A One Year Case Control Study in Rheumatoid Arthritis Patients Indicates a Prevention of Bone Mineral Density Loss in Both Responders and Non Responders to Infliximab

Hubert Marotte; Beatrice Pallot-Prades; Laurent Grange; Philippe Gaudin; Christian Alexandre; Pierre Miossec

Disclosures

Arthritis Res Ther. 2007;9(3) 

In This Article

Results

Demographic, clinical, biological and BMD characteristics of the 189 RA patients enrolled in the study are summarized in Table 1 . Patients exhibited typical clinical and biological features of RA. The study included 152 women and 37 men (mean ± SD age 52.5 ± 14.2 years). The disease duration was 10.6 ± 9.0 years. A total of 118 patients (62%) were on steroids (mean dose 5.30 ± 5.79 mg/day). The DAS28 score was 5.17 ± 1.07. Thirty-five patients (19%) were on biphosphonates. At baseline, serum PTH and 25-OHD levels were 30.5 ± 17.6 pg/ml and 19.7 ± 10.3 ng/ml, respectively. Serum osteocalcin and serum CTX-I were 19.9 ± 11.8 ng/ml and 446 ± 307 pg/ml, respectively. All of these values were in the normal range. The femoral neck and lumbar BMD values were 0.801 ± 0.159 g/cm2 and 0.911 ± 0.143 g/cm2, respectively. According to the BMD values of these patients, 50% had osteopenia and 20% had osteoporosis.

When the two groups were compared, no significant variation was observed. In particular, the BMD levels were similar at the femoral neck and at the spine between the groups ( Table 1 ).

In the control group, after 1 year of follow up femoral neck BMD decreased from 0.797 ± 0.162 g/cm2 to 0.770 ± 0.162 g/cm2 (-2.5%; P < 0.001; Table 2 ). Similarly, lumbar BMD decreased from 0.896 ± 0.142 g/cm2 to 0.861 ± 0.142 g/cm2 (-3.9%; P < 0.001). Regarding markers of bone turnover, the osteocalcin and CTX-I levels remained the same at baseline and after 1 year. The same stability was observed for 25-OHD and PTH levels.

In the infliximab group, femoral and lumbar BMD values exhibited no change between baseline and 1 year. The femoral neck BMD remained stable, being 0.807 ± 0.156 g/cm2 at baseline and 0.809 ± 0.151 g/cm2 at 1 year (+0.2%; not significant). The lumbar BMD was also stable, being 0.930 ± 0.142 g/cm2 at baseline and 0.928 ± 0.136 g/cm2 at 1 year (-0.2%; not significant). Regarding markers of bone turnover, the osteocalcin and CTX-I levels did not change significantly between baseline and 1 year.

Various factors are known to influence BMD, including age, sex, menopause status and steroid use. In order to examine the specific effect of infliximab within the context of these possible confounding factors, various linear regression models were used ( Table 3 ). A direct effect of infliximab in a simple model was observed, leading to an increase in BMD values of 0.029 g/cm2 (P = 0.001) at the femoral neck and 0.037 g/cm2 (P = 0.001) at the lumbar spine (model 1). The effects of infliximab on BMD remained after stratification for male sex and age (model 2). After stratification for male sex, age and steroid use, the effect of infliximab on BMD values at the lumbar spine remained (0.033 g/cm2; P = 0.001), although the effect was not significant at the femoral neck (0.020 g/cm2; P = 0.03; model 3). Stratification for male sex, age, menopause status, steroid status and DAS28 level (model 4) did not change the association of infliximab treatment with increased BMD at the lumbar spine (0.034 g/cm2; P = 0.001) and at the femoral neck (0.023 g/cm2; P = 0.01). When we added biphosphonate status to the model (model 5), similar effects of infliximab on BMD were observed at the lumbar spine (0.035 g/cm2; P = 0.001) and at the femoral neck (0.023 g/cm2; P = 0.02). Infliximab was found to have similar effects on BMD at the lumbar spine (0.041 g/cm2; P = 0.001) and at the femoral neck (0.028 g/cm2; P = 0.002) when baseline BMD values were taken into account (Model 6).

Sixty-four patients (64%) receiving infliximab were classified as good responders, as defined by an improvement of at least 1.2 in DAS28 score at 1 year. At baseline, no significant difference was observed in BMD at the two sites between clinical responders and nonresponders. The change in lumbar spine BMD was +0.4% (0.890 ± 0.163 g/cm2 at baseline and 0.894 ± 0.154 g/cm2 1 year later) for nonresponders and -0.8% (0.955 ± 0.141 g/cm2 at baseline and 0.947 ± 0.127 g/cm2 1 year later) for responders. Similarly, the change in BMD at the femoral neck was +2.0% (0.765 ± 0.157 g/cm2 at baseline and 0.780 ± 0.174 g/cm2 1 year later) in the nonresponders and -0.4% (0.840 ± 0.142 g/cm2 at baseline and 0.836 ± 0.141 g/cm2 1 year later) in responders (Figure 1). Accordingly, there was no significant difference in BMD between responders and nonresponders. These results indicate that there was bone protection in patients classified as nonresponders, in whom a positive response in terms of RA signs and symptoms could not be demonstrated.

Changes in BMD over 1 year. Changes over 1 year in bone mineral density (BMD), for methotrexate (control) and infliximab groups, are represented in percentages at (a) the femoral neck and (b) lumbar spine. For the infliximab group, changes in BMD are also separated according to the clinical response, defined by an improvement of at least 1.2 in Disease Activity Score (DAS)28 score over 1 year. NS = not significant.

On looking for correlation between bone markers and clinical parameters in the whole population (control and infliximab groups), some interesting correlations were observed. Negative correlations were observed between age and BMD at the lumbar spine and femoral neck (r = -0.279, P < 0.001 and r = -0.398, P < 0.001, respectively). Steroid dose correlated with lower levels of bone formation markers, reflected by osteocalcin level (r = -0.322; P < 0.001). Similarly, when we looked for correlation between BMD values and markers of bone turnover in the whole population, we observed a negative correlation between osteocalcin levels and lumbar spine and femoral neck BMD (r = -0.301, P < 0.001 and r = -0.193, P < 0.001, respectively). However, we observed no association between changes in bone markers and changes in BMD values.

Comments

3090D553-9492-4563-8681-AD288FA52ACE
Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as:

processing....