Table 4.

The prediction formulae derived using the multiple linear regression approach to estimate the AUC_0–24 of afatinib.

Sampling Numbers	Sampling Time (h)	Prediction Formula for AUC0–24	Predicted versus Observed AUC0–24		Slope		Intercept 95% CI *	p *
			r2	p	95% CI *	p *
One-point	0	22.3× C0 + 215.9	0.761	<0.001	17.7 to 28.6	0.001	85.5 to 331.2	0.005
	1	16.4 × C1 + 286.1	0.712	<0.001	12.2 to 21.8	0.001	143.7 to 411.0	0.001
	2	14.5 × C2 + 276.9	0.691	<0.001	10.6 to 19.7	0.001	110.3 to 433.1	0.012
	4	13.7 × C4 + 263.4	0.762	<0.001	10.5 to 18.0	0.001	112.1 to 410.0	0.007
	6	14.0 × C6 + 214.6	0.840	<0.001	11.4 to 17.3	0.001	81.7 to 334.6	0.004
	8	17.5 × C8 + 75.9	0.899	<0.001	15.6 to 20.1	0.001	−25.0 to 159.6	0.108
	12	23.8 × C12 + 11.9	0.916	<0.001	21.6 to 26.7	0.001	−75.2 to 84.7	0.770
Two-points †	0	10.6 × C0 + 9.1 × C6 + 135.4	0.911	<0.001	6.1 to 16.5	0.003	38.8 to 228.5	0.022
	6				5.6 to 12.3	0.001

AUC_0–24, area under the plasma concentration–time curve from 0 to 24 h; C_n, plasma concentration at n h after afatinib administration. * Calculated using the bootstrap method. † Best sampling point.