How We Do Harm (23 page)

These approaches—flying by the seat of your pants, combining everything under the sun, and continuing to treat even after the signs of the disease are gone—continued to guide Snuffy
et al.
a quarter of a century after Freireich and Frei, when I arrived at NCI.

Fast-forward another quarter century, and we still use combination therapies, often administering new, expensive, targeted drugs with the old warhorses of oncology, drugs that have now been around longer than some of the oncologists prescribing them.

We still conduct early-phase research to determine the side effects to set the “maximum tolerated dose” of just about every new drug.
We still argue about optimal strategies for stopping treatment: how long do we continue to treat after all visible signs of disease disappear?

Do we continue to treat after therapy appears to have failed?

*

MANY
observers predicted that VAMP would be horrendously toxic and would end in disaster, which indeed seemed to occur when several of the kids treated at NCI ended up in comas and on ventilators.

Yet, the kids recovered, and biopsies of the bone marrow showed that they were free of disease.
Could VAMP have been the cure for childhood leukemia?

There was much rejoicing, which came to a halt after several kids started to develop new neurological problems.
These problems led doctors to discover that in most cases VAMP wasn’t curing leukemia.
Instead, some of the disease was hiding behind the protection of the blood-brain barrier.
These kids relapsed with disease in their brains.
About 5 percent of the children survived for more than a year, and only a small number of treated children, for reasons no one understands, became long-term survivors.

*

THROUGHOUT
his long career, Freireich has sided with organizations that argue that a dying patient should not be impeded from taking any risk he or she chooses.
Predominantly, these are political conservatives who object to the requirement by FDA that clinical researchers obtain “investigational new drug” licenses before administering an experimental therapy.
An IND is given after the drug sponsor has presented data to the FDA demonstrating that the drug can be tested safely.
The FDA does not require any proof of efficacy to grant an IND.
Clinical trials done under an IND ultimately develop the data, which the drug’s developer eventually takes to the FDA to get a new drug approved.
Incredibly, the editorial board of
The Wall Street Journal
has spent the past five decades railing against the FDA requirement that pharmaceutical companies demonstrate that their products are effective in fighting disease.

To conservatives, these requirements, enacted in 1962, are the juncture in history when things went terribly wrong, a kind of a regulatory equivalent of the original sin.
The same groups also mount ethical challenges to placebo-controlled randomized trials, arguing that it’s morally wrong to withhold a treatment from half of your patients.

Freireich has often said that the nanny state personified by the FDA bureaucrats has no business inserting itself into the doctor-patient relationship.
As an octogenarian physician at MD Anderson Cancer Center in Houston, J continues to argue against big government getting involved in the clinic.
A few years ago, J joined a group called the Abigail Alliance for Better Access to Experimental Drugs, which sought to make drugs available immediately after they pass through Phase I testing.

Phase I studies establish the maximally tolerated dose and side effects.
The primary aim of a Phase I trial is not to determine whether the drug benefits the patient.
Indeed, only 5 percent of cancer drugs show anticancer activity in Phase I testing.
Effectiveness is determined in Phase II studies.

The group’s campaign resulted in a court case that was upheld by a three-judge panel at the U.S.
Court of Appeals for the District of Columbia.
Ultimately, a panel of all judges seated on that court rejected that petition, and the Supreme Court refused to reconsider the appellate court’s ruling.

In the heat of that battle, Freireich spoke to a reporter about his frustration with the power of the FDA.
“They can’t regulate the interaction between a physician-scientist and a dying cancer patient,” he said to
The Cancer Letter
in 2005.
“That is something that is in the area of professional expertise, and they don’t know shit about it.
You give power to an agency, the agency has to find people who will do this drone work, so they look for failed oncologists, and the failed oncologists love this position, because now they have power.”

FDA is just one component of regulatory oversight of clinical trials.
In an interview with
The Cancer Letter,
Freireich was particularly angry about the role of his former MD Anderson colleague Richard Pazdur, a colorectal-cancer expert who left for Washington to become head of the FDA’s cancer unit.
Pazdur is a favorite target of attacks by the Abigail Alliance and
The Wall Street Journal
editors.

“Look at Ricky Pazdur!”
Freireich continued.
“He was so-so and not doing much; now he is King Kong!
He gets invited to every talk on every drug at every meeting everywhere in the world.
Who would know what is the highest probability of benefiting a patient and the lowest probability of doing harm?
Is it Dr.
Pazdur, who’s been sitting at a desk for ten years, or is it Dr.
Freireich, who’s in the clinic, beating his ass, taking care of dying cancer patients?
Who knows more about this?

“What is wrong with us?
The system is upside down!”

I find it interesting that Freireich’s political position is rooted in his early success, which indisputably paved the way toward the cure of childhood leukemia.
Had success stories driven by instinct been the norm rather than an exception, Freireich’s argument would have supported elimination of any and all efforts to ensure that patients are told the truth and are protected from exploitation, whether it’s motivated by financial greed on the part of a private doctor or eagerness on the part of an academic to find another subject for a clinical trial.

Alas, such triumphs are not often repeated.
Instead, the scenario most frequently replicated in drug development is better illustrated by suramin.

*

I
got to know J in the early 1990s, when he spent a year on sabbatical at the NCI.
Tall, heavy, solid, and in his seventies, he was approachable and friendly.
He joked that MD Anderson had sent him into exile (this was true), and NCI director Sam Broder had asked him to study the training of medical oncologists.

J’s sense of humor takes a while to get used to.
Once, in front of me, he said to my mentor Peter Greenwald, then director of the NCI Division of Cancer Prevention and Control, “We finally find a good one and you pervert him by convincing him to be interested in this prevention shit.”

The word
one
in “a good one” stood for “Negro.”
This backhanded compliment threw me back on my heels for a moment, as I had rarely experienced racism from doctors.

The second half of J’s joke was equally shocking: the gullible mind of this just-discovered “good one”—Otis—was being corrupted with dangerous teachings about cancer prevention and all the nonsense that goes into it.

For starters, J dismissed the concept of public health.
He went on to expound on his contempt for Phase III trials.
These trials compare two treatments to determine which is better.
“When a treatment works, it is obvious that it works, and you don’t need a study to show it,” he said.
Then he went on to slam prevention, a discipline highly dependent on Phase III trials, many of them involving placebo.

Finally, on the off chance that he was being too subtle, J explained that public health is self-evidently an absurd concept.
“When a group of people is stranded on a desert island, they decide which of them is going to
treat
disease.
They don’t decide who is going to
prevent
disease.”

As the conversation progressed, I let go of my initial feeling of outrage.
I understood that this was J being J.
Our conversation demonstrated another manifestation of his unusual sense of humor and a bit of the outrageousness that has punctuated his career.
Yet, two decades later, Greenwald still remembers being shaken by this conversation.

I am still fond of J and have considered him a friend for more than twenty years.
He is an icon of oncology.
I may even share some of his penchant for in-your-face truth-telling.
Throwing hardballs in the debates of medicine is a fair tactic.
(All is fair in efforts to provoke people to think.) However, I disagree with J’s contention that the research world is upside down.

I don’t know how you can responsibly practice medicine without measuring effectiveness of therapies you are administering.
We have to know that the treatments we are testing are better than standard treatments or nothing at all.
Without the requirement that drugs demonstrate effectiveness, we simply cannot know whether we are doing harm.
In the case of drugs for diseases where we have treatments but no cure, only randomization will give you the answer.

I share J’s frustration that many of our successes increase median survival by three or four months at best.
I have never talked to J about it, but I was thinking of him when the combination of Gemzar and Tarceva was approved by the FDA for cancer of the pancreas, because in a clinical trial it increased median survival by fourteen days.

Our progress is not fast enough.

*

RESPONSIBLE
medical researchers recognize that by definition we are dealing with the unknown.
We put patients at risk while we hope to do good.
Often—particularly in Phase I studies—we struggle with a profound dilemma: Whom are we trying to benefit?
Are we trying to benefit the patient in front of us?
Or are we trying to benefit society, which may benefit from knowledge gained through the experiment we are conducting?
We try to be decent and respectful, and here Father Polakowski’s maxim fits particularly well.
We try to explain the truth about the experiment: what are the known risks, what is not known, and what is believed.

For a Phase I trial, you have to know enough to believe that the experiment has merit.
You can’t cut corners—you have to find out everything you possibly can before you put a patient in harm’s way.
If you are testing a drug, for example, you can’t rush into a study without learning as much about the drug as you can in preclinical studies.

When doing a human study, you also have to have what ethicists call equipoise.
It is unethical to put a patient on a study if you are convinced you know the answer.
If you are convinced that Drug A is better than Drug B, it is unethical to give a patient Drug B.

Ultimately, this ability to say truthfully, genuinely, and sincerely that you don’t know is the principal distinction between responsible research and the wilderness.

The very nature of what we do requires a good clinical researcher to have a certain amount of ego, perhaps even a touch of hubris.
If you are a patient, you do not want an insecure, uncertain physician.
This is even more important when you are in a clinical study.
As a practical matter, it’s important to have objective third parties look over the investigator’s shoulder, to make sure that he hasn’t fallen in love with his own ideas, that he isn’t cutting corners.

It’s okay to have beliefs—you need them to formulate the scientific question, the hypothesis, you are testing—but they have to be labeled as such.
It’s also important to make sure that someone knowledgeable and objective is overseeing the researcher and that this person—or this committee—is willing to ask tough questions.

As cancer doctors, we are in no position to promise the cure.
But we do have the moral obligation to tell the truth.
My stake in a clinical study is different from my patients’.
I stand to gain publication in a medical journal, meet my financial quota for selling clinical services (most institutions have those), maintain my cancer center, receive a federal grant, or perhaps earn consulting fees from a drug company.
A patient could end up experiencing longer survival or a catastrophic toxicity.
If I want to be able to look at myself in the mirror, I have to be certain that my patients have a clear understanding of my scientific justification for conducting the clinical experiment as well as my understanding of its potential risks and benefits.

I have to be certain that I have done all I can to explain uncertainty to my patients.
Any shortcut in that endeavor amounts to denying my patients’ humanity and my own.

Chapter 15

The Guillain-Barré Syndrome
BEFORE SCREENING PATIENTS
for the suramin study, I glance at the protocol and see that it’s a classic Phase I, which means that we would begin by treating three patients at a dose we believe to be too low.
Then, if they do well, we would double the dose and treat another cohort of three patients.
And so forth until we slam into a dose-limiting toxicity.
In the beginning, you hurt people by not giving them enough drug.
In the end, you hurt people by giving too much.
When you get to that point, you scale back the dose to the maximum beneficial amount.

As we screen patients, we verify the staging of their adrenal cancer, review pathology, and generally make sure they do indeed have adrenal cancer.
Those whose disease is metastatic to the liver or lung are put on suramin.

During screening I notice for the first time that patients at NCI are remarkably optimistic about their chances of getting a cure.
This is puzzling in a Phase I trial, where the goal is to find the right dose for later studies.
Benefiting the patients, let alone
curing
them, is not the goal at this stage; sure, it could happen by accident, but we aren’t tailoring the experiment to this end.
The question we are asking is limited: how much of this glumpy substance can we give to a human before triggering a catastrophic event?