Nodal progression in CLL: Real-world predictors of richter transformation and identification of a distinct high-risk NP-CLL subgroup

Abstract Background Aggressive behavior of chronic lymphocytic leukemia (CLL) is relatively uncommon, but in clinical practice, concern for Richter transformation (RT) often prompts PET-CT and biopsy in patients with progressive lymphadenopathy. While risk factors are described in patients progressing on targeted therapy trials, real-world predictors of transformation remain poorly defined. Patients with biopsy-proven nodal progression without transformation (NP-CLL) are less well characterized and may represent a biologically distinct subgroup. Methods We identified patients in the Dana-Farber CLL database who underwent PET-CT followed by biopsy within 3 months to evaluate possible RT, based on clinical concern for nodal progression. Logistic regression was used to assess predictors of transformation. Overall survival (OS) was estimated using Kaplan-Meier method and compared by log-rank tests. Results Of 4,103 CLL patients, 313 were identified via electronic query as having PET-CT and biopsy within 3 months. After manual review to confirm CLL-related indication and availability of scan and pathology reports, 249 were included. Median age at PET-CT was 66 (range 40-90), median interval from diagnosis was 4 years; 73% male, 55.7% treatment-naïve, 29.3% prior targeted therapy. Based on pathology review, 70 (28.1%) were classified as transformed (NP-T) and 179 as NP-CLL. Among NP-CLL cases, 12 (6.7%) had an initial biopsy showing CLL and were diagnosed with RT on repeat biopsy within a median of 8.5 months (range 3.6-19.0). At a median follow-up of 2.3 years, other NP-CLL cases have not transformed. Among transformed cases, diffuse large B-cell lymphoma (DLBCL) was the most common histology (57 cases, 81.4%). Of 22 cases with known cell-of-origin, 16 were non-GCB and 4 GCB. Classical Hodgkin lymphoma (cHL) accounted for 10 cases (14.3%). In patients whose most recent prior therapy included a targeted agent (BTKi, BCL2i, or PI3Ki), whether on or off treatment, early progression defined using ROC analysis as <1.54 years from treatment initiation to PET was strongly associated with transformation (OR: 4.33, 95% CI: 1.05-17.84; p=0.0422, AUC 0.90). Other treatment-related predictors included number of prior therapies (OR: 1.32 per line, 95% CI: 1.07-1.64; p=0.0094), while treatment-naïve status was protective (OR: 0.37, 95% CI: 0.21-0.67; p=0.0010). SUVmax was a predictor of transformation (OR: 2.65 per 5-unit increase, 95% CI: 1.95-3.59; p<0.0001). Trisomy 12 was associated with reduced odds of transformation (OR: 0.45, 95% CI: 0.21-0.93; p=0.0400). TP53 aberrations and IGHV4-39 usage, both of which are associated with RT in prior studies, were more frequent among transformed pts but did not reach statistical significance. Median OS from biopsy was significantly shorter in NP-T compared to NP-CLL (1.75 vs. 7.71 years; p < 0.0001). To evaluate whether NP-CLL represents a higher-risk state, we compared OS from diagnosis in NP-CLL to a treated subset of our broader CLL cohort diagnosed between 2013-2023 (Santos Azevedo et al., ASH 2024). OS was significantly worse in NP-CLL compared to other treated CLL cases (p < 0.0001). At 5 years, OS was 86.0% in NP-CLL vs. 94.9% in the broader CLL cohort; by 10 years, OS was 54.7% vs. 82.7%, respectively. These differences suggest that nodal progression may identify a biologically distinct subgroup of CLL patients with inferior long-term outcomes, even in the absence of transformation. Re-review of NP-CLL biopsies identified 21 cases with features suggestive of accelerated disease (e.g., Ki-67 > 40% or descriptive terms such as “accelerated” or “aggressive”). OS did not differ by histologic features, suggesting these findings are either rare or under-annotated, and that pathology alone may not reliably identify high-risk NP-CLL. Conclusion This study reflects real-world diagnostic evaluation for possible RT and reveals biologic heterogeneity among patients with nodal progression. In addition to SUVmax and treatment-related factors, early progression following targeted therapy emerged as a strong predictor of transformation. Although only a minority of NP-CLL patients have transformed to date, their OS was inferior to the general CLL population, indicating that nodal progression even without transformation may mark a biologically aggressive state. These findings support recognition of NP-CLL as a distinct high-risk subgroup warranting further study.