vulnerable elderly. But nursing homes found the tests a poor fit for their needs.
Many nursing homes worried that the antigen-based tests were not sensitive enough to screen an asymptomatic population and the tests would miss positive cases—a problem in a nursing home, where there are many vulnerable residents. If nursing homes used the tests to screen their staff, and then missed workers who might have asymptomatic infection and be able to spread the virus, the consequences could be grave. It was well established that many outbreaks in nursing homes were triggered by infected staff members who didn’t know they were shedding the virus.
State and local health departments expressed similar concerns. A testing approach in these facilities needs to be as airtight as possible, since the risk of a single introduction of virus can be catastrophic. Thus many of the antigen tests went unused. A federal survey found that about 30 percent of thirteen thousand facilities that were provided the rapid tests hadn’t used the equipment. During a period of time when regulations from the Medicare program required nursing homes to do routine testing if there was an outbreak in adjacent communities, hundreds of facilities didn’t use the antigen tests they were given.9 They stayed with PCR.
Because of the antigen tests’ potentially lower sensitivity, the FDA initially authorized them to be used only to diagnose SARS-CoV-2 infection in people already showing symptoms, because that’s what the data submitted to the FDA supported at the time. Individuals who are already symptomatic generally have a higher viral load in their secretions and are more likely to register a true positive result on these tests. By contrast, patients who aren’t yet showing symptoms, or may never show symptoms, may have less virus in their secretions, so they may be more likely to test negative even though they are still harboring the infection.10
Despite these limitations, a lot of people still sought to use the antigen tests to screen largely asymptomatic populations at nursing homes, including some of the federal officials in charge of expanding the national supply of tests. Manufacturers weren’t seeking emergency use authorizations for screening in asymptomatic populations, so the FDA didn’t have a lot of data to support the reliability of the antigen tests in these settings.11 The companies making these tests were worried that their technologies wouldn’t perform as well under these circumstances. They didn’t want to generate a whole bunch of data showing that the tests weren’t that good at detecting the virus in people who weren’t showing symptoms. In many cases, they didn’t pursue these trials.
