Diffuse large B-cell lymphoma (DLBCL) is increasingly understood not as a single disease, but as a spectrum of biologically diverse entities shaped by molecular features, clinical risk factors, and treatment-related effects. A series of recent analyses highlights how advances in genomic profiling, imaging-based prognostication, and survivorship research are converging to refine patient management across the disease continuum.
In one analysis by Wang et al, 23% of DLBCL harbored genetic subclones expressing B-cell differentiation themes that distinguished them from other malignant cells in the same tumor.1 These themes, ranging from germinal center B-cell and memory B-cell phenotypes to cell cycle and cell growth signatures, were closely linked to outcomes. For example, germinal center–associated signatures correlated with better responses to therapy, whereas cell growth–related themes predicted poorer survival.1
Complementing these findings, another study by Albitar et al identified a distinct subset of DLBCL characterized by an APOBEC-like mutational signature, present in approximately 6% of cases.2 These tumors displayed marked hypermutation and clustered genomic alterations, along with a unique transcriptomic profile. A defined gene expression signature was able to reliably identify this subgroup, raising the possibility of incorporating such markers into routine diagnostics.2
Together, these molecular advances underscore a broader shift toward biologically driven classification systems that move beyond traditional histology. By integrating gene expression signatures and subclonal architecture, clinicians may be better able to stratify risk and tailor treatment strategies in a disease, long recognized for its heterogeneity.
In parallel, efforts to refine prognostication using noninvasive tools are gaining traction, particularly in high-risk populations. In patients with relapsed or refractory TP53-aberrant DLBCL, PET/CT-based radiomic analysis by Pizzuti et al demonstrated a strong predictive value for survival outcomes.3 Quantitative imaging metrics, such as changes in SUVmax, metabolic tumor volume, and total lesional glycolysis, were closely associated with both progression-free and overall survival.3
Notably, these radiomic features outperformed traditional clinical tools such as the Revised International Prognostic Index, which failed to distinguish outcomes in this cohort. A composite model incorporating multiple high-risk imaging features identified a subgroup of patients with a particularly poor prognosis, suggesting that radiomics may offer a more precise and dynamic approach to risk assessment in aggressive disease.3
Although advances in biology and risk prediction are improving front-end decision-making, attention is also turning to survivorship, where treatment-related complications can have a lasting impact. A large population-based study by Ravindran et al examining opioid use in patients treated for DLBCL found that 55% of patients received opioid medications during therapy.4 Among these patients, 6.7% developed new persistent opioid use in the year following treatment.
Risk of persistent use was significantly higher among patients exposed to opioids during treatment, with additional predictors including relapsed or refractory disease, greater comorbidity burden, palliative care involvement, and a history of mental health conditions.4 These findings highlight the importance of balancing symptom control with long-term risk and underscore the need for proactive monitoring and intervention strategies in survivorship care.
Collectively, these studies illustrate how DLBCL management is evolving toward a more comprehensive and individualized model. Molecular profiling is uncovering distinct disease subsets, advanced imaging is enhancing prognostic precision, and real-world data are shedding light on the long-term consequences of treatment.