Familial haemophagocytic lymphohistiocytosis in twin infants

Pathology (January 2013) 45(1), pp. 80–99

CORRESPONDENCE Obtaining expert opinions in diagnostically difficult cases Sir, Klebe et al.1 describe their experience with expert review of ‘difficult’ mesothelial proliferations. They found only 57% complete concordance between the ‘expert’ and referring pathologists’ diagnosis, with some 9.5% of cases having ‘significant’ disagreement (benign versus malignant, or mesothelioma versus another tumour type). Whilst disturbing, this will come as no surprise to those of us with subspecialty interests in other rare tumour types such as sarcoma or GIST, and no doubt in other organ systems as well. These figures are in line with similar studies performed overseas; for example, in their review of 349 soft tissue specimens, Thway and Fisher2 reported minor diagnostic discrepancy in 15.7% and major discrepancy in 10.9%, including a difference of ‘benign versus malignant’ in 5%. Similarly, in a review of 266 referred soft tissue lesions for which a primary diagnosis had been offered by the referring pathologist, there were ‘major’ discrepancies (e.g., benign versus malignant, nonmesenchymal tumour) in 25% of cases, and ‘minor’ discrepancies in 7%.3 Numerous other such studies from a variety of countries support the value of timely expert review of rare and difficult cancers,4–7 with diagnostic discrepancies reportedly as high as 45% or more in some series.6,7 This is more than simply an intellectual exercise: delayed or incorrect diagnosis can lead to profound impacts on the patient who undergoes inappropriate or unnecessary surgery, chemoor radiotherapy, or who is denied potentially life-saving therapy. A review of 1996 musculoskeletal tumours in one specialist unit8 found a diagnostic ‘error’ in 87 cases, 54 of which (2.7%) resulted in a significant change to the patient’s management. In the case of aggressive cancers, even relatively short delays in accurate diagnosis can impact on patient survival.9 Even if the correct diagnosis is reached, errors in grading or risk stratification can also impact on clinical decision-making, e.g., neoadjuvant radiotherapy versus ‘up-front’ surgery in various types of soft tissue sarcoma; adjuvant imatinib versus observation in GIST; ‘wait and see’ versus treatment, or choice of treatment modality, in gastroenteropancreatic neuroendocrine tumour (GEP-NET). It is clear, therefore, that timely expert review of difficult and rare entities is in the best interests of the patient; however, for many pathologists working in non-specialist centres, there are difficulties in obtaining such review. Not only does the referring laboratory incur a cost in terms of packaging and sending the slides (and hopefully a paraffin block or two!) to the ‘expert’, but the receiving laboratory incurs a significant cost in terms of handling the incoming material, making the diagnosis and issuing an opinion (which is often complex). This brings its own rewards in that the cases are often interesting, and expertise is only accrued through exposure to difficult and unusual cases, but the cost must be borne somewhere. Many of us choose not to charge a fee for this work, but this is only sustainable if the number of referrals is relatively small (and with the indulgence of our heads of department), and some institutions insist on a fee of perhaps several hundred dollars – a cost which most referring laboratories cannot meet. This Print ISSN 0031-3025/Online ISSN 1465-3931

#

means that the fee is most often passed on to the patient, an added stress at the very time that they are facing a potentially devastating diagnosis. Surely our patients’ access to accurate diagnosis and grading, and subsequent management, should be independent of their personal wealth? There is clearly a need for timely and appropriate referrals of difficult cases for expert review, and this should be funded in an equitable fashion. The relatively small cost of these referrals would be more than compensated by avoiding unnecessary and inappropriate treatment based on an incorrect diagnosis or tumour grade; for example, only one incorrect diagnosis of high risk GIST will cost in excess of AU$45 000 per annum10 in drug costs alone, for adjuvant imatinib which will not benefit the patient. Therefore, funding of expert opinion referrals would not only be good medicine, but would be cost effective. The Royal College of Pathologists of Australasia (RCPA) has prepared two applications to the Medical Services Advisory Committee (MSAC) to fund retrieval of tissues and such expert opinions, which would help the situation at least in Australia, and this initiative should be strongly supported by pathologists and clinicians alike. Patients with rare cancers deserve the same expert diagnosis and management that is afforded to those with common tumours, and that can only be achieved through timely referral for expert advice when it is needed. Chris Hemmings ACT Pathology, Canberra, Australian National University, and Australasian Sarcoma Study Group, Canberra, ACT, Australia Contact Dr C. Hemmings. E-mail: [email protected] 1. Klebe S, Greiggs K, Ely M, Henderson D. Is there a need for expert opinion for biopsy diagnosis of difficult cases of malignant mesothelioma? Pathology 2012; 44: 562–3. 2. Thway K, Fisher C. Histopathological diagnostic discrepancies in soft tissue tumours referred to a specialist centre. Sarcoma 2009; May 27: (Epub ahead of publication). 3. Arbiser Z, Folpe A, Weiss S. Consultative (expert) second opinions in soft tissue pathology: analysis of problem-prone diagnostic situations. Am J Clin Pathol 2001; 116: 473–6. 4. Lehnhardt M, Daigeler A, Hauser J, et al. The value of expert second opinion in diagnosis of soft tissue sarcomas. J Surg Oncol 2007; 97: 40–3. 5. Amant F, Moerman P, Cadron I, et al. The diagnostic problem of endometrial stromal sarcoma: report on six cases. Gynecol Oncol 2003; 90: 37–43. 6. Lurkin A, Ducimetiere F, Vince D, et al. Epidemiological evaluation of concordance between initial diagnosis and central pathology review in a comprehensive and prospective series of sarcoma patients in the RhoneAlpes region. BMC Cancer 2010; 10: 150. 7. Sharif M, Hamdani S. Second opinion and discrepancy in the diagnosis of soft tissue lesions at surgical pathology. Indian J Pathol Microbiol 2010; 53: 466–70. 8. Grimer R, Carter S, Spooner D, Sneath R. Diagnosing musculoskeletal tumours. Sarcoma 2001; 5: 89–94. 9. Kim M, Lee S, Cho W, et al. Prognostic effects of doctor-associated diagnostic delays in osteosarcoma. Arch Orthop Trauma Surg 2009; 129: 1421–5. 10. Australian Government, Department of Health and Ageing. Australian Pharmaceutical Benefits Scheme. Cited Oct 2012. www.pbs.gov.au.

DOI: 10.1097/PAT.0b013e32835baec7

2012 Royal College of Pathologists of Australasia

Copyright © Royal College of pathologists of Australasia. Unauthorized reproduction of this article is prohibited.

CORRESPONDENCE

Rapidly fatal post-allogeneic stem cell transplant lymphoproliferative disorder presenting with skin and bone marrow involvement Sir, Post-transplant lymphoproliferative disorders (PTLD) are an uncommon complication of both solid organ and allogeneic blood or bone marrow transplants (BMT). Early PTLD almost invariably results from uncontrolled proliferation of Epstein– Barr virus (EBV) infected donor derived B cells in these immunosuppressed patients who are unable to mount an EBV specific cytotoxic T-cell response. The main risk factors for developing PTLD are the duration and intensity of immunosuppression and recipient EBV seronegativity at the time of transplant. Additional risk factors in recipients of allogeneic BMT include degree of mismatch between the donor and recipient and T-cell depletion of the graft.1,2 The 2008 World Health Organization Classification3 charts the heterogeneous spectrum of PTLD disorders ranging from infectious mononucleosis-type polyclonal proliferations, which generally resolve with reduction in immunosuppression, to monomorphic PTLD which may present as an aggressive lymphoma and require chemotherapy. Patients who have received lung, small bowel or multiple organ transplants are reported to be at the highest risk of PTLD, with rates of approximately 5%.2 By contrast, the rate of PTLD following allogeneic BMT is reported to be lower, with an overall risk of approximately 1%.1 This report describes a patient who developed a highly aggressive PTLD within 2 months of a second allogeneic BMT for relapsed acute myeloid leukaemia (AML). Clinical progression was rapid and the patient died one week after presentation. Such a fulminant presentation is extremely rare with only three similar reported cases.4–6 A 45-year-old EBV IgG positive Caucasian male with a background of Crohn’s disease was diagnosed with AML in August 2007. Despite an adverse cytogenetic profile (monosomy 5, del7q and multiple additional anomalies) a complete morphological and cytogenetic remission was documented following induction chemotherapy with high dose cytarabine and idarubicin. After two cycles of consolidation chemotherapy he underwent a matched sibling peripheral blood stem cell transplant in January 2008 with cyclophosphamide and total body irradiation conditioning. A bone marrow biopsy on day þ60 showed evidence of a cytogenetic relapse which was successfully managed with withdrawal of immunosuppression. Three years post-transplant a routine bone marrow demonstrated a cytogenetic relapse (1/40 metaphases demonstrating the complex karyotype noted at diagnosis), and 1 month later he was in a morphological relapse with 30% bone marrow myeloblasts. Following two cycles of salvage chemotherapy morphological and cytogenetic remission was again documented. The patient then underwent a fully matched (HLA-A, -B, -C, and -DRB1 by molecular typing) unrelated donor BMT with fludarabine, busulphan and thymogloblulin conditioning (total dose thymoglobulin 4.5 mg/kg). The EBV status of the donor was unknown. Post-transplant immunosuppression was with methotrexate and cyclosporin. The patient engrafted on day þ14 and simultaneously developed a rash and deranged liver function tests. A skin biopsy was consistent with graft versus host disease and he was successfully treated with topical

81

hydrocortisone and 40 mg oral prednisone. A rising cytomegalovirus polymerase chain reaction (PCR) titre on day þ35 was effectively treated with oral valganciclovir. The patient was readmitted on day þ71 with a 1 week history of progressive nausea, vomiting and diarrhoea with associated fever with rigors. Multiple blood cultures were negative although stool Clostridium difficile toxin PCR and norovirus enzyme immunoassay were both positive. His IgG level was low at 3.32 g/L and treatment with metronidazole and intravenous immunoglobulin was commenced. His clinical condition deteriorated with the development of abdominal pain, oliguria and severe lactic acidosis with a lactate of 7.7 mmol/L. A computed tomography (CT) scan showed wall thickening of the descending colon along with intra-abdominal lymph nodes measuring up to 3 cm. An exploratory laparotomy and colonoscopy, performed due to concern of possible ischaemic bowel, showed only mild inflammation of the colon. Cytological analysis of peritoneal fluid demonstrated a small number of atypical cells. On day þ76 a skin rash developed and was biopsied. Simultaneously the patient developed pancytopenia with Hb 103 g/L, WCC 2.7
109/L and platelets 22
109/L and a bone marrow biopsy was performed. The patient died from progressive multiorgan failure on day þ78 before the results of these investigations were available. The skin biopsy demonstrated an abnormal perivascular infiltrate of large, morphologically abnormal lymphoid cells with polymorphic nuclei and prominent nucleoli (Fig. 1A). The abnormal cells were positive for CD20/CD79a/EBER/MUM-1/ PAX-5 and Bcl-2 but negative for CD10/CD30/CD138 and Bcl-6 (Fig. 1B). Occasional plasmacytoid lymphoma cells were noted in the peripheral blood. The bone marrow biopsy was hypocellular with an abnormal infiltrate of similar lymphoid cells (Fig. 2A). Flow cytometry of the bone marrow showed an abnormal population of large CD19 and lambda positive cells. Immunohistochemistry of the trephine showed small clusters of large cells with strong immunoreactivity for EBER (Fig. 2B) and relative lambda light chain restriction. Cytogenetic analysis was normal. Whilst the lymphoma cells showed significant pleomorphism, the clinical picture and immunostains were consistent with monomorphic PTLD, most likely a non-germinal centre diffuse large B-cell lymphoma. Only one previous case has been reported of simultaneous multifocal skin and bone marrow involvement in a patient with PTLD. This developed in a 14-month-old EBV seronegative recipient of a combined small bowel and liver transplant from an EBV seropositive donor.4 The patient presented with cutaneous PTLD associated with circulating lymphoma cells 5 months post-transplant and died rapidly from multiorgan failure. Two other cases of similarly fulminant PTLD have been reported. The first occurred in an EBV seronegative recipient of a renal transplant from an EBV seropositive donor.5 The second developed in an EBV seropositive recipient of a non-myeloblative allogeneic BMT for relapsed anaplastic large cell lymphoma.6 The donor had a positive IgM titre for EBV viral capsid antigen (VCA) in the pre-transplant specimen, consistent with recent infection. ln each of these previous cases the aggressive presentation of the disease was attributed either to the recipient being EBV seronegative or to recent EBV infection in the donor. In our case the recipient was EBV seropositive and we presume the development of PTLD was instead related to the intensive immunosuppression as a result of his previous treatments. Although only a relatively low dose of

Copyright © Royal College of pathologists of Australasia. Unauthorized reproduction of this article is prohibited.

82

Pathology (2013), 45(1), January

CORRESPONDENCE

A A

B B Fig. 1 Skin biopsy. (A) Medium power view showing dermal perivascular infiltrate of lymphoma cells (H&E). (B) EBER staining is strongly positive.

Fig. 2 (A) High power view of two lymphoma cells seen in the bone marrow aspirate with deep blue markedly vacuolated cytoplasm (MGG). (B) Immunohistochemistry of the bone marrow trephine shows a hypocellular marrow with an abnormal infiltrate of strongly EBER positive lymphoma cells.

may allow improved risk stratification and hence individualisation of patient immunosuppressive regimens. thymoglobulin was used in his conditioning regimen, the T-cell depletion associated with this therapy is a known risk factor for the development of PTLD.3 It is also possible that the history of Crohn’s disease is relevant, although his inflammatory bowel disease had been inactive with no specific therapy since his initial diagnosis of AML. Patients with inflammatory bowel disease who are treated with thiopurines are known to have an increased risk of lymphoma which is often EBV positive.7 PTLD is a rare complication of allogeneic stem cell transplant.1 The disease is often extranodal at presentation which may result in a delay in diagnosis.8 Historically the prognosis for patients with PTLD has been poor with survival rates of around 30%.9 The cornerstone of treatment for PTLD postsolid organ transplant is withdrawal of immunosuppressive therapy. However, this is rarely useful post-BMT because the defect is delayed EBV-memory cytotoxic T cell recovery, not suppression of their function.10 Treatment with monoclonal antibodies such as rituximab, chemotherapy or ex vivo generated EBV cytotoxic T cells are resulting in improved outcomes.11 Some groups have advocated serial monitoring of EBV viral load in high risk patients to aid early detection and treatment.12 This case demonstrates the unique clinical and pathological features of an early onset highly aggressive disseminated variant of PTLD post-BMT, highlighting the need for early and ongoing vigilance in high risk patients. Further research into the pathogenesis of this devastating complication

Conflicts of interest and sources of funding: The authors state that there are no conflicts of interest to disclose. Robin Gasiorowski* John Gibson{ Geoff Watson§ Judith Trotman{ Stephen Larsen{ ANZAC Research Institute, and {Department of Haematology, Concord Repatriation General Hospital, Concord, zInstitute of Haematology, §Department of Anatomical Pathology, Royal Price Alfred Hospital, Camperdown, NSW, Australia Contact Dr R. Gasiorowski. E-mail: [email protected] 1. Curtis RE, Travis LB, Rowlings PA, et al. Risk of lymphoproliferative disorders after bone marrow transplantation: a multi-institutional study. Blood 1999; 94: 2208–16. 2. Dharnidharka VR, Tejani AH, Ho PL, Harmon WE. Post-transplant lymphoproliferative disorder in the United States: young Caucasian males are at highest risk. Am J Transplant 2002; 2: 993–8. 3. Swerdlow SH, Campo E, Harris NL, et al. WHO Classification of Tumours of Haematopoietic and Lymphoid Tissues. 4th ed. Lyon: IARC Press, 2008. 4. Apichai S, Rogalska A, Tzvetanov I, Asma Z, Benedetti E, Gaitonde S. Multifocal cutaneous and systemic plasmablastic lymphoma in an infant with combined living donor small bowel and liver transplant. Pediatr Transplant 2009; 13: 628–31.

Copyright © Royal College of pathologists of Australasia. Unauthorized reproduction of this article is prohibited.

CORRESPONDENCE

5. Mathur RV, Kudesia G, Suvarna K, McKane W. Fulminant post-transplant lymphoproliferative disorder presenting with lactic acidosis and acute liver failure. Nephrol Dial Transplant 2004; 19: 1918–20. 6. Zamkoff KW, Bergman S, Beaty MW, Buss DH, Pettenati MJ, Hurd DD. Fatal EBV-related post-transplant lymphoproliferative disorder (LPD) after matched related donor nonmyeloablative peripheral blood progenitor cell transplant. Bone Marrow Transplant 2003; 31: 219–22. 7. Beaugerie L, Brousse N, Bouvier AM, et al. Lymphoproliferative disorders in patients receiving thiopurines for inflammatory bowel disease: a prospective observational cohort study. Lancet 2009; 374: 1617–25. 8. Nalesnik MA, Jaffe R, Starzl TE, et al. The pathology of posttransplant lymphoproliferative disorders occurring in the setting of cyclosporine A-prednisone immunosuppression. Am J Pathol 1988; 133: 173–92. 9. Savage P, Waxman J. Post-transplantation lymphoproliferative disease. QJM 1997; 90: 497–503. 10. Gross TG, Steinbuch M, DeFor T, et al. B cell lymphoproliferative disorders following hematopoietic stem cell transplantation: risk factors, treatment and outcome. Bone Marrow Transplant 1999; 23: 251–8. 11. Heslop HE. How I treat EBV lymphoproliferation. Blood 2009; 114: 4002–8. 12. Stevens SJ, Verschuuren EA, Pronk I, et al. Frequent monitoring of EpsteinBarr virus DNA load in unfractionated whole blood is essential for early detection of post transplant lymphoproliferative disease in high-risk patients. Blood 2001; 97: 1165–71.

DOI: 10.1097/PAT.0b013e32835b5de4

Familial haemophagocytic lymphohistiocytosis in twin infants Sir, Haemophagocytic lymphohistiocytosis (HLH) is an uncommon entity comprising a constellation of symptoms and laboratory findings that together form the clinical inflammatory syndrome. HLH encompasses two separate conditions: a primary hereditary form (familial haemophagocytic lymphohistiocytosis, FHL) and a secondary/acquired form (secondary HLH, sHLH).1 sHLH may follow a variety of stimuli including viral, bacterial, and fungal infections, as well as a number of malignant diseases, notably T-cell lymphomas. FHL is generally inherited in an autosomal recessive or x-linked manner, presents during infancy or early childhood, and is uniformly fatal if left untreated. Here we present an interesting case of FHL presenting in a pair of identical twins in the immediate postnatal period. The patients were diamnionic-dichorionic identical twins born at 35 weeks gestation. Twin 1 was born via normal spontaneous vaginal delivery, while twin 2 was born via emergent caesarean section due to placental abruption. The initial post-natal clinical course was uncomplicated, and the patients were discharged home without complication. At 2 months of age both twins were hospitalised for fever with neutropenia, and severe thrombocytopenia. Liver ultrasound showed hepatomegaly. Laboratory studies (Table 1) showed many abnormalities, including low fibrinogen, elevated D-dimers, and prolonged coagulation times, indicating likely disseminated intravascular coagulation. Blood cultures at this time were negative in both twins. Simultaneous bone marrow aspirates were obtained on both twins. While the bone marrow aspirate from twin 2 showed non-specific findings due to markedly haemodilute aspirate smears (i.e., predominantly mature lymphocytes similar to peripheral blood), the bone marrow aspirate from twin 1 showed significantly increased haemophagocytic histiocytes on the smear (Fig. 1) in addition to erythroid hyperplasia and relative decrease in myeloid cells.

83

Megakaryocytes were unremarkable. Additional laboratory studies (Table 2) showed elevated alanine aminotransferase (ALT), elevated aspartate aminotransferase (AST), increased triglycerides and increased soluble interleukin-2 receptor (CD25) in both twins. Studies for NK-cell functional activity were non-contributory due to technical reasons in twin 1 and not decreased significantly in twin 2. Twin 1 met the following criteria for the diagnosis of haemophagocytic lymphohistiocytosis: (1) fever, (2) fibrinogen level less than 1.5 g/L, (3) increased haemophagocytic lymphohistiocytosis on bone marrow aspirate smear, (4) ferritin level more than 500 ng/mL, and (5) elevated soluble interleukin-2 receptor (soluble CD25) level of more than 2400 U/mL. Twin 2 met the following diagnostic criteria for the diagnosis of haemophagocytic lymphohistiocytosis: (1) fever, (2) cytopenias (haemoglobin less than 9 g/dL and platelet count less than 100 k/mL), (3) fibrinogen level less than 1.5 mg%, (4) ferritin level more than 500 ng/mL, and (5) soluble interleukin-2 receptor (soluble CD25) level more than 2400 U/mL. A diagnosis of haemophagocytic lymophohistiocytosis was made on both the twins, based on these findings (see Table 3).2 Due to the fact that these were identical twins, FHL was considered and subsequent perforin-1 (PRF1) gene mutation analysis showed identical mutations, i.e., homozygous for 50delT (L17fsX50), confirming the diagnosis of FHL type 2 in both the twins. Both patients were started on chemotherapy consisting of dexamethasone, etoposide, and cyclosporine according to the HLH-2004 protocol. Twin 2 progressively deteriorated, developing respiratory and hepatic failure, fungaemia, and worsening coagulopathy, with death occurring 8 days after the start of chemotherapy. Twin 1 did well on chemotherapy and was eventually discharged 4 months after admission. FHL is a rare condition, most often presenting in infancy, with a peak age of presentation between 1 and 6 months.1 The reported incidence ranges from 0.12 per 100 000 cases in studies conducted in Sweden3 to 7.5 per 10 000 among hospitalised patients in Turkish studies.4 The high rate of FHL in the Turkish population may be due to the high rate of consanguineous marriage, fitting with the autosomal recessive inheritance patterns seen in most genetically elucidated types of FHL. In addition, as the incidence of FHL varies amongst people of different ethnic backgrounds, so too do the genetic defects that give rise to the HLH phenotype. The mechanism of immune impairment leading to the syndrome is unclear; however, it is clear that the mechanism involves the release of inflammatory cytokines, T-cell and histiocyte activation, and NK-cell impairment. This leads to the defining clinical and laboratory findings, including fever, splenomegaly, multiple cytopenias, hypertriglyceridaemia Table 1

Results of laboratory studies on both twins

Test (normal range)

Twin 1

Twin 2

Haemoglobin (10–15 g/dL) WBC (5–14 k/mL) Absolute neutrophil count (2–7 k/mL) Platelet count (150–400 k/mL) D-dimer (<1100 ng/mL FEU) Fibrinogen (181–456 mg%) Prothrombin time (10.0–11.7 s)

9.2 5.2 1.2 100 3660.00 <50 140

8.9 4.2 1.1 34 3780.00 <50 30.5

FEU, fibrinogen equivalent units, WBC, white blood cell.

Copyright © Royal College of pathologists of Australasia. Unauthorized reproduction of this article is prohibited.

84

Pathology (2013), 45(1), January

CORRESPONDENCE

Fig. 1 May-Gru¨nwald Giemsa; multiple haemophagocytic histocytes.

and/or hypofibrinogenaemia, haemophagocytosis in the bone marrow, spleen, or lymph nodes, low/absent NK-cell activity, hyperferritinaemia, and increased soluble IL-2 receptor (soluble CD25), in the absence of malignancy (Table 3). Five of the eight criteria are necessary for a diagnosis of HLH, though a diagnosis may still be made if there is molecular evidence consistent with HLH.2 Diagnostic work-up includes complete blood count and blood smear analysis, liver function tests, blood chemistry, triglyceride and cholesterol levels, ferritin, and coagulation tests. Bone marrow biopsy and aspirate may be initially negative, and should be repeated over time to see if evidence of haemophagocytosis appears; biopsies from other organs can help demonstrate haemophagocytic histocytes or evidence of chronic persistent hepatitis in more uncommon instances.2 Microscopic features are identical in the familial and acquired forms. The main histological feature is diffuse infiltration by T-lymphocytes and histiocytes. The organs most frequently examined are the bone marrow, spleen, liver, and central nervous system, and so are the most frequently described; however, almost no organ is spared, and the characteristic findings may be seen in almost any organ. Perhaps most frustrating is the fact that the characteristic haemophagocytic histiocytes (showing phagocytosis of nucleated cells as well as

Table 2

red blood cells) may be absent in some organs, or infrequent at best, thus necessitating repeated biopsies for histological confirmation. Histological findings in the liver, which may be the second most common organ biopsied after the bone marrow, shows a pattern of sinusoidal dilatation, congestion and hyperplasia of Kupffer cells. Haemophagocytic histiocytes are not seen in the portal inflammatory infiltrate, but characteristically in the dilated sinusoids.5 The immunophenotype of the phagocytic histiocytes are also unique. They express common macrophage-associated antigens, S100 protein and also may express CD1a (which are more commonly seen in Langerhans cells or interdigitating dendritic cells).6 Five distinct genetic subtypes of FHL have been identified to date. FHL type 1 has been mapped to chromosome 9 (9p21), although the exact protein and gene remain unknown.7 This mutation is believed to account for 10% of all FHL. FHL type 2 has been mapped to the perforin gene located on chromosome 10 (10q21–22).8 Perforin gene mutations cause either a nonfunctional form or markedly reduced/absent form of the perforin protein (PRF1), resulting in decreased granzyme mediated toxicity by NK cells and cytotoxic T-cells. Missense mutations of PRF1 result in a non-functional form of the protein through conformational changes that inhibit proteolytic processing of the protein precursors; nonsense mutations result in absent

Results of additional laboratory studies on both twins

Test (normal range) Triglycerides (33–85 mg/dL) Ferritin (322–468 ng/mL) ALT (2–30 IU/L) AST (16–43 IU/L) Soluble interleukin-2 receptor; CD25 (<970 U/mL) NK-cell activity; NK-cell functional assay (8–170 LU30) Perforin-1 gene mutation analysis

Twin 1

Twin 2

125 11460 437 384 >6500 Non-contributory Homozygous for 50 del T(L17fsX50) mutation

140 11057 435 1704 >6500 15 Homozygous for 50 del T(L17fsX50) mutation

ALT, alanine aminotransferase; AST, aspartate aminotransferase.

Copyright © Royal College of pathologists of Australasia. Unauthorized reproduction of this article is prohibited.

CORRESPONDENCE

Table 3 Revised diagnostic criteria for haemophagocytic lymphohistiocytosis2 The diagnosis of HLH can be established if either 1 or 2 below is fulfilled: 1. A molecular diagnosis consistent with HLH 2. Diagnostic criteria for HLH fulfilled (5/8 criteria below) Initial diagnostic criteria: 1. Fever 2. Splenomegaly 3. Cytopenias (affecting 2 or more of 3 lineages in the peripheral blood) a. Haemoglobin <90 g/L (in infants <4 weeks of age: hemoglobin <100 g/L) b. Platelets <100
109/L c. Neutrophils <1.0
109/L 4. Hypertriglyceridaemia and/or hypofibrinogenaemia: a. Fasting triglycerides 3.0 mmol/L (265 mg/dL) b. Fibrinogen 1.5 g/L (150 mg/dL) 5. Haemophagocytosis in bone marrow, spleen, or lymph nodes 6. No evidence of malignancy New diagnostic criteria: 1. Low or absent NK-cell activity (according to local laboratory reference) 2. Ferritin 500 mg/L 3. Soluble CD25 (soluble IL-2 receptor) 2400 U/mL Additional comments: 1. If haemophagocytic activity is not proven, further search for haemophagocytic activity, including additional biopsies from other organs and serial aspirates over time may be helpful 2. Supportive evidence may be for the diagnosis: a. Spinal fluid pleocytosis (mononuclear cells) and/or elevated spinal fluid protein b. Liver biopsy with findings of chronic persistent hepatitis 3. Other clinical and laboratory findings that may support the diagnosis: cerebromeningeal symptoms, lymphadenopathy, jaundice, oedema, skin rash, hepatic enzyme abnormalities, hypoproteinaemia, hyponatraemia, increased VLDL, decreased HDL HDL, high density lipoprotein; HLH, haemophagocytic lymphohistiocytosis; VLDL, very low density lipoprotein.

protein production.9,10 Further, the type of mutation affects the clinical presentation with missense mutations having a later age of onset than nonsense mutations.11 Mutations in PRF1 account for 20–40% of FHL. While there have been over 50 different PRF1 mutations identified, a subset of mutations have been found to be more prevalent in certain ethnicities. The 1122G!A (W374X) mutation appears to be more prevalent in Turkish cohorts, while the 272C!T (A91 V) mutation appears to dominate in Italian populations.11 The 50delT mutation, also seen in our described patients, is interestingly the single mutation seen in African/African American individuals. FHL type 3 is traced to the hMunc13–4 gene, encoding the Munc13–4 protein, which is involved in vesicle priming.12 The granules containing perforin and granzymes A and B are normal, however priming prior to release is abnormal, resulting in the abnormal FHL phenotype. FHL type 4 results from abnormalities in the Syntaxin 11 (STX11) gene,13 found only in patients of a Turkish/Kurdish background thus far. Syntaxin 11 is also involved in vesicle priming. FHL type 5 has been mapped to chromosome 19p, which encodes the STXBP2 gene encoding Munc18–2 protein (syntaxin binding protein 2).14 This protein is involved in intracellular trafficking and granule exocytosis. Regardless of the mutated protein, or specific type of mutation, patients with FHL tend to have a clinically similar disease. Prior to the use of cytotoxic chemotherapy FHL was uniformly fatal, with patients succumbing to infection and multi-organ failure. With the use of chemotherapeutic agents, including etoposide, anti-thymocyte globulin, and CSA, as well as steroids, the majority of patients were able to show

85

symptomatic improvement; however, a true cure was not obtained without haematopoietic stem cell transplantation. The regimen currently in use, as produced by the 2004 revision of the Histiocyte Society consensus study protocol (HLH2004), uses etoposide, dexamethasone, cyclosporine, and intrathecal methotrexate.4 In conclusion, haemophagocytic lymphohistiocytosis is a rare disease that may present in a phenotypically indistinguishable acquired or familial form. The familial form is a genetically heterogeneous and phenotypically homogeneous disease, of which five genetically distinct subtypes have been described. Diagnosis rests on a combination of clinical and histological features, and chemotherapy along with haematopoietic stem cell transplant is the mainstay of therapy. Conflicts of interest and sources of funding: The authors state that there are no conflicts of interest to disclose. Bryce Higa Milind Velankar Department of Pathology, Loyola University Medical Center, Maywood, IL, USA Contact Dr B. Higa. E-mail: [email protected]

1. Henter JI, Samuelsson-Horne AC, Arico M, et al. Treatment of hemophagocytic lymphohistiocytosis with HLH-94 immunotherapy and bone marrow transplantation. Blood 2002; 100: 2367–73. 2. Henter J, Horne A, Arico M, et al. HLH-2004: Diagnostic and therapeutic guidelines for hemophagocytic lymphohistiocytosis. Pediatr Blood Cancer 2007; 48: 124–31. 3. Henter J, Elinder G, Soder O, Ost A. Incidence in Sweden and clinical features of familial hemophagocytic lymphohistiocytosis. Acta Paediatr Scand 1991; 80: 428–35. 4. Gurgey A, Gogus S, Ozyurek E, Aslan E, et al. Primary hemophagocytic lymphohistiocytosis in Turkish children. Pediatr Hematol Oncol 2003; 20: 367–71. 5. de Kerguenec C, Hillaire S, Molinie V, Gardin C, et al. Hepatic manifestations of hemophagocytic syndrome: a study of 30 cases. Am J Gastroenterol 2001; 96: 852–7. 6. Herlin T, Pallesen G, Kristensen T, Clausen N. Unusual immunophenotype displayed by histiocytes in haemophagocytic lymphohistiocytosis. J Clin Pathol 1987; 40: 1413–7. 7. Ohadi M, Lalloz M, Sham P, et al. Localization of a gene for familial hemophagocytic lymphohistiocytosis at chromosome 9p21.3–22 by homozygosity mapping. Am J Hum Genet 1999; 64: 165–71. 8. Stepp S, Dufourcq-Lagelouse R, Le Deist F, et al. Perforin gene defects in familial hemophagocytic lymphohistiocytosis. Science 1999; 286: 1957–9. 9. Risma KA, Frayer RW, Filipovich AH, Sumegi J. Aberrant maturation of mutant perforin underlies the clinical diversity of hemophagocytic lymphohistiocytosis. J Clin Invest 2006; 116: 182–92. 10. Molleran S, Villanueva J, Sumegi J, et al. Characterization of diverse PRF1 mutations leading to decreased natural killer cell activity in North American Families with hemophagocytic lymphohistiocytosis. J Med Genet 2004; 41: 137–44. 11. Trizzino A, Stadt U, Ueda I, Risma K, et al. Genotype-phenotype study of familial hemophagocytic lymphohistiocytosis due to perforin mutations. J Med Genet 2008; 45: 15–21. 12. Feldmann J, Callebaut I, Raposo G, et al. Munc13-4 is essential for cytolytic granules fusion and is mutated in a form of familial hemophagocytic lymphohistiocytosis (FHL3). Cell 2003; 115: 461–73. 13. Glolam C, Grigoriadou S, Gilmour KC, Gaspar HB. Familial hemophagocyt5ic lymphohistiocytosis: advances in the genetic basis, diagnosis and management. Clin Exp Immunol 2011; 163: 271–83. 14. Cote M, Menager MM, Burgess A, et al. Munc18-2 deficiency causes familial hemophagocytic lymphohistiocytosis type 5 and impairs cytotoxic granule exocytosis in patient NK cells. J Clin Invest 2009; 119: 3765–73.

DOI: 10.1097/PAT.0b013e32835b5db2

Copyright © Royal College of pathologists of Australasia. Unauthorized reproduction of this article is prohibited.

86

Pathology (2013), 45(1), January

CORRESPONDENCE

A novel oncocytoid papillary renal cell carcinoma, type 2, with aberrant cytogenetic abnormalities

Sir, Renal cell carcinoma is classified by the World Health Organization (2004)1 on the basis of histological patterns correlated with cytogenetic, genetic and both familial and sporadic cases. Papillary renal cell carcinoma (PRCC) is usually multifocal, comprises 10–15% of renal cell carcinomas, and is postulated to arise from the intercalated cells of the distal tubule.1–4 Benign oncocytoma accounts for 5–15% of surgically resected renal tumours and is believed to arise from the intercalated cells of collecting ducts. There are case reports of hybrid tumours oncocytoma/chromophobe RCC,2,4,5 but only four cases of combination or hybrid tumour oncocytoma/PRCC have been reported,2,4–6 of which three cases are <5 mm and would qualify as adenomas. The present case is the second case to qualify as a hybrid tumour oncocytoma/PRCC. A 64-year-old man, with a known medical history of autoimmune haemolytic anaemia, hypertension and hepatitis B, presented with an incidental finding of a right lower pole renal mass during an ultrasound scan performed for his hepatitis. A subsequent computed tomography (CT) scan demonstrated a 66
63
47 mm complex renal mass and multiple small cysts scattered in both kidneys. There was no evidence of hydronephrosis or hydroureter. There was no extension of tumour into veins and no lymphadenopathy was seen. The patient had no history of hereditary polycystic kidney disease. A core biopsy of the renal mass showed a PRCC. Due to other medical conditions, a partial nephrectomy was not performed until a year later, and the partial nephrectomy measured 72
50
47 mm and a separate piece of perinephric fat 65
50
45 mm. Part of two simple cysts were seen at the resection margin, and measured 30
23 mm and 32
25 mm each. On sectioning, there was a circumscribed tumour measuring 72
50
46 mm. The tumour was protuberant and had two distinct areas: one area had a tan/fleshy appearance and was multicystic at one edge, and the other area was haemorrhagic and with a red-brown colour and areas of golden-yellow. The tumour was situated 1 mm from the closest resection margin, and showed attenuation, but no invasion through the renal capsule. Microscopically, the tumour showed two distinct morphological features corresponding to the macroscopic appearance. The darker brown/haemorrhagic areas showed papillary tubulopapillary and cystic growth pattern with fibrovascular stalks

A

B

covered by pseudostratified cells with irregular vesicular nuclei, small nucleoli and abundant eosinophilic cytoplasm. Aggregates of foamy macrophages were present within the stalks as well as in cystic spaces, and psammoma bodies within the papillary cores and adjacent desmoplastic stroma (Fig. 1A). There were focal areas of haemorrhage in the neoplasm. These features supported PRCC, type 2, and comprised approximately 50% of the entire renal tumour. The tan/fleshy area showed tubulocystic architecture lined by cuboidal cells with uniform round vesicular nuclei with well defined nucleoli and granular or eosinophilic cytoplasm (Fig. 1B). The intervening stroma was oedematous and myxoid. Microcysts, some filled with red blood cells were present. These features supported oncocytoma and comprised the other 50% of the renal tumour. There were also areas of clear demarcation between the two components: the PRCC and the oncocytoma (Fig. 1C). The PRCC component showed positive staining for CK7, vimentin (Fig. 2A) and CD10 (Fig. 2B), and was negative for CD117 (Fig. 2C). The oncocytoma component showed positive staining for CD117 (Fig. 2C) and negative for CK7, vimentin and CD10. Electron microscopy showed the fleshy part of the tumour to be composed of tubules with a single layer of epithelial cells, central lumen and few scattered microvilli on the epithelial surface. The cytoplasm was filled with enlarged, swollen mitochondria and well-defined cristae, and other organelles and microvesicles were scanty. There was a continuous basal lamina and cell junctions, features which were characteristic of an oncocytoma (Fig. 3A).7 The PRCC component showed the epithelial cells with mitochondria admixed with other cell organelles (Fig. 3B). Fluorescence in situ hybridisation (FISH) with centromeric probes for chromosomes X, Y, 1, 7 and 17 was carried out on formalin fixed, paraffin embedded tumour sections from areas with each type of morphology. Areas with an oncocytoma-like morphology showed gains of chromosome 1 (30% of interphase nuclei) and low level gain of chromosome 7 (8% of nuclei) as well as low level loss of chromosome 17 (32% of nuclei). Areas with papillary morphology showed a similar pattern but with significantly more nuclei showing þ7 (50%) and 17 (58%). No alterations to sex chromosome numbers were seen in either area. To our knowledge, there are only four reported cases of hybrid oncocytoma/PRCC. PRCCs are grossly well-circumscribed, and areas of haemorrhage, necrosis and cystic degeneration are frequently present. There are two morphological types based on the nuclear features and growth pattern. Type 1 is composed of small cells with scanty, pale basophilic cytoplasm arranged in a single layer covering papillary cores, and in Type 2 there are larger papillae with pseudostratification of higher nuclear grade and abundant eosinophilic cytoplasm. In our case the PRCC showed positivity for CD10, vimentin and CK7, and was

C

Fig. 1 (A) The heterogeneous part of the renal tumour shows features of a PRCC with well-formed papillae, interstitial foam cells and plentiful calcified psammoma bodies (H&E). (B) The fleshy tumour area shows the typical features of oncocytoma with rounded nuclei and plentiful eosinophilic cytoplasm. There are also myxoid areas (H&E). (C) The renal tumour shows a well demarcated zone between the PRCC (left) and the oncocytoma (right) (H&E).

Copyright © Royal College of pathologists of Australasia. Unauthorized reproduction of this article is prohibited.

CORRESPONDENCE

CD10

Vimentin

A

B

87

C-kit

C

Fig. 2 Immunohistochemistry. The PRCC (left) shows positive staining for (A) vimentin and (B) CD10. The oncocytoma (right) is negative for these markers. (C) CD117 stain is positive for the oncocytoma and negative for the PRCC.

negative for CD117. The electron microscopy showed mitochondria admixed with other cell organelles and microvesicles. The one component of the renal tumour showed all the immunohistological features of a PRCC, type 2. Oncocytomas are benign well-circumscribed neoplasms which may have a central, stellate scar of hyalinised connective tissue and are mahogany brown. Haemorrhage and cystic degeneration may be present, but tumour necrosis is not a feature. They are arranged in nests, tubules, cords or microcysts and the tumour cells have dense eosinophilic cytoplasm, round nuclei and small nucleolus. Onocytomas are positive for CD117, variable positivity for CD10, and are negative for CK7 and vimentin. Electron microscopy showed a mitochondria rich cytoplasm and scanty other organelles. The second component of the renal tumour in our case supported the features of an oncocytoma.

A 10 µm

B 10 µm Fig. 3 Electron microscopy. (A) The fleshy area of the tumour shows the features of oncocytoma: a tubular structure with plentiful mitochondria and a well formed basal lamina (
6000). (B) The PRCC shows cells with variably sized luminal microvilli resting on a basal lamina. The cytoplasm contains variable numbers of mitochondria and other cell organelles (
6000).

The papillary carcinoma component in the four cases reported as hybrid tumour were small in size and in two were reported to be ‘a small focus’6 and ‘numerous small nests’2 and in the other two were reported to be 1.5 mm4 and 7 mm.5 The oncocytoma component of these tumours ranged in size from 15 mm to 36 mm. As three of the four reported cases were <5 mm they should be classified as papillary adenomas rather than papillary carcinomas. The tumour in our case was 72 mm in maximum dimension, with approximately equal proportions of PRCC and oncocytoma, and is the largest of the hybrid tumours, and qualifies as a hybrid or combined oncocytoma/PRCC, type 2. In the three cases2,4,5 the papillary component was small and was found within the larger oncocytoma. In the case reported by Vasuri and Fellegara,6 the two tumours were found adjacent to each other, whereas our case showed the PRCC and oncocytoma as a mixed tumour, both within and adjacent to each other. Vasuri and Fellegara6 believed these neoplasms were two distinct colliding neoplasms, and Rowsell et al.5 postulated that their case ‘represents divergent differentiation of either oncocytoma into papillary RCC or papillary RCC into oncocytoma’. Al-Saleem et al.2 demonstrated gains of chromosome 7 in the oncocytoma part of their tumour, which was a feature of PRCC, and hypothesised that this was a secondary event leading to the evolution of PRCC within the oncocytoma. The cytogenetic patterns seen in our case were not characteristic of any common renal tumour, particularly not of oncocytoma in which loss of a sex chromosome and loss of chromosome 1 are the common cytogenetic changes, or for PRCC in which trisomy 7 and 17 are the most common abnormalities.1,8,9 The gain of chromosome 1 and loss of chromosome 17 are not characteristic of either renal oncocytoma or PRCC and gain of chromosome 7 is not common in oncocytoma. The close relationship between the cytogenetic abnormalities detected in the two differing morphological areas of tumour strongly suggests a common origin with the increasing proportion of cells with trisomy 7 in the papillary component suggesting tumour progression. The unusual pattern of cytogenetic abnormalities but more consistent histological patterns seen suggests we may be dealing with a novel tumour entity, an oncocytoid PRCC, type 2, with aberrant cytogenetic expression. These reports of mixed tumours with oncocytoma and PRCC components, and also the hybrid morphology between oncocytoma and chromophobe RCC,10 indicate that further studies are needed to show that different tumours have different molecular pathways, and the possible evolution of a subset of oncocytomas to papillary and chromophobe RCC has diagnostic and therapeutic implications.

Copyright © Royal College of pathologists of Australasia. Unauthorized reproduction of this article is prohibited.

88

Pathology (2013), 45(1), January

CORRESPONDENCE

Conflicts of interest and sources of funding: The authors state that there are no conflicts of interest to disclose. Rajalingam Sinniah* Angeline Teo* Ashleigh Murch{ *Department of Anatomical Pathology, PathWest Laboratory Medicine, Royal Perth Hospital, Wellington Street, Perth, and, and {Cytogenetics Department, Path West Laboratory Medicine, King Edward Memorial Hospital, Subiaco, WA, Australia Contact Professor R. Sinniah. E-mail: [email protected] 1. Eble JN, Sauter G, Epstein JI, Sesterhenn IA, editors. World Health Organisation Classification of Tumours. Pathology and Genetics of Tumours of the Urinary System and Male Genital Organs. Lyon: IARC Press, 2004. 2. Al-Saleem T, Balsara BR, Liu Z, et al. Renal oncocytoma with loss of chromosomes Y and 1 evolving to papillary carcinoma in connection with gain of chromosome 7. Coincidence or progression? Cancer Genet Cytogenet 2005; 163: 81–5. 3. Murphy WM, Girgnon DJ, Perlman EJ. Atlas of Tumour Pathology. Tumours of the Kidney, Bladder, and Related Urinary Structures. 4th series, fascicle 1. Washington DC: AFIP, 2004; 164–74. 4. Floyd MS Jr, Javed S, Pradeep KE, De Bolla AR. Composite oncocytoma and papillary renal cell carcinoma of the kidney treated by partial nephrectomy: a case report. Scientific World Journal 2011; 11: 1173–7. 5. Rowsell C, Fleshner N, Marrano P, Squire J, Evans A. Papillary renal cell carcinoma within a renal oncocytoma: case report of an incidental finding of a tumour within a tumour. J Clin Pathol 2007; 60: 426–8. 6. Vasuri F, Fellegara G. Collision renal tumours. Int J Surg Pathol 2009; 17: 338–9. 7. Erlandon RA, Shek TW, Reuter VE. Diagnostic significance of mitochondria in four types of renal epithelial neoplasms: an ultrastructural study of 60 tumours. Ultrastruct Pathol 1997; 21: 409–17. 8. Linehan WM, Walther MM, Zbar B. The genetic basis of cancer of the kidney. J Urol 2003; 170: 2163–72. 9. Yang XJ, Tan MH, Kim HL, et al. A molecular classification of papillary renal cell carcinoma. Cancer Res 2005; 65: 5628–37. 10. Tickoo SK, Reuter VE, Amin MB, et al. Renal oncocytosis, a morphologic study of fourteen cases. Am J Surg Pathol 1999; 23: 1094–101.

DOI: 10.1097/PAT.0b013e32835b682e

internal surfaces, the latter with adherent aggregates of hair. No solid or papillary areas were seen. Microscopic examination showed a multilocular cyst with constituent tissues from the three embryonic layers including adipose tissue, skeletal muscle, bone, cartilage with keratinising squamous epithelium and associated skin adnexal structures, pseudostratified ciliated respiratory type epithelium and ciliated ependymal cells with associated choroid plexus lining the cystic spaces. Focally the choroid plexus tissue formed a well demarcated, non-invasive papillary tumour characterised by complex branching fibrovascular structures lined by mildly crowded, elongated and focally stratified columnar ependymal cells without nuclear atypia or mitotic activity. The features were diagnostic of a choroid plexus papilloma (WHO Grade I) (Fig. 1). The basic components of the choroid plexus are glomerular vascular tufts derived from the vascular leptomeninges that are covered by a cobblestoned secretory ependymal epithelium.1 Small nests of meningothelial cells are native constituents of the choroid plexus and can give origin to psammoma bodies.1 Choroid plexus papillomata are rare primary central nervous system tumours that grossly appear as a circumscribed, cauliflower-like mass within ventricles.2,3 Microscopically, columnar non-ciliated cells that show nuclear crowding and stratification are arranged in an orderly fashion around delicate well formed fibrovascular

A

Choroid plexus papilloma arising in a mature cystic teratoma of a 32-year-old female Sir, Herein, to the best of our knowledge, we report the third published case of a choroid plexus papilloma arising in a mature cystic teratoma. The female patient, aged 32 years, presented with a 6 month history of left loin pain post-partum. She underwent investigation with pelvic ultrasound and computed tomography (CT) of the abdomen revealing a 10 cm complex mass arising from the left ovary with a large cystic component and elements of fat, all consistent with a dermoid cyst. The affected ovary was removed and submitted for pathological examination. Macroscopically the specimen consisted of an opened cyst, 80
30 mm when laid flat, with smooth external and

B Fig. 1 Choroid plexus papilloma in a mature cystic teratoma. (A) Low power view of a circumscribed papillary lesion with delicate, well formed fibrovascular stalks projecting into a cystic space. (B) A high power view shows the fibrovascular stalks are lined by non-ciliated columnar cells that show nuclear crowding and stratification.

Copyright © Royal College of pathologists of Australasia. Unauthorized reproduction of this article is prohibited.

CORRESPONDENCE

stalks.2,3 Calcified bodies are often seen in the stroma of papillomas.3 An atypical papilloma is defined by the presence of mitoses >2 per 10 high power fields (HPF) and two or more of increased cellularity, nuclear pleomorphism, solid growth and areas of necrosis.3 There are only two other reports of choroid plexus papillomata arising in mature cystic teratomas in the literature. In one report a choroid plexus papilloma developed in the wall of an ovarian dermoid in a 14-year-old female.4 In the other, an atypical choroid plexus papilloma with cytologic atypia, focal necrosis and mitoses up to 3/10 HPF, was described arising in the wall of a dermoid in a 26-year-old female.5 Follow-up is limited in both cases. Usual WHO grade I choroid plexus papillomata arising in the central nervous system are considered benign and complete excision in these cases is considered curative. By extrapolation, the oophrectomies performed to remove the teratoma in the current case and the other reported cases are considered curative. Importantly, choroid plexus papillomata can be confused with other ovarian papillary epithelial neoplasms that may mimic primary or metastatic tumours including serous, clear cell or endometroid tumours, especially if associated with psammomatous calcifications.6–8 Metastasis from lung adenocarcinoma, papillary thyroid carcinoma, urothelial carcinoma and malignant mesothelioma would also enter the differential diagnosis.9 In conclusion, we describe another example of the rare phenomenon of a choroid plexus papilloma arising in a mature cystic teratoma of the ovary. Pathologists need to be aware of this entity so as not to confuse this with other well differentiated ovarian papillary epithelial neoplasms that may be primary or metastatic. Conflicts of interest and sources of funding: The authors state that there are no conflicts of interest to disclose. Benjamin F. Dessauvagie*{ Sukeerat Ruba*{ Peter D. Robbins{ *Histopathology Department, PathWest, King Edward Memorial Hospital, Subiaco, and {Division of Tissue Pathology, PathWest, QEII Medical Centre, Nedlands, WA, Australia Contact Dr B. Dessauvagie. E-mail: [email protected] 1. Sternberg S. Histopathology for Pathologists. 2nd ed. Philadelphia: Lippincott Raven, 1997; 271–3. 2. Tena-Suck T, Salinas-Lara C, Rembao-Bojo´rquez D, Castillejos M. Clinicopathologic and immunohistochemical study of choroid plexus tumours: single-institution experience in Mexican population. J Neurooncol 2010; 98: 537–65. 3. Nelsen J, Mena H, Parisi J, Schochet S. Principles and Practice of Neuropathology. 2nd ed. Oxford: Oxford University Press, 2003; 332–4. 4. von Gunten M, Burger H, Vajtai I. Choroid plexus papilloma developing in a dermoid cyst of ovary. Histopathology 2006; 49: 204–5. 5. Quadri A, Ganesan R, Hock Y, Karim S, Hirschowitz L. Malignant transformation in mature cystic teratoma of the ovary: three cases mimicking primary ovarian epithelial tumors. Int J Surg Pathol 2011; 19: 718–23. 6. Terada T. Immature teratoma of ovary composed largely of choroid plexus. Int J Gynecol Cancer 2010; 20: 1101–2.

89

7. Robboy S, Mutter G, Prat J, Bentley R, Russell P, Anderson M. Robboy’s Pathology of the Female Reproductive Tract. 2nd ed. London: Churchill Livingstone, 2009; 749–54. 8. Crum C, Nucci M, Lee K. Diagnostic Gynecologic and Obstetric Pathology. 2nd ed. Philadelphia: Elsevier Saunders, 2011; 905–7. 9. Ikota H, Tanaka Y, Yokoo H, Nakazato Y. Clinicopathological and immunohistochemical study of 20 choroid plexus tumours: their histological diversity and the expression of markers useful for differentiation from metastatic cancer. Brain Tumor Pathol 2011; 28: 215–21.

DOI: 10.1097/PAT.0b013e32835b6855

Metastatic papillary thyroid carcinoma to the kidney: report of two cases mimicking primary renal cell carcinoma and review of the literature Sir, Papillary carcinoma of the thyroid is the most common subtype, accounting for up to 86% of thyroid cancers. Mean age at diagnosis ranges from 31 to 49 years, with a female to male ratio between 2:1 and 3:1. The main pattern of spread is to cervical lymph nodes, with distant metastases occurring uncommonly and having an adverse impact on survival. Distant metastases from papillary carcinoma of the thyroid can occur at any time during the course of the disease: initial presentation of metastatic disease has been reported in 1–12% of differentiated thyroid tumours, being less frequent in papillary (2%) than in follicular (10%) thyroid carcinoma, whereas cumulative incidence, including metastatic disease following initial treatment, varies between 10 and 35%, depending upon the histology, again being least in well-differentiated papillary thyroid carcinoma.1 Increasing age and primary size, male sex, extrathyroidal extension, and histological subtypes, including tall-cell, columnar-cell, diffuse sclerosing and solid variants, have been associated with adverse prognosis.2 When haematogenous spread occurs, it is usually to bone, brain, lungs and soft tissue. Metastasis to the kidney is found in 2.8–3.8% and 6–20% of thyroid papillary and follicular cancer cases, respectively, but clinically detectable differentiated thyroid cancer metastatic to the kidney is exceedingly rare.3 Herein we report two cases of papillary thyroid carcinoma metastatic to the kidney. The literature on metastatic thyroid tumours to the kidney is also reviewed and differential diagnoses are discussed. Case 1 was a 63-year-old man who presented in February 1995 with gastrointestinal symptoms, 10 lb weight loss and dizziness. He had a history of papillary thyroid carcinoma status post-thyroidectomy (1980), radical neck dissection and bilateral lung metastases (1982), iodine-131 treatment (1986), as well as chemotherapy (1991) and radiation therapy (1992). On admission, abdominal computed tomography (CT) scan showed a partially cystic mass in the right abdomen. The patient underwent angio-infarction and subsequent (5 months later) resection of a large right kidney mass eroding into the duodenum and forming a pyelo-duodenal fistula. Grossly, most of the kidney was involved by a partially solid, cystic and extensively necrotic neoplastic process, 13 cm in largest dimension. Microscopically, the lesion showed a papillary architecture and characteristic nuclear changes (Fig. 1).

Copyright © Royal College of pathologists of Australasia. Unauthorized reproduction of this article is prohibited.

90

Pathology (2013), 45(1), January

CORRESPONDENCE

A

B

C

D

Fig. 1 Papillary carcinoma of the thyroid, tall-cell variant, metastatic to the kidney (H&E). (A) Neoplastic papillae are characterised by a central fibrovascular stalk of loose connective tissue and thin-walled vessels, and (B) in some cases show abundant acellular hyaline material. The papillae are lined by cells that often show abundant eosinophylic cytoplasm and are at least twice as tall as large. The nuclei (A, C, D) sometimes conform the elongated cells in which they are contained, and show the characteristic abnormalities of papillary thyroid carcinoma: optical clearing (arrows), indentations, folds and grooves (arrowhead).

Papillae showed a central fibrovascular stalk lined by neoplastic epithelium. The better developed papillae were long with a complex arborising pattern. Some were straight and slender, arranged in a parallel, regimented fashion; others were short and stubby. The papillary stalk was mainly composed by loose connective tissue and variously sized thin-walled vessels; in some cases, it was swollen by oedematous fluid or occupied by an abundant acellular hyaline material. Occasionally, it was infiltrated by lymphocytes or clusters of foamy or haemosiderin-laden macrophages. The neoplastic cells lining the papillae often showed tall cell features; they were at least twice as long (tall) as wide, with large eosinophilic cytoplasm and round to slightly oval nuclei. Nuclear contour appeared smooth on superficial examination, although closer inspection revealed subtle irregularities in the form of indentations, folds, and grooves. Another peculiar feature of neoplastic nuclei was the empty appearance of the nucleoplasm, which seemed almost totally devoid of chromatin strands. These nuclei were similar to the ones previously described as pale, optically clear, watery, empty, ground glass, or ‘Orphan Annie’s eyes’. No psammoma bodies or other concretions were noted. Immunohistochemically, neoplastic epithelial cells showed positive staining for cytokeratin (CK)19, CK7, CD57, thyroglobulin, and thyroid transcription factor-1 (TTF-1) (Fig. 2A–D). Cells did not react with CK20, racemase, HMWCK, p63, CD10, and RCC marker. The histological and immunohistochemical findings supported the diagnosis of metastatic papillary thyroid carcinoma, tall-cell variant. The patient was lost to follow-up. Case 2 was a 56-year-old woman who presented in December 2008 with haematuria and left flank pain due to an enlarging left lower pole renal mass. Her past medical history included metastatic papillary carcinoma of the thyroid, follicular variant to the brain (resected), thyroidectomy, resection of liver metastasis, and metastatic disease to the left arm treated

by external beam radiation therapy. At ultrasound, a 7.9
7.6
6.0 cm heterogeneous mass was identified at the inferior pole of the left kidney. Needle biopsy of the renal mass showed variably-sized neoplastic follicles generally filled with homogeneous eosinophilic colloid. Follicles were lined by cuboidal epithelial cells with overlapping irregular nuclei that focally harboured characteristic changes of papillary thyroid carcinoma, such as grooves and intranuclear pseudoinclusions. No psammoma bodies or true papillations were identified. By immunoperoxidase staining, performed to further evaluate the lesion, the tumour cells showed positive reactivity for CK7, thyroglobulin, TTF-1, and CD57. In view of the previous diagnoses the findings were considered consistent with metastatic papillary thyroid carcinoma, follicular variant. The patient underwent microsphere embolisation of the renal mass and died of disease, 7 years after her first diagnosis. Metastatic disease to the kidney is observed frequently at autopsy, but is rarely found clinically in living patients.3 Although there are no modern autopsy series devoted to kidney metastases, studies performed in the past have reported that renal metastases outnumber primary renal tumours by 4:1, and that up to 12.6% of cancer patients had metastatic disease to the kidney, with frequencies of bilaterality and multiplicity being as high as 71–81%. By contrast, primary renal cell carcinomas are rarely bilateral. Renal metastasis should be suspected whenever there is a known primary, even in cases of unilateral solitary renal masses. Although, theoretically, all solid tumours may give rise to renal metastasis, secondary lesions to the kidney occur more commonly in patients with lung and breast cancer, melanoma, gastric carcinoma and lymphoma. Reports in the literature suggest rates of epithelial (non-lymphoma) renal

Copyright © Royal College of pathologists of Australasia. Unauthorized reproduction of this article is prohibited.

CORRESPONDENCE

CK7

Thyroglobulin

A

CD57

B

C

TTF-1

D

91

Fig. 2 Papillary carcinoma of the thyroid metastatic to the kidney (immunohistochemistry). Tumour cells show strong positivity for (A) CK7, (B) CD57, (C) thyroglobulin and (D) TTF-1.

metastases of 1.5–1.8% of the general population.4 Imaging features are rarely pathognomonic. Metastasis to the kidney of differentiated thyroid cancer is an uncommon event; in an autopsy series of 161 fatal primary malignant thyroid tumours, only four (6%) differentiated thyroid tumours metastasised to the kidney, whereas 35 (54%) and 40 (61%) metastasised to the lungs and bones, respectively.5 In another autopsy study, Silliphant et al. found that six of 44 (14%) differentiated papillary and follicular thyroid cancers metastasised to kidney, compared to 22 (50%) and 17 (26%) cases metastasising to the lungs and bones, respectively.6 On the other hand, thyroid cancer represents only 1.0–2.5% of primary tumours metastasising to the kidneys.3 Detection of thyroid carcinoma with clinically apparent kidney metastases is very rare, with less than 30 cases reported in the literature so far (21 of them referenced in Malhotra et al.).7–12 Clinical and pathological features were only accessible in a subset of cases (n ¼ 28) (Table 1). Of the 28 cases of renal metastases associated with differentiated thyroid carcinoma, 13 were from papillary carcinoma (7 of which were follicular-variant papillary carcinomas), and 15 were from follicular carcinoma. In five cases there was bilateral involvement. In three patients the disease was discovered incidentally during intravenous pyelography or ultrasound. In the vast majority of cases, patients had known thyroid tumours at the time the renal metastases were identified. However, in some instances, metastases to the kidney preceded the knowledge of the primary thyroid neoplasm and were treated surgically as primary renal tumours. Ruggiero et al. reported a case of a 25-year-old woman who underwent radical nephrectomy for a right renal mass.7 The tumour was diagnosed as papillary thyroid carcinoma, follicular variant. The patient had no previous history of thyroid disease. During subsequent evaluation, metastatic disease was also identified in the patient’s lungs. More recently, Gupta et al.10 reported a case of metastatic papillary thyroid carcinoma that presented with flank

pain and haematuria and was treated by radical nephrectomy as a primary renal malignancy. The patient had neither history nor signs and symptoms of thyroid disease. Later work-up of the patient for thyroid disease revealed a nodule of 0.6 cm in the right lobe of the thyroid, which was confirmed as a papillary thyroid carcinoma by ultrasound-guided fine needle aspiration. No other metastatic sites were identified. Herein we describe two cases of thyroid papillary carcinoma, a tall-cell variant and a follicular variant, metastatic to the kidney. The former (Case 1) to our knowledge is the first case of tall-cell variant papillary thyroid carcinoma metastatic to the kidney described to date. Although both patients initially presented with disseminated disease, the renal metastasis presented as a unilateral, large heterogeneous mass located at the lower pole of the kidney. The differential diagnoses for tall-cell variant papillary thyroid carcinoma metastatic to the kidney include primary papillary renal cell carcinoma, particularly type 2, micropapillary urothelial carcinoma of the upper urinary tract, and metastasis of papillary carcinomas from other organs. Papillary renal cell carcinoma comprises approximately 10% of renal cell carcinomas. The male to female ratio ranges between 1.8:1 and 3.8:1. Papillary renal cell carcinoma frequently contains areas of haemorrhage, necrosis and cystic degeneration. A pseudo-capsule may be identified. Bilateral and multifocal tumours are common. Neoplastic cells typically form varying proportions of papillae and tubules. Papillae contain a delicate fibrovascular core and aggregates of foamy macrophages; cholesterol crystals may be present. Calcified concretions are common in papillary cores and adjacent desmoplastic stroma. Papillary renal cell carcinoma has been subclassified into two morphological variants, types 1 and 2. Papillary renal cell carcinoma type 2 is composed of tall cell with eosinophilic cytoplasm and less frequently shows microcalcifications. Neoplastic cells exhibit large and spherical nuclei, prominent nucleoli, and varying degrees of nuclear pseudostratification,

Copyright © Royal College of pathologists of Australasia. Unauthorized reproduction of this article is prohibited.

92

Pathology (2013), 45(1), January

CORRESPONDENCE

Table 1

Clinical and pathological features of renal metastases from differentiated thyroid carcinoma

Reference

Sex/Age

Tumour type

Presentation

Years after detection

Site

Takayasu et al., 19687 Davis et al., 19797 Johnson et al., 19827 Marino et al., 19917 Sardi et al., 199212 Tur et al., 19947 Ro et al., 19957 Graham et al., 19957 Lam et al., 199611 Benchekroun et al., 19997 Garcia-Sanchis et al., 19997 Gamboa-Dominguez et al., 19999 Muller et al., 20007 Smallridge et al., 20017 Smallridge et al., 20017 Abe et al., 20027 Ferrer Garcia et al., 20027 Moudouni et al., 20027 Insabato et al., 20037 Matei et al., 20037 Iwai et al., 20057 Liou et al., 20057 Ruggiero et al., 20057 Kumar et al., 20057 von Falck et al., 20077 Gupta et al., 200810 Malhotra et al., 20107 Borde et al., 20118 Present cases: Case 1 Case 2

F/44 F/49 F/66 F/– M/53 F/72 F/47 M/75 F/91 M/56 F/65 F/50 F/58 F/53 F/61 M/37 F/58 M/62 F/64 F/78 F/76 F/50 F/25 F/66 F/64 M/50 M/30 M/56

Follicular Follicular Follicular Follicular Papillary PTC-FV Follicular PTC-FV Follicular Papillary Follicular PTC-FV Follicular PTC-FV PTC-FV Papillary Follicular Follicular Follicular Follicular Follicular PTC-FV PTC-FV Follicular Follicular Papillary Papillary Papillary

Abdominal mass Incidental finding (IVP) Gross haematuria Neck nodule Haematuria No complaints; liver mass Haematuria Gross haematuria Incidental autopsy finding Low back pain Neck and sternal mass Haematuria and flank pain Dyspnoea Back pain Upper back mass Incidental finding (US) Lumbalgia Left upper abdominal discomfort Incidental finding (US) Microscopic haematuria and flank pain Haematuria and flank pain Low back pain Abdominal/flank pain Scalp mass Constant increase in serum thyroglobulin levels Flank pain and haematuria Low backache radiating to the lower limbs Post radioiodine-131 treatment scanning

3 18 37 23 7 3 7 No No 3 No No 11 No No No 5 9 35 10 13 No No No 20 No 20 No

Bilateral Bilateral Left kidney Right kidney Right kidney Right kidney Right kidney Left kidney Left kidney Left kidney Left kidney Left kidney Bilateral Right kidney Left kidney Left kidney Left kidney Left kidney Right kidney Right kidney Right kidney Right kidney Right kidney Left kidney Left kidney Right kidney Bilateral Bilateral

M/63 F/56

Papillary PTC-FV

Nausea, vomiting, easy satiety Haematuria and flank pain

15 6

previous history* previous history previous history previous history* previous history previous history previous history*

previous history previous history* previous history previous history* previous history

Right kidney Left kidney

*

Treated surgically as primary renal cell tumour. IVP, intravenous pyelogram; PTC-FV, papillary thyroid carcinoma - follicular variant; US, ultrasound.

although typical nuclear changes of papillary thyroid carcinoma are not present. Papillary renal cell carcinomas typically express CK7, CK8, CK18, CK19, CAM 5.2, RCC marker, CD10, and racemase (Table 2). Micropapillary urothelial carcinoma of the upper urinary tract is a rare variant of urothelial carcinoma. There is a male predominance (M:F ¼ 3.2:1). Almost all the reported cases occur in the urinary bladder, but it may also involve the renal pelvis and the ureter. It consists of slender, delicate fine papillary and filiform processes, with central fibrovascular cores. Papillae are lined by high nuclear grade cells with eosinophilic cytoplasm. Psammoma bodies are infrequent. Micropapillary carcinomas are immunoreactive for CK7, EMA, CK20, Leu M1, and CEA. The main differential diagnosis for the follicular variant of papillary thyroid carcinoma metastatic to the kidney is a Table 2 Immunohistochemical comparison between papillary thyroid carcinoma and papillary renal cell carcinoma

TTF-1 Thyroglobulin CK17 CD57 Racemase CD117 RCC marker CD15 CK7 CK19 EMA

PTC

PRCC

þ þ þ þ /þ þ  /þ þ þ þ

    þ þ/ þ þ þ þ þ

PRCC, papillary renal cell carcinoma; PTC, papillary thyroid carcinoma.

recently described entity called ‘primary thyroid-like follicular carcinoma of the kidney’. Other primary renal cell tumours, including oncocytoma, papillary renal cell carcinoma with tubular architecture, and metanephric adenoma should also be considered, although the presence of inspissated colloidlike material and/or follicular architecture is rare and patchy in these tumours. First reported in 2004, thyroid-like follicular carcinoma of the kidney is an extremely rare variant of renal cell carcinoma with only eight accepted cases described in the literature. This tumour shows a slight female predominance (M:F ¼ 1:2). Histologically, these tumours are well circumscribed with a distinct fibrous capsule, a striking follicular architecture with micro- and macro-follicles filled with inspissated colloid-like material. The cells lining the follicles have moderate to scant amphophilic to eosinophilic cytoplasm. Nuclei are round to oval, with uniform chromatin and inconspicuous nucleoli. Although these tumours are usually incidentally detected, have a relatively small size, and a predominantly indolent behaviour, a distinct malignant potential is supported by reported metastatic disease in two patients. Thyroid-like follicular carcinomas of the kidney lack key histological features of papillary carcinoma of thyroid, such as papillary architecture and classic nuclear changes. Immunohistochemically, tumour cells are negative for classical markers of thyroid neoplasms such as thyroglobulin, TTF-1 and CD57. Pax2, RCC marker, CD10, WT1, Ksp-cadherin, racemase, vimentin, and CD56 have been reported to be negative. In a case recently diagnosed at the authors’ institution, thyroid-like follicular carcinoma of the kidney showed strong nuclear staining for PAX8 (unpublished data, personal communication).

Copyright © Royal College of pathologists of Australasia. Unauthorized reproduction of this article is prohibited.

CORRESPONDENCE

In conclusion, thyroid carcinoma should be considered in the differential diagnosis of a renal mass, particularly in patients with a high serum thyroglobulin level, even if the mass is solitary and unilateral, or no history of thyroid cancer is present. The overlapping profile between papillary renal cell carcinoma and metastatic papillary thyroid carcinoma highlights the importance of clinicopathological correlation, and demonstrates the importance of using a panel of antibodies in differentiating these tumours. Conflicts of interest and sources of funding: The authors state that there are no conflicts of interest to disclose. Sara M. Falzarano*{ Deborah J. Chute* Cristina Magi-Galluzzi*{ *Pathology and Laboratory Medicine Institute, and {Glickman Urological and Kidney Institute, Cleveland Clinic, Cleveland, OH, United States; zDepartment of Pathology and Human Oncology, University of Siena, Siena, Italy Contact Dr C. Magi-Galluzzi. E-mail: [email protected] 1. Mihailovic J, Stefanovic L, Malesevic M. Differentiated thyroid carcinoma with distant metastases: probability of survival and its predicting factors. Cancer Biother Radiopharm 2007; 22: 250–5. 2. Clark JR, Lai P, Hall F, et al. Variables predicting distant metastases in thyroid cancer. Laryngoscope 2005; 115: 661–7. 3. Patel U, Ramachandran N, Halls J, et al. Synchronous renal masses in patients with a nonrenal malignancy: incidence of metastasis to the kidney versus primary renal neoplasia and differentiating features on CT. Am J Roentgenol 2011; 197: W680–6. 4. Pascal RR. Renal manifestations of extrarenal neoplasms. Hum Pathol 1980; 11: 7–17. 5. Heitz P, Moser H, Staub JJ. Thyroid cancer: a study of 573 thyroid tumors and 161 autopsy cases observed over a thirty-year period. Cancer 1976; 37: 2329–37. 6. Silliphant WM, Klinck GH, Levitin MS. Thyroid carcinoma and death. A clinicopathological study of 193 autopsies. Cancer 1964; 17: 513–25. 7. Malhotra G, Upadhye TS, Sridhar E, et al. Unusual case of adrenal and renal metastases from papillary carcinoma of thyroid. Clin Nucl Med 2010; 35: 731–6. 8. Borde C, Basu S, Kand P, et al. Bilateral renal metastases from papillary thyroid carcinoma on post 131I treatment scan: flip-flop sign, radioiodine SPET, 18F-FDG PET, CECT and histopathological correlation. Hell J Nucl Med 2011; 14: 72–3. 9. Gamboa-Dominguez A, Tenorio-Villalvazo A. Metastatic follicular variant of papillary thyroid carcinoma manifested as a primary renal neoplasm. Endocr Pathol 1999; 10: 256–68. 10. Gupta R, Viswanathan S, D’Cruz A, et al. Metastatic papillary carcinoma of thyroid masquerading as a renal tumour. J Clin Pathol 2008; 61: 143. 11. Lam KY, Ng WK. Follicular carcinoma of the thyroid appearing as a solitary renal mass. Nephron 1996; 73: 323–4. 12. Sardi A, Agnone CM, Pellegrini A. Renal metastases from papillary thyroid carcinoma. J La State Med Soc 1992; 144: 416–20.

DOI: 10.1097/PAT.0b013e32835b5dcc

Somatostatin receptor expression in prostate carcinoma: the urological pathologist’s role in the era of personalised medicine Sir, Somatostatin (SST) is known to inhibit the secretion of a wide range of hormones, exocrine glands, and gastrointestinal

93

motility. Among other actions, SST has revealed an antiproliferative potential, reversing the impact of mitogenic signals delivered by substances such as epidermal growth factor. The actions of SST are mediated by membrane-associated receptors that comprise five distinct subtypes (termed SSTR1 to 5). Frequently multiple subtypes coexist in the same cell. After binding their ligand, SSTR-ligand complexes undergo cellular internalisation with intracytoplasmic and intranuclear translocation. Reubi et al.1 showed that the degree of internalisation, i.e., the ratio of internalised SSTR2 to membranous SSTR2, varied greatly from one patient to the other. Although generally found in endosome-like structures, internalised SSTR2 were also identified to a small extent in lysosomes, as seen in colocalisation experiments. Very recently Waser et al.2 showed that phosphorylated SSTR2 was present in most gastrointestinal neuroendocrine tumours from patients treated with octreotide but that a striking variability existed in the subcellular distribution of phosphorylated receptors among such tumours. Cloning of the five SSTRs has led to the development of subtype-selective ligands.3 In the era of personalised medicine and targeted therapies, SSTR profiling is an important prerequisite for successful in vivo somatostatin receptor targeting for imaging or therapeutic purposes in an individual patient. Therefore, localisation and expression of the five SSTRs in a tumour must be determined to decide whether the patient is eligible for these applications. Several methods have been used to determine the expression of SSRTs. Tissue somatostatin receptors can be measured directly in vivo by performing a OctreoScan or 68 Ga-DOTATOC positron emission tomography/computed tomography scan. Molecular techniques such as in situ hybridisation histochemistry and autoradiography have been used in a limited number of studies.4 The former basically investigates SSTR mRNA expression in cryostat sections. The latter also utilises cryostat sections and is based on radioligands, i.e., 125I-labelled somatostatin ligands, such as octreotide. Previous studies have dealt with only some of the subtypes, therefore information is limited. The type of information obtained using these two techniques is not always comparable to that obtained with immunohistochemical analysis in formalin fixed, paraffin embedded (FFPE) tissue, in which the architecture and the cytology in the background are well preserved. In addition, the immunohistochemical technique is widely available, and faster, easier and cheaper to apply than in situ hybridisation histochemistry and autoradiography. It can be used even retrospectively on archival material. We read with great interest the recent publication by Ko¨rner et al.5 This study was performed on neuroendocrine tumours from various gastrointestinal and extragastrointestinal sites and in a small group of non-neuroendocrine tumours. The aim of the investigation was to correlate FFPE-based immunohistochemistry using the monoclonal anti-somatostatin receptor subtype 2A antibody UMB-1 (Biotrend Chemikalien, Germany; or Epitomics, USA), with the gold standard in vitro method quantifying somatostatin receptor levels in tumour tissues. The results obtained by comparing the UMB-1 immunohistochemistry with tumoural in vitro 125I-[Tyr3]-octreotide binding site levels allowed recommendations for the use of SSTR immunohistochemistry in daily diagnostics for optimally tailored patient management. Data on the immunohistochemical patterns of the five SSTRs in prostate cancer (PCa), its precursor high-grade prostatic intraepithelial neoplasia (HGPIN) and normal prostatic

Copyright © Royal College of pathologists of Australasia. Unauthorized reproduction of this article is prohibited.

94

CORRESPONDENCE

epithelium were obtained by our group in FFPE archival tissue material.6–9 Data were collected separately for the luminal/ secretory and basal epithelial cells, for the latter when present, as well as for the smooth muscle cells of the stroma and for the endothelial cells (Fig. 1). For the secretory or luminal cells, differences were found between normal, HGPIN and PCa (Fig. 2A), and between incidentally-detected and clinically-detected acinar PCa, therefore between insignificant or indolent and significant or aggressive cancers.9 Typical patterns in terms of localisation and expression of the five SSTRs in PCa with neuroendocrine differentiation,8 PCa following complete androgen ablation (CAA)6 and hormone refractory PCa7 were identified, with differences among these three groups and from untreated acinar adenocarcinoma9 (Fig. 2B). For the basal cells in normal prostate and HGPIN, immunoreactivity was primarily detected in the cytoplasm in all the five subtypes. In subtypes 1 and 3 the mean proportions of positive cells were higher than in the other three subtypes. The proportions were higher in normal prostate compared with HGPIN. Immunoreactivity for the five SSTRs in the groups of cases with neuroendocrine tumours was

Pathology (2013), 45(1), January

similar in terms of expression and localisation to that seen in the group of untreated HGPIN. The values in the patients under complete androgen ablation were lower than in the untreated patients. For the smooth muscle and endothelial cells, there were no cases with a distinct positivity in the cell membrane. Subtype 1 showed a strong immunoreactivity in the cytoplasm in the majority of the smooth muscle cells and the endothelial cells. Nuclear staining was seen only with subtypes 4 and 5. Neuroendocrine differentiation in PCa well as CAA and hormone refractory PCa did not affect SSTR expression and localisation in the in the endothelial and smooth muscle cells. The limitation of our study was that the immunohistochemical results were not compared with data obtained with a molecular technique or with a gold standard in vitro method quantifying somatostatin receptor levels in tissues, as was in the case of the Ko¨rner paper.5 However, specificity of the antibodies used in our studies (Rabbit polyclonal anti-SSTR subtype antibodies from Chemicon International, USA) was assessed. Western blot experiments on a prostate tissue extract were performed. Western blot analysis of prostate tissue

Fig. 1 Immunoreactivity for (A) SSTR4 in the luminal and basal cells in normal prostate, (B) SSTR3 in untreated HGPIN, (C) SSTR4 in untreated PCa, (D) SSTR4 in prostate cancer with neuroendocrine differentiation, (E) SSTR4 in complete androgen ablated prostate cancer, and (F) for SSTR4 in hormone refractory prostate cancer.

Copyright © Royal College of pathologists of Australasia. Unauthorized reproduction of this article is prohibited.

CORRESPONDENCE

%

95

15

10 Nep HGPIN PCa 5

0 SSTR1

A %

SSRT2

SSRT3

SSTR4

SSTR5

100 90 80 70 60 50

Membrane

40

Cytoplasm

30

Nucleus

20 10 0

B

GS 6 PCa

GS 8 PCa

PCa NE

CAA PCa

HR PCa

Fig. 2 (A) Mean percentages of intensely immunostained luminal (secretory) cells in the cytoplasm for the five SSTR subtypes in normal epithelium (Nep), high-grade prostatic intraepithelial neoplasia (HGPIN) and prostate cancer (PCa) from untreated patients. (B) Mean percentages of immunostained luminal cells for SSTR4 in PCa, separately for cell membrane, cytoplasm and nucleus from all groups. The mean values of SSTR expression in the cytoplasm, membrane and nuclei of the epithelial cells of hormone refractory PCa are lower than in the cancer areas of the PCa of the other two groups. The highest values are seen in the cytoplasm, the difference being significant. (GS 6, Gleason score 3þ3¼6; GS 8, Gleason score 4þ4¼8; NE, NE differentiation; CAA, PCa following complete androgen ablation; HR, hormone refractory PCa.)

performed with the panel of five polyclonal anti-SSTR antibodies yielded single bands as previously described by Helboe et al.10 We agree with Ko¨rner et al.5 that the type of antibodies can have an influence on the result to the point that the cytoplasmic and nuclear localisation, i.e., cellular internalisation, of the SSTRs can be seen intriguing and controversial. For instance, the immunohistochemical expression at the cell level of the five SSTRs in normal and pathological prostate tissue was also investigated by Dizeyi et al.11 There were differences between our investigation and that by Dizeyi et al., probably due to the types of antibodies used (in Dizeyi’s investigation the antibodies were from a private source and not commercially available). In conclusion, our data showed that SSTR profiling in an individual patient with HGPIN and the multifaceted PCa is feasible. Even though there is no clinical application for a somatostatin-based diagnostic test for prostate pathology at present, as opposed to neuroendocrine tumour, this should be of relevance to better tailor somatostatin analogue-based

diagnostic or therapeutic procedures in neoplasms other than neuroendocrine tumours.12 This is particularly important in the era of personalised medicine and targeted therapies. Rodolfo Montironi*ô Marina Scarpelli*ô Liang Cheng{ Antonio Lopez-Beltran§ Francesco Montorsi{ Ziya Kirkalijj *Section of Pathological Anatomy, Polytechnic University of the Marche Region, School of Medicine, United Hospitals, Ancona, and {Department of Urology, University VitaSalute, Scientific Institute H San Raffaele, Milan, Italy; zDepartment of Pathology and Laboratory Medicine, Indiana University School of Medicine, Indianapolis, IN, USA; §Department of Surgery, Cordoba University Medical School, Cordoba, Spain; jjDepartments of Urology, School of

Copyright © Royal College of pathologists of Australasia. Unauthorized reproduction of this article is prohibited.

96

CORRESPONDENCE

Medicine, Dokuz Eylu¨l University, Izmir, Turkey; ôthese authors contributed equally to this work Contact Professor R. Montironi. E-mail: [email protected] 1. Reubi JC, Waser B, Cescato R, et al. Internalized somatostatin receptor subtype 2 in neuroendocrine tumors of octreotide-treated patients. J Clin Endocrinol Metab 2010; 95: 2343–50. 2. Waser B, Cescato R, Liu Q, et al. Phosphorylation of sst2 receptors in neuroendocrine tumors after octreotide treatment of patients. Am J Pathol 2012; 180: 1942–9. 3. Hofland LJ, van der Hoek J, Feelders R, et al. Pre-clinical and clinical experiences with novel somatostatin ligands: advantages, disadvantages and new prospects. J Endocrinol Invest 2005; 28: 36–42. 4. Montironi R, Cheng L, Mazzucchelli R, et al. Immunohistochemical detection and localization of somatostatin receptor subtypes in prostate tissue from patients with bladder outlet obstruction. Cell Oncol 2008; 30: 473–82. 5. Ko¨rner M, Waser B, Schonbrunn A, et al. Somatostatin receptor subtype 2A immunohistochemistry using a new monoclonal antibody selects tumors suitable for in vivo somatostatin receptor targeting. Am J Surg Pathol 2012; 36: 242–52. 6. Mazzucchelli R, Morichetti D, Santinelli A, et al. Immunohistochemical expression and localization of somatostatin receptor subtypes in androgen ablated prostate cancer. Cell Oncol (Dordr) 2011; 34: 235–43. 7. Mazzucchelli R, Morichetti D, Scarpelli M, et al. Somatostatin receptor subtypes in hormone-refractory (castration-resistant) prostatic carcinoma. Asian J Androl 2011; 13: 242–7. 8. Morichetti D, Mazzucchelli R, Santinelli A, et al. Immunohistochemical expression and localization of somatostatin receptor subtypes in prostate cancer with neuroendocrine differentiation. Int J Immunopathol Pharmacol 2010; 23: 511–22. 9. Morichetti D, Mazzucchelli R, Stramazzotti D, et al. Immunohistochemical expression of somatostatin receptor subtypes in prostate tissue from cystoprostatectomies with incidental prostate cancer. BJU Int 2010; 106: 1072–80. 10. Helboe L, Møller M, Nørregaard L, et al. Development of selective antibodies against the human somatostatin receptor subtypes sst1-sst5. Brain Res Mol Brain Res 1997; 49: 82–8. 11. Dizeyi N, Konrad L, Bjartell A, et al. Localization and mRNA expression of somatostatin receptor subtypes in human prostatic tissue and prostate cancer cell lines. Urol Oncol 2002; 7: 91–8. 12. Mazzucchelli R, Scarpelli M, Lopez-Beltran A, et al. Immunohistochemical expression and localization of somatostatin receptors in normal prostate, high grade prostatic intraepithelial neoplasia and prostate cancer and its many faces. J Biol Regul Homeost Agents 2012; 26: 181–92.

DOI: 10.1097/PAT.0b013e32835bae76

Reticular and microcystic schwannoma of the parotid gland Sir, Reticular and microcystic schwannoma is a recently described, rare, distinctive variant of schwannoma with a predilection for visceral organs, particularly the gastrointestinal tract.1,2 Fewer than 20 cases have been described in the literature and involvement of the head and neck has not been reported.1–6 Herein we report a case of reticular and microcystic variant of schwannoma in the parotid gland and discuss its unique histomorphological features and the relevant differential diagnoses in this location. Microcystic/reticular variant of schwannoma demonstrates similar biological behaviour to usual schwannoma and must be distinguished from other parotid gland tumours that may recur aggressively if incompletely excised. A 59-year-old woman presented with a small pre-auricular swelling. She did not have any stigmata of neurofibromatosis

Pathology (2013), 45(1), January

(NF) type I or II. Fine needle aspiration smears showed loosely cohesive aggregates of cytologically bland spindle shaped cells admixed with myxoid stroma (Fig. 1A). The cellular aggregates showed swirling patterns (Fig. 1B) and occasional cells showed intracytoplasmic vacuoles (Fig. 1C). The cytological features were of a benign lesion and a diagnosis of pleomorphic adenoma was made on cytology. At surgery, the tumour was found closely associated with facial nerve and was resected with nerve preservation. The surgical specimen comprised a portion of the parotid gland with a macroscopically well circumscribed nodule measuring 28
20
16 mm. On cut section, the tumour appeared pale grey, firm and had a gelatinous texture (Fig. 2A). Histologically, the tumour was well demarcated, but unencapsulated with entrapped ductal structures at the periphery (Fig. 2B). Occasional lymphoid aggregates were also seen at the periphery (Fig. 2C). The tumour was composed of slender spindle shaped cells arranged in an anastomosing lattice-like pattern amidst myxoid stroma. Microcystic and reticular formations were seen (Fig. 2D). The cells showed cytoplasmic vacuolation and uniform slender spindle shaped to ovoid nuclei (Fig. 2E). Foci of extravasated red blood cells were seen. Focally, the extravasated red blood cells overlying cytoplasmic vacuoles appeared reminiscent of intracytoplasmic lumina with red blood cells (Fig. 2F). The lesion displayed relatively uniform cellularity and hypoand hypercellular areas, or areas with nuclear palisading were not seen. Hyalinised, thick walled blood vessels were absent. Degenerative changes, cyst formation, haemorrhage, or foamy macrophages were not seen. Epithelial elements, other than the peripheral entrapped ducts, were not identified within the tumour. Cytological atypia, mitoses or necrosis were not present. Special stains with periodic acid-Schiff reagent, with and without diastase digestion, and mucicarmine did not show any intracytoplasmic glycogen or mucin. Immunohistochemical staining of the tumour showed diffuse and strong nuclear and cytoplasmic immunoreactivity with S100 (Fig. 3A). The tumour also showed immunoreactivity for GFAP (Fig. 3B) and CD34. The tumour lacked immunoreactivity with multiple epithelial markers including Pan CK, AE1/AE3, EMA, Cam5.2, mCEA, and CK7. The tumour also lacked immunoreactivity with myoepithelial markers such as smooth muscle actin (SMA), p63, calponin, smooth muscle myosin heavy chain (SMMHC) and vascular markers such as CD31, D240 and factor VIII. The tumour did not demonstrate immunoreactivity with desmin and myogenin. Electron microscopy performed on formalin fixed, paraffin embedded tissue showed scattered elongated spindle cells in an amorphous matrix with thin cellular processes, scant cytoplasmic organelles and focal basal lamina-like material (Fig. 3C). Epithelial, myoid or endothelial differentiation was lacking. Schwannomas are generally benign, non-recurring tumours that can show a plethora of histological appearances and several variants are well described in the literature. Liegl et al. have recently described reticular and microcystic variant of schwannoma in visceral organs.1 As observed in the current case, these tumours show a predilection for women in the seventh decade and tend to be relatively small in size.2 The tumours are typically well circumscribed and show spindle shaped cells embedded in myxoid matrix as seen in this case.1,2 The current case also showed extensive lattice-like patterns with cribriform microcystic areas as described in the literature.1,2 Areas with

Copyright © Royal College of pathologists of Australasia. Unauthorized reproduction of this article is prohibited.

CORRESPONDENCE

97

Figure 1 (A) Fine needle aspiration smears showed loosely cohesive aggregates of cytologically bland spindle shaped cells admixed with myxochondroid matrix (DiffQuik). (B) Swirling pattern within the cellular aggragates (PAP). (C) Intracytoplasmic vacuoles (PAP).

extravasated red blood cells were also present. An unusual feature observed in the current case and not described previously in literature was the presence of intracytoplasmic vacuoles within most of the spindle shaped cells. These vacuoles could also be observed in the fine needle aspiration material. A high index of suspicion and awareness of the histological features of this variant are essential for accurate diagnosis as the tumour lacks several typical features seen in a usual schwannoma. Visceral examples of microcystic and reticular variant of schwannomas lack encapsulation and may show entrapment of normal structures at the periphery as seen in the current case.1 A

peripheral cuff of lymphoid aggregates, as seen in the current case, may be helpful when present; however, Liegl et al. observed this feature in only about 10% of their cases.1 The tumours also generally lack the classic organisation into Antoni A and Antoni B areas, and thick walled hyalinised vessels may be rare.1,2,4 Immunohistochemistry is of diagnostic utility as microcystic and reticular variant of schwannoma demonstrates immunoreactivity for S100, GFAP and occasionally for CD34, as is the usual case in schwannomas. As expected, the tumours lack immunoreactivity for all epithelial, myoepithelial, muscle and vascular markers such as a broad panel of cytokeratins, p63, calponin, SMA, desmin, CD31, D240 and factor VIII.

Figure 2 (A) Parotid gland with a well circumscribed nodule measuring pale grey, firm tumour with a gelatinous texture. (B) Well demarcated, but unencapsulated tumour with entrapped ductal structures at the periphery (H&E). (C) Lymphoid aggregates at the periphery (H&E). Slender spindle shaped cells arranged in a anastomosing lattice-like pattern amidst myxoid stroma with formation of microcystic and reticular areas (H&E). Spindle shaped cells with cytoplasmic vacuolation (H&E). Foci of extravasated red overlying cytoplasmic vacuoles reminiscent of intracytoplasmic lumina with red blood cells (H&E).

Copyright © Royal College of pathologists of Australasia. Unauthorized reproduction of this article is prohibited.

98

Pathology (2013), 45(1), January

CORRESPONDENCE

Figure 3 (A) Diffuse nuclear and cytoplasmic immunoreactivity with S100. (B) Immunoreactivity with GFAP. (C) Electron micrograph showing schwannian features such as slender, elongated interdigitating cell processes covered by a thin, continuous layer of basement membrane material and scant cytoplasmic organelles.

The histopathological appearance of these tumours raises a wide spectrum of differential diagnoses, particularly in the salivary gland where extratemporal schwannomas arising from the intraparotid facial nerve are very rare.7 The differential diagnostic considerations in this location include pleomorphic adenoma and myoepithelial neoplasms. Given the remarkable overlap in the histopathological features of these three entities, the lack of immunoreactivity for cytokeratins and myoepithelial markers such as calponin, p63, and SMA, as would be expected in pleomorphic adenoma and myoepithelial tumours, aids in making the distinction. The present case of microcystic and reticular schwannoma showed cord-like arrangement of cells and extensive cytoplasmic vacuolation. Artefactual overlapping of extravasated red blood cells in some areas (Fig. 2) was reminiscent of intracytoplasmic vascular lumina, raising the differential diagnosis of epithelioid haemangioendothelioma. The presence of cords and nests of cells with extensive cytoplasmic vacuolation were also suggestive of a parachordoma, a rare and unusual entity in the salivary gland. Lack of immunohistochemical reactivity for several vascular markers and cytokeratins can aid in excluding these unusual entities. The constellation of the clinical, histological and immunohistochemical features are generally sufficient for accurate diagnosis without resorting to electron microscopy, which shows features including long slender interdigitating cell processes and basement membrane material, in keeping with schwannian differentiation as seen in the present case. Preservation of the facial nerve is an important consideration during parotid surgery, potentially leading to incomplete resection of the lesion. Schwannomas, including the microcystic and reticular variant, are benign tumours that generally do not recur even after incomplete resection. Thus, accurate distinction of schwannoma from other benign entities such as pleomorphic adenoma is important, as the latter shows propensity for multiple and multifocal recurrences that are difficult to control surgically following incomplete resection.8 In summary, we describe microcystic and reticular variant of schwannoma in the parotid gland with an emphasis on some of its unusual histopathological features warranting a high

index of suspicion and ancillary techniques for accurate diagnosis. Jia-Min Pang* Annabelle Mahar* Kerwin Shannon{ James Kench* Charles Chan{ Ruta Gupta* *Tissue Pathology and Diagnostic Oncology, Royal Prince Alfred Hospital, Camperdown, {Sydney Head and Neck Cancer Institute, Royal Prince Alfred Hospital, Camperdown, zElectron Microscopy Unit, Anatomical Pathology, Concord Hospital, and Discipline of Pathology, the University of Sydney, Sydney, NSW, Australia Contact Dr R. Gupta. E-mail: [email protected] 1. Liegl B, Bennett MW, Fletcher CD. Microcystic/reticular schwannoma: a distinct variant with predilection for visceral locations. Am J Surg Pathol 2008; 32: 1080–7. 2. Chetty R. Reticular and microcystic schwannoma: a distinctive tumor of the gastrointestinal tract. Ann Diagn Pathol 2011; 15: 198–201. 3. Agaimy A, Markl B, Kitz J. Peripheral nerve sheath tumors of the gastrointestinal tract: a multicenter study of 58 patients including NF1-associated gastric schwannoma and unusual morphologic variants. Virchows Arch 2010; 456: 411–22. 4. Lee SM, Goldblum J, Kim KM. Microcystic/reticular schwannoma in the colon. Pathology 2009; 41: 595–6. 5. Kienemund J, Liegl B, Siebert F, Jagoditsch M, Spuller E, Langner C. Microcystic reticular schwannoma of the colon. Endoscopy 2010; 42 (Suppl 2): E247. 6. Liegl B, Bodo K, Martin D, Tsybrovskyy O, Lackner K, Beham A. Microcystic/reticular schwannoma of the pancreas: a potential diagnostic pitfall. Pathol Int 2011; 61: 88–92. 7. Eisele DW, Johns ME. Salivary gland neoplasms. In: Bailey BJ, editor. Head and Neck Surgery–Otolaryngology. Philadelphia: Lippincott Williams and Wilkins, 2001; 1279–97. 8. Witt RL. The significance of the margin in parotid surgery for pleomorphic adenoma. Laryngoscope 2002; 112: 2141–54.

DOI: 10.1097/PAT.0b013e32835be3ec

Copyright © Royal College of pathologists of Australasia. Unauthorized reproduction of this article is prohibited.

CORRESPONDENCE

99

Predominance of VREfm ST203 subgroup in Queensland

Table 1

Correlation of MelT with ST

MelT

Possible ST*

CC

Sir, The molecular epidemiology of Enterococcus faecium in an Australian setting has recently been described for the first time.1 Johnson et al. described the epidemiology of 85 E. faecium isolates in blood culture over a 12 year period at a single institution in Victoria, Australia (Austin Health, Melbourne).1 This comprised 34 vancomycin resistant E. faecium (VREfm) and 51 vancomycin susceptible E. faecium (VSEfm) isolates. They defined 17 different sequence types (STs) amongst 85 E. faecium isolates using multilocus sequence typing (MLST) and found three dominant STs (ST17, ST252 and ST203). Amongst the VREfm isolates, all but one carried the resistance gene, vanB.1 ST17, the putative founder of clonal complex 17 (CC17), was stable and predominated in VREfm and VSEfm for the first 10 years of the 12 year study period. From 2007 to 2009, a significant increase in VREfm bacteraemia was identified.1 The predominant ST was ST203, accounting for 76% and 81.8% of VREfm bacteraemia isolates in 2007 and 2009, respectively. ST203 is a double-locus variant of ST17 and both STs belong to CC17. We have investigated 765 VREfm isolates (39 clinical and 726 screening isolates) from hospitalised patients across 26 hospitals in Queensland, collected from January to November 2010. Van genotyping was performed on all isolates. A total of 758 isolates (99.1%) had a vanB genotype while seven isolates were positive for vanA. Ninety-one vanB VREfm were selected to study molecular epidemiology using repetitive polymerase chain reaction (PCR) (DiversiLab; bioMe´rieux, France), comprising both clinical (n ¼ 39) and screening (n ¼ 62) specimens. Results revealed that the majority of isolates from Queensland (n ¼ 88) were very closely related (>92% similarity). Fourteen isolates from this major group were further analysed by high resolution melting (HRM) genotyping as described.2 Four melting types (MelTs) were identified, MelT11 (n ¼ 1), MelT34 (n ¼ 1), MelT121 (n ¼ 11) and one novel MelT. MelT11, MelT34 and MelT121 represent STs which are clustered in CC17. MelT11 and MelT121 include various STs from each subgroup founded by ST17 and ST203, respectively. MelT34 incorporated multiple subgroups of CC17 as well as the singleton ST51 (Table 1). MLST was performed in all 14 isolates as previously described;3 of these, 12 isolates were ST203. The results from MLST were correlated with HRM genotyping except the novel MelT that was identified as ST203 (Table 1). In conclusion, our results indicate that 88 of 91 isolates (97%) were closely related. HRM genotyping and MLST of representative isolates from this cluster revealed that ST203 was predominant. This suggests that ST203 may be responsible for VREfm in Queensland. These findings correspond with what was found by Johnson et al. The same ST causing outbreaks in two geographically non-contiguous states suggests

11

17, 63, 103, 180, 187, 234, 267, 295, 307, 308, 357, 460, 475, 480, 538, 543, 554, 578, 584 49, 51, 177, 232, 341, 547, 548, 556 203, 365, 412, 478, 483, 577

17

34 121

17 17

* Generated by Enterococcus faecium MLST and MelT key as described by Tong et al.2 and found at http://menzies.edu.au/node/43174. Bold type indicates ST identified by MLST. CC, clonal complex; MelT, melting type; MLST, multilocus sequence typing; ST, sequence type.

that other regions in Australia may be similarly affected. ST203 has also been reported in Korea, Japan China, Germany, Denmark, The Netherlands and Serbia (http://efaecium.mlst.net/). Interestingly, ST203 isolates reported from these countries almost entirely possessed the vanA gene. On the contrary, the majority of isolates from Victoria and Queensland possessed vanB. This highlights that CC17, especially ST203, has a great ability for hospital adaptation. CC17 has caused outbreaks in multiple continents including Australia.4 Further study and surveillance of this subgroup is necessary to understand its persistence in the hospital environment. Conflicts of interest and sources of funding: The authors state that there are no conflicts of interest to disclose. Witchuda Kamolvit* Hanna E. Sidjabat* Graeme R. Nimmo*{ Snehal N. Anuj{ Haakon Bergh{ Leisha J. Richardson* David L. Paterson*{ *The University of Queensland Centre for Clinical Research, Herston, Brisbane, and {Division of Microbiology, Pathology Queensland Central Laboratory, Brisbane, Qld, Australia Contact Dr W. Kamolvit. E-mail: [email protected] 1. Johnson PD, Ballard SA, Grabsch EA, et al. A sustained hospital outbreak of vancomycin-resistant Enterococcus faecium bacteremia due to emergence of vanB E. faecium sequence type 203. J Infect Dis 2010; 202: 1278–86. 2. Tong SY, Xie S, Richardson LJ, et al. High-resolution melting genotyping of Enterococcus faecium based on multilocus sequence typing derived single nucleotide polymorphisms. PloS ONE 2011; 6: e29189. 3. Homan WL, Tribe D, Poznanski S, et al. Multilocus sequence typing scheme for Enterococcus faecium. J Clin Microbiol 2002; 40: 1963–71. 4. Willems RJ, Top J, van Santen M, et al. Global spread of vancomycinresistant Enterococcus faecium from distinct nosocomial genetic complex. Emerg Infect Dis 2005; 11: 821–8.

DOI: 10.1097/PAT.0b013e32835b68d2

Copyright © Royal College of pathologists of Australasia. Unauthorized reproduction of this article is prohibited.