Fax: (011) 49-6221-56-4920
Prognostic value of the Follicular Lymphoma International Prognostic Index score in marginal zone lymphoma
An analysis of clinical presentation and outcome in 144 patients
Version of Record online: 26 JUN 2012
Copyright © 2012 American Cancer Society
Volume 119, Issue 1, pages 99–106, 1 January 2013
How to Cite
Heilgeist, A., McClanahan, F., Ho, A. D. and Witzens-Harig, M. (2013), Prognostic value of the Follicular Lymphoma International Prognostic Index score in marginal zone lymphoma. Cancer, 119: 99–106. doi: 10.1002/cncr.27704
- Issue online: 17 DEC 2012
- Version of Record online: 26 JUN 2012
- Manuscript Accepted: 11 MAY 2012
- Manuscript Revised: 4 MAY 2012
- Manuscript Received: 22 MAR 2012
- marginal zone lymphoma;
- mucosa-associated lymphoid tissue;
- mucosa associated lymphoid tissue;
- Follicular Lymphoma International Prognostic Index;
In marginal zone lymphoma (MZL), clinical and follow-up data on large cohorts of patients are difficult to obtain. The objective of this single-center, retrospective analysis of a large cohort of 144 patients with MZL was to elucidate the role of prognostic markers, treatments, and outcomes in this disease.
In total, 144 patients were identified who were diagnosed with MZL at the authors' institution between 2003 and 2010. Data on clinical parameters, treatments, response, and survival were analyzed. In addition, the validity of the International Prognostic Index (IPI) and Follicular Lymphoma International Prognostic Index (FLIPI) prognostic scores were tested in patients with MZL.
Among 144 patients with MZL, 96 patients (67%) had extralymph node (extranodal) MZL, 32 patients (22%) had lymph node (nodal) MZL, and 16 patients (11%) had splenic MZL. The 5-year progression-free survival rate was 82% in the nodal MZL group, 88% in the extranodal MZL group, and 74% in the splenic MZL group and did not different between the 3 groups (P = .60). The 5-year overall survival rate was excellent in all 3 MZL groups (nodal MZL, 89%; extranodal MZL, 92%; splenic MZL, 82%; P = .46). In our cohort, the FLIPI score was a significant prognostic marker: The 5-year progression-free survival rate for patients who had FLIPI scores of 0 to 2 (low or intermediate risk) was excellent at 92%, whereas it was only 62% for patients who had FLIPI scores of 3 to 5 (poor risk; P = .003). Similarly, the 5-year overall survival rate for patients who had FLIPI scores of 0 to 2 was 95%, whereas it was only 62% for patients who had FLIPI scores of 3 to 5 (P = .0009).
The FLIPI score had strong prognostic value in patients with MZL. Patients who have low-risk or intermediate-risk FLIPI scores have an excellent prognosis, whereas patients with poor-risk FLIPI scores are candidates for novel treatment approaches. Cancer 2013. © 2012 American Cancer Society.
Marginal zone lymphomas (MZLs) are a non-Hodgkin lymphoma (NHL) subgroup that arises from postgerminal center marginal zone B cells and shares a similar immunophenotype, with positivity for the B-cell markers cluster of differentiation 19 (CD19) (B-lymphocyte antigen CD19; a protein encoded by the CD19 gene), B-lymphocyte antigen CD20 (an activated-glycosylated phosphoprotein), and CD22 (a transmembrane protein that specifically binds sialic acid) and negativity for CD5 (a type I transmembrane protein present on B-1 cells and T cells), CD10 (neprilysin; also known as membrane metalloendopeptidase, neutral endopeptidase, and common acute lymphoblastic leukemia antigen [CALLA]), and usually CD23 (Fc epsilon RII or FcϵRII; a C-type lectin). According to the 2008 World Health Organization (WHO) Classification of Tumors of Hematopoietic and Lymphoid Tissues, 3 types of MZL can be distinguished: extralymph node (extranodal) MZL, splenic MZL, and lymph node (nodal) MZL.1 Recurrent genetic abnormalities are common. Gain of chromosomes 3, 12, and 18 and loss of bands 23 and 24 in the long arm of chromosome 6 (6q23-q24) in tumor necrosis factor, alpha-induced protein 3 (TNFAIP3) were identified in all types of MZL entities. Losses of band 14.3 in the long arm of chromosome 13 (13q14.3) in microRNA 15a (MIRN15A) and MIRN16-1 and of bands 13.3 and 12 in the long arm of chromosome 17 (17p13.3-p12) in tumor protein p53 (TP53) were observed in splenic MZL. Loss of 11q21-q22 in ataxia telangiectasia mutated (ATM) was observed in nodal and splenic MZL, and loss of 7q32.1-q33 in ATM was observed in extranodal and splenic MZL.2 The extranodal type constitutes approximately 70% of all MZLs and also is classified as B-cell lymphoma of mucosa-associated lymphoid tissue (MALT).1, 3 It may arise in epithelial tissues, such as stomach, salivary glands, lung, or thyroid gland. MALT lymphomas may be related to chronic immune stimulation, and patients may display autoimmune disorders or infections with Helicobacter pylori, Chlamydophila psittaci, or Campylobacter jejuni.4-7 When it affects the stomach, the translocation involving 11p22 and 18q32 (t[11;18][p22;q32]) is associated with the involvement of regional lymph nodes and a highly impaired response to the eradication of Helicobacter pylori.8 Approximately 20% of MZLs are the splenic type,9 and bone marrow involvement and leukemic manifestation are common. Genetic abnormalities, such as deletion of 7q31 (del7q31) or complete or partial trisomy 3 have been reported.10 A recently published study describes the loss of 7q32.1-q33 in both splenic and extranodal MZL.2 Splenectomy or rituximab monotherapy are effective therapeutic options in splenic MZL.11, 12 Approximately 10% of MZLs are the nodal type without involvement of extranodal tissue.13 Depending on the site and the extent of involvement, therapeutic options include radiotherapy or systemic immunochemotherapy administered according to guidelines for treating follicular lymphoma.14 To date, only a few studies have analyzed and compared the prognosis of patients with the 3 common types of MZL.15 In addition, the prognostic factors that affect the outcome of patients with MZL are poorly understood. In the current study, we sought to elucidate the role of prognostic factors, treatments, and outcomes in a single-center, retrospective analysis of a large cohort of 144 patients with MZL.
MATERIALS AND METHODS
We reviewed the clinical database from the University Hospital of Heidelberg, Germany, for patients who presented with a hematologic malignancy between 2003 and 2010 to identify patients who had MZL according to WHO classification criteria.1 Data on sex, age, Ann Arbor stage, and clinical symptoms; serum lactate dehydrogenase (LDH), soluble CD25 (sCD25) (a type I transmembrane protein; the alpha chain of the interleukin-2 receptor),16 and β2 microglobulin levels before therapy; Eastern Cooperative Oncology Group performance status; and International Prognostic Index (IPI) and Follicular Lymphoma International Prognostic Index (FLIPI) scores were included in the analysis. The study was approved by the local ethics committee of the University of Heidelberg. Clinical responses were classified according to the International Workshop to Standardize Response Criteria for Non-Hodgkin Lymphomas.17 A descriptive statistical analysis to determine baseline characteristics, response rates, progression-free survival (PFS), and overall survival (OS) was performed. The Kaplan-Meier method was used to calculate estimates of PFS and OS. P values < .05 were considered statistically significant.
In total, 144 patients with MZL were identified and classified into 3 different groups: extranodal MZL, nodal MZL, and splenic MZL. Information on type of treatment, response to treatment, and survival was available for 108 patients. Clinical characteristics are summarized in Table 1. There were 96 patients (67%) with extranodal MZL, 32 patients with nodal MZL (22%), and 16 patients with splenic MZL (11%). There was a slight predominance of women in the extranodal group (63%); whereas, in the other groups, both sexes were equally affected. The median age was 60 years (range, 29-79 years) for the extranodal MZL group, 67 years (range, 38-88 years) for the nodal MZL group, and 66.5 years (range, 34-79 years) for the splenic MZL group. Most nodal MZLs (88%) and splenic MZLs (75%) were diagnosed at an advanced clinical stage (Ann Arbor stage III/IV), whereas the distribution of Ann Arbor stage I/II compared with stage III/IV in the extranodal group was almost even. B symptoms were rare in all 3 groups. An elevated serum LDH level was reported in 44% of patients with splenic MZL and in 25% of patients with nodal MZL but in only 3% of patients with extranodal MZL. Elevated sCD25 levels were observed in 75% of splenic MZLs, in 63% of nodal MZLs, and in 25% of extranodal MZLs. The median follow-up was 32 months for patients with splenic MZL, 34.5 months for patients with nodal MZL, and 48 months for patients with extranodal MZL.
|No. of Patients (%)|
|Characteristic||All Patients||Extranodal MZL||Nodal MZL||Splenic MZL|
|Total||144 (100)||96 (67)||32 (22)||16 (11)|
|Men||76 (53)||49 (51)||20 (63)||7 (44)|
|Women||68 (47)||47 (49)||12 (37)||9 (56)|
|Median age [range], y||62 [29-88]||60 [29-79]||67 [38-88]||66.5 [34-79]|
|Ann Arbor stage|
|I-II||58 (40)||51 (53)||3 (9)||4 (25)|
|III-IV||79 (55)||39 (41)||28 (88)||12 (75)|
|Missing||7 (5)||6 (6)||1 (3)||0 (0)|
|B symptoms: ≥1||9 (6)||2 (2)||4 (13)||3 (19)|
|Above normal||18 (13)||3 (3)||8 (25)||7 (44)|
|Normal||90 (63)||68 (71)||15 (47)||7 (44)|
|Missing||36 (25)||25 (26)||9 (28)||2 (13)|
|Above normal||46 (32)||24 (25)||20 (63)||12 (75)|
|Normal||58 (40)||43 (45)||3 (9)||2 (13)|
|Missing||40 (28)||29 (30)||9 (28)||2 (13)|
|No. of sites|
|1||72 (50)||55 (57)||4 (13)||13 (81)|
|2||24 (17)||16 (17)||6 (19)||2 (13)|
|>2||37 (25)||22 (23)||14 (44)||1 (6)|
|Missing||11 (8)||3 (3)||8 (25)||0 (0)|
|0-1||93 (65)||68 (71)||13 (41)||12 (75)|
|>1||20 (14)||6 (6)||10 (31)||4 (25)|
|Missing||31 (22)||22 (23)||9 (28)||0 (0)|
|0-2||63 (44)||51 (53)||7 (22)||9 (56)|
|3-5||45 (31)||20 (21)||16 (50)||5 (31)|
|Missing||36 (25)||25 (26)||9 (28)||2 (13)|
|Low||53 (37)||47 (49)||4 (13)||2 (13)|
|Intermediate||32 (22)||19 (20)||6 (19)||7 (44)|
|High||23 (16)||5 (5)||13 (41)||5 (31)|
|Missing||36 (25)||25 (26)||9 (28)||2 (13)|
|All||144 (100)||96 (100)||32 (100)||16 (100)|
Patients with extranodal MZLs received either radiotherapy alone (24%) or systemic therapy (31%) (Table 2). If systemic therapy was chosen, then the majority of regimens (87%) included rituximab. Thirty percent of patients received rituximab monotherapy (Table 3). In the nodal MZL group, 61% of patients received systemic therapy (Table 2). The regimens most commonly received were combined cyclophosphamide, vincristine, doxorubicin, and prednisone (CHOP) and combined pentostatin and cyclophosphamide (PC). Rituximab was part of systemic therapy for 79% of patients (Table 4). For patients who received radiotherapy, the most frequent site of radiation was the eye (33% of patients). Other sites were salivary glands, stomach, skin, lymph nodes, bones, and bladder (Table 5). Sixty-four percent of patients with splenic MZL underwent splenectomy (Table 2).
|No. of Patients. (%)|
|Therapy||All Patients||Extranodal MZL||Nodal MZL||Splenic MZL|
|No therapy||13 (12)||8 (11)||3 (13)||2 (14)|
|Systemic treatment alone||39 (36)||22 (31)||14 (61)||3 (21)|
|Surgery alone||17 (16)||6 (9)||2 (9)||9 (64)|
|Radiation alone||18 (17)||17 (24)||1 (4)||0 (0)|
|HP-Er alone||3 (3)||2 (3)||1 (4)||0 (0)|
|Radiation and chemotherapy||4 (4)||3 (4)||1 (4)||0 (0)|
|Systemic treatment plus HP-Er||4 (4)||4 (6)||0 (0)||0 (0)|
|Surgery and radiation||4 (4)||4 (6)||0 (0)||0 (0)|
|Other||6 (6)||5 (7)||1 (4)||0 (0)|
|All||108 (100)||71 (100)||23 (100)||14 (100)|
|Chemotherapy||Percentage of Patients|
|Rituximab and bendamustine||13|
|R-COP and fludarabine||4.3|
|CHOP and fludarabine||4.3|
|Chlorambucil and fludarabine||4.3|
|Chemotherapy||Percentage of Patients|
|Rituximab and bendamustine||7.1|
|Site of Radiation||Percentage of Patients|
In the extranodal group, 41% of patients achieved complete remission (CR) or unconfirmed CR (CRu) after chemotherapy alone, 32% achieved partial remission (PR), 23% had stable disease (SD), and only 4.5% had progressive disease (PD) or recurrent disease (RD) (Table 6). Radiotherapy alone led to CR or CRu in 75% of patients in the extranodal group, 6% achieved PR, 19% had SD, and no patient had a response worse than SD (Table 7). In the nodal group, 29% of patients achieved CR or CRu with chemotherapy alone, another 29% achieved PR, 21% had SD, and 22% had PD or RD (Table 6).
|Percentage of Patients|
|Response||Extranodal MZL||Nodal MZL|
|Complete remission (CR)||36.4||14.3|
|Response||Extranodal MZL, %|
|Complete remission (CR)||76.5|
Survival and Prognostic Scores
The 5-year PFS rates for the extranodal, nodal, and splenic MZL groups were 88%, 82%, and 74%, respectively (P = .60) (Fig. 1, top). The 5-year OS rates for the extranodal, nodal, and splenic MZL groups were excellent at 92%, 89%, and 82%, respectively (P = .46) (Fig. 1, bottom). By applying the FLIPI score,18 we were able to divide extranodal and splenic MZL types into 2 groups with significantly different OS and PFS. Because there was no significant difference between the low-risk and intermediate-risk groups according to the FLIPI, we combined these 2 groups and compared them with the poor-risk group. The 5-year PFS rate for patients who had FLIPI scores of 0 to 2 (low or intermediate risk) was excellent at 92%, but it was only 62% for patients who had FLIPI scores of 3 to 5 (poor risk; P = .003). Similarly, the 5-year OS for patients who had FLIPI scores of 0 to 2 was 95%, but it was only 62% for patients who had FLIPI scores of 3 to 5 (P = .0009) (Fig. 2). In a subgroup analysis of the extranodal MZL group, we observed significant differences in PFS (HR, 4.82; P = .03) and OS (HR, 5.94; P = .01) between patients who had FLIPI scores of 0 to 2 and those who had FLIPI scores of 3 to 5 (Fig. 3). In the splenic MZL group, there was a trend toward a nonsignificant difference between patients who had FLIPI scores of 0 to 2 and those who had FLIPI scores of 3 to 5 in terms of PFS (P = .07) and OS (P = .06), (Fig. 4). In a subgroup analysis of the nodal MZL group, the FLIPI score was not able to separate 2 significantly different groups with regard to PFS or OS (Fig. 5). The IPI score, which was developed for aggressive lymphoma, was not able to identify different risk groups of patients with extranodal, nodal, or splenic MZL (data not shown).
MZL is a rare type of malignant B-cell lymphoma. Because of the paucity of large clinical trials, the standard treatment for MZL is still a matter of debate. The purpose of this study was to analyze the role of prognostic markers, treatments, and outcomes in a large cohort of 144 patients with MZL who were diagnosed at our institution between 2003 and 2010. Most of our patients (67%) were diagnosed with extranodal MZL, whereas splenic MZL (11%) was the rarest type. Patient with localized disease received radiotherapy and achieved high response rates (CR, 76%). Like in other indolent lymphomas,19 rituximab has demonstrated effectiveness in the treatment of MZL.20-22 However, a prospective randomized trial on this issue is still missing. In our cohort, among those who chose systemic therapy, 79% of patients with nodal MZL and 87% of patients with extranodal MZL received rituximab with or without chemotherapy. Because of the lack of prospective, randomized studies in MZL, the optimal chemotherapeutic partner of rituximab is not clear. In our patients, CHOP was chosen most frequently in accordance with data from prospective, randomized trials in follicular lymphoma.23 Recent data, however, indicate that bendamustine currently is also a reasonable choice.24 Patients who had splenic MZL predominantly underwent splenectomy. In our cohort, the response rates were high, and 5-year PFS and OS were excellent for all types of MZL, as reported previously.15, 25-27 Although MZL in general is an indolent disease, the clinical course can be highly variable, and prognostic factors are not well defined. The FLIPI score was developed as a prognostic scoring system for follicular lymphoma and includes age (>60 years vs ≤60 years), Ann Arbor stage (III-IV vs I-II), hemoglobin level (<120 g/L vs ≤120 g/L), the number of involved lymph node areas (>4 vs ≤4), and serum LDH level (above normal vs normal or below).18 In 2 studies of nodal MZL, a higher FLIPI score predicted poorer survival.28, 29 In extranodal MZL, only 1 retrospective study has attempted to identify prognostic scoring systems: The investigators reported that the FLIPI score was not predictive in extranodal MZL and proposed a novel prognostic score for MZL. That score, however, was based on data from a single institution and has not been evaluated in other cohorts.15 In our study, we observed that the FLIPI score had high prognostic significance. The 5-year PFS and OS rates for patients who had FLIPI scores of 0 to 2 (low or intermediate risk) were excellent at 92% and 95%, respectively; whereas the rates were only 56% and 62%, respectively, for poor-risk patients who had FLIPI scores of 3 to 5. Despite the heterogeneous treatment of patients during the long reported period from 2003 to 2010, the FLIPI prognostic score was able to identify different prognostic subgroups. This finding implies that the prognostic significance of the FLIPI score may reflect biologic and pathologic disease parameters rather than the impact of a particular treatment regimen. In a subgroup analysis, the FLIPI score had prognostic significance for patients with extranodal MZL and splenic MZL, but not for patients with nodal MZL, probably not because of particular biologic characteristics of the disease entity but, more likely, because of the low patient numbers in this subgroup. However, these results were obtained in a retrospective analysis and should be confirmed in a prospective, randomized trial. In our cohort, the IPI score, which was developed for aggressive lymphoma, was not able to identify different risk groups.30 We conclude that, although the prognosis of patients with MZL in general is excellent, patients who have high-risk FLIPI scores have a relatively poor prognosis and may be candidates for novel treatment approaches.
No specific funding was disclosed.
CONFLICT OF INTEREST DISCLOSURES
The authors made no disclosures.
- 1Swerdlow SH, Campo E, Harris NL, et al. eds. WHO Classification of Tumours of Haematopoietic and Lymphoid Tissues. Lyon, France: IARC Press; 2008.
- 22Combination therapy with rituximab and intravenous or oral fludarabine in the first-line, systemic treatment of patients with extranodal marginal zone B-cell lymphoma of the mucosa-associated lymphoid tissue type. Cancer. 2009; 115: 5210-5217., , , et al.
- 23Frontline therapy with rituximab added to the combination of cyclophosphamide, doxorubicin, vincristine, and prednisone (CHOP) significantly improves the outcome for patients with advanced-stage follicular lymphoma compared with therapy with CHOP alone: results of a prospective randomized study of the German Low-Grade Lymphoma Study Group. Blood. 2005; 106: 3725-3732., , , et al.
- 24Bendamustine plus rituximab is superior in respect of progression free survival and CR rate when compared with CHOP plus rituximab as first-line treatment of patients with advanced follicular, indolent, and mantle cell lymphomas: final results of a randomized phase III study of the StiL (Study Group Indolent Lymphomas, Germany) [abstract]. Blood (ASH Annual Meeting Abstracts). 2009; 114. Abstract 405., , , et al.
- 30[No authors listed] A predictive model for aggressive non-Hodgkin's lymphoma. The International Non-Hodgkin's Lymphoma Prognostic Factors Project. N Engl J Med. 1993; 329: 987-994.