Hematological Oncology Hematol Oncol 2016; 34: 36–41 Published online 13 October 2014 in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/hon.2159
Case Report
Risk factors for the evaluation of potential central nervous system metastasis in Burkitt’s lymphoma: a case study and literature review Yue-Ting Tang2†, Xiao-Yang Jiao1,2*, Xiao-Lan Chang1 and Dong-Yang Huang1* 1 2
Department of cell biology and genetics, Shantou University Medical College, Guangdong, China Department of Hematology Laboratory, First Affiliated Hospital Of Shantou University Medical College, Guangdong, China
*Correspondence to: Xiao-Yang Jiao, PhD, and Dong-Yang Huang, PhD, Department of Cell Biology and Genetics, Shantou University Medical College, 22 Xinling Road, Shantou, Guangdong, China. E-mail: [email protected]; [email protected] † Yue-Ting Tang and Xiao-Yang Jiao contributed equally to this study.
Received 12 January 2014 Revised 10 May 2014 Accepted 10 July 2014
Abstract Burkitt’s lymphoma (BL) is a malignancy of B lymphocytes. The rapid growth rate and frequent systemic spread result in most patients presenting with advanced disease at diagnosis. Cerebrospinal fluid cytology is the gold standard (with very high accuracy) for diagnosing BL central nervous system (CNS) metastasis; however, the low sensitivity of this method limits its clinical applications. Here, we report a case of BL with CNS metastasis. The levels of vascular endothelial growth factor (VEGF)-A and VEGF-C in the serum and cerebrospinal fluid were used to evaluate the status of BL remission and recurrence. Comparisons were made between VEGF and the other risk factors used in evaluating CNS metastasis. Although not in strict accordance, VEGF levels mirrored the disease course. Therefore, VEGF may reflect the status of BL CNS metastasis. Understanding the role of VEGF in CNS metastasis may help to improve the staging and risk classification of BL as well as the investigation of targeted therapy. Copyright © 2014 John Wiley & Sons, Ltd. Keywords: Central nervous system leukaemia; Burkitt’s lymphoma; VEGF-A; VEGF-C; Cerebrospinal fluid
Introduction Burkitt’s lymphoma (BL) is a hyperproliferative disorder of B lymphocytes, characterized by the over-expression of the c-myc proto-oncogene [1,2]. The rapid growth rate and frequent systemic spread result in 70% to 80% of patients exhibiting advanced stages of disease at diagnosis. Most commonly, the sporadic form is associated with spread to the bone marrow (BM) and/or the central nervous system (CNS) [3]. CNS and BM involvement, advanced age, elevated L-lactate dehydrogenase (LDH) levels, poor response to chemotherapy and cytogenetics are risk factors in evaluating the prognosis of BL. These factors are also used to stratify patients into different therapeutic regimens. However, there are no systematic prospective evaluation systems for BL patients with CNS metastasis. Cerebrospinal fluid (CSF) cytology is the gold standard to diagnose malignant meningitis, and its accuracy is approximately 95% [4]. However, conventional cytology has a low sensitivity because of the paucity of mononuclear cells in CSF [5]. The methods of Flow cytometric immunophenotyping and/or PCR may compensate for this by enhancing the sensitivity. However, a high rate of discordant cytomorphology and PCR results has been reported due to the considerable false-positive rate in PCR [6]. Copyright © 2014 John Wiley & Sons, Ltd.
To date, evidence has suggested the relevance of angiogenesis in lymphoma progression and clinical outcomes, and pathological angiogenesis is increasingly recognized as an important feature of BL pathogenesis [7,8]. Vascular endothelial growth factor (VEGF) was identified as a promoter of tumour angiogenesis and vascular permeability. In animal studies, the finding that vascular permeability parameters detected by dynamic contrast-enhanced magnetic resonance imaging were positively correlated with VEGF expression detected by immunohistochemistry supported the identification of VEGF as a vascular permeability factor [9,10]. Research determined that VEGF levels in the CSF may be a useful marker for both the diagnosis and evaluation of treatment response in carcinomatous meningitis [11]. Therefore, VEGF is thought to participate in BL CNS metastasis. Here, we reported a case of BL with CNS metastasis. The diagnosis of CNS metastasis was according to the observation of BL cells in the CSF, the typical clinical symptoms of meningitis, and the result of MRI. From the early onset of BL through 6-month followup, this patient has undergone three remissions and four CNS recurrences. We collected the samples at each time point and measured the levels of VEGF-A/C (CUSABIO BIOTECH, Newark, New Jersey) in the serum and CSF, as well as the other conventional risk factors, to evaluate their clinical significance in BL CNS metastasis. As the CSF of
Risk factors of CNS metastasis in BL
37
healthy controls was extremely difficult to obtain, control samples in our study were obtained from 10 patients (three males, seven females; median age: 31 ± 5.16 years) with acute injures or tension headache who exhibited no neoplastic, infective, inflammatory, cerebrovascular, neurodegenerative or mental disease. This study was approved by the Ethics Committee of Shantou University Medical College, and informed consent was obtained from the patient prior to the study.
Case report A 50-year-old woman presented with a left cervical lump for 20 days. The painless lump measured approximately 4–5 cm, and the overlying skin was not red. Physical examination also demonstrated dysphonia, but no other abnormalities were present. The patient’s HIV and Epstein–Barr virus tests were negative. The initial complete blood count and the biochemistry parameters were normal. CSF cytological examination was normal, with no tumour cells. Computed tomography and MRI head scans revealed no significant findings. Biopsies of the left cervical mass and lymph nodes were consistent with BL. Histopathological examination revealed distinct starry-sky pattern large cells with marked nucleoli. Immunohistochemical analysis revealed that the tumour cells were positive for CD20, CD79a, CD10 and Bcl-6 and negative for CD3, CD7, CD21, CD5, Bcl2, TDT and CyclinD1. The tumour cells also exhibited strong and diffuse nuclear Ki-67 with a proliferation index of 90%. Therefore, the final diagnosis was made based on the histology and immunophenotyping. A BM aspirate test indicated that there were a few large abnormal lymphoid cells with variation in cell and nuclear size microscopically. The cells exhibited basophilic cytoplasm with numerous vacuoles and multiple small basophilic nucleoli. The patient underwent chemotherapy with a regimen comprised initially of vinblastine, theprubicin, cyclophosphamide and prednisone treatment, followed by a course of doxorubicin, cyclophosphamide, vincristine, prednisone and etoposide and a final regimen of doxorubicin, cyclophosphamide, vincristine and prednisone, resulting in complete remission. Three months after initial presentation, the patient returned with headache, vomiting, cervical stiffness and diplopia. Computed tomography scan revealed a few low-density areas around the right lateral ventricle. MRI revealed gross round-shaped intra-ventricular lesions located in the lateral ventricle, two subcortical areas and the basal nuclei area, exhibiting slight hypo-intensity on both T1-weighted and T2-weighted sequences (Figure 1). Cytology revealed lymphoma cells in the CSF (Figure 2). Routine lab investigation revealed that the serum LDH level and other biochemical indexes were within normal ranges (Table 1). Neurological symptoms disappeared, and CSF biochemical and cytological indicators returned to normal after intravenous high-dose methotrexate and Copyright © 2014 John Wiley & Sons, Ltd.
Figure 1. Magnetic resonance imaging revealed multiple lesions in the brain: gross round-shaped intra-ventricular lesions located in the lateral ventricle, two subcortical areas and the basal nuclei area, exhibiting slight hypointensity on both T1-weighted and T2weighted sequences
Figure 2. Lymphoma cells (on the two sides) and mature lymphocytes (in the middle) in cerebrospinal fluid (CSF) (Wright Staining, ×1000): Cytology of CSF revealed the large abnormal lymphoid cells with variation in cell and nuclear sizes. These cells exhibited loose karyoplasm and small basophilic nucleoli
intrathecal methotrexate treatment. However, after 5 weeks, a first CNS relapse developed. The patient then underwent whole-brain irradiation at a dose of 3400 cGy divided into 20 sessions, plus high-dose methotrexate therapy. The patient achieved remission again but only maintained remission for 1 month. In the next 3 months, the patient experienced numerous CNS relapses and remissions. The VEGF levels in the serum and CSF were monitored throughout the period. Figure 3 demonstrates how VEGF levels and other parameters changed with the remission and recurrence of disease. During the period, the patient’s CSFWBC fluctuated with disease course, remission after chemotherapy, and recurrence (Figure 3C). ALBCSF, Hematol Oncol 2016; 34: 36–41 DOI: 10.1002/hon
Y-T Tang et al.
38
Table 1. General characteristics of the patient Stages
WBCCSF (×106/L) CLCSF (mmol/L) GLUCSF (mmol/L) ALBCSF (g/L) ALBserum(g/L) BBB Value LDHserum (U/L) β2-MGserum (ng/ml) VEGF-ACSF (pg/ml) VEGF-Aserum (pg/ml) VEGF-Cserum (pg/ml) VEGF-A Index VEGF-CCSF (pg/ml)
ID
CNSL1
Remis1
CNSL2
Remis2
CNSL3
Remis3
CNSL4
Control
2 127 3.39 0.03 33.0 0.91 147 2630.5 ---------
360 129 4.56 0.04 29.5 1.36 144 2308.5 33.57 64.56 89.62 383.5
5 120 3.43 0.03 34.3 0.88 180 --52.38 68.74 33.29 874.6
120 124 2.98 0.36 35.7 10.08 277 3696.0 22.23 10.41 40.18 211.8
4 129 3.65 0.21 33.1 6.34 250 --25.22 21.45 51.24 185.3
150 123 3.50 0.16 40.3 3.97 2584 3785.3 218.24 21.97 371.32 2502.0
5 120 3.16 0.08 31.5 2.54 186 --72.25 8.04 22.17 3538.0
80 121 3.16 0.37 28.5 12.98 380 5011.7 174.35 120.59 47.10 111.4
4.2 ± 1.3 123.8 ± 1.4 3.82 ± 0.12 0.102 ± 0.006 40.0 ± 1.11 2.60 ± 0.23 199 ± 17.4 --16.63 ± 2.58 12.99 ± 3.25 58.25 ± 17.58 658.9 ± 113.9
CSF, cerebrospinal fluid; CL, chlorine; ALB albumin; GLU, glucose; WBC, white blood cell; LDH, L-lactate dehydrogenase; BBB, blood brain barrier. ID, initial onset of Burkitt’s lymphoma without relapse in the central nervous system (CNS); CNSL, Burkitt’s lymphoma relapse in CNS; ---, not detected; , below detection Data of the control group are presented as the mean ± SD. CNSL 1/2/3/4, diagnosis of CNSL recurrence for the first/second/third/fourth time. Remis 1/2/3, remission after therapy for the first/second/third time. β2-MG, beta2-microglobulin; BBB Value, ALBCSF × 103/ALBSerum; VEGF-A Index, (VEGF-ACSF/VEGF-ASerum)/(ALBCSF/ALBSerum).
LDHserumβ2-MGserum and the blood brain barrier (BBB) value also fluctuated with the disease course. However, other parameters, such as CLCSF and GLUCSF, remained unchanged (Table 1). The VEGF and LDH levels were elevated when compared with the control group, especially after the second remission (Figure 3B, D, E, F). During the second remission, lymphoma cells disappeared from the CSF after treatment with high-dose methotrexate. However, VEGF levels were not strictly in accordance with this trend, with a slight increase before the second remission. Thereafter, a sudden increase of VEGF-A levels in the serum (120.6 pg/ml) and in the CSF (174.4 pg/ml) was observed during the last recurrence, associated with severe CNS symptoms. At this phase, no clinical improvement was achieved, although the treatment remained the same, representing a turning point in the disease process.
Discussion The lymph node is the most common site for malignant solid tumour metastasis, and lymphangiogenesis plays a crucial role in this process. BL often appears in extranodal sites involving the CNS and is manifested as pachymeningeal or leptomeningeal infiltrates [12]. It is unclear how tumour cells enter the CNS during metastasis. A key event in CNS metastasis is the migration of lymphoma cells through the BBB. Recently, several parameters have been developed to assess the BBB integrity in individuals with CNS metastatic potential. In this study, we used the ratio of CSF/serum albumin as an indicator of BBB integrity [13], in which the quotient of CSF albumin Copyright © 2014 John Wiley & Sons, Ltd.
(Qalbumin) was equal to (albuminCSF) × 103/ (albuminserum) = (Qalbumin). In this case, we observed the patient’s Qalbumin generally fluctuated with the development of CNS symptoms. The BBB value was lower during the remission period than during the relapse phase, suggesting that the integrity of the BBB was impaired. The leaky BBB allows cells (either WBC or tumour cells) to penetrate into the CNS. Therefore, we found that the BBB value was positively associated with the number of WBC/tumour cells in the CSF, suggesting that leakage of the BBB was involved in WBC/tumour cells entry into the CNS. Vascular endothelial growth factor, including VEGF-A and VEGF-C, are potential mediators of endothelial permeability by acting as capillary permeability-enhancing agents. Consequently, VEGF may affect the integrity of the BBB. A previous study found that the loss of BBB integrity was a result of the down-regulation of the tight junction proteins claudin-5 and occludin in response to increased VEGF [14]. VEGF-C has been implicated as a specific regulator of lymphangiogenesis, and VEGF-A has also been demonstrated in recent studies to promote lymphangiogenesis in addition to promoting angiogenesis [15,16]. Therefore, VEGF is essential for angiogenesis, lymphangiogenesis and BBB function [17]. Increased VEGF correlates closely with the increased incidence of regional lymph nodes and distant metastases in both humans and animals [18,19]. To evaluate the role of VEGF in BL CNS metastasis, we assessed VEGF levels in the CSF and serum of our BL patient. Our data indicated that the levels of VEGF-ACSF, VEGF-Aserum and VEGF-Cserum increased with the severity of the disease and were higher corresponding control levels, especially after the second Hematol Oncol 2016; 34: 36–41 DOI: 10.1002/hon
Risk factors of CNS metastasis in BL
39
Figure 3. Vascular endothelial growth factor (VEGF) levels and the other lab data changed with remission and recurrence of disease. The CSFWBC count fluctuated with the disease course, elevating at recurrence and decreasing at remission (C). The levels of L-lactate dehydrogenase, VEGF-A CSF and VEGF-CSerum exhibited elevated tendencies when compared with the control group during the disease course, especially after the second remission (B, D, F). Blood brain barrier value and VEGF-ASerum also exhibited rising tendencies when compared with the control group and varied with remission and recurrence of disease (A, E). In conclusion, although not in strict accordance, VEGF levels fluctuated with the disease course, including the status of remission and recurrence. VEGF may participate in Burkitt’s lymphoma central nervous system metastasis
relapse. However, VEGF-CCSF was below the detection limit (3.90 pg/ml). We further explored the ratio of VEGFCSF/VEGFserum to evaluate its potential as an indicator reflecting the efficiency of peripheral blood VEGF leaking into the CSF and hence affecting BBB integrity. The ratio of VEGFCSF/VEGF serum was calculated in the succeeding text: Vascular endothelial growth factor index = (VEGFCSF/ VEGFserum)/(albuminCSF/albuminserum) [20]. A previous study utilized the VEGF index as an indicator to evaluate the source of VEGFCSF, either from local production or from peripheral blood by passive influx. A high VEGF index suggests that VEGF is predominantly produced intrathecally [11]. In our case, the high VEGF-A index suggested that VEGF-A in the CSF was indeed produced locally. High VEGF-A levels increased vascular permeability Copyright © 2014 John Wiley & Sons, Ltd.
and damaged the BBB integrity, as reflected by the high BBB value. The impaired BBB could allow more molecules and cells to enter into the CNS. At the last recurrence, the VEGF-A index was low, which may due to unusually high VEGF-A production in the serum during the marked deterioration of the patient’s condition. High levels of VEGF-C have been correlated with lymphatic vessel invasion, the emergence of sentinel and distant lymph node metastasis, and overall poor prognosis in tumours [21]. In this case, the median time from onset to CNS metastases was approximately 3 months. The patient had an overall response to high-dose intravenous methotrexate, but relapsed soon after. Given the high VEGF levels and frequent recurrences in this patient, our data suggest a correlation between VEGF and CNS metastasis in BL may exist, and the mechanism should be further explored. Hematol Oncol 2016; 34: 36–41 DOI: 10.1002/hon
Y-T Tang et al.
40
In some cases of CSF cytomorphology tests, lymphoma cells may be confused with atypical lymphocytes caused by reactive lymphocytosis; the latter may be significantly enlarged with blast-like nuclei as observed in lymphoma cells [22], leading to false-positive results. In addition, the low sensitivity of conventional cytology has raised concerns about the quality of cytospin preparations. Lymphoma cells confirmed by morphology were only observed in one of nine cases where 99% of cells are already deemed neoplastic [5]. In this case, WBCs in the CSF fluctuated greatly during the 6-month follow-up. Lymphoma cells increased in the CSF during relapse, and the cells were subsequently cleared with chemotherapy. After treatment, CNS metastasis was largely underestimated because the WBCs in the CSF were less than 5 × 106/L, even in the presence of severe neurological symptoms. The intriguing point was that the CSF was free of tumor cells yet contained high VEGF levels. During the second remission, no WBCs were detected in the CSF, whereas the levels of VEGF and the BBB value remained persistently high, suggesting there may be occult damage in the BBB integrity. The early rise in VEGF levels during disease onset and a positive correlation between VEGF levels and WBCCSF during the later stage suggest that VEGF levels may be useful in predicting disease relapse. VEGF-ACSF may help discriminate patients who may potentially benefit from interventions. In addition to VEGF, other factors can increase the risk of BL metastasis to the CNS, such as LDH. In this case, the patient exhibited elevated levels of serum LDH, and LDH levels were higher during relapse than in remission. There was a trend of LDH elevation during the disease course, indicating the increased risk of CNS involvement in the patient. Although the patient exhibited an increasing risk of BL CNS involvement with the high levels of VEGF-ACSF, VEGF-Aserum and VEGF-Cserum, the mechanisms underlying this relationship must be further studied in more patients. Over the past decade, the well-documented role of VEGF in tumour angiogenesis has led VEGFs to become one of the leading therapeutic targets for the treatment of cancer [23]. Understanding the role of VEGF in CNS metastasis, combined with other indicators, may help to improve staging and risk classification, reduce cytotoxic chemotherapy and promote the investigation of targeted therapy.
Conflict of interest The authors have no competing interest.
Acknowledgements This study was funded by the Natural Science Foundation of Guangdong Province, China (S2012030006289, NO9151008-901000043) and the Science and Technology Program of Guangdong Province, Copyright © 2014 John Wiley & Sons, Ltd.
China (NO2010B031600321, No. 2012B031800217), and we would like to thank Dr Frieda Law for editing the manuscript.
References 1. Blum KA, Lozanski G, Byrd JC. Adult Burkitt leukemia and lymphoma. Blood 2004; 104: 3009–3020. 2. Sandlund JT, Downing JR, Crist WM. Non-Hodgkin’s lymphoma in childhood. N Engl J Med 1996; 334: 1238–1248. 3. Miles RR, Arnold S, Cairo MS. Risk factors and treatment of childhood and adolescent Burkitt lymphoma/leukaemia. Br J Haematol 2012; 156: 730–743. 4. Chamberlain MC, Glantz M, Groves MD, Wilson WH. Diagnostic tools for neoplastic meningitis: detecting disease, identifying patient risk, and determining benefit of treatment. Semin Oncol 2009; 36: S35–S45. 5. Hegde U, Filie A, Little RF, et al. High incidence of occult leptomeningeal disease detected by flow cytometry in newly diagnosed aggressive B-cell lymphomas at risk for central nervous system involvement: the role of flow cytometry versus cytology. Blood 2005; 105: 496–502. 6. Fischer L, Martus P, Weller M, et al. Meningeal dissemination in primary CNS lymphoma: prospective evaluation of 282 patients. Neurology 2008; 71: 1102–1108. 7. Ruan J. Antiangiogenic therapies in non-Hodgkin’s lymphoma. Curr Cancer Drug Targets 2011; 11: 1030–1043. 8. Guedez L, McMarlin AJ, Bennett TA, Bennett TA, StetlerStevenson M, Stetler-Stevenson WG. Tissue inhibitor of metalloproteinase-1 alters the tumorigenicity of Burkitt’s lymphoma via divergent effects on tumor growth and angiogenesis. Am J Pathol 2001; 158: 1207–1215. 9. Li KA, Zhang RM, Zhang F, et al. Studies of pathology and VEGF expression in rabbit cerebrospinal fluid metastasis: application of dynamic contrast-enhanced MRI. Magn Reson Imaging 2011; 129: 1101–1109. 10. Dafni H, Kim SJ, Bankson JA, Sankaranarayanapillai M, Ronen SM. Macromolecular dynamic contrast-enhanced (DCE)-MRI detects reduced vascular permeability in a prostate cancer bone metastasis model following anti-platelet-derived growth factor receptor (PDGFR) therapy, indicating a drop in vascular endothelial growth factor receptor (VEGFR) activation. Magnet Reson Med 2008; 60: 822–833. 11. Stockhammer G, Poewe W, Burgstaller S, et al. Vascular endothelial growth factor in CSF: a biological marker for carcinomatous meningitis. Neurology 2000; 54: 1670–1676. 12. Gu Y, Hou YY, Zhang XB, Hu F. Primary central nervous system Burkitt lymphoma as concomitant lesions in the third and the left ventricles: a case study and literature review. J Neurooncol 2010; 99: 277–281. 13. Hirohata S, Isshi K, Oguchi H, et al. Cerebrospinal fluid interleukin-6 in progressive Neuro-Behcet’s syndrome. Clin Immunol Immunopathol 1997; 82: 12–17. 14. Argaw AT, Gurfein BT, Zhang Y, Zameer A, John GR. VEGFmediated disruption of endothelial CLN-5 promotes blood– brain barrier breakdown. Proc Natl Acad Sci U S A 2009; 106: 1977–1982. 15. Stacker SA, Achen MG, Jussila L, Baldwin ME, Alitalo K. Lymphangiogenesis and cancer metastasis. Nat Rev Cancer 2002; 2: 573–583. 16. Joukov V, Pajusola K, Kaipainen A, et al. A novel vascular endothelial growth factor, VEGF-C, is a ligand for the Flt4 (VEGFR-3) and KDR (VEGFR-2) receptor tyrosine kinases. EMBO J 1996; 15: 1751. Hematol Oncol 2016; 34: 36–41 DOI: 10.1002/hon
Risk factors of CNS metastasis in BL
17. Arshad F, Wang L, Sy C, Avraham S, Avraham HK. Blood– brain barrier integrity and breast cancer metastasis to the brain. Patholog Res Int 2010; 2011: 920509. 18. Skobe M, Hawighorst T, Jackson DG, et al. Induction of tumor lymphangiogenesis by VEGF-C promotes breast cancer metastasis. Nat Med 2001; 7: 192–198. 19. Hirakawa S, Kodama S, Kunstfeld R, Kajiya K, Brown LF, Detmar M. VEGF-A induces tumor and sentinel lymph node lymphangiogenesis and promotes lymphatic metastasis. J Exp Med 2005; 201: 1089–1099. 20. van der Flier M, Stockhammer G, Vonk GJ, et al. Vascular endothelial growth factor in bacterial meningitis: detection in
Copyright © 2014 John Wiley & Sons, Ltd.
41
cerebrospinal fluid and localization in postmortem brain. J Infect Dis 2001; 183: 149–153. 21. Rinderknecht M, Detmar M. Tumor lymphangiogenesis and melanoma metastasis. J Cell Physiol 2008; 216: 347–354. 22. Baraniskin A, Deckert M, Schulte-Altedorneburg G, Schlegel U, Schroers R. Current strategies in the diagnosis of diffuse large Bcell lymphoma of the central nervous system. Br J Haematol 2012; 156: 421–432. 23. Vecchiarelli-Federico LM, Cervi D, Haeri M, Li Y, Nagy A, Ben-David Y. Vascular endothelial growth factor--a positive and negative regulator of tumor growth. Cancer Res 2010; 70: 863–867.
Hematol Oncol 2016; 34: 36–41 DOI: 10.1002/hon
Case Report
Risk factors for the evaluation of potential central nervous system metastasis in Burkitt’s lymphoma: a case study and literature review Yue-Ting Tang2†, Xiao-Yang Jiao1,2*, Xiao-Lan Chang1 and Dong-Yang Huang1* 1 2
Department of cell biology and genetics, Shantou University Medical College, Guangdong, China Department of Hematology Laboratory, First Affiliated Hospital Of Shantou University Medical College, Guangdong, China
*Correspondence to: Xiao-Yang Jiao, PhD, and Dong-Yang Huang, PhD, Department of Cell Biology and Genetics, Shantou University Medical College, 22 Xinling Road, Shantou, Guangdong, China. E-mail: [email protected]; [email protected] † Yue-Ting Tang and Xiao-Yang Jiao contributed equally to this study.
Received 12 January 2014 Revised 10 May 2014 Accepted 10 July 2014
Abstract Burkitt’s lymphoma (BL) is a malignancy of B lymphocytes. The rapid growth rate and frequent systemic spread result in most patients presenting with advanced disease at diagnosis. Cerebrospinal fluid cytology is the gold standard (with very high accuracy) for diagnosing BL central nervous system (CNS) metastasis; however, the low sensitivity of this method limits its clinical applications. Here, we report a case of BL with CNS metastasis. The levels of vascular endothelial growth factor (VEGF)-A and VEGF-C in the serum and cerebrospinal fluid were used to evaluate the status of BL remission and recurrence. Comparisons were made between VEGF and the other risk factors used in evaluating CNS metastasis. Although not in strict accordance, VEGF levels mirrored the disease course. Therefore, VEGF may reflect the status of BL CNS metastasis. Understanding the role of VEGF in CNS metastasis may help to improve the staging and risk classification of BL as well as the investigation of targeted therapy. Copyright © 2014 John Wiley & Sons, Ltd. Keywords: Central nervous system leukaemia; Burkitt’s lymphoma; VEGF-A; VEGF-C; Cerebrospinal fluid
Introduction Burkitt’s lymphoma (BL) is a hyperproliferative disorder of B lymphocytes, characterized by the over-expression of the c-myc proto-oncogene [1,2]. The rapid growth rate and frequent systemic spread result in 70% to 80% of patients exhibiting advanced stages of disease at diagnosis. Most commonly, the sporadic form is associated with spread to the bone marrow (BM) and/or the central nervous system (CNS) [3]. CNS and BM involvement, advanced age, elevated L-lactate dehydrogenase (LDH) levels, poor response to chemotherapy and cytogenetics are risk factors in evaluating the prognosis of BL. These factors are also used to stratify patients into different therapeutic regimens. However, there are no systematic prospective evaluation systems for BL patients with CNS metastasis. Cerebrospinal fluid (CSF) cytology is the gold standard to diagnose malignant meningitis, and its accuracy is approximately 95% [4]. However, conventional cytology has a low sensitivity because of the paucity of mononuclear cells in CSF [5]. The methods of Flow cytometric immunophenotyping and/or PCR may compensate for this by enhancing the sensitivity. However, a high rate of discordant cytomorphology and PCR results has been reported due to the considerable false-positive rate in PCR [6]. Copyright © 2014 John Wiley & Sons, Ltd.
To date, evidence has suggested the relevance of angiogenesis in lymphoma progression and clinical outcomes, and pathological angiogenesis is increasingly recognized as an important feature of BL pathogenesis [7,8]. Vascular endothelial growth factor (VEGF) was identified as a promoter of tumour angiogenesis and vascular permeability. In animal studies, the finding that vascular permeability parameters detected by dynamic contrast-enhanced magnetic resonance imaging were positively correlated with VEGF expression detected by immunohistochemistry supported the identification of VEGF as a vascular permeability factor [9,10]. Research determined that VEGF levels in the CSF may be a useful marker for both the diagnosis and evaluation of treatment response in carcinomatous meningitis [11]. Therefore, VEGF is thought to participate in BL CNS metastasis. Here, we reported a case of BL with CNS metastasis. The diagnosis of CNS metastasis was according to the observation of BL cells in the CSF, the typical clinical symptoms of meningitis, and the result of MRI. From the early onset of BL through 6-month followup, this patient has undergone three remissions and four CNS recurrences. We collected the samples at each time point and measured the levels of VEGF-A/C (CUSABIO BIOTECH, Newark, New Jersey) in the serum and CSF, as well as the other conventional risk factors, to evaluate their clinical significance in BL CNS metastasis. As the CSF of
Risk factors of CNS metastasis in BL
37
healthy controls was extremely difficult to obtain, control samples in our study were obtained from 10 patients (three males, seven females; median age: 31 ± 5.16 years) with acute injures or tension headache who exhibited no neoplastic, infective, inflammatory, cerebrovascular, neurodegenerative or mental disease. This study was approved by the Ethics Committee of Shantou University Medical College, and informed consent was obtained from the patient prior to the study.
Case report A 50-year-old woman presented with a left cervical lump for 20 days. The painless lump measured approximately 4–5 cm, and the overlying skin was not red. Physical examination also demonstrated dysphonia, but no other abnormalities were present. The patient’s HIV and Epstein–Barr virus tests were negative. The initial complete blood count and the biochemistry parameters were normal. CSF cytological examination was normal, with no tumour cells. Computed tomography and MRI head scans revealed no significant findings. Biopsies of the left cervical mass and lymph nodes were consistent with BL. Histopathological examination revealed distinct starry-sky pattern large cells with marked nucleoli. Immunohistochemical analysis revealed that the tumour cells were positive for CD20, CD79a, CD10 and Bcl-6 and negative for CD3, CD7, CD21, CD5, Bcl2, TDT and CyclinD1. The tumour cells also exhibited strong and diffuse nuclear Ki-67 with a proliferation index of 90%. Therefore, the final diagnosis was made based on the histology and immunophenotyping. A BM aspirate test indicated that there were a few large abnormal lymphoid cells with variation in cell and nuclear size microscopically. The cells exhibited basophilic cytoplasm with numerous vacuoles and multiple small basophilic nucleoli. The patient underwent chemotherapy with a regimen comprised initially of vinblastine, theprubicin, cyclophosphamide and prednisone treatment, followed by a course of doxorubicin, cyclophosphamide, vincristine, prednisone and etoposide and a final regimen of doxorubicin, cyclophosphamide, vincristine and prednisone, resulting in complete remission. Three months after initial presentation, the patient returned with headache, vomiting, cervical stiffness and diplopia. Computed tomography scan revealed a few low-density areas around the right lateral ventricle. MRI revealed gross round-shaped intra-ventricular lesions located in the lateral ventricle, two subcortical areas and the basal nuclei area, exhibiting slight hypo-intensity on both T1-weighted and T2-weighted sequences (Figure 1). Cytology revealed lymphoma cells in the CSF (Figure 2). Routine lab investigation revealed that the serum LDH level and other biochemical indexes were within normal ranges (Table 1). Neurological symptoms disappeared, and CSF biochemical and cytological indicators returned to normal after intravenous high-dose methotrexate and Copyright © 2014 John Wiley & Sons, Ltd.
Figure 1. Magnetic resonance imaging revealed multiple lesions in the brain: gross round-shaped intra-ventricular lesions located in the lateral ventricle, two subcortical areas and the basal nuclei area, exhibiting slight hypointensity on both T1-weighted and T2weighted sequences
Figure 2. Lymphoma cells (on the two sides) and mature lymphocytes (in the middle) in cerebrospinal fluid (CSF) (Wright Staining, ×1000): Cytology of CSF revealed the large abnormal lymphoid cells with variation in cell and nuclear sizes. These cells exhibited loose karyoplasm and small basophilic nucleoli
intrathecal methotrexate treatment. However, after 5 weeks, a first CNS relapse developed. The patient then underwent whole-brain irradiation at a dose of 3400 cGy divided into 20 sessions, plus high-dose methotrexate therapy. The patient achieved remission again but only maintained remission for 1 month. In the next 3 months, the patient experienced numerous CNS relapses and remissions. The VEGF levels in the serum and CSF were monitored throughout the period. Figure 3 demonstrates how VEGF levels and other parameters changed with the remission and recurrence of disease. During the period, the patient’s CSFWBC fluctuated with disease course, remission after chemotherapy, and recurrence (Figure 3C). ALBCSF, Hematol Oncol 2016; 34: 36–41 DOI: 10.1002/hon
Y-T Tang et al.
38
Table 1. General characteristics of the patient Stages
WBCCSF (×106/L) CLCSF (mmol/L) GLUCSF (mmol/L) ALBCSF (g/L) ALBserum(g/L) BBB Value LDHserum (U/L) β2-MGserum (ng/ml) VEGF-ACSF (pg/ml) VEGF-Aserum (pg/ml) VEGF-Cserum (pg/ml) VEGF-A Index VEGF-CCSF (pg/ml)
ID
CNSL1
Remis1
CNSL2
Remis2
CNSL3
Remis3
CNSL4
Control
2 127 3.39 0.03 33.0 0.91 147 2630.5 ---------
360 129 4.56 0.04 29.5 1.36 144 2308.5 33.57 64.56 89.62 383.5
5 120 3.43 0.03 34.3 0.88 180 --52.38 68.74 33.29 874.6
120 124 2.98 0.36 35.7 10.08 277 3696.0 22.23 10.41 40.18 211.8
4 129 3.65 0.21 33.1 6.34 250 --25.22 21.45 51.24 185.3
150 123 3.50 0.16 40.3 3.97 2584 3785.3 218.24 21.97 371.32 2502.0
5 120 3.16 0.08 31.5 2.54 186 --72.25 8.04 22.17 3538.0
80 121 3.16 0.37 28.5 12.98 380 5011.7 174.35 120.59 47.10 111.4
4.2 ± 1.3 123.8 ± 1.4 3.82 ± 0.12 0.102 ± 0.006 40.0 ± 1.11 2.60 ± 0.23 199 ± 17.4 --16.63 ± 2.58 12.99 ± 3.25 58.25 ± 17.58 658.9 ± 113.9
CSF, cerebrospinal fluid; CL, chlorine; ALB albumin; GLU, glucose; WBC, white blood cell; LDH, L-lactate dehydrogenase; BBB, blood brain barrier. ID, initial onset of Burkitt’s lymphoma without relapse in the central nervous system (CNS); CNSL, Burkitt’s lymphoma relapse in CNS; ---, not detected; , below detection Data of the control group are presented as the mean ± SD. CNSL 1/2/3/4, diagnosis of CNSL recurrence for the first/second/third/fourth time. Remis 1/2/3, remission after therapy for the first/second/third time. β2-MG, beta2-microglobulin; BBB Value, ALBCSF × 103/ALBSerum; VEGF-A Index, (VEGF-ACSF/VEGF-ASerum)/(ALBCSF/ALBSerum).
LDHserumβ2-MGserum and the blood brain barrier (BBB) value also fluctuated with the disease course. However, other parameters, such as CLCSF and GLUCSF, remained unchanged (Table 1). The VEGF and LDH levels were elevated when compared with the control group, especially after the second remission (Figure 3B, D, E, F). During the second remission, lymphoma cells disappeared from the CSF after treatment with high-dose methotrexate. However, VEGF levels were not strictly in accordance with this trend, with a slight increase before the second remission. Thereafter, a sudden increase of VEGF-A levels in the serum (120.6 pg/ml) and in the CSF (174.4 pg/ml) was observed during the last recurrence, associated with severe CNS symptoms. At this phase, no clinical improvement was achieved, although the treatment remained the same, representing a turning point in the disease process.
Discussion The lymph node is the most common site for malignant solid tumour metastasis, and lymphangiogenesis plays a crucial role in this process. BL often appears in extranodal sites involving the CNS and is manifested as pachymeningeal or leptomeningeal infiltrates [12]. It is unclear how tumour cells enter the CNS during metastasis. A key event in CNS metastasis is the migration of lymphoma cells through the BBB. Recently, several parameters have been developed to assess the BBB integrity in individuals with CNS metastatic potential. In this study, we used the ratio of CSF/serum albumin as an indicator of BBB integrity [13], in which the quotient of CSF albumin Copyright © 2014 John Wiley & Sons, Ltd.
(Qalbumin) was equal to (albuminCSF) × 103/ (albuminserum) = (Qalbumin). In this case, we observed the patient’s Qalbumin generally fluctuated with the development of CNS symptoms. The BBB value was lower during the remission period than during the relapse phase, suggesting that the integrity of the BBB was impaired. The leaky BBB allows cells (either WBC or tumour cells) to penetrate into the CNS. Therefore, we found that the BBB value was positively associated with the number of WBC/tumour cells in the CSF, suggesting that leakage of the BBB was involved in WBC/tumour cells entry into the CNS. Vascular endothelial growth factor, including VEGF-A and VEGF-C, are potential mediators of endothelial permeability by acting as capillary permeability-enhancing agents. Consequently, VEGF may affect the integrity of the BBB. A previous study found that the loss of BBB integrity was a result of the down-regulation of the tight junction proteins claudin-5 and occludin in response to increased VEGF [14]. VEGF-C has been implicated as a specific regulator of lymphangiogenesis, and VEGF-A has also been demonstrated in recent studies to promote lymphangiogenesis in addition to promoting angiogenesis [15,16]. Therefore, VEGF is essential for angiogenesis, lymphangiogenesis and BBB function [17]. Increased VEGF correlates closely with the increased incidence of regional lymph nodes and distant metastases in both humans and animals [18,19]. To evaluate the role of VEGF in BL CNS metastasis, we assessed VEGF levels in the CSF and serum of our BL patient. Our data indicated that the levels of VEGF-ACSF, VEGF-Aserum and VEGF-Cserum increased with the severity of the disease and were higher corresponding control levels, especially after the second Hematol Oncol 2016; 34: 36–41 DOI: 10.1002/hon
Risk factors of CNS metastasis in BL
39
Figure 3. Vascular endothelial growth factor (VEGF) levels and the other lab data changed with remission and recurrence of disease. The CSFWBC count fluctuated with the disease course, elevating at recurrence and decreasing at remission (C). The levels of L-lactate dehydrogenase, VEGF-A CSF and VEGF-CSerum exhibited elevated tendencies when compared with the control group during the disease course, especially after the second remission (B, D, F). Blood brain barrier value and VEGF-ASerum also exhibited rising tendencies when compared with the control group and varied with remission and recurrence of disease (A, E). In conclusion, although not in strict accordance, VEGF levels fluctuated with the disease course, including the status of remission and recurrence. VEGF may participate in Burkitt’s lymphoma central nervous system metastasis
relapse. However, VEGF-CCSF was below the detection limit (3.90 pg/ml). We further explored the ratio of VEGFCSF/VEGFserum to evaluate its potential as an indicator reflecting the efficiency of peripheral blood VEGF leaking into the CSF and hence affecting BBB integrity. The ratio of VEGFCSF/VEGF serum was calculated in the succeeding text: Vascular endothelial growth factor index = (VEGFCSF/ VEGFserum)/(albuminCSF/albuminserum) [20]. A previous study utilized the VEGF index as an indicator to evaluate the source of VEGFCSF, either from local production or from peripheral blood by passive influx. A high VEGF index suggests that VEGF is predominantly produced intrathecally [11]. In our case, the high VEGF-A index suggested that VEGF-A in the CSF was indeed produced locally. High VEGF-A levels increased vascular permeability Copyright © 2014 John Wiley & Sons, Ltd.
and damaged the BBB integrity, as reflected by the high BBB value. The impaired BBB could allow more molecules and cells to enter into the CNS. At the last recurrence, the VEGF-A index was low, which may due to unusually high VEGF-A production in the serum during the marked deterioration of the patient’s condition. High levels of VEGF-C have been correlated with lymphatic vessel invasion, the emergence of sentinel and distant lymph node metastasis, and overall poor prognosis in tumours [21]. In this case, the median time from onset to CNS metastases was approximately 3 months. The patient had an overall response to high-dose intravenous methotrexate, but relapsed soon after. Given the high VEGF levels and frequent recurrences in this patient, our data suggest a correlation between VEGF and CNS metastasis in BL may exist, and the mechanism should be further explored. Hematol Oncol 2016; 34: 36–41 DOI: 10.1002/hon
Y-T Tang et al.
40
In some cases of CSF cytomorphology tests, lymphoma cells may be confused with atypical lymphocytes caused by reactive lymphocytosis; the latter may be significantly enlarged with blast-like nuclei as observed in lymphoma cells [22], leading to false-positive results. In addition, the low sensitivity of conventional cytology has raised concerns about the quality of cytospin preparations. Lymphoma cells confirmed by morphology were only observed in one of nine cases where 99% of cells are already deemed neoplastic [5]. In this case, WBCs in the CSF fluctuated greatly during the 6-month follow-up. Lymphoma cells increased in the CSF during relapse, and the cells were subsequently cleared with chemotherapy. After treatment, CNS metastasis was largely underestimated because the WBCs in the CSF were less than 5 × 106/L, even in the presence of severe neurological symptoms. The intriguing point was that the CSF was free of tumor cells yet contained high VEGF levels. During the second remission, no WBCs were detected in the CSF, whereas the levels of VEGF and the BBB value remained persistently high, suggesting there may be occult damage in the BBB integrity. The early rise in VEGF levels during disease onset and a positive correlation between VEGF levels and WBCCSF during the later stage suggest that VEGF levels may be useful in predicting disease relapse. VEGF-ACSF may help discriminate patients who may potentially benefit from interventions. In addition to VEGF, other factors can increase the risk of BL metastasis to the CNS, such as LDH. In this case, the patient exhibited elevated levels of serum LDH, and LDH levels were higher during relapse than in remission. There was a trend of LDH elevation during the disease course, indicating the increased risk of CNS involvement in the patient. Although the patient exhibited an increasing risk of BL CNS involvement with the high levels of VEGF-ACSF, VEGF-Aserum and VEGF-Cserum, the mechanisms underlying this relationship must be further studied in more patients. Over the past decade, the well-documented role of VEGF in tumour angiogenesis has led VEGFs to become one of the leading therapeutic targets for the treatment of cancer [23]. Understanding the role of VEGF in CNS metastasis, combined with other indicators, may help to improve staging and risk classification, reduce cytotoxic chemotherapy and promote the investigation of targeted therapy.
Conflict of interest The authors have no competing interest.
Acknowledgements This study was funded by the Natural Science Foundation of Guangdong Province, China (S2012030006289, NO9151008-901000043) and the Science and Technology Program of Guangdong Province, Copyright © 2014 John Wiley & Sons, Ltd.
China (NO2010B031600321, No. 2012B031800217), and we would like to thank Dr Frieda Law for editing the manuscript.
References 1. Blum KA, Lozanski G, Byrd JC. Adult Burkitt leukemia and lymphoma. Blood 2004; 104: 3009–3020. 2. Sandlund JT, Downing JR, Crist WM. Non-Hodgkin’s lymphoma in childhood. N Engl J Med 1996; 334: 1238–1248. 3. Miles RR, Arnold S, Cairo MS. Risk factors and treatment of childhood and adolescent Burkitt lymphoma/leukaemia. Br J Haematol 2012; 156: 730–743. 4. Chamberlain MC, Glantz M, Groves MD, Wilson WH. Diagnostic tools for neoplastic meningitis: detecting disease, identifying patient risk, and determining benefit of treatment. Semin Oncol 2009; 36: S35–S45. 5. Hegde U, Filie A, Little RF, et al. High incidence of occult leptomeningeal disease detected by flow cytometry in newly diagnosed aggressive B-cell lymphomas at risk for central nervous system involvement: the role of flow cytometry versus cytology. Blood 2005; 105: 496–502. 6. Fischer L, Martus P, Weller M, et al. Meningeal dissemination in primary CNS lymphoma: prospective evaluation of 282 patients. Neurology 2008; 71: 1102–1108. 7. Ruan J. Antiangiogenic therapies in non-Hodgkin’s lymphoma. Curr Cancer Drug Targets 2011; 11: 1030–1043. 8. Guedez L, McMarlin AJ, Bennett TA, Bennett TA, StetlerStevenson M, Stetler-Stevenson WG. Tissue inhibitor of metalloproteinase-1 alters the tumorigenicity of Burkitt’s lymphoma via divergent effects on tumor growth and angiogenesis. Am J Pathol 2001; 158: 1207–1215. 9. Li KA, Zhang RM, Zhang F, et al. Studies of pathology and VEGF expression in rabbit cerebrospinal fluid metastasis: application of dynamic contrast-enhanced MRI. Magn Reson Imaging 2011; 129: 1101–1109. 10. Dafni H, Kim SJ, Bankson JA, Sankaranarayanapillai M, Ronen SM. Macromolecular dynamic contrast-enhanced (DCE)-MRI detects reduced vascular permeability in a prostate cancer bone metastasis model following anti-platelet-derived growth factor receptor (PDGFR) therapy, indicating a drop in vascular endothelial growth factor receptor (VEGFR) activation. Magnet Reson Med 2008; 60: 822–833. 11. Stockhammer G, Poewe W, Burgstaller S, et al. Vascular endothelial growth factor in CSF: a biological marker for carcinomatous meningitis. Neurology 2000; 54: 1670–1676. 12. Gu Y, Hou YY, Zhang XB, Hu F. Primary central nervous system Burkitt lymphoma as concomitant lesions in the third and the left ventricles: a case study and literature review. J Neurooncol 2010; 99: 277–281. 13. Hirohata S, Isshi K, Oguchi H, et al. Cerebrospinal fluid interleukin-6 in progressive Neuro-Behcet’s syndrome. Clin Immunol Immunopathol 1997; 82: 12–17. 14. Argaw AT, Gurfein BT, Zhang Y, Zameer A, John GR. VEGFmediated disruption of endothelial CLN-5 promotes blood– brain barrier breakdown. Proc Natl Acad Sci U S A 2009; 106: 1977–1982. 15. Stacker SA, Achen MG, Jussila L, Baldwin ME, Alitalo K. Lymphangiogenesis and cancer metastasis. Nat Rev Cancer 2002; 2: 573–583. 16. Joukov V, Pajusola K, Kaipainen A, et al. A novel vascular endothelial growth factor, VEGF-C, is a ligand for the Flt4 (VEGFR-3) and KDR (VEGFR-2) receptor tyrosine kinases. EMBO J 1996; 15: 1751. Hematol Oncol 2016; 34: 36–41 DOI: 10.1002/hon
Risk factors of CNS metastasis in BL
17. Arshad F, Wang L, Sy C, Avraham S, Avraham HK. Blood– brain barrier integrity and breast cancer metastasis to the brain. Patholog Res Int 2010; 2011: 920509. 18. Skobe M, Hawighorst T, Jackson DG, et al. Induction of tumor lymphangiogenesis by VEGF-C promotes breast cancer metastasis. Nat Med 2001; 7: 192–198. 19. Hirakawa S, Kodama S, Kunstfeld R, Kajiya K, Brown LF, Detmar M. VEGF-A induces tumor and sentinel lymph node lymphangiogenesis and promotes lymphatic metastasis. J Exp Med 2005; 201: 1089–1099. 20. van der Flier M, Stockhammer G, Vonk GJ, et al. Vascular endothelial growth factor in bacterial meningitis: detection in
Copyright © 2014 John Wiley & Sons, Ltd.
41
cerebrospinal fluid and localization in postmortem brain. J Infect Dis 2001; 183: 149–153. 21. Rinderknecht M, Detmar M. Tumor lymphangiogenesis and melanoma metastasis. J Cell Physiol 2008; 216: 347–354. 22. Baraniskin A, Deckert M, Schulte-Altedorneburg G, Schlegel U, Schroers R. Current strategies in the diagnosis of diffuse large Bcell lymphoma of the central nervous system. Br J Haematol 2012; 156: 421–432. 23. Vecchiarelli-Federico LM, Cervi D, Haeri M, Li Y, Nagy A, Ben-David Y. Vascular endothelial growth factor--a positive and negative regulator of tumor growth. Cancer Res 2010; 70: 863–867.
Hematol Oncol 2016; 34: 36–41 DOI: 10.1002/hon
