Exophytic schneiderian carcinoma of the lacrimal sac: first case report.

Pathology (August 2014) 46(5), pp. 450–472

CORRESPONDENCE Expression of MAP 2 by haemangioblastomas: an immunohistochemical study with implications for diagnosis

Sir, Haemangioblastomas are uncommon central nervous system (CNS) tumours that usually occur in the cerebellum, brain stem or spinal cord of adults.1 The tumours are composed of neoplastic stromal cells and abundant blood vessels, are low grade and occur sporadically or in association with von Hippel–Lindau syndrome. The histogenesis of the stromal cells is uncertain with glia, endothelial cells, arachnoid cells, fibrohistiocytic cells, neuroendocrine cells and neuroectodermal cells all proposed as origins.2–6 Since it has been suggested that haemangioblastomas are glial or neuroepithelial in origin, we hypothesised that the stromal cells would express microtubule associated protein 2 (MAP 2), an antigen that is consistently expressed by neural and neoplastic glial cells. In addition, since studies have suggested that non-neuroendocrine carcinomas usually do not express MAP 2,7,8 we hypothesised that its expression would distinguish haemangioblastoma from metastatic clear cell renal carcinoma. We further hypothesised that MAP 2 expression would distinguish haemangioblastoma from clear cell and microcystic meningiomas, two other tumours in the differential diagnosis. With the approval of the Institutional Review Board of the University of Texas Southwestern Medical Center, formalin fixed, paraffin embedded sections of 48 haemangioblastomas, seven metastatic clear cell renal carcinomas, seven clear cell meningiomas and eight microcystic meningiomas were retrieved from the archives of the Department of Pathology, Division of Neuropathology (Table 1). Additional sections of the tumours were stained with an antibody to MAP 2 (monoclonal mouse IgG, 1:4000; Sigma-Aldrich, USA) using a Leica Bond-III automated stainer (Leica Biosystems, Germany). Expression of MAP 2 was defined as cytoplasmic staining of any intensity. MAP 2 was expressed by stromal cells of 32 haemangioblastomas (67%); staining varied from focal to multifocal and weak to strong (Fig. 1). MAP 2 was not expressed by the neoplastic cells of any of the renal carcinomas or microcystic or clear cell meningiomas (Table 2). Microtubule associated proteins are structural microtubulebinding proteins that bind to tubulin polymers to contribute to the regulation of microtubule functions.9 The MAP 2 family is abundant in the mammalian CNS and composed of two groups of isoforms, high molecular weight, MAP 2A and 2B and low molecular weight, 2C and 2D. In the nervous system, high molecular weight forms are expressed in neurons while low Table 1

Fig. 1 (A) A cerebellar haemangioblastoma in a 47-year-old man (H&E). (B) Immunohistochemistry for MAP 2 with diaminobenzidine.

molecular weight forms are also expressed in glia.9 MAP 2 has also been identified by immunohistochemistry in cells of the adrenal medulla, pancreatic islets, exocrine pancreas, salivary glands and thyroid follicles, cells of monocyte/macrophage lineage, follicular dendritic cells and skeletal muscle.7 Among neoplasms, MAP 2 is expressed by cells of most neural, glial, neuroectodermal, neuroendocrine and neural crest tumours.7,8 In this regard, MAP 2 has emerged as a sensitive marker of primary and also metastatic neuroblastoma, particularly in bone marrow.7 Expression has also been observed in tumours of the haemangiopericytoma/solitary fibrous tumour family, glomus tumours, Kaposi’s sarcoma, isolated cases of gastrointestinal stromal tumour, endometrial stromal sarcoma,

Clinicopathological features of 70 CNS clear cell tumours

Haemangioblastoma Renal cell carcinoma Clear cell meningioma Microcystic meningioma

M/F

Age, years

26/22 4/3 3/4 3/5

15–78 46–69 35–66 44–72

Print ISSN 0031-3025/Online ISSN 1465-3931

#

Cerebellum

Brain stem

Spinal cord

Supratentorial

28 1 2 0

2 0 0 0

11 4 3 0

6 2 2 8

2014 Royal College of Pathologists of Australasia

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

CORRESPONDENCE

Table 2

Expression of MAP 2 by CNS clear cell tumours

Haemangioblastoma Renal cell carcinoma Clear cell meningioma Microcystic meningioma

Positive

Negative

% Positive

32 0 0 0

16 7 7 8

67 0 0 0

liposarcoma and leiomyosarcoma and in tumours containing histiocytes/macrophages.7 Among non-neuroendocrine carcinomas, focal expression of MAP 2 was observed in six of 50 (12%) non-small cell lung carcinomas, although areas of neuroendocrine differentiation (confirmed by synaptophysin expression) were noted in four of the six tumours.8 Other investigators noted punctate, apical expression in five of five papillary thyroid carcinomas and weak expression in one of five adenocarcinomas of the lung.7 MAP 2 was not expressed by neoplastic cells in five of five papillary carcinomas of the kidney.7 We demonstrated MAP 2 expression by stromal cells in 32 of 48 (67%) haemangioblastomas by immunohistochemistry; MAP 2 was not expressed by neoplastic cells of seven clear cell meningiomas, eight microcystic meningiomas or seven metastatic clear cell renal carcinomas. In their study of the utility of MAP 2 expression in the diagnosis of low grade neuroepithelial tumours, Blumcke et al. detected MAP 2 in seven of 10 (70%) haemangioblastomas.10 The studies suggest that MAP 2 is a moderately sensitive marker of haemangioblastoma and, in the CNS, highly specific inside of the usual differential diagnosis. Compared with inhibin, recent series11–13 suggest that MAP 2 is less sensitive but, with regard to differentiation of CNS clear cell neoplasms, perhaps more specific.13,14 We conclude that addition of MAP 2 to a panel of immunohistochemical stains may aid in the diagnosis of clear cell tumours of the brain and spinal cord. Acknowledgements: The authors thank Ping Shang for assistance with immunohistochemical staining. Conflicts of interest and sources of funding: The authors state that there are no conflicts of interest to disclose. Patrick J. Malafronte Ella M. LePage Kimmo J. Hatanpaa Dennis K. Burns Charles L. White III Jack Raisanen Department of Pathology, University of Texas Southwestern Medical Center, Dallas, TX, USA Contact Dr J. Raisanen. E-mail: [email protected] 1. Aldape KD, Plate KH, Vortmeyer AO, et al. Haemangioblastoma. In: Louis DN, Ohgaki H, Wiestler OD, Cavenee WK, editors. WHO Classification of Tumours of the Central Nervous System. 4th ed. Lyon: IARC, 2007; 184–6. 2. Alles JU, Bosslet K, Schachenmayr W. Hemangioblastoma of the cerebellum – an immunohistochemical study. Clin Neuropathol 1986; 5: 238– 41.

451

3. Becker I, Paulus W, Roggendorf W. Histogenesis of stromal cells in cerebellar hemangioblastomas. An immunohistochemical study. Am J Pathol 1989; 134: 271–5. 4. Jurco S III, Nadji M, Harvey DG, et al. Hemangioblastomas: histogenesis of the stromal cell studied by immunocytochemistry. Hum Pathol 1982; 13: 13–8. 5. Mizuno J, Iwata K, Takei Y. Immunohistochemical study of hemangioblastoma with special reference to its cytogenesis. Neurol Med Chir 1993; 33: 420–4. 6. Nemes Z. Fibrohistiocytic differentiation in capillary hemangioblastoma. Hum Pathol 1992; 23: 805–10. 7. Krishnan C, Higgins JP, West RB, et al. Microtubule-associated protein-2 is a sensitive marker of primary and metastatic neuroblastoma. Am J Surg Pathol 2009; 33: 1695–704. 8. Liu Y, Saad RS, Shen SS, et al. Diagnostic value of microtubule-associated protein-2 (MAP-2) for neuroendocrine neoplasms. Adv Anat Pathol 2003; 10: 101–6. 9. Sanchez C, Diaz-Nido J, Avila J. Phosphorylation of microtubule-associated protein 2 (MAP2) and its relevance for the regulation of the neuronal cytoskeletal function. Prog Neurobiol 2000; 61: 133–68. 10. Blumcke I, Muller S, Buslei R, et al. Microtubule-associated protein-2 immunoreactivity: a useful tool in the differential diagnosis of low-grade neuroepithelial tumors. Acta Neuropathol 2004; 108: 89–96. 11. Carney EM, Banerjee P, Ellis CL, et al. PAX2(-)/PAX8(-)/inhibin A(þ) immunoprofile in hemangioblastoma: A helpful combination in the differential diagnosis with metastatic clear cell renal cell carcinoma to the central nervous system. Am J Surg Pathol 2013; 35: 262–7. 12. Doyle LA, Fletcher CD. Peripheral hemangioblastoma: clinicopathologic characterization in a series of 22 cases. Am J Surg Pathol 2014; 38: 119–27. 13. Rivera AL, Takei H, Zhai J, et al. Useful immunohistochemical markers in differentiating hemangioblastoma versus metastatic renal cell carcinoma. Neuropathology 2010; 30: 580–5. 14. Gurses I, Scheithauer BW. Inhibin-A immunoreactivity in nervous system lesions. Appl Immunohistochem Mol Morphol 2012; 20: 277–84.

DOI: 10.1097/PAT.0000000000000138

Primary pulmonary hyalinising spindle cell tumour with giant rosettes Sir, Hyalinising spindle cell tumour with giant rosettes (HSCT) is an uncommon variant of low grade fibromyxoid sarcoma (LGFMS) which usually occurs in the deep soft tissues of the lower extremities in young to middle aged adults.1 Primary soft tissue HSCT can metastasise to the lung2 but primary pulmonary HSCT are very rare with only two cases described previously in the literature.3,4 Herein we report a third case. A 66-year-old woman, with a history of breast carcinoma and bronchiectasis, had a right lung nodule diagnosed as an incidental finding on chest X-ray. A positron emission tomography (PET) scan showed the nodule to be FDG avid with no FDG avid lesions elsewhere in the body. A right lower lobe thoracotomy and wedge excision was performed. The gross specimen showed a very well circumscribed, lobulated, firm, cream coloured tumour, 26 mm in maximum dimension, which had a gritty texture on sectioning. The histological features were consistent with a HSCT. The tumour had well circumscribed margins and was composed of variably sized, well defined, prominent rosettes scattered throughout a relatively loose spindle cell background. The rosettes consisted of central hyalinsed collagenous material surrounded by palisading, mildly anisomorphic epithelioid cells (Fig. 1A). Areas of the tumour resembled a sclerosing epithelioid fibrosarcoma with epithelioid cells dispersed within a densely sclerotic hyalinsed stroma with pericellular clearing


452

Pathology (2014), 46(5), August

CORRESPONDENCE

Fig. 2 FUS fluorescence in situ hybridisation. A FUS (16p11) break apart probe showing a split red and green signal in many of the cells indicating a FUS gene rearrangement.

fibrosarcoma is considered to be a separate distinct variant of fibrosarcoma9 which does not have the FUS rearrangement.10 The behaviour of HSCT is identical to LGFMS6 and should be considered as a low grade sarcoma.7 Acknowledgements: We would like to thank the patient for kindly giving consent to publish this case, her surgeon Dr John Tharion, The Canberra Hospital, and also Associate Professor S. O’Toole, Royal Prince Alfred Hospital, for kindly providing the FISH images. Conflicts of interest and sources of funding: The authors state that there are no conflicts of interest to disclose. Fig. 1 Hyalinising spindle cell tumour with giant rosettes. (A) Prominent rosettes with a central collagenous core surrounded by palisaded epithelioid cells were present throughout the tumour (H&E; bar ¼ 200 mm). (B) Areas of the tumour resembled a sclerosing epithelioid fibrosarcoma (H&E; bar ¼ 50 mm).

(Fig. 1B). There was no necrosis and mitotic figures were infrequent. The spindled cells were positive for vimentin and negative for S100, AE1/AE3, CK8/18, CD34, desmin and smooth muscle actin. The epithelioid cells palisading around the rosettes were focally S100 positive. The Ki-67 proliferation index was up to 2%. In the absence of a soft tissue primary on imaging, and no previous history of soft tissue surgery, this tumour was considered to be consistent with a primary pulmonary HSCT. Fluorescence in situ hybridisation (FISH) was performed using a FUS (16p11) break apart probe on unstained slides of formalin fixed, paraffin embedded tissue which detected a FUS gene rearrangement (Fig. 2). HSCT is a variant of LGFMS with the same morphology as LGFMS except for the presence of well formed collagen rosettes.5,6 Immunohistochemically, the spindle cells of HSCT express vimentin, consistent with fibroblastic differentiation.1 The cells forming the rosettes have a neural phenotype with S100, Leu-7 and NSE expression1,5 and ultrastructural features of Schwann cells.2 Although the morphology of HSCT appears distinctive, giant collagen containing rosettes are also seen in other mesenchymal tumours.1 The diagnosis of LGFMS/HSCT is confirmed by characteristic FUS-CREB3L2 and FUSCREB3L1 fusion genes from a t(7;16)(q32-34;p11) balanced translocation which is not seen in other mesenchymal tumours.7 HSCT can have areas with morphology similar to sclerosing epithelioid fibrosarcoma8 but sclerosing epithelioid

Karen Whale Genevieve Bennett Department of Anatomical Pathology, ACT Pathology, The Canberra Hospital, Canberra, ACT, Australia Contact Dr K. Whale. E-mail: [email protected] 1. Lane KL, Shannon RJ, Weiss SW. Hyalinizing spindle cell tumor with giant rosettes: a distinctive tumor closely resembling low-grade fibromyxoid sarcoma. Am J Surg Pathol 1997; 12: 1481–8. 2. O’Sullivan MJ, Sirgi KE, Dehner LP. Low-grade fibrosarcoma (hyalinizing spindle cell tumor with giant rosettes) with pulmonary metastases at presentation: case report and review of the literature. Int J Surg Pathol 2002; 10: 211–6. 3. Kim L, Yoon YH, Choi SJ, et al. Hyalinizing spindle cell tumor with giant rosettes arising in the lung: report of a case with FUS-CREB3L2 fusion transcripts. Pathol Int 2007; 57: 153–7. 4. Magro G, Fraggetta F, Manusia M, et al. Hyalinizing spindle cell tumor with giant rosettes: a previously undescribed lesion of the lung. Am J Surg Pathol 1998; 22: 1431–3. 5. Ramaswamy AS, Chatura KR. Hyalinizing spindle cell tumor with giant rosettes. Ann Saudi Med 2011; 31: 83–6. 6. Folpe A, van den Berg E, Molenaar WM. Low grade fibromyxoid sarcoma. In: Fletcher CD, Unni KK, Mertens F, editors. World Health Organization Classification of Tumours. Pathology and Genetics of Soft Tissue and Bone. Lyon: IARC Press, 2002; 104–5. 7. Rekhi B, Deshmukh M, Jambhekar NA. Low-grade fibromyxoid sarcoma: a clinicopathologic study of 18 cases, including histopathologic relationship with sclerosing epithelioid fibrosarcoma in a subset of cases. Ann Diagn Pathol 2011; 15: 303–11. 8. Reid R, de Silva MV, Paterson L, et al. Low-grade fibromyxoid sarcoma and hyalinizing spindle cell tumor with giant rosettes share a common t(7;16)(q34;p11) translocation. Am J Surg Pathol 2003; 27: 1229–36.


CORRESPONDENCE

9. Meis-Kindblom, van den Berg E, Kindblom LG, et al. Sclerosing epithelioid fibrosarcoma. In: Fletcher CD, Unni KK, Mertens F, editors. World Health Organization Classification of Tumours. Pathology and Genetics of Soft Tissue and Bone. Lyon: IARC Press, 2002; 106–7. 10. Wang WL, Evans HL, Meis JM, et al. FUS rearrangements are rare in ’pure’ sclerosing epithelioid fibrosarcoma. Mod Pathol 2012; 25: 846–53.

DOI: 10.1097/PAT.0000000000000128

Megakaryocytes in axillary lymph nodes mimicking metastatic breast carcinoma following neoadjuvant chemotherapy and Herceptin Sir, The presence of isolated tumour cells within sentinel lymph nodes in patients who have had neoadjuvant chemotherapy could be taken to imply a prior heavier nodal tumour burden and result in excision of further lymph nodes and/or postoperative radiotherapy. Herein reported is a case in which megakaryocytes were present in axillary sentinel lymph nodes following neoadjuvant chemotherapy and Herceptin administration mimicked nodal metastasis. A 46-year-old woman presented with a mass in her left breast visible on mammogram, ultrasound and corresponding to a 40 mm area of abnormal enhancement on magenetic resonance imaging (MRI). Core biopsy revealed a grade 3 invasive ductal carcinoma of no special type (Fig. 1A) and accompanying high nuclear grade ductal carcinoma in situ (DCIS). The tumour expressed oestrogen receptor and was determined to be Her2 amplified by silver in situ hybridisation (SISH) method. A brisk lymphoplasmacytic infiltrate was present within the tumour. MRI highlighted a 1 cm ‘suspicious’ axillary lymph node, but on ultrasound examination this was shown to have a normal hilum and no cortical thickening. Fine needle aspiration cytology revealed only mixed lymphoid cells. The patient subsequently underwent three cycles of neoadjuvant chemotherapy with 5-fluorouracil, epirubicin and cyclophosphamide followed by three cycles of docetaxel and trastuzumab (Herceptin). One further cycle of trastuzumab was administered due to surgical delay at the request of the patient. Six weeks after completion of the chemotherapy regimen wide local excision and sentinel node biopsy was performed. Neoadjuvant therapy was accompanied by administration of the granulocyte-colony stimulating factor (G-CSF) pegfilgrastim (Neulasta). Pretreatment haematological parameters were normal apart from borderline mild lymphopenia. During treatment, this worsened and the patient also became mildly anaemic. Preoperative ultrasound examination showed no obvious residual tumour at the site of an earlier placed clip. The surgical excision specimen revealed a 32 mm scar, but no residual invasive carcinoma and only a single duct-lobular unit displayed residual immunohistochemically Her2 positive (3þ) DCIS. Within each of four of the five sentinel lymph nodes [assessed at four haematoxylin and eosin (H&E) stained levels and with an AE1/AE3 immunohistochemical marker performed on a serial section of level 2], there were a few prominent large cells with voluminous eosinophilic cytoplasm and large convoluted, variably polylobated hyperchromatic nuclei (Fig. 1B,C). They were present within the sinusoids

453

of the cortex and medulla, but not in the subcapsular sinus. There was focal minimal haemosiderin deposition but no scarring was seen. These cells were predominantly absent in the AE1/AE3 immunostained levels of each node so a lack of cytokeratin expression was interpreted as a sampling issue. Initially they were interpreted as metastatic isolated tumour cells (ITC) and their unusual appearance explained by prior chemotherapy and Herceptin administration. Post-operative radiotherapy was discussed as an option at a multidisciplinary team meeting, but in view of this unusual appearance and the complete pathological response in the breast for the invasive component, further assessment was undertaken to clarify their nature, prior to any further treatment decision. The possibility that they were megakaryocytes was raised. Levels of the nodes stained with H&E containing these cells were then de-stained and immunohistochemical markers for AE1/AE3, Her2 and CD61 were performed. The cells were negative for the first two markers but showed strong CD61 expression supporting that they were indeed megakaryocytes (Fig. 1D). Review of the initial AE1/AE3 levels revealed in one, several non-reactive smaller examples of these cells now determined to be megakaryocytes. No other convincing haematopoietic precursors were noted. There were a few eosinophils within the lymph nodes. The presence of megakaryocytes in axillary lymph nodes in patients with breast carcinoma is rare and has been documented on only five prior occasions in seven individuals (all female, 28–53 years old) and recorded as examples of nodal megakaryocytes or extramedullary haematopoiesis (EMH).1–5 Hoda et al.1 first reported the presence of megakaryocytes in axillary lymph nodes following mastectomy and axillary node sampling for DCIS with microinvasion. Unlike with this individual, all subsequent cases have been reported following neoadjuvant chemotherapy2–5 with the additional administration of G-CSF documented in two of these cases.3,4 Megakaryocytes were detected during intraoperative lymph node assessment in one instance.2 Similar to the current case, one previously reported case was Her2 amplified, but in contrast the Herceptin was administered post-operatively. Takhar et al.4 reported a case of nodal EMH histologically following six cycles of chemotherapy with concurrent G-CSF. However, preoperative fine needle aspiration of an axillary lymph node, as well as showing evidence of metastasis, revealed the presence of multinucleated cells at that time, suggesting the nodal megakaryocytes predated the neoadjuvant chemotherapy. Megakaryocytes found in vessels of organs such as the liver and spleen and in lymph node parenchyma usually are seen as part of EMH. However, because up to at least half of all megakaryocytes migrate into the systemic blood stream, many migrating to the lungs which is a reservoir for megakaryocytes,6 it is not inconceivable that they may circulate through other organs including nodes in a similar manner and therefore their detection may not necessarily represent EMH, as they may be discovered ‘in transit’.1,6 In four previous individuals, the presence of other haematopoietic precursors was documented with all three lineages noted in three,5 and the presence of myeloid precursors along with increased eosinophils in the fourth.3 The presence of only rare eosinophils was documented in another case.1 In the current case similarly there were a few accompanying eosinophils, but no other haematopoietic lineages were evident. The explanation for the presence of megakaryocytes and/or EMH in axillary lymph nodes in breast cancer patients has been suggested to relate to bone marrow suppressing


454


CORRESPONDENCE

A

B

C

D

Fig. 1 (A) Primary breast carcinoma on initial diagnostic core biopsy (H&E). (B) Megakaryocyte within a lymph node (H&E). (C) Megakaryocytes within a lymph node (H&E). (D) CD61 expression by two adjacent megakaryocytes.

chemotherapeutic drugs (such as cyclophosphamide) and/or GCSF administration.2–5 The patient reported by Hoda et al.,1 did not receive neoadjuvant chemotherapy and had no documented haematopoietic disorder. Bone marrow examination was even performed in that case as part of the subsequent diagnostic workup. EMH in lymph nodes may mimic metastasis in several ways. Erythroid precursors may potentially be confused with lobular carcinoma. Megakaryocytes may be more likely to mimic metastasis from carcinomas with a metaplastic component (squamous or sarcomatous).5 However, the latter may also mimic Her2 amplified tumours, which typically are high grade and have voluminous eosinophilic cytoplasm as was suggested in this case. Moreover, neoadjuvant treatment can alter the appearance of tumour cells such that they appear cytologically different to those in the pre-treatment biopsy material, mostly of higher grade. Although the therapeutic response in breast and lymph nodes is generally similar, various changes in lymph nodes may occur.7 A clue to the non-epithelial nature of the cells within the node in this case was the location of the cells within cortical and medullary sinusoids but not subcapsular sinuses.1 Sentinel lymph node assessment with cytokeratins increases the efficiency of detecting small metastases. However, nonspecific cytokeratin expression occurs in lymph nodes by plasma cells and reticulum cells particularly with use of pan-cytokeratin and CK8/18 (cam5.2) which may lead to over diagnosis, although this is not documented with AE1/AE3.8 On the other hand, not all genuinely keratin positive cells in lymph nodes may be metastatic or even epithelial due to the presence of the rare occurrence of axillary nodal endosalpingiosis, and benign epithelial or mesothelial inclusions. Other benign processes that may mimic metastasis but are keratin negative include naevus cells, and multinucleated or signet ring histiocytes. 1,4,5 Understanding the various processes that may mimic axillary or other nodal metastasis in breast cancer patients with

or without a history of neoadjuvant therapy will help reduce errors and guide appropriate patient management. Acknowledgements: Enormous thanks are owed to Frances Boyle and Ivan Burchett. Conflicts of interest and sources of funding: The authors state that there are no conflicts of interest to disclose. M. J. Wilsher S. F. Bonar Douglass Hanly Moir Pathology, Macquarie Park, NSW, Australia Contact Dr M. J. Wilsher. E-mail: [email protected] 1. Hoda SA, Resetkova E, Yusuf Y, Cahan A, Rosen PP. Megakaryocytes mimicking metastatic breast carcinoma. Arch Pathol Lab Med 2002; 126: 618–20. 2. Zafar N. Megakaryocytes in sentinel lymph node — a potential source for diagnostic error. Breast J 2007; 13: 308–9. 3. Millar EK, Inder S, Lynch J. Extramedullary haematopoiesis in axillary lymph nodes following neoadjuvant chemotherapy for locally advanced breast cancer—a potential diagnostic pitfall. Histopathology 2009; 54: 622– 3. 4. Takhar AS, Ney A, Patel M, Sharma A. Extramedullary haematopoiesis in axillary lymph nodes following neoadjuvant chemotherapy for locally advanced breast cancer. BMJ Case Rep 2013; May 22: pii: bcr2013008943. 5. Prieto-Granada C, Setia N, Otis CN. Lymph node extramedullary hematopoiesis in breast cancer patients receiving neoadjuvant therapy: a potential diagnostic pitfall. Int J Surg Pathol 2013; 21: 264–6. 6. Kaufman RM, Airo R, Pollack S, Crosby WH. Circulating megakaryocytes and platelet release in the lung. Blood 1965; 26: 720–31. 7. Sahoo S, Lester SC. Pathology of breast carcinomas after neoadjuvant chemotherapy: an overview with recommendations on specimen processing and reporting. Arch Pathol Lab Med 2009; 133: 633–42.


CORRESPONDENCE

8. Xu X, Roberts SA, Pasha TL, Zhang PJ. Undesirable cytokeratin immunoreactivity of native nonepithelial cells in sentinel lymph nodes from patients with breast carcinoma. Arch Pathol Lab Med 2000; 124: 1310–3.

DOI: 10.1097/PAT.0000000000000130

Phyllodes tumour of the nipple presenting with nipple discharge Sir, Phyllodes tumour (PT) rarely presents with nipple discharge, a finding that typically implies its origin within or in the vicinity of the nipple and its communication with a lactiferous duct.1–5 Herein reported is a case of phyllodes tumour arising within the nipple, presenting as a left nipple mass with nipple discharge. The patient, an otherwise well 43-year-old female, presented with a left nipple mass and nipple discharge. Cytological assessment of the nipple discharge showed abundant foamy macrophages with scattered papilliform groups of epithelial cells with cytological atypia (Fig. 1A). Bilateral mammograms showed moderately dense fibroglandular tissue only. Ultrasound scanning revealed an unusual solid lesion within the left nipple, comprising a main lobulated area 14 mm across with an inferior lobulation at the 6 o’clock position measuring 6 mm across (Fig. 1B). An ultrasound guided fine needle aspiration (FNA) biopsy of the inferior portion was performed (Fig. 1C,D). The direct smears were cellular and contained cohesive bimodal groups with a papilliform configuration as well as occasional bimodal sheets, and a few bare bipolar nuclei in the background along with foamy macrophages. A few tight clusters of cells with cytological atypia were seen, similar to

455

those seen within the preceding nipple smear. No stromal component was included. The features were thought to be suggestive of a papillary lesion, particularly given the location of the lesion. The specimen was excised along with associated nipple ducts through a nipple incision. The lesion protruded through the incision at surgery (Fig. 2A). The smooth surfaced, protruding mass and associated ducts were excised in several pieces, with separate excision of overlying skin. In fact the lesion was so superficial that a scalpel was required to dissect it off overlying skin. The specimens were fixed in 10% neutral buffered formalin. The larger specimen (Fig. 2B) was smooth and lobulated, similar to its ultrasound image outline (Fig. 1B); it measured 18 9 8 mm after fixation and its cut surface was described as mucoid. The smaller duct excision comprised two pieces of firm pale tissue 10 8 5 mm and 7 4 3 mm, respectively, and the skin, 16 5 3 mm, was unremarkable. Histologically, there was a fibroepithelial lesion with an exaggerated leaf-like (phylloid) growth pattern, arising from and occupying dilated nipple ducts (Fig. 2C). There was only focal subepithelial stromal condensation and mitoses were inconspicuous. The surface epithelium frequently was hyperplastic with papillary excrescences (Fig. 2D), which corresponded to the epithelium seen in the nipple smear and FNA samples (Fig. 1A,C,D). The overall features were considered consistent with a benign, predominantly intraductal/cystic phyllodes tumour. The patient developed a small amount of ischaemic nipple skin necrosis post-operatively that soon healed. PTs are rare breast tumours with a characteristic phylloid architecture. Favoured treatment is surgical excision with clear margins, as even benign cases may recur when inadequately resected. They are subclassified as benign, low grade or high grade malignant based on stromal characteristics. They may occasionally contain microcysts, but rarely display an intracystic pattern.3,6 Horiguchi et al.6 described such a case

Fig. 1 (A) Nipple smear cytology with cellular epithelial clusters with cytological atypia (Papanicolaou stain). (B) Nipple ultrasound showing lobulated nipple mass, with main and second lobulated areas indicated. (C) Nipple mass, fine needle aspiration cytology (FNAC) showing proteinaceous background containing foamy macrophages, with clusters of epithelial cells (Giemsa). (D) Nipple mass, FNAC showing epithelial sheets and papilliform clusters and background foamy macrophages (Papanicolaou stain).


456


CORRESPONDENCE

Fig. 2 (A) Intraoperative photograph of lesion in situ revealed after incision of nipple skin. (B) Photograph of fresh main specimen taken immediately after removal. (C) Phyllodes tumour of nipple occupying nipple duct. Smooth muscle of nipple at arrow (H&E). (D) Surface epithelial hyperplasia corresponding to the epithelium seen in nipple smear and FNA specimens (H&E).

arising in the upper outer right breast of a 62-year-old female, measuring over 11 cm, characterised by phylloid papillary protrusions into a central cystic cavity within the lesion itself. Chung et al.7 described a case of ‘intraductal fibroadenomatosis’ arising in the lower outer quadrant of the left breast of a 33-year-old female in which there was an ill defined lesion measuring up to 1.6 cm on ultrasound, comprising broad based fibroadenoma-like polypoid protrusions into the lumens of breast ducts. Although a fibroadenoma was favoured in view of low stromal cellularity, there were areas with stromal overgrowth and phylloid architecture and features also overlapped with an intraduct papilloma and ductal adenoma. In describing two cases of fibroepithelial proliferation admixed with intraduct papilloma, Cummings et al.8 postulated that fibroadenoma and intraduct papilloma may share a common origin. The lesions each arose in the right breast of two female individuals aged 13 and 60 years, respectively. Whereas intraduct papillomata are connected to the walls of ducts by a narrow stalk, these lesions included broad based fibroepithelial lesions protruding into ducts, some with a phylloid architecture. The current case was regarded as a phyllodes tumour in view of its uniform phylloid growth pattern and benign in view of its low stromal cellularity, and lack of cytological atypia or mitoses. PTs usually present within the breast parenchyma, and are often located within the superior breast, are mostly benign and not associated with nipple discharge.1,3,4 PT presenting with nipple discharge has been rarely described and most often is associated with a PT arising within a large subareolar duct, with growth protruding into its lumen.1–5 PT is not surprisingly unexpected in this circumstance. Bloody nipple discharge in particular is more commonly associated with intraduct papilloma and carcinoma.1 In the current case the mass occupied the nipple and was associated with non-bloody nipple discharge. PT at the nipple may even protrude through the nipple itself.2,4

Apart from a case of a 15 mm diameter intraductal benign PT occurring in a 45-year-old female, presenting with clear nipple discharge and in which there was a separate focus of atypical ductal hyperplasia,3 other cases presenting with (brown or bloody) nipple discharge have been of intraductal PT arising in juvenile female patients aged 11–13 years. These ranged in size from 21 mm to 45 mm, with no recurrence at variable but often short follow-up of 6 months to, in one case, 8 years and 10 months.1–5 Complete excision was recorded in two cases, with short follow-up of 18 months and 6 months only.4,5 In one case, the bloody nipple discharge was associated with partial haemorrhagic infarction of a benign PT.1 Ultrasound heterogeneity and lesional cystic change may help distinguish PT from a fibroadenoma.1 However, presentation as a lesion in or near the nipple means intraduct papilloma is still more likely to be considered. Lian et al.3 stressed that PTs presenting as intraductal lesions means they mimic intraduct papilloma on ultrasound and any resultant microdochotomy is not ideal management for lesions that require clear margins, no matter the histological grade, and that in arising along duct walls, their delineation may be difficult to appreciate surgically and even histologically. The current case extended to surgical margins and minimal non-lesional tissue was resected because an intraduct papilloma, not a PT, was expected. Acknowledging there is a possibility of recurrence, because the features were resolutely benign, the patient will be closely monitored to assess for any future recurrence. Conflicts of interest and sources of funding: The authors state that there are no conflicts of interest to disclose. Mark Wilsher1 Merran McKessar2 Cindy Mak3


CORRESPONDENCE 1

Douglass Hanly Moir Pathology, Macquarie Park, 2Mater Imaging, Wollstonecraft, and 3Poche Centre, North Sydney, NSW, Australia Contact Dr M. Wilsher. E-mail: [email protected] 1. Tagaya N, Kodaira H, Kogure H, et al. A case of phyllodes tumour with bloody nipple discharge in a juvenile patient. Breast Cancer 1999; 6: 207–10. 2. Martino A, Zamparelli A, Santinelli G, et al. Unusual clinical presentation of a rare case of phyllodes tumor of the breast in an adolescent girl. J Pediatr Surg 2001; 36: 941–3. 3. Lian D, Cheah E, Tan PH, et al. Phyllodes tumour with intraductal growth: a rare cause of nipple discharge. Histopathology 2007; 50: 666–9. 4. Kanka KC, Sawicki JE, Svahn JD. An unusual case of phyllodes tumour presenting as a trans-nipple tumour in a 13-year-old female. Breast J 2009; 15: 657–8. 5. Hayano F, Yamada S, Nakano S, et al. Intraductal fibroadenoma under the nipple in an 11-year-old female. Diag Pathol 2014; 9: 32. 6. Horiguchi J, Lino Y, Aiba S, et al. Phyllodes tumor showing intracystic growth: a case report. Jpn J Clin Oncol 1998; 28: 705–8. 7. Chung A, Scharre K, Wilson M. Intraductal fibroadenomatosis: an unusual variant of fibroadenoma. Breast J 2008; 14: 193–5. 8. Cummings MC, da Silva L, Papadimos DJ, et al. Fibroadenoma and intraduct papilloma – a common pathogenesis. Virchows Arch 2009; 455: 271–5.

457

small amount of decidualised endometrium, normal superficial myometrium and squamous epithelium. Immunohistochemistry showed CD68 and MAC 387 were positive in the epithelioid macrophages and giant cells and CD10 and oestrogen receptor were negative (Fig. 1C). Ziehl–Neelsen, Fite, methenamine silver and PAS stains were negative for organisms and no birefringent material was seen on polarised light. Case 2 was a 50-year-old para 2, gravida 2 who was referred with a vulval lesion diagnosed as VIN 3 on biopsy. She had a past history of CIN 3 managed 10 years previously by cone

DOI: 10.1097/PAT.0000000000000136

Endometrial granulomas with Levonorgestrel Intrauterine System (Mirena) Sir, The Levonorgestrel Intrauterine System (Mirena) is associated with various morphological changes in the endometrium. Most commonly, endometrial glands are atrophic, the stroma is decidualised, localised changes of inflammatory infiltrates and micropolyps attributed to the presence of an intrauterine device may be present and the endocervix shows microglandular hyperplasia.1 Less common changes in the endometrial glands are various types of metaplasia, atypia resembling the Arias-Stella phenomenon, proliferative or secretory activity or hyperplasia. Less common stromal responses are necrosis or infarcted decidua, mucinosis, hyaline nodules and fibrosis. Its widespread use as both a contraceptive and in the management of menorrhagia requires pathologists be aware of these changes. We have recently observed endometrial granulomas in two women exposed to Levonorgestrel Intrauterine System (IUS), a change to our knowledge not previously described. Case 1 was a 41-year-old, para 1, gravida 1 with irregular vaginal bleeding. She had been using a levonorgestrel IUS for contraception and management of menorraghia for 8 years. A dilatation and curettage (D&C) specimen before the first levonorgestrel IUS was inserted showed normal proliferative endometrium and a second D&C 5 years later showed weakly secretory endometrium. Upon her current presentation, she was managed with dilatation and curettage with replacement of her levonorgestrel IUS. On 6 week follow-up her abnormal uterine bleeding had resolved. The curettings included a 1.5 mm nodule of amorphous eosinophilic material. The amorphous eosinophilic material was rimmed by several layers of epithelioid macrophages and occasional giant cells (Fig. 1A,B). Also present were a

Fig. 1 Case 1 shows (A) amorphous eosinophilic material lined by a rim of histiocytes (H&E), (B) a multinucleate giant cell within the eosinophilic material (H&E), and (C) macrophages confirmed by positive CD68 staining (immunoperoxidase).


458


CORRESPONDENCE

biopsy. Following cervical stenosis and menorrhagia nonresponsive to medroxyprogesterone (Provera) 9 years ago, she was managed with a levonorgestrel IUS. She has remained on her levonorgestrel IUS with oral and topical vaginal oestrogen since, with no complaints. The VIN was excised and at the same time a D&C with replacement of the levonorgestrel IUS was performed. The curettings included three nodules of amorphous eosinophilic material up to 2 mm in diameter. Each nodule was rimmed by a wall of several layers of epithelioid macrophages with occasional giant cells. Lymphocytes, plasma cells, eosinophils and neutrophils were seen amongst the macrophages. A fourth nodule consisted of a similar multilayered wall of epithelioid macrophages but these lined a collapsed space. Also present were a small amount of decidualised endometrium, normal superficial myometrium and inflamed endocervical tissue. The same immunohistochemistry results, negative special stains for organisms and absence of polarisable material as in Case 1 were found. In both cases a diagnosis of endometrial granulomas was made on the findings of epithelioid histiocytes surrounding amorphous eosinophilic material. The histiocytic nature of the cells was confirmed on immunohistochemistry (IHC) by positive CD68 and MAC 387 and negative CD10 and ER. There was no evidence of infection clinically or on special stains. Previous studies of the effects of levonorgestrel IUS on endometrial morphology have reported hyaline nodules and infarcted pseudodecidualisation in 7.5% and 19.8% of cases, respectively.2 We hypothesise that the amorphous eosinophilic material is the residuum of necrotic decidua in the setting of very high endometrial progestagen concentration found with levonorgestrel.3 The lack of menstruation in patients on levonorgestrel IUS is the reason why this amorphous eosinophilic material remains in the endometrium. In some cases, it attracts a granulomatous reaction analogous to other granulomatous reactions to necrotic tissue, e.g., post-diathermy granulomatous reactions. So as not to confuse them with other causes of endometrial granulomas, pathologists need to be aware of this type of granuloma in patients on levonorgestrel IUS. Conflicts of interest and sources of funding: The authors state that there are no conflicts of interest to disclose. Fletcher Charlton1 Geoffrey Otton2 Andrew Hedges3 James Scurry4 1

School of Medicine and Public Health, University of Newcastle, Callaghan, 2Hunter Centre for Gynaecological Cancer, 3Department of Gynaecology, and 4Division of Anatomical Pathology, Hunter Area Pathology Service, John Hunter Hospital, New Lambton Heights, NSW, Australia Contact Associate Professor J. Scurry. E-mail: [email protected] 1. Silverberg SG, Haukkamaa M, Arko H, Nilsson CG, Luukkainen T. Endometrial morphology during long-term use of levonorgestrel-releasing intrauterine devices. Int J Gynecol Pathol 1986; 5: 235–41. 2. Hejmadi RK, Chaudhri S, Ganesan R, Rollason TP. Morphologic changes in the endometrium associated with the use of the mirena coil: a retrospective study of 106 cases. Int J Surg Pathol 2007; 15: 148–54.

3. Nilsson C, Haukkamaa M, Vierola H, Luukkainen T, Arcangeli P. Tissue concentrations of levonorgestrel in women using a levonorgestrel-releasing IUD. Clin Endocrinol 1982; 17: 529–36.

DOI: 10.1097/PAT.0000000000000127

Multiple squamoproliferative lesions arising in the setting of chronic graft-versus-host disease Sir, Graft-versus-host disease (GvHD) is an immune-mediated inflammatory disease. It is a manifestation of histoincompatibility which occurs when immunologically competent cells are introduced into an immunologically incompetent host. The greater the histoincompatibility of the donor and recipient match, the greater the severity of the disease. Although GvHD generally develops after allogeneic bone marrow transplantation, it can also occur after transfusion of non-irradiated blood products or as a consequence of passenger leukocytes after solid organ transplantation. In GvHD a response is mounted by incompatible immune competent donor cells against certain organs of the host such as the liver, the gastrointestinal system and the skin.1 The reaction can be either acute or chronic phases. Acute GvHD is a syndrome characterised by dermatitis, hepatitis and enteritis which develops within 100 days after allogeneic bone marrow transplantation. Chronic GvHD, on the other hand, is a more complex syndrome which develops after day 100. Chronic cutaneous GvHD has a multitude of clinical presentations with lichenoid papules and sclerodermoid plaques being the most common manifestations.1 Of patients who develop chronic GvHD, 70–80% will previously have had acute GvHD. De novo chronic GvHD represents the remainder. De novo chronic GvHD does not appear to have a negative impact on patient survival.2 This report describes a case of multiple squamous cell carcinomas and benign squamoproliferative lesions in a patient with chronic lichenoid and sclerodermoid chronic GvHD, details the cutaneous changes in this condition and discusses the possible association between chronic GvHD and the squamoproliferative lesions. A 64-year-old man underwent stem cell transplantation in 2002 for diffuse large B-cell non-Hodgkin lymphoma. Stem cell transplantation was required because the lymphoma which had been originally diagnosed in 1998 was refractory to standard chemotherapy. Following stem cell transplantation, he developed GvHD which was initially treated with immunosuppression (cyclosporin) and corticosteroids. He also received Palquenil. He is currently on corticosteroids only. During the last 10 years he has had multiple skin biopsies and excisions for skin tumours, mostly on his lower limbs. Only lesions with tumour-like features have been biopsied. Currently he also has numerous skin plaques on his shoulders, arms and legs. The lymphoma is under control and he is currently in remission. Histologically, the skin tumours were squamous cancers with varying degrees of differentiation (Fig. 1). All the skin biopsies, particularly the ones from the right lower leg also showed chronic GvHD of both lichenoid and sclerodermoid variants


CORRESPONDENCE

459

Fig. 1 (A,B) The 2004 biopsy of a nose lesion showing an invasive well differentiated squamous cell carcinoma. (A) H&E stained section showing the overall crateriform architecture. The adjacent skin shows slight epidermal thickening and some dermal elastosis. A dark zone representing a lymphocytic infiltrate is seen at the base of the tumour. (B) H&E stained section, detailing the infiltrating squamous cell and the lymphocytic infiltrate at the periphery and within the squamous islands. The peritumoural lymphocytic infiltrate has a lichenoid pattern and a few apoptotic tumour cells can also be seen. (C,D) The 2005 biopsy of anterior leg lesion showing an invasive poorly differentiated squamous cell carcinoma. (C) H&E stained section showing surface ulceration (left) and dermal infiltration by squamous cancer cells. A heavy lymphocytic infiltrate which merges with tumour can also be seen. The adjacent skin (right) shows marked epidermal hyperplasia and a heavy lymphocytic infiltrate with a lichenoid pattern. This reaction which represents lichenoid chronic GvHD resembles hypertrophic lichen planus. (D) H&E stained section, showing the pleomorphic squamous cells and the heavy lymphocytic infiltrate at the periphery and within the malignant squamous islands.

(Fig. 2). The first skin cancer which appeared in 2004 was on the skin of the nose. It was a well differentiated squamous cell carcinoma (Fig. 1A,B). The tumour had a keratoacanthomatous growth pattern with an infiltrating deep margin and extension into the deep dermis. Peritumoural and intratumoural lymphocytic infiltrates were also seen. The uninvolved skin showed mild solar damage (epidermal keratosis and solar elastosis). This was followed in 2005 by poorly differentiated squamous cell carcinoma on the right lower leg. This tumour also showed lymphocytic infiltration around and within the tumour (Fig. 1C,D). The uninvolved skin showed lichenoid chronic GvHD but no signs of solar damage. Between 2005 and 2013 the patient had a number of other proliferative skin lesions, ranging from pseudoepitheliomatous hyperplasia (Fig. 2B) to squamous cell carcinoma. The non-neoplastic skin showed a mixture of lichenoid and sclerodermoid changes (Fig. 2). The lichenoid tissue response was characterised by interface dermatitis, taking the form of lymphocytic inflammation at the epidermodermal junction, associated with keratinocyte necrosis/apoptosis (Fig. 2C). Marked apoptotic activity was focally present. Lymphocyte satellitosis (lymphocytes surrounding apoptotic keratinocytes in the epidermis or the skin appendages) and vacuolar change in the basal layer were also seen (Fig. 2C). Other epidermal changes (thickening/hyperplasia with saw-tooth rete ridges and at times pseudoepitheliomatous hyperplasia) were also noted (Fig. 2B). The lymphocytic infiltrate was polyclonal, made up of a mixture of helper and cytotoxic T cells and B cells. Lymphocyte markers were performed to exclude recurrent lymphoma or emergence of new lymphoma. Infective causes were excluded by special stains for

ordinary bacteria, acid-fast bacilli and fungi by Gram, Ziehl– Neelsen and Fite, and periodic acid-Schiff and Grocott stains, respectively. The sclerodermoid changes which were always associated with some lichenoid inflammation, were characterised by diffuse dermal sclerosis (Fig. 2A). The dermis and the underlying subcutaneous tissue were homogenised and sclerotic. The adnexal units had almost completely disappeared (Fig. 2A) The histological manifestations of chronic cutaneous GvHD evolve over time. These manifestations are often modified by treatment and to some extent overlap with those of acute GvHD.3,4 Although in most patients GvHD starts as an acute phase during the first 3 months after transplantation, not all cases of chronic GvHD are preceded by acute GvHD. In this patient no biopsies were taken during the acute phase of the GvHD but the chronic disease has been biopsied several times, mainly because of the squamoproliferative changes that have ranged from pseudoepitheliomatous epidermal hyperplasia to squamous cancers. The cancers varied from well differentiated with keratoacanthomatous growth pattern to poorly differentiated (Fig. 1). Most of the cancers have arisen in non-solar damaged skin. This case is unusual and to some extent unique because a number squamous neoplasms have arisen in non-solar damaged skin affected by chronic GvHD. The other important aspect of the case is the exuberant squamous epithelial hyperplasia. While some of this was pseudoepitheliomatous, others appeared to be associated with lichenoid inflammation and part of the linchenoid chronic GvHD. The differential diagnosis of this type of tissue reaction includes hypertrophic lichen


460

CORRESPONDENCE


Fig. 2 Another leg lesion with chronic GvHD showing lichenoid and sclerodermoid patterns. (A) H&E stained section showing marked epidermal hyperplasia (right), a moderate lymphocytic infiltrate at the epidermodermal junction (left of centre) with a lichenoid reaction pattern and dermal hyalinisation and loss of skin adnexal units (sclerodermoid pattern). No solar damage is apparent. (B) H&E stained section showing the overall squamoproliferative architecture. A dark zone of lymphocytic cells is seen at the interface between the proliferative squamous cells and the dermal stroma (right). (C) H&E stained section showing interface inflammation and focal epidermal cell necrosis (apoptosis) with accumulation of lymphocytes around apoptotic keratinocytes (satellitosis).

planus.5 This variant of lichen planus which has similar morphological features also affects the anterior leg. In hypertrophic lichen planus the basal cell damage is usually confined to the tips of the rete ridges and may be missed on casual observation. The infiltrate in lichenoid chronic GvHD is not as dense or as band-like as in the usual lesions of lichen planus. A few eosinophils and plasma cells may be seen in some cases, particularly when the patient is taking medications such as betablockers.5 Hypertrophic lupus erythematosus is another consideration when assessing chronic GvHD. In this subtype of lupus, characteristic microscopic features of lupus (hyperkeratosis with follicular plugging, hydropic degeneration of the basal layer, scattered apoptotic keratinocytes in the basal layer or in the epithelium and thickening of the basement membrane), together with acanthosis and hyperkeratosis of the epidermis, are usually present. The dermal lichenoid or patchy lymphocytic infiltrate is accentuated around the pilosebaceous follicles. There is interstitial mucin deposition and oedema and usually no eosinophils and neutrophils are present. Direct immunofluorescent examination of the affected skin shows deposition of IgG and C3, and to a lesser extent of IgM, IgA and C1 in a diffuse irregular band at the dermoepidermal junction. Lupus has a predilection for light-exposed areas such as the forehead, nose, cheeks, upper part of the back, upper

chest, and dorsal aspects of the hands and feet and eyelids and ears.6 Cancer is becoming an important cause of morbidity and mortality after transplantation. Several types of cancers are encountered in the post-transplant setting.7–9 Rarely, the donor has an undiagnosed cancer which can pass to the recipient. Relapse of the original cancer and post-transplant lymphoproliferative disorder are relatively common. Relapse can happen a few months to a few years after transplantation. It is rarely seen 5 or more years after transplantation. Risk of de novo cancer after transplantation is 2–3-fold compared with the general population of the same age and sex. Younger recipients experience a greater relative increase in risk compared with older recipients. Post-transplant secondary cancers may include solid tumour cancers, often of the skin, mouth, brain, liver, cervix, thyroid, breast and bone.7–9 Reasons for the increased risk are likely to be due to the interplay of several factors, including the nature of the transplanted organ and exposure to viral infections. Infection with viruses such as Epstein–Barr (EBV), cytomegalovirus (CMV), hepatitis B (HBV) or hepatitis C (HCV) predispose to cancer development. Other risk factors include total body irradiation and high dose chemotherapy either before transplantation or as part of the conditioning treatment. Chronic GvHD has been shown to be an independent risk factor for all solid cancers and especially those of the oral


CORRESPONDENCE

cavity.9 In the current case, chronic GvHD and immunosuppression are likely to be important in cancer development. Pseudoepitheliomatous hyperplasia (PEH) is a benign process characterised by hyperplasia of the epidermal and adnexal epithelium. PEH closely simulates squamous cell carcinoma, clinically and pathologically.10 PEH may be present in a number of conditions characterised by prolonged inflammation or infection as well as in association with many cutaneous neoplasms. PEH can be especially difficult to distinguish from squamous cell carcinoma, even after examining multiple levels and additional samples.11 Special stains and immunoperoxidases are usually unhelpful. p53 protein overexpression can help distinguish between the two conditions but interpretation of p53 is not clear cut.10 No single histological feature is pathognomonic of GvHD. Furthermore, the diagnosis of GvHD can be complicated by the fact that other eruptions, such as engraftment syndrome, viral exanthems and drug eruptions can also occur after transplantation. These can have similar histopathological features to GvHD, making clinical correlation necessary and complex. The timing and administration of all medications is important when a diagnosis of GvHD is considered. However, a number of histological patterns are recognised in GvHD. The minimal histological criteria for active GvHD require apoptosis within the lower layers of the epidermis.11 The hallmark of both acute and chronic GvHD is interface dermatitis with a lichenoid inflammatory reaction pattern. Lichenoid tissue reaction takes the form of lymphocytic inflammation at the epidermodermal junction, usually associated with keratinocyte necrosis (apoptosis). Marked apoptotic activity, which is defined by the presence of more than five epidermal apoptotic bodies per section from a 4 mm punch biopsy, is often present. Lymphocyte satellitosis recognised by lymphocytes surrounding apoptotic keratinocytes in the epidermis or the skin appendages is an important feature of the lichenoid tissue response of GvHD and vacuolar change in the basal layer can be an important clue to the diagnosis. The epidermal changes in lichenoid reaction include thickening (hyperplasia) with saw-tooth rete ridges.3,5,6 The skin lesions of chronic GvHD are not synchronous. The presence or absence of chronic GvHD features in a biopsy can be influenced by sampling or partial thickness sample. A lesser degree of this combination can occur occasionally in severe clinical acute GvHD. Therefore, clinical correlation is crucial when skin biopsies from patient with GvHD are being evaluated.12 In the sclerotic phase which is a late stage in chronic GvHD, fibrosis has a top-down progression from the papillary through the reticular dermis. Some patients develop diffuse dermal sclerosis without an apparent inflammatory lichenoid phase. However, some interface inflammation is usually noted. The suggested minimal criterion for the diagnosis of cutaneous sclerotic chronic GvHD is homogenisation (sclerosis) of most of the papillary dermis or reticular dermis. In the morphoeic variant of chronic cutaneous GvHD, sclerosis may be largely confined to the reticular dermis and underlying fascia with little or no epidermal disease. In fasciitis type, which is another variant of sclerotic GvHD, there is fibrous thickening only in the fascia, with adjacent inflammation but without any epidermal or dermal involvement.12 Sampling and technical factors can cause a false negative histological assessment of GvHD because biopsy represents a snapshot in time of a complex and dynamic biological process. The duration of disease activity, the use of immunosuppressive

461

therapy, the possibility of existence of more than one process at a given time, the location and the quality of the sample and the histological preparation can influence biopsy interpretation. The timing of biopsies and their relationship to treatment are highly variable. The key indicator of activity is often blunted by the potent immunosuppressive treatment. Infections and drug reactions can mimic chronic GvHD. Residual destruction of epithelia or glandular structures and irreversible fibrosis pose problems in separating old damage from ongoing disease activity. Given the high prevalence of chronic GvHD in the population of interest with up to 80% of allogenic stem cell recipients surviving beyond 100 days, exhibiting signs of chronic GvHD, the positive predictive value of a positive biopsy for GvHD is high.2 The negative predictive value of a biopsy, on the other hand, can be lower. After treatment, skin biopsy samples display a combination of residual damage, with loss of rete ridges and dermal appendages. An increase in papillary or dermal sclerosis and a reduction in the lichenoid inflammatory response are also seen. Immunosuppression with corticosteroids forms the basis of firstline therapy in both acute and chronic GvHD, producing sustained responses. In steroid resistant cases antithymocyte globulin, the humanised monoclonal anti-CD52 antibody (alemtuzumab), anticytokine inhibitors, globulin, extracorporeal photophoresis, and sirolimus, pentostatin, low-dose methotrexate, rituximab, thalidomide and imatinib are also used, particularly in refractory chronic GvHD.1,2,12 The hallmark of an incomplete response is residual apoptotic changes in the epidermis or appendages. After treatment, minor residual perivascular lymphocytic inflammation or persistent epidermal vacuolar degeneration requires additional study, as does the assessment of activity in patients who have received treatment (psoralen and UVA irradiation) or who have established deep dermal sclerosis or morphoeic chronic GvHD. Apoptosis of keratinocyte without other chronic GvHD in screening skin biopsies does not indicate chronic GvHD and does not necessarily predict that a flare may follow cessation of therapy.11 GvHD remains a very important source of morbidity and mortality in allogeneic transplant patients. In order to understand prognosis and the type of response to treatment in patients with GvHD, it is important to objectively identify and validate the markers capable of predicting the pathological processes in patients with chronic GvHD. To confirm the diagnosis and establish the stage of the disease, biopsies of the affected organ are often needed. Serial skin biopsies are sometimes required because certain histological findings classically attributed to GvHD can also be seen in patients not clinically affected by GvHD and no single parameter has been shown to achieve statistical significance. The utility of serial biopsies in judging the response to treatment has not been fully determined. Therefore, skin biopsies alone after allogeneic BMT are of limited use in predicting progression of clinical disease.11 Conflicts of interest and sources of funding: The author states that there are no conflicts of interest to disclose. Ibrahim Zardawi University of Newcastle, and Douglass Hanly Moir Pathology, Gateshead, NSW, Australia Contact Professor I. Zardawi. E-mail: [email protected]; izardawi@dhm. com.au


462

CORRESPONDENCE

1. Ratanatharathorn V, Ayash L, Lazarus HM, et al. Chronic graft-versus-host disease: clinical manifestation and therapy. Bone Marrow Transplant 2001; 28: 121–9. 2. Wagner JL, Seidel K, Boeckh M, Storb R. De novo chronic graft-versushost disease in marrow graft recipients given methotrexate and cyclosporine: risk factors and survival. Biol Blood Marrow Transplant 2000; 6: 633– 9. 3. Shulman HM, Kleiner D, Lee SJ, et al. Histopathologic diagnosis of chronic graft-versus-host disease: National Institutes of Health Consensus Development Project on Criteria for Clinical Trials in Chronic Graft-versus-Host Disease: II. Pathology Working Group Report. Biol Blood Marrow Transplant 2006; 12: 31–47. 4. Pen˜as PF, Zaman S. Many faces of graft-versus-host disease. Australas J Dermatol 2010; 51: 1–10. 5. Murphy GF, Saavedra AP, Mihm MC. Atlas of Non-tumour Pathology: Inflammatory Disorders of the Skin. Silver Spring: American Registry of Pathology, 2012; 190–226. 6. Bardek I, Basta-Juzbasˇ ic´ A, Marinovic´ B, et al. Hypertrophic lupus erythematosus: case report. Acta Dermatovenerol Croat 2009; 17: 134–8. 7. Brown SJ, Jackson GH, Reynolds NJ, Lawrence CM. Chronic cutaneous graft-versus-host disease associated with multiple cutaneous squamous cell carcinomas. Clin Exp Dermatol 2006; 31: 472–3. 8. Yokota A, Ozawa S, Masanori T, et al., Kanto Study Group for Cell Therapy (KSGCT). Secondary solid tumors after allogeneic hematopoietic SCT in Japan. Bone Marrow Transplant 2012; 47: 95–100. 9. Demarosi F, Soligo D, Lodi G, et al. Squamous cell carcinoma of the oral cavity associated with graft versus host disease: report of a case and review of the literature. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2005; 100: 63–9. 10. Zayour M, Lazova R. Pseudoepitheliomatous hyperplasia: a review. Am J Dermatopathol 2011; 33: 112–22. 11. Massi D, Fondi C, Nozzoli C, et al. The impact of histopathologic examination of graft-versus-host disease in the era of reduced-intensity conditioning regimen: a study from the Gruppo Italiano Trapianto di Midollo Osseo. Hum Pathol 2011; 42: 254–68. 12. Skert C, Patriarca F, Sperotto A, et al. Sclerodermatous chronic graftversus-host disease after allogeneic hematopoietic stem cell transplantation: incidence, predictors and outcome. Haematologica 2006; 91: 258–61.

DOI: 10.1097/PAT.0000000000000134

Exophytic schneiderian carcinoma of the lacrimal sac: first case report Sir, A 66-year-old woman presented at the department of otorhinolaryngology of the Saarland University Hospital with a slowly growing tumescence at the medial corner of the left eye, noticed by her for the first time about 1 year before visiting our clinic (Fig. 1A). During the medical history she denied any pain or visual disturbance but reported incremental epiphora for several months. Besides hypertension treated with captopril and hydrochlorothiazide the patient was free of other diseases. On ultrasound examination a 29 24 22 mm, equally margined tumour could be found with a homogeneous interior structure and a high level of perfusion (Fig. 1B). Computed tomography (CT) scan of the head revealed a tumour completely filling the lacrimal sac with a small branch reaching into the nasolacrimal duct. However, the tumour showed no signs of infiltrating orbital or adjacent osseous structures indicating a non-destructive growth pattern (Fig. 1C). Considering the patient’s enduring complaints and radiological findings it was decided on surgical treatment without ensuring histopathological diagnosis by a pre-operative biopsy. In detail, after having opened the lacrimal sac by a palpebral margin incision, the tumour was completely removed, followed


by an endonasal dacryocystorhinostomy and U-intubation of the nasolacrimal duct. Morphologically, the resected tumour showed a papillomatous, homogeneously beige coloured surface and a tuberous rigidity (Fig. 1D). Histological sections of the formalin fixed, paraffin embedded (FFPE) tumour tissue depicted a predominantly exophytic and partially inverted papillomatous growth pattern of a keratinising squamous epithelium partially appearing as transitional with a distinct disturbance of maturation (Fig. 2A). Cells showed a clearly deranged nucleus to plasma ratio up to superficial layers without any signs of basement membrane infiltration. According to the predominant growth pattern and cytomophological characteristics, the tumour was classified as exophytic schneiderian carinoma in situ. The results of immunhistochemical analyses are shown in Table 1. Confirming the morphological diagnosis of a schneiderian carcinoma, the tumour cells showed moderate positivity for CK5/6 without any expression of CK7 or CK20. The strong expression of p63 as p53 target gene as well as the overexpression of p53 itself has repeatedly been described before in the literature for this tumour entity1,2 and thus confirmed the histological diagnosis. Interestingly, the tumour cells also showed an intense BCL2 expression, which is actually a characteristic of lymphomas and has not yet been described in sinunasal papillomas or carcinomas. In order to evaluate the influence of human papillomavirus (HPV) in the carcinogenesis of the tumour, a simultaneous p16-Ki-67 immunostaining was performed using the CINtecPLUS kit (Roche, Switzerland) and showed moderate positivity for p16, especially basal cells as well as Ki-67 positive cells spread over all cell layers. However, there were no cells showing simultaneous expression of both markers as would be typical for HPV induced neoplasias (Fig. 2B). Accordingly, HPV DNA could neither be detected by polymerase chain reaction (PCR) nor by in situ hybridisation of the FFPE tumour tissue. After surgical treatment, the patient stayed in hospital for 2 days with no post-operative complications. After having received the histopathological results confirming all side tumour free resection borders, the post-operative plan was discussed in an interdisciplinary tumour conference. Considering the R0 status and the histopathological classification of the tumour as carcinoma in situ, adjuvant radiation or chemotherapy were omitted. Generally, one can distinguish between three different subtypes of schneiderian papillomas of the sinunasal and lacrimal system: the most frequent inverted type accounting for 62%, an exophytic type accouting for 32% and a very rare oncocytic type accounting for 6% of all cases.3 Malignant transformation into schneiderian carcinomas occurs in a minority of cases with an emphasis on the inverted and oncocytic subtype.4 Malignant transformation of exophytic papillomas is extremely rare and there are only three cases of exophytic schneiderian carcinomas of the sinunasal cavity described in literature.5–7 So far, no case of an exophytic schneiderian carcinoma of the lacrimal sac has been reported. Concerning the carcinogenesis of this tumour entity, several studies exist that could identify HPV infection as a causative agent in the development of schneiderian papillomas with an overall prevalence of HPV DNA in 10–12% of all cases.8,9 Therefore, HPV positivity seems to be associated with a lower risk of malignant transformation and indicates a better prognosis.9 In contrast to other HPV associated malignancies,


CORRESPONDENCE

463

Fig. 1 (A) Clinical presentation of the tumour as slowly growing tumescence at the medial corner of the left eye. (B) On ultrasound examination we found a hypervascularised tumour, measuring 22 29 34 mm, (C) which was confirmed by computed tomography of the head. (D) After surgical resection, the tumour showed a beige polypoid surface.

A

B

Fig. 2 (A) H&E stained histological preparations of the tumour tissue showed a predominantly exophytic and partially inverted papillomatous growth pattern of a keratinising squamous epithelium partially appearing as transitional with a distinct disturbance of maturation. (B) Immunohistochemical dual stain of Ki-67 and p16 showed a distinct positivity for p16, especially in the basal cell layers and Ki-67 positive cells spread out to all cell layers. Notably, there were no p16-Ki-67 double positive cells, indicating that the increased cellular proliferation rate was not induced by human papillomavirus infection.

there is no significant correlation of p16 expression and HPV DNA detection in schneiderian papillomas2 which is supported by the results of p16 immunohistochemistry and HPV PCR in the reported case. The clinical management of schneiderian papillomas is quite challenging as the risk of transformation into carcinoma is Table 1 Antigen P16 Ki-67 CK5/6 CK7 CK20 P63 P53 Bcl2

Immunohistochemical characterisation Reactivity þþ þ þþ þþþ þþþ þþþ

, negative; þ, weakly positive; þþ, moderately positive; þþþ, strongly positive.

difficult to predict considering the versatile influence of histological subtype, HPV status and molecular markers.1,2 Given that the risk of malignant transformation is estimated at about 10%,4,5 most authors prefer a therapeutic concept comparable to that of malignancies of the nasolacrimal system, with an emphasis on surgical treatment. In the case of incompletely resected or metastasised tumours, radiation or radiochemotherapy is recommended.10,11 For clinical follow-up, the patient will be examined including ultrasound imaging, while applying the same intervals as malignancies because the risk of recurrence in this case is presumably high considering (1) the negative HPV status, (2) the histological finding of high grade dysplasia, (3) the overexpression of p53 and (4) the high rate of recurrence in this entity in general.12 For the first 4 months of follow-up, the patient has been free of recurrence. Acknowledgements: The authors would like to acknowledge the contribution of Ulrike Bechtel and Monika Hoffmann in performing immunohistochemical staining.


464

CORRESPONDENCE

Conflicts of interest and sources of funding: The authors state that there are no conflicts of interest to disclose. Maximilian Linxweiler1 Bernhard Schick1 Andrea Hasenfus2 Sigrun Smola3 Basel Al-Kadah1 1

Department of Otorhinolaryngology, 2Department of General and Surgical Pathology, and 3Institute of Virology, Saarland University Medical Center, Homburg/Saar, Germany Contact Dr M. Linxweiler. E-mail: [email protected] 1. Gujrathi C, Pathak I, Freeman J, Asa S. Expression of p53 in inverted papilloma and malignancy associated with inverted papilloma. J Otolaryngol 2003; 32: 48–50. 2. Cheung MF, Lau TWS, Cheung LKN, et al. Schneiderian papillomas and carcinomas: A retrospective study with special reference to p53 and p16 tumor suppressor gene expression and association with HPV. Ear Nose Throat J 2010; 89: E5–12. 3. Barnes L, Tse LLY, Hunt JL. Schneiderian papillomas. In: Barnes L, Eveson JW, Reichart P, Sidransky D, editors. World Health Organization Classification of Tumors. Pathology and Genetics of Head and Neck Tumors. Lyon: IARC Press, 2005; 28–32. 4. Vorasubin N, Vira D, Suh JD, et al. Schneiderian papillomas: comparative reviw of exophytic, oncocytic, and inverted types. Am J Rhinol Allergy 2013; 27: 287–92. 5. Terada T. Malignant transformation of exophytic Schneiderian papilloma of the nasal cavity. Pathol Int 2012; 62: 199–203. 6. Buchwald C, Franzmann MB, Jacobsen GK, et al. Carcinomas occuring in papillomas of the nasal septum associated with human papilloma virus (HPV). Rhinology 1997; 35: 74–8. 7. Norris HJ. Papillary lesions of the nasal cavity and paranasal sinuses. Part 1. Exophytic (squamous) papilloma. A study of 28 cases. Laryngoscope 1962; 72: 1797–801. 8. Hoffmann M, Klose N, Gottschlich S, et al. Detection of human papillomavirus DNA in benign and malignant sinunasal neoplasms. Cancer Lett 2006; 239: 64–70. 9. Kim JY, Yoon JK, Citardi MJ, et al. The prevalence of human papillomavirus infection in sinunasal inverted papilloma specimens classified by histological grade. Am J Rhinol 2007; 21: 664–9. 10. Parmar DN, Rose GE. Management of lacrimal sac tumours. Eye (Lond) 2003; 17: 599–606. 11. El-Sawy T, Frank SJ, Hanna E, et al. Multidisciplinary management of lacrimal sac/nasolacrimal duct carcinomas. Ophthal Plast Reconstr Surg 2013; 29: 454–7. 12. Elner VM, Burnstine MA, Goodman ML, Dortzbach RK. Inverted papillomas that invade the orbit. Arch Ophthalmol 1995; 113: 1178–83.

DOI: 10.1097/PAT.0000000000000129

IgG4-related dacryoadenitis evolving into an extra-nodal, marginal zone B-cell lymphoma (EMZL): a tale of two lacrimal glands Sir, IgG4-related disease (IgG4-RD) is a distinct clinicopathological entity characterised by abundance of IgG4-positive plasma cells in the affected tissues and sometimes with raised serum IgG4 levels. Recent literature suggests an association between IgG4-RD and risk of malignancies.1,2 The relationship between lymphoma and IgG4-RD is unclear. We describe a case of bilateral IgG4-related dacryoadenitis which evolved into an extra-nodal marginal zone lymphoma (EMZL) on one side.


A 65-year-old female patient presented with a bilateral, painless proptosis and swelling of the upper eyelids which had been present for 6 months. Examination revealed bilateral upper lid oedema, lacrimal gland enlargement (Fig. 1A,B), and proptosis of 20.5 mm and 20 mm in the right and left eyes, respectively. Computed tomography showed a heterogenous mass of intermediate intensity involving both lacrimal glands (Fig. 1C). Incision biopsy from both glands showed storiform fibrosis and a dense lymphoplasmacytic infiltrate (Fig. 1D,E). Immunohistochemical staining revealed 28 and 31 IgG4 positive plasma cells per high power field (HPF) (Fig. 1F) and an IgG4/IgG positive plasma cell ratio of 60% and 57% in the right and left glands, respectively. The right lacrimal gland also showed diffusely infiltrating, monotonous, small lymphoid cells (Fig. 2A) which stained positively with CD20 (Fig. 2B) and Bcl2 and were negative for CD3, CD5, CD10 and CD23. Light chain restriction was seen (positive for kappa and negative for lambda), thus confirming their monoclonal nature (Fig. 2C). The Ki-67 labelling index was 15% (Fig. 2D). A diagnosis of bilateral IgG4-related dacryoadenitis evolving into an EMZL (MALT-type) on the right side was made. Bone marrow and systemic examination was normal. The patient was administered six cycles of cyclophosphamide, hydroxydanorubicin and oncovin and prednisolone (CHOP-regimen). At 15 months follow-up, the patient was disease free. Since IgG4-RD was first identified in the pancreas at the beginning of this century, several organs in the human body including the orbit have been shown to be affected by this disease.3,4 Most patients of orbital IgG4-RD present with proptosis and/or eyelid swelling.4,5 Microscopically, it is characterised by storiform fibrosis, dense lymphoplasmacytic infiltrate and obliterative phlebitis although the latter two may be inconspicuous or absent in the lacrimal glands. Obliterative phlebitis was absent in our patient. The diagnosis of IgG4-RD requires at least 10 IgG4 positive plasma cells per HPF and IgG4/IgG positive plasma cell ratio in excess of 40%. Both these criteria were fulfilled in our patient. Although the exact relation between IgG4-RD and lymphoma is uncertain, they are close mimics of each other in the orbit. IgG4-positive plasma cells have been identified in orbital lymphomas.6 Further, lymphomas may arise in a background of IgG4-RD.7 Lymphoma may arise either concurrently with or subsequent to the diagnosis of IgG4-RD.8,9 Conversely, IgG4-RD may follow a lymphoma.10 In our patient, lymphoma and IgG4-related dacryoadenitis occurred concurrently. Speculations suggest dysregulation of B lymphocytes associated with autoimmune disease, leading to proliferation of abnormal B lymphocytes, culminating in the occurrence of malignant B-cell lymphoma.11 Orbital IgG4-RD associated lymphomas are usually of EMZL subtype and those occurring at extraocular sites in association with IgG4-RD are usually of diffuse large B-cell subtypes.7,12 Our patient had an EMZL of mucosaassociated lymphoid tissue (MALT) type. This case is unique because we show, for the first time, a bilateral IgG4-related dacryoadenitis evolving into a lymphoma on one side. Our case upholds previous reports in the literature describing increased risk of malignancies associated with IgG4-RD.2 Moreover, our case also highlights the importance of performing a bilateral lacrimal gland biopsy to ensure accurate diagnosis and appropriate management in patients with bilateral lacrimal gland lesions. To conclude, we report the first documented case of bilatera IgG4-related dacryoadenitis evolving into an EMZL


CORRESPONDENCE

465

Fig. 1 (A,B) IgG4-related dacryoadenitis presenting with bilateral lacrimal gland enlargement and eyelid swelling. (C) Computed tomography showing a diffuse and heterogenous mass involving both lacrimal glands. (D) Storiform fibrosis seen in IgG4-related dacryoadenitis (H&E). (E) Splaying of lacrimal acini and a dense lymphoplasmacytic infiltrate (H&E). (F) IgG4 positive plasma cells.

Fig. 2 (A) Diffuse infiltrate of small lymphoid cells (H&E). Inset: Small and monotonous lymphoid cells with scant cytoplasm and slightly irregular nuclear borders (H&E). (B) CD20 positivity, (C) kappa light chain staining, and (D) Ki-67 staining in tumour cells.

on one side. Since IgG4-RD patients are at a greater risk of developing a lymphoma, they need to be followed for a longer duration. Kaustubh Mulay1 Ekta Aggarwal2 1

National Reporting Centre for Ophthalmic Pathology (NRCOP), Centre For Sight, Hyderabad, and 2Department of Ophthalmic Plastics, Vasan Eye Care Hospitals, Hyderabad, India Contact Dr K. Mulay. E-mail: [email protected] 1. Lightfoot N, Kalnins R. Diffuse large B-cell lymphoma/follicular lymphoma arising in a background of IgG4-related pachymeningitis. Pathology 2013; 45: 696–8.

2. Yamamoto M, Takahashi H, Tabeya T, et al. Risk of malignancies in IgG4related disease. Mod Rheumatol 2012; 22: 414–8. 3. Hamano H, Kawa S, Horiuchi A, et al. High serum IgG concentrations in patients with sclerosing pancreatitis. N England J Med 2001; 344: 732–8. 4. Kubota T, Moritani S. Orbital IgG4-related disease: clinical features and diagnosis. ISRN Rheumatology 2012; 2012: 412896. 5. Mulay K, Aggarwal E, Jariwala M, Honavar SG. Orbital immunoglobulinG4-related disease: case series and literature review. Clin Experiment Ophthalmol 2013; Dec 12: (Epub ahead of print). 6. Kubota T, Moritani S, Yoshino T, et al. Ocular adnexal marginal zone B cell lymphoma infiltrated by IgG4-positive plasma cells. J Clin Pathol 2010; 63: 1059–65. 7. Cheuk W, Yuen HK, Chan AC, et al. Ocular adnexal lymphoma associated with IgG4þ chronic sclerosing dacryoadenitis: a previously undescribed complication of IgG4-related sclerosing disease. Am J Surg Pathol 2008; 32: 1159–67. 8. Takahashi N, Ghazale AH, Smyrk TC, et al. Possible association between IgG4-associated systemic disease with or without autoimmune pancreatitis and non-Hodgkin lymphoma. Pancreas 2009; 38: 523–6. 9. Kim T, Grobmyer SR, Dixon LR, et al. Autoimmune pancreatitis and concurrent small lymphocytic lymphoma: not just a coincidence? J Gastrointest Surg 2008; 12: 1566–70.


466

CORRESPONDENCE


10. Oshima Y, Usui R, Manabe S, et al. IgG4-related tubulointerstitial nephritis and lymphadenopathy after therapy for malignant lymphoma. Intern Med 2012; 51: 1221–6. 11. Hansen A, Lipsky PE, Do¨rner T. B-cell lymphoproliferation in chronic inflammatory rheumatic diseases. Nat Clin Pract Rheumatol 2007; 3: 561–9. 12. Ishida M, Hodohara K, Yoshida K, et al. Occurrence of anaplastic large cell lymphoma following IgG4-related autoimmune pancreatitis and cholecystitis and diffuse large B-cell lymphoma. Int J Clin Exp Pathol 2013; 6: 2560–8.

DOI: 10.1097/PAT.0000000000000132

Malassezia pachydermatis fungaemia in an adult on posaconazole prophylaxis for acute myeloid leukaemia Fig. 1 Direct Gram stain from the aerobic blood culture broth.

Sir, Malassezia pachydermatis is a relatively rare agent of bloodstream infections. We describe a fungaemia with this predominantly neonatal pathogen in a geriatric male, only the second case of its kind documented in the adult patient population. This case is novel in being the first super infection described with this yeast on posaconazole, a primarily mould active antifungal agent. A 69-year-old Chinese male on oral posaconazole prophylaxis (200 mg every 8 h for 9 days) while on induction chemotherapy (intravenous daunorubicin 80 mg/d, cytarabine 180 mg/d) for acute myeloid leukaemia (AML) presented with febrile neutropenia of 12 days. Oral temperature was 39.18C, with WBC count 0.2 109/L, absolute neutrophil count 0.02 109/L and lymphocyte count 0.17 109/L, and raised inflammatory markers (procalcitonin 2.04 mg/L and CRP 72.4 mg/L). Patient co-morbidities included moderate anaemia (Hb 8 g/dL) and diabetes mellitus (HbA1c 8.6%). Chest radiography revealed bibasal air space shadows and left mid zone interstitial markings, while diffuse ground glass changes were observed on chest computed tomography (CT) scan with interlobular interstitial thickening. The patient spiked a temperature of 39.48C on an empirical regimen of piperacillin-tazobactam (4.5 g every 8 h) and vancomycin (1–1.5 g every 24 h) initiated for the febrile episode above. The antimicrobial therapy was hence switched to meropenem (1 g every 8 h). An aerobic blood culture (BACTEC Plus Aerobic/F bottles; BD Diagnostic Systems, USA) collected from a peripherally inserted central catheter (in situ for 2 weeks) yielded unipolar budding yeast cells, some with collarettes on the fourth day of incubation (Fig. 1). Subcultures from the broth were performed onto 5% sheep blood agar, chocolate agar, MacConkey agar and Sabourad’s Dextrose agar (BBL; Bio-Media) which were incubated aerobically at 378C with 5% CO2. Cream coloured, 0.5 mm dry colonies were observed on the first day of incubation on all plates including Sabourad’s (Fig. 2). Consequently, the above therapy was discontinued and IV amphotericin B 1 mg/kg once daily was initiated. An identification was performed on the matrix-assisted laser desorption/ionisation time-of flight (MALDI-TOF MS; Bru¨ker Daltonik, Germany) employing a standard ethanol/formic acid protein extraction protocol.1 The spectrum generated was transferred to the Biotyper software and compared to the

MBT database DB 5627, analysed by the software-defined algorithm. This yielded a score acceptable to the species level (2.233) for M. pachydermatis which correlated with the identification from sequencing of the Internal Transcribed Spacer (ITS) gene.2 Using published primers an amplicon of 529 bp was generated. Nucleotide sequencing of both strands was performed using an ABI 3730XL DNA sequencer (PerkinElmer, Applied Biosystems Division, USA). The sequenced product was 100% identical to the ITS rRNA gene M. pachydermatis GenBank entry (GU 373662, GU 373660 and AY 743637) in a region of 529 determined base pair positions. A phenotypic identification was not attempted for the above isolate, neither was an antifungal susceptibility performed in the absence of current Clinical and Laboratory Standards Institute (CLSI) interpretive guidelines. The incriminating peripherally inserted central catheter line removed 5 days following the positive blood culture did not yield a growth on culture. Multiple repeat blood cultures collected on therapy were sterile. The clinical improvement was accompanied by a resolution of the inflammatory markers including procalcitonin (0.21 mg/L) and CRP (13.7 mg/L).

Fig. 2 Cream coloured colonies of Malassezia pachydermatis on Sabourad’s agar (24 h).


CORRESPONDENCE

The genus Malassezia has been implicated as a cause of disseminated infections, typically against a backdrop of profound immunosuppression with M. furfur3 being more commonly encountered than its non-lipophilic relatives. To the best of our knowledge only a single case of fungaemia due to this pathogen has been described to date in the adult population.4 The mode of acquisition is subject to speculation in the absence of known risk factors (contact with dogs/skin lesions/recent surgery or total parenteral nutrition) in our patient. A definitive diagnosis of catheter-related bloodstream infection is also dubious in the context of sterile peripheral blood cultures and catheter tip.5 However, the patient was initiated on a polyene antifungal, amphotericin B (MIC 1 mg/mL), being the drug of choice for invasive infections with azoles being relatively disappointing and echinocandins inactive.6 Serum posaconazole levels were incidentally not performed for the above patient. Thus a vulnerability to super infection exists in light of a recent transition from amphotericin B to posaconazole in the choice of AML prophylaxis,7 particularly against a background of poor accessibility and a lengthy turnaround time to serum level monitoring facilities. Bioavailability may be a contributory factor being subject to variables such as concomitant food intake, gastric pH or gut motility. Further on, a potential role for biofilms in antifungal resistance has recently been described for this novel pathogen.8 Until recently, time consuming conventional methods of speciation including Vitek 2 and API ID 32 C (both bioMerieux, France) have formed the backbone of identification for unusually encountered fungi including Geotrichum, Kodamaea ohmeri and Saccharomyces species. A rapid identification is thus pertinent. Contemporary non-biochemical techniques like the MALDI-TOF9 have recently served to abbreviate the turnaround time to speciation. Cumbersome molecular manoeuvres such as ITS sequencing of the rDNA (genes 1 and 2) or the D1/D2 domain of the large subunit rDNA10 may represent an alternative, despite the attendant turnaround time, expense and accessibility. A protracted antifungal course complicated by its toxicity underscores a role for an expedited identification of such uncommonly encountered fastidious yeasts. Conflicts of interest and sources of funding: The authors state that there are no conflicts of interest to disclose. Saugata Choudhury1 Ryan Llorin Marte2 Department of Laboratory Medicine, and 2Department of Infectious Diseases, Tan Tock Seng Hospital, Singapore 1

Contact Dr S. Choudhury. E-mail: [email protected] 1. Mellmann A, Bimet F, Bizet C, et al. High interlaboratory reproducibility of matrix-assisted laser desorption ionization-time of flight mass spectrometry-based species identification of nonfermenting bacteria. J Clin Microbiol 2009; 47: 3732–4. 2. Innis MA, Gelfand DH, Sninsky JJ, White TJ, editors. PCR Protocols: A Guide to Methods and Applications. New York: Academic Press, 1990. 3. Fine RN, Salusky IB, Hall T, et al. Peritonitis in children undergoing continuous ambulatory peritoneal dialysis. Pediatrics 1983; 71: 806–9. 4. Lautenbach E, Nachamkin I, Schuster MG. Malassezia pachydermatis infections. N Engl J Med 1998; 339: 270. 5. Mermel LA, Allon M, Bouza E, et al. Clinical practice guidelines for the diagnosis and management of intravascular catheter-related infection: 2009 Update by the Infectious Diseases Society of America. Clin Infect Dis 2009; 49: 1–45.

467

6. Tragiannidis A, Bisping G, Koehler G, Groll AH. Minireview: Malassezia infections in immunocompromised patients. Mycoses 2010; 53: 187–95. 7. Freifeld AG, Bow EJ, Sepkowitz KA, et al. Clinical practice guideline for the use of antimicrobial agents in neutropenic patients with cancer: 2010 Update by the Infectious Diseases Society of America. Clin Infect Dis 2011; 52: 427–31. 8. Figueredo LA, Cafarchia C, Otranto D. Antifungal susceptibility of Malassezia pachydermatis biofilm. Med Mycol 2013; 51: 863–7. 9. Kolecka A, Khayhan K, Arabatzis M, et al. Efficient identification of Malassezia yeasts by matrix-assisted laser desorption ionization-time of flight mass spectrometry (MALDI-TOF MS). Br J Dermatol 2014; 170: 332–41. 10. Arendrup MC, Boekhout T, Akova M, et al. ESCMID and ECMM joint clinical guidelines for the diagnosis and management of rare invasive yeast infections. Clin Microbiol Infect 2014; 20 (Suppl 3): 76–98.

DOI: 10.1097/PAT.0000000000000139

Male carrier of haemophilia A Sir, Haemophilia A is an X-linked bleeding disorder that results from a decrease or absence of coagulation factor VIII activity. The severity of the disorder ranges from mild to severe, with the severity being consistent within each family. Hemizygosity for the X-chromosome means that affected males far outnumber affected females who in the majority are carriers. Klinefelter syndrome, defined as X chromosome excess in males, has a birth prevalence of approximately 2 per 1000 in Australia.1 The 47, XXY karyotype usually results from meiotic non-disjunction during gamete formation. The principal effects of Klinefelter syndrome include hypogonadism and reduced fertility. There are other possible physical and behavioural issues with varying severity. Up to 50% of males with Klinefelter syndrome remain undiagnosed throughout life, with many undiagnosed until a delay in puberty or infertility investigations, although in retrospect clues to the diagnosis may have been present in these patients, including specific learning difficulties and gynaecomastia.1 The presence of the second X chromosome may provide protection against X-linked disorders. The child in this study was born to a haemophilia carrier of a missense mutation in exon 11 of the F8 (coagulation factor VIII) gene (NM_000132.3 c.1649G>A, p.Arg550His), responsible for mild haemophilia A. Prenatal testing for the mutation was declined with factor VIII levels and mutation analysis arranged on a cord blood sample to assess whether the child was affected and would require management. Analysis of the cord blood showed a factor VIII level of 60IU/dL indicating the child was unaffected by haemophilia A. Mutation analysis detected heterozygosity, an anomalous result in a male. An obvious explanation for this result is maternal cell contamination, however STR analysis of the cord sample (15 loci were examined using the AmpFISTR Indentifier kit; Life Technologies, USA) eliminated maternal cell contamination as a possibility. Heterozygosity in the baby was confirmed by repeat sequence analysis on DNA prepared from a peripheral blood sample. Chromosome analysis by karyotyping and chromosome array confirmed a 47XXY karyotype consistent with a diagnosis of non-mosaic Klinefelter syndrome. The additional X chromosome containing a normal copy of the F8 gene inherited either from the father or the mother


468


CORRESPONDENCE

due to a non-disjunction event in gamete formation, has allowed a factor VIII level within the normal range, providing protection from symptomatic haemophilia. This indicates that there has not been significant skewing of X inactivation directed against the X chromosome containing the normal F8 gene. Protection from the impact of mutations in X linked genes in males with Klinefelter syndrome is a well-established concept, particularly in conditions which are usually considered to be lethal in hemizygous males. For example, small numbers of cases have been reported of males affected by Aicardi syndrome,2 incontinenti pigmenti,3 X-linked chondrodysplasia punctata,4 and Rett syndrome,5 who have later been identified to also be affected by Klinefelter syndrome. Other X-linked conditions reported in males with Klinefelter syndrome with potentially ameliorated symptoms dependent on X inactivation status include Becker muscular dystrophy,6 fragile X syndrome,7 X-linked Alport syndrome,8 X-linked chronic granulomatous disorder,9 and properdin deficiency.10 While neither haemophilia A nor Klinefelter syndrome are rare in themselves, to our knowledge this is only the third case recorded in the literature of an association of haemophilia with Klinefelter syndrome. Interestingly, the previously two reported cases indicated the child was phenotypically a haemophiliac. Chipail et al.11 in 1965 described a case of Klinefelter syndrome with an incidental finding of haemophilia A. The expression of the haemophilia was presumed to be associated with the inactivation of the normal X chromosome. Vencesla´ et al.12 reported the case of a Klinefelter patient inheriting both X chromosomes from the mother with the appearance of a de novo mutation resulting in a deletion of exons 1–12 in one of these chromosomes. The unfortunate outcome for this patient was skewed X inactivation of the normal F8 allele resulting in the expression of severe haemophilia A. In this present case the fact that one of the X chromosomes was carrying a mutation in the F8 gene may have been masked had the child not been the son of a known haemophilia carrier. While unlikely, it is possible that the ‘carrier’ status in this child could be found in other Klinefelter boys. The diagnosis of Klinefelter syndrome in this patient at such a young age, although challenging for the family, is potentially fortuitous. It allows the potential for issues related to Klinefelter syndrome to be monitored for in childhood, and preventive management arranged such as hormone treatment to help initiate puberty. In fact, there is an increasing call for instituting newborn screening programs for this condition given the potential advantages of an early diagnosis and targeted therapies. Until relatively recently, the vast majority of males with Klinefelter syndrome were considered infertile due to azoospermia, with sperm donation being the only opportunity for parenting. However, with current assisted reproductive technology, testicular biopsy may identify viable sperm in up to 50% of affected males which in conjunction with intracytoplasmic sperm injection (ICSI) may allow males to father their own biological children. The risk to offspring of inheriting an extra copy of the X chromosome appears to be low (about 1%). Should our present patient be fertile, unlike the usual scenario where the daughter of a haemophilia A male is an obligate carrier, in this case there is a 50% chance of a daughter being a carrier. Thus, molecular testing should be offered to any

daughters of reproductive age to allow appropriate reproductive risks to be discussed. Conflicts of interest and sources of funding: The authors state that there are no conflicts of interest to disclose. Vaughan K. Williams1 Jan Liebelt2 Catherine Nicholls3 Samantha Reardon1 Ram Suppiah1 1

Haematology Department, Women’s and Children’s Hospital, North Adelaide, 2SA Clinical Genetics Service, Women’s and Children’s Hospital, North Adelaide, and 3Genetic Pathology Department, SA Pathology, Adelaide, SA, Australia Contact Dr V. Williams. E-mail: [email protected] 1. Herlihy AS, Halliday JL, Cock ML, McLachlan RI. The prevalence and diagnosis rates of Klinefelter syndrome: an Australian comparison. Med J Aust 2011; 194: 24–8. 2. Chen TH, Chao MC, Lin LC, et al. Aicardi syndrome in a 47, XXY male neonate with lissencephaly and holoprosencephaly. J Neurol Sci 2009; 278: 138–40. 3. Buinauskaite E, Buinauskiene J, Kucinskiene V, Strazdiene D, Valiukeviciene S. Incontinentia pigmenti in a male infant with Klinefelter syndrome: a case report and review of the literature. Pediatr Dermatol 2010; 27: 492–5. 4. Sutphen R, Amar MJ, Kousseff BG, Toomey KE. XXY male with X-linked dominant chondrodysplasia punctata (Happle syndrome). Am J Med Genet 1995; 57: 489–92. 5. Schwartzman JS, Bernardino A, Nishimura A, Gomes RR, Zatz M. Rett syndrome in a boy with a 47,XXY karyotype confirmed by a rare mutation in the MECP2 gene. Neuropediatrics 2001; 32: 162–4. 6. Suthers GK, Manson JI, Stern LM, Haan EA, Mulley JC. Becker muscular dystrophy (BMD) and Kinefelter’s syndrome: a possible cause of variable expression of BMD within a pedigree. J Med Genet 1989; 26: 251–4. 7. Mavrou A, Syrrou M, Tsenghi C, Agelakis M, Youroukos S, Metaxotou C. Martin-Bell syndrome in Greece, with report of another 47,XXY fragile X patient. Am J Med Genet 1988; 31: 735–9. 8. Ars E, Ta´zon-Vega B, Ruiz P, et al. Male-to-male transmission of X-linked Alport syndrome in a boy with a 47,XXY karyotype. Eur J Hum Genet 2005; 13: 1040–6. 9. Gill HK, Kumar HC, Cheng CK, et al. X-linked chronic granulomatous disease in a male child with an X-CGD carrier, Klinefelter brother. Asian Pac J Allergy Immunol 2013; 31: 167–72. 10. Schejbel L, Rosenfeldt V, Marquart H, Valerius NH, Garred P. Properdin deficiency associated with recurrent otitis media and pneumonia, and identification of male carrier with Klinefelter syndrome. Clin Immunol 2009; 131: 456–62. 11. Chipail A, Leu EG, Dragomir C. Sindrom Klinefelter asociat cu hemofilie A la un copil de 8 ani. Pediatria (Bucur) 1965; XIV: 461–5. 12. Vencesla´ A, Baena M, Garrido RP, et al. F8 gene dosage defects in atypical patients with severe haemophilia A. Haemophilia 2012; 18: 708–13.

DOI: 10.1097/PAT.0000000000000131

Is it feasible for patients to estimate their own 24 hour urine volume? Sir, Twenty-four hour urine collections are considered to be the ‘gold standard’ for many assays including urinary albumin, catecholamines, and for the evaluation of proteinuria.1–4 This was advocated as a 24 h urine collection overcame the diurnal


CORRESPONDENCE

variation in the various analytes measured.5 Yet, although the volume of urine produced in the period of 24 h needs to be known to determine the excretion rate of a given analyte, the laboratories only need approximately 10 mL of urine to complete the assays.6 Previous authors have examined the difficulty in the collection of 24 h urine samples; however, none of these authors have studied the possibility of estimation of urine volume at home.5,7,8 This study examined the feasibility of the measurement of urine volume in the patient’s home by comparing self-assessed urine volume with laboratory assessed urine volume. The study included two parts: a preliminary survey and the actual experiment. The study was approved by the Bond University Human Research Ethics Committee. A preliminary survey of 104 adults was conducted which involved the subject estimating an unknown volume of water in a 24 h urine bottle. Subjects included 32 hospital and laboratory staff members, 48 consecutive hospital ward patients and 24 consecutive collection centre patients at Sullivan Nicolaides Pathology in September 2012. The 24 h urine bottle contained a set volume of water with yellow food colouring mimicking urine and had the same measurement scale on the bottle as the one included in the study (Fig. 1). Between 1 December 2012 and 1 December 2013, adult patients presenting to collection centres on the Gold Coast and Coffs Harbour with a request from their doctor for a 24 h urine assay for management of their clinical condition were asked to participate in the study. Participants were invited to complete a questionnaire regarding participant demographics and perceived difficulties in addition to the 24 h urine collection and were also asked to estimate the volume of urine collected by using a measurement scale on the side of the bottle (Fig. 1). The measurement scale included three markings: one for each litre, up to a volume of three litres (Fig. 1). A written explanatory statement was given to all participants and consent was obtained. The collected sample was then processed as per routine protocol in the laboratory. The final volume was obtained using standard laboratory practice of weighing on a calibrated balance. At the end of the study, data in the questionnaires was collated and patient estimated urine volume was compared with the laboratory measured volume. All

3L

469

questionnaires were de-identified by a random nine digit number and participants remained anonymous throughout the study. A total of 22 urine samples were collected from the study. Twenty of the collected samples had completed consent forms and were included in our results (Table 1). Of the 20 participants, 17 had previously collected a 24 h urine collection prior to this study. Analysis of the study questionnaires revealed that 17 participants found the collection easy, including two whom had not previously performed the collection. There were mixed participants’ views in terms of how easy it was to miss a sample; eight found it easy to miss a sample, six found it difficult to miss a sample and five did not report ease or difficulty. Ten participants believed it would be easier to measure the 24 h urine volume at home and only bring a small sample to the collection centre whilst eight did not, one was undecided and one did not answer the question. Of particular interest was the fact that 16 of the 20 participants found it easy to transport the 24-h urine sample to the collection centre. In our preliminary survey, the laboratory measured volume was 1.25 L and the mean volume estimated was 1.33 L [standard deviation (SD) 0.230, coefficient of variation (CV) 0.172]. Thirty-nine participants’ estimates were within 2% of the actual volume and 86 participants had an error margin of less than 10% (Fig. 2A). It is unclear why only five participants had estimates which were within 3–5% of the actual volume whilst there were 39 within 2% and 42 within 5–10% (Fig. 2A). There were two participants whose error margin was more than 50% of the actual volume. Of the 86 participants whose estimates were within 10% of the actual volume, 65 were female and 21 were male. Twenty-six were less than 40 years of age, 34 were between 40 and 60 years old and 26 were more than 60 years old. Seventeen of the 22 participants estimated the volume of urine collected. Paired t test for patient estimation of urine volume was 2.12 ( p ¼ 0.878). Six participants’ estimations were within 2% of the actual laboratory measured volume, and 12 were within 10% of the actual volume. Fig. 2B shows the estimated volume compared with the actual laboratory measured volume. Of the 12 participants whose estimates were within 10% of the actual volume, eight were male and four were female. Eight of the 12 participants were older than 60 years of age and seven had received a tertiary education, two had secondary education and three participants’ highest education level was primary school. The two participants whose estimated volume was more than 25% of the actual volume were both male and were more than 50 years of age with secondary and tertiary qualifications, respectively. Table 1

Participant demographics in preliminary survey and 24 h urine study

2L

1L

Gender Male Female Age, years 71 Total participants

Preliminary survey

Study

53 51

11 9

32 23 14 12 23 104

4 2 1 4 9 20

Fig. 1 Measurement scale on side of 24 h urine bottle.


470


Number of participants (n = 104)

CORRESPONDENCE

42

45 40 35 30 25 20 15 10 5 0

39

8

5

50%

A Volume (L)

5

Percentage error (%) 4 3 2 1 0 1

2

3

4

5

6

7

8

9 10 11 12 13 14 15 16 17

Sample number (n = 17)

B

Estimated volume

Fig. 2 (A) Preliminary survey of simulated urine volume estimation in 104 participants. (B) Estimated versus laboratory measured urine volume in 17 patients collecting a 24 h urine sample.

From these results, it appears that it may be feasible for certain patients to estimate their own urine volume at home and bring a small aliquot for laboratory testing using their estimated urine volume. To identify patients who may fall into the poor estimator group (that is, those who were greater than 10% from the true value) it may be necessary to have several test urine bottles with different volumes in the collection centre and a brief test before the decision is taken to accept the patient estimate. This would also test the consistency of estimates between and within patients. Both the preliminary survey and study results did not show that age or educational level were a factor in the participants’ ability to estimate urine volume. The preliminary survey showed that females were better estimators than males; however, the study found males were better estimators. Nonetheless, patient preference from the questionnaire for this method was very polarised with half preferring the proposed method of estimating one’s urine volume and the other half preferring the current method of laboratory measured urine volume. It is unfortunate that our sample population is small, as this limits our ability to comment on how likely these results reflect the general population. It was interesting to note that this project was met with some resistance both from the patients and collection staff. Feedback from various collection centre staff voiced difficulty in recruiting patients as many potential participants especially first time collectors did not want to complicate the already cumbersome process of a 24 h urine collection by participating in a study. A total of 85 urine bottles were delivered to collection centres, with 65 bottles given out to patients, but 22 bottles were returned. Although there were 43 lost urine samples, of the 22 recruited participants, 20 had complete consent forms and questionnaires. Despite handing out 65 bottles to patients, the low response rate perhaps is a sign of the reluctance of patients to perform this test even when it is requested by their doctor as a necessary part of the management of their condition. It has been reported elsewhere that there was a 30% non-compliance with 24 h urine collections in a similar population so perhaps these

results are not unexpected.9 Of the 20 participants, only three participants were first time collectors, which may have influenced the results. Patients who had previously completed a 24 h urine test were much more comfortable to collect another 24 h urine test; of the 17 participants who had previously performed this collection, 15 found the collection easy. Although it would have been preferable to have more gradations on the measurement scale, the selected type of bottle and measurement scale was chosen as they are identical to the ones used in the public health system on the Gold Coast and are currently commercially available. Despite the broad gradations, patient estimates and actual measurements showed very close agreement. The majority of participants were able to estimate their 24 h urine volume adequately. There was no correlation with age or educational level in the ability of participants to accurately estimate their urine volume. It may be feasible for selected patients to measure their urine volume at home and bring an aliquot for testing in the laboratory in the future. Patients who may be poor estimators may be identified with a test urine bottle in the collection centre. However, further studies with larger participant recruitment are needed. Acknowledgements: We would like to thank Professor Chris Del Mar for his input in developing the patient questionnaires and Sullivan Nicolaides Pathology especially Sue Kehrer for her help in recruiting participants from Coffs Harbour. Conflicts of interest and sources of funding: This work was supported by a Scholarship in Pathology for Medical Schools from the Royal College of Pathologists of Australasia. The authors state that there are no conflicts of interest to disclose. Amy Millicent Cao1 Robert Gordon Wright1 Tony Badrick2 Faculty of Health Sciences and Medicine, and 2Faculty of Biomedical Science and Medicine, Bond University, Robina, Qld, Australia 1

Contact Dr A. M. Cao. E-mail: [email protected] 1. Royal College of Pathologists of Australasia. RCPA Manual. Catecholamines – urine. 6 Aug 2007, cited 11 Dec 2013. http://www.rcpamanual. edu.au/index.php?option=com_pttests&task=show_test&id=441&Itemid= 34 2. Gansevoort RT, Brinkman J, Bakker S, De Jong PE, de Zeeuw D. Evaluation of measures of urinary albumin excretion. Am J Epidemiol 2006; 164: 725–7. 3. Shidham G, Hebert LA. Timed urine collections are not needed to measure urine protein excretion in clinical practice. Am J Kidney Dis 2006; 1: 8–14. 4. Cote AM, Firoz T, Mattman A, Lam EM, von Dadelszen P, Magee LA. The 24-hour urine collection: gold standard or historical practice? Am J Obstet Gynecol 2008; 199: 625. 5. Lambers Heerspink HJ, Brantsma AH, de Zeeuw D, Bakker SJ, de Jong PE, Gansevoort RT. Albuminuria assessed from first-morning-void urine samples versus 24-hour urine collections as a predictor of cardiovascular morbidity and mortality. Am J Epidemiol 2008; 168: 897–905. 6. Mayo Clinic. Mayo Foundation for Medical Education and Research. Test ID:UIOD. Cited 11 Dec 2013. http://www.mayomedicallaboratories.com/ test-catalog/Specimen/9549 7. Miler M, Simundic AM. Low levels of adherence to instructions for 24-hour urine collection among hospital outpatients. Biochem Med 2013; 23: 316– 20.


CORRESPONDENCE

8. Price CP, Newall RG, Boyd JC. Use of protein:creatinine ratio measurements on random urine samples for prediction of significant proteinuria: a systematic review. Clin Chem 2005; 51: 1577–86. 9. Badrick T, Hobson P, Langguth D. Changing the way referring general practitioners request urinary albumin studies. Pathology 2013; 45: 626–8.

DOI: 10.1097/PAT.0000000000000126

C4d FlowPRA is a useful tool in live related renal transplants Sir, C4d FlowPRA is an uncommon test, which has not been utilised frequently as a part of renal transplant work-up or post-transplant monitoring. Flow cytometric cross match (FCXM) is a standard and sensitive tool, routinely used to detect donor specific antibodies prior to renal transplantation.1 It is known to be more sensitive than the traditional complement-dependant cytotoxicity cross match (CDC-XM), and also helps to identify if the positivity is for T cells or B cells or both. However, there are instances where there may be false positive FCXM, and in such cases IgG FlowPRA is a useful tool. A thorough pre-transplant work-up is more important, especially in the Indian scenario, where most of the donors are live and related, unlike the West where there is predominance of deceased donor transplants. An IgG FlowPRA test utilises commercially available beads coated with 30 antigens of class I and II each. This test thus helps to identify whether the positivity is actually due to anti-HLA antibodies. There are multiple components of this test, i.e., screening, specific and single beads, with each step helping to identify the exact type of the anti-HLA antibody. However, this test cannot distinguish between complement-fixing and non-complement-fixing antibodies. It is a known fact that complement-fixing antibodies are the most detrimental factor and implicative in cases of antibody-mediated rejection (AMR). C4d-PRA testing helps to overcome this handicap.2 C4d deposition in the peri-tubular capillaries has been considered to be a diagnostic criterion for the detection of antibody-mediated rejection as per the Banff schema.3 C4d is a product of C4, which is formed as a result of antibody mediated classic complement activation pathway. In the present case the patient was found to be positive for alloantibodies by FCXM and IgG FlowPRA, but the same was negative by C4d FlowPRA, thus denoting the antibodies to be non-complement fixing. The results were corroborative with the negativity for CDC-XM. We report a case of a 45-year-old male patient, a known case of chronic kidney disease and hypertension, with a proposed renal transplant from a live donor (wife). The patient had a history of multiple blood transfusions taken at different centres/ hospitals. There was no history of any prior transplant. The transplant physician suggested the patient to undertake CDC-XM, IgG FlowPRA and FCXM, as a part of routine pre-transplant work-up. The FCXM was performed using threecolour flow cytometry and anti-human IgG (Jackson ImmuoResearch Laboratories, USA) after discriminating T and B cells using CD3 and CD22 (BD Biosciences, USA). This was found to be positive for both T and B cells. IgG FlowPRA screening test (One Lambda, USA) was performed immediately using

471

both class I and class II beads. The results were calculated using the cut-off set on negative control. It showed positivity for both class I and class II with the values of about 60% and 40%. However, since the CDC-XM was negative, we wanted to explore it further and decided to perform C4d FlowPRA, to see whether the antibodies were complement-fixing or not. The test was performed as per the protocol provided using the FITC conjugated anti-human C4d antibody (C4dpAb2) (Biomedica, Austria). A known serum, with normal C4 levels, was used as a source of complement in the test. The result of this test was negative for both class I and class II beads (Fig. 1). This case proves that C4d FlowPRA assay should be used more routinely in cases where FCXM and IgG FlowPRA are positive, to prove if the alloantibodies are complementfixing or not. CDC-XM, although a gold standard, suffers from the problem of subjectivity and inter-observer variation. This test may also help in identifying low levels of complement-fixing antibodies that CDC-XM may miss, especially in the Indian scenario where most of the transplants are liverelated. The presence of alloantibodies against the HLA antigens of the donor in the serum of a transplant recipient is considered to be a risk factor for antibody mediated rejection.4 Traditionally, CDC-XM is considered a gold standard to identify DSA, which is a risk factor for graft rejection or loss. However, this test is limited by the fact that it may miss out on low concentrations of antibody. Karpinski et al., demonstrated that patients that were negative even with the more sensitive AHG-CDC, but positive FCXM and FlowPRA, had poor graft outcome and adverse immunological events.5 However, it is also known that noncomplement fixing antibodies are less harmful than the complement fixing antibodies. The usually performed FCXM and IgG FlowPRA assays are not able to distinguish between the two. The C4d FlowPRA assay was developed by Bartel et al.6 that helped to detect HLA alloantibody-triggered C4d formation in vitro. It was found that recipients with complement-fixing pre-sensitisation were found to have worse graft survival as compared to those who had non-complement-fixing antibodies. In this study the authors found that C4d FlowPRA had a high level of specificity (compared to capillary C4d staining on biopsy), while the sensitivity was found to be higher with IgG FlowPRA. They also demonstrated more consistent findings with IgG FlowPRA and C4d FlowPRA as compared to biopsy C4d, over a period of time. The author also felt that antibody or complement detection in absence of capillary C4d deposition could reflect C4d positive AMR. Some degree of discordance between tissue biopsy C4d expression and flow cytometry based C4d FlowPRA, is assumed to be due to the sub-threshold level of HLA antigen expression by endothelial cells. Our case proves that C4d FlowPRA is an important test that should be included in the work-up of all flow cytometry laboratories that are undertaking transplantation work-up. It may not be required in all cases; however, it should be performed in cases where FCXM and/or IgG FlowPRA is positive. This is especially important in cases where the donors are live and related, which is the case in our setting. Conflicts of interest and sources of funding: The authors state that there are no conflicts of interest to disclose.



1000

CORRESPONDENCE

600 400

SSC-Height

600

R2

R3

200

400

R1

0

0

200

SSC-Height

800

800

1000

472

0

200

400

600

800

0

1000

200

400

600

800

1000

Class II PE

40

30

FSC-Height

CLASS II

15

Counts

20

Counts

20

30

25

CLASS I

10

M2 M1 10

M1

0

0

5

M2

0

200

400

600

800

1000

C4D -lgG

0

200

400

600

800

1000

C4D -lgG

Tube: Untitled Gated Events: 5874 Total Events: 10000

Tube: Untitled Gated Events: 2692 Total Events: 10000

Marker % Gated 100.00 All

Marker % Gated 100.00 All

M1

99.37

M1

98.25

M2

0.68

M2

1.75

Fig. 1 Flow cytometry results showing negativity for C4d FlowPRA, on the basis of the markers set with negative control.

Pranav Dorwal Amit Pande Simmi Mehra Neetu Tyagi Dharmendra Jain Vimarsh Raina Medanta The Medicity, Gurgaon, India Contact Dr P. Dorwal. E-mail: [email protected] 1. Rebibou JM, Chabod J, Bittencourt MC, et al. FlowPRA evaluation for antibody screening in patients awaiting kidney transplantation. Transpl Immunol 2000; 8: 125–8.

2. Wahrmann M, Exner M, Haidbauer B, et al. [C4d]FlowPRA screening – a specific assay for selective detection of complement-activating anti-HLA alloantibodies. Hum Immunol 2005; 66: 526–34. 3. Racusen LC, Colvin RB, Solez K, et al. Antibody-mediated rejection criteria – an addition to the Banff 97 classification of renal allograft rejection. Am J Transpl 2003; 3: 708–14. 4. Gebel HM, Bray RA. The evolution and clinical impact of human leukocyte antigen technology. Curr Opin Nephrol Hypertens 2010; 19: 598–602. 5. Karpinski M, Rush D, Jeffery J, et al. Flow cytometric cross match in primary renal transplant recipients with a negative anti-human globulin enhanced cytotoxicity crossmatch. J Am Soc Nephrol 2001; 12: 2807–14. 6. Bartel G, Wahrmann M, Exner M, et al. In vitro detection of C4d-fixing HLA alloantibodies: association with capillary C4d deposition in kidney allografts. Am J Transpl 2008; 8: 41–9.

DOI: 10.1097/PAT.0000000000000133


Lacrimal sac lymphoproliferative lesion: case report.

Malignant melanoma of the lacrimal sac: a case report.

Cytokeratin-Negative Undifferentiated (Lymphoepithelial) Carcinoma of the Lacrimal Sac.

Bilateral squamous cell carcinoma of the lacrimal sac.

Non-keratinising undifferentiated nasopharyngeal-type carcinoma of the lacrimal sac.

Inverted papilloma of lacrimal sac invading into the orbit: case report and review of literature.

Infestation of the lacrimal sac by Rhinosporidium seeberi: a clinicopathological case report.

Allergic fungal sinusitis involving the lacrimal sac: a case report and review.

Lacrimal sac mucocele.

Hepatocellular carcinoma metastasis to the lacrimal gland: A case report.

Congenital lacrimal sac diverticulum.

Schneiderian inverted papilloma: report of a case.

Lacrimal sac tumors.

Adult alveolar rhabdomyosarcoma of the lacrimal sac.

Pyogenic granulomas of the lacrimal sac.

IgG4-related lacrimal sac diverticulitis.

Benign mixed tumor of the lacrimal sac.

Malignant melanoma of the lacrimal sac.

Occult oncocytoma of the lacrimal sac.

Low-grade papillary schneiderian carcinoma, a unique and deceptively bland malignant neoplasm: report of a case.

Primary Systemic Amyloidosis Involving the Lacrimal Sac.

Low flow venous malformation lesion presented with medial canthal swelling simulating swelling of the lacrimal sac origin: A case report.

Squamous cell carcinoma arising from inverted schneiderian papilloma: a case report with oral involvement.

Transitional cell-type papillary carcinoma of the lacrimal sac manifesting as a medial canthal skin lesion.