790
Prevalence of Vitamin B12 Deficiency in Patients with Plasma Cell Dyscrasias A Retrospective Review
Rachid Baz, M.D.1 Carlos Alemany, M.D.2 Ralph Green, M.D.3 Mohamad A. Hussein,
M.D.
2
1 Internal Medicine Residency Program, The Cleveland Clinic Foundation, Cleveland, Ohio. 2
The Cleveland Clinic Myeloma Research Program, The Cleveland Clinic Foundation, Cleveland, Ohio.
3
Department of Medical Pathology, University of California Davis, Davis, California.
BACKGROUND. To the authors’ knowledge, the prevalence of vitamin B12 deficiency among patients with plasma cell dyscrasias (PCD) is largely unknown. Identifying this vitamin deficiency in such patients could help improve their anemia and increase their tolerance to potentially neurotoxic agents. METHODS. The authors retrospectively reviewed the charts and laboratory results of 664 consecutive patients diagnosed with PCD who had their vitamin B12 and folate status evaluated between 1997 and 2001 at the Cleveland Clinic Multiple Myeloma Research Program (Cleveland, OH). The patients were screened for vitamin B12 deficiency using serum vitamin B12 and methylmalonic acid. RESULTS. Of the 664 patients whose medical charts were reviewed, information on vitamin B12 status was available for 522 patients (78%). Among these 522 patients, 71 (13.6%) had laboratory-defined vitamin B12 deficiency and the remaining 451 patients (86.4%) did not. On univariate analysis, vitamin B12 deficiency correlated with immunoglobulin A (IgA) PCD (P ⫽ 0.04), higher mean corpuscular volume (P ⫽ 0.008), and longer follow-up (P ⫽ 0.048). In a covariate adjusted model, only the presence of IgA PCD was associated with an increased prevalence of vitamin B12 deficiency (P ⫽ 0.003). CONCLUSIONS. Vitamin B12 deficiency was prevalent in patients with PCD, especially in patients with the IgA subtype. Serum vitamin B12 measurements should be part of the initial evaluation and subsequent workups for anemia in patients with PCD. Cancer 2004;101:790 –5. © 2004 American Cancer Society.
KEYWORDS: vitamin B12, immunoglobulin A, plasma cell dyscrasia, multiple myeloma, monoclonal gammopathy of undetermined significance.
T
Presented in poster form at the 44th Annual Meeting of the American Society of Hematology, Philadelphia, December 2002. Supported in part by The Myeloma Foundation of America and by The Pastore foundation. Address for reprints: Mohamad A. Hussein, M.D., The Cleveland Clinic Multiple Myeloma Research Program, The Cleveland Clinic Taussig Cancer Center, 9500 Euclid Avenue, R35, Cleveland, OH 44195; Fax: (216) 445-3434; E-mail:
[email protected] Received February 23, 2004; revision received May 11, 2004; accepted May 18, 2004.
o our knowledge, the prevalence of vitamin B12 deficiency among patients with plasma cell dyscrasias (PCD) is largely unknown. In 1962, Larsson1 estimated that the prevalence of pernicious anemia in patients with multiple myeloma (MM) ranged from 4.3% to 5.8%. Case reports from 1970 further strengthened the association between MM, or monoclonal gammopathy of undetermined significance (MGUS), and vitamin B12 deficiency. However, these case reports were comprised of small sample sizes and lacked systematic diagnostic criteria for pernicious anemia.2–7 In 1993, Hsing et al.8 noted an increased incidence of MM among female patients with pernicious anemia, which further strengthened the association. Although pernicious anemia is believed to be the most common cause of vitamin B12 deficiency, the latter can occur separately.9,10 The laboratory diagnosis of vitamin B12 deficiency is not a straightforward one and is often made only when patients exhibit signs and symptoms, which can be particularly common in patients
© 2004 American Cancer Society DOI 10.1002/cncr.20441 Published online 7 July 2004 in Wiley InterScience (www.interscience.wiley.com).
Vitamin B12 Deficiency in PCD/Baz et al.
with PCD. Anemia and neuropathy are frequently associated with both vitamin B12 deficiency and PCD.11–16 Identifying vitamin B12 deficiency in patients with PCD could help to improve their anemia and increase their tolerance to potentially neurotoxic agents. The incidence of symptomatic neuropathy associated with thalidomide is approximately 20%, and the incidence of Grade 3 neuropathy associated with bortezomib was approximately 17% in a recent Phase II trial.17,18 The use of potentially neurotoxic agents for MM makes it especially important to eliminate background causes of neuropathy such as vitamin B12 deficiency. To our knowledge, the prevalence of vitamin B12 deficiency in patients with PCD has not been evaluated in a large study. Therefore, we retrospectively reviewed the charts and laboratory results of patients diagnosed with PCD who had their vitamin B12 and folate status evaluated at The Cleveland Clinic Foundation (Cleveland, OH) over a 5-year period.
MATERIALS AND METHODS We retrospectively reviewed the charts of 664 consecutive patients with PCD who were enrolled at the Cleveland Clinic Multiple Myeloma Research Program between 1997 and 2001. Patients with a diagnosis of PCD had levels of serum folate, red blood cell folate, serum vitamin B12, serum methylmalonic acid (MMA), and plasma homocysteine measured as part of their initial evaluation and as indicated thereafter. Laboratory measurements were performed in the routine clinical laboratory at the Cleveland Clinic Foundation. Between 1997 and January 1998, folate and vitamin B12 assays were performed by microparticle enzyme immunoassay and ion capture, respectively. Since 1998, these assays have been performed by electrochemiluminescent immunoassay (ACS 180). Serum MMA level was measured by gas chromatography mass spectrometry. Before July 1999, plasma homocysteine level was performed by high-performance liquid chromatography. Since July 1999, it was performed by fluorescence polarization immunoassay. We also reviewed standard laboratory results (e.g., myeloma parameters), therapeutic agents, and patient demographic information.
Criteria for Diagnosis of Plasma Cell Dyscrasia Type Patients with MGUS had a monoclonal protein, ⬍ 10% plasma cells in the bone marrow biopsy specimen in the absence of anemia, renal failure, hypercalcemia, or bone lesions. Patients with MM had a monoclonal protein with ⬎ 10% plasma cells in the bone marrow biopsy specimen and at least one of the following:
791
anemia, bone lesion, renal failure, or hypercalcemia. Patients with amyloidosis had amyloid deposits identified in biopsy specimens. Patients with smoldering MM had a monoclonal protein with ⬎ 10% plasma cells in the bone marrow biopsy specimen and did not have renal failure, anemia, or hypercalcemia. Patients with plasma cell leukemia had ⬎ 2000 peripheral plasma cells per microliter in the presence of a monoclonal plasma cell proliferation. Patients with solitary plasmacytoma had histologic evidence of a plasma cell tumor involving the bone in the absence of other bone lesions, anemia, renal failure, hypercalcemia, or monoclonal protein and ⬎ 10% plasma cells in the bone marrow biopsy specimen.
Vitamin B12 Status A laboratory diagnosis of vitamin B12 deficiency was made when the serum vitamin B12 level was ⬍ 200 pg/mL or between 200 and 300 pg/mL and the serum MMA level was elevated in the absence of renal impairment (defined as serum creatinine level ⬎ 1.4 mg/dL). Three patients were diagnosed with vitamin B12 deficiency before their evaluation at the Cleveland Clinic and were prescribed B12 supplementation by the referring physician. Therefore, their initial values at our institution were based on B12 supplementation, and they were considered to have laboratory-defined vitamin B12 deficiency in the context of the current study. Vitamin B12 deficiency was considered unlikely when the serum vitamin B12 level was ⬎ 300 pg/mL or between 200 and 300 pg/mL and the MMA level was not elevated. The vitamin B12 status was deemed unknown when the necessary laboratory tests were not available for review based on the above criteria. We used the aforementioned laboratory criteria to define vitamin B12 deficiency for three reasons. First, ⬍ 5% of patients with serum vitamin B12 levels ⬎ 300 pg/mL have biochemical evidence of vitamin B12 deficiency (defined as an elevated level of MMA and/or homocysteine responsive to vitamin B12 therapy).11 Second, a serum vitamin B12 level ⬍ 200 pg/mL has a high specificity for vitamin B12 deficiency.13,19,20 Third, patients with indeterminate serum vitamin B12 levels (vitamin B12 level ⬎ 200 pg/mL and ⬍ 300 pg/mL) had their diagnosis of vitamin B12 deficiency confirmed by an elevated levels of MMA and serum homocysteine when available.
Statistical Analysis Continuous variables were compared with the Student t test or the Wilcoxon rank-sum test when appropriate
792
CANCER August 15, 2004 / Volume 101 / Number 4
TABLE 1 Laboratory Values for 71 Patients with Plasma Cell Dyscrasias and B12 Deficiency
Characteristics
B12 < 200 (pg/mL) (n ⴝ 48)
200 < serum B12 < 300 pg/mL and MMA > 375 nmol/L (n ⴝ 20)
Mean serum vitamin B12 (pg/mL) Mean serum MMA (nmol/L) Mean serum homocysteine (mol/L)
160.6 349.5 13.1
250.4 568.7 11.85
Patients receiving vitamin B12 injections (n ⴝ 3) 421.0 361.6 9.75
MMA: methylmalonic acid.
TABLE 2 Patient and Laboratory Data by Type of PCD Characteristics
MGUS (n ⴝ 132)
MM (n ⴝ 320)
Amyloid (n ⴝ 43)
Plasmacytoma (n ⴝ 14)
SMM (n ⴝ 10)
All PCD (N ⴝ 522)a
PCD (%) Age (yrs) Follow-up (mos) Gender (% Male) Ig chain (%) IgG IgM IgA Light-chain lambda (%) Hemoglobin level (g/dL) MCV (fL) Calcium level (mg/dL) -2-microglobulin level (mg/L) Serum B12 level (pg/mL) Serum MMA level (nmol/L) Homocysteine level (mol/L) Vitamin B12 deficiency (%)
25 65.5 32.7 49
61 60.2 36.8 59
8 61.3 25.9 67
3 57.9 33.6 64
2 59.6 53.6 30
61.5 35.0 57
63 18 19 45 12.9 91.4 9.2 2.7 523.6 268.0 12.0 12 (9)
73 4 23 32 11.0 93.7 9.2 5.1 453.7 329.7 12.1 54 (17)
71 14 14 71 12.6 88.8 8.8 3.5 522.12 406.4 11.7 4 (9)
56
89
44 44 14.0 88.2 9.4 2.0 459.5 244.8 13.5 1 (7)
11 22 11.2 91.2 8.9 6.4 621.5 305.0 10.0 0 (0)
70 8 22 37 11.7 92.5 9.2 4.3 483.8 321.5 12.0 71 (13)
MGUS: monoclonal gammopathy of undetermined significance; MM: multiple myeloma; SMM: smoldering multiple myeloma; MCV: mean corpuscular volume; PCD: plasma cell dyscrasia; Ig: immunoglobulin; MMA: methylmalonic acid. a Three patients with plasma cell leukemia were not included, and none was vitamin B12 deficient. Reference ranges for the following variables were hemoglobin, 13.5–17.5 g/dL; mean corpuscular volume, 80–100 fL; serum B12, 221–700 pg/mL; serum methylmalonic acid, 79–376 nmol/L; serum homocysteine, 7.4–15.7 mol/L; and calcium, 8.5–10.5 mg/dL. Laboratory values and age are reported as means.
and are expressed as mean ⫾ standard deviation. Categoric variables were compared with the Pearson chisquare or Fisher exact tests when appropriate and are expressed as proportions. Variables with significant associations on univariate analysis were included for multivariate analysis alongside age and gender. Multivariate model selection was performed using backward and stepwise selection with exit criteria at P ⬍ 0.2. Second-level interactions were investigated in the multivariate model after bivariate analysis using the Cochrane–Mante– Haenszel chi-square test. No interaction was noted. All variables with P ⬍ 0.2 were included in the multivariate analysis. Goodness of fit of the logistic model was satisfied by the Hosmer and Lemeshow test. P ⬍ 0.05 was statistically significant. All analyses were
conducted using SAS statistical software, version 8 (SAS Institute, Inc., Cary, NC).
RESULTS Of the 664 patients whose medical charts were reviewed, information regarding vitamin B12 status was available for 522 patients (78%). The remaining 142 patients (22%) were excluded from the statistical analysis. Among the 522 study participants, 71 (13.6%) had laboratory-defined vitamin B12 deficiency and the remaining 451 patients (86.4%) did not. Table 1 contains the mean serum levels of vitamin B12, MMA, and homocysteine of the 71 patients with vitamin B12 deficiency. Table 2 summarizes patient demographics, myeloma typing, and B12 status by PCD type for all of the
Vitamin B12 Deficiency in PCD/Baz et al.
793
TABLE 3 Univariate Analysis of Predictors of Vitamin B12 Deficiencya Characteristics Age (SD) Gender (% males) Follow-up (mos) MCV (fL) Hemoglobin level (g/dL) Leukocyte count (103/L) Platelet count (103/L) -2-microglobulin level (mg/L) Ig (%) IgA IgG IgM Conventional chemotherapy Melphalan and prednisone (MP) Vincristine, doxorubicin, decadron (VAD) Both MP and VAD Neither VAD nor MP PCD type (%) MGUS SMM MM Plasmacytoma Plasma cell leukemia Amyloid Light chain (% lambda) Treatment type (%) XRT Thalidomide Arsenic trioxide
Patients with vitamin B12 deficiency (n ⴝ 71)
Patients without B12 deficiency (n ⴝ 451)
P value
59.9 (11.5) 56.3 42.9 (49.4) 94.5 (6.8) 11.3 (2.5) 6.2 (3.2) 215.6 (90.0) 4.9 (7.4)
61.8 (12.3) 56.8 33.8 (33.3) 92.2 (6.8) 11.7 (2.2) 6.6 (3.3) 224.3 (101.7) 4.2 (5.7)
⬎ 0.2 ⬎ 0.2 0.048 0.008 0.12 ⬎ 0.2 ⬎ 0.2 ⬎ 0.2
34.5 63.6 1.8
19.8 70.9 9.3
0.04
33.8 9.9 18.3 38.0
23.7 8.4 10.4 57.4
12 (16.9) 0 54 (76.0) 1 (1.4) 0 4 (5.6) 35.4
120 (26.7) 10 (2.2) 266 (59.1) 13 (2.9) 2 (0.4) 39 (8.7) 36.8
⬎ 0.2
28.6 8.7 4.3
27.5 12.9 2.0
⬎ 0.2 ⬎ 0.2 ⬎ 0.2
⬎ 0.2
0.19
SD: standard deviation; MCV: mean corpuscular volume; Ig: immunoglobulin; PCD: plasma cell dyscrasia; MGUS: monoclonal gammopathy of undetermined significance; MM: multiple myeloma; SMM: smoldering multiple myeloma; XRT: radiotherapy. a Laboratory values and age are reported as means.
522 study participants. Although patients with MM had a higher prevalence of vitamin B12 deficiency compared with patients with MGUS or amyloidosis (17% vs. 9% and 9%, respectively), this difference was not statistically significant (P ⫽ 0.19). The mean corpuscular volume (MCV) was of particular interest. For example, 17 of the 522 study participants (3.3%) had a MCV ⬍ 80 fL whereas 60 participants (11.6%) had an MCV ⬎ 100 fL. Table 3 reports the results of the univariate analysis. A statistically higher MCV was noted among the patients with vitamin B12 deficiency compared with those without (94.5 fL vs. 92.2 fL, respectively; P ⫽ 0.008). However, only 15 of the 60 patients with macrocytosis had vitamin B12 deficiency whereas no patient with microcytic MCV had vitamin B12 deficiency. Univariate analysis also revealed a statistically greater prevalence of vitamin B12 deficiency among the patients who had immunoglobulin (Ig)A PCD compared
TABLE 4 Multivariate Analysis of Predictors of Vitamin B12 Deficiency Characteristics
Odds ratio (95% confidence interval)
P valuea
Age Gender Follow-up time Calcium MCV Light chain (lambda vs. kappa) Heavy chain (IgG or IgM vs. IgA)
0.99 (0.96–1.02) 0.82 (0.43–1.55) 1.00 (0.99–1.01) 0.74 (0.51–1.08) 1.03 (0.98–1.08) 1.69 (0.83–3.43) 0.37 (0.19–0.72)
0.55 0.53 0.10 0.12 0.22 0.16 0.003
MCV: mean corpuscular volume; Ig; immunoglobulin. a The P value reflects the influence of the variable on the prevalence of vitamin B12 deficiency.
with those who had IgG or IgM PCD (22.3% vs. 11.8%, respectively; P ⫽ 0.014). In addition to the MCV, two other variables were associated with an increased prevalence of definite B12 deficiency, namely, longer follow-up time and IgA PCD (vs. IgG or IgM PCD).
794
CANCER August 15, 2004 / Volume 101 / Number 4
In the covariate adjusted model, only the presence of IgA PCD retained statistical significance (Table 4).
DISCUSSION The purpose of the current study was to determine the prevalence of vitamin B12 deficiency among patients with PCD. The results showed that 13.6% of our patients with PCD had this vitamin deficiency. Although this estimate is considerably higher than that reported in the study by Larsson,1 the Larsson study was comprised of a population that was North European (as opposed to our North American patients) and was smaller (69 patients vs. 522 patients). In addition, Larsson evaluated the prevalence of pernicious anemia alone and not of vitamin B12 deficiency in general. The clinical diagnosis of vitamin B12 deficiency is based on laboratory results, clinical symptoms, and the finding of a potential cause for such a deficiency. The current study evaluated only the laboratory diagnosis of vitamin B12 deficiency. Although this methodology may result in an estimate that is greater than the number of patients with a clinical deficiency, the clinical diagnosis of vitamin B12 deficiency may be masked by the presence of the PCD. Therefore, detection must depend on the laboratory determination. Although there is no clear etiology for the increased prevalence of vitamin B12 deficiency among patients with PCD, several hypotheses might be reasonable to consider. Herbert21 proposed that MM paraproteins interfere with the assay for serum vitamin B12, yielding lower results and potentially identifying a greater number of patients with low vitamin B12 results but without a true vitamin deficiency. The interpretation of serum levels of MMA and homocysteine, when available, overcomes this possible limitation in the absence of renal impairment. Vitamin B12 deficiency and PCD both become more prevalent with increasing age. However, in the current study, age was not associated with a higher prevalence of vitamin B12 deficiency. Vitamin B12 deficiency is most commonly the result of an autoimmune antibody-mediated process (pernicious anemia), and PCD involves abnormal clonal production of Ig.6,22 Indeed, monoclonal proteins have been found on occasion to be directed against human IgG, streptolysin O, and lipoprotein (a).23 It is therefore possible that an M protein could have antiintrinsic factorlike activity or may in some other way interfere with the normal vitamin B12 absorptive process. The association between IgA PCD and vitamin B12 deficiency would support such a hypothesis because IgA is the secretory form of Ig. The association seems to suggest a role for IgA in the pathogenesis of the vitamin B12
deficiency. Further characterization of the Ig may help to further characterize this finding. Malignant plasma cells may more rapidly consume the body’s store of vitamin B12 and, hence, increase the likelihood that a patient develops vitamin B12 deficiency.6 Bone marrow-derived MM cells were shown to have increased uptake and accumulation of vitamin B12 in culture.24 If this is the mechanism responsible for B12 deficiency, then one would expect a higher prevalence of vitamin B12 deficiency among patients who have MM and among patients who have larger myeloma burden compared with MGUS. This was not the case in the current study as neither type of PCD nor plasma cell burden was associated with a higher prevalence of vitamin B12 deficiency. Traditionally, anemia, macrocytosis, and/or neuropathy are the signs and symptoms that signal a need for a workup that results in the diagnosis of vitamin B12 deficiency. In the current study, macrocytosis and anemia were not associated with an increased prevalence of vitamin B12 deficiency. In addition, when patients with PCD develop neuropathy, it is often attributed to their PCD or to the side effects of chemotherapeutic agents used for its therapy. Vitamin B12 deficiency is not primarily considered a causative factor. Conversely, peripheral neuropathy often limits the use of active antimyeloma agents in patients with recurrent or refractory MM. Eliminating confounding neuropathy caused by vitamin B12 deficiency would increase the clinical use and decrease the discontinuation of active antimyeloma agents such as thalidomide, vincristine, arsenic trioxide, and bortezomib. Hyperhomocystinemia is a recognized independent risk factor associated with venous thrombosis. Patients with PCD, and specifically MM, are at greater risk for developing deep venous thrombosis regardless of the treatment regimen, especially those who are being treated with a combination of immunomodulators, high-dose steroids, and anthracyclines.25 Among 51 patients undergoing anticoagulation with warfarin, 56% had an elevated serum homocysteine level without assay for vitamin B12 deficiency.26 Vitamin B12 deficiency is emerging as a common, but modifiable, risk factor for hyperhomocysteinemia in patients with PCD. Determining their vitamin B12 status might help to prevent thromboembolism and increase the effectiveness of anticoagulation.27 The current study had several limitations. The patient population is representative of patients referred to a tertiary care facility as evidenced by the higher percentage of patients with MM compared with MGUS. In addition, laboratory methodology changed during the study period. Although this change could theoretically bias the laboratory results, we did not
Vitamin B12 Deficiency in PCD/Baz et al.
find a change in the prevalence of vitamin B12 deficiency before or after the methodology changes. We believe that the change in methodology did not bias the study results. Finally, a cause of the vitamin B12 deficiency was not recorded in the current study. The findings of the current study indicate that vitamin B12 deficiency is prevalent among patients with PCD, especially in patients with IgA as the monoclonal protein. Neither MCV nor the presence of anemia is predictive of this deficiency. Serum vitamin B12 measurement should be part of the initial evaluation of patients with PCD and should not be restricted to patients with anemia, macrocytosis, and unexplained neuropathy. Identifying vitamin B12 deficiency is especially important in patients receiving active therapy for MM that has potential neurotoxic side effects.
REFERENCES Larsson SO. Myeloma and pernicious anemia. Acta Med Scand. 1962;172:195–205. 2. Fraser KJ. Multiple myeloma and pernicious anaemia. Med J Aust. 1969;1:298 –299. 3. Ginsberg A, Mullinax F. Pernicious anemia and monoclonal gammopathy in a patient with IgA deficiency. Am J Med. 1970;48:787–791. 4. Gomez AR, Harley JB. Multiple myeloma and pernicious anemia (case report). W V Med J. 1970;66:38 – 41. 5. Hrncir Z, Mazak J, Mateja F, Vanasek J, Tichy M. Coincidence of pernicious anaemia and IgA-K myeloma. A clinical model of the relation between autoimmune disease and tumour from immunocompetent tissue. Neoplasma. 1970; 17:197–204. 6. Perillie PE. Myeloma and pernicious anemia. Am J Med Sci. 1978;275:93–98. 7. Hoffbrand AV, Hobbs JR, Kremenchuzky S, Mollin DL. Incidence and pathogenesis of megaloblastic erythropoiesis in multiple myeloma. J Clin Pathol. 1967;20:699 –705. 8. Hsing AW, Hansson LE, McLaughlin JK, et al. Pernicious anemia and subsequent cancer. A population-based cohort study. Cancer. 1993;71:745–750. 9. Carmel R. Prevalence of undiagnosed pernicious anemia in the elderly. Arch Intern Med. 1996;156:1097–1100. 10. Carmel R. Subtle and atypical cobalamin deficiency states. Am J Hematol. 1990;34:108 –114. 11. Lindenbaum J, Savage DG, Stabler SP, Allen RH. Diagnosis of cobalamin deficiency: II. Relative sensitivities of serum
12.
13.
14.
15. 16.
17.
18.
1.
19. 20. 21. 22. 23. 24.
25.
26.
27.
795
cobalamin, methylmalonic acid, and total homocysteine concentrations. Am J Hematol. 1990;34:99 –107. Stabler SP, Savage DG, Lindenbaum J. Diagnosis of cobalamin deficiency: II. Relative sensitivities of serum cobalamin, methylmalonic acid, and total homocysteine concentrations. Am J Hematol. 1990;34:90 –98. Matchar DB, McCrory DC, Millington DS, Feussner JR. Performance of the serum cobalamin assay for diagnosis of cobalamin deficiency. Am J Med Sci. 1994;308:276 –283. Ilyas AA, Quarles RH, Dalakas MC, Brady RO. Polyneuropathy with monoclonal gammopathy: glycolipids are frequently antigens for IgM paraproteins. Proc Natl Acad Sci U S A. 1985;82:6697– 6700. Latov N. Monoclonal gammopathy and neuropathy. Ann Neurol. 1992;31:690 – 691. Latov N. Pathogenesis and therapy of neuropathies associated with monoclonal gammopathies. Ann Neurol. 1995;37: S32– 42. Richardson PG, Barlogie B, Berenson J, et al. A phase 2 study of bortezomib in relapsed, refractory myeloma. N Engl J Med. 2003;348:2609 –2617. Ochonisky S, Verroust J, Bastuji-Garin S, Gherardi R, Revuz J. Thalidomide neuropathy incidence and clinico-electrophysiologic findings in 42 patients. Arch Dermatol. 1994;130: 66 – 69. Green R. Metabolite assays in cobalamin and folate deficiency. Baillieres Clin Haematol. 1995;8:533–566. Green R, Kinsella LJ. Current concepts in the diagnosis of cobalamin deficiency. Neurology. 1995;45:1435–1440. Herbert V. Vitamin B-12: plant sources, requirements, and assay. Am J Clin Nutr. 1988;48(3 Suppl.):852– 858. Fraser KJ. IgA immunoglobulins and autoimmunity. Lancet. 1969;2:804 – 805. Potter M. Myeloma proteins (M-components) with antibody-like activity. N Engl J Med. 1971;284:831– 838. Ermens AA, Sonneveld P, Michiels JJ, Lindemans J. Increased uptake and accumulation of cobalamin by multiple myeloma bone marrow cells as a possible cause of low serum cobalamin. Eur J Haematol. 1993;50:57–59. Hussein M, Elson P, Tsoe E, Karam M, Srkaloci G. Doxil (D), vincristine (V), decadron (D) and thalidomide (T) (DVd-T) for relapsed/refractory multiple myeloma (RMM) [abstract 1566]. Blood. 2002;100:403a. Weber D, Ginsberg C, Walker P, et al. Correlation of thrombotic/embolic events (T/E) with features of hypercoagulability in previously untreated patients before and after treatment with thalidomide (T) or thalidomide dexamethasone (TD) [abstract 787]. Blood. 2002;100:209a. Miller JW, Green R, Ramos MI, et al. Homocysteine and cognitive function in the Sacramento area Latino study on aging. Am J Clin Nutr. 2003;78:441– 447.
