This is the authors' abstract. We don't add key points for this paper.
Primary effusion lymphoma (PEL) is a human herpesvirus 8 (HHV-8)-associated malignancy that presents in the serous body cavities as lymphomatous effusions, usually without extracavitary tumour masses (Carbone & Gloghini, 2008). Most cases occur in human immunodeficiency virus type-1 (HIV)-infected patients, but PEL can also develop in HIV-seronegative individuals following prolonged immunosuppressive therapy and/or in association to age-related immunodepression (Carbone & Gloghini, 2008). Other agents, including hepatitis C virus (HCV), may be associated to PEL and HHV-8 negative cases have been also described as HHV-8-unrelated PEL-like lymphoma (Paner et al, 2003; Kobayashi et al, 2007; Carbone & Gloghini, 2008). This lymphoma arises from mature post-germinal centre B-cells which, despite having a pre-plasma cell genotype and gene expression signature (Jenner et al, 2003; Carbone & Gloghini, 2008), rarely express surface immunoglobulins (Igs) due to the defective expression of some B-cell-specific transcription factors (Arguello et al, 2003; Di Bartolo et al, 2009). By applying highly sensitive techniques, however, light chain mRNA was frequently demonstrated and a minority (10%) of tumour cells was found to display low levels of cytoplasmic monoclonal light chains (Wakely et al, 2002; Di Bartolo et al, 2009). The prognosis of PEL is poor with a median survival of 6 months also due to the lack of specific prognostic factors and tumour markers to optimize therapy (Nador et al, 1996; Boulanger et al, 2005). This report shows that elevated levels of clonally-restricted serum free light chain (sFLC) can be found in some PEL cases and that variations in such levels may reflect therapy-related changes in the tumour cell mass, so representing a possible marker for treatment monitoring/modulation. The sFLC assay specifically quantifies circulating light chains unbound to heavy chains into intact Ig molecules and allows disease monitoring in non-secretory/oligo-secretory (NS-OS) multiple myeloma and other situations where, at the tumour plasma cells level, the compromised heavy chains production leads to unbalanced, excess production of clonal light chains or to expression of light chains only (Pratt, 2008). Patients’ characteristics are shown in Table I. Patient 1 had HCV-related cirrhosis and a previous left nephrectomy due to kidney carcinoma. He presented with bulky ascites and computed tomography disclosed a left pleural effusion and a diffuse thickening of the peritoneal membrane with mesenteric nodulations. Peritoneal fluid was removed and a laparoscopic peritoneal biopsy performed. Morphology (Fig 1A, B), immunophenotypic (Fig 1C, D) and molecular studies (Fig 1E) were consistent with HHV-8-unrelated PEL-like lymphoma. The patient was given bortezomib, cyclophosphamide and steroids, which resulted, in complete effusions control after 3 weeks. Abdominal ultrasounds confirmed the clinical response but, 2 weeks after the second course of treatment, effusions recurred leading to death with progressive lymphoma. Patient 2 had been receiving continuous immunosuppression for 9 years after kidney transplantation. He presented with fever, right pleural effusion and bulky ascites. Tumour cell features were consistent with HHV-8+ PEL. The patient refused chemotherapy and was given oral thalidomide (200 mg/d), which resulted in ascites and pleural effusion control lasting 3 months. Due to neurotoxicity, thalidomide was reduced to 100 mg/d for 1 month and then withdrawn. Bulky effusions recurred, leading to death despite further treatment. Patient 3 had HCV-related cirrhosis, a severe cardiomyopathy and presented with pleural, pericardial and abdominal effusions containing HHV-8+ PEL cells (Table I, Fig 1E). The patient received two courses of cyclophosphamide, vincristine and prednisone leading to substantial control of effusions but bulky disease recurred shortly after the third course of treatment, leading to death 1 month later. (A–D) Patient 1: (A) Morphological appearance of tumour cells from ascitic fluid (Papanicolaou stain of cytocentrifuge preparation, original magnification ×60); (B) Haematoxylin and eosin stain of laparoscopic peritoneal biopsy, original magnification ×25; positive immunohistochemical stains of tumour cells from the peritoneal biopsy for CD138 (original magnification ×25) (C) and cytoplasmic lambda chains (original magnification ×40) (D). (E) PCR demonstration of HHV-8 and EBV genomes in DNA extracted from ascitic fluid tumour cells of Patients 1 (HHV-8−, EBV+) and 3 (HHV-8+, EBV−). MW, molecular weight (100 bp) DNA markers. As positive controls (+ve CNT) for amplification, DNA from the PEL cell line BC-1 (HHV8+) and the lymphoma cell line DoHH2 (EBV+), was used. DNA integrity was controlled by amplification of the human albumin reference gene (ALB). (F) Changes in serum free lambda light chain (sFLC) levels in PEL and PEL-like lymphoma patients before treatment (diagnosis), at maximal treatment response (Patient 1: after the first course of bortezomib/cyclophosphamide; Patient 2: after 2·5 months of continuous oral thalidomide at 200 mg/d; Patient 3: after two courses of cyclophosphamide plus vincristine) and at disease progression. NR, upper limit of normal range for lambda sFLC levels (i.e. 26·30 mg/l). (G) Changes in sFLC κ/λ ratio in PEL and PEL-like lymphoma patients at diagnosis, response and disease progression. Normal reference range: 0·26–1·65. The free κ and λ concentrations (normal ranges, κ: 3·3–19·4 mg/l; λ: 5·71–26·3 mg/l) were assayed by immunonephelometry (Freelite; The Binding Site, Ltd., Birmingham, UK) on cryopreserved (−80°C) serum samples and the FLC κ/λ ratio was calculated (free κ concentration divided by free λ concentration, reference range 0·26–1·65). All of the patients displayed an abnormal baseline sFLC κ/λ ratio (ranging from 0·04 to 0·15) due to a λ chain clonal involvement, with absolute free λ levels ranging from 143·5 to 790 mg/l (Table I). Cytoplasmic λ chains were documented in tumour cells in Patients 1 and 2, while imunohistochemistry was negative in the remaining case (Table I, Fig 1D). Most interestingly, serial measurements during the course of disease showed changes in sFLC levels that correlated with clinical progress. At the time of maximal response to treatment, levels of free λ chains displayed a higher than 75% drop from baseline levels in all patients followed by a rise at tumour progression (Fig 1F). This was paralleled by corresponding modifications of the sFLC κ/λ ratio (Fig 1G). Interestingly, while the last sample from Patient 1 was obtained at recurrence, free λ levels were increased at least 3 weeks before clinical evidence of progression in Patients 2 and 3. While comorbidity- and age-related renal impairment might have increased sFLC half-life, facilitating their straightforward detection in our cases (Pratt, 2008), the results indicate that elevated levels of clonally-restricted sFLC can be found in some PEL patients and that variation in such levels may parallel treatment-related changes of the tumour cell mass. Primary effusion lymphoma cells from HIV+ patients express intact Igs in only 20% of cases (Nador et al, 1996) but monoclonal light chains have been more frequently detected in tumour cells from HIV−PEL and PEL-like lymphoma cases (Kobayashi et al, 2007). The hampered Ig production is partly due to the lack of specific transcription factors (i.e. Oct-2, OCA-B), whose ectopic expression in PEL cells restores Ig heavy chain reporter activity and increases endogenous Ig heavy chain transcription (Di Bartolo et al, 2009). Also, in some PEL cases, the dysregulation in heavy and light chain production could be so extreme that only free light chains are produced at detectable levels. This situation, although involving different cellular mechanisms, is functionally reminiscent of NS-OS myeloma and of the ‘light chain escape’ phenomenon occurring in some myeloma patients (Pratt, 2008). Whilst needing confirmation in a larger series, including HIV+ cases, our findings suggest that sFLC testing could prove useful for therapeutic monitoring of some PEL patients and early evaluation of investigational agents. Dr Barbara Amoroso is medical director at The Binding Site Ltd, Rome, Italy. All of the other authors declare no competing financial interests. This work was supported in part by Ministero della Salute, Ricerca Finalizzata FSN, IRCCS, Rome, Italy. We thank Dr Alessandro Marchei for help in correlative studies.
Answers are grounded in this paper and cite their sources.
No takes yet. Share an insight, caveat, or question.
Filippi et al. (2009) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: