Comments

Health Affairs encourages readers to engage in discussion via comments on our Web site.

• To RESPOND to a particular article: Click on the link "Submit a response to this article" in the box at the top right-hand corner of the article.
• To READ responses to a particular article: Click on the link "View responses" in the box at the top right-hand corner of the article.

Comments to:

Joshua T. Cohen and Peter J. Neumann What’s More Dangerous, Your Aspirin Or Your Car? Thinking Rationally About Drug Risks (And Benefits) Health Affairs, May/June 2007; 26(3): 636-646. [Abstract] [Full Text] [PDF] [Reprints & Permissions]

Comments:Submit a response to this article

Comments published:

Facts on Tysabri

David E. Kaplan   ( 9 May 2007 )

Response To Tysabri Critique

Joshua T. Cohen, Peter J. Neumann   ( 9 May 2007 )

Majority of the MS Community's Views on Tysabri

Lauren M. Roberts   ( 10 May 2007 )

An Opportunity For Clarity

Harold Katlin   ( 10 May 2007 )

Risks And Benefits Of Tysabri

Joshua T. Cohen, Peter J. Neumann   ( 10 May 2007 )

Tysabri Reference In The Cohen/Neumann Paper

Chris Chan   ( 10 May 2007 )

Root-Cause Failure Analysis Of Drug-Induced PML Morbidity

Mike Barron   ( 14 May 2007 )

Valuable Statistics

Andrew H. White   ( 14 May 2007 )

Facts on Tysabri 9 May 2007
David E. Kaplan, MS Volunteer www.MSPatientForChoice.org Send comment to journal: Re: Facts on Tysabri davidksra{at}comcast.net David E. Kaplan	I admittedly have read only the abstract and the Reuters article by Maggie Fox based on the full paper by Neumann and colleagues, but I must comment based on that, as those presentations of your article have received widespread publicity. While no expert on risk, I have followed (and been involved in) the Tysabri suspension and reapproval, on behalf of thousands of multiple sclerosis (MS) patients. Although I am sure the paper meant to be informative, the brief reference to Tysabri in the Reuters article (I presume based on the Health Affairs article) was terribly inaccurate and misleading to the millions of MS and Crohn’s disease patients, their family members, and the neurologists and gastroenterologists who read Reuters’ reports. The Reuters article summarized the Tysabri portion of Neumann and colleagues' paper by stating that “Using Tysabri, known generically as natalizumab, to treat multiple sclerosis raises the death rate by 65 per 100,000 people a year. Biogen Idec's and Elan Corp's Tysabri was withdrawn from the U.S. market last year after three patients contracted a rare brain disease, but the Food and Drug Administration is reconsidering it after many patients said they would use it despite the risks.” First, Tysabri was withdrawn from the U.S. market more than two years ago, on February 28, 2005 (not last year) when two cases of PML (one confirmed, the other suspected and later confirmed) were encountered in clinical trial patients on the combination of both Tysabri and Biogen Idec’s other MS drug, Avonex, for a period of over two years. One of these patients died, but the other patient survived. Since Tysabri had only been on the market commercially in the U.S. for three months, out of prudence Biogen Idec and its partner, Elan, in consultation with the FDA, decided to withdraw the drug while an independent, exhaustive safety evaluation was completed. In the course of that investigation, a third case of PML was found in a Belgian patient who had died more than a year earlier during the clinical trial for Crohn’s disease, for which Tysabri was also being evaluated. That patient’s death had been attributed to other causes but was determined to have been from PML, although, importantly, that patient had a long history of immunosuppressant drug use for his Crohn’s disease and was immunosuppressed (a large risk factor for PML). Thus, none of the cases of PML were in patients on Tysabri monotherapy (for either MS or Crohn’s disease) who had reasonably normal immune systems -- the vast majority of patients. All three cases of PML were found in patients on either concurrent Avonex or immunosuppressant drugs, essentially drug interactions in a very small percentage of similarly situated patients. The FDA and its independent advisory committee concluded that it was prudent to assess the risk of PML from the use of Tysabri at one in 1,000 (the risk of contracting PML, not the risk of death) but that it was imperative to evaluate that risk in the context of the benefit from this tremendously effective MS drug, the most effective MS drug currently available. Second, the FDA decided nearly a year ago (June 5, 2006) to reapprove Tysabri for relapsing forms of MS, but it required a detailed risk management program designed to mitigate the risk of PML by requiring that Tysabri be used as an MS monotherapy (i.e., alone and not in combination with the other major MS drugs) and only in patients with reasonably healthy immune systems. Thus, the FDA is not in the process of reconsidering the approval of Tysabri –- the agency made that decision nearly a year ago, and I am sure in light of what has transpired since that decision was made, the FDA is quite pleased to have made Tysabri available to appropriate patients. While there can be no guarantee that the risk of PML among MS patients in the required risk management plan will be less than one in 1,000, many believe that the major risk factors have been eliminated, likely to reduce the PML risk significantly. Third, just last week at the American Academy of Neurologists annual meeting in Boston, data derived from every patient on Tysabri were released, showing that there have been no additional cases of PML (or any other opportunistic infection, for that matter) in the approximately 10,000 patients on Tysabri, some for over a year (as the FDA permitted the resumption of the clinical trials well over a year ago). Thus, to date, there have been two deaths among approximately 13,000 patients using Tysabri (some in conjunction with other drugs), translating to just 15 out of 100,000, and zero deaths among the approximately 11,000 patients using Tysabri as monotherapy (which is now required), translating of course to zero in 100,000. While many neurologists will feel even more comfortable in assessing any risk of PML from Tysabri monotherapy after two years of widespread use, the fact that to date not one single case of PML has been encountered in patients on Tysabri monotherapy with reasonably healthy immune systems is very encouraging. Should this trend continue as the worldwide use of Tysabri surpasses 20,000 then 30,000 patients over the next year or so, it should become clear to all that the PML encountered was due to oversuppression of the immune system from unfortunate drug interactions in a very small percentage of those patients, something not anticipated but fairly easily avoided. Fourth, now that PML is known as a possible risk, efforts are under way to have protocols in place to monitor patients for PML, even though the risk management plan appears to significantly reduce any chance of getting PML, and to immediately suspend Tysabri therapy and remove Tysabri from the blood to improve the chance of recovery. These protocols will further reduce the risk of PML and death. Finally, as I am sure the authors acknowledge, the risk from any one act must be placed in context. If one stays in bed hoping to avoid any risk, there are other risks that arise (bedsores, depression, homelessness, etc.). The risk from poorly treated MS, or from untreated MS, is far greater than any risk from Tysabri. It is for this reason that FDA approved Tysabri (the risk/benefit assessment) and thousands of patients have made the fully informed decision to begin Tysabri treatment. Tysabri has already allowed thousands of MS patients to recover a more normal life, improving their balance, walking, vision, energy levels, and overall quality of life. I invite readers to visit our all-volunteer Web site, www.MSPatientsForChoice.org, started and run by MS patients for the MS community, to read firsthand reports about the benefits many have already obtained from Tysabri. Please be more careful about selecting examples based only on misleading headlines or incorrect memory. I ask that the journal run a correction to this otherwise interesting article.
Response To Tysabri Critique 9 May 2007
Joshua T. Cohen, Research Associate Professor of Medicine Tufts-New England Medical Center, Peter J. Neumann Send comment to journal: Re: Response To Tysabri Critique jcohen{at}tufts-nemc.org Joshua T. Cohen, et al.	David Kaplan's eLetter raises several important issues that must be emphasized to properly understand our article's message that risks and benefits must be viewed in context in order to ensure that health care and regulatory decisions are as rational as possible. First, Kaplan points out that the three PML cases associated with Tysabri were among patients who were also receiving other therapies that affected their immune status. The implication is that their risk for contracting PML was higher than the risk faced by patients receiving Tysabri alone. Our paper emphasizes that the risks were average values for the entire exposed population. Specifically, we noted (p. 643) that "mortality probabilities can differ among members of exposed populations.… Drug risks vary by age, health status, concomitant medications, and other factors." For a patient evaluating the trade-offs, this heterogeneity is clearly important. For policymakers, this information can play a key role in the evaluation of risk-management strategies. Nonetheless, even when variability is taken into account, the need to compare risks and benefits will remain. Second, as Kaplan points out, the risk assessment of medications can be a fast-moving target. Our paper describes the assumptions that we used to develop our risk estimates and notes that "the assumptions inevitably introduce uncertainty" (p. 644). The new data presented last week at the American Academy of Neurologists annual meeting were, of course, not available to us at the time we wrote our article. Regulatory decisions and individual decisions about appropriate treatments must be made in real time, however. The FDA and health care providers must endeavor to ensure that their decisions and recommendations are consistent with the best available science because the science can change over time, perhaps making it necessary to revise these decisions and recommendations. The best that can be hoped for at any point in time is to make decisions "with the available information on the table" (p. 645). Third, Kaplan notes that "the risk from any one act must be placed in context." That is the central message of our article. We state that the risk comparisons described in our article "underline a crucial point: Risks should not be evaluated without also considering benefits" (p. 645). In the case of Tysabri, we state that "the risk comparisons described here point to the need to carefully consider the benefit side of the equation. The perceived acceptability of any risk is clearly linked to the associated advantages of the activity. Consider the case of Tysabri for the treatment of MS. Tysabri does not (at least directly) reduce mortality risk for MS patients, but it does mitigate symptoms for many" (p. 644). We conclude by urging the FDA to emphasize systematic evaluation of benefit data in addition to its evaluation of risk data. Finally, we note that there is a factual issue that may have been incorrectly reported by the press as part of their coverage of our work. Our article does not state that Tysabri was withdrawn last year. It instead states -- as Kaplan notes -- that Biogen Idec and Elan suspended sales of Tysabri in 2005. We believe that a full reading of our article will reveal a balanced and accurate treatment of this issue.
Majority of the MS Community's Views on Tysabri 10 May 2007
Lauren M. Roberts, Retired MS Patient Send comment to journal: Re: Majority of the MS Community's Views on Tysabri LGLBGL2003{at}AOL.COM Lauren M. Roberts	As an MS patient for over 31 years, current Tysabri patient, and active member of the MS community that encourages patients to be their own advocates, please note that I am only one of many who are extremely tired of Tysabri being unfairly maligned by these largely avoidable drug interactions; Tysabri's RiskMap/TOUCH protocol mitigates the 0.1% risk almost completely. And we do understand that no drug is without risk, not even aspirin. Furthermore, we are aware of the very significant fact that Tysabri alone quite clearly did not present any extraordinary risk. It was the combination of Tysabri and other immune modifiers or immunosuppressant drugs that created the risk that caused Tysabri to be voluntarily withdrawn from the market for 15 months. Many MS patients become disabled to the point that they cannot work and/or experience a very poor quality of life due to increasing and accumulating disabilities. Our patient population also has a very high suicide rate due to no hope left for many to hold on to, despite family and professional support. Having Tysabri unfairly labeled by many in the media and even professionals as a risky drug discourages many uninformed patients and neurologists from fully considering Tysabri's use, which in turn deprives those patients and their families from its benefits that could accrue if the true and full facts were represented. A large majority of MS patients are quite well-versed in Tysabri, and they view media reports such as the one resulting from this article as biased and nothing more than "throwing the baby out with the bathwater." Not authoring a corrected version of your article for the many other media outlets to follow will only lead to uninformed MS patients suffering new and further (and quite possibly permanent) disabilities, many of which could have clearly been avoided with Tysabri's superior efficacy of 67%.
An Opportunity For Clarity 10 May 2007
Harold Katlin, Retired/Disabled MS Volunteer Send comment to journal: Re: An Opportunity For Clarity haroldnk{at}yahoo.com Harold Katlin	As both the letter from David Kaplan and the authors' response illustrate, there clearly is significant confusion when talking about Tysabri and the "risks vs. benefits" scenario. Why not take this opportunity to help 2.5 million MS patients worldwide better understand the fact-based risk of developing PML as a direct result from using Tysabri in monotherapy. For the sake of this discussion, I propose that you consider Tysabri patients to have an immune system that is considered to be reasonably healthy, and Tysabri would not be contraindicated because of other issues. I truly believe that such a "risk analysis" of monotherapy Tysabri use would not only help millions of MS patients better understand the true risk of developing PML from using Tysabri, it would also give you the opportunity to be the first person/organization to tackle this important question in this manner. I can not begin to emphasize adequately just how significant this issue is. Prior to the "PML scare," Tysabri was considered to be the most significant breakthrough in the treatment of MS since the introduction of Betaseron. With the PML issue now causing so much worry and concern amongst MS patients and their neurologists, such an impartial evaluation of the facts would be greatly appreciated by all. Thank you. Note: My first Tysabri infusion is scheduled for May 14, 2007.
Risks And Benefits Of Tysabri 10 May 2007
Joshua T. Cohen, Research Associate Professor of Medicine Tufts-New England Medical Center, Peter J. Neumann Send comment to journal: Re: Risks And Benefits Of Tysabri jcohen{at}tufts-nemc.org Joshua T. Cohen, et al.	We want to clarify that the intent of our article was not to portray Tysabri as unacceptably risky. Instead, our purpose was to point out that even when some drugs do pose notable risks, their substantial benefits must also be taken into consideration. We thought that Tysabri was a good example of a treatment that especially deserved this type of analysis. As it turns out, an article published two weeks ago [Dorsey ER et al., Neurology 2007; 68:1524-1528] has conducted exactly the type of analysis we have called for in our article. The recent article by Dorsey et al. suggests that the benefits of Tysabri far exceed its risks. Interestingly, this analysis assumes virtually the same mortality risk that we reported in our work. We encourage people to keep in mind the broader point of our article, which is that there should be a thoughtful discussion of risks and benefits of all interventions, including Tysabri.
Tysabri Reference In The Cohen/Neumann Paper 10 May 2007
Chris Chan, Associate Director, FP&A MS Volunteer Send comment to journal: Re: Tysabri Reference In The Cohen/Neumann Paper cchan7556{at}yahoo.com Chris Chan	I commend Cohen and Neumann on an informative article, and I believe the article’s intent is timely and warranted. Indeed, it is quite clear that the paper’s primary intent is to put the risk profiles of drugs in a proper perspective, and for the most part I believe that Cohen and Neumann succeed in doing so. However, it would have been better to have selected a different example than Tysabri to illustrate drug risk, as the basis for determining a reasonably precise risk for this new drug is still quite limited. Other drugs with a longer history and more widespread use would have provided a much clearer basis for determining theoretical risk levels. Given the very limited history of Tysabri, particularly at the time the article appears to have been drafted, the specified Tysabri risk of 65 fatalities per 100,000 person-years is problematical and misleading, and thus detracts from the paper’s overall effectiveness. Because David Kaplan (in his “Facts on Tysabri” eLetter) has already adequately specified the reasons why this risk figure is likely significantly overstated, I will not rehash the details. But I believe it is safe to generally say that a risk figure based on the unfortunate deaths of two patients who took Tysabri in combination with other immunosuppressants in an environment in which PML was completely unexpected does not accurately reflect today’s reality, i.e., one in which Tysabri is administered as monotherapy only and in which patients are required to enroll in a comprehensive risk management program designed to minimize the risk of PML and other opportunistic infections, monitoring to permit the possible detection of any early indication of PML while it might be treatable, and other safeguards. The authors responded to Kaplan’s point that the three Tysabri PML cases were all in patients on combination therapy by noting that “the (specified) risks were average values for the entire exposed population. . . . For a patient evaluating the trade-offs, this heterogeneity is clearly important.” With all due respect, that response completely misses Kaplan’s point, and I maintain that the authors' calculation of a risk (presumably going forward) of 65 fatalities per 100,000 person-years figure is misleading. One could make the case that the figure should really be “0” fatalities, since no Tysabri monotherapy patient has contracted PML through the most recent reporting date (April 23, 2007). Based on everything I have learned, I do not believe that there is NO risk of PML in Tysabri monotherapy, but I do believe that the risk is substantially lower than that for Tysabri administered with other immunomodulators or immunosuppressants. I note that one of the deceased patients was also taking Imuran and Remicade, each of which has been linked to PML in and of itself. It may simply be that the combination of Tysabri and these other drugs was too much for a small percentage of patients being treated with those combinations, and that given the mitigating factors of the risk management plan that is now in place, Tysabri will be as safe as the older-generation MS drugs while being about twice as effective (thus reducing MS risk) and more tolerable (no flu-like effect or injection site reactions suffered by many patients on most of the older generation drugs). Although the authors' Tysabri risk assertion is only a small portion of their paper, it unfortunately adds to the misinformation that has been and is still in circulation about Tysabri. One result of the myriad of inaccurate information is increased patient confusion and stress, a contributor to MS relapses. This results in such phenomena as the one the paper suggests in which “one recent survey reported that more than half of the MS respondents using Tysabri would probably continue to take the drug if it posed a fatality risk of 1 in 1,000,” and yet “(in a Reuters survey) most respondents said that they would stop use of the drug if there were an additional two or more PML deaths associated with it.” The confusion and disconnect are apparent, since the “1 in 1,000” estimate used by the FDA’s independent advisory committee in March 2006 (before the Agency’s reapproval of Tysabri in June 2006 and the approximately 10,000 patients currently on Tysabri with zero incidence of PML) implies that patients would accept five to ten PML deaths [if considering patient years] at this time. In conclusion, I commend Cohen and Neumann for their worthy paper, but I also caution them on using risk figures that are arguably misleading and hence detract from the paper’s main intent. In the case of Tysabri, such arguably misleading figures may also have a tangible effect on MS patients and physicians seeking to make the best treatment choice, as such figures may factor into their risk vs. reward analyses. Thank you, and I look forward to future enlightening articles from these authors.
Root-Cause Failure Analysis Of Drug-Induced PML Morbidity 14 May 2007
Mike Barron, Retired/disabled nuclear facilities operator Founder and Volunteer co-contributor of msPatientsforChoice.org Send comment to journal: Re: Root-Cause Failure Analysis Of Drug-Induced PML Morbidity mike{at}mspatientsforchoice.org Mike Barron	Thank you for allowing your viewing audience to present comments to the Health Affairs editor. I just received my 9th TOUCH infusion of natalizumab on May 7th and admit that I am no longer afraid of active MS. I had forgotten how I had been held hostage to this horrible disease for over 12 years since diagnosis. I am so fortunate and very grateful to live in a time where a novel drug discovery came to fruition in a desperate time of need for me and so many others. I am willing to accept any small unknown risk of continuing this wonderful therapy to sustain my stable remission from active MS in exchange for the numerous benefits I have already been receiving since my first two of 11 total infusions I have received thus far. I only have one comment regarding FDA-approved drug safety and the natalizumab-associated PML occurrences. An extensive root-cause failure analysis of the three incidents of PML in the natalizumab clinical trials reveals another possible PML causative agent or culprit other than natalizumab and the various immune-modulating therapies prescribed to the patients. There is an interesting correlation and commonality among the three cases of natalizumab PML reported in the New England Journal of Medicine that has been disregarded or overlooked as a subject of debate among the experts. All three of the natalizumab PML patients were also given high-dose methylprednisolone concomitantly with natalizumab and other disease-modifying agents preceding the eventual differential diagnosis of actual PML. In most PML cases, might it be considered harmful or imprudent medical practice to intravenously infuse a patient with anti- inflammatory high-dose glucosteroids transiently rendering the patient autoimmune incompetent or defenseless against a DNA viral infection of the brain for several days? If in fact that is true, it doesn’t surprise me that recent medical study has revealed the mortality rate from PML in technologically advanced societies is two to three times higher than in underdeveloped and third-world countries. With the expected occurrence of PML increasing at an alarming rate because of the ubiquitous JCV and the AIDS/HIV global pandemic, might it be a thought-leading idea here in the United States for the Neuropharmacology safety and drug product divisions of the FDA to investigate and then mandate a revised black-box warning and label contraindication for progressive multi-focal leukoencephalopathy on all formulations of high-dose glucosteroids or corticosteroids such as methylprednisolone? High-dose and pulse glucosteroids are often prescribed for inflammatory-type activity clinically observed on MRI, but that could also be differentially caused by an encephalopathy such as activated JCV or a latent PML infection. Better drug labeling on immune-suppressing steroidal medications in effect would provide an additional safety barrier for differentially diagnosed neurological disease patients presenting to any clinician with unisolated PML, drug-induced PML symptoms, or possibly undisclosed AIDS/HIV/PML co-morbidity while avoiding further medical harm to those unfortunate patients.
Valuable Statistics 14 May 2007
Andrew H. White, Vice president Mortgage lender, Massachusetts Send comment to journal: Re: Valuable Statistics andy.stepstone{at}comcast.net Andrew H. White	While people may confuse or misunderstand perceived risk, isn't the important lesson here the differential benefit obtained by taking an aspirin versus not? It's nice to know that taking one daily offers a male similar risk to driving a car, but I want to know the benefit of taking an aspirin or not taking an aspirin. Are you saying that risk is 10.4 per 100,000 by adding the aspirin to a daily regimen, and 1 per 100,000 for those who don't, thus indicating that taking an aspirin actually elevates risk of death?