A Waiting Trial
In the summer of 2006, Doug Bergman had a needle plunged into his heart 10 times for science. Now he has leukemia.
When I knock on the door of room 733 of the MeritCare Roger Maris Cancer Center in Fargo, ND, Doug Bergman is sitting up in his bed. He's been expecting me, but it's not me he's waiting for. On this mid-October day, wearing flannel pajama bottoms and a brown T-shirt, Bergman, 55, is anxiously awaiting the results of a bone marrow biopsy he had this morning.
The biopsy wasn't pleasant. "'Geez, you've got hard bones,' that's what the guy doing the biopsy said," says Bergman. "The worst of it is when they draw it out. They make you...
MeritCare's room 733 is the third hospital room I've sat in with Bergman. In August 2006, I spent the better part of two days with him at the Minneapolis Heart Institute, where he underwent a clinical trial, ACT34-CMI. The trial tested whether injections of autologous CD34+ cells directly into the myocardium can reduce the number of anginal episodes in patients with refractory chronic myocardial ischemia (CMI). (See our October 2006 issue for a report on the trial.)
For five days that month, Bergman received shots of Neupogen, or granulocyte colony stimulating factor (G-CSF), designed to boost his white cell count. It worked: When doctors hooked him up to an apheresis machine that sucked the white cells out of his blood, his white cell count was 36,800 per microliter of blood, more than four times his normal.
Researchers then brought his white cells to the lab, where Baxter's Isolex 300i Magnetic Cell Selection System plucked out the cells with CD34 proteins on their surfaces. (Baxter is the trial's sponsor.) The next day, doctors jabbed a needle filled with either saline or Bergman's own CD34+ cells - the trial is placebo-controlled and double-blinded, so he doesn't know which - into his heart 10 times to deliver the cells.
For a while, Bergman didn't notice any changes in his symptoms. At his six-month follow-up appointment in late February, he was still having the same number of anginal episodes as before the procedure. Then, "about a month after that I noticed I wasn't getting them as often." He was pleased, and hopeful.
Then, last April, his legs started to feel tired. When I spoke with him in July, he didn't even mention it, and he also didn't mention it to the ACT34-CMI team during his one-year follow-up visit in Minneapolis in late August. But soon, others noticed, too. "I remember one day I took him and my grandson over to Como Park," Bergman's sister, Jo Ann Wilber, told me. "His legs were so heavy, he couldn't even walk."
When doctors checked the arteries in his legs in August for plaque buildup, which might have explained the fatigue, they found nothing. But they also ran a routine blood test that showed very low white blood cell counts. Doctors repeated that test, found that his white cell count was just 600 per microliter, and referred him to MeritCare Cancer Center.
In September, doctors diagnosed Bergman with acute myelogenous leukemia (AML).
Bergman's sister e-mailed me on September 24 to tell me the news. "I'm not real sure why Doug wants me to tell you this, but he does. He said if you wanted you could call me."
That month, MeritCare had admitted Bergman for his first round of chemotherapy, with cytarabine, also known as Ara-C. It was six rough days. Bergman had all the classic side effects: He lost 24 pounds, hallucinated, vomited, had sores in his mouth, lost his hair. "They are hoping to kill cells so he can get a bone marrow transplant," Wilbur wrote. "Praying that I am a match. ..."
Three weeks later, I was standing at the doorway of room 733, looking at a man with much thinner hair than I remembered and a catheter coming out of his neck. The sign on the door urges visitors to wash their hands; patients with leukemia are at high risk of infections because of their low white blood cell counts.
I sit down with Bergman and his sister, and we look out the picture window at the fall foliage surrounding Fargo. It's a beautiful time of the year in Rochert, 50 miles east of Fargo, where Bergman lives with his wife of 35 years, Sharon. They ran a truck repair shop, with Doug as the mechanic and Sharon as the bookkeeper. They had always wanted to live on a lake, so in 1995 they bought 45 acres on a lake in Rochert. Three years ago, they bought another 120 acres. It's the kind of place where you can sit at the window and watch deer graze in the backyard. The house is filled with animals, including the couple's three dogs, Timber, Augie, and Chloe.
Bergman smiles as he tells me his three-year-old grandson, Spencer, thinks his grandmother is just being mean. Spencer "thinks that if she would just come pick me up, I could come home."
AML is a very unpredictable disease. Most patients have none of the usual risk factors for cancer, such as a history of smoking or exposure to radiation. I wanted to know: Could Bergman's cancer be related to the trial?
The ACT34-CMI consent form, approved by the Allina Health System's Institutional Review Board on July 14, 2006, is 17 pages long. Pages 9 through 11 detail the potential risks of each stage of the study. Some have to do with the various tests performed during the run-up to the trial: chest pain or shortness of breath during stress testing, and allergic reactions to the dyes used during MRI and SPECT nuclear scanning. Others are what patients would typically be warned about before a cardiac catheterization: bleeding, blood clot, arrhythmia, and even heart attack. The form describes other more unique potential risks of having a needle jammed into the heart: inflammation, a needle going through the heart, or "currently unknown side effects that might be permanently disabling or fatal."
After the first e-mail from Bergman's sister, it was the section describing "potential risks of stem cell mobilization" that interested me. After all, G-CSF stimulates white blood cell colonies, so some of those colonies might be AML progenitors. It seemed to me that it was the only potential aspect of the trial that might feasibly have led to his cancer. Researchers have shown that G-CSF binds to AML cells with certain receptors (such as transferrin and G-CSF), and can boost those colonies (Blood, 74:2668-73, 1989; Br J Haematol, 77:54-9, 1991; J Cell Physiol, 148:421-5, 1991). And, there's a sort of natural experiment to test the theory: Stem cell donors are routinely given G-CSF to boost their white blood cell counts before they donate, to maximize the number of stem cells circulating and therefore available for donation.
I asked Stephanie Lee, who studies the before-and-after effects of stem cell transplants for donors and recipients at the Fred Hutchinson Cancer Research Center in Seattle, about the risk of AML following G-CSF administration. The Hutchinson is a major stem cell transplant center, performing more than 400 transplants per year. The AML risk is theoretical, and not yet proven, she said. "There are a couple of reports of AML arising in people who received G-CSF prior to collection," she told me by E-mail. "However, since these are such rare events, it's impossible to know if this is just coincidence." In at least one of the cases, the donor was related to the patient receiving the transplant for AML, so it's possible that they both had a genetic predisposition. Still, she wrote, "Most transplant physicians probably mention leukemia development as a possible risk to healthy donors, and say that the risk is theoretical without substantial support in the literature."
"There are some laboratory studies suggesting that growth factors induce epigenetic changes," Lee wrote. (I found one such study in the January 2004 issue of ExperimentalHematology in which researchers found altered replication timing of alleles.) "These results are concerning to the transplant community, but without the clinical correlate of leukemia development, it's hard to know how to interpret them." If she were going to be a donor, Lee told me, she would have G-CSF, unless there was some reason why the other option, bone marrow biopsy, would be a better choice.
The section of the ACT34-CMI consent form describing the potential risks of G-CSF is one paragraph long. It notes that patients who are allergic to Escherichia coli-derived proteins, or who have sickle cell traits or disease, should not get G-CSF. "Other side effects include bone pain, headache, increased heart disease symptoms, and changes in your blood cell counts or anemia," it continues. "In extremely rare cases, rupture of the spleen has occurred." I checked the study's Investigational New Drug Protocol. It describes the same risks, in more clinical detail and with statistics on each side effect.
Nowhere does either document mention AML.
There have been other cancers in Bergman's family - small-cell lung cancer killed his father - but never a case of leukemia. That's not unusual; the rates of familial AML are "diminishingly small," according to University of Washington medical geneticist Marshall Horwitz. However, as is true for many rare diseases, first-degree relatives of those with the disease - Bergman's sister, Jo Ann, is now in that category - are at a three- to-five-fold higher risk than the general population. This suggests the existence of rare hereditary mutations.
Bergman has also been exposed to two substances that could, at least theoretically, increase his risk: Agent Orange, when he served in Vietnam, and vascular endothelial growth factor (VEGF), as part of a different trial he enrolled in two years before ACT34-CMI. However, research has never shown conclusively that Agent Orange increases the risk of AML in veterans, and studies that identified an increased risk in children of veterans were shown to be based on a miscalculation (Lancet Oncol, 3:199, 2002). VEGF, another growth factor, might also promote cancer growth. But given its mechanism of action, says Ralph Levitt, Bergman's oncologist, VEGF is probably more likely to retard cancers than it is to promote them.
Levitt says Baxter called him and asked whether he thought Bergman's AML was related to the trial. "I said 'probably not,'" Levitt tells me when I meet with him in MeritCare's oncology clinic. (Bergman gave Levitt permission to discuss the case with me.) He notes that oncologists use G-CSF to support patients with leukemia, to boost their white blood counts. "We don't know, 99% of the time, what causes leukemia. These types of toxic exposures, many times, take many years to manifest themselves." Since Bergman's cancer developed about a year after the trial began, the time frame doesn't make sense, he says.
The two-year-survival rate for AML is about 25% without a stem cell transplant, which boosts the survival rate to 40% or 50%. Bergman has a few things working in his favor: His particular chromosomal abnormality - a translocation of bits of chromosomes 8 and 21 - gives him a slightly more favorable prognosis than average. And he's under the age of 60, after which the prognosis worsens.
His heart makes treating his AML a challenge. Oncologists typically blast the bone marrow with chemotherapy to kill all the AML cells, and then give patients stem cell transplants, usually someone else's. Such transplants - "the best chance Bergman has," Levitt tells me - take their toll on the heart, which in Bergman's case was compromised to begin with.
Plus, the standard drug used to treat AML, idarubicin, is highly toxic to the heart. That means that if Levitt used it, Bergman's ejection fraction - the percentage of blood in his left ventricle that leaves the heart with every beat - might dip below 45%, making him ineligible for a stem cell transplant. Instead, Levitt used Ara-C, which doesn't have cardiac side effects. So far, Bergman's cardiac function is holding up. His last ejection fraction reading was 50%. "His heart's doing pretty well," says Levitt. "At this point, his echocardiogram is borderline normal, not in the abnormal range."
All of that means that Bergman is healthy enough to have a bone marrow transplant. Now the question is where the stem cells will come from. Unfortunately, his sister was not a close enough match. (Had Jo Ann - who is now considered higher-risk for AML - been a match, she would have had to consider being treated with G-CSF before the donation.)
As much as he wants to have a transplant, Bergman isn't looking forward to the four to six weeks it will keep him in a Minneapolis hospital, away from his lake in Rochert. However, he may have other options: The ACT34-CMI researchers may have saved some of his own stem cells when they were collected at the beginning of the trial. "The intriguing part here is that it may be that he has some of his own cells in Minneapolis," Levitt says. "The caveat is: Are there any clones of early leukemia that may be transplanted back?" Ideally, the ACT34-CMI researchers would analyze the stored cells, both to see if they're safe to transplant, and to determine whether Bergman had leukemia before the trial started.
Obviously, this is information Bergman would like to have. "We need the help now," his sister says. "How many people get a chance to use their own cells?" But that information hasn't been easy to get.
Bergman, because he had taken part in the VEGF trial, was familiar with the concepts of "randomized" and "blind," and he knew that neither he nor the researchers would know whether he had his own CD34+ cells or placebo stuck into his heart. But page 2 of the ACT34-CMI consent form also contains this sentence: "However, in the case of a medical emergency, this information can be made available to you and your study doctor."
As of early December, the Heart Institute hadn't told Bergman what he'd received. There had been conflicting reports, according to him and his sister, about whether the ACT34-CMI researchers would be able to tell him, or give him any cells stored in Minneapolis. If there were any cells left, the decision seems to be Baxter's to make, since the consent form says leftover cells are either donated to Baxter or destroyed.
According to the study protocol, hospitalization qualifies as a serious adverse event that must be reported to Baxter. But the study requires such reporting for only 12 months, and Bergman's AML was diagnosed 13 months after his procedure. Did the company have a position on whether his AML was a potential side effect of any part of the protocol? Did any other participants (now more than 100 patients, at 20 sites) develop AML, or other significant adverse events? The day that Bergman started his first round of chemotherapy, Columbia University Medical Center in New York announced it had enrolled the first New York City patient in the ACT34-CMI trial. At the time of Bergman's diagnosis, one of the trial's principal investigators, Jay Traverse, told me that there hadn't been any other cases, and he and his colleagues knew of no relationship between the protocol and Bergman's AML.
In October, Baxter spokesperson Jed Weiner told me that Baxter couldn't discuss any trial subject's personal information or any safety data. "We need to maintain the clinical integrity of this and our future clinical trials," he said. I pressed him on whether a diagnosis of AML constituted the kind of medical emergency that would convince researchers to unblind Bergman's records. The protocol says a subject may be "withdrawn/prematurely terminated from the study" for five reasons, which include "subject or caregiver request" and "withdrawal due to an adverse event (physician's judgment)."
"The study has not been unblinded, and so therefore if Mr. Bergman's diagnosis is accurate, and if the study and his participation has not been unblinded, you can draw an inference from that," was all that Weiner would tell me. "I wish I could share more with you. From a personal standpoint I hope he has good recovery."
Baxter plans to enroll 150 patients in the ACT34-CMI trial by the end of March 2008, and, given the 12 month follow up, present results some time in 2009. Results from similar trials have been mixed, at best. Three of the trials so far haven't shown any positive effects, and the one that did was questionable. According to Norwegian cardiologists who had been investigators on one trial, the one positive study used poor methodology, and its positive results were due to poor outcomes in the placebo group, rather than real success, they wrote in The Lancet last June. They urged others to use caution before testing stem cell therapies in patients following heart attacks.
When I spoke to him last July, before the diagnosis of AML, Bergman told me that he felt "very, very good" and that he was glad he had enrolled in the trial, for which he was paid $400 (much of which he spent on gas driving to and from the heart institute). But when I visited him a few months later in October, his opinion had changed. I asked him whether he would enroll in the VEGF and ACT34-CMI clinical trials if he had the chance to go back in time. He thought for a minute before answering. "I would say no, to both trials." Does he think the ACT34-CMI trial caused his leukemia? "I don't know how to answer that question. It's almost as if [the researchers] don't want it to be connected. I think they're being a little secretive," he said.
Soon after I saw him, Bergman got some good news: The biopsy he had in mid-October had come back clean. (The results hadn't shown up before I had to leave Fargo.) "Dr. Levitt said 'the blasts disappeared,'" he told me when I called him. "'Your blood right now is looking good, but it's not going to stay that way.'" When I ask if he's looking forward to getting started with the transplant process, he laughs heartily and says, not surprisingly, no. "I'd rather stay home, but I guess Dr. Levitt convinced me I've gotta get going or it's going to be not good. I have to get it over with, I guess."
Then, in early December, he got more good news. There were two matches, one adult human and one umbilical cord. Bergman had still been in remission as of Thanksgiving, so the transplant was tentatively scheduled for early this month. And there was even a possibility that he wouldn't need a transplant. When I spoke to him just before we went to press, however, he was back in the hospital. He hadn't been feeling well, and had spiked a fever.