Table 2

Time to loss of therapeutic responsea

Measurement

Treatment group, n

Time to LTR, days (95% CI)b

Hazard ratio

(95% CI)c

P-valued

Patients with LTR at end of DB treatment, n (%)


25th percentile

50th percentile


Primary analysis

PBO, 50

18 (16 to 29)

56 (28 to 85)

0.44 (0.27 to 0.71)

<0.001

32 (64)

MLN, 100

45 (29 to 59)

NC

35 (35)

Sensitivity analysis I

PBO, 50

18 (16 to 29)

56 (28 to 85)

0.46 (0.29 to 0.74)

0.001

32 (64)

MLN, 100

36 (29 to 57)

NC

37 (37)

Sensitivity analysis II

PBO, 50

28 (18 to 63)

85 (60 to NL)

0.35 (0.19 to 0.65)

<0.001

21 (42)

MLN, 100

NC

NC

19 (19)

Sensitivity analysis III

PBO, 50

29 (17 to NL)

NC

0.59 (0.32 to 1.08)

0.08

18 (36)

MLN, 100

86 (43 to NL)

NC

24 (24)

PGIC

PBO, 50

22 (15 to 31)

86 (30 to NL)

0.36 (0.20 to 0.63)

<0.001

25 (50)

MLN, 100

NC

NC

22 (22)

MAF global fatigue

PBO, 50

18 (16 to 68)

NC

0.80 (0.46 to 1.38)

0.41

20 (40)

MLN, 100

30 (26 to 59)

NC

36 (36)

SF-36 PCS

PBO, 50

15 (15 to 22)

68 (18 to NL)

0.81 (0.50 to 1.30)

0.36

26 (52)

MLN, 100

23 (15 to 33)

87 (57 to NL)

47 (47)

SF-36 MCS

PBO, 50

28 (16 to 46)

57 (29 to NL)

0.74 (0.44 to 1.24)

0.24

23 (46)

MLN, 100

29 (23 to 60)

90 (85 to NL)

40 (40)


aLoss of therapeutic response was defined as follows: (1) primary analysis: <30% reduction in Visual Analogue Scale (VAS) pain score from pre-milnacipran exposure or worsening of fibromyalgia requiring alternative treatment; (2) sensitivity analysis I: <30% reduction in VAS pain score from pre-milnacipran exposure or withdrawal from the study for any reason; (3) sensitivity analysis II: <30% reduction in VAS pain score from pre-milnacipran exposure; or (4) sensitivity analysis III: worsening of fibromyalgia requiring alternative treatment. DB = double-blind; LTR = loss of therapeutic response; MAF = Multidimensional Assessment of Fatigue; MCS = Mental Component Summary; MLN = milnacipran; NC = not calculable (no patients in quartile); NL = no limit; PBO = placebo; PCS = Physical Component Summary; PGIC = Patient Global Impression of Change; SF-36 = Short Form-36 Health Survey. bQuartiles (that is, 25th percentile, 50th percentile) based on Kaplan-Meier estimates. cHazard ratio of milnacipran versus placebo based on Cox proportional hazards regression model with treatment group as an explanatory variable. dP-values based on logrank comparison between milnacipran and placebo.

Clauw et al. Arthritis Research & Therapy 2013 15:R88   doi:10.1186/ar4268

Open Data