Expression of CD99 in Multiple Myeloma: A Clinicopathologic and Immunohistochemical Study of 170 Cases

Abstract

Background: Multiple myeloma (MM) is a heterogeneous and ultimately fatal disease. Risk stratification using prognostic biomarkers is crucial to individualize treatments. We sought to investigate the role of CD99, a transmembrane protein highly expressed in many hematopoietic cells including subpopulations of normal and neoplastic plasma cells, for MM risk stratification. Methods: CD99 expression was measured in paraffin samples of bone marrow and extramedullary biopsies of 170 patients with MM. Patients were divided into those with high score (moderately and strongly positive) and low score (negative and weakly positive), with all staining being cytoplasmic and/or membranous. Results: High anti-CD99 immunostaining was observed in 72 of 136 (52.9%) bone marrow biopsies and 24 of 87 (27.6%) extramedullary biopsies in MM. High CD99 expression of extramedullary specimens was associated with significantly longer overall survival (OS; p=.016). High CD99 expression of extramedullary specimens was also associated with better prognosis in the nonautologous stem cell transplantation group of MM patients (p=.044). In multivariate analysis, International Staging System stage was an independent prognostic factor, whereas CD99 expression was no longer statistically significant. Conclusions: Expression of CD99 in extramedullary specimens was correlated with longer OS, suggesting that CD99 may be a helpful immunohistochemical marker for risk stratification.

Headnote

Background: Multiple myeloma (MM) is a heterogeneous and ultimately fatal disease. Risk stratification using prognostic biomarkers is crucial to individualize treatments. We sought to investigate the role of CD99, a transmembrane protein highly expressed in many hematopoietic cells including subpopulations of normal and neoplastic plasma cells, for MM risk stratification. Methods: CD99 expression was measured in paraffin samples of bone marrow and extramedullary biopsies of 170 patients with MM. Patients were divided into those with high score (moderately and strongly positive) and low score (negative and weakly positive), with all staining being cytoplasmic and/or membranous. Results: High anti-CD99 immunostaining was observed in 72 of 136 (52.9%) bone marrow biopsies and 24 of 87 (27.6%) extramedullary biopsies in MM. High CD99 expression of extramedullary specimens was associated with significantly longer overall survival (OS; p=.016). High CD99 expression of extramedullary specimens was also associated with better prognosis in the nonautologous stem cell transplantation group of MM patients (p=.044). In multivariate analysis, International Staging System stage was an independent prognostic factor, whereas CD99 expression was no longer statistically significant. Conclusions: Expression of CD99 in extramedullary specimens was correlated with longer OS, suggesting that CD99 may be a helpful immunohistochemical marker for risk stratification.

Key Words: CD99; Multiple myeloma; Immunostaining; Autologous stem cell transplantation

Multiple myeloma comprises 10% to 15% of patients with hematologic malignancies.1 It is a slowly progressive, but fatal lesion (median survival, 3 to 4 years) without efficient curative regimens established to date. The standard treatment approach for multiple myeloma patients aged ≤ 65 years consists of high- dose chemotherapy followed by autologous stem cell transplan- tation (ASCT).2 The widespread use of ASCT and the introduc- tion of the novel agents bortezomib and the immunomodulato- ry derivatives, thalidomide and lenalidomide, have significantly improved survival of patients with multiple myeloma.3,4 How- ever, patients eventually become resistant to treatment and suc- cumb to this disease.

One of the major obstacles in developing effective treatments for multiple myeloma is the heterogeneity of this disease. Al- though patients often present with the same histological char- acteristics, multiple myelomas have been classified into seven or eight subgroups by gene expression profiling or by cyclin D ex- pression and chromosome translocation patterns.5,6 Each sub- group showed marked variation in clinical characteristics and treatment responses, requiring different therapeutic approach- es.7 Although application of more aggressive treatment strate- gies in select high risk patient groups is crucial, few markers have been identified for subtyping based on risk stratification. Prognostic factors of multiple myeloma include International Staging System (ISS) score,8 extent of organ involvement,9 cyto- genetic abnormalities,10-12 and some hematologic indices.13,14 However, there are only a few immunohistochemical markers in pathological specimens that are known to predict patient outcome.

CD99 is known to be a useful marker in the diagnosis of Ew- ing sarcoma and primitive neuroectodermal tumors.15 It is also expressed in many hematopoietic neoplasms, including lympho- blastic lymphoma/leukemia,16,17 acute myelogenous leukemia,18 anaplastic large cell lymphoma (ALCL),19,20 and diffuse large B- cell lymphoma (DLBCL).21 Furthermore, expression of CD99 has been reported to be significantly higher in DLBCL patients with high risk.21 Additionally, expression of CD99 was found in some plasma cell neoplasms,22 but the prognostic impact of CD99 in these tumors has not yet been fully investigated.

Based on this previous data, we analyzed expression of CD99 in a large number of plasma cell neoplasms and correlated ex- pression with various clinicopathologic parameters and progno- sis. We found that higher CD99 expression in extramedullary specimens was associated with longer overall survival (OS) in patients with multiple myeloma, particularly those who did not undergo ASCT.

MATERIALS AND METHODS

Patients and samples

Samples were collected from 170 patients diagnosed with multiple myeloma (plasma cell myeloma) and 15 patients diag- nosed with plasmacytoma of extramedullary tissue at Asan Medical Center from 2001 to 2008. To minimize the effects of potential confounders, only specimens obtained at initial diag- nosis were collected. The specimens included 136 bone marrow biopsy specimens and 102 extramedullary tissue biopsy speci- mens (87 cases of multiple myeloma and 15 cases of plasmacy- toma) and both bone marrow and extramedullary biopsy speci- mens were available from 53 patients with multiple myeloma. Medical records were reviewed for clinical features, radiologic, laboratory, and pathologic findings, as well as treatment applied and clinical outcomes. Histologic morphology was considered to be of the plasmablastic type when > 30% of tumor cells were plasmablastic (patient characteristics are summarized in Table 1).

Immunohistochemistry and pathologic evaluation

Paraffin-embedded tissue samples were immunohistochemi- cally stained for CD99 using DAKO Envision-plus kits (Dako, Glostrup, Denmark) following manufacturer's protocols. Briefly, formalin-fixed, paraffin-embedded sections with 4-μm thick- ness were de-paraffinized and dehydrated through a graded al- cohol series, and antigen was retrieved by incubation in citrate buffer for 1 hour at 95°C. The sections were incubated with hy- drogen peroxide for 5 minutes and with Dako cytomation pro- tein block for 5 minutes at room temperature. The tissue sam- ples were subsequently incubated with anti-CD99 antibody (1:100, DN16, DiNonA, Suwon, Korea) for 1 hour at room temperature, washed, and incubated with secondary antibody for 1 hour. Antibody binding was visualized by incubation in diaminobenzidine solution for 1 minute, and the sections were counterstained with hematoxylin.

Immunohistochemical staining for CD99 was evaluated semi-quantitatively by assessing both the intensity and percent- age of positive cells. The intensity of membranous or cytoplas- mic staining was scored as negative (0), weakly positive (1), moderately positive (2), and strongly positive (3) (Fig. 1). Scor- ing based on percentage of positive cells was categorized as 0 (<10%), 1 (10% to 40%), 2 (41% to 70%), and 3 (71% to 100%). The final score was obtained by adding the scores for intensity and percentage. Each sample (bone marrow specimen or extramedullary specimen) was subsequently divided into two groups: those with low CD99 expression (score, 0 to 2; negative and weakly positive) and those with high CD99 expression (score, 3 to 6; moderately and strongly positive).

Statistical analysis

Correlations between expression of CD99 and categorical variables (age, sex, ISS stage, laboratory results, and bone lesion) were analyzed using Pearson's χ-square test or Fisher's exact test. Survival was analyzed by the Kaplan-Meier method. OS was defined as the time from the date of diagnosis by bone mar- row or extramedullary tissue biopsy to the date of death, and recurrence-free survival was defined as the time from the date of diagnosis to the date of first recurrence. Patients lost to follow up or who died of other causes were left out of the analysis. Sur- vival curves of two or more groups were compared by the log- rank test and Cox proportional hazards model. A p<.05 was considered statistically significant. All statistical analyses were performed using SPSS ver. 18 (SPSS Inc., Chicago, IL, USA).

Ethical permission

The Institutional Review Board (IRB) of Asan Medical Cen- ter (Seoul, Korea) approved the study protocol and provided all necessary ethical permissions.

RESULTS

Patient characteristics in plasma cell neoplasm

Of the 185 plasma cell neoplasms, 170 cases were multiple myelomas (plasma cell myeloma) and 15 cases were plasmacy- toma, including nine extramedullary (extraosseous) plasmacyto- mas and six solitary plasmacytomas of the bone (solitary osseous plasmacytoma). The mean age of patients with multiple myelo- ma was 60 years (range, 29 to 84 years). The mean ages of pa- tients with extramedullary (extraosseous) and solitary osseous plasmacytomas were 56 years (range, 28 to 74 years) and 57 years (range, 24 to 67 years), respectively. The male to female ratio of patients with multiple myeloma was 1.09:1.

The clinical characteristics of the 170 multiple myeloma pa- tients are summarized in Table 1. Laboratory results, including serum calcium, protein, and β2-microglobulin concentrations and 24-hour urinary protein, differed widely.

CD99 expression in plasma cell neoplasms

Immunohistochemical staining for CD99 showed a membra- nous and/or cytoplasmic staining pattern. Most CD99 positive bone marrow biopsies showed both cytoplasmic and membra- nous patterns, whereas most extramedullary biopsies showed a membranous pattern. In multiple myeloma samples, high anti- CD99 immunostaining was observed in 72 of 136 (52.9%) bone marrow and 24 of 87 (27.6%) extramedullary biopsies and there was no significant difference in CD99 expression (p=.365). Both bone marrow and extramedullary biopsy specimens were available from 53 patients with multiple myeloma, and there was no significant distinction or concordance of CD99 expres- sion pattern between consecutive extramedullary and bone mar- row samples. In plasmacytoma samples, high anti-CD99 immu- nostaining was observed in six out of 15 (40%).

CD99 expression and survival in multiple myeloma patients

The median follow-up period was 999 days (range, 2 to 4,686 days). At the time of analysis, 129 patients (75.9%) with multiple myeloma had died. When the patients were divided into those with low (score, 0 to 2; negative and weakly positive) and high (score, 3 to 9; moderately and strongly positive) CD99 expression, high CD99 expression in extramedullary biopsies of patients with multiple myeloma tended to be associated with low ISS grade (p=.072). No other clinical prognostic markers were associated with CD99 expression in either bone marrow or extramedullary biopsies (Table 2).

In the multiple myeloma patients with extramedullary biop- sies (n=87), OS was significantly longer in the high CD99 ex- pression (p=.016) group. However, OS was not associated with CD99 expression in multiple myeloma patients with bone mar- row biopsies (n=136) (Fig. 2). Univariate analyses of multiple myeloma patients with extramedullary biopsy (n= 87) showed that other factors related to poor OS included older age (p= .044), male sex (p= .022), plasmablastic morphology (p= .043), high ISS stage (p<.0001), and high Durie-Salmon stage (p<.0001). Multivariate analysis showed that ISS stage (serum β 2-microglobulin) was an independent prognostic factor, whereas high CD99 expression in extramedullary biopsies was not statistically significant (p= .864) (Table 3).

We also assessed CD99 expression related to ASCT response. Subdivision of patients with low CD99 expression in extramed- ullary biopsies demonstrated that OS was significantly longer in patients who had undergone ASCT (p=.004). Among patients who did not undergo ASCT, those with high CD99 expression had better OS than those with low CD99 expression (p= .044). Of patients who underwent ASCT, those with high CD99 ex- pression also showed longer OS, but this difference was not sta- tistically significant (p= .476). In patients with high CD99 ex- pression, there was no survival difference between patients who did and did not undergo ASCT (p= .710) (Fig. 3A).

When we assessed OS in patients from whom we obtained bone marrow biopsies, we found that CD99 expression did not significantly affect patient survival, both in patients with (p= .419) and without (p=.332) ASCT. OS was significantly longer in patients who underwent ASCT than in those who did not, ir- respective of CD99 expression (high CD99 group with ASCT vs high CD99 group with no ASCT, p<.0001; low CD99 group with ASCT vs low CD99 with no ASCT, p=.028) (Fig. 3B).

DISCUSSION

Recently, the expression and prognostic impact of CD99 were demonstrated in hematolymphoid neoplasms, including DLB- CL, Burkitt lymphoma, CD30 negative peripheral T-cell lym- phoma, and ALCL.19 We have investigated the prognostic sig- nificance of CD99 expression in multiple myeloma and assessed the utility of CD99 immunostaining as a histologic marker for risk stratification. In evaluating the correlation of CD99 expres- sion with various clinical parameters and patient prognosis, we found that OS in patients with multiple myeloma was signifi- cantly associated with high CD99 expression in extramedullary specimens, particularly in patients without ASCT. Further, low CD99 expression in extramedullary specimens was associated with high ISS stage.

Our results partly correspond with previous studies. CD99 im- munoreactivity has been associated with better prognosis or low- er clinicopathological stages with urothelial carcinoma23 and neu- roendocrine carcinomas of the pancreas, stomach, and lung.24,25 However, in lymphoid malignancy, CD99 positivity was report- ed to be associated with higher risks in patients with DLBCL21 and the non-germinal center B cell-like (GCB) subgroup of dif- fuse large B-cell lymphoma.26 This discrepancy may be due to several reasons. Multiple myelomas are indolent tumors that grow slowly, whereas DLBCLs are aggressive tumors that grow rapidly. The biological function of CD99 may therefore differ in these tumors, as the functions of CD99 are very diverse. CD99 expression was found to induce apoptosis,27 inhibit anchorage- independent growth, and result in anoikis resistance, and high expression of CD99 was found to contribute to the malignant properties of sarcomas by promoting growth.28 In hematopoiet- ic cells, CD99 mediates T-cell interactions and induces apopto- sis,29 whereas, in CD34+ progenitor cells, CD99 promotes cell proliferation and migration.30 Differences in the prognostic as- sociations of CD99 may be due to differences in CD99 signal- ing pathways among cancer cell type. The function of CD99 and its signaling pathway are not completely understood, indi- cating the need for further studies of the function and biologi- cal significance of CD99 in multiple myeloma and plasma cells.

We also observed that the association between CD99 expres- sion and patient survival of multiple myeloma differed in bone marrow and extramedullary specimens. There was no significant association between CD99 expression and patient survival in bone marrow samples of multiple myeloma. Discordant prog- nostic association between CD99 expression and different cancer subtypes has been reported previously. For example, CD99 ex- pression was found to correlate with OS in patients with the GCB subtype of DLBCL, but showed an inverse correlation in patients with the non-GCB subtype.26 According to 2008 World Health Organization (WHO) classification of plasma cell neo- plasms, extramedullary involvement of multiple myeloma is generally a manifestation of advanced disease, possibly represent- ing different biological properties.13 CD99 may be involved dif- ferently in myeloma-stromal interactions in medullary and extra- medullary sites.

Interestingly, we found that ASCT had no survival benefit in multiple myeloma patients with high CD99 expression in ex- tramedullary specimens, whereas ASCT significantly improved OS in multiple myeloma patients with low CD99 expression in extramedullary specimens. By contrast, ASCT significantly en- hanced OS in multiple myeloma patients with both high and low CD99 expression of bone marrow specimens. Aside from age, there are no criteria for selecting candidates for ASCT. Our finding that ASCT is associated with different survival out- comes in patients with high and low extramedullary expression of CD99 suggests that CD99 immunostaining in extramedul- lary specimens may be helpful in selecting patients for ASCT.

In conclusion, we have shown that extramedullary CD99 ex- pression in multiple myeloma correlates with good prognosis. CD99 may be a useful immunohistochemical marker in risk stratification.

Conflicts of Interest

No potential conflict of interest relevant to this article was reported.

Acknowledgments

This study was supported by the Korea government (The na- tional Research Foundation of Korea, MRC grant no. 2008- 0062286) and the Asan Institute for Life Sciences (no. 2011- 0794, Seoul, Korea).

References

REFERENCES

1. Jemal A, Siegel R, Ward E, Murray T, Xu J, Thun MJ. Cancer statis- tics, 2007. CA Cancer J Clin 2007; 57: 43-66.

2. Kumar L, Verma R, Radhakrishnan VR. Recent advances in the management of multiple myeloma. Natl Med J India 2010; 23: 210-8.

3. Cavo M, Baccarani M. The changing landscape of myeloma thera- py. N Engl J Med 2006; 354: 1076-8.

4. Kumar SK, Rajkumar SV, Dispenzieri A, et al. Improved survival in multiple myeloma and the impact of novel therapies. Blood 2008; 111: 2516-20.

5. Bergsagel PL, Kuehl WM, Zhan F, Sawyer J, Barlogie B, Shaugh- nessy J Jr. Cyclin D dysregulation: an early and unifying pathogen- ic event in multiple myeloma. Blood 2005; 106: 296-303.

6. Zhan F, Huang Y, Colla S, et al. The molecular classification of mul- tiple myeloma. Blood 2006; 108: 2020-8.

7. Barlogie B, Shaughnessy J, Munshi N, Epstein J. Plasma cell myelo- ma. In: Beutler E, Lichtman M, Coller BS, Kipps TJ, Seligsohn U, eds. Williams hematology. 6th ed. New York: McGraw-Hill, 2001; 1279-304.

8. Inamoto Y, Kurahashi S, Imahashi N, et al. Combinations of cytoge- netics and international scoring system can predict poor prognosis in multiple myeloma after high-dose chemotherapy and autolo- gous stem cell transplantation. Am J Hematol 2009; 84: 283-6.

9. Kleber M, Ihorst G, Udi J, Koch B, Wäsch R, Engelhardt M. Prog- nostic risk factor evaluation in patients with relapsed or refractory multiple myeloma receiving lenalidomide treatment: analysis of re- nal function by eGFR and of additional comorbidities by comorbid- ity appraisal. Clin Lymphoma Myeloma Leuk 2012; 12: 38-48.

10. Gutiérrez NC, Castellanos MV, Martín ML, et al. Prognostic and bi- ological implications of genetic abnormalities in multiple myeloma undergoing autologous stem cell transplantation: t(4;14) is the most relevant adverse prognostic factor, whereas RB deletion as a unique abnormality is not associated with adverse prognosis. Leukemia 2007; 21: 143-50.

11. Rajkumar S, Fonseca R, Lacy M, et al. Abnormal cytogenetics pre- dict poor survival after high-dose therapy and autologous blood cell transplantation in multiple myeloma. Bone Marrow Transplant 1999; 24: 497-503.

12. Fonseca R, Blood E, Rue M, et al. Clinical and biologic implications of recurrent genomic aberrations in myeloma. Blood 2003; 101: 4569-75.

13. Swerdlow SH, Campo E, Harris NL, et al. WHO classification of tu- mours of haematopoietic and lymphoid tissues. 4th ed. Lyon: IARC Press, 2008; 202.

14. Stewart AK, Bergsagel PL, Greipp PR, et al. A practical guide to de- fining high-risk myeloma for clinical trials, patient counseling and choice of therapy. Leukemia 2007; 21: 529-34.

15. Hamilton G, Fellinger EJ, Schratter I, Fritsch A. Characterization of a human endocrine tissue and tumor-associated Ewing's sarcoma antigen. Cancer Res 1988; 48: 6127-31.

16. Pan Y, Liu WP, Li JF, et al. A clinicopathological study of 96 cases of lymphoblastic lymphoma. Zhonghua Xue Ye Xue Za Zhi 2005; 26: 218-22.

17. Riopel M, Dickman PS, Link MP, Perlman EJ. MIC2 analysis in pe- diatric lymphomas and leukemias. Hum Pathol 1994; 25: 396-9.

18. Zhang PJ, Barcos M, Stewart CC, Block AW, Sait S, Brooks JJ. Im- munoreactivity of MIC2 (CD99) in acute myelogenous leukemia and related diseases. Mod Pathol 2000; 13: 452-8.

19. Buxton D, Bacchi CE, Gualco G, et al. Frequent expression of CD99 in anaplastic large cell lymphoma: a clinicopathologic and immuno- histochemical study of 160 cases. Am J Clin Pathol 2009; 131: 574-9.

20. Sung CO, Ko YH, Park S, Kim K, Kim W. Immunoreactivity of CD99 in non-Hodgkin's lymphoma: unexpected frequent expres- sion in ALK-positive anaplastic large cell lymphoma. J Korean Med Sci 2005; 20: 952-6.

21. Lee SP, Park S, Park J, Hong J, Ko YH. Clinicopathologic character- istics of CD99-positive diffuse large B-cell lymphoma. Acta Haema- tol 2011; 125: 167-74.

22. Park CK, Shin YK, Kim TJ, Park SH, Ahn GH. High CD99 expres- sion in memory T and B cells in reactive lymph nodes. J Korean Med Sci 1999; 14: 600-6.

23. Xuan Y, Kim S, Lin Z. Protein expression and gene promoter hy- permethylation of CD99 in transitional cell carcinoma of urinary bladder. J Cancer Res Clin Oncol 2011; 137: 49-54.

24. Goto A, Niki T, Terado Y, Fukushima J, Fukayama M. Prevalence of CD99 protein expression in pancreatic endocrine tumours (PETs). Histopathology 2004; 45: 384-92.

25. Pelosi G, Fraggetta F, Sonzogni A, Fazio N, Cavallon A, Viale G. CD99 immunoreactivity in gastrointestinal and pulmonary neuro- endocrine tumours. Virchows Arch 2000; 437: 270-4.

26. Hong J, Park S, Park J, et al. CD99 expression and newly diagnosed diffuse large B-cell lymphoma treated with rituximab-CHOP im- munochemotherapy. Ann Hematol 2012; 91: 1897-906.

27. Scotlandi K, Baldini N, Cerisano V, et al. CD99 engagement: an ef- fective therapeutic strategy for Ewing tumors. Cancer Res 2000; 60: 5134-42.

28. Kreppel M, Aryee DN, Schaefer KL, et al. Suppression of KCMF1 by constitutive high CD99 expression is involved in the migratory ability of Ewing's sarcoma cells. Oncogene 2006; 25: 2795-800.

29. Kim MK, Choi YL, Kim MK, et al. MHC class II engagement inhib- its CD99-induced apoptosis and up-regulation of T cell receptor and MHC molecules in human thymocytes and T cell line. FEBS Lett 2003; 546: 379-84.

30. Imbert AM, Belaaloui G, Bardin F, Tonnelle C, Lopez M, Chaban- non C. CD99 expressed on human mobilized peripheral blood CD34+ cells is involved in transendothelial migration. Blood 2006; 108: 2578-86.