evaluation of monoclonal gammopathies in the “well” elderly

Evaluation of Monoclonal Gammopathies in the “Well” Elderly

JEFFREY CRAWFORD, M.D. MARY KATHLEEN EYE, B.S. HARVEY JAY COHEN, M.D.

Durham. North Carolina

From the Department of Medicine and Geriatric Research, Education, and Clinical Center, Veter- ans Administration Medical Center, and the Divi- sion of Geriatric and Hematology/Oncology, De- partment of Medicine, and the Center for the Study of Aging and Human Development, Duke Universitv Medical Center. Durham. North Caro- lina. This-work was support in part by the Geriat- ric Fellowship Program and the General Medi- cine Research Service of the Veterans Adminis- tration. Requests for reprints should be addressed to Dr. Jeffrey Crawford, Gero-oncology Research, Building 10 (182A), Veteran Ad- ministration Medical Center, 508 Fulton Street, Durham, North Carolina 27705. Manuscript sub- mitted January 14, 1986, and accepted February 14, 1986.

The study of monoclonal gammopathies in the elderly provides an opportunity to define immunologic and neoplastic changes with aging. Previous reports using paper and cellulose acetate electrophoresis have documented an age-related increase in monoclonal gammopathies. In this study, the more sensitive techniques of high-resolution agarose gel electrophoresis and immunofixation were used, in con- junction with other protein studies, to further evaluate the prevalence of monoclonal gammopathies in 111 ambulatory residents (aged 62 to 95) of a retirement home. Eleven of the 111 residents (10 percent) were found to have a monoclonal gammopathy, ranging in concentration from 0.2 to 1.6 g/dl. All monoclonal gammopathies were confirmed by immunofixation, which also documented the presence of additional unsuspected monoclonal components in three of the 11 residents. The prevalence of monoclonal gammopathies by age ranged from 6 percent in the group younger than 60 years of age to 14 percent in the group older than 90 years of age. Only one of the 11 residents had any clinical or routine laboratory suggestion of a monoclonal gammopathy. The other IO had normal ratios of total protein and albumin to globulin. Five of the 11 (45 percent) had an otherwise clinically unexplained erythrocyte sedimentation rate of more than 20 mm/hour, compared with only two of 100 in the group without monoclonal gammopathies. Follow-up studies one to three years after initial evaluation revealed that five of the 11 patients had died, two with evidence of disease progression. In the other six patients, monoclonal protein concentration and other protein values remained stable. An unexplained elevation of the erythrocyte sedimentation rate in the elderly warrants investigation for the presence of a monoclonal gammopathy. Agarose gel electrophoresis and immunofixation identify a higher percent of monoclonal gammopathies in the elderly than has previously been recognized. Identification of monoclonal components in this population is useful for the subsequent study of plasma cell dyscrasias, neoplastic disease, or other immune dysfunction in the aged.

Multiple myeloma is one of the most common hematologic malignancies that show a marked increase in incidence with age [ 11. It represents one part of the spectrum of monoclonal gammopathies that includes the much larger category of idiopathic monoclonal gammopathies or monoclonal gammopathies of undetermined significance, which are also age- related [2]. Previous studies using paper electrophoresis (and cellulose acetate electrophoresis) have suggested that the prevalence of these disorders is one percent over the age of 65 and 5 percent over the age of

January 1987 The American Journal of Medicine Volume 82 39

MONOCLONAL GAMMOPATHIES IN THE ELDERLY-CRAWFORD ET AL

80 [3]. Longitudinal follow-up studies in these patients suggest a significant incidence of development of other plasma cell dyscrasias, including multiple myeloma, Wal- denstrom’s macroglobulinemia, and amyloidosis, as well as lymphoproliferative disorders such as chronic lympho- cytic leukemia and lymphoma. In one series of 241 patients with monoclonal gammopathies of undetermined significance, a lyrnphoproliferative or plasma cell disorder subsequently developed in one third of the group within 15 years of the time of diagnosis [4]. In addition, these monoclona! gammopathies are frequently associated with other forms of neoplastic disease, although the biologic relationship between these two disorders is less clear [5].

The presence of a monbclonal gammopathy in the elderly represents one opportunity to explore the complex relationshtps of changing immune function and risk of development of neoplastic disease in the aged. The prev- atence of monoclonal gammopathies in the elderly suggest they are one of the most common age-related laboratory abnormalities with important clinical implications. However, studies in the “well” elderly have been limited and used less ‘sensitive techniques than are currently available. Therefore, to better define the prevalence of monoclonal proteins in an elderly population using more recently available methods of protein analysis, and to explore the relationship of monoclonal proteins to clinical course in this population, we studied 111 residents of defined health status living in a retirement community.

SUBJECTS AND METHODS

Residents of the Methodist Retirement Home, affiliated with the Durham Veterans Administration and Duke Medical Centers, undergo a voluntary annual physical examination and laboratory screening, including complete blood cell counts and automated chemical profiles. At the time of this evaluation, protein electrophoresis, by high-resolution agarose gel, with correlative immunofixation for suspected monoclonal gammopathies, and determination of quantitative immunoglobulin levels, plasma viscosity, and erythrocyte sedimentation rates by both Wintrobe and Westergren methods were performed in our laboratory. This report presents the results in those residents who where.found to have monoclonal gammopathies. Details of protein studies in the rest of the population are reviewed elsewhere [6]. Agarose Gel Electrophoresis. High-resolution agarose gel electrophoresis of serum samples was performed with the Panagel system (Worthington DiagnosticIs, Cooper Bio- medical; Malvern, Pennsylvania). This system utilizes a cooling block in the migration unit enabling the application of 206 volts across the gel bed. The combination of high voltage and a modified barbiturate buffer (pH 8.6) facilitates separation of the serum into at least 11 different protein bands [7]. Optimal results in our laboratory involved application of’ 15 pl of serum for three minutes, followed by blotting, followed by an electrophoresis time,of 40 minutes at 200 volts. The gels were then fixed and stained with

amido black solution. Preliminary identification of monoclonal bands in the beta and gamma regions was made by qualitative examination. lmmunofixation Electrophoresis. The confirmation of possible monoclonal or oligoclonal gammopathies was subsequently accomplished by immunofixatioti techniques [8]. Serum samples and appropriate dilutions of these samples were run simultaneously using the high-resolution electrophoresis equipment previously described. All sample dilutions for immunofixation electropboresis were made with 0.9 percent saline solution to adjust immunoglobulin concentrations to 20 to 25 mg/dl. Following electrophoresis, the upper portion of the gel with the serum sample lane was removed and stained as described previously. The remaining sample lanes of the gel were then overlayed with antiserum-saturated strips. These were prepared by satu- rating individual cellulose acetate strips, 1 cm by 5 cm (Sepharose Ill, Gelman Sciences; Ann Arbor, Michigan) with I:2 dilutions of different antiserum samples including goat antihuman IgG, IgA, and IgM and burro antihuman kappa and lambda free and bound light chains (ICL Scientif- ic; Fountain Valley, California). Additional identification of light chains of some smaller monoclonal components was made with high-titer Dako antiserum (Accurate Chemical and Scientific; Westbury, New York). After a two-hour incu- bation in a covered humidity chamber, the gels were washed in saline overnight’and then stained with Coomas- sie blue R-250 (Eastman Kodak; Rochester, New York). By comparison with the routinely stained agarose gels, the suspected monoclonal components could be definitively identified as to heavy and light chain type by this method. Densitometer Correlation. Sampies definitively identified as having monoclonal or oligocolonal components were subsequently run using the high-resolution Paragon SPE II agarose gel electrophoresis system (Beckman lmmuno Systems; Brea, California), using the described procedure [9]. Following processing and staining, densitometer tracings of these gels were made to quantitate individual protein fractions, using a Beckman CDS-200 densitometer. For this purpose, densitometer tracings identified albumin, alpha-l, alpha-2, beta, gamma, and monoclonal protein regions. Cellulose Acetate Electrophoresis. Residents with documented monoclonal’gammopathies by agarose gel electrophoresis underwent standard serum protein electrophoresis as performed in the Duke Clinical Special Hematology Laboratory using standard cellulose acetate electrophoresis (Beckman Instruments; Fullerton, California) [lo]. Cel- lulose acetate electrophoresis gels were scanned vtsually and by densitometer as just described and correlated with other agarose gels for the same residents. Quantitative lmmunoglobulin Levels.. Serum IgG, IgA, and IgM were individuatly quantitated using a Hyland PDQ laser nephelometer (Hyland Diagnostics; Deerfield, Illinois) accompanied by the ICL-AIM kinetic software for rate nephelometric determinations (ICL Scientific) and a Hew- lett-Packard HP9815A programmable calculator. Samples of nephelometric-grade goat antiserum to human .IgA, IgG, and IgM were obtained from ICL Scientific. These proce- dures are described thoroughly elsewhere [ 1 I]. Briefly, however, the reaction kinetics of antigen-antibody complex

40 January 1987 The American Journal of Medicine Volume 82


formation are detected by the forward scatter of laser light and calculated in rate-of-change units. Antiserum to IgG, IgA, or IgM was added to a polyethylene glycol buffer and equilibrated for one hour. Aliquots of serum (1 to 10 ~1) were added, and the reaction kinetics of antigen-antibody complex formation were measured by the nephelometer as forward light scatter and expressed as rate-of-change units over one minute. For each assay, a standard curve for IgG, IgA, and IgM was generated to convert rate-of-change units to mg/dl for each sample. Five-point standard curves al- ways yielded a correlation coefficient greater than 0.95, and standardized control samples (Hyland Diagnostics) were included to assure day-to-day consistency of results. The normal ranges for this assay are IgG, 700 to 1,700 mg/ dl; IgA, 70 to 350 mg/dl; and IgM, 70 to 210 mg/dl (ICL Scientific). Erythrocyte Sedimentation Rates. Samples of venous blood were collected in potassium EDTA tubes (Vacutainer Systems; Rutherford, New Jersey) for determinations of erythrocyte sedimentation rates. The Wintrobe and Modi- fied Westergren methods were performed simultaneously on the same day the samples were obtained. In the Win- trobe methods [ 121, blood was pipeted directly into 110 mm tubes calibrated from 0 to 100 mm (American Dade Company; Miami, Florida). In the modified Westergren method [12], the “dispette” method was used. This in- volves disposable 200 mm tubes graduated from 0 to 150 mm (Ulster Scientific; Highland, New York). Before the tubes were filled, 1.2 ml of blood was diluted with 0.5 ml of 0.9 g/dl saline solution. In both methods, the tubes were placed in a vertical rack and left undisturbed at room temperature. At the end of one hour, the distance from the bottom of the surface meniscus to the top of the column of red cells was recorded in mm/hour as the erythrocyte sedimentation rate. Duplicate determinations were within 1 to 3 mm/hour in all cases. Plasma Viscosity. The remainder of the EDTA-anticoagu- lated blood not used for the erythrocyte sedimentation rate determination was separated by centrifugation at 2,000 rpms (800 x g) for 20 minutes, and the plasma layer was removed. The plasma was kept at room temperature prior to the viscosity measurement, with results reproducible and consistent for up to two weeks. Plasma viscosity was measured in duplicate using an automated capillary viscometer (Coulter Electronic Limited; Luton, England), as previously described [ 131. Briefly, 0.5 ml sample volumes were equilibrated in the viscometer at 25’C, and the flow was elec- tronically timed and compared with the flow time of 3.6 g/d1 saline solution of known viscosity, resulting in the calcula- tion of an absolute viscosity expressed in centipoise (cp). The normal range for both males and females .with this instrument is 1.50 to 1.72 cp, with a mean of 1.64 f 0.05 at 25“C; adjustment for age is not necessary [ 141. Analysis. All clinical information was obtained by thorough chart review and discussion of individual laboratory findings with the care givers of the Methodist Retirement Home. All identifying clinical information and laboratory results were entered on the Duke Cancer Center computer. Secondary analysis used the time-oriented record for oncology (TORO) system [ 151.

TABLE I Age Distribution and Age-Specific Prevalence of Monoclonal Gammopathies in Retirement Home Population

Age Total (years) Number

Monoclonal Gammopathies Number Percent of Age Group _

60-69 5 0 70-79 27 2 80-89 65 7 10.7

290 14 2 14.2 Total 111 11 10.0

RESULTS

The population studied included 111 ambulatory residents (18 men and 93 women) with an age range of 62 to 95 and a median age of 82. Ninety percent of the population was over age 75. Over a three-year period, all residents whose blood samples were sent to our laboratory at the time of their annual examination were included- in the study. At the conclusion of the study, all charts were reviewed to correlate clinical diagnosis and laboratory measurements. Thus, data were available from the time of the initial laboratory sampling, as well as follow-up information ranging from three months to 37 months, with a median follow-up time of 16 months: This provided a significant follow-up period to observe the clinical course of the population with monoclonal gammopathies as well as the population without morioclonal gammopathies. Of the 11 patients with documented monoclonal’ gammop: athies, four had died. The other sevenunderwent follow- up laboratory evaluation of their monoclonal gammopathies. Prevalence of Monoclonal Garnmopathies. Table I outlines the age distribution for the study population as well as the age-specific prevalence of monpclonai gammopathies, with the range from 7.4 percent for the age group 70 to 79 years (or 6.2 ‘percent for the group younger than 80 years) to 14.2 percent for the population over the age of 90 years. The overall prevafence’in this population was 19 percent or 11 of 111 residents. Immunologic Evaluation of Monoclongl Gammopathies. Table II documents the monoclonal protein concentration by densitometer tracing of agarose gels as well as the identjfication of the immunoglobulin type and quantitative immunoglobulin levels for the residents with identified monoclonal gammopathies. The monoclonal protein concentration ranged fr.om 0.2 to lb8 g/dl. Eight of the II residents had, IgG lambda components, and three had IgG kappa components, defined by immunofixation, arevet? sa! of the usual 2: 1 ratio of kappa to lambda. Additionally, immunofixation detected additional monoclonal components that were not initially suspected by standard agarose gel electrophoresis in three of the 11 residents. These



TABLE II Monoclonal Protein Concentrations, lmmunoglobulin Types, and Serum Quantitative lmmunoglobulin Levels for Residents with Monoclonal Gammopathies

Subject

1 2 3 4 5 6 7 8 9

10 11

hlonoclonal Protein Concentration (gldl)

0.2 0.3 0.4 0.4 0.4 0.6 0.6 0.8 0.8 1.5 1.8

Quantitative lmmunoglobulins (mgldl)

lmmunoglobulin Type(s) IgG I@ w

IgG lambda 1,469 199 83 IgG lambda 720 111 58 IgG lambda 1,186 286 70 IgG kappa, IgG kappa 1,326 416 71 IgG lambda 876 162 221 IgG kappa 1,089 122 ,53 IgG kappa, IgA lambda 733 256 46 IgG lambda, IgA lambda, IgA lambda 1,289 153 76 IgG lambda 1,670 183 79 IgG lambda 2,656 52 57 IgG lambda 2,196 86 55

TABLE Ill Other Identlfying Clinical and Laboratory Information for Residents with Monoclonal Gammopathies

Serum Total Serum

Erythrocyte Sedimentation Rate Plasma

Subject Age/Sex Protein Albumin Hematocrit (mm/hour) Viscosity WI) WI) (percent) Wintrobe Westergren (cp)

1 86/M 7.4 4.6 37 28 12 1.66 2 78/F 6.8 4.8 41 15 12 1.50 3 84/F 6.6 4.0 40 10 2 1.48 4 72/F 7.4 4.6 43 30 31 1.70 5 89/F 6.7 4.8 35 48 74 1.72 6 85/M 7.0 4.8 41 25 15 1.65 7 81/F 6.4 4.3 40 32 28 1.55 8 84/F 7.6 4.8 40 16 18 1.66 9 84/F 7.5 4.2 48 36 30 1.91

10 92/M 8.1 4.5 45 23 11 1.81 11 94/F 7.5 3.7 40 50 84 1.86

included a second IgG kappa component in Subject 4, an IgA lambda component (associated with IgG kappa) in Subject 7, and two additional IgA lambda components in Subject 8. Serum samples from 20 other residents who did not have visible monoclonal components by agarose gel electrophoresis were chosen randomly and studied further by immunofixation. No unsuspected monoclonal components were detected.

Subsequent cellulose acetate electrophoresis readily identified monoclonal components in six of the 11 residents (Subjects 2, 6, 8 9, 10, and 11). Thus, the only monoclonal component clearly detectable by cellulose acetate electrophoresis at a concentration of less than 0.6 mg/dl was in Subject 2, for whom identification was aided by accompanying borderline low quantitative immunoglobulin levels. The pattern of immunoglobulin values outlined in Table II showed some correlation of IgG levels at higher monoclonal protein concentrations, but otherwise showed no specific pattern with reference to IgA and

IgM levels. lmmunoglobulin levels were not helpful in the prediction of the subsequent clinical course in these patients. Other Laboratory Features of the Group with Monoclonal Gammopathies. Table Ill outlines other identifying clinical and laboratory information for this group with monoclonal gammopathies. The sex distribution of three males and eight females was consistent with the sex distribution of the overall study population. Total serum protein levels were within normal limits for all residents, as were serum albumin levels. Subject 11, however, did show a reversal of the ratio of albumin to globulin. Only two of 11 residents had hematocrit values of less than 40 percent. The erythrocyte sedimentation rate was more than 20 mm/hour in seven of the 11 residents by the Wintrobe method and in five of the 11 residents by the Westergren method. There was no clear correlation of the size of the monoclonal component with the degree of elevation of the erythrocyte sedimentation rate. However, in the group with mono-


clonal gammopathies, an elevated Westergren etythrocyte sedimentation rate was otherwise unexplained in five residents, or 45 percent of the group. In contrast, in the group without monoclonal gammopathies, only two of 100 residents had elevations of the erythrocyte sedimentation rate to more than 20 mm/hour (Westergren method) without an obvious clinical cause for the elevation. Plasma viscosity was increased (more than 1.72 cp) only in the three residents with the largest monoclonal components and, therefore, was not useful in screening for the presence of a monoclonal component. Clinical Correlation and Follow-Up. At the time of initial study, no clinical conditions relevant to the presence of a monoclonal gammopathy were noted in these subjects, and no specific therapy was initiated. Table IV outlines clinical associations noted during follow-up study of the residents with monoclonal gammopathies over one to three years from initial detection by agarose gel electrophoresis. One resident died of an unrelated myocardial infarction. Two residents died with the clinical diagnoses of renal and pancreatic carcinoma, neither confirmed histologically due to the patients’ and families’ desire to avoid hospitalization. Subject 3 had a flank mass and hematuria several months prior to death. Subject 10 had obstructive jaundice with a palpable epigastric mass one month prior to death. One resident died from dementia, but a pathologic lytic fracture of the humerus, presumably related to myeloma, developed prior to her death. Another resident had marked progression of the monclonal component from 0.8 to 3.3 g/dl over three years, along with the development of an amyotrophic lateral sclerosis-like disorder, and subsequently died. In the other six residents, the monoclonal components were all m-identified by agarose gel electrophoresis between one and two years after initial evaluation, without a significant change in the size of the component, or the development of clinical consequences suggesting an evolution of a plasma cell dyscrasia or lymphoproliferative disorder.

COMMENTS

The studies of the prevalence of monoclonal gammopathies in the elderly were pioneered by the thorough evaluation of an elderly population in Malmo, Sweden, by paper electrophoresis in 1963 [ 161, in which 3.1 percent of patients with a mean age of 80 were noted to have monoclonal components present. Using the same tech- nique, subsequent evaluation of more than 2,000 people in Sweden older than 60 years demonstrated a prevalence of 1.7 percent for the group 60 to 69 years of age, rising to 5.7 percent for the group 80 to 89 years of age c171.

Subsequent studies of an elderly population of approxi- mately 750 residents over the age of 60 in Olmstead County, Minnesota, were performed using cellulose acetate electrophoresis. The prevalence of monoclonal gam-


TABLE IV Clinical Associations and Follow-Up Information for Residents with Monoclonal Gammopathies (one to three years from initial detection)

Subjecl Clinical Correlation/Follow-Up

1 Died of myocardial infarction 2 Stable M component 3 Died of probable renal carcinoma 4 Stable M component 5 Stable M component 6 Stable M component 7 Stable M component 8 Increase of M component 0.8 to 3.3 g/dl,

development of ALS-like disorder, death 9 Died of dementia with pathologic

fracture of humerus 10 Died of probable pancreatic carcinoma 11 Stable M component

mopathies in this group was 0.3 percent for the group 60 to 69 years of age, increasing to 4.8 percent for the group older than 80 years [IS]. A subsequent study from France of 12,000 patients over the age of 60 resulted in a similar prevalence of 1.6 percent for the age group 60 to 69 and 4.2 percent for the age group over 80 [ 191.

Compared with cellulose acetate systems, high-resolution agarose gel electrophoresis achieves better separation of proteins, with an enhanced sensitivity for the detection of small monoclonal or oligoclonal components [7,9]. One study, using agarose gel electrophoresis, detected 5 percent clonal bands in healthy adult donors aged 22 to 65 years, compared with only 0.3 percent by cellulose acetate [20]. However, 28 of 30 bands were faint by visual examination, and studies to prove these were of monoclonal origin were not done. In this regard, immunofixation electrophoresis has been shown to have an enhanced sensitivity for the documentation of these small monoclonal components compared with standard immunoelectrophoresis [2 1,221.

This study represents the first attempt to define the prevalence of monoclonal gammopathies in a “well” elderly population using these sensitive techniques of protein electrophoresis. In comparison with cellulose acetate electrophoresis in the same population, these techniques appear to enhance the ability to detect monoclonal components in the elderly. Moreover, data from this one small series suggest an age-related prevalence of monoclonal gammopathies at least twice that generally identified by prior techniques and reported in the literature. In addition, multiple components were detected in three of the 11 patients with monoclonal gammopathies-a fea- ture not previously well recognized in age-related monoclonal gammopathies.

In addition, this study undertook a more detailed analysis of plasma proteins to assess if these were useful



discriminant features between the subjects with monoclonal gammopathies and others. Quantitative immunoglobulin levels were not helpful in differentiating residents with or without monoclonal gammopathies. Total serum protein levels were normal in all 11 patients and only one had reversal of the ratio of albumin to globulin, which has often been used as a clue to the presence of a monoclonal or polyclonal gammopathy. Likewise, only two of 11 residents with monoclonal gammopathies had hematocrit values of less than 40 percent. Plasma viscosity measurements were abnormal only in the three residents with the highest monoclonal protein concentrations, limit- ing the usefulness of this test in screening for the presence of a monoclonal gammopathy in this population. Etythrocyte sedimentation rates were frequently abnormal in the patients with monoclonal gammopathies but without clear relation to the size of the monoclonal component, suggesting variability in the interaction of individual monoclonal immunoglobulins with red cells to create rouleaux formation. Of interest was the fact that five of 11 (45 percent) of the patients with monoclonal gammopathies had otherwise unexplained increases in the erythrocyte sedimentation rate; in contrast, in the population without monoclonal gammopathies, only two of 100 (two percent) had otherwise clinically unexplained erythrocyte sedimentation rates. Therefore, the presence of an occult monoclonal gammopathy should be considered as one possible cause of an otherwise unexplained increase in the erythrocyte sedimentation rate in the elderly.

Further chart review of the residents with monoclonal gammopathies versus those without monoclonal gammopathies failed to identify any features allowing clinical recognition of the presence of a monoclonal gammopathy. Thus, despite thorough clinical evaluation and routine laboratory testing, no test short of protein electrophoresis was consistently useful in the identification of the presence of a monoclonal gammopathy.

Our current follow-up population is too small to know whether or not persons with monoclonal gammopathies detected by agarose gel electrophoresis (but not cellulose acetate electrophoresis) have the same risk of subsequent clinical progression or evolution as do persons in whom monoclonal gammopathies are defined by less sensitive techniques. These previous studies have suggested that development of a plasma cell or lymphoproliferative disorder may occur in 20 to 28 percent, followed for three to 14 years in one study [23], and for more than 10 years in the second [4]. Preliminary follow-up would suggest that these clinical consequences may be similar in our population. Prompt identification of such disease progression may allow early intervention with chemo- therapy, which, in the case of multiple myeloma, may be as effective and no more toxic in the elderly than it is in the young [24,25]. It would be premature, however, to recommend routine screening of elderly populations by

the techniques outlined until the long-term utility and cost/ benefit ratio of such an undertaking are known.

Detailed protein evaluation of the elderly, as performed in this study, suggests that monoclonal gammopathies may be one of the most prevalent immunologic markers of aging. Radl et al’s [26] studies of 73 persons all over the age of 95 demonstrated a prevalence of monoclonal gammopathies of 19 percent, defined by agar electrophoresis and immunoelectrophoresis. Because of the prior reported prevalence of 5 percent in persons older than 80 [17-191, a rapid increase in monoclonal gammopathies over the age of 90 has been postulated [27]. Our results, using sensitive techniques more comparable to Radl’s, would suggest that this age-related increase in monoclonal gammopathies may be more linear. Such a linear mode of increase is paralleled by the findings in several strains of mice in which age-related monoclonal gammopathies may reach a prevalence of 60 to 70 percent, with the increase beginning in middle age [27].

The explanation for this rather dramatic increase in monoclonal gammopathies with increasing age is un- known, but may relate to more general age-related de- fects in the immune system. In both humans and C57 black mice, benign monoclonal gammopathies develop 100 to 200 times more frequently than gammopathies associated with full malignant proliferation, such as multiple myeloma or lymphoma. This parallels the findings of increasing antibodies to a number of antigens (thyroid, red blood cells) in older persons, associated with a less striking increase in autoimmune disease in this population [3]. The emergence of such monoclonal immunoglobulins as we have described may be an indicator of a fundamen- tal dysregulation of the immune system occurring with age, such as decreased T cell control of B cell function [27,28]. For example, thymectomy in young C57 black mice is associated with an increase in monoclonal components during subsequent aging [29]. In humans, 30 percent of renal transplant recipients subjected to immu- nosuppression demonstrate at least transient monoclonal components, which increase in frequency with age over .the 20- to 60-year age range of the population [30]. Although the numbers in our study are small, the imbal- ance in light chain types, with lambda predominance in these older subjects, not previously described, may also be indicative of such dysregulation. Further studies of monoclonal gammopathies are needed to provide a larger population in which to explore the relation between this disorder and immune dysregulation, as well as the emergence of lymphoproliferative and plasma cell dyscrasias in the elderly.

ACKNOWLEDGMENT

We thank Judy Robertson for her assistance with the computerized evaluation of the database, and Mildred Kennedy and the staff of the Methodist Retirement Home.


2.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

Pottern LM, Blattner WA: Etiology and epidemiology of multiple myeloma and related disorders. In: Wiernik P, Canel- 10s G, Kyle R, Schiffner C, eds. Neoplastic diseases of the blood, vol 1. New York: Churchill Livingstone, 1985; 413- 429.

Kyle RA: Monoclonal gammopathy of undetermined significance-natural history in 241 cases. Am J Med 1978; 64: 814-826.

Cohen HJ: Multiple myeloma in the elderly. Clin Geriatr Med 1985; 1: 827-855.

Kyle RA: “Benign” monoclonal gammopathy - misnomer? JAMA 1984; 251: 1949-1854.

Crawford J, Cohen HJ: An approach to monoclonal gammopathies in the elderly. Geriatrics 1982; 37: 97-l 12.

Crawford J, Eye MK, Cohen HJ: The clinical utility of erythrocyte sedimentation rate and plasma protein analysis in the elderly. Am J Med (in press).


REFERENCES

Jeppsson JO, Laureel CB, Franzen B: Agarose gel electro- ohoresis. Clin Chem 1979: 25: 629-638.

Sun T: lmmunofixation electrophoresis. In: Ritzman S, ed. Prote in abnormalities, vol I. Physiology of immunoglobulins: diagnostic and clinical aspects. New York: Alan R. Liss, 1982; 97-118.

Jolitt CR: Classification and interpretation of Paragon serum protein electrophoresis patterns. Brea, California: Beck- man, 1983; l-30.

Ritzman SE, Daniels JC: Serum electrophoresis and total serum proteins. In: Ritzman S, Daniels J, eds. Serum protein abnormalities: diagnostic and clinical aspects. Boston: Little, Brown, 1975; 3-26.

Ritchie RF, ed: Automated immunoanalysis, part 2. New York: Marcel Dekker, 1978.

Henry JB, ed: Clinical diagnosis and management by laboratory methods. Philadelphia: WB Saunders, 1979.

Crawford J, Cox EB, Cohen HJ: Evaluation of hyperviscosity in monoclonal gammopathies. Am J Med 1985; 79: 13- 22.

Coulter Harkness Viscometer Instruction Manual. Luton, En- gland: Coulter Electronics Limited, Issue B, 1982.

Cox EB, Stanley WE: Schemia-driven time-oriented record system on a minicomputer. Comput Biomed Res 1979; 12: 503-515.

Hallen J: Frequency of “abnormal” serum globulins (M- components) in the aged. Acta Med Stand 1963; 173: 737-744.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

Axelsson U, Bachmann R, Hallen J: Frequency of pathologi- cal proteins (M-components) in 6,995 sera from an adult population. Acta Med Stand 1966; 179: 235-247.

Kyle RA, Finkelstein S, Elveback LR, Kurland LT: Incidence of monoclonal proteins in a Minnesota community with a cluster of multiple myeloma. Blood 1972; 40: 719-724.

Fine JM, Lambin P, Leroux P: Frequency of monoclonal gammopathy (M-components) in 13,400 sera from blood donors. Box Sang 1972; 23: 336-343.

Papadopoulos NM, Elin RJ, Wilson DM: Incidence of gamma globulin banding in a healthy population by high-resolution electrophoresis. Clin Chem 1982; 28: 707-708.

Ritchie RF, Smith R: Immunofixation. Ill. Application to the study of monoclonal proteins. Clin Chem 1976; 22: 1982- 1985.

Sun T, Lein YY, Degnan T: Study of gammopathies with immunofixation electrophoresis. Am J Clin Pathol 1979; 72: 5-11.

Fine JM, Lambin P, Muller JY: The evolution of asymptomat- ic monoclonal gammopathies. Acta Med Stand 1979; 205: 339-341.

Cohen HJ, Silberman HR, Forman W, Bartolucci A, Liu C: Effects of age on responses to treatment and survival of patients with multiple myeloma. J Am Geriatr Sot 1983; 31: 272-277.

Cohen HJ, Bartolucci A: Age and the treatment of multiple myeloma. Southeastern Cancer Study Group experience. Am J Med 1985; 79: 316-324.

Radl J, Sepers JM, Skvaril F, Morel1 A, Hijmans W: Immuno- globulin patterns in humans over 95 years of age. Clin Exp lmmunol 1975; 22: 84-90.

Radl J: Differences among the three major categories of para-proteinaemias in aging man and the mouse. A mini review. Mech Ageing Dev 1984; 28: 167-170.

Radl J: Monoclonal gammopathies. An attempt at a new classification. Neth J Med 1985; 28: 134-137.

Radl J, De Glopper E, Van Den Berg P, Van Zwieten MJ: Idiopathic paraproteinemia. Ill. Increased frequency of paraproteinemia in thymectomized aging C57BLIKaLwRij and CBA/BrARij mice. J lmmunol 1980; 125: 31-35.

Radl J, Valentijn RM, Haaijman JJ, van Es LA: Increased incidence of monoclonal gammopathies in patients under immunosuppressive treatment after renal transplantation is age-related. Clin lmmunol lmmunopathol 1985; 37: 98- 102.


evaluation of monoclonal gammopathies in the “well” elderly

Documents