Predicting response and survival in chemotherapy-treated triple-negative breast cancer.

BACKGROUND: In this study, we evaluated the ability of gene expression profiles to predict chemotherapy response and survival in triple-negative breast cancer (TNBC). METHODS: Gene expression and clinical-pathological data were evaluated in five independent cohorts, including three randomised clinical trials for a total of 1055 patients with TNBC, basal-like disease (BLBC) or both. Previously defined intrinsic molecular subtype and a proliferation signature were determined and tested. Each signature was tested using multivariable logistic regression models (for pCR (pathological complete response)) and Cox models (for survival). Within TNBC, interactions between each signature and the basal-like subtype (vs other subtypes) for predicting either pCR or survival were investigated. RESULTS: Within TNBC, all intrinsic subtypes were identified but BLBC predominated (55-81%). Significant associations between genomic signatures and response and survival after chemotherapy were only identified within BLBC and not within TNBC as a whole. In particular, high expression of a previously identified proliferation signature, or low expression of the luminal A signature, was found independently associated with pCR and improved survival following chemotherapy across different cohorts. Significant interaction tests were only obtained between each signature and the BLBC subtype for prediction of chemotherapy response or survival. CONCLUSIONS: The proliferation signature predicts response and improved survival after chemotherapy, but only within BLBC. This highlights the clinical implications of TNBC heterogeneity, and suggests that future clinical trials focused on this phenotypic subtype should consider stratifying patients as having BLBC or not.

Epstein-Barr virus microRNAs and lung cancer.

BACKGROUND: We conducted the first analysis of viral microRNAs (miRNAs) in lung cancer, with a focus on Epstein-Barr virus (EBV). METHODS: We evaluated viral miRs with a two-channel oligo-array targeting mature, anti-sense miRNAs in 290 cases. In 48 cases, we compared microarray and real-time quantitative PCR (qPCR) expression for three EBV miRNAs. We tested for EBV DNA, RNA, and protein in tumour tissue from six cases with and six cases without strong qPCR-based evidence of EBV miRNAs. RESULTS: The EBV miRNAs strongly differentiated between adenocarcinoma and squamous cell carcinoma using the microarray (P<0.01 for 9 out of 16 EBV miRNAs). However, microarray and qPCR measurements of BART1, BART2, and BHRF1-3 expression were not significantly correlated (P=0.53, 0.94, and 0.47, respectively). Although qPCR provided substantial evidence of EBV miRNAs in 7 out of 48 cases, only 1 of these 7 cases had detectable EBV DNA in tumour tissue. None had detectable EBV RNA or protein by histochemical stains. CONCLUSION: In a comprehensive evaluation of EBV miRNA, DNA, RNA, and protein in lung cancer, we found little evidence of EBV in lung tumour tissue. Discrepancies between microarray- and qPCR-based strategies highlight the difficulty of validating molecular markers of disease. Our results do not support a role of EBV in lung cancer.

Epstein-Barr virus in gastric adenocarcinomas: association with ethnicity and CDKN2A promoter methylation.

AIMS: It has been shown previously (by immunohistochemistry) that gastric adenocarcinomas harbouring Epstein-Barr virus (EBV) frequently lose p16 protein. This study aimed to examine the mechanisms of inactivation of the CDKN2A gene and correlate the results with clinicopathological features. METHODS: Methylation specific polymerase chain reaction was used to detect CDKN2A promoter methylation in gastric adenocarcinomas from American patients. In addition, immunohistochemistry was used to detect the loss of the p16 protein and in situ hybridisation was used to detect the presence of EBV. The tumours were also analysed for the presence of microsatellite instability. RESULTS: Eleven (10%) of 107 tumours harboured EBV in the malignant cells. In gastric cancers without EBV, 32% exhibited CDKN2A promoter methylation and 26% had p16 protein loss. In contrast, 91% of the tumours containing EBV had CDKN2A promoter methylation (p = 0.0003) and 90% showed p16 protein loss (p = 0.0001). The presence of EBV was also associated with male sex (p = 0.03) and was more common in tumours from Texas Hispanics than from non-Hispanic whites or African-Americans (p = 0.01). EBV was not associated with microsatellite instability, histological subtype, stage, or grade of the tumour, or age or survival time of the patient. CONCLUSIONS: The presence of EBV in gastric adenocarcinomas is strongly associated with CDKN2A inactivation by promoter methylation. In addition, these findings suggest that there are ethnic differences in tumour virology and pathogenesis.

Comparative analysis of the expression of the Epstein-Barr virus (EBV) anti-apoptotic gene BHRF1 in nasopharyngeal carcinoma and EBV-related lymphoid diseases.

Epstein-Barr virus (EBV) has been identified in a wide range of neoplastic and non-neoplastic disorders. The EBV open reading frame BHRF1 encodes a protein with partial sequence and functional homology to the anti-apoptotic onco-protein Bcl-2 and may therefore have a role in the proliferation of EBV positive cells. We have developed a rat monoclonal antibody against pBHRF1, which can detect BHRF1 in paraffin sections. While a number of mutant versions of BHRF1 were recognised, the monoclonal did not detect the BHRF1 homologue encoded by Herpesvirus papio or two mutants with deletions in the BH2 region. This novel rat monoclonal antibody (6A9) was used to examine tissue sections from 39 cases of non-keratinising undifferentiated nasopharyngeal carcinoma (NPC), 6 cases of metastatic NPC, 7 cases of EBV-positive NPC with squamous differentiation from Chinese patients, 15 cases of EBV-positive post-transplant lymphoproliferative disorder (PTLD), 6 EBV-containing lymphoblastoid cell lines, and 2 cases of oral hairy leukoplakia (OHL). In 11 cases of undifferentiated NPC, RT-PCR data were available for comparison with the immunohistochemistry. Both cases of OHL and two cases of LCL were positive for BHRF1 but none of the PTLD showed positive staining. All cases of undifferentiated NPC were positive for Bcl-2 but only one BHRF1 positive cell was identified in 1 of 39 cases of primary undifferentiated NPC. The 6A9 antibody produced less background staining and no nuclear positivity compared with the commercially available mouse monoclonal 5B11. It is concluded that BHRF1 can not be detected by immunohistochemistry in NPC and therefore it appears not to play a significant anti-apoptotic role in the progression of this EBV-associated tumour. The 6A9 monoclonal appears to be superior to 5B11 for the detection of pBHRF1 in tissue sections.

The Burkitt-like lymphomas: a Southwest Oncology Group study delineating phenotypic, genotypic, and clinical features.

The Revised European-American Lymphoma classification gives Burkitt-like lymphoma (BLL) provisional status, leaving unresolved the differential diagnosis with Burkitt lymphoma (BL) and diffuse large B-cell lymphoma (DLBCL). This study compared the biologic features of adult BLL and DLBCL. The phenotypic distinction between BLL and DLBCL was determined by immunohistochemical staining of frozen tissue from 13 patients with BLL and 55 patients with DLBCL by using an extensive antibody panel including Ki-67, CD10, CD11a/lymphocyte function-associated antigen 1alpha (LFA-1alpha), CD18/LFA-1beta, CD58/LFA-3, and CD54/intercellular adhesion molecule, CD8 for tumor-infiltrating cytotoxic T cells (T-TILs), CD44 homing receptor, and p53 and Bcl-2 oncogenic proteins. Compared with DLBCL, BLL had a higher proliferative rate (mean Ki-67, 88% versus 53%), greater expression of CD10 and p53 antigens, and decreased expression of Bcl-2. BLL cases had a consistent absence of one or more cell adhesion molecules (92% versus 27%), low T-TIL numbers, and absence of CD44 homing receptor (92% versus 14%). The t(8;14) translocation was identified in 80% of BLL cases, but no patients with BLL had the t(14;18) translocation. In a 10-year analysis, median survival of patients with BLL was 1.2 years, and that of patients with DLBCL was 2.5 years. Although the proportion of patients cured was similar in the 2 groups, BLL patients had an increased risk of early death. We conclude that BLL can be recognized by its combined morphologic and phenotypic features and that it represents a high-grade lymphoma much closer to BL than DLBCL. Retention of the BLL category or inclusion of BLL as a variant of BL is biologically and clinically more appropriate than absorbing the category of BLL into DLBCL. (Blood. 2001;97:3713-3720)

Absence of Epstein-Barr virus in smooth muscle cells of idiopathic hypertrophic pyloric stenosis.

CONTEXT: The etiology of idiopathic hypertrophic pyloric stenosis (IHPS) is unknown. Epstein-Barr virus (EBV) infects smooth muscle cells and is associated with leiomyomas and leiomyosarcomas of immunocompromised persons, including persons with the acquired immunodeficiency syndrome. OBJECTIVE: To determine whether EBV is causally associated with IHPS. DESIGN: Biopsy samples of the pylorus were obtained from 10 infants with projectile vomiting and pyloric hypertrophy on ultrasound, with confirmation of hypertrophy at the time of pyloromyotomy. The presence of EBV infection was tested by in situ hybridization for EBV-encoded RNA 1 (EBER1) in smooth muscle cells of IHPS. SETTING: Biopsy specimens were obtained from children treated for IHPS at a tertiary referral hospital and were tested in a clinical molecular diagnostics laboratory. RESULTS: All of the 10 smooth muscle biopsies were negative for EBER1. Cellular U6 RNA was detected in all smooth muscle samples, confirming that the RNA in the specimens was intact and capable of detection by in situ hybridization. CONCLUSIONS: The absence of EBER1 in 10 cases of clinically diagnosed and histopathologically confirmed cases of IHPS effectively excludes EBV infection of smooth muscle cells as a causal factor in the pathogenesis of IHPS.

Latent intracellular Epstein-Barr Virus DNA demonstrated in ocular posttransplant lymphoproliferative disorder mimicking granulomatous uveitis with iris nodules in a child.

A case of intraocular posttransplant lymphoproliferative disorder (PTLD) is described in a 9-year-old female who underwent orthotopic liver transplantation at the age of 18 months. The prevalence of ophthalmic involvement in PTLD can be expected to rise with the increasing number of major organ transplantations, as well as improved survivorship. Children are particularly at risk for this posttransplant complication because they are usually seronegative for the Epstein-Barr virus (EBV) prior to transplant. Accurate diagnostic classification of PTLD to include confirmation of EBV infection carries significant therapeutic and prognostic implications.

Hepatosplenic alphabeta T-cell lymphomas: a report of 14 cases and comparison with hepatosplenic gammadelta T-cell lymphomas.

Hepatosplenic gammadelta T-cell lymphoma is a distinct entity, characterized by occurrence in young adult males with hepatosplenomegaly, B-symptoms, peripheral blood cytopenias, and no lymphadenopathy; lymphomatous infiltrates in the splenic red pulp, hepatic sinusoids, and bone marrow sinuses; T-cell receptor (TCR) gammadelta chains and a cytotoxic T-cell phenotype; isochromosome 7q; and an aggressive clinical course. In comparison, this study describes the clinicopathologic features of 14 hepatosplenic T-cell lymphomas expressing TCR alphabeta chains. They occurred in 11 women and 3 men with a median age of 36 years. Clinical presentation was similar to that described previously for hepatosplenic gammadelta T-cell lymphomas, except for the female preponderance and age distribution (5 patients younger than 13 years of age and 5 patients older than 50 years of age). Disease distribution was primarily in the splenic red pulp and hepatic sinusoids, although liver infiltrates were largely periportal in four cases. Bone marrow involvement, observed in eight patients, was usually interstitial and/or within the sinuses. Lymph nodes were involved in five patients, although lymphadenopathy was demonstrable in only two. Ten cases were composed of intermediate-size tumor cells with round/oval nuclei, slightly dispersed chromatin, inconspicuous nucleoli, and scant to moderate amounts of cytoplasm. Four lymphomas contained primarily large cells with irregular nuclei, dispersed chromatin, discernible nucleoli, and moderate to abundant cytoplasm. Tumor cells in all 14 lymphomas were cytotoxic alphabeta T-cells; 13 co-expressed natural killer cell-associated antigens and showed T-cell clonality. Three lymphomas were associated with Epstein-Barr virus. Two of four cases had an isochromosome 7q. Eleven patients are dead, eight within a year of diagnosis, and two patients have maintained complete remissions after combination chemotherapy. These data show that hepatosplenic T-cell lymphomas include an alphabeta-subtype. This group, along with the previously recognized gammadelta group, should be recognized as phenotypically heterogeneous subtypes of the same disease entity.

Molecular diagnosis of Epstein-Barr virus-related diseases.

Epstein-Barr virus (EBV) is the causative agent of infectious mononucleosis, and it may also be found in a wide variety of benign and malignant lesions including oral hairy leukoplakia, inflammatory pseudotumor, Hodgkin's disease, non-Hodgkin's lymphoma, nasopharyngeal carcinoma, and gastric carcinoma. Molecular testing is increasingly important in the diagnosis and monitoring of patients affected by these diseases. In biopsy tissues, molecular detection of EBV-encoded RNA transcripts by in situ hybridization remains the gold standard for proving that a histopathological lesion is EBV-related. EBV-encoded RNA hybridization and EBV LMP1 immunostains are used routinely to detect latent EBV in tissues affected by posttransplant lymphoproliferative disorder (PTLD) or in enlarged nodes from patients with infectious mononucleosis. Traditional serology is the best test for evaluating acute versus remote infection in healthy individuals. High serological titers serve as a tumor marker for some EBV-related malignancies, but titers are not a dependable tumor marker in immunocompromised hosts. EBV viral load testing by quantitative DNA amplification of blood samples is a promising new laboratory test that has proven useful for early diagnosis and monitoring patients with PTLD. Recent studies suggest a role for EBV viral load testing in nasopharyngeal carcinoma, Hodgkin's disease, and AIDS patients with brain lymphoma. Further research is needed to define more fully the clinical utility of viral load tests in the full spectrum of EBV-associated diseases. Gene expression profiling is on the horizon as a means to improve subclassification of EBV-related diseases and to predict response to therapy.

Analytic validation of a competitive polymerase chain reaction assay for measuring Epstein-Barr viral load.

Epstein-Barr virus (EBV) is associated with several benign and malignant diseases, and blood tests for EBV viral load show promise as markers of disease burden in affected patients. A commercial quantitative PCR method (BioSource International) was recently introduced to facilitate measuring viral load. It relies on coamplification of EBV DNA and a spiked competitor in plasma or serum, followed by semiautomated product detection on enzyme-linked immunosorbent assay (ELISA) plates. In the current study, analytic performance characteristics were assessed, and the authors describe several methodologic improvements to facilitate laboratory implementation. Rapid DNA extraction was accomplished using commercial silica spin columns, heat-labile uracil-N-glycosylase was used to inhibit amplicon contamination, and inexpensive agarose gels were used to screen for polymerase chain reaction products requiring ELISA plate quantitation. Accuracy and precision were verified using EBV DNA standards derived from two cell lines and plasmid containing viral sequences. The assay was sensitive to as few as five template copies per polymerase chain reaction and was linear across four orders of magnitude (correlation coefficient 0.995). When applied to matched plasma and serum samples from 15 patients with nasopharyngeal carcinoma, both sample types yielded similar viral load results. This commercial EBV viral load assay provides sensitive and quantitative detection of EBV DNA using equipment already available in many molecular diagnostic laboratories.

Epstein-Barr viral load measurement as a marker of EBV-related disease.

Epstein-Barr virus (EBV) is associated with a wide variety of benign and neoplastic diseases. EBV viral load assays that may prove useful in rapid assessment of disease status are now available. The two most common approaches to viral load measurement are quantitative, competitive PCR, and real-time PCR. Laboratory studies have shown that these assays are sensitive and specific for measuring EBV DNA in blood samples. Clinical investigations suggest a role for viral load measurement in predicting and monitoring EBV-associated tumors, including nasopharyngeal carcinoma, post-transplant lymphoproliferative disorder, Hodgkin's disease, and AIDS-related lymphoma. These new laboratory tools show promise in improving clinical management of affected patients.

DNA extraction from paraffin-embedded tissues.

In routine histopathology, most tissues are fixed in formalin and embedded in paraffin for long-term preservation. DNA can be extracted from these tissues for subsequent molecular analysis by amplification methods. We describe herein a protocol for DNA preparation from paraffin-embedded tissues based on published procedures (1-3). In brief, tissue sections are placed into microfuge tubes, then deparaffinized with xylene. The xylene is removed with ethanol washes, and the tissue is treated with proteinase K to make DNA available for amplification.

DNA extraction from fresh or frozen tissues.

The first step in molecular analysis of patient tissues is preparation of purified, high molecular weight DNA. A number of methods and commercial kits are available for DNA isolation. Traditional organic extraction protocols (1,2) are based on the fact that DNA is soluble in water whereas lipids are soluble in phenol. In these protocols, tissues are disaggregated and then treated with detergent to lyse cell membranes followed by proteinase to digest proteins. Phenol, an organic solvent, is added to help separate the lipids and protein remnants from the DNA. Chloroform is then used to facilitate the removal of phenol. DNA is subsequently concentrated and further purified by precipitation in a cold mixture of salt and ethanol. Finally, DNA is resolubilized in Tris-EDTA buffer.

RNA extraction from fresh or frozen tissues.

RNA can be isolated from fresh or frozen tissue, then purified and quantified for subsequent molecular analysis. RNA is quite labile compared to DNA for good reason: RNA is the transient message that transmits information from activated genes. Once gene transcription is turned off, no more RNA is produced, and preexisting RNA is rapidly degraded to prevent continued translation of proteins. Because RNA is so labile, special care is required at all steps of RNA extraction to prevent RNA degradation.

Molecular Methods for Detecting Epstein-Barr Virus (Part I) : In Situ Hybridization to Epstein-Barr Virus-Encoded RNA (EBER) Transcripts.

In situ hybridization (ISH) to Epstein-Barr virus (EBV)-encoded RNA (EBER) is considered the "gold standard" for detecting and localizing latent EBV in biopsy samples. Transcripts from the EBER1 and EBER2 genes are an appropriate target because they are the most abundant viral RNA in latently infected cells, exceeding 1 million transcripts per cell. Furthermore, EBERs are consistently expressed in virtually all EB V-infected tumors and in the lymphoid tissues of infectious mononucleosis (1-5). As a result, EBERs represent a naturally amplified target for detecting and localizing latent EBV in histologic samples. The only EBV-related disorder in which EBER is consistently absent is oral hairy leukoplakia, an infection of squamous epithelial cells in which the virus undergoes lytic viral replication rather than latent infection (6).

Molecular Methods for Detecting Epstein-Barr Virus (Part II) : Structural Analysis of Epstein-Barr Virus DNA as a Marker of Clonality.

The Southern blot technique can be used to determine the clonality of Epstein-Barr virus (EBV) infected cells (1,2). This clonality assay capitalizes on measurable polymorphisms in EBV genomic structure, namely, the variable number of tandem repeats lying at either end of the linear viral genome. Thus, the size of genome varies from virion to virion depending on the number of terminal repeat sequences. When a particular virion infects a cell, its linear genome circularizes by fusing the terminal repeats to form an episome containing from 1-20 tandem repeat sequences. If an infected cell undergoes malignant transformation, the viral DNA replicates along with cell DNA during mitosis, and the same terminal repeat structure is inherited by all tumor cell progeny. Therefore, Southern blot analysis of the EBV terminal restriction fragment serves as a marker of cellular clonality.

Molecular Methods for Detecting Epstein-Barr Virus (Part III) : EBV Viral Load by Competitive Polymerase Chain Reaction.

Epstein-Barr virus (EBV) viral load testing is rapidly gaining acceptance in the diagnosis and monitoring of patients with EBV-related neoplasia, including transplant recipients, autoimmune deficiency syndrome (AIDS) patients with brain lymphoma, and patients with nasopharyngeal carcinoma. In transplant recipients, several studies have shown that EBV viral load in blood is a useful marker of EBV-driven posttransplant lymphoproliferative disease (PTLD), both in terms of predicting disease and in monitoring efficacy of therapy (1-4). In AIDS patients, high EBV viral load in cerebrospinal fluid (CSF) is so characteristic of brain lymphoma that the assay has been touted as sufficient for making that diagnosis without the need for brain biopsy, assuming that there is also clinical and radiographic support for the diagnosis (5). In nasopharyngeal carcinoma patients, a recent study showed that plasma EBV viral load is nearly always elevated, and the degree of elevation is higher in those with distant metastases (6). Additional studies are needed to more completely define the utility of EBV viral load measurement in monitoring residual disease following therapy, and to evaluate clinical utility in other EB V-associated diseases.

Accumulation of p53 in infectious mononucleosis tissues.

Epstein-Barr virus (EBV) infects lymphocytes, where it persists indefinitely for the life of the host; whether the virus interacts with p53 to maintain itself in these cells is unknown. Lymphoid biopsy samples from 10 patients with infectious mononucleosis (IM) were examined for expression of p53 by immunohistochemistry. Accumulation of p53 was detected in all 10 cases, primarily in large lymphocytes of the expanded paracortex. The presence of EBV was confirmed in all 10 cases by EBER1 (EBV-encoded RNA) in situ hybridization, whereas 11 non-IM control samples lacked significant EBER1 and did not express p53 in paracortical lymphocytes. Interestingly, EBV infection alone does not cause accumulation of intracellular p53, because many more cells expressed EBER1 than p53 in the IM tissues. To determine whether p53 was confined to the subset of infected cells in which viral replication was occurring, BZLF1 immunostains were performed. Viral BZLF1 was detected in 8 of 10 IM tissues; however, the paucity and small size of the BZLF1-expressing lymphocytes suggests that they are not the same cells overexpressing p53. To further examine the relationship between p53 and EBV gene expression, the tissues were studied for latent membrane protein 1 (LMP1) expression by immunohistochemistry. Viral LMP1 was observed in the large paracortical lymphocytes of all 10 cases of IM, indicating co-localization of p53 and LMP1 in these cells. Our findings confirm that p53 overexpression is not specific for nodal malignancy and that p53 accumulation is characteristic of IM. Because p53 was not coexpressed in the same cells as BZLF1, it appears that BZLF1 is not directly responsible for p53 accumulation. Nevertheless, co-localization of p53 and LMP1 in activated-appearing lymphocytes suggests that EBV infection is responsible for p53 accumulation. HUM PATHOL 31:1397-1403.

Prothrombin gene mutation uncommon in pulmonary embolism.

BACKGROUND: Venous thrombosis followed by pulmonary embolism is one of the most common causes of sudden death among middle-aged adults. Several inherited polymorphisms are associated with heightened risk of venous thrombosis, including mutation at position 20210 of the prothrombin gene and mutation at codon 506 of the factor V gene. METHODS: We studied mutation prevalence in 67 individuals who died of pulmonary embolism and were autopsied in a medical examiner's facility over a 5-year period. Mutations were identified by polymerase chain reaction followed by allele-specific endonuclease digestion. RESULTS: Traditional risk factors for pulmonary embolism (eg, immobility, oral contraceptive use, cancer) were identified in 75%. Heterozygous mutation of the prothrombin gene was found in 3/67 (4%), and heterozygous mutation of the factor V gene was identified in 3/66 (4%). No homozygotes or compound heterozygotes were identified. The prevalence of mutation was not significantly different from that of the general population. CONCLUSIONS: Individuals who die suddenly from pulmonary embolism are not often affected by prothrombin or factor V gene mutations. Therefore, medical examiners need not routinely test for these mutations in individuals who die of pulmonary embolism.

Necrotizing ulcerative stomatitis in human immunodeficiency virus-seropositive individuals: a review of the histopathologic, immunohistochemical, and virologic characteristics of 18 cases.

OBJECTIVE: The purpose of this retrospective study was to delineate the histopathologic, immunohistochemical, and virologic characteristics of 18 cases of necrotizing ulcerative stomatitis. STUDY DESIGN: Eighteen examples or oral ulcerations in human immunodeficiency virus-seropositive individuals were identified that displayed unique histopathologic features. Immunohistochemic staining for CD1a, CD3, CD23, CD68, HLA-DR, p24, cytomegalovirus, HSV-1, and HSV-2 was performed, along with in situ hybridization for Epstein-Barr virus RNA and special staining for bacteria and fungi. RESULTS: The lesions demonstrated ulceration, extensive necrosis, leukocytoclasia, histiocytic vasculitis with luminal fibrin clots, and a prominent infiltrate of large atypical cells with amphophilic cytoplasm, vesicular nuclei, and prominent nucleoli, interspersed with crescentic histiocytes, a histologic picture resembling extranodal Kikuchi's disease. Immunohistochemical findings suggested that the large atypical cells were histiocytes. Fifty-six percent (10/18) of the cases were immunoreactive for human immunodeficiency virus p24 within focal histiocytes, whereas Epstein-Barr virus RNA was identified in 1 (6%) of 17 cases. CONCLUSIONS: Necrotizing ulcerative stomatitis is an inflammatory disease characterized by specific, reproducible microscopic features. We postulate that the histopathologic resemblance of necrotizing ulcerative stomatitis to extranodal Kikuchi's disease reflects a similar immune response to differing pathogens.