Real news. Real stories. Real voices.
Play Live Radio
Next Up:
0:00 0:00
Available On Air Stations
Supported by

Facing cancer? Here's when to consider experimental therapies, and when not to

Experimental therapies for cancer can be tempting when you're sick, but many fail to offer any benefit.
Cavan Images
Getty Images/Cavan Images RF
Experimental therapies for cancer can be tempting when you're sick, but many fail to offer any benefit.

Note: Molecular biologist and author Jeff Stewart has worked more than 15 years as a consultant to drugmakers, scrutinizing data on new treatments to fight cancer. Last July, the 50-year-old father of seven was diagnosed with stomach cancer himself. He spent much of the next 10 months in treatment, while also writing the newly published Living: Inspiration from a Father with Cancer.

His book is a compilation of life lessons and reflections that "helped me endure hard times and avoid harder ones," he says. Framed as a life guide for his kids, it also includes insider advice for anyone facing a cancer diagnosis. The following excerpts have been edited for length and clarity. --Editors

As a cancer patient, I'm getting forwarded many, many articles about early stage cancer treatments and alternative therapies. I think every cancer patient gets these. I'm public about my cancer, so I'm getting these from more than friends and family. I'm getting these also from people I've never met but who are trying to help.

Sponsor Message

Cancer treatments are not just a personal interest. Part of my job for over 15 years has been to advise pharma companies on cancer drugs. My clients have included big pharma and small biotechs. You'd know the big-pharma names. I've interviewed hundreds of oncologists over the years. Figuring out the scientific and commercial potential of a cancer drug is a normal day on the job for me.

How to assess experimental treatments? Follow the evidence

Basic rule of cancer treatments: Evidence wins. We need evidence to believe anything works. That's especially true for cancer.

Step one: Is the drug FDA-approved?

Step two: If the drug is approved, is the drug also recommended in cancer guidelines? If the approved drug is not in cancer guidelines, insurance companies aren't going to pay.

Sponsor Message

Step three: If the drug is neither approved nor in guidelines, is the drug in late-stage clinical trials? That usually means phase III. If so, then maybe a cancer patient can join those. If not in late-stage clinical trials, the drug is too early in testing to help most people who have cancer now.

Amazing result in a test tube? Talk to me in 15 years if I'm still here. Just started phase I clinical trials in people? Still too early to help me. Cures mice? Of all the oncologists I've interviewed about mouse data, I've been told by at least a quarter of them this exact punchline: "I've never treated a mouse for cancer." Yuk, yuk. Funny oncologists. It's not just snark. They have a point buried under their tired joke. Most things that cure cancer in mice don't work in people.

The author signs copies of his new book.
Carissa Wadsack-Stewart / Jeff Stewart
Jeff Stewart
The author signs copies of his new book.

It's worse than you imagine. Nearly every new thing coming out of a university is too early to help anyone who has cancer now. Worse, nearly everything fails. If that sounds jaded, I'm sorry. This is oncologists' lived reality.

Yes, there have been great strides in cancer treatment. The really promising drugs that can do anything in the short term are already in late-stage clinical trials. Oncologists read. They know what's coming. Anything early-stage will not, cannot cure someone who has cancer now. I have to think one of the worst parts of an oncologist's job is to explain why a miracle cure in early development holds no promise at all for a cancer patient today.

Hard truth: Most experimental cancer treatments fail

Sponsor Message

Here's what most people not immersed in oncology don't get. Even the most promising cancer drugs fail. Cancer drugs have the second-worst failure rate of any disease. Only Alzheimer's is worse.

Think of the tens of millions of dollars spent to get one cancer drug out of a university, into cell lines and mice, and finally into patients to be tested in clinical trials. That's a huge effort. It might take a decade. Those drugs that get tested in people have won a biotech lottery. For any cancer drug to be tested in people, the science has to be amazing. Scientists working on the drug believe it's a lock to work. There may be talk of a Nobel Prize or at least the Lasker Award. Everything seems sure to succeed. What could go wrong?

Do you want to take a guess at how many of those "sure winners" end up passing clinical trials? Seven percent. That's 7% of the best drugs that emerged from the best science and were so promising that a pharma company invested $10 million to more than $1 billion to test the drugs in patients. Ninety-three percent of the "winners" fail.

What about repurposed approved drugs? Approved drugs can be used off label by physicians. What if, say, an anti-parasite drug cured cancer? Why not take that?

The question is, again, where is the evidence? Cancer drugs are special. State laws require insurance companies to pay for cancer drugs any time independent cancer guidelines say the drugs should be used. Even if the drug is not FDA-indicated for the cancer, so long as the evidence shows the drug works, insurance companies must pay. Leading oncologists update cancer guidelines whenever the evidence gets good enough.

Why cancer guidelines are your friend

You see where this is going? For an approved drug not to be on cancer guidelines, the evidence sucks.

This is what I do when I'm forwarded information about nonstandard, alternative, or early cancer therapies: I hit delete. I know, even without reading, the evidence isn't there yet. Things that look fantastic almost always fail. Anything early-stage is not helpful for anyone who has cancer now.

Snake-oil sellers are all over cancer patients. They are all over me. These hucksters will make a buck ripping off cancer patients if they can. These hucksters are vultures (or optimistic to the point of delusion). They don't have evidence. See above.

Even legitimate innovators have a hard time imagining it's possible their cancer drug will fail. But their cancer drug will fail most of the time. It's not something scientists like to admit to themselves.

If you want to take an unproven libido booster, that's one thing. But cancer? Don't waste the time you have left.

What is a cancer patient supposed to do when the standard treatments seem to be pointless? What if the odds with standard treatments are so bad that there might as well be no treatment at all? I'd say to ask your oncologist if there are promising, late-stage clinical trials you can join. This is a perfect question.

A late-stage clinical trial is the best chance a cancer patient has to get a next-generation treatment before approval. We're in a golden age of cancer immunotherapy. There are promising immunotherapies in late-stage clinical trials. If you're enrolled in a trial, not only do you get a chance for a new treatment, you will help move the science along so future patients may benefit.

Understanding the disease and its treatment can ease fears

If you or a loved one has received a cancer diagnosis, I'm sorry. I'm sorry this has happened to you. Cancer is frightening. It's all so complicated that, when we get the diagnosis, we don't know what to think. We barely know what to feel. Understanding cancer and its treatment — even the hard things to hear — helped me be less afraid.

I hope, I pray, my story helps you even though your cancer, your experience, may be different from mine. I'm not going to pretend to be an oncologist and give treatment advice — listen to your oncologist — but if you need to talk, I'm at I'll respond if I'm able.

What's next? "Pre-bunking" instead of debunking bad information about cancer treatment

To my colleagues in the health care industry: There is an opportunity to do good here. The cancer patient needs a trusted, friendly voice to help explain things — on call, 24/7. The health care system isn't prepared to do this. The vacuum is filled now by fraud and fear.

Cancer patients today are not in a neutral information environment. Instead, cancer patients are flooded with false facts and quacks who promise 100% cure rates. That's the reality we live in.

There is one defense against misinformation that we know works: pre-bunking. We must fill cancer patients with facts in forms they can understand before the frauds get to them. How do we do this without hiring a call-center army of oncologists? I'm hopeful that artificial intelligence trained on the best evidence will be a "cancer counselor" that will be there to explain things to patients anytime, day or night.

There is upside for us all: Patients who follow evidence-based medicine have double the chance of surviving their cancer, research has found. Demonstrating an AI cancer counselor has a positive effect on medication compliance or even overall survival in a clinical setting should be possible with a modest number of clinical-trial patients.

The pieces are there. Done right, an AI cancer counselor could save more lives than many cancer treatments. If I survive my cancer, I hope to join you in the effort.

Jeff Stewart is a managing director at Syneos Health Consulting. All views, thoughts and opinions expressed here are his own, and not necessarily those of his employer or others. This essay was adapted from the book, Living: Inspiration from a Father with Cancer, published by Wadsak-Stewart Press on May 15, 2023. He can be reached at

Copyright 2023 NPR. To see more, visit

Jeff Stewart