What ever happened to blood substitutes?

Remember the mid-1980s? The donor questionnaires were exploding as we added new and unbelievably nosy questions. We began excluding new demographic sectors of our former donors and everyone was worried about the shortages that the intensified screening and testing would create. (Nice to know that some things don’t change!) At the same time, both the medical and the general public were in a state of near panic about the infectious hazards of transfusion. Autologous donation programs grew exponentially and we brought expensive devices into the operating rooms to scavenge every last milliliter of shed blood. The only good transfusion was the one avoided! Against this background of concern about the adequacy and safety of the blood supply in the mid-1980s, 11 companies, from one-product start-ups to industry giants, launched the development and testing of blood substitutes based on perfluorocarbons or modified hemoglobin (Hb). Many of these products entered human clinical trials in the 1990s. But today, 20 years later, none of the perfluorocarbon products, and only a handful of the Hb-based oxygen carriers (HBOCs), are still being tested clinically, and none of them has been licensed for human use in the US, Canada, or Europe. Whatever happened? This issue of TRANSFUSION features a review by Buehler and Alayash that helps us understand the journey that the candidate “blood substitutes” are attempting to make from concept to clinic. In particular, they take a step back and look critically at the preclinical testing of the HBOCs and the toxicities that brought several of them to grief in clinical trials. They offer several explanations for the discrepancies between the promising preclinical studies and the disappointing clinical trials, but one comment is particularly telling: “. . . preclinical models were actually predictive of clinical outcome . . . but were not fully understood due to the novelty of the product.” When work on the HBOCs began in the 1980s, the goals were to produce a material with a number of key characteristics, among them, better shelf stability than banked red cells (RBCs), useful vascular t1/2, absence of infectious agents, avoidance of the known toxicities related to residual stroma and renal impairment, and replication of the oxygen delivery behavior of erythrocytic Hb. The manufacturers have in fact been quite successful in achieving the first four goals. The HBOCs under development all have vascular 1/2 values in the 18- to 24-hr range and can be stored at 4°C or room temperature for at least 1 year. All of them have been successfully processed to eliminate the presence of microorganisms (although the jury is still out with respect to prion removal). None of them produce the acute renal injury seen when unmodified Hb is present in the vascular space. Replicating the oxygen delivery behavior of the RBC has proven to be a tall order, however. In the 1980s, this appeared to be relatively straightforward problem. All we had to do was to obtain a product with an O2 binding capacity and a P50 similar to that of native Hb. Most of the HBOCs were therefore designed or modified to achieve this end. But what we found out over the next several years was that although the Hb concentration, P50, and Hill coefficient may have been adequate descriptors of the oxygen delivering properties of intraerythrocytic Hb, the rules were different when Hb was removed from the RBC. The work with the HBOCs not only challenged our assumptions about oxygen delivery, but indicated the importance of hitherto overlooked factors that affect their performance in vivo. We can start with the P50. The working assumption was that a Hb-based oxygen carrier with a P50 similar to that of Hb in the RBC (around 28 torr) would deliver oxygen better than a low-P50 (high-affinity) product. That may not always be the case, however, particularly with extraerythrocytic Hb. The picture of oxygen delivery and its regulation that has emerged in the past 10 years is more dynamic and complex than what we understood previously. It appears that tissues, perhaps the vascular endothelium at the level of the precapillary arterioles, sense the local oxygen supply and respond by regulating regional blood flow. This autoregulatory theory1 predicts that low-affinity Hbs, which would tend to “dump” their O2 load at the relatively high ambient PO2 levels of the precapillary arterioles, might trigger vasoconstriction and limit local blood flow, hence impairing downstream tissue oxygenation. By contrast, a stingier, low-P50 Hb would tend to retain its O2 until it encountered lower PO2 levels such as those found in the capillaries, thereby slipping past the arteriolar gate keepers and delivering its O2 load through the porous capillary endothelium to the tissues. Although this model is attractive and turns the received wisdom about oxygen delivery on its ear, P50 is only one of the players, and maybe even a bit one, in this story. It has also been suggested that ability of the HBOC to facilitate or hinder the diffusion of O2 through the vascular space to the surface of the endothelial cells may be a key factor.2 The diffusion constant of a macromolecule such as oxyhemoglobin is inversely related to its molecular radius; thus large molecules have lower diffusion constants than small molecules. Therefore, a HBOC with a large molecular radius (e.g., a polymerized oxyhemoglobin or one that is conjugated to several long polymers of polyethyleneglycol) would tend to retard the diffusion of oxygen relative to unadorned, tetrameric oxyhemoglobin. The relatively sluggish diffusion of O2 from large HBOCs would also have the effect of delaying its delivery until it had reached the capillaries, again defeating the autoregulatory mechanisms. Consistent with this hypothesis, there is evidence in model systems that large-radius HBOCs with low diffusion constants selectively deliver O2 at the level of the capillaries, rather than the arterioles.3 In addition to this autoregulatory model, other theories have been proposed to explain the vasoconstrictive effects noted with many (though not all) of the HBOCs. Hb is an avid scavenger of nitric oxide (NO), the vasodilator constitutively released by vascular endothelial cells. Nevertheless, the imperfect correlation between NO binding and vasoconstrictive properties of different Hb-based oxygen carriers suggests that this property of Hb may play a relatively small part in its vasoactivity.4 Direct toxic effects of the oxidation products of Hb on the vascular endothelium may also be a factor.5 Some of the physical characteristics of modified Hbs may be as important as its chemical properties.6 Hb solutions, particularly at the concentration of many of the HBOCs (approx., 10 g/dL), exhibit high colloid oncotic pressure, which directly affects intravascular volume and cardiac output. Colloid oncotic pressure in turn is related to the amount of protein and the number of molecules (which is reduced by polymerization if the protein concentration is held constant) and is affected by the charge of the protein and its propensity to interact with solvent water. The different HBOCs vary markedly in degree of polymerization, effective molecular radius, and tendency to bind water and therefore colloid oncotic pressure. Their variable volume expansion properties, and effects on blood flow, may also affect their oxygen delivery characteristics. The effect of the low viscosity of Hb solutions (relative to RBCs) is more straightforward, if counterintuitive. Although it would seem that the addition of a low-viscosity oxygen carrier to the vascular space might increase blood flow (cardiac output) due to reduced resistance (systemic vascular resistance), in fact the opposite appears to occur, at least at the level of the microvasculature. Endothelial cells are sensitive to shear stress, a property of a moving fluid that is directly related to viscosity. As the viscosity of a moving fluid decreases, so too does the shear stress. In response to decreased shear stress, endothelial cells down regulate the production of the vasodilators endothelin-1 and NO. The influence of viscosity on the vasoconstrictive response has been demonstrated in animal model systems.7 Despite the fact that these oxygen carriers are based on modified Hb, they are not in any sense “just like blood.” As pointed out by Buehler and Alayash, this message was brought home forcefully when the toxicities of these preparations began to manifest themselves, but was also signaled by the early preclinical and in vitro work. The development of the candidate blood substitutes has been a fascinating and at times frustrating journey. The transfusion medicine community will continue to follow developments in this field with active interest. Although we may still be several years away from having a product on our shelves, the insights into oxygen delivery and its regulation provided by HBOC research have been priceless.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.