Plasmablastic Lymphoma Descriptors and Clinicopathologic Determinants of Survival: Analysis of a Pooled Database

Background

Plasmablastic Lymphoma (PBL) is a rare and aggressive subtype of non-Hodgkin's lymphoma. It tends to be associated with immunosuppressive clinical contexts, such as HIV infection and immunosuppressive therapies for autoimmune disorders and organ transplants. However, due to the rarity of the disease, which is poorly understood, varied treatment approaches have been implemented, and no optimal data or guidelines for managing PBL are available. Furthermore, a few small case series reported conflicting prognosticators. Therefore, we conducted this pooled database analysis to identify the prognostic factors, clinicopathological characteristics, and therapeutic strategies influencing survival in this rare disease.

Methods

To study the demographic characteristics, molecular and immunohistochemical signatures, therapeutic interventions, survival, and prognostic factors, we compiled a pooled database of 300 cases that fit the diagnostic criteria for PBL. Kaplan-Meier survival curves were constructed. Cox proportional hazards model and Log-rank tests were used to assess the influence of demographic and clinicopathologic factors on overall survival (OS).

Results

A total of 300 patients with confirmed PBL were identified. The median age was 49. There was a male preponderance with M:F of 2.7. The median duration of symptoms before diagnosis was 1 month. PBL involved the H&N (39%), Lymph nodes (LN) (33%), GI tract (31%), bone marrow (17%), bones (14%), lungs and pleura (10%), skin (9%), spleen (7%), liver (6%), CNS (6%), and testes (2%). Constitutional symptoms were reported in 35%. Stage III/IV accounted for 51% of the cohort. Sixty-two percent were immunosuppressed, 47% due to HIV. While 58% were Kappa restricted, 40% were Lambda, and 2% null. The median OS and DFS of the whole group were 25 and 15 months, respectively. The involvement of BM (p<0.0001), liver (p=0.003), lungs/pleura (p<0.0001), and upper GI tract (0.001) was associated with worse OS. Moreover, CD4<100 (p=0.006), HHV8+ (p=0.02), EBER-negative (p=0.01), LDH>500 (p=0.04), and stage>2 (p<0.0001) were also associated with worse OS. PBL involving the H&N had better OS (p=0.0001). Though CNS involvement, presence of constitutional symptoms, and MYC+ had numerically worse OS, it did not reach statistical significance. Age, sex, light chain restriction, type of immunosuppression, other organ involvement, and other IHC phenotype did not impact OS. Compared to no treatment, chemotherapy and stem cell transplant had incrementally superior median OS (2 vs. 27 vs. NR months, p<0.0001). Frontline intensive chemotherapy yielded better OS than CHOP-like and myeloma regimens in decreasing order (p<0.0001). Patients who attained CR as their best response also had a superior median OS (p<0.0001).

Conclusion

This study presents clinicopathologic data from the largest pooled cohort of patients with PBL. It identifies the type of organ involvement, CD4 counts, EBER and HHV8 status, LDH levels, stage, type of therapy, and quality of response to treatment as critical determinants of OS.