Bone Mineral Density Remains Stable in Pyruvate Kinase Deficiency Patients Receiving Long-Term Treatment with Mitapivat

Background: Hereditary pyruvate kinase (PK) deficiency results in chronic hemolysis that may lead to reduced bone mineral density (BMD). Mitapivat (AG-348) is an investigational, first-in-class, allosteric activator of PKR that has been shown to improve hemoglobin and other hemolysis markers in non-regularly transfused patients (pts) and reduce transfusion burden in regularly transfused pts. Mitapivat has mild aromatase inhibition effects; however, a recent study of 30 non-regularly transfused pts with PK deficiency showed that mitapivat did not appear to promote progression of BMD abnormalities for up to 56 months (mos). The large-pooled analysis presented here assessed BMD over time in a broader PK deficiency population including both non-regularly transfused and regularly transfused pts receiving long-term treatment with mitapivat.

Methods: This study included pts enrolled in DRIVE-PK (NCT02476916), ACTIVATE (NCT03548220), ACTIVATE-T (NCT03559699), and the long-term extension (LTE) study (NCT03853798). All trials included pts ≥ 18 years (yrs) with a confirmed diagnosis of PK deficiency treated with mitapivat. DRIVE-PK (N = 52) was a Phase (Ph.) 2, global, randomized open-label study. ACTIVATE (N = 80) was a Ph. 3, randomized, double-blind, placebo-controlled study. ACTIVATE-T (N = 27) was a Ph. 3, open-label, single-arm study. Pts in DRIVE-PK and ACTIVATE were not regularly transfused, defined as ≤ 3 and ≤ 4 units of red blood cells in the prior 12 mos, and no transfusions in the prior 4 and 3 mos, respectively. Pts in ACTIVATE-T were regularly transfused, defined as ≥ 6 transfusion episodes in the prior 12 mos. Data from 90 pts who received mitapivat for > 12 mos were pooled from these 3 trials along with the LTE study of pts from ACTIVATE and ACTIVATE-T who elected to continue or crossover (if on placebo) to mitapivat. BMD changes over time were measured using dual-energy X-ray absorptiometry (DXA) scans at 3 locations (total femur [combined femoral neck and total hip], femoral neck, and spine). Based on DXA T-scores, pts were classified according to standard definitions as having normal BMD (≥ -1.0 at all locations), osteopenia (> -2.5 to < -1.0 at ≥ 1 locations), or osteoporosis (≤ -2.5 at ≥ 1 locations). DXA scans were obtained locally for all 3 studies and interpreted locally for DRIVE-PK and centrally for ACTIVATE, ACTIVATE-T, and LTE. Scans were collected at baseline (BL), between week (wk) 24 and 28, every 6 mos through mo 30, and then annually for DRIVE-PK; at BL and wk 24 for ACTIVATE; at BL, wk 16, and wk 40 for ACTIVATE-T; and at BL, every 24 wks until wk 96, and then every 48 wks for the LTE study.

Results: Of 90 pts assessed (median age 34.5 yrs [range 18-71]; 50.0% female), median exposure to mitapivat was 20 mos (range 12.2-64.9). Eight pts (8.9%) had concomitant use of bisphosphonate medications, 8 pts (8.9%) had a prior medical history of fracture, and 6 pts (6.7%) reported fractures or joint injuries during the study period. BL BMD T-scores were collected for 88 pts. At BL, 24 pts (26.7%) had a T-score ≥ -1.0 (normal BMD), 48 pts (53.3%) had a worst T-score > -2.5 to < -1.0 (osteopenia), 16 pts (17.8%) had a worst T-score ≤ -2.5 (osteoporosis), and 2 pts (2.2%) were missing T-score data. At last BMD assessment (data cut-off 03Mar2021), 76 of 88 pts (86.4%) remained stable, 7 pts (7.5%) improved, and 5 pts (5.7%) worsened (Figure 1a). As of the last assessment, 27 (30.0%), 46 (51.1%), and 17 (18.9%) of 90 pts had a T-score indicating normal BMD, osteopenia, and osteoporosis, respectively (Table 1b). Among 24 pts with BL T-scores ≥ -1.0, 21 pts remained stable, and 3 pts showed worsening BMD. Of 48 pts with a BL T-score > -2.5 to < -1.0, 6 pts improved to a T-score indicating normal BMD, 40 pts remained stable, and 2 pts worsened. Of 16 pts with a BL T-score of ≤ -2.5, 1 pt improved to a T-score of > -2.5 to < -1.0.

Conclusions: This large-pooled analysis showed that BMD remained largely stable over time in adult pts with PK deficiency receiving long-term treatment with mitapivat for up to 65 mos. As these pts had a substantial degree of reduced BMD at BL which did not progress, it is hypothesized that mitapivat may halt bone loss in PK deficiency patients via its mechanism of action by decreasing hemolysis and ineffective erythropoiesis and stabilizing iron homeostasis. BMD data will continue to be monitored as part of the ongoing LTE study to further understand this process.

Figure 1

View largeDownload slide

Figure 1

View largeDownload slide

Close modal