What do oncology and orphan diseases have in common? Even ten years ago, that question might have sounded like the opening to a joke with a bad punch line. ‘Not much’ was the honest answer. Cancer is a true force of nature, affecting an estimated 29 million people on the planet at any given time and earning it the nickname the ‘emperor of all maladies’, as Siddhartha Mukherjee dubbed it in his Pulitzer Prize-winning book of the same name.
On the other hand, orphan diseases are by definition very rare. In the United States, an orphan disease is one affecting <200,000 persons (1 in every 1,500). In Europe, the definition is a bit narrower, with <5 in 10,000 (1 in every 2,000) people affected. Japan is stricter still, with a prevalence of <50,000 (1 in every 2,500) required for a disease to be considered ‘orphan’.
Why would the exact definition of orphan diseases matter? Because market incentives to develop treatments for ‘rare’ diseases are colliding with recent advances in both our ability to understand what is driving a disease and our ability to create ‘personalised’ drugs to hinder the process. The result is a shift in pharmaceutical R&D strategy towards development of treatments that target small populations and charge high per capita prices.
Oncology is orphaned
It may sound strange to call a disease an ‘orphan’, but the term was not chosen flippantly. Due to the small population affected, drug manufacturers had little hope of recouping the costs of developing a drug with such small market potential, and thus little R&D was conducted in these areas. In 1983, the US Congress stepped in to try to change that by passing the Orphan Diseases Act. It confers on them a number of incentives, including seven-year market exclusivity, tax credits, grants for drug development, fast-track approvals, and waived fees.
Initially, the legislation was used exactly as intended, with companies focusing on truly rare indications such as Fabry Disease, Gaucher Disease, Cystic Fibrosis and Cushing Syndrome. A list of rare diseases at the time included only a few oncology indications – rare tumours such as Barrett’s oesophagus (a pre-cancerous condition caused by acid reflux to the oesophagus) and Wilms Tumour (a type of kidney cancer primarily affecting children).
However, over the past twenty years, two forces have collided to shift that balance, and now over 40 per cent of orphan drugs are indicated to treat various cancers. In fact, some of the top grossing oncology drugs – ‘household’ brand names like Rituxan, Herceptin, Avastin – have orphan status. How did we get here? A combination of better disease understanding coupled with the end of the ‘blockbuster’ patent era that has driven pharmaceutical companies to find new ways to profit.
A few years ago, ‘targeted therapy’ was the buzzword in oncology. It started with Glivec (imatinib mesylate), an oral small molecule tyrosine kinase inhibitor (TKI) manufactured by Novartis that appeared to stop chronic myeloid leukaemia (CML) in its tracks. Patients who previously could expect to live only 2-3 years after diagnosis were still going strong after 5-10 years. The drug was a true breakthrough because it targeted the source of the disease – a translocated chromosome (dubbed the Philadelphia chromosome after its city of discovery) which churned out a faulty protein (Bcr-Abl) that drives tumour genesis. At long last, a ‘cure’ for a particular type of cancer seemed to be available in the form of a little orange pill.
Around the same time across the country, Genentech was busy developing a monoclonal antibody, trastuzumab, which seemed to target the HER2 receptor protein which is over-expressed on some breast cancer cells. By blocking the receptors, the drug appeared to cut off the signal that drove tumour growth. That drug went on to become the blockbuster known as Herceptin.
What is interesting to note is that both of these drugs almost never made it to market: CML is a rare indication, with roughly only 6,000 new cases diagnosed in the US every year – 95 per cent of which are driven by the Philadelphia chromosome. Researchers at Novartis worked tirelessly for years to gain traction for their project. Similarly, HER2+ breast cancer accounts for only roughly a quarter of all breast tumours, and patients must undergo expensive testing to determine whether or not their tumours over-express the protein in order to be eligible to take the drug.
These breakthroughs took the world of oncology by storm. Targeted therapy seemed to be the long-sought answer to the cancer conundrum, and companies began looking to develop both monoclonal antibodies and TKIs to target the root causes of various tumours. It seemed that an end to the War on Cancer that US President Richard Nixon had declared in 1971 might truly be on the horizon. However, the case of AstraZeneca’s Iressa (gefitinib) proved a cautionary tale of how much we still do not yet understand.
As scientists started to uncover the drivers of individual tumour types, they discovered that many types of lung cancer seem to over-express a receptor called EGFR, much like some breast cancers over-express the receptor HER2. AstraZeneca seized on this insight to develop the small molecule TKI Iressa. As lung cancer is one of the most common tumours worldwide, Iressa held amazing promise to be a global blockbuster when it was first launched in Japan in 2002.
Unfortunately, the biology was not that straightforward. Subsequent trials showed little benefit from the drug, and there were a number of patient deaths. After much trial and error, it was found that EGFR over-expression itself does not predict response; rather, only those patients who had a specific ‘biomarker’ – a mutation to the gene responsible the EGFR gene – seem to gain significant benefit. In fact, those without the mutation seemed to fare far worse on Iressa than on traditional chemotherapy.
Further subset analysis revealed that this characteristic seems to be prevalent among female Asian non-smokers – a very specific patient population indeed! And, unfortunately, a much smaller population than the drug’s original indication, giving it a thus greatly reduced market potential.
Iressa taught us a number of important things – it matters what you measure, how you measure it, and how you define positivity for the purposes of drug eligibility. Before Iressa, we used to think of a tumour such as lung cancer purely in terms of histology, or cell type. There were two types – small cell (SCLC) and non-small cell (NSCLC), and each type was treated with different cytotoxic chemotherapies.
As our understanding of the drivers of different tumours has grown, we have learned that within a particular tumour type such as lung cancer, there are myriad different potential genetic drivers of the disease each forming its own subset. Thus, ‘lung cancer’ is actually an umbrella term for a number of genetically distinct diseases, each with its own unique potential genetic targets for drug therapy.
Pfizer’s Xalkori (crizotinib), a TKI that targets the protein that is created due to a specific chromosome translocation, is a perfect example of translating this scientific understanding into therapy that makes a difference. Patients who have the ALK translocation in their tumours experience significantly better outcomes on Xalkori than they do on chemotherapy.
Xalkori, like Glivec and Herceptin before it, has been hailed as an example of the dawn of truly ‘personalised’ medicine – a drug designed specifically for a patient’s individual disease, as opposed to the ‘one size fits all’ approach that has been chemotherapy. This is certainly very good news for both patients and pharmaceutical companies.
The personalised medicine punchline – affordability
However, there is a catch – the punchline to our opening bad joke. What do oncology and orphan diseases have in common? Increasingly, a lot: the ALK mutation is only found in around 3 per cent of all NSCLC patients, making them a very small subset of the tumour type. In order for Pfizer to recoup its development costs for the drug, it needs to ensure that all diagnosed NSCLC patients are tested for the ALK mutation.
This sounds easy enough, but in actual practice it is quite difficult – testing is expensive and tissue samples are limited. And as more companies pursue specific therapeutic targets for each tumour type, competition to get a ‘bite’ of limited biopsy tissue for biomarker testing is poised to become fierce.
The reality is, the more ‘personalised’ the medicine, the fewer candidates there are for that medicine and that means fewer patients from whom companies can recoup the development and production costs. These therapeutic advances leading to personalised medicines are taking place just as healthcare systems around the world are groaning under the burden of ageing populations coupled with soaring expenses. As such, personalised medicines, many of which have Orphan Drug status, are increasingly falling under the spotlight of regulators due to their high cost:benefit ratio. Pressure to reduce prices is being felt on both sides of the Atlantic. Sanofi was recently forced to halve the price of its colorectal cancer drug, Zaltrap (zif-aflibercept), after Sloan-Kettering rejected it on a cost basis. Meanwhile, UK reimbursement strategies are increasing moving towards ‘no effect no pay’ in oncology.
There is speculation that targeted therapies and personalised medicine in orphan diseases and oncology could lead to a shake-up in orphan drug incentive schemes, and treatments may be increasingly restricted to segmented patient populations. In other words, scientific advancement may potentially be the victim of its own success, as societies find they struggle to afford the cost of developing specific drugs for each genetically distinct disease.
‘Personalised’ challenges
As we have seen, the nature of oncology therapy is changing rapidly, and the industry needs to adapt its approach to the area. It will need to be able to answer questions such as, will a biomarker test be needed? What will it do to the price of the drug? Will doctors test for our client’s drug over the competition when numerous potential targeted therapies exist for an indication?
Other key areas to investigate are whether potential secondary indications provide strong enough markets in which to expand; what price point the market can support; and how can the patient perspective be integrated into a drug’s development.
Finally, don’t assume that an orphan oncology indication writes its own positioning – cancer brands often span a number of distinct and quite different indications, so it is critical to build brands from the beginning of development, rather than just market products for specific indications.
