Free digital image analysis software helps to resolve equivocal scores in HER2 immunohistochemistry.

Evaluation of human epidermal growth factor receptor 2 (HER2) immunohistochemistry (IHC) is subject to interobserver variation and lack of reproducibility. Digital image analysis (DIA) has been shown to improve the consistency and accuracy of the evaluation and its use is encouraged in current testing guidelines. We studied whether digital image analysis using a free software application (ImmunoMembrane) can assist in interpreting HER2 IHC in equivocal 2+ cases. We also compared digital photomicrographs with whole-slide images (WSI) as material for ImmunoMembrane DIA. We stained 750 surgical resection specimens of invasive breast cancers immunohistochemically for HER2 and analysed staining with ImmunoMembrane. The ImmunoMembrane DIA scores were compared with the originally responsible pathologists' visual scores, a researcher's visual scores and in situ hybridisation (ISH) results. The originally responsible pathologists reported 9.1 % positive 3+ IHC scores, for the researcher this was 8.4 % and for ImmunoMembrane 9.5 %. Equivocal 2+ scores were 34 % for the pathologists, 43.7 % for the researcher and 10.1 % for ImmunoMembrane. Negative 0/1+ scores were 57.6 % for the pathologists, 46.8 % for the researcher and 80.8 % for ImmunoMembrane. There were six false positive cases, which were classified as 3+ by ImmunoMembrane and negative by ISH. Six cases were false negative defined as 0/1+ by IHC and positive by ISH. ImmunoMembrane DIA using digital photomicrographs and WSI showed almost perfect agreement. In conclusion, digital image analysis by ImmunoMembrane can help to resolve a majority of equivocal 2+ cases in HER2 IHC, which reduces the need for ISH testing.

Grading of neuroendocrine tumors with Ki-67 requires high-quality assessment practices.

Gastrointestinal and pancreatic neuroendocrine tumors (NETs) arise from disseminated neuroendocrine cells, expressing general and specific neuroendocrine markers. The World Health Organization 2010 classification of NETs is based on grading them according to the proliferation index (PI), which is determined by immunohistochemical staining of the nuclear antigen Ki-67. The classification introduces Ki-67 as the most important criterion for tumor grading, influencing patients' prognoses and the choice of treatment. The aim of this study was to evaluate the assessment of PI value in NETs and its influence on tumor grading. The tumor material consisted of 51 NETs from the pancreas (n=31) and ileum (n=20). The slides were stained with the Ki-67 antibody and visualized using a polymer kit. PI was assessed visually by microscope oculars and using a public domain image analysis software, ImmunoRatio. The PI was measured from the most proliferative areas of the tumor. The PI values and tumor grade by ImmunoRatio were highly reproducible as compared with conventional assessment, which suffered from variation especially if ascertained by different observers. Computer-aided assessments had almost perfect correlation (r=0.985, r=0.987, and r=0.995) (P=0.000) and reproducibility (κ=0.886, κ=0.886, and κ=1.000) (P=0.000) in PI values and tumor grades, respectively. The PI values and tumor grade between conventional and ImmunoRatio assessments by a qualified observer were in good agreement. ImmunoRatio is a qualified diagnostic aid to more objectively analyze Ki-67 PI-based tumor grade in NETs.

Amplification and overexpression of the ABCC3 (MRP3) gene in primary breast cancer.

The ATP-binding cassette (ABC) of active transporters comprises a group of proteins that which facilitate efflux of anticancer drugs from cancer cells. We focused on the gene amplification and protein expression of ABCC3 (also known as MRP3) in breast cancer cell lines and clinical tumor samples. Fluorescence and chromogenic in situ hybridization, using an ABCC3-specific probe, was used to analyze 11 breast cancer cell lines and 112 clinical tumor samples. The results of ABCC3 were correlated with the amplification status of HER2 and topoisomerase II alpha (TOP2A), which are located close to ABCC3 at 17q12-q21. Immunohistochemistry was used to assess ABCC3 protein overexpression. Of the cell lines studied 6 HER2-positive lines and 1 HER2-negative line exhibited amplification of ABCC3. In the HER-2-negative clinical tumor samples, only 4/55 (7.3%) exhibited ABCC3 amplification. In the HER2-positive tumors, ABCC3 was amplified in 16/57 tumors (28.1%, P=0.0059). TOP2A did not exhibit any consistent coamplification pattern. ABCC3 (MRP3) protein overexpression was more common in tumors with gene amplification (P=0.069). In silico analysis of 804 breast cancers with matched gene expression and copy number microarray data revealed significant differences ABCC3 across the molecular subtypes. Specifically, increased ABCC3 mRNA and gene copy numbers were most prominent in HER2 amplified and/or HER2-enriched classified tumors. Moreover, differential ABCC3 mRNA levels were found within the HER-2 amplified subset when stratified by the estrogen receptor status. We conclude that ABCC3 is frequently amplified and overexpressed in HER2-positive breast cancer, and something that warrants further studies correlating the results with therapeutic outcome.

Chemical castration and anti-androgens induce differential gene expression in prostate cancer.

Endocrine therapy by castration or anti-androgens is the gold standard treatment for advanced prostate cancer. Although it has been used for decades, the molecular consequences of androgen deprivation are incompletely known and biomarkers of its resistance are lacking. In this study, we studied the molecular mechanisms of hormonal therapy by comparing the effect of bicalutamide (anti-androgen), goserelin (GnRH agonist) and no therapy, followed by radical prostatectomy. For this purpose, 28 men were randomly assigned to treatment groups. Freshly frozen specimens were used for gene expression profiling for all known protein-coding genes. An in silico Bayesian modelling tool was used to assess cancer-specific gene expression from heterogeneous tissue specimens. The expression of 128 genes was > two-fold reduced by the treatments. Only 16% of the altered genes were common in both treatment groups. Of the 128 genes, only 24 were directly androgen-regulated genes, according to re-analysis of previous data on gene expression, androgen receptor-binding sites and histone modifications in prostate cancer cell line models. The tumours containing TMPRSS2-ERG fusion showed higher gene expression of genes related to proliferation compared to the fusion-negative tumours in untreated cases. Interestingly, endocrine therapy reduced the expression of one-half of these genes and thus diminished the differences between the fusion-positive and -negative samples. This study reports the significantly different effects of an anti-androgen and a GnRH agonist on gene expression in prostate cancer cells. TMPRSS2-ERG fusion seems to bring many proliferation-related genes under androgen regulation.

ImmunoMembrane: a publicly available web application for digital image analysis of HER2 immunohistochemistry.

AIMS: Assessment of the human epidermal growth factor receptor 2 (HER2) with immunohistochemistry (IHC) is routine practice in clinical pathology laboratories. Visual classification of the staining reaction (usually into 0/1+, 2+ or 3+) is subjective and prone to significant inter- and intra-observer variation. In this study, we describe ImmunoMembrane, an easy-to-use HER2 IHC analysis software, which is freely available as a web application, requiring no download or installation. METHODS AND RESULTS: ImmunoMembrane uses colour deconvolution for stain separation and a customized algorithm for cell membrane segmentation. A quantitative score (IM-score, 0-20 points) is generated according to the membrane staining intensity and completeness. Specimens are classified into 0/1+, 2+ or 3+ based on IM-score cut-offs defined using a training set. The classification and membrane segmentation are presented as a pseudo-coloured overlay image. With a validation set (144 HercepTest(®) -stained whole tissue sections), ImmunoMembrane matched well with the pathologist's visual classification (weighted kappa κ(w) =0.80), as well as fluorescence in-situ hybridization (FISH) (IHC disagreement 3.5%, n=144) and chromogenic in-situ hybridization (CISH) (IHC disagreement 2.8%, n=144). CONCLUSIONS: We anticipate that publicly available web applications, such as ImmunoMembrane, will accelerate the adoption of automated image analysis in clinical diagnostics of HER2 IHC. ImmunoMembrane is freely accessible at: http://jvsmicroscope.uta.fi/immunomembrane/.

Characterization of non-specific cytotoxic cell receptor protein 1: a new member of the lectin-type subfamily of F-box proteins.

Our previous microarray study showed that the non-specific cytotoxic cell receptor protein 1 (Nccrp1) transcript is significantly upregulated in the gastric mucosa of carbonic anhydrase IX (CA IX)-deficient (Car9(-/-)) mice. In this paper, we aimed to characterize human NCCRP1 and to elucidate its relationship to CA IX. Recombinant NCCRP1 protein was expressed in Escherichia coli, and a novel polyclonal antiserum was raised against the purified full-length protein. Immunocytochemistry showed that NCCRP1 is expressed intracellularly, even though it has previously been described as a transmembrane protein. Using bioinformatic analyses, we identified orthologs of NCCRP1 in 35 vertebrate genomes, and up to five paralogs per genome. These paralogs are FBXO genes whose protein products are components of the E3 ubiquitin ligase complexes. NCCRP1 proteins have no signal peptides or transmembrane domains. NCCRP1 has mainly been studied in fish and was thought to be responsible for the cytolytic function of nonspecific cytotoxic cells (NCCs). Our analyses showed that in humans, NCCRP1 mRNA is expressed in tissues containing squamous epithelium, whereas it shows a more ubiquitous tissue expression pattern in mice. Neither human nor mouse NCCRP1 expression is specific to immune tissues. Silencing CA9 using siRNAs did not affect NCCRP1 levels, indicating that its expression is not directly regulated by CA9. Interestingly, silencing NCCRP1 caused a statistically significant decrease in the growth of HeLa cells. These studies provide ample evidence that the current name, "non-specific cytotoxic cell receptor protein 1," is not appropriate. We therefore propose that the gene name be changed to FBXO50.

Overall and worst gleason scores are equally good predictors of prostate cancer progression.

BACKGROUND: Gleason scoring has experienced several modifications during the past decade. So far, only one study has compared the prognostic abilities of worst (WGS) and overall (OGS) modified Gleason scores after the ISUP 2005 conference. Prostatic needle biopsies are individually paraffin-embedded in 57% of European pathology laboratories, whereas the rest of laboratories embed multiple (2 - 6) biopsies per one paraffin-block. Differences in the processing method can have a far-reaching effect, because reporting of the Gleason score (GS) is different for individually embedded and pooled biopsies, and GS is one of the most important factors when selecting treatment for patients. METHODS: The study material consisted of needle biopsies from 236 prostate cancer patients that were endocrine-treated in 1999-2003. Biopsies from left side and right side were embedded separately. Haematoxylin-eosin-stained slides were scanned and analyzed on web-based virtual microscopy. Worst and overall Gleason scores were assessed according to the modified Gleason score schema after analyzing each biopsy separately. The compound Gleason scores (CGS) were obtained from the original pathology reports. Two different grade groupings were used: GS 6 or less vs. 7 vs. 8 or above; and GS 7(3 + 4) or less vs. 7(4 + 3) and 8 vs. 9-10. The prognostic ability of the three scoring methods to predict biochemical progression was compared with Kaplan-Meier survival analysis and univariate and multivariate Cox regression analyses. RESULTS: The median follow-up time of the patients was 64.5 months (range 0-118). The modified GS criteria led to upgrading of the Gleason sums compared to the original CGS from the pathology reports 1999-2003 (mean 7.0 for CGS, 7.5 for OGS, 7.6 for WGS). In 43 cases WGS was > OGS. In a univariate analysis the relative risks were 2.1 (95%-confidence interval 1.8-2.4) for CGS, 2.5 (2.1-2.8) for OGS, and 2.6 (2.2-2.9) for WGS. In a multivariate analysis, OGS was the only independent prognostic factor. CONCLUSIONS: All of the three Gleason scoring methods are strong predictors of biochemical recurrence. The use of modified Gleason scoring leads to upgrading of GS, but also improves the prognostic value of the scoring. No significant prognostic differences between OGS and WGS could be shown, which may relate to the apparent narrowing of the GS scale from 2-10 to 5-10 due to the recent modifications.

Histopathological variables and biomarkers enhancer of zeste homologue 2, Ki-67 and minichromosome maintenance protein 7 as prognosticators in primarily endocrine-treated prostate cancer.

OBJECTIVE: • To evaluate the prognostic value of histopathological variables and immunostainings of biomarkers enhancer of zeste homologue 2 (EZH2), Ki-67 and minichromosome maintenance protein 7 (MCM7) from core biopsies of hormonally treated patients with prostate cancer. PATIENTS AND METHODS: • Biopsies of 247 primarily endocrine-treated patients were analysed for histopathological characteristics and Gleason scores (GS) according to the revised guidelines of International Society of Urologic Pathology (ISUP) consensus conference 2005. • Immunohistochemical stainings were analysed with the aid of digital image analysis. • The prognostic value of the histopathological variables and the biomarkers was analysed with univariate and multivariate Cox regression analysis, with biochemical recurrence as an endpoint. RESULTS: • Biomarkers EZH2 (relative risk [RR] 2.0, 95% confidence interval 1.2-3.3), Ki-67 (3.4, 2.1-5.5) and MCM7 (2.4, 1.5-3.9) were significantly associated with progression-free survival in a univariate analysis. • Ki-67 immunostaining index detected high-risk patients with GS of 7 (9.1, 8.0-10.3). • In a multivariate analysis with non-conventional GS groups 5-7 (3 + 4), 7(4 + 3)-8, and 9-10, the independent prognostic markers were pretreatment GS (2.2, 1.5-3.2), prostate-specific antigen (PSA) level (2.1, 1.1-4.2), perineural invasion (PNI) (1.6, 1.2-2.2), and clinical T-stage (cT) (1.9, 1.0-3.7). • Combination of the independent markers (PSA level > 20 ng/mL or GS >3 + 4 or PNI >3 or cT >2) yielded best risk stratification (RR 11.6, 10.4-12.7). CONCLUSIONS: • GS remains one of the most important prognostic factors in prostate cancer. However, the refined guidelines by ISUP 2005 might have shifted the threshold between low-grade and high-grade cancers from GS 6 vs 7 to GS 3 + 4 vs 4 + 3. • PNI is an independent prognostic marker superior to cT. • Ki-67 is the most useful biomarker in detecting patients with GS = 7 at high risk for progression.

Routine dual-color immunostaining with a 3-antibody cocktail improves the detection of small cancers in prostate needle biopsies.

We performed dual-color immunostaining with a 3-antibody cocktail (α-methylacyl coenzyme-A racemase, CK34betaE12, and p63) on prostate biopsies from 200 patients. Current practice (hematoxylin and eosin staining followed by dual-color immunostaining on selected cases) was compared with a protocol in which routine dual-color immunostaining was provided in all cases. In the original pathology reports, adenocarcinoma was diagnosed in 87/200 (43%) patients. Small foci interpreted as putative cancers were detected with dual-color immunostaining in 14/113 patients who were originally diagnosed with a nonmalignant lesion. All of the suggested cancerous foci were independently reevaluated by 5 pathologists. A diagnosis of adenocarcinoma was assessed by consensus in 8 cases, and atypical small acinar proliferation was diagnosed in 1 case. Consensus was not reached in 5 cases. Six of the foci reclassified as cancer were of Gleason score 3 + 3 = 6, while 2 were graded as Gleason score 4 + 4 = 8. The feasibility of routine dual-color immunostaining was also tested by analyzing the time spent on microscopic assessment. Because small, atypical lesions expressing α-methylacyl coenzyme-A racemase (blue chromogen) were easy to detect using dual-color immunostaining, the microscopic analysis of dual-color immunostaining and hematoxylin-eosin staining was faster than that of hematoxylin-eosin staining alone that was later followed by dual-color immunostaining in selected cases (median 251 seconds versus 299 seconds, P < .0001). We concluded that routine dual-color immunostaining of all prostate biopsies would produce better diagnostic sensitivity with a smaller microscopy workload for the pathologist. However, minute foci interpreted as cancer with dual-color immunostaining need to be confirmed with hematoxylin-eosin staining, and minimal criteria for a definitive diagnosis of cancer are still lacking.

ImmunoRatio: a publicly available web application for quantitative image analysis of estrogen receptor (ER), progesterone receptor (PR), and Ki-67.

INTRODUCTION: Accurate assessment of estrogen receptor (ER), progesterone receptor (PR), and Ki-67 is essential in the histopathologic diagnostics of breast cancer. Commercially available image analysis systems are usually bundled with dedicated analysis hardware and, to our knowledge, no easily installable, free software for immunostained slide scoring has been described. In this study, we describe a free, Internet-based web application for quantitative image analysis of ER, PR, and Ki-67 immunohistochemistry in breast cancer tissue sections. METHODS: The application, named ImmunoRatio, calculates the percentage of positively stained nuclear area (labeling index) by using a color deconvolution algorithm for separating the staining components (diaminobenzidine and hematoxylin) and adaptive thresholding for nuclear area segmentation. ImmunoRatio was calibrated using cell counts defined visually as the gold standard (training set, n = 50). Validation was done using a separate set of 50 ER, PR, and Ki-67 stained slides (test set, n = 50). In addition, Ki-67 labeling indexes determined by ImmunoRatio were studied for their prognostic value in a retrospective cohort of 123 breast cancer patients. RESULTS: The labeling indexes by calibrated ImmunoRatio analyses correlated well with those defined visually in the test set (correlation coefficient r = 0.98). Using the median Ki-67 labeling index (20%) as a cutoff, a hazard ratio of 2.2 was obtained in the survival analysis (n = 123, P = 0.01). ImmunoRatio was shown to adapt to various staining protocols, microscope setups, digital camera models, and image acquisition settings. The application can be used directly with web browsers running on modern operating systems (e.g., Microsoft Windows, Linux distributions, and Mac OS). No software downloads or installations are required. ImmunoRatio is open source software, and the web application is publicly accessible on our website. CONCLUSIONS: We anticipate that free web applications, such as ImmunoRatio, will make the quantitative image analysis of ER, PR, and Ki-67 easy and straightforward in the diagnostic assessment of breast cancer specimens.

Parallel inhibition of cell growth and induction of cell migration and invasion in breast cancer cells by bone morphogenetic protein 4.

Bone morphogenetic proteins (BMP) are extracellular signaling molecules that belong to the transforming growth factor ß (TGFß) superfamily. Bone morphogenetic proteins have diverse roles during development where they regulate proliferation, differentiation, and apoptosis in many different cell types by modulating the transcription of specific target genes. BMPs have also been implicated in both promotion and inhibition of cancer progression. We have recently shown that BMP4 is commonly expressed in breast cancer but its functional significance has not been previously explored. Our data demonstrate that in all nine breast cancer cell lines studied, BMP4 treatment leads to a dramatic growth suppression as a result of the induction of G1 arrest of the cell cycle. At the same time, BMP4 stimulates cell migration and invasion in a subset of these breast cancer cell lines. The BMP4-induced phenotypic changes were mediated through the activation of the canonical SMAD signaling pathway whereas no activation of MAP-kinases ERK1/2 or p38 was detected. Our results thus implicate that BMP4 is an important regulator of key phenotypic characteristics of cancer cells, cell growth, cell migration, and invasion, and that, similar to TGFß, it possesses both tumor suppressive and oncogenic properties in breast cancer.

Linking whole-slide microscope images with DICOM by using JPEG2000 interactive protocol.

The use of digitized histopathologic specimens (also known as whole-slide images (WSIs)) in clinical medicine requires compatibility with the Digital Imaging and Communications in Medicine (DICOM) standard. Unfortunately, WSIs usually exceed DICOM image object size limit, making it impossible to store and exchange them in a straightforward way. Moreover, transmitting the entire DICOM image for viewing is ineffective for WSIs. With the JPEG2000 Interactive Protocol (JPIP), WSIs can be linked with DICOM by transmitting image data over an auxiliary connection, apart from patient data. In this study, we explored the feasibility of using JPIP to link JPEG2000 WSIs with a DICOM-based Picture Archiving and Communications System (PACS). We first modified an open-source DICOM library by adding support for JPIP as described in the existing DICOM Supplement 106. Second, the modified library was used as a basis for a software package (JVSdicom), which provides a proof-of-concept for a DICOM client-server system that can transmit patient data, conventional DICOM imagery (e.g., radiological), and JPIP-linked JPEG2000 WSIs. The software package consists of a compression application (JVSdicom Compressor) for producing DICOM-compatible JPEG2000 WSIs, a DICOM PACS server application (JVSdicom Server), and a DICOM PACS client application (JVSdicom Workstation). JVSdicom is available for free from our Web site ( http://jvsmicroscope.uta.fi/ ), which also features a public JVSdicom Server, containing example X-ray images and histopathology WSIs of breast cancer cases. The software developed indicates that JPEG2000 and JPIP provide a well-working solution for linking WSIs with DICOM, requiring only minor modifications to current DICOM standard specification.

The application of JPEG2000 in virtual microscopy.

Virtual microscopy (i.e., the viewing of entire microscope specimens on a computer display) is becoming widely applied in microscopy teaching and clinical laboratory medicine. Despite rapidly increasing use, virtual microscopy currently lacks of a universally accepted image format. A promising candidate is JPEG2000, which has potential advantages for handling gigabyte-sized virtual slides. To date, no JPEG2000-based software has been specifically suited for virtual microscopy. To study the utility of JPEG2000 in virtual microscopy, we first optimized JPEG2000 code-stream parameters for virtual slide viewing (i.e., fast navigation, zooming, and use of an overview window). Compression using ratios 25:1-30:1 with the irreversible wavelet filter were found to provide the best compromise between file size and image quality. Optimal code-stream parameters also consisted of 10 wavelet decomposition levels, progression order Resolution-Position-Component-Layer (RPCL), a precinct size of 128 x 128, and code-block size of 64 x 64. Tiling and the use of multiple quality layers were deemed unnecessary. A compression application (JVScomp) was developed for creating optimally parameterized JPEG2000 virtual slides. A viewing application (JVSview) was developed specifically for virtual microscopy, offering all of the basic viewing functions. JVSview also supports viewing of focus stacks, embedding of textual descriptions, and defining regions of interest as metadata. Combined with our server application (JVSserv), virtual slides can be viewed over networks by employing the JPEG2000 Interactive Protocol (JPIP). The software can be tested using virtual slide examples located on our public JPIP server ( http://jvsmicroscope.uta.fi/ ). The software package is freely downloadable and usable for noncommercial purposes.