JAK inhibitors are the current standard of care in myelofibrosis, but many do not address and may worsen anemia; thus, anemia-related responses have traditionally been overlooked as efficacy end points in pivotal clinical trials, leading to a lack of consistency and analytic detail in their reporting. Here we apply our experiences in the phase III trials of momelotinib, a JAK1/JAK2/ACVR1 inhibitor and the first therapy indicated by the US FDA for myelofibrosis patients with anemia, to highlight how application of different criteria impacts the anemia-related benefits reported for any potential treatment in myelofibrosis. We advocate for a convention of a new expert consensus panel to bring consistency and transparency to the definition of anemia-related response in myelofibrosis.
What is this Perspective about? Anemia (too few healthy red blood cells) is common in patients with myelofibrosis. While it is becoming more common to measure the anemia benefits associated with potential treatments for myelofibrosis in clinical trials, different definitions of anemia benefit are available. This Perspective reviews these definitions, the differences between them, and why consistency and clarity in measuring anemia benefit matter. The definitions used in clinical trials of momelotinib, a treatment for patients with myelofibrosis and anemia, are also explained to show how the anemia benefit observed in these trials could have changed if different definitions were used. What does this Perspective show? Definitions of anemia benefit may include the number of red blood cell transfusions a patient receives, the amount of hemoglobin (a red blood cell protein) in their blood, or a combination thereof. Considerations such as timing, the types of patients included, and other factors are not consistent across definitions and not always clearly reported. Results when different definitions of anemia benefit were followed in the momelotinib clinical trials show that the amount of benefit observed with treatments changes depending on which definition is used. What conclusions can be drawn from this Perspective? More consistency and clarity in the definitions of anemia benefit in myelofibrosis clinical trials are needed, suggesting that a new panel of experts should come together to discuss this topic.

Current WHO serum ferritin (SF) thresholds for iron deficiency (ID) in children (<12 μg/L) and women (<15 μg/L) are derived from expert opinion based on radiometric assays in use decades ago. Using a contemporary immunoturbidimetry assay, higher thresholds (children, <20 μg/L; women, <25 μg/L) were identified from physiologically based analyses.
We examined relationships of SF measured using an immunoradiometric assay from the era of expert opinion with 2 independently measured indicators of ID, hemoglobin (Hb) and erythrocyte zinc protoporphyrin (eZnPP), using data from the Third National Health and Nutrition Examination Survey (NHANES III, 1988-1994). The SF at which circulating Hb begins to decrease and eZnPP begins to increase provides a physiological basis for identifying the onset of iron-deficient erythropoiesis.
We analyzed NHANES III cross-sectional data from 2616 apparently healthy children, aged 12-59 mo, and 4639 apparently healthy nonpregnant women, aged 15-49 y. We used restricted cubic spline regression models to determine SF thresholds for ID.
SF thresholds identified by Hb and eZnPP did not differ significantly in children, 21.2 μg/L (95% confidence interval: 18.5, 26.5) and 18.7 μg/L (17.9, 19.7), and, in women, were similar although significantly different, 24.8 μg/L (23.4, 26.9) and 22.5 μg/L (21.7, 23.3).
These NHANES results suggest that physiologically based SF thresholds are higher than the thresholds from expert opinion established during the same era. SF thresholds found using physiological indicators detect the onset of iron-deficient erythropoiesis, whereas the WHO thresholds identify a later, more severe stage of ID.

Our study aims to provide a synthesis of the best available evidence on the hemoglobin (hgb) red blood cell (RBC) transfusion thresholds in adult traumatic brain injury (TBI) patients, as well as describing the risk factors and outcomes associated with RBC transfusion in this population. A systematic review and meta-analysis was conducted using PubMed, Google Scholar, and Web of Science electronic databases according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines to assess articles discussing RBC transfusion thresholds and describe complications secondary to transfusion in adult TBI patients in the perioperative period. Fifteen articles met search criteria and were reviewed for analysis. Compared to non-transfused, TBI patients who received transfusion tended to be primarily male patients with worse Injury Severity Score (ISS) and Glasgow Coma Scale. Further, the meta-analysis corroborated that transfused TBI patients are older (p = 0.04), have worse ISS scores (p = 0.001), receive more units of RBCs (p = 0.02), and have both higher mortality (p < 0.001) and complication rates (p < 0.0001). There were no differences identified in rates of hypertension, diabetes mellitus, and Abbreviated Injury Scale scores. Additionally, whereas many studies support restrictive (hgb <7 g/dL) transfusion thresholds over liberal (hgb <10 g/dL), our meta-analysis revealed no significant difference in mortality between those thresholds (p = 0.79). Current Class B/C level III evidence predominantly recommends against a liberal transfusion threshold of 10 g/dL for TBI patients (Class B/C level III), but our meta-analysis found no difference in survival between groups. There is evidence suggesting that an intermediate threshold between 7 and 9 g/dL, reflecting the physiological oxygen needs of cerebral tissue, may be worth exploring.

β-Thalassemia is an inherited hematologic disorder caused by various mutations of the β-globin gene, thus resulting in a significant decrease in adult hemoglobin (HbA) production. An increase in fetal hemoglobin (HbF) levels by drug molecules is considered of great potential in β-thalassemia treatment and is expected to counterbalance the impaired production of HbA. In this work, based on a set of 129 experimentally tested biological inhibitors, we developed and validated a computational model for the prediction of K562 functional inhibition, possibly associated with HbF induction. To facilitate future advancements in the field, we incorporated our model into Enalos Cloud Platform, which enabled online access to our computational scheme (http://enalos.insilicotox.com/K562) through a user-friendly interface. This web service is offered to the wider community to promote in silico drug discovery through fast and reliable predictions.