What started out yesterday as a routine request from a journalist to comment on a routine press release about a scientific report became one of the more interesting examples of why what you read may not always be what you think it is.
Several months ago, I discussed the information process and how it influences the behaviors of patients and physicians in cancer treatment. An important element of that discussion is the need for everyone to be responsible in the process. In fact, not just responsible, but exceptionally responsible, since the reports of cancer research outcomes may affect someone’s life very directly.
It is not just the doctors and the patients who are interested in what the media has to say about cancer treatments and breakthroughs. So are the financial folks, who look for clues that a company has developed something important that may influence the performance of that company, and eventually impact their profits and their stock prices.
How a story is presented can influence those very important perceptions. Presentations don’t have to be wrong to be misleading.
The “process” used to work differently years ago. Companies or research institutes would develop a drug that would be a candidate for a new cancer treatment. The compound would be put through a series of laboratory experiments, and if those were successful the drug would find its way into clinical trials in cancer patients.
These trials would be conducted generally in three “phases”: phase I was the initial use in patients, generally people who had failed other potentially useful drugs, and who had no other options. The drugs would be administered in increasing doses, with an eye primarily on the toxicity of the drug, and to some degree which cancers (if any) responded to the new drug.
The process was very deliberative, and detailed observations were made.
After the phase I trial was completed, the investigators would move on to phase II, where they would try the drug in cancers that either theoretically might respond to the new drug, or in those cancers were some response may have been seen in phase II.
If responses were noted in a specific cancer, the drug would then move on to phase III, usually with a pretty good understanding of what the tolerable and effective doses would be, and with an eye towards treating larger numbers of patients with specific cancers where the drug had shown some benefit in phase II.
At each of these junctures, if the trials were successful, the investigators would prepare reports to be presented at meetings of their peers, and submit their articles describing their work to medical journals. These journals would have the reports and the data carefully scrutinized by peers of the researchers to be certain the data was complete and fairly presented. If the research paper passed muster, it would be published for all to see.
This wasn’t a perfect system, but it wasn’t bad either. It relied on trust in reporting the data, and the investigators usually didn’t have much direct personal benefit in the outcome of the study, other than the accolades and glory that would come from doing good work that may have had significant impact.
A couple of years ago, when I was working as a medical editor, I began to take note of a trend that was somewhat disconcerting to me. Once again I need to stress that there was no wrongdoing or implied wrongdoing, but I found it interesting nonetheless. My observation was that a number of the reports of Phase I and Phase II trials were coming directly from the drug companies in the form of a press release. This would happen, not infrequently, before any data had been presented publicly in either a presentation or a journal article.
On several occasions, since the press release had received notice, I would have to prepare a report for our web news stories. I would call the phone number on the press release, ask for the data, and was told that it was not available. Or an abstract had been presented and I couldn’t get access to the slides from the meeting.
The result was that, not infrequently, I didn’t write the story or if I did I was not comfortable that I had all of the information I needed to judge for myself how significant the findings really were (I won’t get into detail here how numbers can be made to look good, but you better believe they can be presented in a number of ways that makes a benefit look larger than it is. For those of you out there who are statistic mavens, think in terms of absolute and relative risk.).
That is what brings me to yesterday’s story.
As I mentioned, it started out as a routine request from a competent reporter who I have talked with previously. A company had issued a press release announcing outstanding results in the treatment of an aggressive, non-curable form of lymphoma. They implied that the results were related to the use of a patient specific vaccine the company manufactured. The company’s stock, though thinly traded, went up after a number of the financial news services spread the word.
Vaccines are a hot topic these days, and I have commented on recent reports regarding melanoma and pancreatic cancer vaccines in late 2005. One of these reports was from the company, and another was reported in a presentation at a meeting. Both were promising.
There were other recent research reports about the immune response to vaccines. One in particular described how a kidney cancer vaccine produced an immune response that most of us believe is important to getting good results in cancer treatment. But, and this is a very important but, those researchers did not say in their reports that this implied a breakthrough in treating patients with kidney cancer. That would be the subject of continuing clinical research.
So here we have a headline that implied new research results (performed, I might add, by very reputable researchers from very reputable cancer centers and reported in a very reputable medical journal) showed amazing responses in a disease called mantle cell lymphoma, which is notoriously difficult to treat and cannot be cured.
As I mentioned, the company had issued a press release on the topic. Again, nothing wrong with that. They quoted the number of people who had a response to the “new” treatment with chemotherapy and vaccine. Nothing amiss there either. If the headlines were right, this was a significant piece of research and the information was important.
But, being the somewhat stubborn person that I am, I asked for a copy of the article reporting these results.
Shortly thereafter, with the able help of my colleagues at the Society, I had an email copy of the journal article on my computer.
But, I thought, there must be some mistake. This was an article from September 2005, not January 24th. And, it did not have any mention in the title that this was about a new treatment for patients with mantle cell lymphoma. No, this was an article describing in elegant detail how the researchers measured the immune response of cancer patients in a unique situation where there immune cells had been destroyed by another generally accepted cancer treatment called Rituxan, and how the patients developed (or didn’t develop, as the case was for several of them) an immune response to another vaccine that was administered at a time when their immune system was severely compromised.
That’s not to say the article didn’t say how many patients were treated with the chemotherapy/Rituxan/vaccine combination, nor is it to say that the article didn’t mention in passing how many of the patients responded to the chemotherapy (the vaccine in question was administered, in most cases, several months after the patients had completed the initial two phases of treatment). And, in another part of the paper, the authors indicated that most of the patients had relapsed.
No, this was not a paper being presented to show the results of a new cancer treatment. This was a paper about the immune system, and whether or not it had to be completely intact to develop antibodies to the vaccine. End of story.
Now my interest was piqued even further. As is usually the situation, these stories land on my desk in the middle of my other daily activities. So I must emphasize that I did not have a lot of time to do a lot of research. But, based on the references in the journal article, and some other search techniques that I commonly use, I did a quick evaluation of what had been published on mantle cell lymphoma treatment with a specific eye towards seeing whether anyone has reported recently success with this particular form of treatment.
The answer? I couldn’t find anything suggesting that this particular vaccine had made any significant contribution to the treatment of mantle cell lymphoma. Review articles in reputable cancer treatment journals didn’t mention the vaccine. An article published in 2004 in a cancer journal written by one of the investigators who participated in the original research on the vaccine reported in 1999 said that the vaccine merits consideration “as another important option for investigation in MCL (mantle cell lymphoma)”, and references a phase II trial report in the same journal—from 2002 (I couldn’t access that report since we do not have a subscription to that journal, and the journal did not make the paper available on line when I tried to access it).
According to the National Cancer Institute, mantle cell lymphoma is considered an aggressive lymphoma that is not curable with chemotherapy. They report that the median survival of patients with this lymphoma is 3-5 years (median means that half the patients live less than 3-5 years, and half live more). Articles on MCL in a reputable cancer journal say that the median overall survival for patients with this cancer is approximately 3-4 years. Another editorial said that the median “progression free survival” was 20 months.
So, I went back to the more recent article that led to the press release. What information that was available in the article said that with the chemotherapy, 92% of the folks treated with the chemotherapy had a complete remission, and 8% had a partial response. Those are indeed striking numbers. But, looking further at the data, all of those responses had occurred before the vaccine was given. In fact, in an article published in 2002, the researchers reported that this particular form of chemotherapy did indeed have an excellent response in a variety of lymphoma patients. And, based on the actual observation of the patients, their median overall survival was probably going to be 89% at 49 months, which means it was projected and not actual.
There was more information. Patients for the most part didn’t get the vaccine until at least 3 months after the completion of their chemotherapy—and the administration of chemotherapy occurred over several months after the patients were diagnosed.
Then there was the information in the article about how long these patients remained in complete remission. That is the measure of real success in treating aggressive lymphomas, where the idea is to achieve a complete remission that may result in a cure. That number tells a different story, since most of the patients had relapsed. The median progression free survival was 22 months. But that is not much different than has been reported in other recent studies.
The bottom line of this is that these patients responded well to chemotherapy. They also may live longer than has been the situation in other studies. But that does not seem to be because their initial treatment was remarkably successful. In fact, it may mean that subsequent treatments have had excellent results. The report itself made no claims that this was a breakthrough treatment.
What does all of this mean? First, in my opinion, one must always be very cautious when reading media reports touting a new breakthrough, especially when the information is based on a press release from the company.
Second, it is incumbent on the press to read these press releases critically if they are doing something more than just passing on a press release (and they indicate that in fact is what they are doing).
But my real concern is the message this may send to patients and families and others who are coping with a serious life-threatening disease. There must always be foremost in our minds the impacts that press releases and media reports are going to have on patients facing a life threatening illness. We should be held to a high standard in our research, our reports, and our representations.
I have not been able to find the type of report in the literature or referenced in the article that would lead me to change my recommendations for the treatment of a patient with mantle cell lymphoma, supporting a breakthrough as a result of the vaccine. In fact, the major breakthrough in this situation appears to be the chemotherapy itself. But despite that success, it hasn’t yet resulted in cures or a significant advance in the treatment of these patients according to the experts in the field.
This won’t be the first time someone presented information in the most favorable light. And, technically, there was nothing wrong in what was done here.
But, again in my personal opinion, we really owe our patients a higher standard than to suggest an article describes a breakthrough therapy, when it is clearly not the intent of the article nor justified by the results over time.
And, as always, if there is information that is available to refute my conclusions or clarify the information I have presented and discussed here, please, please let me know. I would welcome being shown that I erred in my conclusions.
But, if I’m on target, please be careful in the future and understand the responsibility to not only get it correct, but to get it right.