Alonso-Álvarez Sara, Alcoceba Miguel, García-Álvarez María, Blanco Oscar, Rodríguez Marta, Baile Mónica, Caballero Juan Carlos, Dávila Julio, Vidriales María Belén, Esteban Carmen, Arias Piedad, Díaz Luis G, Tamayo Pilar, Caballero María Dolores, Gutiérrez Norma C, González Marcos, Martín Alejandro
Department of Hematology, University Hospital of Salamanca (HUS-IBSAL), CIBERONC, and Cancer Research Institute of Salamanca-IBMCC (CSIC-USAL University), 37007 Salamanca, Spain.
Department of Pathology, University Hospital of Salamanca (HUS/IBSAL), 37007 Salamanca, Spain.
Cancers (Basel). 2020 Feb 18;12(2):474. doi: 10.3390/cancers12020474.
The biology and clinical impact of bone marrow (BM) infiltration in patients with diffuse large B-cell lymphoma (DLBCL) remains unclear in the rituximab era. We retrospectively analyzed 232 patients diagnosed with DLBCL at our center between 1999 and 2014. Concordant-presence of large cells similar to those of the lymph node biopsy- and discordant-infiltration by small cells forming lymphoid aggregates, lacking cytological atypia-BM infiltration was defined by histological criteria and further characterized by flow cytometry (FCM). Cell of origin (COO) was determined using Hans' algorithm. For the clonal relationship between tumor and discordant BM, the VDJH rearrangement was analyzed. Survival analyses were restricted to 189 patients treated with rituximab and chemotherapy. Thirty-six (16%) had concordant, and 37 (16%) discordant BM infiltration. FCM described different indolent lymphomas among discordant cases, clonally related with DLBCL in 10/13 available samples. Median follow-up was 58 months. 5-year-progression-free survival (PFS) for non-infiltrated, discordant and concordant groups was 68%, 65% and 30%, respectively ( < 0.001). Combining COO and BM infiltration, patients with discordant BM and non-germinal center B-cell COO also had decreased 5-year-PFS (41.9%). In multivariate analysis, concordant BM had an independent effect on PFS (HR 2.5, p = 0.01). Five-year cumulative incidence of central nervous system (CNS) relapse was 21%, 4% and 1% in concordant, discordant and non-infiltrated groups, respectively ( < 0.001). In conclusion, concordant BM infiltration represents a subset with poor prognosis, whereas the prognostic impact of discordant BM infiltration could be limited to non-CGB cases.
在利妥昔单抗时代，弥漫性大B细胞淋巴瘤（DLBCL）患者骨髓（BM）浸润的生物学特性及临床影响仍不明确。我们回顾性分析了1999年至2014年间在本中心确诊为DLBCL的232例患者。根据组织学标准定义与淋巴结活检中相似大细胞的一致性存在以及由形成淋巴样聚集的小细胞构成的不一致性浸润（缺乏细胞学异型性），并通过流式细胞术（FCM）进一步进行特征分析。采用汉斯算法确定细胞起源（COO）。对于肿瘤与不一致性骨髓之间的克隆关系，分析VDJH重排。生存分析仅限于189例接受利妥昔单抗和化疗的患者。36例（16%）存在一致性骨髓浸润，37例（16%）存在不一致性骨髓浸润。FCM在不一致性病例中描述了不同的惰性淋巴瘤，在13份可用样本中有10份与DLBCL存在克隆相关性。中位随访时间为58个月。非浸润组、不一致性浸润组和一致性浸润组的5年无进展生存率（PFS）分别为68%、65%和30%（<0.001）。结合COO和骨髓浸润情况，不一致性骨髓浸润且非生发中心B细胞COO的患者5年PFS也降低（41.9%）。在多变量分析中，一致性骨髓浸润对PFS有独立影响（风险比2.5，p = 0.01）。一致性浸润组、不一致性浸润组和非浸润组中枢神经系统（CNS）复发的5年累积发生率分别为21%、4%和1%（<0.001）。总之，一致性骨髓浸润代表预后不良的一个亚组，而不一致性骨髓浸润的预后影响可能仅限于非CGB病例。
