General Awareness of Biologics and Biosimilars
Overall, most physicians are aware of which drugs used within their specialties are biologics; however, there was a sizeable minority that did not know this information (Fig. 1). Some physicians mistook
chemical drugs for biologics and many selected “none of the above,” which indicated that they did not recognize any of the suggested therapies as biologics. These data reflect a lack of understanding of what defines a biologic, which was surprising since all physicians included in the survey responded that they prescribe biologic drugs. If physicians do not know that a given drug is a biologic, it will create an educational challenge to explain that, after patent expiry, copies of these therapies will be marketed as biosimilars instead of as generic drugs.
Although the vast majority of respondents across all specialties have heard about biosimilars (Fig. S1 in Supplementary Materials), their actual knowledge of the fundamentals of biosimilars, as demonstrated in greater detail below, was low. For example, a sizable minority was unaware that a biosimilar was available in the US (Fig. 2).
Most respondents (75%) appear to trust FDA approval decisions; however, they will still want additional information when making treatment decisions (Fig. S2 in Supplementary Materials). A small group of respondents (13%), predominantly dermatologists and rheumatologists, will not solely trust the FDA’s assessments and decision to approve the biosimilar and will seek more information. Presumably, these physicians will decide which medications to use based on their own assessments. The survey did not probe further into how those decisions would be made.
Understanding and Perceptions of Key Concepts Related to Biosimilars
Only half of respondents are knowledgeable of the concept of totality of evidence, which is a new data evaluation paradigm used to assess a biosimilar (Fig. 3) . This insight is not surprising because innovator drug development is typically not based on analytical data and has always included clinical data for all indications. Physicians have been taught to expect as such when assessing new drugs.
One fundamental premise of biosimilars is that a product will be clinically evaluated in one or more indications of use, and the FDA may ‘extrapolate’ the totality of data to approve the biosimilar for use with other indications for which the originator biologic is approved. While only 12% of respondents surveyed are comfortable with the concept of extrapolation (Fig. S3 in Supplementary Materials), separate questions that assessed details of extrapolation reveal that most physicians are not familiar with the concept. This lack of knowledge was further highlighted at a US Advisory Committee meeting held on July 12–13, 2016,  whereby the panelists became more accepting of the concept after it was explained that extrapolation was based on totality of evidence data from the originator biologic molecule to the biosimilar molecule and not from the clinical aspects of one indication to another indication.
Almost 60% of respondents correctly understood that to be approved as “interchangeable,” a biosimilar must be shown to be safe and effective for back-and-forth switching with no negative impacts to safety or efficacy (Fig. S4a in Supplementary Materials). However, there is a lack of awareness or misinformation about the term ‘interchangeable’ (Fig. S4b–e). Overall, one-third of respondents believe that the term means that approval of a biosimilar implies interchangeability with the originator biologic (Fig. S4b). In fact, an additional category was created within the US regulatory pathway for an interchangeable product. As of the time of writing of this article, the exact nature of the additional information that will be needed has not yet been defined, but is thought to include additional clinical data beyond that provided to establish biosimilarity. An overwhelming 80% of the respondents were not aware that interchangeability would enable a pharmacist to switch between originator biologic and biosimilar, and vice versa (Fig. S4c).
Perceptions of Safety and Efficacy
More than half of respondents are aware that biosimilars must be comparable in safety and efficacy to the originator biologic, although that leaves a sizeable minority who are uncertain about these important approval criteria (Fig. 4). Physicians who are uncertain of this foundational fact are more likely to have reservations and may be cautious in adopting biosimilars.
Nonetheless, a large percentage of respondents are unsure or concerned about the safety of biosimilar medicines (Fig. 5), although the FDA can only approve a biosimilar if the agency is convinced that the safety profile is comparable to that of the originator biologic. This misconception could simply be due to a lack of knowledge, although the higher levels of uncertainty among rheumatologists compared to other specialties suggests that this concern may be caused, at least in part, by coordinated activities that raise questions about biosimilars and the biosimilar pathway, as biosimilars are being developed for several leading rheumatology products marketed in the US.
About half of respondents across all specialties specified that they wanted to have access to data directly evaluating the safety of switching patient populations from the originator biologic to its biosimilar (Fig. 6). This highlights a potential concern of many physicians that the very act of switching may be dangerous because of concerns that the biosimilars may not be structurally identical to their reference product. It is likely that some physicians are unaware of two factors: (1) the primary sequence of a biosimilar must be identical to the reference product, and (2) that the quality attributes of all biologics (including both originator biologics and biosimilars) often vary from batch to batch within specified limitations, and that this variability has not led to safety concerns . The critical quality attributes of batches of biosimilars must also be within the very same limits (pending FDA issuance of a guidance on statistical considerations for the analysis of analytic similarity). In addition, while not specified in the Biologics Price Competition Act of 2009 (BPCIA)  or in US guidance documents, as of September 2016, all biosimilar products approved by the FDA have incorporated at least one switch from originator biologic to biosimilar into their biosimilar clinical development programs. These important factors must be better communicated to physicians.
More than one-third of respondents believe that abbreviated approval translates to a greater safety risk (Fig. 7), which may reveal a basic misunderstanding that the “abbreviated” program refers to less supportive data. In fact, the analytical data requirement is much greater and forms the basis for the totality of evidence supporting biosimilarity. The “abbreviated” portion refers only to a requirement for fewer and smaller efficacy and safety studies because the purpose of biosimilar clinical studies is to confirm high similarity to the originator biologic and not to re-establish the safety and efficacy profile that was already established by the originator biologic. Of the specialties, dermatologists and rheumatologists expressed significantly higher concerns at 43% and 48%, respectively.
Finally, the majority of physicians (85%) appreciate that, as is true for all biologics, the immunogenicity profile of a biosimilar will not be completely known pre-approval (Fig. S5 in Supplementary Materials). To alleviate possible concerns about immunogenicity of a given biosimilar, it will be important to communicate the safety data that will be obtained from post-approval pharmacovigilance surveillance. It also may be useful to better communicate the safety data obtained with use of biosimilars in Europe, where there have been more than 400 million patient-days of experience in which no new immunogenicity concerns have been observed .
Preferred Information Sources
Peer-reviewed literature is by far the most trusted and preferred information source for biosimilars among physicians (82%; Fig. S6 in Supplementary Materials). The top trusted information sources include scientific journals (88%), FDA (73%), and physician peers (64%); however, dermatologists and rheumatologists indicated less trust in the FDA as an information source compared with other specialties at 64% and 58%, respectively (Fig. S7 in Supplementary Materials). Physician trust in media was very low across all specialties (<5%). Journals, physician peers, and congresses/symposia are the primary trusted information sources for current information on biosimilars across specialties.
Physician Interests and Overall Thoughts
Approximately 60% of respondents are concerned about patient compliance with originator biologics and/or access to treatment options (Fig. 8). A large majority of respondents believe that biosimilars will enable increased access and utilization, expand treatment options, and provide savings to the US healthcare system (Fig. S8a, b in Supplementary Materials). Although a large majority of respondents (91%) are open to switching patients to a biosimilar (Fig. S9 in Supplementary Materials), the concerns noted above in several separate questions must be taken into account as they may impact uptake and utilization of biosimilars. Primarily, physicians are open to the concept of using biosimilars for their patients but would like to see the supportive data for their indication. They would also like to see switching data. When the FDA requires one-time switch data from originator biologic to biosimilar for chronic conditions, the switching data will be available but must be communicated.
Finally, most physicians are interested in learning more about key biosimilar concepts, suggesting an openness to learn. Topics of interest to physicians included product attributes of safety and efficacy and key concepts of biosimilars (totality of evidence, extrapolation, interchangeability), as well as information on access and the economics of biosimilars (Fig. 9).