We report on a primary mediastinal large B-cell lymphoma with aberrant expression of beta-human chorionic gonadotropin (beta-hCG). The patient, a 33-year-old man, had cough, dyspnea, fever, superior vena cava syndrome, and a mediastinal bulky tumor. A biopsy showed that the latter was characterized by large cells, sclerosis, and compartmentalization. The neoplastic elements expressed CD45, CD20, CD79a and, partially, CD30, whereas they were negative for CD3, epithelial membrane antigen and cytokeratins. Surprisingly, they displayed a clear-cut positivity for beta-hCG. The remaining oncofetal markers applied (PLAP and alpha1-fetoprotein) were negative. Electron microscopy demonstrated the presence of numerous nuclear pockets and the lack of intercellular junctions. DNA analysis by polymerase chain reaction showed clonal rearrangement of Ig heavy-chain genes. The patient responded promptly to the administration of MACOP-B. To the best of our knowledge, this is the first example of B-cell lymphoma showing positivity for beta-hCG; a similar aberrant expression was previously observed only in three Japanese patients with human T-cell lymphotropic virus type I+ adult T-cell lymphoma/leukemia. Because primary mediastinal large B-cell lymphoma has in the past been frequently confused with germ cell tumors, pathologists should be aware of possible beta-hCG expression by lymphomatous cells to avoid the risk of misdiagnosis.
We describe a patient who was referred to us with the diagnosis of pancreatic cancer but who had eosinophilic fibrosis of the pancreas and other organs, including the sub-mandibular salivary glands, retro-orbital tissue, liver, kidneys, and surrounding the abdominal aorta. He had no pain or other symptoms. After treatment with methylprednisolone, all lesions disappeared and now he seems to be cured of this apparently immune-mediated disorder. To our knowledge, involvement of all these particular organs by eosinophilic fibrosis in the absence of symptoms has never before been described.Eur J Gastroenterol Hepatol12:1037-1039
Background The presence of ulceration has been recognized as an adverse prognostic factor in primary cutaneous melanoma (PCM). Objectives To investigate whether the extent of ulceration (EoU) predicts relapse-free survival (RFS) and overall survival (OS) in PCM. Materials and methods We retrieved data for 477 patients with ulcerated PCM from databases of the Italian Melanoma Intergroup. Univariate and multivariable Cox proportional hazard models were used to assess the independent prognostic impact of EoU. Results A significant interaction emerged between Breslow thickness (BT) and EoU, considering both RFS (P < 0·0001) and OS (P = 0·0006). At multivariable analysis, a significant negative impact of EoU on RFS [hazard ratio (HR) (1-mm increase) 1·26, 95% confidence interval (CI) 1·08–1·48, P = 0·0047] and OS [HR (1-mm increase) 1·25, 95% CI 1·05–1·48, P = 0·0120] was found in patients with BT ≤ 2 mm, after adjusting for BT, age, tumour-infiltrating lymphocytes, sentinel lymph node status and mitotic rate. No impact of EoU was found in patients with 2·01–4 mm and > 4 mm BT. Conclusions This study demonstrates that EoU has an independent prognostic impact in PCM and should be recorded as a required element in pathology reports.
Hairy cell leukaemia-variant (HCL-v) is an indolent mature B-cell neoplasm, distinct from HCL and characterized by shorter survival and poorer response to purine analogues (Tiacci et al, 2006). Both HCL and HCL-v are CD11c+/CD103+, but CD25 is usually expressed only in HCL. In addition, the BRAF-V600E mutation, as well as annexin-A1 (ANXA1) and phospho-ERK (pERK) expression, are typically present in HCL and absent in HCL-v (Tiacci et al, 2011, 2012, 2013). Interestingly, some HCL and HCL-v cases carry a specific immunoglobulin (Ig) heavy chain variable gene rearrangement (IGHV4-34) with no or little somatic hypermutation, and show a lower complete response rate and progression-free survival after cladribine than corresponding HCL and HCL-v patients carrying other rearrangements (Arons et al, 2009). The development of large B-cell lymphoma after HCL-v has been reported in only four patients, all of whom subsequently died within a year (Matutes et al, 2001; Razaq et al, 2006). These cases, occurring 4–14 years after HCL-v diagnosis, were only briefly described clinically, with little pathological and no genetic characterization, leaving uncertainty as to whether they represented HCL-v transformation or the development of a clonally independent (e.g., therapy-related) high-grade lymphoma. The present study clarifies this issue for the first time, in a case extensively analysed with several immunophenotyping and genetic studies (Data S1 and Table SI), and provide the first evidence that the poorer prognosis linked to a germline or lowly mutated IGHV4-34 rearrangement in HCL-v can also extend to disease transformation. In April 2006, a 67-year-old male presented with fatigue, weight loss, splenomegaly (20 cm) without lymphadenopathy and thrombocytopenia (30 × 109/l). Although his white blood cell (WBC) count was normal, 16% of WBCs displayed an atypical hairy morphology and abundant cytoplasm. Flow cytometry showed 40% of B cells CD11c+/CD103+/CD25-/CD5-/CD23-/CD10-. A bone marrow (BM) biopsy revealed a mostly intra-sinusoidal infiltration by lymphoid cells with wide cytoplasm and hairy projections, comprising 25% of the marrow cellularity and being CD20+/ANXA1-/CD5-/CD23-/BCL6-/CD10-/cyclinD1-/IRF4- (Fig 1A–B). A diagnosis of HCL-v was made, and chemotherapy with cladribine (5 mg/m2/day for 5 days) delivered. This resulted in splenomegaly regression and clearance of the BM leukaemic infiltrate (down to 2%), but only minor platelet count improvement (up to 69 × 109/l). This response lasted only until March 2007 when the patient relapsed with constitutional symptoms, splenomegaly (21 cm) and BM infiltration (20%) by HCL-v. The patient received 4 weekly doses of rituximab, without response. A splenectomy, performed in September 2007, confirmed the diagnosis of HCL-v, showing an atrophic white pulp and an expanded red pulp infiltrated by lymphoid cells with wide cytoplasm and small nucleolus (Fig 1C) that were CD20+/CD103+/CD5-/CD23-/CD10-/IRF4-/BCL6-/CD25-/pERK-/ANXA1- (Fig 1C/inset-1D–1E). CyclinD1 was weakly expressed in a minor fraction of tumour cells and the proliferative index was low (~5%, Fig 1F). As expected, BRAF-V600E was not detected in the spleen DNA by allele-specific polymerase chain reaction. HCL-v can harbour MAP2K1 activating mutations (Waterfall et al, 2014), which however were not detected by Sanger sequencing of the spleen DNA, in keeping with the lack of pERK (the target of MAP2K1/MEK1 kinase) on immunohistochemistry. Conversely, Sanger sequencing identified a clonal heterozygous hot-spot missense mutation (S34F) of U2AF1 (Fig S1). U2AF1 encodes U2 small nuclear RNA auxiliary factor 1, a component of the RNA splicing machinery, and is recurrently mutated in myelodysplastic syndromes and other myeloid neoplasms (like other splicing factor genes) (Je et al, 2013), as well as, at lower frequency, in lung adenocarcinomas (Imielinski et al, 2012). More recently, the U2AF1 S34F mutation was observed in 3/24 patients with HCL-v (Waterfall et al, 2014), the only lymphoid malignancy where this mutation has been reported so far. Finally, by Sanger sequencing, we detected in the splenic sample a clonal productive IGHV4-34 rearrangement with low mutation load (0·94%, 2/213 mutated nucleotides). After splenectomy, the patient gained weight and platelets rose to 275 × 109/l. However, in August 2013, he presented again with constitutional symptoms, thrombocytopenia (49 × 109/l) and leukaemic lymphocytosis (WBC 14·05 × 109/l, lymphocytes 70%). At this time, however, he also had fever, hepatomegaly and a generalized lymphadenopathy with a high standardized-uptake value (SUVmax 20) on positron emission tomography–computerized tomography (PET-CT). A BM biopsy documented 80% infiltration by typical HCL-v. However, a node biopsy showed a complete effacement of the normal architecture by medium-to-large lymphoma cells with prominent nucleolus (Fig 2A) that were CD20+/CD103+/CD5-/CD23-/CD10-/CD25-/IRF4-/CD10-/ANXA1- (Fig 2B–C). These cells differed from the BM and splenic HCL-v cells not only by the larger nucleolus and cell size, but also for their strong BCL6 and cyclinD1 expression (Fig 2D–E) in the absence of CCND1 gene alterations by fluorescence in situ hybridization. Furthermore, the proliferative index was much higher in lymphoma cells (up to 60%) than in HCL-v cells (~5%; Figs 1F and 2F). The lymph node DNA carried the same clonal heterozygous U2AF1 point mutation as the spleen DNA (Fig S1). Furthermore, DNA fragment length analysis and sequencing of the immunoglobulin genes documented identically sized kappa-light chain and heavy-chain rearrangements (Fig S1), and an identical sequence of the clonal IGHV4-34+ rearrangement in the spleen and lymph node. Thus, the lymphoma was genetically related to the relapsed HCL; yet, it diverged clinically with nodal involvement and fever, and phenotypically with the acquisition of strong expression of BCL6 and CCND1, two oncogenes whose deregulated expression plays key roles in de novo diffuse large B-cell and mantle cell lymphomagenesis respectively. No clonal TP53 mutations were observed in the lymphoma by massively parallel sequencing, ruling out that a TP53 mutation, potentially present as rare subclonal event in HCL-v, expanded under the selection pressure of the subsequent chemotherapy and contributed to lymphoma transformation. After one cycle of R-CVP (rituximab/cyclophosphamide/vincristine/prednisolone) and 5 cycles of R-COMP (R-CVP plus liposomal doxorubicin), the leukaemic lymphocytosis disappeared, a BM biopsy was negative and PET-CT documented the resolution of splenomegaly and lymphadenopathies. Complete remission was confirmed at two subsequent CT scans, 4 and 9 months later. The patient was alive and well in remission at the last follow-up, 16 months after transformation. This contrasts with the four high-grade lymphomas reported so far after HCL-v, all leading to death within a year and none characterized for its clonal relationship with the preceding indolent leukaemia. Supported by grants from Associazione Italiana per la Ricerca sul Cancro (AIRC IG-14447 to E.T.), Hairy Cell Leukaemia Foundation (to B.F. and E.T.), Ministero dell'Istruzione, Università e Ricerca (MIUR; Futuro in Ricerca 2010-RBFR10WT2K to E.T.; PRIN 20104HBZ8E to B.F), and the Associazione Umbra contro le Leucemie e i Linfomi (AULL, to B.F.). E.T. is a Scholar in Clinical Research of the Leukaemia and Lymphoma Society (Contract no. 2030-14). MZ and MR selected the case, analysed and interpreted the data generated from the corresponding biopsy samples, and wrote the manuscript. RV contributed to writing the manuscript. SP, EF and RL performed the laboratory work to generate the data. MIADC provided the clinical data of the case. GFO contributed to generate the immunohistological data. SA analysed and interpreted the data, and contributed to writing the manuscript. ET and BF analysed and interpreted the data, wrote the manuscript and coordinated the work. All the authors have approved the final draft. The authors wish to thank Dr. Gianluca Schiavoni, Dr. Alessandra Venanzi, Barbara Bigerna, Alessandra Pucciarini, Roberta Pacini and Alessia Tabarrini at the Institute of Haematology in Perugia (Italy) for technical assistance; and Dr. Monica Pessino at University of California, Santa Barbara, (USA) and Veronica Pessino PhD candidate in Biophysic at University of California San Francisco (USA), for help in the linguistic revision of the manuscript. The authors report no potential conflict of interest. Data S1. Materials and methods. Fig S1. Clonal relatedness of the HCL-v with the subsequent aggressive lymphoma. Table SI. Immunohistochemical and genetic analyses. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.
