Fine-Needle Aspiration Biopsy as a Diagnostic Modality for Orbital Adnexal Lymphoma

When diagnosing orbital lesions, it is important to use a diagnostic modality with a high chance of yielding a diagnosis combined with a low risk of complication [1]. Incisional biopsy in the orbit is a procedure associated with inherent risks and complications such as diplopia, orbital hemorrhage, and nerve damage all of which could result in significant sequelae [2, 3]. A fine-needle aspiration biopsy (FNAB) is in comparison minimally invasive and with few complications, often performed in an outpatient setting, making it both time and cost-effective [4]. However, the role of FNAB in diagnosing lymphoproliferative disease has long been the subject of debate due to the limited amount of cellular material collected, and the assumed loss of histological structure in the sample. Incisional biopsy is thus still the main diagnostic tool worldwide for suspected lymphoproliferative disease in the orbit [5]. In recent years, the results of FNAB have become more precise and reliable due to increased experience in cytomorphological assessments in combination with advancements in ancillary techniques such as flow cytometry and immunocytochemistry [6].

Orbital adnexal lymphomas (OALs) are more commonly primary extra-nodal neoplasms, while around 27% originate from disseminated lymphomas, or invasion of the adnexa, i.e., secondary orbital lymphomas. It is important to distinguish between primary and secondary orbital lymphomas as they require different treatment. The majority of OALs are B-cell type non-Hodgkin lymphoma, and the majority are low-grade neoplasms. The most common form of primary OAL is the low-grade malignant, extra-nodal, marginal zone lymphoma (MZL), also called mucosa-associated lymphoid tissue (MALT) lymphoma if the epithelium of the conjunctiva or lacrimal gland is involved. Low-grade malignant follicular lymphoma (FL) is the second most common primary subtype, but the most common subtype of secondary orbital lymphomas (Fig. 1, 2) [7‒9].

Lymphomas are a fairly heterogeneous group of diseases, and therapeutic strategies are based on an increasingly precise nosology of the World health Organization classification of 2008 and the revised 2016 classification of lymphoid neoplasms. The Ann-Arbor classification, which is an anatomical staging system, includes the presence (A) or absence (B) of symptoms, and the TNM classification for OAL [10, 11]. The number of classification systems for lymphomas has led to difficulties when deciding the best treatment [12].

Morphological grades IIIA and IIIB of FL have a higher percentage of centroblasts than grades I–II and a higher proliferation rate (>30%). Therefore, there is minimal risk of misdiagnosing a high-grade FL as a low-grade malignant lymphoma when using FNAB [13, 14]. It is important to distinguish between morphological grades I–II and grades III–IV in both MZL/MALT lymphoma and FL as they require different treatment strategies. The localized high grade and disseminated orbital lymphomas are treated with both chemotherapy and radiotherapy in contrast to the localized low-grade primary orbital lymphomas which are treated with radiotherapy as a monotherapy [15].

FNAB has been one of the tools used to diagnose orbital tumors, including orbital lymphomas, at Karolinska University Hospital in Stockholm, Sweden since the 1970s [6]. The aim of this study was to evaluate FNAB as a tool specifically for the diagnosis of lymphoproliferative orbital lesions in light of the improved cytomorphological and immunological assessment methods that have evolved over recent decades.

Data were retrieved retrospectively for all adult patients who had undergone orbital FNAB at Karolinska University Hospital Stockholm, Sweden, between January 2005 and December 2015. Between the years 2005 and 2015, a total of 314 orbital FNAB were performed. Forty-eight patients were assigned a diagnosis of definitive or suspected lymphoproliferative disease, and out of these did 38 patients (12%) received a definitive lymphoma diagnosis. All patients were referred from the Oculoplastic and Orbital Services, Sankt Erik Eye Hospital, Stockholm, Sweden because the lesion under investigation was deemed suitable for FNAB. Lesions confined to the posterior part of the orbit were not sent for FNAB.

The lymphomas included were either primary or secondary orbital lymphomas. The FNABs were obtained from the orbit, including the lacrimal gland and lacrimal drainage system, but did not include lesions only involving the conjunctiva or the skin of the eyelids.

The following data were collected: age and gender, location of the tumor, FNAB date and result, repeat FNAB (reFNAB) date and result, incisional biopsy date and result, type of treatment, last follow-up, and recurrence of orbital lymphoma. Patients without follow-up data were excluded.

FNAB was performed without anesthesia in an outpatient setting, at Karolinska University Hospital, by an experienced cytopathologist. The mass was localized by palpation and reached transcutaneously using a 27-gauge needle attached to a syringe (Fig. 3). Non-palpable and deep-seated lesions were biopsied with a 25-gauge (0.53 mm) needle with a stylet, and with the assistance of a radiologist using ultrasound or computed tomography for guidance. One to three passes of cytology collection were made in every case. The procedure for FNAB has been described in detail previously [16].

Approximately half of the aspirate was ejected onto at least 2 glass slides and the smears were air dried. Rapid onsite evaluation was done with all the aspirates, one smear was stained with May-Grünwald-Giemsa (MGG), for cytomorphological examination and one smear was formalin fixed for proliferation analysis with immunological staining with monoclonal antibody, MIB-1 (Ki67). The remaining material was suspended in 1.5 mL, phosphate buffered saline, for other forms of analysis, such as immunostaining on cytospin preparations and/or flow cytometry cases when dominating cells were small mature lymphocytes the cell suspensions were sent to flow cytometry analyses which were performed by hematopathologist (Fig. 4).

Immunocytochemistry on cytospin preparations was performed with three step alkaline phosphatase anti-alkaline phosphatase method. The monoclonal antibodies used were anti-kappa, anti-lambda, CD20, CD5, CD10, and CD3. Flow cytometry was initially performed with a four-color cytometer and after 2008 an eight-color cytometer was available using a panel of monoclonal antibodies for kappa, lambda, CD19, CD5, CD23, CD10, CD20, CD4, CD7, CD8, CD3, and Bcl2.

A total of 48 patients underwent diagnostic FNAB leading to a diagnosis of definitive or suspected lymphoproliferative disease at Karolinska University Hospital, between 2005 and 2015. The median follow-up time was 61 months, range 1 month–182 months. Ten patients were excluded due to lack of follow-up data. Thus, 38 patients were included, 23 females and 15 males. The median age was 69 years at the time of diagnosis (range 44–92 years). All lymphomas were B-cell non-Hodgkin lymphomas, the majority (31 cases, 82%) being low grade (Fig. 5).

Of the 38 patients included in the study, 31 (82%) were conclusively diagnosed as having low-grade (proliferation ≤30%) or high-grade (proliferation ≥30%) lymphomas according to the first FNAB. The disease was subclassified in 20 patients (65%). The diagnosis in 7 patients (18%) was either inconclusive, suspected lymphoma, or reactive lymphatic infiltrate. The patients whom underwent the first performed FNAB were afterward cytologically categorized, as per the proposed Sydney system, for reporting lymph node cytopathology (Table 1). Of these, four underwent incisional biopsy, two underwent reFNAB plus incisional biopsy, and 1 patient reFNAB only. Reinvestigation confirmed the initial diagnosis of malignant lymphoma in 3 patients, but the diagnosis was changed in the remaining three. One of these underwent incisional biopsy as a second diagnostic modality, which also gave inconclusive results, and therefore underwent a second incisional biopsy, which showed MALT lymphoma. The patient originally diagnosed as having lymphatic infiltrate was diagnosed as having low-grade malignant B-cell lymphoma following reFNAB, and the diagnosis of the third patient who had been assigned an initial diagnosis of reactive lymphatic infiltrate was changed to MALT lymphoma. Thus, 2 patients initially diagnosed as having non-lymphoproliferative disease were found to have lymphoproliferative disease after reinvestigation.

Within the group of definitive lymphoma diagnosis, the largest group was the non-subclassified low malignant B-cell lymphoma, followed by the subclassified FL, MZL/MALT lymphoma, mantle cell lymphoma, small lymphocytic lymphoma, and high-grade B-cell lymphoma (Table 2). Seventeen out of 38 patients (45%) were diagnosed with primary orbital lymphoma, while the remaining 21 were found to have disseminated disease, i.e., secondary orbital lymphoma. All patients underwent systemic workup for lymphoma. In all the cases, final diagnosis was reported by a Cytopathologist.

Regarding treatment, 22 patients (58%) received radiotherapy only, 11 patients (29%) chemotherapy only, and 4 patients (11%) received both. The remaining patient declined treatment. Fourteen of the 17 patients with primary orbital lymphoma received radiotherapy only.

Orbital recurrence was diagnosed in 3 patients. One patient was diagnosed with low malignant B-cell lymphoma and two with FL, out of the three one, with FL, died prior to treatment.

Among all 48 FNABs performed, only one complication was reported, a periorbital swelling that was caused by a small orbital hemorrhage diagnosed by a computed tomography scan. The patient was kept under observation, the symptoms eventually subsided, and the patient was discharged.

Incisional biopsy is still the preferred and most common choice of primary diagnostic tool in suspected lymphoproliferative disease in the orbit. This is partly due to the perception that incisional biopsy is more reliable, and that there is a risk of obtaining an insufficient amount of tissue with FNAB. Furthermore, it is believed that FNAB provides too little tissue to allow for complete cytomorphological evaluation. However, advancements in both the technique of FNAB and immunocytochemistry have made FNAB a safe and effective method [17].

In the present study, a definitive lymphoma diagnosis could be made in 31 of 38 patients (85%) using FNAB, and in 65% of these cases, subclassification of the lymphoma was possible. Most importantly, all cases could be divided in low and high grade with the assessment of the proliferation rate with KI67 immunostaining. This level of diagnostic accuracy is in line with the improved accuracy reported for FNAB of lesions in other parts of the body. One needs to remember that a limitation to FNAB, as a diagnostic modality, is that the lesions need to be accessible by a 25- or 27-gauge needle. With this in mind, when the FNAB is successful and a sufficient amount of material is retrieved, a diagnosis is made in the majority of cases [6, 17].

Fifteen patients with secondary lymphoma were known to have a lymphoma diagnosis before undergoing FNAB and no knowledge of previous lymphoma in 6 patients. In those patients with previously known lymphoma, one needs to mention the possibility of bias regarding the FNAB diagnosis by the pathologist.

Even with an incisional biopsy, it may prove difficult to subclassify low-grade malignant lymphomas, especially differentiating between the most common types, MZL/MALT lymphoma, and FL. Thus, neither incisional biopsy nor FNAB ensures a conclusive diagnosis in a 100% of cases. In a large study of 353 cases of incisional-biopsy-proven ocular adnexal lymphomas, by Ferry et al. [7], 5% of the patients had low-grade malignant B-cell lymphomas that could not be subclassified. T-cell lymphomas may be difficult to diagnose with both immunostaining and cytomorphology and often further molecular analysis is required. These orbital lymphomas are uncommon so there is limited clinical experience. The mature B-cell neoplasms are more easily subclassified if the cytomorphology is corroborated by an immunocytochemistry and evaluation of cell proliferation with Ki67/MIB-1 [18].

The 2016 revision of the WHO classification of lymphoid neoplasms defines the diagnostic criteria for lymphomas and highlights advances in genetic and molecular techniques that can have clinical implications for treatment choices [11]. In August 2022, the 5th Edition of the classification of hematolymphoid tumors (5th WHO-Hem) was published. In combination with the Sydney classification system, the WHO-Hem provides a more accessible tool when diagnosing lymphoproliferative diseases. The widely used Sydney classification system categorizes lymph node cytopathology. It offers a great level of reassurance as it provides both categorization and management algorithms. Because of the Sydney system and due to the advancements and accuracy of cytopathology, FNAB has become an efficient diagnostic tool. The updated WHO-Hem and the Sydney system have caused a pivotal shift within the cytopathology field connecting the histomorphology with the underlying biology. In an article by Gupta et al. [19], the Sydney system was used to assess the risk of malignancy (ROM) for each diagnostic category. The ROM could give us additional knowledge of the accountability and prognosis of the cytopathology [19, 20].

A combination of information including immunophenotype, clinical data, and cytomorphological features is often needed for the complete classification of lymphoma. The small and mixed cell size was, according to the cytopathologist, associated with most challenges. Despite the fact that subclassification of the disease was not possible in all patients in this study, therapy could be initiated as there are no targeted therapeutic strategies for the different subtypes of primary low-grade B-cell lymphoma in the orbit [14]. Fourteen of 17 patients with primary orbital lymphoma (82%) showed no orbital recurrence during follow-up. Out of these, 8 patients underwent radiotherapy as monotherapy, 2 patients received only chemotherapy, and 1 patient received both. This is a similar success rate of FNAB-proven OAL to that reported by Pinnix et al. [21], who observed an 86% complete remission rate after EBRT in 22 patients with incisional-biopsy-proven OAL. Prognosis is highly dependent on subclassification. The low-grade subtypes when compared to the high-grade subtypes have a better prognosis. After 5 years, the mortality was: 12% for EMZL, 19% for LPL, 22% for FL, 48% for DLBCL [7].

In a previous study carried out at the same clinic, Wiktorin et al. [4] reported six minor complications among 225 orbital FNABs. Wiktorin et al. [4] included some of the patients included in the present study (for the years 2005–2013), including the single patient with a minor complication. No additional complications were observed among the patients included in the present study from 2014 to 2015. FNAB carried out with thin needles and by an experienced cytopathologist or radiologist thus appears to be safe. Due to this, and in view of the potential risk of morbidity associated with incisional biopsies in the orbit, we suggest that FNAB be used for diagnosis in suspected orbital lymphoma. If the systemic workup shows disseminated disease, or if the lymphoma has bilateral orbital involvement and subclassification of a low-grade malignant lymphoma is not possible with FNAB, an incisional biopsy or reFNAB should be carried out to obtain a definitive diagnosis to ensure that the most appropriate treatment is given. If the lymphoma is disseminated, the incisional biopsy can be taken from another, more easily accessible lymph node, if possible, rather than the orbit, for the sake of the patient and to ensure better sampling.

The results of this study show that FNAB, in conjunction with immunological workup and proliferation analysis, has a very high diagnostic accuracy. Subclassification was possible in 65% of the cases studied. All cases could be subdivided between low and high grade with the proliferation analysis. Furthermore, orbital FNAB causes less discomfort to the patient, is cost-effective, and safe when performed by an experienced cytopathologist or radiologist. The above findings indicate that FNAB is a valid first option for the diagnosis of suspected lymphoproliferative disease in the anterior half of the orbit that would otherwise require orbitotomy and incisional biopsy.

This study protocol was reviewed and approved by the Ethics Committee of Stockholm, approval number 2016/15-31/1, and was performed in agreement with institutional guidelines and the principles of the Declaration of Helsinki. Exempt from obtaining written informed consent was granted by the Ethics Committee of Stockholm as only retrospective records were reviewed. Written informed consent regarding depicting the person in Figure 3 is retrieved by the authors.

The authors have no conflicts of interest to declare.

No funding or grants were received for this study.

Nicole Dan reviewed data, performed data analysis, and drafted the manuscript; Cecilia Norin designed the study, contributed to data collection, and revision of the manuscript. Eva Dafgård Kopp participated in the design of the study and revision of the manuscript. Olga Strömberg contributed to data analysis and hematological aspects of lymphoma in the manuscript. Edneia Tani contributed to data analyzes and authored paragraphs regarding cytomorphological and immunologic analyses as well as revision of the whole manuscript. Elin Bohman supervised data collection, analysis, and interpretation, as well as performed main manuscript revision.

Nicole Dan and Cecilia Norin contributed equally to this work as first authors.

All relevant data generated or analyzed during this study are included in this article. Further inquiries can be directed to the corresponding author.