DISCUSSION
PEL-like or HHV8 and EBV-negative PEBL is a rare B-cell neoplasm that typically presents as a malignant serous effusion in non-immunocompromised patients. In the last decade, several reports of HHV8-negative effusion lymphomas have been published demonstrating remarkably similar cytomorphologic characteristics to PEL but differences in immunophenotype, demographics, response to treatment and clinical outcome; suggesting that PEL-like is a distinct entity. As mentioned before, to date only HHV8-positive PEL is a recognized entity in the current WHO classification1; but HHV8 and EBV-negative PEBL has been recently suggested as a provisional entity by the International Consensus Classification of Mature Lymphoid Neoplasms.2
Our case is of particular interest because of the Caucasian origin of the patient as most of the published data are case series reported from Japan6. Despite its different origin, the other characteristics are similar to those described in the Japanese cases. Its prevalence is unknown and its diagnosis is challenging (especially because it has not been described in the current WHO classification).
HHV8 and EBV-negative PEL is considered to have a distinct etiopathogenic mechanism than HHV8-positive PEL. Different theories have emerged with no conclusive data to date. Kobashi et al7 suggested in 2007 that immunosenescence related to age could play a major role, as most of the patients were older (median age of 70 years) compared with PEL individuals. In the same line, Ohshima et al8 concluded that multi-step genomic abnormalities found in PEL-like cases could be related to age. In their study, most of the patients were elderly men with all cases showing complex karyotypes but without identification of a driver genomic abnormality. Suggested mechanisms of lymphomagenesis in HHV8 and EBV-negative PEL include impaired immune surveillance and dysregulatory haemopoietic pathway. Unlike PEL, in which HHV8 acts as an exogenous stimulus for local B-cell expansion, no viral mechanism has yet been identified in PEL-like lymphomas. In our case, Epstein-Barr virus-encoded small RNA in situ hybridization was negative but the patient had seropositivity for EBV. In the literature, around one third of cases of PEL-like lymphomas are infected with EBV.9However, whether EBV is simply a prior infection or a trigger of PEL-like lymphoma remains unclear and further investigation is needed. Regarding HCV infection in PEL-like lymphomas, there is a slightly higher prevalence of its infection among these patients (30% to 40% of patients). Association between HCV and B-cell non-Hodgkin lymphoma is well reported in the literature. Thus, it has been suggested that fluid overload related to cirrhosis due to HCV infection plays a role in PEL-like lymphomas. Moreover, underlying medical condition leading to fluid overload has been observed too in almost one-half of the patients.7 The ICC also recognizes chronic serosal stimulation by fluid overload in its pathogenesis.2Although this last hypothesis needs further confirmation and not all the authors agree on that.6
Tumor cells in HHV8 and EBV-negative PEL express pan-B-cell markers and they are negative for plasma cell markers in contrast with PEL. CD45 antigen expression is present in both neoplasms indicating hematological derivation of the effusion cells.10
According to the reported series, the site of the effusion varies but the most common location was pleural, followed by ascites and being the pericardial one the less common.7 Nonetheless, approximately 40% of the cases showed multiple body cavity lesions including pericardial in all of them.6 In our case, probably both pericardial and pleural effusions had a neoplastic involvement according to images from PET scan (figure 2b). Nonetheless, the reason why this B-cell neoplasm is exclusively confined to serous membranes without developing tumor masses is still non-explained.
Diffuse Large B-cell lymphoma associated with chronic inflammation is another B-cell neoplasm which also arises in the serous membranes but unlike our case, the pathogenic mechanism and localization of this entity is well understood. Chronic inflammation causes local immunosuppression permitting EBV-infected memory B-cells evade immunosurveillance, accumulate genetic alterations and evolve to lymphoma.11
There is no standard treatment for this entity due to the small number of cases although the majority of them are treated with chemotherapy including anthracycline and rituximab (as tumor cells express CD20). In a retrospective study conducted by Kaji et al, 86% of patients received CHOP-like regimens.6 Other systemic treatments consisted of rituximab in monotherapy or in combination with etoposide. Perhaps, due to the introduction of rituximab, the prognosis seems to be better in PEL-like lymphomas.12 In this study conducted in Japan6, the 2-year overall survival was 84.7% (95% confidence interval, 71.3-92.1) compared to 1-year overall survival of 30% in PEL. Moreover, there are some cases that were successfully treated only with drainage of the effusion.13 Its survival might be conditioned by the advanced age at diagnosis and patients’ comorbidity.14