Abstract
Objective. To retrospectively analyze the clinical course of patients with mucosa-associated lymphoid tissue (MALT)-type lymphoma of the parotid gland and associated Sjögren’s syndrome (SS).
Methods. All consecutive patients with SS and MALT lymphoma (MALT-SS) diagnosed in the University Medical Center Groningen between January 1997 and January 2009 were analyzed. Clinical course and treatment outcome of SS and MALT lymphoma were evaluated.
Results. From a total of 329 patients with SS, 35 MALT-SS patients were identified, with a median followup of 76 months (range 16–153 mo). MALT lymphoma was localized in the parotid gland in all cases. Treatment consisted of “watchful waiting” (n = 10), surgery (n = 3), radiotherapy (n = 1), surgery combined with radiotherapy (n = 2), rituximab only (n = 13), or rituximab combined with chemotherapy (n = 6). Complete response was observed in 14 patients, partial response in 1 patient, and stable disease in 20 patients. In 6 of 7 patients with initially high SS disease activity (M-protein, cryoglobulins, IgM rheumatoid factor > 100 KIU/l, severe extraglandular manifestations), MALT lymphoma progressed and/or SS disease activity increased after a median followup of 39 months (range 4–98 mo), necessitating retreatment. Only 1 patient with MALT who had low SS disease activity showed progression of lymphoma when left untreated.
Conclusion. An initially high SS disease activity likely constitutes an adverse prognostic factor for progression of lymphoma and/or SS. Such patients may require treatment for both MALT lymphoma and SS. In SS patients with localized asymptomatic MALT lymphoma and low SS disease activity, a “watchful waiting” strategy seems justified.
Sjögren’s syndrome (SS) is a systemic autoimmune disease characterized by chronic inflammation of the salivary and lacrimal glands. SS is frequently accompanied by systemic symptoms. Four percent to 7%1,2 of patients with SS develop malignant B cell lymphoma, 48%–75% of which are of the mucosa-associated lymphoid tissue (MALT)-type. These B cell lymphomas are most frequently located in the parotid gland3,4,5. A study showed a 6.6-fold increase of non-Hodgkin’s lymphoma (NHL) in patients with SS compared to controls6. MALT lymphoma of the parotid gland was almost exclusively associated with SS, as there was a 1000-fold increase in the relative risk of having SS in the case of a MALT lymphoma presenting in the parotid gland6. In patients with SS, parotid gland enlargement is frequently present but varies in time. The change from variable to persistent enlargement of glands is an important clinical sign, indicating the possible development of MALT lymphoma. Further, the emergence of lymphoma in SS may be heralded by extraglandular manifestations of SS (e.g., palpable purpura, vasculitis, renal involvement, peripheral neuropathy). None of these features is specific for MALT lymphoma in SS, but any of them should raise suspicion, particularly if accompanied by features such as monoclonal gammopathy, reduced levels of complement C4, CD4+ T lymphocytopenia, a sharp increase in IgG levels, or cryoglobulinemia2,7,8,9,10.
Assessment of SS patients who may have developed a MALT lymphoma is not always easy, but an incisional biopsy of the parotid gland can safely be performed under local anesthesia11.
MALT lymphoma in general is an indolent disease, with a reported 5-year overall survival between 86% and 95%, without significant difference in clinical course between localized and disseminated disease12,13. Recurrences may involve extranodal or nodal sites. Progression to aggressive lymphoma is rare, occurring in fewer than 10% of cases14.
MALT lymphoma in patients with SS is often localized at 1 or more salivary glands [usually the parotid gland(s)], but can also occur in other extranodal sites, such as the orbital adnexa15 and stomach16. Dissemination of MALT-SS may be observed in local draining lymph nodes or sometimes distant nodes, and occasionally other mucosal sites and bone marrow17. Prognosis does not seem to be influenced by dissemination to other MALT organs, although involvement of lymph nodes might be an adverse prognostic factor13,17.
The relative infrequency and heterogeneity of MALT lymphomas, with their different biology, clinical presentation, and behavior, make it difficult to define optimal treatment of these patients in general. Antibiotics are widely used as initial treatment of MALT lymphoma associated with microbial pathogens, in particular gastric MALT lymphoma associated with Helicobacter pylori18,19,20,21. In other MALT lymphomas with symptomatic local disease, local treatment (surgery or radiotherapy) results in excellent disease control22. However, in patients with SS, conventional radiotherapy of the salivary glands (25 to 39 Gy) may lead to significant further xerostomia23,24. An alternative approach might be low-dose 2 × 2 Gy involved-field radiotherapy. This therapy is very effective in follicular lymphoma25 and data in MALT lymphoma seem promising26,27. For symptomatic disseminated disease, chemotherapy has commonly been used, with a 75% complete remission rate and 5-year event-free survival and overall survival rates of 50% and 75%, respectively12,13,28,29,30. More recently, rituximab, a chimeric murine/human anti-CD20 monoclonal antibody that is highly efficacious in patients with B cell lymphoma, alone or in combination with chemotherapy31,32,33,34, has also been used effectively in patients with MALT lymphoma, with or without associated SS35,36,37,38,39,40.
At present, no clear guidelines exist for the management of patients with MALT-SS. In this retrospective study we report our experience in 35 consecutive patients with MALT-SS treated in our center.
MATERIALS AND METHODS
Patients
Our hospital, the University Medical Center Groningen, is a referral institute for SS in The Netherlands. Patients are referred to either the Department of Rheumatology or Oral Medicine. Subsequently, all patients with suspicion of SS are routinely evaluated in both departments. In cases of MALT lymphoma, patients are also seen by a hematologist. A retrospective analysis was performed of all consecutive patients diagnosed with SS and MALT lymphoma in our hospital, between January 1997 and January 2009. Patients with a diagnosis of extranodal marginal zone (MALT) lymphoma according to the World Health Organization classification41 and a concomitant diagnosis of SS according to the American-European consensus criteria42 were included. All patients with MALT-SS were seen at 6-month intervals for routine followup. Followup ended February 1, 2010 (median followup 76 mo, minimal followup of 12 mo).
Diagnosis of SS and MALT lymphoma
A complete routine diagnostic investigation for SS was carried out in all patients at their first visit as described38. In brief, investigation included subjective complaints of ocular and oral dryness, eye tests (Rose bengal staining and Schirmer tear test), measurement of unstimulated whole salivary flow, parotid sialography, and serology (anti-SSA/SSB antibodies). In all patients an incision biopsy of the parotid gland was part of the investigation and was performed under local anesthesia11,43. All patients in our cohort were diagnosed using the SS classification criteria of the American-European consensus group; findings of hepatitis C and HIV infections were excluded accordingly42.
All biopsies were reviewed by a hematopathologist (PMK). In all 35 cases immunohistochemistry was performed for at least CD3, CD5, CD20, CD79a, CD10, BCL6, and cytoplasmic immunoglobulins (kappa, lambda, IgM, IgG, and IgA). In 21/35 cases DNA analysis of clonality was performed using polymerase chain reaction (PCR) for FR1, FR2, and FR3 of the immunoglobulin heavy-chain gene complex, as described44,45. All FR PCR reactions were performed in 4-fold at 2 dilutions. All 21 cases revealed a dominant and reproducible monoclonal population of B cells in 1 or more framework PCR reactions. To distinguish between reactive benign lymphoproliferation and MALT lymphoma, the classification of Quintana, et al46 was used. All lymphomas were classified according to the WHO classification of 200841.
Staging of MALT lymphoma and disease activity score of SS
Patients were staged according to a standard lymphoma protocol including computed tomography (CT) or magnetic resonance imaging (MRI) scans of head/neck, thorax, and abdomen, and a bone marrow biopsy47. For SS-associated MALT lymphomas located in the salivary gland, we used a relatively simple staging system based on the Ann Arbor classification47 and the modification for primary gastric lymphoma by Musshoff48 as follows: (1) Localized disease: lymphoma located in 1 or more salivary glands (unilateral or bilateral), without enlargement of lymph nodes. (2) Locally disseminated disease: lymphoma localized in 1 or more salivary glands (unilateral or bilateral) with 1 or more enlarged regional lymph nodes (> 1 cm). (3) Disseminated disease: localization of lymphoma in 1 or more salivary glands (unilateral or bilateral) with 1 or more enlarged distant lymph nodes (> 1 cm), and/or bone marrow, spleen, liver, or other extranodal site than the salivary gland, or localization of lymphoma in multiple extranodal sites.
SS disease activity was evaluated based on the following variables: the presence of extraglandular manifestations (e.g., arthritis, fatigue, vasculitis, glomerulonephritis), subjective sicca symptoms (using a 100-mm visual analog scale), salivary gland function, and serological measures [levels of total IgG and IgM rheumatoid factor (RF), C4, cryoglobulins, M-protein]. Based on these measures, a global impression of disease activity (low, moderate, high) was determined by the physicians (RPEP, JP, HB, FKLS, GWvI) participating in the multidisciplinary group for patients with SS and MALT lymphoma.
Treatment and treatment evaluation
The choice of treatment modality was decided by a team of experts (rheumatologist, oral surgeon, pathologist, hematologist) based on clinical, serological, and radiographic data. In general the following treatment regimens were used: watchful waiting, surgery, radiotherapy, or surgery combined with radiotherapy or cyclophosphamide-prednisone. Rituximab was available from 2002 onward and was added to the treatment regimens. Rituximab monotherapy was given as 4 infusions of 375 mg/m2 given once weekly as described38; rituximab with cyclophosphamide and prednisone (R-CP) was given as 6–8 intravenous infusions of 375 mg/m2 of rituximab and 750 mg/m2 of cyclophosphamide. One infusion was given every 3–4 weeks, in combination with 100 mg prednisone PO for 5 days. After initial rituximab or R-CP treatment, no maintenance immunotherapy was given.
Data on time between diagnosis of SS and presentation of lymphoma, response to treatment, and clinical course during followup were retrieved from the medical records. Tumor responses were classified as complete response (CR), partial response (PR), stable disease (SD), or recurring/progressive disease (PD), according to the standardized response criteria for malignant lymphomas49,50. In brief, CR required the absence of palpable swelling of salivary glands and reduction of size of all nodes to < 1 cm on computed tomography/magnetic resonance imaging (MRI), and normalization of bone marrow, spleen, liver, or other extranodal sites, if initially involved. In patients with localized disease, CR was considered when no evidence of disease was present after diagnostic surgical excision with or without subsequent radiotherapy or after R-CP objectified by MRI. Seven patients treated with rituximab monotherapy took part in a prospective clinical trial in which a repeated parotid gland biopsy was performed after treatment as defined by protocol38. In these patients, a repeated biopsy of the involved parotid gland had to show complete disappearance of the lymphoma infiltrate in order to classify them as having achieved CR. PR required regression of initial tumor mass by ≥ 50% without development of new lesions. SD was defined as < 50% regression or < 50% increase of the known sites of disease. PD recurrence required ≥ 50% increase of any previously identified abnormal lesions or any new lesion (including recurrence in case of previous CR), irrespective of simultaneous responses at other sites.
Since there are no validated disease activity response criteria for SS51,52, deterioration of SS was arbitrarily defined as the occurrence of 1 or more of the following: an increase in levels of IgG, increase in levels of IgM-RF, decrease in salivary gland function, increase in subjective oral or ocular symptoms, and/or the development of extraglandular manifestations (arthritis or vasculitis; pulmonary, hepatic, or renal SS involvement). Improvement of SS required 1 or more of the following: a decrease in levels of IgG and/or IgM-RF, improvement of salivary gland function53, improvement of subjective symptoms, and/or disappearance of extraglandular manifestations. Stable disease was defined as the absence of deterioration or improvement according to the criteria noted above. The criteria are in agreement with the outcome criteria for SS for clinical trials as proposed by Pillemer, et al54 and the disease activity scales that were developed later and remain to be validated51,52.
RESULTS
Patient characteristics
From a total of 329 patients with SS diagnosed in our hospital between January 1997 and January 2009, 35 (11%) patients with MALT-SS were identified. Two SS patients developed extraglandular lymphoproliferative malignancies without parotid gland involvement; 1 patient had an inguinal extra-ossal plasmacytoma, the other a rectal MALT lymphoma. These patients were both excluded from analysis. Characteristics of MALT lymphoma and SS of the 35 patients in the study are listed in Tables 1 and 2.
In 11 (31%) out of 35 patients, MALT lymphoma was detected in a parotid biopsy during routine diagnostic investigation of SS. Eighteen MALT lymphomas (51%) were detected in patients presenting with active SS and suspicion of MALT lymphoma development based on persistent glandular swelling, confirmed after parotid biopsy. Six patients (17%) presented with a parotid gland tumor initially; in these patients a diagnosis of MALT lymphoma in association with SS was made following the parotid biopsy.
Median age at MALT lymphoma diagnosis was 55 years (range 26–84 yrs). MALT lymphoma was localized in the parotid gland in all cases (n = 35). The majority of patients, 26 (74%) out of 35, had localized disease. Five patients (14%) had locally disseminated disease, and 4 patients (11%) had disseminated disease (bone marrow in 1 patient, lacrimal gland involvement in 2, and involvement of the stomach in 1). Seven patients had high SS disease activity initially, as exemplified by monoclonal gammopathy/cryoglobulins, increased IgM-RF, and 1 or more severe extraglandular manifestations (arthritis, vasculitis; pulmonary, hepatic, or renal SS involvement).
Treatment and response of MALT lymphoma
Ten patients (28%) received no initial treatment because the lymphoma was asymptomatic. These patients were closely monitored (“watchful waiting”; Tables 3 and 4). In 5 patients a diagnostic superficial parotidectomy was performed because of persistent symptomatic unilateral parotid gland swelling. The treatment resulted in complete excision of the lymphoma in 3. Three patients received radiotherapy, including the 2 patients with incomplete excision of the lymphoma. Thirteen patients (37%) were treated with rituximab only; 7 of those patients participated in a phase II study with rituximab38. Six patients (17%) were treated with R-CP.
Lymphoma response in the 25 patients treated with surgery, radiotherapy, rituximab, or R-CP was as follows: CR in 14 (56%) patients, PR in 1 patient (4%), and SD in 10 patients (40%). No serious side effects were observed.
A reduction in extraglandular manifestations (arthritis, vasculitis; pulmonary, hepatic, or renal SS involvement) was seen in all systemic-treated patients (rituximab or R-CP).
Improvement in serologic measures (increased C4, no presence of cryoglobulins and M-protein) was observed in the majority of patients (9 out of 13) treated with rituximab. Three of the 4 patients that did not show improvement in serologic measures after rituximab treatment initially had high SS disease activity. All 6 patients treated with R-CP showed normalization of these serologic measures after treatment (Table 5).
Longterm outcome and followup
After a median followup of 76 mo (range 16–153 mo) after initial diagnosis of MALT-SS, progression or recurrence of MALT lymphoma was observed in 10 out of 35 patients at a median time of 45 mo after diagnosis (range 4–98 mo; Table 6). Five of the 10 patients with progression or recurrence of lymphoma had high SS disease activity initially (Table 6). The difference in disease-free survival of patients with initial low vs high SS disease activity was significant (p < 0.05; Figure 1).
In the “watchful waiting” group, 2 of the 10 patients showed MALT lymphoma progression, after 34 and 81 mo. One of these patients also had high SS disease activity at the initial diagnosis of MALT. All other patients in the “watchful waiting” group had low SS disease activity.
Increased extraglandular SS activity without progression of MALT lymphoma necessitating immunosuppressive retreatment was observed in 1 patient treated with rituximab at 27 mo after MALT diagnosis (Table 6).
Transformation to high-grade lymphoma was not observed. Thirty-four patients are alive at a median followup of 76 mo (range 16–153 mo); 1 patient died of pneumonia unrelated to MALT lymphoma.
DISCUSSION
In our retrospective study we analyzed the clinical course of patients with SS and associated MALT lymphoma. Patients with SS who had MALT lymphoma of the parotid gland usually had localized disease. High SS disease activity at presentation (i.e., multiple extraglandular SS manifestations, low C4, presence of cryoglobulins and/or M-protein) was associated with clinical progression or recurrence of MALT lymphoma and/or deterioration of SS during followup necessitating (re)treatment. Although the median followup of 6 years is still relatively short, transformation to high-grade lymphoma was not observed and no patient died from lymphoma.
MALT lymphoma in patients with SS is part of a spectrum ranging from indolent asymptomatic lymphoma and low SS disease activity to locally disseminated or disseminated lymphoma and severe extraglandular SS manifestations. According to the latest diagnostic consensus criteria, preexistent lymphoma is considered to be an exclusion criterion of SS, because lymphoma of the parotid gland itself can cause mouth dryness and parotid gland swelling42. In our opinion, however, MALT lymphoma of the salivary gland should not be considered as an exclusion criterion for the diagnosis of SS. As also shown in the study by Ekström Smedby, et al6, the great majority of these lymphomas are associated with SS or other autoimmune diseases17,55. According to that study, vasculitis, peripheral nerve involvement, glomerulonephritis, fever, anemia, and lymphopenia are observed significantly more often in patients with MALT-SS than in the general SS population5. In our study the majority of patients (71%) had no severe extraglandular manifestations. This discrepancy might be explained by the inclusion of 11 patients in whom a MALT lymphoma was detected during diagnostic investigation for SS, which in our institution included a parotid biopsy instead of a labial biopsy. The parotid biopsy is preferred because of its association with less morbidity and diagnostic potential comparable to that of a labial biopsy in the diagnosis of SS43. Other studies have found an incidence of 3.4% to 7% of MALT lymphoma within their SS cohorts1,2,5. Since the parotid is the gland most commonly involved in MALT lymphoma, routine use of the parotid biopsy for diagnosis of SS likely explains a slightly higher incidence of MALT lymphoma in our SS cohort (11%)56, as well as the higher frequency of lymphoma patients without severe extraglandular manifestations of SS.
The high survival rate in our cohort (97%; 1 patient died of pneumonia unrelated to MALT lymphoma) is in accord with reports of MALT lymphoma not associated with SS13,14. Progression/relapse was seen in 29% of our patients; this finding is also in accord with the 30% progression/relapse rate reported for MALT lymphoma not associated with SS57.
The staging system for MALT-SS used in this study (see Materials and Methods) may provide better prognostic information at diagnosis than the traditional Ann Arbor staging. Although MALT lymphoma in SS can localize in other mucosal sites, it usually localizes in the main target of the autoimmune disease, i.e., the parotid gland16,56. It is debatable whether it is necessary to perform full staging in patients with MALT-SS, including CT scans of thorax and abdomen and bone marrow biopsy. Bone marrow involvement is rare in the patients described, and probably does not influence prognosis or treatment12.
In some patients with locally disseminated or disseminated disease, it may be difficult to decide whether symptoms should be attributed to lymphoma activity or to SS activity. For example, weight loss might be attributed to lymphoma activity, but could also be part of SS disease activity. In these patients, both lymphoma and SS disease activity need to be addressed: not only the clinical characteristics of the lymphoma, but also the severity of SS manifestations might determine the choice of treatment. We are aware that an international standardized activity score is needed for evaluating SS disease activity. However, in the time frame within which this cohort was diagnosed, no standardized activity score was available51,52. At that time the disease activity was assessed by a team of experts according to our own standardized methods (as above).
As observed in this analysis, “watchful waiting” seems a suitable option in patients with asymptomatic MALT lymphoma in the absence of high SS disease activity, since most patients remained asymptomatic for a long period of time (Figure 1). In patients with symptomatic MALT lymphoma, such as a persistent disabling parotid gland swelling, but with low SS disease activity, local treatment with low-dose involved-field radiotherapy to spare remaining salivary function (2 × 2 or 1 × 4 Gy) might be sufficient. However, experience with low-dose involved-field radiotherapy in extranodal MALT lymphoma is limited.
Our study also suggests that in patients with MALT-SS with initial high SS disease activity, rituximab monotherapy might not be sufficient, because these patients required retreatment due to recurrence of MALT lymphoma and/or development of SS disease activity. Also, normalization of serological variables was not seen (low C4 levels and presence of cryoglobulins and/or M-protein). In these patients, treatment might have to include more intensive immunosuppressive therapy, for instance a combination of rituximab with cyclophosphamide and prednisone (R-CP). This combination therapy is effective in the treatment of both indolent lymphoma and autoimmune disease58,59.
Current guidelines for management and treatment of patients with MALT-SS in our center are based on the treatment experience set out in this article (Figure 2): (1) Asymptomatic MALT and low SS disease activity: watchful waiting. (2) Symptomatic local MALT, no or low SS disease activity: radiotherapy. (3) High SS disease activity and asymptomatic MALT: rituximab only (phase II trial) or immunochemotherapy of R-CP. And (4) Symptomatic MALT and high SS disease activity: R-CP.
Although our study indicates that most treated patients fare well, recurrences may occur. It remains to be determined whether these patients might benefit from maintenance B cell depletion therapy, as in indolent B cell lymphoma and autoimmune diseases such as rheumatoid arthritis and systemic lupus erythematosus.
- Accepted for publication June 7, 2011.