Feature Pharmaceuticals

Monoclonal antibodies: magic bullets with a hefty price tag

BMJ 2012; 345 doi: http://dx.doi.org/10.1136/bmj.e8346 (Published 12 December 2012) Cite this as: BMJ 2012;345:e8346
  1. Allen F Shaughnessy, professor
  1. 1Tufts University School of Medicine, Boston, Massachusetts, USA
  1. Allen.Shaughnessy{at}Tufts.edu

Sales of monoclonal antibodies are projected to reach more than $160bn in the US alone over the next few years so is it any wonder that drug companies fiercely protect their profits? Allen Shaughnessy considers why these unique and complex drugs are so eyewateringly expensive

The pharmaceutical industry just doesn’t feel the love these days. Mistrust abounds among the general public as well as within the medical industry. It’s no surprise, then, that eyebrows raise and fingers point when a company withdraws an effective drug from the market shortly after proof is published of its benefit in a completely new treatment arena. What is it up to?

Alemtuzumab is a monoclonal antibody marketed to treat chronic lymphocytic leukaemia. It has also been used, off-label, for multiple sclerosis, and last month two phase III studies were published showing its efficacy and superiority over interferon beta-1a.1 The US Food and Drug Administration and the European Medicines Agency are considering approval for this indication.

By now many patients with multiple sclerosis should be taking the drug, even though it hasn’t yet been sanctioned for this use by regulatory agencies. But Genzyme, a Sanofi company, withdrew it from the market two months before these studies were published. Some predict that once it is approved for multiple sclerosis alemtuzumab will be re-released under a new brand name (and at a much lower dose than that used for leukaemia), this time at a much higher cost.2

From a business perspective, this move makes sense. The cost of Campath, the brand name for the leukaemia version, was about $60 000 (£37 000; $46 000) a year. Lowering the dose to that used to treat multiple sclerosis would have reduced the price to $6000 a year.

This would have been a bargain basement price for immunomodulator treatment of multiple sclerosis. Natalizumab, another monoclonal antibody used for multiple sclerosis, is about $55 000 a year.2

While it might be a boon for patients and their healthcare payers, it does not make fiscal sense to produce a medicine that has been shown to be superior to existing treatments at a lower cost than the current price point. Corporations, including drug companies, are fiduciaries of shareholders and have a legal requirement to maximise the return on their investment.

Perhaps the manufacturer is taking its cue from Genentech, a subsidiary of Roche Pharmaceuticals, which sells bevacizumab for colon and other cancers and ranibizumab to treat patients with age related macular degeneration.

Although ranibizumab has theoretical advantages, and bevacizumab is not licensed for macular generation, clinicians around the world use bevacizumab rather than ranibizumab for the eye disorder because it is much cheaper. The cost differential was so striking that one primary care trust authorised the off-label use of bevacizumab rather than pay for the higher priced option, reversing its stance only after the company offered price concessions.3

Yet it seems inherently unfair to take a product, lower the dose, and inflate the cost several orders of magnitude. The uniqueness of monoclonal antibodies and the complexity of their development and production are touted as reasons why these products are so expensive. But is it really justified?

Monoclonal antibodies: new way to treat disease

In the late 1800s Paul Ehrlich, the German scientist and physician, imagined the development of a “magic bullet” that would selectively target a disease causing organism. Monoclonal antibodies could be the quintessence of his dream since it is possible to produce one that binds uniquely to almost any substance.

Sometimes monoclonal antibodies are used to stimulate the immune system to recognise the cell as foreign. In other situations they can block specific cell receptors required for tumour growth. They can even be used as radioimmunotherapy, delivering radiation to specific cell targets.

The first commercially marketed monoclonal antibody was muromonab-CD3, released in 1986 and used to suppress T cells to prevent rejection of organ transplants. Currently over 20 monoclonal antibodies are used as medical treatments for conditions ranging from cancer to autoimmune diseases to extremely rare inherited disorders. Monoclonal antibodies are also used for diagnostic testing—for example, home pregnancy tests use this technology.4

Even though the antibodies are highly targeted, they can often be used for different disorders that share the same pathology. For example, bevacizumab exerts its effect in patients with colorectal cancer by blocking the ability of vascular endothelial growth factor A to stimulate angiogenesis, which is required to support tumour growth. When injected into the eye, this same mechanism prevents the abnormal growth of blood vessels that results in age related macular degeneration and diabetic retinopathy.

Monoclonal antibodies are specifically produced from a single, cloned B lymphocyte cell line (hence “monoclonal”). To start the process, a mouse or rabbit is injected with an antigen that will stimulate B cells to produce an antibody specific to that antigen (each B cell produces an antibody to a single antigen).

These B cells are collected from the spleen of the mouse and mixed with myeloma cells, which grow continuously. This mixture is chemically manipulated to fuse the B cells with the myeloma cells, forming “hybridomas.”

Antigen screening is used to select the hybridomas that produce the right antibody, and these can then be cultured indefinitely in large bioreactors.

Multiple sclerosis occurs when T and B lymphocytes mistakenly attack the myelinated axons in the central nervous system, destroying the myelin and axon to varying degrees. Alemtuzumab targets T and B lymphocytes while sparing other immune system elements. The antibody binds to the CD52 protein found on the surface of mature lymphocytes but not the stem cells that produce them. After treatment, these CD52 lymphocytes, now tagged with the antibody, are destroyed by the immune system. Depletion of these lymphocytes is pronounced and long lasting, with a median recovery time to normal levels of 35 months.

How good is the evidence?

Two recently published phase III studies have shown that alemtuzumab is effective in patients with relapsing-remitting multiple sclerosis. The studies, called Comparison of Alemtuzumab and Rebif Efficacy in Multiple Sclerosis (CARE-MS I and II),5 6 enrolled previously untreated patients with low disability levels (CARE-MS I) and patients with a history of disease activity despite immunomodulator treatment (CARE-MS II). Alemtuzumab was more effective than interferon beta-1a in preventing relapses over the two years of study, producing a 54.9% improvement in previously untreated patients and 49.4% improvement in patients who had had treatment. In patients with more advanced disease, alemtuzumab also decreased the number of patients experiencing sustained accumulation of disability (hazard ratio 0.58, 95% confidence interval 0.38 to 0.87).

Treatment is not without risk. Immunosuppression related infections occur with alemtuzumab. In the CARE-MS I study, infections occurred in two thirds of alemtuzumab treated patients compared with 45% of patients treated with interferon. Herpes infection was also common (16-18%) despite antiviral prophylaxis in the CARE-MS I study.

Almost all patients receiving alemtuzumab experienced infusion related reactions despite pretreatment with high dose methylprednisolone. Other adverse effects reported in the studies were immune thrombocytopenia (1% in CARE-MS I) and thyroid disorders.

Why so expensive?

The top 12 biological products in the United States brought in combined revenue in 2010 of $30bn. By 2014, sales are expected to increase to $166bn, comprising about 30% of the branded prescription drug market.7

The average cost for the top nine biologicals is more than $200 000 a year in the US.8 The most expensive drug in the world is eculizumab, used to treat the extremely rare paroxysmal nocturnal haemoglobinuria, which affects about 5000 patients in the US and 1000 in the UK, costing a whopping $409 500 a year for the average patient.

Monoclonal antibodies are so expensive in part because of the cost and complexity of manufacture, the need for relatively high doses, and the price point set by early innovative treatments. Although raw material costs are low—$2 per gram of product produced9—the process itself is extremely expensive. A typical production run takes 10-14 days and may produce only 5-25 kg of antibody. The process involves 10 distinct steps, from initial culture of the cells through three separations to isolate the antibody.9 Royalties, research and development, and marketing costs add to the overall price. However, as the scale of production has increased over the past decade, production costs have decreased by two thirds.9

Governments in the US and UK have taken different approaches to dealing with the high cost. Since it is almost impossible to make the exact replica of a monoclonal antibody in a different processing plant, US lawmakers passed legislation in 2010 to promote competition in biological drugs by allowing “biosimilar” products to be marketed after the patent period. This ruling allows for production of generic monoclonal antibodies.10 Given the cost and complexity of developing and manufacturing these drugs, though, the price reductions are expected to be in the range of 20-30% rather than the 80% reduction that occurs when generic versions of typical medicines are marketed.11

End result versus perceived value?

If I need to keep track of time, I can buy a watch on a street corner in downtown Boston for $15 or a branded “tourbillon chronometer” for several thousand dollars (or much more). Both tell the time as accurately as I need, neither will turn my wrist green, and, at a casual glance, both look like a typical watch.

Yet my perception of the pricey watch is that its components and assembly are much more costly than those of the cheaper one. If I were to find out that both had essentially the same internal movements, I would be much less likely to fancy the expensive timepiece.

Had alemtuzumab not been previously available for the treatment of cancer at a lower price, there would have been no expectation regarding its cost, other than it would be in the range of existing multiple sclerosis treatments.

How do we decide what a product is worth? It is a human perversity that we are willing to pay a lot for something—until we know the cost of production. Then, we resent paying more, despite the fact that the product still is useful to us.

Monoclonal antibodies in numbers

  • $30bn: the combined revenue from the top 12 biological products in the United States in 2010

  • 453%: expected increase in US sales of the above top 12 from 2010 to 2014. It will bring annual revenues to $166bn, comprising about 30% of the branded prescription drug market

  • $200 000: average cost of treating a patient for a year with one of the top nine biologicals in the US

  • $409 500: average cost per year for the average patient on the most expensive drug in the world, eculizumab, used to treat paroxysmal nocturnal haemoglobinuria

  • $2: average raw material cost per gram of product produced

Notes

Cite this as: BMJ 2012;345:e8346

Footnotes

  • Competing interests: The author has completed the ICMJE unified disclosure form at www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declares no support from any organisation for the submitted work; no financial relationships with any organisation that might have an interest in the submitted work in the previous three years; and no other relationships or activities that could appear to have influenced the submitted work.

  • Provenance and peer review: Commissioned; not externally peer reviewed.

References