Change in hemoglobin according to PKLR genotype and hemoglobin response according to PK protein level at baseline in the DRIVEPK trial. (A) The mean change from baseline in the hemoglobin level according to the patient's PKLR genotype category. Of the 52 patients who received mitapivat, 20 (38%)—all of whom had at least 1 missense mutation—had a mean hemoglobin increase >1  g/dL (indicated by a thick gray line); 19 patients had a mean hemoglobin increase ≥1.5  g/dL (thin gray line). A mean hemoglobin increase >1  g/dL did not occur in any of the 5 patients who were homozygous for the R479H mutation (as indicated by an asterisk) or in any of the 10 patients who were homozygous for non-missense mutations (as indicated by a red bar). (B) Hemoglobin response according to PK protein level at baseline. Hemoglobin response is defined as >1  g/dL increase in the hemoglobin at >50% of the assessments during the core period. Baseline pyruvate kinase protein level in red cells has been normalized to the hemoglobin level (an approximation of the total red-cell protein level) and to the pyruvate kinase protein level measured in a healthy control (to allow for interassay comparisons). Patients with an increased baseline level of PK protein were more likely to have a hemoglobin response to mitapivat. The data bars in panels A and B are not aligned for each patient, so a 1:1 comparison of the individual data bars in the 2 panels is not possible. From The New England Journal of Medicine, RF Grace et al., Safety and Efficacy of Mitapivat in Pyruvate Kinase Deficiency, 381(1):933-944. Copyright 2019 Massachusetts Medical Society. Reprinted with permission.