Usage of IGHV4-39 with Stereotypic B Cell Receptor Is An Independent Risk Factor of Chronic Lymphocytic Leukemia Transformation to Richter Syndrome

Richter’s syndrome (RS) represents the transformation of chronic lymphocytic leukemia (CLL) to aggressive lymphoma, mainly occurring as diffuse large B-cell lymphoma (DLBCL). The biology of CLL transformation to RS is poorly understood and knowledge on risk factors of RS development is scant. We tested whether IGHV gene usage and stereotypic B cell receptor (BCR) at CLL diagnosis have an impact on RS transformation. The first step of the study consisted of a case-control analysis comparing IGHV gene usage and prevalence of stereotypic HCDR3 in RS (n=69; all DLBCL) versus a control group (n=715) of CLL that had not transformed to RS. The second step consisted of an actuarial assessment of the impact of IGHV gene usage and stereotypic HCDR3 at CLL diagnosis, on the risk of subsequent transformation to RS in a cohort of 754 CLL, of which 39 had transformed to RS. Comparison of IGHV usage in unmutated RS versus unmutated control CLL documented that IGHV4-39 was the sole gene preferentially utilized (6/48, 12.5% vs 5/277, 1.8%, respectively, p=.002) by RS. Prevalence of stereotypic HCDR3 was significantly higher in RS compared to non-transformed CLL when considering all cases (RS: 50.7% vs non-transformed CLL: 22.2%; p<.000001), unmutated cases only (RS: 58.3% vs non-transformed CLL: 35.7%; p=.003), and mutated cases only (RS: 33.3% vs non-transformed CLL: 13.7%; p=.022). Compared to non-transformed CLL, RS preferentially utilized BCR belonging to a subset characterized by rearrangement of unmutated IGHV4-39/IGHD6-13/IGHJ5 genes (2/159, 1.2% vs 5/35, 14.3%, respectively; p=.002). All cases with stereotypic IGHV4-39 carried +12 as the sole FISH abnormality. After a median follow-up of 41.1 months, 39/754 CLL had transformed to RS. Univariate analysis documented:

• shorter time to transformation in CLL utilizing IGHV4-39 (5-year risk: 35.4%) compared to CLL utilizing other IGHV genes (5-year risk: 5.6%) (p<.000001);

• higher risk of RS in CLL utilizing stereotypic HCDR3 (5-year risk: 14.2%) compared to CLL without stereotypic HCDR3 (5-year risk: 3.9%) (p<.00001).

CLL with stereotypic HCDR3 and IGHV homology 98% showed a significantly higher risk of transformation (5-year risk: 18.4%) compared to CLL with IGHV homology 98% but without stereotypic HCDR3 (5-year risk: 6.8%) (p=.006). Also, stereotypic HCDR3 identified a CLL subgroup that, despite presenting with IGHV homology <98%, showed an increased risk of RS (p=.040). This observation indicates that stereotypic HCDR3 is not a surrogate of IGHV homology for RS prediction. We then tested the independent predictive value for RS transformation of IGHV4-39 usage and of stereotypic HCDR3. Multivariate analysis selected IGHV4-39 usage (HR: 4.25; p=.002) and stereotypic HCDR3 at CLL diagnosis (HR: 3.08; p=.002) as independent predictors of RS transformation. The observation that all RS utilizing IGHV4-39 carried stereotypic HCDR3 prompted investigation of the interaction between IGHV4-39 usage and stereotypic HCDR3 in the model. Multivariate analysis selected the interaction between IGHV4-39 usage and stereotypic HCDR3 at CLL diagnosis as the strongest independent predictor of RS transformation (HR: 5.13; p=.001). The relevance of the interaction between IGHV4-39 and stereotypic HCDR3 was confirmed by bivariate log rank analysis. Accordingly, CLL utilizing both IGHV4-39 and stereotypic HCDR3 were identified as the disease category with highest risk of transformation (5-year risk: 68.7%). Transformation to RS and progression to symptomatic disease according to NCI Working Group guidelines are distinct events in CLL. Accordingly, neither IGHV4-39 usage nor stereotypic HCDR3 affected the risk of CLL progression occurring without transformation to RS. IGHV4-39 usage and stereotypic HCDR3 may be appropriate biological markers for RS prediction since:

• these markers predict RS in a fashion that is independent of other clinical and biological features;

• given the widespread use of IGHV sequencing for CLL prognostication, information on IGHV4-39 and stereotypic HCDR3 may be obtained at CLL diagnosis without additional testing; and importantly

• all CLL with concomitant IGHV4-39 usage and stereotypic HCDR3 ultimately transform to RS.

A close monitoring and a careful biopsy policy may be of help for early recognition of RS transformation in patients carrying IGHV4-39 usage and stereotypic HCDR3.