Abstract
Abstract 1595
Immunohistochemistry for Cyclin D1 (CCND1) expression is routine in cases suggestive of mantle cell lymphoma (MCL). Most MCL are t(11;14)/IGH-CCND1-positive by FISH. PCR based detection of the fusion transcript is hampered by widespread breakpoints. Only few data is available on quantitative real-time PCR (RQ-PCR) for CCND1 expression measurement.
To assess CCND1 mRNA expression and correlate it with t(11;14) in mature B-cell neoplasms.
We established a RQ-PCR assay for CCND1 mRNA measurement and investigated 451 cases: 142 MCL (in all cases IGH-CCND1 confirmed by FISH), 76 chronic lymphocytic leukemia (43 typical CLL, 33 CLL/PL), 20 hairy cell leukemia (HCL), 13 hairy cell leukemia-variant (HCL-v), 20 splenic marginal zone lymphoma (SMZL), 91 other mature B-cell neoplasms. CCND1 background expression was assessed in 29 pts with other hematological neoplasms and 60 healthy individuals. FISH and/or chromosome banding analysis for the t(11;14) was available in 364 pts.
Bone marrow (BM, n=267) or peripheral blood (PB; n=184) samples were analyzed by cytomorphology, multiparameter flow cytometry (MFC), FISH, and RQ-PCR. CCND1 mRNA expression was given by RQ-PCR in comparison to ABL1 mRNA expression (%CCND1/ABL1). Limited dilution of high expressers into cDNA of healthy controls revealed a sensitivity of the assay of up to 0.1 %.
IGH-CCND1 translocation carriers had higher %CCND1/ABL1 than those without which hold true in the total cohort (mean±SD, 420.4±740.3 vs 17.8±128.3; p<0.001), separately in all B-cell neoplasms (420.4±740.3 vs. 20.6±140.5; p<0.001), in all CLL (352.9±302.9 vs. 28.8±201.6; p<0.001), and in all SMZL (178.7±236.9 vs. 3.4±4.8; p=n.s.). In 150 IGH-CCND1- positive samples (142 MCL, 3 CLL, 2 SMZL, 3 other mature B-Cell neoplasms) median %CCND1/ABL1 level was 178.2 (range, 1.5–4,152.0). Normalized by the number of FISH-positive cells assuming 100% lymphoma infiltration, IGH-CCND1- positive samples showed a median %CCND1/ABL1 of 445.8 (range, 17.9–4848.5). Therefore, we chose a normalized %CCND1/ABL1 expression of 17.0 as threshold for CCND1 positivity. In total, 131/150 IGH-CCND1- positive samples had %CCND1/ABL1 expression levels above this threshold resulting in a sensitivity of 87.3% (not normalized) for detection of the t(11;14) by CCND1 expression. In the IGH-CCND1-positive samples, %CCND1/ABL1 was higher in PB than BM (p=0.003) corresponding to a higher infiltration degree in PB by MFC (p=0.002). In IGH-CCND1 negative samples (n=214), the median %CCND1/ABL1 was 2.4 (range, 0.0–1,721.5). In total, 204/214 of t(11;14)/IGH-CCND1 negative cases had low CCDN1 expression (<17.0), resulting in a negative predictive value of 95.3%. In the group of other mature B-cell neoplasms median %CCND1/ABL1 was 1.1 (range, 0.1–72.4), in healthy individuals 1.1 (range, 0.0–7.8). Mean %CCND1/ABL1 was higher in MCL than in other mature B-cell neoplasms pts (mean±SD, 392.9±685.3 vs. 51.5±324.1; p<0.001). The mean expression levels were considerably lower in CLL, CLL/PL, SMZL, and other mature B-cell neoplasms, but the maximum values in these subgroups were comparable to MCL, thus reflecting a significant overlap and indicating that IGH-CCND1-positive lymphomas other than MCL can have increased CCND1 expression. HCL and HCL-v showed low CCND1 expression. Of 141 cases with %CCND1/ABL1 values above the threshold of 17.0, in which FISH was available, 131 cases had a t(11;14)/IGH-CCND1 (92.9%), and 10 were translocation-negative. Thus, the specificity of %CCND1/ABL1 to indicate t(11;14) was 92.9%. Performing Spearman rank correlation in MCL, the %CCND1/ABL1 ratio was correlating with % aberrant nuclei by FISH (R=0.688; p<0.001) and lymphoma infiltration by MFC (R=0.700; p<0.001). 34 cases were analyzed during follow up, CCDN1 expression level was stable when untreated, showed 2.5–3.5 log reduction after therapy, and increase at relapse.
CCND1 expression is highly correlated with t(11;14)/IGH-CCND1 and significantly increased in MCL as compared to other B-cell lymphoma subtypes. Increased CCND1 expression can also be found in other lymphoma subtypes. CCND1 expression measurement by RQ-PCR may be considered a useful adjunct for diagnosis in pts with suspected MCL or in lymphoma pts with suspected t(11;14)/IGH-CCND1 but ambiguous results by other techniques. This technique may also be useful to monitor therapy response.
Bacher:MLL Munich Leukemia Laboratory: Employment. Kern:MLL Munich Leukemia Laboratory: Equity Ownership. Haferlach:MLL Munich Leukemia Laboratory: Equity Ownership. Alpermann:MLL Munich Leukemia Laboratory: Employment. Haferlach:MLL Munich Leukemia Laboratory: Equity Ownership. Schnittger:MLL Munich Leukemia Laboratory: Equity Ownership.
Author notes
Asterisk with author names denotes non-ASH members.