The 12-month learning curve has been agonizingly steep, but it has had its rewards. Researchers have gleaned new insights about how to manage the disease and its complications—knowledge that lays the groundwork for understanding and treating infections caused by other novel viruses.
In March 2020, the World Health Organization declared SARS-CoV-2 a global pandemic. On the eve of its one-year anniversary, two Harvard Medical School COVID-19 experts—Ingrid Bassett and Lindsey Baden—pondered some of the most valuable lessons learned.
Baden is an associate professor of medicine at Harvard Medical School and director of clinical research in the Division of Infectious Diseases at Brigham and Women’s Hospital. Bassett is an associate professor of medicine at HMS and an infectious diseases expert at Massachusetts General Hospital.
They are also members of the Harvard Medical School-led Massachusetts Consortium on Pathogen Readiness (MassCPR), a multi-institutional, cross-disciplinary, international research effort established on March 2, 2020, at Harvard Medical School to help combat the current pandemic and set the stage for fighting future ones.
Together with colleagues Karen Jacobson and Rajesh Gandhi, Baden and Bassett lead MassCPR’s Clinical Disease Management and Outcomes group, charged with understanding the disease caused by SARS-CoV-2, identifying optimal therapies to treat and manage its complications, and developing ways to improve patient outcomes.
Baden and Bassett outlined the following key lessons from Year One of the pandemic:
A Dr. Jekyll/Mr. Hyde of diseases
One of the truisms in the study of infectious diseases is that the same pathogen can affect individuals differently—a variability predicated on individual host factors, including genetic predispositions, differences in immune response, underlying conditions, and overall health.
But COVID-19 has taken this notion to a new level.
Beyond the well-defined differences in risk from age and underlying disease, there lies yet-unexplained variability. A small portion of young, otherwise healthy people can experience rapid progression to severe disease, while some nonagenarians and centenarians have survived COVID-19. Age remains the greatest predictor of disease severity.
“What is super tricky about this bug is that there are early pre-symptomatic and asymptomatic phases,” Baden said. “You can be infected for a week and not know it.”
Pre-symptomatic and asymptomatic transmission emerged as one of the key differences between SARS-CoV-2 and its older cousin SARS-CoV-1, the virus that caused the SARS outbreak of 2003, Baden said.
“With SARS-1, an infected individual sheds virus several days into symptomatic illness,” Baden said. “With SARS-CoV-2, most people shed without even knowing they are infected.”
While this feature may not be entirely unique to SARS-CoV-2, Baden said, it presents a major obstacle to halting the spread of the virus. It also proved to be a blind spot early in the pandemic when it was still unclear whether individuals with asymptomatic infections could transmit the pathogen.
But the unpredictability of COVID-19 goes beyond these person-to-person variations in disease presentation. What makes COVID-19 particularly challenging is its ability to turn on a dime within the same individual.
The disease could start as a mild, even asymptomatic, infection that may swiftly pivot into severe disease that affects a number of different organs. Making things trickier yet, the presence and severity of symptoms early in the disease do not neatly track with disease severity later on. People who have mild initial symptoms may swiftly progress to severe disease. Conversely, those with intense initial presentation may be spared complications.
“The variability of presentation is incredibly vexing, the timing of illness is incredibly vexing, and not knowing where in the disease someone is is also incredibly vexing,” Baden said.
Right patient, right time, right drug
When an infection causes illness serious enough to land someone in the hospital, the single most important pearl of clinical wisdom gleaned over the past year may be this: Treat the right patient with the right drug at the right time.
The axiom of individualized treatment is hardly new. While it also applies to many other diseases, it may be especially relevant in COVID-19 for several reasons.
COVID-19 is a shapeshifting disease that goes through an acute and an inflammatory phase, each demanding a different treatment.
Timing is critical in treatment choice, but staging infection is not always straightforward because of the pre-symptomatic phase and the great variability in presentation.
“The vast majority of people do not have very many symptoms early on, so when you come to the hospital with progressing disease, we do not know whether you’ve been infected for three days or 13 days,” Baden said.
During the acute phase—typically the first seven days of infection—the virus enters the body and starts cranking out copies of itself. As the virus begins to spread, the immune system kicks into high gear to combat the invader. Viral replication slows down and eventually stops. The disease transitions to its next phase, defined by immune-fueled inflammation. The most serious complications of COVID-19 tend to occur precisely during this second phase, largely as a result of an overly exuberant immune response that may damage various organs and tissues. Which organ gets damaged in the crossfire between host and pathogen varies from person to person. For one it could be the lungs, for another the kidneys or liver, and for a third the heart or the brain.
The cardinal features of serious COVID-19 are: Respiratory illness marked by low oxygen levels due to lung damage, endothelial dysfunction—damage to the blood vessels that promotes clotting—and immune-driven complications that could affect any number of organs.
The acute phase requires therapies that target viral replication or antibody-based treatments that block the virus’s entry into cells. This is the window of opportunity to halt progression and prevent complications that might otherwise occur during the subsequent inflammatory phase. Once the disease reaches the inflammatory stage, the treatment needs to target the sources of inflammation and tame the overactive immune response.
No two patients with COVID-19 are alike. The virus replicates at different speed and for different lengths of time in different people, and it triggers different immune responses in each person. For example, it is now known that patients with compromised immunity have longer periods of viral replication, which could extend beyond the seventh day of infection and could lead to more severe disease. But these variations do not always align with the presence of underlying disorders.
The individual variability in immune response even among those without underlying conditions remains a black box in understanding disease progression, Bassett said.
One question that researchers will be looking into is whether there may be various immune profiles among infected patients, depending on which arm of the immune system and which immune pathway is abnormally activated.
An evolving arsenal of therapies
COVID-19 treatments fall in three broad categories: those targeting the virus, those targeting the systemic inflammation and those aimed at organ-specific dysfunctions. This latter group of supportive treatments includes breathing-support, such as noninvasive supplemental oxygen to intubation for those with critical illness.
Treatments for the acute phase include antiviral drugs, which work by disrupting viral replication, and antibody-based therapies, which work by blocking viral entry into cells.
Antiviral and antibody-based treatments work best when delivered in the early days of infection—preferably within five or seven days of infection. One way to determine who might benefit from these treatments is to check a patient’s blood for antibodies. If a patient is already developing antibodies, it means their immune system has marshalled defenses. Giving antivirals or antibody treatments to these patients is much less likely to offer much benefit, Bassett and Baden said.
“I would use the presence of antibodies as a better marker of when you were infected than when you had your first snivel,” Baden said. “To me, antibodies are the best clock for timing infection.”
Antivirals: A scarcity of options
Despite early hopes that several drugs could interfere with viral replication, thus far only one medication—the Ebola drug remdesivir—has shown therapeutic benefit in high-quality randomized clinical trials. It works by interfering with the virus’s ability to make protein and replicate its genome. Despite this, multiple studies have converged on a verdict that for most patients, remdesivir doesn’t appear to decrease mortality.
The drug does, however, shorten hospital stay, which is an important therapeutic benefit: Shorter hospitalization means more beds for sick patients—a critical consideration in the midst of a raging pandemic. The absence of clear survival benefit from remdesivir underscores the need to identify new therapies that curtail disease severity and lead to better survival, Bassett said, as well as therapies that prevent hospitalization in the first place.
Emerging evidence suggests that for some hospitalized patients—those with serious enough illness to require supplemental oxygen but not intubation—targeting both the virus and the resulting inflammation may be beneficial. A new study demonstrated that combining remdesivir with the anti-inflammatory drug baricitinib sped up recovery time and resulted in improved clinical status.
Antibody treatments: A Goldilocks challenge
SARS-CoV-2 enters human cells using its spike protein—the primary target of current antibody therapies. Antibodies latch onto a part of the spike known as the receptor-binding domain and prevent the virus from entering cells. Antibody therapies include plasma from recovered individuals and lab-grown monoclonal antibodies.
Convalescent plasma is not new. Used for more than 100 years, it involves the infusion of antibody-rich plasma from survivors into ill patients. A recent study showed that plasma appeared to halt disease progression among elderly patients at high risk for severe disease, but only if administered within three days of symptom onset. And not all convalescent plasma is the same when it comes to COVID-19. Studies have demonstrated that plasma from COVID-19 survivors has benefit in some patients but not in others. One reason for this variability could be that people donating plasma may have different levels of antibodies to the virus. Plasma with high titers of antibodies is likely to be more beneficial.
The utility of convalescent plasma is also limited by logistics. The process of collecting it and infusing it is too cumbersome and time-consuming to deploy at scale, particularly amid a pandemic.
These days, scientists can circumvent the limitations of plasma by making antibodies in the lab. To do this they use human B cells—the antibody factories of our immune system—and clone them to produce antibodies.
So far, the FDA has issued emergency-use approval for two pharmaceutically made antibodies: bamlanivimab (Eli Lilly) and casirivimab+imdevimab (Regeneron). The authorizations are for outpatient treatment of people at risk for severe disease. Clinical trials have demonstrated that monoclonal antibodies, when administered early in an infection, reduce viral load, can help patients get rid of the virus faster, and may mitigate symptoms.
Recently, Eli Lilly released preliminary data suggesting that giving its monoclonal antibody to nursing home patients helped reduce infections among other residents and staff. But researchers caution that these results have yet to be verified through independent peer view.
But antibody-based therapies are no panacea, Baden said.
Two factors that limit the utility of monoclonal antibodies. First, they appear to have the most benefit for a narrow patient population: those early in their infection and at risk for severe disease but not yet hospitalized. Second, there are logistical hurdles. The treatments are authorized for IV infusion at outpatient facilities, but there are currently not many sites that can handle antibody infusions. To help alleviate this problem, the U.S. Department of Health and Human Services recently launched a website that will help patients find infusion sites near them.
In sum, Baden said, antibody-based treatments present a Goldilocks challenge.
“I need to give you the therapy at the time this therapy is likely to work,” Baden said. “For plasma, this means we need to make sure it has a higher titer of antibodies, not just ‘You said you had COVID, I’ll collect your blood and give it to someone else.’”
As new SARS-CoV-2 variants emerge, it’s unclear whether the virus might eventually develop mutations that would allow it dodge lab-grown or convalescent antibodies.
Again, Baden said, this speaks to the need for more science to inform the best decisions as the pathogen changes.
“Our countermeasures are growing but we need to continue with the science into the pathobiology of the disease and when in that cycle each intervention may or may not work,” Baden said.
Treating the inflammatory phase: A need for refinement
Treatments for this stage involve the use of inflammation-taming therapies that target the aberrant immune response that sets in after the acute phase and that can lead to immune-mediated organ damage.
The exuberant immune response is marked by elevations in many inflammatory markers, Baden said. An important gap in the current knowledge is which inflammatory pathways drive which part of the disease. Without this piece of the puzzle physicians cannot “toggle” the activity of individual immune pathways, while sparing others.
Drugs that target one such inflammatory pathway offer a telling example.
IL-6 inhibitors block the activity of an inflammatory chemical called interleukin-6 (IL-6). Inflammatory chemicals, or cytokines, such as IL-6 have been of longstanding interest as possible drivers of inflammation-fueled damage in a number of disorders.
In the case of COVID-19, cytokines can induce lung tissue damage and have been linked to increased mortality. Yet, there is lingering uncertainty about IL-6’s role in COVID-19. Scientists are still trying to understand whether the elevation in IL-6 that occurs early in an infection actually fuels tissue damage or whether it’s a bystander inflammatory marker. Studies of one IL-6 blocker, tocilizumab, have yielded messy and confusing results. The bottom line thus far appears to be that while IL-6 blockers may slow progression to end-stage disease, they do not appear to improve overall survival. Moreover, they seem to increase the risk of other infections in some patients.
In contrast to the narrowly targeted IL-6 blockers, dexamethasone is a broadly acting steroid that dampens overall immunity. The drug appears to reduce risk of death among patients with severe respiratory illness—those on mechanical ventilation or noninvasive breathing support. However, the treatment carries the traditional risks of corticosteroids, which dampen immune system protection across the board.
“Rather than targeting the immunopathology with a scalpel and surgical precision, it covers the whole waterfront of inflammation,” Baden said.
The drug’s broad immune-dampening activity is why it should be used with extreme caution in patients with COVID-19 in whom the immune system is already compromised, Baden said.
In a recent report, Baden and colleague Jonathan Li described the case of an immunosuppressed patient with COVID-19 who was receiving both antiviral therapy and immunosuppressive drugs. The patient harbored SARS-CoV-2 for months. Sequential samples showed that the virus evolved over time with changes observed in the spike protein, an observation that underscores the virus’s adaptability to immune pressure. The worry is, Baden said, that patients who cannot mount an immune response robust enough to clear the virus could become a source of drug-resistant virus in the community.
“If you are a sick person in the hospital with low oxygen levels, heading to the ICU, you get dexamethasone. Period,” Baden said. “The one patient group I hesitate in are severely immunosuppressed patients in whom the drug may help with the inflammation, but if they can’t clear the virus, do we create more problems?”
Other problems with the use of corticosteroids that have been well established include elevated blood sugar and avascular necrosis—bone death due to lack of blood supply. Thus, treatment with these drugs requires careful observation for these side effects.
Supportive care generally entails treatments that target specific organs affected by COVID-19. These therapies can include breathing support—from noninvasive low-flow oxygen supplementation for those with moderate illness to ventilators for those with the most severe disease. Generally, such therapies are neither new nor specific to COVID-19. They are aimed at treating the injured organ, such as, for example, the use of dialysis to support damaged kidneys in patients with COVID-19.
An important cornerstone of supportive care in COVID-19 is addressing the elevated risk for blood clotting, either with medications that prevent clot formation in the first place or therapies that dissolve clots when they do form.
In COVID-19, blood clots arise in two ways: from the damaged lining of the blood vessels, which is a target of the virus, and from the activation of clotting factors. People with COVID-19 lung disease often develop micro-clots in the lungs, but clots can also travel to the heart, brain, or kidneys.
Current recommendations on who should receive anticoagulation therapy continue to evolve as new data emerge.
People with mild to moderate disease who are at home should not receive preventive anticlotting treatment. However, those with serious illness who have established risk for clots—those with clotting disorders or history of clots—may benefit from anticlotting medications. Those with severe enough disease to require hospitalization should receive a prophylactic dose of anticlotting therapy. Beyond that, physicians should individualize the decision on anticlotting treatment based on several factors, including levels of certain biomarkers that portend aberrant clotting, Baden said.
The road ahead
The amount of knowledge about COVID-19 generated in a year is without precedent in the history of medicine. Yet, as is the case with any complex mystery, the more one learns, the more questions crop up, and the thicker the plot becomes.
Some of the lingering challenges in treating COVID-19 include:
- Developing better ways to profile risk for severe disease that go beyond age and underlying diseases
- Developing more accurate ways to determine disease stage to better inform timing and type of therapy
- Identifying treatments that prevent hospitalizations in the first place, rather than therapies that merely speed up recovery for those hospitalized
- Optimizing the design and delivery of antibody-based therapies, including the design of lab-made antibodies that target multiple parts of the virus, which will be particularly important as the virus continues to mutate
- Developing therapies that don’t rely solely on antibodies but also boost cellular immunity, another critical form of protection that stems from the body’s T cells, whose role in COVID-19 is not yet well understood
- Developing refined, precision-targeted anti-inflammatory therapies that do not dampen overall immunity but modify specific inflammatory pathways
- Generating better ways to understand the drivers of disparate outcomes among patients of different ethnic and racial backgrounds and tailoring timing and use of therapies accordingly
A telling data point on the drivers of disparate outcomes, Bassett said, was the striking differences in hospitalizations at Mass General.
“One of the things that really blew my socks off was that, normally, in a given year, at MGH we have about 8 percent of inpatients discharged from the hospital who identify as Hispanic. During that first COVID-19 surge, 35 percent of the people hospitalized at MGH with COVID identified as Hispanic,” said Bassett, who led a study detailing these differences.
And the age difference between hospitalized white and self-identified Hispanic patients was also staggering—Hispanic patients were, on average, 20 years younger, she added. Yet, she noted, while Hispanic patients had similar rates of ICU admission or death to those of white patients, Hispanic patients tended to recover faster.
Teasing out the role of factors that magnify risk—from comorbidities to social and economic variables—will occupy public health experts’ attention for years to come.
Another gap in the current knowledge is what happens in the post-recovery phase, Bassett said, including the short- and long-term perturbations that arise from COVID-19. The subpopulation of COVID-19 survivors with lingering symptoms that have come to be known as “long-haulers” represent a physiological enigma.
Understanding the long-term effects of the disease will likely come with time, as the world moves past the acute phase of what is, within human history, a still-young disease.
Other vexing questions: How emerging viral mutations and new strains might affect our ability to prevent transmission and treat disease, and which mutations will matter and which will not.
“When it comes to mutations, we don’t want to be Chicken Little running after every new mutant, but we also don’t want to be an ostrich with its head in the sand,” Baden said.
Lessons beyond the clinic
The one question that looms the largest for Bassett and Baden is this: Knowing what we know now, what should we do differently for the next pandemic?
“When we are done with this, when we’re on the other side of it, which I am hopeful we’d get to with vaccines, how do we prepare for the next one?” Baden said. “What investments should we make? I think the investments need to be in the science capacity to position ourselves to respond to pathogen X and how do we do that even faster next time around?”
Bassett says that during this interpandemic period, it would be important to take stock of successes, failures, and missed opportunities to prepare for the next pandemic.
One early success came from the nimble pivot to adapt existing clinical tools to study COVID-19, such as the development of a rapidly searchable database based on MassGeneral Brigham’s electronic health records. The tool enabled researchers to rapidly query the avalanche of medical records in near-real time, allowing them to quickly identify patterns of disease among hospitalized patients—who is being admitted, with what symptoms, and who is dying.
“I think that was an innovation and a major step forward in terms of being able to answer questions on a very short time horizon about what’s happening now,” Bassett said.
Another success—and a silver lining of the pandemic—for Bassett has been the never-before-seen level of collaboration across researchers and institutions.
“The sense of urgency and mission caused people to cooperate in a way that I’ve really never seen before in my academic career,” Bassett said. “Academics have a certain streak of competition because we’re applying for the same grants and trying to publish papers in the same journals. It was very refreshing for all of us, all over the city, to come together to solve this one problem. We shared our best practices openly right from the beginning.”
This new collaborative infrastructure will continue to yield benefits as various institutions converge to study the long-term outcomes in COVID-19 survivors.
“For example, if we’re looking at rare outcomes or looking for specific subpopulations where any one academic center may not have enough people, but if we have a multi-institutional cohort, we could answer questions much more robustly,” Bassett said.
But for Bassett, there is also a “never again” element in preparing for future outbreaks.
“When I think back to the beginning, the fact that people were being sent home with symptoms and told ‘Go and spend two weeks at home and stop working and self-quarantine,’ without actually knowing how many people were infected, or that states were forced to bid against one another for personal protective equipment…Putting them in conflict should not happen again,” Bassett said.
Avoiding these failures requires a centralized, federally managed response, the very thing that was missing for most of the current pandemic, with lethal consequences.
The most important lesson then may be this: Biomedical discovery should go hand in hand with enhancing public health infrastructure and other systems for spreading the results of new scientific knowledge for maximum impact.
“I am an implementation science researcher,” Bassett said. “For me, it’s all about how do we close the gap between what we know is effective and works in the lab and in clinical trials and how this knowledge gets delivered into the real world.”