Nocebo Effect May Have Role in Some Treatment Failures, Adverse Events in Nonmedical Switching

Researchers reviewed randomized controlled trials and real-world evidence studies on nonmedical switching; that is, switching patients who are doing well on their current originator therapy to a biosimilar.

The nocebo effect may contribute to treatment failures upon nonmedical switching from an originator to a biosimilar, according to a review article, whose authors recommend rigorous studies to evaluate the nocebo effect, better patient education on biosimilars, and involving patients in the decision to switch.

Researchers reviewed randomized controlled trials (RCTs) and real-world evidence (RWE) studies on nonmedical switching, ie, switching patients who are doing well on their current originator therapy to a biosimilar.1

Although head-to-head trials in biologic-naive patients have generally found similar efficacy and safety of originator and biosimilar, discontinuation of biosimilar therapy due to loss of response or adverse events is somewhat common in switch studies. The authors noted the reasons for these treatment failures have not been investigated, and they explore the possibility that some of these failures are due to negative expectations of the patient who is aware they are switching from the originator to the biosimilar—the nocebo effect.

According to the authors, the currently available data does not allow for distinguishing between a true difference in pharmacologic activity and the nocebo effect as potential reasons for loss of efficacy or adverse events. They acknowledge that the clinical effects of subtle differences between originators and biosimilars are not fully known, and could cause certain individuals to respond differently; furthermore these subtle differences in the molecule potentially could trigger an immunogenic response to a biosimilar upon switching. The authors call for well-designed, blinded clinical trials on nonmedical switching with appropriate control groups and attention to the causes of therapeutic failure to accurately evaluate how much loss of response is attributable to the nocebo effect.

Nonmedical switching is most often motivated by cost. However, additional physician or hospital visits due to adverse effects or failure of treatment following nonmedical switching could negate the cost savings of switching to a biosimilar.

Negative views of biosimilars by physicians and patients remain a barrier to reducing costs. The authors cite a survey in which 84% of US physicians said they would not support switching a stable patient from an originator to a biosimilar,2 highlighting the need for more well-designed studies to improve physicians’ views of nonmedical switching. Also, since patients’ negative expectations are the driver of the nocebo effect, the authors suggest better patient education could help to reduce misconceptions about biosimilars, thereby reducing the potential for the nocebo effect.

The authors also recommend involving patients in the choice to switch. They note that in some countries and by some insurers, nonmedical switching is mandated for patients who are doing well on originator therapy. They recommend switching not be mandated, as involuntary switching may exacerbate patients’ negative expectations, leading to greater likelihood of the nocebo effect. In their review of the literature, they found the rate of discontinuation of biosimilars was lower in RWE studies in which patients were able to choose whether to switch.

References

1. Fleischmann R, Jairath V, Mysler E, Nicholls D, Declerck P. Nonmedical switching from originators to biosimilars: Does the nocebo effect explain treatment failures and adverse events in rheumatology and gastroenterology? [published online ahead of print, January 16, 2020]. Rheumatol Ther. 2020;10.1007/s40744-019-00190-7. doi:10.1007/s40744-019-00190-7

2. Teeple A, Ellis LA, Huff L, et al. Physician attitudes about non-medical switching to biosimilars: results from an online physician survey in the United States. Curr Med Res Opin. 2019;35(4):611—617. doi:10.1080/03007995.2019.1571296.