|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Originally published as JCO Early Release 10.1200/JCO.2009.22.5359 on April 27 2009 © 2009 American Society of Clinical Oncology.
The Amiens Strategy: Small Phase III Trials for Clinically Relevant Progress in the War Against CancerDepartment of Urology, Vrije Universiteit (VU) University Medical Center, Amsterdam, the Netherlands
Institute of Immunology, Vilnius University, Vilnius, Lithuania
Department of Pathology, Vrije Universiteit (VU) University Medical Center, Amsterdam, the Netherlands
National Association of Dairy Farmers, Harare, Zimbabwe
Department of Urology, Vrije Universiteit (VU) University Medical Center, Amsterdam, the Netherlands
Clinic of Gastroenterology, Faculty of Medicine, University Hospital St Ivan Rilski, Sofia, Bulgaria
Department of Urology, Vrije Universiteit (VU) University Medical Center, Amsterdam, the Netherlands To the Editor: We admire the plea of Stewart and Kuzrock1 for the use of the Amiens strategy in the war against cancer. Cancer research should focus on clinical relevance that can be proven mathematically rather than on statistical significance alone. Clinical relevance implies considerable improvement (eg, > 50% or > 6 months) in clinically relevant parameters like disease-free or progression-free survival. We would like to add some points in line with the ideas of Stewart and Kuzrock. Large clinical trials are performed to reduce false-positive and false-negative results (ie, so-called type I and II errors). Regarding the first, the same P value has the same probability of being a false-positive result, whether it is associated with a large trial of 2,000 patients or a small trial with 20 patients per arm. Thus, large trials are not necessarily more resistant to false-positive statistical results. Regarding the second, false-negative results occur more often in small trials, especially if the therapeutic intervention yields only a limited clinical improvement. Thus large clinical trials can reach statistically significant conclusions that have no clinical relevance. Reducing the size of trials would greatly enhance the number of clinical trials performed. This would be great for progress against cancer, but it might also have a less desired effect. A publication bias could occur because trials without statistically significant results are less likely to be published. This problem could be circumvented by the registration of all clinical phase III trials.2 Careful preclinical testing can prevent many large, expensive, and unsuccessful clinical phase III trials. Laboratory animal experiments can have predictive value for clinical studies, but they must be performed and evaluated in an appropriate way.3 Experimental tumors should comprise approximately 1% of body weight (250 mg = approximately 2.5 x 108 cells in a mouse), in contrast to the small tumor loads of 103 to 106 cells that are often used. Therapeutic effects should be measured and evaluated in clinical terms (eg, complete response, partial response, stable disease, and progressive disease), not on the basis of significantly slower tumor growth. In preclinical research, there is an inclination to regard slower growth as a promising effect, whereas clinicians would regard the same result as unpromising progressive disease. In theory, the therapeutic effect of a drug may differ in mice and men, because the biology of rodents is different from that of men. Yet such differences in therapeutic effect are rare.3 In large trials using heterogeneous patient populations, the risk of overlooking clinically relevant improvements for subgroups exists. In retrospective analyses of subgroups, the risk of data dredging exists.4 Thus, the most appropriate method is small-scale trials with homogeneous patient populations. In the case of renal cell carcinoma, positive and negative clinical effects of interferon treatment could be distinguished on the basis of cellular immune status.5 A single trial group, regardless of its size, would not yield these contradictory results between different patient groups. Finally, different application routes may result in huge differences in clinical effects of chemotherapy6 and immunotherapy.7 The pharmacology of drugs determines their bioavailability and clinical effects. For example, intratumoral application of interleukin-2 can have enhanced antitumor effect as a result of intratumoral vascular leakage leading to massive necrosis.8 This has been applied in the phase II setting in 10 patients undergoing irradiation for stage III to IV nasopharyngeal carcinoma in a nonendemic area. Even in this small study,9 we were able to show statistical significance (P = .01) and clinical relevance by a huge increase (from 8% to 63%) in 5-year disease-free survival rates compared with historical controls. This clinically relevant improvement calls for a randomized phase III study. However, the current pre-Amiens paradigm in cancer research points us toward a much larger clinical trial than is required. Like the stagnant trench warfare strategy, large clinical trials hamper progress toward victory in the war against cancer. Therefore, we cheer the initiative of Stewart and Kuzrock1 to follow the strategy of the Battle of Amiens in performing small-scale trials aiming for clinically relevant improvements. AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST The author(s) indicated no potential conflicts of interest. REFERENCES
1. Stewart DJ, Kurzrock R: Cancer: The road to Amiens. J Clin Oncol 27:328–333, 2009. 2. De Angelis C, Drazen JM, Frizelle FA, et al: Clinical trial registration: A statement from the international committee of medical journal editors. N Engl J Med 351:1250–1251, 2004. 3. Den Otter W, Steerenberg PA, Van der Laan JW: Testing therapeutic potency of anticancer drugs in animal studies: A commentary. Regul Toxicol Pharmacol 35:266–272, 2002.[CrossRef][Medline] 4. Maggioni AP, Darne B, Atar D, et al: FDA and CPMP rulings on subgroup analyses. Cardiology 107:97–102, 2007.[CrossRef][Medline] 5. Characiejus D, Pasukoniene V, Kazlauskaite N, et al: Predictive value of CD8highCD57+ lymphocyte subset in interferon therapy of patients with renal cell carcinoma. Anticancer Res 22:3679–3683, 2002.[Medline] 6. Scheper RJ, Vos A, De Groot J, et al: Evaluation of various cytostatic drugs as local immunotherapeutic agents. Invest New Drugs 2:221–225, 1984.[Medline] 7. Shaker MA, Younes HM: Interleukin-2: Evaluation of routes of administration and current delivery systems in cancer therapy. J Pharm Sci epub ahead of print on November 13, 2008. 8. Jacobs JJ, Sparendam D, Den Otter W: Local interleukin 2 therapy is most effective against cancer when injected intratumourally. Cancer Immunol Immunother 54:647–654, 2005.[CrossRef][Medline] 9. Jacobs JJ, Hordijk GJ, Jürgenliemk-Schulz IM, et al: Treatment of stage III-IV nasopharyngeal carcinomas by external beam irradiation and local low doses of IL-2. Cancer Immunol Immunother 54:792–798, 2005.[CrossRef][Medline]
Related Article
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2009 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
