Pharma’s evolving relationship with NICE

pharmafile | November 1, 2006 | Feature | Research and Development, Sales and Marketing |  NHS, NICE, approval process, fast track, healthcare 

Back when New Labour was only halfway through its first term of office and when the Millennium Bug was feared rather than forgotten, a new NHS body was launched.

The National Institute of Clinical Excellence, as it was then, was unveiled on April Fools' Day 1999, with a promise to put an end to postcode prescribing and actively aid the introduction of new drugs. With hindsight, some would say it hasn't quite lived up to such ideals.

Wyeth EMEA's director for market access and reimbursement, Peter Conway, participated in the consultation process when NICE was being set up. "Its title then was Faster Access To Modern Medicines," he says."It seems ironic now, because I don't really perceive that it's actually given faster access to therapies in that many situations."

He goes on to say: "However, one fundamental point to make is that at least NICE gives us, as an industry,a transparent system to work with."

The transparency of NICE decisions has certainly been a mark in its favour, although the pharma industry would sometimes like more detailed information on the reasoning behind them – a search on the website, however, will often throw more light on these.

NICE's formative years

In its early years, there was considerable confusion over what impact NICE would have on the pharma industry, and companies took different approaches to the Institute says Paul Catchpole, healthcare management director at Roche and chair of the ABPI's health technology assessment user group.

"There were two sorts of views on NICE, initially – that it would go away if people waited long enough, or that it was here to stay and, therefore, it would be important to try to look at how to engage with it. That differential affected the UK industry's rate of progress in getting to grips with what NICE was all about.

"Now most companies have a pretty good understanding of what is required in a NICE submission and what is required in terms of working with NICE."

Nevertheless, the difficult decisions the Institute takes are often controversial. A prominent example of this was its guidance in February 2002 that severely limited the multiple sclerosis treatment, beta-interferon, leading to a unique 'risk-sharing' agreement between the government and manufacturers to make the drugs available. More recently, decisions this year on Pfizer's inhaled insulin Exubera and the Alzheimer's disease drug Aricept, which it jointly markets with Eisai, have provoked anger.

The Aricept decision denied the treatment to patients with mild Alzheimer's disease, yet allowed their treatment to be funded once they had progressed to the moderately-severe form of the disease, leading Eisai's managing director, Dr Paul Hooper, to label the decision perverse.

A further complaint from the industry is that the nature of NICE decisions have meant that it effectively works as a de-facto rationing body, whether that is the intention or not.

Wyeth's Peter Conway notes: "The NHS default reaction to NICE has been to use appraisals as an option to delay funding.

"Of course, NICE has tried to put steps into place to stop that happening, but it's inevitable that it happens that way."

The situation where NHS bodies await NICE guidance before agreeing to fund a drug occurs with enough regularity that it has even acquired its own name – 'NICE blight'.

Matters came to a head last year with unprecedented public, political and media pressure for access to Roche's Herceptin to treat early breast cancer, an indication for which the drug had not yet been licensed. Already under fire for being too slow to appraise new treatments, the Herceptin row was instrumental in the development of the new, fast-track single technology appraisal (STA) process and the drug was one of the first to be referred under the process to NICE by ministers.

External affairs director at Bristol-Myers Squibb Richard Marsh says: "Clearly the issue of NICE blight was becoming more serious. The time it was taking to get NICE appraisals was seemingly acting as a barrier to access to our products to patients, because the NHS was holding on until it got a recommendation from NICE before funding the products. So we welcome the fact that the STA process has come along."

On the right track

Unveiled last November, the new process is NICE's answer to critics' charges that it appraises drugs too slowly. NICE, or the National Institute for Health and Clinical Excellence as it is now, will still run its multiple technology appraisal (MTA) process for some drugs and others will benefit from STAs, which as the name suggests, focuses on assessing a single drug at a time.

The main differences for pharma between the two processes are speed of review and the evidence on which NICE bases its decisions. Evidence for an MTA comes mainly from an independent academic health technology assessment group, which produces an independent systematic synthesis of the evidence base – including but not limited to a manufacturer's submission. In the STA process, the manufacturer's submission becomes the primary piece of evidence and is, instead, critiqued by an evidence review group.

The other headline difference between the two processes is that STAs are speedier and can produce guidance much closer to launch. At its fastest, with no appeals, NICE says it will take eight months from initiation of an appraisal to publication of guidance. In contrast, the MTA process has generally taken between 12 and 24 months in the past.

The first five drugs to be appraised under the STA process are all breakthrough cancer treatments, among them Roche's Herceptin, in early stage breast cancer, and MabThera, in non-Hodgkin's lymphoma, both of which were recommended by NICE. Roche's Paul Catchpole notes that NICE had a relatively short period of time to get the STA process up and running, but says the process looks promising.

He says: "NICE has managed to reduce the time taken for appraisals by half, which is very welcome, because UK patients generally get slower access to new, effective medicines than  those in Europe.

"Now that NICE is established as an additional hurdle to access, it's important it doesn't take too long to address it and the new STA process is very welcome in that respect."

One of NICE's aims with STAs is to be able, where possible, to issue guidance in a matter of weeks, rather than months or, even, years  after a new drug receives regulatory approval. This is a huge shift from the way it has worked in the past and, indeed, the way the industry previously wanted it to work.

Pharma had argued that new drugs needed to be on the market for a number of years before they could develop a sufficiently mature evidence base for an appraisal to be conducted. But that was before the consequences of NICE blight were felt and the re-think it demanded on the timing of appraisals.

In the case of Herceptin, the first drug to be appraised under the STA process, NICE's preliminary recommendation came less than three weeks after it received European approval to treat early breast cancer. "Herceptin is the perfect example of the STA process enabling appraisals to be done much closer to the time of launch," comments Paul Catchpole.

"But with the evidence about a produc's clinical effectiveness maturing after it's been on the market for some time, you have to make more assumptions in the economic modelling undertaken to demonstrate its economic effectiveness."

This will require NICE to take a more pragmatic view of the available evidence and the modelling assumptions used – something, which Catchpole says will be essential if the process is going to be effective.

Professor David Barnett, who chairs one of NICE's two appraisal committees and who oversaw the STAs of Herceptin, Taxotere and Taxol, says the Herceptin STA is the 'gold standard' for such a review. "It ran almost perfectly – we knew what the licence was likely to be, we engaged with the manufacturer well in advance of that – and the process came together beautifully."

As such, it provides important points for other companies, yet to go through an STA, to take note of. Dave Pinnington, business development director with specialist healthcare and medical education agency Harvey Walsh, says it is critical for pharma companies to have a full understanding of NICE's requirements and agenda in order to work with the Institute.

"Some key considerations must be: the quality of the data being supplied, having a focus on disease and not just cost modelling and re-aligning brand strategy around the NHS agenda for long-term disease management."

Transitional issues

With such a new and important process, it is inevitable the industry has some reservations about it. "The flaw in the process," says BMS Richard Marsh, "is that there is no opportunity for input into the report, which the ERG puts together from our submission for the appraisal committee before it is made public on NICE's website. That remains a flaw with the process and we're talking to NICE about it."

Roche, too, has voiced concerns about this issue. However, NICE says  allowing manufacturers to comment will be considered in its forthcoming review of the STA programme as a whole.

The scope of the review has yet to be decided, but Professor Barnett, who will participate in it, suspects it may take in such issues as whether the evidence NICE uses is appropriate and if there are ways to include an independent assessment of the evidence.

Some of this should become clearer as more drugs are appraised. The first wave of STAs focused on oncology, but the second larger wave, announced in August, of 14 products includes asthma, MS, arthritis, glaucoma and heart failure treatments, alongside a number of cancer drugs.

To the second, expanded wave of STAs was also recently added a new work programme for assisting the Institute to assess the effectiveness of treatments and technologies currently used by the NHS, reviving concerns that NICE is being asked to do more work with fewer resources. The expanded workload follows last year's £3.5 million cut in NICE's budget and the reduction of its appraisal committees from three to two. However, a third appraisal committee will be restored this month and sited in Manchester, and like the two London-based committees, it will conduct STAs and MTAs.

Working together

The STA processes promise to alleviate the problem of NICE blight through faster appraisals, but relations between the industry and NICE are likely to remain variable because of the importance of decisions to both the NHS and pharma.

To some extent, this was always the case. "There were elements of uncertainty, mistrust and ignorance on both sides when NICE first started, in terms of what it would mean for people. Now the relationship has improved dramatically and it's much, much better, although there are still occasions of significant and, at times, heated debate," says Professor Barnett.

This change follows industry recognition that not only is the Institute here to stay, but that constructive engagement will produce more results than always squaring up to it.

Says BMS Richard Marsh: "We should have the confidence to say that if we are doing what we say we do – which is to produce genuinely innovative medicines that bring great benefits to patients – this will be recognised by an objective system."

Rational use, not rationing

An exclusive interview with NICE Appraisal Committee chair Prof David Barnett

Are flawed submissions from pharma a major problem for NICE?

We have not suggested that the manufacturers' submissions are either frequently or infrequently incomplete. It is a submission and, as such, it is up to manufacturers to make it as they see fit and there are guidelines on what should be included. The completeness of that is entirely within their purview.

What view does NICE take of the evidence pharma submits?

The current approach is that all evidence has value, but the degree to which that evidence is weighted will depend upon its place in a quality hierarchy. At one extreme would be the best, well set-up, randomised, controlled trial and the at the other extreme would be one individual's personal opinion on how they use this drug. That is the same for the STA or the MTA process.

If a pivotal randomised clinical trial was important for a drug's regulatory application, then obviously, it's a very important piece of evidence. Further down the list would be phase II studies, case series, confidential evidence held on file by the manufacturer – all of which would be taken into consideration, but weighted according to its quality.

Are you sympathetic to pharma concerns on proving cost effectiveness nearer to launch?

I'm aware of the difficulty expressed by the industry  on this issue, but I'm not not sure that we should have to wait for four years' use of the drug in the NHS before any consideration cost effectiveness is possible.

When NICE began seven and a half years ago – and I was there at the inception – companies were concerned that, in their view, it was difficult to identify cost effectiveness without knowing how the drug would be used in clinical practice.

Now we've moved on a bit and everybody says "Why are you so slow?" So, it's been a complete turn-around. I don't have an opinion on which is better, but there's certainly a dichotomy of views.

The timing of the process is not just for the benefit of the pharma companies or just for the benefits of clinicians and patients, but for the NHS as a whole. If the NHS says we need guidance from you as soon as possible because this is an important technology, then NICE has to respond, which is what were trying to do do with the STA process.

Could NICE issue more conditional approvals?

It's an interesting thought, although a decision almost implicitly is just that – the degree to which that is predicated is in the shelf life, the review date, given with guidance. So if, in fact, it was apparent that there will be new data coming along quite quickly, then the review date is set to make sure that is taken into account.

Does an alternative to the QALY (quality adjusted life year) need to be developed?

The QALY represents the best available method of both measuring and comparing healthcare benefits from a variety of different technologies across different diseases. I wouldn't say it's perfect, but then nobody is suggesting it is.

Health economics, as a science, including the use of the QALY, has dramatically improved in its ability to define the process of cost effectiveness assessment over the last seven years, and I believe that NICE has catalysed that change.

Does the QALY have any limitations?

Some of the limitations include the non-availability of quality of life evidence in some of the trials that are carried out, but that has changed over the years as it becomes almost a sine qua non for all submissions to NICE. It would be unusual for manufacturers to now say we don't have any evidence on quality of life or we don't believe it matters.

There are also issues around QALY weighting – some people, though I'm not one of them, would say that you should apply QALYs in different diseases, in different people, in different circumstances differentially, but the evidence and the science to back that is not available. Research into the area of QALY weighting has been initiated.

…and what are its strengths?

It is internationally established, as is the general realisation of the need to provide evidence on quality of life as part of any assessment of clinical effectiveness. The bottom line of what matters to patients is not just that the treatment works, but that they actually live longer and feel better. The QALY has brought that to the forefront and I believe that's what patients want and it's certainly what clinicians want for their patients.

Is there an official £30,000 per QALY threshold?

Clearly, the committee does not work to a single threshold – this would be inappropriate because it's difficult to substantiate other than on the basis of case law. More relevantly, if one expressed a specific cost effectiveness threshold, then drugs could be priced just below it and that would be the end of the process – why would we then need an appraisal?

However, there is a range of cost-effectiveness figures that are used in the process of evaluation of cost per QALY  over which the committee makes its value judgements. They are quite clear and are published in our methods guide.

If a drug's cost effectiveness is less than £20,000 per QALY, it is likely that this would be considered appropriate for the NHS as a rational use of resources. If it's between £20-30,000 per QALY, then increasing evidence to support it is required, and if it's above £30,000 per QALY, even more evidence is required and the judgements that need to be made around these decisions are expressed very clearly in our methods guide, which is published on the NICE website.

Has NICE become a de-facto rationing body?

NICE is not set up to ration healthcare, it is set up to ensure the most rational use of the available resources. You might think that means the same thing, but they are very different objectives. Assessing cost effectiveness is part of NICE's mandate to ensure the rational use of available resources. NICE is not responsible for deciding on the amount of resources available to the NHS or for making more resources available for healthcare.

In any case, I don't think rationing is the issue – some sections of the media find that this is the best way to describe what we do, but that description is misleading and inappropriate.

The DoH has said local commissioners of healthcare should not refuse to fund a drug on the basis of the absence of NICE guidance.

What horizon scanning work is NICE undertaking for the STA process?

NICE will not issue guidance until a marketing authorisation is in place. However, in order to provide guidance as soon as possible after a licence is granted, we would like to do as much work as we can on the valuation of the drug in advance of that point. That depends upon information provided by the company, for example what they predict the licence will say and when it's going to be granted. All of these things may be uncertain until close to the licensing date.

Has NICE been under pressure to achieve more with less funding over the last year?

I'm satisfied that we did what we proposed to do – namely set up a more rapid process. The new process was up and running very quickly and it showed the ability of NICE to respond to the needs of the NHS and the ability of the team to deliver.

What could the pharmaceutical industry do to improve the way it works with NICE?

The simple answer is to continue in the same way as it is – keep working with NICE, keep supporting the process and keep seeing its potential benefits to the NHS as a whole.

Professor Barnett chairs one of NICEs two appraisal committees. He is  Professor of Clinical Pharmacology at the University of Leicester Medical School and an honorary consultant physician at the Leicester Royal Infirmary with a special interest in cardiovascular disease.

Dominic Tyer is deputy editor of Pharmafocus.

E-mail: pharmafocus@wiley.co.uk