Alpha-chain disease and related small-intestinal : a Memorandum *

Primary intestinal lymphotias are remarkably frequent in the Mediterranean region and South-West Asia. They are usually found in young persons from. the lower socio- economic strata of the population. These conditions sometimes present a premalignant phase characterized by plasmacytic infiltration of the small intestine. It has been reported that early treatment of cases with antibiotics is followed by complete remission, suggesting that some environmental factors may be responsible for the disease. Some patients have an abnormal alpha-chain protein in their serum. This Memorandum reviews the present knowledge of the clinical, immunological, epidemiological, and therapeutic aspects of this condition.

Primary intestinal are remarkably pre- lymphoma in man. Uniquely, the association of a valent in the Mediterranean region and South-West disordered immune response and the development Asia (1, 12, 26) and have a number of characteristic of immunoblastic sarcoma, possibly from the same features. They are most frequent in young patients clone of cells, can be studied in a cell population from underprivileged backgrounds who present with that synthesizes a distinctive marker protein. a malabsorption syndrome resulting from extensive Over the last decade data have accumulated from and diffuse infiltration of the wall of the small intes- isolated case studies in diverse regions and the time tine, predominantly by plasma cells. Clinical studies now seems opportune to initiate a new phase of suggest evolution from a premalignant cellular infil- research into this condition. As a first step, a co- trate to frank neoplasia involving more primitive ordinated international collaborative effort is essen- immunoblasts. tial to collate the data available and to establish The frequency of primary intestinal lymphomas standard nomenclatures, diagnostic criteria and among populations exposed to conditions of poor therapeutic responses so as to lay the groundwork hygiene (35) and evidence for complete remissions for systematic research into the etiopathogenesis and induced in early cases by oral antibiotic therapy (30), epidemiology of the disease. This Memorandum implicate environmental factors in the pathogenesis reviews the present knowledge about this disease. of this disorder. Since ingested microorganisms are Since in the early stage it does not appear to be a a powerful proliferative stimulus to the secretory truly malignant lymphoma, in this Memorandum IgA system (7), the premalignant phase of the dis- the condition is termed " immunoproliferative small- ease may represent an aberrant immune response by intestinal disease" (IPSID). the secretory IgA system to sustained, topical, anti- genic stimulation. The study of diffuse primary intestinal lymphomas therefore offers important op- CLINICAL FEATURES portunities for research into the pathogenesis of The major clinical features of IPSID are remark- ably uniform. They are (4, 8, 22, 24, 37): chronic * This Memorandum was drafted by the signatories listed diarrhoea, abdominal pain, abdominal distension, on pages 622-623. Reprints of the Memorandum and full de- nausea and vomiting, marked weight loss. tails of a proposed programme to study the disease (including protocols for the study of patients) can be obtained from Less common features include (17, 23, 24): oedema, , World Health Organization, Avenue Appia, tetany, low-grade fever (high temperatures are rare), 1211 Geneva 27, Switzerland. A French translation of this Memorandum will be published in a future issue of the melaena. Bulletini. An occasional presentation is that of an acute

3558 - 615 - BULL. WORLD HEALTH ORGAN., Vol. 54, 1976 616 MEMORANDUM abdomen due to obstruction, perforation, or intus- (i) limited bacteriological studies of stools have susception of the small intesine (1, 3, 10). not revealed any consistent pathogens: in several The diarrhoea may have features of steatorrhoea cases, bacterial overgrowth associated with bile-salt or may be very watery and at times so copious as to deconjugation in jejunal fluid and an antibiotic require urgent fluid and electrolyte replacement. The responsive 14C-glycocholate breath test has been severity of diarrhoea has on occasions led to hypo- observed (J. C. Rambaud, unpublished results). kalaemic nephropathy with polyuria. Radiological findings Population at risk Barium studies in IPSID patients frequently reveal Although IPSID is seen mainly in the second and an abnormal small-intestinal pattern, usually more third decade, cases have been noted in all age groups. marked in the upper small intestine (2, 4, 10, 26). There are no clear sex differences. The disease affects Findings include: (a) a characteristic coarse mucosal populations of diverse ethnic and geographical ori- pattern with a pseudopolypoid appearance, (b) stric- gins (34). There is a predilection for certain ethnic tures and segmentation, and (c) impression by groups in the same geographical region (26). The extrinsic nodes. patients described have mostly been from the low These findings are suggestive but are not diag- socioeconomic strata (33). nostic for this disease. Occasionally the appearances of osteoarthropathy are also noted (10). Absence of Presenting signs mediastinal adenopathy and radiological features of The common presenting signs include (8, 22, 24): osteomalacia should be emphasized. Abdominal emaciation, clubbing of fingers and toes, abdominal lymphoangiography may be abnormal but this inves- mass, abdominal tenderness, oedema. Occasionally tigation has limited value in the investigation of this tetany is found. Palpable peripheral nodes are a rare disease. presenting sign. It should be emphasized that hepatosplenomegaly Biopsy findings is not a presenting feature of the disease. Peroral mucosal biopsies at multiple sites have proved essential to the diagnosis and management Laboratory findings of the disease. Plasmacytic infiltration may occasion- Routine laboratory tests are not diagnostic. Abnor- ally be noted in the colon, rectum, stomach, naso- malities usually relate to malabsorption or loss of pharynx, bone marrow, or peripheral blood. substances through the . These include: Laparotomy Although laparotomy often reveals thickening of (a) malabsorption of fat, xylose, folate, and vita- the B12 with intrinsic factor. and vita- small-intestinal wall, mesenteric lymph node min Fat, xylose, enlargement, and a small spleen absence of are at least (19), min B12 malabsorption partially respon- these findings on gross examination does not exclude sive to antibiotics; the diagnosis. (b) hypocalcaemia, hypomagnesaemia and hypo- kalaemia; Course of the disease (c) hypocholesterolaemia; The course of the disease in the absence of ther- (d) mild anaemia, usually hypochromic-eosino- apy is progressive deterioration, usually culminating philia and lymphopenia are not usually found; in frank malignancy with consequent obstruction, intussusception, or perforation of the small intes- (e) hypoalbuminaemia and protein-losing enter- tine (1). The course may be interrupted by periods opathy; of improvement. Death is usually due to either (f) elevated alkaline phosphatase, often of intes- cachexia and infection or complications of localized tinal origin (8, 24, 27); tumour. Dissemination of the immunoblastic tumour outside the abdomen is a rare, late event of the (g) positive occult blood in stool may be found; disease (1, 3). (h) a variety of parasites have been associated The time lapse between initial symptoms and diag- with the disease, Giardia being the most frequent; nosis ranges from one month to six years. ALPHA-CHAIN DISEASE AND RELATED SMALL-INTESTINAL LYMPHOMA 617

HISTOPATHOLOGY OF IMMUNOPROLIFERATIVE that is evident in the gut is not clear, since it may SMALL-INTESTINAL DISEASE signify either a reaction pattern identical to that in ( MEDITERRANEAN TYPE ") (IPSID) the intestinal mucosa or actual dissemination of malignant cells to the mesenteric lymph nodes. The histological features of this disorder are A histopathological classification a of this disorder characteristic of the clinical syndrome of the so- is as follows: called Mediterranean lymphoma with malabsorp- (a) diffuse, dense, compact, and apparently benign tion, which is commonly associated with the presence lympho-proliferative mucosal infiltration, in the serum of alpha-heavy-chain disease (aHCD) (i) pure plasmacytic protein (15, 29). The aHCD protein may occasionally (ii) mixed lympho-plasmacytic; be found in immunoblastic intestinal lymphomas not apparently associated with diffuse lympho-plasma- (b) as in (a), plus circumscribed " immunoblastic" cytic infiltration of the lamina propria of the mucosa. sarcoma(s), in either the intestine and/or mesen- The diffuse plasma-cell infiltration at the time of teric lymph nodes; clinical presentation is usually massive, causing wide (c) diffuse " immunoblastic" or plasmacytic sar- separation of the crypts of Lieberkiihn and obliter- coma with or without demonstrable, apparently ation of the villous architecture, without, however, benign lympho-plasmacytic infiltration. significant impairment of the integrity of the surface epithelium (24). In some instances, obliteration of IMMUNOGLOBULIN ABNORMALITIES IN IPSID the villi is not complete, and focal areas in which the villous architecture is preserved but in which A characteristic immunoglobulin abnormality is the villous stroma is heavily infiltrated with plasma found in the majority of patients suffering from cells and/or lymphocytes may be evident (3). immunoproliferative small-intestinal disease (IPSID) Topographically, this form of the immunoprolifer- whose intestinal biopsies show the characteristic ative disease usually begins in the upper small intes- diffuse lympho-plasmacytic infiltration, with or with- tine (22), in contrast to the solitary lymphomas not out the presence of an immunoblastic lymphoma. associated with diffuse plasma-cell infiltration that This abnormality is very rarely found in intestinal have a predilection for the stomach, terminal ileum, lymphomas not associated with IPSID. and colorectal areas. The plasma-cell infiltration of The aberrant protein comprises a population of IPSID is diffuse and extensive, frequently involving molecules consisting of incomplete heavy a-chains, the entire length of the small intestine. devoid of light chains (33). The molecular weight of the monomeric polypeptide subunit varies be- Histologically, the diffuse plasma-cell infiltrates, The of these the tween 29 000 and 34 000 (11). length as well as mixed lympho-plasmacytic infiltrates, is thus than half but less than three- are usually confined to the lamina propria of the chains greater quarters of normal a, heavy chains. Antigenic analy- mucosa and may lack conclusive cytological features that the of the sis (34) and chemical studies (35) indicate neoplasia. Occasionally, they penetrate deeply Fc is present in aHCD proteins, into the submucosa and show cellular con- entire fragment atypia their C-terminus is identical to that of normal with sarcoma. that sistent plasmacytic a, chains and that the heavy-light peptide is missing. The morphology of the diffuse lympho-plasmacytic The hinge region has been shown by chemical meth- infiltrations is in keeping with the view that a pre- ods to be present in all eight proteins so far studied. malignant stage of the disease may exist, a view that All attempts to raise specific to individ- appears to be supported by regression of the clinico- uals aHCD proteins have failed, indicating the histological and immunological abnormalities follow- paucity of antigenic determinants in the N-terminal ing oral antibiotic therapy (30). In some cases, aggregates of lymphocytes contain- a Explanatory notes: (a) (i) & (ii): atypical plasma cells ing large, scattered, atypical " immunoblastic" cells or large lymphoid cells (" immunoblasts ") may be scattered within the lympho-plasmacytic infiltration. (b) & (c): Cells are present within the diffuse plasmacytic infiltration. resembling Sternberg-Reed cells may be part of the tumour The significance of these cells is not clear but they infiltrate. A diffuse follicular lymphoid hyperplasia involving the have usually been associated with, or may represent, intestinal mucosa, with or without a malignant lymphoma neoplastic transformation. of the poorly differentiated lymphocytic type, may at times be encountered in patients with the clinical syndrome of The significance of mesenteric lymph node involve- Mediterranean lymphoma. This type, however, is not usually ment by the same lympho-plasmacytic proliferation associated with aHCD. 618 MEMORANDUM portion of the fragment. In view of these results in position 12. Protein Def is therefore possibly syn- and of the molecular weight data, the missing por- thesized as an intracellularly deleted a1 heavy chain. tion of the chain is located in the Fd segment and Similar conclusions were reached from the study of involves both the VH and C1 regions. another aHCD protein (41). It is possible, however, The N-terminal sequences of several aHCD pro- that the N-terminal sequence is not a portion of the teins are heterogeneous (35). Even for those proteins V region but rather a non-cleaved polypeptide. It is with a single N-terminal amino acid, marked heter- of interest that valine at position 9 of the hinge ogeneity became apparent after two steps in degra- peptide, where the identity with a normal a1 chain dation. Attempts to obtain the N-terminal sequence starts, could be the equivalent of glutamine at posi- on an automated sequencer were unsuccessful. The tion 216 of y-chains, the site where normal synthesis N-terminal residues are different from those found resumes in several yHCD proteins with internal in any of the subgroups of the variable regions of deletions (14). normal heavy chains. The most likely explanation Thus the primary defect in aHCD proteins appears of this heterogeneity is that it is the consequence of to be a deletion affecting the variable and first con- intracellular proteolysis occurring after synthesis. stant regions of the heavy chains, which are under The fact that the N-terminal residues found in the independent control. Any genetic hypothesis about seven proteins studied were valine and/or isoleucine aHCD proteins should also take into account the suggests that the postulated degradation may stop fact that immunofluorescent and radioimmuno- at this level for some reason, which could be enzyme electrophoretic studies of proteins synthesized in specificity, steric hindrance, or the presence of a vitro have failed to detect any light-chain production carbohydrate moiety. An analogy for intracellular in the cells which secrete aHCD proteins (34). This proteolysis of an incomplete protein has been found failure of light-chain synthesis has been confirmed in alkaline phosphatase of the amber and ochre by biosynthesis studies of nascent Ig subunits in mutants of Escherichia coli (39). Alternatively, post- such patients (5). Since light and heavy chains are synthetic cleavage may occur extracellularly. In one under the control of unlinked genes, this peculiar case of gamma heavy chain disease (yHCD) (where situation raises a puzzling problem for the cellular similar N-terminal heterogeneity may be found), the geneticist. The possibility remains that the light chain nascent abnormal chain in the cytoplasma and in is transcribed but not translated. Studies looking for the culture fluid during biosynthetic studies had a the presence or absence of its messenger are war- molecular weight of 36 000; by contrast the molecu- ranted. lar weight of the serum yHCD protein was 28 000 (5). All 50 allCD proteins examined to date belong The demonstration of a large internal deletion in to subclass 1 (36), an unexpected and unexplained a yHCD protein (13) led to the postulate that in finding since 30-40% of normal secretory IgA pro- aHCD there was a similar primary deletion followed tein is subclass 2 (16). This finding, together with and obscured by a secondary, limited intracellular amino acid substitution in position 12 of Def, sug- proteolysis. This hypothesis has been supported to gests that a HCD proteins may be monoclonal but some extent by biosynthetic and structural studies. this cannot be confirmed because the other essential Cellular biosynthetic studies excluded the possibility criteria for monoclonality, i.e. light chain and homo- of the synthesis of a normal a-chain with subsequent geneity of V region leading to shared idiotypic degradation to a smaller fragment after its release specifity and activity, cannot be tested. from the ribosomes (5). Comparison of the amino Several aHCD proteins have been found to con- acid sequence of the hinge region of aHCD protein tain a J chain. The secretory piece binds to aHCD from patient Def with that of a normal IgAl showed protein but the nature of the binding is unknown. that, after a short segment, thought to correspond Whether aHCD proteins should be considered as to the variable region, protein Def displays a gap "abnormal" is still an open question. Polypeptides which comprises almost the whole Fd segment includ- analogous to the proteins of yHCD were recently ing the CH, domain (40). Normal synthesis resumes reported to be present in very small amounts in at a valine residue in the hinge region just preceding normal plasma (18). However, if heavy-chain dis- a segment that contains a partially-duplicated frag- ease arises from proliferation of a clone of cells ment and the interheavy chain disulfide bonds. From producing such polypeptides (which could be prone there on the molecule is apparently normal with the to neoplastic transformation) it is necessary to pos- exception of a substitution of threonine for serine tulate a wide variety of cells carrying such deletions ALPHA-CHAIN DISEASE AND RELATED SMALL-INTESTINAL LYMPHOMA 619 in normal individuals since the site and length of acrylamide gel electrophoresis or by gel filtration the deletion appears to vary from one aHCD protein after dissociating the molecule. to another. It should be emphasized that deletions The striking and unexpected electrophoretic heter- are not confined to HCD proteins and that various ogeneity of these presumably monoclonal aHCD other types of heavy- and/or light-chain deletions proteins is certainly due in part to the heterogeneity have now been described in human (as well as of their N-terminal sequences as discussed above. murine) myeloma proteins. It may also be related to two other features, the The diagnosis of aHCD relies entirely upon lab- high carbohydrate content of most aHCD pro- oratory studies including immunochemical analysis teins (35) and their propensity for polymerization. of the serum proteins and may be difficult in a Indeed, on ultracentrifugation aHCD proteins appear routine laboratory. It can easily be missed on serum to consist of dimers with a 34S sedimentation con- protein electrophoresis, and electrophoresis did not stant and, in most instances, of larger polymers of demontrate the pathological protein in half of the various sizes (34). 80 sera studied in Paris. When detectable by electro- The serum levels of normal IgA, IgG, and IgM phoresis, the aHCD protein shows as an abnormal are usually depressed. These decreases are not solely broad band usually in the a2 or ,B region. The charac- due to a protein-losing enteropathy, as shown by teristic narrow band, suggestive of a monoclonal Ig the disproportionate depression of serum immuno- abnormality, is absent. In most of the cases where globulin levels relative to the serum albumin con- the pathological protein was not detectable, serum centration. electrophoresis showed only a decrease in serum albu- The diagnosis of aHCD is made more difficult by min and a moderate to severe hypogammaglobu- the very low concentration of pathological protein in linaemia. the urine. In most patients, however, it could be The diagnosis is usually suspected or established detected in concentrated urine and had the same by immunoelectrophoretic analysis of the serum of electrophoretic and immunochemical characteristics these patients. In many cases the protein abnormal- as that found in the serum. Bence-Jones proteinuria ity escaped detection by routine immunoelectro- was never found. The pathological protein was also phoresis using polyvalent antiserum to human nor- found in significant amounts in jejunal fluid, as ex- mal serum; analysis with monospecific antisera for pected from the involvement of the intestine, whereas IgA is therefore essential. the IgA in the parotid saliva of these patients was aHCD protein usually gives an abnormal pre- normal (34). cipitin line either extending from the a1 globulins to In order to establish the true incidence of aHCD the slow f2 region or showing a faster electro- in IPSID patients, it will be necessary to study sys- phoretic mobility than normal IgA. However, in a tematically the immunoglobulin content of the intes- few patients the aHCD protein had a slow electro- tinal fluid and to perform biosynthesis studies of phoretic mobility. The anomalous component does intestinal biopsies on those patients without detect- not of course precipitate with antisera to light chains. able aHCD protein in the serum. However, it should be emphasized that this lack of Interesting exceptions to the usually characteristic precipitation with anti-K and anti-A antisera is not protein findings have been reported. A small homo- a sufficient criterion for the diagnosis of aHCD since geneous IgG component has been found in addition many IgA myeloma proteins, even though they con- to aHCD protein in serum in two patients. In sera tain light chains (mainly A chains) fail to precipitate from these patients who presented with the typical with such antisera. Selected antisera to IgA which clinico-pathological features of the intestinal form contain antibodies related to the conformational of aHCD an entire IgA myeloma globulin was found specificity of the Fab region and which precipitate and Bence-Jones protein was present in the urine only with a- and light-chains combined, were found in one of these patients. yHCD protein was demon- to be very useful for the diagnosis of aHCD by strated in the serum and in an intestinal biopsy from immunoelectrophoresis or the Ouchterlony tech- another patient with the typical clinico-pathological nique (34). Alternatively, the immunoselection plate pattern of IPSID. The aHCD protein present in the method of Radl can be used (8). In all doubtful serum of three patients apparently suffering from a cases the pathological protein should be purified, respiratory form of aHCD was indistinguishable reduced, and alkylated, and the lack of light chains from that described above for the much commoner should be demonstrated directly by starch or poly- intestinal form (38). 620 MEMORANDUM

OTHER IMMUNOLOGICAL STUDIES found in association with IPSID and, in many instances, no evidence of parasitic infestation was Although IPSID may be considered the conse- present at the time of diagnosis. These findings do quence of a basic underlying immunodeficiency, not necessarily militate against a causal role for bac- tests of immune competence in patients have received teria or parasites, specific or nonspecific. Infection scant attention. No studies of immunoglobulin levels or infestation may be present only during the early in jejunal fluid have been reported and no functional phase ofaHCD, especially during early infancy, when tests of the secretory immune system have been per- the intestinal immune system is immature, and may formed in these patients. not be manifest in identifiable form years later at Leucopenia is not a feature of IPSID and limited the time of diagnosis. Unfortunately, the absence of testing of T-cell function using mitogenic responses Fab in aHCD protein (33) precludes its use for to phytohaemagglutinin and skin testing has not identifying putative antigenic stimuli. revealed evidence of a gross defect in cell-mediated immunity. No antigen-specific tests of T-cell function have been made. EPIDEMIOLOGICAL AND FAMILY STUDIES No studies of tumour-associated antigens have Epidemiological knowledge of this disorder is been made in IPSID patients. limited. What little is known comes from case infor- mation accumulated in a small number of clinical CELL CULTURE-VIROLOGY centres, notably in Algeria, France, Iran, Iraq, Israel, Lebanon, South Africa, Tunisia, and the United Attempts to establish a continuous line of aHCD- Kingdom. secreting cells in culture have failed. However, cells from the lympho-plasmacytic infiltrate can be cloned General incidence in short-term culture and usually remain viable for While certain regional patterns are suggested by several weeks. the available clinical data, the general incidence of Limited serological studies for antibodies against the disease is not clearly defined, either for the world Epstein Barr virus, cytomegalovirus and herpes as a whole or for any given country. Limited surveys simplex virus have not provided any significant by small-intestinal biopsy in the general population findings. of South-West Asia leave unresolved the suggestion that a large reservoir of subclinical disease may BACTERIOLOGY AND PARASITOLOGY exist.

In a survey of 60 case reports, bacteriological data Age from stool culture is available on only 5 patients. The disorder appears to affect primarily young The only pathogen cultured was a Salmonella in one adults 20-40 years of age, although occasional cases patient. In several patients, overgrowth of aerobes have been described in younger and older persons. and anaerobes has been demonstrated in jejunal This young age distribution contrasts sharply with fluid. In most of these patients, steatorrhoea, vita- the older pattern seen in other forms of small-intes- min B12 absorption (Schilling test) and the 14C- tinal lymphoma (the so-called " western type"). glycocholate breath test improved markedly follow- ing a short course of tetracycline given orally Sex (J. C. Rambaud, unpublished results). These findings Both sexes seem to be about equally affected. Again suggest that a stagnant loop syndrome may con- this stands in contrast to the usual picture of lympho- tribute to the malabsorption in these patients. In matous disease, in which male cases predominate. two cases, however, no evidence for a stagnant loop syndrome was observed (6, 25). Racial or ethnic incidence Parasitic infestation was found in 21 patients of Although at first the disease seemed primarily to the 60 studied (34%) (31). Giardiasis was the com- affect Arab populations of the eastern and southern monest infestation found (31) but many other para- Mediterranean regions (hence the term " Mediter- sites have been reported including Entamoeba histoly- ranean lymphoma "), there now appears to be a tica, Trichomonas hominias, Trichuris trichuria, hook- wide spectrum of racial or ethnic incidence, reflected worm, coccidia, and Ascaris lumbricoides. Thus, no in an emerging pattern of wide geographical distri- particular bacterial pathogen or parasite has been bution. Notably this includes non-Ashkenazi Jews ALPHA-CHAIN DISEASE AND RELATED SMALL-INTESTINAL LYMPHOMA 621 in Israel, but not Jews of European or American tive case-control or prospective cohort approach) origin (26). In South Africa, Cape Coloureds and have yet been conducted to assess the wide range of mulattoes are affected, although apparently not the etiological factors that may contribute to this pos- more numerous Bantu population in that coun- sibly multifactorial disease. What information exists try (23). comes from limited and anecdotal sources: incon- clusive data regarding dietary habits in a few Leba- Socioeconomic status nese cases and scattered bacterial cultures of intestinal A striking and generally consistent feature of the flora. There has not been any definitive work con- disease is its occurrence in low socioeconomic strata cerning relationships with potentially oncogenic of the population. Occasional cases, however, come viruses, particularly the Epstein-Barr herpesvirus. from higher social levels. Geographical distribution THERAPY While most cases are still recorded in South-West Asia and northern Africa, cases have now been de- Reports of intensively studied IPSID patients are scribed in many parts of the world including South still uncommon and mainly concern those published Africa, Europe (especially along the Mediterranean as cases of aHCD. Although the natural history of coast), the Indian subcontinent, and South and this disorder is not fully known, provisional thera- Central America (35). While the disease appears to peutic guidelines may be obtained from the data be extremely rare in Europe and North America, available on two main groups of patients published the report of an atypical intestinal case from Finland in the literature. suggests that no region of the world is exempt (32). Firstly, those patients whose intestinal mucosa No tendencies for cases to be clustered discretely showed a dense predominantly plasma-cell infil- within particular regions of particular countries have tration confined to the lamina propria; 16 patient been described. are classified in this group. Seven of these patients were treated with oral anti- Secular trends biotics-mainly tetracycline and ampicillin-alone. Four achieved complete clinical, histological, and It is not known whether the disease is increasing immunological remission following 2-10 months of or declining in incidence in particular populations, treatment (disappearance of aHCD protein from except that, in Israel, the disease appears to be serum and intestinal juice, and normal plasma cells declining in frequency. in the lamina propria of the gut as assessed by Familial incidence immunofluorescence studies) (21, 24, 30). In two patients this complete remission lasted 6 and While only limited family studies have been made, 2 years, respectively, after the cessation of all there appears to be no strong predilection for cases treatment. The clinical condition of a fifth patient to cluster in families. Only one instance of disease is much improved after 6 months' treatment, in primary relatives has been described, two sibs but the intestinal histology is unchanged and in southern Iran. No obvious HLA patterns were aHCD protein is still present in serum. In a suggested in a small series of patients from Iran; sixth patient 2 years of continuous oral tetra- surveys of la antigens have not been carried out. cycline therapy was required to achieve remission No clear immunoglobulin abnormalities have been (disappearance of malabsorption and of aHCD found in a limited number of sera from close rela- protein from serum, and radiological changes in tives of patients surveyed in Israel (28). The obser- the small intestine) (B. Ramot, unpublished results). vation that family members of an IPSID propositus A seventh patient died from intercurrent infection have the intestinal isoenzyme of alkaline phospha- during therapy for IPSID. tase in serum (27)-a feature of IPSID (see page 616) Six other patients of this group were treated by warrants further study. chemotherapy alone-melphalan, cyclophosphamide, chlorambucil, vincristine, and prednisone in various Potential etiological factors combinations, sequences, and dosages. Only one No formal epidemiological or joint epidemiologi- patient in this group achieved complete clinical, cal/laboratory studies (whether using the retrospec- histological, and immunological remission after an 622 MEMORANDUM

8-month course of chlorambucil, and this remission intestine after treatment was unchanged in one case is still maintained 6 years after ceasing therapy. and was unavailable in the second case. A second patient was almost completely asympto- In 3 patients the immunoblastic cells were scat- matic, with no detectable aHCD protein in serum, tered within the small intestine and lymph node 3 years after beginning intermittent treatment with plasmacytic proliferation (3). Transient clinical im- cyclophosphamide and prednisone, preceded by an provement was obtained in these 3 patients with ineffective course of vincristine. A third patient various types of chemotherapy. Complete clinical showed no improvement after 6 months of similar and histological remission was obtained in one case treatment. Three patients had only a temporary after two courses of the MOPP a regimen but there- improvement (with complete clinical remission in after clinical relapse was observed. one case) and died after the development of an immunoblastic sarcoma. * * Three patients received combined therapy with oral antibiotics and ch;emotherapy. One of these F. Ala, Iranian National Blood Transfusion Service, achieved complete clinical, histological, and immuno- Teheran, Iran. logical remission after 15 months of treatment with W. F. Doe, Department of Immunopathology, Scripps tetracycline, oleandomycin, and cyclophosphamide. Clinic and Research Foundation, La Jolla, CA, USA (Rapporteur). This remission was maintained 12 months after all B. S. Drasar, Bacterial Metabolism Research Laboratory, treatment ended. Subsequently, an ileal immuno- Public Health Laboratory Service, London, England. blastoma developed in this patient (20). A second Present address: Department of Microbiology, London patient was clinically asymptomatic after 6 months School of Hygiene and Tropical Medicine, Keppel of treatment with tetracycline and melphalan (histo- Street, London, England. logical and immunological status unknown) (8). One W. Dutz, Medical College of Virginia, Virginia Common- patient experienced complete clinical remission with wealth University, Richmond, VA, USA. incomplete histological and immunological improve- F. S. Farah, WHO Immunology Research and Training ment during a one-year treatment, but relapsed Centre, American University ofBeirut, Beirut, Lebanon. N. Goldblum, The Chanock Centre for Virology, (after stopping therapy) and eventually developed Hadassah Medical School, Hebrew University, Jeru- an immunoblastic sarcoma (9). salem, Israel. The delay between the first symptoms and the A. I. Goussev, Immunology, World Health Organization, beginning of treatment, and the initial clinical and Geneva, Switzerland. histological status appeared to bear no relationship L. Haghighi, Department of Microbiology, Pahlavi Uni- to whether or not the patients responded to therapy. versity Medical School, Shiraz, Iran. However, it should be emphasized that the histology P. Haghighi, Department of Pathology, Pahlavi Univer- of the mesenteric lymph nodes was known in only sity Medical School, Shiraz, Iran. C. W. Heath, Center for Disease Control, Atlanta, GA, 5 cases, and that information on the small-intestinal USA. pathology was incomplete, in terms of length and A. Kharazmi, Department of Microbiology, Pahlavi depth of involvement, in all patients except one. University Medical School, Shiraz, Iran. Among the 6 patients who achieved complete remis- J.-P. Lamelin, International Agency for Research on sion, 5 had initial low concentrations of aHCD pro- Cancer, Lyon, France. tein in serum, compared with 2 of 5 patients who Ramona Lunt, Cancer, World Health Organization, did not respond to treatment. Geneva, Switzerland. The second group provides the results of therapy Lindsay J. Mackey, WHO Immunology Research and on 6 from Training Centre, Hopital Cantonal, Geneva, Switzer- IPSID patients suffering immunoblastoma. land. Complete sustained remission has apparently been F. Modabber, School of Public Health and Institute of achieved in 2 cases using deep X-ray therapy. Two Public Health Research, University of Teheran, patients with mesenteric lymph node " histiocytic Teheran, Iran. sarcoma" coexisting with a pure plasmacytic infil- A. Modjtabai, Taj Pahlavi Cancer Institute, Teheran, tration of the gut, were treated with total abdominal Iran. irradiation of 30 gy (3000 rad) and sterioids (23). H. Moroz, Beilinson Medical Centre, Tel Aviv, Israel. They achieved complete clinical and immunological (serum only) remission, lasting 2 and 1 years, a nitrogen mustard, vincristine (oncovin), procarbazine, respectively. However, the histology of the small and prednisone. ALPHA-CHAIN DISEASE AND RELATED SMALL-INTESTINAL LYMPHOMA 623

V. Nassar, Department of Pathology, Emory University, P. Salem, American University of Beirut Hospital, Atlanta, GA, USA. Beirut, Lebanon. Present address: Department of K. Nasr, Pahlavi University Medical School, Shiraz, Iran. Developmental Therapeutics, The University of Texas G. T. O'Conor, International Affairs, National Cancer System, Cancer Center, Texas Medical Center, Hous- Institute, Bethesda, MD, USA. ton, TX, USA. M. Potter, Laboratory of Cell Biology, National Cancer M. Seligmann, Immunochemistry Laboratory, Institut Institute, Bethesda, MD, USA. de Recherches sur les Maladies du Sang, Hopital J.-L. Preud'homme, Institut de Recherches sur les Mala- Saint-Louis, Paris, France (Chairman). dies du Sang, Hopital Saint-Louis, Paris, France. L. H. Sobin, Cancer, World Health Organization, Geneva, N. Racoveanu, Cancer, World Health Organization, Switzerland. Regional Office for the Eastern Mediterranean, Alexan- J. N. Soloviev, Institute of Experimental Oncology, dria, Egypt. Moscow, USSR. J.-C. Rambaud, Department of Gastroenterology, H6pi- G. B. de The, International Agency for Research on tal Saint-Lazare, Paris, France. Cancer, Lyon, France. B. Ramot, Institute of Haematology, Chaim Sheba G. Torrigiani, Immunology, World Health Organization, Medical Centre, Tel-Hashomer, Israel. Geneva, Switzerland. H. Rappaport, City of Hope National Medical Center, T. Waldmann, Departement of Immunology, National Duarte, CA, USA. Cancer Institute, Bethesda, MD, USA.

ACKNOWLEDGEMENTS

The advice of Professor A. Capron, Institut Pasteur, Lille, France; Dr J. F. Fraumeni, National Cancer Institute, Bethesda, MD, USA; Dr G. Klein, Karolinska Institute, Stockholm, Sweden; Dr P. H. Levine, National Cancer Institute, Bethesda, MD, USA; and Dr A. S. Rabson, National Cancer Institute, Bethesda, MD, USA, is gratefully acknowledged. This Memorandum was drafted following a meeting that was mainly financed from a grant from the National Cancer Institute, National Institutes of Health, Department of Health, Education and Welfare, USA.

REFERENCES

1. AL-BAHRANI, Z. R. & BAKIR, F. Annals of the Royal 14. FRANGIONE, B. & FRANKLIN, F. C. Seminars in College of Surgeons of England, 48, 103 (1971). , 10: 53 (1973). 2. BALIKIAN, J. ET AL. Americanjournalofroentgenology, 15. GALLN, A. ET AL. Cancer (Philad.) (in press). radium therapy and nuclear medicine, 107: 131 (1969). 16. GREY, H. M. ET AL. Journal ofexperimental medicine, 3. BOGNEL, J. C. ET AL. Revue europekenne d'e'tudes cli- 128: 1223 (1968). niques et biologiques, 17: 362 (1972). 17. IRUNBERRY, J. ET AL. Nouvelle revuefranfaise d'hema- 4. BONOMO, L. ET AL American journal of medicine, tologie, 10: 609 (1970). 52: 73 (1972). 18. LAM, C. W. K. & STEVENSON, G. T. Nature (London), 5. BUXBAUM, J. N. & PREUD'HOMME, J. L. Journal of 246: 419 (1973). immunology, 109: 1131 (1972). 19. LARocHE, C. ET AL. Presse medicale, 78: 55 (1970). 6. CHERNOV, A. J. ET AL. Gut, 13: 103 (1972). 20. MANOUSOS, 0. N. ET AL. British medical journal, 2: 7. CRABBE, P. A. ET AL. Laboratory investigation, 22: 409 (1974). 448 (1970). 21. MONGES, H. ET AL. Archives franraises des maladies 8. DOE, W. F. ET AL. Gut, 13: 947 (1972). de l'appareil digestif (in press). 9. DOE, W. F. British journal of cancer, 31 (Suppl. 11): 22. NASR, K. ET AL. Gut, 11: 673 (1970). 350 (1975). 23. Novis, B. H. ET AL. American journal of digestive 10. DOE, W. F. ET AL. British journal ofradiology, 49: 3 diseases, 18: 679 (1973). (1976). 24. RAMBAUD, J. C. ET AL. Digestion, 1: 321 (1968). 11. DORRINGTON, K. J. ET AL. Biochimica et biophysica 25. RAMBAUD, J. C. ET AL. Annales de medecine interne, acta, 221: 647 (1970). 121: 135 (1970). 12. EIDELMAN, S. ET AL. Medicine, 45: 111 (1966). 26. RAMOT, B. ET AL. Israel journal of medical sciences, 13. FRANGIONE, B. & MILSTEIN, C. Nature (London), 1: 221 (1965). 224: 597 (1969). 27. RAMOT, B. & STREIFLER, C. Lanzcet, 2: 587 (1966). 624 MEMORANDUM

28. RAMOT, B. & HULU, N. British journal of cancer, 36. SELIGMANN, M. Archives of internal nmedicine, 135: 31 (SuppI. 11): 343 (1975). 78 (1975). 29. RAPPAPORT, H. ET AL. Cancer (Philad.), 29: 1502 37. SHAHID, M. J. ET AL. Cancer (Philad.), 35: 848 (1972). (1975). 30. ROGE, J. ET AL. Pathologie et biologie, 18: 851 (1970). 38. STOOP, J. W. ET AL Clinical and experimental immu- 31. SALEM, P. ET AL. Cancer (Philad.) (in press). mology, 9: 625 (1971). 32. SAVILAHTI, E. & BRANDTZAEG, P. Scandinavian jour- 39. SUZUKI, T. & GAREN, A. Journal ofmolecular biology, nal of immunology, 2: 322 (1973). 45: 549 (1969). 33. SELIGMANN, M. ET AL. Science, 162: 1396 (1968). 40. WOLFENSTEIN-TODEL, C. ET AL. Proceedings of the 34. SELIGMANN, M. ET AL. Journal of clinical investi- National Academy of Sciences, 71: 974 (1974). gation, 48: 2374 (1969). 41. WOLFENSTEIN-TODEL, C. ET AL. Biochemical and 35. SELIGMANN, M. ET AL. Annals ofthe New York Acad- biophysical research communications, 65: 47 emy of Sciences, 190: 487 (1971). (1975).