34th Resident Symposium - Abstracts Submitted

   
 

The following abstracts/papers were submitted for consideration (* indicates those residents/fellows who were chosen to present by the Board of Directors):  

 

MUCINOUS ADENOCARCINOMAS OF THE THYMUS: REPORT OF 2 CASES AND REVIEW OF THE LITERATURE

 

Seong H. Ra, Michael C. Fishbein, Tamar Baruch-Oren, Peter Shintaku, Sophia K. Apple, Robert B. Cameron, Chi K. Lai

 

ABSTRACT:

 

Background: Most adenocarcinomas of the mediastinum are metastatic lesions.  Primary thymic adenocarcinomas are extremely rare neoplasms.  We could find only 12 cases reported in the literature; of these 12, only 4 were of the mucinous subtype.

 

Design: We report two additional cases of the mucinous subtype, including a previously unreported mucinous variant with numerous psammoma bodies.

 

Results: The first case in a 61-year-old female resembled a mucinous (colloid) carcinoma of other organs such as the breast and colon. It consisted of islands and strips of tumor cells floating in large pools of extracellular mucin. A unique feature of this tumor was the presence of numerous psammoma bodies. Immunohistochemically, the tumor cells were positive for cytokeratin (CK) 7 and negative for CD5. The second case in an 82-year-old female was a mucinous adenocarcinoma arising from a thymic cyst with areas of transition from benign to dysplastic epithelium. The tumor cells formed dilated glands, cords, and small nests that infiltrated the thymic cyst wall and exhibited evidence of mucin production. Immunohistochemically, the tumor cells were positive for CK 7 and focally positive for both CD5 and CK 5/6.

 

Conclusion: Mucinous adenocarcinoma, with or without, psammoma bodies, may be of primary thymic origin and should be considered in the differential diagnosis of malignant mediastinal tumors. These two cases provide further documentation of the rare occurrence of primary mucinous adenocarcinomas of the thymic gland.

 

 

RECURRENT HEPATIC LYMPHANGIOMATOSIS

 

Seong H. Ra, Robert F. Bradley, Michael C. Fishbein, Charles R. Lassman

 

ABSTRACT:

 

Hepatic lymphangiomatosis is a rare disease characterized by an abnormal lymphatic proliferation involving the liver alone, liver and spleen, or multiple organs.  Hepatic lymphangiomatosis becomes symptomatic secondary to compression of vital structures due to progressive growth or replacement of the normal parenchyma which can lead to liver failure.  Resection and orthotopic liver transplantation (OLT) can be used as treatment for this disease.  We herein describe a 42 year old female who had undergone successful OLT for hepatic lymphangiomatosis with recurrent disease detected 19 years later in the transplanted liver.  This is, to our knowledge, the first described case of recurrent hepatic lymphangiomatosis after OLT.  We discuss the clinical, radiologic, pathologic, and immunohistochemical findings and review the cases of hepatic lymphangiomatosis that have undergone OLT.

 

 

MUCINOUS CYSTADENOMA OF BORDERLINE POTENTIAL: REPORT OF A CASE AND REVIEW OF THE LITERATURE

 

Seong H. Ra, Gladell Parker, Marcello Franco, Nathan Pinheiro, Mahul B. Amin

 

ABSTRACT: 

 

Mucinous cystadenoma of borderline potential arising within the renal pelvis is a rare occurrence with only two cases reported to date.  This tumor represents a transition from a mucinous cystadenoma to adenocarcinoma, similar to borderline tumors within the ovary.  We report a case of a renal pelvic mucinous cystadenoma of borderline potential producing massive hydronephrosis secondary to mucin accumulation.  A 52 year old male presented with several month duration of painless hematuria and mucosuria.  A nephrectomy revealed a markedly dilated renal pelvis and calyces filled with mucin.  Microscopic sections revealed widespread intestinal metaplasia with foci of high grade dysplasia arising in extensive adenomatous change.  We review the clinical, radiological, immunohistochemical, and pathologic features of this case and review the differential diagnosis and distinguishing characteristics of this lesion.  

 

 

A Neurofibromatosis Type 1 Patient with Multiple Concurrent Gastrointestinal Tumors

 

Renuka Agrawal1, Mariza Deparalta-Venturina2, Jon D.Wilson2

1Department of Pathology, Loma Linda University Medical Center, Loma Linda

2Department of Pathology, William Beaumont Hospital, Royal Oak, Michigan

 

ABSTRACT:

 

Neurofibromatosis type1 (NF1) is one of the most common inherited diseases with an estimated birth incidence of 1:3,000 and autosomal dominant transmission. In addition to cutaneous, soft tissue, and visceral neurofibromas, patients with NF1 have an increased incidence of other tumors. The gastrointestinal manifestations of NF1 are not uncommon, although they are less frequent than neurocutaneous manifestations. The spectrum of NF1-associated gastrointestinal lesions includes hyperplastic lesions of gastrointestinal neural tissue and its supporting structures, gastrointestinal stromal tumors (GISTs), and endocrine cell tumors of the duodenum and periampullary region.

We present the clinical and pathological findings in a 63-year-old female NF1 patient with prior history of multiple cutaneous neurofibromas and a sarcoma of the thigh. The patient presented with obstructive jaundice. Endoscopy revealed an ampullary mass which was subsequently resected. Pathologic evaluation of the pancreaticoduodenectomy specimen revealed multiple concurrent neoplasms. These included an obstructing ampullary carcinoid, a separate duodenal mixed carcinoid/adenocarcinoid, multiple neurofibromas and multiple C-KIT positive GISTs.

These lesions have been previously reported in association with NF1. However, to our knowledge the concurrent nature of this patientís tumors is distinctly unusual. This case further highlights the importance of a thorough clinical and pathological evaluation in NF1 patients as they may show multiple concurrent gastrointestinal neoplasms.

 

 

Loss of the Y Chromosome: An Age-related or Clonal Phenomenon in Acute Myelogenous Leukemia/Myelodysplastic Syndrome?

 

Anna Wong, MD (anna.wong@cshs.org), Belle Fang, MD, MS, Ling Zhang, MD, Xiuping Guo, PhD, Stephen Lee, MD, Rhona Schreck, PhD.

Department of Pathology, Laboratory Medicine and Medical Genetics Institute, Cedars Sinai Medical Center, Los Angeles, California.

 

ABSTRACT:

 

Context:  Various publications report that loss of the Y (-Y) chromosome is an age related phenomenon. It is also known that the incidence of acute myelogenous leukemia/myelodysplastic syndrome (AML/MDS) increases with age.  The clinical association between -Y chromosome and AML/MDS disorders has been difficult to delineate because they are both related to aging.  One comprehensive report suggests that -Y chromosome in >75% of cells may indicate a clonal phenomenon that could be a marker for AML/MDS.   This study attempts to evaluate this relationship.

 

Design:  A retrospective review of cytogenetic reports of male patients from 1996 to 2007 was performed.   Karyotypes with ĖY chromosome were stratified based on the percentage of cells missing the Y.  Age and bone marrow biopsy diagnosis was collected.  Association between myeloid disorder and ĖY chromosome was evaluated by logistic regression analysis, with and without adjusting for age effect.

 

Results:  144 patients showed ĖY chromosome.  An increased incidence (P <0.05) of AML/MDS were seen only in patients with -Y chromosome in 100% of cells (Table 1).

 

TABLE 1: Summary statistics for loss of Y chromosome.

 

Mean Age

Total Patients

AML/MDS

P value- (Compared to normal)

P value Ė Age Adjusted

 

(Compared to normal)

Normal

73.09

99

13 (13.13%)

-

-

<25% -Y

74.57

51

3 (5.88%)

0.18

0.16

25-74% -Y

78.22

54

5 (9.26%)

0.48

0.34

75-99% -Y

79.11

19

1 (5.26%)

0.35

0.28

100% -Y

82.35

20

8 (40.00%)

0.01

0.03

 

Conclusions: Loss of the Y chromosome, in most cases, appears to be primarily an age-related phenomenon, with the percent of cells missing the Y chromosome increasing with age.  However, in individuals where all cells in the bone marrow show ĖY chromosome, there is a statistically significant increase in AML/MDS compared to patients with some or all 46,XY cells, even after adjusting for age, suggesting that the absence of any normal cells may be more indicative of AML/MDS.    

 

 

Has the 2004 Revision of the International Society of Heart and Lung Transplantation (ISHLT) Grading System Improved the Reproducibility of the Diagnosis and Grading of Cardiac Transplant Rejection?

 

H Yang1, C Lai1, T Baruch-Oren1, S Ra1, W Watts1, WD Wallace1, DW Gjertson1 and MC Fishbein

Pathology and Laboratory Medicine, UCLA David Geffen School of Medicine, Los Angeles, CA, United States.

 

ABSTRACT:

 

Background: In 1990, the ISHLT created a working formulation to standardize the diagnosis and grading of cardiac transplant rejection. In 2004, the grading scheme was revised, such that prior grades 1A, 1B, and 2 were incorporated into a new grade 1R, prior grade 3A became grade 2R, and prior grade 3B and 4 became grade 3R. With better defined and fewer grades of cellular rejection, reproducibility was expected to improve.

 

Design: To test this hypothesis, we examined the interobserver reproducibility of both the 1990 and the 2004 Revised ISHLT Classification for Cardiac Allograft Rejection using Kappa statistics, generally regarded as the statistic of choice for measuring inter-observer agreement. Six independent observers (2 residents, 2 fellows, and 2 attending pathologists) graded the H&E stained slides of 175 endomyocardial biopsies according to the 1990 grading system. These grades were then converted into the 2004 revised grading system. The evaluation was carried out blindly and the agreement of the diagnoses was analyzed.

 

Result: The combined Kappa value of all grades diagnosed by all six reviewers was 0.369 for the 1990 grading system and 0.402 for the 2004 grading system (not statistically significant). Kappa values for grades 1B (0.265)/1R (0.397) and 3A(0.230)/2R(0.230) were lower than for all other grades, except grade 4 (0 = 0.466; 1A = 0.352; 2 = 0.508; 3B = 0.424; 4 = 0.239). Of note, combining the grades 3B and 4 (kappa values of 0.424 and 0.239, respectively) into one grade, 3R, did improve the agreement for high-grade lesions (grade 3R) (kappa value of 0.524).

 

Conclusion: In this study, the new 2004 ISHLT grading system for cardiac transplant rejection did not improve the inter-observer reproducibility when compared to the 1990 grading system, in large part because pathologists have difficulty in agreeing upon grades 1B/1R and 3A/2R rejection. In order to achieve better reproducibility with the current grading system, better criteria to define grades 1B/1R & 3A/2R are needed.

 

 

*GENE EXPRESSION MICROARRAY SYSTEM USED AS A MOLECULAR DIAGNOSTIC TOOL TO DISTINGUISH MALIGNANT MELANOMA VS NEVUS FROM FFPE TISSUE.

 

1Koh, S.S.; 1Abrishami, P.; 1Wei, J.; 2Opel, M.; 2Yau, K; 2Shaw, R.; 3Shibata, P.; 3Tau, Y.; and 1Binder, S.

1UCLA Medical Center, Los Angeles, CA, USA., 2Combimatrix Molecular Diagnostics, Irvine CA, USA, and 3Pathology Inc., Torrance, CA, USA.

 

ABSTRACT:

 

Background: Distinction of melanoma from a benign nevus may be very difficult when only standard histologic criteria are utilized.  A study of 11 expert pathologists reviewing 37 classic melanocytic lesions showed that they were in agreement in only 30% of the cases (Hum. Pathol. 1996;27:1115-1116).  Since the clinical management and prognosis of patients is entirely dependent on pathologic diagnostic accuracy, the goal of our study is to develop an objective molecular diagnostic method using the gene expression microarray system to distinguish melanoma from nevus of all types.

 

Design: 62 melanomas and 58 common nevi were retrieved from the archival files.  Slides were created from formalin fixed paraffin embedded (FFPE) blocks and cells of interest were isolated with laser capture microdissection.  Total mRNA was isolated, amplified using a modified T7-T3 amplification protocol (Genisphere), and labeled.  The samples were then hybridized to a 12K CustomArray (Combinatrix, Seattle, WA, USA) using a two color format and a Stratagene universal RNA reference.  Data analysis was performed to create a gene list database.

 

Results: There was a significant difference in gene expression of the most altered genes distinguishing melanoma from nevus.  An initial 120-gene signature was created and found to be consistent with some previously identified genes that had strong correlation with melanoma and melanocytic lesions.  Our method showed melanomas, relative to nevi, had increased expression of some genes such as PCNA, GSPT1, PHACTR1, Stat1, and ARPC2 (p=1.19E-05 to 5.43E-07) and decreased expression of FABP7, DLC1, GPX3, and Ellsl (p=8.06E-06 to 1.30E-09), to name a few within our  initial gene signature.

 

Conclusions: The distinctive gene expression profile of melanoma and nevus offers the ability to distinguish them by an objective molecular measure.  This offers the possibility of utilizing the gene expression microarray as a future molecular diagnostic tool to distinguish melanoma from melanocytic lesions with uncertain biologic behavior, with high sensitivity and specificity.  Additionally, the use of FFPE tissues increases practicality since the majority of clinical specimens are formalin fixed/paraffin embedded, facilitating the movement of microarray based technologies from the lab bench to the clinical bedside.

 

*DOES SIZE MATTER?  COMPARISON STUDY BETWEEN MRI, GROSS, AND MICROSCOPIC TUMOR SIZES IN BREAST CANCER IN LUMPECTOMY SPECIMENS

 

Bita Behjatnia MD1, Julie Sim MD2, Lawrence W. Bassett MD2, Neda A. Moatamed MD1, and Sophia K. Apple MD1

1 Pathology & Laboratory Medicine, University of California in Los Angeles, 10833 Le Conte Avenue, Box 951732, A7-149 CHS, Los Angeles, CA 90095-1732

2 Radiologic Sciences, University of California in Los Angeles, Box 951721, BL-428 CHS, Los Angeles, CA 90095-1721

 

ABSTRACT:

 

Background: In invasive breast cancer, tumor size is one of the most important factors to determine disease-free and cause-specific survival.  MRI is being used with increasing frequency for detection and measurement of breast lesions and is known to be the most sensitive imaging analysis in assessing tumor size.  Size of the invasive lesion is not only essential in staging the cancer but also in determining the type and extent of the subsequent surgical and oncological management.  Despite accurate technology used in radiologic fields, the true size of the lesion(s) can be overestimated or underestimated in some patients and may change the subsequent course of action.  Equally, tumor size as determined by gross examination of the specimen can be underestimated or overestimated depending on the manner of sectioning and or by including contiguous areas of non-invasive lesions. 

 

Objectives: 1) To determine the accuracy of MRI and gross pathology in estimating the tumor size using microscopic pathology as the gold standard; 2) To assess the change in the T stage of the cancer when size of tumor by MRI and gross pathology are different from microscopic pathology.

 

Methods: A retrospective study was done on thirty-seven cases from thirty-three female patients, some with multiple lesions excised within the same patient. Patients were between 30 and 85 years and all underwent bilateral breast MRI with subsequent lumpectomy between 2002 and 2006. All mastectomy cases were excluded from the study. All patients had diagnosis of invasive breast cancer with or without in-situ component.  The histological tumor size was used as the gold standard by submitting the entire lumpectomy specimen in a sequential manner.  The size of the lesion(s) on MRI and gross pathology were compared with the microscopic tumor size.  The percentage of cases where MRI and gross had overestimated or underestimated the tumor size of the invasive cancer was calculated.   In each case, based on the maximum dimension of the invasive tumor from MRI, gross, and microscopy, a determination was made whether the T stage of the cancer was affected and the percentage of such changes were calculated for both invasive lobular and ductal carcinomas.

 

Results: Among thirty-seven cases, twenty-seven (73%) of the cases had invasive ductal (IDC) and 10 (29%) invasive lobular carcinoma (ILC).  Overall, tumor size by MRI matched exactly the same histological size in only 3%, underestimated 27%, and overestimated 70% of cases. By MRI analysis, T stage was altered 24% of the time, 6% into higher T-stage and 18% into lower T-stage.  Overall, tumor size by gross examination matched exactly the same histological size in 22%, underestimated 58%, and overestimated 22% of cases.  By gross pathologic size, T stage was altered 58% of the time, 39% into higher T-stage and 3% into lower T-stage.

MRI

 

IDC

ILC

T-stage

IDC

ILC

 

Underestimated

15%

60%

Changed

21%

33%

 

Overestimated

81%

40%

into higher

0

67%

 

Same size

4%

0%

into lower

100%

33%

Gross

Underestimated

52%

70%

Changed

40%

44%

 

overestimated

30%

0

into higher

90%

100%

 

Same size

18%

30%

into lower

10%

0

Conclusions: Our study shows that MRI and gross pathology are both less accurate in predicting the tumor size when compared with microscopic tumor size.

 

1) T-stage of the invasive tumor size was more accurately predicted by MRI (76%) than gross pathology measurement (42%).

2) Both MRI and gross pathology tumor size are better in predicting the actual tumor size in IDC than ILC.

3) MRI tends to overestimate IDC tumor size (81%) affecting T-stage in 21% of the cases.

4) MRI tends to underestimate ILC tumor size (60%) affecting T-stage in 33% of the cases.

5) Gross measurement more commonly underestimates (52%) than overestimate (30%) IDC tumor size, affecting T-stage in 40% of the cases.

6) Gross measurement tends to underestimate ILC tumor size (70%) affecting T-stage in 44% of the cases.

7) MRI as an imaging modality for invasive breast lesions as well as gross measurement of tumor size has significant limitations particularly in cases of invasive breast carcinoma.  Our results show that the microscopic measurement of the tumor size is necessary for accurate T-stage.

 

References:

1) Liberman L, Mason G, Morris EA, Dershaw DD. Does Size Matter? Positive Predictive Value of MRI-Detected Breast Lesion Size. AJR 2006; 186: 426-430.

2) Lagios MD. Problems in the assessment of tumor size: an elusive grail in current practice. Seminars in Breast Disease 2005; 8:24-30.

3) Sundararajan S, Tohno E, Kamma H, Ueno E, Minami M. Detection of Intraductal     component around Invasive Breast Cancer Using Ultrasound: Correlation with MRI and Histopathological Findings. Radiation Medicine 2006; 24(2): 108-114.

4) Hata T, Takahashi H, Watanabe K, Takahashi M, Taguchi K, Itoh T, Todo S. Magnetic Resonance Imaging for Prospective Evaluation of Breast Cancer: A comparative Study with Mammography and Ultrasonography. J Am Coll Surg 2004; 198:190-197.

5) Wiberg MK, Aspelin P, Sylvan M, Bone B. Comparison of Lesion Size Estimated by Dynamic MR Imaging, Mammography, and Histopathology in Breast Neoplasms.

Eur Radiol 2003; 13: 1207-1212.

 

 

An Online Decision Support System for Diagnosing Hematopoietic Malignancies By Flow Cytometry Immunophenotyping 

 

You-Wen Qian, MD1, Dinesh Mital PhD2, Stephen. Lee, MD1

1 Cedars Sinai Medical Center, Los Angeles, CA

2University of Medicine and Dentistry in New Jersey, Newark, NJ

 

ABSTRACT:

 

An online decision support system for hematopoietic neoplasm based on virtual flow has been developed. Rules were implemented via extensible Markup Language (XML). 153 cases representing 28 different hematopoietic neoplasms were correctly classified. Further testing for unknown cases is undergoing.

 

Introduction: Flow cytometry is an essential tool to characterize hematologic malignances. Almost all laboratories use stand-along software to analyze flow cytometry data, coupled with manual interpretation which is time-consuming since multi-steps need to be taken. The process is error-prone due to manual data input. For pathologists who are not subspecialized in hematopathology, an online automated interpretation system will be useful in making diagnosis. For expert Hematopathologist, such system can be used for screening assessment.  It can also sever as teaching tool for pathology residents.

 

Methods and Results: Several stand-alone systems have been built for automated analysis of flow cytometry immunophenotyping results for malignant lymphoma and leukemia,1 2.

 

In this study, a knowledge-based decision support system to interpret online flow cytometry results for hematopoietic neoplasms have been developed as a complete Client-Server application by using Java programming and XML technology. The listmode flow cytometry data files are imported to the online decision support system where gating, dotplot, histogram and contour plot can been performed. Upon gating, the cell clusters of designation (CD markers) results are generated in percentage of cell population, based on which the positive or negative designation for CD markers is automatically assigned. By analyzing 273 of flow cytometry data profiles including both normal condition and hematopoietic neoplasms, we have set the threshold for CD markersí positivity and B cell clonality (kappa and lambda ratio less than 0.5 or more than 4). Our expertise on diagnosing hematologic disorder is utilized in developing a semantic network of knowledge base.  Confidence level of final differential diagnosis is calculated through a formula from the confidence factor of CD markers in making a particular diagnosis. Inference engine was implemented by JAVA programming (version 1.5.0) and XML (version 1.0) where tree structure and search algorithm are employed.  The decision support system is hosted in an Apache-Jakarta Tomcat server container (version 5.5.20).  A set of 153 flow cytometry listmode data files following flow cytometry data standard 2.0 with distinctive immunphenotyping profile covering both normal and 28 different hematopoietic neoplasms using World Health OrganizAtion (WHO) criteria are fed into the system and correct diagnosis was made. To further validate this online decision support system, a large case volume with unknown diagnosis will be tested. This is the first reported online decision support system implemented with XML and JAVA programming to interpret flow cytometry immunophenotyping results for hematopoietic neoplasms.

 

Conclusion: The website (http://www.flowcytometryonline.com) has been setup for online decision support where listmode data files located either in Sever side or Client side can be accessed and analyzed. Stand-alone application is also provided when internet access is limited. The system is expected to facilitate clinical diagnosis of hematologic neoplasms.

 

References

1.    Cualing HD. Automated analysis in flow cytometry. Cytometry Communications in Clinical Cytometry. 2000;42:110Ė113.

2.    Thews O, Thews A, Huber C, and Vaupel P. Computer-assisted interpretation of flow cytometry data in hematology. Cytometry 1996;23:140-149.

 

 

ProExC As A Marker Of HPV-Associated Squamous Lesions Of The Cervix

 

Riem El-Sabbagh Badr, AE Walts and S Bose. Department of Pathology & Laboratory Medicine, Cedars-Sinai Medical Center, Los Angeles, CA, US

 

ABSTRACT:

 

Background: Based upon DNA microarray analysis, TriPath Imaging (Burlington, NC) recently developed ProExC, an antibody that targets minichromosome maintenance protein 2 and topoisomerase II, two novel biomarkers associated with cervical neoplasia. Initial studies suggesting high specificity and sensitivity for detection of high grade squamous intraepithelial lesions (HSIL) in cytology liquid preparations have been reported. Our study was designed to assess ProExC staining as an adjunct in the diagnosis and grading of cervical biopsies for HPV-associated SILs.

 

Design: Slides of 98 cervical biopsies were retrieved from our files. Patients ranged in age from 19 to 83 years (mean: 37 yrs). Based on H&E stains, the diagnoses were: negative (38), condyloma &/or CIN I (23), CIN II/III (37). Each biopsy and appropriate controls were immunostained for ProExC in accordance with standard protocols and manufacturerís recommendations. All the cases were also immunostained for p16 (Biocare Medical), 96 cases for Ki67 (Ventana) and 56 cases were subjected to in-situ hybridization (ISH) utilizing the Inform HPV Family 6 and 16 probes (Ventana). ProExC was recorded positive when more than 50% of lesional nuclei stained. P16 was recorded positive when more than 25% of lesional nuclei (spotty) stained or when there was band-like staining while Ki67 was recorded positive when more than 50% of lesional nuclei stained as previously described by us in a study of anal biopsies. ProExC staining was correlated with H&E diagnoses, p16/Ki67 stains, and HPV ISH.

 

Results: ProExC positivity was present in 48% of the condyloma/CIN I, and in 92% of the CIN II/III. 3% of the negative cases were positive for ProExC. ProExC positivity was strongly correlated with the presence of HPV-associated lesions, high risk HPV-DNA, Ki67 positivity, and p16 positivity (p<0.001 for each). As defined in our study, ProExC was associated with 96% sensitivity and 93% specificity in detecting HSIL. In detecting HPV as determined by ISH, ProExC was associated with 82% sensitivity and 75% specificity.

 

Conclusions: ProExC positivity strongly correlates with HPV-associated lesions and with the presence of HPV DNA as determined by ISH. Nevertheless it is not helpful in distinguishing low from high grade SILs. Sensitivity, Specificity, PPV and NPV of ProExC in detecting HPV-associated lesions are similar to that of p16 & Ki67 immunostaining.

 

 

*The Spleen in Patients with HIV and Castlemanís disease

 

Gretchen Galliano, M.D. and Randa Alsabeh, M.D.

Department of Pathology, Cedars-Sinai Medical Center, Los Angeles, CA

 

ABSTRACT:

 

Background:  Castlemanís disease is a lymph node hyperplasia with two histologic variants: hyaline-vascular type and plasma cell type.1 Hyaline-vascular type is classically a benign, isolated lymphadenopathy, and plasma-cell type is characterized by large numbers of plasma cells.1,2 Multicentric Castlemanís disease (MCD) is a lymphoproliferative disorder characterized by lymphadenopathy which often consists of histologic features of both variants and systemic symptoms related to cytokine dysregulation such as fevers and chills3,4. Multicentric Castlemanís is associated with Human Herpes Virus 8 (HHV8) infection, and in the setting of HIV infection signifies a poor prognosis despite treatment with immune modulators or chemotherapeutic agents.4,5,6,7 There are no reports characterizing the spleen in patients with HIV and MCD even though there is an associated splenomegaly.  In this abstract, we describe the first series of Castlemanís disease in the spleen of HIV patients.

 

Methods:    The spleens of five patients with MCD and HIV were examined and key clinical, gross, and histologic findings characterized.  HHV8 immunostaining and clonality by Kappa and lambda was performed on all five spleens. 

 

Results:  Fours spleens from patients with HIV and MCD were removed during life, and one spleen was examined where the diagnosis of MCD was made at autopsy.  (See clinical data table 1).

 

Table1. Clinical data

Patient

Age (y)

Symptoms/Signs

CD4

PB

HAART

Reason

Treatment

Outcome

1

40

F, N, H

-

A

+

Lymphoma

C,S,E

D (5 m)

2

56

F,N, H

154

A,T

+

Autopsy

S

D (2 m)

3

37

F,N,V, R,H

<100

A,T,L

+

hypersplenism

-

D (2 m)

4

33

F

7

A,T

+

ITP, torsion

S,E

D (2 m)

5

36

F, W, R,V

6

A,T,L

+

hypersplenism

S

D (1 m)

F = fever; N = night sweats; W = weight loss; V= vomiting/diarrhea; R= Rash; H = hepatomegaly; PB = peripheral blood; A = anemia; T = thrombocytopenia; L = leukocytosis; Reason = reason for splenectomy; C = CHOP; S = steroids; E = etoposide; D = death; m = months after presentation

 

All spleens were enlarged with an average maximum dimension of 24.2 cm (SD = 4.1 cm). The average weight was 1700 g (SD = 560 g).  There were several common abnormalities seen on histology (see table 2).

 

Table 2. Key histologic features

Patient

Size (cm)

Weight (g)

HHV8

Clonality

EPS

Infarct

Lymphoma

1

26

2100

+

Lambda

-

-

+

2

-

1000

+

Polyclonal

+

+

-

3

25.5

2270

+

Polyclonal

+

+

-

4

18.1

1200

+

Lambda

+

+

+

5

27.2

1910

+

Lambda

+

-

+

EPS = erythrophagocytosis

 

All spleens showed collections of plasmablastic cells. The plasmablasts were located primarily around follicles; however, plasmablasts were occasionally seen in the red pulp. All spleens showed numerous plasma cells primarily noted within the red pulp. The white pulp is typically poorly defined or shows regression. Infarcts were often single infarcts, and ranged from 0.3 cm to 4.5 cm.

 

Cases 1, 4, and 5 had lymphoma within the spleen, and all three were lambda clonal.  Case 1 had high-grade lymphoma with immunoblastic and plasmacytoid differentiation and stained positive for CD20, CD43, and lambda. This patient had confirmed Castlemanís disease in a biopsy of the right femoral and left axillary lymph nodes.  Case 4 showed plasmablastic lymphoma positive for lambda light chain and negative for CD20, CD3, EBV-LMP, and EBER. Case 5 had a microscopic focus of plasmablastic lymphoma in a background of plasmablastic variant of Castlemanís disease in the spleen. The lymphoma was CD138 and lambda positive and negative for CD3 and CD20.

 

Discussion:  We present the largest series of spleens to date in patients with MCD and HIV. All spleens examined were involved with Castlemanís disease, and Human herpes virus 8 infection was evident in all cases with strong staining in cells with plasmablastic morphology. Castlemanís disease with HHV8 positive plasmablasts was also present in the lymph nodes examined in these patients in other biopsies performed.  Erythrophagocytosis was a common finding, and was present to a marked degree.

 

Three out of five patients had lymphoma upon examination, and two of these cases were unsuspected prior to splenectomy. Lambda clonality was present in the three spleens with lymphoma, and staining with kappa and lambda may be essential in diagnosing lymphoma in these patients.

 

All patients died within 5 months after presentation for splenectomy.  The patient with MCD diagnosed at autopsy was admitted for pneumonia.

 

This series indicates that splenomegaly in a patient with Castlemanís disease and HHV8 is an extremely poor prognostic sign, and the spleen in these patients may harbor occult lymphoma.

 

References:

1. Waterson A, Bower M. Fifty years of Multicentric Castlemanís Disease. Acta Oncologica 2004;43:698-704.

2. Hoffman C. HIV Medicine. 14th edition, 2006

3. Oskenhendler E, Duarte M, Soulier J, et al. Multicentric Castlemanís Disease in HIV infection: a clinical and pathological study of 20 patients. AIDS 1996; 10:61-7.

4, Nishinmoto et al. Humanized anti-IL6 treatment of Multicentric Castlemanís Disease. Blood 2005; 1-6:2627-2632.

5. Segolene N, Agbalika F, Rabian C et al. Failure of rituximab in HIV associated Multicentric Castlemanís Disease. Am J Hematol 2005; 79:337-339.

6. Bacon C et al. Pathology of bone marrow in Human Herpes 8 associated Multicentric Castlemanís Disease. Brit J Haematol 2004; 127: 585-591.

7. Oskenhandler et al. Failure of cidofovir in HIV associated Castlemanís Disease. Blood 2004; 103:4368

 

 

JAK2 V617F Mutation Is Infrequent in "5q-Syndrome" and 5q-Associated MDS.

 

Ling Zhang, Saskia Gueller, Xuemin Li, Phillip H. Koeffler, Stephen Lee, Qin Huang, Pathology and Laboratory Medicine, Cedars Sinai Medical Center, Los Angeles, CA, USA; Experimental Hematology, Cedars Sinai Medical Center, and Division of Pathology, City of Hope National Medical Center, Los Angeles, CA, USA

 

ABSTRACT:

 

Background: V617F mutation in Janus Kinase 2 (JAK2) gene has been found in chronic myeloproliferative disorders (MPD) including polycythemia vera (90%), essential thrombocythemia and chronic idiopathic myelofibrosis (30-50%), and occasionally in myelodysplastic syndromes (MDS). "5q- Syndrome" is a MDS that shares features with MPD and characterized by an atypical megakaryocytic hyperplasia in bone marrow and usually thrombocytosis in peripheral blood. The most common deleted region for this syndrome is 5q13.3q33.1. An interstitial deletion with variable proximal (5q12-14) and distal (5q31-33) breakpoints has been found in other MDS with/without additional chromosomal abnormalities beyond "5q- Syndrome". To date, JAK2 mutation was detected in 6/97(6.2%) of patients having diagnosis of MDS with "5q- Syndrome".

 

Design: In our study 21 MDS patients (10 with "5q- Syndrome" and 11 MDS with isolated or complex 5q-) whose diagnosis by both bone marrow aspiration/biopsy and conventional chromosomal analysis were confirmed.

 

Materials and Method: Both regular PCR and allele-specific PCR (AS-PCR) were performed. 1). Regular PCR: Primers were created to amplify a 460 bp fragment containing the site of JAK2 V617F mutation in exon 12 of chromosome 9. Forty-five cycles of PCR were performed at an annealing temperature of 570 C. Resulting PCR product was digested with 2 U BsaXI for 16 hours and with an additional 2 U BsaXI for another 16 hours at 370 C, then analyzed on a 2% agarose gel. The mutant allele remained undigested whereas the wild-type allele was digested into 241 bp, 189 bp and 30 bp fragments. All experiments included a positive (HEL cells) and negative (K562 cells) control. 2). an allele-specific polymerase chain reaction (AS-PCR) with high sensitivity to detect JAK2 mutation was also performed. In a single-tube multiplex system, two pairs of primers that are designed specifically to amplify the normal (wild-type JAK2 gene with two peaks at positions 126 and 228) and mutant allele (peak at position 156). In addition, the forward primer from one set and the reverse from the other are able to amplify the wild type JAK2 gene as internal control.  The PCR products are separated and detected by capillary electrophoresis in the ABI 3100 Genetic Analyzer.

 

Results: PCR results showed clear wild type PCR patterns in all 21 cases. AS-PCR also confirmed the no JAK2 mutations in all cases with deletion of 5q.

 

Conclusion: No JAK2 mutations were detected in 21 patients either with "5q- Syndrome"or other 5q- associated MDS suggesting that JAK2 mutations are infrequent in these MDS patients.

 

 

A Flow Panel in Distinguishing Hypoplastic MDS from Hypoplastic Marrow

 

Ling Zhang1, Jean R Lopategui1, Sharon E Kelly1, Tobi Neer1 and Stephen Lee1. 1Pathology and Clinical Medicine, Cedars Sinai Medical Center, Los Angeles, CA, United States.

 

ABSTRACT:

 

Background: Hypoplastic MDS (hMDS) appears to be a distinct clinicopathologic entity, accounting for 15% of all MDS. It can be difficult to diagnose based on morphologic bone marrow examination or cytogentic analysis due to paucicellularity. Flow cytometry (FCM) approaches have not been well described. We devised an 8-parameter FCM panel, mainly including myeloid and erythroid maturation markers, to differentiate hMDS marrows from normal marrows.

 

Design: 65 patients with cytopenia(s) or anemia were included (10/2005-07/2006). Cases were divided into 3 groups: A) normorcellular with normomorphology, 41 cases, B) hypocellualr with normal or unequivocal morphology, 14 cases. C) Morphologically diagnostic of MDS, 10 cases. An 8 FCM was performed and scored: 1) hypogranularity, 2) aberrant expression of CD56, 3) lack of CD10 expression, 4) decreased CD64 expression, 5) and 6) lack of CD13 or CD33 expression, 7) CD 34 expression on all cells excluding erythroid, 8) decreased expression of CD71/Glycophorin gating on erythroid precursors. Karyotypings were analyzed.

 

Result: Our FCM panel can discriminate hypocellualr marrow from hMDS (Table 12). Normal karyotypes were found in both group A (39) and B (10), but abnormal in 3 of 9 in group C.

 

Conclusion: In the 8-parameter FCM panel, score > 3 more suggested hMDS. Hypogranularity, increased CD34 might be useful in discriminating hMDS from hypoplastic marrow. Decreased CD71 was frequently found in hMDS with erythoid dysplasia.

 

Table 1. Comparison of 8 Flow Parameters in Patients with/without hMDS

Parameters

Normal (A)

Hypocellular (B)

hMDS (C)

*P Value (B vs C)

Hypogranularity

4/41 (9.8%)

1/14 (7.1%)

7/10 (70%)

0.001

CD56

0/41 (0%)

1/14 (7.1%)

3/10 (30%)

0.14

CD10

5/41 (12.1%)

4/14 (28.5%)

6/10 (60%)

0.12

CD64

5/41 (12.1%)

4/14 (28.5%)

3/10 (30%)

0.94

CD13

0/41 (0%)

0/14 ( 0%)

0/10 (0%)

1

CD33

0/41 (0%)

1/14 (7.1%)

1/10 (10%)

0.80

CD34

1/41 (2.4%)

1/14 (7.1%)

4/10 (40%)

0.05

CD71

2/41 (4.9%)

2/14 (14.3%)

6/10 (60%)

0.02

*Chi-square Test
 

Table 2. Flow Scoring in Patients with/without hMDS

Scores

Normal (A)

Hypocellular (B)

hMDS (C)

Score <3

41/41 (100%)

12/14 (85.7%)

1/10 (10%)

Score >=3

0/41 (0%)

2/14 (14.3%)

9/10 (90%)

X+/-SD

0.66 +/- 0.69

1.29 +/- 1.14

3.30 +/- 1.06

             

Student t-test: P Value: A vs B=0.07, B vs C<0.001, A vs C <0.0001

 

 

Loss of Y Chromosome with Trisomy 15 in Elderly Not Associated with Myelodysplastic Syndrome

 

Ling Zhang1, Anna Wong1, Rhona Schreck2 and Stephen Lee1. 1Department of Pathology and Clinical Medicine, Cedars Sinai Medical Center, Los Angeles, CA, United States and 2Department of Cytogenetics, Cytogenetic Institute, Cedars Sinai Medical Center, Los Angeles, CA, United States.

 

ABSTRACT:

 

Background: Loss of the Y chromosome is not an unusual finding in the bone marrow of elderly males1 although it has been associated with acute non lymphocytic leukemia2. Trisomy 15 is found in some hematological malignancies including MDS3. Y loss in conjunction with trisomy 15 is a rare genetic combination. There was controversy as to whether the specific chromosomal alteration (loss of Y and trisomy 15) is an effect of aging, as opposed to being related to the development of MDS or AML3, 4 .Our study tries to identify if the loss of Y chromosone and trisomy 15 in our group of patients is a result of MDS or not.

 

Design: Retrospective study was designed. Archival cytogenetic data was retrieved for patients who had both a loss of Y chromosone and trisomy 15 from 03/1990 to 05/2006. Seven male patients with age ranging form 71 to 91(average 83.4) year old were available for analysis. Medical charts were reviewed for clinical history, laboratory data and bone marrow biopsy reports coinciding with the date of karyotype analysis.

 

Result: All 7 patients with the loss of Y chromosome and trisomy 15 showed no clinical history of MDS or morphological findings suggesting MDS in their peripheral blood or bone marrow examination. See table 1.

 

Conclusion: Loss of the Y chromosone and trisomy 15 is a rare karyotypic combination. In our study all 7 patients with loss of Y and trisomy 15 were elderly (>75 year old) but had no evidence of morphologic MDS. We assumed that this karyotype is related to aging effect but not MDS.

 

Table 1. Summary of Clinical, Pathologic and Cytogenetic Resultsd>

ID

Age

Clinical Diagnosis

Specimen

Pathologic Diagnosis

Karyotyping

RF

77

Anemia, thrombocytopenia

BM

no MDS

45 X,-Y[2]/46 X,-Y,+15[5]/46;XY[18]

HK

86

Iron deficiency anemia, diverticulosis

BM

no MDS

46 X,-Y,+15[2]/46 XY[33]

MW

90

Anemia, GI polyps

BM

no MDS

46 X,-Y,+15[5]/46 XY[15]

IH

91

Anemia

BM

no MDS

45 X,-Y[4]/46 X,-Y,+15[5]/46 XY[11]

MB

82

Anemia, thrombocytopenia, chronic renal failure

BM

no MDS

46 X,-Y,+15[20]46 XY[1]

DP

83

Anemia, demensia, multiple stroke attacks

BM

no MDS

46 X,-Y,+15[20]

WD

75

Anemia, diabetic, chronic renal failure

BM

no MDS

45 X,-Y[1]/46 X,-Y,+15[1]/46 XY[23]

 

References:

1. Kigdon Cancer Cytogentics Grouop: Genes Chromosone Cancer 1992

2. Wiktor A et al: Genes Chromosomes Cancer 2000

3. Smith A et al: Cancer Genet Cytogenet 1999

4. Sinclair EJ et al: Cancer Genet Cytogenet 1998

 

 

Unusual Parotid tumors: Rare combination of Sebaceous lymphadenoma with other neoplasms

 

Renuka Agrawal, MD Loma Linda Medical University Center

Mia Perez, MD Loma Linda Medical University Center MPerez@ahs.llumc.edu

C Odell, MD Park View Community Hospital

 

ABSTRACT:

 

BACKGROUND & AIMS: Primary sebaceous lymphadenomas of the parotid gland are unusual benign tumors in adults. These, in combination with other existing lesions, are even more rare.

 

METHODS: Two specimens of superficial parotidectomy were received & stained with H&E. We present gross & histological findings.

 

RESULTS: Case 1: An 86 year old male had a stable mass in the angle of the jaw since 20 years. The specimen showed a prominent 3 cm tan nodule & separate distinct 7 mm brown nodule, both well circumscribed. Microscopically, the larger nodule was composed of nests of sebaceous gland like bland epithelial cells, surrounded by dense lymphoid stroma. We identified it as sebaceous lymhadenoma. The smaller nodule showed an unencapsulated collection of cells with abundant eosinophilic granular cytoplasm & a small round bland central nucleus. More such smaller collections of cells were seen in other parts of the gland, some with cystic areas. We identified these multiple lesions as nodular & cystic oncocytosis, with a dominant 7 mm nodule.

 

Case 2: a 72 year-old male had a slowly growing, mobile, asymptomatic parotid mass. FNA diagnosis was epidermal inclusion cyst. The specimen showed 2 distinct lesions; a 2 cm firm brown nodule & a 3.5 cm partly cystic lesion with fleshy tan areas, both well circumscribed. Microscopically, the smaller firm nodule showed a partly cystic tumor lined by a bilayered papillary oncocytic epithelium & lymphoid stroma. The epithelial cells had abundant eosinophilic granular cytoplasm & a small round bland central nucleus. The cystic spaces were filled with cholesterol clefts. we identified it as Warthin tumor. The larger lesion showed keratin filled cysts lined predominantly by bland squamous cells with dense lymhoid stroma showing numerous islands, ducts, & smaller cysts lined by similar cells. Focal sebaceous differentiation was seen. The prominent cystic component clinically simulated an epidermal inclusion cyst. We identified it as sebaceous lymhadenoma.

 

There was no evidence of malignancy in either of these cases.

 

CONCLUSIONS: Although sebaceous glands are present in the parotid, primary sebaceous tumors are rare. To our knowledge, very few cases of sebaceous lymhadenoma have been reported in literature. These tumors are benign, have a very low recurrence rate & are amenable to conservative surgery. It is important to recognise them to avoid overtreatment.

 

 

Utility of a comprehensive immunohistochemical (IHC) panel in the differential diagnosis of spindle cell lesions of the urinary bladder

 

D Westfall1, AL Folpe2, GP Paner1, E Oliva3, AM Gown4 and MB Amin

1 Cedars-Sinai Medical Center, Los Angeles, CA, United States;

2 Mayo Clinic, Rochester, MN, United States;

3 Massachusetts General Hospital, Boston, MA, United States; and

4 PhenoPath Laboratories, Seattle, WA, United States.

 

ABSTRACT:

 

Background: Spindle cell lesions of the urinary bladder are relatively uncommon, but pose a significant diagnostic challenge, as the morphology and immunoprofile of pseudosarcomatous processes in this location significantly overlap with spindled epithelial and mesenchymal tumors. The utility of a panel including antibodies to the recently described basal cell markers p63 and cytokeratin 5/6 (CK5/6) along with those to pan-cytokeratin (CK), smooth muscle actin (SMA) and ALK1 protein has not been analyzed in this diagnostic context.

 

Design: 46 spindle cell lesions including 10 pseudosarcomatous myofibroblastic proliferations (PMP) (related and unrelated to prior procedure), 23 sarcomatoid urothelial carcinomas (Sarc Ca), 12 leiomyosarcomas (LMS) and 1 leiomyoma were immunostained for p63, CK5/6, CK, SMA, and ALK1.

 

Result: IHC profile of spindle cell lesions of the urinary bladder:

 

Grade

p63

Alk-1

SMA

CK5/6

PMP (n=10)

0

3/10

9/9

8/10

0/10

9/9

 

1+

0/10

0/9

1/10

1/10

0/9

 

2+

4/10

0/9

1/10

4/10

0/9

 

3+

3/10

0/9

0/10

5/10

0/9

 

Total + (%)

7/10 (70)

0/9 (0)

2/10 (20)

10/10 (100)

0/9 (0)

Sarc Ca (n=23)

0

7/23

12/23

23/23

6/23

17/23

 

1+

2/23

3/23

0/23

8/23

1/23

 

2+

2/23

3/23

0/23

6/23

1/23

 

3+

12/23

5/23

0/23

3/23

4/23

 

Total + (%)

14/23 (61)

11/23 (49)

0/23 (0)

17/23 (74)

6/23 (26)

Smooth muscle tumors (n=13)

0

6/13

10/13

13/13

2/13

12/12

 

1+

3/13

2/13

0/13

2/13

0/12

 

2+

3/13

1/13

0/13

2/13

0/12

 

3+

1/13

0/13

0/13

7/13

0/12

 

Total + (%)

7/13 (54)

3/13 (23)

0/13 (0)

11/13 (85)

0/12 (0)

Grade: 0, < 5% +; 1+, 5-10% +; 2+, 11-50% +; 3+, >50% +

 

Conclusion: Although clinically and biologically distinct, PMP, Sarc CA and LMS of the bladder show significant immunophenotypic overlap, even using an extended panel of antibodies. Although clinical and morphologic correlation remain the cornerstones of this differential diagnosis, an IHC panel composed of CK, SMA, ALK1, p63 and CK5/6 is a useful diagnostic adjunct: the combination of CK-SMA-ALK1 (+) / p63-CK5/6 (-) supports PMP; CK-p63-CK5/6 (+) / SMA-ALK1(-) favors Sarc Ca, and SMA (+) / CK-p63-CK5/6-ALK1 (-) is seen most often in LMS.

 

 

Diagnosis of myelodysplastic syndrome using flow cytometry glycophorin A/CD71 pattern in bone marrow cases with erythroid hyperplasia with dysplasia

 

Danielle Westfall M.D., Ling Zhang M.D., Stephen Lee M.D.

Department of Pathology and Laboratory Medicine, Cedars Sinai Medical Center, Los Angeles, CA

 

ABSTRACT:

 

Introduction: The diagnosis of myelodysplastic syndrome (MDS) is challenging in bone marrow (BM) conditions when erythroid hyperplasia/dysplasia predominates. Cytogenetics can be helpful, but is not sensitive. Also, refractory anemia (RA) and refractory anemia with ringed sideroblasts (RARS) rarely show cytogenetic abnormalities.  There are a few useful flow cytometry (FC) markers to evaluate erythroid hyperplasia/dysplasia, including CD71, CD105, cytosolic H-ferritin (HF), cytosolic L-ferritin (LF), mitochondrial ferritin (MtF) and CD95. The FC pattern analysis of CD71/glycophorin A (GA) in dyserythropoiesis has not been well reported.

 

Design: BMs from 131 patients (average age 67.4 years, range 20-95) with cytopenia(s) were included in the study (10/2005-12/2006). Cases were divided into 3 groups: 1) normal morphology (68 cases), 2) equivocal, morphologically suggestive but not diagnostic of MDS (31 cases), and 3) morphologically diagnostic of MDS (32 cases). Normal FC pattern (bright CD71/ medium bright GA) gating on erythroid precursors was established based on normal BMs. Decreased expression of either CD71 or GA, or both, in over 50% of the erythroid cell population was identified as an abnormal pattern. The degree of morphologic dyserythropoiesis was divided into: mild (<15% of cells with megaloblastic changes) and severe (>15% cells showing megaloblastic changes). Over 5%, but <15% RS was considered increased, while >15% was classified as RARS or refractory cytopenia with multilineage dysplasia with ringed sideroblasts (RCMLD-RS).

 

Results: 1). The three groups showed no significant difference of erythroid hyperplasia.  2). Both erythroid dysplasia and percentage of RS were more frequently seen in MDS. 3) Erythroid dysplasia, but not erythroid hyperplasia, showed a greater incidence of altered CD71/GA pattern.  4). Altered CD71/GA was more useful in predicting severe dysplasia than no or mild dysplasia regardless of coexistence of other lineage dysplasia (Details refer to Tables 1 and 2).

 

Conclusion: 1). Combined semi-quantitative and pattern approach to CD71/GA by FC may distinguish altered antigen expression in cases of MDS with dyserythropoiesis from erythroid hyperplasia.

 

Table 1. Correlation between BM morphologic diagnosis and Variable Factors

Factors

BM morphology groups

P-value*

Normal

(N = 68)

Borderline

(N = 31)

MDS

(N = 32)

Age

68.7 Ī 16.2

65.5 Ī 17.9

66.6 Ī 17.3

0.64

Erythroid Hyperplasia

35% (24)

32% (10)

41% (13)

0.67

Erythroid Dysplasia

21% (14)

58% (18)

78% (25)

<0.0001

Ringed Sideroblasts

0% (0)

6% (2)

31% (10)

<0.0001

Altered CD71/GA pattern

7% (5)

29% (9)

63% (20)

<0.0001

*ANOVA (Age) or Cochran-Armitage Trend Test (other factors)

 

Table 2. Correlation between normal or altered CD71/GA FC pattern and BM Erythroid morphology

BM morphology in Erythroid Lineage

CD71/GA FC pattern

Fisherís Exact Test

Normal *

(n= 96)

Abnormal *

(n = 34)

No Dysplasia (n=73)td>

63% (61)

35% (12)

 

Mild Dysplasia (n=38))

29% (28)

29% (10)

 

Severe Dysplasia (n=19))

7% (7)

35% (12)

p=0.0003

 

 

Intrathoracic Frozen Sections in Patients With a History of Breast Cancer. A Review of "Best Evidence" With Assessment of Pre-test Odds.

 

Jonathon Herbst, MD1, Alberto Marchevsky, MD1, Robert A. Jenders, MD, MS2

1 Pathology, Cedars-Sinai Medical Center, Los Angeles, CA, USA,

2 Medicine, Cedars-Sinai Medical Center, University of California-Los Angeles, Los Angeles, CA, USA.

 

ABSTRACT:

 

Context: Pulmonary nodules in patients with a history of breast cancer are generally thought to be metastatic, although lung cancer and other conditions are common. It is difficult to reliably distinguish primary from metastatic breast neoplasm at the time of frozen section.

 

Design: A systematic review of the literature from 1970-2007 was conducted to query for "best evidence" regarding this question. One hundred and twenty-nine cases of intrathoracic frozen sections in patients with a history of breast cancer performed at Cedars-Sinai Medical Center from 1989 to 2006 were reviewed. Data were analyzed with meta-analysis.

 

Results: 334 cases were reported in the English literature; the odds for metastatic carcinoma were .662 (.190 - 9), for primary lung cancer were .590 (.098 - 2.33) and for benign conditions were .285 (0 - 3). The odds for metastatic carcinoma in our patients were .408, for primary lung were .923, and for benign conditions were .149. One of our patients underwent an unnecessary lobectomy for metastatic breast cancer. Seven other patients with primary lung cancer were misidentified at the time of frozen section and required a second thoracotomy.

 

Conclusion(s): In most institutions the odds of a metastatic carcinoma are similar to those of a lung cancer in patients undergoing frozen section for a single lung nodule. At our institution the odds ratio is 2.3 primary lung/metastatic carcinoma. Having an awareness of the pre-test odds may be useful to thoracic surgeons and pathologists at the time of frozen section at a particular institution.

 

 

Plasmablastic lymphoma with c-MYC rearrangement in HIV-positive patients

 

Hung S. Luu, MD (luuh@cshs.org); Ling Zhang, MD; Randa Alsabeh, MD. 

Department of Pathology, Cedars-Sinai Medical Center, Los Angeles, CA.

 

ABSTRACT:

 

Plasmablastic lymphoma (PBL) is an uncommon, aggressive lymphoma characterized morphologically by the plasmatoid and blastoid appearance of cells and immunophenotypically by loss of pan B markers and positive CD138 reactivity.  PBL is considered a variant of diffuse large B-cell lymphoma.  Translocation associated with c-MYC gene(8q24) is a hallmark of Burkittís lymphoma and is also associated with 10-15% of diffuse large B-cell lymphoma as well as multiple myeloma and other solid tumors. C-MYC positivity may portend more advanced stage or aggressive clinical behavior and may have value as a prognostic indicator.  However, up to date, the c-MYC rearrangement in PBL has not been reported.  We report 2 cases of PBL ages 39 and 42 diagnosed in 2006.  An immunohistochemical panel including CD20, CD30, CD79a (Dako, Carpenteria, CA), CD4, CD138, Bcl-2, Bcl-6, Ki67, p53, EBER (Ventana, Tucson, Ariz), BOB-1, OCT-2 (Santa Cruz Biotechnology, Santa Cruz, CA), CD3, PAX5 (Biocare Medical, Concord, CA), and HHV8 (Advanced Biotech, Columbia, MD) was performed. Both cases showed CD138 immuoreactivity and complete loss of pan B markers including CD20, PAX5 and CD79a by IHC. Aberrant expression of T-cell markers was noted in 1 case with expression of CD4.  Fluorescent in situ hybridization (Vysis, Des Plaines, IL) break-apart for the 8q24 MYC rearrangement was performed and identified in both cases.  Additional clinical and pathologic features are summarized in the table.  Fluorescent in situ hybridization for c-MYC rearrangement should be considered as a routine application in PBL and may be predictive of aggressive clinical behavior.vior.

 

Pathologic features and clinical outcomes in 2-case PBL

Age/

 

Gender/HIV status

Site

Key IHC Stains

c-MYC

 

FISH

Outcomes- Follow up at 4 months

39/M/+

Right Thigh

CD138+

BOB-1+

OCT-2+

P53+

Ki-67 >95%

EBER-

HHV8-

+

Deceased

42/M/+

Rectum

CD138+

BOB-1-

OCT-2+

P53+

Ki-67 >95%

EBER-

HHV8-

+

Undergoing radiation treatment

 

 

*Placental Mesenchymal Dysplasia Is Associated With High Rates of Intrauterine Growth Restriction and Fetal Demise

 

Truc Pham, M.D., Julie Steele, M.D., Carla Stayboldt, M.D., Linda Chan, M.D., and Kurt Bernirschke, M.D.

 

ABSTRACT:

 

Placental mesenchymal dysplasia (PMD) is a rare condition of placentomegaly and abnormal chorionic villi often clinically mistakenly as partial hydatidiform mole.  However, it is clinicopathologically distinct with high incidence of intrauterine growth restriction (IUGR) and fetal death.  This study presents 11 new PMD cases and provides a meta-analysis of the associated IUGR and fetal death rates.  The cases were identified between 1971 and 2005, mostly  from consultation files.  To our knowledge, 71 PMD cases have previously been reported: 15 of these were associated with Beckwith-Wiedemann syndrome (BWS).  With the addition of our new results, among all cases without BWS, 50% had IUGR and 43% had intrauterine fetal demise (IUFD) or neonatal death.  Females represented 82% of cases.  Thus, PMD is associated with high IUGR and IUFD/neonatal death rates an disproportionally affects females.  The cause and pathogenesis are yet unknown.  The current understanding and hypotheses involving PMD are discussed.

 

For complete published discussion see Am. J. Clinical Pathology 2006

 
 
 

© 2006-2017 Los Angeles Society of Pathologists, Inc.