Skip to main content
. 2020 Dec 1;3(12):e2027754. doi: 10.1001/jamanetworkopen.2020.27754

Table 2. Association of the Baseline MMP and TIMP Levels With Treatment Outcomes.

Marker Univariate model Multivariable model
OR (95% CI) P value aOR (95% CI)a P value
MMP-1 1.36 (1.01-1.85) .04 1.32 (0.95-1.84) .10
MMP-2 2.76 (1.40-5.48) .004 3.26 (1.51-7.03) .003
MMP-3 0.48 (0.27-0.88) .02 0.45 (0.22-0.92) .03
MMP-7 1.75 (1.18-2.61) .006 1.71 (1.09-2.68) .02
MMP-8 2.04 (1.33-3.14) .001 2.16 (1.34-3.47) .001
MMP-9 1.80 (0.89-3.64) .10 2.16 (0.99-4.67) .051
MMP-12 1.15 (0.30-4.41) .84 1.29 (0.30-5.50) .73
MMP-13 0.20 (0.02-2.39) .20 0.18 (0.01-2.77) .22
TIMP-1 7.55 (2.85-19.9) <.001 8.23 (2.92-23.22) <.001
TIMP-2 12.50 (3.98-41.82) <.001 14.31 (3.83-53.39) <.001
TIMP-3 0.94 (0.29-3.02) .92 0.75 (0.23-2.47) .64
TIMP-4 1.01 (0.62-1.63) .98 1.00 (0.59-1.70) >.99

Abbreviations: aOR, adjusted odds ratio; MMP, matrix metalloproteinase; OR, odds ratio; TIMP, tissue inhibitor of matrix metalloproteinases.

a

Multivariable conditional logistic regression models were used to study the association of biomarker with treatment outcomes (unfavorable) and are adjusted for age in years, sex, body mass index, diabetes status, smoking status, alcohol status, and smear grading.