Cancer Drugs Fund requires further reformBMJ 2016; 354 doi: https://doi.org/10.1136/bmj.i5090 (Published 27 September 2016) Cite this as: BMJ 2016;354:i5090
All rapid responses
Although we welcome the authors’ creative ideas on generating high-quality evidence to inform NICE’s appraisal of cancer treatments, the article does not do the current arrangements justice, and is wrong to imply that the Cancer Drugs Fund (CDF) exists only to support observational data collection.
NHS England’s new approach to the appraisal and funding of cancer drugs in England, which came into force on 29 July 2016, states its aims are to provide patients with faster access to cancer treatments, drive stronger value for money for taxpayers and to offer a new fast-track route to NHS funding to pharmaceutical companies with promising new cancer treatments, providing they are priced responsibly.1 With the pharmaceutical industry playing its part through fair pricing, patients can access NICE-recommended treatments up to 8 months sooner than if the new CDF didn’t exist.
Despite the authors’ scepticism, the novel fast-track route to NHS funding through the CDF will offer what they correctly identify as ‘an excellent opportunity to generate evidence on the effectiveness of new cancer drugs’. It is incorrect to suggest that the new CDF process is based on an overreliance on collecting ‘real-world’ data, although this data collection route is an option in the new arrangements. As described in NICE’s specification for CDF data collection arrangements,2 ongoing or new clinical studies, which would of course encompass randomised controlled trials, are one of two main options for collecting data on drugs that are recommended for use within the CDF. The other is the Systemic Anti-cancer Therapy (SACT) dataset, which is not necessarily a standalone option and is potentially complementary to clinical studies.
When a treatment is recommended for use within the CDF, a formal data collection arrangement will be drawn up as part of a managed access agreement.1,2 It will outline the key areas of uncertainty and explain how these will be addressed through data collection. Longer-term overall survival data from an ongoing randomised clinical trial can be taken into account, reducing the uncertainty associated with the extrapolations used to inform the cost-effectiveness estimates.
The other key component of the managed access approach is a commercial access agreement, which will determine the level of reimbursement during the data collection period. Companies will receive less for the drug in the CDF than they would if had it been recommended for routine commissioning. This will reflect the decision uncertainty: greater uncertainty will result in a lower level of reimbursement during the data collection period. To limit the financial impact on taxpayers, the CDF budget will be tightly controlled under the new financial control mechanisms put in place by NHS England. In addition, a proportional rebate will be paid by all pharmaceutical companies if the overall CDF allocation is overspending at the financial year end.
Concerns about NICE’s rigour in its decision-making are also misplaced. Any marketing authorisation application by a company will still be rigorously assessed by the medicines regulators so it is unthinkable that randomised clinical trials will cease to underpin NICE’s recommendations on new drugs. The CDF gives patients a mechanism for more rapid access to the most promising treatments at a time when clinical uncertainties remain about their longer-term effectiveness. These include drugs with a conditional marketing authorisation or involved in the Medicines and Healthcare products Regulatory Agency Early Access to Medicines Scheme, which have promise, but come to market with a limited evidence base when they are initially appraised by NICE. For these medicines, it is even more likely that data collection through ongoing randomised clinical trials used to obtain regulatory approval will play a leading role in the CDF mechanisms designed to fill the evidence gaps.
The limitations associated with types of evidence other than randomised controlled trials, including observational studies, have long been recognised by NICE and others, and are described in a technical support document on using real-world data in NICE decision-making and NICE’s guide to methods of technology appraisal.3,4 The relevant sections of the guide remain unchanged following the publication of the addendum5 to our methods and process that supports the appraisal of cancer treatments using the new CDF process. Regardless of whether the evidence is from clinical trials, observational data collection or a combination of the two, it is the degree to which the evidence is robust, of adequate quality and fit for purpose to address the decision question that drives NICE’s use of the information and the degree of scrutiny we apply in its appraisal.
The prospect of collecting additional randomised controlled data within the NHS is an exciting one but, as noted by the authors in their article, is fraught with difficulties including funding, ethics, clinical equipoise, and clinician and patient willingness to participate. When a treatment is recommended for use in the CDF, the NICE managed access approach offers both a robust and pragmatic solution to the dilemma of how to make promising treatments available to NHS patients and at a price that reflects the clinical uncertainty.
1. NHS England Cancer Drugs Fund Team. Appraisal and funding of cancer drugs from July 2016 (including the new Cancer Drugs Fund): a new deal for patients, taxpayers and industry. Published Jul 2016 [cited 2016 Oct 6]. Available from https://www.england.nhs.uk/wp-content/uploads/2013/04/cdf-sop.pdf.
2. National Institute for Health and Care Excellence. Specification for CDF data collection arrangements. Published Jul 2016 [cited 2016 Oct 6]. Available from https://www.nice.org.uk/Media/Default/About/what-we-do/NICE-guidance/NIC....
3. Bell H, Wailoo AJ, Hernandez M, et al. The use of real world data for the estimation of treatment effects in NICE decision making. Published Oct 2016 [cited 2016 Oct 6]. Available from http://www.nicedsu.org.uk/RWD%20DSU%20REPORT%20Final%2017JUNE2016.pdf.
4. National Institute for Health and Care Excellence. Guide to methods of technology appraisal. Published Apr 2013 [cited 2016 Oct 6]. Available from https://www.nice.org.uk/process/pmg9/chapter/foreword.
5. National Institute for Health and Care Excellence. PMG19 Addendum A - Final amendments to the NICE technology appraisal processes and methods guides to support the proposed new Cancer Drugs Fund arrangements. Published Mar 2016 [cited 2016 Oct 6]. Available at https://www.nice.org.uk/Media/Default/About/what-we-do/NICE-guidance/NIC....
Competing interests: All of the authors are employed by the National Institute for Health and Care Excellence.
The proposal from Grieve et al. to allocate CDF funding to randomised studies in place of the collection of observational data is one with which I agree, though for different reasons to the authors.
Whilst the collection of observational data (the new CDF) may be helpful in some cases, in the vast majority of instances this will not address they key uncertainties in health economic models - which in turn determine whether a drug is accepted for use. On the whole these uncertainties relate to the long term outcomes (such as survival at 5 or 10 years). With observational data collection starting from the point of a NICE decision (and then time to enrol patients), this data is unlikely to arrive before the patent on drugs expires, or it is replaced by an even more effective treatment.
This illustrates the main issue with the revised fund; it solves the wrong problem. At the point of NICE assessment (having worked on over 30), we generally have very good information on how the drug will perform over the first couple of years (what the CDF will collect), and more data here has minimal value. What we often lack however is information on the comparative effectiveness of the drug (as pointed out by Grieve et al.), and the long term outcomes with the drug (which would be better collected in extension studies to the original clinical trials).
It may be a useful political sticking plaster to defer difficult decisions, but the new CDF will rarely aid decision making, and therefore is unlikely to improve the health of the population. As implied by Grieve et al. it should therefore be regarded as a missed opportunity.
Competing interests: No competing interests
I have no argument with the proposals in this article, but would wish to draw attention to another "real world " problem for Clinicians trying to manage Oncology budgets to the best of their ability, namely new indicators for old drugs.
Emotive and Headline grabbing decisions on new drugs distract from the business of getting value for money, etc.
But if the indication for an established regime of adjuvant therapy changed, say, from women aged 55 and over to 45 and over, it's easy to imagine how this might slip under the radar of new drug expenditure, until the inevitable overspend at ground floor level is discovered.
Ownership of that problem may be hard to pin down, and inevitably, more pressure applied to the introduction of truly new drugs.
Competing interests: No competing interests
I wholeheartedly agree with Grieve et al, that the best way to collect data as well as assess cost-effectiveness of cancer drugs approved for subsidy by the Cancer Drugs Fund is to incorporate the drug usage into timely randomised controlled trials within routinely collected data sources, to establish which drugs are relatively effective.
This will also ensure that users of these drugs are aware that the drugs are under clinical assessment and may be withdrawn if the drug is not found to be improve patient outcome.
However, the professors involved will have to ensure timeliness in the design and implementation of these trials. They should be reminded that the Cancer Drugs Fund was created by politicians based on populist policies to address lobby groups on a very emotionally charged and physically exhausting /involved group of diagnoses. Any hint of prolonged delay in the release of subsidised drugs while researchers design and develop a clinical trial will result in outcries from the community and lobby groups and such plans may end up being bypassed if the political will bends to vox populi.
Thus it is important that the clinicians involved are prompt in trial development to put appropriate studies in place to appease the public.
Competing interests: No competing interests