The International Myeloma Working Group has defi ned MGUS as the presence of a monoclonal protein in patients without the evidence of multiple myeloma, amyloidosis, Waldenström macroglob-
Trang 1InXtroduction
Oswald Steward
Reeve-Irvine research center, departments of anatomy & nurobiology, nurobiology & behavior,
and neurosurgery, university of california at irvine, Irvine, CA 92697
Introduction
Monoclonal gammopathy of unknown signifi cance
(MGUS) affects up to 2% of persons aged 50 years
or over, and about 3% of those older than 70 years
The aim of this chapter is to highlight some of the
features of this disease process and its relationship
to aging It will focus on (1) epidemiology, (2)
biol-ogy of MGUS and the role of various cytokines in
the pathophysiology, (3) correlation of the biology
of MGUS with aging, (4) diagnosis and follow-up
of patients with MGUS, and (5) natural history and
predictors of progression in patients with MGUS
Epidemiology
The monoclonal gammopathies are a group of
dis-orders associated with proliferation of a single clone
(monoclonal) of plasma cells They are characterized
by the secretion of an immunologically homogenous
monoclonal protein (M-protein, M-component,
M-spike, or paraprotein) Each M-protein consists
of two heavy (H) polypeptide chains of the same
class and subclass, and two light (L) chains of the
same type The heavy chains are IgG, IgM, IgA, IgD,
and IgE, while the light-chain types are kappa (κ),
and lambda (λ) In contrast, a polyclonal
gammopa-thy is characterized by an increase in one or more
heavy chains, and in both types of light chains, and
is usually associated with an infl ammatory or
reac-tive process [1] The term monoclonal gammopathy
of unknown signifi cance (MGUS) was fi rst coined by
Biological and clinical signifi cance of monoclonal
gammopathy
Arati V Rao, Harvey Jay Cohen
Kyle et al., to replace the term benign monoclonal
gammopathy, which was misleading because, at diagnosis, it is not known if the disease process will remain stable and asymptomatic, or evolve into symptomatic multiple myeloma (MM) [2]
The International Myeloma Working Group has defi ned MGUS as the presence of a monoclonal protein in patients without the evidence of multiple myeloma, amyloidosis, Waldenström macroglob-ulinemia (WM), or any other B-cell lymphoprolif-erative disorder More specifi cally it is defi ned as an M-spike of 3.0 g/dL or trace or no light chains in a 24-hour urine collection, less than 10% plasma cells
in the bone marrow, and no related organ or sue impairment, i.e., no lytic bone lesions, and the absence of anemia, hypercalcemia, and renal insuf-
tis-fi ciency [1] As for most cancers, especially logic malignancies, this condition demonstrates an increased incidence with age and affects up to 2% of persons50 years and about 3% of those older than
hemato-70 years [3] A screening study conducted in Sweden demonstrated MGUS in 0.1–0.2% of persons aged 30–49 years, 1.1–2.0% of persons 50–79 years, and in 5.7% of persons 80–89 years [4] In a cluster of cases
of MM, Kyle et al were able to detect an M-protein
in 15 of 1200 persons 50 years or older (1.25%) [3], and in France, 303 of 17 968 persons 50 years or
older (1.7%) had an M-protein [5] Crawford et al.
have reported that 10% of 111 persons older than
80 years had an M-protein ranging in concentration from 0.2 to 1.8 g/dL [6] This has also been reiterated
by Cohen et al., who found that 3.6% of 816 persons
70 years or older had an M-protein [7] As with
11
Blood Disorders in the Elderly, ed Lodovico Balducci, William Ershler, Giovanni de Gaetano
Trang 2MM, the incidence of MGUS is higher in African–
Americans than in whites, and in one study the
prevalence of an M-protein was 8.4% in 916 African–
Americans [7] In contrast, the incidence of MGUS is
only about 2.7% in elderly Japanese patients [8] The
monoclonal protein in MGUS is most commonly
IgG (73%), followed by IgM (14%) and IgA (11%) The
light chains in MGUS most commonly involve the
κ molecules (62%) [9].
MGUS is frequently a single abnormality, but it
may be associated with many other diseases, as
would be expected in the elderly populations Most
common associations have been with the B-cell
lymphoproliferative disorders like chronic
lym-phocytic leukemia, non-Hodgkin lymphoma, and
hairy-cell leukemia [10] One prospective study
showed that MGUS was detected in 1.1% of patients
with solid tumors referred for systemic
chemother-apy [11] A third of patients with chronic neutrophilic
leukemia, which is a rare disorder characterized by
persistent leukocytosis of mature neutrophils, have
an elevated M-protein [12] It has also been seen
in Gaucher’s disease [13], myelofi brosis, hepatitis
C infection, HIV infection [14], rheumatoid
arthri-tis [15], and other related disorders Interestingly,
MGUS has been observed after liver, renal, and
bone-marrow transplantation, and in these patients
the development of an M-protein correlated with
the presence of a viral infection, e.g.,
cytomegalovi-rus infection [16,17]
The gammopathies can be further classifi ed as:
• benign (IgG, IgM, IgA, IgD)
• associated with malignancies that are not known
to produce monoclonal proteins
• idiopathic Bence-Jones proteinuria [18,19]
• biclonal gammopathies [20]
• triclonal gammopathies [21]
Idiopathic Bence-Jones proteinuria is a condition
in which patients excrete large amounts of
mono-clonal light chains (Bence-Jones protein) and follow
a benign course A small series of seven patients
revealed no evidence of malignant plasma-cell
dis-order, and no serum M-protein, but urine light-chain
excretion of 1 g/day [19] In all these patients the
plasma cell labeling index was low After a follow-up
of 7–28 years, three of the seven patients, developed
MM, while two other patients developed matic MM, and evolving MM One patient devel-oped primary amyloidosis after 12 years, and two patients continued to have stable levels of Bence-Jones proteins The authors suggested that patients with idiopathic Bence-Jones proteinuria should be monitored regularly and indefi nitely
asympto-Of patients with a gammopathy of unknown signifi cance, 3–4% have a biclonal gammopathy, characterized by the production of two different M-proteins This may be due to the proliferation of two different clones of plasma cells each producing
an unrelated monoclonal protein, or it may result from a single clone of plasma cells producing two M-proteins In a series of 57 biclonal gammopathy patients, the most common diagnoses were biclonal gammopathy of undetermined signifi cance (65%), multiple myeloma (16%), and lymphoproliferative disease (19%) [20] The clinical features and response
to therapy were similar to patients with monoclonal gammopathy Of note, serum protein electrophore-sis (SPEP) may produce only a single band on the acetate strip, and the biclonal gammopathy may be recognized only by immunofi xation
Triclonal gammopathy has also been reported, and these patients may also have underlying lym-phoproliferative disorder, or a nonhematologic condition causing production of three different immunoglobulins [21]
Biology of MGUS and the role of various cytokines
It is well known that germinal-center B cells uniquely modify the DNA of immunoglobulin (Ig)genes through sequential rounds of somatic hyper-mutation, antigen selection, and IgH switch recom-bination Post-germinal-center plasmablasts can generate plasmablasts that have successfully com-pleted somatic hypermutation and IgH switching before migrating to the bone marrow, where stromal cells enable terminal differentiation into plasma
Trang 3140 Arati V Rao, Harvey Jay Cohen
cells MGUS and MM are both characterized by
the accumulation of transformed plasmablasts or
plasma cells in the bone marrow [22,23] However,
MGUS is less proliferative than MM, with 1% cells
synthesizing DNA [24] Gene expression profi ling
data has demonstrated a higher level of cyclin D1,
D2, or D3 mRNA in patients with MGUS and
sub-sequently with MM, when compared to normal
plasma cells [25] This allows the plasma cells to be
more susceptible to proliferative stimuli, with
selec-tive expansion, after interacting with bone-marrow
stromal cells that produce interleukin 6 (IL-6) and
other cytokines There is also some evidence that the
Rb protein which controls the cell-cycle restriction
point from G1 to S phase might be dysregulated due
to methylation of p16, which can inhibit cyclinD/
CDK4 and thus prevent phosphorylation of Rb [26]
This, along with deletion of chromosome 13, may
be the earliest change in MGUS that allows
progres-sion to MM [24] In addition, activating mutations of
N-ras and K-ras are absent with MGUS but are seen
in 30–40% patients with MM [27]
Role of IL-6, IL-6R, IL-1 β, and TNF-α
The function, differentiation, and survival of
hematopoietic cells are governed by the presence
of certain cytokines These cytokines in turn require
expression of an appropriate cellular receptor to
exert their many biologic effects The
develop-ment of MGUS and MM is dependent upon
differ-ent cytokines like granulocyte colony-stimulating
factor (G-CSF), interferon alpha (IFN-α), leukemia
inhibitory factor (LIF), IL-11, tumor necrosis
fac-tor alpha (TNF-α), and IL-6 IL-6, a multifunctional
cytokine, has been thoroughly investigated in MM
and MGUS and may possess the most biologic and
clinical signifi cance [28] The primary function of
IL-6 is to stimulate the differentiation of mature B
cells into plasma cells and also to allow proliferation
of plasmablasts in the bone marrow [29] In addition,
IL-6 is known to inhibit fas- and
dexamethesone-induced plasma-cell apoptosis in vitro [30,31]
Initial clinical observations have suggested that
high serum IL-6 levels correlate with advanced
disease, aggressive disease, and chemotherapy refractoriness [32] Multiple studies have been per-formed to demonstrate that IL-6 stimulates pro-liferation of myeloma cells in vitro, and anti-IL-6 antibodies or IL-6 antisense oligonucleotides can inhibit IL-6-stimulated growth of myeloma cells The autocrine and paracrine functions of IL-6 in myeloma, along with IL-6 transgenic mouse mod-els, has also been studied The expression of IL-6 receptors (IL-6R) by myeloma cells, and responses
in patients treated with anti-IL-6 antibodies, have also been examined in order to study the role of IL-6
in the pathophysiology of MGUS and MM [33,34] More recently, similar fi ndings have also been con-
fi rmed in patients with MGUS Sati et al developed
a dual-color fl uorescence in-situ hybridization
(FISH) technique to investigate the expression of IL-6 mRNA in bone-marrow cells of patients with
MM, with MGUS, and in healthy bone-marrow donors [35] The IL-6 protein could be detected by direct immunofl uorescence in all plasma cells from all patients with MM, and in those with MGUS, with lower levels of expression in patients with MGUS than in those with MM However, neither the IL-6 mRNA nor protein could be detected in normal plasma cells from healthy subjects These data demonstrated that patients with MGUS and MM express the IL-6 mRNA, and support the hypothesis
of autocrine synthesis of IL-6 in these patients Most investigators, however, agree that the contribution
of autocrine IL-6 is minimal, and there are emerging data that the paracrine secretion of IL-6 is the major factor in the pathogenesis of MGUS, MM, and other monoclonal gammopathies [36]
IL-6 production has been detected by Th2 T cells, monocytes, endothelial cells, fi broblasts, and bone-marrow stromal cells, and the latter is probably the major source of IL-6 in monoclonal gammopathies
[37] This has been confi rmed by Klein et al., who
attributed the high production of IL-6 to ent cells of the bone-marrow microenvironment
adher-by demonstrating a spontaneous proliferation of myeloma cells in vitro Recombinant IL-6 was able
to amplify this proliferation, and anti-IL-6 ies were able to inhibit these cells [28]
Trang 4antibod-Of note, the signaling of IL-6 is mediated via a
spe-cifi c heterodimer receptor made up of an α chain of
80 kD (IL-6R) and a β transducer chain of 130 kD
(gp130) A remarkable feature of the IL-6 receptor is
the agonist role of its soluble form (sIL-6R), which
is able to bind IL-6 with an affi nity similar to that
of membrane IL-6R Also, the IL-6/sIL-6R complex
is able to bind and activate the gp130 transducer
chain [38,39] Stasi et al investigated the clinical
sig-nifi cance of serum sIL-6R in 81 patients with MGUS,
and 164 patients with MM, and found higher levels
of sIL-6R in the MM patients In a univariate
analy-sis, sIL-6R was a signifi cant but weak prognostic
indicator, and higher levels were associated with
shorter survival [40]
The relationship of IL-6 to other cytokines like
TNF-α and IL-1β has also been well studied TNF-α
and IL-1β are potent inducers of IL-6 production and
play a role in paracrine secretion of IL-6 [41] These
two cytokines, especially IL-1β, are potent osteoclast-
activating factors and play a role in the development
of lytic lesions (Fig 11.1) Lacy et al performed
in-situ hybridization for IL-1β using bone-marrow
aspirates from 51 patients with MM, 7 with
smol-dering myeloma, 21 with MGUS, and 5 healthy
subjects [42] IL-1β mRNA was detected in plasma
cells from a majority of patients with MM (49 of 51) and smoldering myeloma, but only 5 of 21 patients with MGUS, and none of the normal sub-jects had any detectable IL-1β mRNA This contrast
in cytokine expression of IL-6 and IL-1β between
patients with MGUS and MM has also been
demon-strated by Donovan et al [43] TNF- α plays a role in
the production of IL-6 in a dose-dependent fashion
Blade et al mea sured serum levels of IL-6 and TNF- α
in 38 healthy subjects and 100 patients with MGUS IL-6 levels were signifi cantly higher in MGUS than
in healthy controls (p 0.0001) Similarly, TNF-α
levels were signifi cantly higher in MGUS than in
control populations (p 0.015) [44]
More recently, the role of adhesion molecules
in the biology of myeloma has been studied [45] Normal B cells are able to home to certain tissues due to the presence of surface adhesion molecules Myeloma cells may express a variety of surface adhesion molecules such as NCAM (CD56), ICAM (CD54), HCAM (CD44), and others A recent study has also demonstrated impaired osteoblastogenesis
in myeloma, thought to be due to increased levels
of cytokines like IL-1β, TNF-α, and IL-6 which in
MGUS MM-Low LI
IL-6IL-1b
MM-High LI
M-protein
adhesionmolecules
lytic bonelesions
paracrineIL-6
OAFIL-1b
gammopathy of unknown signifi cance; LI, labeling index; OAF, osteoclast-activating factor Adapted from Lacy MQ et al.,
Blood 1999; 93: 300–5 [42].
Trang 5142 Arati V Rao, Harvey Jay Cohen
turn led to upregulation of ICAM-1 [46] It has also
been hypothesized that acquisition of NCAM
expres-sion in myeloma is a malignancy-related
phenom-enon NCAM (CD56) is strongly expressed on most
myeloma cells but is not found on normal plasma
cells In one study, CD56 expression in high density
was present in 43 of 57 patients with untreated MM
and in none of 23 patients with MGUS [47] IL-6 has
been shown to increase HCAM (CD44) gene
expres-sion and cause overexpresexpres-sion of all CD44 variant
exons [48] It is thought that these adhesion
mol-ecules play a role in cell-to-cell contact between
myeloma cells and marrow stromal cells, and this
maybe leads to the homing of myeloma cells to the
bone marrow and to the development of osteolytic
bone lesions, and may also play a crucial role in
myeloma cell survival
IL-6 and bone-marrow angiogenesis
There has been a study which demonstrated that
vascular endothelial growth factor (VEGF) expressed
and secreted by myeloma cells stimulates the
expres-sion of IL-6 by microvascular endothelial cells and
the bone-marrow stromal cells In turn, IL-6
stimu-lates the expression of VEGF, which as we know is a
potent stimulator of angiogenesis [49] Numerous
studies have now demonstrated that marrow
angio-genesis parallels tumor progression and correlates
with tumor growth and metastatic potential in
mul-tiple myeloma patients We also have evidence that
bone-marrow angiogenesis progressively increases
along the spectrum of plasma-cell disorders, from
MGUS to advanced myeloma This has been studied
in the bone-marrow samples of 400 patients (76 with
MGUS) by immunohistochemical staining for CD34
to identify microvessels, and compared to normal
bone-marrow samples [50] The median microvessel
density per 400 high-power fi eld was 1.3 in controls,
3 in MGUS, 11 in newly diagnosed MM, and 20 in
relapsed MM Higher-grade angiogenesis was noted
with more advanced disease, and this correlated with
the bone-marrow plasma-cell percentage,
bone-mar-row plasma-cell labeling index, and survival
Role of HHV-8
Human herpesvirus 8 (HHV-8), also known as Kaposi sarcoma-associated herpesvirus (KSHV), was originally described after isolation from a patient with Kaposi sarcoma [51] In these patients
it has been isolated from primary sarcoma cells as well as B cells, macrophages, and dendritic cells HHV-8 has also been shown to be associated with systemic Castleman disease, and primary effu-sion lymphoma where it is localized to just the malignant cells The viral genome encodes a large number of homologs of cellular genes, includ-ing genes functioning in cell regulation (cyclin D), control of apoptosis (bcl-2, death effector domain proteins), cell–cell interaction, immunoregulation, and cytokine signaling, especially IL-6 [52] IL-6, which is considered an important growth factor for myeloma, and a biologically active homolog to human IL-6, termed vIL-6, has been identifi ed in the HHV-8 genome [53] This vIL-6 binds to gp130 directly, suggesting that this molecule may directly activate IL-6R signal transduc tion without binding
to the IL-6R α chain HHV-8 also contains the viral
homolog for interferon regulatory factor (vIRF), which has been detected in patients with multiple myeloma Fibroblasts transfected with vIRF develop into stromal tumors when injected into nude mice, thus suggesting vIRF has properties of a viral onco-
gene [54] Rettig et al have demonstrated vIL-6 RNA
transcripts in cultured KSHV-infected bone-marrow dendritic cells, thus suggesting a role in producing paracrine stimulation of plasma-cell growth and the possibility of transformation of MGUS to MM [55] However, a follow-up study refuted these fi ndings
by using PCR analysis for multiple regions of the HHV-8 genome and serologic studies on patients with MM and found no role of HHV-8 in the etiol-
ogy of MM [56] Ablashi et al performed serologic
assays (whole-virus ELISA) to detect IgG antibody to HHV-8 in 362 patients with MGUS and 110 patients with MM Only 7.8% of the MGUS sera contained HHV-8 antibody to lytic proteins, and no differ-ences were noted in the distribution of antibody
Trang 6to HHV-8 in sera from MGUS patients who
pro-gressed to MM The seroprevalence of HHV-8
in MGUS (7.8%), MM (5.4%), and healthy donors
(5.9%) was similar, thus arguing for the lack of
epi-demiologic evidence of HHV-8 in the pathogenesis
of MM Currently, it is unclear if MGUS patients with
HHV-8 infection will progress and go on to develop
overt MM
Relationship between aging and the
development of monoclonal gammopathy
The prevalence fi ndings discussed previously in
the Epidemiology section suggest that there may
be some fundamental changes that occur with the
process of aging that make individuals more
sus-ceptible to developing a monoclonal gammopathy
Animal models have provided some clues in
under-standing the pathophysiology of age-related
mono-clonal gammopathy In aging C57BL/KaLwRij mice,
80% of aged animals will develop a monoclonal
gammopathy that is essentially indistinguishable
from an MGUS in humans [57,58] Also, plasma-cell
dyscrasias such as MGUS, MM, or WM are rarely
seen in C57BL/KaLwRij mice less than two years old
It is hypothesized that these animals may have a
dysregulated immune system that predisposes them
to develop a monoclonal gammopathy Radl has
demonstrated his fi ndings in the C57BL/KaLwRij
mice as an imbalance between a failing T-cell
com-partment (due to an involuted thymus) with an
oth-erwise intact B-cell compartment [57] The loss of
a balanced T-cell/B-cell dichotomy in the immune
system may lead to a restriction of the B-cell
reper-toire and thus to excessive B-cell clonal
prolifera-tion, excessive immunoglobulin producprolifera-tion, and
ultimately to the development of a monoclonal
gammopathy
More recently, Ellis et al have demonstrated that
the relative numbers of the CD30 T-cell subset and
levels of CD30 expression are elevated in activated
lymphocytes from normal aged individuals (60
years) and in MGUS patients, when compared to
younger controls [59] Peripheral blood lymphocytes from MGUS patients and age-matched controls produced comparable levels of IL-6 when activated with anti-CD3 plus IL-2, and costimulation with a soluble form of CD30 ligand (sCD30L/CD8alpha) augmented anti-CD3 inducible IL-6 production similarly in both groups However, peripheral blood lymphocytes from MGUS patients also produced measurable IL-6 when activated with sCD30L/CD8alpha alone This capability was associated with the unique presence of CD30 T cells in the peripheral blood of MGUS patients Furthermore, a higher percentage of activated MGUS T cells express CD30 when activated by incubation with idiotype-expressing autologous serum than those activated
by anti-CD3 plus IL-2 These results indicate that quantitative alterations in CD30 T cells accom-pany aging and MGUS, and that these cells may con-tribute to the chronic activation of B cells though the production of IL-6
In addition to the above murine data, there is also
a wealth of data to indicate that IL-6 gene sion, along with tissue and serum levels of IL-6, all increase with age As indicated before, IL-6 is the chief cytokine implicated in the development of
expres-MM [60–64] Early observations demonstrated an age-associated rise in IL-6 in autoimmune prone mice [63] However, subsequent studies have dem-onstrated a similar age-associated increase in IL-6
in “normal” (without any disease) mice Similarly, an age-associated increase in IL-6 has been described
in healthy older humans and in older adults with coincident age-associated diseases like Alzheimer dementia, osteoporosis, and lymphoproliferative disorders [65] One proposed mechanism for this age-associated increase in IL-6 is the reduced infl u-ence of normally inhibiting sex steroids on endog-enous IL-6 The ability of estrogen to repress IL-6 expression has been studied in human endometrial stromal cells and from observations that menopause
or oophorectomy resulted in increased IL-6 levels [66,67] Similarly, dihydrotestosterone also inhib-its IL-6, albeit to a lesser extent than estrogen [68]
In one study orchiectomy induced bone-marrow
Trang 7144 Arati V Rao, Harvey Jay Cohen
IL-6 protein and mRNA expression and led to
increased replication of bone-marrow osteoclast
progenitors, which was prevented by administration
of IL-6-neutralizing antibody or implantation of a
slow-release form of testosterone [69] Thus it seems
likely that at the time of menopause or andropause,
IL-6 gene expression is not that tightly regulated,
leading to inappropriate expression in some
tis-sues and a rise in serum levels This age-associated
rise in IL-6 is of physiologic consequence,
render-ing an individual susceptible to a number of disease
processes induced by pro-infl ammatory signals,
including MM osteoclast stimulation,
lymphopro-liferative disorders, decreased functional status,
and frailty [70] One might hypothesize that in an
older patient with MGUS, the rise in IL-6 levels may
potentiate the usual molecular changes seen in an
aging individual, like decreased immune
surveil-lance, decreased DNA repair, telomere shortening,
and decreased chromosomal stability, and thus lead
A polyclonal increase in the immunoglobulins will produce a broad band that is limited to the gamma region An M-protein may be present even when the total protein concentration, beta- and gamma-globulin levels, and quantitative serum immu-noglobulin levels are all within normal limits A small M-protein may be concealed in the normal
AgeTime
Host resistance
Growth factors
Promotion
MM
Figure 11.2 Relationship between age, HHV-8, and IL-6 in the pathogenesis of MGUS and MM Adapted from Cohen HJ
et al in Balducci L et al., eds, Comprehensive Geriatric Oncology (London: Taylor and Francis, 2004), 194–203 [71].
Trang 9146 Arati V Rao, Harvey Jay Cohen
beta or gamma areas and thus easily overlooked
[73] Immunofi xation should be performed when
a peak or band is seen on SPEP, and this confi rms
the presence of and type of M-protein [74,75]
Immunofi xation can detect a serum M-protein of
0.02 g/dL and a urine M-protein of 0.004 g/dL
Urine protein electrophoresis (UPEP) is also
important in the evaluation of monoclonal
gam-mopathy Ideally, immunofi xation of a 24-hour urine
sample is recommended but it can be performed on
a random sample or the fi rst morning specimen It is
not uncommon for a patient to have a normal SPEP
with no M-protein, but for urine immunofi xation
to show a monoclonal light chain [76] Other
stud-ies might include a bone-marrow aspiration and
biopsy, a radiological skeletal bone survey,
quanti-tative serum immunoglobulins, and 24-hour urine
collection for protein quantitation and
electro-phoresis Light chains may not be detected in the
urine because of reabsorption by the proximal renal
tubules Also, there is variation in glomerular fi
ltra-tion and tubular funcltra-tion, and this is relevant in
patients with non-secretory myeloma, solitary
plas-macytoma, or primary systemic amyloidosis These
issues may be circumvented by measuring serum
free light chains, which has been shown to be a very
sensitive method of detecting and monitoring light
chains [77] The quantitative immunoassays can
detect less than 0.1 mg/dL κ and λ chains compared
to 15–50 mg/dL by immunofi xation, and 50–200 mg/
dL by SPEP [78]
It is important to differentiate a patient with
MGUS from one with MM or WM Most patients
with MGUS are asymptomatic, and may be
diag-nosed incidentally by their primary-care physicians
performing a workup for anemia or other related
conditions The size of the M-protein is helpful, and
has been a matter of debate, with the International
Myeloma Working Group using the level of 3 g/dL
as the cutoff value [1] Patients with MGUS do not
have any signs or symptoms from related organ
and tissue damage, as cited above Most patients
with MM or WM have a reduction in polyclonal or
background immunoglobulins, while only 30% of
patients with MGUS have a decrease in polyclonal
immunoglobulins The morphologic appearance of the bone marrow might help in differentiation, and this was illustrated in a study where bone-marrow aspirates from 154 patients were examined by blinded cytologists [79] These patients underwent bone-marrow aspiration as part of a workup for sus-pected myeloma The single morphologic charac-teristic that strongly differentiated MM from MGUS was the presence of large nucleoli in the plasma cells
of patients with MM Higher percentage of plasma cells (mean 48% in MM vs 10% in MGUS), irregu-lar cytoplasmic contour of plasma cells, presence of cartwheel chromatin and vacuolization, more ani-socytosis and plasma cells in clusters were features more prominent with MM bone-marrow specimens
Of note, the popular and commonly used beta-2 microglobulin level is thought not to be useful in differentiating normal individuals from those with MGUS or with early MM [80]
Plasma cells in MM are phenotypically distinct from their normal counterparts, and this has been studied
by fl ow cytometry in patients with MGUS and MM [81] The clonal plasma cells of patients with MGUS show a phenotypic profi le similar to that of myelo-matous plasma cells (CD38, CD56, and CD19), although the proportion of phenotypically normal plasma cells is higher in patients with MGUS than
in those with myeloma Thus, there are actually two populations of plasma cells in persons with MGUS: one is normal and polyclonal (CD38, CD56,CD19), and the other is clonal and has an abnor-mal immunophenotype (CD38, CD56, CD19) This study also demonstrated that the proportion of bone-marrow plasma cells that was polyclonal (as assessed by fl ow cytometry of bone-marrow aspirate with the use of four monoclonal antibodies – CD38
or CD138, CD56, CD19, and CD45) was the best single factor for distinguishing between MGUS and multiple myeloma Only 1.5% of patients with MM had more than 3% of normal plasma cells, whereas 98% of patients with MGUS had more than 3%.Conventional cytogenetics is not very useful in differentiating MGUS from MM, because of the low number of cells in metaphase in MGUS It is now thought that MGUS patients already have
Trang 10the chromosomal characteristics of a plasma-cell
malignancy, and this was confi rmed in a study in
which interphase FISH was performed on
bone-marrow plasma cells of 36 patients with MGUS
Chromosomal abnormalities were identifi ed in 53%
patients, with gains in chromosomes 3, 11, 7, and
18 most commonly seen [82] The deletion of
chro-mosome 13q is a clinically relevant feature in MM,
and one study utilized interphase FISH to
demon-strate deletion of the 13q14 locus in 45% patients
with MGUS [83] This was confi rmed by a long-term
follow-up study (median follow-up 30 months)
using conventional cytogenetics and interphase/
metaphase FISH in 18 asymptomatic, untreated
MGUS patients [84] Deletion of 13q14 was
identi-fi ed in identi-fi ve of these patients, and all identi-fi ve progressed
to MM 6 to 12 months after identifi cation of the 13q
anomaly The authors concluded that the extent of
13q deletion does not vary with clinical outcome,
and plays a crucial role in the pathogenesis of MM by
conferring a proliferative advantage to clonal plasma
cells However, these results must be interpreted
with caution since transition from MGUS to MM can
also occur in patients with normal karyotype, as
sug-gested by two patients in the same study
Another tool in distinguishing the two conditions
might be the presence and amount of circulating
plasma cells that can be seen in MM, MGUS, and
smoldering myeloma However, a recent study of 327
patients with MGUS has suggested that patients who
had detectable plasma cells in the peripheral blood
had a shorter median progression-free survival,
shorter median survival, and shorter time to
initia-tion of any therapy for progressive disease [85]
Finally, the plasma-cell labeling index, which
measures the synthesis of DNA, is a useful test for
differentiating MGUS from MM This was
evalu-ated in one study of 80 patients (59 MM, 20 MGUS,
1 plasma-cell leukemia) where plasma-cell
pro-liferation analysis was performed after
bromode-oxyuridine (BRD-URD) incorporation and double
immunoenzymatic labeling on cytological smears
[86] The BRD-URD is incorporated into the nucleus
of cells synthesizing DNA The plasma-cell labeling
index (percentage of cells in S phase) was 0.25 for
MGUS, 0.4 for stage I MM, 2.4 for stage III MM, and 3.7 for plasma-cell leukemia While there was no correlation between the labeling index and beta-2 microglobulin levels between the MGUS and MM patients, there was a correlation of these variables within the different stages of MM
Thus, making the right diagnosis of MGUS is important, as these patients need regular follow-up
in order to detect the development of the nant form, i.e., MM or other related disorders It has been suggested by Kyle & Rajkumar that if a patient has a serum M-protein value of 1.5 g/dL and no other features suggestive of a plasma-cell dyscrasia,(i.e., no anemia, hypercalcemia, renal failure) a bone-marrow examination or skeletal bone survey is not required and an SPEP should be repeated annu-ally [87] Patients with M-protein levels between 1.5 and 2.5 g/dL who are asymptomatic should have additional studies performed, including quantita-tive immunoglobulins and 24-hour UPEP, but do not need a bone-marrow biopsy or skeletal survey The SPEP should be repeated every 3–6 months for
malig-a yemalig-ar malig-and if stmalig-able the durmalig-ation between the tests can be increased to 6–12 months and then annu-ally or if any symptoms occur If the M-protein level
is 2.5 g/dL, complete workup, including titative immunoglobulins, 24-hour UPEP, bone-marrow aspiration and biopsy, and skeletal bone survey, must be performed Given the data we have from recent studies, it might be useful to check serum free light chains in patients with M-protein
quan-1.5 mg/dL and use it along with the kappa/lambda free light chain ratio to risk-stratify patients with MGUS in order to predict progression to more malignant disease (Table 11.1) [88] If the M-protein
is IgM, a bone-marrow aspiration and biopsy is cated to rule out WM or any other lymphoprolifera-tive disorder If all the studies are normal, the SPEP can be repeated every 2–3 months for a year, and if stable can be repeated at 6- to 12-month intervals
indi-It is unusual for a serum monoclonal protein to appear during long-term follow-up Formerly, if the M-protein remained stable for 3–5 years, the process was assumed to be benign and additional follow-up was not mandatory However, the most recent data
Trang 11dis-148 Arati V Rao, Harvey Jay Cohen
from Kyle et al have demonstrated an average risk
of development of a malignant process of 1% per
year [90] This study found a relative risk of 7.3 for
MM, WM, primary amyloidosis, chronic lymphocytic
leukmia, IgM lymphoma, and plasmacytoma in the
patients with MGUS, as compared to white subjects
in the Iowa SEER registry from 1973 to 1997
Natural history and predictors of
progression
There have been multiple studies conducted on the
natural history of this relatively indolent disease
This chapter, however, will focus on the most recent
studies, with large patient populations and long
durations of follow-up In one study, Baldini et al
fol-lowed 335 patients with MGUS for a median period
of 70 months [91] The frequency of malignant
trans-formation was 6.8%, and there was no difference in
patients with IgM, IgG, or IgA monoclonal protein In
a univariate analysis of the IgG cases, the relative risks
for developing multiple myeloma were as follows: 2.4
for each 1 g/dL increase in IgG serum monoclonal
component, 3.2 for detectable light chain
protein-uria, 4.4 for an increase of one unit log bone-marrow
plasma-cell percentage, 6.1 for age 70 years, 3.6 and
13.1 for a reduction in one or two polyclonal
immu-noglobulins Patients with an M-spike of 1.5 g/dL,
bone-marrow plasma cells 5%, and no urinary light
chains were at lowest risk, and it was suggested in
this paper that patients do not need a skeletal bone
survey or bone-marrow examination until there is
evidence of progressive disease In another study
by Rajkumar et al [89], 1384 patients with MGUS were
followed for 34 years (median 15.4 years) and ing this follow-up MM, lymphoma with an IgM serum monoclonal protein, primary amyloidosis, macroglobulinemia, chronic lymphocytic leukemia,
dur-or plasmacytoma developed in 115 (8%) patients The cumulative probability of progression to one of these disorders was about 10% at 10 years, 21% at 20 years, and 26% at 25 years The overall risk of pro-gression was about 1% per year, and patients were
at risk of progression even after 25 years or more
of stable MGUS Of note, patients with MGUS had shorter median survival than expected for age- and
sex-matched controls (8.1 years vs 11.8 years, p0.001) After analysis of all baseline factors only the concentration and type of monoclonal protein were independent predictors of progression In this study patients with IgM or IgA monoclonal protein had an increased risk of progression as compared to patients
with IgG monoclonal protein (p 0.001), and the risk
of progression was directly related to the tion of monoclonal protein in the serum at the time
concentra-of diagnosis concentra-of MGUS (p 0.001) Finally, a study
by Cesana et al followed 1104 patients with MGUS
for 33 years (median follow-up 65 months) [92] Cumulative transformation probability at 10 and 15 years was 14% and 30% respectively, with 64 patients (5.8%) evolving to multiple myeloma or other related disorders Marrow plasmacytosis greater than 5%, detectable Bence-Jones proteinuria, polyclonal serum immunoglobulin reaction, and a high eryth-rocyte sedimentation rate were independent factors infl uencing MGUS transformation
One very elegant study has hypothesized that the presence of free light chains in the serum, with
Table 11.1 Risk stratifi cation using the size of the serum M-protein and free light chain (FLC) ratio [89].
Absolute risk of progressionRisk group No of patients Hazard ratio at 20 years (%)
Low (serum M-protein 1.5 g/dL and 606 1 7
normal FLC ratio [0.26–1.65])
Intermediate (either risk factor present) 373 3.5 26
High (serum M-protein 1.5 g/dL and 169 6.8 46
abnormal FLC ratio [0.26 or 1.65])
Trang 12abnormal kappa/lambda ratio, would indicate clonal
evolution in the neoplastic plasma cell and thus
increased risk of progression in MGUS [93] These
investigators utilized a novel, highly sensitive serum
free light chain assay that has recently been
introduced into clinical practice It enables the quantifi
-cation of free kappa and lambda chains secreted by
plasma cells, i.e., those not bound to intact
immu-noglobulin Monoclonal elevations can be reliably
distinguished from polyclonal elevations using the
kappa/lambda ratio The free light chain ratios were
determined in 47 patients with MGUS who had
doc-umented progression to MM or related malignancy,
and compared to 50 patients with MGUS who had
no evidence of progression The presence of higher
kappa/lambda free light chain ratio was associated
with a higher risk of MGUS progression (relative
risk 2.5, 95% CI 1.6–4.0, p 0.001) The same group
has now utilized the kappa/lambda free light chain
ratio along with the serum M-protein level in 1384
patients with MGUS to risk-stratify these patients
and thus predict the absolute risk of progression at
20 years (Table 11.1) An abnormal kappa/lambda
free light chain ratio of 0.26 or 1.65 was
consid-ered a major independent risk factor for progression
of MGUS to MM or other related malignancy
Thus, in general MGUS can be characterized as
a pre-malignant condition with a limited and
con-trolled lymphoplasmacytic expansion and a benign
course However, transformation to a more
malig-nant condition like MM can occur, and generally is
associated with higher tumor burden and a
progres-sively downward course
Summary
Monoclonal gammopathy of unknown signifi cance,
a pre-malignant condition, is the most common
plasma-cell dyscrasia, and its incidence increases
with age It is characterized by a serum M-protein of
3.0 g/dL or trace or no light chains in a 24-hour
urine collection, less than 10% plasma cells in the
bone marrow, and the absence of lytic bone lesions,
anemia, hypercalcemia, and renal insuffi ciency
There is a clear association between aging and the development of monoclonal gammopathy, and this may be due to increased levels of IL-6 and other related cytokines Long-term follow-up studies have demonstrated the rate of transformation of MGUS
to MM or related disorder to be 1% per year Risk factors for progression have been studied by multi-ple groups, and those that place the patient at high risk include size of M-protein, presence of urinary light chains, and bone-marrow plasmacytosis More sophisticated tests like the plasma-cell labeling index, and FISH to detect 13q deletion, might also be helpful in determining risk of transformation More recently, very sensitive assays for serum free light chains have been developed, enabling us to calculate the kappa/lambda chain ratio and use this to predict progression of MGUS to more serious MM While there are no absolute fi ndings at the time of diagno-sis of MGUS that allow us to distinguish patients who will remain stable from those who will develop more malignant disease, we do have many diagnostic tests
in our armamentarium to help follow these patients
REFERENCES
1 International Myeloma Working Group Criteria for the classifi cation of monoclonal gammopathies, mul-tiple myeloma and related disorders: a report of the
International Myeloma Working Group Br J Haematol
2003; 121: 749–57.
2 Kyle RA Monoclonal gammopathy of undetermined
sig-nifi cance: natural history in 241 cases Am J Med 1978;
64: 814–26.
3 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: 719–24.
4 Axelsson U, Bachmann R, Hallen J Frequency of logical proteins (M-components) from 6,995 sera from
patho-an adult population Acta Med Scpatho-and 1966; 179: 235–47.
5 Saleun JP, Vicariot M, Deroff P, Morin JF Monoclonal gammopathies in the adult population of Finistere,
France J Clin Pathol 1982; 35: 63–8.
6 Crawford J, Eye MK, Cohen HJ Evaluation of monoclonal
gammopathies in the “well” elderly Am J Med 1987; 82:
39–45
Trang 13150 Arati V Rao, Harvey Jay Cohen
7 Cohen HJ, Crawford J, Rao MK, Pieper CF, Currie MS
Racial differences in the prevalence of monoclonal
gammopathy in a community-based sample of the
eld-erly Am J Med 1998; 104: 439–44.
8 Bowden M, Crawford J, Cohen HJ, Noyama O A
com-parative study of monoclonal gammopathies and
immunoglobulin levels in Japanese and United States
elderly J Am Geriatr Soc 1993; 41: 11–14.
9 Kyle RA Monoclonal gammopathy of undetermined
signifi cance and solitary plasmacytoma Implications
for progression to overt multiple myeloma Hematol
Oncol Clin North Am 1997; 11: 71–87.
10 Hansen DA, Robbins BA, Bylund DJ, Piro LD, Saven A,
Ellison DJ Identifi cation of monoclonal
immunoglob-ulins and quantitative immunoglobulin abnormalities
in hairy cell leukemia and chronic lymphocytic
leuke-mia Am J Clin Pathol 1994; 102: 580–5.
11 Anagnostopoulos A, Galani E, Gika D, Sotou D,
Evangelopoulou A, Dimopoulos MA Monoclonal
gam-mopathy of undetermined signifi cance (MGUS) in
patients with solid tumors: effects of chemotherapy on
the monoclonal protein Ann Hematol 2004; 83: 658–60.
12 Rovira M, Cervantes F, Nomdedeu B, Rozman C
Chronic neutrophilic leukaemia preceding for seven
years the development of multiple myeloma Acta
Haematol 1990; 83: 94–5.
13 Shoenfeld Y, Berliner S, Pinkhas J, Beutler E The
asso-ciation of Gaucher’s disease and dysproteinemias Acta
Haematol 1980; 64: 241–3.
14 Hamazaki K, Baba M, Hasegawa H, et al Chronic
hepatitis C associated with monoclonal gammopathy
of undetermined signifi cance J Gastroenterol Hepatol
2003; 18: 459–60.
15 Youinou P, Le Goff P, Renier JC, Hurez D, Miossec P,
Morrow WJ Relationship between rheumatoid arthritis
and monoclonal gammopathy J Rheumatol 1983; 10:
210–15
16 Hashino S, Imamura M, Kobayashi S, Kobayashi H,
Kasai M, Miyazaki T Monoclonal gammopathy (IgA:
lambda) after autologous T-cell-depleted bone marrow
transplantation in a patient with non-Hodgkin’s
lym-phoma Ann Hematol 1993; 67: 135–7.
17 Passweg J, Thiel G, Bock HA Monoclonal gammopathy
after intense induction immunosuppression in renal
transplant patients Nephrol Dial Transplant 1996; 11:
2461–5
18 Kyle RA, Maldonado JE, Bayrd ED Idiopathic Bence
Jones proteinuria: a distinct entity? Am J Med 1973; 55:
222–6
19 Kyle RA, Greipp PR “Idiopathic” Bence Jones
protein-uria: long-term follow-up in seven patients N Engl J
23 Kuehl WM, Bergsagel PL Multiple myeloma: evolving
genetic events and host interactions Nat Rev Cancer
2002; 2: 175–87.
24 Fonseca R, Barlogie B, Bataille R, et al Genetics and
cytogenetics of multiple myeloma: a workshop report
26 Guillerm G, Gyan E, Wolowiec D, et al p16(INK4a) and
p15(INK4b) gene methylations in plasma cells from monoclonal gammopathy of undetermined signifi -
cance Blood 2001; 98: 244–6.
27 Liu P, Leong T, Quam L, et al Activating mutations of
N- and K-ras in multiple myeloma show different cal associations: analysis of the Eastern Co-operative
clini-Oncology Group Phase III Trial Blood 1996; 88:
2699–706
28 Klein B, Zhang XG, Lu ZY, Bataille R Interleukin-6 in
human multiple myeloma Blood 1995; 85: 863–72.
29 Lotz M Interleukin-6: a comprehensive review Cancer
Treat Res 1995; 80: 209–33.
30 Chauhan D, Pandey P, Ogata A, et al Dexamethasone
induces apoptosis of multiple myeloma cells in a JNK/
SAP kinase independent mechanism Oncogene 1997;
15: 837–43.
31 Chauhan D, Kharbanda S, Ogata A, et al
Interleukin-6 inhibits Fas-induced apoptosis and stress-activated protein kinase activation in multiple myeloma cells
Blood 1997; 89: 227–34.
32 Zhang XG, Klein B, Bataille R Interleukin-6 is a potent myeloma-cell growth factor in patients with aggressive
multiple myeloma Blood 1989; 74: 11–13.
33 Hallek M, Bergsagel PL, Anderson KC Multiple loma: increasing evidence for a multistep transforma-
mye-tion process Blood 1998; 91: 3–21.
Trang 1434 Chen YH, Shiao RT, Labayog JM, Modi S, Lavelle D
Modulation of interleukin-6/interleukin-6 receptor
cytokine loop in the treatment of multiple myeloma
Leuk Lymphoma 1997; 27: 11–23.
35 Sati HI, Apperley JF, Greaves M, et al Interleukin-6
is expressed by plasma cells from patients with multiple
myeloma and monoclonal gammopathy of
undeter-mined signifi cance Br J Haematol 1998; 101: 287–95.
36 Klein B, Zhang XG, Jourdan M, et al Paracrine rather
than autocrine regulation of myeloma-cell growth
and differentiation by interleukin-6 Blood 1989; 73:
517–26
37 Treon SP, Anderson KC Interleukin-6 in multiple
mye-loma and related plasma cell dyscrasias Curr Opin
Hematol 1998; 5: 42–8.
38 Gaillard JP, Mani JC, Liautard J, Klein B, Brochier J
Interleukin-6 receptor signaling, gp80 and gp130
receptor interaction in the absence of interleukin-6
Eur Cytokine Netw 1999; 10: 43–8.
39 Gaillard JP, Liautard J, Klein B, Brochier J Major role of
the soluble interleukin-6/interleukin-6 receptor
com-plex for the proliferation of interleukin-6-dependent
human myeloma cell lines Eur J Immunol 1997; 27:
3332–40
40 Stasi R, Brunetti M, Parma A, Di Giulio C, Terzoli E,
Pagano A The prognostic value of soluble interleukin-6
receptor in patients with multiple myeloma Cancer
1998; 82: 1860–6.
41 Sati HI, Greaves M, Apperley JF, Russell RG,
Croucher PI Expression of interleukin-1 beta and
tumour necrosis factor-alpha in plasma cells from
patients with multiple myeloma Br J Haematol 1999;
104: 350–7.
42 Lacy MQ, Donovan KA, Heimbach JK, Ahmann GJ,
Lust JA Comparison of interleukin-1 beta expression
by in situ hybridization in monoclonal gammopathy
of undetermined signifi cance and multiple myeloma
Blood 1999; 93: 300–5.
43 Donovan KA, Lacy MQ, Kline MP, et al Contrast in
cytokine expression between patients with
mono-clonal gammopathy of undetermined signifi cance or
multiple myeloma Leukemia 1998; 12: 593–600.
44 Blade J, Filella X, Montoto S, et al Interleukin 6 and
tumour necrosis factor alpha serum levels in
mono-clonal gammopathy of undetermined signifi cance Br
J Haematol 2002; 117: 387–9.
45 Van Riet I, Van Camp B The involvement of adhesion
molecules in the biology of multiple myeloma Leuk
Lymphoma 1993; 9: 441–52.
46 Silvestris F, Cafforio P, Calvani N, Dammacco F Impaired osteoblastogenesis in myeloma bone disease: role of upregulated apoptosis by cytokines and malig-
nant plasma cells Br J Haematol 2004; 126: 475–86.
47 Sonneveld P, Durie BG, Lokhorst HM, Frutiger Y, Schoester M, Vela EE Analysis of multidrug-resistance (MDR-1) glycoprotein and CD56 expression to separate
monoclonal gammopathy from multiple myeloma Br J
Haematol 1993; 83: 63–7.
48 Vincent T, Mechti N IL-6 regulates CD44 cell surface
expression on human myeloma cells Leukemia 2004;
18: 967–75.
49 Dankbar B, Padro T, Leo R, et al Vascular endothelial
growth factor and interleukin-6 in paracrine
tumor-stromal cell interactions in multiple myeloma Blood
2000; 95: 2630–6.
50 Rajkumar SV, Mesa RA, Fonseca R, et al Bone marrow
angiogenesis in 400 patients with monoclonal mopathy of undetermined signifi cance, multiple mye-
gam-loma, and primary amyloidosis Clin Cancer Res 2002;
pathway genes by KSHV Science 1996; 274: 1739–44.
54 Gao SJ, Boshoff C, Jayachandra S, Weiss RA, Chang Y, Moore PS KSHV ORF K9 (vIRF) is an oncogene which
inhibits the interferon signaling pathway Oncogene
1997; 15: 1979–85.
55 Rettig MB, Ma HJ, Vescio RA, et al Kaposi’s
sarcoma-associated herpesvirus infection of bone marrow
den-dritic cells from multiple myeloma patients Science
1997; 276: 1851–4.
56 Ablashi DV, Chatlynne L, Thomas D, et al Lack of
sero-logic association of human herpesvirus-8 (KSHV) in patients with monoclonal gammopathy of undeter-mined signifi cance with and without progression to
multiple myeloma Blood 2000; 96: 2304–6.
57 Radl J Age-related monoclonal gammapathies: clinical
lessons from the aging C57BL mouse Immunol Today
1990; 11: 234–6.
58 Radl J Aging and proliferative homeostasis:
mono-clonal gammopathies in mice and men Lab Anim Sci
1992; 42: 138–41.
Trang 15152 Arati V Rao, Harvey Jay Cohen
59 Ellis TM, Le PT, DeVries G, Stubbs E, Fisher M,
Bhoopalam N Alterations in CD30() T cells in
mon-oclonal gammopathy of undetermined signifi cance
Clin Immunol 2001; 98: 301–7.
60 Ershler WB Interleukin-6: a cytokine for gerontologists
J Am Geriatr Soc 1993; 41: 176–81.
61 Ershler WB, Sun WH, Binkley N, et al Interleukin-6 and
aging: blood levels and mononuclear cell production
increase with advancing age and in vitro production is
modifi able by dietary restriction Lymphokine Cytokine
Res 1993; 12: 225–30.
62 Suzuki H, Yasukawa K, Saito T, et al Serum soluble
interleukin-6 receptor in MRL/lpr mice is elevated
with age and mediates the interleukin-6 signal Eur J
Immunol 1993; 23: 1078–82.
63 Tang B, Matsuda T, Akira S, et al Age-associated
increase in interleukin 6 in MRL/lpr mice Int Immunol
1991; 3: 273–8.
64 Wei J, Xu H, Davies JL, Hemmings GP Increase of
plasma IL-6 concentration with age in healthy
sub-jects Life Sci 1992; 51: 1953–6.
65 Ershler WB, Sun WH, Binkley N The role of
interleukin-6 in certain age-related diseases Drugs Aging 1994; 5:
358–65
66 Kania DM, Binkley N, Checovich M, Havighurst T,
Schilling M, Ershler WB Elevated plasma levels of
interleukin-6 in postmenopausal women do not
cor-relate with bone density J Am Geriatr Soc 1995; 43:
236–9
67 Tabibzadeh SS, Poubouridis D, May LT, Sehgal PB
Interleukin-6 immunoreactivity in human tumors Am
J Pathol 1989; 135: 427–33.
68 Daynes RA, Araneo BA, Ershler WB, Maloney C, Li GZ,
Ryu SY Altered regulation of IL-6 production with
normal aging: possible linkage to the age-associated
decline in dehydroepiandrosterone and its sulfated
derivative J Immunol 1993; 150: 5219–30.
69 Zhang J, Pugh TD, Stebler B, Ershler WB, Keller ET
Orchiectomy increases bone marrow interleukin-6
lev-els in mice Calcif Tissue Int 1998; 62: 219–26.
70 Ershler WB, Keller ET Age-associated increased
inter-leukin-6 gene expression, late-life diseases, and frailty
Ann Rev Med 2000; 51: 245–70.
71 Cohen HJ, Nikcevich D Natural history and
epide-miology of monoclonal gammopathies In Balducci
L, Lyman GH, Ershler WB, Extermann M, eds,
Comprehensive Geriatric Oncology, 2nd edn (London:
Taylor and Francis, 2004), 194–203
72 Wu J, Seo PH, Cohen HJ Approach to monoclonal mopathy of unknown signifi cance in the older patient
gam-Clin Geriatr 2005; 13: 18–24.
73 O’Connell TX, Horita TJ, Kasravi B Understanding and
interpreting serum protein electrophoresis Am Fam
Physician 2005; 71: 105–12.
74 Attaelmannan M, Levinson SS Understanding and
identifying monoclonal gammopathies Clin Chem
2000; 46: 1230–8.
75 Keren DF, Warren JS, Lowe JB Strategy to diagnose monoclonal gammopathies in serum: high-resolution electrophoresis, immunofi xation, and kappa/lambda
quantifi cation Clin Chem 1988; 34: 2196–201.
76 Keren DF, Alexanian R, Goeken JA, Gorevic PD, Kyle RA, Tomar RH Guidelines for clinical and laboratory evalu-
ation patients with monoclonal gammopathies Arch
Pathol Lab Med 1999; 123: 106–7.
77 Drayson M, Tang LX, Drew R, Mead GP, Carr-Smith H, Bradwell AR Serum free light-chain measurements for identifying and monitoring patients with nonsecretory
multiple myeloma Blood 2001; 97: 2900–2.
78 Bradwell AR, Carr-Smith HD, Mead GP, et al Highly
sensitive, automated immunoassay for
immunoglob-ulin free light chains in serum and urine Clin Chem
2001; 47: 673–80.
79 Milla F, Oriol A, Aguilar J, et al Usefulness and
reproduc-ibility of cytomorphologic evaluations to differentiate myeloma from monoclonal gammopathies of unknown
signifi cance Am J Clin Pathol 2001; 115: 127–35.
80 Bataille R, Grenier J, Sany J Beta-2-microglobulin in myeloma: optimal use for staging, prognosis, and treat-
ment A prospective study of 160 patients Blood 1984;
63: 468–76.
81 Ocqueteau M, Orfao A, Almeida J, et al
Immuno-phenotypic characterization of plasma cells from mono clonal gammopathy of undetermined signifi -cance patients: implications for the differential diag-
nosis between MGUS and multiple myeloma Am J
Pathol 1998; 152: 1655–65.
82 Drach J, Angerler J, Schuster J, et al Interphase fl
uo-rescence in situ hybridization identifi es chromosomal abnormalities in plasma cells from patients with mono-clonal gammopathy of undetermined signifi cance
Blood 1995; 86: 3915–21.
83 Konigsberg R, Ackermann J, Kaufmann H, et al
Dele-tions of chromosome 13q in monoclonal gammopathy
of undetermined signifi cance Leukemia 2000; 14:
1975–9
Trang 1684 Bernasconi P, Cavigliano PM, Boni M, et al Long-term
follow up with conventional cytogenetics and band
13q14 interphase/metaphase in situ hybridization
monitoring in monoclonal gammopathies of
undeter-mined signifi cance Br J Haematol 2002; 118: 545–9.
85 Kumar S, Rajkumar SV, Kyle RA, et al Prognostic value
of circulating plasma cells in monoclonal gammopathy
of undetermined signifi cance J Clin Oncol 2005; 23:
5668–74
86 Ffrench M, Ffrench P, Remy F, et al Plasma cell
pro-liferation in monoclonal gammopathy: relations with
other biologic variables: diagnostic and prognostic
sig-nifi cance Am J Med 1995; 98: 60–6.
87 Kyle RA, Rajkumar SV Monoclonal gammopathies of
undetermined signifi cance Hematol Oncol Clin North
Am 1999; 13: 1181–202.
88 Rajkumar SV, Kyle RA, Therneau TM, et al Serum free
light chain ratio is an independent risk factor for
pro-gression in monoclonal gammopathy of undetermined
signifi cance Blood 2005; 106: 812–17.
89 Rajkumar SV, Kyle RA, Therneau TM, et al Presence of
an abnormal serum free light ratio is an independent risk factor for progression in monoclonal gammopa-
thy of undetermined signifi cance Blood (ASH Annual
Meeting Abstracts), Nov 2004; 104: 3647.
90 Kyle RA, Therneau TM, Rajkumar SV, et al A long-term
study of prognosis in monoclonal gammopathy of
unde-termined signifi cance N Engl J Med 2002; 346: 564–9.
91 Baldini L, Guffanti A, Cesana BM, et al Role of different
hematologic variables in defi ning the risk of malignant
transformation in monoclonal gammopathy Blood
1996; 87: 912–18.
92 Cesana C, Klersy C, Barbarano L, et al Prognostic
fac-tors for malignant transformation in monoclonal mopathy of undetermined signifi cance and smoldering
gam-multiple myeloma J Clin Oncol 2002; 20: 1625–34.
93 Rajkumar SV, Kyle RA, Therneau TM, et al Presence of
monoclonal free light chains in the serum predicts risk
of progression in monoclonal gammopathy of
undeter-mined signifi cance Br J Haematol 2004; 127: 308–10.
Trang 18Anemia of aging
Trang 20Introduxtion
Oswald StewardReeve-Irvine research center, departments of anatomy & nurobiology, nurobiology & behavior,
and neurosurgery, university of california at irvine, Irvine, CA 92697
Introduction
Anemia represents a common problem among
the elderly, with a prevalence of 5–10% for the
community-dwelling elderly between 65 and 74
years, and over 20% for seniors 85 years and over In
hospitalized elderly patients and in skilled nursing
facilities, anemia prevalence ranges from 40 to 50%
The recent appreciation of the numerous adverse
consequences of anemia has generated interest in
a more complete understanding of anemia in the
elderly Erythropoietin (EPO) is a hormone central
to the regulation of red-blood-cell production that
increases in response to falling hemoglobin
con-centration Paradoxically, although EPO levels rise
slightly with age in non-anemic elderly people,
the expected EPO response to anemia appears
sig-nifi cantly blunted in the elderly, supporting a
rela-tive endogenous EPO defi ciency This relarela-tive EPO
defi ciency, possibly attributable to occult renal
insuffi ciency, may play a central role in the rising
prevalence of anemia with advancing age and
unex-plained anemia in the elderly
Erythropoiesis
Hematopoiesis, the production of blood elements,
occurs in an orderly hierarchical fashion
Mainte-nance of mature peripheral blood cells (i.e.,
plate-lets, red blood cells or erythrocytes, neutrophils,
eosinophils, basophils, monocytes, lymphocytes,
natural killer cells, and dendritic cells) demands
Erythropoietin and aging
Andrew S Artz
ongoing production to meet losses and respond to stresses A pluripotent hematopoietic stem cell pro-duces committed progenitors of myeloid, erythroid, and megakaryocytic lineages Erythropoiesis denotes the process of forming mature red blood cells The earliest erythroid lineage progenitors include the BFU-E (burst-forming unit, erythroid), which later give rise to CFU-E (colony-forming unit, erythroid) Normal erythropoiesis in adults occurs exclusively in the bone marrow
Erythropoietin physiology
Various hematopoietic growth factors support proliferation, differentiation, and survival of stem cells Erythropoietin, a glycoprotein hematopoietic growth factor composed of 165 amino acids after modifi cation from the 193-amino-acid molecule encoded by mRNA, serves as a primary regulator of red-cell production [1,2] The markedly elevated cir-culating EPO levels in patients with aplastic anemia allowed initial EPO isolation from the urine [3] The glycosylation present on EPO delays clearance of the molecule, but is not necessary for biologic activ-ity [4] Synthesis and EPO regulation occur prima-rily through the kidney, from interstitial fi broblasts and/or from proximal tubular cells, with a smaller contribution by liver hepatocytes [5–9] As a con-sequence, renal failure inexorably leads to anemia from impaired EPO production
Under normal homeostasis, suppressed tion of EPO leads to a minimal constitutive pro-duction Reduced tissue oxygenation (rather than
transcrip-Blood Disorders in the Elderly, ed Lodovico Balducci, William Ershler, Giovanni de Gaetano
Trang 21158 Andrew S Artz
diminished red-cell production), typically from
anemia or hypoxia, potently stimulates synthesis
with logarithmically elevated serum EPO levels [10]
Hypoxic conditions promote activation of the hypoxia
inducible factor 1 (HIF-1) pathway, with expression
of the alpha and beta subunits, and eventual
down-stream EPO gene transcription [11] With reduced
oxygen delivery, the number of renal cells producing
EPO may also increase [9] Elevated serum EPO levels
enhance erythrocyte production primarily by
inhib-iting apoptosis of erythroid progenitor cells, although
EPO also promotes erythroid progenitor proliferation
and differentiation [12]
Erythropoiesis with aging
Early laboratory models helped guide subsequent
human studies describing EPO kinetics with aging
Basal erythropoiesis and red-cell mass in aged mice
appears similar to younger mice, although measured
hematocrit may be slightly lower secondary to
dilu-tion [13,14] In response to stressors, such as hypoxia,
bleeding, or environmental perturbation, aged mice
have an impaired erythropoietic response [13,15,16]
Human studies in older subjects with unexplained
anemia have shown normal BFU-E cells but
dimin-ished CFU-E, the erythroid precursors most replete
with EPO receptors, suggesting a block in
differenti-ation possibly mediated by inadequate endogenous
EPO levels Further, CFU-E viability requires EPO [17]
Erythropoietin receptors
Hematopoietic growth factors in general bind with
high affi nity to specifi c receptors on target cells
EPO activity necessitates binding to erythropoietin
receptors (EPO-R) Within the marrow
compart-ment, EPO receptors (EPO-R) occur on the two
classes of committed erythroid precursors, the
BFU-E and CFU-BFU-E [18] While BFU-BFU-E cells have a high
proliferative capacity, they only have low EPO-R
expression The CFU-E cells, progeny of the BFU-E
cells, have a lower proliferative capacity but increased
EPO-R density After EPO attaches to EPO-R, signal
transduction permits cell survival The heterogeneity
of EPO sensitivity at the molecular level of such cells may also facilitate erythropoietic regulation [19]
Non-marrow activity of erythropoietin and erythropoietin receptors
Recognition of EPO in non-hematopoietic tissue has pointed to the potentially pleiotropic effects of the hormone The presence of EPO-R on vascular endothelial cells [20], various renal cells [21,22], human tumor cells [23], and the central nervous system (CNS) refl ect a multiplicity of non-hematopoietic roles for EPO, and possibly therapeutic implications Research into the understanding of EPO in CNS development and homeostasis has been particularly fruitful Similar to the bone-marrow compartment, hypoxia-regulated EPO and EPO-R expression occur within the central nervous system [24–27] Endogenous EPO does not signifi cantly cross the blood–brain barrier [28], although large pharmacologic doses may pene-trate the CNS EPO may protect the CNS after hypoxic injury [29], motivating studies of pharmacologic EPO
as neuroprotective therapy after strokes, with ing preliminary results [30] Changes with advancing age have not been studied
promis-Normal erythropoietin response
to anemia
Serum endogenous EPO levels typically remain within a narrow reference range for young healthy patients at normal hemoglobin concentrations Isolation of extremely high concentrations of EPO from the urine of patients with aplastic anemia helped uncover the dynamic nature of EPO regula-tion in response to anemia The advent of immu-noassays to measure serum EPO further promoted research clarifying serum EPO kinetics [31]
Relative erythropoietin defi ciency
Determining the expected rise in serum EPO to anemia remains problematic and complicates both
Trang 22clinical and research data interpretation Reference
ranges reported by laboratories describe serum EPO
levels from non-anemic healthy populations and do
not refl ect the dynamic nature of EPO responses
Interpretation of these values may result in
con-fusion, as a serum EPO level within the reference
range (approximately 4–28 mIU/mL, depending on
the laboratory) for an anemic patient often
signi-fi es a relative EPO designi-fi ciency Multiple methods of
quantifying EPO responsiveness have further
com-plicated evaluating studies, including mean and
median EPO levels, with and without adjustment for
hemoglobin concentration Generating a standard
curve documenting the expected serum EPO for a
given hemoglobin concentration enables
appropri-ate statistical modeling but also requires logarithmic
transformation of EPO levels Such transformations
further complicate clinical interpretation
Among younger subjects with signifi cant
ane-mia, a normal EPO response has been characterized
as a semilogarithmic increase In general, serum
EPO levels inversely correlate with hemoglobin, at
least with signifi cant degrees of anemia in younger
subjects experiencing blood loss, iron defi ciency,
hemolysis, or erythroid marrow aplasia Assuming
this represents the normal kinetics of serum EPO in
response to anemia, a curve of the appropriate EPO
concentration for the degree of anemia has been
proposed [32,33] Lack of precision represents a
major obstacle to comparing EPO levels to this
“nor-mal curve.” For example, it remains unknown what
EPO concentration below the predicted level should
be considered low Nevertheless, the predicted EPO
level for a given hemoglobin concentration may
approximate an appropriate EPO response and
per-mit defi ning a relative EPO defi ciency
Renal failure exemplifi es the prototypical
ane-mia from impaired EPO production In many other
conditions, EPO levels are elevated above the
ref-erence range for non-anemic controls but remain
signifi cantly below the expected concentration For
example, an EPO level of 50 mIU/mL at a
hemo-globin of 9.0 g/dL, although far above the reference
range, would still be considered a suboptimal EPO
response This inadequate EPO response for the
degree of anemia commonly occurs with tions such as cancer, HIV, and rheumatoid arthritis [34–36] The mechanism may be multifactorial, but it
condi-is likely that there condi-is a contribution from reduced renal EPO secretion and impaired marrow responsiveness
Erythropoietin levels in the elderly
Non-anemic elderly
Multiple studies have assessed endogenous serum EPO levels in older subjects without anemia (Table 12.1)
Japanese older subjects living in a “home for aged” were compared to younger healthy controls [37] The non-anemic older subjects had numerically but statistically non-signifi cant higher mean EPO levels (mean EPO of 20.4 mIU/mL and 15.7 mIU/
mL for older and younger, respectively) In a study
of older adults with a mean age of 77.8 years ing in a home for the elderly, Kario and colleagues ascertained serum EPO levels [39] An elevated mean EPO level of 26.9 mIU/mL ( 15.2) was observed among the 116 non-anemic elderly, compared to
resid-a meresid-an of 15.8 mIU/mL in younger controls Kresid-ario
et al subsequently reported higher EPO levels
among non-anemic elderly (mean EPO of 24.3 mIU/mL) compared to younger subjects (mean EPO of 14.8 mIU/mL) [40] Also, they showed a signifi cant inverse relationship between EPO and hemoglobin (Hb) concentrations in non-anemic elderly subjects
(r –0.302, p 0.001) Further, a signifi cant
associa-tion existed between advancing age and rising EPO
(r 0.220, p 0.01) These studies provided early
evidence for mildly increased EPO occurring with advancing age and/or slight reductions of hemo-globin concentration, even within the normal range.Powers and colleagues found EPO levels in
25 healthy elderly people, with an age range of 60–82 years, to be similar to those in younger con-trols (means of 10.8 and 13.1 mIU/mL, respectively) [38] In a larger cohort, the same author showed similar EPO levels in younger and older subjects, after adjusting for hematocrit [41] Joosten and
Trang 23160 Andrew S Artz
colleagues quantifi ed endogenous EPO levels in
hospitalized non-anemic elderly patients with a
mean age of 81.5 years and mean hemoglobin of
14.2 g/dL, and compared them to young controls
with a mean hemoglobin of 14.6 g/dL [42] They
showed mean EPO levels did not differ by age
cohort, with levels of 7.5 and 9.5 mIU/mL for older
and younger subjects, respectively These studies by
Powers and Joosten appear contradictory to earlier
data on rising EPO levels with age in those without
anemia However, Powers’ subjects primarily
car-ried a diagnosis of rheumatoid arthritis, a condition
known to blunt EPO responses, and the participants
in the study by Joosten were hospitalized patients,
presumably having other comorbid conditions also
blunting EPO responsiveness Thus, one could not isolate age-related changes in EPO responsiveness
To identify the diurnal variations in endogenous EPO, Pasqualetti and colleagues measured serum EPO in 20 younger adults with a mean age of 45 years and mean hemoglobin of 13.9 g/dL and compared them to older adults with a mean age of 65 years and mean hemoglobin of 13.5 g/dL [46] In general, EPO levels peaked around 6 pm and were at their lowest around midnight Young and old subjects exhibited
a similar diurnal variation, although younger
sub-jects had higher mean daily levels (p 0.05) and
more diurnal variation (p 0.05) in EPO relative
to older subjects The signifi cance of these fi ndings remains unclear, although one may postulate that
Table 12.1 EPO levels in non-anemic elderly adults compared to younger adults.
Mori et al 1988 [37] n 78 (13 with chronic disease, n 127 Elderly mean EPO 20.4 10.4
16 with cancer, 10 without Age 22–46 years Younger 15.7 1.3etiology), home for the aged
Age 70–89 years
Powers et al 1989 [38] n 25 n 30 Elderly mean EPO 10.8 6.4
Age 60–82 years Age 65 years Younger 13.1 5.5
Kario et al 1991 [39] n 116 n 26, volunteers Elderly mean EPO 26.9 15.2
Mean age 77.8 years for all Mean age 42.4 Younger 15.8 EPO rose with age elderly
Kario et al 1992 [40] n 150, ambulatory n 111 Elderly mean EPO 24.3 (17.2–34.5) Age 60 years Mean age 46 years Younger 14.8
Age-related increase in EPO
Nafziger et al 1993 [41] n 30 Not provided Elderly mean EPO 11 3.5
Levels “similar” to non-anemic
Joosten et al 1993 [42] n 27, hospitalized n 30, hospital staff Elderly mean EPO 7.5
Mean age 81.5 years Mean age 32.5 years Younger mean EPO 9.5
Goodnough et al n 31, autologous blood n 40, from same Elderly with higher EPO using
1995 [43] donors prior to surgery population Log EPO and age
Mean age 71.1 years Age 65 years, mean No difference in EPO in subset of
47.6 years 18 patients
Ershler et al 2005 [44] n 143, healthy elderly None Mean EPO 13.0 and rose with
EPO, serum endogenous erythropoietin level (mIU/mL).
Trang 24more tightly regulated EPO in the young resulted in
stronger temporal changes
Endogenous EPO response to anemia
in the elderly
As previously stated, determining a normal EPO
response to hemoglobin changes represents a major
challenge in accurate characterization of EPO
kinet-ics Clearly, young subjects with hemolytic anemia or
subjected to phlebotomy have a brisk rise in serum
EPO levels An absolute EPO defi ciency causes
ane-mia in renal failure, and a relative EPO defi ciency
contributes to anemia in various conditions such as
cancer, autoimmune conditions, prematurity, and
human immunodefi ciency Studies evaluating serum
endogenous EPO in anemic elderly subjects have
allowed a better understanding of changes in serum
EPO and erythropoiesis with aging
Studies in anemic elderly
Mori et al examined iron-defi cient anemic older
subjects from ages 70 to 89 years, and found elevated
endogenous EPO levels with lower hemoglobin
concentrations [37] The preserved EPO
respon-siveness in elderly iron-defi cient anemic subjects
indicated that age does not necessarily confer an
absolute EPO defi ciency
In one of the earliest studies describing a relative
EPO defi ciency in the elderly, Carpenter and
col-leagues detailed hematologic parameters and EPO
concentration for a large group of younger and older
subjects (aged 70 years and over) [47] EPO levels
rose among the iron-defi cient elderly with
worsen-ing anemia In contrast, for the elderly with
normo-cytic anemia (n 375), the EPO response to anemia
was reduced relative to the elderly with microcytic or
macrocytic anemia and relative to all iron-defi cient
anemic subjects Similarly, Kario and others reported
an inverse semilogarithmic relationship between
EPO and Hb concentrations in subjects with Hb
con-centrations less than 12.0 g/dL (r –0.559, p 0.001)
EPO concentrations in the elderly were lower than
those in young subjects with iron-defi ciency anemia
with the same Hb level [39] In a follow-up study, the
same group showed elevated EPO levels with falling hemoglobin, but this was restricted to the elderly with iron-defi ciency anemia The authors suggested that while elderly subjects with iron defi ciency may have a degree of preserved EPO responsiveness
to falling hemoglobin, anemia of advancing age is characterized by a relative EPO defi ciency, and they hypothesized that pharmacologic EPO replacement may reverse the defi cit [40]
Joosten and others investigated hospitalized older subjects and found that among 24 iron-defi cient anemic older persons with a mean age of 83 years and mean hemoglobin of 10.0 g/dL, lower hemo-globin correlated with higher EPO levels [42] How-ever, in the older subjects with anemia due to chronic disease or acute infection, the lack of correlation between EPO and hemoglobin concentration sug-gested a blunted EPO response Whether conditions characterized by infl ammation or aging in this hos-pitalized cohort blunted the EPO response could not
be determined
Among 31 iron-defi cient elderly subjects, Nafziger and colleagues observed the expected inverse correlation, with lower hemoglobin and higher EPO concentrations [41] However, com-pared to 33 subjects less than 60 years of age, the correlation of rising EPO with lower hemoglobin values was less pronounced Moreover, there was
a trend towards lower EPO levels in older defi cient anemic patients relative to younger anemic subjects, reaching signifi cance when hemo-globin concentration fell below 10 g/dL, suggesting that iron defi ciency and severe anemia represent additional stimuli to EPO secretion that remain sub-optimal in the elderly
iron-In a study of older Japanese subjects compared
to patients under 60 years of age, Matsuo and leagues enumerated EPO levels and the immature reticulocyte count, a measure of marrow erythro-poietic response, among iron-defi cient subjects [48] Higher reticulocyte counts refl ect more brisk erythropoiesis Although median EPO levels did not differ in the young and old, the reticulocyte count/log EPO concentration ratio was 3.3 for older
col-and 8.1 for younger subjects (p 0.01), suggesting
Trang 25162 Andrew S Artz
a diminished marrow response to circulating
endogenous EPO levels among the elderly The lack
of adequate iron for red-cell synthesis confounds a
reliable determination of the marrow
erythropoi-etic responsiveness to endogenous EPO secretion
Alternatively, fewer iron stores or reduced capacity
with aging to mobilize iron may have accounted for
reduced erythropoiesis
Pieroni and colleagues studied anemia of chronic
infl ammation in 56 older subjects with cancer,
infec-tion, or another infl ammatory disorder, and
com-pared them to a control group of older subjects In
general, they observed blunted EPO
responsive-ness to anemia, until moderate to severe anemia of
8.5 g/dL [49] The underlying conditions may also
have impaired EPO responsiveness, resulting in
blunted EPO levels Nevertheless, these data, in
con-cert with the study by Joosten and colleagues [42],
demonstrate that markedly low hemoglobin levels
provide a stronger EPO stimulus
Kamenetz and colleagues evaluated 17 elderly
sub-jects with unexplained anemia, ranging in age from
65 to 90 years, and compared them to older
non-anemic controls and subjects with anemia and recent
stroke [50] Anemia was defi ned as a hemoglobin
con-centration 12 g/dL EPO levels did not signifi cantly
differ across the three different groups, refl ecting
impaired EPO responsiveness in the anemic patients,
as one would expect elevated EPO levels in anemic
subjects
In a study of 71 healthy blood donors undergoing
aggressive phlebotomy, Goodnough and colleagues
characterized the endogenous EPO response [43]
Older subjects had similar EPO responsiveness
to phlebotomy Red-cell expansion in response to
phlebotomy showed a non-signifi cant trend for
being reduced in the elderly (p 0.10)
In a case series of patients referred to a
hematol-ogy practice, seven patients were described with
unexplained anemia who later received
pharma-cologic therapy for the anemia At baseline,
hemo-globin and endogenous EPO were 10.0 g/dL and
10.8 mIU/mL, respectively, providing evidence for
a blunted EPO response in patients primarily with
unexplained anemia [51]
We analyzed 900 elderly subjects drawn from fi ve nursing homes to determine anemia prevalence [52] Eighty-one subjects with anemia or borderline low hemoglobin from the cohort were randomly selected for further characterization [53] Among 60 patients with confi rmed anemia, iron defi ciency and unex-plained anemia accounted for 23% and 45% of cases, respectively (Table 12.2) Correlating serum EPO lev-els to hemoglobin concentration revealed increased EPO levels with iron defi ciency but no correlation for seniors with unexplained anemia (Fig 12.1) The rise
in EPO among those with iron defi ciency appeared less than expected, suggesting that a blunted EPO response, albeit to a lesser degree than in unexplained anemia, existed with iron defi ciency In the anemic residents with adequate renal function (glomerular
fi ltration rate 30 mL/minute and serum creatinine
2.0 g/dL), calculated creatinine clearance remained independently associated with lower EPO levels when
controlling for age, gender, and Hb value (p 0.008), suggesting that even mild renal insuffi ciency promotes impaired EPO responsiveness
Although a relative EPO defi ciency for the degree of anemia has been described for numerous conditions, such as cancer, prematurity, human immunodefi -ciency, autoimmune conditions, and the intensive care unit, limited data assess the impact of older age
in these diseases on endogenous EPO ness In a small cohort of 20 cancer patients aged
responsive-70 years and over undergoing platinum-based therapy, Cascinu and others reported mean endog-enous EPO levels similar to those in patients younger than 70 years (51 and 62 mIU/mL, respectively) [54] Absent correlations of serum EPO for hemoglobin, limited inferences regarding age can be made
chemo-EPO responsiveness in the elderly with anemia: conclusions
The studies of anemic older patients permit several conclusions A relative EPO defi ciency characterizes the anemia associated with aging (unexplained ane-mia) Chronic disease does not adequately explain the blunted responses, suggesting EPO defi ciency may be a feature of advancing age and/or occult