Health Committee on March 11th, 2008

[The Chair, Conservative, Joy Smith]

Welcome, everybody. I would like to bring the meeting to order.

Welcome to our guests who are here today. We have with us the Cancer Advocacy Coalition of Canada, the Expert Advisory Committee on the Vigilance of Health Products, the Institute for Safe Medication Practices Canada, and representation from the University of Montreal.

Pursuant to Standing Order 108(2), I'd like to welcome you all to the seventh meeting of the post-market surveillance of pharmaceutical products, prescriptions, and non-prescriptions.

I'm going to ask the witnesses to give a 10-minute presentation. After all the presentations are finished, we'll get the questions from the committee.

We're very grateful that you took the time to join us this morning, and we're eagerly listening to all the insightful information that each one of you will tell us. Could we please begin with the Cancer Advocacy Coalition of Canada?

[Dr. James Gowing, Chair of the Board, Cancer Advocacy Coalition of Canada]

Thank you for recognizing the importance of post-market surveillance and for providing the opportunity to many groups like ours to present suggestions for improvements.

We are very gratified that you've seen fit to act on one of our previous recommendations, made when we presented our views to you during your hearings on the common drug review last year. Please accept our congratulations on your December 2007 report regarding CDR. In our view, your recommendations, if implemented, will substantially improve the cancer drug approval process in Canada.

The Cancer Advocacy Coalition of Canada is a full-time registered non-profit group, comprised of physicians, patients, and business executives from across the country, who are all unpaid volunteers. We publish the annual Report Card on Cancer in Canada, the only independent evaluation of cancer system performance in the country. In the just-released 2007 report card—which you should have in a day or two, as it's in the mail—we examined the allocation of research funds, the needs of young adults with cancer, the clinical trial research process, health human resources, access to diagnostics and drugs, and the role of nurses in supportive care. We publish these reports to identify barriers to better cancer control and to offer constructive solutions that are also realistic and implementable.

For the past three years, the articles about access to cancer drugs have recommended phase four, or post-approval, trials to confirm treatment results in the cancer population at large; and for two years in a row we have recommended increased translational research to identify the subsets of patients that benefit from these new drugs. Post-market surveillance is an appropriate portal to initiate these proposals, and it could encompass a more comprehensive system of information that is analyzed, shared, and disseminated to benefit patients.

When a new cancer drug is approved, the only obligation at present is the reporting of unexpected adverse events—and even this is not done comprehensively. From the physician's perspective, lack of effectiveness is an adverse event, when the treatment has added toxicities and other health risks to the patients, with no good result.

The information available to physicians about a new cancer drug is often not very helpful in the real world. Patients in preapproval clinical trials represent, at best, 3% of the wider population; and when patients at late stages of disease are treated in these trials, they generally achieve a response rate of 20% or less. Evidence suggests that response rates in the real world can drop to 10%. But for responders, the outcome could be a cure.

There are notable exceptions, such as Herceptin, where a tumour marker has been identified to improve patient selection. In such cases, the response rate is impressive and can include cure. However, most drugs do not have biomarkers, and for the majority of patients who do not respond, we are subjecting them to a drug with potential severe adverse effects with no hope of benefit. The objective, then, is to use every means at our disposal to identify the patients who will benefit and to spare those who will not benefit from the wasted time and avoidable adverse effects.

If post-market surveillance were formalized to capture information on positive results, as well as adverse events, then the characteristics of responders and non-responders could be captured for analysis. The immediate benefits would be remarkable. Better patient selection for the use of these expensive new drugs would save a huge amount of money that is otherwise fruitlessly spent. To accomplish that goal, post-market surveillance must collect additional reporting, and much of this must come from physicians, who can offer the important clinical detail necessary for further analysis. Reports ought to be submitted on every patient treated with a newly approved cancer drug within 12 weeks of initiating the treatment to ensure the accuracy of the information.

Physicians and others will not cooperate with an onerous paperwork exercise that produces nothing useful for their knowledge or their patients' well-being. The reporting we envisage could be done on a single computer screen; indeed, the proposal is undermined by anything more time-consuming. The information system required has to be linked to appropriate research by qualified entities, such as the National Cancer Institute of Canada, and have some assurance of timely outputs that are readily available online. Then the effort is worthwhile.

The data gathered should be readily available online and be presented in a manner that offers greater meaning than mere compilations of reported events. Indeed, if this undertaking is to serve its intended purpose, physicians and health administrators would find many other valuable uses for the data, including guideline writing and biomarker research.

In the final analysis, as this new knowledge becomes available, the true potential of each cancer drug would be realized, patient access would be increased, and treatment results improved.

A three-month increase in average survival translates into years for the 15% or so who respond to a drug. Three patients are alive well beyond three years after taking one of these new drugs and have reported their stories in our report card this year. Their written reports are very compelling stories. I would encourage you, when you get this report in a few days, to look at their proposals.

The Cancer Advocacy Coalition encourages you to focus on what is possible and needed, rather than what is already done. A robust post-market surveillance system could be an integral system providing valuable information and education, benefiting all Canadians and saving money.

I want thank you for your time and interest.

[The Chair, Conservative, Joy Smith]

Thank you so much for your presentation, Mr. Gowing.

We're now going to hear from Ms. Diane Brideau-Laughlin, chair of the Expert Advisory Committee on the Vigilance of Health Products.

[Diane Brideau-Laughlin, Chair, Expert Advisory Committee on the Vigilance of Health Products]

Bonjour, madame. Thank you very much. My presentation will not be very long.

Thank you for the opportunity to participate in the proceedings of this committee.

My name is Diane Brideau-Laughlin and I am the Chair of the Expert Advisory Committee on the Vigilance of Health Products. I would like to take a few moments to discuss the mandate of the Committee, as well as its current composition.

The Expert Advisory Committee on the Vigilance of Health Products constitutes an integral part of the health product and food branch's post-market surveillance strategy. Its mandate is to provide the branch with ongoing external expert advice on broad strategic policy in regard to the safety and therapeutic effectiveness of marketed health products for human use. It also provides a mechanism to involve the public, providing them with a forum to have their views heard by experts, who can discuss their input and incorporate the discussions into the recommendations provided. In other words, it consists of real people who work in the real world and who have the opportunity to speak to real people about real issues regarding their health.

The committee was established in November 2007 and has met twice so far. The inaugural meeting was an orientation session on the branch's various post-market strategies. The second meeting, held recently in February, in Longueuil, Quebec, discussed issues relating to adverse reaction reporting primarily.

The current membership is composed of eight women and nine men from across the country. We have representation from the Atlantic region, Quebec, Ontario, and the western provinces. We also have an individual with aboriginal expertise on reserve and in rural settings. The membership includes health professionals, patient and consumer advocates, and researchers in academia as well as in industry. It has individuals with interests in ethics, epidemiology, biological sciences, human medicine, health product vigilance, public health, social sciences, risk assessment, as well as communication and risk management.

It's a privilege to be the chair of this committee, and thank you.

[The Chair, Conservative, Joy Smith]

Thank you so much for your presentation, Ms. Laughlin.

Our next witness is Ms. Sylvia Hyland, vice-president of the Institute for Safe Medication Practices Canada.

Please, go ahead.

[Sylvia Hyland, Vice-President, Institute for Safe Medication Practices Canada]

Thank you.

Carmen, do you have the handouts?

[The Chair, Conservative, Joy Smith]

Everyone has a copy of them. Thank you.

[Sylvia Hyland]

Thank you, Madam Chair.

On behalf of the Institute for Safe Medication Practices Canada, I would like to thank you for inviting us to be part of your study on post-market surveillance. ISMP Canada is an independent national not-for-profit organization established for the analysis of medication error or incident reports. Our purpose is to identify underlying contributing factors or causes and to make system-based recommendations for enhanced patient safety.

We are working together with the Canadian Institute for Health Information and Health Canada to develop the Canadian medication incident reporting and prevention system. We also work closely with the Canadian Patient Safety Institute and other organizations and associations working to advance medication safety and patient safety at a provincial, national, and international level.

Our area of expertise is analysis of the preventable subset of adverse drug events. We know from a number of studies, including the Canadian adverse events study and two studies in the U.K., that a large proportion of adverse drug events are in fact preventable.

Through the shared learning from reports of medication incidents that we have received, we've made approximately 195 recommendations for consideration by hospitals as safeguards in their medication use systems. Of those, approximately 50 have now been adopted by CCHSA, the Canadian Council on Health Services Accreditation, as required practices.

[The Chair, Conservative, Joy Smith]

You have a very insightful presentation, but you're going a little bit too fast for the translators.

[Sylvia Hyland]

Oh, I'm so sorry.

[The Chair, Conservative, Joy Smith]

Could you slow down just a little bit? Thank you.

[Sylvia Hyland]

Thank you. I will.

Having 50 of our recommendations adopted by the accrediting body influences 3,700 health care sites and exemplifies how learning from reports and their analysis can lead to knowledge translation and proactive system enhancements.

I would like to take this opportunity to provide you with some examples relevant to Health Canada's regulatory mandate.

The first photograph I have for you is a picture of a transdermal patch introduced into the market. Because the patch was almost invisible, practitioners in emergency departments reported not knowing that patients were receiving a highly potent narcotic; the patches were being missed. The manufacturer responded very quickly and added both the name and colour to the patch, as seen on the next page. Medication patches are relatively new, and these incidents inform the need to include regulatory guidelines for labelling of patches.

On the next page you'll see two bags. The bag on the right, which is intended for pharmacy use only, was inadvertently infused intravenously instead of the IV solution on the left. We received three reports, one of an incident leading to serious harm, and we knew that in the U.S. there had been a fatality reported due to a similar error. We worked with the manufacturer, and they changed their label, as seen on the next page. This is a significant improvement.

There is still opportunity to improve the packaging of the bag. Until we make it impossible to connect a product not for IV use to an IV line, such dangers continue to exist.

On the next page you'll see two ampoules that were mixed up and reported as a near miss. The risk for harm is high. We have worked with individual manufacturers to improve such labelling, and we've provided an example of how the labelling can be improved on the next page, where you see the upper ampoule with print on glass and the lower ampoule.

We think printing critical information directly on glass should not be permitted. Ultimately this learning needs to be incorporated into regulations or guidelines so that the knowledge is translated into current and future manufacturing practices.

The next example is interesting because the label meets regulatory requirements for display of concentration. However, the product is not prescribed in millimoles; it is ordered by the physician in grams, and the conversion is not an easy calculation. We acted after only one report. The manufacturer quickly changed its label and thanked us for not just identifying the problem, but for providing recommendations for improvement. Note that this company took the extraordinary step of removing their logo from the label in order to give prominence to critical information.

In the last example we show five neuromuscular blocking agents. Neuromuscular blocking agents have been identified as high-alert drugs, which means that the drugs, when involved in an error, have a high incidence of causing harm. Three of the manufacturers voluntarily chose to follow our advice to place a warning on the top of the vial and two did not. This is a good example of a situation in which we could evaluate the recommendation together and, if it is reasonable, make it a requirement, again so that the learning is not lost on future products.

The key message is that underlying practice errors are opportunities for system enhancements to help prevent recurrence. Unless underlying risks are mitigated, it could be predicted that the errors will repeat again and again.

A key component of reporting programs is the follow-up and analysis and being responsive. For medication incidents, this includes identifying whether there are preventive strategies. It is also important to demonstrate to reporters that they have made a difference through their efforts to report, and this will in turn increase reporting.

ISMP Canada is aware of Health Canada's plans and initiatives to strengthen post-marketing surveillance, such as additional regulatory authorities post-market with labelling, and we would like to offer our assistance where possible.

Thank you.

[The Chair, Conservative, Joy Smith]

Thank you so much.

Our final witness is Dr. Yola Moride, associate professor in the Faculty of Pharmacy, University of Montreal.

Welcome, Doctor.

[Dr. Yola Moride, Associate Professor, Faculty of Pharmacy, Université de Montréal]

Thank you very much for providing me with this opportunity to present my views. I would also like to mention that I'm the current president of the International Society for Pharmacoepidemiology, and for several years I have been an expert consultant for regulatory authorities such as Health Canada as well as the European Medicines Agency, EMEA, as well as for the pharmaceutical industry. It's important also that I say that my presentation presents my views and not those of the organizations I'm affiliated with.

At present in the post-marketing setting, safety surveillance is mainly conducted through spontaneous reporting. While it is recognized that spontaneous reporting is the best method to detect a previously unknown safety issue at the level of the entire population, it is not suitable to quantify a risk. Spontaneous reporting is highly efficient to generate a signal, but not to evaluate the risk. Therefore, at the present time drug safety data originate mainly from randomized controlled clinical trials at the time of approval and spontaneous reporting thereafter as the main safety net to monitor drug harms.

What we have observed under this model is that over the past 30 years in Canada there have been 121 drug withdrawals due to safety issues. This represents extreme, if not catastrophic, regulatory decisions that should be avoided because of the major negative impacts at many levels, such as drug companies, regulatory agencies, and eventually loss of confidence by the public.

In other jurisdictions, such as the U.S. and the E.U., it has been recognized that the current model is insufficient to properly monitor the benefits and harms of medicines. Hence, there has been a major shift in paradigm for drug safety surveillance. Instead of relying entirely on randomized clinical trial data to assess the benefit-risk of a drug and afterwards on the spontaneous reporting system, authorities have introduced in their regulations risk management in all phases of drug development.

Since 2005, a pharmaceutical product is authorized on the basis that in the specified indication at the time of authorization, the benefit-risk is judged positive for the target population. However, it is recognized that not all actual or potential risks will have been identified when an initial authorization is sought. In addition, there may be subsets of patients for whom the risk is greater than that for the target population as a whole, or there may be subsets of patients for whom we are willing to accept greater risks because the condition for which they are treated is serious and they do not respond to any other available therapy.

Furthermore, there may be potential risks that need to be addressed, and to conduct additional randomized trials prior to the submissions will probably not bring the answers that are needed because some of those adverse events are extremely rare. Instead, an active surveillance system from the time of marketing would allow us to properly monitor these potential risks.

Finally, the benefits of a drug must also be monitored in the post-marketing setting, because even though a drug has been judged efficacious according to clinical trial results, the benefits may be much lower in the real-life setting. This is the case, for example, with anti-depressants, whereby more than 50% of patients discontinue their treatment before the minimum recommended duration of six months. Hence, patients are exposed to the risks that occur early in the treatment and to very little benefit since the drug must be taken for an extended period of time. Again, this would not have been reported in clinical trials.

Tools are available, such as observational epidemiological studies, to bridge this information gap and ensure that the benefit risk of a new drug is maintained. Should problems be identified in the usage of these drugs, or at-risk patients be identified, then interventions can be rapidly implemented to maintain the benefit-risk within the acceptable range. Such interventions are referred to as risk minimization action plans.

So the current risk management model involves various phases that can be summarized into detection, evaluation, minimization, and communication.

In the EU, all new drug applications must be accompanied by risk management plans. In the U.S., although not mandatory, such plans are expected by the FDA. In fact, drug approval may be delayed if the plan is judged not satisfactory.

Pharmaceutical companies are now realizing the huge economic consequences of not taking full responsibility for properly managing the risk. We are moving away from a reactive process to a much more proactive approach involving a broader evidence base and a widening of expertise, resources, and methodologies. Studies are conducted in the context of conditional approvals, and denial of marketing occurs if commitments are not met.

For many years Canada has been a leader in the area of pharmaco-epidemiological research, with one of the highest concentrations of experts in the world and access to invaluable resources, such as prescription and medical services databases that are available through our public health care system. Yet studies conducted in Canada are mainly investigator-driven. In the absence of legislation, those studies are not being implemented right at the time of marketing to ensure that real-life data on drug safety is generated as soon as possible and fed back to the regulators, who can then reassess the benefits and risks of a drug. In the absence of such legislation, regulatory authorities such Health Canada have very little leverage to request these studies.

Finally, an important element of risk management is risk communication. In addition to the package inserts, the evidence being generated must be fed back to the health care professionals and patients as soon as possible. The process by which the risk is being managed must be transparent and no longer a top-down approach.

In conclusion, the current model, such as the one used in Canada, is deemed insufficient to appropriately optimize the benefits and risks of medicines in the post-marketing setting. Risk management is the new paradigm that will use complementary sources of data and methods. Although the methods and expertise have been available in Canada for many years, they have not yet been packaged to be part of the drug regulatory process.

There are many resources in Canada that remain underutilized, such as claims databases, that could be useful in improving the analysis of spontaneous reporting of data and in evaluating the risk in the post-marketing setting.

Thank you.

[The Chair, Conservative, Joy Smith]

Thank you very much.

Obviously this panel of witnesses today is extremely good, and we have appreciated the comments you've made.

We'll now go to questioning, starting with Monsieur Thibault.

[Robert Thibault, Liberal, West Nova, NS]

Merci, madame la présidente.

Thank you all for your presentations. They were very informative.

There seem to be recurring themes in the presentations on this subject. As I think Dr. Moride pointed out, it's not necessarily that we have a lack of information in the country; it's how we use it and how it's interlinked. We also have the question put forward on whether practitioners would willingly provide information on adverse effects—or even the larger question of post-market surveillance, I presume—if there were feedback, if valuable information came back to them that helped them in their practices. It's a shame that we haven't got to that point yet.

We are wrestling with whether there should be legislation. Should mandatory reporting be necessary? Would that assist us? It's a difficult question. I think it's made a little more complicated by Dr. Gowing, because his definition of adverse effects is a little different from what we've been dealing with so far. It's beyond somebody being rushed to the hospital; it includes somebody not receiving the effect you would expect from a therapy.

I want to open the question to all of you. We've heard of two possible elements here of change, aside from the technological aspect. One is progressive licensing, where medication would come on the market at different levels and be progressively licensed as information was available. The second one that we heard a lot about is off-label use. We've heard of the necessity for off-label use because of the way we license our drugs and the therapies they eventually become useful for.

It appears to me that both of those instances would be a great way to get a lot of the data you were talking about, Dr. Moride, because the uses wouldn't necessarily be the ones they were licensed for. So you could expect the community to feed back more at that time if the health treatment community was getting useful information from it.

I open it to all of you to comment.

[Dr. Yola Moride]

The issue of off-label use has been recognized as extremely important. In the current setting, spontaneous reporting would address the issue of off-label use, because it's all adverse events that should be reported, regardless of the indication.

[Robert Thibault, Liberal, West Nova, NS]

But if we're hearing that only 10% of adverse events are reported and about the wide use of off-label therapies, then we can presume that it's not efficient.

[Dr. Yola Moride]

Do you mean spontaneous reporting? It's extremely efficient because it does what it intends to do, which is generate a signal. You don't want the system to be swamped with the common and not serious events.

This component, I find, is working efficiently. It covers the entire population, but it doesn't address the other issues related to drug safety. We need data on the risks and on the at-risk populations.

The issue of off-label use actually is included in risk management planning. There need to be provisions and specific actions planned right from phase two with regard to what will be done regarding the potential for off-label use.

[Dr. James Gowing]

Off-label use is a very good thing, I think. In the case of one of the drugs we use to control pain, gabapentin, 90% is used off label. The label is for epilepsy, but in terms of controlling cancer pain, patients receiving gabapentin often need only about 10% of the amount of morphine and other drugs because they use it.

We should be collecting that information, though. This is why I think, for our proposal on new cancer drugs coming out, whether they're used on label or off label, I wouldn't want to interfere with that. But we should be collecting that data, and we're not. So we're not learning anything from it. Some of the uses of off label are actually a whole lot better than what they're labelled for.

[Robert Thibault, Liberal, West Nova, NS]

That's the point I want to make. We've heard examples of drugs licensed for adults being very useful for children, but they are not licensed. Almost all of that therapeutic use is off label because of the clinical trial problems and all those things.

I'm just wondering if the exchange of information across the medical practice community on these off-label uses is very good. Are people learning of the potential of these drugs?

We know that the pharmaceutical manufacturers will promote the label use, but it's more difficult to promote the off-label use.

[Dr. James Gowing]

Well, they're not allowed to promote the off-label use. Certainly in the oncology community there aren't that many of us across the country, and we soon learn what works and what doesn't. We should be collecting that knowledge and disseminating it more widely, and we're not.

[Robert Thibault, Liberal, West Nova, NS]

We heard.... Sorry.

[Dr. William Hryniuk, Past Chair, Cancer Advocacy Coalition of Canada]

The additional consideration is that there's some reticence to publish the use of drugs off label, so in fact the communication between professionals about off-label use and effectiveness is inhibited. But if this mechanism were introduced, that would automatically overcome that impediment.

[Robert Thibault, Liberal, West Nova, NS]

That would provide that information.

We had one presenter here last week who suggested that a lot of the drugs that were on the market now had low efficacy; they are of value to a small percentage of the population. There is 15% efficacy in some cases, and you'd be led to think that for 85% of the people taking the drug it has no value or it only increases the positive results by 15%.

Could you explain this matter in terms of your oncology practice? How do you see this?