TRIALS AND TRUST: This is the first part of a series on health equity and diversity in clinical trials and cancer research.
SAN FRANCISCO — In her living room, Laura Esserman, a breast cancer surgeon at the University of California, San Francisco, was singing. The first time Esserman sang in the OR, she watched an anxious patient’s blood pressure gently calm down on the vitals monitor. Now Esserman, a lifelong musical theater performer, serenades them with the song of their choice as the anesthesiologist guides them into blackness.
Wearing a silky dark kimono and bright red spectacles, Esserman sang the song she’d want sung to her, if she were the one on the operating table. The bands of bead bracelets on Esserman’s arms that normally cackle and clack when she speaks were silent even as her wrists danced in the air — a surgeon’s steady hands. The song is “For Good,” from the musical “Wicked.” “It goes ‘Because I knew you, I have been changed for good.’ That’s the story of medicine,” she said. “As a physician, as you take care of all these people, their stories become part of your story and how you learn. It changes how you treat the next person.”
Esserman’s calling is to collect stories of thousands of patients through her clinic and the cancer research and clinical trials she runs — which, in turn, might change the practice of medicine. But like so many medical researchers today, Esserman realized in the last few years that medicine hasn’t been doing enough good by patients of color. The vast majority of cancer studies, including some of hers, didn’t have significant representation from Black or brown patients. She resolved to do better.
Several years ago, Esserman started a clinical trial called the WISDOM study about an issue that makes her fume: the recommendations that every woman 40 and over get an annual mammogram. Her hypothesis is that it would save more lives and result in less harm if each woman was instead given a personalized screening schedule based on factors including age, genetics, family history, and breast density.
“Why are we still screening the way we did in 1980, when we didn’t know what an estrogen receptor is?” she said. Her glare could melt stone. “We’re looking at data 30 years old, now. My goodness. Treatments have changed. It’s time for a new approach.”
“For Good,” sung by Laura Esserman
She was determined from the start that the trial would have broad representation from people of color. Instead, she has learned a lot about how big the gap is between good intentions and good results. Without a host of strategies tailored to gain the participation of specific communities, a clinical trial may need what Consuelo Wilkins, a professor of medicine and a health equity expert at Vanderbilt University, calls a rescue.
Esserman and her team aimed to recruit 100,000 women to the trial in the first year. The plan was to send an email invitation out through the Athena Breast Health Network, a group of researchers, clinicians, and patients across the University of California schools, which had a racially diverse group of 140,000 patients.
“We anticipated most everyone would participate in this study, and we’d get to 100,000 pretty quickly based on the experience we had in the Athena Breast Health Network,” said Allison Fiscalini, the director of the Athena Breast Health Network and one of the WISDOM study managers. “But that was not accurate. And as we started to look at who is coming in, it was a more white, non-Hispanic population.”
Instead, in the first three years of the trial, roughly 21,000 women signed up to participate in WISDOM. Just 1.7% of them were Black. Esserman had to admit, they were doing something wrong. She needed help.
Oncologists have increasingly seen mammograms as a double-edged scalpel. Screening has indisputably saved lives — breast cancer mortalities have fallen roughly 40% since mammography’s introduction — but it can also catch abnormalities that aren’t dangerous. That can lead to unnecessary treatment, procedures, and stress and anxiety. “I’ve had people come in and say, ‘This is my third biopsy,’” she said. “More is just more, not better. Sometimes more is worse.”
That’s led to disagreement among experts as to how frequent mammograms should be. The American College of Radiology recommends all women get a mammogram every year starting at age 40. Other institutions, like the U.S. Preventive Services Task Force and the American Cancer Society, espouse less intense recommendations, but Esserman said most major breast centers adopt the radiologists’ suggestion. “Eighty-eight percent of them,” she said. “And 65% of women screen every year. Radiologists haven’t let go of annual screening, and the public hasn’t let go of it.”
Esserman, who is also a member of the American College of Radiology, thinks that might be too often for most women. Breast screening shouldn’t be one-size-fits-all since women have different chances of developing breast cancer, she said.
The trick is figuring out exactly how much each woman ought to be screened. Some factors are well established — for example, having certain mutations like BRCA1 and BRCA2 make breast cancer very likely — but everyone else falls along a gradient. “The mutation carriers, they’re like bright red. Then, we have the pink sea of women who are not really high risk. Some women are a darker or lighter shade of pink,” said Ruth Etzioni, a biostatistician and cancer researcher at the Fred Hutchinson Cancer Center who does not work on WISDOM. “For a long time, epidemiologists have been trying to differentiate that big pink sea.”
Aside from those key mutations, scientists have a few tools to calculate someone’s personal risk for breast cancer. There are also individual genetic variations that may contribute only a slight amount to breast cancer risk. Some women may have enough of these mutations that they add up to a substantial additional risk to breast cancer, while others only have a few. Scientists call this a polygenic risk score.
“Then we have more generic factors like age, reproductive history, family history, and breast density,” Etzioni said. Aside from breast density, most of these factors don’t differentiate women’s risk dramatically; but they do all add a little bit of definition to who’s at more or less risk of breast cancer.
WISDOM is testing whether combining all of those factors — plus race-specific algorithms for polygenic risk — into a single overall score might begin to differentiate risk enough to affect health outcomes with individualized screening recommendations. Based on WISDOM’s risk calculator, all women are recommended to either alternate mammography and MRI every six months, a mammogram once a year, or a mammogram once every two years.
Women who enroll in the study can choose annual mammography, personalized screening, or to be randomly assigned one of those two options. At the end of the study, the team will scrutinize the data to see if those who got a personalized screening recommendation were able to avert more later-stage breast cancers while also experiencing fewer biopsies or other unnecessary procedures.
And if not, Esserman said, at least the study will be a launching point for future studies. If WISDOM is able to enroll a substantial number of women, there will be a wealth of data to begin refining and improving personalized breast cancer screening. “And if this doesn’t work, you try something else,” she said. “You just keep learning and working. Trials are how we learn.”
It’s in studies like WISDOM, where scientists are advancing breast cancer screening technology and methods, that Black women desperately need to be included, said Kirstin Bibbins-Domingo, the editor-in-chief of JAMA and a professor of medicine at UCSF who does not work on WISDOM. Breast cancer mortality has fallen for everyone, but it’s still 40% higher for Black women than for white women. According to a recent article in the New England Journal of Medicine, that disparity has only increased over time.
That used to not be the case. Breast cancer mortality for Black and white women was roughly equal before 1980 — then mammography and several other advancements in breast cancer medicine arose. Unequal access to those advancements explains part of the disparity, but a pervasive exclusion of Black women from participating in clinical trials probably explains another part of it, Bibbins-Domingo said.
The vast majority of clinical studies, including those investigating breast cancer, have been overwhelmingly made of white participants, Bibbins-Domingo said. That’s left a dearth of clinical knowledge around how new medicines affect many Black women.
“When we fail to represent Black women in studies of screening in high enough numbers, we just don’t have the scientifically right answer to what should we do and what role does screening play in this very important health problem,” she said. “The great rationale for the WISDOM trial is to say, can we think about something else that tells us about the risk for you as an individual. That’s the advance, the new tech, and unless you build in equity in the trials, it actually may not help Black women and in fact make their outcomes worse.”
Polygenic risk scores, like the kind the WISDOM trial hopes to test, is a prime example of tools based overwhelmingly on data from white populations, Bibbins-Domingo pointed out. In a recent paper published in Nature Medicine, scientists found that polygenic risk scores in general didn’t perform as well in people with African ancestry. “Polygenic risk scores should be derived from populations where they’re intended to be used,” Bibbins-Domingo said.
Federal agencies and universities have long said that increasing diversity in research is a priority, but it’s largely been “lip service,” she said. In a National Academies report on health equity in clinical trials that Bibbins-Domingo chaired, the committee predicted that the lack of diversity in clinical trials would cost hundreds of billions of dollars over the next three decades measured by life expectancy, disability-free life, and years in the labor force. Despite that, the report noted no progress.
“There’s been some improvements in representation of women, but mostly white women,” she said. “But we found that progress has largely stalled.”
In the big pink sea of breast cancer risk, Black women have little idea where along the pink gradient their risk falls.
When Esserman started recruiting for the WISDOM study, she imagined women from all racial backgrounds would flock to the trial. “We just opened it up and advertised it. We thought everyone would want to just jump on, and we’d get diversity,” she said. “But that’s not what happened. It felt like a complete failure to be diverse.”
After a couple of years of trying to enroll people into WISDOM, only 48 Black women had volunteered to participate, compared to thousands of white women. The last thing Esserman wanted to do was run another cancer trial with scant participation from Black people. Early in Esserman’s career as a surgeon, she operated on a young Black woman named Lynnea. “She had metastatic breast cancer in her 30s,” Esserman recalled. Esserman cut out all the cancer she could, but it roared back five years later, ending Lynnea’s life. “We didn’t have the tools to find out what kind of cancer she was at risk for and know in time to screen her.”
If trials like hers didn’t enroll more Black women, Esserman thought, then she’d never know. “I promised Lynnea’s mother that I’d get to the bottom of this,” she said.
But she didn’t know what to do. So she phoned a friend.
“Forty-eight, that’s all?” said Funmi Olopade, a breast oncologist at the University of Chicago.
“Help me,” Esserman pleaded.
Olopade said later she was shaken but not that surprised. Esserman had mainly reached out to the populations she’d worked with in the past — women who were connected through University of California sites — using email blasts.
“It was mostly elite academic centers in California,” Olopade said. “Those have traditionally excluded Black and brown patients. If you look at anything that starts in an academic center, it’s not going to reach the most vulnerable populations.”
The WISDOM study was one that Olopade already believed in deeply, plus she’d been friends with Esserman for over two decades and collaborated with her on several other projects in the past.
“It was clear the way WISDOM was started was about rich white women who are always going to be the first to sign up. Women who understand technology,” Olopade said. “But I really wanted to motivate other women like me to participate.”
She would help Esserman lead the study.
Olopade wasn’t the only researcher of color Esserman asked for help. Friends and colleagues who were experts in diversity and equity issues began giving her advice and helping her reach out to Black community groups like sororities and community health organizations.
“I participated in some of these,” said Jennifer James, an ethicist and sociologist at UCSF. Her work, an ethical study of the WISDOM study, is separately funded.
She and Esserman were reviewing some of the slides that Esserman had prepared, and the points weren’t sitting well with James.
“If you don’t join the study, we’re not going to have enough information on your community. It’s up to you to do this,” Esserman wanted to say. “The onus is on people to participate.”
“No, not us. It’s on you,” James shot back.
“How’s it on me? I’m trying to invite people,” Esserman said.
“The question shouldn’t be why aren’t people joining the trial, but what’s wrong with me if people aren’t joining the trial because my trial isn’t working for them. It’s not meeting their needs, and maybe I’m not communicating well about the trial, or I’m not going to the right places and talking to the right people,” James said.
That conversation stuck with Esserman. She was learning that she had to change things about the study — like expanding the enrollment sites beyond University of California clinics — but James made her realize she also had to change her mindset.
“You have to be willing to say, I’m using the wrong language. I’m thinking about it differently. And because I was in a hurry, I thought all that was needed was just making sure you got the word out,” Esserman said. “But that’s not it. You have to work on it.”
This is a process that Wilkins sees all the time. Wilkins, who also sees patients as a physician, runs the Vanderbilt Recruitment Innovation Center, which helps researchers on improving diversity in their clinical trials. Recently, when scientists come to her for consultations, she’s been seeing the same patterns over and over.
“Investigators who are experienced in research and clinical trials assume the strategies they use to recruit largely white populations will work when they try to recruit a more diverse population,” Wilkins said. “They’re walking with this confidence like, ‘oh I’m a good person, I’m not racist. I’m going to design the trial, and they’re just going to show up.’”
It never works out that way, Wilkins said. Instead, “we get contacted later — sometimes by researchers we gave advice to already, and they didn’t implement it,” she said. “We’re getting called to rescue the study. Then they’re in rescue mode, and it’s much harder to do this work if you’re changing and having to implement things much later than if you did it in the beginning.”
To successfully recruit racially diverse participants, Wilkins said, researchers should think about three main categories: the study’s inclusion and exclusion criteria, trust and distrust, and access to the study and health care. Making sure each of these things are equitable and just, Wilkins said, takes a lot of time, effort, and money.
“Typically, when I’m talking to people about recruiting from racialized groups, I tell them it’s going to cost 2 to 3 times as much as it costs to recruit the general population,” Wilkins said. “And it costs more to rescue. That’s if the rescue is even possible.”
Much of the advice that Wilkins gives scientists is the same that Esserman got from her collaborators like U. Chicago’s Olopade and UCSF’s James. The WISDOM study created a Community Leadership Advisory Board composed of leaders of color like nurse navigators, survivors, patient advocates, and directors of health organizations who began guiding how the study used language and communicated with different communities of color and broadened WISDOM’s reach in the breast health community. The team also created a monthly virtual forum that’s open to the public where participants can discuss the WISDOM trial, give feedback, and learn more in general about breast health and cancer. That helps the study give back, at least a little, to the communities that scientists want to recruit from, Wilkins said.
Esserman and her team also applied for additional funding from the National Cancer Institute to expand the study to several other sites with large communities of color. These included Louisiana State University, University of Alabama at Birmingham, and the University of Chicago where Esserman knew researchers like Olopade had already built key relationships with community leaders. Then, WISDOM began making a concerted effort to be as visible as possible, tabling and canvassing people at health fairs, community events, or outside local clinics.
“One of my secret weapons is I happen to be friends with Valerie Jarrett, who’s a neighborhood girlfriend and serves as president of the Obama Foundation,” Olopade said. “She joined WISDOM. I joined it. On Mother’s Day, we talked to people about WISDOM. My recruiter sits in front of the hospital as people walk by and talks to people about WISDOM.”
That consistent, in-person presence helps people feel more connected to the recruiter and the study, Olopade said, and makes it more likely that they’ll participate. It’s something that they’ve been encouraging recruiters and clinical research coordinators to do across all the study sites.
Since WISDOM began these steps in 2018, the study’s enrollment has changed. From 2019 to the beginning of 2022, WISDOM’s new enrollments went from 81% white to 68% white. New Black participants changed from less than 2% of all new enrollees to roughly 10%.
Le’Andrea Anderson-Tolbert is WISDOM’s clinical research coordinator at the University of Alabama at Birmingham. Both Anderson-Tolbert’s mother and one of her co-workers, Kathy Levy, joined the WISDOM study themselves, and they’d all come down to Selma to recruit for the study at the annual Selma Bridge Crossing Jubilee. Music thumped and the smell of grilled meat filled the air.
Anderson-Tolbert was born and raised in Alabama and graduated from Tuskegee University in 2018. “Alabama? It’s just kind of like home,” she said. She waved a hand at the town’s dusty center.
“A lot of Black people in Alabama have roots here. My grandma is from here.” Her voice dropped slightly. “Because, honestly? You know, slave trade and everything and the Alabama River is right there. Most people came on ships through here.”
As people passed by the table, Anderson-Tolbert would strike up conversations with them.
“I haven’t heard of the WISDOM study,” said one woman, Faye Greene from Montgomery.
“It’s a national study, we’re trying to see how often women should get screening for breast cancer. Especially Black women, because us Black women, we get breast cancer more often,” Anderson-Tolbert replied. “We do a genetic test, so we have some people in a personalized arm with the genetic test, and we get some family history and stuff like that. We give you a recommendation on how often to get screened.”
Greene wanted to know more. A friend of hers had two siblings pass away the year before from cancer, and they’d always suspected there was something genetic related to the family’s history of cancer.
“Like BRCA1 — it’s basically like a gene change and it’s passed down through the family line. It increases your risk for breast cancer, ovarian cancer. My mother-in-law had it,” Anderson-Tolbert said, nodding her head. “She didn’t know until she got older that she had the BRCA gene.”
If Greene or her daughter or anyone in her friend’s family wanted to join the WISDOM trial, Anderson-Tolbert added, they could learn if they had any genetic risks for breast cancer, too.
“I think that’s a really good thing,” Greene said. “So, they can find out why is this happening and is it something that you’re going to be passing on down to your children.”
She took some of the pamphlets, and Anderson-Tolbert promised to follow up with a phone call later in case Greene wanted any help signing up for the study. “Here’s my number, too,” she said. “Reach out to me.”
Anderson-Tolbert says she usually gets a couple of people to sign up — or at least take some of the WISDOM information home — each time she goes out to a community event like this. People don’t always join the study, “but at least we get the word out,” she said.
Later, another woman stopped by Anderson-Tolbert’s table. Sheila Evans, from Selma, was interested in the study for many of the same reasons. “It’s good to know more” about your breast cancer risk, she said. Later, though, when she read on the consent form that the study didn’t pay for the mammograms, she got cold feet. If the study recommended extra screenings beyond just an annual mammogram, she knew she wouldn’t be able to afford it.
“I’m on Medicaid,” she said. “And to be honest with you, it’s not going to pay for all this extra stuff.”
Everything that the WISDOM study has done to increase diversity in the last couple of years is commendable, Vanderbilt’s Wilkins said. “It sounds like they’ve been very thoughtful and taken a step back to re-evaluate what they could change,” she said, looking at the WISDOM enrollment data. “I would say congratulations to them on making some improvement.”
Still, she added, “it sounds like they are doing the right strategies, but they’ve implemented them in a rescue mode.” That often leaves gaps in equity that might not have been there had the researchers designed the study from the ground up to be inclusive.
“For example, a lot of studies have inclusion, exclusion criteria that are going to keep people from wanting to participate. It may have nothing to do with the science. Like I got consulted on a study where they require a social security number. Right away, people are suspicious. Why do you have to have that?” Wilkins said. “Then let’s just say I’m a woman who meets all the eligibility criteria, and then I show up and then I learn that I’m actually too poor to participate in the research, because I don’t have insurance. Or I’m not documented.”
That boxes out a significant proportion of the very people that scientists want to recruit to studies, Wilkins said, since people without insurance or documentation are disproportionately people of color. Requiring people to have certain health insurance that will cover certain costs in the trial is also a barrier to forming partnerships with community health organizations, Wilkins added.
“If you’re talking about working with community centers where they may have a third or a half of their population uninsured, they’re not taking that study,” Wilkins said. “You can’t tell me to support your study when I have to tell half of my patients that they wouldn’t be eligible.”
When WISDOM first launched, people needed health insurance to enroll. Later, Esserman removed it as a requirement and raised money so that the trial would cover any additional costs that weren’t part of routine treatment. WISDOM now pays for the genetic testing and any follow-up genetic counseling if participants have a pathogenic mutation like BRCA1 or 2.
“We started out where we were going to try to force the payers to cover it, and then we realized that was going to make the trial more inequitable and more unjust,” Esserman said. “We had to just suck it up and raise the money for the genetic testing.”
Esserman also said she fought with insurance companies for months to get them to cover the cost of the trial. Some came on board, she said, including Blue Cross Blue Shield. Others didn’t. She wasn’t able to raise enough money to cover participants’ routine breast screenings.
If a participant has trouble getting extra recommended screenings covered, though, she urges them to call WISDOM. “We will advocate for you,” she said. WISDOM coordinators also try to work with participants to find low or no-cost screenings.
That’s an important commitment, Wilkins said, but there should still be more. There may be other barriers to participating in WISDOM that the trial doesn’t overcome. “Is child care available? Are there transportation resources?” she said. “Often those things are missed.”
Esserman has taken to calling the trial WISDOM 1.0. She promised the next iteration — WISDOM 2.0 — is coming. She wants everything — the science and the diversity — to be better. This time, she said, she doesn’t want to miss anything crucial when it comes to making sure the study is equitable and just.
With WISDOM 2.0, Esserman plans to take everything she’s learned from over the last few years to make sure everyone is included in the trial. It’s the only way, she said, to make sure that people of color benefit from the advances in medicine, genetics, and screening. The trial itself, she added, will hopefully be a platform for women of color to benefit from the most cutting-edge screening science and inform themselves about their own risk.
“I’m building a coalition of African American leaders to get 10,000 African Americans into the trial, 10,000 Latina women, Asian women,” she said. “We want to double down, triple down, and build on all of our relationships.”
Crucially, she wants WISDOM 2.0 to enroll women as young as 30, partly because African American women are more likely to get aggressive forms of breast cancer at younger ages. “Lots of people I’ve taken care of were 29, 32, 35, 37, with metastatic breast cancer. We’re going to identify those young people who are really at risk and screen them earlier,” Esserman said.
That would make the WISDOM study “a godsend,” said Ricki Fairley, a breast cancer survivor who has partnered with WISDOM and runs WhenWeTrial, a nonprofit dedicated to increasing Black participation in breast cancer trials. “Black women get triple-negative breast cancer at three times the rate of white women, at much younger ages, at much later stages, and we don’t know why,” she said. “Anything that can help Black families understand our genetic profile is critical, just critical.”
That’s exactly the point, Esserman said. Clinical trials are an opportunity for the people who join them to access the medicine of the future, today. And as long as she and her team can make sure those participants come from diverse backgrounds, it will mean that their stories will have a shot at changing the next generation of screening — personalized screening — for the better.
This series was supported by the USC Annenberg Center for Health Journalism’s 2022 Impact Fund for Reporting on Health Equity and Health Systems.