Author(s): Vulink NC, Denys D, Fluitman SB, Meinardi JC, Westenberg HG
Affiliation(s): Rudolf Magnus Institute of Neuroscience, Department of Psychiatry, University Medical Centre Utrecht, Utrecht, The Netherlands. N. C.C. vulink@umcutrecht. nl
OBJECTIVE: To assess the efficacy of quetiapine addition to citalopram in treatment-naive or medication-free obsessive-compulsive disorder (OCD) patients. METHOD: Seventy-six patients who met DSM-IV criteria for OCD and who were drug-free or drug-naive at entry were randomly assigned in a 10-week, double-blind trial with citalopram (60 mg/day) plus quetiapine (300-450 mg/day) or placebo; treatment-refractory OCD patients were excluded. Of the 76 eligible patients, 66 patients completed the trial-31 in the quetiapine and 35 in the placebo group. The change from baseline to endpoint on the total Yale-Brown Obsessive Compulsive Scale (YBOCS) and the response to treatment in the quetiapine addition compared with the placebo addition group were the primary outcome measures. Response was defined as a 35% or greater reduction on the YBOCS and a Clinical Global Impressions-Improvement (CGI-I) score at endpoint of 1 or 2. The study was conducted from November 2003 to June 2005 at the University Medical Centre Utrecht, The Netherlands. RESULTS: As measured by the mean reduction in YBOCS scores following an intent-to-treat, last-observation-carried-forward analysis, quetiapine addition (11.9) was significantly superior to placebo (7.8; p = .009). Quetiapine addition was also significantly superior to placebo on the CGI-I scale, with a mean +/- SD CGI-I score of 2.1 +/- 1.3 versus 1.4 +/- 1.2, respectively (p = .023). Quetiapine addition (N = 22, 69%) was also associated with a significantly greater number of patients responding to treatment compared with placebo addition (N = 15, 41%; p = .019). More patients receiving quetiapine (N = 8) than placebo (N = 2; NS) discontinued treatment due to adverse events. CONCLUSIONS: The combination of quetiapine and citalopram was more effective than citalopram alone in reducing OCD symptoms in treatment-naive or medication-free OCD patients. TRIAL REGISTRATION: www. trialregister. nl Identifier NTR116.
Placebo or active control? Either, as long as it is in the patient's interest*
* Authors: Silvio Garattini, Vittorio Bertele, Luca LiBassi, "Mario Negri" Institute for Pharmacological Research, Regulatory Policies Laboratory, Milan, Italy
The appropriate use of placebo in controlled clinical trials is still debated. This article briefly reviews different conditions of applicability of the placebo dilemma in the light of patients' interests and needs. We also test our assumptions that (a) what should justify the authorization of new drugs is their proven superiority over the best available comparators; (b) superiority to placebo is only clinically meaningful in patients who cannot benefit from the standard active comparator; (c) non-inferiority of new drugs compared to available treatments can only be assessed when additional benefits over active controls are foreseeable and adequately tested. Ethical committees should advocate these principles in defence of patients' interests. Public funds should support independent clinical research, which would help address questions of high priority for public health, but be of no commercial appeal for industry
The discussion raised by the latest version of the Declaration of Helsinki (1) about the legitimacy of using placebo in controlled clinical trials is still alive. Temple and Ellenberg (2) state that the acceptability of placebo depends on whether the patient will be harmed by deferral of effective treatment. Emanuel and Miller (3) advocate a middle-ground alternative, which allows the use of placebo when no therapy is available for a given indication or when the lack of benefit (not necessarily harm) or the discomfort to patients receiving placebo is negligible. Lewis et al (4) argue that placebo-controlled trials remain the only means of assessing the efficacy of new medicines, whose potential advantages over recognised alternatives lie in areas other than efficacy.
The debate focuses on patients' well-being as the only aim of clinical investigation. Life, however, is different and the placebo or active-control dilemma can hardly be resolved in an environment essentially driven by the strong economic interests of the pharmaceutical industry where patients' needs and interests are not the only sources of clinical research hypotheses. In our opinion what matters is not so much the methodology as the aim and independent conduct of trials. Once identified, any clinical question that is truly relevant for patients and independently pursued would in itself imply the best way to be addressed. When the questions identified have only relative importance for patients, any methodological discussion on how best to answer them is an exercise of limited value.
Use of placebo not targeted to patients' needs
In order to maintain its level of profitability, the drug industry has to contain its risks and costs, and increase its revenues and market. One way to limit risks and costs is to use placebo trials, whenever possible, in order to obtain a slice of an existing market rather than to risk an expensive failure trying to prove an advantage over active comparators. Most placebo-control trials are done not because there are no alternatives, but because it is easier to show an effect and therefore to claim efficacy to the regulatory authorities.
The strategy is to position a new product so that its superiority can be suggested with advertising and incentives to doctors and pharmacists without having to prove it scientifically. A drug that is scientifically proven to be superior to its comparator is unlikely to need such efforts to sustain its sales and convince doctors to prescribe it. But efforts are certainly needed with most copies, which are by definition very similar: for most ACE-inhibitors, sartans, antidepressive agents or nonsteroidal anti-inflammatory drugs it is very difficult to make a choice on the basis of patients' needs (5), as these products were not developed to provide a better treatment in the first place.
Clearly, me-too drugs aim at finding a place in the market rather than in therapy. Therefore, the whole issue of clinical trials requires discussion of a fundamental question that is preliminary to any others about study design and controls: "Is it ethical to submit patients to clinical trials whose sole purpose is clearly to gain access to the market?"
Placebo is sometimes not used when it should be
Placebo-controlled studies in depressive or hypertensive patients resistant or intolerant to available therapies would be appropriate, as they would respond to a real need; patients resistant to current therapy would lose out from the lack of these studies. But in the eyes of drug companies this approach would mean gaining approval for a restricted indication and, consequently, a limited market and profits. This is why such trials are not usually done. That the use of placebo in these circumstances is not a satisfactory solution for pharmaceutical companies is demonstrated by the fact that often they do not avoid comparison with active controls, but only test the equivalence or - more often nowadays - the non-inferiority of new drugs with respect to the available comparators. Non-inferiority studies raise two major questions: one methodological, one substantial.
Placebo does not add to non-inferiority tests
The arguments raised (2 - 4) on the methodological deficiencies and inconsistent messages of non-inferiority trials are fully endorsed (6 - 8). What we object to here is that inclusion of a placebo arm would overcome the potential lack of sensitivity of this kind of study (2, 4). As was the case with trials for depression (9), the inclusion of placebo may show whether the test and control treatments are truly effective. However, even so, the claimed similarity of the test drug and the active comparator would remain an artefact: because of the weak power of the study in view of the small number of patients, this similarity may in fact hide important clinical differences (7 - 10).
According to pre-specified definitions, non-inferiority limits include an excess of outcome events associated with the test treatment. Two, five, or ten more deaths, strokes, fractures, interventions every hundred treated patients may not be considered enough to mark a difference between the new and the control drug, thus permitting the conclusion of non-inferiority of the former. A poorer outcome with the new drug than with current available treatment is not acceptable, even if better than in a concurrent placebo arm. Poor sensitivity is intrinsic to non-inferiority trials, which deliberately aim at overlooking differences rather than highlighting them.
As with superiority over placebo, non-inferiority to active comparators might also allow onto the market drugs that in fact are less active (or safe, tolerable, convenient, etc.) than those already available, usually with consolidated properties and a lower cost. Moreover, these approaches do not meet patients' or physicians' needs to define the place in therapy and respective roles of new and available treatments.
Active controls are not used as they should be
The solution of a methodological problem (sensitivity of non-inferiority trials) does not in any event dignify a hypothesis (non-inferiority of the test drug) that has no or little importance for patients. Besides sensitivity, the main problem with these trials is that they lack ethics (8). Here the point is not even about placebo or active control. Simply, these studies deliberately disregard patients' interests in favour of commercial ones. Non-inferiority studies do not provide any possible advantage to patients. Like placebo-controlled studies, they aim at claiming efficacy, and possibly additional advantages, without providing proof.
These trials only have an economic interest and we believe few patients would agree to participate if the industrial sponsor's message were clearly conveyed in the "informed consent" as follows: "I want to recruit a number of patients and let chance decide whether they should go on taking the effective treatment they are currently given, or try my new drug, which is not expected to be any better. To me it is enough to establish that my drug is equivalent to or not worse than the other one". It is surprising that these trials obtain clearance by Ethical Committees in the absence of other advantages.
The excuse for carrying out these trials is usually that physicians need several alternatives because not all patients respond the same way. But again, if this is the case, why not use placebo-controlled trials in patients not adequately responding to other treatments? In contrast, just as it does not solve methodological problems, the inclusion of a placebo arm in non-inferiority trials is even less likely to solve these ethical problems.
Placebo does not help distinguish true and false innovation
There may still be a place for non-inferiority trials when there are potential advantages for patients and/or the community, besides similar efficacy. A non-inferiority study can be regarded as ethical when a drug similar to its comparator in terms of efficacy may nevertheless have better safety, convenience, or cost-effectiveness: for example, a Cox-2 selective inhibitor with fewer gastrointestinal adverse effects, a sartan inducing less cough than ACE-inhibitors, an orally available iron chelator for children with thalassaemia, an estrogen patch replacing daily pills, a vaccine combination. What is important is that the claimed advantages and their impact on patients and/or the community be appropriately tested and documented, as is required for efficacy. This approach in fact implies documentation of superiority.
It is hard to envisage here any role for placebo (4): potential advantages should be assessed in comparison with any recognised active control, of course. The advantage should be pre-specified and serve as a guide for establishing an acceptable limit of non-inferiority, meaning the difference in outcome allowed for two products still to be considered similar. The excess of events allowed to document similar clinical efficacy while looking for other potential advantages should be explicitly justified: how many more deaths, infarctions, interventions, etc would we consider acceptable in testing (and possibly proving) that the new drug is actually, safer, better tolerated, easier to use, or cheaper than its comparator?
More independent clinical research as a solution Most divergences on the use of placebo or active control would be minimised if public health interest were constantly kept as the primary objective. If the study plan and conduct reflect this, investigators' intellectual independence with respect to the tested hypothesis would be assured too. Financial interests can equally be covered when proven advantages are provided to individual patients or their community. In this respect, medicinal products cannot be regarded like any other product, because their value should only lie in the public health gains they provide, which need to be documented objectively.
Ethical Committees are urged to look more carefully at trial protocols to establish not only that no harm will result to patients involved in the trial but also whether the trials are appropriately designed to demonstrate any foreseeable benefit for patients for whom the drug is intended after its approval.
Public funds supporting independent clinical research play a critical role in helping the scientific community keep its intellectual freedom with respect to priority objectives, identification of clinical hypotheses and appropriate trial design. It also enables public health institutions to address questions that are not in the direct interests of industry (11), which cannot be expected to pursue, and fund, scientific and public health objectives that are not in line with - and may even be against - its commercial aims. In the United States health policy makers have correctly understood these principles and ensured an adequate level of funding for independent research.
Good examples are the results achieved with public funds for HIV/AIDS treatments at the time the public health interest for this area was particularly important (12). The same cannot be said at present for the European Community, which still has to develop its own policy in this area and decide on the level of funding it considers useful to cover the public health interest. It is important to note that independent research would also indicate to industry what kind of products public health needs, and could therefore be a powerful stimulus to re-orientate industrial clinical research too.
References
1. The World Medical Association, Inc. The Declaration of Helsinki (2000). Available on http://www. wma. net.
2. Temple, R. Ellenberg, S. S. Placebo-controlled trials and active-control trials in the evaluation of new treatments. Part 1: ethical and scientific issues. Annals of Internal Medicine. 133. 455 - 463 (2000).
3. Emanuel, E. J. Miller, F. G. The ethics of placebo-controlled trials - A middle ground. New England Journal of Medicine. 345. 915 - 919 (2001).
4. Lewis, J. A. Jonsson, B. Kreutz, G. et al. Placebo-controlled trials and the Declaration of Helsinki. Lancet. 359. 1337 - 1340 (2002).
5. Anonymous. Drugs in 2001: a number of ruses unveiled. Prescrire International. 11. 58 - 60 (2002).
6. Bertele, V. Torri, V. Garattini, S. Inconclusive messages from equivalence trials in thrombolysis. Heart. 81. 675 - 676 (1999).
7. Barbui, C. Violante, A. Garattini, S. Does placebo help establish equivalence in trials of new antidepressants? European Psychiatry. 15. 1 - 6 (2000).
8. Garattini, S. Bertele, V. Adjusting Europe's drug regulation to public health needs. Lancet. 358. 64 - 67 (2001).
9. Kupfer, D. J. Franck, E. Placebo in clinical trials for depression. Complexity and necessity. Journal of the American Medical Association. 287. 1853 - 1854 (2002).
10. Greene, P. J. La Vaque, T. J. Postmarketing surveillance and black box warnings. Journal of the American Medical Association, 288. 957 (2002).
11. Writing group for the women's health initiative investigators. Risks and benefits of estrogen plus progestin in healthy postmenopausal women. Journal of the American Medical Association, 288. 321 - 333 (2002).
12. Consumer Project on Technology backgrounder on the development of fourteen FDA-approved HIV/AIDS drugs. Available on http://www. cptech. org/ip/health/aids/druginfo. html
(Views expressed in signed contributions remain the responsibility of the authors.)
No comments:
Post a Comment