A 66-year-old Caucasian man presented with complaints of vertigo, diplopia, and ataxia in August 2016. Physical exam was notable for horizontal and vertical nystagmus, mild left dysmetria, and wide-based staggering gait. There was no palpable lymphadenopathy or organomegaly. The complete blood count, renal function, and liver function tests were within normal limits. He subsequently underwent a contrast enhanced computed tomography (CT) and magnetic resonance imaging (MRI) of the brain that was remarkable for a 1.7×1.3 cm cerebellar mass abutting the roof of the 4th ventricle without any evidence of obstructive hydrocephalus. A slit-lamp eye exam showed no evidence of involvement in the eyes. A scrotal ultrasound examination did not show any evidence of any mass lesions in either testis. He had a contrast enhanced whole body CT scan that showed no evidence of primary malignancy in the chest, abdomen, or pelvis. The patient subsequently had a suboccipital craniotomy and resection of the cerebellar mass.

Hematoxylin and eosin staining (H & E) of the excisional brain tumor sample showed sheets of large lymphoid cells with open chromatin and irregular nuclei along with prominent nucleoli and a small amount of cytoplasm. The immunohistochemical (IHC) staining on the neoplastic cells demonstrated atypical lymphoid cells that strongly and diffusely expressed CD45, CD20 CD 79a, MUM1, and BCL2, were weakly positive for BCL6, and had a Ki-67 index of >95%, suggesting a rapidly proliferating lymphoma with an activated B-cell immunophenotype. Small foci of necrosis and frequent mitosis were present (Fig. 1A, B). The tumor cells did not express CD3, CD10, CD30, CD43, CD 138, EMA, and ALK-1. MYC stain showed homogeneous nuclear expression in >90% of cells. (Fig. 1C-I). Epstein-Barr virus (EBV) in situ hybridization was negative. Flow cytometry of the brain tumor cells revealed a medium- to large-sized lymphoid population of kappa light chain restricted B-cells expressing CD19 and CD20, and negative for CD5 and CD10.

Interphase fluorescence in situ hybridization (FISH) studies were performed on paraffin-embedded tissue sections using dual color break apart probes for MYC and BCL6, and dual color, dual fusion IGH/BCL2 probes. A total of 50 interphase nuclei were scored for the presence of these rearrangements using an automated image analysis platform. Signal patterns consistent with rearrangements of the 3q27 (BCL6) and 8q24 (MYC) loci were identified in 64% and 84% of examined nuclei, respectively. Although there was no evidence of the t(14;18)(q32;q21) translocation, a signal pattern was observed indicating a deletion of the 18q21 (BCL2) locus in 28% of nuclei examined (Fig. 2). This finding was consistent with a complete loss of chromosome 18 (monosomy 18) or a specific deletion of the 18q21 (BCL2) locus. Subsequent to these results, the tumor was formally diagnosed as high-grade B-cell lymphoma, activated B-cell (ABC) subtype, with MYC and BCL6 gene rearrangements (double-hit primary CNS lymphoma), with the International Extranodal Lymphoma Study Group (IELSG) prognostic score of 1 (for the location of the tumor). Cytological and flow cytometry analysis on cerebrospinal fluid (CSF) showed small B-lymphocytes accounting for less than 0.1% of total events.

The patient began treatment with the MATRix regimen (intravenous high-dose methotrexate, cytarabine, thiotepa, and rituximab) [4]. The patient has received 3 cycles of the MATRix regimen and the atest MRI showed complete remission of the lymphoma. We have had a follow-up of 8 months since the patient completed the MATRix regimen and he continues to be in remission.

Double-hit (DHL) and triple-hit lymphomas are being classified as “high grade B-cell lymphoma (HGBL) with rearrangements of MYC and BCL2 and/or BCL6” within the 2016 revision of the lymphoma classification by the World Health Organization (WHO) [5]. In addition to gene rearrangements, PCNSL with increased expression of MYC and BCL2 by IHC are associated with poor prognosis as well, and these are commonly known as “Double expresser (DE)” lymphomas [6]. Increased expressions of MYC alone [7, 8] or BCL6 alone [9] are independent markers of poor prognosis in PCNSL. A true double-hit PCNSL has been described only once in the past by Sakurai et al. [10]. In this particular case, the patient had CCND1/MYC translocation on FISH testing and the disease had an aggressive course. To the best of our knowledge, the case presented here is the second case of DHL of primary CNS origin and the first one describing dual translocations of MYC and BCL6.

Management of PCNSL is cumbersome and still needs to be standardized as there are very few prospective studies done in this regard to guide the management. Although this patient had DHL, his IELSG prognostic score was surprisingly low at 1, indicating the limitations of this prognostic score, which does not include cytogenetic or molecular risk factors. The patient received the MATRix regimen (methotrexate, cytarabine, thiotepa, rituximab) and after 3 rounds, he achieved complete remission with complete resolution of the CNS mass. We omitted thiotepa from the second round because of prolonged cytopenias. A recent phase 2 trial done by Ferreri et al. [4] showed significant improvement in the complete remission rate and overall survival in PCNSL patients receiving the MATRix combination as compared to methotrexate/cytarabine and methotrexate/cytarabine/rituximab. Whole brain radiation and intrathecal chemotherapy have not shown any survival benefit in such cases. High-dose chemotherapy with autologous stem-cell transplantation (HDC-ASCT) is still under investigation for its use in primary CNS lymphomas [2]. Interestingly, our patient had a deletion of BCL2 and low BCL2 expression makes the prognosis slightly favorable [11], which may be negating some of the adverse effects of the rearrangement of MYC and BCL6.

DHL among PCNSL is a rare entity; however, it might be possible that we are not testing diligently for gene rearrangements in PCNSL. The current IELSG prognostic score can be misleading in such cases and DHL or DE lymphoma of primary CNS origin needs to be treated aggressively at the outset. Future studies, which include molecular profiling, are needed to standardize the treatment of PCNSL.