Ocular adnexal mucosa-associated lymphoid tissue lymphoma: a single center experience of 32 patients from China

Abstract

MALT represents the most common subtype in ocular adnexa lymphoma. Little data, however, has been reported in China. We consecutively analyzed 32 patients from April 2008 to January 2016. Median age at diagnosis was 57 (32-90) years. Most patients presented with stage IE (30/32, 87.5%) and the remaining 2 staged IVE. Treatment and followup data were available for 29 patients. Other than resection and biopsy for diagnosis, most of them (20/29, 68.9%) received radiotherapy. The rest underwent chemotherapy (1, 3.5%), both radiotherapy and chemotherapy (2, 6.9%), or ‘watch and wait’ (6, 20.7%). With a median followup of 28 (6-94) months, all of the patients are alive. Two cases progressed at 24 months and 30 months. Comparing the PFS for four treatment arms, no significant difference can be observed (P=0.147). Our data demonstrated the chronic clinical course and excellent prognosis of OAML. For localized cases, surgery combined with radiotherapy is recommended.

Introduction

Lymphoma is the most common subtype of malignancy among ocular adnexa malignancies. It can involve conjunctiva, lacrimal gland, orbit, eyelids, and orbital muscles. Ocular adnexa lymphoma (OAL) was reported for the first time in 1952 by Feinstein and Krause [1]. The incidence of OAL was described as 0.18-0.31 per 100,000 per year in the 1970s [2] and it has increased gradually in the last forty years [3].

The histology of OALs are mostly of low-grade (84%) with only 16% high-grade [4]. Ocular adnexa MALT lymphoma (OAML) is the most common subtype of OAL and others less common include diffuse large B cell lymphoma (DLBCL), follicular lymphoma (FL), lymphoplasmacytic lymphoma (LPL), and mantle cell lymphoma (MCL) [5-7].

However, little data about OAML cohorts have been reported in China. Thus, we consecutively analyzed 32 OAML cases diagnosed and treated in our hospital to assess their clinical characteristic, treatment and outcomes.

Material and methods

Patient selection and diagnosis

Thirty-two OAML cases diagnosed and treated in our hospital from April 2008 to January 2016 were enrolled in our research. All of them were provided informed consent for our institutional guidelines at their first visit, according to the 1975 guidelines of the Declaration of Helsinki.

Data of patient gender, age, Ann Arbor stage, ECOG performance status, serum lactate dehydrogenase (LDH), the number of extra nodal sites involved (ESI), β2-microglobulin (β2-MG), serum ferritin (SF), serum paraprotein, erythrocyte sedimentation rate (ESR), C-reactive protein (CRP) and virus (EBV/HBV) status were collected. The International Prognostic Index (IPI) was used for prognostic stratification. All of the patients underwent bone marrow aspiration and radiological scanning of not only orbit, but also the whole body.

Histological diagnosis of all of the patients with OAML was performed, according to the World Health Organization classification for hematologic and lymphoid tissues tumors (2008 version), by reference pathologists.

Statistical analysis

The statistical analyses were performed using the SPSS software (version 18.0) and Graphpad Prism 6. Chi square-test was used to test association between two categorical variables. Progression-free survival (PFS) was defined as the period from the date of diagnosis of OAML to disease progression after the first treatment or to the last followup date. Overall survival (OS) was defined as the period from diagnosis to the date of last followup or death. We used the Kaplan-Meier curve to describe PFS and OS and log-rank analysis to test the differences.

Results

Baseline clinical characteristics

Thirty-two patients with histologically verified OAML who were diagnosed and treated at our center between April 2008 to January 2016 were enrolled in this study. As shown in Table 1, the median age at diagnosis was 57 years (range, 32-90 years) with a male predominance (male/female, 21/22). The median followup time was 28 months (range, 6-94 months).

Table 1.

Characteristics of OAML patients at diagnosis and first-line therapy

		Number of patients	Proportion
Gender	Male	21	65.6%
	Female	11	34.4%
Age	Median/Range	57 (32-90)
Involved sites	Conjunctiva	3	9.4%
	Lacrimal gland	2	6.3%
	Orbit	15	46.8%
	Eyelid	10	31.2%
	Orbital muscles	2	6.3%
Stage	IE	30	93.8%
	IVE	2	6.2%
ECOG	0	31	96.8%
	≥1	1	3.2%
LDH	Normal	31	96.8%
	Over ULN	1	3.2%
IPI	0-1	30	93.8%
	>1	2	6.2%
First-line therapy (n=29)	Surgery	6	20.7%
	Surgery/Radiotherapy	20	68.9%
	Surgery/Chemotherapy	1	3.5%
	Surgery/Chemotherapy/Radiotherapy	2	6.9%

IPI = International Prognosis Index; ECOG = Eastern Cooperative. Oncology Group; LDH = lactate dehydrogenase.

OAML usually presented with asymptomatic ocular adnexal mass and ophthalmologists practiced biopsy or surgical excision to confirm the diagnosis. Typical histological changes and immunophenotypic finding of the tumors is CD20 (+), PAX5 (+), BCL-2 (+), CD3 (-), CD5 (-), CD10 (-), BCL-6 (-), CCND-1 (-), and Ki-67<10%, as shown in Figure 1. Most common site of OAML was orbit (15/32, 46.8%), followed by eyelid (10/32, 31.2%), conjunctiva (3/32, 9.4%), lacrimal gland (2/32, 6.3%), and orbital muscles (2/32, 6.3%). Most of the patients had localized disease, that is, stage IE according to Ann Arbor (30/32, 93.7%). Two (6.3%) presented with stage IVE, due to bone marrow involvement. Bone biopsy and immunohistochemistry (IHC) of patient No.11 was: CD5-, CD10-, CD20+, PAX5+, CD23-, BCL-2 weakly positive, and B-cell lymphoma bone marrow invasion. Bone marrow flow cytometry of patient No.14: atypical lymphocytes were positive for CD19 and CD20; negative for CD5, CD10, and CD23. CD19 positive cells expressed lambda light chain restriction. B symptoms were observed in only one case. During followup, no central nervous system involvement was found.

Figure 1.

Typical histological changes and immunophenotypic findings of OAML. (A) Hematoxylin and eosin, × 40; (B) Hematoxylin and eosin, × 100; (C) Hematoxylin and eosin, × 400. Staining with an immunohistochemical marker of (D) CD20 (original magnification × 200); (E) PAX-5 (original magnification × 200); (F) BCL-2 (original magnification × 200); (G) BCL-6 (original magnification × 200); (H) CCND-1 (original magnification × 200); (I) Ki-67 (original magnification × 00); (J) CD30 (original magnification × 200); (K) CD5 (original magnification × 200); (L) CD10 (original magnification × 200).

The International Prognostic Index (IPI) was available in all 32 patients and 30 of them (93.8%) were at low risk (IPI 0-1). Thirty-one (96.8%) patients were rated zero for European Cooperative Oncology Group (ECOG) performance status at diagnosis. Beta 2-MG level was measured in 29/32 patients and was elevated in 17.2% (5/29). SF was increased in 17.9% (5/28), ESR (erythrocyte sedimentation rate) in 25.0% (7/28), LDH (lactate dehydrogenase) in 3.2% (1/32), and CRP (C-reactive protein) in 10.7% (3/28). EBV-DNA copy numbers were available in 22 patients and 1 of them (4.55%) was over the upper limits of normal (5000 copy number). HBsAg was positive in 9.4% (3/32). No evidence of hepatitis C infection was found (0/32). Thyroid function test was performed on 16 patients. Three had abnormalities but no one could meet the diagnostic criteria of Grave’s disease or chronic autoimmune thyroiditis (Hashimoto’s thyroiditis). cANCA/pANCA/PR3-cANCA/MPO-ANCA were performed on 11 patients and all of them were negative. We also had paraprotein tested in 30 patients. IgM was elevated in 10/30 (33.3%), IgA in 3/30 (10%), and IgG in 1/30 (3.3%).

Treatment and outcomes

Different types of treatment were practiced on OAML patients and treatment details were available in 29 of 32 patients. Treatment details and brief clinical characteristics of patients are shown in Table 2. After diagnosis, 20 of 21 (68.9%) patients received radiotherapy, 1/29 (3.4%) patients were given chemotherapy, and 2/29 (6.9%) received both radiotherapy and chemo-immunotherapy/chemotherapy. The remaining six (20.7%) did not undergo any further treatment besides surgery and decided to watch and wait.

Table 2.

Characteristics, treatment, and survival of OAML patients

No	Gender	Age	Localization	Ann arbor Stage	AJCC TNM stage	1st-line treatment	Chemotherapy (first-line)	Progression	PFS (months)	OS (months)	Alive
1	M	46	Orbit	IE	T2NXM0	Surgery/Radiotherapy		N	81.77	81.77	Y
2	F	60	Orbit	IE	T2NXM0	-		-		-	-
3	F	65	Lacrimal gland	IE	T1NXM0	-		-		-	-
4	M	90	Eyelid	IE	T3NXM0	Surgery		Y	24.5	43.77	Y
5	M	60	Orbit	IE	T2NXM0	Surgery		N	68.43	68.43	Y
6	F	58	Eyelid	IE	T3NXM0	Surgery/Radiotherapy		N	58.83	58.83	Y
7	M	69	Orbit	IE	T2NXM0	Surgery/Radiotherapy		N	58.13	58.13	Y
8	M	53	Conjunctiva	IE	T1NXM0	Surgery/Radiotherapy		N	56.8	56.8	Y
9	M	57	Orbit	IE	T2NXM0	-		-		-	-
10	M	63	Eyelid	IE	T3NXM0	Surgery		N	54.6	54.6	Y
11	F	55	Orbit	IVE	T2NXM1	Surgery/Chemoimmunotherapy/Radiotherapy	R-CHOP*4	N	42.1	42.1	Y
12	F	52	Eyelid	IE	T3NXM0	Surgery/Radiotherapy		N	42.07	42.07	Y
13	F	50	Orbit	IE	T2NXM0	Surgery		N	41.77	41.77	Y
14	M	74	Eyelid	IVE	T3NXM1	Surgery/Chemotherapy	miniCHOP*6	N	28.3	28.3	Y
15	M	78	Orbit	IE	T2NXM0	Surgery/Radiotherapy		N	27.67	27.67	Y
16	M	42	Orbit	IE	T2NXM0	Surgery/Chemotherapy/Radiotherapy	CHOP*3	N	26	26	Y
17	F	44	Eyelid	IE	T3NXM0	Surgery/Radiotherapy		N	26.47	26.47	Y
18	F	52	Oorbit	IE	T2NXM0	Surgery/Radiotherapy		N	16	16	Y
19	M	63	Orbit	IE	T2NXM0	Surgery		N	15.73	15.73	Y
20	M	55	Orbit	IE	T2NXM0	Surgery/Radiotherapy		N	12.67	12.67	Y
21	F	40	Orbital muscles	IE	T2NXM0	Surgery/Radiotherapy		N	27.07	27.07	Y
22	F	69	Orbital muscles	IE	T2NXM0	Surgery/Radiotherapy		N	20.33	20.33	Y
23	M	61	Eyelid	IE	T3NXM0	Surgery		Y	30.47	100.43	Y
24	M	57	Orbit	IE	T2NXM0	Surgery/Radiotherapy		N	17.3	17.3	Y
25	M	52	Eyelid	IE	T3NXM0	Surgery/Radiotherapy		N	12.47	12.47	Y
26	M	66	Lacrimal gland	IE	T1NXM0	Surgery/Radiotherapy		N	6.33	6.33	Y
27	M	48	Orbit	IE	T2NXM0	Surgery/Radiotherapy		N	30.3	30.3	Y
28	M	60	Eyelid	IE	T3NXM0	Surgery/Radiotherapy		N	25.97	25.97	Y
29	F	63	Orbit	IE	T2NXM0	Surgery/Radiotherapy		N	17.6	17.6	Y
30	M	63	Conjunctiva	IE	T1NXM0	Surgery/Radiotherapy		N	6.07	6.07	Y
31	F	32	Conjunctiva	IE	T1NXM0	Surgery/Radiotherapy		N	7.5	7.5	Y
32	M	53	Eyelid	IE	T3NXM0	Surgery/Radiotherapy`		N	6.83	6.83	Y

The majority of the patients (20/29, 68.9%) received local treatment which included both surgical resection and radiotherapy. All of them responded well and none progressed during followup. The total radiation dose ranged from 30 Gy to 55 Gy, mostly 34 Gy/17 f. Complications of radiotherapy included vision loss (1/20, 5.0%), dry eye syndrome (2/20, 10.0%), and cataracts (1/20, 5.0%). No other complications, such as keratitis and retinopathy, were observed.

A total of 6 (20.7%) patients were followed up without any further treatment. One patient with stage IE was diagnosed with left eye lesion (patient No.23). Thirty months after surgical resection, his eye lesion recurred. Moreover, a mass on his right eye was observed and also histologically diagnosed OAML. Considering his bilateral involvement, 6 cycles of CHOP (cyclophosphamide 1.2 g d1, epirubicin 80 mg d1, vincristine 2 mg d1, prednisone 50 mg bid d1-5) were administered and partial remission (PR) was achieved. The patient then received second-line chemo-immunotherapy of GDP (gemcitabine 1.6 g d1 d8, cis-Dichlorodiamineplatinum 100 mg d1, dexamethasone: 40 mg d1-4) for 4 cycles. With his lymph nodes and the eye lesion remaining in PR, he refused to take any further treatment. No progression had been observed until the end of our followup. One 90-year-old patient with stage IE was progressed 24 months after surgery. Considering his personal status, age, and history of prostate cancer we did not conduct any further therapy. Until the last followup, his eye lesion had been growing slowly and no evidence of other organ involvement had been observed.

One of the 29 (3.4%) patients received surgery and chemotherapy (patient No.14). This patient staged IVE received 6 cycles of mini CHOP (cyclophosphamide 0.7 g d1, epirubicin 60 mg d1, vincristine 1 mg d1, prednisone 35 mg bid d1-5). After the first-line treatment, PR was achieved and the patient refused to take further therapy. No progression had been observed until the end of our followup.

Two (6.9%) patients underwent surgery, radiotherapy and chemoimmunotherapy/chemothe-rapy. One was a 55 year old woman staged IVE (patient No.11). After surgical resection, she received chemoimmunotherapy (R-CHOP for 4 cycles: Rituximab 500 mg d0, cyclophosphamide 1.1 g d1, epirubicin 80 mg d1, vindesine 4 mg d1, prednisone 45 mg bid d1-5). Imaging assessment demonstrated PR of her eye lesion and then she received radiotherapy for 32 Gy/17 f. The other patient received CHOP (cyclophosphamide 1.4 g d1, epirubicin 90 mg d1, vincristine 2 mg d1, prednisone 50 mg bid d1-5) for 3 cycles and radiotherapy for 34.2 Gy/19 f (patient No.16). No progression was observed for both of them during our followup.

Up to our followup point, all of the 29 patients were alive with 100% of 3-years OS rate. Two patients progressed at 24 and 30 months after diagnosis, respectively, and the 3 year PFS rate was 85.0% (Figure 2). Comparing four treatment arms, no significant difference could be observed in PFS (P=0.147) (Figure 3).

Figure 2.

Kaplan-Meier curve for progression free survival of OAML patients.

Figure 3.

Kaplan-Meier curve for progression free survival of OAML patients in four treatment arms.

Discussion

MALT lymphoma can involve salivary glands, the gastrointestinal tract, bladder, and pulmonary. Ocular adnexa are one of the most common sites where MALT lymphoma can occur. We reported a series of 32 patients with OAML, comprising a relatively large cohort of OAML in China to date.

According to researchers from Korea and Japan, OAML presents at a median age of 45 years old at diagnosis and has a male preponderance [8-10]. However, reports from western countries including the United States and Italy, show an older age (median age, ~65 years) and a female preponderance (M/F=1/1.5) [11-13]. The data of our series is quite consistent with those from Korea and Japan that males make up the majority, while the median age of diagnosis is older (57 years) but still younger than the western population. It has been demonstrated that the predilection age and gender constitution of OAML may have regional variation but Asians share some similarities. Unilateral involvement is more common than bilateral. According to Tan moto [14], Decaudin [15], and Ferry [16] the proportion of bilateral cases is no more than 25%. In our cohort, no bilateral involvement was observed at diagnosis. Orbit was reported to be the most common involved site among OAML, which is also confirmed in our data. Most of the patients had local disease, usually staged IE (Ann Arbor Staging system), and IPI score was mostly at low risk. This is also confirmed in our cohort. Two patients staged IVE had bone marrow involvement at first visit which may imply that bone marrow biopsy should be a routine test for patients with OAML.

It has been frequently reported that MALT lymphoma is usually accompanied by autoimmune disorders [17,18]. Thyroid function test and ANCA series test (including cANCA/pANCA/PR3-cANCA/MPO-ANCA) were conducted in part of our cases. Three of 16 had abnormal thyroid function (lower T3/lower TG/higher A-TG and anti-T). According to a reference endocrinologist, none of them can be diagnosed with Grave’s disease or chronic autoimmune thyroiditis (Hashimoto’s thyroiditis). Among the 11 patients who had ANCA series test, the results were negative. In our series, OAML does not seem to be related to autoimmune disorders. However, more than a half of our cohort did not take these lab tests, thus more data are still needed to demonstrate the relationship between OAML and autoimmune disorders. Other risk factors evaluated, such as elevated β2-MG, TK1, ESR, CRP, SF, CA125, presence of paraproteinemia, also did not have prognosis significance in our cohort.

In order to reduce possible impairment of vision, the management of OAML requires a multidisciplinary cooperation which involves an ophthalmologist, hematologist, radiotherapist, and pathologist. Currently, treatments often applied for OAML patients include: surgery, radiotherapy, chemoimmunotherapy, and the ‘watch and wait’ strategy. In our center, patients with OAML are usually diagnosed in the Department of Ophthalmology via biopsy or resection. Subsequently, experienced hematologists will conduct an overall examination to assess the stage, prognosis, and to make a precision treatment protocol. According to certain protocol, patients will be transferred to the corresponding department (hematology or radiology). However, sometimes the protocol cannot be implemented thoroughly due to poor patient compliance.

Before any treatment, a surgical resection or biopsy of the eye lesion is of great importance. Doctors can conduct more precise management after pathological and immunophenotyping analysis. However, in situ or disseminated extra ocular relapse occurs in the surgery group more frequently than the radiotherapy group [19,20]. Thus, among the clinical course of OAML, the role of surgery is more likely to biopsy and figure out the diagnosis. In our series, six patients only received surgical resection as first-line therapy. Two of them (66.7%) relapsed at 24 and 40 months after diagnosis, respectively. Obviously, subsequent treatment after surgery is quite necessary incase of relapse.

For localized MALT lymphoma, radiotherapy is recognized as the most common management and a safe and effective treatment. Radiotherapy is reported to contribute to a local control rate of 100% and relapse rate ranging between 0%-15% [21-24]. However, radiotherapy is known not only for its high local control ratebut also for complications such as cataracts, keratitis, dry eye syndrome, and retinopathy [25-29]. In our cohorts, four cases in the radiotherapy group presented with complications after radiotherapy (one for amblyopia, two for dry eye syndrome, and the other for cataracts). All of these four cases reported that complications after radiotherapy did not affect their daily life. Recent reports have emphasized the importance of using a lead shield to protect the cornea during radiotherapy. With adequate protection, it seems that complications of radiotherapy are acceptable and manageable.

It was reported in the study by Avilés et al. that OAML patients did not benefit from the adjunct of chemoimmunotherapy to radiotherapy and the toxicities rates wassimilar in two treatment arms [30]. Another study illustrated that patients treated with radiotherapy with single reagent chemotherapy/immunotherapy and combination chemo-immunotherapy did not differ in OS rates [19]. In our center, chemo-immunotherapyis usually applied for OAML patients with systemic involvement. Radiotherapy will be added when complete remission of the eye lesion was not achieved.

For early stage OAML patients who do not receive treatment, survival rate is similar to those immediately treated. Tanimoto et al. conducted a study including 36 localized OAML cases [31]. At a median followup of 7 years, almost 70% patients did not require treatment. Forty-seven percent of patients progressed and 6% died because of progressive lymphoma. Thus, in the management of elderly and frail patients with resected lesions or asymptomatic disease, the ‘watch and wait’ strategy is strongly recommended.

There are some limitations of our series. OAL is a rare disease, likely representing as many as 8% of all extra nodal NHLs [2]. Its incidence is approximately 0.2 per 100,000 which makes it reasonable that, even though MALT lymphoma is the most common subtype of malignant lymphoma in the orbit, the incidence of OAML is quite low. Even taking this into account, our cohort is relatively small. Moreover, for indolent lymphoma, the followup time is relatively short to illustrate the prognosis precisely. It has been well demonstrated that H.pylori plays a vital role in the development of gastric MALT lymphoma. The relationship between C.psittaci and OAML is quite controversial and confusing. In 2004, Ferreri et al. first reported the presence of C.psittaci in 80% of OAML biopsies and doxycycline treatment led to positive remission of eye lesions in more than half the patients [32]. Yoo C et al., from Korea, also successfully isolated C.psittaci in approximately 80% of OAML patients [33]. Nevertheless, some research from Germany, China, Japan, USA, Cuba, Netherlands and France reported that no relationship was observed between C.psittaci and OAML [12,34-38]. Thus, there appears to be a strong regional variability in the association of C.psittaci and OAML. As a limiting condition, the test of C.psittaci has not been conducted in our center. Further future research is still needed for the field of chlamydia-lymphoma associations.

Taken together, OAML is a relatively rare subtype of non-Hodgkin lymphoma with localized and indolent clinical course, although a fraction of them slowly progressed. Our data illustrates the excellent prognosis of OAML despite various arms. Future investigations on OAML, especially prospective ones with long followingup time, are needed.

Disclosure of conflict of interest

None.