Cross-over designs

There are two main types of clinical trial design, parallel and cross-over. In a parallel study, subjects are assigned to one of two or more treatments, e.g. active and placebo, and proceed through the trial concurrently. In a cross-over design, subjects act as their own controls, undergoing two or more treatments in sequence (see Fig. 12.1). 

Because they require fewer subjects and are usually less costly to complete, cross-over trials are commonly used to investigate the effects of phytochemical products. Cross-over designs may be used effectively for studying conditions that are relatively stable so that a similar baseline status can be established at... 

Some clinical trials can be completed with only a few visits. Others require more frequent contact with the study staff. As an example of the former, our clinic has conducted many studies intended to assess blood insulin and glucose responses to test products such as snack bars and beverages. These are usually conducted using a cross-over design and may require only three visits one for screening, one for consumption of the control product, and one for consumption of the active product. In contrast, we have also completed several trials to assess dietary and pharmaceutical interventions intended to promote weight loss. These usually require frequent clinic visits over a period of at least 12 weeks and sometimes as long as two years. 

There are occasions when it is not possible to use a cross-over design in a bioequivalency determination. For example, if the half-life of the drug is very long, the required washout period between the two treatment periods may be several months. Obviously, it is quite impracticable to consider such a long washout period. Test subjects are unlikely to wait patiently for long periods of time, and thus if we tried to conduct a bioequivalency study with, say, a 3-month washout period, we would probably find that a significant number of our test subjects would not be available for the second dose. 

In those instances when the cross-over design is not possible, a parallel design may be required. In such a protocol the test subjects are divided into... 

A practical approach for assessment of bioquivalence under selected higher-order cross-over designs, J. Vuorinem, Stat. Med., 16, 2229 (1997). 

Y. LeRoux, C. Guimart, and M. Tenenhaus, Use of repeated cross-over design in assessing bioequivalence, Eur J. Drug Metab. Pharmacokinet., 23, 339 (1998). 

There is the possibility of carry-over effects. This is more crucial in Latin square and other cross-over designs. Knowledge of pharmacokinetics and metabolism of a compound under study generally helps in avoiding this problem. 

Two studies have been performed so far on women with breast cancer complaining of hot flushes - neither showed an improvement (Table 4.4). Quella et al. (2000) did not show any reduction in hot flushes in breast cancer survivors using 150 mg of phytoestrogen in tablets. The study was a cross-over design and had two phases lasting only four weeks, which were not separated by a wash-out period thus a carry-over effect cannot be excluded. 

To test the hypothesis that the addition of CCM reduces dental erosion, the erosive effects of four different drinks was compared a citric acid-based orange-flavored soft drink fortified with CCM (pH 4.0, 1344 mg Ca/liter) the same drink without CCM (pH 3.6, 72 mg Ca/liter) and positive and negative controls consisting of a diet phosphoric acid-based cola (pH 3.1, 35 mg Ca/liter) and distilled water, respectively (Rugg-Gurm et ah, 1998). In a randomized cross-over design comprised of four 6-day periods, 11 subjects were required to wear a palatal... 

Cross-over design trials represent an adaptation of randomized control trials in which each participant acts as his/her own control. In the simplest scenario, each participant will receive either a placebo or the test drug for the first half of the trial and will receive the alternative treatment for the second half. The order of placebo versus test drug for any individual is randomized. Hence, at any time point, approximately half the test participants will be receiving the placebo and the other half the test substance. 

Choosing the most appropriate design (for example using a cross-over design rather than a parallel group design where this is appropriate)... 

Another example of the within-patient design is the cross-over design. Again each subject receives each of the treatments but now sequentially in time with some subjects receiving the treatments in the order A followed by B and some in the order B followed by A. 

In both the paired design and the cross-over design there is, of course, randomisation in the second paired design example above, it is according to which forearm receives A and which receives B and randomisation is to treatment order, A/B or B/A, in the cross-over design. 

It is important therefore to only use these designs when you can be sure that carry-over effects will not be seen. Introducing a washout period between period I and period II can help to eliminate carry-over so that when the subject enters period II their disease condition is similar to what it was at the start of period I. Cross-over designs should not be used where there is the potential to affect the underlying disease state. ICH E9 is very clear on the use of these designs. 

The cross-over design is used extensively in phase I trials in healthy volunteers to compare different formulations in terms of their bioequivalence (where there is no underlying disease to affect). They can also be considered in diseases, for... 

The Respective Advantages of the Parallel Group and Cross-Over Designs... 

Assay performance criteria for biopharmaceuticals are often highly variable therefore strict statistical criteria that attempt to rigorously establish traditional in vivo bioequivalence may not always be appropriate. In some cases an assessment of rate and extent of absorption as indicated by the maximum concentration (Cmax), time of maximum concentration (Tmax) and area under the curve (AUC) may be needed. In other cases complicating factors related to binding proteins, endogenous concentration, and unusual concentration-time profiles may need to be considered [15]. In cases where complications may arise from immune response to heterologous proteins, cross-over designs are inappropriate. 

Studies can be conducted with a group comparison or with cross-over design. Individual parameters can be averaged over predefined time periods but it may not always be possible to predict the time of drug-induced effects. Also, in freely moving animals, drug-induced changes may not be continuous but there may be... 

This type of hADME study will always have an explorative character, and typically comprises less than the normal minimum of 12 subjects for a PK study. If the compound is expected to show different and unpredictable ADME characteristics in special populations, then it might be necessary to include these populations in addition to normal, healthy subjects. Or, if other conditions might influence ADME in an unpredictable manner (e.g. food effects), then it might even be necessary to run the study in a cross-over design. The inclusion of females (being not of childbearing potential) was discussed several times, but it seems to be a rare exception for hADME studies. 

Fourteen subjects order of periods arranged according to a double, cross-over design. 

Spinach is also an excellent source of manganese. However, spinach contains high amounts of soluble fiber and oxalic acid. Both of these factors have been found to inhibit the utilization of iron. Using two 5-day periods in a cross-over design, manganese utilization from spinach was determined. Subjects (12) consumed their normal, self-selected, self-recorded diets. During one of the two randomly arranged periods, subjects were asked to eat one 8 oz. can of spinach. 

Su K-P, Shen WW, Chuang C-L, Chen K-P, Chen CC. A pilot cross-over design study on QTc interval prolongation associated with sulpiride and haloperidol. Schizophr Res 2002 59 93 1. 

Figure 4 Mean it-verapamil S-verapamil ratio observed following intravenous and oral dosing of racemic verapamil. A, After intravenous administration of a single 15 mg dose in 8 volunteers (personal communication from A. Rasymas, Univ. of Toronto, Canada). B, After oral administration of two different 120-mg immediate release formulations dosed every 8 hr to 22 normal volunteers in a crossover design study and measured at steady state over two dosing intervals (- -, test formulation u, reference formulation). C, After oral administration of two different lots of a 180-mg once daily sustained-release formulation to 48 normal volunteers in a cross-over design study and measured at steady state new manufacturing site n, reference manufacturing site).

Figure 7 Stereospedfic comparison of concentration-time profiles of verapamil and norverapamil obtained from two lots of sustained-release verapamil manufactured at different manufacturing sites (—new manufacturing site —, reference manufacturing site). Sustained-release racemic verapamil was administered 240 mg once daily to the 44 normal volunteers in a cross-over design study plasma samples were collected at steady state on d s 7-8. (On days 6 and 7 verapamil was administered with food, and on day 8 it was administered while fasting.)...

Phenotyping could also be important for safety reasons, determination of sampling times and wash-out periods in cross-over design studies. 

A cross-over design can be used. Where this is not appropriate a parallel group study design should be chosen. 

The better-designed studies included double blind, placebo-control conditions within a parallel research design. However, most studies did not provide any data on how they randomized or counter-balanced their experimental and placebo or control groups. No studies explicated how they were able to match the two groups in terms of the severity or duration of the subjects anxiety states. Future studies should use a more robust research methodology, such as a double blind, placebo-controlled, cross over design. 

---