I Used to Report on Clinical Trials, Now I’m In One; Here’s Why I Volunteered

Mike Huckman

I am not brave.

Once, I passed out after a routine blood draw for a diagnostic test. Ever since that first scary and embarrassing episode at a LabCorp location in Manhattan several years ago, the act of giving a blood sample, no matter how small, has been a source of anxiety.

Will I break out in a cold sweat? Feel lightheaded? Collapse on the floor?

Fearing a repeat performance, I get myself tied up in knots. It also should be evident by now that I’m not fond of needles. I’m 58 years old, but I still look away like a squeamish kid whenever I get a flu shot or some other routine vaccination. Just make it quick and tell me when it’s over.

All of the above makes me an unlikely candidate for a COVID-19 vaccine clinical trial. But a few months ago, when I read in my local newspaper that vaccine developers Pfizer and Moderna needed more volunteers, I decided to participate—or at least see if I could enroll in one of the studies.

My first call happened to be to a Wilmington, North Carolina clinic working on the Pfizer test. It is here that I want to make a disclosure. Pfizer is a client of W2O Group, the communications agency where I work. One has nothing to do with the other. I volunteered like the tens of thousands of others who want to support the global vaccine research and development effort.

The coordinator at the local study site called me back pretty quickly. How about 7 am the very next day? I was in.

First shot.

There was some trepidation. More than once, I considered backing out.

But I kept the appointment. I walked in the door at the clinic, and signed the required consent agreement. I sat down, took a deep breath, and looked away when the nurse swabbed my arm with antiseptic before taking a routine blood draw.

Everything was OK, so far. Next, they stuck the swab up my nose to collect a sample to look for signs of the SARS-CoV-2 virus.

Then, I waited. About 45 minutes later — time for the experimental vaccine to thaw out from deep freeze — I got either the experimental vaccine or the saltwater dummy vaccine. A nurse came in to give the mystery injection.

After 30 minutes of observation by a staffer, seeing no serious adverse effects, I was sent home.

Three weeks later, I returned on schedule for the second injection. Same deal as before. One week after that, they asked that I return for a follow-up blood draw.

Second shot.

Participants in this clinical trial are not told about the blood sample test results, in keeping with the stringent study protocol. I have not gotten an antibody test on my own since receiving the second shot, and still don’t know if I have developed neutralizing antibodies to SARS-CoV-2.

Here’s what I do know. Several hours after receiving each injection, I experienced what I would refer to as a hot flash lasting a few minutes or so. There was a touch of nausea, too—but only after the first shot.

Both sensations passed and never came back. As Pfizer seeks an Emergency Use Authorization from the FDA, who got what remains a secret to those of us in this more than 43,000-person study. All we know is that half of the participants received the real vaccine and half were given the placebo.

Did I experience a so-called placebo effect where people who receive the dummy version of a treatment or vaccine trick themselves into thinking they’re on the real thing or did I have a reaction to the genuine vaccine, one that might even suggest my immune system was being effectively primed against the virus?

I am not a doctor or a vaccine expert, and I will not hazard a guess. I have to wait for Pfizer to “unblind” the study to find out for sure.

Every Friday since receiving the first shot, I also have to go into a clinical trial app and input whether I have or had any COVID-19 symptoms. Thankfully my answer, so far, has been no.

After each clinic visit, I get $119 loaded onto a special debit card the office provides. I am not doing this for the pocket money. I am doing it because I have trust and faith in science and in the scientists—the people—who prove or disprove that science.

For nearly 10 years, I covered the biopharmaceutical industry as an on-air reporter at the business cable network, CNBC. I regularly interviewed scientists, researchers, therapeutics developers, and corporate executives.

For the past 10 years, since transitioning to become a communications consultant, I have worked alongside dozens of biopharmaceutical company scientists, academic researchers and clinicians, helping them talk about their passion, ingenuity, and data. 

They are the most intelligent, disciplined, rigorous, uncompromising, fact-driven people I have ever met.

If I received the vaccine, I’ll have some peace of mind. But I will continue to wear a mask, wash my hands, and maintain physical distance at least until the public health experts say we can get back to normal.

If I was injected with the saline solution, I will wait my turn to receive a vaccine, if or when it’s approved by the FDA. Since I’m not in a high-risk group, an essential worker or on the front lines, that could be a while.

People tell me on social media that I am brave and courageous for volunteering to be in a COVID-19 vaccine study. I am neither. I’m just a needle-phobic guy who wants to play a very small role in getting us out of this hell. Each of us has a role to play, large or small.

From 2000-2010, Mike Huckman was CNBC’s pharmaceuticals reporter.  He is currently the Global Practice Leader, Executive Communications at W2O Group.


Getting Married in 2020. One of Many Difficult Family Decisions

Carolyn Ng, managing director, Vertex Ventures HC

So there it was, hanging in the closet.

My carefully chosen bridal gown, with its lace in a delicate shade of ivory. Beside it was my bridal veil, featherlight, translucent, and freshly purchased just the day before the wedding.

When Anna, my girlfriend, gingerly pinned it on my head, I heard the words of my mother sounding in my head. She had once told me, sweet and soft, “Don’t forget your veil. There is nothing more bridal and more beautiful than a bridal veil.”

My heart sank for a moment. Because I know my mother would have given anything to see me walk down the aisle as a bride, veil and all.

Yet I had to tell her, no, mum, Matt and I are getting married, but you will not be able to attend in person. We won’t allow you, or other family members to attend.

It was a difficult conversation, and 2020 was full of all the most difficult conversations, for us as individuals, as a family, as a country.

As the years pass, and as Matt and I grow old together, what would we want our wedding day to be remembered for? I want it to be centered on us and our love. But for it to be about us and our love, it has to, by my definition, naturally include all the important people around us in our lives. Friends and family who have shaped us, challenged us, cared for us, and molded Matt and I to be who we are today, choosing each other as life partners.

There is no celebration of “us,” if it does not include “them.”

And that is why this COVID-19 pandemic has taken such a toll on us all. It is not only the headlines-grabbing devastation that has robbed the US of more than 250,000 lives and over 20 million jobs. What is also draining and painful is the multitude of mundane daily decisions we have all been deliberating over and over in the past nine months, calculating the risks of a deadly virus exposure against the very human need for togetherness.

After all, it is this need for social interactions and physical closeness that define us all as human beings.

Weddings present a special conundrum in the pandemic. We got engaged in November 2019, and were originally planning to be married in July of 2020.

How did we change plans? Well, one thing for sure, we certainly did NOT want our parents and aunts and uncles and any of our lovely guests catching the SARS-CoV-2 virus from our wedding. Given our professions — we are both healthcare venture capitalists — we have front row access to scientific and clinical knowledge pertaining to the SARS-CoV-2 virus and what it is doing to public health and the economy.

They say with knowledge comes power. And I would add, the power to act responsibly.

So, some tears were shed. Some hearts broken.

But eventually all of our parents and family members agreed to stay in the safety of their respective homes and not fly to attend our wedding in person. Instead of the original 150-person guest list, we had a couple of our dearest local friends witness (and operate the Zoom for) our wedding, which they did with aplomb. Our wedding was held outdoors in a beautiful private garden, and we also made sure that all of us involved had a negative COVID-19 test within that week of our wedding. Face masks were given to our in-person guests as wedding favors.

It was not completely risk-free. We recognized that. But we could conscientiously say that utmost effort was taken to reduce COVID-19 risk to as low as possible, while allowing for a figment of magic to still be conjured for our special day.

I would like to think there was still magic in the air, and that all of our Zoom wedding guests could feel it. It was as magical as a video technology like Zoom could allow you to feel. We had a couple of technical glitches of course, as any respectable wedding would have. But save for a few hangovers, we were immensely fortunate that our beautiful wedding concluded with no real sequelae. I mean the COVID-19 types.

Not every wedding couple has been as lucky as we were. Some took bigger risks and faced dire and deadly consequences. Barely two weeks after our nuptials, I read with a heavy heart an article authored by the CDC on the wedding in August which turned into a nightmarish super-spreader event.

In that case, the couple from California held a 55-person indoor wedding in rural Maine. It was reported that almost none of the guests wore masks indoors, nor did they maintain physical distancing of at least 6 feet apart. At least 177 infection cases were linked to this single event, including outbreaks at a long-term care facility and a correctional facility, which were 100 and 200 miles away from the wedding venue!

Tragically, seven people died from COVID-19 as a result of this wedding.

And even more tragically, perhaps, none of those who died were even in attendance. They were all secondary and tertiary infection cases. The virus has won the R0 game.

To have your wedding be the cause of seven innocent people’s deaths? I cannot imagine the pain and anguish of the newlywed couple. That burden of guilt and despair that will forever lace the memory of the day they tied the knot. An event of negligence, a lifetime of regrets.

Regrets. We desperately try to avoid them. Every critical decision, step, action that we make in our lives, we are propelled to take the ones that will render the highest probability of “no regrets”.

In a pandemic, these decisions that would give us “no regrets” become way more difficult to make.

Pre-COVID-19, we had been talking fondly, for the longest time, of a visit to Kentucky. There, we would visit Matt’s grandmother. He calls her “Mommom” in the most endearing way, and I hear lovely stories of how he grew up spending the most wonderful time with her when she was still living in Philadelphia.

And then the pandemic hit and turned the world upside down, killing people and infecting many more. We put our Kentucky trip plans on ice. We could visit her in 2021, we said. Vaccines would be coming and we would be visiting her safely, we said. We would also do a bourbon tasting tour, we said.

Mommom will love you so much when she meets you, Matt said.

Mommom, 91, passed away this past weekend.

We never made it out to Kentucky. Matt did not get to say his good-bye to his closest grandmother. I never got to meet this incredible woman who was part of what made Matt who he is today.

Regrets and also sorrow. How desperately we avoid them. Our scientist/doctor selves keep reminding us that it was a prudent decision we made not flying over this year and introducing additional virus risk to Mommom who was obviously in a high-risk age group. But nothing can stop our human hearts from bleeding.

We can’t help but ask ourselves, “Could we have done this?” or “Could we have done that?” to see her one last time?

We felt the tug again to see family this Thanksgiving. But seeing the surge to more than 180,000 cases a day in the US, we decided to cancel our flights from San Francisco to Philadelphia. It is cold this time of the year in Philly, rendering outdoor dining impractical and social distancing difficult. Flying also introduces additional risks to our parents and we are not confident how we can mitigate them without quarantines and testing. In short, we will not be tasting Matt’s mum’s famous 25-pound turkey this year.

So, with Thanksgiving and Christmas around the corner, how would you like this year’s special family holidays to be remembered?

There are trade-offs in every arrangement we make in the midst of a global pandemic. But may we choose actions that we may live with for years to come even post-pandemic, conscientiously, responsibly and lovingly. And without regrets, if we are so blessed.

Carolyn Ng and Matt McAviney at their wedding. (Credit: iPhone photography by Kevin Dai, Vivo Capital)


Written, in loving tribute, to my husband Matt McAviney’s grandmother “Mommom”, who left us this past weekend peacefully. I wish I have had the chance to get to know you.


Medicines Based on Unusual Genetic Traits: Andrew Farnum on The Long Run

Today’s guest on The Long Run is Andrew Farnum.

Andrew is the CEO of Seattle-based Variant Bio.

Andrew Farnum, CEO, Variant Bio

Variant Bio is a startup seeking to discover new drugs, by finding gene variants in rare ethnic groups. It’s especially interested in what can be learned by sequencing exceptional groups of people in countries where there hasn’t been much sequencing.

This is a company seeking to bring the benefits of an exciting new technology to a wider group of traditionally underserved people. If Variant is successful, it could help improve health and wellbeing in these rare ethnic groups, as well as wider groups of people all around the world.

Andrew comes to this challenge after spending 9 years working on the $2 billion strategic investment fund at the Bill & Melinda Gates Foundation. That gave him an up-close look at some of the most exciting technologies, the best biomedical entrepreneurs, and the classic challenges of how to broadly improve health outcomes in poor countries.

He has thought long and hard about how to build trust with many different players to execute on a vision like the one at Variant Bio.

Please join me and Andrew Farnum on The Long Run.


A New Model for Vaccine Communications Grounded in Science and Empathy

Mike Kuczkowski, founder and CEO, Orangefiery

With COVID-19 cases, hospitalizations and deaths surging, the impressive vaccine results from Pfizer/BioNTech, Moderna and now AstraZeneca arrive just in time to provide some needed hope.

But for these vaccines to bring the pandemic to an end, enough people need to be willing to take them. That’s not a given.

Various polls have told a story this year about a rising tide of hesitancy. Polls in September showed a slim majority of Americans – 51 percent in one survey – are willing to take COVID-19 vaccines. More recent surveys show a slight uptick in confidence.

About 65 to 70 percent of people will need to either survive COVID-19 infections or get vaccinated in order for us to achieve herd immunity, according to the World Health Organization. In the U.S., that’s 182 million adults. (Based on 70 percent of 260 million US adults, according to US Census Bureau.)

While vaccines for infectious diseases such as measles, polio, tetanus and smallpox are widely regarded as the greatest public health interventions in history after clean drinking water, global vaccination rates have been flattening or declining in many parts of the world for the past two decades.

Vaccine hesitancy is a complex phenomenon. It is rooted in social and cultural issues like parental authority, historical experiences, religious views and confidence in various information sources. Anyone can be vaccine hesitant — there’s not a clear correlation between levels of education, wealth or access to information and vaccine hesitancy.

The traditional public health communications response in the face of these trends has been to speak louder and more emphatically about the importance of vaccination and to highlight the risks of vaccine refusal on communities. It hasn’t worked.

Heidi Larson, director, Vaccine Confidence Project

That’s because the conversation about vaccines isn’t really about vaccines. Writing about cases where communities have refused vaccination, Heidi J. Larson, Ph.D., director of the Vaccine Confidence Project in London and author of Stuck: How Vaccine Rumors Start—and Why They Don’t Go Away says:

“Vaccine revolts unleash underlying sentiments about personal and collective histories, relationships with government, big business and international bodies.”

Vaccine hesitancy, it seems, is about power, fear and trust. Overlay these trends on our pandemic moment – with declining trust in our public health institutions, rising misinformation, a polarized media and social media environment –and it’s a recipe for resistance.

So how do we respond? I’ve seen people ask who this generation’s Elvis Presley is, suggesting that a well-lit photo of the right celebrity will turn the tide of public opinion. There’s certainly a role for celebrity influencers, but if vaccine hesitancy is rooted in issues of power and control, something more will be needed. COVID-19 presents an opportunity to do something dramatically different in vaccine communications, an approach that abandons a paternalistic tone in favor of a shared decision-making model and potentially reframes the problem of vaccine hesitancy.

There are some efforts like this underway. In July, an interdisciplinary working group convened by the Johns Hopkins Center for Health Security produced a report urging the groups involved in vaccine development to take a ‘design thinking’ approach that put the public at the center of the effort. The recommendations emphasized the need to understand public expectations, earn public confidence in the fairness and even-handedness of vaccine allotment and availability, make vaccines available in familiar settings and communicate in meaningful, relevant and personal terms.

We can do this. But it means rethinking how we talk about vaccines and how we engage people around them. From a positioning standpoint, vaccination messaging has to be grounded in the science, but it has to focus on things that are more personal and meaningful than efficacy and safety information. Think about common everyday concerns — jobs, schools, restaurant dining, travel — that will be enabled by a successful vaccination program.

It also requires a dynamic set of outreach and communications strategies that will be rooted not in selling but in education and engagement.

Here are a few examples of things various stakeholders should consider doing:

  • Vaccine manufacturers should produce robust analyses of their data in peer-reviewed publications and ensure wide distribution of their findings to key opinion leaders (KOLs), relevant medical societies, allied health professionals and patients, the latter using patient-accessible language
  • The federal government should lead a strong vaccine distribution effort, prioritizing the geographies and communities where the vaccine can be most effective in mitigating the spread of COVID-19 and protecting frontline workers and vulnerable populations
  • State and local governments, in coordination with federal authorities, should create awareness campaigns that position vaccines broadly as an offer, not a mandate, but that emphasize their value in ending the pandemic
  • Insurers and policymakers should set minimal copayments or waive cost-sharing provisions for vaccines to ensure broad access, particularly for the most vulnerable people
  • Healthcare professionals should counsel people with patience, recognizing that in many cases the concerns people express in rational terms may often reflect emotional issues such as fear, vulnerability and powerlessness
  • Advocacy groups should create tailored communications for their audiences, particularly patients with health conditions that put them at higher risk for severe COVID-19
  • Employers, who have found themselves in the unaccustomed position of communicating about health-related topics to employees and customers, should embrace an appropriate role in vaccine communications, educating leadership teams on vaccine data and recommending vaccination to employees, particularly those in roles that place them at risk of infection
  • Nonprofits that serve at-risk communities should redouble outreach efforts to reach those whom the pandemic has shown are so clearly disproportionately impacted by COVID-19
  • Vaccine experts, public health experts and epidemiologists in academia and government should serve as trusted sources to media outlets and other communities (including social media) to address misinformation and disinformation in clear but compassionate terms
  • Media outlets should write extensively about vaccine issues — the good and if necessary, the bad — and aspire to provide a fully accurate picture of the risks and benefits, knowing that in past vaccine experience there has been an overemphasis on stories that lacked a scientific basis in fact
  • Government should make an effort to engage and solicit involvement from influencers, even non-scientific ones, who have a strong following in key communities and on social media. In so doing, there should be an effort to provide them with access to trusted, scientific sources and resources that can help them shape their comments in accurate ways
  • Social media platforms should monitor and flag factually inaccurate posts and point participants in the direction of accurate information
  • Celebrities who appeal to different age, gender and demographic groups should join the cause, reinforcing trust in trustworthy sources and encouraging people to ask the right questions and take appropriate actions

In short, it will take a team effort. A big one. One that is less reliant on authority and more invested in transparency, authenticity and two-way communications. An effort that treats people like they’re smart, have agency and can make their own decisions. Which, after all, is true.

We all have a stake in ending this pandemic. We appear to have worked out crucial aspects of the science. That, in turn, gives us the opportunity to talk about the science in transparent, accessible ways, and engage people in the decision about getting a vaccine with dignity. In so doing, we can achieve the goal of herd immunity that would end the pandemic.

Over the long term, this new model for communication about COVID-19 vaccines could set the stage for more successful vaccination campaigns. That’s worth our valuable time and energy.


Mike Kuczkowski is the founder and CEO of Orangefiery, a consulting and communications firm that serves biotech, pharmaceutical and nonprofit clients in healthcare. With more than 25 years of experience in consulting, communications, politics and journalism, he and his team help organizations navigate complex issues of public interest and create growth and change through a combination of insights, messaging, engagement and organizational learning. He is a trustee of the Institute for Public Relations.


Why the Operation Warp Speed Vaccine Studies Aren’t Limited to Severe Disease

Dr. Larry Corey is an internationally renowned expert in virology, immunology, and vaccine development and a leader of the COVID-19 Prevention Network (CoVPN), which was formed by the National Institute of Allergy and Infectious Diseases at the U.S. National Institutes of Health to respond to the global pandemic. He is a Professor of Medicine and Virology at University of Washington and a Professor in the Vaccine and Infectious Disease Division and past President and Director of Fred Hutchinson Cancer Research Center.

[Editor’s Note: a version of this article was first published on Nov. 13 on the Johns Hopkins University Coronavirus Resource Center. —LT ]

The COVID-19 Operation Warp Speed (OWS) trials have taken some criticism in the medical press, and lay press, for evaluating what some consider to be “trivial” characteristics of mild COVID-19 disease.

Some are arguing that it would make more sense to evaluate vaccine efficacy entirely on ability to prevent severe COVID-19 disease, hospitalizations and deaths.

There’s a reason why the trials weren’t focused entirely on those narrow parameters. Allow me a few minutes to explain.

The trial protocols for four major vaccine developers – Pfizer, Moderna, AstraZeneca and Janssen (Johnson & Johnson) — show that the primary endpoint of the OWS vaccine trials is a reduction of coronavirus disease.

The disease is defined broadly in this case, and includes reducing the signs and symptoms of mild COVID-19 illness (e.g., headache, cough, fever, myalgias, signs of loss of taste and smell).

By including mild COVID-19 disease within a primary endpoint, the trials can capture the wide range of symptoms, including sometimes surprising symptoms, observed in patients with COVID-19.

This approach will ultimately allow for vaccines, if successful, to be given to the widest group of people possible – not just a couple important subpopulations. The trials should provide the kind of evidence necessary for efficient real-world rapid assessment of the vaccines’ effect on COVID-19.

For companies developing vaccines, inclusion of mild COVID-19 disease within a primary endpoint for the OWS trials makes sense; these companies recognize the wide range of COVID-19 disease, from mild to severe, and want to produce an effective vaccine with widespread benefit.

Reading further into the vaccine trial protocols shows that a wealth of information is being collected for analysis. Within each trial, there is a deeper evaluation of disease severity: the trials have an endpoint that includes measurement of disease severity at diagnosis and over three weeks of follow-up for every participant who develops COVID-19.

If one looks in the tables at the end of the protocols, (Table 17 in the Moderna protocol, Table 4 in the AstraZeneca protocol, and Appendices 2, 6-8 in the Janssen protocol), one sees that every person in the trials with COVID-19 infection is intensively followed from the time of diagnosis over the next three weeks with standardized assessments of their signs and symptoms. The assessment includes hard objective measurements, such as the use of medical facilities, hospitalizations, and what as investigators we all hope is rare – death.

At the time the vaccine trials were designed in the spring and summer, there was no consensus about the frequency with which those who volunteered to enroll would actually get COVID-19. More importantly, it was unknown what the spectrum of COVID-19 disease among trial participants would be.

In fact, we were pretty certain we would not see what, at that point, had been the norm in the pandemic (i.e., large numbers of people presenting to hospital emergency rooms with full-blown end-stage pneumonia). We suspected that as testing became more widely available in the summer and fall, the vaccine trials would begin to reflect some of this change. We expected participants would report to their investigators earlier, with milder COVID-19 symptoms than what was, at that time, being seen in medical practice or in any published literature.

Healthy volunteers who enroll in clinical trials, especially ones with the complexity and follow-up required in the COVID-19 vaccine efficacy trials, generally have to be interested and concerned about the disease.

The design of the OWS trials involves active disease surveillance. There is essentially no barrier to COVID-19 testing for the 30,000 participants per study. Moreover, these 30,000 participants are knowledgeable about COVID-19, likely anxious about getting the disease, and the experience of serving as a trial participant makes them attuned to any aspect of the disease, all the time. Participants in the trials receive weekly calls/texts about coming to the clinic at the first signs of any respiratory illness or fever. Importantly, trial participants have essentially unlimited access to COVID-19 testing.

Thus, we felt that a clinical classification of COVID-19 disease severity at the time of diagnosis would not really provide an accurate reflection of a participant’s COVID-19 illness. We expected that most trial participants would get worse after an initial diagnosis. As such, the real endpoint evaluation of disease severity – and hence the ultimate “primary endpoint” – would be the constellation of signs and symptoms that were collected over the two to three weeks post-diagnosis. The section outlining the detailed follow-up of COVID-19 cases within each study had its own schematic diagrams/tables and was “costed out” for site funding separately.

One of the harmonizing aspects across all the vaccine trials is that participants with COVID-19 are queried daily with a standardized questionnaire evaluating their signs and symptoms on a mild-to-moderate-to-severe scale. Samples for viral shedding in saliva or the nose are taken every second to fourth day, depending on the protocol, to evaluate the duration for which the virus is shed. Duration of fever and pulse oximetry data are also recorded.

These data allow one to grade the rate of progression, duration of symptoms, duration of signs, duration and severity of systemic effects, and need for any follow-up medical interventions (e.g., physician visits, telemedicine visits, ER visits, hospitalizations, and any complications of hospitalizations). Evaluations of these data are blinded and will be analyzed between the vaccinated and placebo recipients to determine whether disease duration and severity are reduced.

Data outlined in the prior paragraph are listed as secondary endpoints in the trials. While “relegated” to this section, it does not mean their importance is secondary. This was highlighted in a recent paper on using burden of disease as the primary endpoint in COVID-19 vaccine studies by Mehrotra and colleagues, as well as recent analyses of data from the monoclonal antibody studies conducted by Lilly and Regeneron.

The monoclonal antibody studies do a nice job of aggregating medically complicated disease to show the benefits of early use of the antibodies. They also show one approach to compiling recorded signs and symptoms of COVID-19 to show clinical benefit to patients over time. This offers a model to structure analyses both within and between individual and collated vaccine studies. In addition, the prospective close follow-up that will occur in the vaccine studies lays the groundwork for future studies and maximizes what we learn from the 150 expected COVID-19 cases in each vaccine trial of 30,000 participants.

It is unlikely that, in any individual trial, we will see enough hospitalization and death to evaluate efficacy of a vaccine versus placebo for severe COVID-19 disease. Obtaining this information within a trial would require lengthening the time to determine whether a vaccine works by 18 months.

That is simply too long to wait in a fast-moving pandemic.

Vaccines for other diseases offer a useful perspective about the inclusion of mild illness as a primary endpoint within a trial: data from other vaccines show that those capable of modifying outpatient/modest illnesses have almost universally been capable of modifying more severe illness. With most vaccines, efficacy in reducing the severe spectrum of a disease is actually somewhat higher than for mild aspects of a disease.

Let us hope that vaccination will easily produce discernible, measurable, medically important differences in COVID-19 disease in the populations enrolled in the trials. The current surge in activity in the U.S. reminds us of the urgent need for vaccines.

Dr. Larry Corey is an internationally renowned expert in virology, immunology, and vaccine development and a leader of the COVID-19 Prevention Network (CoVPN), which was formed by the National Institute of Allergy and Infectious Diseases at the U.S. National Institutes of Health to respond to the global pandemic. He is a Professor of Medicine and Virology at University of Washington and a Professor in the Vaccine and Infectious Disease Division and past President and Director of Fred Hutchinson Cancer Research Center.


1 Pfizer. PF-07302048 (BNT162 RNA-Based COVID-19 Vaccines) Protocol C4591001. 2020. https://pfe-pfizercom-d8-prod.s3.amazonaws.com/2020-09/C4591001_Clinical_Protocol.pdf.

2 Moderna TX. Protocol mRNA-1273-P301, Amendment 3. 2020. https://www.modernatx.com/sites/default/files/mRNA-1273-P301-Protocol.pdf.

3 AstraZeneca. Clinical Study Protocol – Amendment 2 AZD1222- D8110C00001. 2020. https://s3.amazonaws.com/ctr-med-7111/D8110C00001/52bec400-80f6-4c1b-8791-0483923d0867/c8070a4e-6a9d-46f9-8c32-cece903592b9/D8110C00001_CSP-v2.pdf.

4 Janssen Vaccines & Prevention BV. VAC31518 (JNJ-78436735) clinical protocol VAC31518COV3001 amendment 1. 2020. https://www.jnj.com/coronavirus/covid-19-phase-3-study-clinical-protocol.

5 Moderna TX. Protocol mRNA-1273-P301, Amendment 3. 2020. https://www.modernatx.com/sites/default/files/mRNA-1273-P301-Protocol.pdf.

6 AstraZeneca. Clinical Study Protocol – Amendment 2 AZD1222- D8110C00001. 2020. https://s3.amazonaws.com/ctr-med-7111/D8110C00001/52bec400-80f6-4c1b-8791-0483923d0867/c8070a4e-6a9d-46f9-8c32-cece903592b9/D8110C00001_CSP-v2.pdf.

7 Janssen Vaccines & Prevention BV. VAC31518 (JNJ-78436735) clinical protocol VAC31518COV3001 amendment 1. 2020. https://www.jnj.com/coronavirus/covid-19-phase-3-study-clinical-protocol.

8 Mehrotra, DV, et al. “Clinical Endpoints for Evaluating Efficacy in COVID-19 Vaccine Trials.” Annals of Internal Medicine. https://doi.org/10.7326/M20-6169. Published: 22 October 2020(https://doi.org/10.7326/M20-6169. Published: 22 October 2020).

9 Chen, Peter, et al. “SARS-CoV-2 Neutralizing Antibody LY-CoV555 in Outpatients with Covid-19.” The New England Journal of Medicine. https://www.nejm.org/doi/full/10.1056/NEJMoa2029849. Published: 28 October 2020(https://www.nejm.org/doi/full/10.1056/NEJMoa2029849. Published: 28 October 2020).


The mRNA Vaccine News is Good. But Let’s Keep Masks for Now

Dr. Larry Corey is an internationally renowned expert in virology, immunology, and vaccine development and a leader of the COVID-19 Prevention Network (CoVPN), which was formed by the National Institute of Allergy and Infectious Diseases at the U.S. National Institutes of Health to respond to the global pandemic. He is a Professor of Medicine and Virology at University of Washington and a Professor in the Vaccine and Infectious Disease Division and past President and Director of Fred Hutchinson Cancer Research Center.

[Editor’s Note: a version of this article was first published on the Johns Hopkins University Coronavirus Resource Center. —LT ]

Clinical trials are usually designed to answer one or two specific questions.

For the pivotal COVID-19 trials evaluating messenger RNA vaccines from Pfizer and Moderna, researchers are looking at whether these vaccines prevent a person from getting sick, keep them from suffering a severe, prolonged illness, or keep them from having to be hospitalized.

These are important questions. And the initial trial results are an ironclad demonstration of the power of science. But there are still many more questions we’d like to answer, and more studies will need to be done to allow us to return to a more normal life.

Importantly, these clinical trials are not focused on whether a vaccine boosts immunity to the point that it prevents someone from getting infected with the SARS-CoV-2 virus at all.

It’s possible that a COVID-19 vaccine may benefit the individual who gets vaccinated by protecting them from getting sick. But we know that many people who get infected with the virus can spread it for several days without displaying symptoms, without even knowing they are ill. That means it’s theoretically possible – if not likely – that the virus could invade the body of a vaccinated person, and that vaccinated person might still unwittingly be able to spread the virus to others.

This is a critical distinction that has received little attention.

If you think about it, infecting someone without them knowing it is a great strategy for a virus whose evolutionary goal is to “replicate and survive.”

HIV is a master of this approach. The HIV virus infects most people while causing minimal symptoms. It persists for years before an individual develops the symptoms of AIDS. While the virus is hiding out, unnoticed, the HIV can spread through sexual contact.

SARS-CoV-2 differs from HIV in several ways. For starters, the new coronavirus is most infectious in the first few days after a person has gotten infected. For those first few days, the virus has explosive infectivity — until the body’s immune defense start to kick in against this novel virus it has never seen before.

And, SARS-CoV-2 isn’t transmitted through sex. This virus spreads through much less intimate interactions such as breathing, talking and sneezing around people in your immediate vicinity. This ability to spread through the air makes SARS-CoV-2 way more infectious than HIV. Get close – inches – to someone who’s infected and SARS-CoV-2 is highly contagious. Step a few feet away, and the risk goes down considerably. Wear a mask and it’s a heck of a lot less risky.

So, enter a vaccine. What does it do to this dynamic? Answer… we don’t know yet.

Some of these open questions are quite personal.

People will ask:

If I get vaccinated, do I lessen my risk of actually getting infected? Will I walk around with the virus and not know it?

Do I still shed the virus from my nose, and therefore pose a risk to others if I sneeze? Is my risk of spreading the virus to others reduced by the vaccine, or is it at the same level as if I were not vaccinated at all? How long might the vaccine protect me from getting sick? How long might it keep me from infecting others? Could I be one of those super-spreaders without me knowing that I am sick?

Scientists want to know those answers, because they will be informative for how we adjust our public health response.

If, for example, we have vaccines that are protective against illness, but ineffective against the spread, then we could have a new public health – and communications – challenge.

Imagine if we have hundreds of vaccinated people walking around with the virus unknowingly, behaving with a false sense of security, and yet putting unvaccinated people at an even higher risk of contracting COVID-19?

In medical terms: will vaccinated persons be asymptomatic carriers who can still transmit throughout their household, their community, their school? Will the virus still end up preying on clusters of vulnerable people, like in nursing homes, as it has this year? This is what we call onward transmission.

These questions are of obvious importance. If vaccinated individuals are capable of transmitting infection, then anybody who is not vaccinated faces the same risks of infection as they did before the arrival of the COVID-19 vaccine.

With vaccine hesitancy resulting in fewer people agreeing to be vaccinated, we do not yet know whether and when we will be able to markedly reduce the public health implications of COVID-19 and reduce its circulation in the workplace, in close communities, and stop super-spreading events.

With the promise of a vaccine, many continue to ask: will I still have to wear a mask?

The answer is yes, for now. At least until we know more about infectiousness among the vaccinated.

The way we generally look at a vaccine preventing infection is whether a vaccinated person makes a neutralizing antibody. We want to see increasing levels of those neutralizing antibodies, and for those biological markers to line up with clinical measurements that look at whether people get infected throughout an entire trial. We look at antibodies pre- and post-vaccination to gain this crucial insight into whether the vaccines are doing what they are supposed to do.

We also have tests that determine whether the placebo groups get asymptomatically infected more than the vaccine groups. For example, the most common respiratory virus vaccine – influenza vaccine – sometimes prevents someone from getting a severe case of flu but doesn’t often prevent people from getting infected.

Put differently, one can still acquire the infection even after getting vaccinated. The main goal of most vaccines is to prevent people from getting sick. Preventing a person from getting infected, and therefore preventing them from potentially transmitting the disease, is usually considered to be a higher hurdle to clear.

So, at this point on Nov. 18, we do not know – even with the high efficacy reported in reducing symptomatic disease – whether after vaccination individuals are still infectious. Once a vaccine is widely available, do we still need to be careful about whom we invite to our home for dinner? Do we need to continue to wear masks with other people around, especially casual acquaintances or strangers?

At the moment, the answer is still yes until we unravel these issues.

Mathematical modelers: vaccination and reduction in transmission

Mathematicians have modeled and made predictions about COVID-19, and they have envisioned all sorts of scenarios. And, one scenario they have created imagines that if we have a vaccine that makes everybody asymptomatic, but individuals remain infectious, then whoa! We may actually end up with more people who get infected!

Scenarios like these raise the concern that a lot of people will shrug their shoulders, not get vaccinated, and they will get infected. We will still see a lot of hospitalizations because of COVID-19. There will be more waves of death.

This realization helps explain why we must maximize the number of people in the population who get vaccinated. This requires a concentrated push to overcome vaccine hesitancy, especially in persons who are at high risk. Because we certainly are going to see behavioral changes once people get vaccinated. They’ll think, “I got vaccinated, why do I need to wear a mask?”

So, those modelers, they’ve forewarned us that we do need to learn about infectivity after vaccination.

In the end, to answer the “mask or no mask” question, we need to do a study. That’s where I am today. I’m discussing the issue with colleagues and advocating for a study. Households and college campuses, I think, are a good place to consider for this. These are places where we can vaccinate and sample people every other day and see if they acquire COVID-19. And if they do get vaccinated and give periodic samples, what’s the amount of virus showing up in their noses, their airways, or their bloodstreams?

With that kind of information in hand, we can do some contact tracing to see if people spread it to their roommates and classmates. Colleges are starting to do this anyway. Let’s piggyback on their efforts.

We should do studies that answer the question once and for all: mask or no mask?

Dr. Larry Corey is an internationally renowned expert in virology, immunology, and vaccine development and a leader of the COVID-19 Prevention Network (CoVPN), which was formed by the National Institute of Allergy and Infectious Diseases at the U.S. National Institutes of Health to respond to the global pandemic. He is a Professor of Medicine and Virology at University of Washington and a Professor in the Vaccine and Infectious Disease Division and past President and Director of Fred Hutchinson Cancer Research Center.


Pfizer, BioNTech’s Watershed Moment, Lilly Antibody Gets EUA, & The Rebuilding Begins

Luke Timmerman, founder & editor, Timmerman Report

First thing Monday, we all woke up to the brightest ray of light in this dark year.

Pfizer and Germany-based BioNTech reported that their vaccine candidate was found to be more than 90 percent effective at preventing COVID-19. The report was via press release, not peer-reviewed journal, but this was still a moment to celebrate.

The interim analysis wasn’t based on a thin-gruel data set of 30 events. The companies enrolled 43,538 people in the pivotal study, and derived the efficacy calculation off of 94 confirmed cases of COVID-19. This is a more solid persuasive set of data than I expected, based on reading the Phase III clinical trial protocol in September. The protection number might dip when the full analysis is run on 164 confirmed cases, but still, this is meaningful.

What’s more, the study wasn’t just a bunch of white people. It had diverse participants. Pfizer expanded the enrollment population to make sure this trial didn’t suffer from that Achilles heel.

The sheer magnitude of benefit, and the fact it was distributed across diverse populations, ought to go a long way toward allaying the fears of so many people – which we now have a popular name for (vaccine hesitancy). My worst nightmare for months has been that the scientific enterprise would deliver something important – a safe vaccine with 60-65 percent protection – but that it would be pointless because people would be afraid to take it, out of fear that it wasn’t proven safe.

I don’t believe that will happen now. People will likely line up to take a vaccine if they know it’s 90 percent effective, that’s been shown safe in more than 40,000 people, and that will — along with mask-wearing and some more social distancing — help bring this global nightmare to an end in 2021.

What’s more, this is the kind of scientific validation we all needed to see for mRNA vaccines as a platform, and for our efforts at targeting the SARS-CoV-2 spike protein. The Pfizer / BioNTech news bodes especially well for Moderna, and for other vaccine candidates in the pipeline. With multiple companies going all-out on manufacturing and distribution, and with some having different refrigeration requirements, we have a good chance to achieve the vaccination coverage across the population that we need to break the chain of transmission once and for all.

Between now and that day, a lot is up to us. It’s hard for people to see how seemingly innocuous individual actions, from people with no apparent symptoms, can add up to so much compound harm. But if you have family members or relatives that have any doubt, please steer them to this chilling CDC report about the ripple effect from a wedding in rural Maine in August, when people weren’t wearing masks.

You can read the full CDC report for yourself.

This ought to make everyone stop and think about family plans for the holidays.

It’s a sad thing to miss family gatherings this year. But there is hope, and good reason for us to stick it out a few months longer.

Everyone reading this publication probably doesn’t need to be reminded of what works. The virus spreads through the air. Cloth masks, physical distancing, crowd avoidance, and hand-washing are still the best non-pharmaceutical interventions we have until the vaccine cavalry arrives. We have good reason to have confidence in the neutralizing antibody therapies for patients who get mild to moderate illness.

Our country has been deeply damaged by pernicious efforts to undermine science and truth itself. Millions of people have been told repeatedly, that COVID-19 was a hoax, that it was magically going away, and it was cooked up by the media and Democrats to undermine the President in an election year.

That’s, of course, a lie. We had 150,000 new cases reported yesterday, and 66,000 people hospitalized. This is uncontained national spread, not just a cluster here or there. More than 1,000 people are dying every day, and estimates are 100,000-150,000 will die before Inauguration Day.

A huge amount of suffering and death could have been avoided, if the pernicious lies hadn’t been told over and over and over.

The result is that we now have to sift through the ashes to build a new society where people can talk to each other again like rational adults. If we can start there, maybe we can begin to listen to each other, find toeholds of common ground, and begin to cobble together some more of the social glue we call trust.

The election of Nov. 3 was a clear and convincing start down that path toward a better world. Now, if people of good faith are willing to do the work, with eyes on the North Star of creating a better world for ourselves and for future generations, then we have a chance to wake up to some more bright mornings like we did when Pfizer and BioNTech surprised us on Nov. 9.


Eli Lilly got an Emergency Use Authorization for bamlanivimab as a therapeutic neutralizing antibody for newly diagnosed patients with mild-to-moderate COVID-19. Now we’ll see if Lilly can make enough of the antibody to help patients who are rapidly filling up hospitals.

Regulatory Action

AstraZeneca won FDA clearance for ticagrelor (Brilinta) to reduce the risk of stroke. The twice-daily medication also comes with a slightly elevated risk of bleeding.

An FDA advisory committee voted against approval of Biogen’s aducanumab for Alzheimer’s disease. The agency’s advisors were clearly unimpressed with the application, despite a rather unusually rosy appraisal of the drug by the FDA staff. The agency has a Mar. 7, 2021 deadline to complete its review and make a decision.

Foundation Medicine won FDA clearance for its FoundationOne Liquid CDx as a companion diagnostic for olaparib (Lynparza), a treatment for BRCA1/2 mutated prostate cancer.


  • SARS-CoV-2 D614G Variant Exhibits Efficient Replication Ex Vivo and Transmission In Vivo. Science. Nov. 12. (Yixuan J. Hou et al)


  • SARS-CoV-2 Transmission Among Marine Recruits During Quarantine. NEJM. Nov. 11. (Andrew Letizia et al)
  • Multiple COVID-19 Outbreaks Linked to a Wedding Reception in Rural Maine. CDC Morbidity and Mortality Weekly Report. Nov. 13. (Parag Mahale et al)
  • Cellphone Mobility Network Models of COVID-19 Explain Inequities and Inform Reopening. Nature. Nov. 10. (Serina Chang of Stanford University et al)


  • After COVID Diagnosis, Nearly 1 in 5 Are Diagnosed With a Mental Disorder. NPR. Nov. 11. (Laurel Wamsley)
  • 60-Day Outcomes Among Patients Hospitalized with COVID-19. Annals of Internal Medicine. Nov. 11. (Vineet Chopra et al, summary by Atul Gawande below)

Public Health

  • Maryland Launches Targeted Wastewater Sampling for Coronavirus. Maryland Department of the Environment. Nov. 12. ( Larry Hogan)
  • Community Use of Cloth Masks to Control the Spread of SARS-CoV-2. Nov. 10. CDC Scientific Brief.


  • The Hidden Superpower of Strategic Focus. Drug Baron. Nov. 9. (David Grainger)


  • Why Do COVID Death Rates Seem to Be Falling? Nature. Nov. 11. (Heidi Ledford)
  • New T-Cell Test from Adaptive Biotech May Better Determine Immunity to Coronavirus. NYT. Nov. 10. (Apoorva Mandavilli)
  • Covalent Drugs Go From Fringe Field to Fashionable Endeavor. C&EN. Nov. 8. (Bethany Halford)
  • Scientists Relieved as Joe Biden Wins US Presidential Election. Nature. Nov. 7. (Jeff Tollefson)
  • Scientist Behind Pfizer / BioNTech Vaccine Says It Can End Pandemic. The Guardian. Nov. 12. (Philip Oltermann)
  • COVID Hell. A Humanitarian Disaster. Experts Sound the Alarm About the Outbreak. Nov. 12. Washington Post. (Marisa Iati)

Our Shared Humanity

  • Will the World Be Generous Enough to Defeat COVID-19? Speech at the Paris Peace Forum. Nov. 12. (Melinda Gates)
  • Some Lives Matter: The Dirty Little Secret of the US Healthcare System. The Hastings Center Report. Oct. 23. (Leonard Fleck)
  • Lucille Bridges, 86, Dies. Led Her Daughter Across a Color Line. NYT. Nov. 11. (John Ismay)
  • How to Minimize COVID Risk and Enjoy the Holidays. Scientific American. Nov. 10. (Christie Aschwanden)


  • America Is Behind on COVID-19 Vaccine Messaging. Experts Call for Honest, Straight Talk. USA Today. Nov. 10. (Elizabeth Weise)
  • We’re All Science Communicators. Here’s How to Do it Better. C&EN. Nov. 4. (Jen Heemstra)

Policy Responses


Boston-based Decibel Therapeutics, the developer of gene therapies for hearing loss, raised $82.2 million in a Series D financing. OrbiMed led.

Cambridge, Mass. and Suzhou, China-based Kira Pharmaceuticals said it raised $46 million to advance work on complement-targeted therapies to treat immune-mediated diseases. Quan Capital led. Frederick Beddingfield was named CEO.

Emeryville, Calif. and Germany-based Metagenomi raised $65 million in a Series A financing to do CRISPR-based gene editing of microorganisms. Leaps by Bayer and Humboldt Fund co-led.

Bedford, Mass.-based Homology Medicines, a genetic medicines company, raised $60 million in an equity investment from Pfizer. The big drugmaker bought 5 million shares at $12.

Shanghai-based Immagene Biopharmaceuticals, an immunology-based drug developer, closed a $21 million Series B financing led by Vertex Ventures China.

Personnel File

South San Francisco-based Graphite Bio hired Katherine Vega Stultz as chief operating officer and Philip P. Gutry as chief business officer and head of finance & investor relations. (See TR coverage of Graphite Bio’s $45 million Series A, September 2020).

Vancouver, BC and Seattle-based Chinook Therapeutics, the developer of precision kidney medicines, hired Eric Bjerkholt as Chief Financial Officer.

Cambridge, Mass.-based Foghorn Therapeutics hired Michael LaCascia as chief legal officer.

Seattle-based Qiming Venture Partners USA said it promoted Anna French and Colin Walsh from principals to partners. French is in the Boston office and Walsh is in the San Francisco office.

Cambridge, Mass.-based Magenta Therapeutics hired Steve Mahoney as chief financial and chief operating officer.

Data That Mattered

San Diego-based Arena Pharmaceuticals failed to hit the primary endpoint in a Phase II trial with its once-daily, oral sphingosine 1-phosphate (S1P) receptor modulator, etrasimod, for moderate to severe atopic dermatitis. The primary endpoint was Eczema Area and Severity Index (EASI), evaluated for the drug and placebo groups at 12 weeks. The company, however, slapped a positive headline on its press release, saying the drug did better on the validated Investigator Global Assessment (vIGA). Despite missing the primary endpoint in Phase II, the company said it will advance the drug into Phase III anyway. Shares fell.

South San Francisco-based Five Prime Therapeutics showed in a randomized, placebo-controlled Phase II trial that bemarituzumab in combination with mFOLFOX6 chemotherapy was able to improve progression-free survival, and overall survival time, for patients with front-line advanced gastric and gastroesophageal cancer. The drug is designed to block activation of fibroblast growth factor receptor 2b. Five Prime stock boomed from $5.34 a share before the news to $22.43 in the two days after. It announced plans to strike while the iron is hot, doing a new offering of 5 million shares to raise cash.

AstraZeneca and Amgen reported positive results from a Phase III trial of patients with severe asthma. The drug, Tezepelumab, is designed to inhibit TSLP, a cytokine in epithelial cells that is thought to play a key role in multiple inflammatory cascades that cause airways to inflame. The drug showed an ability to reduce asthma exacerbations over 52 weeks. Data will be presented at a medical meeting.

ViiV Healthcare, the HIV drug developer established by GSK and Pfizer, said its every-other-month injectable treatment for HIV, cabotegravir, beat Gilead’s emtricitabine/tenofovir disoproxil fumarate (Truvada) in a head-to-head study. Researchers with the HIV Prevention Trials Network found the injectable medicine was 89 percent more effective than the once-daily oral regimen at prevention women from getting HIV.


Vifor Pharma agreed to pay $30 million upfront, another $30 million equity investment, and $20 million in clinical study milestone payments to Angion Biomedica. The deal is part of a collaboration to deveop a small-molecule hepatocyte growth factor (HGF) mimetic to treat acute kidney injuries.


Creating the Future of Microbiome-Based Therapies: Simba Gill on The Long Run

Today’s guest on The Long Run is Simba Gill.

Simba is the CEO of Cambridge, Mass.-based Evelo Biosciences.

Simba Gill, CEO, Evelo Biosciences

Evelo is part of a new generation of biotech companies seeking to make medicines based on new understanding of the microbiome. The science here is fascinating. Evelo’s drug candidates are biologics designed to be taken orally, to act directly in the gut, and to have therapeutic effects throughout the body.

In this conversation, Simba describes what the key scientific discoveries were that inspired the founding of Evelo, and how the company created a strategy for therapeutic interventions (which is no obvious thing). Evelo is in the clinical development stages with drug candidates for chronic inflammatory diseases like psoriasis and for cancer. Based on some of its clinical observations, there’s even a rationale for testing one of its candidates for COVID-19 – clinical studies are ongoing.

Simba grew up in the UK, and got his scientific training as an immunologist. He is a global citizen, and has consistently gravitated to big new ideas when the biology risk was high, and entrepreneurs have to be a little extra comfortable entering the unknown.

Simba also thinks a lot about the intersection between science and society. Much of his drive is about creating therapies that can help the largest number of people possible around the world.

As you’ll hear from the start, Simba was part of the 27-member Kilimanjaro Climb to Fight Cancer team in 2019. He and I got to know each other, and discovered our shared interest in science and society, while on that trip to climb the highest mountain in Africa. He’s someone with a big heart.

Now, please join me and Simba Gill on The Long Run.


Biotech Companies Acting on Promises to Increase Racial Diversity

Karl Simpson, CEO, Liftstream

First came the pandemic. Then the economic slump. Then the push for racial justice. Taken together, you have three major challenges for business leaders to tackle all at once.

Because the pandemic has illuminated many racial inequalities, these issues have become intertwined.

In the biopharma sector, many company leaders stepped up. Decisive moves were made to reorganize working patterns to protect their staff, while continuing important work. In some cases, work that is vital to meet the need for new therapies to treat COVID-19 patients, and find a possible vaccine at “warp speed”.

Many leaders have also made statements, and created action plans, to address racial inequality in their companies. While the individual motives may differ, it has provided the industry with an opportunity to re-establish its bonds of trust with citizens; a chance CEOs appear keen to take.

Most employees in biotech and pharma are motivated by the mission to help patients, regardless of their race, gender or socioeconomic circumstance. This north star has burned brightly throughout the pandemic. As the industry has risen to the COVID-19 challenge, it has done so while paying attention to the need for conducting diverse clinical trials.

Pfizer/BioNTech reported Nov. 9 that 42% of their 43,538 trial participants who enrolled in their COVID-19 vaccine pivotal trial have racially or ethnically diverse backgrounds. Moderna, another front runner in the race for a COVID-19 vaccine, said that 37 percent of the people enrolled in its pivotal study are from diverse communities — 11,000 out of the 30,000 total. Specifically, the trial included 6,000 Hispanic or Latinx participants, and more than 3,000 African American participants.

Diversity in clinical trials is one area where the industry can take direct, immediate, and positive action. It’s a crucial step toward, delivering better patient outcomes and reducing inequities in healthcare. The industry must also address health equity more broadly, including the issue of access to new treatments, especially for people with no health insurance, or inadequate health insurance.

Many of the large companies in biopharma have already been focused for some time on improving access to medicines for low-income and underserved populations.

That’s one place to start. But for biopharma to restore trust with patients and minority communities it will require some internal changes too. These factors rely on the range of perspectives contributing to the process of innovation and the management of companies. For this reason, diversity is becoming imperative in all aspects of the life sciences sector.

What are companies doing? How have they been responding to the intersecting challenges of the racial justice issue and a pandemic? – Here, we look at a sample of companies, large and small, and explore what action they have taken.

Bristol-Myers Squibb

The converging opportunities of more diverse clinical trials, improved health equity, and a more diverse and equal workforce has led BMS to announce a $300 million, 5-year investment. BMS aims to double executive roles for Black and Latinx employees by 2022 and to level up on pay.

Like many other companies, BMS will look to impact the supply chain too, through a pledge to spend $1 billion globally by 2025 with black and diverse-owned suppliers. Plus, BMS said it will double-match donations its employees make to non-profits dedicated to fighting discrimination or inequities in the health system.

This ambitious initiative came about six weeks after BMS added Paula Price and Derica Rice, both black candidates, to its board of directors. BMS, along with other biopharma companies, has also committed to the NYC Comptroller, Scott Stringer, to disclose its consolidated EEO-1 data, which details some useful standardized data on its employment of racial and ethnic workplace minorities.


Biogen has a long been committed toward diversity, equity and inclusion. This June, Biogen built on its foundational Diversity, Equity & Inclusion (DE&I) program to increase its breadth and depth, both internally in areas like hiring and development, and externally in things like clinical trial engagement strategy.

Minita Shah-Mara, head of organization effectiveness & diversity and inclusion

“Our equity and inclusivity journey is ongoing,” said Minita Shah-Mara, Head of Diversity, Equity & Inclusion at Biogen, “but we should also be excited about where we are on this journey. It requires collaboration, learning, vulnerability and holding ourselves accountable to meet our goals.”

Biogen is aiming for a 30% increase in Black, African American and Latinx people employed in its leadership ranks by the end of next year. Biogen’s plan also focuses on Asian employees where underrepresented, and women too, until parity or population demographics is achieved by level and function. Biogen is also looking to increase the representation of veterans, people of disability, and LGBTQ individuals. The plan doesn’t stop with hiring. It includes tailored development programming for underrepresented employees who may be ready to move into first-time supervisor roles. Research shows this is a pivotal career stage where the opportunity gap widens, and Biogen should be commended for designing a targeted intervention.  

Biogen’s plan contains specific data-driven goals, and a framework for measuring progress against the goals. By collecting richer sets of data, companies can create dashboards to measure progress. This new analytical approach allows leaders to see how the company is performing by geography, by job function and seniority level. Gathering and analyzing this data also paves the way for a better understanding of employee advancement and promotion, hiring, and team composition. Periodic updates will be communicated within the entire organization.

The DE&I strategy, encompassing extensive staff training on inclusive hiring, promotion and retention, extends beyond the walls of the company’s offices. For one thing, Biogen recognizes its ability to boost financial empowerment for minorities. Biogen is increasing its spending with minority-owned businesses. This financial empowerment strategy has already resulted in Biogen depositing $10 million with OneUnited Bank, the largest Black-owned bank in the United States.

Biogen recognizes that serving patients is the priority, and that is embedded in its DE&I strategy. There are multiple components to the external side of Biogen’s strategy. It is seeking to boost minority participation in its clinical trials, to conduct more studies of underrepresented groups in disease areas relevant to the company, and to invest more in patient engagement, education, and access to medicines in underserved communities. As with BMS, Biogen is willing to submit to external scrutiny on how it’s doing on these goals. Biogen is one of the companies that has agreed to disclose its consolidated EEO-1 data. Taken together, with the internal and external components, this is a serious plan.

Sutro Biopharma

Smaller companies are getting involved as well. South San Francisco-based Sutro Biopharma, a developer of drugs for cancer and autoimmune disease, has been following through on the CEO’s open letter to employees and stakeholders in the aftermath of George Floyd’s killing. CEO William Newell has committed company resources to improve the company’s connections with underrepresented and lower socioeconomic communities, having linked in with local organizations like Students Rising Above and Peninsula Bridge. Sutro opened its doors this past summer to nine interns from diverse backgrounds, offering these students real work experience, despite the operational challenges of COVID-19.

Bill Newell, CEO, Sutro Biopharma

Newell says Sutro is just beginning. Sutro is becoming more intentional about addressing racial diversity at all levels of the company, from the board of directors down. The company has created a diversity section on its website with curated content, and all the employees have T-shirts emblazoned with the word SAWUBONA, a Zulu word meaning ‘I see you, you are important to me and I value you’. Newell said several factors have compelled him to become more active. He’s been feeling the call of the broader movement for racial equity. He’s been working to learn more about the subject. And Sutro has grown to a point where it has resources to take meaningful action.  

Newell, a member of the BIO board, is serving on the trade group’s Workforce Development, Diversity and Inclusion committee. He is also on the board at the California Life Sciences Association, which is also adopting a strategy for DE&I. He is very engaged in the topic, and his early efforts are commendable. There is still a lot for Sutro to do, but Newell is showing what can happen at a development-stage biotech company when the CEO is personally committed.

Xilio Therapeutics

Waltham, Mass.-based Xilio Therapeutics is a venture-backed cancer immunotherapy startup, at an early phase of development, like many companies in the Timmerman Report readership. The CEO, Rene Russo, is an advocate for diversity. She signed MassBio’s Open Letter 2.0, which asked leaders to commit to several commitments on racial and ethnic diversity.

Rene Russo, CEO, Xilio Therapeutics

Russo’s public pledge coincided with the company concentrating its DE&I efforts in the past few months. Xilio hired a chief human resources officer, Jason Robart, to help formulate a human capital strategy, of which DE&I is an integral part, rather than somehow being separate. The view at Xilio is that it has to be part of the corporate planning process. 

As a relatively young company with good intentions, Xilio has an opportunity to turn DE&I into an integral part of how it grows up. This is a good time to act intentionally. In Xilio, as with many small and early-stage ventures, the reliance on networks for hiring is high. This network dependence has the potential to seed homogeny. People tend to hire other people who look like them. To counteract this well-known unconscious bias, Xilio is being very intentional about reviewing and assessing the job profiles for which it is recruiting, thereby widening the aperture through which to view prospective employees. More diligence about the relevant experience of candidates too, will support this more expansive hiring approach.

“Xilio’s success in developing therapeutics that serve the needs of a diverse patient population will be enhanced by having a diverse workforce,” Russo noted. About one-third of the company’s current workforce identifies as racial or ethnic minorities, while about half are women.

Xilio has experienced a groundswell of interest in DE&I by employees since the racial injustice protests swept the country in June. The company has conducted some initial survey work to understand better the composition and needs and preferences of the workforce. As a science-based organization, data will be a key component of measuring the effectiveness of its DE&I efforts, but the company will also monitor some of the less quantifiable indicators. “Successful DE&I efforts will show up in the numbers, but perhaps equally important, they will also show up in how we do business,” Russo said. “When we see a material and consistent change in the nature of conversations and in the behaviors of employees, we’ll know that we’re making progress.”

To help develop a comprehensive DE&I program, Xilio has formed a DE&I council made up of employees, including senior leaders. This council will also have members from outside the company too, to introduce external accountability as well as infuse different ideas.

The company has been working with Project OnRamp, an initiative started by Life Science Cares, which gives interns from underrepresented communities opportunities to work in life sciences companies.

One challenge for Xilio is to reduce its reliance on existing networks, and to be open to hiring more diverse candidates from unfamiliar sources. The company is in a building phase, and diversity often gives way to just-in-time hiring. Xilio has been busily hiring in recent months, adding an independent Chairman, two independent directors, a Chief Medical Officer and a Chief Human Resources Officer. While only one of these individuals is a racial minority, Xilio deserves credit for having three women on the board. 

Genocea Biosciences

When COVID-19 forced everyone to change how their organizations work, Genocea CEO Chip Clark had employees working from home like everyone else. But this dislocation has meant that Clark has had to place extra emphasis on internal communication to keep everyone on the team engaged and connected. Clark has been sending daily notes to staff, in addition to inviting employees to attend the usual company happy hour, if they choose.

Chip Clark, CEO, Genocea Biosciences

In the immediate aftermath of George Floyd’s death, Clark, a black man, had to search for the right response. The Monday morning note-to-all had to strike the right balance in terms of what the event meant to him, the employees, the company and the other stakeholders, not least the patients they’re all trying to help. 

The company made a statement on its website, decrying racism and claiming that it cannot be neutral in the face of such events. Clark’s note to staff, and the public statement, would begin an internal dialogue about the issue and what to do about it. This conversation needed pragmatism and balance, and could not be casual or overly political, although keeping these separate isn’t easy. Clark was keen to remind his team that they were privileged to do what they do, get paid for it, and to be part of a vibrant and dynamic local bioscience economy in Massachusetts.

This dialogue led to the first of the company’s deliberate actions. People needed an outlet for their anger, frustration, and desire to improve the situation. Genocea agreed to provide a company match when employees donate to non-profits dedicated to tackling racial and social justice. While donations are a demonstration of this, the company is also encouraging employees to invest their time and talent, through taking time off for community service days.

The company formed a D&I committee. Clark believes that the company is performing well on workforce composition, relative to peers, but acknowledges there is also more the company can, and must, do. One way he is zeroing in on this is to apply a renewed focus on hiring practices, including widening the scope of recruiting efforts, accessing more diverse candidate populations, and monitoring measurable benchmarks; all supported by a documented process.

Clark says the social context of the current environment has increased his recent drive for change, but that the company’s stronger financial position today and the accountability of external groups driving for progress have all made it possible to set positive changes in motion. He admits much of this DE&I work is hard, but he is committed on continuing to build a company where a range of views is welcome. These include political views, which can be a source of tension on such issues as DE&I, but creating an environment where people can live their values and feel connected to the culture of the company is crucial.


Management teams always have many competing priorities to juggle, but it’s clear that racial equity and diversity is now a top-of-mind priority at many companies. Each company was already making efforts and progress on diversity and inclusion, but the events of this summer have provided new impetus.  

Every company has its starting place, with different end goals in mind, and they’ll approach it in their way, mainly using tools, processes, systems and strategies that have been proven to work, yet need tailoring to their needs. No doubt, each will have ways they can improve on their scorecards too.

All of the companies reviewed are aware of the social context their respective firms operate within. They are working for all their stakeholders, including their employees of today, and tomorrow. Every one of these companies understands that by bringing equity, diversity and inclusion to their firms, they reduce risks which jeopardize their long-term success and sustainability.

Leaders are making a risk/benefit calculus on DE&I, and the scales seem to be tilting ever increasingly towards the considerable benefits that being more inclusive offers. Most leaders recognize they are reliant on the developing generation of talents currently making the bioscience community the force that it is. Future generations of biotech workers will be increasingly diverse. To be an attractive employer, companies will need to be fully inclusive of the workforce, and in their interactions with society.

The slow increase in the representation of Black, Latinx and other racial workplace minorities in the industry is still a problem; no one can pretend otherwise. These companies know it. It should not take the tragic racial events of 2020 to cause momentum to shift, but it appears the wheels are in motion to bring about the change we need to see.


Karl Simpson is CEO of Liftstream, an executive search and leadership consulting practice that exclusively serves clients in the life sciences industry by supporting board and executive appointments, and consulting on diversity and governance. Karl is co-founder of BioDirector, a diverse international network of board director which is improving corporate governance and diversity in the boardrooms of innovative healthcare companies. He is also the founder of the Bioscience & Investor Inclusion Group, an industry coalition developing DE&I solutions for VCs and innovator companies.


Reflections from a Wisconsin Boy

Luke Timmerman, founder & editor, Timmerman Report

My first real journalism job flashed to mind this week.

It was 1998-1999. I was a kid reporter fresh out of the University of Wisconsin. My job was to cover Dane County government for The Capital Times, the progressive newspaper in Madison.

Dane County had about 400,000 people. Half lived in the beating liberal heart of the City of Madison – home to one of America’s great state universities and many government workers.

The other half of the county’s residents lived in suburban villages and rural farmland. Former Wisconsin Gov. Lee Dreyfus once famously needled Madison as “30 square miles surrounded by reality.” Think dairy farms, feed mills, rolling hills, and the occasional small-town Main Street.

The Dane County executive, who I was supposed to watch like a hawk and report on daily, was a former environmental activist. The editors of my newspaper were supportive of her political positions.

I loved that job. What an education this was for 23-year-old me.

Kathleen Falk struck me as a bright and dedicated public servant. But she had her work cut out. This was a tense, pressure cooker of a job.

After years of being a legal activist on the outside, advocating fiercely and intelligently for clean water and clean air and all kinds of good environmental causes, she had weighty responsibility.

As County Executive, she had to manage the classic urban-versus-rural land use battles. Wisconsin was losing an average of 2-3 family farms per day in those years (my Mom and Dad’s family farm in Grant County included). Corporate agriculture was ascendant. Family farmers, left with no economic legs to stand on and not enough in the younger generation willing to carry on with that way of life, felt compelled to sell their land to housing developers in order to retire.

Farmland was disappearing under suburban cul de sacs. Wildlife – deer, squirrels, rabbits, pheasants, ruffed grouse, Canada geese – were being left with less habitat.

Falk had a North Star. She wanted to protect the land. But she also governed at a time when Money magazine ranked Madison as one of America’s Best Places to Live. Newcomers were coming, like it or not.

Falk had to balance competing interests — do we allow more housing subdivisions on the periphery of the City of Madison? What about rebuilding some run-down neighborhoods? Would people want to live there? Where exactly were all these new people supposed to live, anyway? How were they supposed to get to and from work, without spewing too much CO2, or jamming up the roads and disrupting the wonderful quality of life which drew people to Madison in the first place?

County Executive Falk, to my mind, never abandoned her principles. But she realized that there is another side to the story, she needed to listen, and needed to take it into account. Representing her tribe in Madison, and ignoring the rest of Dane County, wasn’t going to fly. It was her job to mediate, to carve out pragmatic solutions that both sides could live with.

This was not easy. These issues were emotionally heated with partisan rhetoric, even then.

To get anything of substance done, Falk had to work with a 39-member legislative branch, the Dane County Board. I covered their meetings that ran late into the night. Firebrand 1960s-style liberals from Madison (picture Vermont Senator Bernie Sanders) would step to the microphone and let it rip. When they were done, their counterparts from the rural parts of Dane County (picture Iowa Senator Chuck Grassley) would dish out their counterarguments, sometimes dripping with contempt about the liberals and their lack of “common sense.”

Some of these people really didn’t like each other. They’d come by the press gallery off to the side of the main chamber, and fill this reporter’s ear with gossip about how such-and-such supervisor was full of baloney, or how he or she was in the pocket of special interests.

Yet, these people had to find ways to work together to do basic things, like build and maintain roads, preserve the environment, take care of the most vulnerable citizens, and keep taxes in line with what people could stomach. Falk liked to describe her approach as “progressive policies with fiscal restraint.”

There was always the tension between individual rights and the common good. When you have a retiring farmer who’s worked hard his whole life and struggled to make a living with weak commodity crop prices, how do you tell him that he can’t sell his land to a housing developer?

What else is the farmer supposed to do?

Lord knows, our world needs activists to focus on injustice and demand change. But we’re all citizens in a democracy. We need to recognize this is a fractious country, and we have to balance competing interests. None of us is 100 percent right 100 percent of the time. None of us can win every battle. We need to seek common ground with people we strongly disagree with on some things if we want to accomplish anything.

Falk did her best. She clearly disappointed friends on the left at times. She never won over her skeptics on the right. But when she announced her retirement in 2010, as the longest-running Dane County Executive ever, a local newspaper wrote:

Part of what played into her decision to step down is the fact she fulfilled the two promises she made when re-elected in April 2009: building the nation’s first community manure digester and launching real cultural change around the big costs and wholesale suffering from the community’s abuse of alcohol.

“Getting this manure digester and changing the paradigm of how we clean up our lakes was very important, and I am really proud that we have broken ground on that and will be flipping the switch in a few months,” Falk said.

That’s no punch line. Building a manure digester means dealing with waste from farm livestock in a way that allows farmers to continue making a living, while keeping Dane County’s beloved lakes clean.

That’s smart problem-solving, smart balancing of competing interests. This is what we want, and what we should demand, from our more competent public representatives.

We are in a very dark chapter in our country’s history, worse even than the Red Scare of the 1950s McCarthy era. We can’t go on like this.

I grew up in Wisconsin and still have friends and family there. Some are staunch conservatives. There are things on which I will not yield, and which they won’t either. But we do have things in common.

This might take the rest of our lives to dig ourselves out of this ditch, and get our society back on a stable footing. Our society isn’t structured this way in the social media era, but if we talk less, listen more, and assume good faith from others, we can work together again and start solving some of our biggest problems.


Boston-based Atea Pharmaceuticals, a developer of antiviral therapies, raised $300 million in an IPO at $24 a share.

Seattle-based Icosavax secured $16.5 million in financing, including a $10 million grant from the Bill & Melinda Gates Foundation, to advance work on a virus-like particle approach to a COVID-19 vaccine.


  • An Ultrapotent Synthetic Nanobody Neutralizes SARS-CoV-2 by Stabilizing Inactive Spike. Science. Nov. 4. (Michael Schoof et al, UCSF)
  • Versatile and Multivalent Nanobodies Efficiently Neutralize SARS-CoV-2. Science. Nov. 4. (Yufei Xiang et al, University of Pittsburgh)
  • Regeneron Data Safety Monitoring Board Recommends Clinical Trial for Antibody Cocktail be Modified to Exclude Patients on Ventilation, or High-Flow Oxygen. Oct. 30. (Regeneron statement)


  • De Novo Design of Potent and Resilient hACE2 Decoys to Neutralize SARS-CoV-2. Science. Nov. 5. (Thomas Linsky et al)
  • Multi-omics Resolves a Sharp Disease-State Shift Between Mild and Moderate COVID-19. Cell. Oct. 28. (Yapeng Su et al)
  • Viral genome sequencing places White House COVID-19 outbreak into phylogenetic context. MedRxiv. Nov. 1. (Trevor Bedford et al)


Merck agreed to acquire San Diego-based VelosBio for $2.75 billion in cash. The private company is developing and antibody-drug conjugate aimed at ROR1. It’s a Phase 1 trial for hematologic malignancies, and a Phase 2 clinical trial for solid tumors. Merck cited some encouraging preliminary data that’s set to be presented at the American Society of Hematology meeting next month. VelosBio was founded in 2017, and raised $202 million in its history, including a $137 million Series B in July. Investors who are cashing in on the Merck deal include: Matrix Capital Management, Surveyor Capital, Adage Capital Management, Cormorant Asset Management, Farallon, Foresite Capital, Janus Henderson Investors, Logos Capital, OrbiMed, T. Rowe Price Associates, Venrock Healthcare Capital Partners, Viking Global Investors, Wellington Management, Arix Bioscience, Decheng Capital, Pappas Capital, Sofinnova Investments, and Takeda Ventures.

Sanofi agreed to acquire Netherlands-based Kiadis Pharma for 308 million EUR. The company is developing an off the shelf’ K-NK cell technology platform.

Seattle-based Sana Biotechnology agreed to acquire Oscine Corp., a company developing cell therapies for the brain and central nervous system. Terms weren’t disclosed. (See my October interview with Sana Biotechnology CEO Steve Harr, for Geekwire.)

PerkinElmer agreed to acquire UK-based Horizon Discovery for $383 million. Horizon provides CRISPR and RNAi reagents, cell models, cell engineering and base editing offerings.

Science Features

Regulatory Action

FDA staff issued briefing documents for an advisory committee hearing on Biogen’s aducanumab treatment for Alzheimer’s disease. The staff review was surprisingly positive of the application, especially since FDA reviewers tend to raise pointed questions and (appropriately) probe for weaknesses in sponsor presentation. The FDA advisory committee will review the data presented on Friday Nov. 6, and the staff deadline to review the application is in March. Biogen shares surged (chart below) on what looks like a surprisingly easy pass from the FDA.


Philip Lee, a former chancellor of UCSF and an assistant secretary of health during the critical early days of Medicare, died at 96. Reading obituaries like this, of people who you’ve never heard of but who lived meaningful lives, is the kind of thing that can restore one’s faith in humanity. These people exist in our country, even if they tend not to attract the lion’s share of attention.

Personnel File

BeiGene CFO and chief strategy officer Howard Liang announced he’s retiring in the first quarter of 2021.

Deep Genomics, an AI for therapeutics company, hired Amanda Kay as chief business officer. Tom Hughes, the CEO of Navitor Pharmaceuticals, joined the board.

Novavax added Gregg Alton, the veteran biotech executive formerly of Gilead Sciences, to its board of directors.

Data That Mattered

South San Francisco-based Allogene Therapeutics reported on some preliminary Phase I dose-escalation data from its anti-BCMA cell therapy. Four out of 17 patients (24 percent) had Grade 3 or 4 (severe) infections.

The company then went on to describe another adverse event in what could only be described as so-clinical-as-to-be-obfuscatory language. Quote below:

“The fourth was a Grade 5 event of suspected fungal pneumonia that occurred on day eight post-ALLO-715 infusion. The suspected fungal pneumonia was diagnosed on the day after cell infusion in this patient with advanced and rapidly progressing disease who had failed multiple lines of therapy. This event occurred in the CA cohort, and it was assessed by the investigator as related to progressive disease and the CA conditioning.”

Translation: The fourth patient died. The patient was pretty sick already when he or she entered the trial. When the patient died, the doctor figured it was probably because of a combination of the patient’s illness, and the chemotherapy conditioning agent. Nobody, nobody, nobody is saying here in black and white ink that our experimental cell therapy might have something to do with sad result.

Come on, guys. I suppose it’s possible that Allogene is correct in this innocuous interpretation of events. It may be true that no one really knows at this point what really happened. But the standard clinical language here serves to confuse people, deflect blame, or perhaps lull people to sleep so they just move on.

This isn’t the kind of communication that builds trust. This industry, like so many parts of our modern world, needs to think harder about how to build trust, not further erode it.    


Why Learning From Electronic Health Records Is So Appealing – And So Hard

David Shaywitz

The application of technology to medicine offers the promise of better, more intelligent care; yet success has proved elusive. 

To better understand this, we will consider, first the broad ambition of the “learning health system,” understand the general challenges presented by electronic health records (EHRs), and then finally, consider the complexity of a topical use case: a consortia’s effort to use EHR data to advance the understanding of COVID-19.

EHRs and the Learning Health System Ideal

For years, ability to utilize information from EHRs has been regarded as the cornerstone of the “learning health system (LHS),” where the “feedback gap” I’ve described is successfully closed, providers are able to efficiently learn from their shared experiences, and the standard of care is iteratively improved. The concept, first introduced in 2007, has been celebrated worldwide, and is generally viewed as the ideal towards which care systems should aspire.

Unfortunately, as a 2016 systematic review of the literature by Norwegian researchers Andrius Budrionis and Johan Bellika somewhat inconveniently revealed, the LHS “mostly remains described in theory,” universally venerated, yet rarely implemented.  

Of the many published papers addressing the topic, only a handful described “actual implementations,” according to the review’s authors, who describe their findings as “rather alarming,” observing:

“It seems like the interest in exploiting the potential of LHS is global; however, it remains expressed in words rather than action. Only 13 publications present initial results and support the impact of LHS by implementation. Many emphasize the potential of the novel paradigm for healthcare delivery; however, empirical results are lacking.  LHS aims to shorten the reported 17 year timespan required to put positive research results into practice. But how long does it take to adopt LHS itself?”

As Budrionis and Bellika archly conclude, “The LHS concept is reaching a level of maturity that puts pressure on impact evaluation” – or to translate from the clever prose of academia, “less talk, more action.” While acknowledging the many theoretical challenges associated with the effort, they argue – exactly as I have emphasized – for the prioritization of concrete, palpable utility.  The challenges, they write, “may be easier to solve if the actual impact is clearly visible.”

EHR Challenges

At the core of the LHS is the EHR.  As I’ve discussed, decoding phenotype – and more broadly, the opportunity to capture phenotype at scale– represents a profound opportunity for clinical care and medical science that the EHR should enable, getting there isn’t easy.

The difficulty of actually using data from EHRs was highlighted in a pointed review two Columbia University bioinformaticists, George Hripcsak and David Albers, wrote in 2013. 

While noting that the “national push for EHRs” makes an “unprecedented amount of clinical information available for research,” the authors explain that extracting value is remarkably hard.  For instance, EHR datasets are generally fragmentary and incomplete, recording only select aspects of a patient’s interaction with a given care center; to the extent it’s even possible to reconstruct a patient’s longitudinal journey, such a “time series …is very far from the rigorous data collection normally employed in formal experiments.”

EHR data is also notoriously inaccurate, and many of the errors are systematic, rather than random – representing, for example, influence “by billing requirements and the avoidance of liability,” according to Hripcsak and Alberts. Sometimes, erroneous data are reflexively entered and subsequently copied.  For instance, the authors point out, in one EHR database, “2% of patients who were missing one eye were documented in a narrative note as being PERRLA [a common acronym for a normal bilateral eye exam] – an impossibility.”

The authors also cite the complexity of healthcare, and note the challenge of discerning clinical reasoning from computable EHR data. Moreover, “healthcare data reflect a complex set of processes with many feedback loops,” with the ultimate effect that “the EHR is not a direct reflection of the patient and physiology, but a reflection of the recording process inherent in healthcare with noise and feedback loops.”

(To this point: Harvard researcher Griffin Weber and colleagues actually demonstrated in 2018 that the presence and timing of many laboratory test orders – rather than the results of these tests – were actually more accurate in predicting survival, highlighting the importance of understanding the care processes associated with data in a given EHR.)

Conclude Hripcsak and Albers, “the full challenge of phenotyping is not broadly recognized,” and note that while “interoperability and privacy” are often cited as key challenges, the more difficult problems, arguably, involve the actual, foundational data of which the EHR is comprised.

A Timely Example: EHR Data For COVID-19

A just-published research paper nicely captures both the hopes and challenges of leveraging hospital medical records for evidence generation.

The 4CE consortium is an international group of nearly a hundred hospitals, across five countries, that formed in response to COVID-19, with the goal of leveraging electronic health record (EHR) information to provide timely clinical and public health information about the pandemic. Most of the group were already participating in an ongoing collaborative effort around EHR use built around the “i2b2” platform, and thus were poised to turn their collective attention to the coronavirus challenge.

Data sharing consortia tend to use one of two approaches: either de-identified data are transferred to a common site and analyzed there collectively (this is also the approach used in traditional, multi-site randomized clinical trials), or the analysis is initially performed locally, in a distributed fashion, using agreed-upon protocols, and the results are then aggregated. 4CE utilizes the distributed model, which individual hospitals often view as offering greater protection to their EHR data.

The just-published research sought to evaluate the disease course and outcomes of hospitalized COVID-19 patients who were severely ill – i.e. required admission to an intensive care unit and/or who died. Yet, such seemingly basic information was “not readily available in all environments,” the investigators report – meaning that at many hospitals, it was not possible to reliably extract these data from the EHR. For example, especially during the early crush of COVID-19, many traditional medical floors (and on occasion, even hallways) were converted into ad-hoc ICUs, so “standardized EHR data elements such as ‘transfer to ICU,’” the authors note, often “would not be properly recorded.”

To get around this, the researchers devised a way to triangulate severity based on a combination of medication codes, diagnosis codes, and lab test orders, and procedures codes – information accessible in the EHRs of all participating hospitals. Such an approach is called a “computable phenotype,” and is often used to assess a clinical state when there isn’t a direct measure reliably recorded in the medical record. According to the authors, “reliable mentions of diseases are rare in the clinic record, and individual diagnosis codes are mediocre predictors of the actual presence of a disease.”

Think about what this means for a moment; although it would seem that there’s nothing more fundamental in medicine than the diagnosis of disease, figuring out from the EHR whether a patient truly has a disease can be surprisingly challenging. 

It gets worse. A computable phenotype turns out to be challenging to define as well, even if this is done (and ideally it should be) with input from astute clinicians.  “A phenotype can make sense clinically yet have more performance due to coding anomalies and variation between sites,” the authors explain, emphasizing the importance of validating the computable phenotype. 

Thus, the entire purpose of the initial study was to develop and validate a computable phenotype to assess whether or not a particular patient’s hospital course was consistent with severe disease, a formula that could be applied effectively at each individual hospital.

Ultimately, the group was able to develop a computable phenotype that had reasonably good characteristics (compared to the best available gold standard — ideally the [painstaking] process of manual review of patient charts) across all the test sites. In the process of constructing the algorithm, the researchers were struck by the differences in use of standardized lab test, diagnostic and procedure codes across the many hospitals, highlighting the challenge inherent in efforts to utilize EHR data from multiple care systems.  Billing codes denoting ICU admission, the authors note, were particularly imprecise; many ICU stays were missed.

The exceptional amount of work required to robustly determine, from EHR data, a patient attribute as seemingly basic as severity of a COVID hospitalization, and the complexity associated with this undertaking, highlights the gap between the apparent richness and bounty of EHR data, and the incredible difficulty of extracting even fairly rudimentary insights. 

(In this context, it is perhaps easier to appreciate why many informaticists were immediately skeptical of the improbably rich EHR-derived datasets underlying two high-profile COVID publications this summer – both of which were subsequently retracted, as I’ve discussed.)

Significantly, in a point that is often overlooked, the success the 4CE researchers were able to achieve, they note, was buttressed by local clinical expertise – specifically, doctors at the individual hospitals “who understood the vagaries of hospital coding” and helped improve “data extraction and analysis, thereby contributing to the data quality of the 4CE initiative.” 

The importance of such expertise was explicitly highlighted by consortium authors in an earlier paper as well: “Most importantly,” they write, “at each site there were biomedical informatics experts who understood both the technical characteristics of the data and their clinical relevance.”

Bottom Line

The information in electronic health records offers great promise in the care of patients and the understanding of disease, but the exceptional difficulty of fulfilling this promise is widely underappreciated. 

While improved technologies will almost certainly prove useful, it’s critical to understand the often very local clinical care process around which EHR data are generated.  It was Amy Abernethy’s astute recognition of the primacy of high quality, EHR-derived, clinical expert-curated datasets, for instance, that arguably created the fundamental value in Flatiron’s oncology data platform, subsequently acquired by Roche for around $2B.  (Abernethy is now the Principal Deputy Commissioner of the FDA.) 

The true value of EHR data lies not only, or even primarily, in the volume of data captured or computed.  Rather, it’s in the ability to render, organize, and utilize these data in a fashion that captures and reflects the clinical circumstances and care processes associated with the data’s original generation.


Rebuffed as Overlords, AI Experts Return in Peace, Seeking Partnership with Clinicians

David Shaywitz

Why not healthcare?

That’s the core question at the heart of efforts to apply emerging digital and data technologies to healthcare and life science. 

As Suchi Saria, an entrepreneur and a computer scientist at John Hopkins, where she directs the Machine Learning and Healthcare Lab, puts it, in the 2000s, these technologies transformed sectors, such as banking, in a fashion that was “kind of amazing.” 

Artificial intelligence (AI), operating on these rich data, profoundly changed and improved the way business is done. 

Take, for example, banking fraud detection. The industry “can’t imagine doing it without AI, and with AI they’ve increased sensitivity dramatically, timeliness dramatically” and have far improved specificity, Saria notes.

Contrast this with the AI experience in healthcare. In the last decade, healthcare has “basically spent a ton of our investments” to go “from no data to data, going from no digital infrastructure to additional infrastructure,” says Saria, but yet, “when we think of AI, we think of it as a thing that could be transformative, that has the potential, that is in the future.”

Suchi Saria, Founder, Bayesian Health; John C. Malone Associate Professor and Director of AI & Healthcare, Johns Hopkins University

What we’re missing, insists Saria, is that the future is now – “in reality, today. Now that the data exist, the use of AI in deriving value from the data that’s being collected is the single biggest opportunity in healthcare.”

It’s a hopeful perspective, though of course not universally shared.  But it is hotly debated, perhaps nowhere more intelligently than at the recent, inaugural SAIL pre-symposium (virtual, of course) focused on AI and health, and featuring many of the field’s most thoughtful voices, including data scientists, clinicians, administrators, and even the Editor-in-Chief of the august New England Journal of Medicine (NEJM). All offered comments that were almost invariably germane, focused, and informative, representing the best of what conferences can be. You can watch the whole thing yourself here

Six major themes emerged in my notes:

  1. The challenge of outcomes – what should we, and can we, seek to optimize?
  2. The centrality of bias – the need to ensure AI isn’t perpetuating and exacerbating inequities.
  3. The consensus for the “doctor and AI” mindset – rather than “doctor or
  4. Promising use cases – the “green shoots.”
  5. Opportunities in evidence generation – and why leveraging electronic medical record data remains so hard.
  6. Stubborn hurdles and implementation challenges – including interoperability, data access, and conflating “interesting” and “important.”
The Challenge of Outcomes

Almost by definition, the goal of medicine is to improve outcomes. As NEJM editor Eric Rubin puts it, “we are interested in the impact on the patient,” adding “the closer we can get to something that we care about, the better off we are.”  

Similarly, the lens through which UnitedHealth Group’s Chief Scientific Officer, Ken Ehlert, views potential AI solutions is “are we getting a better outcome?” Entrepreneur and academic ophthalmologist Michael Abramoff also stresses the importance of focusing on outcomes.

But such focus turns out to be easy to say but far more difficult to operationalize, as Harvard’s Zak Kohane points out.

“We’re not very good at looking at outcomes because the systematized capture [of outcomes], whether in trials or EHRs, is noisy, confounded.” He predicts that “many, if not all the AI programs that are going to be deployed in the next 10 years will be poor with respect to outcomes and rich with respect to either human labels or intermediate process measures.”

Zak Kohane, Chair of the Department of Biomedical Informatics, Harvard Medical School

These endpoints can seduce and mislead us, he suggests, leading us to optimize for something we regard as a proxy for meaningful outcomes, yet which may ultimately not be linked to the outcomes as closely as we’d like to imagine.

Kohane (a pediatric endocrinologist) cites the example of diabetologists seeking to improve microvascular disease by focusing on driving down the levels of glycosylated hemoglobin (HbA1c); this turns out to work, to a point, in terms of reducing kidney damage, but pushing “too” intensively for very low HbA1c levels was ultimately found to increase the risk of death. We were “misled by the process outcome in this case,” Kohane says. “For adults, minimizing glycohemoglobin was actually the wrong thing.”

“Medicine,” reflects Kohane, “is a beautiful art but it’s barely a science.  As a result, many of our intuitions of what constitutes a solid correlate to outcomes, again and again, gets proven to us to be wrong.”

It’s also important to recognize, as University of Utah Health’s Chief Medical Information Officer, Maia Hightower, points out, that “the outcomes that we as clinicians may see as important may be different than what our communities see as important.”

The Ubiquity of Bias

Despite, or perhaps because, of a series of high-profile failures (like an AI-powered image classifying program that was able to recognize categories as fine as “Graduation,” yet mislabeled people of color as “Gorillas”), the AI community has tackled this challenge. The AI community has transformed itself from laggards to leaders, as Brian Christian captures in a captivating new book, The Alignment Problem, that I recently reviewed for the Wall Street Journal.

Duke University computer scientist Cynthia Rudin highlights a prismatic example of bias – an insurance company algorithm that aimed to predict which patients might need more care in the future, and thus might benefit from extra (anticipatory) services today. 

The company used “cost as a proxy for care,” in their modeling, Rudin says. “The only problem is that black patients were receiving lower cost health care. They weren’t less ill. They were just receiving lower cost health care.” But these patients would have been systematically underserved by the algorithm’s recommendations. 

Anant Madabhushi, who directs the Center for Computational Imaging and Personalized Diagnostics at Case Western University, offers another example from his own research on prostate cancer.  Black men “tend to have more severe disease,” Madabhushi says, and “potentially higher incidence of prostate cancer,” yet “a lot of the existing risk models that we currently have for prostate cancer have been built largely with a plurality of non-black men represented in those datasets.” 

Attuned to the possibility of racial differences in the disease – as IBM Watson’s Tiffani Bright points out, “you can’t measure what you don’t know about” – Madabhushi uncovered “actual differences in the area around the tumor” in pathology specimens taken from black men and white men.  From this, they “created a dedicated model” that “resulted in a much higher accuracy in predicting risk of recurrence,” compared to a “population-agnostic model.”

At one point, there might have been a collective sense that the best way to avoid bias is to avoid collecting data that might predispose to bias, like race. But a key theme emerging from both this discussion and Christian’s book is that appropriately collecting and thoughtfully considering these data can be essential and invaluable.

“From an operations perspective,” explains Hightower, there’s now the “expectation within the healthcare system [that we’re] capturing all other types of data to tell the complete story of our patients.” She says they “do gender pretty well,” but are “not as good with race. And definitely when we talk about preferred language and LGBTQ+ status, it starts to deteriorate even more.”

As Kohane points out, such information can be critical. Consider hereditary breast and ovarian cancer (HBOC), associated with mutations in the BRCA1 and BRCA2 genes. Ashkenazi Jews are at significantly greater risk of carrying one of these mutations; according to a genetic counselor at Jackson Laboratory, “one in 40 Ashkenazi Jewish individuals versus one in 400 people in the general population carry a mutation in BRCA1 or BRCA2.” If a provider does not customize their calculation, “if they do not discriminate based on the ethnicity of the patient being a Jewish woman,” Kohane notes, “they are actually underserving that patient in a very unfortunate way. And I think that [customizing treatment based on factors that, where appropriate, include ethnicity, for example] is going to be increasingly true as we get more precise about our medicine.”

Marzyeh Ghassemi, Assistant Professor, University of Toronto, Computer Science and Medicine

Marzeyeh Ghassemi, a Canadian computer scientist focused on the application of machine learning to healthcare, and whose lab will be moving to MIT next summer, has thought deeply about fairness and bias. She is especially concerned about “unjust bias” – bias that “perpetuates systemic structural injustice that’s been visited upon a certain group for many reasons.” 

One example she cites: the historical tendency for women’s pain to be “ignored when they go to the doctor.” Consequently, she says, if we feed that data into a model, “we make an algorithm that perpetuates that structure, systemic injustice. That’s a bias that’s bad, and we don’t want to do that.”

One specific suggestion Ghassemi offers: regulators requiring “performance guarantees across different subgroups.” She emphasizes considering such performance proactively, and demonstrating it, represents a better solution than perhaps the more convenient alternative of “narrowing the scope of your claims” – i.e. seeking approval only for a single group. Otherwise, she says, “we going to end up with a lot of devices and algorithms and bells and whistles and treatments that only work on wealthy white people.”

Kohane points out that the computer science community was two decades ahead of medicine in embracing open access publishing. “There is an interesting set of precedents where good societal behavior has actually been pioneered by the computational community,” he says, suggesting that perhaps computer scientists, as they seek to bring AI to medicine, could set another good example here as well.

Doctor and AI

While many journalists seem permanently stuck on the “will AI replace doctors?” storyline, the field itself moved on a long time ago, driven, it seems, less by political expediency – the idea that AI will be an easier sell if doctors are less threatened by it – than by authentic scientific humility.

Many leading AI practitioners have recognized the limitations as well as the power of their computational tools, and see in partnerships with people an opportunity to at once bring out the best from both computer and human while also guarding against some very real concerns.

Among the top worries: the unexpected fragility of AI algorithms. Approaches that seem to work brilliantly in a defined set of circumstances may fail catastrophically when the situation is changed – even imperceptibly. 

For example, fascinating studies involving “adversarial attacks” have revealed that a seemingly sophisticated image recognition algorithm can be tricked by simply altering, in some cases, a single (well-chosen) pixel. Similarly, AI researchers working in healthcare have become increasingly worried about black box algorithms delivering misguided recommendations based on subtle flaws that may lurk undetected – as I discussed at the start of my recent WSJ review of Brian Christian’s The Alignment Problem.

As Duke’s Rudin (also one of the stars of Christian’s book) explains, right now, when thinking about the application of AI to many aspects of healthcare, “we don’t trust our models.” They might be “reasoning about things the wrong way.” 

To leverage the power of AI while mitigating the risks, Rudin’s group is focused on using AI to develop clinical decision aids that are based on simple point schemes, to derive the sorts of scores that physicians are already accustomed to calculating – to ballpark a patient’s cardiovascular risk, for example.  

The trick is using the sophisticated AI to figure out the most relevant variables to measure (a computationally difficult problem) and distill these parameters into simple integers, which a busy physician can still add up and evaluate in the context of the patient.

Cynthia Rudin, professor of computer science, Duke University

The scores produced by this approach, Rudin says, “are just as accurate as any model you can construct.” Plus, they offer the conspicuous advantage of being interpretable – the physician can “really understand how the variables work together jointly to form a final prediction,” she adds. Furthermore, “being able to have the human in the loop actually helps you with the uncertainty that you can’t quantify – the ‘unknown unknowns.’”

There’s also a hope that AI can help aggregate, organize, and prioritize the huge amount of information physicians and other health providers need to contend with, helping to distill for them the information they need – when they need it.

Both Microsoft’s head of research Peter Lee, and UnitedHealth’s Ehlert, for example, envision AI “augmenting what humans can do, absorbing and integrating knowledge for better decisions” as Lee puts it. The ability to process health information in real time, Saria believes, will enable medicine to (finally…) transition from a “reactionary paradigm to an anticipatory paradigm,” anticipating disease in time to prevent it or at least head it off at an early stage.  For example, argues Saria, “AI is pretty much the only way to identify conditions like sepsis, and patients at risk of sepsis, early and precisely.” 

Saria cites the management of stroke as another example, where AI can rapidly identify likely occlusive events, which a radiologist can immediately review and potentially validate, facilitating the timely triage of patients to a comprehensive stroke center for appropriate treatment.

Both Saria and Ehlert also flag the opportunity for AI to offer providers reference values for measurement that are personalized and contextualized for each patient, rather than based on average values for the population as a whole.

Columbia University biomedical informaticist Nick Tatonetti may have captured the shared sentiment the best, observing:

“Medicine is really about human interactions. Caring for somebody and curing someone of a disease is an extremely human activity. And humans should be centered in that process. A lot of technology that’s been introduced in health care has been rightly criticized for getting in the way of the patient doctor relationship, that human connection. There really is an opportunity for technology not to get in the way any longer, but start to disappear into the background and really put that interaction in the center.”

Uses Cases

Several speakers offered concrete examples of the application of AI in medicine. A particularly intriguing example, presented by Greg Hager, a computer scientist and director of the Malone Center for Engineering in Healthcare at Johns Hopkins, focused on surgical training. 

Hager explains that the widespread adoption of the da Vinci surgical system, which enables surgeons to operate with robotic assistance, almost as if they’re playing a video game, offered a remarkable opportunity. The da Vinci’s recording of all aspects of a surgical procedure, from stereo videos to the force applied to the instruments, Hager realized, generates a fantastically useful dataset.  By thoughtfully bringing AI to bear on these data, Hager and his colleagues break procedures down into steps, and “evaluate the quality performance of those steps.”

This analysis can determine “whether it’s an attending [senior physician] or a trainee who’s operating, just by the quality of the performance in that data.” Plus they can feed the data back into training.  “Once we know where you lie in the skill scale,” Hager says, “we can understand where your potential deficits are, and we can turn that into a training regime so we can now say, look, here are the things that would be most useful for you to work on to improve your surgical technique.”

Perhaps not surprisingly, other use cases involved imaging. 

Hager, for example, described the development of an algorithm intended not to replace radiologists, but rather, to enable them to use their skills most effectively. The approach he described would analyze mammograms, and sort them “extremely reliably” into two categories: clearly normal and everything else. This would “use the machine to replace the drudge work, the over and over again work,” and instead “allow radiologists to focus more at the tip of the pyramid, the place where there’s really high value and [the critical need for] human input.”  As he summarizes, “We should be thinking about augmenting people,” and says the way to do this is “to allow them to focus on the place where people have the most value.”

Pathology offers another promising opportunity for the application of AI, Case Western’s Madabhushi points out. In an approach similar to Hager’s, pathology slides might be pre-filtered by a measure of complexity, with the most difficult cases presented to the pathologist when she is the most alert.

He also cited an Israeli company whose software provided “second opinion” reads, reviewing slides that pathologists had already identified as benign. This (theoretically) minimizes the downside risk, while enabling the identification of lesions that initially escaped human detection. (Of course, the concern would be the Peltzman Effect – the worry that pathologists might become less diligent in their initial reads if they thought a computer was likely to double-check their work.)

Ken Ehlert, chief scientific officer, UnitedHealth Group

More generally, both UnitedHealth’s Ehlert and Microsoft’s Lee express hope that AI could also improve our understanding of biology, and complex biological networks. Ehlert also suggests a useful function for AI could be producing for each patient a “patients like mine” function. The idea is that it would be enormously empowering for physicians if an algorithm could review data from millions of patients, presenting the doctor, in real time, with information about how similar patients fared, and what treatment approaches worked best. (This capability, as I’ve described, is painfully absent today.)  

Lee, meanwhile, points to the opportunity for AI to provide doctors with “an intelligent assistant” function, “ambient clinical intelligence” that could listen to a physician interact with a patient and set up a clinical note for her accordingly. 

From EHR to Evidence (?)

The digitization of the electronic health record would seem to present an enormous opportunity for learning and care improvement. Yet, as I’ve repeatedly and perhaps obsessively discussed (see here, here,  and references therein), delivering on this promise has proved exceptionally difficult.

One issue seems to be a misunderstanding of what an EHR is, and isn’t. Essentially, we tend to think we are directly learning about patients, yet what we’re really learning, Kohane reminds us, “is the behavior of doctors.” He adds, “most events, most of the data items, are created by the doctor. So you’re actually learning from the doctor. You’re not learning from the biology.” We need to recognize the difference between the two, he cautions. 

Kohane draws a contrast with the relative feasibility of using images as a substrate for AI.  While acknowledging potential sources of variability in images (a biopsy of a heterogeneous tumor, for example, may happen to catch an unrepresentative sample), Kohane explains,

“I’m going to be much more confident about image-based metrics than I am about  time series, EHR-based metrics, because I just know how much more variation there is [compared to] the slab of tissue that’s obtained in the OR or the retinal image. I assure you, it’s less than the practice of medicine in different cities, and how current, aggressive or venal different doctors are in different systems. That’s going to make our evaluation process for those algorithms that are very doctor-in-the-loop dependent, quite tricky to evaluate.”

Rubin of the NEJM also points to challenges of relying on EHR data. “As someone who contributes to the charts all the time, I’d say that a lot of them are driven by insurance claims rather than caring for the patient or getting the most complete collection of information on that patient.” 

Adds Rubin, we need to think about “the purpose of the data that we’re going to collect, because if we can anticipate that purpose, we can do a better job of collecting data that fulfills that purpose.”

Yet even with their limitations, EHRs still capture data that would seem to be valuable, and provide the opportunity for insight, albeit not through the traditional, gold-standard mechanism of a typical randomized control trial (RCT), with its distinct, highly specified methods of data collection and analysis. 

This presents a dilemma. On the one hand, as Saria asserts, “our reliance on RCTs alone for evidence generation is dramatically slowing down the rate at which we can learn from our data.” 

Rubin’s response (effectively representing the broader medical establishment) is measured. He agrees both that RCTs are “the gold standard right now,” and that they “have tremendous limitations…right now, you can really only ask one question, and it can take 10 years and $100 million to get the answer to that question.” He notes that data from EHRs — “real world data” — is “fundamentally different, and it’s a work in progress to figure out how to make that rigorous,” adding “we have to figure out how to bring rigor, and how to understand the rigor within trials that are not traditionally designed.” Emerging disciplines, Rubin said, need to think about “what do they consider as rigorous,” and to develop “reasonable standards” that can be used for evaluation.

The challenge of leveraging EHR data for evidence generation was experienced directly by Microsoft’s Lee, in the context of his work with the Mayo Clinic on their effort to evaluate convalescent plasma for COVID-19 therapy. This effort enrolled over 100,000 patients, treated over 71,000, and resulted (famously or infamously, depending on your point of view) in the FDA granting Emergency Use Authorization for this treatment.

Peter Lee, Corporate Vice President,
Microsoft Research & Incubations

As Lee candidly describes it;

“For the vast majority of those 71,000 patients, there was tremendous access to clinical histories in electronic form. And so it was almost a perfect situation in modern era where we ought to be able to take all of that clinical experience in a compressed time frame and extract information about safety and efficacy of that experimental therapy…but the process was extremely difficult. And in fact, ultimately, the data was pretty impoverished. And so the amount of instrumentation that we need, the amount of forethought in clinical practice so that the digital exhaust of what we can learn from that clinical practice really feeds into advancing science and ultimately regulatory approvals. Altogether, this is still, in my mind, absolutely the future, but is just much more difficult and subtle than at least I had realized, even as short as one year ago.”

While the spirit to learn from EHR is willing, the flesh (or at least the requite infrastructure), it seems, remains weak.

Hurdles and Barriers

Perhaps predictably, two impediments to the productive application of AI to medicine emerged from the discussion: data sharing and implementation.

The challenge of data access, long lamented, remains a serious problem – perhaps the most significant problem, according to a clearly exasperated Rudin – in bringing AI to medicine. 

Rudin notes:

“The thing that’s stopping a huge amount of scientific research in health care and AI is lack of data. It’s not an AI question, but if we could solve it, it would give a lot of AI answers…you can’t even reproduce a lot of the scientific papers from a few years ago. Sometimes you e-mail the authors, like the lead author of the paper, and they say, well, I never actually had access to that data in the first place. That was done somewhere else by somebody else. Then you email that person and they don’t have the data. It’s impossible to get it. So how are you going to estimate the effect of drugs? How are you going to, you know, reproduce any scientific study and do a better job of it using machine learning if you don’t have access to the data?”

While acknowledging the “tradeoff with privacy,” Rudin says that “if we cannot figure out ways to make data available for scientists to use, then AI is just going to continue to not be used in hospitals, that’s all I can say.”

Asked if the culture, perhaps, is starting to shift, Rudin is blunt: “No, it’s not.” 

Even COVID-19, it seems, couldn’t motivate the necessary change.

Rudin expected that when the virus hit:

“We would be getting e-mails from everywhere saying, hey, we’ve got a bunch of data on COVID patients, here you go — [a dataset that has] a whole medical record for everybody with the COVID information and their survival and all this stuff. No. There’s a few databases that supposedly are available. But the truth is, they’re not. There’s a lot of barriers to even to get into those databases.”

The gap between the many high-profile, data sharing consortia that have sprung up with great fanfare in response to the pandemic, and the apparent difficultly experienced by a top academic computer scientist trying to use these data, seems disappointing (though I might add: hardly surprising).

Microsoft’s Lee also lamented the challenge of accessing the data needed during the COVID-19 crisis.  Working with hospitals in Seattle and elsewhere, Lee says, “we saw this really completely, vividly.”

As Lee tells it, in the early days of COVID-19, it was “critically important” for hospitals and hospital systems “to understand what patients are being seen, what COVID-19 encounters were taking place, what capacity do we have to treat those patients properly? And that capacity is hospital beds, ICUs [intensive care units], PPE [personal protective equipment], testing and so on. And then how is that capacity being utilized?”

Yet, “despite the incredible digitization over the past decade and fifteen years in all manner of health care operations, what we found was we still had frustrating inability to sort of connect the digital dots here,” Lee says. “We had things like PPE tracked in digital ERP [enterprise resource planning] systems, we had encounters uncoded, but in free form text in EHR systems. And we had very little understanding of utilization.”

Even worse, explains Lee, “in terms of fundamental data interoperability standards, we couldn’t quite connect rapidly the identities of people across these various digital silos.”

While Lee insists he’s “optimistic about the future, because the world, and particularly the US, is moving rapidly to address these interoperability issues,” he emphasizes that the crisis forced many to “confront firsthand…some of the work that we still have to do.”

Beyond the (palpably traumatic) interoperability challenges several speakers highlighted, an additional difficulty apparent in the translation of AI into practice concerns. 

In particular, both Saria and Ehlert emphasize that just because a problem is either “interesting” or “solvable,” and might represent an attractive academic research project, that “doesn’t mean it’s actually interesting to be solved in regular practice and regular life,” as Ehlert put it.  He cited the example of a startup that “had put an enormous amount of energy” into developing a device that would use an automated system to instantly measure your height as you walked through the door. “I didn’t realize we had a height problem in health care,” he quips.

Observes Saria “in academia, we think a lot about publishing papers that show new models and evaluating performance of models. But when you start turning into practice,” you need to start “really thinking through all of the uses cases and thinking about harms versus benefit analysis.”

Related more broadly to implementation, Ehlert also highlighted what may be the most significant challenge of all: alignment.  The issue, he points out, isn’t for-profit vs not-for-profit institutions.  “The reality,” he astutely observes, “is everybody has a stakeholder.” He continues, “a hospital does well when there’s more admissions. A physician does well when they treat more patients. A pharmaceutical does well when they sell more pills. An insurance company does well when they manage the risk better.”

Noting the many different business models these different organizations have, it’s perhaps not surprising that, as Ehlert suggests, “One of the biggest issues I think that we all struggle with is how do we get alignment across those things.”

Perhaps, Ehlert says, we need to “look at our fellow humans” and agree “that our real goal” is for “people to have a better health outcome, and have the highest quality of life for the maximum number of years possible.”  

If we’re all aiming for that shared goal, adds Ehlert hopefully, we should be able to “actually align people” to ensure that the “data is collected the way that it needs to be collected to actually make that happen.”

He’s right, of course. But it’s a big “if” – and a big “should.” 

Note: some quotes have been very lightly edited for clarity.


What Happened in Switzerland?

Alex Mayweg, managing director, Versant Ventures

Back in March — during the first wave — I reflected on the COVID-19 situation in Switzerland. This small country, at that time, was managing its outbreak and quickly getting it under control.

This was just as the federal government had begun coordinating a response, which had previously been left to local authorities. As I mentioned back then, while Switzerland is rarely the first to take measures or adopt something new, when the country does, it tends to take action extremely well.

Switzerland never formally and fully locked down, but the population adhered to the social distancing rules. And the results spoke for themselves — the numbers during May and the early summer faded to nearly insignificant digits. Switzerland’s flattened curve became the envy of many European countries.

While many Swiss citizens had cancelled the summer holidays they had planned to beautiful locations all around Europe, most (including me!) regretted doing so when the summer arrived and almost seemed — well — sort of normal. Perhaps a bit too normal.

The youth took the opportunity and enjoyed the warm sunny weather in Switzerland, hanging out with friends on the beautiful rivers, lake shores, and mountain resorts. And who would blame them after such a spring?

The mountain resorts recorded strong occupancy in the summer months. It almost seemed like most of Europe decided that hiking in the Swiss mountains might be a reasonable vacation destination, where social distancing — if needed — might be easier. Outdoor and camping equipment became extremely popular purchases.

And while some social distancing and hygiene measures certainly persisted, very few wore masks (except on public transport), as the federal government also returned the power back to the 26 local authorities – known as cantons. Each canton implemented its own measures — or lack of — which could be well justified as the case numbers remained low.

In August and September the numbers started creeping up. Cases rose particularly among the 20-29 year old age group. The famous R-value crept up and hovered just above 1, but after some weeks of increase, there were also brief periods of decreasing case numbers. Were we controlling the situation due to our heightened sensitivity, education and Swiss-efficiency hand washing?

Schools re-opened and mask wearing was still minimal. That changed when some cantons, such as Basel Stadt and a few others, reintroduced masks in supermarkets and shops but not in all public areas. Surprisingly, some cantons didn’t require masks in most public areas. The number of new cases was steadily increasing, but the curve was still pretty flat.

Since most of the cases were in young people, there was very little added pressure being put on hospitals. But even if corrected by age-group, the hospitalization and mortality rate looked much better now than it did during the initial first wave. Doctors were learning how to better treat patients in the hospital, and a couple of new medicines – dexamethasone and remdesivir – were available for those in need. Things seemed to be relatively under control.

Then the autumn school holiday break arrived. Again, the mountain resorts were packed. In many areas mask wearing was minimal. With cooling temperatures, guests were enjoying their meals or afternoon refreshments indoors or in mountain huts. Northern Italy was also a popular destination as the numbers there were even lower than in some parts of Switzerland — so why not go? You’ll never see Venice with this few tourists!

About two weeks later,  cases shot up sharply. Suddenly, we saw a switch from local clusters of COVID-19 to a nationwide surge of cases. Hospitalizations spiked.

What were the factors driving this new nationwide spread?

  1. Mixing of people to new locations during holiday time, which also allows further mixing between age groups
  2. Colder weather with more people spending time indoors together and beginning of the heating season providing drying air and less circulation indoors.
  3. Patchy COVID-19 measures across different cantons, generally insufficient social distancing and limited use of masks.

Within two weeks, Switzerland went from one of the best in Europe to one of the worst, eclipsing the UK, Spain, the Netherlands and also France in new cases per 100K inhabitants.

About 10 days ago, when the sharp incline of the case curve surpassed many European countries, the federal government once again took back control from the inconsistent local authorities and introduced mask law in most public indoor spaces. Once again, the federal government issued a strong recommendation to work from home.

While this was important, I could not help notice that many other European countries had already taken these measures, or even tougher ones, a while ago.

On Wednesday, Oct. 28, the federal government announced further measures, such as requirement for mask wearing in certain public open-air areas, early closing times for bars and restaurants and further limitations for gatherings of certain group sizes. Rapid Covid-19 tests will also be rolled out.

The federal government stopped short of the anticipated “circuit breaker lock downs” that may still yet be implemented next to help flatten the curve as cases are spreading into higher age groups and hospital occupancy is going up.

The number of deaths, which usually increases a few weeks after a surge in new cases, is unfortunately once again following this sad, predictable pattern.  

It is hard to know if too little is being done too late. The lack of measures and discipline in the summer are certainly an issue, but also understandable given the very low numbers of infections at the time. It’s difficult for people to remain vigilant when the threat isn’t visible.

The next weeks and months look to be a bumpy ride. We must do everything now to get this back to some reasonable control. But once again I have faith in my fellow citizens that with good discipline and by working together (albeit socially distanced!) we can hopefully manage this next wave as well or even better than the last.

My hope is that with our united best efforts over the next months and, at some point next year with the aid of emerging vaccines, new drugs, further refinement of treatment protocols, and perhaps a bit of “earlier than expected herd immunity” we can get on top of this virus.

If we do, we can eventually start the Roaring Twenties of this millennium.


The Battle for the Soul of Biopharma: Peter Kolchinsky on The Long Run

Today’s guest on The Long Run is Peter Kolchinsky.

Peter is the managing partner at RA Capital Management. The Boston-based firm invests in public and private life sciences companies with a total of $6.8 billion under management.

Peter Kolchinsky, managing partner, RA Capital

Peter is a virologist by training at Harvard University. It’s obviously a valuable set of skills to have in a year like this. But he’s been spending more time lately on the future viability of the biopharmaceutical ecosystem itself.

The industry, as everyone ought to know, is in deep trouble. Public anger over high drug prices has been festering for years, with no real solution being offered by either party in Washington DC. Peter has seen the good that industry can do with creating new medicines, as well as the misdeeds that have been committed by those in the insurance industry and in pharmaceuticals.

In January, he wrote about some proposed solutions in his book, “The Great American Drug Deal.” It’s brilliant, unorthodox in its solutions, and easy for a layperson to read – unlike a lot of health policy books.

Peter hasn’t been content to just say his piece in a book – he’s now spearheading a nonprofit called No Patient Left Behind. It seeks to advocate for some of the ideas described in the book, urging biotech executives, leaders of industry trade groups, and members of Congress to get on board. He sometimes seeks to advance his thoughts as an occasional contributing writer for Timmerman Report.

In this conversation, I skipped Peter’s backstory. We talked about the crisis in biopharma and the broken profit incentives that have created this mess. He sees an internal battle within the industry between builders who develop innovative new drugs and other products, versus the landlords who are more interested in rent-seeking behavior to maximize profits of old blockbusters. He calls this a “battle for the soul of the industry” – perhaps adapting a phrase from a certain candidate’s rhetorical framework this election year.

One small note before we start: Peter was joined on this call by Chris Morrison, a veteran biotech journalist now working as an editor at RA Capital. You’ll hear Chris chime in at a couple of points, once to clarify a patent expiration date, and another time to remind Peter of the name of an author he was trying to cite.

Peter is brilliant and fiery in this conversation. Anyone who cares about the future of biotech ought to give this a listen.

Now, please join me and Peter Kolchinsky on The Long Run.


How Many COVID-19 Deaths Could Have Been Avoided? More Than 150,000

Ruth Etzioni, PhD, Full Member, Division of Public Health Sciences, Fred Hutchinson Cancer Research Center

For a numbers person like myself, the SARS-CoV-2 pandemic is supremely frustrating. Even after 8 months, there are so many numbers that we still do not know.  Even though there are thousands of researchers trying to fill in the gaps in our knowledge, the gaps persist.

As the officially recorded US death toll from SARS-CoV2 approached 200,000 last month (it’s now over 224,000), one missing number in particular began to disturb me. While I agreed with the many voices blaming the abysmal national response for the grim death count, I knew that even if we had responded optimally, the number of COVID-related deaths would not have been zero.

But how many deaths might we have expected had we responded quickly with sufficient testing, effective contact tracing and compliance with masking and social distancing?

I couldn’t find any published guesses. So I did it myself (with some help, acknowledged below).

I took as my model the country of Germany. It is frequently cited as having “crushed the curve” in March and April, because of a combination of rigorous contact tracing, quarantining, and near universal compliance with local and national lockdowns. In fact, Germany managed to keep cases and deaths ultralow through the summer.

But COVID-19 is a virus that tends to resist tidy narratives and easy explanations. Like much of Europe, despite its early accomplishments, Germany, too, is coping with a new spike in COVID-19 cases. On Sept. 22, when the official US death toll surpassed 200,000, Germany was still in good shape, recording new cases at levels only slightly higher than its summer nadir.

I asked the following question: If the US had been as successful as Germany in managing the virus, how many deaths would we have expected by September 22? By “as successful as Germany.” I meant, “if the likelihood of COVID-19 death in the US were the same as in Germany.”

It is not enough to simply translate the COVID-19 death rate per capita in Germany into a number of US deaths, because the age and demographic structures of the two countries are quite different. The vast majority of Germans are of European ethnicity. In the US, 18 percent of the population is Hispanic or Latino and 13 percent are African American. And the German population skews slightly older.

To accommodate the demographic differences between the two countries I used age-specific cumulative death rates in Germany from the Robert Koch Institute as my baseline for non-Hispanic whites. But African American and Hispanic/Latino populations in the US have had markedly higher COVID-19 mortality rates than non-Hispanic whites. Even if the US had responded properly to the pandemic, I do not expect that it would also have solved its persistent problem of racial disparities in underlying health, which makes minority populations especially vulnerable to a pandemic.

So, for these minority populations I inflated the Germany-based death rate using race-specific death rate ratios (relative to non-Hispanic whites) from the CDC. Putting this all together, I (and my co-authors) concluded that, had we been as successful as Germany at managing the virus, the US would have had 43,187 deaths rather than the recorded cumulative total of 200,000 on Sept. 22. You can read the abstract and full manuscript here at MedRxiv.

I stared for a while at this shocking result. It was much worse than anticipated. These findings suggested that almost 80% of the 200,000 lives lost – 160,000 people! – could perhaps have been saved by a response that met the gravity of the threat, a response that went all out to break the chain of community spread in order to protect American lives. Even though my result was just an approximation, I felt like it was an important data point. I wanted to publish it right away. But I knew better.

I have seen over and over again how science has been politicized during the pandemic. I wrote in these pages about inappropriate, politically motivated interpretations of the changing IHME COVID-19 model projections, and the appropriation of the Mayo Clinic convalescent plasma study for political expediency. I knew that my result could be used to support a variety of political stories.  Mostly I was concerned that it would be dismissed by Administration supporters because of its limitations, chief among them being that the US is not Germany.

So I tried to make it politicization proof. I went through all of the relevant differences that I could think of between the two countries and explained why my result was still defensible. Yes, there are cultural differences, and greater trust in the government in Germany. Yes, Americans are more likely to be obese. And yes, it is possible and even likely that what counts as a COVID death in Germany is not the same as in the US. I re-classified my work as a “thought experiment designed to provide a first quantification of a best-case scenario in this country.” I still thought it deserved to be published.

Unfortunately, the journals that I submitted it to did not agree.  I quickly received two flat-out rejections from journals that are known for publishing commentaries on the state of public health in this country.

Was the work simply too hot to handle? Were my methods at fault? Were the limitations deal breakers?  Did the editors disagree with my message?

I’ll never know. But at least my science has been validated. This weekend CNN published an article titled: “Faulty US COVID-19 response meant 130,000 to 210,000 avoidable deaths, report finds.”

Finally! I breathlessly followed the link and arrived at a report from a very reputable group of researchers at Columbia University. They had exactly the same idea as I had, but they looked not only at Germany but also at South Korea, Japan, Australia, Canada, and France.

By the time they posted their report, the US death toll stood at 217,000. The report concluded that “had the U.S. government implemented an “averaged” approach that mirrored these countries, the U.S. might have limited fatalities to between 38,000 to 85,000 lives – suggesting that a minimum of 130,000 COVID-19 deaths might have been avoidable given alternate policies, implementation, and leadership.”

In academia, there is a term for when someone publishes your idea before you have a chance to do so. It’s called getting “scooped.” Normally, it’s painful. But this time I am not at all perturbed. I’m delighted that others agree that this information is urgently needed and that they are equally determined to unearth and disseminate it.

We have a number for the human cost of our disastrous national response to COVID-19.

Now we just need to make sure everyone knows it.

Thanks to my co-authors on the article, Ivor S Douglas MD and Elan Markowiz, future PhD.


Machine Learning for Drug Discovery: Daphne Koller on The Long Run

Today’s guest on The Long Run is Daphne Koller.

Daphne is the CEO of South San Francisco-based insitro. The company is seeking to develop a new platform for drug discovery that leans on a combination of wet labs and machine learning algorithms to spot new biological targets for drug discovery.

Daphne Koller, founder and CEO, insitro

Artificial intelligence and machine learning have been stirring imaginations in biopharma, and generating more than a little hype, for a few years now. Smart people who know how complex drug discovery is, are pining for something to make drug discovery less risky, less time-consuming, and less expensive. Insitro represents one of the prominent wagers in the AI for drug discovery category, as it raised a $143 million Series B financing in May led by Andreessen Horowitz, and which included earlier investors like ARCH Venture Partners, Foresite Capital, GV, and Third Rock Ventures.

Daphne comes to this challenge with a fresh set of eyes. She was a math prodigy as a kid. She became a tenured professor of computer science at Stanford University. She became a successful entrepreneur, co-founding Coursera, the online learning platform. But when it came time to find challenging new problems to solve intellectually, and important problems to solve for humanity, she decided that tackling disease was the place to be for her next act.

We talked about her journey, and some of the key aspects of her strategy to make insitro succeed in this new frontier.

Now, please join me and Daphne Koller on The Long Run.

1 2 3 43