Until quite recently, the prevailing view, adopted by the WHO 2008 Classification, was that mantle-cell lymphoma (MCL) originates from a peripheral B cell located within the inner mantle zone, an area comprised of naïve pre-germinal center (GC) type B cells. However, this notion has been challenged by molecular and functional evidence. Indeed, MCL is characterized by a skewed repertoire of immunoglobulin heavy variable (IGHV) genes and by some imprint of somatic hypermutation (SHM) in the clonotypic IGHV genes of the great majority (~70%) of cases, indicating antigen selection. Furthermore, both relapsed/refractory and treatment-naïve patients with MCL exhibit remarkable responses to B-cell receptor signaling inhibitors, strongly supporting a role for microenvironmental triggering in the natural history of MCL. In the present study, we sought to obtain additional insight into MCL ontogeny through a combined morphologic, immunohistochemical and immunogenetic analysis of 230 patients with a diagnosis of MCL according to the 2008 WHO Classification criteria. The study group included 139 nodal, 32 extranodal, 18 primary splenic MCLs as well as 41 bone marrow biopsies (BMB) infiltrated by MCL. Morphologically, 144/206 (70%) cases were ascribed to the common variety, while 48/206 (23.3%) and 14/206 (6.7%) were characterized as blastoid or pleomorphic variant, respectively. The immunohistochemical analysis (on paraffin sections) focused on CD27, DBA.44 and IRF4 (MUM1), markers not normally expressed by the naïve pre-germinal centre B-cell of the inner mantle zone. The results were as follows: (i) 117/214 (54.7%) cases positive for CD27 expression; (ii) 18/176 (10.2%) cases positive for DBA.44; (iii) 53/98 (54%) cases positive for IRF4. Amongst CD27+ cases, 10/86 (11.6%) were also positive for DBA.44, whereas 27/51 (52.9%) were also positive for IRF4. Immunogenetic information regarding IGHV-IGHD-IGHJ gene rearrangements was available for 167 cases of the study. Fifty of 167 cases (30%) carried IGHV genes with no SHM (100% identity to the germline, GI), whereas the remaining 117 cases (70%) bore some imprint of SHM: in particular, 95/167 cases (56.8%) carried IGHV genes with 97-99.9% GI, while 22/167 cases (13.2%) carried IGHV genes with <97% GI. In keeping with the literature, the IGHV gene repertoire of the present cohort was remarkably biased, with the IGHV3-21, IGHV4-34, IGHV3-23, IGHV1-8 and IGHV5-51 genes accounting for 51% of cases (85/167). The following noteworthy observations were made from the combined assessment of morphological, immunohistochemical and immunogenetic results. (1) DBA.44 was not detected in any of the 21 extranodal MCL cases analyzed, was rare in nodal MCL (5/108, 4.6%), whereas, in contrast, was significantly enriched among primary splenic MCL (6/14, 42.8%; p<0.01 for both comparisons). (2) Unexpectedly, CD27 expression was more prevalent among cases with minimal/borderline SHM (56/91, 61.5%) or no SHM (100% GI: 23/46, 50%), whereas it was less frequent among cases with a significant SHM load (<97% GI: 6/20, 30%; p=0.01 for comparison to 100% GI cases). (3) CD27 expression was significantly (p<0.05) more frequent amongst cases classified as pleomorphic (11/14 cases, 78.6%) versus either blastoid (38/46 cases, 60.8%) or common (65/131 cases, 49.6%). (4) IRF4 was detected in cases from all SHM categories: 10/16 (62.5%) cases with 100 GI%, 14/30 (46.6%) cases with 97-99.9% GI and 3/6 (50%) cases with <97% GI. (5) Blastoid cytology was less frequent in primary splenic MCL (2/18 cases, 11.1%) compared to either nodal (33/126 cases, 26.2%) or extranodal MCL (8/28 cases, 28.5%), however the difference did not reach significance likely due to small numbers. In conclusion, we document the remarkable immunohistochemical and immunogenetic heterogeneity of MCL. Certain profiles, identified here for the first time, are in sharp contrast to those of the naïve pre-germinal centre B-cell of the inner mantle zone (IG-unmutated, CD27-, IRF4-, DBA.44-), which is the postulated MCL progenitor according to the WHO 2008 classification. These findings strongly support antigen drive in a significant fraction of MCL cases. Furthermore, they raise the intriguing possibility that many ontogenetic pathways may give rise to MCL or, alternatively, that several types of normal B cells may serve as MCL progenitors.

Disclosures

Stamatopoulos:Janssen Pharmaceuticals: Honoraria, Membership on an entity's Board of Directors or advisory committees.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution