Funding

Clinical Research

Updated September 10, 2007

Description of Doxycycline Trial Biospecimens

Aliquots of serial biomarker samples (urine and EDTA plasma) are available for analyses from 2 cohorts of subjects with knee OA that were evaluated in recent NIH-funded studies: (1) the 431 subjects who participated in the randomized multicenter, placebo-controlled trial of doxycycline (doxy) and (2) an observational cohort of 253 subjects with knee OA who were recruited from a variety of community and clinic sources in central Indiana for a single-center study of the natural history of knee OA. For each subject, the biomarker material is accompanied by the demographic and clinical data (e.g. WOMAC, concomitant medications) in the case report form and the radiographic analyses of the data (e.g., K and L grade, quantitative measurements of joint space width, atlas-based gradings of joint space width and osteophytes). The radiographic images themselves will be digitized by NIH but digitized images are not yet available.

From the doxy cohort, samples of plasma and urine were obtained at baseline and every 6 months thereafter, for 30 months. From the progression cohort, biomarker samples were obtained in identical fashion at baseline, 16 months and 30 months. All samples have been maintained at -70°C.

Subjects in both cohorts underwent serial knee radiography, performed with the semi-flexed AP view, with correction for radiographic magnification, and quantitative measurements of medial tibiofemoral joint space narrowing. In both cohorts, lateral views of the knee and sunrise (Hughston) views of the patellofemoral compartment were also obtained. In both cohorts, knee x-rays were obtained at baseline, 16 months and 30 months.

Eligibility criteria for the doxy study required that the subject was female, between 45 and 64 years old, in the upper tertile of the population for body mass index and in a conventional standing AP view of the knees, exhibited a K and L grade of 2-3 in the index knee and 0-1 in the contralateral knee.

The OA progression study enrolled subjects of both sexes, at least 45 years old, who had mild-to-moderate radiographic knee OA, based on the presence of a marginal tibiofemoral osteophyte in either a standing AP or semiflexed AP view and JSW ≥ 2.0 mm in the semiflexed AP view.

The following table provides baseline characteristics of the 2 cohorts.

Characteristics
Doxy
Progression Study
Sex, N (%) female 431 (100) 201 (79)
Age, years (mean ± SD) 54.9 ± 5.6 61.6 ± 9.8
BMI, kg/m2 (mean ± SD) 36.7 ± 6.2 33.0 ± 7.9
Race, (%) African-American 66 (15) 62 (25)
Dur'n of symptoms, years, (mean ± SD) 6.5 ± 8.0 8.4 ± 8.9
Overall severity of tibiofemoral OA*    
   Grade 0
   Grade 2
   Grade 3
   Grade 4		 93 (22)
97 (23)
207 (48)
34 (8)		 14 (6)
31 (12)
192 (76)
14 (6)
WOMAC OA Index    
   Knee pain score (mean ± SD)¹ 10.8 ± 4.2 10.7 ± 4.4
   Function score (mean + SD)² 38.3 ± 13.0 40.9 ± 14.2
Subject retention, 16-mo x-ray, N (%) 381 (88) 220(87)
Subject retention, 30-mo x-ray, N (%) 367 (85) 208 (82)

*=Based upon consensus readings of JSN and osteophyte score in the semiflexed AP view
¹ = range, 5-25
² =range, 17-85

Detailed results of the doxy trial have been published in Arthritis Rheum 52: 2015-25, 2005. In brief, doxy significantly slowed the rate of joint space narrowing in the index knee at both 16 months and 30 months, relative to placebo (40%, 33%, respectively), but did not affect the rate of JSN in the contralateral knee.

A paper in Ann Rheum Dis ( 65: 64-68, 2006) presents a comparison of the utility of quantitative and semiquantitative indicators of progression of JSN in the serial radiographs from the 431 subjects in the doxy trial. Although the two were highly related, the DMOAD effect of doxy was more easily detected with quantitative measurements.

The findings of the progression study are published in Ann Rheum Dis 65: 515-519, 2006. In brief, the results indicated that progression of both osteophtye growth and joint space narrowing were predicted by the severity of the respective radiographic features in the baseline x-ray, and by the presence of tibiofemoral OA.