Multi-strain studies using isogenic strains in which a quantitative end-point is measured are more powerful than those using a single outbred stock for three reasons:
- They average out any strain differences so that it is less likely that an effect will be missed because the strain was resistant.
- By using isogenic strains the within-group variation is reduced, so the response (in standard deviation units) is increased, leading to more statistical power.
- The show whether the response is under genetic control.
As an example, the haematological response of mice to chloramphenicol (CAPS) at six dose levels was studied using two mice of each of four isogenic strains (C3H/He, CBA/Ca, BALB/c and C57BL/6 and 6x2x4=48 mice) versus eight mice of the CD-1 outbred stock at each dose level (6×8=48 mice). (data from Festing et al 2001).
Eight haematological parameters were studied in each case. For each parameter the absolute (i.e. regardless of sign) difference between the control group mean and treated group means (averaging across strains in the case of the multi-strain experiment) were calculated and this was divided by the standard deviation of each character. These are shown graphically below. A dotted line has been drawn at 1.75 standard deviations. This is the size of effect which, in some future experiment, would be detectable with a sample size of eight per group, a 90% power and 5% significance level using a two sided, two-sample t-test.
Note that with the CD-1 stock the reticulocytes were affected at the 2000 mg/kg level and the HGB and platelets at the 2500 level. However, with the multi-strain study 4/8 characters were affected at the 1500mg/kg level and 6/8 were affected at the highest level. This shows that the multi-strain study was more sensitive to the toxic effects than the one using just the CD-1 stock.