CHAPTER 109 HEAVY-CHAIN DISEASES
CHAPTER 109 HEAVY-CHAIN DISEASES
JOEL N. BUXBAUM
Definition and History
Etiology and Pathogenesis
Chronic Antigenic Stimulation
Defective Immunoglobulin Molecules
Delta- and Epsilon-Heavy-Chain Disease
Detection of Abnormal Immunoglobulin Protein
Therapy, Course, and Prognosis
The heavy-chain diseases are rare B-cell proliferative disorders of varying degrees of malignancy. Their characteristic feature is the production of a monoclonal immunoglobulin molecule in which the heavy chain is truncated and there is no covalent attachment of light chains, either because of absent L-chain production or failure of formation of H-L disulfide bonds. The analysis of the domain structure of the heavy-chain disease proteins provided one of the early clues to the exon-intron structure of immunoglobulin genes. Immunoelectrophoresis or immunofixation of serum, urine, or secretory fluids or immunohistologic analysis of the proliferating cells establishes the diagnosis. The diseases behave clinically as B-cell lymphomas with the class of their monoclonal protein generally reflective of the major site of involvement. Gamma heavy-chain disease (g-heavy-chain disease) behaves as a systemic lymphoma, alpha heavy-chain disease (a-heavy-chain disease) as a predominantly gut-associated lymphoproliferative disorder, and mu heavy-chain disease (µ-heavy-chain disease) as a systemic lymphoma with some patients presenting features of chronic lymphocytic leukemia. Prognosis is variable and definitive effective treatment regimens have not yet been established, except perhaps in the case of a-heavy-chain disease occurring in nonindustrialized environments.
Acronyms and abbreviations that appear in this chapter include: a-HCD, alpha-heavy-chain disease; BIP, heavy-chain binding protein; CLL, chronic lymphocytic leukemia; EBV, Epstein-Barr virus; Ig, immunoglobulin; IPSID, immunoproliferative small intestinal disease.
DEFINITION AND HISTORY
The heavy-chain diseases are proliferative disorders of B cells that synthesize and secrete incomplete immunoglobulin (Ig) heavy chains. These initially were recognized as gammopathies by the presence of monoclonal proteins in the patients’ serum or urine. The disorders were defined in terms of the production of structurally aberrant immunoglobulin molecules. Since the neoplastic cells displayed plasmacytic features, they were viewed as myeloma variants and studied for the insights they could provide to immunoglobulin structure. The immunochemical description of the first heavy-chain disease protein suggested that the gamma chain was divided into domains. Moreover, it suggested the possibility that the domains were encoded by separate genes (or gene segments).1
The original definition, which remains valid for clinical diagnostic purposes, demanded that patients’ serum or urine contain a deleted Ig H-chain without a bound L-chain.2,3 Later work demonstrated that heavy-chain molecules with similar structures could be identified within neoplastic cells of some cases of so-called nonsecretory myeloma or lymphoma. Also, in some instances defective immunoglobulin light chains could be identified.4 In µ-heavy-chain disease, despite the absence of immunoglobulin light chains in the serum protein, L-chain production by the µ-fragment-synthesizing cells is seen in the majority of cases. Even in the first case of µ-heavy-chain disease, intact immunoglobulin light chains were identified in the proliferating cells, but they were not linked to the µ fragment via a covalent bond in either the cells or the serum.5,6 While the identification of heavy-chain disease proteins is still largely consistent with the original criteria, variations on the theme, detectable by more precise molecular techniques, have allowed an expanded view of the circumstances leading to the production of the benchmark proteins.
ETIOLOGY AND PATHOGENESIS
The etiology of heavy-chain disease is not known. Some of the factors responsible for this disease may be similar to those involved in the etiology and pathogenesis of plasma cell myeloma or chronic lymphocytic leukemia. The etiology of plasmacytic and lymphocytic disorders is discussed in Chap. 106 and Chap. 98.
The observations that none of the heavy-chain disease disorders resembled plasma cell myeloma or Waldenström macroglobulinemia stimulated investigations into the nature of the heavy-chain disease cell and attempts to place it somewhere in the relatively orderly scheme of B-cell differentiation (see Chap. 82). While the infiltrates contain plasma cells, the predominant cell is a lymphocyte, perhaps more precisely reflecting the nature of the tumor stem cell that retains some capacity to mature.
CHRONIC ANTIGENIC STIMULATION
Infection is thought to play a role, at least for some types of heavy-chain disease, particularly a-heavy-chain disease. The latter is more frequently observed in nonindustrialized societies in which gastrointestinal infections are common. A causal relationship between infection and pathogenesis is supported by the observation that early stages of this disease can be treated successfully with antibiotics alone (see discussion below).
There is a high frequency of autoimmune disorders preceding or concurrent with the diagnosis of heavy chain disease, particularly g-heavy-chain disease. This may be related to disease pathogenesis of g-heavy-chain disease in the same fashion as the exposure to gastrointestinal organisms is to the pathogenesis of a-heavy-chain disease.
DEFECTIVE IMMUNOGLOBULIN MOLECULES
Defective Gamma Heavy Chains Structural analysis of the defective monoclonal gamma heavy chains of 23 patients with g-heavy-chain disease reveals several characteristic features (Fig. 109-1). In two cases of g-heavy-chain disease, OMM7 and RIV,8 cDNA and genomic sequence data are available (Fig. 109-2). The proteins usually initiate with a normal variable region amino acid sequence. In most cases, this sequence is short and abruptly interrupted by a large deletion encompassing the remainder of the V region, although four of the proteins shown in Fig. 109-1 appear to have retained most or all of their V, D, and J sequences.9,10,11 and 12 In all g-heavy-chain disease proteins, the entire CH1 domain is also deleted, with normal sequence reinitiating at the hinge (or occasionally the CH2 domain). No light chains are associated with the defective heavy chains, which usually exist in the serum as disulfide-linked dimers.
FIGURE 109-1 A and B
*Structures shown are primary synthetic protein products synthesized by the heavy chain disease cells. Serum proteins were modified after synthesis and did not contain any amino acids before the hinge.
**Structures shown are deduced amino acid sequences determined by cDNA sequencing.
Indicates unusual and heterogeneous amino acid sequences.
Indicates unusual amino acid sequences.
Boxes indicate coding regions.
Lines indicate deletions.
Dashed lines indicate likely structures for which sequence data are missing.
? indicates probable missing domain based on molecular weight and partial protein structure analysis.
V = variable region, D = diversity segment, J = joining region, H = hinge region, CH1, CH2, CH3, CH4 = constant regions of heavy chains, Memb. = membrane exon OMM,7 WIS,50 CHI,51 SPA,52 ZUC,53 CHA,54 gBUR,9 GIF,10 LEA,11 HI,12 HAR,11 BAZ,55 PAR,56 ZAN,57 HAL,58 VAU, LEB,59 WIN,17 UD,57 CRA,61 YOK,62 RIV,8 EST,63 YAO,4 MAL,18 DEF,65 AIT,66 SEC,3 BEN, ARF, MEC, LTE, HAR, AYO,41 BOT,67 DAG,68 GLI,69 BW,14 ROUL,16 µBUR.70
FIGURE 109-2 Boxes indicate coding regions.
Indicates switch region.
Indicates inserted coding sequence.
Indicates inserted non-coding sequence.
Lines indicate intervening (non-coding) sequences.
L = leader region, V = variable region, D = diversity segment, J = joining region, I = inserted sequence, Del = deleted sequence, S = switch region, H = hinge region, CH1, CH2, CH3, CH4 = constant regions of heavy chains, Memb. = membrane exon.
BW,14 OMM,7 RIV, 8 YAO,64 MAL,18 SEC.13
Defective Alpha Heavy Chains As a group, the alpha heavy chains of patients with a-heavy-chain disease have several general features in common with those of the defective gamma heavy chains of patients with g-heavy-chain disease. These include deleted V regions, missing CH1 domains, and the absence of associated light chains. In the cases where amino acid or nucleotide-sequence data are available (Fig. 109-1), most a-heavy-chain disease proteins were shown to contain short unusual sequences of unknown origin at the amino-terminus.
The complete sequences of the genes encoding three a-heavy-chain disease proteins are shown in Fig. 109-2. As in the g-heavy-chain disease genes, OMM and RIV, strikingly similar noncontiguous deletions are present in the V/J and switch/CH1 regions of the a-heavy-chain disease genes. The genomic structures show that the unusual coding regions are actually part of larger regions containing unusual sequences of varying length, which include noncoding as well as coding nucleotides. These inserted regions show no homology to any known sequences, and their mechanism of origin is unclear.
Defective Mu Heavy Chains In common with the defective heavy chains of patients with other types of heavy-chain disease, the defective mu heavy chains of patients with µ-heavy-chain disease contain large variable region deletions (Fig. 109-1). In contrast to the other classes, the µ proteins often contain normal constant regions, including CH1 domains.
There is sequence for only one gene coding for a µ-heavy-chain disease protein (Fig. 109-2). This gene encodes a normal µ constant region containing CH1. The VDJ region is present but was shown to contain a single base deletion generating three stop-codons.14 An inserted stretch of nucleotides immediately 3′ to J destroyed the J donor splice site, causing the cell to splice the 3′ donor site of the leader directly to the CH1 domain, eliminating variable region sequences from the mature mRNA.
Immunoglobulin Light Chains Many patients with µ-heavy-chain disease also synthesize monoclonal light chains detectable as Bence Jones proteins in serum and/or urine. In some cases, immunofluoresence studies showed that the same cells produced both the light and heavy chains.15,16 The light chains can associate in the serum with the short µ heavy chains but do not form covalent bridges even in cases where the CH1 domain, containing the H—L interchain disulfide bond, was shown to be present.16 Monoclonal immunoglobulin light-chain production is rare in patients with g- or a-heavy-chain disease but synthesis of truncated fragments has been noted.4,17,18,19 and 20
GENETIC BASIS FOR DEFECTIVE IMMUNOGLOBULINS
The structural features of the heavy-chain disease proteins can be interpreted in terms of the current knowledge of immunoglobulin gene structure. As expected for secretory proteins, heavy-chain disease genes encode a leader peptide, which is cleaved from the molecule before secretion (see Chap. 9 and Chap. 83). If the heavy-chain disease gene contains intact V and J splice sites, those regions will be present in the mature mRNA and primary synthetic product, even if they contain extensive deletions, insertions, or other in-frame mutations. If there are no intact V or J splice sequences, the donor splice site of the leader peptide will be spliced to the next available acceptor site (Fig. 109-2).
The presence of large deletions in the switch/CH1 regions of the five g- and a-heavy-chain disease genes sequenced so far explain why their corresponding heavy-chain disease proteins lack CH1. Since the normal CH1 acceptor splice site is deleted in these genes, the donor splice site of the leader or J region is alternatively spliced directly to the next available site at the hinge or CH2 domain. It would be expected that other heavy-chain disease proteins with missing CH1 domains contain similar switch/CH1 deletions in their genes. Thus, deletions and mutations of coding regions and splice sites explain why entire domains can be missing from the cytoplasmic mRNA and protein.
Any explanation for the origin of heavy-chain disease proteins must account for the three main defects in heavy-chain disease cells: the two noncontiguous deletions in the V/J and switch/CH1 regions of the heavy-chain genes and the absence of associated light chains.
If the heavy-chain disease abnormalities occurred at random, it should be possible to isolate molecules from serum, which contain any combination of the three defects. Normal heavy chains unassociated with light chains have never been isolated from serum. Biosynthetic studies of immunoglobulin secreting cells have shown that in the absence of light chains, the CH1 domain of heavy chains will bind to the heavy-chain binding protein (BIP) inside the cell or undergo degradation rather than secretion.21 If light chains are present, they will bind to the CH1 domain of the heavy chains and prevent the attachment to BIP. Four g-heavy-chain disease proteins have been described which contain CH1 deletions and appear to have intact VDJ regions.9,10,11 and 12 In these cells, the absence of light chains does not prevent secretion of the aberrant heavy chains, because they are missing CH1 domains and cannot bind BIP; g- and a-heavy-chain disease serum proteins containing only the V/J deletion and intact CH1 domains have not been described. Such structures might be synthesized but would not be secreted from the cell in the absence of light chains due to intracellular binding of BIP at the CH1 domain. This was demonstrated in a study of a nonsecreting myeloma, which produced no light chains but synthesized deleted heavy chains that contained CH1 domains and missing VDJ regions.22 The abnormal H chains were shown to undergo intracellular degradation rather than secretion. Thus, ascertainment bias may partly explain the observed association of the three defects in heavy-chain disease.
It is unclear whether the V/J or switch/CH1 deletion occurs first in heavy-chain disease cells, and it is not known whether the first deletion influences the second. Abnormal VDJ recombination, as an initial event, is suggested by the finding of a significant fraction of cells with aberrant joins when normal human peripheral mononuclear cells are exposed to Epstein-Barr virus (EBV).23 If the cells persist after EBV infection or are generated at some rate throughout life, they could represent potential tumor stem cells. However, deletion of the V region during VDJ joining would render the early B cell unable to synthesize membrane immunoglobulin containing a normal antigen-combining site, and such cells could no longer be stimulated by the antigen cognate for the original antibody. Since switching is generally considered an antigen-dependent event, and most heavy-chain disease cells have switched from µ to g or a, the occurrence of a V region deletion before switching would require the isotype change to be driven by an antigen-independent mechanism such as cytokine stimulation or a proliferative oncogenic event. In only one instance has an activating oncogene (N-ras) mutation been identified in a patient with heavy-chain disease.24
The switch region deletion as the initial abnormality is suggested by the existence of four g-heavy-chain disease proteins (Fig. 109-1) that appear to contain intact variable regions and switch/CH1 deletions.9,10,11 and 12 Alternatively, these molecules may represent a category of heavy-chain disease proteins arising by a different, independent mechanism, either related to or a result of oncogenic transformation.
The variable region deletions and insertions in heavy-chain disease may be a consequence of the hypermutation process. During affinity maturation of normal B cells, the visible consequences of hypermutation are single base substitutions rather than deletions or insertions. However, a study involving single-cell PCR analysis of normal human lymphocytes showed that deletions and insertions were also common in germinal-center–derived B cells undergoing hypermutation but were rare in naive pregerminal center B cells.25,26 The role of this mechanism in the generation of V region deletions in heavy-chain disease would be strengthened if the cells synthesizing the abnormal proteins were shown to carry germinal center markers.
Heavy-chain disease proteins of the g-3 and a-1 classes are overrepresented relative to their frequency among intact human IgG proteins. The reason for this is not known. However, it is noteworthy that the Ig H-chain genes encoding these two heavy chains are arranged in two closely linked clusters in the germ line DNA (see Chap. 83). The first cluster contains the µ/d pair followed sequentially by the gene encoding the g3, g1, and a1 genes, while the second cluster, located 3′ to the first, contains g2, g4, and a2. The preference for isotypes of the first cluster in heavy-chain disease may be a function of their proximity to µ. Switching from µ to an isotype in the second cluster would require the switching mechanism to operate over greater distances, possibly decreasing the likelihood of forming a functional gene. Alternatively, the switch/CH1 deletion in heavy-chain disease proteins may prematurely terminate a sequential switching process, in which cells switch initially from µ to another isotype in the first cluster and then switch again to a gene from the second cluster. Support for this concept comes from studies in which cells were shown to undergo sequential switching, selecting constant region genes in a 5′ to 3′ direction.27
At first it appeared as if each class of heavy-chain disease had a characteristic clinical phenotype. Patients with g-heavy-chain disease typically present with a systemic lymphoma particularly involving lymphoid structures in the head and neck.28 Alpha-heavy-chain disease (a-HCD) is primarily a gut-associated lymphoproliferative state sometimes responding to antibiotics, while µ-heavy-chain disease is an apparent aberrant form of chronic lymphocytic leukemia.29,30 Although the original associations were correct, time has revealed that the deleted proteins are associated with a broader spectrum of clinical syndromes.
From the initial report in 1964 to 1989 almost one hundred cases of g-heavy-chain disease were reported.31,32 Since 1989 only 6 new instances have been recorded, suggesting that knowledge concerning the clinical aspects of the disorder has been incorporated into the body of medical practice; g-heavy-chain disease appears to have the broadest range of clinical presentations among the heavy-chain diseases. As in µ-heavy-chain disease, the most common initial symptom complex is systemic with anemia, weight loss, and occasionally fever. In a significant fraction of patients the detection of lymphadenopathy or splenomegaly is the event that triggers further diagnostic investigation. Some form of lymphoproliferative disease is the final diagnosis in approximately three-quarters of the patients.
The waxing and waning lymphadenopathy and palatal and uvular swelling on the basis of involvement of Waldeyer’s ring, described in the early cases of the disease, are not as frequent as originally believed. While it is characteristic when it occurs, it is found in fewer than 20 percent of g-heavy-chain disease patients.
There has been a striking association of the occurrence of g-heavy-chain disease in patients with preexisting or concurrent autoimmune disease.33 Fully one-third of the reported patients have had some form of immunologically based inflammatory disease, most commonly rheumatoid arthritis. Autoimmune hemolytic anemia, systemic and discoid lupus erythematosus, and Sjögrens syndrome also have been seen associated with this disease
The clinical features of g-heavy-chain disease differ from multiple myeloma (Table 109-1), as those of µ-heavy-chain disease differ from those of Waldenström’s disease (Table 109-2). Renal disease and osteolytic lesions are far less common in g-heavy-chain disease. Renal failure secondary to Bence-Jones proteinuria or AL amyloid is rare, owing to the lack of production or secretion of immunoglobulin light chains.
TABLE 109-1 CLINICAL FEATURES OF MULTIPLE MYELOMA AND GAMMA-HEAVY-CHAIN DISEASE
TABLE 109-2 CLINICAL FEATURES OF WALDENSTRÖM’S MACROGLOBULINEMIA AND MU-HEAVY-CHAIN DISEASE
Lymphadenopathy is present in approximately half the cases; splenomegaly is detectable in between half and three-quarters; hepatomegaly occurs in about one-third. Extralymphoid presentations have been seen in 10 to 15 percent, with cutaneous or thyroid infiltrates specifically reported. In contrast to µ-heavy-chain disease, only two individuals with g-heavy-chain disease proteins have been noted to have chronic lymphocytic leukemia (CLL).
Unlike the other heavy-chain diseases, most cases of a-HCD arise in a characteristic environmental setting. The patients are usually young, in their teens and twenties, and live in nonindustrialized locales with relatively poor sanitation. The disease has been described as “little more than an academic curiosity” in industrialized nations, but it is a widely prevalent, debilitating illness in developing countries.34 Thus significant clinical series and systematic approaches to therapy have been reported from Tunisia, Iran, Pakistan, India, Thailand, South Africa, and Mexico. The index case and the cases in which the structural features of the proteins have been best characterized have been from North Africa.
The most common clinical presentation is that of recurrent or chronic diarrhea with abdominal pain and weight loss due to malabsorption. Fever is common. For a period of time there was confusion over whether Mediterranean lymphoma with malabsorption and a-heavy-chain disease were different conditions, primarily because the heavy-chain disease protein could not be demonstrated in many cases of the lymphoma. It is now clear that most patients with the clinical picture of Mediterranean lymphoma produce the defective IgA proteins at some time in their course. In many cases the protein can only be found in the intestinal secretions, and in others it can only be found within the cytoplasm of the infiltrating plasma cells. All of these patients are now classified as having immunoproliferative small intestinal disease (IPSID).35,36
In addition to the diarrhea, malabsorption, and weight loss, growth retardation is common, and digital clubbing is found in 33 to 66 percent of patients. Mesenteric lymphadenopathy is frequently manifested as an intestinal mass, but extraabdominal lymphadenopathy is rare. Splenic involvement is also uncommon. Moderate hepatomegaly occurs in 20 to 25 percent of the patients. Ascites is sometimes seen, as is peripheral edema, both on the basis of hypoalbuminemia.
Of the major Ig classes, µ-heavy-chain disease is the least common. Approximately 30 cases have been documented in the literature since its original description.5,37 It is likely that more cases have been identified but not reported, particularly in the last 10 years. Alternatively, since it is now apparent that the clinical picture is not specific, it is possible that many additional cases have been diagnosed as CLL- or B-cell lymphoma, and the protein abnormality was not recognized. Initially the reported cases all had a clinical picture consistent with CLL; however, with better diagnostic procedures and a higher index of suspicion the defining protein abnormality has been noted in a broader range of clinical settings. Currently, it appears that only about one-third of the patients with µ-heavy-chain disease have CLL. Thorough investigation of serum proteins in over 150 consecutive patients with CLL failed to identify a single instance of µ-heavy-chain disease, suggesting that fewer than 1 percent of cases of CLL actually are associated with production of an abnormal µ fragment.38 Other clinical presentations varied from essential monoclonal gammopathy with little or no clinical symptoms of B-cell lymphoma (20 percent). Ten percent of the cases were associated with the simultaneous detection of an intact IgM protein in the serum, and another 10 percent with clinical multiple myeloma or extramedullary plasmacytoma.
The presentation is usually that of a systemic disease with weight loss, sometimes fever, anemia, and recurrent infections. Splenomegaly is found in almost all cases, hepatomegaly in three-quarters, and peripheral lymphadenopathy in approximately 40 percent of patients with µ-heavy-chain disease.
DELTA- AND EPSILON-HEAVY-CHAIN DISEASE
No cases have been reported in which an incomplete epsilon chain has been identified. A single case of delta-heavy-chain disease has been described. The patient presented with the clinical features of multiple myeloma with osteolytic lesions of the skull, marrow plasmacytosis, and the rapid development of renal failure and death. While the immunochemical analysis of the patient’s serum was consistent with the presence of a polymerized d fragment, no definitive sequence data were obtained.39
DETECTION OF ABNORMAL IMMUNOGLOBULIN PROTEIN
As a rule the neoplastic cells do not produce large amounts of immunoglobulin. This makes it sometimes difficult to detect the abnormal immunoglobulin produced in heavy-chain disease. Nevertheless, a combination of electrophoretic, immunoelectrophoretic, and immunofixation techniques40 can establish the diagnosis. In a minority of cases the proteins can be initially identified as a discrete homogeneous band of b mobility on serum or urine electrophoresis. The immunoelectrophoretic pattern, when developed with specific anti-heavy- and anti-light-chain antisera, reveals an H-chain specific arc not reacting with either the anti-k or anti-l antisera. This is most readily apparent in proteins of the g class, since occasionally intact monoclonal IgA or IgM M-proteins will not react with some anti-L-chain sera. This is more common with immunoelectrophoresis than with immunofixation techniques. In those cases it may be necessary to separate the monoclonal proteins from the sera, treat them with reducing agents to cleave the disulfide bonds, and subject them to gel electrophoresis to determine the size of the immunoglobulin heavy chain polypeptide.
More commonly the proteins are present in smaller amounts and give heterogeneous patterns on electrophoresis, either because of N-terminal proteolysis or partial digestion of the glycosyl-chains attached to the heavy chains. Again, immunoelectrophoresis or immunofixation with development of the patterns with a panel of anti-H and anti-L antibodies can strongly suggest the diagnosis. More detailed, structural analysis can be performed on the isolated, reduced, and alkylated H-chain monomer and confirm the presence or absence of immunoglobulin light chains in specialized laboratories.
As in patients without µ-heavy-chain disease, marrow aspirates or biopsies often show a pleiomorphic histologic picture with lymphocytes, plasma cells, and cells called lymphocytoid plasma cells, or plasmacytoid lymphocytes, making up a significant proportion of the infiltrate. In 25 to 30 percent of cases the marrow is nondiagnostic or normal. Immunophenotyping of the marrow aspirate has been performed in a minority of patients. In experienced laboratories such studies have usually revealed a lymphoid population staining only for the gamma H-chain with no staining for immunoglobulin light chains. In some cases in which the heavy-chain disease protein could not be detected in the serum or urine, or in which the concentration of heavy-chain disease protein was too low or heterogeneous by electrophoresis, these studies provided the definitive diagnostic test. In a few the analyses revealed the presence of more than a single proliferating clone, and in at least three instances demonstrated a major population of proliferating T cells and a quantitatively minor B-cell clone producing the g-heavy-chain disease protein.
Occasionally, most commonly in a-heavy-chain disease, the protein cannot be seen in the serum, urine, or even the gastrointestinal secretion. Instead a population of plasmacytoid cells may be found in the intestinal epithelium, lymph nodes, or the marrow that stain for only H-chain determinants. In these cases the typical molecular abnormality of heavy-chain disease (see below) can be identified by analysis of cDNA derived from the heavy-chain disease positive cells.41
Serial small-bowel biopsies have shown that the early phases of the disease are characterized by dense mucosal and lamina propria infiltration mature plasma cells producing a monoclonal alpha-chain-fragment.42 The jejunum is the usual site of involvement with the duodenum and ileum involved less often. Involvement may be patchy, and different areas may be in different stages of the disease at the same time, requiring biopsies from multiple sites for accurate staging. As the disease progresses, fewer plasma cells are seen and the dominant cells, still derived from the same clone, are more blastic with little or no production of the alpha fragment. At this point the cells have infiltrated beyond the lamina propria into the muscularis layer. In the latter stage there is extensive infiltration of regional nodes with similar cells and final pathologic diagnoses have included lymphoma of the low, intermediate, or high-grade immunoblastic type, mantle zone lymphoma, or parafollicular B-cell lymphoma. Rarely the tumor cells may be seen in the marrow, although the frequency of marrow involvement has not been precisely determined.
As expected with this sequence of involvement, early imaging of the small bowel reveals a pseudopolypoid appearance. When the muscularis is penetrated a more cobblestone-like picture is found. In the late stages tumor masses are defined, and surface ulceration is detectable. The diagnosis requires small-bowel biopsies from more than one site along the jejunum, and the presence of the alpha fragment in the serum, the intestinal secretions, or in the cells infiltrating the gastric mucosa. If there is demonstrable nodal involvement, mesenteric lymph node biopsy may yield the same findings. The differential diagnosis includes celiac disease, diffuse intestinal giardiasis, Western type intestinal lymphoma (which is not associated with a-fragment production), bacterial overgrowth syndrome, and Whipple disease. Interestingly, many of these patients may have concomitant parasitic infestation or chronic infection with bacteria, such as Helicobacter pylori. The relevance of these infections to pathogenesis is unclear, although they certainly contribute to the symptoms and must be treated.
Instances of patients producing a-heavy-chain disease proteins have been found in other clinical contexts in more industrialized parts of the world. A single three-year-old with frequent infections, hypogammaglobulinemia, and leukopenia was found to have a defective alpha chain,43 but no detailed structural studies were reported. Rare patients in the United States and Japan, with Western type lymphomas involving only the gut or with more widespread disease, have been noted to have a similar circulating protein.44 Single patients with goiter or skin lesions and no detectable systemic disease also have been described. These may represent a different mode of pathogenesis.
Marrow involvement is characterized by infiltration with lymphocytes and plasma cells. Cells that often are described as lymphocytic plasmacytes or plasmacytoid lymphocytes are prominent. Although the marrow of almost all patients contains the multivacuolated plasma cells described in the index case, the vacuoles are not universally apparent.15 Their identification has sometimes been the clinical feature that has suggested the diagnosis, subsequently confirmed by the appropriate electrophoretic and immunoelectrophoretic studies.
The presence or absence of osteolytic lesions or pathologic fractures only has been noted in half the reported patients, with bone involvement occurring in 40 percent of those. It is likely that the true incidence is lower but still approaches the frequency seen in multiple myeloma, rather than that in Waldenström’s macroglobulinemia, where it is quite rare.
Urinary excretion of the µ fragment was noted in only two patients, presumably because in most patients the polymers of the carboxy-terminal µ fragment were too large to be filtered by intact renal glomeruli. Monoclonal light chains have been found in the urine in two-thirds of cases. Nonetheless, renal complications have been infrequent. Cast nephropathy, with renal failure, has been reported in one case, and two patients, including the first reported case, had AL amyloidosis.45 Immunoglobulin light chains capable of producing amyloid are found in approximately 12 percent of cases, an incidence that is not significantly different from that found in patients with multiple myeloma.
THERAPY, COURSE, AND PROGNOSIS
The clinical course has been extremely variable. Survival, ranging from one month to 20 years from the finding of the protein abnormality, has reflected the nature of the associated disease. Patients with a lymphoma on presentation have a more aggressive course than do the patients with little clinical evidence of lymphoproliferative disease. The lymphomas have been treated with a variety of regimens including splenic or other local irradiation and combinations of chemotherapy, usually including cyclophosphamide, or vincristine and prednisone, although other agents have been used in some patients. A single case has been reported in which there was a complete response to fludarabine after a partial remission with epirubicin, chlorambucil, and prednisone.46 Patients with predominant autoimmune disease have been treated with prednisone, and some patients have received no treatment. Disappearance of the g-heavy-chain disease protein in response to treatment has not always been predictive of a good overall therapeutic response.
Since 1965 ten series with more than 10 patients each have assessed the impact of therapy on the course of disease and survival. The total number of patients included is 206. The utility of clinical staging was not recognized until the late 1970s, hence details of responsiveness prior to that time are difficult to interpret. However it appeared from those data that mean overall survival was approximately 8 months in the absence of a therapeutic response and that fewer than half the patients responded to any treatment. Since the early studies, most series have classified patients into those with only mucosal infiltration with plasma cells (stage A) and the remainder with deeper infiltration or nodal spread (stages B and C).47,48 and 49 When stage A patients are treated with antibiotics (usually tetracycline at 1 to 2 g/d for 6 to 8 months), response rates range from 25 percent to 72 percent. The cumulative response rate to antibiotic therapy is 53 percent, with overall survival between 29 percent and 75 percent at 5 years and disease-free survival in one series of 43 percent at 5 years. An occasional stage A patient also received abdominal irradiation. Stage B and C patients were sometimes treated with similar antibiotic regimens and a variety of chemotherapy regimens, usually including cyclophosphamide and prednisone (such as COPP or CHOP, sometimes followed by BACOP, see Chap. 103).
The course is variable in duration with survival ranging from 1 to 11 months after the appearance of symptoms. Because of its rarity no large series of patients treated in a systematic way in a single center has been reported. A variety of drug protocols have been used with most including alkylating agents (chlorambucil, melphalan, and cyclophosphamide) with inconsistent responses. It is possible that more aggressive regimens will be more effective, but to date there is no reported experience bearing on that question.
Kataoka T, Yamawaki-Kataoka Y, Yamagishi H, Honjo T: Cloning immunoglobulin gamma2b chain gene of mouse: characterization and partial sequence determination. Proc Natl Acad Sci USA 76:4240, 1979.
Franklin EC, Lowenstein J, Bigelow B, Meltzer M: Heavy chain disease—a new disorder of serum gamma-globulins. Am J Med 37:332, 1964.
Osserman EF, Takatsuki K: Clinical and immunochemical studies of four cases of heavy (H gamma) chain disease. Am J Med 37:351, 1964.
Matuchansky C, Cogne M, Lemaire M, et al: Nonsecretory alpha chain disease with immunoproliferative small intestinal disease. N Engl J Med 320:1534, 1989.
Ballard HS, Hamilton LM, Marcus AJ, Illes CH: A new variant of heavy chain disease (mu chain disease). N Engl J Med 282:1060, 1970.
Buxbaum J, Franklin EC, Scharff MD: Immunoglobulin M heavy chain disease: intracellular origin of the mu chain fragment. Science 169:770, 1970.
Alexander A, Anicito I, Buxbaum J: Gamma heavy chain disease in man: genomic sequence reveals two noncontiguous deletions in a single gene. J Clin Invest 82:1244, 1988.
Guglielmi P, Bakhshi A, Cogne M, Seligmann M, Korsmeyer SJ: Multiple genomic defects result in an alternative RNA splice creating a human gamma H chain disease protein. J Immunol 141:1762, 1988.
Prelli F, Frangione B: Franklin’s disease: Ig gamma2 H chain mutant BUR. J Immunol 148:949, 1992.
Cooper SM, Franklin EC, Frangione B: Molecular defect in a gamma2 heavy chain. Science 176:187, 1972.
Frangione B, Franklin EC, Smithies O: Unusual genes at the aminoterminus of human immunoglobulin variants. Nature 273:400, 1978.
Terry WD, Ohms J: Implications of heavy chain disease protein sequences for multiple gene theories of immunoglobulin synthesis. Proc Natl Acad Sci USA 66:558, 1970.
Cogne M, Preud’homme J: Gene deletions force nonsecretory alpha-chain disease plasma cells to produce membrane-form alpha-chain only. J Immunol 145:2455, 1990.
Bakhshi A, Guglielmi P, Siebenlist U, Ravetch J, Jensen JP, Korsmeyer SJ: A DNA insertion/deletion necessitates an aberrant RNA splice accounting for a mu heavy chain disease protein. Proc Natl Acad Sci USA 83:2689, 1986.
Zucker-Franklin D, Franklin EC: Ultrastructural and immunofluorescence studies of the cells associated with mu-chain disease. Blood 37:257, 1971.
Cogne M, Aucouturier P, Brizard A, Dreyfus B, Duarte F, Preud’homme J: Complete variable region deletion in a mu heavy chain disease protein (ROUL). Correlation with light chain secretion. Leukemia Res 17:527, 1993.
Adlersberg JB, Grann V, Zucker-Franklin D, Frangione B, Franklin EC: An unusual case of a plasma cell neoplasm with an IgG3-lambda myeloma and a gamma3 heavy chain disease protein. Blood 51:85, 1978.
Tsapis A, Bentaboulet M, Pellet P, et al: The productive gene for alpha-H chain disease protein MAL is highly modified by insertion-deletion processes. J Immunol 143:3821, 1989.
Cogne M, Bakhshi A, Korsmeyer SJ, Guglielmi P: Gene mutations and alternate RNA splicing result in truncated Ig L chains in human gamma H chain disease. J Immunol 141:1738, 1988.
Hauke G, Schiltz E, Bross KJ, Hollmann A, Peter HH, Krawinkel U: Unusual sequence of immunoglobulin L-chain rearrangements in a gamma heavy chain disease patient. Scand J Immunol 36:463, 1992.
Hendershot L, Bole D, Kohler G, Kearney JF: Assembly and secretion of heavy chains that do not associate posttranslationally with immunoglobulin heavy chain-binding protein. J Cell Biol 104:761, 1987.
Cogne M, Guglielmi P: Exon skipping without splice site mutation accounting for abnormal immunoglobulin chains in nonsecreting human myeloma. Eur J Immunol 23:1289, 1993.
Brokaw JL, Wetzel SM, Pollok BA: Conserved patterns of somatic mutation and secondary VH gene rearrangement create aberrant Ig-encoding genes in Epstein-Barr Virus-transformed and normal human B lymphocytes. Int Immunol 4:197, 1992.
Moskovits T, Jacobson DR, Buxbaum JN: N-ras oncogene activation in a patient with gamma heavy chain disease. Am J Hematol 41:302, 1992.
Goossens T, Klein U, Kuppers R: Frequent occurrence of deletions and duplications during somatic hypermutation: implications for oncogene translocations and heavy chain disease. Proc Natl Acad Sci USA 95:2463, 1998.
Klein U, Goossens T, Fischer M, et al: Somatic hypermutation in normal and transformed human B cells. Immunol Rev 162:261, 1998.
Petrini J, Shell B, Hummel M, Dunnick W: The immunoglobulin heavy chain switch: structural features of gamma1 recombinant switch regions. J Immunol 138:1940, 1987.
Buxbaum J: Heavy chain diseases in man. La Ricerca in Clinica e in Laboratorio 6:301, 1976.
Rambaud JC, Bognel C, Prost A, et al: Clinico-pathological study of a patient with “Mediterranean” type of abdominal lymphoma and a new type of IgA abnormality (“Alpha Chain Disease”). Digestion 1:321, 1968.
Franklin EC: µ-chain disease. Arch Intern Med 135:71, 1975.
Fermand J, Brouet J, Danon F, Seligmann M: Gamma heavy chain “disease”: heterogeneity of the clinicopathologic features. Report of 16 cases and review of the literature. Medicine 68:321, 1989.
Kyle RA, Greipp PR, Banks PM: The diverse picture of gamma heavy chain disease. Report of seven cases and review of literature. Mayo Clin Proc 56:439, 1981.
Husby G, Blichfeldt P, Brinch L, et al: Chronic arthritis and gamma heavy chain disease: coincidence or pathogenic link? Scand J Rheumatol 27:257, 1998.
Ryan JC: Premalignant conditions of the small intestine. Semin Gastrointestinal Dis 7:88, 1996.
Rambaud JC, Halphen M, Galian A, Tsapis A: Immunoproliferative small intestinal disease (IPSID): relationships with alpha chain disease and “Mediterranean” lymphomas. Springer Semin Immunopathol 12:239, 1990.
Martin IG, Aldoori MI: Immunoproliferative small intestinal disease: Mediterranean lymphoma and alpha heavy chain disease. Br J Surg 1:20, 1994.
Wahner-Roedler DL, Kyle RA: Mu-heavy chain disease: presentation as a benign monoclonal gammopathy. Am J Hematol 40:56, 1992.
Bonhomme J, Seligmann M, Mihaesco C, et al: Mu-chain disease in an African patient. Blood 43:485, 1974.
Vilpo JA, Irjala K, Viljanen MK, Klemi P, Kouvonen I, Ronnemaa T: Delta-heavy chain disease. A study of a case. Clin Immunol Immunopathol 17:584, 1980.
Franklin EC: The heavy chain diseases. The Harvey Lectures 78:1, 1984.
Fakhfakh F, Dellagi K, Ayadi H, et al: Alpha heavy chain disease alpha mRNA contain nucleotide sequences of unknown origins. Eur J Immunol 22:3037, 1992.
Galian A, Lecestre M, Scotto J, Bognel C, Matuchansky C, Rambaud J: Pathological study of alpha-chain disease, with special emphasis on evolution. Cancer 39:2081, 1977.
Faux JA, Crain JD, Rosen FS, Merler E: An alpha heavy chain abnormality in a child with hypogammaglobulinemia. Clin Immunol Immunopathol 1:282, 1973.
Cohen HJ, Gonzalvo A, Krook J, Thompson TT, Kremer WB: New presentation of alpha heavy chain disease: North American polypoid gastrointestinal lymphoma. Clinical and cellular studies. Cancer 41:1161, 1978.
Preud’homme J, Bauwens M, Dumont G, Goujon JM, Dreyfus B, Touchard G: Cast nephropathy in mu heavy chain disease. Clin Nephrol 48:118, 1997.
Agrawal S, Abboudi Z, Matutes E, Catovsky D: First report of fludarabine in gamma-heavy chain disease. Br J Haematol 88:653, 1994.
Akbulut H, Soykan I, Yakaryilmaz F, et al: Five-year results of the treatment of 23 patients with immunoproliferative small intestinal disease. A Turkish experience. Cancer 80:8, 1997.
Ben-Ayed F, Halphen M, Najjar T, et al: Treatment of alpha chain disease. Results of a prospective study in 21 Tunisian patients by the Tunisian-French intestinal lymphoma study group. Cancer 63:1251, 1989.
Salimi M, Spinelli JJ: Chemotherapy of Mediterranean abdominal lymphoma. Retrospective comparison of chemotherapy protocols in Iranian patients. Am J Clin Oncol 19:18, 1996.
Frangione B, Rosenwasser E, Prelli F, Franklin EC: Primary structure of human gamma3 immunoglobulin deletion mutant: gamma3 heavy-chain disease protein Wis. Biochem 19:4304, 1980.
Frangione B: A new immunoglobulin variant: gamma3 heavy chain disease protein CHI. Proc Natl Acad Sci USA 73:1552, 1976.
Frangione B, Franklin EC: Correlation between fragmented immunoglobulin genes and heavy chain deletion mutants. Nature 281:600, 1979.
Wolfenstein-Todel C, Frangione B, Prelli F, Franklin EC: The amino acid sequence of “heavy chain disease” protein ZUC. Structure of the Fc fragment of immunoglobulin G3. Biochem Biophys Res Commun 71:907, 1976.
Arnaud P, Wang A, Gianazza E, et al: Gamma heavy chain disease protein CHA: immunological and structural studies. Mol Immunol 18:379, 1981.
Smith LL, Barton BP, Garver FA, et al: Physicochemical and immunochemical properties of gamma1 heavy chain disease protein BAZ. Immunochemistry 15:323, 1978.
Rabin BS, Moon J: Clinical findings in a case of newly defined gamma heavy chain disease protein. Clin Exp Immunol 14:563, 1973.
Franklin EC, Prelli F, Frangione B: Human heavy chain disease protein WIS: implications for the organization of immunoglobulin genes. Proc Natl Acad Sci USA 76:452, 1979.
Frangione B, Lee L, Haber E, Bloch KJ: Protein Hal: partial deletion of a “gamma” immunoglobulin gene(s) and apparent reinitiation at an internal AUG codon. Proc Natl Acad Sci USA 70:1073, 1973.
Franklin EC, Kyle R, Seligmann M, Frangione B: Correlation of protein structure and immunoglobulin gene organization in the light of two new deleted heavy chain disease proteins. Mol Immunol 16:919, 1979.
Sala P, Tonutti E, Pizzolitto S, et al: Immunochemical and structural characterization of an IgG1 heavy chain disease. Res Clin Lab 19:59, 1989.
Franklin EC, Frangione B: The molecular defect in a protein (CRA) found in gamma1 heavy chain disease, and its genetic implications. Proc Natl Acad Sci USA 68:187, 1971.
Nabeshima Y, Ikenaka T, Arima T: N- and C-terminal amino acid sequences of a gamma heavy chain disease protein YOK. Immunochemistry 13:245, 1976.
Biewenga J, Frangione B, Franklin EC, Van Loghem E: A gamma1 heavy-chain disease protein (EST) lacking the entire VH and CH1 domains. Scand J Immunol 11:601, 1980.
Bentaboulet M, Mihaesco E, Gendron M, Brouet J, Tsapis A: Genomic alterations in a case of alpha heavy chain disease leading to the generation of composite exons from the JH region. Eur J Immunol 19:2093, 1989.
Wolfenstein-Todel C, Mihaesco E, Frangione B: “Alpha chain disease” protein DEF: Internal deletion of a human immunoglobulin A1 heavy chain. Proc Natl Acad Sci USA 71:974, 1974.
Wolfenstein-Todel C, Mihaesco E, Frangione B: Variant of a human immunoglobulin: “alpha chain disease” protein AIT. Biochem Biophys Res Commun 65:47, 1975.
Barnikol-Watanabe S, Mihaesco E, Mihaesco C, Barnikol HU, Hilschmann N: The primary structure of mu-chain-disease protein BOT. Hoppe-Seyler’s Z Physiol Chem 365:105, 1984.
Mihaesco C, Ferrara P, Guillemot JC, et al: A new extra sequence at the amino terminal of a mu heavy chain disease protein (DAG). Mol Immunl 27:771, 1990.
Franklin EC, Frangione B, Prelli F: The defect in mu heavy chain disease protein GLI. J Immunol 116:1194, 1976.
Lebreton JP, Ropartz C, Rousseaux J, Roussel P, Dautrevaux M, Biserte G: Immunochemical and biochemical study of a human Fc mu-like fragment (mu-chain disease). Eur J Immunol 5:179, 1975.
Copyright © 2001 McGraw-Hill
Ernest Beutler, Marshall A. Lichtman, Barry S. Coller, Thomas J. Kipps, and Uri Seligsohn