Article Text


  1. Miguel Martin
  1. Instituto de Investigación Sanitaria Gregorio Marañón, Universidad Complutense, CiberOnc, GEICAM, Madrid, Spain
  1. Correspondence to Dr Miguel Martin; mmartin{at}

Statistics from

A biosimilar is a biotechnological product that has shown a high similarity to a reference biological product already approved by a medicine agency for a particular indication or indications. The similarity must have been demonstrated in a sensitive population, using sensitive efficacy and safety endpoints that are clinically relevant. Complex biosimilars (mainly monoclonal antibodies, moAbs) are entering the oncology market now in Europe, and this has caught many oncologists off-guard in terms of knowledge of the real characteristics and efficacy of these products and their regulatory framework. Many oncologists are already familiar with simpler biosimilar products mainly used as supportive therapies (such as granulocyte colony-stimulating factors and erythropoietins). They are confident in these products since its efficacy can be easily assessed in the individual patient (ie, an increase in or red blood cell counts). This is not the case with the more complex anticancer moAb, like trastuzumab or bevacizumab. The oncologist cannot establish whether an individual patient is responding to one of these biosimilars as effectively as one could have expected treating the patient with the reference drug. This fact can create an initial distrust and considerable uncertainty in a significant proportion of oncologists. The distrust can be still more marked in the curative setting, for instance, when using a trastuzumab biosimilar in the neoadjuvant/adjuvant setting. Thus, an educational effort, led by the European Society for Medical Oncology (ESMO) educational committee, is necessary to clarify the characteristics of biosimilar products and the advantages and opportunities that they bring to the overall management of cancer in Europe. ESMO has recently published a position paper on that matter, recognising the topicality of the issue.1

The production of a complex moAb biosimilar is neither rock science nor a cheap and easy task. The goal of the production of a biosimilar moAb is generating a protein structurally and molecularly as similar as possible to the original one, assuming that if this goal is reached, both products will have a similar efficacy and toxicity. In the era of modern recombinant DNA technology, molecular cloning is a relatively easy task. On the other hand, producing a biosimilar has little to do with producing a generic drug. The cost of the research process and production of a biosimilar is high (as much as US$200 million).

The European Medicines Agency (EMA) recommends a stepwise approach for the development of biosimilars, starting with a careful structural and functional characterisation in order to establish similarity between the biosimilar and the reference product.2 Comparative non-clinical studies to assess differences in biological properties should be conducted first. In the case of moAb, particular attention should be paid to demonstrate similarity in binding to target antigen, binding to representative isoforms of the Fc gamma receptors, FcRn and complement (C1q), Fab-associated functions (eg, neutralisation of a soluble ligand, receptor activation or blockade) and Fc-associated functions (eg, antibody-dependent cell-mediated cytotoxicity, complement-dependent cytotoxicity and complement activation). In vitro cell-proliferation studies should be also performed and can be followed by in vivo studies if there is a relevant animal model. The conduct of toxicological studies in animals is not mandatory. Comparative pharmacokinetic studies in healthy volunteers or patients, including pharmacokinetic/pharmacodynamic and comparative immunogenicity studies, should be performed to confirm the similarity between the biosimilar and the original product. In single-dose studies, the primary parameter should be the area under the curve (AUC)(0-inf), with Cmax, Tmax, volume of distribution and half-life as secondary parameters. In multiple-dose studies, the primary endpoints are the truncated AUC after the first administration until the second administration (AUC-0-t) and AUC over a dosage interval at steady state (AUCτ), with Cmax and Ctrough at steady state as secondary parameters. These studies are crucial to establish similarity between the two proteins.

The preclinical and pharmacokinetic parts of the development of a biosimilar moAb are crucial for its development and approval. An important concept to keep in mind is that the aim of the development of a biosimilar is to demonstrate similarity in terms of biological properties, clinical efficacy and safety compared with the reference product, but not patient benefit, which has been already demonstrated in large phase III trials by the reference product. An acceptable level of similarity in clinical efficacy can be demonstrated by moderately sized (500–900 patients), randomised, double-blind, comparative clinical trials with equivalence as main goal. The endpoints of such trials differ from those required for the registration of the original product (ie, progression-free survival (PFS), disease-free survival (DFS) or overall survival). Surrogate endpoints like response rate of pathological response rate (pCR) (in the case of breast cancer neoadjuvant trials)3 are considered appropriate endpoints for trials aimed at demonstrating the similarity of a moAb biosimilar to the original product.

At first sight, this is sometimes difficult to understand for some oncologists, used to see more clinically relevant endpoints (ie, PFS or DFS) in registration clinical trials. The approval of moAb biosimilars is based on the overall picture of development in which preclinical studies aimed at showing biological similarity play a major role; clinical trials are only needed to confirm that the similarity in biological characteristics is translated into an equivalent antitumor activity.

The arrival of biosimilar has been accompanied by some debate, still open, about extrapolation, automatic substitution and switching/interchangeability of products. Extrapolation is the translation of the clinical efficacy and safety data seen in one indication (ie, neoadjuvant therapy of breast cancer for trastuzumab biosimilars) to other indications of the reference drug (ie, metastatic breast cancer or gastric cancer). It is admitted in most countries, provided both products have proven bioequivalence, even though no specific trials in the indications to be subject to extrapolation have been performed.

Automatic substitution of an original moAb for a biosimilar by the pharmacist (without consulting the prescribing physician) is completely prohibited in 9 out of 28 EU Member States,1 while some other countries have more flexible policies.

Interchangeability (switch from an original moAb to a biosimilar, or from a biosimilar to another one, implemented by the prescribing physician) is a matter of debate, based on concerns about immunogenicity. In any case, the decision of switching should be made by a physician having a good understanding of the products. The switch should also be discussed with the patient. The safety of a switching strategy with moAb biosimilars has been proven in rheumatoid diseases. In a large trial, patients who were switched from the original infliximab (Janssen’s Remicade) to CT-P13, an infliximab biosimilar (marketed by Celltrion as Remsima and Inflectra), did not show any unexpected differences in the immunogenicity, as compared with the group treated with CT-P13 alone.4 There is little experience on interchangeability of moAb in the oncology field. The Lilac study, a neoadjuvant/adjuvant trial, compared trastuzumab with a biosimilar, AMGEN’s ABP 980, in combination with chemotherapy (four cycles of adryamicin-cyclophosphamide (AC) followed by weekly paclitaxel plus either trastuzumab or ABP 980) as neoadjuvant treatment for patients with HER2-positive early breast cancer ( Identifier NCT01901146). The pCR rate was similar with both therapies. Patients allocated to trastuzumab in the neoadjuvant part were randomised to adjuvant trastuzumab or a switch to ABP 980 in the postsurgical adjuvant phase.5 No patients developed neutralising antibodies in any arm at any study time point, suggesting that the switch is safe. However, the interchangeability strategy poses an additional issue: the difficulty of tracking and allocating infrequent side effects of the new biosimilars in the post-authorisation period. The EMA has not expressed a clear position or recommendation on oncology moAb biosimilar interchangeability so far.

In an era in which the increasing cost of the new oncology drugs is becoming a real concern even in developed, rich countries, the arrival of biosimilar antitumor moAbs is likely to be of great help to maintain health system sustainability. Since the introduction of these drugs means a new scenario for oncologists, a great educational effort aimed at providing an unbiased and complete information about moAb characteristics and its regulatory framework is necessary. The clear and documented ESMO position paper1 is a good start, but additional educational work is still needed to provide oncologists with a better understanding of the complex world of moAb biosimilars.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
View Abstract


  • Funding None declared.

  • Provenance and peer review Not commissioned; internally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.