Q&A: Rethinking cancer, from the basics to the breakthroughs

Dr. Otis Brawley, chief medical officer for the American Cancer Society, says disparities in our health system undermine the scientific progress made in the fight against cancer.
(Todd Williamson Archive / WireImage)
Share via

Dr. Otis W. Brawley considers himself one of the luckiest cancer doctors around.

As chief medical officer for the American Cancer Society, Brawley says, “I’m sort of like the baseball fan who gets to sit in the bleachers with the scorecard and keep score on all the players.”

The game is a matter of life and death. But the players — cancer researchers and doctors — seem to be hitting more doubles and triples and striking out less. More patients are living longer after being diagnosed, and cures are increasingly common.

But new scientific insights will take us only so far. Brawley sees stubborn obstacles to progress in his research on health disparities and in his clinical practice at Emory University. If only our health system were more equal and less fragmented, if the latest discoveries about cancer prevention and treatment spread more quickly, more people could be reaping the benefits.

Brawley elaborated on these ideas in an interview with The Times.

Is cancer the same scourge it’s always been?

It really depends on how you count it. Mortality rates attributed to cancer were 60 per 100,000 people in 1900 and peaked at 215 per 100,000 in 1991. Those mortality rates are 25% lower today, about 160 per 100,000. That’s the good news.

In absolute terms, though, cancer remains one of mankind’s greatest threats. We’re going to lose over 600,000 people to cancer this year. That’s greater than it has been at any other point in our history.

What’s behind the improvements seen in the last 25 years?

Prevention. Major advances in smoking cessation in the 1960s and ’70s translated into people who did not die in the ’90s and beyond.

Also, cancer treatment is more effective today than it’s ever been. In breast cancer, we’ve had a 40% reduction in risk of death. From the 1980s onward, we’ve seen a 45% to 50% reduction in the risk of death from colon cancer. We’ve had a marked decrease in the risk of dying from prostate cancer. This is all progress.

But our progress could have been better if we had been more organized and much more attentive to what the science was telling us about cancer prevention and treatment.

What do you mean?

One of the great American problems is, how do you get the fruits of cancer research and treatment improvements to everybody?

For instance, there are 13 states — Alaska, Montana, Wyoming, South Dakota, Utah, New Mexico, Oklahoma, Missouri, Arkansas, Louisiana, Mississippi, Alabama and Vermont — where the breast cancer death rate has fallen by less than 10%. If you’re a woman diagnosed with breast cancer in any one of those 13 states, you have not reaped all the benefits from all that we know about treatment and prevention of breast cancer. These aren’t scientific problems: They’re societal problems.

Is this a matter of economic inequality? Racial disparities?

Not entirely. A substantial number of Americans — not just black or Hispanic, or even poor Americans — get less than optimal care. In breast cancer alone, about 20% to 25% of white middle-class women in the U.S. get less than optimal care.


Some of it is that physicians are not up-to-date. Some of it is that hospitals are not up-to-date.

Some of it is that patients get lost in the system. You’ll literally have someone seeing a medical oncologist who gets a referral to see a radiation oncologist, and the patient doesn’t understand or forgets to go. The result is the patient gets no radiation, or they get it six months after they should have. That sort of thing is so, so common.

What about new medications? Are they helping?

I am incredibly optimistic about our science. I see clear progress, and I’m excited about things we are learning, about the things we’re able to do.

But I don’t want to exaggerate our progress. I think we could have done better by now, and I think we can do better.

The American Society for Clinical Oncology says that 19% of the cancer drugs approved over the last 10 years have boosted survival outcomes in clinically meaningful ways. That means we still have a long way to go.

What about all the talk about cancer breakthroughs?

There’s this tendency that we see time and time again in oncology — people make folks think there’s light at the end of the tunnel. Patients come to us and think, “Ah, this is THE drug for a cancer we’ve been trying to get a handle on for the last two centuries.”

Sometimes it is, by the way. The first person to get cisplatin for testicular cancer, he got cured 40 years ago, and I just saw him at a meeting recently. One of the first guys to get crizotinib had widely metastatic lung cancer and was about to go to hospice — I would not have anticipated he would live six months — and now he’s a financial adviser here in Atlanta. But there is this tendency to simplify and exaggerate that progress.

Remember about 18 years ago when the growth-inhibitor drugs — antiangiogenesis drugs like Avastin — were thought to be the answer to cancer? A lot of patients got jerked around emotionally and were promised something that we just didn’t have. That’s a bad thing to do. I struggle with how to share my enthusiasm but also give people a subtle dose of sobriety at same time.

If cancer could be made into a chronic disease, would you consider that a victory?

Many of us in oncology are starting to think this way. As opposed to reaching for the word “cure,” we’re thinking of cancer like we think of HIV.

I started taking care of a young lady in the early 2000s with widely metastatic breast cancer. She’s still alive. The way to think about it is not that we gave her drugs, and she went into remission and was cured. We’re giving her Herceptin every couple of weeks, and it’s like we’re throwing a little water on a smoldering fire. She continues to work. She has a good quality of life.

This is the kind of goal that is very achievable in the near future for a lot more patients.

What do you wish Americans understood better about cancer?

The role of obesity. Most people are aware that tobacco is the major risk factor for cancer. But the second most common risk factor is the combination of obesity and lack of physical activity.

Why does obesity matter?

Very early on, a cell with a cancer mutation has to implant somewhere to grow. Once it gets to be more than 64 or 128 cells, it needs to recruit some blood vessels to get oxygen and nutrients. High levels of circulating insulin — a common metabolic consequence of obesity and sedentary lifestyle — help to recruit blood vessels for the tumor. That’s how obesity helps create the environment in which cancer can grow.

Not only is obesity a risk factor for cancer, it’s a risk factor for relapsing cancer once someone goes into remission.

Finally, obesity is actually interfering with a lot of treatment in the United States. I see doctors who are afraid to dose chemotherapy in patients who are very obese. They’re afraid to give a patient the dose called for because they’re afraid of side effects.

Let’s talk about screening and “catching cancer early.” Has the thinking changed?

Yes. Screening for cancer is one of those issues where we really have to balance enthusiasm and hope with an appreciation of the problems.

What kinds of problems?

If you take 1,000 women at age 40 and start screening for breast cancer every year for the next 10 years, more than half of those women are going to have a false positive. That means you’re going to do more mammography, which is more radiation. It is a fact that radiation causes breast cancer, and it does so more effectively in women who are in their 40s than in women who are in their 60s.

So, if you do annual mammography screening in women from 40 to 74 and you do that for 100,000 women, every year you’re going to cause 125 breast cancers and 16 breast cancer deaths. But you’ll also prevent 968 breast cancer deaths. There is a real trade-off.

That sounds like a pretty good trade-off, no?

Well, if you start screening at age 50, the number of breast cancers you cause goes down to about 40.

Now, a substantial number of women in their 40s develop breast cancer and die. That is a fact. But mammography is a terrible test for women in their 40s. It just doesn’t work nearly as well as we wish it would to find their breast cancers. But it increases their risk of developing it.

We need a better test for women in their 40s. That’s a statement I don’t hear enough.

Are there problems for older women too?

Yes. Let’s say you find a pea-sized lesion in a woman’s breast. A pathologist will look at it under scope and write a report that says, “This woman has cancer.”

The question is, is that cancer genetically programmed to grow, spread and kill? Epidemiologists estimate that somewhere around 15% to 25% of all breast cancers detected through screening are never going to grow, spread and kill. It’s just going to stay there, and it may even get smaller over time. Some women who are treated would have done well if it had been left alone.

We call this overdiagnosis, and it accounts for more than half of screening-detected prostate cancer, a goodly number of ultrasound-detected thyroid cancers, and a good 10% to 15% of lung cancer we find.

How can you reduce overdiagnosis?

We’re starting to develop genomic tests to predict the biologic behavior of cancer. The new 21st century definition of cancer diagnosis will need to account for the growing appreciation that cancer comes in a lot of different levels of aggressiveness. The slow-growing cancers we’ll watch, and the fast-growing cancers we’ll treat aggressively.

How much of cancer is “bad luck,” and how much of it would be preventable if we just made better choices?

The message I would like people to take away is that 30% to 35% of cancers are due to smoking. Almost 30% are somehow related to bad diet and lack of physical activity and obesity. We know that pollution causes some cancers, that sun exposure causes some cancers. (Don’t try to add these probabilities up — there’s a lot of overlap.)

But there are also lots of people who will get cancer even though they don’t smoke, they exercise, they’re not fat, they live in a nonpolluted environment. And they ask, ”Why did I get cancer?”

What do you tell them?

I like to explain it this way: When you have a computer program on a thumb drive, and you copy that program from one thumb drive to another, you get an exact copy. But at some point, if you copy it enough, you end up with a message that there’s been a corruption.

The average age of a person with a new cancer diagnosis in the United States is close to 70 years old. When you copy DNA enough, which we do billions of times over a lifetime, eventually you wind up with a corrupted copy.

So being older is a form of bad luck?

There’s another element of chance: If you end up with a corruption in a quiet part of the DNA, that may be no big deal. If you end up with a corruption in an area that turns on DNA replication — or worse yet, that turns it off — then it becomes a bigger deal. If you lose the function of the tumor-suppressor gene, then you’re going to end up having a tumor.

If mutated DNA can’t be repaired, we have another system that causes a cell to kill itself if it’s about to become a cancer cell. It uses a self-destruct switch called P53.

A person in his or her 50s probably has between four and 12 mutations per day. But your DNA repair enzymes and P53 take care of most of them. If they don’t, we have our immune system cells trying to zap those cells and kill them.

We have all these overlapping systems to prevent cancer from happening, and shut it down when it does. But sometimes there’s a failure. I hate the phrase ”bad luck mutations.” But that’s how people end up getting cancer.

This interview has been edited for length and clarity.



With cancer, it's not necessarily where it starts but how it starts

Life after cancer: “The real battle happened after treatment”

Your guide to the latest in cancer screening tests: Which do you need, which can you skip?