Document 91Zb0GrpkkLG4KqZOoE6bJjN5

The Incidence and Epidemiology of Plasma Cell Neoplasms Daniel Bergscigel Ontario Cancer InstitutelPrincess Margaret Hospital. Department of Medicine. Toronto. Ontario. Canada Key Words. Myeloma ~lacroglobulinemia Monoclonal gamrnopathy of undetermined significance Prevalence Incidence Age Race Predi5posing factors Abstract. Plasma cell neoplasia includesmonoclonal developing MM.Previous evidence sbggesting that gammopathy of undetermined significance(MGUS), radiation and benzene may play a role in the multiple myeloma (JIJI), and Waldenstrom's pathogenesis of MM has been greatly weakened macroglobulinemia (WW.In MGCS. a large, stable by subsequent follow-up studies. clone does not cause s>mptoms;additional change(s) Marked differences in the incidence of spon- Mare required to con\ert this clone into a progres- taneous MGUS in inbred strains of mice, differences I sively expanding tumor that becomes sjmptomatic, in the susceptibility of these mice to the induction as in MM or WM. The prevalence of MGUS (Le., of plasmacytomas by the intraperitoneal injection the number of cases in a defined population at a of mineral oil, the racial differences previously men- certain time) is 20 times greater than hlM. The tioned in the incidence of human MM, the associa- incidence (i.e., the number of cases developing in a tion of M M with the HLA Cw2 allele, and the defined population in a defined period) has not familial occurrence of MGUS, MM and WM, sug- been determined for MGUS. Between 1960 and 1969, the average. annual, age-adjusted (1950stan- dard) incidence of JIM in Jlalmo, Sweden was 3.4/105. The incidence of SIM is strongly influenced by the age and race of the population, and the diagnostic services available. gest that genetic factors are important in the genesis of plasma cell neoplasms. Chronic antigenic stimulation (CAS) has long been thought to play a role, but epidemiologic studies have only established an association of MM with rheumatoid arthritis. MM is a disease of old age; it rarely occurs before the age of 40. The incidence of MM increases rapidly with age, is lowest among the Chinese and Introduction Japanese, intermediate among Caucasians in America and Europe. and highest among blacks in the USA. The striking differences in the incidence of MM in different countries appears to be due to racial rather than en\ ironmental differences, since the low incidence among the Chinese and Japanese Plasma cell neoplasms are initiated when an early B cell precursor-with an idiotypic, preswitch rearrangement of the immunoglobulin gene-undergoes a malignant transformation and in Asia has migrated with them to the Bay area of proliferates to form a clone [ I ] . This clone must I California and to Hawaii. The high incidence of increase to about 5 x 10' cells before it can pro- MM in USA black males (10.8/105) and females (7.2/105) is more than twice the rate for whites in the same regions. A striking increase in MM mortality rates has been noted in many countries duce enough of the monoclonal immunoglobulin to be recognized as a "spike" (M-protein) in a serum electrophoresis pattern. Most subjects with ! between 1968and 1986. There is much debate about an M-protein are asymptomatic. If other causes of whether this increase is real, or the result of steadily an M-protein can be ruled out. they are labeled as improving case ascertainment. Epidemiologists habe not firmly identified an) hazardous materials that increase the risk of having monoclonal gammopathy of undetermined significance (MGUS). The clone in MGUS is stable. by definition. and the serum M-protein ~~ Correspondence: Dr. Daniel E. Bergsagel. Department of Medicine,,Ontario Cancer Institute/ Princess Margaret Hospital. 500 Sherbourne Street. Toronto. ONT M1X 1K9. Canada. Received September 1. 1994: accepted for publication September 1. 1994.OAlphaMed Press 1066- concentration remains on a plateau for many years. However. prolonged follow-up of 711 subjects with MGUS at the Mayo Clinic has shown that about two percent of these patients per year progress to develop multiple myeloma (MM). Waldenstrom's macroglobulinemia (WM). a 5099/95/%5.OO/O malignant lymphoma. chronic lymphocytic STEM CELLS 1995;13(suppl2):1-9 Plasma Cell Neoplasms: Incidence and Epidemiology leukemia or amyloidosis [?I. MGUS is consid- ered to be a premalignant lesion because the clone does not grow progressively. Additional neoplastic change(s) idare required to convert this large. stable clone into a progressively expanding tumor with malignant characteristics. such as MM or WM. since 1973. For the 1973-77 period. the average. annual. age-adjusted ( 1970 standard) incidence rates per 100.000 for all races was 3.9 for both sexes. 4.7 for males and 3.3 for females [6]. The frequency of MM in a population is strongly influenced by age. race and acce\sibility of good diagnostic facilities. Prevalence and Incidence The prevalence of MGUS and MM (i.e.. the number of cases in a defined population at a certain time) in the Varmland district of Sweden in 1964. and the incidence (Le.. the number of cases developing in a defined population over a defined period of time) of MM i n Malmo. Sweden from 1960-69. ire shown in Table 1. The prevalence of these diseases was determined by doing a serum electrophoresis on 6,995 consecutive adult blood donors over the age of 25. in the spring of 1964 [3]. An M-protein was detected in the sera of 64 individuals (0.9 percent). Only one of this group had MM; 63 were classified as MGUS. Among the 6.93 1 subjects without serum M-proteins in 1964. two were diagnosed as MM (one had light chain MM) within three years [4]. The population of Varmland in 1964 was about 10,000.In the 70 percent of the adult population of this district who had a serum electrophoresis done in the spring of 1964, the prevalence of MGUS was 9O1/1Os, and of MM 43/1@. MGUS is encountered much more frequently than MM because these subjects are well. and in contrast to MM patients, who have a median survival of only 32 months from the onset of treatment, MGUS patients tend to survike for prolonged periods, and thus accumulate in the population. The incidence of MGUS has not been determined. The average. annual. age-adjusted ( 1950 standard) incidence of MM in Malmo. Sweden was 3.4/10' [SI. In the USA, MM incidence data has been provided by the Surveillance. Epidemiology and End Results (SEER)program Age The age-specific incidence rates for Malmo. Sweden, 1970-79. are shown in Figure 1. MM was not detected in patients under the age of 40 in this sample. Thereafter the incidence rose progressively with age. to reach 64.5/105 in males. and 36.6110' in females over the age of 80 years [5]. The frequency of MGUS also increases with age (Table 11). Race The average. age-adjusted (world standard population) incidence of MM from selected, population-based cancer registries around the world ranges from 0.5 in Hawaiian Japanese males to 8.2 in USA Bay area black males. per 100.000person years. The incidence of MM is distinctly lower ( 1.4/105,or less) for the Chinese of Shanghai and Singapore, and the Japanese of Osaka and Hawaii. than in the Caucasian populations of North America and Europe. Comparisons between incidence rates from cancer registries in different countries must be made cautiously. since the populations served may vary in terms of the availability of diagnostic services, the completeness of case ascertainment, and the calculation of age-adjusted incidence rates. Still, there do appear to be real differences in the incidence of MM in different races. The observation that the incidence of MM. age-adjusted to the 1970 U.S. standard. in male Chinese (2.3/10() and Japanese (1.7/1@)living in San Francisco-Oakland and Hawaii is distinctly lower than for white males (4.6/105) in these same regions supports this view [7]. Furthermore. the MM incidence rates for blacks in America are more than double the rates for whites. at Table I. Prc\ alence and incidence of MGUS and M b l in Sweden Disease [Reference] MGUS [3] M.M 14. 51 No./Total 6416995 316995 Prevalence. 1964 per 100.000 90 I 43 Incidence. 1960-69 per 100.000 ? 3.4 Bergsagel Age-Spec -& t 4 10 I 0 8 4c Fig. 1. Averagc of multiple my< Rates ape-adju tion and the Eu brackets): male sexes 3.1 (4.3) 10.8/105for t females [8]. 1 blacks makes ical malignan United States The pret. subjects over communities 111) [101. The I Japan is simil in this coun immunoglob nephelometr! vation that, fc were signif Japanese. Thi 1 mitted for e Veterans Ad1 Table 11. The Author [Refer Enylisoi*d er (r Rad1 er ul. [S? d Epidemiology )d. the average, lard) incidence as 3.9 for both emales [6]. population is e and accessies. ttes for Malmo. Figure I . MM :r the age of 40 ncidence rose ch 64.5/10-' in w e r the age of I f MGUS also world standard from selected. ies around the iiian Japanese ack males. per ence of M M is for the Chinese 3 the Japanese the Caucasian nd Europe. fnce rates from intries must be ilations served lity of diagnohcase ascenain- age-adjusted ,pear to be real 1M in different :idence of MM. mdard, in male .7/10') living in aii is distinctly d l Os)in these '1. Furthermore. ks in America for whites. at dence. 1960-69 per 1OO.OOO 9 3.4 Bergsagel 3 Age-Specific Incidence of Multiple Myeloma to be much higher in blacks than in whites. especially in the group over the age of 79 [ 1 I ] . In Malmb. Sweden, 1970-79 8 t again in agreement with the higher incidence of MM i n blacks. Increasing Mortality R u m f o r M M -2 10 2 I Trends in the MM mortality rates for four age groups in the U.S.between 1968 and 1986 - are shown in Figure 2. A striking increase in these rates occurred during this period. espe- cially i n the older age groups [ 111. Similar A s40 o; $0 o; increases were noted in the reported. MM death rates from the United Kingdom. France. Age llW. Germany, Japan and Italy [ 12. 131. The rate of increase in the MM mortality rate is greatest in the group over the age of 85. and falls continu- Fig. 1. Average. annual. age-specific incidence rates of multiple myeloma in Malmo. Sweden. 1970-1979. Rates age-adjusted to the 1950 U.S.(total) population and the European standard population (values in brackets):males 4.6 (6.2);females 2.2 (3.0);and both sexes 3.2 (4.3)[ 5 ] . ously with decreasing age. The rate of increase is substantial by the age of 70. more than doubling in both sexes and in all countries between 1968 and 1986. There is much debate about whether the increasing mortality rate from MM is real. or the result of steadily improving case ascertainment. 10.8/10sfor black males and 7.2/1OSfor black The fact that this marked increase in the MM females [SI. The increased incidence of MM in mortality rate has not been observed in three blacks makes this the most common hematolog- communities with a high standard of medical ical malignancy in the black population of the care and a long interest in MM (Olmstead United States [9]. County, Minnesota: Malmo, Sweden and the can- The prevalence of MGUS is also lower in ton of Vaud, Switzerland) suggests that the rising subjects over the age of 60 living in retirement MM mortality rates in other parts of the world communities in Japan than in the U.S.(Table are largely due to improving case ascertainment 111) [IO]. The reduced prevalence of MGUS in [ 5 , 14-16]. Kyle et ai. [ 161 attribute their failure I Japan is similar to the lower incidence of MM to detect an increase in the incidence of MM in in this country. Measurements of normal immunoglobulins in these subjects by laser Olmstead county between 1945 and 1990 to the fact that the Mayo Clinic provides most of the nephelometry provided the surprising obser- medical care in this county. Hematologists and vation that, for unknown reasons, IgG and IgA epidemiologists at the Mayo Clinic have long were significantly higher in the elderly had a major interest in MM. and it seems likely Japanese. The frequency of MGUS in sera sub- that they have succeeded in detecting. and accu- mitted for electrophoresis at the Houston rately diagnosing. almost all of the cases of MM Veterans Administration Hospital was found during the 45-year study period. Table 11. The prevalence of MGUS increases with age Author [Reference] Asclssori et a/. 131 Eiiglisoni er al. [51] Rad1 efal. [ 5 5 ] Age group 30-19 50-69 70-89 290 295 MGUSlTotal 512826 401293 I 191735 5/26 14/73 Percent 0.2 1.4 2.6 19.2 19.2 .- 1 Plasma Cell Neoplasms: Incidence and Epidemiology .-Table 111. The prevalence of MGUS is lower in Japan than in the United States Prevalence of MGL'S in subjects over age 60 in retirement communities Country Japan Normal IpG Normal IgA ,MGUS/Total 4/146 mg/ml .685ir 520 283 t I16 Percent 2.7 United States Normal IgG Normal IgA 11/111 . I 18 ir-102 226 f I I6 10.0 From [IO]. What Causes Plasma Cell Neoplasia? The short answer to this question is that we don't know! Epidemiologists are still in the process of searching for clues. The major tools these detectives use are studies of cohorts. and comparisons of the exposures of cases and controls [ 171. In a cohort study, defined populations exposed to factors suspected of increasing or decreasing the incidence of a cancer. and an unexposed control group (which should be as much like the exposed group as possible) are followed for an interval long enough for a nurnber of cancers to develop. The incidence of cancer in the exposed and control groups is then compared. A cohort study is often the first broad-range attempt to detect an exposure that may be involved in the pathogenesis of a cancer. Suspicious exposures are then investigated in more detail. perhaps with the aid of a hypothesis about the biologic mode of action of the agent in question. These studies require very large numbers of subjects, and prolonged observation, in order to have a reasonable chance of detecting P change in the incidence of cancer following exposure. Cohort studies are not very sensitive. and are rarely specific enough to identify the causative agent. Case-control studies begin with a group of patients with a malignancy e.g.. myeloma. and compare the exposure of these cases to various agents with the exposure of a control group. selgted to match the cases in terms of important confounding variables such as age. race. sex. socioeconomic factors. smoking, etc. The exposure in question is assessed by a questionnaire administered to both the cases and controls. These studies can be done quickly, and require fewer subjects to detect a given level of risk. However, they are dependent on the ability of the cases and controls to remember their exposures equally well. There are also many potential biases in the selection of cases and controls, and in the measurement of exposure. Riedef et al. E91 have summarized all of the reported positive associations between occupational exposures and MM. Agricultural employment (predominantly fanning) is the occupation most frequently associated with MM. Most studies have detected this association. but some, including a recent large ( 1.222 cases and 4.888 controls) case-control study of occupational risk factors for MM in Danish men [ 181. have not. However. none of these studies have been able to identify the aspect of agricultural work. be MultiinpltehMe yUeSloAm, a19M6o8r-tal1it9y8R6ates & 10 -.6*2e- o " I 10 0 1968 19701972 1974 19761978 1980 1982 19841986 Year 1tw Fig. 2. Myeloma mortality rates per 1OO.OOO for four age groups in the USA for 1968-1986 [I?]. Bergsagel j t contact with pesticides or er farmers for M Instead 01 ciations. and i i and case-cont environmentai to first exarni agents. radiat: one time thou; genesis of Mbv' dence that gent stimulation are cell neoplasia. Radiation An exce. observed in so tion, includir workers in n u bomb survivoi radium-dial p; analysis of car bomb survivor thought to pro! ciation betweer of MM. This d lyzed. using o nosed among Hiroshima. wh of the bomb. a estimates of b This new ana excess risk of h Because this ne the earlier posi tality analyses and an earlier i analyzed sever explain the rea change is thc increased follo ferences in dia, to consider onl: the risk estirn analyses to thc mates below 4 on the finding5 Benzene Case repor patients who h zene suggested nd Epidemiology ~~ ts over age 60 in tities Percent 2.7 10.0 kly, and require In level of risk. ,n the ability of nber their expoISO many poten.es and controls, osure. larized all of the etween occupa:ultural employs the occupation MM. Most studtion, but some, cases and 4,888 ccupational risk [181, have not. have been able Aural work, be -tality Rates .a01982 1984 1986 ,*PI 1OO.OOO for four '56[12]. I .A ! Bergsagel it contact with animals, grains. dust. fertilizers, pesticides or engines, which increases the risk of farmers for MM. lnstead of reviewing the rather weak associations, and inconsistent results of most cohort and case-control studies of occupational and environmental risk factors for MM. 1 propose to first examine the role of exposure to two agents, radiation and benzene, which were at one time thought to be important in the pathogenesis of MM. Finally. I will consider the evidence that genetic factors and chronic antigenic stimulation are involved in the genesis of plasma cell neoplasia. Radiation An excess frequency of MM has been observed in some populations exposed to radiation, including American radiologists [ 191, workers in nuclear facilities [20,211, atomic bomb survivors [27], thorotrast patients 1231, radium-dial painters [24] and others [25]. An analysis of causes of death among the atomic bomb survivors in Japan [ 2 2 ] was, for a time. thought to provide strong evidence for an association between radiation dose and the incidence of MM. This database has recently been reanalyzed, using only primary cases of MM diagnosed among residents of Nagasaki and Hiroshima, who were in these cities at the time of the bomb, and had Dosimetry System 1986 estimates of between 0 and 4 Gy kerma (261. This new analysis found no evidence of an excess risk of MM in the atomic bomb survivors. Because this negative result differs sharply from the earlier positive association reported in mortality analyses based on death certificates [27]. and an earlier incidence study [22]. the data was analyzed several different ways, in an effort to explain the reasons for the different result. The change is thought to be due primarily to increased follow-up, adding 12 years. But differences in diagnostic criteria, and the decision to consider only first-primary MM. also affected the risk estimates. The decision to limit the analyses to the cohort with DS86 kerma estimates below 3 Gy did not have a major impact on the findings. Ben:erie Case reports of MM developing in Turkish patients who had been heavily exposed to benzene suggested that this agent might be involved 5 in the pathogenesis of the disease [?ti]. A cohort analysis of workers (predominantly white males) employed in the manufacture of rubber hydrochloride in pliofilm plants, by a process that includes the dissolution of natural rubber in benzene. is one of the best studies of benzene-exposed workers. This cohort has been analyzed on four occasions by the National Institute for Occupational Safety and Health. In the analyses of 1977 and 1981. an'increased risk of acute myelogenous leukemia was recognized. but there was no increase in the incidence of MM [29. 301. In the third'analysis of 1987. four cases of MM were detectid [31]. The expected number of MM deaths in an unexposed population of the same age. sex and race was one. However, three of the workers who developed MM in this study had minimal exposure to benzene (one had worked at the plant for only four days), and there was no trend towards an increasing incidence of MM with greater exposure. A fourth update has added six years of follow-up (321. No new cases of MM have developed. and the standardized mortality ratio for MM now is no longer significantly increased in the benzene-exposed group. Genetic Factors Striking differences in the incidence of monoclonal gammopathies in inbred strains of mice. racial differences in the incidence of MGUS and MM in humans. the association of an increased risk of developing MM with certain human leukocyte antigens (HLA), and the occurrence of familial MGUS. MM and monoclonal macroglobulinemia all suggest that genetic factors play an important role in the pathogenesis of plasma cell neoplasms. Plasma Cell Neoplasia in Inbred Strains of Mice Benign monoclonal pammopathies. closely resembling the MGUS found in humans. develop spontaneously in various inbred strains of mice (Table IV). The C57BL/Ka variety has been studied extensively [33]. A monoclonal gammopathy (usually IgG) can be detected in about three percent of these mice by three months of age. The frequency increases to more than 60 percent of mice who live to 14 months. A characteristic M-protein persists for the life of the mouse, but additional M-proteins can appear in the same mouse. The frequency of 4 Plasma Cell Neoplasms: Incidence and Epidemiology Table IV. Mouse plasma cell neoplasms A. Spontaneous monoclonal gammopathieb [Reference] CBA 1331 BXLB/c [33] SZB [??I C?H [33] CS7BL/Ka [33] Sude [35] B. Mineral-oil induced plabmacytomas [56] BALBlc NZB C57BL DBN2 C3H AL/N Major Ig IgM IgM IgG IgA Frequency 3mo. 14mo. 0 1% 0 t 0% 0 26% 18 mo. 0 36% 3 >604 0 75% 62.54 20-304 Rare Rare Rare Rare multiple M-proteins increases with age. Few of the monoclonal gammopathies progress to become malignant tumors, but Rad1 et al. [34] have shown that a few old mice do develop M-protein-secreting tumors. which can be transplanted by the intravenous injection of bone marrow cells. Most of these transplantable tumors cause osteolytic lesions in the recipients. making this a good mouse model of MM. About 75 percent of homozygous nude mice maintained in a sterile environment develop an M-protein (usually IgG) by 24 months [35]. These mice lack a thymus and are T cell deficient. Nude mice raised in a conventional environment develop a monoclonal IgM at an earlier age. Only 47 percent of the heterozygous littermates of the nude mice developed M-proteins. In contrast, 26 percent of NZB mice. and 36 percent of C3H mice develop an IgM monoclonal gammopathy by 24 months. whereas in DBA and BALBk mice, spontaneous M-proteins are uncommon. These observations indicate that the incidence and type of spontaneous M-protein in inbred strains of mice are influenced strongly by the mouse genotype. The specific genes have not been identified yet. Plasmacytomas can be induced in some strains of mice by the intraperitoneal injection of mineral oil (Table IV). It is of interest that susceptibility to the induction of peritoneal plasma- cytomas does not correlate with the strains that have a high incidence of spontaneous M-proteins. Again. the _renotypeinfluences the frequency of mineral oil-induced plasmacytomas. and the penes involved have not been identified yet. Racial Differences in the Incidence of MGUS and MM The striking differences in the incidence of MGUS and MM in different parts of the world, mentioned earlier, appear to be due to racial rather than environmental differences, since the low incidence of these disorders among the Chinese and Japanese in Asia has traveled with them to the Bay area of California, and to Hawaii. The genes that determine these racial differences have yet to be identified. Certain HLA Types Are Associated with an Increased Risk of M M A large population-based study of 46 black male MM cases (with 88 black male controls) and 85 white male MM cases (with 122 white male controls) has been done to determine whether there is any association of HLAs of Class I (HLA-A, HLA-B, HLA-C) and Class I1 (HLA-DR, HLA-DQ) with the disease [36]. Black MM patients had significantly higher gene frequencies than their controls for Bw65. Cw2 and DRwI4, while white MM cases had higher gene frequencies than controls for A3 and Cw2. and blanks at the DR and DQ loci. The frequency of CwZ in the black and white controls was similar. These findings suggest that the Cw2 allele. or a gene close to the C loci, confers susceptibility to the development of MM, but does not explain the higher risk among blacks. The authors also suggest that undefined Class 11 antigens may play an etiologic role. New molecular techniques. using genomic DNA, may lead to the identification of these alleles. Bergsagel -Familial P l m Several MM and MG -. families rep' added two brs [38]. and ani with IgCt Mh and two broi these -43 fami in seven firstspring), was r relatives (sib more distant r one third-de; fourth-d eg re families witt Renicr et al. [ second-degre The occui nancy in a fa pattern of inh may be expost ard. However. ognized in an have been in! discovery that have inheritel 40,4 1 ] suggci these plasma c lod score (the lihood of link by the LIPED of Waldenst ~ autoimmune segregating H some six. The favoring chro susceptibility Chronic Anti Clinician about the PO> genesis of M!v stimulates a immune syste of the responc era1 attempts t with a past hi terial infecti allergy desens disease: but tt J Epidemiology -requency 1. 21 mo. 18 10% 16% 18 mo. 36% >608 75% 67.58 20-30% Rare Rare Rare Rare lice of MGUS the incidence it parts of the ir to be due to 11 differences. ese disorders se in Asia has a of California, etermine these identified. red with an Idy of 46 black male controls) vith 122 white to determine I n of HLAs of C) and Class I1 disease [36]. tly higher gene or Bw65. Cw? ses had higher r A3 and Cw2. loci. The frewhite controls ,\t that the Cw2 ci. confers susMM. but does ~ i gblacks. The -.dClass I1 antiNew molecuJNA. may lead les. Bergsagel Fnmiliul P l u s m i Cell Neoplusio Several families with multiple cases of M M and MGUS have been reported. To the 41 families reported by Lorh er 01. (371 can be added two brothers who developed IgG/k MM [38]. and another family i n which a female with 1gG MM was found to have two sisters and two brothers with 1gG MGUS [39]. In these 43 families. MM or MGUS was detected in seven first-degree relations (parent and offspring), was most frequent in 32 second-degree relatives (siblings). and much less frequent in more distant relatives. with occurrences in only one third-degree (aunt and niece), and three fourth-degree (cousins) relations. In the 25 families with monoclonal IgM reviewed by Renier et ul. [39]. nine were first-, and 16 were second-degree relations. The Occurrence of multiple cases of a malignancy in a family, without a clear Mendelian pattern of inheritance. suggests that the family may be exposed to the same environmental hazard. However. no hazardous factor has been recognized in any of the M-protein families that have been investigated. On the other hand. the discovery that several affected family members have inherited identical HLA haplotypes [37, 40. 411 suggests that the tendency to develop these plasma cell neoplasms may be inherited. A lod score (the logarithm of the ratio of the likelihood of linkage to no linkage) was calculated by the LIPED program to test if the occurrence of Waldenstrom's macroglobulinemia and autoimmune manifestations in a family was segregating with the HLA region on chromo- some six. The lod score turned out to be 4.86, favoring chromosomal linkage of a postulated susceptibility gene to the HLA complex [41]. Chronic Antigenic Stimulation (CAS) Clinicians have speculated for many years about the possible role of CAS i n the pathogenesis of MM [42]. The hypothesis is that CAS stimulates a proliferative response in the immune system. and that MM develops in one of the responding cells. There have been several attempts to corrdate the occurrence of MM with a past history of exposure to viral or bacterial infections. immunizations. allergies, allergy desensitization therapy and autoimmune disease: but the results are inconsistent [43]. ~ 1: 7 There does. however. appear to be an association between rheumatoid arthritis and MM. At least two follow-up studies of patients with rheumatoid arthritis have detected a subsequent increased incidence of M M [44-46]. and an excess of rheumatoid arthritis has been detected i n case-control studies i n New Zealand and northern Sweden [47. 481. A n examination of the frequency of autoimmune diseases among first-degree relatives of MM patients discovered a significantly increased risk of rheumatoid arthritis, as compared to the incidence in first-degree relatives of the controls [J9]. If this finding is confirmed, it will suggest that genetic factors underlie the association. A case-control study of patients with Gaucher's Disease in Israel found that these patients have an increased risk of developing hematologic malignancies. including MM [50].A possible mechanism is that the accumulating cerebroside acts as a chronic antigenic stimulant of the immune system. as was first suggested by Shoenfeld et al. [5 I]. These investigators found that patients with Gaucher`s Disease had elevated levels of serum immunoglobulins. that increased with age. In addition. Marti et al. [52] detected diffuse, polyclonal hyperglobulinemia in IO. oligoclonal hyperglobulinemia in six. and MGUS in two of 23 patients with Gaucher's Disease. The sequence may be that CAS leads to a polyclonal, followed by an oligoclonal response, from which a monoclonal proliferation emerges. resulting in MGUS. MM. lymphoma or leukemia. The discovery of an HIV- 1-seropositive patient with MM. whose IgG/k M-protein specifically recognized the HIV- I p24 gag anti- gen. suggests that the antigen-driven response to this viral infection did play a role in the pathogenesis of MM in this patient [53]. j i L i I I I i References I Bakkus MHC. Vanrict I. Vancamp B et al. Evidence that the clonogenic cell in multiple-myeloma originates from a pre-switched but \omatically mutated B-cell. Br J Haematol I99487:68-71. 2 Kyle RA. Lust JA. .Monoclonal gammopathics of undetermined significance. In: Wiernik PH. Canellos GP. Kyle RA. Schiffer CA. eds. Neoplastic Diseases of the Blood. Vol I. New 8 Plasma Cell Neoplasms: Incidence and Epidemiology York. Edinburgh. London, Melbourne. Tokyo: Churchill Livingstone. 199I :571-596. 3 Axelsson L.. Bachmann R. Hallen J. Frequency of pathological proteins (M-components)in 6.995 sera from an adult population. Acta Med Scand 1966:1?9:235-247. 4 Axelsson U. Hiillen J. A population study on monoclonal gammopathy: follow-up after 5 1/2 years on 61 subjects detected by electrophoresis of 6995 sera. Acta Med Scand 1972:191:1II-I 13. 5 Turesson I. Zettervall 0. Cuzik J et al. Comparison of trends in the incidence of multiple myeloma in Malmo. Sweden. and other countries. 1950-79.N Engl J Med 1984:310:421-424. 6 Young JL Jr., Percy CL. Asire AJ. eds. Surveillance. epidemiology. and end results: incidence and mortality data. 1973-77. Bethesda: National Institutes of Health. publication 81-2330. 1981:1082. National Cancer Institute Monograph 57 7 Devesa SS. Descriptive epidemiology of multiple myeloma. In: Obrams GI, Potter M, eds. Epidemiology and Biology of Multiple Myeloma. Berlin: Springer-Verlag, 1991:3-12. 8 Sondik EJ. Cancer Statistics Review 1973-1986. Washington. D.C.: US Government Printing Office. I989DHHS publication no. NIH 89-2789. 9 Riedel DA. Pottern LM, Blattner WA. Epidemiologyof multiple myeloma. In: Wiernik PH. Canellos GP, Kyle RA. Schiffer CA. eds. Neoplastic Diseases of the Blood Vol.1. New York. Edinburgh. London, Melbourne, Tokyo: Churchill Livingston. 199I :347-372. IO Bowden M. Crawford J. Cohen HJ et al. A comparative study of monoclonal gammopathiesand immunoglobulin levels in Japanese and United Stateselderly. J Am Geriatr SOC1993:11:11-14. 1 1 Singh J. Dudley AJ. Kulig KA. Increased incidence of monoclonal gammopathy of undetermined significance in blacks and its age-related differences with whites on the basis of a study of 397 men and one woman in a hospital setting. J Lab Clin Med 1990:l 16:785-789. 12 Davis DL. Hoe1 D. Fox J et al. International trends i n cancer mortality i n France. West Germany. Italy. Japan. England and Wales. and the United States. A n n N Y Acad Sci 1990:60L):5 -48. 13 Schwartz J. Multinational trends in multiple myeloma. Ann N Y Acad Sci 1990;6O9:215-224. 14 Linos A. Kyle RA. O'Fallon WM et al. Incidence and secular trend of multiple myeloma in Omstead County. .Minnesota: 1965-77. J Nat Cancer Institute 198I :66:17-20, 15 Levi F. La Vecchia C. Trends in multiple myeloma [letter].Int J Cancer 1990;16:755-756. 16 Kyle R A . Beard CM. O'Fallon WM et al. Incidence of multiple myeloma i n Olmstead County. Minnesota: 1978 through 1990. with a review of the trend since 1945. J Clin Oncol 1994:12:1577-1583. 17 Boyd NF. The epidemiology of cancer: In: Tannock IF. Hill RP. eds. The Basic Science of Oncology (Second Edition). Toronto: Per,oamon Press. 19927-22. 18 Heineman EF. Olsen JH. Pottern LM et al. Occupational risk factors for multiple myeloma among Danish men. Cancer Causes Control 199?:3:555-568. 19 Lewis EB. Leukemia. multiple myeloma and aplastic anemia in American radiologists. Science 1963~421:492-1494. 20 Dolphin GW. A comparison of the observed and expected cancers of the haematopoietic and lymphatic systems among workers at Windscale. London: Her Majesty's Stationery Office. 1976. 21 Gilbert ES. Omohundro E, Buchanan JA et al. Mortality of workers at the Hanford site: 1945-1986. Health Phys 1993:64:577-590. 22 lchimaru M. lshimaru T, Mikami M et al. Multiple myeloma among atomic bomb survivors. Hiroshima and Nagasaki. 1950-76. N Engl J Med 1979:301:439-440. 23 Faber M. Malignancies i n Danish thorotrast patients. Health Phys 1978:35: 153-158. 24 Poledenak AP. Stehney AF, Rowland RE. Mortality among women first employed before 1930 in the U.S. radium dial painting industry. A group ascertainment from employment lists. Am J Epidemiol 1978:107:179-195. 25 Cuzick J. Radiation-induced myelomatosis. N Engl J Med 1981:301:20J-210. 26 Preston DL. Kusumi S. Tomonaga M et ai. Cancer incidence in atomic-bomb survivors. Part 111: Leukemia. lymphoma and multiple-myeloma. 1950-1987. Radiat Res 1994:137(suppl2):S68-S97. 27 Shimizu Y, Kato H. Schull WJ et al. Life Span Study Report I 1. Part I . Comparison of risk cwfficients for site-specificcancer mortality based on the DS86 and T65DR shielded kerma and organ doses. Radiat Res 1989:118:502-524. 28 Aksoy M. Erdem S. Dincol G et al. Some etiologic factors in the etiology of multiple myeloma. A study in 7 patients. Acta Haematol 1984:71:116-120. Bergsagel . 29 Infante P Leukemi 1977:ii:7( 30 Rinsky R benzene v 3 I Rinsky R.? and leukrrN Engl J ?. 32 Paxton \l Leukemia i the pliotilr sure distrit 33 Rad1 I. H ' Idiopathic model - t Journal of 34 Rad1 J . Cr model of h , J Pathol IC) 35 Rad1 J. Mir frequency the sera of Immunopar 36 Pottern L\ multiple m. evidence t Epidemiol 37 Loth TS. t aspects of ! 199I :I56:-l 38 Bizzaro X. tiple myek undeterm i Haematolg. 39 Renier G. If ers with \+ Cancer 198 40 Grosboi.; f Multiple m! chemical 3; Cancer 198t 41 Bliittner Walden\ t ri' mune dihe. 1980:9? :83 42 lsobe T. 0. associated \I 806 cases. A 43 Riedel DA. multiple m! Am 1992:6:. I Epidemiology 1965-77. J Nat Is in multiple '90:46:755-756. on WM et al. a in Olmstead !h 1990. with a . J Clin Oncol of cancer: In: asic Science of into: Pergamon tern LM et al. ltiple myeloma ausrs Control ' myeloma and ilogists. Science f the observed natopoietic and *sat Windscale. 'y Office, 1976. hanan JA et al. Hanford site: L:577-590. kami M et al. bomb survivors, 5. N Engl J Med Rish thorotrast 53- 158. Rowland RE. nployed before inting industry. iployment lists. 95. yelomatosis. N I M et ai. Cancer ivors. Part Ill: .iple-myeloma. ~ppl2):S68-S97. t t al. Life Span >onof risk coefIrtality based on erma and organ -524. t al. Some etioiltiple myeloma. cta Haematol Berpsagel 29 Infante PF. Rinsky RA. Wagoner J K et ai. Leukemia in benzene workers. Lancet 1977:ii:76-78. 30 Rinsky RA. Young RJ. Smith AB. Leukemia in benzene workers. Am J Ind Med I98 1 2 2 17-245. 3 I Rinsky RA. Smith AB. Hornung R et PI.Benzene and leukemia: an epidemiologic risk asbessment. N Engl J Med 1987:316:1014-1050. 32 Paxton MB. Chinchilli VM. Brett SM et al. Leukemia risk associated with benzene exposure in the pliofilm cohort: 1. Mortality update and exposure distribution. Risk Anal 199414:147-154. 33 Radl J. Hollander CF. Van den Berg P et al. Idiopathic paraproteinemia. I. Studies in an animal model - the aging C57BLIKaLwRij mouse. Journal of Exp lmmunol 1978:33:395-402. 34 Radl J. Croese JW. Zurcher C et al. Animal model of human disease. Multiple myeloma. Am J Pathol 1988:1 3 3 9 3 - 5 9 7 . 35 Rad1 J. Mink JG. van den Berg P et al. Increased frequency of homogeneous immunoglobulins in the sera of nude mice with age. Clin lmmunol Immunopathol 1980:17:469-476. 36 Pottern LM. Gart JJ. Nam JM et al. HLA and multiple myeloma among black and white men: evidence of a genetic association. Cancer Epidemiol Biomarkers Prev 1992:1:177- 182. 37 Loth TS, Perrotta AL, Lima J et ai. Genetic aspects of familial multiple myeloma. Mil Med 1991 :156:430-133. 38 Bizzaro N,Pasini P. Familial occurrence of multiple myeloma and monoclonal gammopathy of undetermined significance in 5 siblings. Haematolgica 1990:75:58-63. 39 Renier G. lfrah N. Chevailler A et al. Four brothers with Waldenstrom's macroglobulinemia. Cancer 1989:64:1554-1559. 40Grosbois F, Gueguen M. Fauchet R et al. Multiple myeloma in two brothers. An immunochemical and immunogenetic familial study. Cancer 1986:58:2417-2421. 41 Blattner WA. Garber JE. Mann DL et al. Waldenstrom's macroglobulinemia and autoimmune disease in a family. Ann Intern Med 1980:93:830-832. 42 Isobe T. Osserman EF. Pathologic conditions associated with plasma cell dyscrasia: a study of 806 cases. Ann NY Acad Sci 197 I :90:507-518. 43 Riedel DA. Pottern LM. The epidemiology of multiple myeloma. Hematol Oncol Clin North Am 1992;6:225-247. 9 44 lsomaki HA. Hakulinen T. Joutsenlahti L'. Excess risk of lymphomas. leukemia and myeloma i n patients with rheumatoid arthritis. J Chron Dix I978:3I 169 1-696. 45 Katusic S. Beard CM. Kurlund LT et al. Occurrence of neoplasms i n the Rochester. Minnesota. rheumatoid arthritis cohort. Am J Med 1985:78(1A):50-55. 46 Hakulinen T. lsomaki H. Knekt P. Rheumatoid arthritis and cancer studies based on linking nationwide registries i n Finland. Am J Med 1985:78(lA):29-32. 47 Pearce NE. Smith AH. Howard-JK et al. Case-control study of multiple myeloma and farming. Br J Cancer 1986:5-1:193-500. 48 Eriksson M. Rheumatoid arthritis as P risk-factor for multiple myeloma: a case-control study. Eur J Cancer 1993:29A:259-263. 49 Linet MS. McLaughlin J K . Harlow SD et al. Family history of autoimmune disorders and cancer in multiple myeloma. I n t J Epidemiol 1988:17:512-513. 50 Shiran A. Brenner B. Laor A et al. Increased risk of cancer in patients with Gaucher's disease. Cancer 1993:72:219-224. 51 Shoenfeld Y. Gallant LA. Shaklai M et al. Gaucher's disease: a disease with chronic stimulation of the immune system. Arch Pathol Lab Med 1982;106:388-391. 52 Marti GE. Ryan ET, Papadopoulas NM et al. Polyclonal B-cell lymphocytosis and hyperglobulinemia in patients with Gaucher disease. Am J Hematol 1988:29:189-194. 53 Konrad RJ. Kricka LJ. Goodman DRP et al. Brief report: myeloma-associated paraprotein directed against the HIV-I p24 antigen in an HIV-1-seropositive patient. N Engl J Med 1993:328:1817-1819. 54 Englisova M. Englis M. Kyral V et al. Changes in immunoglobulin synthesis in old people. Exp Geront 1968:3:125-127. 55 Radl J, Sepers JM, Skvaril F et al. Immunoglobulin patterns in humans over 95 years of age. Clin Exp Immunol 1975:22:84-90. 56 Potter M. Experimental plasma cell tumors and other immunoglobulin-producing lymphoreticular neoplasms in mice. In: Azar HA. Potter M. eds. Multiple Myeloma and Related Disorders VoPl. Hagerstown. Maryland: Harper & Ron. 1973:153-194.