Hydroxyurea is a widely available disease-modifying therapy for sickle cell disease (SCD), but its effectiveness is currently limited by inadequate utilization, and less than optimal response. Research is needed to improve adherence to this evidence-based therapy and emphasis needs to be placed on determining whether therapy with hydroxyurea can prevent or even reverse organ dysfunction. In addition, research identifying new adjunct therapies to blood transfusion and hydroxyurea, as well as disease-specific therapies for co-morbidities such as kidney disease, hypertension, obstructive lung disease, and pulmonary hypertension will be valuable in the management and treatment of SCD.

Blood is the most prescribed drug in the PICU, and 95% of all neonates are transfused during their stay in the NICU. Currently, there are no evidence-based guidelines for the optimal hemoglobin levels or platelet counts for these populations. There is a balance that must be achieved between hemostasis and thrombosis for this vulnerable population.

Clinical trials have begun to assess the optimal hemoglobin levels in neonates, but there are no trial to asses the optimal platelet count. Neonatologists, pediatric intensivists, and transfusion medicine physicians are beginning to come together to work on solutions to these problems.

Bleeding is frequently encountered in the ICU and is associated with substantial morbidity and associated mortality. When bleeding occurs, coagulopathy is often present and the optimal coagulation factor management regimen remains a matter of debate. Traditionally (and presently in the US), plasma transfusion has been a cornerstone therapy for the replacement of coagulation factor content in this setting. Moreover, recent evidence supporting the use of plasma in the setting of trauma-related hemorrhage seems to have also generated a renewed enthusiasm for plasma transfusion in other critical care settings.

In many locations, there is interest in alternatives to plasma transfusion such as four-factor prothrombin complex concentrates (PCC4) for ICU patients with bleeding. In some locations, factor concentrates have entirely replaced plasma transfusion. However, evidence regarding the benefits and risks of PCC4 versus plasma in ICU patients is lacking. Therefore, we would aim to study the comparative efficacy and risks of a hemostatic strategy relying on PCC4 versus plasma for ICU patients with coagulopathy and bleeding.

Coagulopathy and associated bleeding are common in the intensive care unit environment. Therefore, we believe a multicenter clinical trial evaluating the knowledge gap identified above would be feasible with NHLBI support. Of note, due to the labeled contraindication of disseminated intravascular coagulation for PCC4, such patients would need to be excluded from this trial. Similarly, coagulation abnormalities resulting from congenital coagulation factor deficiencies for which there is a specific coagulation factor product available would also be excluded.

Notably, improved management of coagulopathic bleeding has the potential to impact both clinical outcomes and healthcare resource utilization. Therefore, the outcomes of a trial addressing this knowledge gap would include patient-important outcomes (e.g. mortality, length of hospital stay, bleeding, transfusion-related respiratory complications, thromboembolic complications) as well as outcomes related to healthcare utilization (e.g. product cost, total blood products consumed, interventions required to achieve hemostasis such as surgery or embolization).

PLT count is almost always the only laboratory result considered in deciding to transfuse PLTs. But PLT counts provide no information about PLT hemostatic function and its contribution to bleeding. A variety of in vitro coagulation tests have been developed: viscoelastography, whole blood PLT aggregometry, etc. But while testing-based transfusion algorithms may reduce blood product utilization, it has not been established that any in vitro test can either predict or help reduce bleeding. There is a gold standard method to assess clinical efficacy of transfused PLTs: incidence of grade 2 or higher bleeding in clinical trials of thrombocytopenic hematology-oncology patients receiving prophylactic PLT transfusions. No analogous gold standard of PLT hemostatic efficacy exists for therapeutic PLT transfusions to treat active bleeding. There is a pressing need to develop such a standard. Establishing reliable methods for evaluating the effects of PLT transfusion in actively bleeding patients will improve our understanding of how different factors (storage conditions, pathogen reduction etc.) affect the functional performance of PLTs.

PLT transfusions are administered routinely to support bleeding pediatric and adult medical and surgical patients. Opportunities to conduct clinical trials in various settings (cardiac surgery, neurosurgery, orthopedic surgery, trauma, etc.) are widely available. PLT transfusion is commonly used to support bleeding patients receiving perioperative supportive therapies such as extracorporeal membrane oxygenation (ECMO). These clinical situations represent critical opportunities to improve the care of bleeding patients. This approach will simultaneously facilitate comprehensive evaluation and validation of both current and novel in vitro tests of hemostasis. If a given in vitro test were reproducibly shown to correlate strongly with bleeding reduction caused by PLT transfusion, then by definition that would be a clinically meaningful test. Finally, this line of inquiry will allow assessment of the adverse effects of PLT transfusion in bleeding patients.

Cancer patients undergo intensive medical and surgical therapies to treat their underlying disease. Treatment commonly results in anemia requiring RBC and platelet transfusions to support the patient through the hypoproliferative phase of chemotherapy. This is particularly true for those patients requiring hematopoetic stem cell transplantation (HSCT). Following therapy, cancer outpatients commonly receive RBC transfusions for weeks to months to maintain their functional status.

Common causes of death in patients with hematological malignancies and other cancers are infections and bleeding. A meta-analysis of clinical trials suggested that liberal transfusion is associated with greater risk of infection. Conversely, restrictive transfusion could adversely affect quality of life and functional status in oncology populations. In addition, pre-clinical and clinical studies support that concomitant anemia and thrombocytopenia significantly compound bleeding risk, and that hemostasis can be optimized in thrombocytopenia by maintaining a higher hematocrit. Although bleeding risks in relation to platelet transfusion thresholds are well studied in patients with hematological malignancy, optimal hemoglobin levels in thrombocytopenic patients are not known. Despite the significant allocation of blood components to cancer patients as a whole, RBC transfusion practices are not well studied within this group.

Randomized controlled clinical trials and other studies investigating optimal transfusion thresholds and other measures of practice are required to provide health care providers with evidence to guide one of the most common therapies administered in the setting of malignancy. The clinically important end points of well-designed studies could include: 1) quality of life and functional status for both inpatients and outpatients; 2) neurocognitive development in pediatric populations; 3) bleeding events / bleeding scores; 4) impact on immunity including immunomodulation and infection; 5) reconstitution of hematopoiesis; and 6) survival and/or recurrence of disease.. Besides a generalizable study population, certain target populations of interest are those with high risk disease, HSCT patients, patients undergoing radiation therapy, and pediatric patients.

There are >1.6 million new cases of cancer annually in the USA, including >50,000 with leukemia and >6,000 with HSCT. Cancer therapies are rapidly advancing in the era of genomics and immunotherapy. Capitalizing on the tradition of research in cancer, single and multicenter studies of RBC transfusion are feasible using randomized controlled designs in conjunction with clinical trials of chemotherapeutic regimens. The results of these studies will impact a large patient population’s quality of life, and may ultimately impact healthcare cost and blood demand.

The reprogramming of adult stem cells has resulted in the generation of induced pluripotent stem cells (iPSCs) that can develop into any tissue of the body. These iPSCs ultimately may be used as a transplantable source of stem cells for a variety of hematologic diseases. Although this technology has enabled the generation of patient-specific or disease-specific stem cells that are also amenable to genetic manipulation, the major scientific hurdle has been the ability to create clinically meaningful functional blood products, including transplantable HSCs from differentiating iPSCs. The production of clinically functional blood products -- i.e. red blood cells derived from autologous iPSCs --could replace allogeneic products in highly immunized patients and the generation of megakaryocytes for patient-specific platelet production from iPSCs could drive significant progress in this area. Furthermore, disease-specific iPSCs could serve as targets for both drug development and drug screening in patients with rare hematologic disorders. In addition, support for scale-up and GMP processes, which are difficult to fund via the R01 mechanism will require specific grant opportunities tailored to infrastructure and process development.

Transfusion of RBC is major adjunct in the management of trauma, acute and chronic illness. Issues in blood transfusion include availability of donors, RBC typing and crossmatching, cold storage of donor cells, and limited viability of stored RBC. Globally, in many situations where blood is critically needed, these systems are not available.
An increasing percentage of people with SCD require regular RBC transfusion to prevent stroke and other major complications. In addition, RBC transfusion is employed repeatedly in the management of serious acute complications of SCD. Transfusion of normal RBC to replace or supplement the patient’s defective RBC is the most effective intervention in the management of SCD.

Impacts:
• Molecular genotyping of RBC will reduce alloimmunization.
• Use of new generation HBOCs that do not require blood typing, crossmatching, refrigeration, and that do not transmit infection, would save lives in conditions of severe hemorrhage, stroke, possibly heart attack, especially where there is no immediate access to adequate medical facilities.
• In SCD, HBOCs could prevent pain or reduce its severity and duration, prevent stroke, reduce severity of acute chest syndrome, and other vasoocclusive complications. Finally, HBOCs have the potential to alter the pathogenesis of SCD.

The problems in managing chronic RBC transfusion in SCD remain the same as they have been for decades: all immunization, iron overload, and infection transmission. It is clear that traditional serological RBC phenotyping is unable to detect several variants of RBC antigens, especially those in the Rh system, in populations of African descent. This leads to erroneous phenotyping and the appearance of “auto antibodies” that are truly alloantibodies resulting from transfusion of mismatched blood. As a result, people with SCD are the most frequent users of the American Rare Donor Program.

Blood transfusion is the most commonly employed therapeutic procedure in the US affecting about 5 million patients per year. Having evidence-based transfusion practice guidelines would result in optimized clinical management of patients who need blood transfusions (red blood cells, platelets, plasma) whether for specific conditions (e.g., acute coronary syndromes, sickle cell disease, prenatally) or in specific age groups. Factors that need further evaluation include transfusion trigger in various patient populations, component type and volume transfused, storage age of blood products, patient blood management tools, donor factors that impact on product characteristics, and component processing and storage.

It is increasingly being recognized that much of the transfusion practice has been empirical, without vigorous scientific evaluation, and thus should be evaluated and optimized. For example, it is not clear if patients with acute coronary syndrome would benefit more from a liberal than a more restrictive red blood cell transfusion strategy; and it is not clear whether plasma transfusion benefits patients with slightly elevated INR (international normalized ratios). Additionally, research capabilities both at the molecular level and at the population level have now made such scientific evaluations possible.

Millions of plasma units continue to be transfused to non-bleeding critically ill patients. This practice persists despite the lack of high quality evidence indicating improved patient outcomes with prophylactic plasma transfusion triggered by the INR. Studies testing restrictive versus liberal transfusion triggers are now well-established in academic medicine for Red Blood Cell products. More recently, multicenter clinical trials of prophylactic platelet transfusion vs no-prophylaxis have been conducted as well. However, there remains substantial equipoise on the topic of prophylactic plasma transfusion. A large definitive clinical trial that aims to identify the INR threshold at which prophylactic plasma transfusion prevents bleeding complications in non-bleeding critically ill patients would impact clinical practice in a very meaningful way.

Abnormal coagulation tests are common in the ICU and plasma is frequently administered in this specific clinical setting. Indeed, more plasma is administered in the ICU environment than in any other clinical area. Therefore, a multi-center clinical trial addressing the knowledge gap identified above would be feasible with NHLBI support.

Importantly, healthcare providers become increasingly uncomfortable with the avoidance of coagulation factor replacement as coagulation screening test results drift further from the normal range. To address this concern, an adaptive clinical trial may be desirable. In this design, a targeted cohort with modest elevations in the INR measurement would be screened for enrollment (e.g. INR results ranging from 1.5 - 2.0). If the avoidance of plasma transfusion is noted to be safe, the study would advance to the next stage of recruitment, targeting a higher range of INR values (e.g. 2.0 - 2.5).

Finally, the value of the INR in predicting bleeding complications is increasingly scrutinized. Therefore, it can be argued that the INR is not the ideal "trigger" for plasma transfusion. Despite this, it must be acknowledged that the INR remains the primary driver of decisions related to plasma transfusion. Therefore, a large definitive trial specifically detailing the the efficacy of the current practice versus a more conservative approach to prophylactic plasma transfusion would be practice changing and potentially very transformative.

The PT/INR are frequently used to guide hemostatic interventions, particularly in the ICU where the risk for hemorrhage is increased. In part, this is the result of long standing misconceptions regarding the relationship between PT/INR values with in vivo coagulation factor concentrations and clinical coagulation or hemostasis. Indeed, this misinformation almost certainly contributes to the well-documented inappropriate use of plasma products. To this point, it has now been demonstrated in a variety of clinical settings that the PT/INR is often poorly predictive of clinical coagulopathy or peri-procedural bleeding complications.

The last decade has seen a resurgence of interest in viscoelastic tests as alternatives assessments of hemostasis and guides to blood component use. Indeed, a body of literature supports the potential value of these tests in specific clinical settings. However, the value of such testing strategies has not been well described in the setting of the intensive care unit. Ongoing equipoise regarding the optimal assessment of coagulation status in the ICU suggests that a well conducted trial in this domain would have the potential to be truly transformative.

The assessment of coagulation status and decisions related to plasma transfusion occurs multiple times throughout the day in essentially every intensive care unit across the US. Therefore, an adequately sized study population is clearly present and therefore, a multicenter trial is feasible with NHLBI support.

A potential challenge with a study evaluating the optimal coagulation testing strategy is the incomplete penetration of viscoelastic testing in current clinical practice raising concerns related to the resource and educational requirements if such a trial were pursued. Additionally, viscoelastic testing can take several forms, each of which may impose specific confounders. Therefore, specific technologies and transfusion algorithms would likely need to be employed.

Importantly, innovative trial designs may also help us to address some of the feasibility concerns noted above. One such design that may prove valuable in this setting is the stepped-wedge cluster randomization design. Implementation of such a design might be expected to improve investigator/clinician cooperation/compliance and may help reduce treatment group contamination.

Limitations of current storage methods challenge our ability to meet the increasing demands of cancer chemotherapy and complex surgeries as well as provide platelets to remote areas where trauma frequently occurs. Platelets are transfused to prevent or control bleeding without robust translational or clinical trial evidence of efficacy and safety. In patients with hypoproliferative thrombocytopenia receiving chronic platelet transfusion predominantly for prophylaxis, post-transfusion increments, survival time, and durable hemostatic efficacy are important; however, platelet transfusion alone does not prevent bleeding. The overall state of hemostasis, the endothelium and donor characteristics that affect platelet quality may be important considerations in selecting the best platelet product tailored to individual patients. Better laboratory and clinical measures of platelet function, efficacy and safety are essential for best use of these limited resources. These outcomes should span and link in-vitro testing, animal models, and clinical effects such as thrombotic events, immune, hemostatic, and endothelial effects. Extrinsic, recipient factors (e.g., immune deficiency, platelet or endothelial dysfunction) and intrinsic factors (e.g., donor specific or platelet preparation) should be considered. Systematic analysis of recipients and donors should be broad and include considerations of glycomics, metabolomics, proteomics, genomics, in vivo microscopy, and advanced imaging.

Radiolabeled platelet recovery and survival methods are well established and feasible to perform at several centers within the US. Currently 4-6 million units of platelets and whole blood are transfused per year. It is very feasible to conduct clinical trials that compare methods of platelet preparation, processing, and storage in children and adults who require platelets for either prophylaxis or active bleeding. Multiple research networks that focus on both pediatric and adult transfusion medicine are motivated to perform the pre-clinical and clinical trials required to establish improved efficacy and safety of platelet-containing blood products.

Perioperative hemorrhage remains a significant concern. Coagulopathy frequently develops in the setting of surgical hemorrhage and the optimal strategy for addressing this issue remains a matter of debate.
Military experience suggests the application of ratio-based plasma transfusion practices may improve clinical outcomes in the setting of massive hemorrhage. However, numerous concerns (e.g. survival bias, uniqueness of a military population, risk for adverse events related to high-volume plasma transfusion) preclude the broad generalization of such strategies to non-trauma surgical populations.
Goal-directed hemostatic resuscitation strategies (e.g. coagulation management based upon the results of hemostasis testing) have shown promise in specific surgical environments such as cardiac surgery and liver transplantation. However, definitive direct comparisons between goal-directed strategies and alternative approaches such as fixed-ratio plasma transfusion have not been performed. Moreover, the role of such strategies in more heterogeneous surgical populations remains uncertain. To better define the optimal approach to plasma transfusion strategies in the setting of intraoperative (non-trauma) hemorrhage, we propose a multicenter clinical trial directly comparing fixed-ratio and goal-directed plasma resuscitation strategies in the operating room environment.

Surgical patients with increased risk of hemorrhage may be targeted to improve subject recruitment (e.g. cardiac surgery, liver transplantation, aortic vascular surgery, multi-level instrumented spinal fusions or tumor resections, etc). Therefore, an adequately sized study population is likely to be present in a multicenter study and we believe such a trial would be feasible with NHLBI support.

To further enhance the feasibility of identifying a surgical population experiencing significant hemorrhage in a time-efficient manner, predictive algorithms such as the critical RBC administration threshold (CAT) could be also be employed.

The optimization of decisions related to plasma transfusion has the potential to improve not only clinical outcomes, but also healthcare resource utilization. Therefore, the outcomes of a trial in this domain could include both patient-important outcomes (e.g. mortality, bleeding complications), as well as outcomes related to healthcare utilization (e.g. total blood products consumed, hospital length of stay).

 Clinical trial data show that in multiple patient populations, a 7 to 8 g/dL hemoglobin threshold for red blood cell transfusion is safe. However, equipoise for transfusion thresholds persists in patients with ischemic heart disease and acute neurological conditions where a higher hemoglobin level may decrease ischemic injury to myocardial and cerebral tissues, respectively. Alternatively, liberal red blood cell transfusion may be associated with adverse outcomes such as pulmonary edema or increased rates of heart failure. In other populations, it is possible that even lower transfusion thresholds than 7 g/dL would be safely tolerated. On the other hand, if a 7 g/dL threshold is found to be superior to a lower threshold, this would establish a minimal appropriate threshold to initiate red blood cell transfusion.

Multicenter clinical trials that utilize markers of oxygen consumption are needed to answer these questions. A pilot study in patients with acute myocardial infarction demonstrated that recruitment was feasible for clinical scenarios in which equipoise exists. An international setting where the safety and availability of the blood supply remain significant issues may be best suited to answer the question of whether transfusion thresholds lower than 7 g/dL are safe.

The changes in red blood cells during refrigerated storage have been well documented and associated with negative clinical sequelae in the peer reviewed literature. While the impact of this so-called storage lesion does not impact every patient during every transfusion it is reasonable to expect that when a unit of blood is transfused to a particularly vulnerable patient from a donor that has red blood cells pre-disposed to degradation, stored in a manner that has allowed significant change to occur, the risk of a negative clinical sequelae is increased. In this case it will be important to understand what underlies the likelihood of a donors blood to store poorly, the changes that occur during storage that could impact vulnerable patients and design approaches to mitigate the degradation that could result.

We believe mitigating the impact of the storage lesion is feasible by reducing and controlling the oxygen concentration in the RBC unit prior to refrigerated storage. We are continuing our development of a device to do this and to generate the data demonstrating the effect of deoxygenation and anaerobic storage.