Long-term outcome in localized extranodal mucosa-associated lymphoid tissue lymphomas treated with radiotherapy


  • Jayant S. Goda MD, DNB,

    1. Department of Radiation Oncology, Princess Margaret Hospital, Toronto, Ontario, Canada
    2. University of Toronto, Toronto, Ontario, Canada
    Search for more papers by this author
  • Mary Gospodarowicz MD,

    1. Department of Radiation Oncology, Princess Margaret Hospital, Toronto, Ontario, Canada
    2. University of Toronto, Toronto, Ontario, Canada
    Search for more papers by this author
  • Melania Pintilie PhD,

    1. University of Toronto, Toronto, Ontario, Canada
    2. Department of Biostatistics, Princess Margaret Hospital, Toronto, Ontario, Canada
    Search for more papers by this author
  • Woodrow Wells MD,

    1. Department of Radiation Oncology, Princess Margaret Hospital, Toronto, Ontario, Canada
    2. University of Toronto, Toronto, Ontario, Canada
    Search for more papers by this author
  • David C. Hodgson MD,

    1. Department of Radiation Oncology, Princess Margaret Hospital, Toronto, Ontario, Canada
    2. University of Toronto, Toronto, Ontario, Canada
    Search for more papers by this author
  • Alexander Sun MD,

    1. Department of Radiation Oncology, Princess Margaret Hospital, Toronto, Ontario, Canada
    2. University of Toronto, Toronto, Ontario, Canada
    Search for more papers by this author
  • Micheal Crump MD,

    1. University of Toronto, Toronto, Ontario, Canada
    2. Department of Hematology and Medical Oncology, Princess Margaret Hospital, Toronto, Ontario, Canada
    Search for more papers by this author
  • Richard W. Tsang MD

    Corresponding author
    1. Department of Radiation Oncology, Princess Margaret Hospital, Toronto, Ontario, Canada
    2. University of Toronto, Toronto, Ontario, Canada
    • Radiation Medicine Program, Princess Margaret Hospital, Toronto, Ontario, Canada M5G2M9
    Search for more papers by this author
    • Fax: (416) 946-6556



This study was conducted to evaluate the long-term outcomes in patients with stage IE and IIE mucosa-associated lymphoid tissue (MALT) lymphomas treated with involved field radiotherapy (RT).


Between 1989 and 2004, 192 patients with stage I and II MALT lymphomas were treated. The report focuses on 167 patients who received RT. The median age of patients was 58 years with a female predominance (2:1). Presenting sites were as follows: orbital adnexa in 71 patients, salivary glands in 28 patients, stomach in 25 patients, thyroid in 21 patients, and other sites in 22 patients. The median dose to nonorbital sites was 30 grays (Gy) (range, 17.5-35 Gy) and was 25 Gy for the orbit (range, 25-35 Gy). The median follow-up was 7.4 years (range, 0.67-16.20 years).


Complete response and complete response, unconfirmed (CR/CRu) was noted in 166 (99%) patients. The 10-year recurrence-free rate (RFR) was 76%, the disease-free survival (DFS) rate was 68%, the overall survival (OS) rate was 87%, and the cause-specific survival rate was 98%. According to presenting site, the 10-year RFR was 95% for thyroid, 92% for stomach, 68% for salivary glands, and 67% for orbit. Patients with thyroid and gastric MALTs had better outcome compared with patients with MALTs diagnosed at other sites (P = .004). Among those patients who achieved CR, 19% developed disease recurrence (n = 31), chiefly in distant sites or untreated contralateral-paired organs. At the time of disease recurrence, 7 patients (23%) had transformed to diffuse large B-cell lymphoma, 2 of whom died of lymphoma. The 5-year OS rate after treatment failure was 83%.


Patients with localized MALT lymphomas are reported to have excellent clinical outcome after moderate-dose RT, and some are likely cured. In the current study, thyroid and gastric MALTs were found to have significantly less risk of distant recurrence. Despite disease recurrence, the overall survival remains excellent in these patients. Cancer 2010. © 2010 American Cancer Society.

Initially described by Isaacson and Wright,1 mucosa-associated lymphoid tissue (MALT) lymphoma is now established as a distinct type of non-Hodgkin lymphoma.2 The stomach is the most common mucosal site of involvement.3 Various microorganisms have been associated with MALT lymphomas depending on the site. In the gastric location, the causative role of Helicobacter pylori infection has been established by epidemiologic and molecular studies, and therapeutic success is achieved by eradication of this bacteria with regression of disease.4-6 For other locations, although the use of antibiotic therapy is useful in some instances of orbital MALT lymphoma (Chlamydia psittacci), it is not uniformly successful in achieving durable control. Moreover, for some sites (eg, salivary gland and thyroid MALT lymphomas), there has been no documentation of a role of microorganisms, but rather association with autoimmune disorders such as Sjogren syndrome and Hashimoto thyroiditis.7, 8 Because MALT lymphomas are frequently localized (70-90%) and tend to stay localized for a long time, local treatment approaches are attractive, and radiotherapy (RT) is often a good choice in the interest of organ preservation. Although MALT lymphoma is a radiation-sensitive disease, to the best of our knowledge, it is not known whether local therapy can result in long-term control of the disease, and whether a high recurrence rate is expected as in follicular lymphomas. In this report, we examined the long-term outcome in patients with localized MALT lymphomas treated with involved field RT.


We retrospectively analyzed the charts of 192 consecutive patients who had biopsy-proven stage I and II extranodal MALT lymphomas between 1989 and 2004. This is an updated analysis on an expanded cohort of patients previously reported in 2003 when 103 patients were included with a median follow-up of 5 years.9 Patients who had transformed lymphoma at initial diagnosis were excluded. Patient characteristics are detailed in Table 1. Patients with gastric lymphomas were investigated initially for H. pylori. Five patients had tumors not associated with H. pylori and therefore received definitive treatment with irradiation. Six who presented in the early years of the study (1989-1995) were not tested for H. pylori and did not receive antibiotics. The remaining patients (n = 14) who presented beyond 1995 and were H. pylori-positive were treated with antibiotics first, and RT recommended only if they had persistent or recurrent lymphoma despite successful H. pylori eradication (median duration from antibiotics to irradiation was 10 months; range, 2-17 months). Connective tissue disorders were confirmed by testing specific antibodies.

Table 1. Patient Characteristics
CharacteristicsNo. of Patients (N = 167)Percentage of Patients
  • a

    Nasopharynx (3 patients), maxillary sinus (1 patient), larynx (1 patient), and hypopharynx (1 patient).

  • b

    Rectum (1 patient) and meninges (1 patient).

Age at diagnosis, y
Ann Arbor stage
Maximum tumor bulk
 ≤5 cm13883
 >5 cm127
 Not stated1710
Anatomical sites
 Orbital adnexa7142.5
 Salivary glands2816.8
 Other head and neck sitesa63.6
 Urinary bladder42.4
 Skin and soft tissue31.8


The treatment guideline followed was that of primary RT for stage I/II disease, except for an uncommon situation of a complete surgical excision of the disease (eg, in the skin site), in which surgery could be regarded as adequate therapy and no adjuvant RT was recommended. Occasionally, combined chemotherapy and RT was recommended at the discretion of the treating physician, usually due to the presence of bulky tumor (>5 cm), or other indication of aggressive clinical behavior, such as rapid tumor growth. Of the 192 patients, 167 received RT and among these 144 (86%) received RT alone. Twenty-three patients (14%) received combined chemotherapy and RT. The remaining 25 patients who did not receive RT were excluded. Their treatments were comprised of surgery alone (n = 14), antibiotics alone (n = 5), chemotherapy alone (n = 1), surgery followed by chemotherapy (n = 2), and those who refused treatment and were observed (n = 3). A detailed description of RT doses is shown in Table 2. Common involved field RT prescriptions were 25 Gy in 10 fractions for orbital lymphomas and 30 Gy for nonorbital sites. Response assessment was performed at 2 to 6 months after RT with clinical examination and appropriate imaging, or interventional studies (eg, endoscopy for gastric MALT). Routine endoscopy was performed every 4 to 6 months for gastric lymphomas up to 2 to 3 years, and thereafter history and physical examination during clinic visits. The median follow-up time was 7.4 years (range, 0.67-16.2 years). A total of 119 patients (71%) had >5 years of follow-up, and 38 (23%) had >10 years.

Table 2. Radiation Doses with Respect to Anatomical Locations for MALT Lymphomas
Anatomic SiteNo. of Patients25 Gy>25-30 Gy35 GyOthera
  • MALT indicates mucosa-associated lymphoid tissue; Gy, grays; H&N, head and neck; GI, gastrointestinal.

  • a

    Others: 1 patient: salivary gland, 17.5 Gy; 2 patients: stomach, 20 Gy; 1 patient, lung, 31 Gy; 1 patient, bladder, 31 Gy; and 1 patient, bladder, 34 Gy.

Orbital adnexa716551 
Salivary glands2822411
Other H & N sites615  
Lung3 2 1
Urinary bladder4  22
Skin and soft tissue3 3  
Breast4 13 
Other GI sites (rectum)1  1 
Meninges1 1  

RT Technique

Patients received involved field RT. For stage IE disease, prophylactic treatment to the regional lymph nodes were not intentionally done, although for some sites (eg, the thyroid gland, parotid gland, and stomach), invariably the direct drainage lymph node basin is almost always covered.

The macroscopic (gross) tumor volume (GTV) consisted of the primary tumor; the clinical target volume (CTV) was individualized but generally included the whole extranodal organ in which the tumor arose. For orbital adnexal lesions, the CTV included the entire orbit. In the first two-thirds of the study period, the orbital lesions were treated with either 60Co or 4-6-megavolt (MV) x-ray beams, either as a single direct anterior port or a wedged pair of fields for the posterior orbital lesions. Later, high-energy electrons were used as a single anterior field, with a suspended 8-mm-diameter pencil-shaped lens shield to protect the lens. When possible, the lacrimal gland was also shielded.

For gastric MALT lymphoma, there was an evolution from simple beam arrangements to conformal techniques (since 1998). Computed tomography (CT)-based planning is the standard with prescribed dose of 30 to 35 Gy to the entire stomach, paragastric, and celiac lymph nodes while keeping exposure of the kidneys and the liver to within tolerance limits. The planning objective for the organs at risk follows the principle of as low as reasonably achievable (ALARA). Patients were given dietary restrictions to minimize distension and improve the day to day targeting. Respiratory motion was dealt with by ensuring adequate coverage under fluoroscopy, and later in the study period by using 4-dimensional CT planning performed during quiet breathing. For stage IE parotid tumors, the entire parotid gland including the deep lobe was treated with emphasis to spare the contralateral uninvolved gland (either with wedged pair technique, or direct mixed photon/electron beam), evolving recently with the use of IMRT technique (since 2004). For patients with stage IIE disease, CTV included the parotid and the ipsilateral cervical lymph nodes. Bilateral parotid lymphomas were treated with bilateral parotid irradiation as single course. Pretreatment dental assessment and meticulous follow-up care was always an integral part of management. For thyroid lymphomas, patients who underwent total thyroidectomy with complete disease removal were given observation as an option, whereas patients who underwent subtotal thyroidectomy received radiation to the residual thyroid and the level 3, 4, and 6 cervical lymph nodes and the superior mediastinal lymph nodes.

Statistical Analysis

The endpoints were overall survival (OS), disease-free survival (DFS), recurrence-free rate (RFR), and cause-specific survival (CSS). Time was calculated from the date of diagnosis to the event of interest. For OS, death due to any cause was considered. For CSS, death as a result of lymphoma or its treatment complication was considered. For RFR, time to first treatment failure was considered whereas for DFS, time to first failure or death due to any cause were considered as events. The survival plots were calculated using the Kaplan-Meier method. CSS and RFR percentages were estimated using cumulative incidence function, and covariates were tested using the Gray test. Multivariate analysis for DFS was performed using Cox proportional hazard model and for RFR, the Fine and Grey model was used.


Tumor characteristics are presented in Table 1. Among the 71 orbital tumors, 40 were located in the conjunctiva, 9 in the lacrimal gland, 6 in periorbital soft tissues, and 11 in eyelids (5 not documented). Twelve patients had bilateral orbital involvement at presentation. Among the 28 parotid gland patients, 20 had unilateral and 4 had bilateral involvement. Three patients had submandibular gland involvement, whereas only 1 patient had sublingual gland involvement.

Treatment Outcome


Survival plots are illustrated in Figure 1. The 5-year OS, CSS, RFR, and DFS rates of the 167 patients of stage I and II MALT lymphomas who received RT and systemic therapy were 95%, 99%, 82%, and 79%, respectively. The corresponding 10-year estimates were 87%, 98%, 76%, and 68%, respectively. The 5-year OS, CSS, and RFR rates of the 144 patients who received RT alone were 94%, 99% and 81% respectively; and the corresponding 10-year estimates were 89%, 98%, and 74%, respectively. Overall, there were 17 deaths (10%) reported at last follow-up, 14 due to unrelated causes and 3 due to lymphoma. OS was found to be significantly different according to age (98% at 10 years for patients aged <60 years and 76% for patients aged ≥60 years; P = .0004). The 10-year RFRs for thyroid and gastric lymphoma were 95% and 92%, respectively, whereas for salivary gland and orbital lymphoma, it was 68% and 67%, respectively. Thyroid and gastric MALT lymphoma together had a more favorable RFR than the other sites (P = .004) (Fig. 2). The 10-year RFR for these 2 sites combined together was 93% as compared with 68% for the other sites. Site of the disease (stomach and thyroid MALT) was found to be the only significant factor for both DFS and RFR on multivariate analysis.

Figure 1.

Kaplan-Meier curves are shown demonstrating the 5-year and 10-year overall survival (OS), cause-specific survival (CSS), and recurrence-free rate (RFR).

Figure 2.

Recurrence-free rates (RFRs) of patients with gastric and thyroid mucosa-associated lymphoid tissue (MALT) lymphoma (indicated by solid line) compared with patients with MALT lymphoma at other sites (indicated by hashed line) are shown.

Patterns of Disease Recurrence and Prognostic Factors for Treatment Failure

A total of 31 patients (19%) had recurrence of disease either at local, distant, or contralateral sites, as detailed in Table 3. The pattern of recurrence is illustrated in Table 3. The majority of the recurrences (25 of 31 patients, 80%) occurred within 5 years, of which 13 patients (52%) developed disease recurrence within the first 2 years. A minority (6 of 31 patients, 19%) developed disease recurrence beyond 5 years. Nine of 31 (29%) patients developed disease recurrence in the contralateral paired organ (isolated contralateral paired organ in 6 patients and contralateral organ followed by distant disease recurrence in 3 patients). Six of 9 patients (67%) who had isolated contralateral paired organ recurrence (orbit in 3 patients and parotid in 3 patients) were in disease remission after treatment for their recurrence, whereas 1 patient (Patient 16 in Table 4) who had developed disease recurrence in the contralateral orbit followed by lung remained free of disease after oral chemotherapy with chlorambucil and prednisone, and 2 patients (Patients 14 and 24 in Table 4) who had developed disease recurrence in the contralateral orbit and parotid followed by brain and lung, respectively, had stable disease at distant sites after treatment with RT and chemotherapy. Distant recurrences were observed in 17 of 31 (55%) patients. Only 2 patients with orbital MALT developed disease recurrence at the same site. Moreover, there was no consistent pattern/location of recurrence evident, other than for a predilection of MALT sites rather than lymph nodes. Common sites of disease recurrence include salivary glands; soft tissues in trunk or extremities; and, occasionally, in lung, gastrointestinal, genitourinary organs, or lymph nodes remote from the primary tumor site (Table 4). The overall survival rate after treatment (or observation) for disease recurrence was 83% at 5 years (Fig. 3).

Figure 3.

Kaplan-Meier curve demonstrating survival rates after treatment failure is shown.

Table 3. Sites of Failure/Disease Recurrence
Primary Site of DiseaseNo. of PatientsNo. of Disease RecurrencesSites of Failure
  • L indicates local; D, distant; C/L, contralateral; L & D, local and distant.

  • a

    Two patients who developed C/L disease recurrence subsequently developed disease recurrence at distant sites (brain and lung).

Salivary glands287142
Other head and neck633
Skin and soft tissue30
Table 4. Details of Patients Who Had Responded to Definitive Therapy and Then Developed Disease Recurrence
Patient No.Age, Years/SexPrimary SitePrimary TreatmentRT Dose, GyaRecurrence Site(s)Time to Recurrence, YearsTreatment of Recurrent DiseaseOutcomeMALT Lymphoma StatusOS, Years
  • RT indicates radiotherapy; Gy, grays; MALT, mucosa-associated lymphoid tissue; OS, overall survival; CHOP, cyclophosphamide, doxorubicin, vincristine, and prednisone; PD, progressive disease; B/L, bilateral; SD, stable disease; Oral CT, chlorambucil, prednisone or cyclophosphamide; R-CVP, rituximab, cyclophosphamide, vinblastine, and prednisone; C/L, contralateral; NED, no evidence of disease; R-CHOP, rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone; RP, retroperitoneal; NA, details not available; NPX, nasopharynx; CNS, central nervous system; CSF, cerebrospinal fluid; CSI, craniospinal irradiation; PEN, prednisone, etoposide, and mitoxantrone.

  • a

    Dose/no. of fractions.

  • b


  • c

    Patient had follicular lymphoma of the breast, parotid, and neck lymph nodes.

  • dDead due to cryoglobulinemia, vasculitis, or amyloidosis.

177/WomanOrbitCHOP, RT35/20Boneb1.3RTDeadPD2.3
244/WomanOrbit (B/L)RT25/10Right orbit1.5ObservationAliveSD4.6
379/WomanOrbit (B/L)RT25/10Neck lymph nodes, (soft tissue thigh, axilla)2.5Oral CT, R-CVPAliveSD5.2
470/WomanOrbitRT25/10C/L Orbit3.1RTAliveNED12.7
563/ManOrbitRT25/10Soft tissue of arm3.2RTAliveNED12.5
685/WomanOrbitRT25/10Small bowelb0.6Resection plus oral CTDeadPD1.7
739/WomanOrbitRT25/10C/L Orbit0.8RTAliveNED4.9
8c53/WomanOrbitRT25/10Soft tissue (temporal region)b0.3R-CHOP and RTAliveNED6.5
951/WomanOrbitRT25/10Parotid (skin, soft tissue)5.5Observation (parotid) RT (arm)AliveSD7.3
1077/ManOrbitRT25/10Paraspinal mass6.2RTAliveSD7.2
1163/WomanOrbitRT25/15Parotid, (lung)3.7RT to parotidAliveSD9.9
1263/ManOrbit (B/L)Oral CT & RT25/10RP lymph nodes, orbit, (ureteric mass & prostateb)2.6R-CHOP and RTAliveSD7.8
1335/ManOrbit (B/L)RT25/10Left orbit4.8Oral CTAliveNED6.2
1463/ManOrbitRT25/10C/L orbit (brain)3.2RT (orbit), methotrexateAliveSD8.6
1524/WomanB/L orbitRT25/10Right orbit (breast)5.2Oral CTAliveSD8.3
1680/WomanOrbitRT25/10C/L orbit (lung)1.3Oral CTDeadNED2.2
1734/WomanOrbitRT25/10Parotid, neck lymph nodesb0.4RT and CHOPAliveNED8.0
1831/WomanOrbitRT25/10C/L orbit1.5RTAliveNED7.7
1972/WomanParotidRT30/20C/L parotid7.6ResectionAliveNED7.8
2037/ManParotidRT30/20C/L parotid9.2ResectionAliveNED10
2142/WomanParotidRT30/20C/L parotid1.6RTAliveNED10.5
2263/WomanParotidRT17.5/10Parotid, (RP and neck lymph nodes), skin3.2RT (neck) and oral CTDeaddSD and multiple causes6.1
2345/WomanParotid & neckRT30/20Neck (inguinal lymph nodes)5.1Observation (neck) surgery (groin)AliveSD5.3
2450/WomanParotidRT30/20C/L parotid, (lung)1.5RT and R-CVPAliveSD10
2569/WomanThyroidRT30/20Mediastinum, lung, B/L hilar lymph nodes2.3NADeadPD2.9
2667/ManParotidRT30/20Parotid, neck lymph nodes (kidneys)0.9NAAliveSD7.3
2774/WomanBreastRT35/20Peritoneum, soft tissue of arm, extradural mass at T3-T73.3Resection and RTAlivePD6.5
2856/WomanNPXRT30/20CNS (CSF) & duodenum0.4CSI plus intrathecal methotrexateAliveNED11.5
2960/WomanNPXRT30/20Skin, soft tissue over scapulab3.4CHOP plus RTAliveNED9.7
3054/WomanLarynxCVP & RT30/20Oral cavity1.2Oral CTAliveNED10.2
3177/WomanBreastRT35/20NPX, RP lymph nodes2.9PENAliveNED5.6

Of the prognostic factors analyzed, the only significant factor was anatomic site. Patients with gastric and thyroid MALT lymphomas had a control rate of 92% and 95% respectively, whereas significantly inferior results were noted for other sites.

RT alone

Among 144 patients treated with RT alone, a complete response (CR/CRu) was observed in 143 (99%). Partial response was noted in 1 patient who had extensive bilateral orbital disease with invasion into the ethmoid sinus, and who was subsequently treated with multiagent chemotherapy. This patient died as a result of chemotherapy-related sepsis 5 years from the time of the initial diagnosis. Among the 143 patients achieving CR, 28 experienced disease recurrence. Only 2 patients developed recurrence exclusively at the local site. Nine patients developed disease recurrence in the nonirradiated contralateral organs, 15 had distant recurrences, and 2 patients developed disease recurrence at both local and distant sites. There were 13 deaths (2 cause specific and 11 unrelated) and 29 treatment failures (recurrence in 28 patients and partial response in 1 patient). Overall local control at 10 years was 95%.

Combined modality therapy

Twenty-three patients (13.8%) were treated with combined modality therapy (CMT). Most received the cyclophosphamide, doxorubicin, vincristine, and prednisone (CHOP) regimen (n = 17; 74%), or cyclophosphamide, vinblastine, and prednisone(CVP) regimen (n = 2; 9%). Four patients (17%) received single-agent chlorambucil. All 23 patients treated with CMT achieved CR. There were 3 recurrences in this group (2 patients developed disease recurrence at distant sites whereas 1 patient developed disease recurrence at both local and distant sites). In the CMT group, there were 4 deaths (1 cause-specific death and 3 due to unrelated causes). Because the patients were selected to receive chemotherapy for various reasons, and the chemotherapy was heterogeneous, a meaningful comparison between the CMT and RT alone groups cannot be made.

Disease Transformation

Seven patients (22.5% of the recurrences) had histologic transformation to diffuse large B-cell lymphoma (Table 5). The median time to transformation (from the initial diagnosis of MALT lymphoma) was 42 months (range, 7-116 months). The original sites of MALT lymphoma were in the nasopharynx (1 patient), breast (1 patient), and orbit (5 patients) for the patients who had transformed. Four patients received salvage chemotherapy with the CHOP ± rituximab regimen either alone or with RT and were doing well with no evidence of disease at the time of last follow-up, whereas 1 patient received the prednisone, etoposide, and mitoxantrone (PEN) regimen.

Table 5. Transformation to High-Grade Lymphoma After Disease Recurrence
Patient No.Primary SiteRecurrent Site With TransformationRecurrence HistologyTime to Transformation, YearsTreatmentOutcomeOS, YearsOther Sites of Disease Recurrence
  • OS indicates overall survival; DLBCL, diffuse large B-cell lymphoma; R-CHOP, rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone; RT, radiotherapy; SD, stable disease; RP, retroperitoneal; NPX, nasopharynx; NED, no evidence of disease; PD, progressive disease; PEN, prednisone, etoposide, and mitoxantrone.

  • a

    Patient had extranodal and lymph node follicular lymphoma that was stable and required no treatment.

  • b

    Zygomatic disease resolved after chemotherapy.

1OrbitProstateDLBCL7.7R-CHOP and RTAlive with SD9.2RP lymph nodes, ureteric mass
2NPXSkin and soft tissueDLBCL9.7CHOP + RTAlive with NED11.2Lung and trachea
3OrbitNeck lymph nodesDLBCL1.1RT & CHOPAlive with NED9.2Parotid
4aOrbitSubcutaneous soft tissue over zygomaDLBCL5.9R-CHOPAlive with NEDb8.2Parotid, breast, neck lymph nodes
5OrbitHip and soft tissueDLBCL1.1RTDead with PD2.3None
6OrbitSmall bowelDLBCL0.6Resection, chlorambucilDead with PD1.8None
7BreastRP lymph nodes, NPXDLBCL2.7PENAlive with NED5.6None

Second Malignancy and Connective Tissue Disorders

Among the entire cohort, 19 patients were documented to have second malignancies. The probability of a second cancer was 6% at 5 years and 11% at 10 years with a relative risk of 1.33 (P value = .23) (Fig 4). Most of the malignancies were at distant sites from the irradiated area, and were of diverse histology. Two patients developed second malignancies within the previous radiation fields 5 and 7 years, respectively, after RT. Another patient who underwent total thyroidectomy had synchronous papillary carcinoma and MALT lymphoma of the thyroid and was treated with radioactive 131iodine and had CR. Two deaths were due to second malignancies (pancreatic cancer and metastatic renal cell cancer).

Figure 4.

Cumulative incidence of second malignancies at 5 years and 10 years in patients with stage I and II mucosa-associated lymphoid tissue (MALT) lymphoma with long-term follow-up are shown.

Connective tissue disorders (CTD) were documented in 16 of 167 (9.6%) patients. Among those with CTD, Sjogren syndrome was the most common, noted in 14 of 16 patients (87%). Other diagnoses included systemic lupus erythematosus (2 of 16 patients) and rheumatoid arthritis (3 of 16 patients). In addition, 2 patients (both of whom had parotid MALT) had amyloid deposition. None of the gastric MALT lymphoma patients had CTD. The majority of patients with CTD had MALT lymphoma in the parotid gland (10 of 16 patients; 63%). Patients with CTD (n = 16) had a poorer disease outcome than those without it (n = 129) (10-year RFR, 48% vs 77%; P = .04). These associations should be interpreted with caution due to the small number of events and the high correlation of Sjogren syndrome with parotid MALT lymphoma.


MALT lymphomas account for 7% to 8% of all newly diagnosed cases of non-Hodgkin lymphoma, ranking third in frequency and only lower to diffuse large B-cell lymphoma and follicular lymphoma. Its incidence has been rising over the past 2 decades.10, 11 These tumors present as localized disease (Ann Arbor stage I and II) in 70- to 90% of cases.9, 12, 13 A minority of patients (15%) will have presented with adverse prognostic factors described by the International Prognostic Index (IPI).14

In this report, the treatment outcome of uniformly treated patients with localized MALT lymphoma was presented. The median age and female predominance are comparable to other published series in the literature.14, 15 The pattern of anatomic sites of distribution in this study differed from other reports, with orbit being the most common followed by salivary gland, gastric, and thyroid MALT lymphoma (Table 1). This reflects the bias in referral pattern because of a geographic centralization of specialized ocular, head, and neck oncology and radiation oncology programs in our institution, and possibly local prevalence of predisposing factors and its successful management in the community (eg, H. pylori infection for gastric lymphomas). Other investigators have found a higher proportion of gastric,13, 14 salivary gland,15 or skin lymphomas.13, 15

The optimal treatment approach for localized nongastric MALT lymphoma is still evolving. Micro-organisms have been implicated in the pathogenesis of MALT lymphoma arising from some locations,16 with successful treatment by using antibiotics, most notably doxycycline for orbital MALT lymphomas.17-19 However, the results are still premature to consider it as a primary modality of therapy because durable remission has only been documented in 48% in a cohort of 27 patients treated19 and that a uniformly high response rate has not been reported by other investigators.20 However, it is a nonmorbid treatment, and therefore an attractive initial treatment approach and can spare or delay the use of cytotoxic therapy such as RT. The use of RT is still considered a standard approach as excellent results have been confirmed by many series with moderate doses of 30 Gy.21-23 This study would support the use of even a slightly lower dose of 25 Gy (in 10-15 fractions) in orbital lymphomas as local control remained excellent. Nongastric MALT lymphomas of other sites also respond extremely well to moderate-dose RT (30-35 Gy) with durable local control. However, there is a pattern of preferential recurrence (or new primary disease) in paired organs and at distant MALT sites in the present study, a finding also reported by other investigators (recurrences at sites such as the spleen,24 bone marrow,13, 14, 25 soft tissues (this study), and distant lymph nodes26). An isolated contralateral recurrence in the paired organ (typically orbit, or salivary gland) can be successfully managed with another course of RT and often does not result in further disease recurrence in distant sites. Perhaps these can be regarded as new disease rather than a recurrence of the original malignant clone, and further studies are warranted in this regard.

If advanced stage of disease is also considered, recurrences after treatment may occur in up to 40% of patients as reported by 2 large European studies.14, 25 The crude recurrence rate in this study was 19%, similar to our previous results for early stage patients, and comparable with a recurrence rate of 22% in 77 early stage patients treated at the Joint Center in Boston.26 Recurrences were common in patients with orbital and salivary gland lymphomas, but there were no disease recurrences noted in patients with gastric MALT (Table 3). The majority of the recurrences in this series occurred within 5 years. This is in contrast to 1 study indicating that it is not uncommon to develop disease recurrence late in the course of follow-up.27 Recurrences of disease typically behave in an indolent manner, and prolonged survival was observed even after disease recurrence (Fig. 3).

Despite the cases of disease recurrence, this study documented excellent survival outcome in patients treated with RT, comparable with other series.13, 15, 26 The excellent outcome of both thyroid and gastric MALT lymphomas was previously reported from our institution, and the data are now more mature with no evidence of late recurrence past 5 years (Fig. 2).

In gastric MALT lymphomas, a very high cure rate with H. pylori eradication therapy28, 29 results in a relatively small proportion of patients requiring subsequent RT for those not controlled with antibiotic therapy. Similar to our experience, Schechter et al30 reported 100% local control rate in 17 patients treated with RT. Similarly good results are reported by Tsai et al,26 Vrieling et al,31 and Yamashita et al.32 Recent evidence indicates t(11;18)(q21;q21) translocations in gastric MALT predict antibiotic resistance33; therefore, in this subgroup, RT is a good treatment option upfront, as well as for those who are H. pylori-negative. Patients with thyroid MALT lymphoma also fared well with RT, with only 1 patient reported to have developed disease recurrence (Table 3). Similar results were reported by Zucca et al15 and Zinzani et al25 in their series, in which no disease recurrences were observed in patients with thyroid MALT. Moreover, in a small series of patients treated with thyroidectomy alone, all had excellent outcome.34 Therefore, the role of RT after total thyroidectomy and complete tumor excision is unclear.

Because MALT lymphomas have an affinity to involve multiple mucosal sites, it is possible that some of the recurrences may in fact represent progression of disease that was not detected initially. In a retrospective study of 36 nongastric MALT lymphomas, endoscopy as a part of staging revealed occult gastric involvement in 12 (33%).35 The high rate of unusual sites of disease recurrences seen in this study has also been commented by others,14, 26 stressing the need for accurate initial staging to improve patient selection for local curative treatment.

The use of moderate-dose RT is tolerated well, and long-term toxicity is infrequent. In the current series, no serious acute toxicity was observed. Patients with orbital lymphoma occasionally developed cataracts 2 to 5 years after treatment, which were successfully treated with surgical extraction. With appropriate lens shielding, the risk of cataracts is reduced to 14%. Patients with salivary gland MALT lymphoma associated with Sjogren syndrome had significant xerostomia, which were often symptomatic and required dietary modifications.

The initial role of chemotherapy cannot be determined in this series, but it must have contributed to control of the disease as all patients had partial response or CRs from the chemotherapy before the initiation of RT. The role of chemotherapy alone has been suggested because there are many active drugs.6, 36, 37 Rituximab has clinical activity with a response rate of 73% and a median response duration of 10.5 months reported in 1 study.38 However, a large retrospective study conducted by the International Extra Nodal Lymphoma Study Group (IELSG) analyzing the results of 180 patients did not find any difference in clinical outcome between initial localized treatment approaches with systemic chemotherapy.15 Consequently, chemotherapy may not be necessary for patients with localized MALT lymphoma.

Transformation in MALT lymphomas is fortunately uncommon. The transformation rates reported in various studies ranges from 3% to 19%.14, 15, 26 In the current series, the rate is 4% which is consistent with the reported literature. Once transformation occurs, the fatality rate is reported to be higher than for MALT lymphoma.14, 15, 26

Because patients tend to have prolonged survival, 1 of the serious long-term morbidities is the development of a second malignancy, which in this series was 11.4%. Although some have reported a higher incidence of second malignancies (20-22%) associated with MALT lymphoma,38, 39 the relative risk was not found to be above a matched normal population in our series of patients, a finding also supported by Au et al.39


Localized extranodal MALT lymphomas can be successfully treated with moderate-dose RT, which results in excellent long-term control rates and minimal toxicity. There is a significant risk of recurrence of MALT lymphomas presenting in sites other than the stomach and thyroid, with unpredictable patterns of recurrence. Despite disease recurrence, the natural history is indolent and the disease remains responsive to further cytotoxic therapy and the overall survival remains excellent after salvage therapy.


The authors made no disclosures.