Letters

Efficacy, safety, and cost of new anticancer drugs

Pessimistic conclusion was not justified

Price needs to be evaluated against effectiveness

Authors' reply

Pessimistic conclusion was not justified

EDITORGarattini and Bertele' conclude that new anticancer drugs introduced during the past six years have not increased survival or quality of life in cancer patients.¹ This is a surprising conclusion to be made at the end of a decade during which unprecedented advances have been made in the treatment of common tumours, in both survival and quality of life. Perhaps if they had examined the contemporary literature and taken account of oncology practice they might have reached a less pessimistic conclusion.

The authors rely heavily for their negative general comments on the value of cancer chemotherapy and chemotherapy trials on the population mortality based study of Bailar and Gornik, which used data collected between 1970 and 1994, pre-dating by many years the period being considered here.

Most of the drugs examined have been evaluated extensively in randomised trials, many with evaluations of quality of life. For example, the taxanes significantly enhance survival in both adjuvant treatment and treating metastatic breast cancer, as well as in lung cancer and ovarian cancer. ^{2 3} Recent results from randomised trials of irinotecan and oxaliplatin (not considered by this article) show extended survival in colon cancer, underlining the significant progress that is being made in the chemotherapy of common cancers.⁴

Although cost considerations are important, they should not affect our judgment of the utility of a new agent. The drugs chosen for cost comparison are mostly inappropriate because they are either not indicated or ineffective in the indication concerned. For example, taxanes are normally used in breast cancer that is refractory to hormones or negative for oestrogen receptors, where the use of tamoxifen would be irrelevant. The comparison of costs of oral versus intravenous treatment (temozolomide and capecitabine) ignores the additional costs of intravenous administration. Giving the oral drug may have economic advantages and be the preference of the patient.⁵ The ephemeral nature of drug pricing as a result of competition by generics or other drugs in the same class has been overlooked in the cost calculations.

The oncology community now routinely performs large international trials with survival and end points of quality of life. It is naive, if not disingenuous, to criticise current drugs on the basis of data available on the EMEA website alone. We all hope that the outlook for future cancer patients will improve further, and that the advent of new targeted agents will contribute to that improvement. Garattini and Bertele' have done these patients no favours.

Hilary Calvert, professor of medical oncology. 
University of Newcastle upon Tyne, Newcastle upon Tyne NE2 4HH hilary.calvert{at}newcastle.ac.uk

Duncan I Jodrell, reader and honorary consultant in medical oncology. 
Western General Hospital, Edinburgh EH4 2XU

James Cassidy, professor of medical oncology. 
Cancer Research UK Department of Medical Oncology, Glasgow G61 1BD

Adrian L Harris, professor of medical oncology. 
Cancer Research UK Medical Oncology Unit, John Radcliffe Hospital, Oxford OX3 9DS

Written with the support of Professor Robert Souhami, director of clinical research, development, and training, Cancer Research UK, London WC2A 3PX.

Competing interests: HC has been reimbursed for attending conferences and has received fees for speaking or consultancy from Bristol-Myers Squibb, GlaxoSmithKline, and Schering-Plough in the past five years. DIJ has received reimbursement for clinical trials relating to the development of capecitabine (Roche) and is a member of the editorial board of the Xeloda website. He has also received support to attend an international conference from Sanofi-Synthelabo, manufacturers of oxaliplatin. JC has received speaker and consultancy fees from major pharmaceutical companies involved in the treatment of colorectal cancer. ALH has received honoraria and research funding from several companies manufacturing anticancer drugs.

Price needs to be evaluated against effectiveness

EDITORGarattini and Bertele' addressed the problem of the cost of the newest anticancer agents.¹ Many innovative anticancer agents have recently become available in Italy, and all have been approved for nationwide reimbursement. The high cost of these agents makes evaluating their price against their effectiveness worth while.

We conducted an analysis on these new anticancer agents, comparing the price approved by Italy's ministry of health with a reference price range determined with a pharmacoeconomic algorithm.² The algorithm values each month of life gained according to current international standards (from 1000 to 5000 ($1010-5050; £633-3165) per month of life gained) and incorporates reductions in hospital stay. We applied it retrospectively to the innovative anticancer agents submitted to the Italian health ministry from February 1999 to August 2001.

Most of the prices approved by the negotiation committee were higher than the reference window given by the pharmacoeconomic algorithm (figure). Hence, at least in Italy, modern innovative anticancer agents are generally assigned a higher price than that suggested by current pharmacoeconomic standards.

View larger version (22K): [in this window] [in a new window]   Innovative anticancer agents approved in Italy: comparison between price range or "negotiation window" proposed by pharmacoeconomic algorithm (horizontal line) and final price approved by the negotiation committee (vertical circle)

In oncology relating the drug price to the survival gain is the most reasonable way to handle the negotiation problems raised by these drugs. Of course, improvements in quality of life could be another important component of clinical effectiveness (and of the consequent price calculations), but we believe that the pharmacoeconomic methods available are not sufficiently mature to propose specific algorithms.

Our findings are likely to be similar to those in other European countries because variations between countries are becoming smaller owing to the multinational marketing strategies of most drug manufacturers (particularly for anticancer treatments). In our analysis no cases were valued at an unreasonably high level of cost effectiveness. In contrast, these exceedingly high levels of (incremental) cost in comparison with (incremental) effectiveness have occasionally been documented in disciplines other than oncologyfor example, preoperative autologous blood donations and interferon beta-1b in secondary progressive multiple sclerosis. ^{3 4}

Andrea Messori, coordinator. 
Sabrina Trippoli, researcher. 
Monica Vaiani, researcher. 
Laboratorio SIFO di Farmacoeconomia, c/o Drug Information Centre, Azienda Ospedaliera Careggi, Viale Morgagni 85, I-50134 Florence, Italy md3439{at}mclink.it

Competing interests: AM and ST have been paid by Eli Lilly and Pfizer for running educational workshops and have been participated in research projects funded by GlaxoWellcome, Sigma-tau, Aventis, and Eli Lilly. MV as acted as a consultant to GlaxoWellcome.

Authors' reply

EDITORContrary to what Calvert et al say, our article did not imply that there has been no gain for patients with cancer during the past six years. It aimed at showing the evidence available when a new drug is approved by relying on the documentation accompanying EMEA approval, while not overlooking the current literature. The fact that we are usually dealing with few patients, phase II trials, open label studies, and lack of comparison makes it difficult to establish how to position new anticancer drugs in the existing therapeutic arsenal.

The taxane issue Calvert et al raised is not reflected in the paper mentioned, which concludes that the impact of these drugs on the natural course of breast cancer has still to be defined.¹ The combination of cisplatin and paclitaxel for advanced ovarian cancer was not validated by studies showing that cisplatin is responsible for the efficacy of the combination. ^{2 3} Irinotecan and oxaliplatin are not even mentioned in the paper supposed to support their merit.⁴

As for the costs, we agree with Calvert et al that direct comparison of tamoxifen and chemotherapy is not pertinent. The table in our article, however, also allowed a comparison of the costs of docetaxel and paclitaxel. For temozolomide it is difficult to accept that its cost (2143 ($2165; £1356) per cycle) is compensated by the avoidance of intravenous injections, since oral formulations of procarbazine and capecitabine are available too. In any case, why should the whole potential saving for the NHS be routed back to industry rather than being invested for better services, supportive care, and independent research? As for the alleged economic advantages, most of the economic analyses of new drugs used in oncology have been criticised on the basis of conflict of interests of the authors.⁵

The analysis by Messori et al supports our view that innovative treatments in oncology are too expensive. Their cost effectiveness may be not "unreasonable," as in other areas, but this depends on how reasonable the assumptions of pharmacoeconomic models used actually are.

We all hope that the future will be brighter, but we must be honest in our promises to patients and objective in information to doctors. Any undue burden for drug expenditure will always be at the expense of other measures that might contribute to better care. Calvert et al may not endorse it, but this is the favour we believe has to be paid to patients with cancer and their families.

Silvio Garattini, director. 
garattini{at}marionegri.it

Vittorio Bertele', head, regulatory policy laboratory. 
Mario Negri Institute for Pharmacological Research, via Eritrea 62, 20157 Milan, Italy

Competing interests: In the past five years VB has received fees for speaking from Schwarz Italia SpA and Instrumentation Laboratory, and for scientific advice from SmithKlineBeecham and Aventis Pharma.