We retrospectively reviewed the records of four patients with IVLBCL from December 2011 to July 2019 at Xiangya Hospital of Central South University. The clinical history, laboratory data, imaging abnormalities, treatment regimens, and outcomes were collected. The diagnosis of each case was confirmed with morphologic examination, immunohistochemical stains, or molecular studies. We also conducted a comprehensive literature review for the cases of IVLBCL published in the last 20 years (between 1999 and 2021) in PubMed (http://www.ncbi.nlm.nih.gov/pubmed/) using combinations of keywords in the title/abstract field, including “intravascular large b-cell lymphoma and IVLBCL”. The cases in the literature were carefully reviewed to extract necessary clinicopathologic data and combined with the cases repeatedly studied in different papers. A total of 311 cases of IVLBCL were retrieved from the literature and were included in our research (3–91).

Computed tomography (CT)–guided percutaneous liver, lung, and adrenal glands biopsies were performed. All biopsied specimens of four cases were formalin-fixed and paraffin-embedded (FFPE) and subsequently sectioned at 4.0 μm for hematoxylin-eosin staining and immunohistochemical study. Appropriate negative and positive controls were performed with satisfactory staining.

Digoxigenin-labeled probe (Dako) was used to detect Epstein–Barr virus–encoded small RNA. The in situ hybridization (ISH) assays were performed on a Ventana Benchmark XT platform with an iViewBlue detection kit according to the manufacture’s manual (Ventana Medical Systems Inc., Tucson, AZ, USA).

Fluorescence ISH (FISH) studies were performed on two cases on FFPE tissue using C-MYC, BCL2, and BCL6 dual-colored break-apart rearrangement probes (Vysis Inc., IL, USA). Hybridization signals were assessed from 100 nuclei of tumor cells with an established cutoff of 15% for rearrangement of the gene locus. When the proportion of tumor cells with separation signals was less than 15%, the gene was interpreted as not rearranged. The FISH images were obtained using an Olympus BX51 fluorescence microscope and were captured on the Olympus Image DP2-BSW Software (Olympus Soft Imaging GmbH, Hamburg, DE).

The major clinical features of the four cases of IVLBCL are presented in Table 1. Patients were middle-aged and elderly with an average age of 56 years at diagnosis ranging from 41 to 72 years and a male-to-female ratio of 1:1. The sites of initial diagnosis of IVLBCL were in the adrenal glands (two cases) and the liver and lung (one each case). All patients presented with B symptoms, including fever, night sweats, weight loss, and fatigue. Laboratory examination results revealed anemia (three of four), hypoalbuminemia (four of four), thrombocytopenia (one of four), and remarkably elevated levels of serum LDH, erythrocyte sedimentation rate (ESR), and C-reactive protein (CRP). Abdominal CT with contrast enhancement showed enlargement of bilateral adrenal glands (three of four) and hepatosplenomegaly (two of four). CT scanning of the chest revealed pleural effusion (three of four), and one case showed ground glass opacities in bilateral lung fields (84). Two patients presented enlarged retroperitoneal lymph nodes at CT scan, which were not biopsied. Fluorine-18 fluorodeoxyglucose (18-FDG) positron emission tomography–CT (PET-CT) study showed high uptake in the liver and bilateral adrenal glands (Figure 1). Bone marrow biopsies were performed in two patients, showing an active proliferation of granulocyte with obvious toxic reaction, but no evidence of lymphoma infiltration was detected. However, a bone marrow biopsy in one of the patients showed hemophagocytosis. In addition, the other two patients without bone marrow biopsy showed no hemophagocytic syndrome in clinical symptoms and serological results. All patients were diagnosed as IVLBCL by percutaneous organ biopsy. Only one patient received systemic chemotherapy consisting of rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone (R-CHOP) and had a complete remission at least initially; two patients received supportive care; and one patient refused any treatment. A total of two patients died of disease during the short period of follow-up.

All cases showed selective growth of large, atypical lymphocytes within the lumen of small blood vessels. Various degrees of cellulose thrombosis were observed in two cases. The neoplastic lymphocytes were large in size with one or more nucleoli, scant cytoplasm, and frequent mitotic figures (Figure 2).

The results of immunohistochemistry and EBER-ISH are summarized in Table 2. Tumor cells were positive for CD20 (four of four), PAX5 (four of four), MUM1 (four of four), BCL6 (three of four), BCL2 (one of four), and CD5 (one of four), but negative for CD3 and CD10. The Ki-67 proliferative index ranged from 70% to 80%. ISH for EBER (Epstein–Barr virus-encoded small RNA) was positive in one patient (Figure 2).

The C-MYC, BCL2, and BCL6 gene translocation were assessed using the dual-colored break-apart rearrangement probes in two cases (Table 2). In our study, rearrangement of BCL6 gene on the IVLBCL cells was detected in case 4, which had been reported previously (83). However, FISH study did not show any rearrangement of C-MYC, BCL2, and BCL6 genes in case 1 (Figure 2).

After an extensive search of the literature, we found 311 cases of IVLBCL. The detailed clinicopathologic data of the 311 cases from the literature are included in Supplementary Tables (Supplemental Digital Content 1, http://links.lww.com/PAS/A438).

IVLBCL mostly affected elderly patients with a male-to-female ratio of 1:1. The average age and median age were 65.2 and 67 years, respectively, ranging from 23 to 92 years. Diagnosis was established in vivo in 258 of 311 (83.0%) patients and postmortem in 51 patients (16.4%). Patients with IVLBCL were diagnosed via RSB (81 of 311, 26.0%), bone marrow biopsy (46 of 311, 14.8%), lung biopsy, and adrenal biopsy. IVLBCL lesions were found in almost all organs including the thyroid, liver, pancreas, prostate, and brain. Bone marrow lymphoma infiltration was detected in 92 patients (92 of 204, 45.1%). Characteristic features included persistent fever (190o of 199, 95.6%), hypoxemia (91 of 92, 99.0%), impaired consciousness (48 of 48, 100%), hypoalbuminemia (41 of 45, 91.1%). The clinical manifestations are mainly divided into two subtypes: one is the classic type and the other is the Asian variant. In Western countries, patients with IVLBCL usually show symptoms of CNS and skin involvement in 311 literature cases. In contrast, patients with IVLBCL in Asian countries usually present hemophagocytic syndrome, mainly including pancytopenia and hepatosplenomegaly. Among the laboratory abnormalities, anemia, and thrombocytopenia were the most frequently observed findings. Most patients exhibited extremely low levels of serum albumin, high levels of serum LDH, ESR, CRP, sIL2R, and ferritin. The increased level of serum LDH was present in almost 97% of cases, whereas anemia was present in two-thirds of cases.

Baseline imaging studies included whole body CT scan, brain MRI, and PET-CT in 180, 73, and 75 patients, respectively (Table S2). Prominent lymphadenopathy was not detected in the majority patients, but small groin or intra-peritoneal lymphadenopathy (<2 cm in diameter) was often observed. Splenomegaly and adrenal tumors were detected in 66 (66 of 180, 36.7%) and 11 (11 of 180, 6.1%) patients, respectively. Chest CT detected pleural effusion in 17 (17 of 180, 9.4%) and ground glass opacity of lung field in 35(35 of 180, 19.4%) patients. Brain MRI showed abnormal findings in 55 patients (55 of 73, 75.3%), with hyperintense lesions in the pons in three (3of 73, 4.1%), non-specific white matter lesions in eight (eight of 73, 11.0%), infarct-like lesions in four (four of 73, 5.5%), and meningeal enhancement in five (five of 73, 6.8%) patients. PET-CT revealed a high incidence of bone marrow (three of 75, 4%), bone (16 of 75, 21.3%), spleen (21 of 75, 28%), lymph node (two o 75, 2.7%), and adrenal gland (11 of 75, 14.7%) involvement.

The results of immunohistochemical analysis of markers for most patients remained consistent. In these statistics, most cases expressed CD20 (273 of 276, 99.0%), CD79a (112 of 114, 98.2%), and other whole B cell markers and expressed MUMI (85 of 100, 85%), BCL2 (62 of 76, 81.6%), CD5 (64 of 142, 45.1%), CD10 (16 of 135, 11.9%), and BCL6 (40 of 90, 44.4%). T cell marker CD3 (2 of 124, 1.6%) is negative. The expression of CD20 is negative, and the expression of CD3 is positive in a very small number of patients, which brings difficulties to the correct diagnosis. The proliferation activity of Ki-67 ranges from 20% to 100%. It has been reported in the literature that the proliferative activity of Ki-67 is an independent risk factor for the prognosis (survival time) of patients with IVL (88). The immunophenotype of most patients (68 of 83, 81.9%) is of non-germinal center origin (non-GCB). Studies have shown that the simultaneous expression of C-MYC and BCL2 may be a useful prognostic indicator. Compared with non-dual expressions, C-MYC/BCL2 dual expressions have a significantly higher mortality rate (88). Positive EBER expression accounts for only four (four of 52, 7.7%) patients.

Molecularly, chromosomal abnormalities were detected in bone marrow samples from the 34 patients in whom karyotype analysis was successful. A common pattern of abnormality could not be defined due to the heterogeneous and complex karyotype pattern of the disease and the small sample size (90). Noticeable, abnormality of chromosome 18 (four of 24, 16.7%) accounts for the largest proportion of chromosome abnormalities in patients with successful karyotype analysis in this study.

Therapeutic management of 220 patients is summarized in Table S2. A total of 220 patients were treated in hospital. Among them, 146 patients received targeted therapy based on the CHOP regimen, and most of them have received different doses of drug adjuvant therapy. Ten patients received at least one course of R-CHOP (median, six cycles), and then, methotrexate (MTX) was injected intrathecally during treatment for each course. In a few cases, hematopoietic SCT, surgery, and radiotherapy were selected as the treatment plan, and the treatment effect was also significantly improved. Because most of the cases did not mention the follow-up time and prognosis of patients after treatment, the overall survival rate of patients could not be calculated in this paper.

This review is the largest to date on clinical presentation forms, pathological features, therapeutic management, and outcome in patients with IVLBCL from the literature.