Concurrent thyroid and abdominal diffuse large B-cell lymphoma arising in the milieu of chronic fibrosing thyroiditis

Abstract

Firm tender thyroid masses present a diagnostic challenge, with differentials including benign fibroinflammatory conditions and neoplastic processes. Imaging modalities cannot reliably distinguish between these differentials. The sampling-dependent nature of both cytopathology and core-biopsy histopathology introduces the potential for indeterminate or misleading results. Surgical biopsy can have high complication rates depending on the underlying condition. Here we report the case of a patient with chronic autoimmune thyroid disease who initially presented with fibroinflammatory thyroid disease and was subsequently diagnosed with B-cell lymphoma. This case highlights the diagnostic challenges associated with these conditions, the limitations of the sampling-dependent tissue assessment and the importance of re-evaluating the diagnosis when the clinical course is discordant with the initial biopsy findings.

Introduction

The differential diagnosis of a firm thyroid mass includes both benign fibroinflammatory conditions and malignancy. When painful, subacute thyroiditis and infection should also be considered. Fibroinflammatory thyroid conditions include Riedel thyroiditis (RT) and fibrosing variant of Hashimoto thyroiditis (FVHT) [1, 2]. Malignant causes include thyroid carcinoma, lymphoma, angiosarcoma, and metastases [3-6].

RT is a rare condition with an estimated incidence of approximately 1 in 100 000 and increasingly recognized as a part of the spectrum of IgG4-related sclerosing disease [7]. RT is characterized by fibrosis of the thyroid parenchyma that extends into the surrounding tissues. In contrast, in FVHT, fibrosis is confined to the thyroid. The two entities can overlap and may coexist with IgG4-related systemic disease [7-10].

Primary thyroid lymphoma (PTL) is a rare malignancy of the thyroid gland, accounting for approximately 1% to 5% of all thyroid malignancies and less than 2% of all extranodal lymphomas [11, 12]. There is a well-established association with preexisting Hashimoto thyroiditis, which increases the risk of PTL by approximately 60-fold [13], with underlying Hashimoto thyroiditis reported in approximately 50% of PTL cases [14]. The most common histologic subtypes of thyroid lymphoma are (1) diffuse large B-cell lymphoma (DLBCL) which accounts for approximately 70% of PTL and typically demonstrates an aggressive behavior requiring prompt systemic therapy; and (2) extranodal marginal zone B-cell lymphoma (MALT type) which accounts for 20% to 25% of PTL and is usually limited to thyroid and follows an indolent course [15, 16]. Less common subtypes are follicular lymphoma, Burkitt lymphoma, and T-cell lymphoma.

The clinical presentation of fibroinflammatory thyroiditis and PTL can overlap considerably with both presenting with rapidly enlarging painful thyroid masses, and compressive symptoms. While imaging techniques such as magnetic resonance, computed tomography (CT), and ultrasound can provide useful circumstantial evidence in fibroinflammatory thyroid disease [2], they cannot reliably differentiate between RT, other thyroid pathologies, and neoplasia [17]. Lesion avidity quantified via maximum standardized uptake value (SUVmax) of fluorine-18 deoxyglucose (FDG) positron emission tomography (PET) is generally higher in malignant than in inflammatory or fibrosing processes; however, overlap in SUVmax values is recognized and the usefulness of FDG-PET in discriminating between etiologies is not yet clear [18, 19]. Ultrasound elastography, an emerging diagnostic technique that evaluates tissue stiffness may be useful in characterizing fibrosing conditions [20], but its use is limited and diagnostic accuracy remains undetermined.

Early recognition and accurate histopathological diagnosis are essential to avoid unnecessary surgical intervention and to initiate appropriate therapy. However, establishing a histopathological diagnosis can be challenging, with as few as 13% of PTL cases having a preoperative diagnosis [14]. Both cytology and core-biopsy histopathology are dependent on obtaining representative tissue samples and can miss the diagnosis of lymphoma if fibrosis and inflammation predominate in the sampled tissue. If RT remains in the list of differential diagnoses, surgical interventions including biopsy are high-risk [2]. In RT, due to the loss of tissue planes, even when experienced surgeons engage in limited surgery, the rate of serious complications (eg, laryngeal nerve injury, hypoparathyroidism) is as high as 39% [21]. This can lead to a situation where diagnosis is uncertain, yet surgical biopsy is perceived as relatively contraindicated, leading to diagnostic delays.

We present a case of a DLBCL developing after the initial diagnosis of fibroinflammatory thyroid disease. Despite initial clinical and histopathological findings suggesting benign fibroinflammatory thyroiditis, the patient subsequently developed rapidly progressive thyroid enlargement and systemic disease, leading to a diagnosis of DLBCL. This case highlights the limitations of sampling-dependent histopathological assessment and underscores the need for re-evaluation when the clinical course is discordant with initial biopsy findings.

Case presentation

A 41-year-old female patient presented in July 2024 with a new painful neck swelling, more prominent on the left, with associated neck pain, odynophagia, fatigue, and fever. She had no shortness of breath or symptoms of hyperthyroidism or hypothyroidism. She had a past medical history notable for Hashimoto thyroiditis (HT), diagnosed during pregnancy at age 31, on levothyroxine replacement. Physical examination demonstrated a bilateral thyroid enlargement, more prominent on the left. A firm, mobile, tender 3-cm left thyroid nodule was palpable. There was no tracheal deviation, cervical lymphadenopathy, or positive Pemberton sign.

Diagnostic assessment

Biochemical evaluation demonstrated mildly elevated serum thyrotropin of 4.9 mIU/L (reference range, 0.5-4 mIU/L) with serum free thyroxine of 1 ng/dL (SI 13 pmol/L) (reference range, 0.8-1.6 ng/dL [SI 10-20 pmol/L]) and serum free triiodothyronine of 2.5 pg/mL (SI 3.8 pmol/L) (reference range, 1.8-4.4 pg/mL [SI 2.8-6.8 pmol/L]). Inflammatory markers were raised with C-reactive protein 24 mg/L (reference range, 0-6 mg/L), and erythrocyte sedimentation rate of 63 mm/h (reference range, 1-20 mm/h). Serum electrolytes, calcium, and creatinine were within normal limits.

Thyroid ultrasound demonstrated a grossly heterogeneous thyroid with increased vascularity, with a mildly suspicious left lobe nodule (Fig. 1: 46 × 26 × 35 mm left mid-pole heterogeneous solid nodule with no calcifications, classified as TI-RADS 3, according to the American College of Radiology's Thyroid Imaging, Reporting, and Data System [TI-RADS] classification system).

Figure 1.

Sonogram of the left thyroid in the longitudinal plane July 2024 showing a semi-solid thyroid nodule.

Fine needle aspiration cytology of the nodule showed degenerate single cells with viable nucleoli, set in a necro-inflammatory background and was reported as indeterminate. Immunocytochemistry was negative for TTF1 (thyroid transcription factor 1), SOX10 (Sry-related HMg-Box gene 10), and P40; and positive for leucocyte common antigen (LCA). LCA positivity in the was potentially indicative of a hemato-lymphoproliferative disorder, noting that immunocytochemistry is unreliable on degenerate cells.

Due to the indeterminate cytological features, a core biopsy of the nodule was performed, which revealed fibrous tissue infiltrated by somewhat crushed small lymphocytes, with no discernable thyroid tissue seen (Fig. 2). Immunohistochemistry showed that the lymphocytes were almost purely T cells, with very low Ki67. No plasma cells were seen. Significant nonspecific binding was present on flow cytometry, precluding any meaningful analysis; nonetheless, B cells appeared polyclonal. The core biopsy was reported as “chronic (fibrosed) lymphocytic thyroiditis” and clinical correlation to exclude RT was recommended. No evidence of a lymphoproliferative disorder was noted. Biochemical evaluation demonstrated a normal IgG4 0.32 g/L (reference range, 0.05-1.30 g/L). FDG-PET CT was considered, but cost was prohibitive and there was no basis for public rebate.

Figure 2.

Histopathologic findings of core biopsy, demonstrating fibrous tissue with small crushed lymphocytes (predominantly T cells on immunohistochemistry staining). A, H&E stain × 10; B, detail of demarcated area; scale bar 100 µm.

Treatment

At this stage, RT was considered the leading differential diagnosis, based on clinical features and elevated inflammatory markers. In September 2024 the patient was commenced on prednisolone 30 mg daily, with a plan to taper this dose by 5 mg weekly. There was good initial response, but symptoms returned on ceasing prednisolone. A second course of 30 mg prednisolone was commenced, with a slower taper of 5 mg fortnightly. Neck pain and swelling recurred on weaning the prednisolone dose to 15 mg.

After discussion in the endocrine/surgical/radiological multidisciplinary team, she was treated with rituximab, 2 × 1-g infusions in March 2025. Subsequently, prednisolone was weaned to 10 mg daily, but symptoms persisted. After further multidisciplinary team discussion, she was referred for consideration of surgical intervention for further progressive symptoms. She was referred to a quaternary hospital head and neck surgical service.

In June 2025, while awaiting review for this opinion, the patient experienced rapidly progressive enlargement of the left neck swelling, with progression of neck pain, odynophagia, and worsening compressive symptoms, particularly dyspnea. She presented to her local emergency department for assessment, and CT demonstrated a large cystic lesion within the left lobe of the thyroid, with marked mass effect, without retrosternal extension (x-ray demonstration of mass effect shown in Fig. 3). Given the rapid growth, internal hemorrhage was considered, and she was admitted under the local ear nose and throat surgical service. She received treatment with dexamethasone and tranexamic acid and was discharged home.

Figure 3.

Serial chest x-rays demonstrating development of mass effect on trachea in neck indicated by arrow: A, October 2024; B, July 2025.

The following week, she developed new abdominal pain and presented for further emergency assessment. A CT chest/abdomen/pelvis demonstrated the known neck lesion, which had grown to 61 × 75 mm, and additionally thick-walled fluid density lesions within the abdomen, one in the right upper quadrant deep to the abdominal wall, and the second protruding exophytically from the upper pole of the left kidney (Fig. 4). FDG-PET scan demonstrated FDG avidity with central photopenic necrotic areas in the left thyroid mass, right upper chest/abdominal wall mass, and left renal mass.

Figure 4.

Cross-sectional imaging findings. A, Sagittal view of fluorine-18 deoxyglucose positron emission tomography–computed tomography (FDG-PET-CT) demonstrating increased avidity within the thyroid gland (standardized uptake value, SUVmax, 26.1) and within the left exophytic renal lesion and right abdominal/chest wall lesion; B, CT June 2025 demonstrating tracheal compression from neck mass; C, axial section of FDG-PET-CT, demonstrating peripheral avidity and central necrotic core of thyroid mass.

Biopsy of the right upper quadrant lesion demonstrated histological features favoring a high-grade B-cell lymphoma, but subtype classification via immunohistochemistry/fluorescence in situ hybridization (FISH) was not possible due to the poorly preserved material in the sample. Repeat core biopsy (x6) of the neck mass demonstrated findings consistent with DLBCL, germinal center B-cell (GCB) subtype (Fig. 5).

Figure 5.

Histopathologic findings of core biopsy of neck mass demonstrating diffuse malignant lymphoid infiltrate with predominant immunoblastic appearance, with extensive apoptotic activity, without normal lymph tissue. A, H&E stain × 200; B, CD10 stain × 200; C, CD20 stain × 200; scale bar 100 µm.

Bone marrow aspiration and trephine did not detect a monoclonal B-cell or aberrant T-cell population. Peripheral flow cytometry demonstrated undetectable B cells, consistent with prior rituximab treatment.

Outcome and follow-up

The patient proceeded to treatment with chemotherapy with dose-adjusted rituximab, etoposide, prednisolone, vincristine, cyclophosphamide, and doxorubicin (DA-R-EPOCH), with subsequent rapid improvement in size of the neck mass and improvement in compressive symptoms. Continuation of chemotherapy is planned.

Discussion

The definitive diagnosis of painful firm thyroid masses is challenging. There are no specific clinical features that distinguish benign fibroinflammatory thyroid disease from neoplastic processes, as they all can present with pressure symptoms, a hard thyroid, and adherence to other structures. RT is classically described as “painless” but pain is common, present in ∼40% of cases [1], and neoplastic processes can be painful if rapidly enlarging. Distinction between conditions is based on histopathology.

The initial presentation in this case was that of a benign fibroinflammatory process. Core biopsy demonstrated fibrosis with predominantly T-cell lymphocytic infiltrate, low lymphocyte Ki-67 levels, and no evidence of lymphoproliferative process. The classical histological features of RT, including storiform fibrosis [2, 22], mixed B & T cell infiltrate, and eosinophilic infiltration, were absent. Nevertheless, given the clinical presentation, elevated inflammatory markers, and absence of an alternative diagnosis, RT remained the working diagnosis and treatment was directed accordingly. The subsequent development of rapidly enlarging painful masses with biopsies consistent with DLBCL highlights the limitations of sampling-dependent tissue analysis. It is possible that an underlying lymphoproliferative process was missed on initial sampling because the sample was not representative and fibrosis and inflammation was predominant. Alternatively, lymphoma may have developed subsequently within a background of chronic fibroinflammatory thyroid disease as has been reported elsewhere in the literature [23]. Regardless of the cause of the missed diagnosis in the first instance, this case demonstrates the importance of reconsidering the diagnosis when disease behavior is discordant with the initial histopathological findings.

The evidence base for treatment of fibroinflammatory thyroid disease is limited. The mainstay in RT is with glucocorticoids, with rituximab and selective estrogen receptor modulators as additional options [1]. There is significant overlap with treatment for B-cell lymphomas, where glucocorticoids and rituximab are components of the usual chemotherapy regimen [24-26]. Consequently, response to treatment cannot reliably distinguish between lymphoma and fibroinflammatory thyroid conditions.

Learning points

• Fibroinflammatory conditions of the thyroid remain a challenging diagnosis due to the overlap of clinical symptoms, inability for imaging to definitively distinguish between underlying processes, and sampling issues inherent to a cytopathologic/histopathologic diagnosis, whereby fine needle aspiration is often nondiagnostic, and even core biopsies may not reveal the diagnosis.

• The clinical response to steroids does not necessarily assist in distinguishing between benign and neoplastic processes as multiple underlying etiologies may respond to glucocorticoid therapy and indeed to rituximab.

• Inflammatory processes involving the thyroid may predispose to the subsequent development of lymphoproliferative malignancy, and a change in clinical symptoms should instigate prompt investigation.

Abbreviations

CT

computed tomography

DLBCL

diffuse large B-cell lymphoma

FDG

fluorine-18 deoxyglucose

FVHT

fibrosing variant of Hashimoto thyroiditis

HT

Hashimoto thyroiditis

PET

positron emission tomography

PTL

primary thyroid lymphoma

RT

Riedel thyroiditis

SUVmax

maximum standardized uptake value

Contributors

All authors made individual contributions to authorship. K.J. was involved in the diagnosis and management of the patient. A.D.A, G.P., A.H., and K.J. were involved in conception of the work. A.D.A was involved in manuscript design, and composition of initial draft. A.D.A, G.P., A.H., and K.J. were involved in critical review of the work for intellectual content. All authors approved of the final draft and agreed to be accountable for all aspects of the work.

Funding

No public or commercial funding.

Disclosures

None declared.

Informed patient consent for publication

Signed informed consent obtained directly from the patient.

Data availability

Data sharing is not applicable to this article as no datasets were generated or analyzed during the current study.