Introduction
Burkitt lymphoma (BL) is a highly aggressive B cell non-Hodgkin lymphoma (NHL) characterized by the translocation and deregulation of the MYC gene on chromosome 8 with the potential to involve multiple organ systems. Three subtypes of BL (sporadic, endemic, and immunodeficiency-associated) are recognized with different epidemiology, risk factors, and clinical presentations.
The sporadic subtype of BL is generally observed in the USA and Western Europe with an overall incidence of three cases per million persons per year in the general population. Sporadic BL is relatively more common in the pediatric population, accounting for 30% of pediatric lymphomas with a peak incidence around the age of 10 years old, while only representing less than 1% of NHL in adults [
1,
2]. Sporadic cases are associated with Epstein-Barr virus (EBV), and the most common site of involvement is within the abdomen, particularly the bowel [
3]. The sporadic subtype will be the focus of this article.
The endemic subtype is found in equatorial Africa and New Guinea with a near 50-fold higher incidence than that seen in the USA [
4]. Endemic BL accounts for up to 50% of all childhood cancer in equatorial Africa with an estimated incidence of 3 to 6 cases per 100,000 children per year [
5]. The most common presentation is a facial tumor, and nearly all cases are associated with EBV [
6]. Immunodeficiency-associated BL is mostly commonly seen in HIV-positive patients, but also may be seen in allograft recipients and patients with congenital immunodeficiency. HIV patients with BL tend to have relatively higher CD4 counts (> 200 cells/μL) and the majority demonstrate EBV positivity. The most common sites of involvement are lymph nodes, bone marrow, and the central nervous system (CNS) [
3].
BL exhibits very heterogeneous presentations and clinical courses among patients. In this article, we review the pathogenesis and classification of BL, illustrate a multimodality imaging approach in evaluating common and uncommon sites of involvement, and highlight important treatment topics.
Imaging evaluation
Although the diagnosis of BL is confirmed pathologically by characteristic biology, immunophenotype, and genetic analysis, imaging plays a critical role throughout the entire clinical course of these patients from initial detection, including emergent presentations, to evaluating treatment response and potential complications. Various modalities are employed in the assessment of BL, each with particular strengths and weaknesses.
For patients presenting in an emergent setting with acute symptoms, either ultrasound or computed tomography (CT) is often utilized. Ultrasound is particularly useful in the pediatric population given the lack of ionizing radiation. Ultrasound can evaluate for the presence of palpable masses or intussusception in pediatric patients and is well suited to evaluate small and superficial sites of involvement such as peripheral lymph nodes, gonads, thyroid, and breast [
11]. Ultrasound is appropriate for the evaluation of lymph node architecture and assessing benign versus malignant nodal features. Ultrasound findings or malignant nodes include loss of an echogenic hilum, asymmetric cortical thickening, distortion of intranodal vascular architecture (i.e., aberrant vessels), and amputated short subcapsular vessels. However, ultrasound may be limited by operator dependence and in its evaluation of deeper abdominal structures, particularly in larger adults. Conversely, CT offers rapid, whole body evaluation (including staging information) with improved soft tissue resolution while overcoming the acoustic limitations of ultrasound [
12]. Exams are preferably performed with intravenous contrast for better soft tissue evaluation, although poor renal function and allergy may prohibit its use. Multiplanar reformatted images should routinely be used. In the setting of osseous involvement, CT is adequate to evaluate for bone destruction. A principle disadvantage of CT is ionizing radiation exposure. It should be emphasized that modalities utilizing ionizing radiation should be performed in accordance with the principle of As Low As Reasonably Achievable (ALARA). Furthermore, interpretation of CT in some pediatric patients may be difficult in the setting of absent intra-abdominal fat planes.
Although not typically employed in the emergency setting, magnetic resonance imaging (MRI) has an increasing role in the evaluation of BL. The strength of MRI is its superior soft tissue characterization, which is of particular value in assessing tumor extension and central nervous system (CNS) involvement. Standard protocols utilize a combination of multiplanar T1- and T2-weighted images, with and without fat saturation, although protocols should be tailored to the anatomy being evaluated. For example, if there is clinical concern for biliary obstruction, magnetic resonance cholangiopancreatography (MRCP) can be obtained. Although data specific to Burkitt lymphoma is lacking, the addition of diffusion-weighted imaging (DWI) has demonstrated added value over conventional sequences by increasing lesion conspicuity and improving accuracy of initial diagnosis in other lymphoma subtypes [
13]. Furthermore, DWI in lymphoma has shown to correlate with higher cellularity and proliferative index and has shown utility in assessing treatment response [
14,
15]. The lack of ionizing radiation is an advantage in assessing pediatric patients, particularly those requiring serial follow-up imaging. However, limitations of MRI include long exam times and relatively limited availability, particularly in emergent settings.
Positron emission tomography (PET), when combined with CT, offers the advantage of combining functional information with anatomic imaging. There is a growing body of literature evaluating the utility of PET/CT in both pediatric and adult BL patients (Table
1) [
16‐
22]. For example, Carrillo-Cruz et al. demonstrated a high discrepancy rate between CT and PET/CT with PET/CT achieving a 100% negative predictive value in predicting treatment response as well 100% positive predictive values in predicting recurrence using a standardized uptake value (SUV) change threshold less than 66% [
19]. Similarly, Wei et al. showed significant reduction in SUV
max on interim and post-therapy PET/CT and that changes in SUV
max greater than 50% was a favorable cutoff point to predict the overall survival of BL patients [
22]. PET/CT may also offer superior staging compared to anatomic imaging such as CT. For example, in the pediatric population, PET/CT has demonstrated better ability to detect both nodal and extranodal sites of disease and greater impact on initial staging compared to CT [
23]. PET/MRI has not yet been investigated in BL patients, although has shown similar performance as PET/CT in staging and follow-up in other lymphoma cohorts [
24]. PET/MRI can simultaneously combine soft tissue characterization strengths of MRI with the functional assessment of PET. Furthermore, substitution of the CT component with MRI allows for radiation dose savings. However, both PET/CT and PET/MRI are subject to relatively long exams and limited availability, particularly PET/MRI. Also, PET exams are subject to added radiation exposure.
Table 1
Utility of PET in Burkitt lymphoma
Albano et al. (2018) [ 16] | Adults (n = 65) | -End of treatment PET/CT results significantly correlate with PFS and OS -Interim PET/CT did not correlate with PFS and OS |
Albano et al. (2019) [ 17] | Adults (n = 65) | -Total metabolic tumor volume and total lesion glycolysis independent prognostic factors for PFS and OS |
Bailly et al. (2014) [ 18] | Children (n = 19) | -Significantly higher NPV of PET (93%) compared to conventional imaging (73%) in detecting CR -PFS significantly higher in patients with negative PET than those with positive PET |
Carrillo-Cruz et al. (2015) [ 19] | Adults and children (n = 32) | −100% NPV in predicting CR −100% PPV of nonresponse with SUV change < 66% after treatment |
Davidson et al. (2018) [ 20] | Adults (n = 20) | -Increased splenic FDG uptake rarely involved at time of staging -Low rate of spleen involvement may serve as a specific characteristic of BL |
Karantanis et al. (2010) [ 21] | Adults and children (n = 15) | -High sensitivity (100%) and specificity (94–96%) for detection of nodal and extranodal disease |
| Adults (n = 29) | -Significant reduction in SUVmax during interim and post-therapy PET/CTs -SUV decease > 50% after post-therapy PET/CT was a favorable cutoff point to predict OS |
According to the National Comprehensive Cancer Network (NCCN) guidelines, initial staging in adults should be performed with diagnostic CT of the chest, abdomen, and pelvis [
25]. Initial imaging studies in children with BL must include chest radiograph, cervical and abdominal ultrasound, CT of the chest, abdomen, and pelvis. Depending on the clinical presentation, contrast-enhanced neck and/or brain MRI may also be warranted. PET/CT should be performed if it is possible, provided it will not delay treatment, which should be started promptly, given the rapid tumor growth [
26]. Also, initial cardiac function evaluation should be performed with either echocardiogram or multi-gated (MUGA) cardiac blood pool scan in anticipation of treatment with an anthracycline or anthracenedione-based regimen. Initial staging PET/CT may be useful, although is not currently routinely recommended, although it should not delay prompt treatment start. Depending on the clinical presentation, contrast-enhanced neck and/or brain MRI may also be warranted. The role of PET/CT imaging prior to completion of therapy has not been well-established. Contrary to evidence demonstrating prognostic value of interim PET/CT in DLBCL, studies to date evaluating this practice in BL have not shown similar results [
16,
22,
27]. If there is a concern for relapsed or refractory disease, PET/CT is recommended as the modality of choice.