The Massive Racial Blind Spot in Cancer-Drug Studies
A new ProPublica investigation sheds light on this considerable problem.
To ensure that a new drug works, and is safe, it must be tested through a clinical trial. Before it can be sold on the market, people with a certain disease are given the drug, and compared to others who receive a placebo. If the people who got the medication do better, the drug is one step closer to being approved.
But clinical trials do more than make sure our medicines meet our standards. They offer patients another option when all available treatments have fallen through, and they can also bring researchers closer to understanding the underlying mechanisms of different diseases. Clinical trials are crucial for research and patients alike. And not everyone is getting equal access to this important resource.
A new article out in ProPublica last week looked at data from clinical trials for cancer drugs and found that Black Americans and other minorities were vastly under-represented in the trial participants.
I spoke to Caroline Chen, one of the journalists who wrote the article, about what's keeping African Americans from clinical trials and why it matters. Read the full article at ProPublica, and also check out (and pass on to family and friends) the accompanying how-to: A Cancer Patient’s Guide to Clinical Trials.
What led you to look into this topic?
This actually started with my colleague, Riley Wong, who was a data fellow with us this summer. Riley wanted to do a data project, looking at minorities in healthcare. Together, we decided to look at clinical trials. What enabled us to do this project was that in 2015, the FDA started publishing data on clinical trials and who's enrolled in them, broken down by gender and by race. Before 2015, it would have been pretty hard to do this project.
We decided to do see who was in the trials, and compare it to who would be most affected by the cancer that the drug was treating. That's where we started. We just had a question: Who's in here, do the people we're testing the drugs on match-up with people who eventually are going to be taking the drug?
You opened the story with a drug called Ninlaro. Can you talk to me a little bit about what that drug treats, and how the story of this drug is representative of a greater issue?
Ninlaro treats a cancer called multiple myeloma, which is a type of blood cancer. Black Americans are more than twice as likely to get multiple myeloma than white Americans. Another way to look at that is: One out of five cases in the US are in black patients, because there's just a smaller raw number of Black Americans altogether. Twenty percent of cases of multiple myeloma in the US are in black patients.
What we found was, in the clinical trials used to get Ninlaro approved, only 1.8 percent of those participants were black. Out of 722 patients who had the drug tested on them, only 13 of them were black. There's this huge gap between the patients who are going to be taking drugs, who is most affected by this disease, and who was actually tested on. That was the pattern that we found, again, and again, and again, in cancer drug trials.
What was the highest proportion of black trial participants that you found for FDA approved cancer drugs?
We were curious as to whether or not anybody was doing a great job with this. But the highest percentage we found out of 31 drugs, was 12 percent. That was for a Johnson & Johnson treatment, and actually, Johnson & Johnson had the highest two percentages: 12 percent and 10 percent. That's still not very high. Just for comparison, in the US about 13.4 percent of people are black.
I thought that the example for prostate cancer was one of the most powerful, because that's one of the ones that is most widely known to affect more African-Americans than any other race, and yet, their percentage was still so low.
Prostate cancer, as you said, affects African Americans more than any other race in the US. These numbers are out of every 100,000 people; 178 out of every 100,000 African-Americans, versus 106 out of every 100,000 white Americans.
I came across a study by Dr. Daniel Spratt, who's at the University of Michigan, and he found from 2009 to 2015, in seven trials that he looked at for new prostate cancer drugs, only 3 percent of participants were black, on average. Then, a more recent example that we found, is a treatment called Erleada, made by Johnson & Johnson, and 66 percent of their participants were white, compared to 6 percent who were black. There is a big disparity there for sure.
What did you see for other minorities?
We focused on Black Americans because there were some gaps in the information for other races and ethnicities. For example, we couldn't really do a good analysis for Hispanics, because the FDA has only started collecting that data as of 2017. There isn't a lot of data in the first place. And it also doesn't break down between white and non-white Hispanics, so we didn't feel like we had enough information to do a good analysis there.
With regards to Asians, they do appear very well-represented in trials overall but what is actually behind that is that drugmakers are often testing their drugs overseas, because they want to get approved in those countries. For example, the Japanese FDA is probably going to want you to have tested your drug on some Japanese patients. You might see a high proportion of Asians, but it doesn't break down what type of Asians are within there, so they could be East Asian, they could be Pacific Islanders, et cetera, and we can't see that information.
The FDA also didn't start gathering information on whether the trials were in or out of the U.S. until 2017. For the trials we could see, Riley found that for drugs with at least 70 percent of their sites being in the US, only 1.7 percent of patients were Asian.
The takeaway is that Asians are well-represented, but they're probably all outside the US. Then, the last group that they had is Native American, or Alaska native. And we also found that they pretty much are nonexistent in these trials. About two-thirds of the trials didn't report any Native American or Alaska Natives at all.
Why does this matter? What do minorities miss out on when they can't participate in clinical trials?
I see a problem both for the present patient and the future patient. For present patients, someone who has cancer right now, it means that these minorities are not being able to access the most cutting-edge treatments. Clinical trials are one of the number of options that a patient has when they're diagnosed. Oftentimes, the already-approved drugs don't work for a patient, and a clinical trial can be their last resort. Or sometimes, clinical trials represent the latest technology, and might be an option for a drug that has less side effects. Particularly in cancer right now, there's been a lot of really great advances that have started to move the treatment away from chemotherapy, which is quite a toxic treatment, generally. The fact that these minorities are not in a trial means that they aren't getting access to those cutting-edge treatments.
I see a second piece of this, which is that while it's really important for us to recognize the social concepts and constructs around race, in some cases, it's shown that drugs can have a different effect on different population groups.
Absolutely. This is what I was talking about with the future patient. If we're not testing the drugs in Black Americans, for example, then there's potential that the drugs would work differently. We don't know this for sure, but we would have to do the trial in that population to know.
For example, with multiple myeloma, Black Americans have more than two times the likelihood of getting the cancer in the first place. They also tend to be diagnosed at a younger age. But their cancer also tends to be less aggressive, and scientists don't totally understand why that is yet. The thinking is that perhaps Black Americans are particularly susceptible to one subtype of multiple myeloma. We don't know this yet, but if we did know this, we could be able to then develop drugs to target those subsets.
I think that's generally where drug development is going anyway. Overall it's becoming more and more targeted. The FDA actually pointed this out as a problem. The FDA themselves published an article, in 2017, where they said meaningful differences may exist in how multiple myeloma affects Black patients and what symptoms they experience and how they respond to medication. The FDA did similar analysis to us, and found, like we did, that fewer than 5 percent of all participants in multiple myeloma trials are black. And they were like, "Yeah. This is a problem."
Is the FDA doing anything to ensure the trials have racial diversity, or enforce it in any way?
This is the tricky part. There are some groups, like the National Black Church Initiative, that I talked to, that want a broad mandate that minorities are required to be represented in trials, to some percentage. We talked to some other researchers, who have a slightly more nuanced view on that, which is, "[We should do that] If there's reason to believe that the disease might present differently"—if it's for genetic reasons, or environmental reasons. Basically: If there's a reason to believe that there is a scientific difference in how the disease presents in some subpopulations, then the FDA should mandate that that population is adequately represented.
The FDA said two things. First, they said, "We don't have the authority to do that." But then they also said, "Well, we do think that we need to make sure that drugs are safe and effective for the patients most likely to use it," and that they can ask drugmakers for additional data, if they aren't sure of that.
We talked to a lot of the drugmakers as well and I think their argument is: There's a trade-off here. If you want our trials to be more diverse, the reality is that it would probably take more time to recruit minorities, and time is money in the drug industry.
Every month that goes by without their drug getting approved can be hundreds of thousands of dollars for a drugmaker. So their argument was like, "If you delay this, it's gonna cost us money. If it costs a lot more money to run the trial, and recruit a certain percentage of minorities, maybe we won't develop another drug, or it means delaying this drug getting to market for everybody."
I feel FDA, right now, is taking much more of an attitude of, "We're gonna do soft persuasion. We're gonna encourage this, but we're not gonna require this." To some people that's not enough.
Why is it hard at all to get minorities into these clinical trials? One might think this is the perfect place for anyone to get treatment because the drug is free.
There are a number of different barriers to access. Financially, I think that one thing a lot of people don't realize is a clinical trial will give you the the experimental drug for free, but everything else still needs to be paid for, usually by your insurance. If the trial says, "We're gonna have you take an already-approved drug, and combine it with a new, experimental drug," the already-approved drug, will have to be paid for by you, or your insurance. Hopefully, you have insurance.
For patients who don't have good insurance, really high copays, really high deductibles, or participating in a trial means they have to go do another biopsy, or they do regular lab tests, or they have to have a hospital stay, they could end up saying, "You know what? It's cheaper for me to just do chemotherapy."
I talked to one doctor who had a patient who opted out of a trial for that reason. There was an already-approved drug, which was part of the trial, and that drug cost $10,000 a month, and the patient's insurance had a 20 percent copay. The patient was like, "That's just too expensive." Unfortunately, that patient then ended up getting chemotherapy, which was cheaper, and he didn't tolerate it well. Now, he's considering hospice.
The other thing I would add, too, is that trials don't pay you a lot. And that's by design, because they don't want to look like they're inducing or enticing patients to join a trial when it's not medically necessary. They'll maybe pay for your parking fees and maybe a meal, but typically, they won't pay for things like childcare, or if you've lost a day's wages, because you had to take a day off work.
What about a person having other existing conditions? How does that factor into participation?
This affects everyone, I would say. Every study has what's called "inclusion and exclusion criteria," which defines who are the patients that they're going to let into trial. Normally what drugmakers are looking for is the "healthiest sick patient." They want a patient who, say, in a cancer trial, has the cancer, but is otherwise very healthy. That candidate is most likely to do well, and it also means that if a side effect turns up in the trial, it's more likely to definitely be because of the drug, and not because the patient has a lot of other medical issues.
The problem with that is you then end up excluding a lot of people, because sometimes they'll be like, "We don't want patients with high cholesterol, we don't want patients with hypertension, we don't want patients who are obese, we don't want patients with diabetes." As it turns out, minorities, in the US tend to have more of these, what's called comorbidities. So having really strict criteria is another barrier that actually, particularly, affects minorities.
One thing we didn't talk about in our story, actually, is the reason why a lot of these criteria exist: It's because drugmakers just copy and paste from the last trial that they've done and aren't really thinking, "Do we really need to have all these criteria?" There's actually a movement right now in the drug industry to be a little bit more thoughtful about what criteria you require, because it would make it easier to recruit patients overall.
What's your response to companies or the FDA saying, "We're just gonna look and see, after we make the drug, to see if it's working." Do you think that's enough?
This is a very common argument that almost everybody we talked to says: "Look, we want to get the drug out to patients as quickly as possible, but we'll watch the drug once it goes on the market. And if it looks like, for whatever reason, it's not performing well, on a specific race, or subgroup of patients, we can do more studying, then."
I have two issues with this. For one, one of the doctors we talked to, Dr. Jonathan Jackson, he said essentially, that this is a really unnecessarily arrogant attitude. Which is that you're saying, "Alright, we'll just wait until someone's hurt to do something about it."
It opens the door to harm for patients who take the drug really early, before any differences are known. I definitely think it is a problematic attitude. I think the other thing that we didn't get into in the story, but I've reported on in the past, is that post-marketing studies are not usually drug makers' top priority, and our post-marketing surveillance systems in the US are still quite flawed.
The FDA has a number of different programs through which they can keep track of unexpected side effects of drugs. One main one really relies on doctors to report that these are happening. But it's not mandatory, and doctors are really busy, so typically, you're going to get a much smaller percentage of reports than actual incidents happening. There could be a drug that's not really working well for a minority group, and it could take potentially years for anybody to notice.
For me, it's like, yes—obviously, we should be watching for unexpected side effects that come up post-market. There's no way that any clinical trial is going to necessarily be perfect, but to use that as an excuse to not enroll minorities, up front, to me, is problematic.
And I imagine it's harder to monitor for when a drug isn't working as well for minorities, rather than monitor for side effects.
Yes. Let's say it doesn't work as well. I feel like there's shades here. Or it could be like, you know what? It doesn't work as well for black Americans as white Americans, and if doctors knew that, they would prescribe a different drug to black Americans. Maybe it's not killing them, but there are better options for them, and we have no way of knowing.
What about the issue of trust—what does the history of medicine and minorities have to do with this issue?
That came up a lot in my interviewing. The US, unfortunately, has a history of not treating minorities very well in medical research. The most famous example is the Tuskegee Study, which was conducted from 1932 to 1972 by the US Public Health Service. Researchers were trying to study the progression of syphilis. They enrolled a bunch of African-Americans sharecroppers, and at some point, after the study started, they knew that there were treatments that would be effective for syphilis, and they knowingly withheld it from the people in this trial because they wanted to learn about the progression of this disease, which is totally unethical.
That study, and other incidents like that from the past—even though that will not happen now, because we now have systems and ethical review boards—has had a really big impact on minority communities. A lot of the doctors I talk to said that they'll bring up a trial, and black patients will say, "Well, I heard about Tuskegee, and I don't want to be a guinea pig."
I think that continues to be a really big component of why black Americans in particular don't take part in clinical trials. I think the only way to address that is through education, but I think it has to come from everyone. We've heard, for example, sometimes doctors will not tell their patients about clinical trial options because they just assume that the patient won't want to be in it.
They would think, "Oh, my patients worry about being guinea pig, so I'm not even gonna bring it up." My stance is: "Well, if you don't even bring it up, how are you going to start changing peoples' minds?" I definitely think this has to happen on all fronts, whether it is the FDA and the drug industry putting actual money and resources into education campaigns, or frontline community oncologists having these conversations with their patients.
The last patient in my story who I talked to, Thomas Goode, has had multiple myeloma since 2005. Multiple myeloma, right now, can't be cured. It can go into remission, with help from drugs, but then it will eventually come back. They're still working on trying to figure out a permanent cure.
Goode has done multiple different drug regiments, he's had three different stem cell transplants, and in between each stem cell transplant, he's taken experimental drugs in clinical trials. He says that the only reason he joined the first trial is because he loves his doctor. He was like, "I didn't know anything about clinical trials, but I really trusted my doctor," and so, that's what bought him into the clinical trials. And that's, he say, what really kept him alive.
Sign up for our newsletter to get the best of Tonic delivered to your inbox.