Mollison’s Blood Transfusion in Clinical Medicine - part 8 pptx

Transfusion of haematopoietic cells The object of transfusing allogeneic haematopoietic progenitor cells is to establish a permanent graft of transfused progenitor cells in the recipient. The fate of allogeneic progenitor cells infused into the venous circulation depends on their ability to traffic to sites of haematopoietic tolerance (‘microenvironment’) and on managing two immunological phenomena: (1) the rejec- tion of donor progenitor cells by the host immuno- logical response and (2) an immunological reaction of grafted immunologically competent cells against the host: GvHD. Both of these reactions depend on the degree of histocompatibility between donor and recipi- ent and also on the immunological competence of the recipient. Engraftment and kinetics also depend on patient age, disease status, the preparative regimen, GvHD prophylaxis and the cellular content of the graft. Bone marrow was the original source of progenitor cells for haematopoietic grafting, but mobilized peripheral blood and cord blood have gradually sup- planted marrow as a source of PC. The engraftment potential of the component is commonly designated in terms of mononuclear cells that express the CD34 antigen, the cluster designation of a transmembrane glycoprotein present on haematopoietic progenitor cells (Krause et al. 1996), although accessory cells in the graft clearly play an important role (Ash et al. 1991). Cells that express CD34 include lineage-committed haematopoietic progenitors, multipotent progenitors and possibly pluripotent stem cells as well. Flow cytometric assays are used to quantify CD34 + cells in both the donor and the component. However, problems with interlaboratory accuracy and repro- ducibility, especially of different PC sources, have been notorious, even with the adoption of a standardized technique (Sutherland et al. 1996; Keeney et al. 1998). Peripheral blood-derived progenitor cells Peripheral blood-derived progenitor cells (PBPCs) were reported to circulate in mammalian blood as early as 1909 (Maximow 1909), and the ability of cir- culating cells to repopulate a lethally irradiated animal was demonstrated in a parabiotic rat model in 1951 (Brecher and Cronkite 1951). However, circulating haematopoietic progenitor cells were not confirmed in human blood until the 1970s (McCredie et al. 1971). Collection of PBPCs obtained from peripheral blood by leucapheresis (see Chapter 17) has now all but replaced infusion of bone marrow. PBPCs have the advantages of engrafting more rapidly and sparing the donor a general anaesthetic, which result in lower morbidity and cost (Kessinger et al. 1989; Azevedo et al. 1995; Bensinger et al. 1995; Korbling et al. 1995; Schmitz et al. 1995). Allogeneic PBPCs have a higher CD34 + cell content than does marrow, which, inde- pendent of stem cell source, increases patient survival while reducing transplant-related mortality and relapse (Mavroudis et al. 1996; Bahceci et al. 2000; Zaucha et al. 2001). A theoretical argument against the use of PBPCs is the greater number of ‘T’ lymphocytes that contaminate these collections, compared with the number of T cells in bone marrow, suggesting the possibility of an increased risk of severe acute GvHD. Indeed, although the risk of acute GVHD after PBPCs is similar to that observed among historic bone mar- row transplant (BMT) controls (Pavletic et al. 1997; Przepiorka et al. 1997), the probability and severity of chronic GVHD (cGVHD) appear to be increased (Bacigalupo et al. 1996; Flowers et al. 2002). Liquid storage and cryopreservation Collection of allogeneic PBPCs is ordinarily scheduled to coincide with the conclusion of the patient’s preparatory regimen, so that the graft can be infused while fresh. Most centres opt to transfuse the cells as soon as possible. Refrigerated storage of unmanip- ulated mobilized collections at 2–6°C for 24 h and as long as 72 h results in little detectable loss of the in vitro functional properties (Beaujean et al. 1996; Moroff et al. 2004). PBPCs, like bone marrow, can be cryopreserved by slow cooling (1–3°C/min) in the presence of the cryprotectant dimethylsulphoxide (DMSO), variable amounts of plasma, with or without hydroxyethyl starch (Hubel 1997). Grafts can be stored at –80°C, but are usually placed in liquid nitrogen at –140°C or colder, at which engraftment potential is preserved for years. Cryopreserved grafts are thawed in a waterbath at 37–40°C and infused through a 170-µ filter. Prolonged post-thaw storage is inadvisable, as prolonged expos- ure to 10% DMSO may harm the cells. Storage up to 1 h does not reduce viability or colony-forming activity (Rowley and Anderson 1993). Rapid infusion of DMSO has been associated with flushing, nausea, vomiting, diarrhoea and hypotension, probably the CHAPTER 14 628 result of histamine release. Reversible encephalopathy has been reported when doses have approached 2 g/kg, so that caution is advisable when large volumes of PBSCs are thawed (Dhodapkar et al. 1994). The graft can be washed free of cryoprotectant, but progenitor cells may be lost in the process. Cord blood progenitor cells Umbilical cord blood is a rich source of progenitor cells (Knudtzon 1974; Broxmeyer et al. 1989a). The use of cord blood progenitor cells (CBPCs) has import- ant real and potential advantages. The number of donors is unlimited, procurement is easy and inexpens- ive and the cells can be HLA typed and preserved in liquid nitrogen. Human CBPCs with high proliferative capacity and NOD/SCID mouse engrafting ability can be stored frozen for > 15 years, and probably remain effective for clinical transplantation (Broxmeyer et al. 2003). In addition, because many of the functions of the immunologically competent cells in cord blood are not fully developed, the chance of their inducing GvHD appears to be diminished (Szabolcs et al. 2003). Even after the transplantation of CBPCs from unrelated donors, mismatched for two, or as many as three, HLA antigens, the risk of severe GvHD seems to be low (Wagner 1995). Umbilical cord blood banking Umbilical cord blood is collected by either the obstetri- cian or the midwife in utero during the third stage of delivery or ex utero after delivery of the placenta by trained nurses or technologists (Wall et al. 1997; Fraser et al. 1998). Collection volume and cell yield appear to be similar with both methods (Lasky et al. 2002). A maternal blood specimen is screened for markers of transmissible disease, and a sample from the unit is cultured, HLA typed, analysed for cell count, viability and in many instances CD34 + cell number and colony count by culture. Suitable units are processed to remove red cells and plasma and are frozen at a controlled rate and stored in liquid nitrogen (Armitage et al. 1999). Related donor transplants The first successful transplantation of CBPCs from an HLA-identical sibling was given to a patient with Fanconi’s anaemia (Broxmeyer et al. 1989a). In 44 paediatric transplantations of CBPCs from sibling donors, patients receiving HLA-identical or 1-antigen mismatched grafts showed an actuarial probability of engraftment of 85% at 50 days after transplantation; there were no instances of late graft failure (Wagner et al. 1995). The median total nucleated cells per kilogram (TNC/kg) was 5.2 × 10 7 . The probability of GvHD at 100 days post transplant was 3% and of chronic GvHD at 1 year was 6%. The probability of survival with a median follow-up of 1.6 years was 72%. Among 102 children with acute leukaemia transplanted by the Eurocord collaborative investiga- tors, 42 received a graft from a related cord blood donor; 12 of these were HLA mismatched (Locatelli et al. 1999). Nucleated cell dose (> 3.7 × 10 7 /kg) cor- related with engraftment; two-year survival was 41%. Rocha and co-workers (2000) compared 113 recipi- ents of HLA-identical sibling CBPC transplants for malignant disease with records of 2052 siblings trans- planted with bone marrow between 1990 and 1997. Although the umbilical cord blood (UCB) had a significantly longer delay in recovery of neutrophil and platelet reconstitution, no significant difference in survival and a significantly lower risk of GvHD and chronic GvHD was reported in the CBPC group. Bone marrow recipients received nearly 10-fold the total nucleated cells per kilogram body weight (TNC/kg). Of 44 children with non-malignant conditions (thalas- saemia, sickle cell disease), two-year survivals were 79% and 90% respectively (Locatelli et al. 2003). One child with sickle cell disease and seven with thalas- saemia failed to sustain engraftment. Four children suffered acute grade II GvHD. Unrelated donor transplants Several thousand CBPC transplants have been per- formed and more than 100 000 umbilical cord com- ponents are available worldwide. With the growth of public (‘unrelated’) CBPC banks, the number of CBPC transplants from unrelated donors now exceeds that from related donors. In one early series, a high rate of engraftment (23 out of 25 cases) was observed in children infused with allogeneic CBPC despite the donor–recipient pair discordance of 1–3 HLA antigens (Kurtzberg et al. 1996). A retrospective analysis of 562 unrelated CBPC transplants found that engraftment exceeded 80% and survival rate was 61%; pre-freeze THE TRANSFUSION OF PLATELETS, LEUCOCYTES, HAEMATOPOIETIC CELLS AND PLASMA COMPONENTS 629 cell count of the graft ranged from 0.7 to 10 TNC/kg (Rubinstein et al. 1998). The number of nucleatedcord blood cells that were transfused per kilogram of the recipient’s weight emerged as the main influence on engraftment. A retrospective analysis of 537 paediatric CBPC transplants from the Eurocord Registry, includ- ing 138 related transplants and 291 unrelated donors, reported similar results (Gluckman and Locatelli 2000). Laughlin and co-workers (2001) reported CBPC transplants in 68 adult recipients who received a median of 2.1 × 10 7 TNC/kg. TNC number per kilogram correlated with rapidity of engraftment and high CD34 + number was associated with event-free survival. Overall survival (22 months) was 28%. As expected from the experience with bone marrow transplantation, GvHD is significantly higher in the setting of grafts from unrelated donors and depends as well upon the age of the recipient, the degree of histo- compatibility between donor and recipient, the nature of the preparatory regimen and a variety of other fac- tors. In this series, grades III and IV GvHD occurred in 20% and chronic GvHD in 36%. Reconstitution of adult recipients of cord blood CBPCs have so far been used primarily for children and doubt has been expressed as to whether the num- ber of progenitor cells in cord blood from a single donor will be sufficient to repopulate the majority of large adults who require a transplant. Most analyses indicate that the key clinical outcomes (days to neutrophil engraftment, platelet engraftment, severe GvHD and disease-free survival) are all superior in younger patients; age-related outcomes are widely attributed to the number of nucleated cells in a single unit of cord blood (Laughlin et al. 2001). There is, as yet, no quantitative assay for the progenitor cell subset that has the capacity for long-term bone mar- row repopulation. On the other hand, the number of progenitor cells that can be assayed (CFU-GM, CFU- GEMM, etc.) is large enough, suggesting that the number of the more primitive progenitor cells may be sufficient (Broxmeyer 1995). There is evidence that the total cellular content of placental cord blood (PCB) grafts is related to the speed of engraftment, though the total nucleated cell (TNC) dose is not a precise pre- dictor of the time of neutrophil or platelet engraft- ment. It is important to understand the reasons for the quantitative association and to improve the criteria for selecting PCB grafts by using indices that are more precisely predictive of engraftment (Rubinstein et al. 1998). The post-transplant course of 204 patients who received grafts evaluated for haematopoietic colony- forming cell (CFC) content among 562 patients reported previously were analysed using univariate and multivariate life-table techniques to determine whether CFC doses predicted haematopoietic engraft- ment speed and risk for transplant-related events more accurately than the TNC dose. Actuarial times to neutrophil and platelet engraftment were shown to correlate with the cell dose, whether estimated as TNC or CFC per kilogram of recipient’s weight. CFC associ- ation with the day of recovery of 500 neutrophils/µl was stronger than that of the TNC. In multivariate tests of speed of platelet and neutrophil engraftment and of probability of post-transplantation events, the inclusion of CFCs in the model displaced the significance of the high relative risks associated with TNC. The CFC content of PCB units is associated more rigorously with the major covariates of post- transplantation survival than is the TNC and is, therefore, a better index of the haematopoietic content of PCB grafts (Migliaccio et al. 2000). A positive cor- relation between CD34 + cells and circulating day-14 colony counts (CFU-GM) has been reported suggesting that with umbilical cord progenitor cells (UCPCs), as with PBPC, CD34 is a reliable measure of haematopoi- etic potential (Payne et al. 1995; Siena et al. 2000). Data from 102 patients identified CD34 + cell dose as the only factor that correlated with rate of engraftment (Wagner et al. 2002). Studies from Spain and Japan of small numbers of adults with haematological malig- nancies report promising rates of engraftment and disease-free survival (Sanz et al. 2001; Ooi et al. 2004). Progenitor cell expansion If the number of progenitor cells in cord blood proves to be scarcely sufficient for repopulation in many adults, the possibility of expanding the number by culture in vitro has been proposed (Apperley 1994; Broxmeyer et al. 1995) and several groups are devel- oping methods to do so (Kogler et al. 1999; Pecora et al. 2000; Jaroscak et al. 2003). Whether the most important primitive progenitor cells are expanded by culture cannot be established in vitro. As yet, no evidence has confirmed that increase in engraftment kinetics or expansion of stem cells has been achieved, CHAPTER 14 630 and the possibility of increased frequency of GvHD with some expansion methods has been raised (Shpall et al. 2002; Jaroscak et al. 2003). Plasma from cord blood has been found to increase the self-renewal capacity of stem cells in vitro (Carow et al. 1993). Cord blood plasma, but not plasma from adults or fetal calf serum, had this effect and cord blood plasma also increased the expansion in vitro of the number of progenitor cells induced by growth fac- tors (Bertolini et al. 1994). Furthermore, CBPCs fully retain this expansion potential after cryopreservation (Bertolini et al. 1994). Use of multiple cord blood collections Because limited cell dose compromises may comprom- ise the outcome of adult UCB transplants, multiple cord blood units have been combined to augment the dose. Zanjani and co-workers (2000) have trans- planted human UCB from multiple donors in a fetal sheep model. Short-term donor engraftment derived from both donors,but for long-term haematopoiesis, a single donor predominated. Multidonor human UCB transplants using up to 12 units have been published (Ende and Ende 1972; Shen et al. 1994). Weinreb and co-workers (1998) reported that a unit that was partially HLA matched predominated in a recipient who received a combina- tion of 12 units. Another patient with advanced acute lymphocytic leukaemia received a mismatched, unre- lated UCB transplant using units from two donors and achieved a complete remission with double chimerism, which persisted until relapse (De Lima et al. 2002). Barker and co-workers (2005) have augmented graft cell dose by combining two partially HLA-matched units. Twenty-three patients with high-risk haemato- logical malignancy received 2 UCB units (median infused dose, 3.5 × 10 7 NC/kg) and 21 evaluable patients engrafted at a median of 23 days. At day 21, engraftment was derived from both donors in 24% of patients and a single donor in 76% of patients. One unit predominated in all patients by day 100. Although neither nucleated or CD34 + cell doses nor HLA match predicted which unit would predominate, the predominating unit had a significantly higher CD34 + dose. The result is similar to the predominant lymphocyte chimerism that persists in trauma patients who receive multiple blood transfusions (Lee et al. 1999). Law, ethics, related banks and genetic selection Controversy continues regarding the propriety of related (‘private’) CBPCs for which the family pays to have the infant’s cells cryopreserved for future use, as contrasted with unrelated (‘public’) banks, in which donated cords are stored for general use (Burgio et al. 2003). Both systems have their adherents and they should not be mutually exclusive. Although most related banks with commercial origins have sought participation from expectant mothers who agree to pay for storage of a cord from their newborn infant, others have been supported by federal grants (Reed et al. 2001). The infrequent utilization of a related cord blood unit does minimize its utility. The probabil- ity that the cord blood will be of use in a family with no history of blood or genetic disease is low (estimated at 1/200 000); moreover, one’s own stem cells may be immunologically less potent than those of an unrelated donor for treating neoplastic diseases. However, several such transplants have been performed success- fully and prohibiting such storage despite appropriate informed consent seems curiously patronizing. The legal issues regarding property rights have been dis- cussed (Munzer 1999). In vitro fertilization and pre- natal genetic diagnosis to select an embryo donor on the basis of specific, desirable disease and HLA charac- teristics have been used successfully to treat a child with Fanconi’s anaemia (Grewal et al. 2004). Effect of ABO incompatibility of grafted cells As ABO and HLA antigens are inherited independ- ently, ABO incompatibility may occur in 20–40% of HLA-matched allogeneic haematopoietic stem cell transplants. ABO incompatibility between donor pro- genitor cells and the recipient’s plasma is not a barrier to successful transplantation (Storb et al. 1977; Buckner et al. 1978). In a series of 12 subjects who received major ABO-incompatible marrow, not one rejected the graft and the incidence of GvHD was no higher than in subjects who received ABO-compatible marrow (Hershko et al. 1980). With major ABO-incompatible marrow grafts, defined as incompatibility of donor ABO antigens with the recipient’s immune system, steps must be taken to prevent an acute haemolytic reaction due to lysis of incompatible red cells contained in the progenitor cell graft. To avoid haemolysis, grafts are purged of red THE TRANSFUSION OF PLATELETS, LEUCOCYTES, HAEMATOPOIETIC CELLS AND PLASMA COMPONENTS 631 cells. A satisfactory method has been described (Warkentin et al. 1985). An alternative method of removing red cells from marrow uses a cell separator (Blacklock et al. 1982). When PBPC or CBPC are used, the number of contaminating red cells is small. Delayed donor red cell engraftment and pure red cell aplasia are well-recognized complications of major ABO-incompatible haematopoietic stem cell trans- plantation (Hows et al. 1983; Sniecinski et al. 1988; Stussi et al. 2002; Griffith et al. 2005). Donor red blood cell chimerism is delayed as long as three-fold (median 114 days) following reduced-intensity non- myeloablative compared with myeloablative condi- tioning for transplant and the delay correlates with the recipient anti-donor isohaemagglutinin titre (Bolan et al. 2001a). Late-onset red cell aplasia, most likely related to delayed lymphoid engraftment, may occur (Au et al. 2004). In some patients, thrombopoiesis may be delayed as well (Sniecinski et al. 1988). After transplants of major ABO-incompatible grafts, the direct antiglobulin test (DAT) may turn positive after about 3 weeks. If substantial numbers of donor red cells enter the circulation, transient immune-mediated haemolysis may result (Sniecinski et al. 1987). Anti-A and anti-B may remain demonstrable in the recipient’s plasma for some months and the DAT may remain positive during this time. In patients with minor ABO-incompatible transplants, defined as those in which the recipient antigens are incompatible with the donor’s immune system, haemolysis may develop 1– 2 weeks after transplantation owing to lysis of ABO- incompatible recipient cells as the donor immune lymphocytes engraft (Hows et al. 1986). This type of haemolysis has been seen only in patients receiving ciclosporin and prednisone GvHD prophylaxis, and may not develop in patients receiving methotrexate (Gajewski et al. 1992). Massive immune haemolysis may occur, and fatalities can be avoided by early, vigorous donor-compatible red cell transfusion until haemolysis subsides (Bolan et al. 2001a). Reactions are most common and severe when the donor is group O and the recipient group A, but neither blood group nor agglutinin titre reliably predict clinical severity. In some of the patients, haemolysis caused by anti-A or anti-B (or both) destroys transfused group O cells, probably as a result of activated complement compon- ents affixing the group O cells (bystander haemolysis). Haemolysis has also been observed when the donor lymphocytes produce anti-D, etc. (see Chapter 11). Special consideration of the ABO group of compon- ents transfused to patients receiving ABO-incompatible grafts should begin with the initiation of the prepar- atory regimen to ensure that blood is compatible with both donor and recipient (Table 14.1). With bi- directional (major–minor) incompatibility, red cell transfusions should be limited to group O. Platelet concentrates administered to adults may be of any blood group, although plasma reduction may be pru- dent, especially for large-volume group O platelets. Plasma-compatible platelets should be used for infants and children. Some centres use the soluble antigens contained in plasma to neutralize isohaemagglutinins. As intravenous immunoglobulin contains variable titres of red cell antibodies, especially of anti-A, some centres screen for alloantibodies, whereas others avoid high-dose IVIG for group A recipients during the post- transplant period. Donor lymphocyte infusion Lymphocytes have been studied more often as blood component contaminants responsible for adverse effects than as therapeutic cells. However, some ostens- ibly adverse effects of mononuclear cell infusions can be exploited for therapeutic benefit. The mechanisms involved in TA-GvHD (see Chapters 13 and 15) are probably responsible for the graft-versus-malignancy effect in allogeneic stem cell transplantation. Studies in animal models are consistent with the observation by Barnes and Loutit (1957) that transplanted bone mar- row has immune activity against residual leukaemia (Kloosterman et al. 1995). Clinical experience with haematopoietic transplantation has been consistent with the presence of antileukaemic activity also in humans, now commonly referred to as the graft-versus-leukaemia effect (GvL). The term ‘adoptive immunotherapy’ was coined by Mathé (1965) who used both marrow transplants and leucocyte infusions to treat acute leukaemia. Kolb and co-workers (1990) provided direct clinical evidence for GvL: the transfusion of donor lymphocytes in conjunction with the adminis- tration of alpha-interferon (IFN-α) induced cytoge- netic remission in three patients with CML in relapse following allogeneic bone marrow transplantation. Numerous independent studies confirm the GvL effect in CML (Slavin et al. 1992; Bar et al. 1993; Porter et al. 1994). In both European and North American reg- istries, more than 90% of the patients received original CHAPTER 14 632 grafts and subsequent donor lymphocyte infusion (DLI) from related donors, typically from an HLA- identical sibling. The results reported by the European Group for Blood and Marrow Transplantation (27 centres, 135 patients, 75 evaluable with CML) are similar to those reported by the North American Multicenter Bone Marrow Transplantation Registry (25 centres, 140 patients, 55 evaluable with CML): DLI-induced clinical remission at a rate approaching 80%, and molecular remission (inability to detect bcr- abl mRNA transcript using polymerase chain reaction) in nearly all patients entering clinical remission (Kolb et al. 1995; Collins et al. 1997). Infusion of small numbers of lymphocytes (10 7 ) (‘bulk dose’) usually suffices, and excess cells collected by leucapheresis are often aliquoted and stored for repeated treatment if needed. The host’s circulation, which often contains a mixture of both donor and host cells during chronic phase relapse, typically converts to cells of only donor origin. The time to remission ranges from 1 to 9 months, with a mean of about 3 months (Kolb et al. 1995; Collins et al. 1997). Nearly all responses are seen within 8 months after DLI, and the probability of remaining in remission at 2 and 3 years is 90% and 87% respectively (Kolb et al. 1995; Collins et al. 1997). Although late relapses occur and toxicity may be significant, DLI efficacy is durable in surviving patients with CML: 26 out of 39 (67%) pati- ents were alive at follow-up with 25 (96% of survivors) remaining in complete remission (Porter et al. 1999). Separating graft-versus-leukaemia from graft-versus-host-disease GvHD occurring after DLI correlates strongly with antileukaemic response. However, the GvL effect and GvHD may be separable, and GvHD may not be required for durable disease remission (Weiss et al. 1994; Rocha et al. 1997; Slavin et al. 2002). Murine studies suggest that the rate of GvHD is inversely THE TRANSFUSION OF PLATELETS, LEUCOCYTES, HAEMATOPOIETIC CELLS AND PLASMA COMPONENTS 633 Blood group Blood products Recipient Donor Red cells* Platelets † FFP* A B O Any AB ‡ A O O Any A § A AB O,A Any AB ‡ B A O Any AB ‡ B O O Any B § B AB O,B Any AB ‡ O A O Any A ‡ O B O Any B ‡ O AB O Any AB ‡ AB A O,A Any AB § AB B O,B Any AB § AB O O Any AB § Rh pos Rh neg Rh neg †† Rh neg** N/A Rh neg Rh pos Rh pos ¶ Rh pos N/A * Restrictions for ABO- and/or Rh-incompatible transplant recipients supported with blood components during pre-transplant conditioning and during the post-transplant period. † Use any ABO group for platelet support for adults. Use plasma-compatible components for children. ‡ Plasma may be transfused to neutralize isohaemagglutinin(s). § Graft plasma depleted, no plasma neutralization required. ¶ Rh-positive components initiated on the day of transplant. ** Rh-negative platelets preferred. †† Rh-negative red cells preferred during the pre-transplant conditioning regimen and post transplant. Table 14.1 ABO and/or Rh- incompatible progenitor cell: transplant transfusion restrictions. proportional to the interval between transplant and DLI (Johnson et al. 1999). These considerations have popularized escalating dose DLI regimens (Dazzi et al. 2000). Although probability of achieving remission in relapsed CML does not differ, the escalating dose regimen is associated with a lower incidence of GvHD. DLI is typically initiated early in disease as soon as disease recurrence is anticipated, and a starting dose of 10 5 T cells/kg is escalated 10-fold at 2- to 4-week intervals (Weiss et al. 1994). Efforts to modify the composition of donor-derived lymphocytes (DDLs) have focused on selective CD8-positive T-cell depletion, which appears to be more effective than non-selective T-cell depletion in reducing GvHD while preserving GvL (Soiffer et al. 2002). Target antigen as the primary determinant of efficacy and toxicity Falkenburg and colleagues (1999) reported the first successful treatment of relapsed accelerated CML using in vitro-expanded leukaemia-specific lymphocytes. Presumably, cell selection and culture restored the anti-tumour activity and specificity against leukaemic cells that weakened with disease progression (Smit et al. 1998). Successful salvage therapy of a child with previously DLI-resistant CML by using DDL pulsed in vitro with a mixture of normal irradiated lymphocytes obtained from the child’s parents has been reported. Efficacy and toxicity in viral diseases Walter and co-workers (1995) have used in vitro- stimulated, culture-expanded, CMV-specific donor- derived cytotoxic T cells to successfully reconstitute cellular immunity against CMV in 11 out of 14 allo- geneic marrow transplant patients. The DLI therapy consisted of four escalating cell doses (0.33, 1.0, 3.3 and 10.0 × 10 8 cells) administered at weekly intervals beginning at days 30–40 after transplantation. DLI- associated toxicity, CMV disease and CMV viraemia were not observed. The results have been confirmed in similar studies (Einsele et al. 2002; Roback et al. 2003). Epstein–Barr virus-related lymphoproliferative disorders The incidence of Epstein–Barr virus-related lympho- proliferative disorders (EBV-LPDs) occurring in T cell- depleted transplants has been estimated at 6–12%, and secondary lymphomas occurring in this clinical setting respond readily to DLI at a dose approximately 10-fold smaller than that typically used for activity against the primary leukaemia. Sustained clinical remissions have been achieved with only mild GvHD, and patients have often required no additional mainten- ance therapy (Papadopoulos et al. 1994; Wagner et al. 2004). EBV-specific lymphocyte infusions have suc- cessfully treated EBV-LPD and EBV-positive Hodgkin disease (Rooney et al. 1998; Bollard et al. 2004). The transfusion of plasma components Fresh-frozen plasma Fresh-frozen plasma (FFP) is plasma obtained from a single donor by normal donation or plasmapheresis and frozen within 6 h of collection to a temperature of –30°C or below. FFP contains all circulating coagu- lation factors in the concentration present in fresh plasma, and haemostatic activity is maintained for a year or longer, depending upon the storage temper- ature. Once thawed, FFP must be stored at 4 ± 2°C for no longer than 24 h before infusion. FFP must not be refrozen, but once thawed (or after 1 year of storage and thaw), it can be used as single-donor plasma, i.e. not to replace labile coagulation factors, for as long as 5 weeks. The concentration of coagulation factor, the citrate concentration and the volume of each unit may vary depending on the characteristics of the donor and of the collection. In 51 units collected by apheresis from plasma donors, factor concentrations at the fifth and ninety-fifth percentile measured: V (690–1270 units/ l); VII (830–1690 units/l); fibrinogen 1800–3700 µg/l; antithrombin (920–1290 units/l) (Beeck et al. 1999). Risks of fresh-frozen plasma Allergic reactions may occur after transfusion of FFP, of which the most serious is severe anaphylaxis, which may develop in IgA-deficient patients with class- specific anti-IgA (Chapter 15). Such reactions are rare. Transfusion-related acute lung injury (TRALI) may occur when the FFP contains strong leucocyte antibod- ies (see Chapter 15). The other main risk of treatment with FFP is the transmission of infectious agents, par- ticularly viruses such as hepatitis B and C viruses, HIV, parvovirus and West Nile virus. Owing to donor selec- CHAPTER 14 634 tion and the availability of methods of inactivating viruses that are used to treat whole plasma in some countries, the risk of transmitting viruses has greatly decreased. However, the problem of inactivating non-lipid-enveloped viruses and the transmission of non-viral agents remains (Chapter 16). Therefore, FFP should still be used only when no safer alternative exists (Shimizu and Robinson 1996). Precautions to be taken before infusion FFP containing potent anti-A or anti-B agglutinins or haemolysins, or FFP that has not been tested for their presence, should not be given to recipients with corres- ponding red cell antigens. Fresh plasma, which is now rarely used, may contain red cells, so that appropriate measures should be taken to prevent immunization of D-negative women of childbearing age. There is no credible evidence that FFP presents such a risk. Indications for fresh-frozen plasma: overused and abused There is no justification for the use of FFP as a volume expander because safer alternatives (colloids and crystalloids) are available. Factor V deficiency. No concentrate of factor V is available and FFP can be used as a source of factor V. Cryoprecipitate-poor plasma contains 80% of the amount of factor V in FFP and can be used as an alternative for FFP (Hellings 1981). Severe liver disease. The liver is the major site of syn- thesis of coagulation factors II, V, VII, IX, X, XI, XII and fibrinogen as well as of factors with potential antithrombotic activity such as proteins C, S and antithrombin. Patients with severe liver disease may experience defects in factor synthesis and increased factor degradation that can result in generalized bleed- ing. Unfortunately, studies regarding the predictability of bleeding and its most effective management by transfusion in the presence of different degrees of hep- atic impairment are old, poorly documented or both. Most recommendations rely upon expert opinion. What seems clear is that no single coagulation assay predicts bleeding (Spector and Corn 1967). Prolonga- tions of the prothrombin time (PT) and activated partial thromboplastin time (aPTT) are the most fre- quent abnormalities among the commonly performed clotting tests in patients with liver disease, and may reflect impaired protein synthesis, vitamin K deficiency or even disseminated intravascular coagulation (DIC). The presence of an abnormal test does not necessitate intervention, especially in the non-bleeding patient. Furthermore, in 30 patients with chronic liver disease, a moderate-dose plasma infusion (12 ml/kg or about 4 units) did not return the PT and aPTT to normal (Mannucci et al. 1982). In the bleeding patient, doses calculated to bring coagulation factor levels to the 20–30% range (20 ml/kg or 6–7 units) may be required as frequently as every 4–6 h to correct the abnormal coagulation tests (Spector et al. 1966). The routine use of FFP as prophylaxis for excessive surgical bleeding in patients with severe liver disease finds few supporters and less evidence of benefit (Oberman 1990). Treatment of acquired deficiencies of factors II, VII, IX and X due to treatment with anticoagulants: warfarin reversal The major risk of anticoagulant therapy is haemor- rhage. For patients treated with the oral vitamin K antagonists, the annual risk of severe haemorrhage ranges from 1–5% (Levine et al. 2001). The intensity of anticoagulation (including poor control), its dura- tion and, in some studies, advanced age and cerebrov- ascular disease all increase the bleeding risk (Landefeld and Goldman 1989). For the bleeding patient or the patient at extreme risk, urgent reversal of vitamin K antagonists can be achieved with plasma infusion to bring factor levels to 30–40%. The volume of plasma can be calculated easily based on the patient’s body weight but as 6 units or more (1500 ml) may be required to reverse anticoagulation in an adult, volume consid- erations may make a prothrombin complex concentrate (PCC) the preferred infusion (Schulman 2003). Recom- binant VIIa has also been used in this situation (Deveras and Kessler 2002) (see Chapter 18). Intravenous vita- min K 1 , the specific warfarin antagonist, may require 12 h or more to be fully effective (Nee et al. 1999). Disseminated intravascular coagulation: a vehicle on the road to multi-organ dysfunction syndrome Disseminated intravascular coagulation is a condition in which the intravascular activation of the clotting THE TRANSFUSION OF PLATELETS, LEUCOCYTES, HAEMATOPOIETIC CELLS AND PLASMA COMPONENTS 635 cascade leads to the final common pathway of sus- tained and excessive thrombin generation. Liberated thrombin and proteolytic enzymes bring about the intravascular production of fibrin and deposition of platelets, with activation of the fibrinolytic system and an increased level of fibrin degradation products (FDPs) (Levi et al. 2001). In mild DIC the platelet count and the levels of clotting factors may be normal due to compensatory increases in production. As DIC becomes more severe, the levels of clotting factors and platelets fall, and a state that may be described as decompensated DIC may lead to multi-organ dysfunc- tion syndrome (MODS). DIC may be precipitated by a wide variety of stimuli, most related to the entry of tissue thromboplastins into the circulation, for example after abruptio placentae, crush injury, head trauma and snake envenomation. Other conditions associated with DIC include infec- tions, malignancies, amniotic fluid embolism, giant haemangioma and intravascular lysis of incompatible red cells (Levi et al. 2004). The cardinal principle of treatment of DIC remains elimination of the underlying cause as, once this has been accomplished, haemostasis usually returns to normal. When the underlying cause cannot be treated effectively, uncontrollable bleeding may result. The transfusion of blood may be essential and the replace- ment of clotting factors has to be considered. This replacement should be guided by coagulation assays and fibrinogen levels. If levels of clotting factors are severely reduced (< 25%), FFP may be given and if the fibrinogen concentration falls below 60 mg/dl, cryo- precipitate may be helpful. An initial dose of 10 bags, to provide 4–6 g of fibrinogen, has been suggested (Prentice 1985). Despite the theoretical objection of adding ‘fuel to the fire’, the administration of fibrino- gen does not seem to be particularly dangerous. Thrombotic thrombocytopenia purpura Before the mechanisms involved in thrombotic throm- bocytopenia purpura (TTP) were suspected, a plasma factor was postulated to correct the syndrome charac- terized by microangiopathic haemolysis and throm- bocytopenia (Upshaw 1978). Relapses in chronic TTP were reversed or prevented by infusions of small volumes of FFP or cryoprecipitate-depleted FFP or by plasma infusion combined with plasmapheresis (Byrnes and Khurana 1977; Bukowski et al. 1981). The plasma factor is not destroyed by the solvent detergent treat- ment of FFP used to inactivate lipid-encapsulated viruses (Moake et al. 1994). In the majority of cases, the plasma factor relates to the activity of a metallo- proteinase that cleaves unusually large multimers of vWF that are associated with the TTP thrombi (Asada et al. 1985; Tsai 1996) (see Chapter 17). Cryoprecipitate-depleted fresh-frozen plasma (cryosupernatant) Cryosupernatant is plasma from which about one-half of the fibrinogen, factor VIII and fibronectin has been removed as cryoprecipitate. The product is also depleted of the largest multimers of vWF, which sediment in the cryoprecipitate fraction and which may be partly responsible for platelet aggregation in TTP (Moake 2004). In some circumstances cryosuper- natant may be more effective than FFP in the treatment of TTP. Seven patients with TTP who failed to respond to intensive plasma exchange with whole plasma responded to plasma exchange with cryosupernatant (Byrnes et al. 1990). Cryoprecipitate When plasma is fast frozen and then thawed slowly at 4–6°C, the small amount of protein precipitated is rich in fibrinogen, factor VIII, vWF and factor XIII. After decanting almost all of the supernatant plasma, the precipitated protein can be dissolved by warming to yield a small volume of solution. The introduction of cryoprecipitate revolutionized the treatment of haemophilia by providing a highly effective, conveni- ent, readily available source of factor VIII. Modern treatment has moved away from cryoprecipitate to pathogen-inactivated factor VIII concentrate and to recombinant factor VIII. Cryoprecipitate is used now as a source of factor VIII and vWF only if safer concen- trates are not available. Cryoprecipitate, containing approximately 200– 250 mg of fibrinogen in a volume of 10–15 ml, pre- pared from a single donor, is used primarily as a source of fibrinogen. The most common indication remains as a replacement for fibrinogen consumed in DIC, although it has been used as a topical haemostatic agent as well (fibrin glue) (Reiss and Oz 1996). Commercial fibrin sealants are safer, better standard- ized and more effective, and avoid the potential risk of CHAPTER 14 636 immunization to contaminant factor V that has been reported when bovine thrombin is used to activate topical cryoprecipitate (Rousou et al. 1989; Rapaport et al. 1992; Atrah 1994) (see also Chapter 18). Thawing by microwave is rapid and preserves fibrino- gen concentration (Bass et al. 1985). Cryoprecipitate also contains about 60% of the vWF and 20–30% of the factor XIII of the original unit of FFP, but the component in not often used as a source of these proteins. ‘Contaminants’ in cryoprecipitate. Cryoprecipitates contain about 30–50% of the original fibrinogen and have about the same titre of anti-A and anti-B as that of the original plasma unit (Rizza and Biggs 1969; Pool 1970). Because of the risk of haemolysis, neonates should receive only ABO-compatible cryoprecipitate. Plasma fractionation The transfusion of whole plasma is unnecessary and usually inefficient if recipients require only a single protein, for example factor VIII. Plasma contains hun- dreds of different proteins, many of which are obvious candidates for replacement therapy, whereas others are well characterized physicochemically, but of un- known function. Commercial plasma fractionation uses dilution, pasteurization and nanofiltration to remove and inactivate most viruses, although no product can be guaranteed ‘pathogen free’. The immunoglobulin (Ig) fraction (predominantly IgG) separated from whole plasma by alcohol fractionation was at first considered virtually free of the risk of transmitting viral hepatitis. However, HCV has been transmitted by both IVIG and anti-D Ig (Bjoro et al. 1994; Meisel et al. 1995; Power et al. 1995). The most widely used method of fractionating plasma is still the cold alcohol precipitation technique described by Cohn and colleagues (1944) or modifica- tions thereof (Kistler and Nitschmann 1962). Cohn fractionation relies on changes in ethanol concentra- tion and pH for bulk precipitation of different protein fractions. An example of a fractionation scheme is shown in Fig. 14.6. Ethanol is removed by lyophiliza- tion or by ultrafiltration. Alcohol fractionation is now combined with glycine precipitation or polyethylene glycol, and with other separation methods such as chromatography to isolate specific proteins, such as coagulation factors and protease inhibitors. Albumin Albumin is available for clinical use either as human albumin in saline containing 4%, 4.5%, 5%, 20% or 25% protein, of which not less than 95% is albumin, or as plasma protein fraction (PPF), available only as a 5% solution, of which at least 83% is albumin. Compared with albumin, most preparations of PPF contain larger amounts of contaminating proteins. Hypotensive reactions attributed to pre-kallikrein activator and acetate have been observed with PPF, but not with albumin (Alving et al. 1978; Ng et al. 1981). For these reasons, most clinicians find little reason to select PPF when an albumin solution is indicated. Albumin preparations are pasteurized by heat treatment at 60°C for 10 h and filtered. When pre- pared in this way, the fraction has proved free of transfusion-transmitted agents such as hepatitis viruses and HIV. Although albumin contributes 75–80% of the col- loid osmotic pressure of the plasma, subjects with a genetically determined total absence of plasma albu- min, in whom the colloid osmotic pressure of plasma is between one-third and one-half of normal, may be completely asymptomatic (Bearn 1978). Such subjects show an increase in various plasma globulins and a slight decrease in blood pressure, changes that are regarded as compensatory. The indications for infu- sions of albumin in hypovolaemic patients are discussed in Chapter 2. Recombinant albumin Recombinant albumin has been synthesized in yeast, in Saccharomyces cerevisiae or Pichia pastoris, and appears to be similar to the plasma-derived protein (Dodsworth et al. 1996). A 20% solution (Recombumin 20%, Aventis Behring) prepared as a pharmaceutical excipient has been tested for safety in doses up to 65 mg in some 500 subjects. It is uncertain when, if ever, recombinant albumin might be commercially available as a product for transfusion. Fibrinogen The rate of disappearance of injected fibrinogen has been studied by giving infusions to patients with the very rare condition, hereditary afibrinogenaemia: in THE TRANSFUSION OF PLATELETS, LEUCOCYTES, HAEMATOPOIETIC CELLS AND PLASMA COMPONENTS 637 [...]... saline-adenine-glucose-mannitol-suspended red cells in a new plastic container: polyvinylchloride plasticized with butyryl-n-trihexyl-citrate Transfusion 31: 26–29 Högman CF, Gong J, Eriksson L et al (1991b) White cells protect donor blood against bacterial contamination Transfusion 31: 620–626 Holme S, Heaton A, Momoda G (1 989 ) Evaluation of a new, more oxygen-permeable, polyvinylchloride container Transfusion. .. severe abdominal infections (Lundsgaard-Hansen et al 1 985 ) Similarly, patients with septic shock or severe injury showed no evidence of improvement after treatment with fibronectin (Rubli et al 1 983 ; Hesselvik et al 1 987 ; Mansberger et al 1 989 ) α1-Antitrypsin α1-Antitrypsin (α1-AT) is a major serine endopeptidase inhibitor in human plasma, which inhibits neutrophil elastase, an enzyme involved in the proteolysis... platelet transfusions associated with ABO or Rh(D) incompatability Transfus Med Rev 17: 57– 68 Lundsgaard-Hansen P, Doran JE, Rubli E (1 985 ) Purified fibronectin administration to patients with severe abdominal infections Ann Surg 202: 745–7 58 Lusher J, Ingerslev J, Roberts H et al (19 98) Clinical experience with recombinant factor VIIa Blood Coagul Fibrinolysis 9: 119–1 28 Lusher JM (1994) Summary of clinical. .. adhesion or interference with thrombin-induced in ammation C1 esterase inhibitor (C1 inh.) Hereditary functional deficiency of C1 inh is due to either a deficiency or a dysfunction of the protein Acquired deficiencies of C1 inh also occur This pro- tease inhibitor is involved in the regulation of several proteolytic systems in plasma, including the complement system, the contact system of intrinsic coagulation... protease-antiprotease imbalance within the alveolar structures of PiZ subjects J Clin Invest 68: 11 58 1165 Gajewski JL, Petz LD, Calhoun L et al (1992) Hemolysis of transfused group O red blood cells in minor ABOincompatible unrelated-donor bone marrow transplants in patients receiving cyclosporine without posttransplant methotrexate Blood 79: 3076–3 085 Gale RP, Winston D (1991) Intravenous immunoglobulin... plasma Blood 81 : 942–949 Cartwright GE, Athens JW, Wintrobe MM (1964) The kinetics of granulopoiesis in normal man Blood 24: 780 Caspar CB, Seger RA, Burger J et al (1993) Effective stimulation of donors for granulocyte transfusions with recombinant methionyl granulocyte colony-stimulating factor Blood 81 : 286 6– 287 1 Castaman G, Lattezada A, Mannucci PM (1995) Factor VIII: C increases after desmopressin in. .. Working Party Chronic Leukemia Blood 86 : 2041–2050 Konietzko N, Becker M, Schmidt EW (1 988 ) Substitution therapy with alpha-1-Pi in patients with alpha-1-Pi deficiency and progressive pulmonary emphysema Dtsch Med Wschr 113: 369–373 Konkle BA, Bauer KA, Weinstein R et al (2003) Use of recombinant human antithrombin in patients with congenital antithrombin deficiency undergoing surgical procedures Transfusion. .. American Pathologists Bode AP, Miller DT (1 988 ) Preservation of in vitro function of platelets stored in the presence of inhibitors of platelet activation and a specific inhibitor of thrombin J Lab Clin Med 111: 1 18 124 Bode AP, Miller DT (1 989 ) The use of thrombin inhibitors and aprotinin in the preservation of platelets stored for transfusion J Lab Clin Med 113: 753–7 58 Bodey GP, Buckley M, Sathe YS et al... (2002) Seven-day storage of single-donor platelets: recovery and survival in an autologous transfusion study Transfusion 42: 84 7 85 4 Dutcher JP, Schiffer CA, Johnston GS (1 981 ) Rapid migration of 111indium-labeled granulocytes to sites of infection N Engl J Med 304: 586 – 589 Edelson RN, Chernik NL, Posner JB (1974) Spinal subdural hematomas complicating lumbar puncture Arch Neurol 31: 134 –137 Einsele H,... was no difference in death rate A single infusion may be insufficient to reduce infection-related mortality for more than a few weeks No serious side-effects occurred in any of the treated infants (Weisman et al 1992) Infection in adults Selected Ig preparations containing antibodies in high titre may have a role in severe viral and bacterial infection (Sawyer 2000) Trials with anti-Pseudomonas Ig prepared . CML): DLI-induced clinical remission at a rate approaching 80 %, and molecular remission (inability to detect bcr- abl mRNA transcript using polymerase chain reaction) in nearly all patients entering clinical. provided direct clinical evidence for GvL: the transfusion of donor lymphocytes in conjunction with the adminis- tration of alpha-interferon (IFN-α) induced cytoge- netic remission in three patients. specific proteins, such as coagulation factors and protease inhibitors. Albumin Albumin is available for clinical use either as human albumin in saline containing 4%, 4.5%, 5%, 20% or 25% protein, of