Abstract
Background
IgM monoclonal gammopathy can be present in a broad spectrum of diseases. We evaluated the value of serum markers in the differential diagnosis of Waldenstrom macroglobulinemia (WM) and other types of IgM monoclonal gammopathies.
Methods
We included patients who were first admitted to hospital and identified as having IgM monoclonal gammopathy by serum immunofixation electrophoresis (sIFE). We evaluated basic clinical features, sIFE, diagnosis, and serum markers. Furthermore, we applied the receiver operating characteristic (ROC) curve to analyze the differential diagnosis value of serum markers for WM. Finally, we used logistic regression and ROC curve to analyze the differential diagnosis value of multimarker combinations to identify WM.
Results
IgM monoclonal gammopathy was most frequently found in patients with Waldenstrom macroglobulinemia, followed by monoclonal gammopathy of undetermined significance (MGUS), B‐cell non‐Hodgkin Lymphoma (B‐NHL), and multiple myeloma (MM). Serum markers showed significant differences among the four diseases. The diagnostic markers LDH, IgM, IgG, IgA, and serum light chain К had higher diagnostic efficiency. Among these markers, serum IgM provided the highest diagnostic efficiency. Additionally, the combined use of all five serum markers provided the most effective diagnosis.
Conclusions
The five serum markers, LDH, IgM, IgG, IgA, and К, each yielded a specific efficacy in differential diagnosis of WM. The single marker with the highest diagnostic efficiency was the serum IgM level. However, a combination of multiple serum markers was better than the use of a single marker in diagnosing WM. The combined use of all five serum markers provided the most effective diagnosis, with an AUC of .952 and sensitivity and specificity of 87.8% and 86.9%, respectively.
Keywords: IgM monoclonal gammopathy, logistic regression, ROC, serum markers, WM
1. INTRODUCTION
IgM monoclonal gammopathy can be present in a broad spectrum of diseases, including Waldenstrom macroglobulinemia (WM), various B‐cell lymphomas, multiple myeloma (MM), amyloidosis (AL), monoclonal gammopathy of undetermined significance (MGUS), polyneuropathy, organomegaly, endocrinopathy, monoclonal protein, skin changes (POEMS) syndrome, and others.1, 2, 3, 4, 5, 6 It is important to differentially diagnose various IgM monoclonal gammopathies clinically. However, to date, there is little information regarding IgM monoclonal gammopathy because IgM monoclonal gammopathy is clinically less common than IgG and IgA types.6, 7, 8, 9, 10, 11 A single‐center retrospective study of the clinical features of 377 IgM monoclonal gammopathies by the Li Jian team showed that the sensitivity and specificity of serum IgM levels > 15.5g/L, as a marker to differentially diagnose WM, were 80.6% and 89.2%, respectively.12 In our study, we retrospectively analyzed the basic clinical features and laboratory serum markers of 182 patients with IgM monoclonal gammopathy who were initially diagnosed by serum immunofixation electrophoresis (sIFE) in our hospital in order to explore the diagnostic value of laboratory serum markers in the differential diagnosis of WM and other IgM monoclonal gammopathies.
2. METHODS
2.1. Patients
We included patients with IgM monoclonal gammopathy, as determined by sIFE, between January 2008 and September 2017 at Union Hospital of Fujian Medical University in this retrospective study. We collected clinical data, including age, sex, sIFE, diagnosis, and routine laboratory serum analyses. Routine laboratory serum markers included hemoglobin (Hb), albumin (Alb), lactate dehydrogenase (LDH), creatinine (CREA), uric acid (URIC), calcium ion (Ca2+), immunoglobulin (IgM, IgA, IgD), serum total light chain (sLC), and kappa and lambda ratio (К/λ). The study was approved by the ethics review board of the Fujian Medical University Union Hospital, and all participants provided informed consent.
2.2. Statistical analysis
The quantitative data in this study were all of non‐normal distribution and were expressed by median (M) values and interquartile range (p 25‐p 75). The Kruskal‐Wallis rank‐sum test was used for comparison among groups, and the Mann‐Whitney rank‐sum test was used for comparison between two groups. All tests were two‐tailed, and a p‐value of less than .05 was considered statistically significant. SPSS version 22.0 statistical software was used for the rank‐sum test, receiver operating characteristic curve (ROC), and logistic regression analysis. GraphPad Prism 5.0 software was used for data mapping.
3. RESULTS
3.1. IgM monoclonal gammopathies distribution and clinical features
In total, there were 4065 patients with sIFE‐confirmed monoclonal gammopathy between January 2008 and September 2017 at Union Hospital of Fujian Medical University, with 182 patients having the IgM type (4.48%). Of these patients, 142 were male (78.02%) and 40 were female (21.98%); the median patient age was 65 years (range, 26‐85 years). Regarding the type of serum light chain, 131 patients (71.98%) were IgM К and 49 patients (26.92%) were IgM λ; additionally, sIFE results showed that two patients had double clones (IgM К + IgM λ and IgM К + IgG λ). Of the 182 IgM monoclonal gammopathy patients, 66 patients were diagnosed with WM, 51 with MGUS, 41 with B‐cell non‐Hodgkin’s lymphoma (B‐NHL), 15 with multiple myeloma (MM), 3 with peripheral neuropathy (PN), 2 with cryoglobulinemia (Cryo), 2 with myelodysplastic syndrome (MDS), 1 with POEMS, and 1 with angioimmunoblastic T‐cell lymphoma (AITL). We outlined the main clinical characteristics and IgM monoclonal gammopathy distribution in Table 1.
Table 1.
IgM monoclonal gammopathies distribution and clinical characteristics
| IgM MG | N (%) | Gender (M/F) | Ages (years) | Age ≥ 65, N (%) | IgM К/λ (n) | Double clone (n) |
|---|---|---|---|---|---|---|
| Total | 182 (100) | 142/40 | 26‐85 | 74 (40.66) | 131/49 | 2 |
| WM | 66 (36.26) | 55/11 | 44‐85 | 31 (46.97) | 50/16 | 0 |
| MGUS | 51 (28.02) | 35/16 | 41‐83 | 14 (34.15) | 31/20 | 0 |
| B‐NHL | 41 (22.53) | 33/8 | 28‐83 | 38 (92.68) | 29/10 | 2 |
| MM | 15 (8.24) | 12/3 | 53‐72 | 5 (33.33) | 12/3 | 0 |
| PN | 3 (1.65) | 2/1 | 69‐76 | 3 (100) | 3/0 | 0 |
| Cryo | 2 (1.10) | 2/0 | 36, 56 | 1 (50) | 2/0 | 0 |
| MDS | 2 (1.10) | 0/2 | 53, 77 | 1 (50) | 2/0 | 0 |
| POEMs | 1 (.55) | 1/0 | 65 | 1 (100) | 1/0 | 0 |
| AITL | 1(.55) | 1/0 | 67 | 1(100) | 1/0 | 0 |
MG, monoclonal gammopathy; WM, Waldenstrom macroglobulinemia; MGUS, monoclonal gammopathy of undetermined significance; B‐NHL, B‐cell non‐Hodgkin’s lymphoma; MM, multiple myeloma; PN, peripheral neuropathy; Cryo, Cryoglobulinemia; MDS, myelodysplastic syndrome; POEMs, POEM syndrome (polyneuropathy, organomegaly, endocrinopathy, monoclonal protein, and skin changes); AITL, angioimmunoblastic T‐cell lymphoma; К, kappa light chain; λ, lambda light chain.
3.2. Laboratory serum marker analysis
3.2.1. Laboratory serum markers in four different IgM monoclonal gammopathies
IgM monoclonal gammopathy was most frequently found in WM, IgM MGUS, B‐NHL, and MM. We analyzed and compared the laboratory serum markers mentioned above in the four different IgM monoclonal gammopathies. Differences in IgM, LDH, IgA, К, IgG, К/λ, Ca2+, Hb, and λ among the various disease groups are statistically significant. Table 2 shows an outline of the serum marker laboratory analysis in four different IgM monoclonal gammopathies.
Table 2.
Serum markers in four different IgM monoclonal gammopathies
| Serum markers | IgM monoclonal gammopathies | P | |||
|---|---|---|---|---|---|
| WM | MGUS | B‐NHL | MM | ||
| IgM | 46.75 | 4.01 | 19.00 | 11.90 | <.001 |
| (g/L) | (24.08‐68.18) | (2.22‐6.99)* | (7.03‐36.85)* | (.40‐24.30)* | |
| LDH | 123 | 226 | 184 | 199 | <.001 |
| (IU/L) | (105‐172) | (165‐421)* | (154‐309)* | (155‐358) | |
| IgA | 0.61 | 1.98 | 1.36 | 0.87 | <.001 |
| (g/L) | (.39‐1.05) | (1.45‐3.72)* | (.57‐2.23) | (.39‐2.13) | |
| К | 35.60 | 12.55 | 21.40 | 13.00 | <.001 |
| (g/L) | (22.30‐64.50) | (9.58‐15.40)* | (10.70‐27.75) | (6.04‐24.10)* | |
| IgG | 7.85 | 12.90 | 11.80 | 6.50 | <.001 |
| (g/L) | (5.54‐9.73) | (9.87‐16.29)* | (6.85‐20.15)* | (4.84‐8.85) | |
| К/λ | 7.72 | 1.97 | 3.53 | 4.76 | .002 |
| (2.87‐15.81) | (1.33‐3.32) | (2.01‐6.21) | (2.32‐10.56) | ||
| Ca2+ | 2.23 | 2.09 | 2.21 | 2.24 | .004 |
| (mmol/L) | (2.08‐2.33) | (1.97‐2.24)* | (2.09‐2.31) | (2.15‐2.33) | |
| Λ | 4.99 | 5.54 | 6.95 | 2.10 | .004 |
| (g/L) | (3.18‐11.30) | (3.92‐8.98) | (3.46‐7.63) | (1.96‐3.43) | |
| Hb | 82 | 101 | 82 | 93 | .006 |
| (g/L) | (65‐99) | (79‐119)* | (68‐104) | (70‐116) | |
| CRP | 59.95 | 41.30 | 40.53 | 30.20 | .071 |
| (mg/L) | (33.00‐81.95) | (11.30‐76.30) | (25.86‐65.04) | (14.95‐51.95) | |
| Alb | 28.0 | 26.9 | 31.1 | 30.4 | .135 |
| (g/L) | (23.7‐34.2) | (23.0‐34.4) | (26.1‐34.7) | (25.7‐37.9) | |
| PLT | 151 | 199 | 194 | 207 | .550 |
| (×109/L) | (98‐262) | (94‐294) | (111‐279) | (124‐244) | |
| URIC | 384 | 346 | 367 | 365 | .603 |
| (μmol/L) | (320‐475) | (250‐483) | (294‐425) | (284‐413) | |
| CREA | 79 | 83 | 76 | 78 | .604 |
| (μmol/L) | (66‐94) | (62‐157) | (61‐93) | (65‐98) | |
| β2‐MG | 3.08 | 3.79 | 3.73 | 3.14 | .945 |
| (μg/mL) | (2.09‐5.37) | (1.86‐6.27) | (1.99‐4.94) | (1.88‐5.00) | |
Serum markers are expressed by median value and interquartile range (p 25‐p 75). WM, Waldenstrom macroglobulinemia; MGUS, monoclonal gammopathy of undetermined significance; B‐NHL, B‐cell non‐Hodgkin’s lymphoma; MM, multiple myeloma; Hb, hemoglobin; Alb, albumin; LDH, lactate dehydrogenase; CREA, creatinine; URIC, uric acid; Ca2+, calcium ion; β2‐MG, β2‐microglobulin; К, kappa light chain; λ, lambda light chain. * compared with WM disease group, there is a significant difference, p < .05.
3.2.2. Diagnostic value of single serum marker for WM diagnosis
An ROC curve was used to analyze the differential diagnosis value of various serum markers (IgM, LDH, IgA, К, IgG, Ca2+, К/λ, λ, and Hb) for WM diagnosis that showed statistically significant differences among the four different IgM monoclonal gammopathies. As shown in Table 3, the five serum diagnostic markers, IgM, LDH, IgA, К, and IgG, had higher diagnostic efficacy for WM than the other serum markers. The area under the curve (AUC) value was .739‐.885, in which the serum IgM level had the highest AUC of .885 (95% CI .837‐.933, p < .001). Levels of serum IgM > 15.20 g/L were 80.6% sensitive and 76.8% specific for diagnosing WM. Figure 1 shows the level of serum IgM in four different IgM monoclonal gammopathies.
Table 3.
The value of laboratory serum markers in differential diagnosis of WM
| Serum markers | AUC | S.E | p | 95%CI | Cutoff value | Sensitivity (%) | Specificity (%) |
|---|---|---|---|---|---|---|---|
| IgM (g/L) | .885 | .024 | <.001 | .837‐.933 | 14.20 | 91.2 | 76.7 |
| К (g/L) | .828 | .035 | <.001 | .759‐.897 | 30.05 | 65.6 | 89.2 |
| LDH (IU/L) | .816 | .033 | <.001 | .751‐.881 | 134 | 91.5 | 61.5 |
| IgA (g/L) | .787 | .035 | <.001 | .719‐.855 | 1.07 | 70.6 | 77.4 |
| IgG (g/L) | .739 | .038 | <.001 | .664‐.814 | 11.60 | 59.3 | 80.3 |
| К/λ | .651 | .049 | .002 | .555‐.747 | 4.99 | 59.0 | 74.7 |
| Hb (g/L) | .614 | .044 | .012 | .528‐.701 | 95 | 50.0 | 71.2 |
| Ca2+ (mmol/L) | .589 | .045 | .052 | .500‐.677 | 2.30 | 40.0 | 77.1 |
| λ(g/L) | .564 | .053 | .221 | .461‐.668 | 10.20 | 30.5 | 90.5 |
IgM, immunoglobulin M; К, kappa light chain; LDH, lactate dehydrogenase; IgA, immunoglobulin A; IgG, immunoglobulin G; К/λ, kappa and lambda ratio; Hb, hemoglobin; Ca2+, calcium ion; λ, lambda light chain; AUC, area under the curve.
Figure 1.
The levels of serum IgM in different IgM MGs groups
3.2.3. Diagnostic value of combined serum markers for WM
We found that combinations of these five serum markers provided higher diagnostic efficacy (IgM, LDH, IgA, К, and IgG). Logistic regression and ROC curve were used to comprehensively analyze the differential diagnosis value of various combination modes. The WM group comprised the disease group, whereas the other three groups (MGUS, B‐NHL, and MM) served as control groups. Diagnostic analysis was performed using 26 combination patterns of the five serum markers, as shown in Table 4. The AUC results obtained for serum marker combinations were greater than that found using a single serum marker, with the exception of the first four serum combination modes in Table 4 that do not include serum marker IgM. In particular, all five serum marker combinations (IgM + LDH + К + IgA + IgG) had the highest diagnostic sensitivity and specificity, with values of 87.8% and 86.9%, respectively. In combinations involving two markers, IgM + К showed the highest diagnostic efficiency, in which the AUC was .912 (95% CI .866‐.958, p < .001), and the sensitivity and specificity were 81.1% and 90.2%, respectively. When combining three markers, the highest diagnostic efficiency was found for IgM + К + IgA, where the AUC was .933 (95% CI .895‐.972, p < .001). When combining four markers, the highest diagnostic efficiency was for the IgM + К + LDH + IgG mode, where the AUC was .951 (95% CI .920‐.982，p < .001). Moreover, combinations including all five markers were the most effective, with an AUC of .952 (95% CI .922‐.982, p < .001) as well as sensitivity and specificity values of 87.8% and 86.9%, respectively (Table 4 and Figure 2).
Table 4.
Various combined marker modes to the differential diagnosis value of WM
| Combined maker modes | AUC | S.E | p | 95%CI | Sensitivity (%) | Specificity (%) |
|---|---|---|---|---|---|---|
| IgA + IgG | .807 | .034 | <.001 | .741‐.873 | 69.9 | 87.3 |
| LDH + IgG | .860 | .028 | <.001 | .774‐.907 | 70.6 | 90.3 |
| LDH + IgA | .868 | .027 | <.001 | .781‐.911 | 83.3 | 74.2 |
| LDH + IgA + IgG | .881 | .026 | <.001 | .830‐.932 | 81.4 | 82.3 |
| К + IgG | .893 | .026 | <.001 | .841‐.944 | 81.1 | 83.6 |
| К + IgA | .898 | .025 | <.001 | .849‐.947 | 70.3 | 93.4 |
| IgM + IgG | .904 | .022 | <.001 | .861‐.948 | 78.4 | 90.3 |
| К + IgA + IgG | .905 | .024 | <.001 | .858‐.952 | 68.9 | 98.4 |
| IgM + LDH | .908 | .023 | <.001 | .863‐.953 | 78.4 | 90.3 |
| IgM + IgA | .910 | .022 | <.001 | .868‐.952 | 70.6 | 93.5 |
| LDH + К | .912 | .023 | <.001 | .866‐.958 | 81.1 | 90.2 |
| IgM + К | .912 | .023 | <.001 | .866‐.958 | 81.1 | 90.2 |
| IgM + IgA + IgG | .915 | .021 | <.001 | .875‐.955 | 76.5 | 88.7 |
| IgM + К + IgG | .926 | .020 | <.001 | .886‐.966 | 77.0 | 90.2 |
| К + LDH + IgA | .927 | .020 | <.001 | .887‐.967 | 74.3 | 96.7 |
| IgM + LDH + IgA | .928 | .019 | <.001 | .891‐.965 | 85.1 | 87.3 |
| К + LDH + IgG | .929 | .020 | <.001 | .890‐.969 | 85.1 | 83.6 |
| IgM + LDH + IgG | .930 | .018 | <.001 | .894‐.966 | 82.4 | 90.3 |
| IgM + К + LDH | .932 | .021 | <.001 | .891‐.972 | 87.8 | 88.5 |
| IgM + К + IgA | .933 | .020 | <.001 | .895‐.972 | 82.4 | 91.8 |
| К + LDH + IgA + IgG | .933 | .019 | <.001 | .896‐.971 | 87.8 | 82.0 |
| IgM + LDH + IgA + IgG | .935 | .017 | <.001 | .900‐.969 | 80.4 | 93.5 |
| IgM + К + IgA + IgG | .936 | .019 | <.001 | .900‐.973 | 73.0 | 98.4 |
| IgM + К + LDH + IgA | .950 | .016 | <.001 | .917‐.982 | 85.1 | 90.2 |
| IgM + К + LDH + IgG | .951 | .016 | <.001 | .920‐.982 | 86.5 | 88.5 |
| IgM + LDH + К + IgA + IgG | .952 | .015 | <.001 | .922‐.982 | 87.8 | 86.9 |
WM group as the disease group, the other three disease groups (MGUS, B‐NHL, and MM) as the control group; IgM, immunoglobulin M; IgA, immunoglobulin A; IgG, immunoglobulin G; К, kappa light chain; LDH, lactate dehydrogenase.
Figure 2.
ROC curve analysis in various combined serum marker modes for diagnosing WM
4. DISCUSSION
IgM monoclonal gammopathy can be present in a broad range of diseases. However, clinically, IgM monoclonal gammopathy is less common than the IgG and IgA types.13 There are few reports on laboratory serum markers for differential diagnosis IgM monoclonal gammopathy, especially for WM, B‐NHL, and MGUS.4, 14, 15 Some reports suggest that the type of light chain can help with differential diagnosis among IgM monoclonal gammopathies. The presence of the К light chain indicates a diagnosis of WM/sWM, IgM MGUS‐associated peripheral neuropathy (PN), cryoglobulinemia, (Cryo), and pCAD; in contrast, the presence of the λ light chain indicates a diagnosis of POEMS and AL.16, 17, 18, 19 Of the 66 WM patients examined in our study, the kappa/lambda ratio (К/λ) was 3.13 (50/16). Of the 51 patients with MGUS, the К/λ was 1.55 (31/20); this value is similar to that observed in previous studies.19 However, the light chain types observed in 3 cases of PN and 2 cases of Cryo in our study were all К type, instead of λ type. However, this may be attributable to the small number of cases included in our study; to confirm these findings, the number of cases should be expanded in future studies.
IgM monoclonal gammopathy was most frequently detected in WM. It is difficult to distinguish WM from other IgM monoclonal gammopathies due to its atypical clinical manifestations. Therefore, our study aimed to explore the potential application value of laboratory tests, especially basic serum markers for differential diagnosis of WM and other IgM monoclonal gammopathies. Our preliminary study found that IgM, LDH, IgA, К, and IgG markers were more effective in diagnosing WM. However, the level of serum β2‐MG, a risk factor for WM, was not significantly different among IgM monoclonal gammopathies, which is the same as that found in other studies.12 Further analysis showed that the best single serum marker for diagnosing WM was IgM, serum IgM levels ≥ 14.20 g/L were used as the cutoff values for diagnosis WM as opposed to other IgM monoclonal gammopathies, and the diagnostic sensitivity and specificity were 91.2% and 76.7%, respectively. Our findings were different from those reported by the previous studies. Their results indicated a cutoff value of >15.5g/L for serum IgM levels in diagnosing WM, as well as a diagnostic sensitivity and specificity of 80.6% and 89.2%, respectively. Therefore, we believe that there is still a deficiency in WM differential diagnosis, when using a single serum marker. We further examined the effectiveness of using combinations of serum markers for the differential diagnosis of WM, as determined by logistic regression and ROC curve analysis.
Laboratory tests involving a combination of serum markers provide an effective measure for improving the accuracy of disease diagnosis. Logistic regression is a kind of probabilistic nonlinear regression with discriminant and predictive functions. The binary logistic regression method analyzes the combined serum markers to get the predicted probability of WM. The predicted probability used to do the ROC curve analysis integrates the effects of multiple serum markers. IgM, LDH, IgA, К, and IgG, the five serum markers with high diagnostic efficacy (AUC, .739‐.885) for differential diagnosis of WM, were selected for combined serum marker analysis. Among all 26 serum markers combination modes, a combination of five markers, IgA + IgG, LDH + IgG, LDH + IgA, LDH + IgA + IgG, and К + IgG, which did not include serum marker IgM, yielded a lower AUC than that obtained with the single IgM marker. The other 21 serum marker combination modes resulted in a greater AUC than that obtained with single serum marker. In particular, all five serum marker combination modes (IgM + LDH + К + IgA + IgG) provided the highest diagnostic sensitivity and specificity, 87.8% and 86.9%, respectively.
In conclusion, IgM monoclonal gammopathy was most frequently found in WM. The five serum markers of LDH, IgM, IgG, IgA, and К had higher efficacy in the differential diagnosis of WM. Incorporating a combination of multiple serum markers was better than single markers for diagnosing WM. Among single serum markers, IgM was the optimal choice for differential diagnosis of WM. For two serum marker combinations, IgM + К allowed for the best diagnosis. Additionally, IgM + К + IgA was preferred for three serum markers combination mode. However, if possible, all five serum markers IgM, LDH, IgA, К, and IgG should be selected. In addition, we will further study how to differentially diagnose MGUS, B‐NHL, and MM from other IgM monoclonal gammopathies.
ACKNOWLEDGMENT
This study was supported by Joint Funds for the innovation of science and Technology, Fujian province (2017Y9051 and 2017Y91040071), Fujian Medicine Innovation Program (2017‐CX‐20 and 2018‐CX‐18) and the government‐funded project of the construction of high‐level laboratory (Min2017 No.4 File).
Zou H, Yang R, Liao Z‐X, et al. Serum markers in the differential diagnosis of Waldenstrom macroglobulinemia and other IgM monoclonal gammopathies. J Clin Lab Anal. 2019;33:e22827 10.1002/jcla.22827
Contributor Information
Ying‐ping Cao, Email: caoyingping@aliyun.com.
Hui‐fang Huang, Email: huanghuif@126.com.
REFERENCES
- 1. Tedeschi A, Conticello C, Rizzi R, et al. Diagnostic framing of IgM monoclonal gammopathy: focus on Waldenstrom macroglobulinemia. Hematol Oncol. 2018;1‐12. [DOI] [PubMed] [Google Scholar]
- 2. Hiroaki Tanaka YS, Ikeda H, Shimizu R, Sugita Y, Iwai R. Characteristics and prognosis of patients with immunoglobulin M monoclonal gammopathy. J Clin Exp Hematopathol. 2017;57:47‐53. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Stork AC, Cats EA, Vlam L, et al. Classical and lectin complement pathway activity in polyneuropathy associated with IgM monoclonal gammopathy. J Neuroimmunol. 2016;290:76‐79. [DOI] [PubMed] [Google Scholar]
- 4. Andrade‐Campos M, Murillo‐Florez I, Garcia‐Sanz R, et al. Immunoparesis in IgM gammopathies as a useful biomarker to predict disease progression. Clin Chem Lab Med. 2017;55:1598‐1604. [DOI] [PubMed] [Google Scholar]
- 5. Higgins L, Nasr SH, Said SM, et al. Kidney involvement of patients with Waldenstrom macroglobulinemia and other IgM‐producing B cell lymphoproliferative disorders. Clin J Am Soc Nephrol. 2018;13:1037‐1046. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Grunenberg AB, Buske C. Monoclonal IgM gammopathy and Waldenstrom’s macroglobulinemia. Deutsches Arzteblatt Int. 2017;114:745‐751. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Castillo JJ, Jurczyszyn A, Brozova L, et al. IgM myeloma: a multicenter retrospective study of 134 patients. Am J Hematol. 2017;92:746‐751. [DOI] [PubMed] [Google Scholar]
- 8. Pangalis GA, Kyrtsonis MC, Kontopidou FN, et al. Differential diagnosis of Waldenstrom’s macroglobulinemia and other B‐cell disorders. Clin Lymphoma. 2005;5:235‐240. [DOI] [PubMed] [Google Scholar]
- 9. Jiang Ye MJ, Qun Y, et al. Analysis of immunoglobulin levels in patients with lymphoproliferative disorders with serum monoclonal IgM. Acta Universitatis Medicinalis Nanjing(Natrual Science). 2013;33:836‐838. [Google Scholar]
- 10. Zhao L, Xu S, Lin P, et al. Monoclonal IgM disease: a retrospective analysis of 107 cases. Chin J Pract Intern Med. 2016;36:137‐140. [Google Scholar]
- 11. Gustine JN, Meid K, Dubeau T, et al. Serum IgM level as predictor of symptomatic hyperviscosity in patients with Waldenstrom macroglobulinaemia. Br J Haematol. 2017;177:717‐725. [DOI] [PubMed] [Google Scholar]
- 12. Cao XX, Meng Q, Mao YY, et al. The clinical spectrum of IgM monoclonal gammopathy: a single center retrospective study of 377 patients. Leuk Res. 2016;46:85‐88. [DOI] [PubMed] [Google Scholar]
- 13. Pontet F. A data base for 3000 monoclonal immunoglobulin cases and a new classification. Clin Chim Acta. 2005;355:13‐21. [DOI] [PubMed] [Google Scholar]
- 14. Boyle E, Manier S, Lejeune J, et al. IgMkappa and IgMlambda measurements for the assessment of patients with Waldenstrom’s macroglobulinaemia. Clinical Cancer Res. 2016;22:5152‐5158. [DOI] [PubMed] [Google Scholar]
- 15. Sarto C, Cappellini F, Giagnacovo M, et al. IgM kappa‐IgM lambda pair quantitation in the clinical laboratory practice. Clin Biochem. 2018;51:56‐60. [DOI] [PubMed] [Google Scholar]
- 16. Talamo M, Pandey MI, et al. IgM MGUS associated with anti‐MAG neuropathy: a single institution experience. Ann Hematol. 2015;94:1011‐1016. [DOI] [PubMed] [Google Scholar]
- 17. Li J, Zhou DB, Huang Z, et al. Clinical characteristics and long‐term outcome of patients with POEMS syndrome in China. Ann Hematol. 2011;90:819‐826. [DOI] [PubMed] [Google Scholar]
- 18. Berentsen E, Langholm R, et al. Primary chronic cold agglutinin disease: a population based clinical study of 86 patients. Haematologica. 2006;91:460‐466. [PubMed] [Google Scholar]
- 19. Ghobrial IM. Are you sure this is Waldenstrom macroglobulinemia? Hematology. American Soc Hematol Educ Program. 2012;2012:586‐594. [DOI] [PubMed] [Google Scholar]