Cytologically targeted next-generation sequencing: a synergy for diagnosing urothelial carcinoma.

INTRODUCTION: Cytology and cystoscopy are used to detect urothelial carcinoma (UC), but together they still fail to detect some UC cases and are not suitable for screening asymptomatic individuals. Mutations are present in more than 98% of UC, mutations have therapeutic significance, and they can be detected by next generation sequencing (NGS) in urine samples. We review the role of NGS in UC detection. MATERIALS AND METHODS: Comprehensive literature review on UC genetics, economics of NGS, and previous reports of UC detection by NGS. RESULTS: The raw costs of NGS have decreased to about 14,000 base pairs per penny, making it appear economically feasible to use NGS widely. Reported NGS assays fall short of predicted sensitivity. Decreased sensitivity is attributed to a low frequency of mutant alleles in many urine samples. Attempts to increase the percentage of mutant alleles, by using cell-free urinary DNA, or by using cell sorting and microfluidics, have been unsuccessful or remain unproven. However, cytologic examination can immediately enable NGS: Urine cytologies with sufficient proportions of abnormal cells could be directly triaged to NGS with high sensitivity for UC detection. For cases with a low proportion of abnormal cells, cytologically targeted microdissection of cells for NGS should maintain sensitivity and decrease sequencing costs. Cytologically targeted urothelial cells for NGS could allow a screening test for low grade UC. CONCLUSIONS: Cytology is immediately poised to allow NGS to improve the diagnosis of UC, allowing NGS to be an ancillary test for atypical cytologies, and potentially allowing a screening test for low-grade UC.

Dedifferentiated Carcinoma of the Endometrium Associated With Low-grade Müllerian Adenosarcoma: A Clinicopathologic Case Report Including the Immunohistochemical and Molecular Profile.

Dedifferentiated carcinoma is defined as undifferentiated carcinoma coexisting with a second component of FIGO grade 1 or 2 endometrioid carcinoma. It is a rare entity with highly aggressive behavior. Dedifferentiated carcinoma combined with another primary uterine tumor is even rarer. We describe a case containing 3 different morphologies comprised of a dedifferentiated carcinoma associated with a low-grade endometrioid carcinoma coexisting with a low-grade Müllerian adenosarcoma. We also used targeted genomic analysis to show all 3 components arise from the same founding clone and identify novel mutations that drive tumor progression.

Somatic molecular analysis augments cytologic evaluation of pancreatic cyst fluids as a diagnostic tool.

Objective: Better tools are needed for early diagnosis and classification of pancreatic cystic lesions (PCL) to trigger intervention before neoplastic precursor lesions progress to adenocarcinoma. We evaluated the capacity of molecular analysis to improve the accuracy of cytologic diagnosis for PCL with an emphasis on non-diagnostic/negative specimens. Design: In a span of 7 years, at a tertiary care hospital, 318 PCL endoscopic ultrasound-guided fine needle aspirations (EUS-FNA) were evaluated by cytologic examination and molecular analysis. Mucinous PCL were identified based on a clinical algorithm and 46 surgical resections were used to verify this approach. The mutation allele frequency (MAF) of commonly altered genes (BRAF, CDKN2A, CTNNB1, GNAS, RAS, PIK3CA, PTEN, SMAD4, TP53 and VHL) was evaluated for their ability to identify and grade mucinous PCL. Results: Cytology showed a diagnostic sensitivity of 43.5% for mucinous PCL due in part to the impact of non-diagnostic (28.8%) and negative (50.5%) specimens. Incorporating an algorithmic approach or molecular analysis markedly increased the accuracy of cytologic evaluation. Detection of mucinous PCL by molecular analysis was 93.3% based on the detection of KRAS and/or GNAS gene mutations (p = 0.0001). Additional genes provided a marginal improvement in sensitivity but were associated with cyst type (e.g. VHL) and grade (e.g. SMAD4). In the surgical cohort, molecular analysis and the proposed algorithm showed comparable sensitivity (88.9% vs. 100%). Conclusions: Incorporating somatic molecular analysis in the cytologic evaluation of EUS-FNA increases diagnostic accuracy for detection, classification and grading of PCL. This approach has the potential to improve patient management.

Heterogeneous mutational profile and prognosis conferred by TP53 mutations in appendiceal mucinous neoplasms.

The eighth edition of American Joint Committee on Cancer (AJCC) advocates a 3-tier grading system for appendiceal mucinous tumors. The mutational profile for each tumor grade and the impact of TP53 mutation on survival are unknown. We classified appendiceal mucinous tumors into 3 grades based on the eighth edition of American Joint Committee on Cancer: 21 G1 low-grade mucinous neoplasms, 21 G2 appendiceal adenocarcinomas, and 26 G3 signet ring cell carcinomas. Mutation profiles were obtained using next-generation sequencing. The impact of TP53 on prognosis was investigated by multivariable analysis. Most G1 tumors harbor KRAS/GNAS mutations with TP53 and SMAD4 in a small subset of cases. G2 and G3 tumors show a more complex mutation pattern carrying PIK3CA, BRAF, or TP53 mutations in addition to KRAS/GNAS. PTEN mutations were detected exclusively in G2 tumors. The prevalence of KRAS and GNAS mutations is significantly lower in G3 tumors relative to G1/G2, whereas TP53, PIK3CA, or BRAF mutations are common. Mutations in NRAS, IDH2, CDH1, RB1, CTNNB1, CDKN2A, PTPN11, and KIT genes were observed in single cases. Patients with TP53-mutated disseminated G2 and G3 tumors had worse progression-free survival than did those with wild-type TP53 tumors (P = .0315). A trend toward worse overall survival was observed in TP53-mutated G3 tumors (P = .102). p53 expression correlated with mutation status. We demonstrate a distinct but overlapping pattern of gene mutations in each grade of appendiceal mucinous tumors and the independent impact of TP53 mutation on progression-free survival but not overall survival.

Immunohistochemical loss of 5-hydroxymethylcytosine expression in acute myeloid leukaemia: relationship to somatic gene mutations affecting epigenetic pathways.

AIMS: Genes affecting epigenetic pathways are frequently mutated in myeloid malignancies, including acute myeloid leukaemia (AML). The genes encoding TET2, IDH1 and IDH2 are among the most commonly mutated genes, and cause defective conversion of 5-methylcytosine into 5-hydroxymethylcytosine (5hmC), impairing demethylation of DNA, and presumably serving as driver mutations in leukaemogenesis. The aim of this study was to correlate 5hmC immunohistochemical loss with the mutation status of genes involved in epigenetic pathways in AML. METHODS AND RESULTS: Immunohistochemical staining with an anti-5hmC antibody was performed on 41 decalcified, formalin-fixed paraffin-embedded (FFPE) bone marrow biopsies from patients with AML. Archived DNA was subjected to next-generation sequencing for analysis of a panel of genes, including TET2, IDH1, IDH2, WT1 and DNMT3A. TET2, IDH1, IDH2, WT1 and DNMT3A mutations were found in 46% (19/41) of the cases. Ten of 15 cases (67%) with TET2, IDH1, IDH2 or WT1 mutations showed deficient 5hmC staining, whereas nine of 26 cases (35%) without a mutation in these genes showed loss of 5hmC. It is of note that all four cases with TET2 mutations showed deficient 5hmC staining. CONCLUSIONS: Overall, somatic mutations in TET2, IDH1, IDH2, WT1 and DNMT3A were common in our cohort of AML cases. Immunohistochemical staining for 5hmC was lost in the majority of cases harbouring mutations in these genes, reflecting the proposed relationship between dysfunctional epigenetic pathways and leukaemogenesis.

Role of Tetrasomy for the Diagnosis of Urothelial Carcinoma Using UroVysion Fluorescent In Situ Hybridization.

CONTEXT: -UroVysion fluorescent in situ hybridization (FISH) is routinely used to detect urothelial carcinoma (UC). A positive threshold is defined as chromosome polysomy in 4 or more cells, which also includes tetrasomy, a natural product of cell division. OBJECTIVES: -To evaluate tetrasomy for UC detection and explore the relation to the surgical diagnosis or patient history. DESIGN: -The FISH was performed on 1532 urine samples from patients with cytology results and 4 or more years of follow-up. We created separate polysomy and tetrasomy categories and constructed receiver operating curves to determine appropriate thresholds using biopsy (n = 194) as the gold standard. Standard FISH and a novel assay integrating cytomorphology and FISH (Target-FISH) were compared. Matching tissue biopsies of urine samples with 10 or more tetrasomy cells were analyzed. RESULTS: -No significant threshold was found for tetrasomy cells. Exclusion of tetrasomy from the polysomy category changed the threshold from 8.5 to 4.5 cells, increased specificity (59.2% to 78.9%), but reduced sensitivity (78.9% to 65.9%). In Target-FISH, the same approach yielded a specificity of 93.7% and sensitivity of 65.2%. Similarly, specificity improved significantly for low- and high-grade UC, but sensitivity decreased for low-grade UC. No evidence of UC was observed in 95% (52 of 55) of the patients referred for screening who had 10 or more tetrasomy cells by FISH. Matching biopsies for urines containing 10 or more tetrasomy cells showed few or no tetrasomy cells. CONCLUSIONS: -Tetrasomy is a nonspecific finding frequently encountered in urine FISH and should be excluded from the polysomy classification. Target-FISH is an optimal approach, offering the ability to detect rare tetrasomy tumors.

Cytology specimens offer an effective alternative to formalin-fixed tissue as demonstrated by novel automated detection for ALK break-apart FISH testing and immunohistochemistry in lung adenocarcinoma.

BACKGROUND: Minimally invasive sampling by cytology or core needle biopsy often provides an initial diagnosis for treatment in patients with lung nodules. From these limited specimens, multiple molecular studies are frequently requested. Current guidelines from the US Food and Drug Administration recommend using formalin-fixed paraffin-embedded tissue sections for the detection of anaplastic lymphoma kinase (ALK) gene rearrangement by fluorescence in situ hybridization (FISH). The authors compared alcohol-fixed and formalin-fixed cytology specimens using a novel automated detection for ALK rearrangements by FISH and immunohistochemistry (IHC). METHODS: ALK FISH testing was performed on 129 lung adenocarcinomas from 71 cytology cases and 58 biopsy/resection specimens using Papanicolaou staining with integrated cytomorphology. IHC with the ALK D5F3 antibody was performed on cases with residual material (88 of 129 cases). RESULTS: The mean age of the patients was 66 years; there were 62 women and 67 men. ALK gene rearrangement was present in 4% of cytology specimens (3 of 71 specimens) and 7% of surgical specimens (4 of 58 specimens). FISH in 13 cases was technically unsuccessful. Of the 7 FISH-positive cases, only 2 cytology cases (4%) and 2 surgical cases (6%) were found to be positive with the ALK antibody, demonstrating 80% concordance. The one case found to be negative for ALK by IHC demonstrated a variant rearrangement of the ALK 2p23 gene locus by FISH. CONCLUSIONS: The results of the current study validate the usefulness of alcohol-fixed and/or formalin-fixed cytology specimens for ALK rearrangement by a novel automated FISH method. IHC using the D5F3 antibody for ALK is specific in this limited cohort. The authors also demonstrated that alcohol-fixed cytology specimens can be used for ALK rearrangement by automated FISH, alone or in conjunction with IHC.

Urine Cytopathology and Ancillary Methods.

Urothelial carcinoma (UC) is the most common malignancy of the urinary tract. Cytology and cystoscopy are two of the most commonly used tests for screening and diagnosis of UC. However, the sensitivity of cytology for UC is less than ideal, while cystoscopy is an invasive and expensive procedure. The search for an accurate, sensitive, noninvasive, and cost-effective method for detecting UC has led to the development of ancillary studies using immunological and molecular methods.

Thymosin beta-NB is the human isoform of rat thymosin beta15.

Thymosin beta15 is a small actin-binding protein upregulated in highly metastatic rat prostate cancer cells, relative to low metastatic cells. We have previously established an important role for thymosin beta15 as a diagnostic marker in human prostate cancer, with potential as a prognostic indicator. We here review the data supporting increased thymosin beta15 expression in other cancer types, including breast, brain, and lung. Human NB thymosin beta is a beta-thymosin originally found in neuroblastoma. New data demonstrate that NB thymosin beta represents the human homolog of rat thymosin beta15; thus we suggest classification as human thymosin beta15. In addition to the previously described gene, thymosin beta15a, we report the discovery of a new isoform of human thymosin beta15, thymosin beta15b, which is transcribed from an independent gene on human chromosome X. The gene structure of thymosin beta15a and beta15b is conserved and the isoforms show 87% identity across the nucleotide sequence. Across the coding sequence the nucleotide differences are silent, resulting in identical proteins. Other thymosin family members have recently been shown to exert potent clinical effects. The functional data available for thymosin beta15, combined with the tumor expression pattern, suggest that thymosin beta15 may play an important role in tumor development and progression in addition to its value as a biomarker in prostate cancer.

Collagen XXIII expression is associated with prostate cancer recurrence and distant metastases.

PURPOSE: We had previously identified a new transmembrane collagen, type XXIII, in metastatic rat prostate carcinoma cells. The purpose of this study was to determine the expression of collagen XXIII in human prostate cancer and investigate its relationship with disease progression. EXPERIMENTAL DESIGN: We investigated collagen XXIII expression in prostate cancer tissue and did a retrospective analysis of association with prostate-specific antigen (PSA)-defined disease recurrence. The presence of collagen XXIII in prostate cancer patient urine was also assessed before and after prostatectomy. RESULTS: Collagen XXIII protein was detected at very low levels in benign prostate tissue and was significantly increased in prostate cancer. Distant metastases exhibited significantly higher collagen XXIII levels compared with either localized prostate cancer or regional (lymph node) metastases. Patients with high collagen XXIII levels had a 2.8-fold higher risk of PSA failure with median time to failure of 8.1 months, compared with low collagen XXIII patients with a median time to failure of 5 years. Multivariate Cox regression showed that the presence of collagen XXIII was significantly associated with time to PSA recurrence, independent of other clinical variables. Collagen XXIII was also detected in prostate cancer patient urine, with reduced levels after prostatectomy, indicating potential as a noninvasive fluid biomarker. CONCLUSIONS: We present the first report demonstrating increased collagen XXIII expression in prostate cancer tissue. We show that collagen XXIII level is a significant independent predictor of PSA-defined disease recurrence, suggesting a potential role as a molecular biomarker of prostate cancer progression and metastasis.

Development of a sensitive and specific enzyme-linked immunosorbent assay for thymosin beta15, a urinary biomarker of human prostate cancer.

OBJECTIVES: In tissue-based assays, thymosin beta15 (Tbeta15) has been shown to correlate with prostate cancer (CaP) malignancy and with future recurrence. To be clinically effective, it must be shown that Tbeta15 is released by the tumor into body fluids in detectable concentrations. Toward this end, we have worked to develop a quantitative high-throughput assay that can accurately measure clinically relevant concentrations of Tbeta15 in human urine. DESIGN AND METHODS: Sixteen antibodies were raised against recombinant Tbeta15 and/or peptide conjugates. One antibody, having stable characteristics over the wide range of pH and salt concentrations found in urine and minimal cross-reactivity with other beta thymosins, was used to develop a competitive enzyme-linked immunosorbent assay (ELISA). Urinary Tbeta15 concentration was determined for control groups; normal (N = 52), prostate intraepithelial neoplasia (PIN, N = 36), and CaP patients; untreated (N = 7) with subsequent biochemical failure, radiation therapy (N = 17) at risk of biochemical recurrence. RESULTS: The operating range of the competition ELISA fell between 2.5 and 625 ng/mL. Recoveries exceeded 75%, and the intra- and inter-assay coefficients of variability were 3.3% and 12.9%, respectively. No cross-reactivity with other urine proteins was observed. A stable Tbeta15 signal was recovered from urine specimens stored at -20 degrees C for up to 1 year. At a threshold of 40 (ng/dL)/mug protein/mg creatinine), the assay had a sensitivity of 58% and a specificity of 94%. Relative to the control groups, Tbeta15 levels were greater than this threshold in a significant fraction of the CaP patients (P < 0.001), including 5 of the 7 patients who later experienced PSA recurrence. CONCLUSIONS: We have established an ELISA that is able to detect Tbeta15 at clinically relevant concentrations in urine from patients with CaP. The assay will provide a tool for future clinical trials to validate urinary Tbeta15 as a predictive marker for recurrent CaP.

Use of thymosin beta15 as a urinary biomarker in human prostate cancer.

BACKGROUND: Additional prostate cancer (CaP) biomarkers are needed to increase the accuracy of diagnosis and to identify patients at risk of recurrence. In tissue-based assays, thymosin beta15 (Tbeta15) has been linked to an aggressive CaP phenotype and correlated with future tumor recurrence. We hypothesized that Tbeta15 may have clinical utility in biological fluids. METHODS: Tbeta15 was measured in urine from CaP patients; untreated (N = 61), prostatectomy (RP, N = 46), androgen deprivation therapy (ADT, N = 14) and control groups; normal (N = 52), genitourinary carcinoma (N = 15), non-malignant prostate disease (N = 81), and other urology (N = 73). We evaluated the utility of urinary Tbeta15 for CaP diagnosis, alone or in combination with prostate-specific antigen (PSA), and the relationship to CaP progression. RESULTS: A normal threshold of 40 (ng/dl)/(mug_protein/mg_creatinine) was defined using receiver operating characteristic analysis and marked the 19th centile for age-matched controls. The proportion of untreated CaP patients with urinary Tbeta15 above the threshold was significantly higher than normal and genitourinary disease controls (P < 0.001). RP caused urinary Tbeta15 to drop significantly (P = 0.005). Pre-surgery Tbeta15 concentrations greater than the normal threshold may confer greater risk of CaP recurrence. Relative to normal controls, patients receiving ADT for aggressive CaP were 12 times more likely to have elevated urinary Tbeta15 (P = 0.001, 95% CI = 2.8, 51.8). Combining PSA and Tbeta15 (PSA > 4, or PSA > 2.5, Tbeta15 > 40, or PSA = 2.5, Tbeta15 > 90) provided the same sensitivity as a 2.5 ng/ml PSA cutoff, but markedly improved diagnostic specificity. CONCLUSIONS: We report that Tbeta15 is a urinary biomarker for CaP and suggest that Tbeta15, in combination with PSA, can be used to improve both the sensitivity and specificity of CaP diagnosis.