Lesprit et la molécule (French Edition)
In the first phase, which had the object of inducing remission, no control group was used. It was the clinician's experience of the disease that served as the reference point. Nevertheless, despite his extensive experience, Bernard did not always consider himself able to distinguish between spontaneous remission and remission due to treatment.
Then, at the beginning of the s, Bernard and his co-workers started organizing comparative trials in an attempt to find therapeutic schedules capable of maintaining remissions. They used no placebo, administering instead competing therapies to each different group of patients. These trials were neither randomized, nor double blind. Even if the Parisian team compared their evaluation criteria regularly with those used by American acute leukaemia specialists, they saw no need to align their practices with those of their American colleagues.
We have already seen that in America, from the mids onwards, clinical trials on acute leukaemia were conducted by the different specialized centres in a similar way, that is to say mainly on the basis of local practices and knowledge and without referring to controlled clinical trial methodology. The new committee was directed by Sidney Farber, and was responsible for co-ordinating the relationship between government, industry, and institutions.
The main mission of this new agency was to supervise pharmacological and other pre-clinical tests, in addition to organizing co-operative clinical trials, thereby bringing together the numerous American teams and, later, foreign teams as well.
Scaling-up had important consequences for the practices involved and the knowledge produced, for standardization was perceived as the condition for fruitful co-operation, transferability of results, and speed of evaluation of candidate molecules. The first co-operative study on leukaemia organized under the aegis of the CCNSC involved four American centres, and included sixty-five patients who were treated between May and October This study aimed at comparing two modes of combination of 6-mercaptopurine and amethopterine.
In the first case, the folic acid antagonist was given continuously, and in the second, only intermittently. The results, published in , showed that remission and survival lasted longer under the continuous regimen. Considering the methodological aspects of their study, the authors of the report, which cited the work of the British statisticians Austin Bradford Hill and Ronald Fisher, insisted upon the adequacy of their approach compared with the principles of the controlled clinical trial.
Patients had been paired according to their age and type of leukaemia, before being randomly allocated to one or other treatment group. For each type of treatment, a detailed protocol had been drawn up and strictly followed. The authors also emphasized the ethical aspects taken into consideration when designing the trial. They argued that this kind of trial was not prejudicial to patients, because the median survival time obtained was similar to that recently published elsewhere in the literature.
Its objectives were to compare treatment with 6-mercaptopurine alone on the one hand, with treatment combining 6-mercaptopurine and azaserine on the other. This randomized controlled trial took place between December and March , and involved patients. Compared to the previous trial, its novelty lay in the fact that patients were receiving their first leukaemia therapy in the context of this trial.
No significant difference was observed between the two forms of treatment. It aimed at determining the therapeutic value of a new compound, 6-azauracil, in cases of leukaemia that were resistant to the usual treatments. The protocol was designed in collaboration with the statistician Edmund Gehan. Paradoxically, the main conclusion reached by this trial was that the use of a placebo was not necessarily required.
Molecular biology in the French tradition? Redefining local traditions and disciplinary patterns
Since 6-azauracil was inactive against the disease, the trial was judged convenient for use as a control in subsequent trials. Another methodological innovation, sequential analysis, was introduced in an ALGB study conducted in and This co-operative trial, also elaborated with Gehan, investigated the effect of 6-mercaptopurine on the duration of the remission induced by steroids. Patients in remission were selected at random, and were given 6-mercaptopurine or a placebo according to a double-blind procedure. When in relapse, the patients who were given the placebo received 6-mercaptopurine.
The sequential analysis of the results allowed the trial to be stopped after remission lengths were observed in twenty-one pairs of patients, since the benefit of a maintenance therapy with 6-mercaptopurine had been proved. On sequential analysis, the paper referred to the work of Peter Armitage, a collaborator of Bradford Hill. Otherwise, the authors of the report also drew attention to the ethical value of controlled trials.
They argued that the use of a placebo had no significant effect on survival time, and that this therapeutic schedule therefore allowed new products to be tested while providing the best treatment. Controlled clinical trials organized by the ALGB had, as already mentioned, very little or no influence on the assessment of new treatments in Paris during the second half of the s.
Surprisingly, they also had little impact on the practices of Burchenal's team, despite its participation in one of the first CCNSC trials. Moreover, the results were compared to those of a local historical control group, and the report did not include any references to statistical methodology, although the criteria for evaluating results were those of the CCNSC. All these observations suggest that at the end of the s, local norms coexisted with standard protocols for the conduct of therapeutic trials both in Europe and in the United States.
Was this also the case in the s, when the NCI model of chemotherapeutic research and the RCT standard seem to have become generalized? Links between the Paris unit and the NCI leukaemia task forces were formalized in the early s.
- Subscribe to RSS.
- Also Available In:!
- Habits of Mind A Brand New Condillac.
- The Peppermint Man?
- La mémoire : De l'esprit aux molécules;
- The Wisdom of Manjusri.
After Michel Boiron contacted the ALGB during a research trip to the United States in , Bernard's team was included in the randomized co-operative trial ALGB , designed to compare the effects of 6-mercaptopurine and methotrexate amethopterine in acute leukaemia in adults. The relationship between Bernard's team and French medical statisticians followed the same chronology.
In , he took advice from the mathematician Daniel Schwartz while preparing a report on the effect of treatments on survival. Meanwhile, similar preliminary trials were being conducted by Burchenal's team, although they differed in two respects from the French ones. First, in , the American group started collecting informed consent forms from participants; second, they continued to recruit patients in relapse.
The Paris protocols were elaborated at the Saint-Louis Hospital by a chemotherapy committee chaired by Bernard. Although the organization of clinical trials in France seems to have been influenced by American practices, the ALGB was not the only model used. Moreover, criteria for assessing treatments resulted from convergence, rather than alignment, with the ALGB. These were: A for marrow, B for peripheral blood, C for physical findings, and D for symptoms. Complete remission implied a reversal to normal, both haematologically and clinically, with a rating of 1 in all four categories A1B1C1D1.
Relapse was considered to occur when the patient's condition deteriorated in any category A3B3C3D3. Other situations were called partial remissions. To qualify for an A1 rating, marrow was not to exceed 5 per cent blast cells in adults, and 10 per cent in children. For Bernard, bone marrow had to contain less than 5 per cent blast cells, even in children.
In order to define responses to treatment with greater accuracy, the co-operative group also modified the categories for evaluation. Within each category, parameters were graded between 1 and 5. A corresponded to a rating of 1 in all categories M, H, P and S. Five terms were then used to describe the response to therapy.
Press corner | European Commission
Finally, during the s, the transformation of the therapeutic trials conducted by Bernard's team was partially inspired by the controlled clinical trial methodology promoted by ALGB statisticians. However, the organization of clinical trials continued to be governed mainly by local norms.
- Wildlife Damage Management.
- Citation Tools;
- Elaboration of a consensual definition of de-escalation allowing a ranking of β-lactams..
This feature was not specific to France, for it also characterized the leading American centres devoted to leukaemia research. Just as in the s, local and international norms coexisted in Europe and the United States in the therapeutic trials in onco-haematology carried out in the s.
Molecular Biology and its Recent Historiography: A Transnational Quest for the ‘Big Picture’
Nevertheless, given the changing organization of cancer clinical research in the United States, the internationalization of the trials, and the circulation of statistical tools, these local and international norms gradually converged. A comparison between the introductory papers co-ordinated by Hill, and the French report written by Schwartz and his colleagues, shows very few differences. All mentioned the same goals, ethical considerations, and methods.
Thus, they agreed that retrospective controls—although widely applied in the past—should be avoided, that treatments should be allotted using tables of random sampling numbers, and that giving one drug in one centre and a competing treatment in another, or giving different treatments concurrently without preliminary randomization, made valid comparisons impossible.
Other common rules were the use of double-blind assessment, and the fact that the decision to enter a patient in a trial had to be made before randomization. Both the British and the French groups tolerated certain departures from the ideal of the controlled trial. These included the use of the patient as his or her own control, concurrent assessment, and sequential analysis. Moreover, clinical trials in malignant disease were considered to be peculiar but valuable despite obvious difficulties.
According to both French and British statisticians, cancer trials required the use of percentage survival for a period of at least five years as an index of curative value. Moreover, they accepted that initial treatment could be determined by random methods and that then, at a later stage, the patient could be given the best treatment. Lastly, double-blind trials were generally found to be impractical for cancer.
The general idea that medical progress depended on the pursuit of systematic, large-scale, co-operative investigations circulated amongst young physicians and health officials in the s, and was largely accepted by the s. However, the notion that medical progress was associated more specifically with controlled trials, conducted on human subjects and investigating the activity of chemotherapeutic agents, was invented and promoted by the physicians. As mentioned above, the reform of medical teaching and hospital organization that was adopted in and implemented in the s played a crucial role in the acceptance of this particular interpretation of biomedicine.
The reform itself resulted from the alliance between medical reformers and state administrators who believed that planning and large-scale socio-technical programmes were the best way to rationalize and modernize French society. On the occasion of Bernard's seventieth birthday, Jean Hamburger wrote of Bernard's role and of what had inspired their self-defined and self-organized group of neo-clinicians:.
Jean Bernard has always been for me—and for many others I guess—a model.
- Opioïdes : contre les overdoses, un antidote sous diffusé.
- The Human Tradition in the Black Atlantic, 1500–2000 (The Human Tradition around the World series)?
- La vision dynamique de Teilhard de Chardin sur l'Univers.
The first reason is that Jean Bernard is a physician who is convinced that medicine can no longer do without the method, humility, and reasoning of science.