Monday, November 23, 2009

Does Morphine Stimulate Cancer Growth?


Over the last week Reuters, ABC news, MSNBC, BBC News, and more than 75 other outlets reported on how two "two new studies add to growing evidence that morphine and other opiate-based painkillers may promote the growth and spread of cancer cells." What was most shocking were the headlines used to promote the stories:
All these articles (most stemming from an initial Reuters report) discussed the recent presentation by a group from the University of Chicago on the mu-opioid antagonist methylnaltrexone (otherwise known as Relistor or “that new opioid induced constipation medication”). This work, as presented at a meeting in Boston, revealed that use methylnaltrexone prevented tumor-cell proliferation and migration in cultured lung cancer cells. The group also presented work revealing that genetically altered rodents lacking the mu-opiate receptor failed to develop tumors after being injected with Lewis lung carcinoma cells, something that did occur in normal mice. This is an interesting line of research and one that gave the makers of Relistor a shot in the arm (as one website put it – “a possible new indication for Progenix's Relistor could revive its fortunes”.) I find the pathophysiology behind this incredibly interesting from an academic standpoint, however it is neither something that would warrant such dramatic headlines nor spur thoughts that it is anything but research in its infancy.

What about the claim that there is “mounting clinical evidence” that morphine may accelerate cancer growth? The only real clinical studies in this regard come from a single institution in Ireland and are basically a couple retrospective cohort studies comparing regional versus general anesthesia. The first one of these articles (Anesthesiology. 2006; 105(4):660-4) showed a beneficial relation between paravertebral block and cancer recurrence in women undergoing breast cancer surgery. The study’s authors acknowledge that “selection bias and the effects of unmeasured confounding variables” could not be excluded, as well as the fact that “relevant information such as the amount of morphine given” was not available in the records. The second study was nearly identical except it was in patients undergoing radical prostate surgery (Anesthesiology. 2008;109(2):180-7)

The most important take home point is that none of these studies actually looked at opioid use (although many of the news articles vaguely cite these studies as evidence for morphine’s deleterious effects on cancer). Even if you want to argue that these were high quality studies and there is a clear benefit of regional anesthesia, opioids should not be singled out as the cause.  Benefit of regional anesthesia may be due to a myriad of other effects of regional anesthesia including the lack of volatile anesthetic agents or an improved stress response. These studies are noteworthy, but I agree with the authors of these articles that they should “be viewed as generating a hypothesis and an estimated effect size for future large randomized controlled trials”.

So if you have a patient on morphine asking you whether “morphine will make my cancer worse”, the answer is no. There is no clinical evidence that morphine, or any other opioids, cause real harm through stimulating cancer growth. There is though significant high quality evidence for their benefit in relieving pain and suffering.


References:

Exadaktylos AK, Buggy DJ, Moriarty DC, Mascha E, & Sessler DI (2006). Can anesthetic technique for primary breast cancer surgery affect recurrence or metastasis? Anesthesiology, 105 (4), 660-4 PMID: 17006061

Biki B, Mascha E, Moriarty DC, Fitzpatrick JM, Sessler DI, & Buggy DJ (2008). Anesthetic technique for radical prostatectomy surgery affects cancer recurrence: a retrospective analysis. Anesthesiology, 109 (2), 180-7 PMID: 18648226
ResearchBlogging.org

12 comments:

LeighSW said...

I hadn't heard about this study but I'm sure a patient will bring it up at some point. Thanks for equipping me to dispel the misinformation. Why does morphine always get such a bad rap?

Eric Widera said...

I worried about even posting this rebuke. Am I just extending the half life of this story? Would this news story die quicker if we just leave it alone?

ken covinsky said...

It is kind of disconcerting how misleading these news reports are. As Eric said, the findings are interesting, but to claim such wide ranging clinical implications is absurd.

It would be interesting to know what role, if any, progenix had in the press coverage of this story. Several aspects of this make me wonder. The press coverage stems from a presentation at a scientific meeting. However, scientific presentations usually use the generic name of the drug (eg, methylnaltrexone), not the brand name (eg, relistor), and seldom mention the company making the drug. In contrast, the press coverage gave very prominent mention of the brand name, as well as the company. This, along with the very unbalanced presentation of the evidence makes one curious. I wonder if the reporters involved had any concept of the mammouth leap between the very narrow biologic finding that they described and the clinical implications they describe in the article.

Christian Sinclair, MD said...

Very awesome post. One I wanted to write as well, but you summed it up very well. I just wrote about your post on Pallimed and linked to it.

Yes I think we need to keep talking about this and get a more informed discussion of the issue out there to counter some of the headlines. Posts like this need to be linked to in the comments section of each of the 75 news outlets that posted the original story. When there are issues like this the next step of blogging is to not only speak your mind but to actively seek out those who may have differing opinions or may have a vested interest in hearing your opinion on the matter.

Alex Smith said...

Completely agree with Christian that we need to get this side of the story out there (and thanks for the link on Pallimed). We need to link to Eric's post in the comment sections of the stories from major news outlets. I'll link to some today.

Alex Smith said...

I did some digging along the lines of Ken Covinsky's comment. Dr. Jonathan Moss, the prominent University of Chicago anesthesiologist and co-investigator on many of the studies of methylnaltrexone (including the one about potential for prevention of cancer growth), has the following disclaimer on a 2007 article in the American Journal of Respiratory and Molecular Biology (37:222):

"J.M. serves as a consultant to Progenetics Pharmaceuticals, has a financial interest in methylnaltrexone as a patent holder through the University of Chicago, and receives stock options from Progenics."

Progenics manufactors Relistor (methylnaltrexone). My take on this information - and I'm interested in others' - this does not mean we should knee jerk toss his studies in the circular file. Rather, this is important information that suggests independent investigation (meaning by people with no financial interest) is warranted. This is a reason to treat these findings with skepticism. We still don't know the extent to which Progenics influenced the press of this story, or "allowed" it to originate from Univerisity of Chicago scientists.

Am I being too generous? Or too pessimistic?

While the potential influence of the pharmaceutical industry is concerning, the bigger story is still how the media portrayed this issue.

sei said...

Alex, kudos for doing the work that the journalists should have been doing, uncovering a possible conflict of interest for the author.

I agree with your closing comment, that the story here is not one about pharmaceutical industry malfeasance, but how poorly the media handled this. The media focuses on sensational headlines requiring minimal work. Unfortunately, new biomedical research often fits this bill.

Taking this full circle, part of the blame for the poor job the media has been doing has to go to blogs and other "new media" that have been stealing eyeballs and ad revenue, leading to layoffs in traditional media. This has forced traditional media to focus on ever-catchier headlines that cost less to produce.

Mike Steinman said...

What can we learn from the case of methylnaltrexone? First, the media should do a better job about not jumping on the bandwagon of a “sexy” new story. However, given the current media climate it is probably unrealistic to expect the sudden emergence of a careful, circumspect approach to reporting. Rather, it is incumbent for us as medical professionals to make our voice heard to provide appropriate balance for such stories. The recent fiasco over the roll-out of new mammography recommendations by the United States Preventative Services Task Force is a case in point – solid science and thoughtful consideration which was profoundly undercut by a failure to shape the media narrative on this topic.

Fortunately, we have a solid base of evidence from which to ground a healthy skepticism toward industry-sponsored work – for example, studies by Lexchin (http://www.bmj.com/cgi/content/full/326/7400/1167) and Als-Nielsen (http://jama.ama-assn.org/cgi/content/full/290/7/9210). Our job is to take such evidence about the impact of industry on research outcomes and translate it to the media and public at large – not as a scare tactic or attack on industry, but to encourage a circumspect approach to interpreting medical evidence. We will succeed when we merge our strongest sources of credibility (as scientists and as caring patient advocates) with our ability to craft a compelling and clear way of “telling the story” that we need to communicate.

Christian Sinclair, MD said...

One easy thing we should do is invite Dr. Moss to the discussion. His email is easily found on the University of Chicago website.

I agree we should not throw out research just because it is linked to industry funded. We would basically deplete a good portion of PubMed. But in good scientific method replication of outcomes is essential with independence.

I agree with sei, in that my major concern is the over-hyped reporting and headline effect and the potential it has to cause some people harm based on failure to take opioids which may be beneficial to them.

I think this issue, like the FDA/Morphine decision is one for the great reasons for blogs to exist. We as medical professionals should be having these thoughtful discussions openly so that professionals and the general public can access this information.

Sadly this story has gotten wide enough coverage that if you type in 'morphine cancer' in Google several of the links on the first page link to news reports that 'Morphine may cause cancer growth.'

I emailed this link to several people today and asked them to share with their teams so they are well-informed if this issue comes up with patients or families.

Great resource links Mike.

Eric Widera said...

The global net sales of Relistor for the third quarter of 2009 was only $3.3 million (Lipitor is about $3 billion a quarter). I’m guessing the push to find alternative revenue streams for this drug is very strong.

Sandi Pniauskas said...

While the media and pharma are often criticized in situations such as this, there is the communication factor - good or bad. The value to patients of harm vs benefit is that the information is available and sufficient enough for patients to question, research and also take the information for discussion to their professional caregivers. The e-patient communities welcome dialogues but not in an exclusive setting which is absent their participation.

Rogue Medic said...

The media deserve the criticism.

If the media made scientific literacy a requirement for those who report on science, they would not be repeatedly criticized for this egregious incompetence.

We expect that the people they hire will have competence in the job. A person hired to drive is expected not just to have a valid license, but to have a good driving record. If they give the keys to someone with a history of bad driving, they should expect criticism, as well as increased insurance premiums and legal costs.

The media put the reporting of science in the hands of people who regularly demonstrate that they do not have a clue about science.

Gary Schwitzer's HealthNewsReview Blog is an excellent blog that focuses on the way the media portray medical stories. He occasionally is able to post something positive, but the poorly reported stories occupy much more of his time.