Should biosimilar drugs be artificially promoted in Canada?

The comments section is closed.

  • Joel Lexchin says:

    The article by Yanick Labrie serves as an example of how omissions can distort the message in an article as much as commissions. By selective use of various studies, Mr. Labrie creates the impression that there is a substantial body of evidence arguing against substituting biosimilars for biologics and that such substitutions can lead to increased health care costs and possible harms to patients. He uses the studies that he cites to support his view that governments should not force patients to switch from biologics to biosimilars.

    He begins by discussing a review completed by Quebec’s Institut national d’excellence en santé et services sociaux (INESSS) on the topic of switching and quotes the study as saying that for certain populations or certain therapeutic fields “very little or no data are available on the safety of switching a biological drug.” What Mr. Labrie neglects to say is that the study also concluded that “These systematic reviews [analyzed by INESSS] did not reveal a statistically significant difference, for all study end points, between patients whose treatment was switched and those who remained on the reference biologic.”

    He then talks about two papers that analyzed a biosimilar switching program in Denmark. He says that the first one showed “that healthcare use (+8 per cent) and total healthcare costs (+7 per cent) have increased for patients with chronic inflammatory diseases (treated with etanercept)”. This information did appear in the paper, but the Discussion section begins with the statement that “we detected no negative impact of the switch on the use and costs of a range of healthcare services during the first year. The fluctuations in the monthly use of services were similar during 1 year before and after the switch, and although there was a minor increase in the use and costs of outpatient services, there were on the other hand decreases in the use of inpatient services and (other than biological) medication costs.”

    His claim about the second paper was that it showed that “45 per cent of the patients followed had increased healthcare use after the switch of infliximab with the biosimilar. On average, the number of service days per patient increased by 7.4 per cent after the mandatory drug switch.” Again, this information was in the paper, but again this is a distortion of what the paper actually concluded. The conclusion was that “In this study of 769 patients with inflammatory arthritis who switched from INX to CT-P13 in routine care, we found only minor, clinically insignificant changes in the use of outpatient health resources during a 6-month time period after the switch compared with before the switch.”

    The Italian study that Mr. Labrie refers to that looked at switching in patients with chronic renal failure requiring hemodialysis did find that patients who used a biosimilar erythropoiesis-stimulating agent required a 40% dose increase to maintain their hemoglobin level compared to those who used the biologic. However, this was a retrospective matched control study, whereas four randomized controlled trials looking at the same two drugs found no difference in efficacy between the two.

    In order to make rational therapeutic decisions we need all of the evidence, not a selected portion that suits the conclusion the author wants to reach.

    • Yanick Labrie says:

      Contrary to Dr. Lexchin’s claims, the study findings reported in my article are consistent with those reported in recent systematic literature reviews. In one of them, researchers at the University of Missouri-Kansas City found that non-medical switching of biological drugs is likely to lead to increased use of medical resources and higher total expenditures in the rest of the health care system (See Yifei Liu et al. 2019 in Advances in Therapy). They surveyed the work of 17 studies that reported an overall increase in costs associated with the non-medical switching of biosimilar drugs. The researchers cite higher rates of surgery (11%), increased use of other medications (13%), and increasing doses of biosimilars (from 6% to 35.4%) as reasons for the increased costs, among others.

      This is also in line with the findings of another group of researchers who recently sought to assess the short-term costs associated with non-medical switching of reference biologics with biosimilars in stable patients with rheumatological, gastroenterological and dermatological autoimmune diseases. Patients who discontinued the initial biologic therapy had higher treatment costs in the first year of follow-up than those who maintained it. For payers, the total estimated costs of substitution amounted to US$771,460 (for 3,609 patients with a non-medical switching rate of 16.6%), mainly due to additional medical visits and laboratory tests that had to be performed. According to the authors, the non-medical switching rate was the main factor responsible for the cost increase (See Allan Gibofsky et al. 2019 in Clin Exp Rheumatol).

      In the Italian study I referred to in my article, I could have added that drug resistance (hyperreactivity to epoetin) was significantly higher in patients who switched to biosimilars than in the control group receiving the reference biologic (41.1% vs. 18.4%). As a result, a higher percentage of patients were unable to achieve the desired hemoglobin concentration despite the higher than usual doses of medication. A consequence of the dosing penalty is more frequent exposure of patients in the group that substituted the biosimilar to higher doses than are clinically recommended. This has the effect of increasing the risk of drug-related adverse events.

      I don’t contend the fact that randomized controlled trials are of great value. But as Dr. Lexchin is well aware, the relevance of using observational studies in real-life settings to complement clinical trials in controlled settings is increasingly recognized within the scientific community. Clinical trials, with their controlled environments, are not able to provide sufficient information to support non-medical switching from biologic originators to their biosimilars. Too often in clinical trials, the number of patients is small, or the duration of follow-up too short. This is important because it is known that adverse events can occur as late as three years after the start of treatment. Furthermore, these trials assess how a patient will respond to substitution once, whereas in everyday life patients are sometimes forced to try more than one drug before finding the one that best suits their needs and condition (see Feagan et al. 2020 in Advances in Therapy).

      Thus, although biosimilar drugs in controlled clinical trials show comparable results with brand-name biologics in terms of efficacy and safety, in real-life situations there may be significant divergences translating to higher adverse events and drug discontinuation rates in patient populations. In 2018, researchers Syed Numan and Freddy Faccin (in Advances in Therapy) conducted a review of studies on patients with chronic inflammatory diseases who substituted a biologic brand-name drug (anti-TNF) for a biosimilar one. A total of 74 real-life studies identified by the authors showed a level of discontinuation of biosimilar treatment very different from that found in controlled clinical trials. For the studies of infliximab biosimilars, these rates ranged from 0% to 87%. For biosimilars of etanercept, discontinuation rates ranged from 8% to 17%.

      Similarly, Bakalos and Zintzara (2019 in Clin Ther) considered 14 observational studies of real-life patients between 2014 and 2018. In half of the selected studies, the discontinuation rate in patients who switched to biosimilars turned out higher than the expected rate in patients receiving therapy with the original biologic drug. Of the 14 studies, 13 reported undesirable effects and lack of efficacy as the main reasons for discontinuing the biosimilar drug. The percentage of discontinuation for these same reasons varies between 2.8% and 28.1% for the totality of the populations concerned.

      Researchers Fleischmann et al (2020 in Rheumatology and Therapy) have highlighted once again the “nocebo effect” resulting from mandatory non-medical switching, which is believed to intensify patients’ negative expectations of biosimilar drugs and contribute to higher discontinuation rates. More observational studies will need to be conducted in the future in order to confirm, or refute, these findings.

      • Joel Lexchin says:

        Mr. Labrie does not deal with the point that he selectively quoted from the INESSS article and the two articles about the Danish system.

      • Yanick Labrie says:

        Dr. Lexchin criticizes me for having “selectively quoted from the INESSS document.” However, he misinterprets my argument as if I had claimed that any biosimilar substitution is to be prohibited, which is incorrect. The argument raised in my article is a warning only against mandatory non-medical switching for purely economic reasons (as implemented in BC and soon in Alberta). I should have made this point clearer.

        The scientific literature review conducted by INESSS led to a 100-page long document (in French). If I had enough space, here’s the whole quote I would have included in my article (taken from the English version of the executive summary):

        “The systematic literature reviews carried out in this report to assess the safety of switching between biologics did not reveal a statistically significant difference, in terms of loss of therapeutic efficacy, immunogenicity, retention rate, and adverse events between patients whose treatment was switched and those who remained on the reference biologic. The levels of evidence associated with these outcomes were considered moderate overall for inflammatory arthritis, diabetes, anemia, and plaque psoriasis. However, the levels of evidence were considered low for loss of therapeutic efficacy in inflammatory bowel disease and insufficient in oncology and for multiple switches. […]

        Non-medical switching in patients already receiving treatment with a reference biologic is generally not accepted by learned societies and the consulted clinicians. The latter are concerned about, among other things, the destabilization of patients who are complex cases and for whom few treatment options are available. They are also concerned about the absence of mechanisms for preventing such patients from being excluded from a non-medical switch, the fact that it is not possible to switch back in the event of a loss of efficacy or significant adverse events following the switch, and the risks associated with multiple switches. […]

        This report identifies certain populations or biologics for which very little or no data are available regarding the safety of biologics switch, and the significant concerns that clinicians have about non-medical switching. Most of the available scientific data have methodological limitations and raise significant uncertainties.”

        Now, let’s address the points raised by Dr. Lexchin with regard to the Danish studies. With regard to the first one, he criticizes me for only referring to the costs of outpatient services (which have increased) and for omitting to consider other health expenses (inpatient services, medication) which have decreased in return. However, as the authors emphasize in their study, “the outpatient costs represented nearly 70% of total costs whereas inpatient costs were 17%, primary sector costs 10% and medication 6%.” Thus, the decreased expenditures in other healthcare categories were not high enough to compensate for the increase in outpatient costs (which explains the slight rise in overall healthcare use and cost post-switch).

        Concerning the second one, he criticizes me for failing to note the authors’ assertion that “only minor, clinically insignificant changes in the use of outpatient health resources [were found] during a 6-month time period after the switch compared with before the switch.” However, the authors did not assess the “clinical” impact of the mandatory non-medical switch but only the “economic” impact, ie whether the use and healthcare costs increased post-switch. The choice of the word “clinical” in this context seems inappropriate to me and should not have appeared in the study, since none of the data analyzes carried out allow any conclusion to be drawn in this regard.

      • Yanick Labrie says:

        Following up on my last comment to Dr. Lexchin, here is a final remark with regard to the Danish study I referred to:

        The conclusion of the study, which no one can dispute, is that the authors found a statistically significant increase (+7%) in the total number of days of services during a 6-month time period after the switch compared with before the switch.

        We don’t know (based on the study results) if and to what extent these changes entailed by the mandatory switch program had a minor or major impact (or no impact at all) on patients from a clinical standpoint.


  • James Dickinson says:

    Mr Labrie’s argument sounds like special pleading on behalf of those who are making obscene profits, as we have come to expect from the Fraser Institute. He needs to recognise that the drug market is not a classic “free market” where prices drop when there is free competition. Too often manipulation of prices leads to inappropriately high prices. Continuing to pay high prices for the originator drug becomes a tax on all of us, and we cannot be satisfied with static drug expenditures: they need to decrease. Too many specialists prescribe the drugs for which they have received “gifts” such as support for conferences, “research” and even clinic staff. While there will be a few patients who do better on the original rather than the biosimilar, most will do just as well and some will do better. This argument is specious nonsense – can we ask who paid the Fraser Institute for it?

    • Yanick Labrie says:

      Mr. Dickinson, neither the CHPI nor the FI have anything to do with the content of this article.

      Yanick Labrie

    • Thank you for your comment James. We do not post sponsored or paid content and take any accusation of such very seriously. This piece was submitted independently by the author as an opinion piece and went through our editorial process. If you have a competing opinion, we would gladly consider publishing it as well.

  • Chris Bonnett says:

    Much of the evidence cited in Mr. Labrie’s article provides specific cautions rather than sweeping, serious and universal concerns based on strong evidence. This is important and helpful to know, but not conclusive.

    The INESSS review summary essentially demonstrates an absence of good quality evidence, but minimal evidence of harm. Of eight conditions and five therapeutic concerns, only three situations of 40 have high quality evidence (p. 3). Sometimes problems are not centred on indiscriminate non-medical switching, but “The loss of a biologic’s efficacy over time…caused by the natural progression of a disease or by a more rapid elimination of the drug” (p.2).

    INESSS also provides perspectives on how non-medical switching could be implemented to minimize risk, including avoiding high risk situations, implementing an appeal procedure and establishing a registry for pharmacovigilance to study real-world patient experience. Health ministries with support from private insurers could invest here in consultation with patients, physicians and manufacturers.

    Regardless of spending trends, specialty drug spending is not trivial, at about 30% of drug budgets and accelerating. We’ll agree there is a need for more and better quality research to guide policy, patient education and physician prescribing.


Yanick Labrie


Yanick Labrie is a health economist and public policy consultant. He is also an affiliated scholar of the Canadian Health policy Institute and a Senior fellow of the Fraser Institute. The views expressed in this article do not necessarily reflect those of these organizations.

Republish this article

Republish this article on your website under the creative commons licence.

Learn more