Patient eligibility

Eligibility criteria can in general be divided into two types - safety and efficacy criteria. Safety criteria are used to select patients who are fit to receive any of the treatments under consideration. Requirements for specific haematological and biochemical values address safety issues as, for example, do the exclusion of patients with immune-deficient diseases from chemotherapy trials, and the tumour size limit on patients being considered for stereotactic cranial irradiation. The efficacy criteria aim to identify patients with the potential to benefit from the new treatment. These include, for example, the common criteria of no previous or concurrent serious disease - if a patient is more likely to die from a disease other than the one the trial is aiming to treat, they are less likely to contribute to the trial than a patient with no obvious, immediate risk of a competing event. There are many other factors which nominally come under this category - things such as age, stage of disease and histology. Trials will often have eligibility criteria based on such things. But it is rare that one of these factors will actually distinguish a patient who will respond from one who will not, it is more a question of degree, or of anticipated quantitative differences, rather than qualitative differences in response. Older patients may be more likely to suffer toxicity limiting the completion of some treatments with consequent effects on their efficacy. But receiving three courses rather than six is not likely to cause harm rather than benefit, rather minimal versus maximal benefit. Hence it is these efficacy criteria which should be considered most carefully in explanatory and pragmatic trials. Explanatory trials might impose stringent limits in an effort to obtain a group of patients thought most likely to respond the way we hope they will. The pragmatic trial will aim to include in the eligibility criteria all those patients who might be considered for treatment outside of a trial setting.

The 'uncertainty principle' gained recognition with the first 'mega trials' in cardiovascular diseases and then cancer [7]. This states that if you feel there is a definite indication or contraindication to one of the treatments under study, then the patient should not be entered into the trial. Only if the doctor and patient are uncertain as to the benefit should a patient be randomized. It is naïve to assume that all trials only involve clinicians with complete uncertainty, and more generally it can be sufficient to have a balance of uncertainty (clinical equipoise) in the clinical community as a whole [8].

Interpreted in its broadest sense, the uncertainty principle is a necessary, if often unwritten, eligibility criterion for any randomized trial. In some situations - for example a widely used treatment - it maybe the only eligibility criterion which is required, since it covers both the safety and efficacy criteria. However, it must be remembered that clinical trial protocols will often be testing new treatments with which many collaborating clinicians will have limited experience. Additional guidance on eligibility may often therefore be necessary. For example, a randomized trial evaluating the role of portal vein infusion of fluorouracil (AXIS) used the uncertainty principle as its only eligibility criterion; however, to steer less experienced clinicians the protocol specified additional guidelines as to contraindications (see Box 4.4).

Eligibility criteria have rightly come under more intense scrutiny in all stages of trials, and the assumptions about homogeneity of anticipated response questioned. For example, it is now uncommon for trials to include an upper age limit. In the past this was often used as a surrogate for fitness to receive treatment and risk of intercurrent

0 0

Post a comment