TRANSLATIONAL AND CLINICAL RESEARCH

Transcription

1 TRANSLATIONAL AND CLINICAL RESEARCH Directed Sibling Cord Blood Banking for Transplantation: The 10-Year Experience in the National Blood Service in England JON SMYTHE, a SUE ARMITAGE, b DOROTHY MCDONALD, a DERWOOD PAMPHILON, c MARTIN GUTTRIDGE, a JULIETTE BROWN, d ANN GREEN, d COLIN BROWN, d RUTH M. WARWICK, e ALAN LANKESTER, a DEIRDRE FEHILY, b MARCELA CONTRERAS, f CRISTINA NAVARRETE, d,g SUZANNE M. WATT a,h a Stem Cells and Immunotherapies Department, b National Health Service Cord Blood Bank, c Bristol Institute of Transfusion Sciences, d Histocompatibility and Immunogenetics Department, e Tissue Services Department, f Division of Diagnostics, Development and Research, National Blood Service, National Health Service Blood and Transplant, United Kingdom; g Department of Immunology and Molecular Pathology, University College London, London, United Kingdom; h Nuffield Department of Clinical Laboratory Sciences, University of Oxford, Oxford, United Kingdom Key Words. Directed cord blood banking Stem cells Transplantation Accreditation ABSTRACT Umbilical cord blood (UCB) is an important source of hematopoietic stem cells for transplantation. Although UCB is often collected from unrelated donors, directed umbilical cord blood (DCB) from sibling donors also provides an important source of UCB for transplantation. This report summarizes the experience in collection, testing, storage, and transplantation of DCB units by the National Blood Service for England and North Wales over 10 years. Eligibility for collection was based on an existing sibling suffering from a disease that may be treated by stem cell transplantation or a family history that could result in the birth of a sibling with a disease that could be treated by stem cell Disclosure of potential conflicts of interest is found at the end of this article. transplantation. Collections were made on the provision that the sibling s clinician was willing to financially support the collection and to take responsibility for medical review of the mother and potential recipient. Given the high investment in UCB banking and the introduction of new regulations and mandatory licensing under the European Union Tissues and Cells Directive and those proposed in the U.S., this report details the procedures that we have used for DCB donations, the outcome data where donations have been used for transplantation, and it provides some timely recommendations for best practices. STEM CELLS 2007;25: INTRODUCTION Umbilical cord blood (UCB) is an important source of hematopoietic stem cells for clinical transplantation. Following the pioneering work of Gluckman et al. [1], cord blood banks now operate in many countries, with more than 10,000 allogeneic transplants performed world-wide [2 5]. Many of the procedures for processing and storage of UCB in use in the U.K. today were established at the New York Cord Blood Bank in the 1990s [6] and modified by the National Blood Service (NBS) [2, 7, 8]. Cord blood transplants have the advantage of less graftversus-host disease (GvHD) than similarly matched non-t-celldepleted bone marrow (BM) transplants, although time to engraftment for single UCB unit transplants may be delayed [2 5, 9 11]. Factors associated with improved UCB engraftment include lower recipient age and weight, closer human leukocyte antigen (HLA) matching, and cell dose [9]. Low UCB cell doses have often restricted their use for transplantation to children or small adults, where recommended doses range from or exceed total nucleated cells (TNC) per kilogram [9]. More recently, the use of multiple UCB units for larger recipients or combinations of UCB with selected bone marrow or mobilized peripheral blood CD34 cells in a haploidentical setting have proven successful in increasing cell dosages for specific disease indications [2, 3, 9, 11 14]. Approximately 25% of patients requiring an allogeneic stem cell transplant have an HLA-matched sibling, and for the majority of the remainder, if white, an appropriate unrelated BM or peripheral blood stem cell (PBSC) harvest or a stored unrelated UCB donation will probably be found [9]. For some children, especially those from ethnic minority groups and those who suffer from inherited disorders such as hemoglobinopathies or immune deficiency, a directed collection of UCB following the birth of a sibling may be the only opportunity for an HLAmatched transplant until the sibling is able to donate BM. In addition, such directed umbilical cord blood (DCB) donations from matched siblings have generally been shown to give better overall results than matched unrelated BM [15 17]. Preimplantation genetic testing in the U.K. to ensure that a pregnancy will result in a child free from a serious inherited disorder known to exist in the family is now accepted practice regulated through the Human Fertilisation and Embryology Authority (HFEA) Correspondence: Jon Smythe, MPhil, Ph.D., Stem Cells and Immunotherapies Department, National Blood Service, NHS Blood and Transplant, Headington, Oxford OX3 9BQ, U.K. Telephone: ; Fax: ; jon.smythe@nbs. nhs.uk Received January 25, 2007; accepted for publication April 20, 2007; first published online in STEM CELLS EXPRESS May 17, AlphaMed Press /2007/$30.00/0 doi: /stemcells STEM CELLS 2007;25:

2 2088 Sibling Donor Cord Blood Transplants [18, 19]. Combined with preimplantation HLA typing, this practice will increase the chances of finding a matched diseasefree DCB collection, and so the number of DCB collections is likely to rise [20, 21]. The National Blood Service in England, through its network of accredited stem cell services and histocompatibility laboratories, has taken a leading role in the provision of a national DCB service for high-risk families. This service encompasses liaison with the obstetric delivery unit to arrange collection, processing, testing, storage, and issue for transplantation. This service has been developed to operate in a GMP-compliant environment as prescribed by the U.K. Department of Health Code of Practice for Tissue Banks 2001 [22]. The DCB service was also developed to meet the evolving standards of the U.K. Blood Transfusion Services Red Book quality standards [23], FACT-NETCORD guidelines [24], the requirements of the U.K. Human Tissue Act [25], and the European Union Directive for Tissues and Cells [26]. Here, we review the service we have provided for DCB collections over 10 years, describing the quality of the units, practices adopted, regulatory issues, and the outcome of the resulting transplants. We also recommend best practices, to ensure a system for DCB collection and storage that is ethical, robust, of high quality, and clinically responsive to patient needs. MATERIALS AND METHODS Eligibility for Directed Cord Blood Donation The NBS accepted a referral for a DCB collection if (a) an existing sibling could be treated by allogeneic transplantation, (b) an earlier birth from the same parents indicated an inherited disorder that could result in a future infant needing an allogeneic transplant, (c) the sibling s clinician and hospital were willing to financially support the banking, and (d) the referring clinician took responsibility for medical review of the mother and potential recipient and obtained written informed consent before the delivery. The referrals were made using NBS forms as signed agreements to cover the elements of consent and collection not performed by NBS personnel. Referring clinicians were those responsible for treating the sibling s condition and so specialized in the relevant disorder. Written Informed Consent Written informed consent was obtained from the donor mother for screening of both the mother and cord blood for mandatory microbiological markers, for tissue typing, and for storage of samples and confidential information. Consent was taken with reference to the U.K. Department of Health s Reference Guide to Consent for Examination or Treatment [27]. An information sheet was latterly provided by the NBS as an aid to those undertaking consent. Consent was also obtained by the NBS from the obstetrician for the delivery hospital to help with the collection. Mandatory Screening Samples from mothers were tested in advance for hepatitis B surface antigen, anti-hepatitis C virus (HCV), HIV 1 and 2, syphilis, and, more recently, human T-lymphotropic virus I and II, HCV, and HIV by polymerase chain reaction (PCR), by the NBS. Potential collections were not necessarily excluded on the basis of positive maternal virology results. Directed Cord Blood Collection and Transport to the NBS Collections were organized and processed by four NBS centers located across England to limit transit times; collection could therefore be performed in the hospital selected by the mother for birth. Initially, NBS staff collected the DCB, but later, for logistical reasons, midwives collected the DCB using detailed instructions provided by the NBS. The instructions covered collection both in utero and ex utero, with additional advice from the NBS provided by telephone when necessary. NBS collection kits were sent to a named individual in the delivery hospital in an insulated box, validated for transport of the DCB at 2 C 8 C using cool pack inserts. The kit included a spare collection pack (Macopharma, Twickenham, U.K., and alcohol and iodine sterile swabs for cleaning the cord. A partially completed label was sent, along with line clips, a sealer, and a labeled bottle for a segment of the umbilical cord as a backup source of DNA for HLA typing. Where the mother s history indicated that the fetus/ infant was at risk for an inherited disorder, the referring hospital liaised directly with the delivery hospital to request neonatal samples for genetic testing. The kit included details of a courier able to deliver at short notice, and shipping records were maintained. The NBS centers were staffed 24 hours a day, and collections were stored at 4 C before cryopreservation within 24 hours of collection. HLA Typing, Including Preimplantation Studies A blood sample from the potential recipient (and family members if required) was requested for HLA typing. Alternatively, a copy of the HLA typing results was provided by the referring hospital. HLA typing by the NBS for HLA-A, B, Cw, DRB1, and DQB1 antigens was carried out by polymerase chain reaction with sequence-specific primers (PCR-SSP), as detailed previously [28]. Before 1997, some HLA typing was done by serology. DNA was stored for any confirmatory HLA typing required subsequently. Prenatal HLA typing from chorionic villus sampling was as described [29]. Preimplantation screening [30], introduced during this study, was the responsibility of the referring clinician and followed U.K. guidelines issued by the HFEA [31]. Directed Cord Blood Processing and Storage If the DCB was badly clotted or of negligible volume, the laboratory head called the referring clinician to discuss whether to proceed. The default position, however, was to proceed, with a final decision on storage being made when all results were available. A unique 12-digit barcode number was used for labeling, paperwork, and files, in addition to other details. A bleedline sample was removed for TNC and CD34 counts, cell viability, and blood grouping. CD34 analysis was performed using a single-platform lyse no-wash flow cytometric protocol [32, 33] that incorporated calibrated beads to quantitate CD34 cell concentration directly and 7-actin-actinomycin to determine viability. Processing was carried out according to European Union GMP standards [34] under grade A air quality conditions in a clean room. A sample was taken prior to processing for tissue typing and virology screening. Cryoprotectant solution was prepared from dimethyl sulfoxide (DMSO) and 10% (wt/vol) dextran-40/saline or 4.5% (wt/vol) human albumin solution [32]. An equal volume of cooled 20% (vol/vol) DMSO solution was added to the cells in a controlled manner. To maximize the cell dose, stored volume reduction was not carried out [23]. A small volume was taken postprocessing for bacteriological screening and to prepare frozen reference samples before the cells were frozen in cryocyte bags (Baxter Healthcare, Newbury, U.K., co.uk). DCB collections were processed individually to avoid inadvertent mixing of bags or samples. A contiguous line sample was left attached to the frozen bags for future confirmatory HLA typing, although this practice varied in the early part of this study. The collections were frozen double-wrapped using a controlled-rate freezer and then transferred to liquid nitrogen-based storage tanks, maintained at less than 150 C and continuously monitored [33]. DCB units from mothers known to be positive for microbiology markers or awaiting results were stored in quarantine tanks. Reference samples were cryopreserved with the DCB unit and stored in the same tank [33]. Reporting to the Referring Clinicians A preliminary report was sent indicating volume, cell counts, and any adverse events, with an additional report sent once all test

3 Smythe, Armitage, McDonald et al Table 1. Diagnostic basis for directed cord blood collections Disease status of sibling or family member Number of cases Hematological malignancies: total 114 Acute lymphoblastic leukemia 74 Acute myeloid leukemia 12 Chronic myeloid leukemia (1) 10 Non-Hodgkin lymphoma 3 Other (2) 15 Anemia: total 68 Thalassemia (1) 36 Diamond-Blackfan anemia 10 Sickle-cell anemia 8 Aplastic anemia 7 Fanconi anemia 6 Congenital dyserythropoietic anemia 1 Immunodeficiency: total 44 Severe combined immunodeficiency (20) 28 Wiskott-Aldrich syndrome (1) 3 Other (1) 13 Enzyme deficiency: total 9 Chronic granulomatous disease (1) 4 Gaucher disease 2 Other (2) 3 Other diseases (3) 9 Grand total 244 A total of 268 collections were made from 244 mothers, including 5 mothers with twins, 13 with two births, and 2 with three births. Of the 268 collections, 35 were obtained from 32 mothers with a family history of an inherited disease. Numbers in parentheses indicate number of collections from mothers with no living child. The majority of these were for severe combined immunodeficiency syndrome (SCID). results were available. Where HLA typing of the potential recipient and family members was done by the NBS, the report included information on the degree of match. NBS histocompatibility staff were available to discuss the HLA typing with the referring hospital s histocompatibility laboratory or transplant medical consultant. Storage Policy The NBS advice on long-term storage of the DCB depended on HLA compatibility, the potential recipient weight, disease progression, and the likelihood and timing of using the DCB unit. Latterly, the NBS policy has been to store units for patients with at least one haplotype match and cell counts greater than TNC. A written statement from the referring clinician was requested to confirm whether or not long-term storage was required. NBS costs (on a not-for-profit basis) were recovered from the referring hospital. Directed Cord Blood Issue for Transplantation All final decisions on transplantation remained the responsibility of the transplant unit. All requests to issue the DCB were confirmed in writing. Frozen reference samples were thawed by the NBS for cell viability, colony-forming unit (CFU) testing, confirmatory HLA typing, and short tandem repeat analysis [33] to confirm identity with the sample that had been HLA typed at the time of banking. Where the reference sample was not the unit bleedline, the sample had been given the same unique barcode number as the unit to avoid incorrect identification. The NBS has shown that viability testing on reference samples from both the bleedline and vials can be used to predict viability in the whole units. CFU assays were carried out latterly using a methylcellulose-based medium according to the manufacturer s instructions (Stem Cell Technologies, Vancouver, BC, Canada, [34]. DCB units were transported in temperature-monitored dry shipper containers at less than 150 C and were thawed by NBS or hospital staff, with instructions provided by the NBS. Table 2. Laboratory characteristics of banked directed cord blood units Volume (ml) Total TNC ( 10 8 ) Transplant Outcome A form requesting information on adverse reactions, primary engraftment data, and transplant outcome was issued with the DCB unit, and additional information was requested by telephone. Each transplant center determined engraftment using its usual protocols. Neutrophil engraftment was defined as the first of three consecutive days to reach neutrophils per liter and platelet engraftment as the first of three consecutive days to reach platelets per liter, with at least 7 days since the last platelet transfusion. RESULTS Total CD34 ( 10 6 ) TNC viability (%) Median Mean SD Range A total of 236 units (88.1%) exceeded 40 ml, and 238 collections (88.8%) exceeded TNC. TNC viability was determined precryopreservation using 7-actin-actinomycin as part of the flow cytometric analysis for cell numbers. Abbreviation: TNC, total nucleated cells. Breakdown of Collections by Patient Disorder Over a 10-year period, 268 collections were made from 244 mothers, including five mothers with twins. Two hundred thirtythree collections (87%) were for an existing sibling, and 35 (13%) were collected where there was a family history of a serious inherited disease. Diagnoses included 114 hematological malignancies, 68 hereditary anemias, 44 immune deficiencies, and 9 metabolic disorders (Table 1). An additional eight requests (2.9%) were received for collections that were not successful, usually due to a damaged cord/placenta at delivery. Collections from twins were made individually ex utero by midwifery staff. The requests for banking changed over time, with a gradual shift from hematological malignancies toward hereditary anemias and immunodeficiencies. Where collections were made from infants who subsequently were found to have inherited the same genetic mutation as the potential recipient, they were discarded. Characteristics of Directed Cord Blood Collections The median volume minus anticoagulant was 73 ml (range, ; Table 2). The volume exceeded 40 ml for 236 (88.1%) of the collections. The median TNC count was (mean, ; range, ) and exceeded for 238 (88.8%) collections. Although mean in utero and ex utero collection volumes and nucleated cell counts differed (a median volume of 75 ml and count of vs. 67 ml and , respectively), this difference was not statistically significant. The median total CD34 cell count per collection was (mean, ; range, ). The median TNC viability prior to freezing for the 50% of collections where this information was collected was 99.0% (mean, 98.2% 3.7%), with only four less than 90% (73%, 78%, 84%, and 84%). Microbiology Results Three collections tested positive for anti-hcv but negative for HCV by PCR, with maternal samples testing positive for HCV.

4 2090 Sibling Donor Cord Blood Transplants Table 3. Transplant outcomes Diagnosis Age at transplant (years) Recipient weight (kg) TNC dose ( 10 7 per kg) DCB CD34 dose ( 10 5 per kg) HLA match Days to engraftment Neutrophils ( l 1 ) Platelets ( l 1 ) Outcome as of December 31, 2005 a Fanconi anemia / Alive 9 years 8 months. Full chimerism. No Neutrophil dysplasia / Alive 9 years 5 months. Full chimerism. No Thalassemia major / Disease returned 4 months. Donor BM transplant later. Thalassemia major / Alive 7 years 1 month. No chr. Thalassemia major / Alive 4 years 1 month. No chr. Thalassemia major / Alive 3 years 5 months. No chr. Thalassemia major /10 NR NR Alive 2 years 9 months. Thalassemia major / Alive 1 year 8 months. No chr. Thalassemia major / Alive 13 months. 70% donor chimerism. DLI for a rejection episode. Diamond-Blackfan / Alive 1 year 5 months. anemia Acute lymphoblastic leukemia / Died from relapse 4 years 5 months post-transplant. Acute lymphoblastic / Died from relapse 2 years 6 months leukemia Acute lymphoblastic leukemia post-transplant (25.9 inc. BM) 2.57 (not inc. BM) 10/ Alive 4 years 5 months. a All surviving patients remain well at the time of submission. GvHD information has been reported, where available, to the National Blood Service. Abbreviations: BM, bone marrow; chr, chronic; DCB, directed umbilical cord blood; DLI, donor lymphocyte infusion; GvHD, graft-versushost disease; HLA, human leukocyte antigen; inc, including; NR, not reported; TNC, total nucleated cells. At the request of the referring clinician, these collections were not discarded but stored in a quarantine vessel pending HLA typing results. Prior to 1999, 7% of collections showed evidence of bacterial contamination. After a change in practice where an iodine swab was used in addition to alcohol wipes, this decreased to 2.4%. The microbiology laboratory indicated that contamination probably took place at the time of collection and provided antibiotic sensitivities. Where the contamination did not preclude use for transplantation, with provision of suitable antibiotics, units were retained in storage. Transplant Characteristics Of the 233 units collected for 217 existing siblings, 65 (28.0%) were a full 10/10 antigen match at HLA A, B, Cw, DR and DQ loci. Of the matched units, 13 were issued for transplantation (Table 3), with all but 3 being used for nonmalignant disorders. The median age of the recipients at transplantation was 5.8 years. This represented median and mean ages of 5.0 years for recipients with malignant disease (n 3) and 6.0 and 6.5 years, respectively, for recipients with nonmalignant disorders (n 10). Cord blood was stored, on average, for 17 months (median, 13; range, 4 47) before transplant. Three units were transplanted for acute lymphoblastic leukemia (ALL), seven for -thalassemia major, one for Fanconi anemia, one for Diamond- Blackfan Anemia (DBA), and one for neutrophil dysplasia. One of the transplanted cords was collected following prenatal HLA typing, and another was collected following preimplantation HLA typing of the embryo prior to pregnancy. A second matched donation was stored following preimplantation typing. None of the units collected from mothers with a family history of inherited disease were transplanted; however, the mothers may yet give birth to affected children who are HLA matches. Transplant Outcomes The median TNC dose for transplantation was TNC per kilogram (mean, ; range, ), with a median recipient body weight of 19.2 kg (mean, ; range, kg) for the 12 patients where DCB units were transplanted alone. One transplant for ALL combined a DCB dose of TNC per kilogram with BM collected from the same donor later to give a combined dose of TNC per kilogram (Table 3, last patient). Total CFU in the frozen units tested was on average CFU per donation (range, ; n 6). This is in line with results for unrelated UCB units stored by the National Health Service CBB ( ) [2]. The median time to neutrophil engraftment, where records were available (n 12), was 18 days (mean, 20 8; range, 14 28), and for platelets, it was 30 days (mean, 30 12; range, 11 59). Engraftment rates did not correlate with TNC or CD34 cell dose, although numbers are small. Two ALL patients engrafted but died from relapse at 2.5 and 4.5 years post-transplant; the third, who received a combined DCB and BM, is alive without evidence of disease. All patients transplanted for inherited disorders engrafted for both neutrophils and platelets, with all but one within 40 days. Although all these patients are alive, one of the seven patients transplanted for -thalassemia, where time to engraftment was the most protracted, relapsed after 4 months and survives following a BM transplant from the same sibling. One of the other -thalassemic patients received a donor lymphocyte infusion to combat a rejection episode. The patient with Diamond-Blackfan anemia, transplanted 6 years after diagnosis, is still alive and well with no chronic GvHD reported. The recipients with neutrophil dysplasia and Fanconi anemia have both survived over 9 years to date with full chimerism and no chronic

5 Smythe, Armitage, McDonald et al DISCUSSION Since the amalgamation of regional services into a National Blood Service in England, the DCB procedures follow standard practice. The collections are now carried out by midwifery staff in the hospital selected for delivery by the mother. The midwifery departments were willing to help because the collections were for high-risk families rather than private companies, a distinction made by the U.K. Royal College of Midwives, the Royal College of Obstetricians and Gynaecologists [2, 37, 38], the American Academy of Pediatrics [39], and the World Marrow Donor Association [40]. We believe our NBS infrastructure, with its uniform standards and policies, is an important factor in the success of the DCB program. It provides the expertise, coordination, and capacity to respond to clinical needs throughout the country and is supported by the extensive NBS quality systems. Over a 10-year period, the NBS has seen an increase in the number of referrals to over 60 per year as awareness among clinicians and mothers increases. Despite collections from more than 60 hospitals, the vast majority were suitable for transplantation based on TNC counts. The proportion of unsuccessful collection attempts (2.9%) is in line with figures for allogeneic UCB banks [15, 16]. The median TNC and CD34 cell counts are comparable to those reported in a multicenter DCB program in the U.S. [15, 16]. The CD34 counts compare very well with the UCB banks that also use this as a measure of suitability [9, 41]. A proportion of unrelated UCB banks [2, 7, 8] adopt a minimum volume of 40 ml and cell count of TNC, and these criteria were met for 88% of the our collections. It is worth noting that six collections of less than 40 ml had TNC counts of more than cells. Of further interest, three of the units transplanted had volumes of 14, 18, and 25 ml, yielding TNC counts of , , and , with doses of , , and TNC per kilogram, respectively, and all three engrafted within 30 days. These results confirm our policy not to automatically exclude units with less than 40 ml, and we agree with Walters et al. [16] that this cut-off is not appropriate for DCB banking because the collection has unique potential. For Walters et al. [16], the choice not to process units less than 20 ml represented a 4.4% loss of their DCB collections, whereas we collected only six (2.2%) units of less than 20 ml. We conclude that collection of a restricted numbers of units by a range of midwifery departments is a viable option for DCB banking. We recommend that the decision to retain the DCB is best made once the cell counts and HLA typing results have been obtained. Other studies using unrelated UCB units indicate that a single HLA antigen mismatch can be compensated for by higher cell doses and recommend that only HLA-A-, -B-, and -DR-matched units be used when cell doses TNC per kilogram are available [9]. However, even for low volume collections, there is the possibility of using the DCB unit in conjunction with small volumes of BM from the same donor once the child has reached sufficient age for this to be an option [13, 14, 16]. There is also the possibility of using donor lymphocytes to counter relapse or infection post-transplantation [9, 42, 43]. Donor lymphocytes were used successfully in this study to combat a rejection episode following one transplant for thalassemia. Those collections that were less than 40 ml were not associated with one hospital or NBS center but were slightly more likely to have been collected ex utero (57%). We and Wall et al. [44] reported little difference in the volume of UCB units collected using either in utero or ex vivo techniques, whereas others [45] have observed higher CD34 cell counts and lower rates of contamination with in utero collections. However, the safety of mother and child is paramount, and the U.K. Royal College of Obstetricians and Gynaecologists [38] has recently suggested all collections should be made ex utero. The median TNC viability prior to cryopreservation was 99%, with only four samples scoring lower than 90%. This suggests that a validated means of transport with a reliable courier is an effective way to avoid loss of viability in transit. Wada et al. [46] have noted lower viabilities of UCB units with increased transit times and lower volumes. However, there was no clear reason (as noted by others [47, 48]) to explain the few low viabilities in our studies. We recommend using a singleplatform CD34 assay that includes viability assessment on fresh samples and combining this with the CFU assay on frozen samples to measure cell numbers accurately and determine viability of the CD34 cell population. UCB banks typically volume-reduce to 21 ml [7, 8, 33, 49], whereas we did not reduce the volume of DCB units. The numbers of DCB units are relatively low, and so the benefit from not losing as much as 20% of cells through volume reduction [49] or even total loss during processing becomes more important than the larger storage space required. Improving umbilical cord cleaning by using alcohol wipes followed by iodine swabs dropped bacterial contamination to 2.4%, which compares very well with other DCB programs [16] (3.3%) and is probably at a minimum [50, 51]. Bacterial contamination can be addressed at the time of transplant with antibiotics, but units contaminated with dangerous organisms are discarded. Units from an HCV-positive mother could be considered where follow-up testing of the donor child indicates there was no vertical HCV transmission. All units are now typed at the time of collection using PCR-SSP for HLA-A, -B, -Cw, -DRB1, and -DQB1 to ensure that the degree of match can be determined fully and so best inform the decision to store the cord blood. NBS Histocompatibility laboratories are all accredited by the European Federation for Immunogenetics. The NBS recommends high-resolution (allele) typing for all potential transplant donors and recipients where any uncertainty regarding HLA compatibility is identified. We found that 28% of collections for a sibling were an HLA 10/10 match, in line with Mendelian segregation. An identical matched sibling donor is preferred as the standard of care to an unrelated donor as a clinical option in many transplant settings, as indicated recently by the fourth European Bone Marrow Transplantation Society (EBMT) report on current practice for transplantation in Europe [52]. Where stem cells from related and unrelated donors are both listed as standard of care options, an unrelated UCB unit may be preferred to a related UCB unit to provide a sufficient cell dose. As the practice of preimplantation HLA testing becomes more widely accepted, it is likely that the proportion of related matched units and transplants will increase [53 55]. UCB collections that are not a full match may be used successfully, as shown for unrelated UCB transplants, where 80% of transplants have one or two mismatches. Indeed, a limited mismatch may be preferable for some malignancies, bearing in mind that cell doses of TNC per kilogram have been recommended for transplants with a one- or two-hla disparity [9]. Where there is a family history of an inherited disorder, the DCB collections are almost always stored for potential future use. Most of the collections for nonmalignant disorders were for severe combined immunodeficiency, and matched UCB units have been successfully used for transplant in immunodeficiency from both related and unrelated donors [9, 56]. Cord blood units have been transplanted after many years [9, 11, 57], so long-term storage may be appropriate. Continued storage should, however, be reviewed at least annually to ensure that units are not stored unnecessarily. A unit with one