V Koljonen E Tukiainen C Haglund 1 T Bo ¨hling 2 Department of Plastic Surgery and 1 Gastroenterological and General Surgery, Helsinki University Hospital, and 2 Department of Pathology, Helsinki University and HUSLAB, Helsinki, Finland 1. Brissett AE, Olsen KD, Kasperbauer JL et al. Merkel cell carcinoma of the head and neck: a retrospective case series. Head Neck 2002; 24; 982–988. 2. Moll I, Zieger W, Schmelz M. Proliferative Merkel cells were not detected in human skin. Arch. Dermatol. Res. 1996; 288; 184– 187. 3. Llombart B, Monteagudo C, Lopez-Guerrero JA et al. Clinicopatho- logical and immunohistochemical analysis of 20 cases of Merkel cell carcinoma in search of prognostic markers. Histopathology 2005; 46; 622–634. 4. Fernandez-Figueras MT, Puig L, Musulen E et al. Prognostic significance of p27Kip1, 45Skp2 and Ki67 expression profiles in Merkel cell carcinoma, extracutaneous small cell carcinoma, and cutaneous squamous cell carcinoma. Histopathology 2005; 46; 614–621. 5. Parrado C, Bjornhagen V, Eusebi V et al. Prognosticating tools in primary neuroendocrine (Merkel-cell) carcinomas of the skin: histopathological subdivision, DNA cytometry, cell proliferation analyses (Ki-67-immunoreactivity) and NCAM immunohisto- chemistry. A clinicopathological study in 25 patients. Pathol. Res. Pract. 1998; 194; 11–23. 6. Erickson LA, Papotti M, Volante M, Jin L, Lewis JE, Lloyd RV. Merkel cell carcinomas: expression of S-phase kinase-associated protein 2 (Skp2), p27, and proliferation markers. Endocr. Pathol. 2003; 14; 221–229. 7. Carson HJ, Reddy V, Taxy JB. Proliferation markers and prognosis in Merkel cell carcinoma. J. Cutan. Pathol. 1998; 25; 16–19. 8. Fernandez-Figueras MT, Puig L, Musulen E et al. Prognostic significance of p27, p45 and Ki67 expression profiles in Merkel cell carcinoma, extracutaneous small cell carcinoma, and cutane- ous squamous cell carcinoma. Histopathology 2005; 46; 614–621. Epstein–Barr virus-induced B-cell proliferation of Hodgkin’s and Reed–Sternberg cell pheno- and genotype may develop in peripheral T-cell lymphomas DOI: 10.1111/j.1365-2559.2006.02490.x Sir: Peripheral T-cell lymphoma (PTCL) can display not only a morphological, but also an immunophenotyp- ical overlap with classical Hodgkin’s lymphoma (cHL), as PTCL may contain CD30+ and CD15+ as well as T-marker-positive Hodgkin, Reed–Sternberg (HRS) cells of cHL type in a mixed inflammatory background. 1 The distinction between PTCL and cHL is further complica- ted by the description of three cases as a rare variant of PTCL with scattered HRS cells being constantly Epstein–Barr virus (EBV)+ and CD30+, negative for T-markers, variably expressing CD15 as well as B-cell markers, but lacking monoclonality. 2 Here we describe three cases of PTCL with similar appearance and phenotype, but with different results of single-cell molecular analyses. In all three lesions diffuse infiltrates composed of small to medium-sized cells with slight to definite nuclear atypia admixed with variable amounts of inflammatory cells without epithelioid cell formation or an arborizing high endothelial venule network were the dominant feature on histology. Throughout the infiltrate scattered mono- and multinucleated giant cells reminiscent of either HRS cell (cases 1 and 3) or rather immunoblast ⁄ centroblast morphology (case 2) could be observed (Figure 1a–c). The diffusely infiltra- ting, small to medium-sized atypical cells with a medium to high proliferation index proved to be positive for at least three T-cell markers (Table 1; Figure 1d–f). Furthermore, using whole tissue DNA extract, monoclonal rearrangement of the T-cell recep- tor gamma (TCRc) gene in all three cases as well as that of immunoglobulin heavy chain (IgH) gene in case 3 could be found. The HRS-like cells exhibited constant EBV association (EBER+, LMP-1+), but two different phenotypes (Table 2; Figure 1g–i,n). In one tumour, the HRS cells exhibited the CD15–, CD20+, CD30+, CD45+ phenotype and a polyclonal IgH gene re- arrangement in a flow-sorted, EBER+ nuclear prepar- ation isolated from the paraffin block. In the other two lesions, however, the HRS-type cells exhibited a CD15– ⁄ +, CD20– ⁄ +, CD30+, CD45– phenotype, lacked any light chain mRNA expression and showed a clonal IgH gene rearrangement (Table 3; Figures 1o–p and 2). The B-cell-specific transcription factor OCT.2 and co- activator OBF.1 ⁄ BOB.1 as well as Bcl-6 and CD138 expression could be analysed in one of the latter two cases (case 3), obtaining positivity for the first marker and negativity for the last three ones (Figure 1j–m). In the same case the IgH gene FRI–CDRIII sequence proved to be functional, displaying 96.2% (FRI–CDIII) and 93.3% (CDRI+CDRII) identity with the IgHV4- 59*01 germ-line gene and no ongoing somatic hypermutation. The R ⁄ S value in the FR regions was 2.0. These data altogether are exclusive for the existence of a variant PTCL with CD15+ and CD30+ HRS-like T cells and identifies these lesions as PTCLs with scattered, EBV+ HRS-like cells either with activated B-cell or cHL-HRS cell phenotype and B-cell genotype. Correspondence 553 Ó 2006 The Authors. Journal compilation Ó 2006 Blackwell Publishing Ltd, Histopathology, 49, 538–558.