This Article Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Drummond, M. F. Articles by Mason, A. R. Search for Related Content PubMed Articles by Drummond, M. F. Articles by Mason, A. R. Social Bookmarking                            What's this?

In Reply

Centre for Health Economics, University of York, United Kingdom

In his comment on our article,¹ Low questions our assertion that cancer drugs reviewed by the National Institute for Health and Clinical Excellence (NICE) in the United Kingdom have "fared quite well, with most recommendations being positive."

He bases his opinion on more recent data than those contained in our article. Specifically, of the 11 NICE recommendations for cancer medications published after our observations and the further 11 technology appraisals (TAs) currently in development, more than one half (13 of 22) have reached a negative decision.

Comparison of Low's data with ours is not straightforward. Our denominator was the number of technology appraisals, whereas his was the number of medications considered. Some TAs consider more than one medication. In addition, Low considers TAs in development. We excluded these on the grounds that, since the process was not yet complete, there was a chance that the recommendations may change, based on stakeholder comments or an appeal.

From June 2006 to March 2007, NICE published nine new appraisals of cancer drug therapies, covering prostate cancer (1), breast cancer (5), follicular lymphoma (1), leukemia (1), and colorectal cancer (1). Table 1 summarizes and compares recent appraisal decisions with those from our previous findings (May 2000 to June 2006).

There are two possible explanations. First, NICE may have changed its procedures for assessment, or the basis for making its decisions. It is difficult to study this from the outside, but one change since our original study is the advent of single technology appraisals (STAs). The procedure for NICE's STAs is slightly different from the original technology appraisals (now called multiple technology appraisals, or MTAs). The main differences are that the process is quicker and the evidence base relies mainly on a submission from the manufacturer, rather than on a detailed evaluation by an independent review team. Of the published (and unpublished) assessments considered by Low, six (four) were STAs, with two (four) being negative.

The other explanation for more negative assessments could be that the medications being appraised are less cost effective. A particularly important trend identified by Low is that many of the more recent medications reviewed by NICE are for rare cancers. It could be that, because the research and development costs have to be recouped from sales of the product to a smaller patient group, the average cost per patient is higher. If this were indeed the reason for more negative appraisals, the future for expensive cancer therapies does not look very bright. In particular, expensive targeted therapies may struggle to obtain reimbursement.

With this in mind, we undertook a more detailed analysis of the negative appraisals published since our original article. We considered only completed appraisals, for the aforementioned reasons, but the same analysis could be applied to the appraisals in development identified by Low. The three appraisals comprised paclitaxel for early node-positive breast cancer (108),² bevacizumab and cetuximab for metastatic colorectal cancer (118),³ and fludarabine for lymphocytic leukemia (119).⁴ Appraisals 108 and 119 were STAs.

In both negative STAs, problems with the manufacturer's model underpinned the NICE recommendation. The key reason for the negative recommendation in appraisal 108 was that the manufacturer's analysis failed to adequately frame the decision problem. Criticisms of the model included the lack of any systematic review to inform effectiveness and cost-effectiveness estimates for paclitaxel, inadequate consideration of toxicities, and the choice of comparator (four cycles of anthracycline, doxorubicin, and cyclophosphamide), which was a regimen not commonly used in the NHS and considered to be less effective than usual National Health Service (NHS) care. Appraisal 119 highlighted a number of flaws in the manufacturer's model. These included the exclusion of the costs of treating adverse events, uncertainty about re-treatment response rates, and strong assumptions about quality of life, the risk of disease progression, and death. The NICE committee considered that the manufacturer's estimates of cost effectiveness were likely to be overly optimistic and that further evidence was unlikely to alter this position. Fludarabine was therefore considered unlikely to be good value for money for the NHS.

Appraisal 118 examined two drugs for colorectal cancer. Despite a range of modeling approaches and sensitivity analyses to explore the impact of key assumptions, neither the model by the manufacturers nor that by the independent assessment group provided convincing evidence that the drugs represented a cost-effective use of NHS resources. At best, there was a .24 probability that one of the drugs (bevacizumab) was cost effective and, at worst, a zero likelihood. Analyses of cetuximab found (at best) a .1 likelihood of cost effectiveness and at worst, the cost per quality-adjusted life year could be as high as £370,000.

It is difficult to draw any firm conclusions at this stage, and it would be useful to revisit this question when more appraisals have been concluded. Certainly, the two drugs examined in the MTA were not found to be very cost effective. However, in both the negative STAs, problems were identified with the manufacturer's model. Under the original MTA approach, it is possible (although not certain) that these methodologic concerns, and their implications for cost effectiveness, would have been addressed by the independent evaluation team. It is also possible that the short time frame allowed for STAs may help explain inadequacies in the manufacturers' models. Time pressures may make it more difficult to ensure the structure of the model is appropriate and that the best available evidence is used to populate the model. As manufacturers acclimatize to the new framework, they may start to prepare their analyses in advance of a formal NICE request. This will be important if their drugs are to meet NICE's standards for demonstrating cost effectiveness and to receive a positive recommendation.

AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST

The author(s) indicated no potential conflicts of interest.

REFERENCE

1. Drummond MF, Mason AR: European perspective on the costs and cost-effectiveness of cancer therapies. J Clin Oncol 25:191-195, 2007[Abstract/Free Full Text]

2. NICE: Paclitaxel for the adjuvant treatment of early node-positive breast cancer. TA108, September 2006. http://guidance.nice.org.uk/TA108

3. NICE: Bevacizumab and cetuximab for the treatment of metastatic colorectal cancer. TA118, January 2007. http://guidance.nice.org.uk/TA118

4. NICE: Fludarabine monotherapy for the first-line treatment of chronic lymphocytic leukaemia. TA119, February 2007. http://guidance.nice.org.uk/TA119

CiteULike    Complore    Connotea    Del.icio.us    Digg    Facebook    Reddit    Technorati    Twitter    What's this?

This Article Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Drummond, M. F. Articles by Mason, A. R. Search for Related Content PubMed Articles by Drummond, M. F. Articles by Mason, A. R. Social Bookmarking                            What's this?