Abstract

Background

“Double-Hit” (DH) lymphomas are most commonly defined as B cell lymphomas demonstrating a MYC gene rearrangement and additional rearrangement(s) involving BCL2 and/or BCL6. DH lymphomas respond poorly to standard immunochemotherapy regimens, often prompting the use of more intensive treatments. DH gene rearrangements can be identified through metaphase cytogenetic testing or more sensitive fluorescence in situ hybridization (FISH) on diagnostic tissue specimens, although these studies are not routinely performed. Here, we analyze a cohort of B cell lymphoma patients to determine whether DH status can be predicted by clinicopathologic features as well as the impact of DH status on survival.

Methods

Fifty-three patients diagnosed with B cell lymphoma treated at the University of Pennsylvania from 2006-2013 who underwent diagnostic FISH for MYC gene rearrangements using probes to detect either an 8q24 split or t(8;14) were included in this analysis. FISH was performed at request of the interpreting pathologist or treating clinician. Patients with classic Burkitt lymphoma were excluded. Cases of DH lymphoma (DH+) were defined as demonstrating at least one of either 8q24 split, t(8;14), t(2;8) or t(8;22) as well as a BCL2, BCL6 and/or BCL1 rearrangement. Therapy was given at the discretion of the treating clinician. Response was defined using the Revised Response Criteria for Malignant Lymphoma (J Clin Oncol. 2007 Feb 10;25(5):579-86.).

Results

DH+ was detected in 17 patients (32%) and a sole MYC gene rearrangement was detected in an additional 9 patients (17%). MYC gene rearrangements were detected by metaphase cytogenetics in 4 (15%) and by FISH in 22 (85%) of these patients. No factor, including age, LDH, stage, International Prognostic Index (IPI) or histology was predictive of DH status (Table I). DH+ patients were treated with R-hyperCVAD (41%), R-CHOP (41%) and other regimens (18%). Complete response was less frequent in DH+ compared to non-DH patients (41% vs. 81%, p=0.002). With a median follow-up of 10.4 months (range 1.2-72.4), the median overall survival was significantly shorter for DH+ compared to non-DH patients (8.2 vs. 56.8 months, p<0.001). Median overall survival was not significantly different for non-DH patients with and without a sole MYC gene rearrangement (50.8 months vs. not yet reached, p=0.33). Univariate Cox regression analysis showed that the presence of a MYC gene rearrangement (MYC+) and DH+ had statistically significant associations with overall survival; however, only DH+ retained statistical significance on multivariate analysis (Table II).

Table I. Baseline Characteristics

 All (n=53) Non-DH (n=36) DH+ (n=17) p value 
Median age (years) at diagnosis (range) 60 (19-81) 61 60 0.75 
Age >60 24 (45%) 18 (50%) 6 (35%) 0.38 
Elevated LDH (>192 U/L) 45 (85%) 29 (81%) 16 (94%) 0.41 
Median %Ki-67+ cells (range) 85 (5-100) 90 80 0.53 
Histology    0.25 
Follicular 5 (9%) 3 (8%) 2 (12%)  
Mantle cell 1 (2%) 1 (3%)  
Diffuse large cell 37 (70%) 27 (75%) 10 (58%)  
B cell lymphoma unclassifiable 8 (15%) 5 (14%) 3 (18%)  
Burkitt-like 2 (4%) 2 (12%)  
Stage (n=52)    1.0 
2 (4%) 2 (5%)  
II 8 (15%) 6 (17%) 2 (12%)  
III 9 (17%) 6 (17%) 3 (19%)  
IV 33 (64%) 22 (61%) 11 (69%)  
IPI (n=46)    0.59 
0-1 9 (20%) 6 (19%) 3 (21%)  
9 (20%) 8 (25%) 1 (8%)  
19 (40%) 12 (37%) 7 (50%)  
4-5 9 (20%) 6 (19%) 3 (21%)  
B symptoms 18 (34%) 11 (31%) 7 (41%) 0.54 
Bone marrow disease (n=49) 22 (45%) 13 (38%) 9 (60%) 0.22 
Extranodal disease 32 (60%) 23 (64%) 9 (53%) 0.56 
 All (n=53) Non-DH (n=36) DH+ (n=17) p value 
Median age (years) at diagnosis (range) 60 (19-81) 61 60 0.75 
Age >60 24 (45%) 18 (50%) 6 (35%) 0.38 
Elevated LDH (>192 U/L) 45 (85%) 29 (81%) 16 (94%) 0.41 
Median %Ki-67+ cells (range) 85 (5-100) 90 80 0.53 
Histology    0.25 
Follicular 5 (9%) 3 (8%) 2 (12%)  
Mantle cell 1 (2%) 1 (3%)  
Diffuse large cell 37 (70%) 27 (75%) 10 (58%)  
B cell lymphoma unclassifiable 8 (15%) 5 (14%) 3 (18%)  
Burkitt-like 2 (4%) 2 (12%)  
Stage (n=52)    1.0 
2 (4%) 2 (5%)  
II 8 (15%) 6 (17%) 2 (12%)  
III 9 (17%) 6 (17%) 3 (19%)  
IV 33 (64%) 22 (61%) 11 (69%)  
IPI (n=46)    0.59 
0-1 9 (20%) 6 (19%) 3 (21%)  
9 (20%) 8 (25%) 1 (8%)  
19 (40%) 12 (37%) 7 (50%)  
4-5 9 (20%) 6 (19%) 3 (21%)  
B symptoms 18 (34%) 11 (31%) 7 (41%) 0.54 
Bone marrow disease (n=49) 22 (45%) 13 (38%) 9 (60%) 0.22 
Extranodal disease 32 (60%) 23 (64%) 9 (53%) 0.56 

Table II. Cox Regression Analysis for Overall Survival

 Univariate Multivariate 
 p Value HR (95% CI) p Value HR (95% CI) 
Age >60 0.08 — — — 
Elevated LDH 0.29 — — — 
Histology 0.18 — — — 
Stage 0.07 — — — 
IPI category 0.17 — — — 
B symptoms 0.07 — — — 
Bone marrow involvement 0.06 — — — 
Extranodal disease 0.34 — — — 
Treatment regimen 0.20 — — — 
MYC+ 0.004 5.0 (1.6-15.1) 0.53 — 
DH+ <0.001 9.2 (3.2-26.8) 0.009 6.7 (1.6-27.7) 
 Univariate Multivariate 
 p Value HR (95% CI) p Value HR (95% CI) 
Age >60 0.08 — — — 
Elevated LDH 0.29 — — — 
Histology 0.18 — — — 
Stage 0.07 — — — 
IPI category 0.17 — — — 
B symptoms 0.07 — — — 
Bone marrow involvement 0.06 — — — 
Extranodal disease 0.34 — — — 
Treatment regimen 0.20 — — — 
MYC+ 0.004 5.0 (1.6-15.1) 0.53 — 
DH+ <0.001 9.2 (3.2-26.8) 0.009 6.7 (1.6-27.7) 
Conclusions

DH status cannot be inferred by baseline disease- or patient-related characteristics and is most predictive of overall survival in this cohort of B cell lymphoma patients. These findings support the practice of routine FISH for DH gene rearrangements in order to better identify DH+ patients who may benefit from risk-adapted and/or targeted therapies. We plan to validate our findings in a larger unselected cohort of diffuse large B cell lymphoma patients.

Disclosures:

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.