Not the Myriad story

An invitation this week to take part in a discussion about the Myriad BRCA patents case on BBC television this Sunday has confirmed that the UK media are as fascinated by this story as their counterparts in the USA. Everyone loves a courtroom drama, and this case has drama in spades, at least for those with a passionate interest in the subject of gene patents. However, I venture to suggest that whatever the outcome of the Myriad / ACLU suit, it will not be year’s the most important decision concerning diagnostic monopolies based on DNA patents. For that we have to turn our gaze across the Atlantic from the USA to the UK; from germline DNA to the somatic mutations present in cancer tumours, and from the USA Supreme Court to the National Institute for Clinical Excellence (NICE).

NICE is not as well known as the US Supreme Court, but for the manufacturers of healthcare products its decisions are just as significant. Within its own domain – Health Technology Assessment – it has a global reputation, and the decisions it makes about whether or not to recommend coverage of new drugs and devices have influence far beyond the UK NHS. In recent years NICE has begun to pay far greater attention to the evaluation of diagnostic devices, and in 2011 it established the Diagnostics Advisory Committee as a focal point for its work in this area. Since then that committee has been involved in what I believe is going to be a landmark evaluation, with profound implications for the molecular diagnostics sector and for public healthcare systems.

The evaluation concerns the relative merits of a number of prognostic tests for post-adjuvant breast cancer patients. There has been a proliferation of such tests in recent years and their exact intended uses vary – they may predict either breast cancer recurrence, risk of metastasis and/or likely response to chemotherapy. Just as BRCA testing became an exemplar for genetic risk prediction, so these tests have become the poster child for molecular tumour profiling based on somatic DNA. Furthermore, the business model adopted by many of the firms entering this space is the same as that used by Myriad Genetics. The traditional in vitro diagnostics (IVD) sector has been a high-volume, low margins business where companies hold intellectual property (IP) in testing platforms, and have not competed over biomarkers.[i] But like Myriad Genetics, the leading companies in the breast cancer prognosis space ( Agendia and Genomic Health) have emerged with a new business model based on exploiting IP in biomarkers and selling their tests not as kits, but as proprietary laboratory-developed tests (LDTs) delivered by the company’s own reference laboratory.

This business model may offer a number of commercial advantages – in the USA firms may sidestep the need for FDA approval (although Agendia’s MammaPrint test is FDA approved); the time to market may also be shorter because technical validation of a test performed in one laboratory may be easier than development of a kit which needs to perform reliably in multiple laboratories; and finally, by creating a proprietary diagnostic monopoly, it may be possible to gain higher reimbursement rates for your tests. This latter trend is perhaps the most significant for hard-pressed healthcare systems in an era of fiscal austerity and  Genomic Health’s Oncotype Dx breast cancer test exemplifies the trend – their 10k annual report filed in March 2013 states that the list price of the test is $4,290.

While other poster children for personalized medicine have yet to garner widespread clinical acceptance, breast cancer prognostic tests are in growing use.  Genomic Health has had significant success – according to their 2012 report they have achieved insurance coverage for 90% of women with node-negative invasive breast cancer in their domestic US market. Increasingly the company is looking for growth overseas – the 2012 report states that they are now providing testing to patients in over 70 countries. However, the report also states that they anticipate that “it will take several years to establish broad coverage and reimbursement” outside the USA. NICE approval would be a significant milestone in their global ambitions.

Three points arising from NICE’s draft decision are worthy of note:

1)      Genomic Health has sought to differentiate itself from its main rival Agendia by predicting likely benefit from chemotherapy as well as likelihood of recurrence. NICE has rejected the data on chemotherapy benefit  as “not robust enough”. This decision seems to confirm the profound challenges associated with demonstrating the utility of using molecular technologies to stratify patients based on their likely response to treatment.

2)      The NICE draft decision to recommend Oncotype is based on a narrowing of the population for which the test would be used, suggesting that it will only be cost-effective  “… in people at intermediate risk of distant recurrence where the decision to prescribe chemotherapy remains unclear … “ In other words NICE deem that it will not be cost-effective to use the test on patients who are classed as at either low or high risk of recurrence using existing protocols.

3)      The cost-effectiveness even in this narrower indication is predicated on a discounted price offered to NICE by Genomic Health late last year. The original price quoted in the NICE evaluation is £2,580, but the sterling equivalent of the current US list price of $4,290 is £2,822 (based on a current exchange rate of 1 USD = 0.662522 GBP), so even the initial price is lower than the company’s current list price, but on top of that Genomic Health have now offered a further discount. How much that discount might be is not known; such information is always treated as commercially confidential by NICE (what level of discounting the company commonly offers to US insurers is also unknown.)

Whatever the rate of the discount it seems safe to conclude that the price will still be considerably greater than the cost of running one of the other tests under consideration – the in-house NHS option of a combination of four markers called IHC4. The NICE evaluation costed IHC4 at £150 per test. One HTA expert I spoke to looked at the draft NICE decision and based on some rough calculations suggested the Genomic Health discount might be as little as £100, giving an NHS price for Oncotype Dx of £2,480. But let us be more generous by a factor of ten and imagine a discount of £1,000 (based on an assumption that winning NICE approval is strategically important to Genomic Health as a stepping stone to market penetration in Europe) –  we are still left with a test which is nearly ten times the cost of the in-house NHS equivalent.

The key advantage which Oncotype Dx enjoys over IHC4 is evidence. The UK test has been in development for a relatively short time, and although NICE deem it promising, it lacks the cumulative weight of evidence from multiple studies which support use of Oncotype Dx. What this suggests is that in the era of proprietary combinations of biomarkers, then significant first mover advantage is gained from being early to build a clinical evidence base. Furthermore, such advantage can gain its own momentum: the two leading tests – Mammaprint and Oncotype Dx – are now benefiting from large publicly-funded trials to test their utility (MINDAct and TailorRx respectively). What seems to have been missing up until now is comparative head-to-head studies of rival technologies. This is set to change with the launch in the UK of the OPTIMA trial, a study which may give IHC4 the chance to prove its worth against its commercial counterparts.

So what does all this boil down to? Let us return to the Myriad Genetics BRCA story. When Myriad threatened the UK government with litigation if it did not respect its patents, there was a groundswell of opposition and BRCA testing remained an NHS service. With similar outcomes in Canada and much of the rest of Europe, it seemed that the blockbuster diagnostic model based on biomarker patents might not be transferable beyond the unique healthcare system of the USA. If NICE hold to their draft decision on Oncotype Dx, then that assumption will have been blown apart. UK patients will have access to a technology commonly viewed as at the cutting-edge of personalised medicine, but the broader ramifications of such transnational diagnostic outsourcing for a cash-strapped NHS increasingly under threat of partial privatisation remains to be seen. Whether other diagnostics companies can benefit from this decision is also as yet unknown, although industry must be watching this case with great interest.

The question I want to end on is this: why is there so much academic and media attention focused on the Myriad BRCA lawsuit and yet practically zero interest in the NICE decision on Oncotype Dx?

[i]Garrison, L and Finley Austin, M ‘Linking Pharmacogenetics-based diagnostics and drugs for personalized medicine’ Health Affairs 25, 1281–1290 (2006)