The Lombardy Rare Donor Programme

Abstract

Background

In 2005, the government of Lombardy, an Italian region with an ethnically varied population of approximately 9.8 million inhabitants including 250,000 blood donors, founded the Lombardy Rare Donor Programme, a regional network of 15 blood transfusion departments coordinated by the Immunohaematology Reference Laboratory of the Ca’ Granda Ospedale Maggiore Policlinico in Milan. During 2005 to 2012, Lombardy funded LORD-P with 14.1 million euros.

Materials and methods

During 2005–2012 the Lombardy Rare Donor Programme members developed a registry of blood donors and a bank of red blood cell units with either rare blood group phenotypes or IgA deficiency. To do this, the Immunohaematology Reference Laboratory performed extensive serological and molecular red blood cell typing in 59,738 group O or A, Rh CCDee, ccdee, ccDEE, ccDee, K− or k− donors aged 18–55 with a record of two or more blood donations, including both Caucasians and ethnic minorities. In parallel, the Immunohaematology Reference Laboratory implemented a 24/7 service of consultation, testing and distribution of rare units for anticipated or emergent transfusion needs in patients developing complex red blood cell alloimmunisation and lacking local compatible red blood cell or showing IgA deficiency.

Results

Red blood cell typing identified 8,747, 538 and 33 donors rare for a combination of common antigens, negative for high-frequency antigens and with a rare Rh phenotype, respectively. In June 2012, the Lombardy Rare Donor Programme frozen inventory included 1,157 red blood cell units. From March 2010 to June 2012 one IgA-deficient donor was detected among 1,941 screened donors and IgA deficiency was confirmed in four previously identified donors. From 2005 to June 2012, the Immunohaematology Reference Laboratory provided 281 complex red blood cell alloimmunisation consultations and distributed 8,008 Lombardy Rare Donor Programme red blood cell units within and outside the region, which were transfused to 2,365 patients with no untoward effects.

Discussion

Lombardy Rare Donor Programme, which recently joined the ISBT Working Party on Rare Donors, contributed to increase blood transfusion safety and efficacy inside and outside Lombardy.

Introduction

Prompt availability of red blood cells (RBC) is a key element of supportive therapy in severely anaemic medical and surgical patients. Timely provision of compatible RBC by transfusion services can be challenging in patients developing RBC alloimmunisation¹, particularly if patients have antibodies to multiple or public RBC antigens^2–4 or belong to ethnic minorities⁵. Data from the literature^1,6,7 and local transfusion recipient records from this department (Table I) show that alloimmunisation rates can be very high in selected groups of regularly transfused patients such as those suffering from thalassaemia or sickle cell disease^8–15, as well as patients with chronic myeloproliferative diseases, aplastic anaemia, recurrent gastrointestinal haemorrage or waiting for solid organ transplantation. An unpublished analysis of 298 chronic recipients from this Blood Transfusion Service, in part originally transfused in other hospitals and transferred to our institution for management of RBC alloimmunisation, showed alloimmunisation rates of 23%, 30% and 30% in patients affected by thalassaemia intermedia, thalassaemia major and sickle cell disease, respectively. The high rates of alloimmunisation in multiply transfused patients are probably due to lack of standardisation of better match policies in our country. We expect that alloimmunisation rates could significantly decrease upon implementation of a national policy on better matching or full phenotype matching. To overcome the above difficulties, a number of regional, national and international rare blood donor programmes have been developed.

Table I.

Frequency distributions of RBC alloimmunisation to common and to high-frequency antigens detected in 2,371 (2.9%) patients out of 82,234 RBC transfusion candidates routinely screened at the LORD-P coordinating transfusion department during 2005–June 2012.

Common antigens (n=2,323; 98%)	Percent
Anti-D	20.0
Anti-K	17.7
Anti-E	14.8
Anti-M	6.2
Anti-c	5.0
Anti-C	5.0
Anti-Jka	4.4
Anti-Cw	2.9
Anti-Lea	2.9
Anti-Lua	2.3
Anti-Fya	2.2
Anti-S	2.2
Anti-Leb	1.1
Anti-s	0.4
Anti-Bga and not identified	12.9
High-frequency antigens (n=48; 2%)	Percent
Anti-Yta	19.6
Anti-k	13.7
Anti-Lub	11.8
Anti-Kpb	9.8
Anti-PP1Pk	9.8
Anti-Vel	9.8
Anti-JMH	5.9
Anti-Ge	3.9
Anti-Lan	3.9
Anti-U	3.9
Anti-Coa	2.0
Anti-Hro	2.0
Anti-Jra	2.0
Anti-Ku	2.0

This article describes the services and outcomes of the Lombardy Rare Donor Programme (LORD-P), which was established in 2005 in the Italian region of Lombardy through the collaboration of the regional network of 15 blood transfusion departments under the coordination of the Immunohaematology Reference Laboratory (IRL) of the Immunohaematology and Transfusion Centre, IRCCS Foundation, Ca’ Granda Ospedale Maggiore Policlinico, Milan¹⁶.

LORD-P data are discussed within the broader framework of other programmes providing similar services at regional, national and international levels^17–25.

Materials and methods

Immunohaematology Reference Laboratory

The IRL was accredited by the AABB in 2003¹⁶. In the period from 2005 to June 2012, the IRL performed the following functions within the LORD-P: (i) studies of complex immunohaematology cases; (ii) studies of rare blood group phenotypes; (iii) quality controls of reagents; (iv) production of biological standards; (v) cryopreservation of samples from donors and patient; (vi) molecular studies of blood group antigens; (vii) development and standardisation of multiple techniques; (viii) preparation of panels of reference cells or DNA; and (ix) management of inter-regional and international exchanges of RBC units and samples from donors and patients.

Lombardy transfusion departments

At the start of the LORD-P, each of the 15 regional transfusion departments selected from their own databases blood donors aged 18–55 with a history of two or more blood donations and a set of pre-defined phenotypes (see below). A small additional blood sample was collected for extensive typing at the IRL after informed consent from the donor obtained at the time of the first donation occurring after preliminary selection from the historical database. Progressive RBC typing batches were scheduled in agreement with the IRL, according to budget allocations.

Red blood cell typing procedures

During the period from 2005 to January 2012, extensive RBC serological and molecular typing was carried out in 59,738 group O or A, Rh CCDee, ccdee, ccDEE, ccDee, K− or k− blood donors aged 18–55 with a record of two or more blood donations.

From 2005 to June 2009, we used a high-productivity solid-phase automated serology system for the first typing (Galileo, Immucor, Norcross, Georgia, USA). The reagent panel included the following antigens: Fy^a, Fy^b; Jk^a, Jk^b; S, s; Co^a; Js^b; Ge2; Lu^b; Kp^b; PP₁P^k; U; Vel and Yt^a. Until December 2006, antigen-negative status was confirmed serologically using different antisera by Galileo or by manual tube testing. Starting from January 2007, all confirmatory typing has been performed by molecular methods employing a local procedure using Luminex technology (Luminex Corp., Austin, Texas, USA) or by commercial polymerase chain reactions with sequence-specific primers (PCR-SSP; Inno-Train Diagnostik, Kronberg, Taunus, Germany). In April 2009, with the availability of new DNA-based technologies, we decided to validate the BeadChip^TM platform (BioArray Solution Ltd., Warren, New Jersey, USA), a high-throughput genotyping method based on DNA microarray technology. These developments allowed us to increase the number of RBC alleles to type. In particular, we performed molecular testing using the BioArray BeadChip HEA kit (Immucor), which simultaneously performs multiple assays on one sample and analyses 33 polymorphisms associated with 11 blood group systems including RH, KEL, JK, FY, MNS, LU, DI, CO, DO, LW and SC²⁶.

Screening for IgA deficiency

We followed the American Rare Donor Program (ARDP) criteria²⁵ for the selection of IgA-deficient donors²⁷ First we used a turbidimetric method (IgA-2, Cobas, Roche/Hitachi, Indianapolis, Indiana, USA) to screen blood samples. Then a second IgA-deficient blood sample (serum sample) was sent to the National American Red Cross Immunohematology Reference Laboratory (Philadelphia, Pennsylvania, USA) to confirm IgA deficiency according to the stringent ARDP criteria (IgA below 0.05 mg/dL confirmed on two separate occasions, whereas the clinical definition of IgA deficiency is less than 7 mg/dL).

Red blood cell cryopreservation

All rare RBC units forming the LORD-P frozen inventory were cryopreserved in the primary bag with a high-glycerol procedure according to Valeri et al.²⁸ and stored in a mechanical freezer at −80 °C. Since 2005 all units have been cryopreserved with the ACP 215 closed system (Haemonetics, Braintree, Massachusetts, USA), which extends the shelf-life of thawed RBC units from 24 hours to 7 days²⁹. In agreement with a written LORD-P policy, the standard 10-year expiry of non-rare cryopreserved RBC units can be overridden for the precious units forming the LORD-P inventory.

Results

Red blood cell typing and the bank of frozen, rare red blood cells

The RBC typing yielded 8,747 donors who were rare for a combination of common antigens (RCA), 538 negative for high-frequency antigens (NHF) and 33 with a rare Rh phenotype (RRH). On 30 June 2012, the LORD-P registry included the following clinically relevant phenotypes: Oh, CCDEE, CCdee, ccdEE, −D−, S–s–U−, Lu(a+b−), Lu(a−b−), Kp^b−, Js^b−, Fy(a−b−), Co^a−, Vel−, Ge:−2, k−, Jr^a−, Yt^a− and PP₁P^k−, whereas SC:−1, Lw(a−b−), Ko, Jk(a−b−), Lan−, I−, P− and P^k− were missing.

On June 30, 2012, the LORD-P frozen inventory included 1,157 cryopreserved units (Table II). The numbers and phenotypes of autologous cryopreserved units are shown in Table III.

Table II.

Inventory of rare RBC units cryopreserved at the LORD-P coordinating transfusion department (June 2012). RRH: rare for Rh phenotype; NHF: negative for high frequency antigens; RCA: rare for combination of common antigens.

N. of RRH units	105
CDE/CDE	22
Cde/Cde	61
cdE/cdE	13
−D−/−D−	9
N. of NHF units	910
Coa−	69
Co(a−b−)	0
Dib−	0
Do(a+b−)	0
Fy(a−b−)	148
Ge:−2	30
Jk(a−b−)	0
Jra−	1
Js(a+b−)	7
k−	158
K:−11	0
Ko	0
Kp(a+b−)	3
Kp(a−b−)	0
Lan−	0
Lu(a+b−)	142
Lu(a−b−)	53
LW(a−b−)	0
LW(a−b+)	0
Oh	19
PP1Pk−	20
Rh:−51	0
Rhnull	0
Sc:−1	0
S–s–U−	2
S–s–U+	0
Vel−	30
Yta−	228
N. of RCA units	142
Total	1,157

Table III.

Inventory of rare autologus RBC units cryopreserved at the LORD-P coordinating transfusion department (June 2012). RRH: rare for Rh phenotype; NHF: negative for high frequency antigens.

RRH units	N. of patients	N. of units
CDE/CDE	1	1
−D−/−D−	3	10
NHF units
Fy(a−b−)	3	19
Ge:−2	1	4
Jra−	5	15
Ko	2	8
Lan−	1	2
Lub−	1	3
PP1Pk−	5	27
Sc:−1	1	1
Vel−	3	16
Total	26	106

Consultations and distribution of units

Since 2005, 281 cases of complex alloimmunisation (defined as patients with ≥3 antibodies, antibodies to high incidence antigens or auto-alloimmunisation with ≥2 antibodies) have been referred to the IRL from our hospital (46%), from other hospitals in the region of Lombardys (25%) and from hospitals in other Italian regions (29%).

A total of 8,008 (7,611 fresh and 397 thawed) RBC units were distributed through the LORD-P during the period from 2005 to June 2012. Table IV shows the numbers and types of NHF RBC units distributed from 2005 to June 2012 for homologous and autologous transfusion. During the period 2005–June 2012, all requests for compatible units were met, with the exception of one request for a patient with anti-P antibodies, who was supported with units retrieved from the Finnish Red Cross.

Table IV.

Number of RRH and NHF RBC units distributed during 2005–June 2012 for homologous and autologous transfusion. RRH: rare for Rh phenotype; NHF: negative for high frequency antigens.

Specificities	Homologous	Autologous	Total
CCdee	8	0	8
ccdEE	15	0	15
CCDEE	1	0	1
Coa−	6	0	6
Fy(a−b−)	77	0	77
k−	34	2	36
Ku−	0	9	9
Lub−	17	0	17
PP1Pk−	14	18	32
Vel−	6	5	11
Yta−	7	0	7
Total	185	34	219

In detail, from November 2011 to June 2012, 12% of LORD-P NHF RBC units were used for a 24-year old African patient with sickle cell disease on the waiting list for liver transplantation who underwent six RBC exchange procedures every 40 days, each with four or five Fy(a−b−) RBC units. FY genotyping of this patients was performed by HEA Beadchip. The patient showed homozygosity for FY*B-67C with GATA box mutation in both alleles³⁰. According to Wilkinson et al.³¹, allowing Fy(b+) products for patients with GATA mutation increases the number of available products in a Blood Centre, although it does not ensure prevention of FY antibodies in all patients. For this reason, the fact that the man was young and in consideration of his enrolment in a liver transplantation programme, we chose to support our patient with Fy(a−b−) units.

With regards to screening for IgA deficiency, from March 2010 to June 2012 we screened 1,941 male AB donors and found only one IgA-deficient donor. In addition, we confirmed IgA deficiency in four donors previously identified by another transfusion department. Anti-IgA antibodies were found in only two blood donors. No requests for IgA-deficient units have been received so far.

Discussion

This article updates and expands a 2005–2006 report³² which described the initial developments and results of the LORD-P. From 2007 to June 2012, through the cooperative efforts of all 15 blood transfusion departments in the region of Lombardy and with the support of a regional government grant, the number of extensively typed repeat blood donors increased from 20,714 to 59,738. This was accompanied by a parallel increase from 2,880 to 9,318 rare blood donors registered in the LORD-P data base and by an increment from 351 to 1,157 of the number of rare RBC units cryopreserved in the LORD-P frozen bank. As expected, the availability of a significantly larger rare donor pool entailed an increase in the average number of rare units distributed per annum by the IRL from 1,012 during 2005–2006 (total: 2,024 units to 142 patients) to 1,456 during 2007–June 2012 (total: 8,008 units to 2,365 patients). Moreover, all the 15 transfusion departments have benefitted from the LORD-P registry of extensively typed donors for the local management of cases of alloimmunisation not requiring IRL consultation. Although we are aware of the possibility of inaccurate or incomplete reporting of adverse events and transfusion reactions, it is encouraging that no reports of acute or delayed haemolytic transfusion reactions were received by the IRL after the transfusion of the rare RBC units collected from LORD-P registered donors or thawed from the LORD-P frozen bank. This evidence supports the quality of the donor RBC typing procedures and patient immunohaematology work-up developed by LORD-P members with the coordination of the IRL.

Besides its value at the regional level, the LORD-P registry and frozen bank has the potential to contribute to resolving planned and urgent transfusion needs of complex transfusion candidates at national and international levels, within the broader framework of other rare donor programmes developed in several countries.

Although Italy still lacks a national rare donor programme, our Bank has already reached a size comparable to that of the rare donor programmes developed in France and UK, both started in the 1960s, which currently include 17,800 and 9,000 rare donors, respectively.

With this perspective, the LORD-P recently joined the Working Party on Rare Donors formed in 1984 by the International Society of Blood Transfusion (ISBT), which currently reports membership from rare donor programmes developed in 16 countries (http://www.isbtweb.org/working-parties/rare-donors/members/, accessed on 31 July 2012). The main outcomes and achievements of this working party were published in 2008 after 24 years of international collaboration²¹.

With regards to RBC typing procedures used in transfusion medicine, assays based on haemagglutination have been and remain the gold-standard methods³¹. Haemagglutination is simple, effective, inexpensive and, when done correctly, serves transfusion medicine in the majority of clinical cases. However, haemagglutination is a subjective test and has limitations, including limited automated high-throughput capability and short supply or unavailability of commercial reagents for all clinically relevant antibodies. In fact, in the last few years licensed serological antisera have become expensive and their availability has decreased^31,33.

Genes encoding a large number of blood group systems have been cloned and sequenced and the molecular bases of most blood group antigens and phenotypes have been determined^34–40. Furthermore, it may be expected that large scale genotyping, if applied to routine blood donors, could significantly change the management of blood provision in many regions^2,41. In fact, extensive typing from each of the 15 blood transfusion departments of Lombardy has become unnecessary following implementation of the LORD-P. In the future, based on our inventory of rare donors blood, we plan to implement a strategy to use full phenotype matching in patients with haemoglobinopathies.

In order to meet the demand of routine blood group genotyping for large numbers of donors and patients, genotyping technologies need to be high throughput and, above all, automated, accurate and cost-effective^42,43. However, cost-effectiveness must not be judged simplistically on the raw cost per test. The potential benefits of having a comprehensive genotype of a donor or patient may minimise transfusion complications as alloimmunisation may be reduced.

A common criticism of blood group genotyping is that genotype does not always reflect phenotype. In fact, under some circumstances genotype does not correlate with serological phenotype. Most DNA-based assays analyse target-specific nucleotides, but cannot determine whether the gene is silenced or not expressed⁴⁴. For this reason, for the foreseeable future haemagglutination methods will remain the primary immunohaematology methods, with DNA-based assays used as adjuncts to them^33,42.

Although the primary aim of rare donor programmes such as the LORD-P is to provide timely and effective treatment to complex transfusion candidates, the study of blood donors with rare variants of gene products on the surface of their RBC offers important perspectives to improve our understanding of blood group functions in physiological and pathological conditions^36,45,46. As a recent example of this work, an elegant article by Saison et al.⁴⁷ entitled “Null alleles of ABCG2 encoding the breast cancer resistance protein define the new blood group system Junior”, discusses the clinical relevance of the fact that “these ABCG2 Jr(a−) individuals are expected to be hypersensitive to the drugs transported by ABCG2” and concludes that “our ability to identify them by blood typing is of major clinical interest in order to predict drug responses and to establish optimal and more personalized drug dosages”. Moreover, with migration in the world becoming more prominent, blood transfusion services will have the opportunity to actively promote blood donation from ethnic minorities, with the ultimate aim of identifying different prevalent antigen patterns.

Finally, additional research perspectives are offered by recent studies reporting preliminary evidence of the possibility of ex-vivo production of mature RBC from stem cells^48–50. Once confirmed and implemented as sustainable industrial processes, these developments could pave the way to ex-vivo production of RBC with rare phenotypes for both diagnostic and therapeutic purposes, which could facilitate timely provision of effective RBC to transfusion candidates with complex RBC alloimmunisation⁵¹.