Rapid Responses to:

EDITORIALS: Paolo Bruzzi Non-drug industry funded research BMJ 2008; 336: 1-2 [Full text]

Rapid Responses: Submit a response to this article

Rapid Responses published:

What influences how research is reported in the BMJ?

Alison Rodgers, Max Watson consultant in palliative medicine Northern Ireland Hospice BT15 3LH   (6 January 2008)

"Clear winner" - depends on your viewpoint.

Hazel Thornton   (9 January 2008)

Structural efforts are being put in place to favour independent research, but more is needed

Research and Development Committee, Garattini S (Chair), Annunziato L, Bonini S, Bernardini R, Del Zompo M, Francavilla A, Garaci E, Liberati A, Rupolo G, Saccani Jotti G -- E mail: Ricerca&sviluppo@aifa.gov.it   (18 January 2008)

What influences how research is reported in the BMJ? 6 January 2008
Alison Rodgers, general practitioner Armagh BT61 7QG, Max Watson consultant in palliative medicine Northern Ireland Hospice BT15 3LH Send response to journal: Re: What influences how research is reported in the BMJ?	Bruzzi comments that improved disease-free survival in early breast cancer in post menopausal women is a “plausible but not validated surrogate end point for overall survival”1. It follows that, when available, actual data relating to overall survival rather than to surrogate end points should not be ignored. This point seems to have been missed by Minerva. Had Minerva, before writing her column, had the benefit of reading this week's issue - in particular the articles by Bruzzi 1 and Lenzer 2 she might have reported the 100 months of follow-up in the ATAC (Arimidex, tamoxifen alone or in combination) trial differently. Rather than “Anastrozole comes out as a clear winner” she might have informed us that "Anastrozole comes out as showing no effect for overall survival and no significant reduction in deaths after recurrence." 3 Had we not read this week's issue we might have accepted Minerva's interpretation instead of going to the original article and seeing that while the "clear winner" does well in the field both competitors (tamoxifen and anastrozole) cross one of the predetermined finishing lines (overall survival) together. We would be disappointed if our trainees’ interpretation of the data was as incomplete, or so closely resembled the sponsoring drug company’s press release.4 1. Bruzzi P. Non-drug industry funded research. BMJ 2008;336(7634):1- 2. 2. Lenzer J, Brownlee S. Doctor takes "march of shame" to atone for drug company payments. BMJ 2008;336(7634):20-21. 3. ATAC. Effect of anastrozole and tamoxifen as adjuvant treatment for early- stage breast cancer: 100-month analysis of the ATAC trial. Lancet Oncology 2008;9:45:53. 4. http://www.atac100.com/29208_31521_0_0_0.aspx?mid=74. Competing interests: None declared
"Clear winner" - depends on your viewpoint. 9 January 2008
Hazel Thornton, Honorary Visiting Fellow, Department of Health Sceinces, University of Leicester. "Saionara", 31 Regent Street, Rowhedge, Colchester, CO5 7EA Send response to journal: Re: "Clear winner" - depends on your viewpoint.	The success of exceptionally rapid accrual of over 9,000 participants to the ATAC trial, coupled with its predetermined event point requirement for triggering an interim analysis, which resulted in closure of the combination arm, led to numerous premature triumphant claims of superiority of anastrozole over tamoxifen when the report was published in 2002, at an average time of only 2.5 years of trial treatment. This triumphalism has not stopped now that the 100-month analysis report of this trial has been published.[1] Bruzzi, however, has drawn attention to the impediments to ever being able to provide reliable information on the long-term efficacy and safety of aromatase inhibitors (AIs), partly due to the crossover of the control group in all but one of the four studies comparing AIs with Tamoxifen, in total involving 20,000 women, but also because of the effects of early publication of interim analyses. He expresses concern that AIs are being used to treat early breast cancer on the basis of inadequate information. [2] But more worryingly, the AI anastrozole is also being used in the IBIS II prevention trial (target number 6,000 healthy but higher risk women) and IBIS II DCIS trial (target number: 4,000 women). Also, in spite of the fact that no significant effect was seen in overall survival, Minerva states that `anastrozole (arimidex) comes out as a clear winner over tamoxifen for treating postmenopausal women with breast cancer who are hormone sensitive (http://oncology.thelancet.com 15 December 2007). [3] The report`s similar interpretation is that the data establish [1] “clearly the long-term efficacy of anastrozole compared with tamoxifen as initial adjuvant treatment for postmenopausal women with hormone- sensitive, early breast cancer.” Serious concerns were expressed in 2002 about drawing premature conclusions; about the effect of anastrozole on cognitive function; and the use of this drug in healthy women within prevention trials. But there is no mention of effects on cognition in the report, [1] although in response to published questions about anastrozole for early breast cancer in The Lancet in 2002, [4] readers were advised that a pilot study had been set up within the study to measure this, and prospective studies were planned. [5] The recommendation was made that this endpoint should be assessed in all future studies of new agents. Patients and clinicians need to be given all the facts in an unbiased manner so that they can make individual choices and shared decisions according to their own health profiles and needs. `Interpretation` that `clearly` favours one drug over another when `no significant effect on survival` was found, requires robust challenging. REFERENCES: [1] The Arimidex, Tamoxifen, Alone or in Combination (ATAC) Trialists` Group. Effect of anastrozole and tamoxifen as adjuvant treatment for early -stage breast cancer: 100-month analysis of the ATAC trial. Lancet Oncology 2008: 9:45-53. [2] Bruzzi P. Non-drug industry funded research. Poor access to drugs leaves important clinical questions unanswered. BMJ 2008; 336:1-2 [3] Minerva. BMJ 2008; 336:52 [4] Roberto Stazi, Edmondo Terzoli; Hazel Thornton. Questions about anastrozole for ealy breast cancer. Lancet 2002; 360:1890-1 [5] Michael Baum on behalf of the ATAC Steering Committee. Author`s reply: Questions about anastrozole for early breast cancer. Lancet 2002; 360:1890-91 Competing interests: None declared
Structural efforts are being put in place to favour independent research, but more is needed 18 January 2008
Research and Development Committee, Italian Drug Agency AIFA (Agenzia Italiana del Farmaco), Roma, Italy, Garattini S (Chair), Annunziato L, Bonini S, Bernardini R, Del Zompo M, Francavilla A, Garaci E, Liberati A, Rupolo G, Saccani Jotti G -- E mail: Ricerca&sviluppo@aifa.gov.it Send response to journal: Re: Structural efforts are being put in place to favour independent research, but more is needed	Berendt et al (1), as well as Bruzzi’s accompanying editorial in the January 5th issue of the BMJ (2) provide fuel the current debate about independent clinical research and the ability of governments and public health agencies to support this research. We would like to inform international readers of two recently implemented initiatives in Italy. First, a regulation intended to support independent clinical research on drugs was adopted. Among the incentives, the regulation foresees - for projects conceived and conducted by independent researchers – the coverage by the National Heath Service. of the costs of marketed drugs, even if used outside approved indications. Secondly, within the law that created the Italian Medicines Agency (AIFA), pharmaceutical companies are required to allocate 5% of their promotional expenditure to a fund for independent research. Finally, a national mechanism for funding independent research on drugs was created and three consecutive bids (for year 2005, 2006 and 2007) have been launched on the following three main areas of interest: orphan drugs for rare diseases and drugs for non-responders; comparative studies on drugs (including head to head comparison of therapeutic strategies); and appropriateness of drug use, pharmacovigilance and outcome research (3) Through a two-phase selection process (based on triage of letters of intent and full independent and external peer review of full protocols for those deemed at highest priority), 105 studies have approved in the first two years (54 in 2005 and 51 in 2006, out of over 400 applications each year), with a funding of 36 and 30 million euro for 2005 and 2006, respectively.. As an example of the limitations of commercial studies, Bruzzi (2) mentioned the commercialisation of aromatase inhibitors in the treatment of early breast cancer, where the approval was mainly based on surrogate end points (disease-free survival), instead of relying on long-term efficacy (overall survival) and safety. Two examples of head to head comparative trials funded through the program are worthy mentioning. The first is an Rct aimed at head-to-head comparison of the long-term (9 years) benefit-risk profile of 3 different aromatase inhibitors for the treatment of early breast cancer in a population of 10,000 women; the study also includes a comparison between up-front and sequential strategies (aromatase inhibitors vs tamoxifen plus aromatase inhibitors). The second is a head to head comparison of interferon to a generic azathioprine in modifying disease activity in patients with relapsing- remitting multiple sclerosis. Both examples – a long term comparison of different drugs and different strategies, together with a comparison including a generic drug – clearly indicate studies that possess only very limited commercial appeal, while are of great public health and patient oriented relevance, besides bearing on the sustainability of health care systems. Different activities should be put in place to promote independent research going from raising-awareness-campaigns to the improvement of research to structural support to those engaged in this research - as, for example, supporting units such as those mentioned by Barendt for Denmark (1) - to the allocation of dedicated research funds as in the AIFA’s example that we referred to. A joint effort at the European level, even within the unavoidable budget constraints, may have an extraordinary impact for independent research, ensuring that patients needs and priorities will be seriously taken into consideration. References 1 Berend L, Hakansson C, Bach FK et al. Effect of the European Trials Directive on academic drug trials in Denmark: retrospective study of applications to the Danish Medicies Agency. BMJ 2008; 336:33-35 2. Bruzzi P. Non drug industry funded research. BMJ 2008; 336: 1-2 3. Breckenridge A. Post marketing strategies of Medicine Clinical Pharmacology & Therapeutics 2008; 83 (1) E mail Ricerca&sviluppo@aifa.gov.it Competing interests: None declared