When a patient with malignant lymphoma develops new lymph node (LN) enlargement, a recurrence of lymphoma is usually first suspected. While LN enlargement is common in children with and without infectious conditions, manifestations of non-Hodgkin’s lymphoma also include painless and rapid LN enlargement . Healthy children and adolescents infected with cytomegalovirus (CMV) are mostly asymptomatic, except for approximately 10% of patients who acquire CMV infections that present as CMV mononucleosis . Herein, we present a case of CMV infection in which computed tomography (CT) and 18F-fluoro-2-deoxy-D-glucose positron emission tomography (FDG-PET) imaging findings suggested tumor recurrence.
A 3-year-old boy was admitted for his one-year post-treatment evaluation after completing multi-agent combination chemotherapy for Burkitt lymphoma (BL). On admission, there were neither specific complaints nor any report of recent gastrointestinal illness from the patient or his parents.
A diagnosis of BL was made 18 months before the patient’s visit. Initial clinical presentation at the time of diagnosis included a 4-5-cm-sized palpable mass in the right upper quadrant of the abdomen without any other LN enlargement. Further imaging work-up, including CT, FDG-PET, and bone scan, along with bone marrow evaluation and cytospin of cerebral spinal fluid showed no metastasis. LN biopsies of the bulky abdominal mass were performed, and the final lymphoma staging was determined to be group B (stage III) according to the FAB/LMB risk stratification . Multi-agent chemotherapy using methotrexate and rituximab was initiated on diagnosis and for 6 months. Rituximab was used off-label based on the modified FAB/LMB 96 regimen . The patient achieved a complete response.
At the time of the present visit, the patient was alert and apparently healthy with soft abdomen, non-tender, and no distention signs. On palpation, there were no signs of an abdominal mass, hepatomegaly, or spleno-megaly. No LN enlargement was observed in the cervical, axillary, or inguinal area. The following examinations were scheduled and performed: complete blood cell (CBC) count with differential; routine admission battery (serum); CT of the thorax, abdomen, and pelvis; bone scan.
The results of the CBC count with differential were as follows: leukocyte count, 7,190 cells/mL with lymphocyte dominancy (58.2%); hemoglobin concentration, 11.9 g/dL; platelet count, 191,000 platelets/mL; absolute neutrophil count, 2,340 cells/mL. A peripheral blood smear revealed microcytic hypochromic red blood cells with anisopoikilocytosis, increased white blood cell number with neutrophilia, and toxic granules without atypical lympho-cytes. Moreover, normal platelet number suggesting the possibility of iron deficiency and reactive neutrophilia. All the routine admission battery results, including the level of lactate dehydrogenase (344 IU/L) were within normal limits.
Abdominal CT showed several enlarged LNs (9-13 mm in size) in the right lower quadrant and small bowel mesentery, whereas thoracic CT showed a few enlarged LNs (8-12 mm) in both the axilla and right paratracheal nodal stations (Fig. 1). Further work-up included FDG-PET, which showed multiple newly developed hypermetabolic, enlarged LNs, in the mesentery (maximum standardized uptake value=3.3, Deauvill Criteria 4) (Fig. 2). Therefore, tumor recurrence was strongly suggested. An ileocecal LN biopsy was performed along with prophylactic appendectomy to confirm the pathology of the lesion. The LN biopsy results revealed reactive hyperplasia with no malignancy evidence. Both the CMV immunohistochem-istry and in situ hybridization test results for Epstein-Barr virus (EBV) were negative, despite CD3 and CD20 antigen reactivity for T and B cells, respectively.
Multiple serologic tests, including those targeting antibodies for CMV, EBV, herpes simplex virus, varicella zoster virus, and hepatitis virus B/C were performed. All other serologic tests were negative except for CMV-Immuno-globulin G (IgG; 19.6 U/mL, negative<12-14 U/mL<positive) and CMV-Immunoglobulin M (IgM; 22.5 U/mL, negative<18-22 U/mL<positive).
Next, qualitative and quantitative polymerase chain reaction (PCR) tests were performed along with CMV cell cultures. Qualitative studies and cultures were performed using blood, urine, and saliva specimens. Quantitative real-time PCR was performed using an ethylenedia-minetetraacetic acid-treated blood sample. All specimens yielded negative culture and qualitative PCR results; however, positive results were obtained for qualitative PCR performed using urine. Real-time PCR showed <1,000 copies/mL of CMV loads, which is within the reference level. Furthermore, a CMV antigenemia assay showed a negative result.
The patient was discharged after asymptomatic CMV infection with no evidence of tumor recurrence. General physical examinations with serology and PCR follow-up were performed in the outpatient department. No abnormal findings were detected on physical examination during the monthly follow-up visits that occurred over the year that followed. Follow-up CT imaging at 4 months showed a decrease in the size and number of LNs, except for a single LN at level 1 of the right axillae, which had decreased in size when evaluated using ultrasonography in the following month (Fig. 3). Further CT imaging at month 8 and 12 remained unchanged with no LN en-largement. However, PCR performed using urine and CMV IgG were persistently positive (Fig. 4). CMV IgM levels peaked (34.4 U/mL) at the 2-month follow-up and decreased thereafter to a negative value at the 10-month follow-up (14.1 U/mL).
Written informed consent was obtained from the patient’s parents for publication of any accompanying images in the case report. This study was approved by the Institutional Review Board of our University Hospital (IRB no. 2019-11-025).
We presented a case of CMV infection in which CT and FDG-PET imaging findings mimicked those of tumor recurrence. Despite performing pathologic work-up, differentiating between acute EBV infection and lymphoma may prove challenging . However, considering that EBV infection may cause certain lymphomas, the two conditions may appear similar . Previous cases of selected infections mimicking the recurrence of certain malignancies have been reported:
We learned two major lessons from our case: (1) the importance of pathologic confirmation of a lesion suggestive of neoplastic conditions, and (2) an individualized approach when determining the etiological agent of suspected recurrent malignancy. After spotting a lesion on imaging, including CT or FDG-PET, the only way to accurately confirm the diagnosis is by performing a biopsy. In the case of a superficial and, hence, easily approachable lesion, image-guided core needle biopsy may be considered . However, for deep lesions, such as in our case, surgery may be required. This is supported by cases of infections mimicking the recurrence of a malignancy that has been previously described in the literature, including a biopsy of lesions suggestive of neoplastic conditions [5-8].
Furthermore, highly prevalent local pathogens, in addition to more general pathogens, should be considered as possible agents mimicking malignancy. In that sense, we believe our case provides a good example. Based on the geographic spread of CMV, it is more common to come across such cases in a country where the seroprevalence of CMV is relatively high. Although Korea is one of the most industrialized member countries of the Or-ganization for Economic Co-operation and Develop-ment, the CMV seroprevalence, which could be an inverse surrogate marker of hygiene and socioeconomic status, was estimated at 94% for all ages in Korea based on a recent study . We believe that our approach may not be limited to specific age groups or geographies.
Despite evidence supporting CMV infection in our patient, we were unable to recognize CMV by antigen immunohistochemistry of the resected LN; this constituted a significant limitation of the study. As studies on LN biopsies in CMV mononucleosis patients are scarce, we believe that this finding is comparable with a study that conducted 6 liver biopsies from previously healthy adult patients with CMV mononucleosis .
In addition, the absence of CMV serology before the identification of CMV infection, and the uncertainty on the primary CMV infection remains the main limitations of the study. The positive results of urine CMV PCR support the presence of the virus. The patient had positive CMV IgM titer for at least 8 months. CMV IgM antibodies are known to be detected for a period of 4 to 6 months after the onset of symptoms . Nevertheless, considering the usual duration of symptomatic patients, over 8 months of positive CMV IgM is still substantial. We speculate that the influences of chemotherapy, including rituximab use, may exist, but further studies exploring this possibility are needed. Increased number of infections has been documented in patients treated with rituximab for lymphoma, and for viral pathogen, hepatitis B reactivation has been well documented . Nevertheless, severe infection by CMV infection has been reported after high-dose chemotherapy with autologous blood stem cell rescue and peritransplantation rituximab . There-fore, even though the case was not a symptomatic severe viral infection, we believe that our case suggests the necessity of considering opportunistic and viral infection for cases of LN enlargement in rituximab-treated pa-tients.
Our findings emphasize the importance of performing an immediate biopsy and using multiple approaches when attempting to explain lesions suggestive of neoplastic conditions.
The authors have no conflict of interest to declare.