Robust, credible, and interpretable AI-based histopathological prostate cancer grading.

Background: Prostate cancer (PCa) is among the most common cancers in men and its diagnosis requires the histopathological evaluation of biopsies by human experts. While several recent artificial intelligence-based (AI) approaches have reached human expert-level PCa grading, they often display significantly reduced performance on external datasets. This reduced performance can be caused by variations in sample preparation, for instance the staining protocol, section thickness, or scanner used. Another limiting factor of contemporary AI-based PCa grading is the prediction of ISUP grades, which leads to the perpetuation of human annotation errors. Methods: We developed the prostate cancer aggressiveness index (PCAI), an AI-based PCa detection and grading framework that is trained on objective patient outcome, rather than subjective ISUP grades. We designed PCAI as a clinical application, containing algorithmic modules that offer robustness to data variation, medical interpretability, and a measure of prediction confidence. To train and evaluate PCAI, we generated a multicentric, retrospective, observational trial consisting of six cohorts with 25,591 patients, 83,864 images, and 5 years of median follow-up from 5 different centers and 3 countries. This includes a high-variance dataset of 8,157 patients and 28,236 images with variations in sample thickness, staining protocol, and scanner, allowing for the systematic evaluation and optimization of model robustness to data variation. The performance of PCAI was assessed on three external test cohorts from two countries, comprising 2,255 patients and 9,437 images. Findings: Using our high-variance datasets, we show how differences in sample processing, particularly slide thickness and staining time, significantly reduce the performance of AI-based PCa grading by up to 6.2 percentage points in the concordance index (C-index). We show how a select set of algorithmic improvements, including domain adversarial training, conferred robustness to data variation, interpretability, and a measure of credibility to PCAI. These changes lead to significant prediction improvement across two biopsy cohorts and one TMA cohort, systematically exceeding expert ISUP grading in C-index and AUROC by up to 22 percentage points. Interpretation: Data variation poses serious risks for AI-based histopathological PCa grading, even when models are trained on large datasets. Algorithmic improvements for model robustness, interpretability, credibility, and training on high-variance data as well as outcome-based severity prediction gives rise to robust models with above ISUP-level PCa grading performance.

AI-based prostate analysis system trained without human supervision to predict patient outcome from tissue samples.

In order to plan the best treatment for prostate cancer patients, the aggressiveness of the tumor is graded based on visual assessment of tissue biopsies according to the Gleason scale. Recently, a number of AI models have been developed that can be trained to do this grading as well as human pathologists. But the accuracy of the AI grading will be limited by the accuracy of the subjective "ground truth" Gleason grades used for the training. We have trained an AI to predict patient outcome directly based on image analysis of a large biobank of tissue samples with known outcome without input of any human knowledge about cancer grading. The model has shown similar and in some cases better ability to predict patient outcome on an independent test-set than expert pathologists doing the conventional grading.

Radical Prostatectomy or Watchful Waiting in Prostate Cancer - 29-Year Follow-up.

BACKGROUND: Radical prostatectomy reduces mortality among men with clinically detected localized prostate cancer, but evidence from randomized trials with long-term follow-up is sparse. METHODS: We randomly assigned 695 men with localized prostate cancer to watchful waiting or radical prostatectomy from October 1989 through February 1999 and collected follow-up data through 2017. Cumulative incidence and relative risks with 95% confidence intervals for death from any cause, death from prostate cancer, and metastasis were estimated in intention-to-treat and per-protocol analyses, and numbers of years of life gained were estimated. We evaluated the prognostic value of histopathological measures with a Cox proportional-hazards model. RESULTS: By December 31, 2017, a total of 261 of the 347 men in the radical-prostatectomy group and 292 of the 348 men in the watchful-waiting group had died; 71 deaths in the radical-prostatectomy group and 110 in the watchful-waiting group were due to prostate cancer (relative risk, 0.55; 95% confidence interval [CI], 0.41 to 0.74; P<0.001; absolute difference in risk, 11.7 percentage points; 95% CI, 5.2 to 18.2). The number needed to treat to avert one death from any cause was 8.4. At 23 years, a mean of 2.9 extra years of life were gained with radical prostatectomy. Among the men who underwent radical prostatectomy, extracapsular extension was associated with a risk of death from prostate cancer that was 5 times as high as that among men without extracapsular extension, and a Gleason score higher than 7 was associated with a risk that was 10 times as high as that with a score of 6 or lower (scores range from 2 to 10, with higher scores indicating more aggressive cancer). CONCLUSIONS: Men with clinically detected, localized prostate cancer and a long life expectancy benefited from radical prostatectomy, with a mean of 2.9 years of life gained. A high Gleason score and the presence of extracapsular extension in the radical prostatectomy specimens were highly predictive of death from prostate cancer. (Funded by the Swedish Cancer Society and others.).

The protein kinase SIK downregulates the polarity protein Par3.

The multifunctional cytokine transforming growth factor ß (TGFß) controls homeostasis and disease during embryonic and adult life. TGFß alters epithelial cell differentiation by inducing epithelial-mesenchymal transition (EMT), which involves downregulation of several cell-cell junctional constituents. Little is understood about the mechanism of tight junction disassembly by TGFß. We found that one of the newly identified gene targets of TGFß, encoding the serine/threonine kinase salt-inducible kinase 1 (SIK), controls tight junction dynamics. We provide bioinformatic and biochemical evidence that SIK can potentially phosphorylate the polarity complex protein Par3, an established regulator of tight junction assembly. SIK associates with Par3, and induces degradation of Par3 that can be prevented by proteasomal and lysosomal inhibition or by mutation of Ser885, a putative phosphorylation site on Par3. Functionally, this mechanism impacts on tight junction downregulation. Furthermore, SIK contributes to the loss of epithelial polarity and examination of advanced and invasive human cancers of diverse origin displayed high levels of SIK expression and a corresponding low expression of Par3 protein. High SIK mRNA expression also correlates with lower chance for survival in various carcinomas. In specific human breast cancer samples, aneuploidy of tumor cells best correlated with cytoplasmic SIK distribution, and SIK expression correlated with TGFß/Smad signaling activity and low or undetectable expression of Par3. Our model suggests that SIK can act directly on the polarity protein Par3 to regulate tight junction assembly.

Analysis of the Human Prostate-Specific Proteome Defined by Transcriptomics and Antibody-Based Profiling Identifies TMEM79 and ACOXL as Two Putative, Diagnostic Markers in Prostate Cancer.

To better understand prostate function and disease, it is important to define and explore the molecular constituents that signify the prostate gland. The aim of this study was to define the prostate specific transcriptome and proteome, in comparison to 26 other human tissues. Deep sequencing of mRNA (RNA-seq) and immunohistochemistry-based protein profiling were combined to identify prostate specific gene expression patterns and to explore tissue biomarkers for potential clinical use in prostate cancer diagnostics. We identified 203 genes with elevated expression in the prostate, 22 of which showed more than five-fold higher expression levels compared to all other tissue types. In addition to previously well-known proteins we identified two poorly characterized proteins, TMEM79 and ACOXL, with potential to differentiate between benign and cancerous prostatic glands in tissue biopsies. In conclusion, we have applied a genome-wide analysis to identify the prostate specific proteome using transcriptomics and antibody-based protein profiling to identify genes with elevated expression in the prostate. Our data provides a starting point for further functional studies to explore the molecular repertoire of normal and diseased prostate including potential prostate cancer markers such as TMEM79 and ACOXL.

Radical prostatectomy or watchful waiting in early prostate cancer.

BACKGROUND: Radical prostatectomy reduces mortality among men with localized prostate cancer; however, important questions regarding long-term benefit remain. METHODS: Between 1989 and 1999, we randomly assigned 695 men with early prostate cancer to watchful waiting or radical prostatectomy and followed them through the end of 2012. The primary end points in the Scandinavian Prostate Cancer Group Study Number 4 (SPCG-4) were death from any cause, death from prostate cancer, and the risk of metastases. Secondary end points included the initiation of androgen-deprivation therapy. RESULTS: During 23.2 years of follow-up, 200 of 347 men in the surgery group and 247 of the 348 men in the watchful-waiting group died. Of the deaths, 63 in the surgery group and 99 in the watchful-waiting group were due to prostate cancer; the relative risk was 0.56 (95% confidence interval [CI], 0.41 to 0.77; P=0.001), and the absolute difference was 11.0 percentage points (95% CI, 4.5 to 17.5). The number needed to treat to prevent one death was 8. One man died after surgery in the radical-prostatectomy group. Androgen-deprivation therapy was used in fewer patients who underwent prostatectomy (a difference of 25.0 percentage points; 95% CI, 17.7 to 32.3). The benefit of surgery with respect to death from prostate cancer was largest in men younger than 65 years of age (relative risk, 0.45) and in those with intermediate-risk prostate cancer (relative risk, 0.38). However, radical prostatectomy was associated with a reduced risk of metastases among older men (relative risk, 0.68; P=0.04). CONCLUSIONS: Extended follow-up confirmed a substantial reduction in mortality after radical prostatectomy; the number needed to treat to prevent one death continued to decrease when the treatment was modified according to age at diagnosis and tumor risk. A large proportion of long-term survivors in the watchful-waiting group have not required any palliative treatment. (Funded by the Swedish Cancer Society and others.).

Stem cell marker-positive stellate cells and mast cells are reduced in benign-appearing bladder tissue in patients with urothelial carcinoma.

Survival after invasive bladder cancer has improved less than that of other common non-skin cancers. In many types of malignancy, treatment failure has been attributed to therapy-resistant stem-like cancer cells. Our aim was therefore to determine identities of stem cell marker-positive cells in bladder cancer tissue and to investigate possible associations between these cells and different forms of bladder neoplasia. We investigated tissue from 52 patients with bladder neoplasia and 18 patients with benign bladder conditions, from a cohort that had been previously described with regard to diagnosis and outcome. The samples were analysed immunohistologically for the stem cell markers aldehyde dehydrogenase 1 A1 (ALDH1) and CD44, and markers of cell differentiation. The majority of stem cell marker-positive cells were located in connective tissue, and a smaller fraction in epithelial tissue. Stem cell marker-positive cells exhibiting possible stem cell characteristics included cells in deeper locations of benign and malignant epithelium, and sub-endothelial cells in patients with or without neoplasia. Stem cell marker-positive cells with non-stem cell character included stellate cells, mast cells, endothelial cells, foamy histiocytes, and neurons. Significantly, ALDH1+ stellate cells and ALDH1+ mast cells were reduced in number in stroma of benign-appearing mucosa of bladder cancer patients. The stem cell markers ALDH1 and CD44 label several types of differentiated cells in bladder tissue. ALDH1+ stellate cells and mast cells appear to be reduced in stroma of normal-appearing mucosa of bladder cancer patients, and may be part of a "field effect" in cancer-near areas.

Histological stain evaluation for machine learning applications.

AIMS: A methodology for quantitative comparison of histological stains based on their classification and clustering performance, which may facilitate the choice of histological stains for automatic pattern and image analysis. BACKGROUND: Machine learning and image analysis are becoming increasingly important in pathology applications for automatic analysis of histological tissue samples. Pathologists rely on multiple, contrasting stains to analyze tissue samples, but histological stains are developed for visual analysis and are not always ideal for automatic analysis. MATERIALS AND METHODS: Thirteen different histological stains were used to stain adjacent prostate tissue sections from radical prostatectomies. We evaluate the stains for both supervised and unsupervised classification of stain/tissue combinations. For supervised classification we measure the error rate of nonlinear support vector machines, and for unsupervised classification we use the Rand index and the F-measure to assess the clustering results of a Gaussian mixture model based on expectation-maximization. Finally, we investigate class separability measures based on scatter criteria. RESULTS: A methodology for quantitative evaluation of histological stains in terms of their classification and clustering efficacy that aims at improving segmentation and color decomposition. We demonstrate that for a specific tissue type, certain stains perform consistently better than others according to objective error criteria. CONCLUSIONS: The choice of histological stain for automatic analysis must be based on its classification and clustering performance, which are indicators of the performance of automatic segmentation of tissue into morphological components, which in turn may be the basis for diagnosis.

Tumour expression of bladder cancer-associated urinary proteins.

UNLABELLED: WHAT'S KNOWN ON THE SUBJECT? AND WHAT DOES THE STUDY ADD?: The current basis for diagnosis and prognosis in urinary bladder cancer is based on the pathologists' assessment of a biopsy of the tumour. Urinary biomarkers are preferable as they can be non-invasively sampled. Urinary cytology is the only test with widespread use but is hampered by poor reproducibility and low sensitivity. By studying the protein expression in bladder tumour tissue samples of proteins previously found in elevated levels in the urine of patients with bladder cancer, we have been able to show that these proteins originate from the tumour. The immunoreactivity of three of the investigated proteins increased with higher stage. Also a serine peptidase inhibitor was found to be predictive of progression from non-muscle-invasive to muscle-invasive tumours. OBJECTIVES: To analyse the expression of five bladder cancer-associated urinary proteins and investigate if expression is related to the malignant phenotype of the tumour. To explore the possible prognostic value of these proteins. PATIENTS AND METHODS: Urine samples, 16 from patients with bladder cancer and 26 from controls, were used in Western Blotting experiments. Tissue microarrays with bladder tissue from 344 patients diagnosed with bladder cancer between 1984 and 2005 was used in immunohistochemistry experiments. The proteins apolipoprotein E (APOE), fibrinogen ß chain precursor (FGB), leucine-rich α2-glycoprotein (LRG1), polymerase (RNA) I polypeptide E (POLR1E), α1-antitrypsin (SERPINA1) and topoisomerase 2A (TOP2A) were probed with antibodies validated by the Human Protein Atlas. RESULTS: Increased expressions of APOE, FGB and POLR1E were correlated with increased tumour stage (P < 0.001). Expression of SERPINA1 in Ta and T1 tumours was found to increase the risk of tumour progression (hazard ratio 2.57, 95% confidence interval 1.13-5.87; P = 0.025) CONCLUSIONS: All proteins previously detected in urine from patients with bladder cancer were also expressed in bladder cancer tissue. The expression of APOE, FGB and POLR1E increased with stage and they are potential diagnostic markers. SERPINA1 was identified as a prognostic marker candidate.

Blind color decomposition of histological images.

Cancer diagnosis is based on visual examination under a microscope of tissue sections from biopsies. But whereas pathologists rely on tissue stains to identify morphological features, automated tissue recognition using color is fraught with problems that stem from image intensity variations due to variations in tissue preparation, variations in spectral signatures of the stained tissue, spectral overlap and spatial aliasing in acquisition, and noise at image acquisition. We present a blind method for color decomposition of histological images. The method decouples intensity from color information and bases the decomposition only on the tissue absorption characteristics of each stain. By modeling the charge-coupled device sensor noise, we improve the method accuracy. We extend current linear decomposition methods to include stained tissues where one spectral signature cannot be separated from all combinations of the other tissues' spectral signatures. We demonstrate both qualitatively and quantitatively that our method results in more accurate decompositions than methods based on non-negative matrix factorization and independent component analysis. The result is one density map for each stained tissue type that classifies portions of pixels into the correct stained tissue allowing accurate identification of morphological features that may be linked to cancer.

Microarray core detection by geometric restoration.

Whole-slide imaging of tissue microarrays (TMAs) holds the promise of automated image analysis of a large number of histopathological samples from a single slide. This demands high-throughput image processing to enable analysis of these tissue samples for diagnosis of cancer and other conditions. In this paper, we present a completely automated method for the accurate detection and localization of tissue cores that is based on geometric restoration of the core shapes without placing any assumptions on grid geometry. The method relies on hierarchical clustering in conjunction with the Davies-Bouldin index for cluster validation in order to estimate the number of cores in the image wherefrom we estimate the core radius and refine this estimate using morphological granulometry. The final stage of the algorithm reconstructs circular discs from core sections such that these discs cover the entire region of each core regardless of the precise shape of the core. The results show that the proposed method is able to reconstruct core locations without any evidence of localization. Furthermore, the algorithm is more efficient than existing methods based on the Hough transform for circle detection. The algorithm's simplicity, accuracy, and computational efficiency allow for automated high-throughput analysis of microarray images.

Diagnosis of intraurothelial neoplasia. Interobserver variation and the value of individual histopathologic attributes.

OBJECTIVE: To determine interobserver variation in histopathologic diagnosis of carcinoma in situ (CIS) and dysplasia (collectively intraurothelial neoplasia [IUN]) of the bladder and identify histomorphologic features important for diagnosis. STUDY DESIGN: A total of 272 consecutive bladder tissue samples were re-evaluated blindly by two general pathologists and one uropathologist for IUN. Discrepancies were resolved jointly. Fifteen histopathologic attributes were evaluated for prediction of diagnosis. Followup revealed recurrence and progression rates for each diagnostic category. RESULTS: Thirty-six percent of specimens contained no evaluable flat mucosa; 51% percent of specimens from papillary urothelial neoplasia (PUN) cases showed CIS. General pathologists detected 56-69% of CIS and 8-42% of dysplasia. Histopathologic features most predictive for CIS were nuclear size, variation in nuclear shape, loss of maturation, loss of polarity, and architectural disorder. None of these individually or in combination exceeded general pathologists' diagnostic accuracy. IUN was not predictive of recurrence or progress. CONCLUSION: Using material mostly consisting of flat mucosa gratuitously provided in PUN resection specimens, IUN carries no prognostic value. General histopathologists detect IUN poorly to moderately, and the five most discriminatory histomorphologic features are insufficient for diagnosis. Interobserver agreement for dysplasia is dismal. Absent flat mucosa in PUN resections predicts recurrence.

Simplification of grading papillary urothelial neoplasia using a reduced set of diagnostic features.

OBJECTIVE: To determine whether a reduced set of the histopathologic features used in internationally accepted classifications is capable of accurately grading papillary urothelial neoplasms (PUN). STUDY DESIGN: All surgical specimens from urinary bladders received during a 2-year period were reexamined by an expert uropathologist for assessing the accuracy of original nonexpert PUN grading and staging. Thirteen histopathologic features entailing 32 attributes were evaluated with regard to prediction of expert grade. Patients were followed for 35-59 months (mean, 47). RESULTS: A total of 88 PUN specimens could be analyzed completely including follow-up specimens. Agreement between original and expert grade was 71% for low-grade and 87% for high-grade PUN, with overall kappa = 0.53. The histomorphologic features most predictive of expert grade were architectural disorder, variability of nuclear enlargement, and absence of umbrella cells. Neither individual histomorphologic attributes nor their combinations were as predictive of expert pathologist grade as original diagnoses. CONCLUSION: Improvements in PUN grading and prognostication are not likely to be accomplished by only reducing the number of histomorphologic features currently recommended by the World Health Organization and International Society of Urological Pathology.

Radical prostatectomy versus watchful waiting in early prostate cancer.

BACKGROUND: In 2008, we reported that radical prostatectomy, as compared with watchful waiting, reduces the rate of death from prostate cancer. After an additional 3 years of follow-up, we now report estimated 15-year results. METHODS: From October 1989 through February 1999, we randomly assigned 695 men with early prostate cancer to watchful waiting or radical prostatectomy. Follow-up was complete through December 2009, with histopathological review of biopsy and radical-prostatectomy specimens and blinded evaluation of causes of death. Relative risks, with 95% confidence intervals, were estimated with the use of a Cox proportional-hazards model. RESULTS: During a median of 12.8 years, 166 of the 347 men in the radical-prostatectomy group and 201 of the 348 in the watchful-waiting group died (P=0.007). In the case of 55 men assigned to surgery and 81 men assigned to watchful waiting, death was due to prostate cancer. This yielded a cumulative incidence of death from prostate cancer at 15 years of 14.6% and 20.7%, respectively (a difference of 6.1 percentage points; 95% confidence interval [CI], 0.2 to 12.0), and a relative risk with surgery of 0.62 (95% CI, 0.44 to 0.87; P=0.01). The survival benefit was similar before and after 9 years of follow-up, was observed also among men with low-risk prostate cancer, and was confined to men younger than 65 years of age. The number needed to treat to avert one death was 15 overall and 7 for men younger than 65 years of age. Among men who underwent radical prostatectomy, those with extracapsular tumor growth had a risk of death from prostate cancer that was 7 times that of men without extracapsular tumor growth (relative risk, 6.9; 95% CI, 2.6 to 18.4). CONCLUSIONS: Radical prostatectomy was associated with a reduction in the rate of death from prostate cancer. Men with extracapsular tumor growth may benefit from adjuvant local or systemic treatment. (Funded by the Swedish Cancer Society and the National Institutes of Health.).

Sequential intravesical chemoimmunotherapy with mitomycin C and bacillus Calmette-Guérin and with bacillus Calmette-Guérin alone in patients with carcinoma in situ of the urinary bladder: results of an EORTC genito-urinary group randomized phase 2 trial (30993).

BACKGROUND: Bacillus Calmette-Guérin (BCG) is the intravesical treatment of choice for carcinoma in situ (CIS). OBJECTIVE: Our aim was to assess if sequential mitomycin C (MMC) plus BCG after transurethral resection (TUR) is worthy of further study in non-muscle-invasive bladder cancer patients with CIS. DESIGN, SETTING, AND PARTICIPANTS: In a noncomparative phase 2 study, 96 patients with primary/secondary/concurrent CIS of the urinary bladder were randomized to sequential MMC plus BCG or to BCG alone after TUR. INTERVENTION: Patients received six weekly instillations of MMC followed by six weekly instillations of BCG or six weekly instillations of BCG, 3 wk rest, and three further weekly instillations of BCG. Complete responders received three weekly maintenance instillations at 6, 12, 18, 24, 30, and 36 mo in accordance with the initial randomization. MEASUREMENTS: End points were complete response (CR) rate at the first control cystoscopy 16-18 wk after start of treatment, disease-free interval, overall survival, and side effects. RESULTS AND LIMITATIONS: Ninety-six patients were randomized, 48 to each treatment group. Ten patients were ineligible, and three did not start treatment. In all randomized patients, CR rates on MMC plus BCG and BCG alone were 70.8% and 66.7%, respectively. In 83 eligible patients who started treatment, CR rates were 75.6% and 73.8%, respectively. Based on a median follow-up of 4.7 yr, 25 patients (52.1%) on MMC plus BCG and 22 patients (45.8%) on BCG alone were disease free. Twelve patients stopped treatment due to toxicity: three during induction (two MMC plus BCG, one BCG) and nine during maintenance (three MMC plus BCG, six BCG). CONCLUSIONS: In the treatment of patients with CIS, sequential chemoimmunotherapy with MMC plus BCG had acceptable toxicity. CR and disease-free rates were similar to those on BCG alone and to previous publications on sequential chemoimmunotherapy. TRIAL REGISTRATION: This study was registered with the US National Cancer Institute clinical trials database (protocol ID: EORTC-30993). http://www.cancer.gov/search/ViewClinicalTrials.aspx?cdrid=68869&version=HealthProfessional&protocolsearchid=7920643.

The metabolic syndrome and the risk of prostate cancer under competing risks of death from other causes.

BACKGROUND: Associations between metabolic syndrome (MetS) components and prostate cancer development have not been studied comprehensively; results have been divergent. Using the National Cholesterol Education Program Adult Treatment panel III (NCEP) and International Diabetes Federation (IDF) definitions of the MetS, we investigated such associations taking competing risks of death into consideration. METHODS: In the prospective Uppsala Longitudinal Study of Adult Men of 2,322 Caucasian men with 34 years of follow-up baseline, MetS measurements at age 50 years were used. Cumulative incidence of prostate cancer and death with/without the MetS were calculated. Competing risk of dying was taken into account by calculating the conditional probability of prostate cancer with/without the MetS. RESULTS: Two hundred and thirty-seven prostate cancers were identified. Prostate cancer probability by age 80 years with baseline MetS compared with without MetS was nonsignificantly higher [5.2 percent units (confidence interval (CI), -0.8% to 11.3%; NCEP); 2.7 percent units (CI, -2.7% to 8.0%; IDF)]; cumulative incidence proportions of death was significantly higher [19.3 percent units (CI, 13.4-25.3%; NCEP); 15.3 percent units (CI, 9.5-21.1%; IDF)]; and conditional probability of prostate cancer considering death from other causes was significantly higher [7.3 percent-units (CI, 0.2-14.5%); odds ratio of 1.64 (CI, 1.03-2.23; NCEP)] and nonsignificantly higher [5.0 percent-units (CI, -1.6% to 11.6%); odds ratio of 1.43 (CI, 0.89-1.90; IDF]. CONCLUSIONS: The MetS by the NCEP definition is associated with prostate cancer, taking the competing risk of early death from other causes into account. IMPACT: The results further highlight the public health effect of the increasing prevalence of MetS and the importance of considering competing risks when studying risk factors for cancer.

COX-2 is overexpressed in primary prostate cancer with metastatic potential and may predict survival. A comparison study between COX-2, TGF-beta, IL-10 and Ki67.

BACKGROUND: The immune modulating molecules cyclooxygenase-2 (COX-2), transforming growth factor-beta (TGF-beta) and interleukin-10 (IL-10) have regulatory roles in cancer progression. There are conflicting data regarding the roles of these molecules in prostate cancer. To elucidate the prognostic impact of these proteins and provide information on prognosis and treatment, we compared the expression of COX-2, TGF-beta, and IL-10 in prostate cancer specimens with or without metastases. Ki67 was included as a measure of growth fraction of tumor cells. METHODS: Digital video analysis images from tumor cell areas and tumor stromal areas were analyzed on formalin fixed, paraffin-embedded and immunohistochemical stained cancer specimens from 59 patients: 32 patients with metastases and 27 patients without clinical, biochemical, or radiological evidence of metastases within 10 years after diagnosis. The expression of COX-2 was scored as negative, weak, moderate, or strong. The expressions of TGF-beta and IL-10 were assessed as proportions of moderately or strongly stained cells. Ki67 was detected as strong nuclear staining in proliferating cells. RESULTS: In primary cancers in the metastatic group, COX-2, TGF-beta and Ki67 were stronger expressed in epithelial tumor cell and tumor stromal areas compared with non-metastatic cancers (for all markers, p<0.0001). High intensity of COX-2 staining in tumor areas was strongly associated with death from prostate cancer in univariate analyses (hazard ratio [HR] 95% CI, 4.0 (1.1-14.5)). In multivariate analyses, the risk estimate was strengthened but did not reach significance. No associations to death were found for the other markers. CONCLUSION: High expression of COX-2, TGF-beta and Ki67 were in metastatic primary prostate carcinoma compared to non-metastatic cancers. High expression of COX-2 was associated to death from prostate carcinoma.

Focal amplifications are associated with high grade and recurrences in stage Ta bladder carcinoma.

Urinary bladder cancer is a heterogeneous disease with tumors ranging from papillary noninvasive (stage Ta) to solid muscle infiltrating tumors (stage T2+). The risk of progression and death for the most frequent diagnosed type, Ta, is low, but the high incidence of recurrences has a significant effect on the patients' quality of life and poses substantial costs for health care systems. Consequently, the purpose of this study was to search for predictive factors of recurrence on the basis of genetic profiling. A clinically well characterized cohort of Ta bladder carcinomas, selected by the presence or absence of recurrences, was evaluated by an integrated analysis of DNA copy number changes and gene expression (clone-based 32K, respectively, U133Plus2.0 arrays). Only a few chromosomal aberrations have previously been defined in superficial bladder cancer. Surprisingly, the profiling of Ta tumors with a high-resolution array showed that DNA copy alterations are relatively common in this tumor type. Furthermore, we observed an overrepresentation of focal amplifications within high-grade and recurrent cases. Known (FGFR3, CCND1, MYC, MDM2) and novel candidate genes were identified within the loci. For example, MYBL2, a nuclear transcription factor involved in cell-cycle progression; YWHAB, an antiapoptotic protein; and SDC4, an important component of focal adhesions represent interesting candidates detected within two amplicons on chromosome 20, for which DNA amplification correlated with transcript up-regulation. The observed overrepresentation of amplicons within high-grade and recurrent cases may be clinically useful for the identification of patients who will benefit from a more aggressive therapy.

Personalized prediction of tumor response and cancer progression on prostate needle biopsy.

PURPOSE: To our knowledge in patients with prostate cancer there are no available tests except clinical variables to determine the likelihood of disease progression. We developed a patient specific, biology driven tool to predict outcome at diagnosis. We also investigated whether biopsy androgen receptor levels predict a durable response to therapy after secondary treatment. MATERIALS AND METHODS: We evaluated paraffin embedded prostate needle biopsy tissue from 1,027 patients with cT1c-T3 prostate cancer treated with surgery and followed a median of 8 years. Machine learning was done to integrate clinical data with biopsy quantitative biometric features. Multivariate models were constructed to predict disease progression with the C index to estimate performance. RESULTS: In a training set of 686 patients (total of 87 progression events) 3 clinical and 3 biopsy tissue characteristics were identified to predict clinical progression within 8 years after prostatectomy with 78% sensitivity, 69% specificity, a C index of 0.74 and a HR of 5.12. Validation in an independent cohort of 341 patients (total of 44 progression events) yielded 76% sensitivity, 64% specificity, a C index of 0.73 and a HR of 3.47. Increased androgen receptor in tumor cells in the biopsy highly significantly predicted resistance to therapy, ie androgen ablation with or without salvage radiotherapy, and clinical failure (p <0.0001). CONCLUSIONS: Morphometry reliably classifies Gleason pattern 3 tumors. When combined with biomarker data, it adds to the hematoxylin and eosin analysis, and prostate specific antigen values currently used to assess outcome at diagnosis. Biopsy androgen receptor levels predict the likelihood of a response to therapy after recurrence and may guide future treatment decisions.

Concordance between Gleason scores of needle biopsies and radical prostatectomy specimens: a population-based study.

OBJECTIVE: To study the concordance between the Gleason scores of needle biopsies and radical prostatectomy (RP) specimens in a population-based registry, to clarify whether the concordance depends on the annual number of RP specimens assessed in the pathology unit, and to identify preoperative clinical factors that predict upgrading from a Gleason score of or=7 in the RP specimen. PATIENTS AND METHODS: Through the Cancer Registry of Norway, we identified 1116 patients with available Gleason scores from biopsy and RP specimens. Concordance was evaluated using the kappa coefficient, and predictors of concordance were assessed in univariate and multivariate logistic regression analyses. RESULTS: The Gleason scores were identical in biopsy and RP specimens in 591 of the 1116 (53%) patients. The biopsy-based Gleason score more often under-graded (38%) than over-graded (9%) the RP-based Gleason score. Pathology units that examined >40 RP specimens annually had a higher concordance between the Gleason score in the biopsy and RP specimen than did lower-volume units. The rate of upgrading from a Gleason score of or=7 in the RP specimen increased with increasing preoperative prostate-specific antigen serum levels, and with increasing intervals between biopsy and RP. CONCLUSIONS: The concordance in Gleason score between biopsy and RP was highest among the pathology departments that regularly evaluated RP specimens. Careful consideration of clinical factors and biopsy grading might improve the identification of patients considered as suitable for active surveillance.