Postmenopausal women with osteoporosis may also have renal insufficiency. We conducted a post hoc analysis of the ARCH study to determine the efficacy and safety of romosozumab (Romo) vs alendronate (ALN) among patients with different levels of baseline renal function.
In ARCH, 4,093 postmenopausal women, 55–90 years old, were randomized 1:1 to receive monthly subcutaneous Romo 210 mg or weekly oral ALN 70 mg for 12 months (double-blind phase [DBP]). Eligible patients had a bone mineral density (BMD) T score of ≤ –2.5 at the total hip (TH) or femoral neck (FN) and either ≥ 1 moderate/severe vertebral fracture (VFx) or ≥ 2 mild VFx; or a T score of ≤ –2.0 at the TH or FN and either ≥ 2 moderate/severe VFx or an Fx of the proximal femur sustained 3–24 months before randomization. Pts were excluded for significantly impaired renal function (eGFR < 35 mL/min/1.73 m2, calculated using the MDRD equation). For the current analysis, patients were categorized by baseline eGFR: normal renal function (eGFR ≥ 90), mild renal insufficiency (eGFR 60–89), or moderate renal insufficiency (eGFR 30–59). The least squares mean (LSM) % change from baseline in BMD at the lumbar spine (LS), TH, and FN; incidence of new VFx; incidence of adverse events (AEs); and changes in renal function were assessed for each eGFR category at month 12 of the DBP.
At baseline, 15% of patients had eGFR ≥ 90, 60% had eGFR 60–89, 24% had eGFR 30–59, and 0.3% had eGFR 15–29. In the overall patient population, LSM % change (95% CI) from baseline in BMD (Romo vs ALN) was 13.7% (13.4–14.0) vs 5.0% (4.7–5.2) for LS, 6.2% (5.9–6.4) vs 2.8% (2.7–3.0) for TH, and 4.9% (4.7–5.2) vs 1.7% (1.5–2.0) for FN (P < 0.001 at each site). Changes in BMD were similar irrespective of baseline eGFR. Among patients with eGFR ≥ 90, eGFR 60–89, and eGFR 30–59, the incidence of new VFx (Romo/ALN) was 3.3%/7.3% (relative risk reduction [RRR] = 57%; 95% CI: 1–81), 3.2%/3.9% (RRR = 19%; 95% CI: -28–49), and 3.4%/6.2% (RRR = 51%; 95% CI: 5–75), respectively. The incidences of AEs, serious AEs, and fatal AEs were similar in both treatment groups within each eGFR category as well as across eGFR categories; there was a higher incidence of positively adjudicated cardiovascular events in the Romo vs ALN group overall and across eGFR categories. One patient in the Romo group with eGFR 60–89 at baseline and 1 in the ALN group with eGFR ≥ 90 at baseline had an AE of mild hypocalcemia. Similar percentages of patients in the Romo and ALN groups had changes in renal function over 12 months of treatment.