Development of musculoskeletal toxicity without clear benefit after administration of PG-116800, a matrix metalloproteinase inhibitor, to patients with knee osteoarthritis: a randomized, 12-month, double-blind, placebo-controlled study

Table 3 One-year change from baseline in joint space width of the signal knee (intent-to-treat population)

Treatment group	Number^a	Joint space width at baseline	Joint space width at month 12	One-year change from baseline	Mean 1-year percentage change from baseline	P value^b (90% CI)
		Mean (SEM)	Mean (SEM)	Least square mean (SEM)
Placebo	71	3.386 (0.085)	3.252 (0.096)	-0.136^c (0.063)	-4.00
25 mg	66	3.350 (0.094)	3.307 (0.105)	-0.044 (0.062)	-1.24	0.127 (-0.012, ∞)
50 mg	67	3.346 (0.094)	3.123 (0.113)	-0.226^c (0.065)	-7.07	0.869 (-0.193, ∞)
100 mg	69	3.413 (0.094)	3.218 (0.129)	-0.200^c (0.065)	-7.61	0.789 (-0.166, ∞)
200 mg	52	3.466 (0.108)	3.278 (0.138)	-0.192^c (0.072)	-6.72	0.740 (-0.166, ∞)

^aNumber of patients with available data at baseline and month 12. ^bOne-sided P value: The mean change from baseline in each treatment group was compared to placebo by means of an analysis of variance after adjusting for pooled centers, baseline use of estrogen replacement therapy or selective estrogen receptor modulator drugs, and baseline joint space width. ^cThe mean change from baseline was significantly different from zero using a paired t test within each treatment group.
CI, confidence interval for mean difference; SEM, standard error of the mean.

ISSN: 1478-6362