Alpha 1-antitrypsin deficiency: memorandum from a WHO meeting.

Memoranda/Memorandums a1-Antitrypsin deficiency: Memorandum from a WHO meeting* a1-Antitrypsin (AAT) deficiency, also known as a1-antiprotease inhibitor deficiency, is a disease caused by genetically determined AAT deficiency. It occurs as a result of inheritance of two protease inhibitor (PI) deficiency alleles from the AAT gene locus (designated Pl) on chromosomal segment 14q32. 1. The most common deficiency allele is Pl*Z and a large majority of individuals with severe AAT deficiency are Pl type Z. The disease occurs predominantly in white persons of European origin and its frequency in Europe and North America is comparable to that of cystic fibrosis (1 in 2000 to 1 in 7000.) Persons with AAT deficiency may have no clinical manifestations. Chronic obstructive pulmonary disease (COPD) with a high frequency ofpanacinar emphysema is the most prevalent clinical disorder associated with AAT deficiency and the most frequent cause of disability and death. Tobacco smoking is the major risk factor for developing COPD, which generally begins by the third decade of life, much earlier than "usual" COPD that occurs in AAT-replete individuals. Liver disease, the second most frequent clinical manifestation of AAT deficiency, typically presents as cholestasis in infancy but is usually not severe and generally remits by adolescence. Chronic liver disease develops infrequently, although AAT deficiency is the commonest cause of chronic liver disease in childhood. Cirrhosis and carcinoma of the liveraffect at least25% ofAAT-deficient adults over the age of 50 years. AAT deficiency appears to be widely underdiagnosed and based on predicted gene frequencies even in the most intensely studied populations, only a small proportion of those predicted to have AAT deficiency have been diagnosed. Human AAT is available in limited quantity for augmentation therapy. This Memorandum summarizes the discussions and recommendations made by participants at a WHO meeting held in Geneva on 18-20 March 1996 to review existing knowledge about this highly prevalent genetic disorder, develop a strategy for enhancing awareness of it among health-care-givers and the general public, and explore new case-finding and disease-prevention strategies. Biology of a1-antitrypsin deficiency: a conformational disease ac-Antitrypsin (AAT), also known as a1-protease inhibitor, is the protease inhibitor present in highest concentration in human plasma (1, 2). Its prime physiological target is neutrophil elastase. The serpin family of serine protease inhibitors, to which AAT belongs, share a relatively complex template * The participants at the meeting were as follows: Dr A. Barker, Portland, OH, USA; Dr M. Brantly, Bethesda, MD, USA; Dr E. Campbell, Salt Lake City, UT, USA; Dr R. Carrell, Cambridge, England; Dr D.W. Cox, Toronto, Ontario, Canada; Dr A. Dirksen, Copenhagen, Denmark; Dr J.A. Dodge (Vice-Chairman), Belfast, United Kingdom; Dr M.A.F. El-Hazmi, Riyadh, Saudi Arabia; Dr S. Eriksson, Malmo, Sweden; Dr E.K. Ginter, Moscow, Russian Fed- eration; Dr N.G. McElvaney, Bethesda, MD, USA; Dr A. Mehta Cleveland, OH, USA; Dr A. Propst, Innsbruck, Austria; Dr R. Reprint No. 5795 structure (3-6), which has the unusual property of being able to change from one conformational form to another (7, 8). This molecular mobility allows these inhibitors to snare a target protease and to enfold it, forming a tight complex that may circu- late in the blood for an extended time before it is catabolically cleared. Mutations affecting the do- mains controlling molecular mobility can allow these conformational changes to take place prematurely with consequent dysfunction and disease (9). Sandhaus, Denver, CO, USA; Dr G.L. Snider (Rapporteur), Bos- ton, MA, USA; Dr R.A. Stockley, Birmingham, England; Dr J.K. Stoller (Chairman), Cleveland, OH, USA; Dr I.C. Verma, New Delhi, India; Mr J. Walsh, Coconut Grove, FL, USA; Dr M. Wencker, Essen, Germany. WHO Secretariat Dr N.P. Napalkov, Dr M. Tsechkovski, Dr V. Boulyjenkov (Secretary), Dr C.-C. Heuck. Requests for reprints of this article should be addressed to Dr V. Boulyjenkov, Human Genetics Programme, World Health Organization, 1211 Geneva 27, Switzerland. Bulletin of the World Health Organization, 1997, 75 (5): 397-415 X World Health Organization 1997 397 Memorandum The most common cause of AAT deficiency is the Z variant of AAT, which is carried by 1 in 25 individuals of Northern European descent (1, 2). This variant results in normal synthesis of antitrypsin but in only 15% secretion; the remaining 85% is blocked in the terminal secretory pathway of the hepatocyte (10). Much of this blocked antitrypsin is degraded but some accumulates as large intracellular inclusions. PI*Z AAT has a mutation (E342K) at the hinge of the reactive loop that affects the entry of the loop into the A-sheet (11, 12). The end result is the loss of the constraints that hold the reactive loop in its external position and the more ready transition of the molecule to the partially incorporated locking form (9). Two variants, Mmalton (FA52) (13,14) and Siiyama (S53F) (15), both have abnormalities in the shutter region that controls the sliding movement of the A-sheet to allow entry of the peptide loop con- taining the reactive centre into the sheet. The consequences of this more ready conforma- tional transition are twofold: there is a partial block in folding of the protein (16) at the time of synthesis; and a tendency of the fully folded protein to sponta- neously flip to the locking form (9). These intermedi- ate forms are susceptible to intracellular catabolism (17) and also to linkage, which results in loop-sheet polymers (18). The tangles of these polymers, visible on electron microscopy, are believed to form the characteristic liver inclusions seen in AAT defi- ciency. Some of the abnormal AAT is released into the plasma as monomers, but the subsequent forma- tions of polymers in plasma, which are particularly marked in the Siiyama variant of the disease (19), contribute to the plasma inhibitory deficiency. Genetics and molecular biology of al-antitrypsin a1-Antitrypsin locus The AAT gene locus, designated PI, is located on chromosomal segment 14q32.1, and resides among a gene cluster that includes cortisol-binding globulin, ac-antichymotrypsin, 19-ac-antitrypsin, al- antitrypsin pseudogene/protease inhibitor-like gene (PI*L) and protein C inhibitor. The entire complex is located within a 280kb region near the immu- noglobulin locus (14q32.33). The arrangement of these five genes within the cluster is as follows: a1-antichymotrypsin/ protein C inhibitor, ac-antitrypsin, PI*L, and cortisol-binding globulin. The a,-antitrypsin and ac-antichymotrypsin/protein C inhibitor loci within the 220kb segment are arranged in opposite orientations to one another (ac-antichymotrypsin and a,-antitrypsin). With the exception of the PI*L pseudogene, these genes code for related proteins from the superfamily of serpins, which have substan- tial amino acid and structural homologies. Cortisol- binding globulin, ac-antichymotrypsin, a,-antitrypsin and PI*L have a similar exon and intron motif. a1-Antitrypsin genomic organization The AAT gene consists of seven exons (Ia, Ib, Ic and II-V) and six introns spanning more than 12kb (Fig. 1). Exon Ia and lb contain elements responsible for macrophage-specific transcription and exon Ic con- tains the promoter necessary for hepatocyte-directed transcription. The coding region includes 1434bp primarily dispersed on four exons. The start codon ATG sequence for the 24 amino acid signal peptide and two of the three N-linked glycosylation sites, N46 and N83, are located in exon II. Exon III con- tains the last of the three glycosylation sites, N247, and the most common polymorphic site within the coding region V213A. Exon V is the site of the most common mutation associated with AAT deficiency E342K, the active site M358, stop codon TAA and the polyadenylation site (ATAA). With the excep- tion of a small portion of the exon Ia, the entire gene is contained within two EcoRI fragments of 4.8kb and 9.6kb. Between the hepatocyte promoter and the start codon there is a 5.3kb intron, which contains a 143 amino acid open reading frame, an Alu family se- quence repeat, and a pseudotranscription start site. The intron-exon boundaries have conserved consen- sus sequences with the exception of the 3'-junction of exon Ic. a1-Antitrypsin gene expression Hepatocyte mRNA is 1.6kb in length and contains exon Ic, exon II-Y and 3'-untranslated se- quence. Three macrophage-derived AAT mRNAs (two 1.8kb and a 2.0kb) have been identified, and are the product of alternative splicing of exons Ia, Ib and Ic. While the principal site of synthesis is the liver parenchymal cell, AAT synthesis also occurs in mononuclear phagocytes, neutrophils, intestinal epi- thelium, kidney parenchyma, and several other sites. Most of the hepatocyte transcription is directed by a 557bp fragment that begins 20 nucleotides up- stream from the transcription start site and contains the TATA box and three elements critical to effi- cient and tissue-specific transcription. These have been designated the X, Y and P elements and share common features with other promoters of acute re- active hepatocyte-derived proteins. Of these ele- WHO Bulletin OMS. Vol 75 1997398 a1-Antitrypsin deficiency Fig. 1. Organization of the a1-antitrypsin gene. Top (5') to bottom (3'), the boxes denote coding elements (exons) of the al-antitrypsin (AAT) gene; the lines be- tween the boxes are intervening sequences (introns). To- gether, these elements span 12 kb. Exons la, lb, and Ic are regulatory elements essential for normal AAT expression. Ia and lb are macrophage-specific regulatory elements and Ic has both macrophage- and hepatocyte-specific regulatory elements. Four kb 3' to exon Ic are the exons responsible for encoding the amino acid backbone of AAT. Exon II encodes the start codon (ATG), signal peptide, and two of the three carbohydrate attachment sites (N46-CHO and N83-CHO). Exon IlIl encodes the third carbohydrate attachment site (N247-CHO). Exon V encodes the active site (M358), the stop codon (TAA) and the polyadenylation signal (ATTAA). 5' Transcriptional start site (macrophages) Transcriptional start site (hepatocytes) ATG start Signal peptide Asn46-CHO Asn83-CHO Asn247-CHO Met358 (active site) TAA (stop) ATTAA (polyadenylation signal) C'Q) 0 3' ments, X and P are essential to normal transcription in hepatocytes. Several transcription factors are involved in the modulation of transcription, including LFB-1/ HNF-1, C/EBP, HNF-3 and HNF-1/LFA-1. Toget- her, both the cis- and trans-isomers of the hepato- cyte promoter play a crucial role in regulating gene expression in a cascade of events that prob- ably includes both cross-regulation and auto-regu- lation. IL-6, which is the major cytokine responsible for the acute phase response, up-regulates hepatocyte AAT transcription. IL-1 and TNF, which are also involved in the acute-phase response, do not in- crease AAT transcription. Up-regulation of AAT transcription mediated by IL-6 probably occurs pri- marily via transcription factor NF-IL6, which has at least two consensus sequences within the a- antitrypsin promoter. Normal protein trafficking of a1-antitrypsin Translation of AAT mRNA results in a 418 amino acid protein including a 24 amino acid signal peptide. Within the membrane of the rough endoplasmic reticulum, the signal peptide is cleaved from the nascent polypeptide, and glycosylation begins at the moment the peptide reaches the lumenal side of the rough endoplasmic reticulum. At the same time as dolichol-linked oligosaccharides are added, protein folding proceeds in the presence of chaperons. After transport to the cis-Golgi, the carbohydrate side- chains are trimmed and complex residues are added. Following processing in the Golgi, the Mr = 52000 globular glycoprotein is rapidly secreted into the serum. This entire process from translation to secre- tion takes less than 90min. Genetic variation at the a,-antitrypsin locus There are at least 49 AAT gene allelic variations characterized by DNA sequencing (as opposed to isoelectric focusing). Approximately 11% of the nu- cleotide sequence codes for AAT protein sequence and not surprisingly, most of the characterized muta- tions are found within the protein coding exons, since screening for mutations is most often initiated based on protein variation. The mutations within the protein coding regions are relatively evenly dis- persed. Two exceptions to this observation are the codon 51-53 region and the codon 361-362 region. The most likely explanation for the concentration of variants in these locations is that these regions represent mutational "hot spots". Analysis of the nucleotide sequence in both of these regions reveals reiterated DNA sequences which may be sites of slippage misalignment-a mechanism of mutation associated with the deletion or insertion of one or more bases since DNA polymerase "slips" on the monotonous DNA template. WHO Bulletin OMS. Vol 75 1997 399 Memorandum Molecular mechanisms of a1-antitrypsin deficiency The molecular mechanisms responsible for AAT deficiency include errors in the expression, transla- tion, and intercellular processing of AAT. Two variants, QOiso di procida and QOriedenburg are associated with the deletion of most of the coding region of AAT. QOwest, QOtrastevere and prob- ably, QObonny blue are associated with abnormali- ties in mRNA splicing or stability. QOhongkong, QOtrastevere, QOclayton and Z are associated with retention in the rough endoplasmic reticulum and intracellular degradation. Molecular epidemiology of a1-antitrypsin deficiency Population studies About 100 genetic variants (protease inhibitior (PI) types) of AAT deficiency are recognizable by isoelectric focusing. The distribution of the genetic types (PI alleles) has been determined for many populations (20-22), with PI*M and its several subtypes, being the most common allele (Table 1). Deficiency due to the Pl*Z allele The most common deficiency allele is PI*Z, and 95% of individuals with AAT deficiency are of PI type ZZ. The highest frequency of PI*Z is in Scandinavia, it occurs throughout white populations including those of the Middle East, but is absent from oriental and black populations, except where there is white admixture, as in the USA. The estimated frequency of the PI*Z allele in North American white popu- lations is 0.0122, corresponding to a frequency of PI*ZZ homozygotes of 1 in 6700. The frequency of PI*Z is higher in Scandinavia: 0.026, as calculated from 200000 Swedish neonates in a newborn screen- ing programme (23). Rare deficiency alleles The 5% of AAT-deficient persons who are not PI*ZZ are accounted for by about 20 rare variants. These variants are important: they are not identified by PI typing; prognosis may differ from that for PI*ZZ; and they provide insight into residues impor- tant for normal protein function. Deficiency alleles may produce amounts of plasma AAT similar to that of the PI*Z allele or may produce no detectable AAT by standard methods. Alleles associated with detectable a1-antitrypsin Table 2 lists those mutations which produce defi- ciency and dysfunctional alleles with plasma concen- trations of generally about 2-15% of normal. These variants have an isoelectric point by PI typing differ- ent from that of the Z variant (24). Family studies, and ultimately DNA analysis, confirm the presence of these rare deficiency alleles. Table 1: Protease inhibitor (PI) allele frequencies for a1-antitrypsin deficiency in selected populationsa PI alleles: No. Population tested Ml M2 M3 Z Other Denmark 909 0.728 0.136 0.082 0.022 0.023 0.009 Netherlands 357 0.679 0.147 0.129b 0.029 0.013 0.003 Portugal 900 0.510 0.260 0.053 0.150 0.009 0.018 U.S. (White) 904 0.724 0.137 0.095 0.023 0.014 0.007 U.S. (Black) 549 0.982 - - 0.015 0.004 - Chinac 1 010 0.709 0.209 0.070 - - 0.012 Japan 746 0.786 0.153 0.062 - - - France 1 653 0.902d - - 0.071 0.014 0.007 Greece 504 0.960d - - 0.028 0.002 0.006 United Kingdom 4042 0Q930d - - 0.052 0.014 0.004 Saudi Arabia 204 0.926d - - 0.052 0.022 - India 430 0.994d - - - 0.006 a See ref. 20-22. b Frequency of PI*M3 plus PI*M4. c Mean of five Chinese populations. d M (all M subtypes). WHO Bulletin OMS. Vol 75 1997400 a1-Antitrypsin deficiency Table 2: Deficient and dysfunctional ac-antitrypsin alleles Allele Basea Exonb Mutation Deficient F M1(Val213) IlIl R223C (CGT TIGT) Z M1(Ala213) V E342K (GAG AAG) T M2 III E264V (GAA - GTA) S M1(Val213) IlIl E264V (GAA - GTA) Mheerlen Ml (Ala213) V P369L (CCC - CTC) Mmalton M2 II F52 (TTC -* Delete) Mmineralsprings M1(Ala213) II G67E (GGG -* GAG) Mprocida M1(Val213) II L41P (CTG -+ CCG) Wbethesda Ml (Val213) V A336T (GCT - ACT) Ml (Val213) II R39C (CGC - TGC) Mpalermo Ml (Val213) F51 (TTC -e Delete) Mnichinan M1(Val213) II F52 (TTC -> Delete)/G148R (GGG- AGG) Plowell Ml (Val213) III D256V (GAT - GTT) Pduarte M4 IlIl D256V (GAT - GTT) Siiyama Ml (Val213) II F53S (TIC -> TCC) Zaugsburg M2 V E342K (GAG -> AAG) Dysfunctional Pittsburg -c V M358R a Indicates the base allele. b Indicates the exon where the mutation is located. c Unknown. The null (QO) alleles The null alleles (PJ*QO, homozygotes designated as PI*QOQO) are associated with no AAT, or <1% of the normal amount of plasma AAT. The null alleles are widespread. Among 112 patients with AAT defi- ciency, Brantly estimated the frequency of all null alleles to be 1.7 x 10-4 , about 1/100th that of the PI*Z allele in a North American white population and similar to that of PI*Mmalton (25). DNA polymorphisms In addition to the extensive variation found in the protein by electrophoretic methods, variation is found in the DNA sequence, as recognized by re- striction enzymes, producing restriction fragment- length polymorphisms (RFLPs). At least seven RFLPs have been identified within or adjacent to the gene, or in the adjacent downstream homologous sequence (26, 27). DNA polymorphisms and haplotypes are of particular interest for evolutionary studies. A spe- cific DNA haplotype is associated with the PI*Z al- lele and indicates a single origin for all PI*Z alleles (28). The DNA haplotype is the same irrespective of ethnic group within northern Europe, and independ- ent of the presence or type of clinical disease. Spe- cific DNA haplotypes are associated with each of the protein variants and can be used to identify the evo- lutionary pathways (26). They have also been useful as a first step for the identification of specific rare deficiency alleles (24). Clinical manifestations of al-antitrypsin deficiency Chronic obstructive pulmonary disease Chronic obstructive pulmonary disease (COPD), i.e. the occurrence of airflow obstruction due to the pres- ence of chronic bronchitis or emphysema, is the most prevalent clinical disorder associated with AAT defi- ciency. About 95% of patients with severe AAT de- ficiency are PI*Z homozygotes. Their serum levels are about 15% of normal (5-6 i.mol/l (30-40mg/dl) compared with normal serum values of 20-53 ,imol/l (150-350mg/dl)). Other phenotypes that may be as- sociated with disease include PI*Null variants and some PI*SZ individuals (29). Emphysema is the dominant lesion in AAT de- ficiency, with basilar predominance a distinctive al- though not diagnostic feature. In the largest avail- able series, of 165 PI*Z homozygotes examined with plain chest radiographs, 85% (n = 140) had some radiographic features of emphysema; and of these individuals, 99.8% had emphysematous changes that included the lung bases, while 24% had emphy- sematous changes confined to the lung bases (30, 31). The severity of airflow obstruction in AAT defi- ciency, age at presentation of respiratory symptoms, physiologically demonstrable airflow obstruction, and mortality vary widely and are strongly related to habitual tobacco smoking. Lung disease rarely oc- curs in childhood (32). Individuals who have never WHO Bulletin OMS. Vol 75 1997 401 Memorandum smoked rarely develop respiratory symptoms before the fifth decade and impairment of forced expiratory spirometry, which is usually mild, is not detectable before the sixth or seventh decade; occasionally, however, severe airflow obstruction develops among those who have never smoked (33). In the U.S. Reg- istry for Patients with Severe Deficiency of Alpha-i- Antitrypsin (35), the mean age ± SD of the 1129 participants was 46 ± 11 years and their mean FEV1 ± SD was 43 ± 30% predicted. In Larsson's series of 246 PI*Z adults with a median age of 52 years, COPD was present in 74.8% of the participants (36). Chronic bronchitis. In contrast to emphysema, which is defined on the basis of its pathology, chronic bron- chitis is defined on a clinical basis as chronic produc- tive cough with no evident etiology, occurring for more than 3 months per year in each of two succes- sive years. In Eriksson's series of 35 patients, 20% had chronic bronchitis (37) and in Brantly's series of 120 patients, 34% had chronic bronchitis (29). In the British Thoracic Association survey, 59% of 132 smokers had chronic bronchitis, compared with only 29% of those who had never smoked (33). Environmental air pollution. There is little informa- tion on the effects of polluted air on the occurrence of COPD among those with AAT deficiency. How- ever, the effects of air pollution on conventional COPD have been extensively studied. The pollutants for which there is most informa- tion are ozone, sulfur dioxide, oxides of nitrogen, acid aerosols, and particles. Changes in morbidity (emergency admissions to hospital) and mortality have been related to changes in ambient sulfur diox- ide levels (38). Most research on particulates has focused on respirable particles (<10[tm diameter, designated PM10). Statistically significant but small positive associations have been described between PM10 levels and increased severity of airway ob- structive disease and asthma (39) and admissions for COPD (40). Having a job with a high degree of exposure to airborne siliceous dusts, metal gases, agents associated with aluminium production, and welding increased the sex-, age-, and smoking-ad- justed odds ratios for obstructive lung disease (asthma and COPD) by 3.6 compared with having a job without such airborne exposure (41). The effects of smoking on symptoms and lung function impair- ment in AAT replete and deficient persons are many times more powerful than the effects of either urban or occupational air pollution. Infection. Lower respiratory infections in early childhood may be associated with COPD in adult life (42, 43). Lower respiratory infection has also been identified as a risk factor for chronic airflow obstruc- tion in adults with severe AAT deficiency (36, 44). The predominant organisms causing infection in per- sons with COPD are Haemophilus influenzae, Strep- tococcus pneumoniae and Moraxella catarrhalis (45). Unidentified genetic factors. In AAT deficiency, the presence of emphysema in a parent or the presence of asthma appear to increase the risk of developing COPD. In a recent series of 52 persons of PI*Z type, 22 individuals who were ascertained because they had COPD (index cases) had a lower FEV1 (28.7% predicted) than 30 persons who were nonindex cases (mean FEV, 76.4% of predicted) (44,46). Only 10 of the 30 nonindex cases had FEV, values -65% predicted. These data suggest that unidentified ge- netic factors may influence disease severity in AAT deficiency. Lung pathology in AAT deficiency. Understanding about the lung pathology ofAAT deficiency is based on descriptions from inflation-fixed specimens from only 14 autopsies. Panacinar emphysema (PAE) in which all elements of the acinus are uniformly in- volved by airspace enlargement and destruction was present in all the lungs. Emphysema, which results in loss of elastic recoil and of airspace attachments to small poorly supported airways, is an established cause of irreversible airflow obstruction in COPD. Bronchiolar pathology (cellular infiltration, goblet cell metaplasia, fibrosis, luminal secretion and in- creased smooth muscle) is a cause of both irrevers- ible and reversible airflow obstruction in COPD (47). Neither bronchial nor bronchiolar pathology has been well described in AAT deficiency, although the high frequency of bronchial hyper-reactivity makes bronchiolar disease and some airflow obstruc- tion on that basis almost certain. Pathological studies of specimens from lung transplantation for AAT de- ficiency provide an opportunity to correct this im- portant hiatus in our knowledge. Pathogenesis of emphysema in AAT deficiency. Experimental studies have shown that elastases, par- ticularly human neutrophil elastase (HNE), which overcome the antielastase defences of the lungs and reach the elastic fibres in the alveolar wall, cause elastin degradation and give rise to emphysema: non-elastolytic enzymes do not have this effect. AAT has the property of permanently inactivating HNE. It has been hypothesized that in AAT defi- ciency, emphysema results from an imbalance between the elastases and the sharply decreased amount of AAT present in the lungs. The induction of emphysema by smoking can be explained by the four- to fivefold increase in total lavageable lung WHO Bulletin OMS. Vol 75 1997402 a,-Antitrypsin deficiency cells (both neutrophils and macrophages) induced by smoking. Thus, the elastase burden of the lungs is increased and the elastase-antielastase imbalance is increased. Oxidants from cigarette smoke and from intrinsic systems in the neutrophil and macrophage can make AAT incapable of inactivating HNE by oxidizing Met358 (the centre of the active site of the molecule), thus further contributing to the elastase- antielastase imbalance. Macrophages in contact with elastic fibres can also degrade elastin by a constitu- tively produced metalloelastase. The metalloelastase is inactivated by tissue inhibitor of metalloprotease, which is also produced by macrophages (48). Desmosine and isodesmosine are cross-linked lysine-containing amino acids that are derived only from elastin and which are not metabolized. Thus, their excretion in the urine is an indicator of excess elastin degradation in the body, although clinical evidence is needed to infer the origin of the damaged elastin in the lungs. Stone (unpublished data) has produced evidence that urinary desmosine is in- creased in AAT deficiency to about the same level as in common COPD, thus providing more direct evi- dence for elastin destruction as the cause of emphy- sema (Fig. 2). Fig. 2. Mean ± SE of urinary desmosine (DES) in three groups: 22 normal individuals who have never smoked (PI*M); 28 A-1-ATD persons, a mix of smokers, ex-smok- ers and those who have never smoked (Pl*Z); and 21 current and ex-smokers with COPD. a: These two patient groups were significantly different from the normal group (P < 0.01). a12- C X 10- 8- 1- a) 6- C') w >, 4-co .C D 2- PI*M a PI*Z a COPD N- :i Pathogenesis of chronic bronchitis in AAT deficiency. Although tobacco smoking is the most impor- tant risk factor for chronic bronchitis, elastase- antielastase imbalance may be its more proximate cause. HNE induces permanent secretory cell metaplasia in the bronchi of hamsters. Secretory leukocyte inhibitor, which is produced by the bron- chial glands, accounts for 80% of the elastase inhibi- tory capacity of sputum in chronic bronchitis (49) and HNE is among the most potent of mucus secretagogues. Asthma and AAT deficiency. As in patients with usual COPD, bronchial hyper-responsiveness is common among AAT-deficient individuals, affecting 20-27% of them (30). The frequency of reported asthma in such individuals has been reported to range from 4% to 34% (33,34,36,37). For a series of 52 subjects, it was reported that chest wheeziness apart from colds, attacks of wheezing and self- reported asthma were all significantly more frequent (P < 0.02) in the 32 persons with FEV, - 65% predicted than in the 20 persons with FEV, > 65% predicted, suggesting that atopic disease might be a risk factor for development of airflow obstruction in AAT deficiency (44). Bronchiectasis and AAT deficiency. Although fewer than 100 cases of bronchiectasis associated with AAT deficiency have been reported (50) and the only available case-control study of bronchiectasis failed to demonstrate an association with AAT defi- ciency, there is little doubt of an association between the condition and AAT deficiency. Brantly (personal communication) reports a 5-10% prevalence of bronchiectasis in 143 patients by tomography; the bronchiectasis involved is most often cylindrical, the type known to occur as part of the lung pathology in usual COPD. Prognosis and natural history of pulmonary disease. Understanding about the natural history of AAT deficiency is incomplete. Lung function is generally well preserved in the first two decades of life. Avail- able estimates of yearly decline in FEV, among adult smokers range from as low as 42 ml per year to as high as 317 ml per year (57, 74-78) (Table 3). In an 11-year follow-up study of 246 PI*ZZ homozygotes, Larsson reported a 37% mortality rate, most frequently ascribed to respiratory failure (59% of deaths), but with a minority of deaths (13%) due to complications of liver disease (36). In a study evaluating survival among 120 PI*Z homozygotes referred to the National Institutes of Health, Brantly et al. reported that the actuarial survival to age 60 years among PI*Z subjects was 16% compared with WHO Bulletin OMS. Vol 75 1997 14 H 1 A- ...... ...... ...... ...... ..... ...... ...... ...... ..... ...... ...... ...... ..... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ....... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ..... ...... ...... ...... ...... ...... ...... ...... ...... ...... ..... ...... ...... ...... ...... ...... ..... ...... ...... ...... ...... ...... ..... ...... ...... ...... ...... ....... ...... ...... ...... 403 Memorandum Table 3: Estimates of the rate of decline of FEV, among a,-antitrypsin deficient individuals Mean FEV1 slope (ml/yr decline) No. potentially Current Never Study, year (ref.) Source eligible PI*ZZ No. recruited smoker Ex-smoker smoked Janus et al., 1985 (34) Laboratory roster 84 35 (42)a 317 61 80 Hutchison et al., 1987 (79) Registry 164 82 (50) 67 44 66 Wu & Eriksson, 1988 (80) Reference laboratory 346 80 (23) 61 81 61 Brantly et al., 1988 (30) N.I.H. referral 120 24 (210) 51 - - Buist et al., 1983 (81) Pooled data 298 105 (35) 104-11 lb (30% < FEV, % predicted < 65%) 42-46 (65% < FEV, % predicted) Seersholm & Kok-Jensen, Danish registry 612 161 (26) 132 58 86 1995 (51) a Figures in parentheses are percentages. b Pooled data. an expected age-matched U.S. survival of 85% (30). While referral bias may cause mortality rates in these studies to be overestimated, it seems clear that se- vere AAT deficiency significantly shortens survival. In the Danish AAT deficiency registry of 591 subjects with PI type ZZ or ZO, of whom 192 were identified by family studies, the overall median life expectancy was 54 years: 52 years for smokers and 67 years for these who had never smoked; 49 years for the index group; and 69 years for the nonindex group. The life expectancy of persons in the nonindex group who had never smoked was not dif- ferent from that of the normal Danish population (52). Under-recognition ofAAT deficiency. There is strong evidence from estimates of the frequency of AAT deficiency in populations of persons with COPD (53) and from the frequency of the gene defect based on direct population studies (1/1575 to 1/5097) (23, 44) that the proportion of AAT-deficient persons who have been identified is a small fraction of those actu- ally in the population: 4.5% in the United Kingdom (33), 6% in Sweden (36), and about 5% in the USA (see Rationale for screening, below). One reason for under-recognition may be failure of physicians to associate manifest COPD with AAT deficiency. A poll of 304 individuals diagnosed to have severe AAT deficiency found an average delay of 7.2 years between the first onset of symptoms and the initial diagnosis of the condition; 43% of respondents re- ported seeing at least three physicians and 12% re- ported seeing 6-10 physicians before the correct diagnosis was made (54). It is not known what pro- portion of severely AAT-deficient persons, particu- larly nonsmokers, have no clinical manifestations. Liver disease Childhood onset. Screening of 200 000 Swedish neonates identified 120 PI*ZZ children (23) and provides the best estimate for risk of liver disease. During the neonatal period, about 70% develop ab- normal liver function tests but only about 10% have clinically significant symptoms. Suggested risk fac- tors for the development of progressive disease among the children who develop hyperbilirubinae- mia during the early weeks of life include being male and having viral hepatitis B, while breast feeding reduces the risk. At 8 years of age, liver cirrhosis with early death occurred in only three children (2.4%), i.e. in 13.6% of those PI*ZZ children with clinical evidence of liver abnormalities in infancy (Fig. 3). However, AAT deficiency and biliary atresia are the two leading indications for liver trans- plantation in children (55). Adult onset. Cirrhosis and fibrosis of the liver were noted in Swedish adults with AAT deficiency (56). Among PI*ZZ patients ascertained through hospital admissions, 12% had liver cirrhosis (36). In a series of 115 adults with AAT deficiency, four had biopsy- or autopsy-proven cirrhosis and two had definite bio- chemical evidence of liver disease, providing a total of 5.2% affected with liver disease (57). The risks for developing liver disease were higher for males than females and reached a peak of 15.4% at age 51-60 years. Pathology of liver disease in AAT deficiency. In con- trast to lung disease, the serum concentration of AAT can be 40% greater than normal in infants or children with liver disease. Liver biopsy in jaundiced WHO Bulletin OMS. Vol 75 1997404 a,-Antitrypsin deficiency Fig. 3. Distribution of liver test abnormalities in AAT-deficient children aged 1-18 years. Reproduced with permission from: Sveger T, Eriksson S. The liver adolescents with al-antitrypsin deficiency. Hepatology, 1995, 22: 514-517. i 70 Pm S-ALAT (D 60- .0 ~~~~~~~~~~~~S-GT ~40- Or-30- c020 N infants typically shows cholestasis, giant cells, and minimal portal inflammation; there may also be fatty infiltration and cellular infiltration around the portal tracts. The development of cirrhosis is presaged by bile duct proliferation and portal tract fibrosis (58). In adults with AAT deficiency, diastase- resistant, periodic-acid-Schiff-positive (PAS-D) globules, proven by specific immunostaining to be AAT, are present within periportal hepatocytes. Liver biopsies of individuals with disease are charac- terized by lymphocytes in close proximity to areas of abundant PAS-D globules. As the disease progresses, the liver becomes fibrotic and occasion- ally displays piecemeal necrosis; bile duct pro- liferation, liver cell dysplasia and hepatocellular carcinoma may occur. The absence of both hepato- cyte inclusions and of liver disease in the null vari- ants of AAT deficiency, in which the liver cells do not produce the protein, provides powerful evidence of a causal association between PAS-D inclusions and liver disease. Natural history of liver disease. The natural history of liver disease in adults is incompletely understood. Environmental factors clearly play a role in the pathogenesis of liver injury in AAT-deficient indi- viduals. Up-regulation of AAT synthesis during an inflammatory response tends to increase steady-state levels of abnormality folded AAT molecules in the hepatocytes and can thereby initiate hepatotoxic consequences. Factors that accumulate during the inflammatory response, such as bacterial endotoxin, interleukin 6, interleukin 1, NF Kb and elastase-AAT complexes, are capable of up-regulating AAT syn- thesis by a distinct mechanism. These findings sug- gest that an infectious agent or episode may induce liver injury, since increased levels of AAT were found in PIZZ patients with chronic liver disease. Much information has come from serial case-control studies in Malmo, where the autopsy frequency has been high (59), and the gene frequency in the popu- lation is well established. There is an increased risk for cirrhosis and hepatocellular carcinoma, particu- larly among males. This risk seems to be unrelated to alcohol abuse or viral hepatitis. The absence of hepa- titis B or hepatitis C viruses among cases, despite careful testing of stored sera in Malmo, is important in view of more recent studies from Innsbruck, Aus- tria, which report that the risk of developing chronic liver disease is unrelated to alcohol abuse, but may be increased by viral hepatitis B or C infection, par- ticularly in heterozygous patients (58, 60). Further- more, a higher susceptibility to viral infection was found in patients with AAT deficiency than in those with no such deficiency. Being male was confirmed to be an independent risk factor for the development of chronic liver disease. The absence of viral co- infection in the Swedish autopsy study on stored sera is controversial and may be explained by a low inci- dence of viral hepatitis B or C in North European countries, but this requires further study. Dermatological manifestations Panniculitis, characterized by inflammatory and necrotizing lesions of the skin, represents the least common of the well-recognized complications of AAT deficiency, with only 26 cases having been re- ported in the English literature up to 1991 and fewer subsequently (61); the prevalence among AAT- deficient individuals is probably less than 1 case per 1000. Panniculitis has accompanied a variety of phe- WHO Bulletin OMS. Vol 75 1997 405 Memorandum notypes, including PI*ZZ, PI*MZ, PI*SS, and PI*MS. The inflammatory skin lesion is thought to result from unopposed proteolysis in the skin, which has been triggered by an unidentified immune phenomenon. Psychosocial impact of AAT deficiency Impact of neonatal screening In the nationwide neonatal screening programme for AAT deficiency in Sweden, identification of the con- dition in neonates had negative psychological effects in some families. The following recommendations resulted from a systematic study of the psychological and psychosocial consequences of screening (62): - parents should be informed in advance of AAT testing of the phenylketonuria sample; - AAT deficiency should be identified and parents informed early; - a doctor's appointment should be specially arranged; - appointments should ideally be attended by both parents; - written information should be made available; - repeated informative appointments should be scheduled; and - the same physician should see the parents at each visit. Impact on patients with AAT deficiency Not surprisingly, AAT deficiency exacts a psycho- social toll on affected persons. In a recent survey of 1730 subscribers to a national AAT newsletter in the USA, 414 questionnaires were returned: 75.3% of respondents with severe deficiency reported at least one adverse effect of their disease: 44.4% retired early, and 19.1% changed to a physically easier job (54). The duration of diagnostic delay, which was frequently several years, correlated with the degree of adverse psychosocial effects. Alpha-1 National Association The need to share support and information provided the impetus for a national patient organization in the USA, the Alpha-1 National Association (A-1NA). The founders of the A-1NA were among the original participants in studies of infusion therapy with puri- fied AAT at the National Institutes of Health. The goal of the association is to improve the lives of those affected by AAT deficiency through support, education, and research. Education and training of board members, patient members, families, and friends has been paramount. Other important activi- ties include dissemination of educational materials, lobbying for research funds, raising funds for ongo- ing operations, facilitation of national support groups, publishing and distribution of the national newsletter (the Alpha-1 News, with about 2000 subscribers), an information hotline (AlphaLine, 1-800-4ALPHA1), free public screenings for AAT deficiency, and an informational CD-ROM for patients and health professionals (AlphaMedia). The guiding principle of the AlNA has been that knowledge about AAT deficiency ameliorates its impact. Augmentation therapy of AAT deficiency Intravenous AAT augmentation therapy Because many PI*SZ individuals whose serum AAT levels exceed 80mg/dl ( 1 [tmol/l) do not develop emphysema, this level was chosen as a target for augmentation therapy of AAT deficiency. Among the 21 patients who received a once-weekly infusion of 60mg per kg body weight of plasma-derived AAT concentrate the mean lowest AAT level was 126mg/ dl compared to a mean pre-therapy level of 30 mg/dl. The half-life of infused AAT, estimated at 4.5 days, was maintained over the 2-year study period, sug- gesting no increase in catabolic breakdown or anti- body formation in response to the AAT infusions. Measurement of serum anti-neutrophil elastase ca- pacity confirmed the functional activity. Monthly bronchoalveolar lavage at 2-6 months showed AAT levels and anti-neutrophil elastase capacities in bronchoalveolar lavage fluid that were above the amount estimated to prevent alveolar destruction. The results of this study formed the basis for the U.S. Food and Drug Administration's approval in 1988 of the pooled plasma concentrate (Prolastin, Bayer), for treatment of severe AAT deficiency in the USA. It seemed reasonable to assume that if enough AAT were given intravenously to maintain serum levels above the target level, emphysema would not develop in a nonsmoker. Approximately 2200 indi- viduals are currently receiving Prolastin in the USA, and its use has been approved in Canada, Germany, and Spain, where a total of approximately 2000 patients are receiving it. In France, 16 patients are receiving a different plasma concentrate. WHO Bulletin OMS. Vol 75 1997406 a1-Antitrypsin deficiency Hubbard et al. evaluated an AAT dose of 250mg/kg given every 28 days to 9 patients for 1 year (63). These patients maintained serum AAT levels of 70-80mg/dl for 20 days of a 28-day infusion in- terval; the mean (+SE) lowest serum AAT level after the 12th infusion was 67 ± 10mg/dl. The bronchoalveolar lavage AAT levels and anti- neutrophil elastase capacities at 28 days were above the target threshold, suggesting maintenance of pro- tection at the alveolar level even though protective serum levels were not sustained for the 28-day cycle. Cammarata et al. were unable to reproduce these results with five patients on monthly augmentation therapy (64). For three patients studied weekly over two monthly cycles, lowest levels were between 25 mg/dl and 40mg/dl. The U.S. Food and Drug Ad- ministration has approved Prolastin use only for in- fusion of 60mg/kg at 7-day intervals. Recombinant DNA-directed AAT is a non- glycosylated protein produced by yeast encoding the normal human AAT except for a terminal methion- ine and three carbohydrate chains. Recombinant AAT has a half-life of a few hours due to rapid renal clearance from the blood and its intravenous use is therefore not feasible (65). Investigations of whether augmentation therapy actually retards emphysema progression have been hampered by the long intervals needed for study, lack of a suitable, noninvasive, objective marker of emphysema, the large numbers of patients needed, and the expense of such trials. The change in FEV, has become the marker for progression or halting of emphysema. It has been estimated that such a trial would need 250 patients in each arm of a placebo versus AAT augmentation trial followed for 3 years. Two preliminary reports address the issue of efficacy of augmentation therapy. Seersholm com- pared the annual change in FEV, among 235 Ger- man patients for 2-6 years on Prolastin and 161 Danish patients for 7 years not on Prolastin. In this retrospective study, the FEV, declined by 63ml per year in the German patients (treated group) and 81ml per year in the Danish patients (untreated group); the difference was not significant (66). A European multicentre double-blind randomized trial which is in its fifth year, consists of 25 patients in a two-arm study: Prolastin (250mg/kg every 28 days) versus placebo (albumin). The end-points are changes in pulmonary function and annual CT scan findings. Finally, the U.S. National Heart, Lung, and Blood Institute (NHLBI) Registry of Patients with Severe Deficiency of AAT (enrolment, 1129 patients) in- cludes individuals both receiving and not receiving Prolastin. Although not a clinical trial, analysis of the decline in FEVy is currently being finalized, and publication of these results is expected soon. Safety of intravenous AAT therapy During the 8 years of clinical experience with Prolastin infusions in the USA, reported side-effects have been relatively few. Headaches, myalgias, arthraligias, and low back pain are the most frequent complaints requiring no treatment or occasional an- algesic use. Deterioration of shortness of breath, probably due to increased protein load with the infu- sion, may occur in patients with severe COPD or heart failure. No confirmed serious viral diseases have occurred as a result of AAT infusion. Never- theless, it is recommended that studies to determine infection status with hepatitis and human immuno- deficiency virsus (HIV) be carried out before initiat- ing infusion; individuals with a nonprotective titre against hepatitis B virus should be immunized. Several lots of Prolastin were recalled in 1994 and 1996 since the plasma used in the purification process had been donated by individuals with sus- pected Creutzfeldt-Jakob disease (CJD), which has, however, not been documented to be transmitted by blood products. Although acute allergic or anaphy- lactic reactions have been few, it is important to be aware that patients who have severe AAT deficiency may also rarely be IgA deficient. Prolastin contains some IgA, and acute anaphylaxis has been well documented in IgA-deficient PI*ZZ patients re- ceiving regular augmentation therapy (Brantly, unpublished observation). Augmentation of AAT by aerosol delivery Direct delivery of AAT to the lung by aerosol has been explored because only 2% of intravenously administered AAT reaches the lungs, with the re- mainder being distributed throughout other body tis- sues. Since the lung is the major organ to be affected by the excessive elastase burden in AAT deficiency, aerosol administration using plasma concentrates of AAT and rAAT is currently under investigation. Synthetic elastase inhibitors for treating AAT deficiency Several commercial enterprises are hoping that the expense, inconvenience, and limited supply of plasma AAT might be overcome by using a synthetic elastase inhibitor to treat AAT deficiency. Synthetic inhibitors of HNE have been based on the identifica- tion of its active site specificity. In general, the small- est peptide inhibitors of HNE have started with the basic Val-Pro-Val or Ala-Pro-Val tripeptide struc- ture identified as being most sensitive and selective for HNE in substrate-based enzymatic assays. Ra- tional chemical design has been based on desired kinetics and route of administration and molecular WHO Bulletin OMS. Vol 75 1997 407 Memorandum modelling has been employed for this purpose. Serendipity has played a role in the development of some inhibitors. For example, nonpeptide com- pounds evolved from an evaluation of the elastase inhibitory properties of cell-wall-bound nucleic acids in bacteria and compounds gaining oral activity and cell penetration by incorporating the lactam moiety of cephalosporin antibiotics. Lung transplantation for AAT deficiency Up to September 1995, more than 4300 patients (about 30% with COPD) had undergone lung trans- plantation worldwide. The 6-year actuarial survival for recipients of all types of lung transplant is ap- proximately 40%; emphysema patients have a 4-year survival rate of 54%, the best level of all groups. Most survivors have experienced useful functional improvement. Because of the shortage of suitable lung donors, most patients now receive a single lung transplant. Approximately 12% of lung transplants are per- formed for emphysema caused by AAT deficiency. AAT-deficient recipients are about 6-7 years younger than other COPD patients and have an ac- tuarial survival of 45% at 5 years. Lung volume reduction surgery Lung volume reduction surgery (LVRS) consists of multiple wedge resections of the most severely in- volved emphysematous lung (target areas), identi- fied by imaging. Surgical decrease of marked hyperinflation is believed to improve both the me- chanical properties of the respiratory system and blood-gas exchange. Those with widespread, uni- form emphysema, pulmonary hypertension (mean pulmonary artery pressure >4.67 kPa (35mmHg)) severe hypercapnia (paCO2 > 7.33kPa (55mmHg)), very low FEV, (<20% of predicted), and skeletal deformities are considered less than ideal candi- dates. These exclusion criteria are, however, arbi- trary and need validation. Reported mortality, usually from respiratory failure, is about 5%. Im- provement in FEV, is highly variable with improve- ment in dyspnoea and 6-minute walking distance often greater than changes in FEV,. Benefits have lasted for 1 year, the maximal length of follow-up to date. Only a few patients with AAT deficiency have undergone LVRS and no data have been reported. Randomized controlled trials to evaluate LVRS are in their early stages. Prospects for gene therapy in AAT deficiency AAT deficiency is theoretically an attractive target for gene therapy since the AAT molecule functions extracelluarly and has a broad safety margin. The phenotypical abnormality, deficiency of AAT, could be corrected by transferring the normal human AAT cDNA to the cells of deficient individuals - and these cells could be anywhere in the body- as long as sufficient AAT reaches the lower respiratory tract. However, gene therapy represents a radical intervention and it is generally regarded that the condition to be treated in this way must have no effective alternative therapy. This is a moot point for AAT deficiency; augmentation therapy with AAT is available but is expensive, inconvenient, and in short supply. Three general transfer strategies have been evaluated in AAT deficiency: retroviruses, adeno- viruses, and nonviral vectors. For gene therapy, the retrovirus is modified to make it able to enter the cell and insert the new -gene into the genome but incapable of directing the cell to produce in- fectious retrovirus. An early study involving ad- ministration of modified fibroblasts intraperitoneally to mice demonstrated that cells not normally pro- ducing AAT could be targeted by retroviral vectors to produce the protein (67-70). Unlike retrovirus, adenovirus DNA functions extrachromosomally (71) so that mutagenesis is not an issue, however, the therapy will have to be given periodically since the genetic modification of the target cell is not passed on to its progeny. No adenoviral-based strategies have been able to produce therapeutic levels of AAT (72). Liposomes are artificial lipid bilayers designed to transfer nucleic acids into the cell cytosol via cell- membrane fusion. A plasmid containing the AAT cDNA and a cytomegalovirus promoter complexed to cationic liposomes has been given intravenously and by aerosol to rabbits (73), showing AAT protein in the pulmonary endothelium. Molecular conjugates represent another non- viral system consisting of plasmid DNA complexed to polylysine and a molecule that targets the DNA complex to a specific cellular receptor. These vectors are not efficient and require further development (74). Current status of worldwide registries for AAT deficiency Three major studies are currently investigating the natural history and efficacy of intravenous augmen- WHO Bulletin OMS. Vol 75 1997408 a,-Antitrypsin deficiency tation therapy with human AAT in patients with severe AAT deficiency and pulmonary emphysema. The U.S. registry is a multicentre surveillance study coordinated by the National Heart, Lung, and Blood Institute of the National Insitutes of Health to assess the natural history of AAT deficiency in individuals receiving and not receiving intravenous AAT aug- mentation therapy (35). Change in FEVy and sur- vival are the end points. Between March 1989 and October 1992, a total of 1129 individuals were en- rolled. The Danish-Dutch study - the European Randomized Placebo-controlled Trial of alpha,- Protease Inhibitor Replacement Therapy (75) was designed as a double-blind, placebo-controlled study using the decline in FEV, as the primary end point; 57 patients were enrolled. The aim of a German multicentre surveillance study (the Westdeutsche Arbeitsgemeinschaft zur Therapie von Lungenerkrankungen (WATL) Multicentre Trial of a,-PI Augmentation Therapy in Patients with a,-PI Deficiency) was to evaluate the efficacy of intravenous AAT augmentation therapy using FEV, and survival as primary end points. This study includes 443 patients with severe AAT deficiency and pulmonary disease, all of whom receive weekly augmentation therapy with 60mg per kg human AAT; a placebo control group was not considered ethical. Because of the small number of known patients with AAT deficiency and the slow progression of pulmonary emphysema as well as other constraints, the planning and performance of each study has been accompanied by doubts about the protocol, the required number of patients, and the time needed to reach a statistically significant result. The three groups discussed above chose different strategies to solve these problems and results from these studies will soon be available. None, however, will answer all the important questions, but together, they should expand our knowledge of the natural history of AAT deficiency and provide the basis for devel- oping new national strategies for managing the disease. Screening for AAT deficiency Rationale for screening AAT deficiency is one of the most prevalent, poten- tially lethal hereditary diseases among Caucasians. Health care providers have often been reluctant to test for "uncommon" hereditary diseases such as AAT deficiency because of the perceived low yield of the test - which in any case depends on the population tested. Siblings of known AAT-deficient subjects provide the highest yield (25% positive), but is the group with the fewest available subjects. Up to 3% of patients with chronic obstructive lung disease have AAT deficiency. As discussed above, most indi- viduals with this disease worldwide are currently undiagnosed, and prolonged delays in making the correct diagnosis are common (54). Over the past few years, there has been renewed interest in testing for AAT deficiency because of the importance of identifying current smokers with the condition and persuading them to quit and of initiating other thera- pies that might limit the development of lung disease. Screening for AAT deficiency may be em- pirically divided into three types of activities: nondirected, or population, screening of adults; di- rected screening of adults who have a higher- than-average risk of having AAT deficiency or of suffering from AAT-deficiency-related lung disease ("case-finding"); and neonatal screening pro- grammes. Population screening of adults has been largely confined to blood donor populations (44) al- though useful for estimating the prevalence of undi- agnosed AAT deficiency in the community, the yield is sufficiently low that it is not cost-effective to screen large populations of adults. Since March 1991, the AAT Deficiency Detec- tion Center, in Salt Lake City, UT, USA, has offered testing to individuals with chronic bronchitis, em- physema, and asthma, and also to those with a family history of AAT deficiency. In the 5 years ending 29 February 1996, 16748 samples had been received for testing: 515 individuals were identified who had im- munoreactive AAT levels <11 itmol/l. Of these, one was phenotype PI*SZ and the remainder were phe- notype PI*Z. Thus, 3.1% of the total samples sub- mitted were from individuals with AAT deficiency, and it is estimated that the centre has diagnosed approximately 15% of the known individuals with AAT deficiency in the USA during its 5 years of operation. Neonatal screening programmes, most notably those undertaken in the 1970s in Sweden and in Or- egon, USA (200000 and 107038 neonates, resp.), enjoyed success in assessing the prevalence of AAT deficiency, and eventually will contribute to our knowledge of the natural history of the disease in adults as well as in early life (23,32). Recently, there has been a small pilot screening effort in New York State (76), as well as some renewed discussion among workers involved in neonatal screening (77). Despite the potential for rearing identified neonates by focusing on a healthy lifestyle, and thus markedly lessening the lung disease morbidity in future gen- erations (32), the advisability of neonatal screening for AAT deficiency has not been widely accepted. WHO Bulletin OMS. Vol 75 1997 409 Memorandum Conclusions and recommendations * Among Caucasian populations, AAT deficiency is as common a genetic disorder as cystic fibrosis. Available evidence indicates that in all countries with a high frequency of Caucasians, the number of diagnosed cases of AAT deficiency is only a small proportion of that predicted from estimates of gene frequency. The proportion of ascertained cases of AAT deficiency ranges from less than 10% in the USA and the United Kingdom to as high as 50% in smaller countries such as Denmark. Although there are insufficient amounts of human AAT available to provide augmentation therapy for all who might need it should all cases be ascertained, many alterna- tive preventive measures could be put in place. These include counselling carriers not to smoke or to stop smoking; provision of adequate immunization against influenza and pneumococcal infection; occu- pational counselling to minimize breathing polluted air; appropriate treatment of respiratory infection and atopic disease; and genetic counselling. It is therefore recommended that all patients with COPD and adults and adolescents with asthma be screened once for AAT deficiency using a quantitative test. Those with abnormal results on screening should undergo PI typing. * The 1995 ATS Standards of Care for COPD repre- sent an important advance but deal with AAT defi- ciency in a limited manner. For example, liver disease in childhood, genetic counselling, and the importance of smoking avoidance programmes and infection control programmes are not stressed. Standards that are specific for the care of persons with severe AAT deficiency should be drafted and will require involvement by gastroenterologists, pae- diatricians and geneticists and perhaps their na- tional societies. * Case finding among adults is too late to prevent COPD in many AAT-deficient persons, since the greatest risk factor for early-onset COPD is tobacco smoking. Because of the powerful addictive proper- ties of smoking, even with maximal support, a small proportion of smokers stop (22% sustained quitters at 5 years in the Lung Health Study). Two studies in which AAT deficiency was ascertained by neonatal screening found much lower frequencies of smoking at adolescence than expected. Furthermore, there are many other preventive measures that can be taken. The following recommendations can there- fore be made: neonatal AAT screening programmes should be undertaken in all developed countries with Caucasian populations; limited programmes of neonatal screening should be undertaken in devel- oping countries to determine the frequency of genes leading to deficiency and the burden of resulting disease in those populations; implementation of a neonatal screening programme requires involve- ment by gastroenterologists, geneticists, paediatricians (especially those involved in neonatal screening) clinical pathologists, ethicists, health care lawyers and the allied health professions, in addition to pulmonary specialists. Before such a screening programme is undertaken, laws or regulations must be in place to protect persons found to be severely AAT deficient from possible negative impacts on insurance coverage. Permission of families for screening must be obtained and strong support measures must be in place to assure appropriate counselling and support to persons found to have AAT deficiency and to their families. * An international a,-antitrypsin registry, with standardized requirements for case acquisition, would be difficult to establish because of differences in country-to-country practices and governmental funding of some existing registries; however, it should be possible to develop an international AAT coordinating registry (IAATCR). Such a registry would have as its function the development of stand- ardized forms for data entry and the development of a centralized database for carrying out appropriate research. Research coordination and dissemination of information to member-country data bases also seems feasible and appropriate. The IAATCR could be funded by a mixture of governmental, foundation, and industry funds. It would require a director, who could be part-time, and who would report to a rotat- ing board of directors made up of members from the national AAT registries. Also needed would be one or two full-time staff members for data manage- ment and computing equipment for data storage and analysis. * There is an urgent need for randomized clinical trials of the efficacy of AAT augmentation therapy in persons with the deficiency. Information from the NHLBI AAT Deficiency Registry may indicate need for fewer subjects in such studies than was hereto- fore considered necessary. Among the needs for such clinical trials are a placebo-controlled, outcome-driven trial of AAT augmentation therapy; determination of the need for adjusting AAT dosing during exacerbations of COPD; determination of the minimum effective replacement dose of AAT; and the most appropriate timing of the regimens. There are many questions as to the best end points to be measured and of the study design. All appropriate groups should be represented, including investiga- tors, representatives of national health care agencies, industry, patient groups and the managed health WHO Bulletin OMS. Vol 75 1997410 a1-Antitrypsin deficiency care industry. These studies could be carried out nationally or internationally, depending on logistics. An international working party on randomized clini- cal trials should be appointed to address these issues and to make specific recommendations. * Specific research needs for AAT deficiency are shown below. - Structural basis of deficiency: determine the precise molecular mechanisms involved in the misfolding and aggregation of Z a1-antitrypsin with a view to developing therapeutic strategies to inhibit aggregate formation and hence prevent liver disease and increase plasma levels of al- antitrypsin. - Risk factors and prognosis of lung disease: iden- tify risk factors for emphysema other than ciga- rette smoking (infection, atopy, familial factors); determine the role of environmental factors as risk factors for emphysema; and establish the natural history of lung disease in AAT deficiency. - Description of the pathology of the lungs in AAT deficiency: parenchyma; bronchi; bronchioles; and blood vessels. - Determination of the frequency, type, clinical and physiological manifestations of bronchiectasis in AAT deficiency. - Specific therapy of lung disease due to AA T defi- ciency: implement a randomized placebo-con- trolled clinical outcome trial of intravenous augmentation therapy; explore the role of mark- ers of elastin degradation or other indicators in evaluating the effects of augmentation therapy; assuming that augmentation therapy is proven to be effective, determine the preferred regimen by head-to-head comparative trials; determine the specific role of augmentation therapy in patients after lung transplantation; further evalu- ate alternative therapies, including aerosolized recombinant AAT and synthetic elastase inhibi- tors; foster continuing work on the safety and efficacy of gene transfer as a possible therapeutic option for AAT deficiency; and determine the relative roles of lung volume reduction surgery and lung transplantation for emphysema due to AAT deficiency. - Liver disease in AA T deficiency: foster research into the basic mechanisms of liver disease in AAT deficiency; determine the role of antioxi- dant therapy in preventing liver disease; deter- mine whether breast-feeding is protective against neonatal liver disease; determine the frequency and risk factors for development of cirrhosis and hepatocellular carcinoma in adults; determine the contribution of inflammation and hyperther- mia to the accelerated accumulation of AAT in the hepatocyte and hence the exacerbation of liver damage; determine the efficacy and advis- ability of vaccination against viral hepatitis B, and when a vaccine becomes available, against viral hepatitis C. Determination of gene frequency of AAT-defi- cient alleles in developing countries and the asso- ciated disease burden. Analysis ofthe costs and benefits ofscreening, as a prelude to implementing neonatal screening for AAT deficiency. Acknowledgements The participants at the meeting thank the following indi- viduals: Mrs S. Brandley, and Mr J. Brandley (representa- tives of the Alpha-1 National Association); Dr K. Allenby, Mrs D.P. Grange, Dr R. Gundel, Dr R.A. Lutes, Dr P. Fernandez, Dr R. Rousell (representatives of Bayer Cor- poration); Dr S. Kee; Dr J.J. Petillo and Dr R. Schwartz (representatives of Centeon). Resume Deficit en a1-antitrypsine: Memorandum d'une reunion de l'OMS Le deficit en a1-antitrypsine (AAT), connu 6gale- ment sous le nom de d6ficit en inhibiteur de l'ac- prot6ase, est une maladie imputable a un deficit g6n6tiquement d6termin6 en AAT. Le d6ficit r6sulte de la transmission de deux alleles codant pour le d6ficit en inhibiteur de protease (PI) au niveau du locus du gene AAT (designe par PI) localis6 sur le segment chromosomique 14q32.1. L'allble codant pour le deficit le plus connu est PlZ, et la majorit6 des sujets atteints de deficit severe en AAT ont un locus PI de type ZZ. La maladie survient essen- tiellement chez des blancs d'origine europ6enne, et sa fr6quence en Europe et en Amerique du Nord est comparable a celle de la mucoviscidose (1 pour 2000 a 1 pour 7000). Une personne atteinte peut ne pas avoir de manifestations cliniques. La pneumo- pathie obstructive chronique (POC) avec un em- physeme panacinaire de frequence 6levee, est le trait clinique le plus fr6quemment associ6 au deficit en AAT et la cause la plus fr6quente d'incapacite et de d6ces. Le tabagisme est le facteur de risque le plus important d'apparition d'une POC, laquelle survient avant la trentaine, c'est-a-dire beaucoup plus tot que la POC classique survenant en I'ab- sence de d6ficit. La maladie de foie, la deuxieme WHO Bulletin OMS. Vol 75 1997 411 Memorandum manifestation par ordre de fr6quence du deficit en AAT, se presente typiquement comme une chole- stase du nourrisson qui en g6n6ral n'est pas grave et r6gresse avant l'adolescence. 11 est rare qu'ap- paraisse une h6patopathie chronique, bien que le d6ficit en AAT soit la cause la plus fr6quente d'h6patopathie chronique de l'enfant. La cirrhose et le cancer du foie touchent au moins 25% des adultes de plus de 50 ans porteurs d'un d6ficit en AAT. Le deficit en AAT paralt largement sous diagnostiqu6, et compte tenu de la frequence attendue des genes, seule une petite partie des cas attendus de d6ficit en AAT on 6t6 diagnostiqu6s, meme dans les populations les mieux 6tudi6es. L'AAT humaine pour le traitement de suppl6men- tation n'est disponible qu'en petite quantite. Une r6union de l'OMS s'est tenue a Geneve, du 18 au 20 mars 1996, pour faire le point des connaissances sur cette affection gen6tique tres fr6quente, pour mettre au point une strat6gie afin de sensibiliser les prestateurs de soins et la popu- lation g6n6rale, et pour rechercher des strat6gies nouvelles de d6pistage et de prevention de la maladie. Les recommandations formul6es par les participants sont r6sumees cidessous. * Parmi les sujets de race blanche, le deficit en AAT est une anomalie g6n6tique aussi fr6quente que la mucoviscidose, mais le nombre de cas diagnostiqu6s ne represente qu'une faible partie des cas attendus d'apres la frequence du gene. Les mesures pr6ventives suivantes pourraient etre mises en place: conseiller aux porteurs de ne pas fumer ou d'arreter le tabagisme; prestations vaccinales adapt6es contre la grippe; conseil, pour diminuer la quantite d'air pollue respir6 en milieu professionnel; traitement approprie des infections respiratoires et de l'atopie; conseil gen6tique. * Elaboration d'un projet de traitement codifi6 sp6cifique des personnes atteintes de d6ficit s6vere en ATT. * Depistage n6onatal du d6ficit en AAT dans tous les pays d6veloppes ayant des populations blanches; programmes limites de depistage n6o- natal dans les pays en d6veloppement pour d6ter- miner la fr6quence des genes entrainant le deficit et la charge pathologique associee pour ces popu- lations; la mise en ceuvre d'un programme neonatal de d6pistage implique le travail de sp6cialistes appartenant a de tres nombreuses disciplines. Avant que ce type de programme soit applique, il est n6cessaire d'instituer des lois et une reglemen- tation pour prot6ger les personnes atteintes de d6ficit severe en AAT contre les effets n6gatifs possibles de ce d6pistage sur la couverture par les assurances. De plus, I'autorisation des familles devra etre obtenue pour pratiquer le depistage, et des mesures importantes de soutien doivent etre mises en place afin d'assurer un conseil approprie et I'aide aux personnes atteintes et a leur famille. * Mise en place d'un registre international de co- ordination pour l'AAT. * Cr6ation d'un groupe international de travail pour les essais cliniques randomis6s sur le d6ficit en AAT. * Des travaux de recherche s'imposent dans les domaines suivants: les bases structurales du d6ficit en AAT; -les facteurs de risque et le pronostic de la pneumopathie chez les personnes atteintes de d6ficit en AAT; la pathologie pulmonaire observ6e avec le deficit en AAT; la fr6quence, le type et les manifestations cliniques et physiologiques de la bronchec- tasie observ6es avec le d6ficit en AAT; le traitement appropri6 de la pneumopathie due au d6ficit en AAT; -I'h6patopathie du d6ficit en AAT; la fr6quence des alleles du deficit en AAT dans les pays en developpement et la charge pathologique associ6e; les coOts et les avantages du d6pistage, en pr6paration du depistage n6onatal du deficit en AAT. References 1. Laurell C-B, Eriksson S. The electrophoretic al- globulin pattern of serum in a1-antitrypsin deficiency. Scandinavian journal of clinical and laboratory investi- gation, 1963, 15: 132-140. 2. Eriksson S. Studies in a,-antitrypsin deficiency. Acta medica scandinavica, 1965, 432: 1-85. 3. Travis J, Salvesen GS. Human plasma proteinase inhibitors. Annual review of biochemistry, 1983, 52: 655-709. 4. Carrell RW, Boswell DR. Serpins: the superfamily of plasma serine proteinase inhibitors. In: Barrett A, Salvesen G, eds. Proteinase inhibitors. Amsterdam, Elsevier Biomedical Press, 1986: 403-419. 5. Potempa J, Korzus E, Travis J. The serpin family of proteinase inhibitors: structure, function and regula- tion. Journal of biological chemistry, 1994, 269: 15957-15960. 6. Huber R, Carrell RW. Implications of the three- dimensional structure of ca,-antitrypsin for structure 412 WHO Bulletin OMS. Vol 75 1997 a,-Antitrypsin deficiency and function of serpins. Biochemistry, 1989, 28: 8951-8966. 7. Engh RA et al. Divining the serpin inhibition mecha- nism: a suicide substrate "springe"? Trends in bio- technology, 1995, B: 503-510. 8. Carrell RW, Stein PE. The biostructural pathology of the serpins. Biological chemistry, Hoppe-Seyler, 1996, 377: 1-17. 9. Stein PE, Carrell RW. What do dysfunctional serpins tell us about molecular mobility and disease? Nature structural biology, 1995, 2: 96-114. 10. Carrell RW et al. Structure and variation of human al- antitrypsin. Nature, 1982, 298: 329-334. 11. Jeppson J-O. Amino acid substitution GluIELys in al- antitrypsin PiZ. FEBS letters, 1976, 65: 195-197. 12. Yoshida A et al. Molecular abnormality of human alpha,-antitrypsin variant (Pi-ZZ) associated with plasma activity deficiency. Proceedings of the Na- tional Academy of Sciences of the USA, 1976, 73: 1324-1328. 13. Graham A et al. Molecular characterisation of three alpha-i-antitrypsin deficiency variants: proteinase inhibitor (Pi) nullca,rdiff (Asp256/EVal): Pi Mmalton (Phe51lEdeletion) and Pi(Arg39IECys). Human genet- ics, 1989, 84: 55-58. 14. Frazier GC et al. In-frame single codon deletion in the Mmalton deficiency allele of a1-antitrypsin. American joumal of human genetics, 1989, 44: 894-902. 15. Seyama K et al. Siiyama (Serine53 (TCC) to phenyl- alanine 53 (TCC)). Journal of biological chemistry, 1991, 266: 12627-12632. 16. Yu M-H, Lee K-N, Kim J. The Z-type variant of human a1-antitrypsin causes a protein folding defect. Nature structural biology, 1995, 2: 363-367. 17. Le A, Graham KS, Sifers RN. Intracellular degrada- tion of the transport impaired human P1Z al- antitrypsin variant. Joumal of biological chemistry, 1990, 265: 14001-14007. 18. Lomas DA et al. The mechanism of Z a1-antitrypsin accumulation in the liver. Nature, 1992, 357: 605-607. 19. Lomas DA et al. a1-Antitrypsin Siiyama (Ser53EPhe): further evidence for intracellular loop-sheet polymeri- sation. Journal of biological chemistry, 1993, 268: 15333-15335. 20. Cox DW. a1-Antitrypsin deficiency. In: Scriver CR et al., eds. The metabolic and molecular bases of inher- ited disease, 7th ed. New York, McGraw-Hill, 1995: 4125-4158. 21. Lieberman J. Alpha1-antitrypsin deficiency and re- lated disorders. In: Emery AEH, Rimolin DL, eds. Prin- ciples and practice of medical genetics, vol. 2. New York, Churchill Livingstone, 1983: 911-935. 22. Kamboh Mi. Biochemical and genetic aspects of hu- man serum a1-proteinase inhibitor protein. Disease markers, 1985, 3: 135. 23. Sveger T. Liver disease in alpha,-antitrypsin defi- ciency detected by screening of 200000 infants. New England journal of medicine, 1976, 294: 1316-1321. 24. Cox DW, Billingsley GD. Rare deficiency types of a1-antitrypsin: electrophoretic variation and DNA haplotypes. American joumal of human genetics, 1989, 44: 844-854. 25. Cox DW, Billingsley GD, Smyth S. Rare types of a1- antitrypsin associated with deficiency. In: Allen RC, Arnaud P, eds. Proceedings of the Third Intemational Conference on Electrophoresis, 1981, New York. New York, de Gruyter, 1981: 505-510. 26. Cox DW, Billingsley GD, Mansfield T. DNA restric- tion site polymorphisms associated with the alpha,- antitrypsin gene. American journal ofhuman genetics, 1987, 41: 891-906. 27. Hodgson I, Kaisheker N. DNA polymorphisms of the human a, antitrypsin gene in normal subjects and in patients with pulmonary emphysema. Journal of medi- cal genetics, 1987, 24: 47-51. 28. Cox DW, Woo SLC, Mansfield T. DNA restriction fragments associated with alpha,-antitrypsin indicate a single origin for deficiency allele PIZ. Nature, 1985, 316: 79-81. 29. Brantly M, Nukiwa T, Crystal RG. Molecular basis of alpha-i-antitrypsin deficiency. American journal of medicine, 1988, 84 (suppl. 6A): 13-31. 30. Brantly ML et al. Clinical features and history of the destructive lung disease associated with alpha-i- antitrypsin deficiency of adults with pulmonary symp- toms. American review of respiratory disease, 1988, 138: 327-336. 31. Gishen P et al. Alpha-i-antitrypsin deficiency: the radiological features of pulmonary emphysema in subjects of PI-type Z and P1-type SZ: a survey by the British Thoracic Association. Clinical radiology, 1982, 33: 371-377. 32. Wall M et al. Long-term follow-up of a cohort of chil- dren with alpha-i-antitrypsin deficiency. Joumal of pediatrics, 1990, 116: 248-251. 33. Tobin MJ, Cook PJL, Hutchison DCS. Alpha1- antitrypsin deficiency: the clinical and physiological features of pulmonary emphysema in subjects ho- mozygous for Pi Type Z. British journal of disease of the chest, 1983, 77: 14-27. 34. Janus ED, Phillips NT, Carrell RW. Smoking lung function, and alpha-i-antitrypsin deficiency. Lancet, 1985, 1:152-154. 35. The Alpha-i-Antitrypsin Deficiency Registry Study Group. A registry of patients with severe defi- ciency of alpha-i-antitrypsin: design and methods. Chest, 1994, 106: 1223-1233. 36. Larsson C. Natural history and life expectancy in severe alpha,-antitrypsin deficiency, PiZ. Acta medica scandinavica, 1978, 204: 345-351. 37. Eriksson S. Studies in alpha-i-antitrypsin deficiency. Acta medica scandinavica, 1965, 177: 1-85. 38. Waller RE. Field investigations of air. In: Holland WW et al., eds. Oxford textbook of public health, vol. 2, 2nd edit. Oxford, Oxford University Press, 1991: 435- 450. 39. Abbey DE et al. Estimated long-term ambient con- centrations of PM10 and development of respiratory symptoms in a nonsmoking population. Archives of environmental health, 1995, 50: 139-152. 40. Schwartz J. Air pollution and hospital admissions for the elderly in Detroit, Michigan. American joumal of respiratory and critical care medicine, 1995, 150: 648-655. WHO Bulletin OMS. Vol 75 1997 413 Memorandum 41. Bakke PS et al. Prevalence of obstructive lung dis- ease in a general population: relation to occupational title and exposure to some airborne agents. Thorax, 1991, 46: 863-870. 42. Shaheen SO et al. The relationship between pneu- monia in early childhood and impaired lung function in late adult life. American journal of respiratory and critical care medicine, 1994, 149: 616-619. 43. Shaheen SO et al. Do lower respiratory tract infec- tions in early childhood cause chronic obstructive pul- monary disease? American journal of respiratory and critical care medicine, 1995, 151: 1649-1652. 44. Silverman EK et al. Alpha-i-antitrypsin deficiency. High prevalence in the St. Louis area determined by direct population screening. American review of respi- ratory disease, 1989, 140: 961-966. 45. Monso E et al. Bacterial infection in chronic obstruc- tive pulmonary disease. A study of stable and exacer- bated outpatients using the protected specimen brush. American journal of respiratory and critical care medicine, 1995, 152: 1316-1320. 46. Silverman EK et al. Variability of pulmonary function in alpha-i-antitrypsin deficiency: residual family re- semblance beyond the effect of the Pi locus. Human heredity, 1990, 40: 340-355. 47. Wright JL et al. Airway inflammation and peri- bronchiolar attachments in the lungs of nonsmokers, current and ex-smokers. Lung, 1988, 166: 277-286. 48. Snider GL. Emphysema: the first two centuries - and beyond; a historical overview, with suggestions for future research: part 2. American review of respira- tory disease, 1992, 146: 1615-1622. 49. Stockley RA et al. Elastase inhibitors of sputum sol phase: variability, relationship to neutrophil elastase inhibition, and effect of corticosteroid treatment. Thorax, 1986, 41: 442-447. 50. Shin MS, Ho K-J. Bronchiectasis in patients with alpha-i-antitrypsin deficiency: a rare occurrence? Chest, 1993, 104: 1384-1386. 51. Seersholm N, Kok-Jensen A. Survival in relation to lung function and smoking cessation in patients with severe hereditary alpha-i-antitrypsin deficiency. American joumal of respiratory and critical care medicine, 1995, 151: 369-373. 52. Seersholm N, Kok-Jensen A, Dirksen A. Survival of patients with severe al1-antitrypsin deficiency with spe- cial reference to non-index cases. Thorax, 1994, 49: 695-698. 53. Lieberman J, Winter B, Sastre A. Alpha-i- antitrypsin P1-types in 965 COPD patients. Chest, 1986, 89: 370-373. 54. Stoller JK et al. Physical and social impact of alpha- 1 -antitrypsin deficiency: results of a mail survey of the readership of a national newsletter. Cleveland Clinicjournal of medicine, 1995, 61: 461-467. 55. Sharp HL et al. Cirrhosis associated with al- antitrypsin deficiency: a previously unrecognized in- herited disorder. Joumal of laboratory and clinical medicine, 1969, 73: 934-939. 56. Berg NO, Eriksson S. Liver disease in adults with alpha,-antitrypsin deficiency. New England journal of medicine, 1972, 287: 1264-1267. 57. Cox DW, Smyth S. Risk for liver disease in adults with a1-antitrypsin deficiency. American journal of medicine, 1983, 74: 221-227. 58. Propst T et al. Prevalence of hepatocellular carci- noma in alpha-i-antitrypsin deficiency. Journal of hepatology, 1994, 21: 1006-101 1. 59. Eriksson S, Carlson J, Velez R. Risks of cirrhosis and primary liver cancer in alpha1-antitrypsin deficiency. New Englandjoumal of medicine, 1986, 314: 736-739. 60. Propst T et al. High prevalence of viral infection in adults with homozygous and heterozygous alpha1- antitrypsin deficiency and chronic liver disease. An- nals of internal medicine, 1992, 117: 641-645. 61. Edmonds BK, Hodge JA, Rietschel RL. Alpha-1 - antitrypsin deficiency-associated panniculitis: Case report and review of the literature. Pediatric dermatol- ogy, 1991, 8: 296-299. 62. McNeil TF, Sveger T, Theelin T. Psychosocial effects of screening for somatic risk: the Swedish al-antitrypsin experience. Thorax, 1988, 43: 505-507. 63. Hubbard RC et al. Biochemical efficacy and safety of monthly augmentation therapy for alpha-1 -antitrypsin deficiency. Journal of the American Medical Associa- tion, 1988, 260: 1259-1264. 64. Cammarata SK et al. Failure to achieve adequate serum levels with monthly replacement therapy in a1-antitrypsin deficiency. Chest, 1994, 106: 651-652. 65. Casolaro MA et al. Augmentation of lung anti- neutrophil elastase capacity with recombinant human al-antitrypsin: Journal of applied physiology, 1987, 63: 2015. 66. Seersholm N et al. Annual decline in FEV, in patients with severe al-PI deficiency: comparison of 161 un- treated Danish and 235 German patients receiving augmentation therapy. American journal of respiratory and critical care medicine, 1995, 151: A835. 67. Boris-Lawrie KA, Temin HM. Recent advances in retrovirus technology. Current opinion in genetic de- velopment, 1993, 301: 353-362. 68. Garver RI et al. Clonal gene therapy: transplanted mouse fibroblast clones express human al-antitrypsin gene in vivo. Science, 1987, 237: 762-764. 69. Kolodka TM, Finegold M, Woo SLC. Hepatic gene therapy: efficient retroviral-mediated gene transfer into rat hepatocytes in vivo. Somatic cell and molecu- lar genetics, 1993, 19: 491-497. 70. Curiel D, Stier L, Crystal RG. Gene therapy for a1- antitrypsin deficiency using lymphocytes as vehicles for al-antitrypsin delivery. Clinical research, 1989, 37: 578A. 71. Strauss SE. In: Ginsberg H, ed. The adenoviruses. New York, Plenum Press, 1984: 451-496. 72. Jaffe HA et al. Adenovirus-mediated in vivo gene transfer and expression in normal rat liver. Nature genetics, 1992, 1: 368-371. 73. Canonico AE et al. Aerosol and intravenous transfection of human al-antitrypsin gene to lungs of rabbits. American journal of respiratory cell and molecular biology, 1994, 10: 24-29. 74. Michael SI, Curiel DT. Strategies to achieve targetted gene delivery via the receptor-mediated endocytosis pathway. Gene therapy, 1994, 1: 223-232. 414 WHO Bulletin OMS. Vol 75 1997 ac-Antitrypsin deficiency 75. Jeppsson JO, Franzen B. Typing of genetic variants of a,c-antitrypsin by electrofocusing. Clinical chemistry, 1982, 28: 219-225. 76. Spence WC, Pass K, Murphy PD. A rapid fluorome- tric assay for newborn screening of a,-antitrypsin fol- lowed by phenotyping of deficient specimens by isoelectric focusing. Screening, 1994, 3: 23-31. 77. Niggemenn B, Paul K. a,-Antitrypsin deficiency: a need for screening? Screening, 1992, 1: 195-201. 78. Sveger T, Eriksson S. The liver adolescents with al- antitrypsin deficiency. Hepatology, 1995, 22: 514- 517. 79. Hutchison DCS et al. Longitudinal studies in alpha-1 - antitrypsin deficiency: a survey by the British Thoracic Society. In Taylor JC, Mittman C, eds. Pulmonary emphysema and proteolysis. New York, Academic Press, 1987: 7-15. 80. Wu MC, Eriksson S. Lung function, smoking, and survival in severe alpha-i-antitrypsin deficiency, PI*ZZ. Journal of clinical epidemiology, 1988, 41: 1157-1165. 81. Buist AS et al. The natural history of air-flow obstruc- tion in PiZ emphysema. American review of respira- tory disease, 1983, 127 (suppl): S43-S50. WHO Bulletin OMS. Vol 75 1997 415