Is It Possible to Omit Bone Marrow Biopsy in Diagnostic Workup in Patients with Newly Diagnosed Primary CNS Lymphoma? a Retrospective Analysis in the PET/CT Era

Abstract

Background: Primary central nervous system lymphoma (PCNSL) is a rare and aggressive extra-nodal non-Hodgkin lymphoma (NHL) confined to the brain, eyes, leptomeninges, and spinal cord. More than 90% of patients belong to diffuse large B-cell lymphoma (DLBCL). The consensus around the extensive systematic staging in patients with suspected or proven PCNSL is still a matter of debate. Current diagnostic and staging guidelines recommend extensive workup to exclude systemic disease with computed tomography (CT) scan of the chest, abdomen, and pelvis, a bone marrow biopsy (BMB), and testicular ultrasonography in males. Whole-body 18F-fluorodeoxyglucose positron emission tomography (PET-CT) is currently considered optional in diagnostic workup in patients with presumed PCNSL. The utility of BMB in the era of PET-CT has been challenged in systemic lymphomas, but data are lacking for patients with PCNSL. Among patients with PCNSL, BMB is often normal or reveals low-grade lymphomas or just clonal B-cells. In rare instances, aggressive lymphoma involves both bone marrow (BM) and the central nervous system (CNS). However, the role of BMB in PCNSL patients in the PET-CT era is yet to be established. This study aimed to analyze whether BMB could be safely omitted in the diagnostic workup in patients with PCNSL of DLBCL histology without signs of systemic lymphoma on PET-CT scans. Methods: A comprehensive nationwide retrospective multicenter study with enrollment of patients from the referent medical centers in Denmark for treatment of PCNSL was performed. Patients included in the study were diagnosed between 2002 and 2020 and identified through the Danish Pathology Register. All patients with CNS involvement were potential candidates. Inclusion criteria were the following: 1) patients should be diagnosed with DLBCL in brain, eyes, leptomeninges, cerebrospinal fluid, and spinal cord; 2) have available BMB; and 3) have available PET-CT without signs of lymphoma outside of CNS. Patients with PCNSL and prior history of other lymphoma were also included in the study. Patients were excluded if they were diagnosed with systemic lymphoma or PCNSL other than DLBCL. Patients who did not have PET-CT and BMB performed or in whom this information could not be retrieved through medical records were also excluded from further analysis. Results: Of 1238 potential candidates identified through the Danish pathology register, only 300 patients fulfilled the inclusion criteria. Most patients were excluded due to a lack of information on PET-CT scans and BMB. All included patients had available BMB and no signs of systemic lymphoma on PET-CT. Of the included patients, 252 (84%) were diagnosed with PCNSL without a previous lymphoma diagnosis, while 48 (16%) had a previous history of lymphoma. The median age of patients was 68 years. There was a slight predomination of men (60%) compared to women (40%). Twenty-three (7.7%) patients were found to have discordant findings in BMB. However, none of these showed infiltration with DLBCL. The histologies included various low-grade lymphomas found in 17 patients (5.7%), monoclonal B-cells in 3 patients (1%), and smoldering multiple myeloma in one patient. Conclusion: In this retrospective study of 300 patients with PCNSL with PET-CT scans without signs of systemic lymphoma outside CNS, we found no cases with BM infiltration of DLBCL. A small proportion of patients with PCNSL have a discordant, often low-grade lymphoma in the bone marrow. With a sample size of 300, the chance of overlooking DLBCL infiltration in the BM is ≤ 1 % (95% CI). Therefore, the results of this, so far the most extensive study evaluating BMB in PCNSL in the PET-CT era, suggest that BMB can safely be omitted in patients with PCNSL of DLBCL histology and no signs of systemic lymphoma on PET-CT scans.