Quantitative fluorescence image analysis in bladder cancer screening

"SMART" cytology: The next generation cytology for precision diagnosis

Cytology plays an important role in diagnosing and managing human diseases, especially cancer, as it is often a simple, low cost yet effective, and non-invasive or minimally invasive diagnostic tool. However, traditional morphology-based cytology practice has limitations, especially in the era of precision diagnosis. Recently there have been tremendous efforts devoted to apply computational tools and to perform molecular analysis on cytological samples for a variety of clinical purposes. Now is probably the appropriate juncture to integrate morphology, machine learning, and molecular analysis together and transform cytology from a morphology-driven practice to the next level - "SMART" Cytology. In this article we will provide a rather brief review of the relevant works for computational analysis on cytology samples, focusing on single-cell-based multiplex quantitative analysis of biomarkers, and introduce the conceptual framework of "SMART (Single cell, Multiplex, AI-driven, and Real Time)" Cytology.

The Drake Health Registry Study: cause-specific mortality experience of workers potentially exposed to beta-naphthylamine

OBJECTIVE

To examine the cause-specific mortality experience of an occupational cohort with probable past exposure to beta-naphythylamine (BNA).

METHODS

Subjects were 374 male and 26 female workers employed at a Pennsylvania chemical plant that produced or used beta-naphthylamine (BNA) between 1940 and 1981. Vital status through 1998 was determined for 97.5% of the cohort and cause of death for 100% of 79 deaths. Limited industrial hygiene data and reports from former employees were used to categorize workers as high, medium, or low risk for BNA exposure. Statistical analyses included US and local county-based standardized mortality ratios (SMRs).

RESULTS

We observed statistically significantly elevated county rate-based SMRs for all causes combined (SMR = 1.98, 95% confidence interval (CI) = 1.56-2.49), all malignant neoplasms combined (28 deaths, SMR = 3.08, 95% CI = 2.05-4.46), respiratory system cancer (12 deaths, SMR = 3.91, 95% CI = 2.02-6.83), and bladder cancer (four deaths, SMR = 16.83, 95% CI = 4.59-43.1). Three bladder cancer cases were classified as high risk (SMR = 26.79, 95% CI = 5.53-78.29). Mortality risks were also elevated for most other malignant and non-malignant cause of death categories examined.

CONCLUSIONS

Bladder cancer risk remains highly elevated among Drake/Kilsdonk workers and appears to be causally related to past BNA exposure. While lifestyle and behavioral risk factors may explain some of the mortality excesses for non-urological cancers, the possibility remains that BNA exposure may have also played a role in these and other observed cancer excesses.

The Drake Health Registry Study: findings from fifteen years of continuous bladder cancer screening

BACKGROUND

The Drake Health Registry Study (DHRS) is an ongoing bladder cancer screening program initiated in 1986 due to workers' probable past exposure to the bladder carcinogen, beta-naphthylamine (BNA).

METHODS

At periodic screening visits, a health survey is administered and three screening tests are applied to a urine sample, urinalysis (UA), papanicolaou (PAP), and quantitative fluorescence image analysis (QFIA). Positive screens are eligible for a free bladder cystoscopy with random biopsies.

RESULTS

Forty of 51 persons eligible for diagnostic evaluation underwent cystoscopy. One person was diagnosed with carcinoma in situ, two with transitional cell papilloma, 14 with dysplasia, two of which developed transitional cell carcinoma; 26 had bladder abnormalities such as chronic inflammation, chronic cystitis, atypical changes, atypia, hyperplasia, or papillary clusters.

CONCLUSIONS

The DHRS continues to identify early stage bladder cancer and other abnormalities among workers exposed to BNA before 1981 and generates useful clinical, psycho-social, and epidemiologic data.

Biomarkers in monitoring for efficacy of immunotherapy and chemoprevention of bladder cancer with dimethylsulfoxide

This study correlated biomarkers expressed in tumor and epithelial field with clinical response and recurrence. Of 25 bladder cancer patients, 11 received 6 weeks of intravesical Bacille Calmette-Guerin (BCG), and 14 were treated weekly with intravesical dimethylsulfoxide (DMSO) for 4 weeks to further modulate biomarker expression. G-actin, DNA aneuploidy, and p300 tumor antigen were evaluated by quantitative fluorescence image analysis on uroepithelial cells from bladder wash samples prior to and immediately following treatment. Excluding patients who did not respond to BCG (and who had persistently abnormal p300 and DNA markers), recurrence correlated with persistent abnormal G-actin findings. Of patients who were G-actin negative following therapy, only 25% recurred during follow-up in contrast to 67% in patients who were positive (p < 0.03 by Fisher's exact test). The odds ratio for recurrence was 6.00 (95% confidence interval: 1.3-28.6). Cytosolic G-actin levels can be an important intermediate end point marker for chemoprevention.

Biomarker risk assessment and bladder cancer detection in a cohort exposed to benzidine

BACKGROUND

Cancer screening with highly sensitive, specific biomarkers that reflect molecular phenotypic alterations is an attractive strategy for cancer control. We examined whether biomarker profiles could be used for risk assessment and cancer detection in a cohort of Chinese workers occupationally exposed to benzidine and at risk for bladder cancer.

METHODS

The cohort consisted of 1788 exposed and 373 nonexposed workers, followed from 1991 through 1997. We assayed urothelial cells from voided urine samples for DNA ploidy (expressed as the 5C-exceeding rate [DNA 5CER]), the bladder tumor-associated antigen p300, and a cytoskeletal protein (G-actin). Workers were stratified into different risk groups (high, moderate, and low risk) at each examination based on a predefined biomarker profile. For workers who developed bladder cancer, tumor risk assessment was analyzed from samples collected 6-12 months before the cancer diagnosis. The associations between risk group and subsequent development of bladder cancer were analyzed by Cox proportional hazards regression analysis and logistic analysis, after adjustment. All statistical tests were two-sided.

RESULTS

Twenty-eight bladder cancers were diagnosed in exposed workers and two in nonexposed workers. For risk assessment, DNA 5CER had 87.5% sensitivity, 86.5% specificity, an odds ratio (OR) of 46.2 (95% confidence interval [CI] = 8.1 to 867.0), and a risk ratio (RR) of 16.2 (95% CI = 7.1 to 37.0); p300 had 50.0% sensitivity, 97.9% specificity, an OR of 40.0 (95% CI = 9.0 to 177.8), and an RR of 37.9 (95% CI = 16.8 to 85.3). The risk of developing bladder cancer was 19.6 (95% CI = 8.0 to 47.9) times higher in workers positive for either the DNA 5CER or p300 biomarkers than in workers negative for both biomarkers and 81.4 (95% CI = 33.3 to 199.3) times higher in workers positive for both biomarkers. G-actin was a poor marker of individual risk.

CONCLUSIONS

Occupationally exposed workers at risk for bladder cancer can be individually stratified, screened, monitored, and diagnosed based on predefined molecular biomarker profiles.

A Fourier-transform infrared spectroscopic comparison of cultured human fibroblast and fibrosarcoma cells: a new method for detection of malignancies

Infrared vibration spectroscopy appears to be a more powerful technique for tumor diagnosis than visible or UV spectroscopy. In the present work, Fourier-transform infrared (FTIR) spectroscopy was used to compare cultured normal fibroblast and fibrosarcoma cells. Significant differences were observed by comparing the spectra of the normal human cells with that of the cancer cells. The PO2 symmetric stretching mode at 1082 cm-1 is shifted to a higher frequency in the cancer cell and a broad band, whose center is located at 1064 cm-1 in the cancer cell is reduced in intensity. In addition, the decrease in intensity of the CH2 bending mode relative to that of CH3 mode is detectable only in the fibrosarcoma cell. This FTIR difference between fibroblast and fibrosarcoma cells suggests that either fatty acid chains or protein side chains of the cancer cells are partially degraded resulting in more terminal carbon (e.g., CH3). It is also possible that changes in the environment upon carcinogenesis induces a change in the relative absorption cross section for CH3 and CH2 bending vibrations. These results suggest that FTIR spectroscopy may become a promising and sensitive technique for tumor identification.

Bladder cancer risk assessment with quantitative fluorescence image analysis of tumor markers in exfoliated bladder cells

BACKGROUND

The detection of potentially highly curable low-grade bladder cancers by noninvasive techniques remains an unsolved problem. Conventional cytology detects such tumors with 50% sensitivity, and addition of DNA measurements to cytology only improves sensitivity incrementally. Tumor-associated antigens potentially offer an additional diagnostic marker.

METHODS

In this study, the M344 antibody against a tumor-associated antigen expressed mainly by low-grade tumor cells was tested for its sensitivity and specificity, alone and in combination with DNA ploidy and cytology. Voided urine samples from 69 asymptomatic control subjects, urines and bladder washings from 59 patients with cancer, and 195 symptomatic control patients were collected. Cells were double-labeled with M344 monoclonal antibody and Hoechst. Each case was blinded, and the number of positive cells was scored by two independent observers.

RESULTS

High-grade and low-grade transitional cell carcinomas (TCC) were detected with equal efficiency (78%, P < 0.001 versus symptomatic control patients). Urine samples proved higher specificity in detecting cancers. Patients being monitored for recurrence, but without current detectable cancer, were intermediates between control subjects and patients with cancer, suggesting that this marker also responds to dysplasia or field disease. Patients with outlet obstruction did not significantly differ from patients with previous TCC (P = 0.95). When combined with DNA ploidy measurements and cytology, the sensitivity for low-grade and high-grade tumors was 88% and 95%, respectively.

CONCLUSIONS

The M344 antibody potentially could improve the specificity and sensitivity of detection of low-grade bladder tumors in symptomatic and asymptomatic patients as well as monitoring for recurrence, therapeutic response, and assessment of individual risk.

Alterations in phenotypic biochemical markers in bladder epithelium during tumorigenesis

Phenotypic biochemical markers of oncogenesis and differentiation were mapped in bladder biopsies to investigate changes that occur in bladder tumorigenesis and to identify markers for increased bladder cancer risk. Touch preparations from biopsy specimens from 30 patients were obtained from tumors, the adjacent bladder epithelium, and random distant bladder epithelium. Markers, including DNA ploidy, epidermal growth factor receptor (EGFR), and oncoproteins, were quantified in individual cells by using quantitative fluorescence image analysis. Cluster analysis revealed the markers fell into three independent groups: (i) G-actin and EGFR; (ii) ploidy, cytology, and p185 (HER-2/neu oncoprotein) (ERBB2); and (iii) p300, a low-grade tumor antigen. Each marker displayed a gradient of abnormality from distant field to adjacent field to tumor. Different patterns for each marker suggested a developmental sequence of bladder cancer oncogenesis; G-actin was altered in 58% of distant biopsies (vs. 0/6 normals, P < 0.001), ploidy and cytology were altered in < 20% of distant fields and approximately 80% of tumors, and the other markers were intermediate. Patterns of EGFR and p185 suggest low-and high-grade tracks diverge early (P < 0.05 by Mann-Whitney U test for EGFR and ANOVA for p185). In conclusion, this study shows that a sequence of phenotypic changes accompanies development and progression of bladder cancers. Biochemical alterations in cells of the bladder field are often detectable before abnormal pathology, and markers previously thought to be limited to tumors were found in the field. The hierarchy of expression may be useful in identifying high-risk patients, assessing completeness of response to therapy, and monitoring and predicting recurrence.

Treated history of noninvasive grade 1 transitional cell carcinoma. The National Bladder Cancer Group

A total of 178 patients with grade 1 noninvasive (stage Ta) bladder tumors followed from 1 to 10 years (median 58 months) was prospectively evaluated by cystoscopy, transurethral resection, mucosal biopsies, cytology, size and number of tumors at diagnosis, recurrences, progression in grade and stage, number of negative or positive cystoscopies and death from all causes. Histopathological and cytological studies were confirmed by a Central Pathology Laboratory using the criteria for grade 1 as described previously. Of the patients 122 (68.5%) had a single tumor. Three-quarters of the patients had tumors of less than 2 cm., 95% had mild or no urothelial dysplasia and 1 had positive cytology results. There were 419 recurrent tumors in 109 patients (61%). Patients with multiple tumors were at a significantly greater risk for recurrences (p < 0.001). Size of tumor significantly affected the rate of recurrence in the first 2 years after initial diagnosis in single tumor patients only. Of the multiple tumor patients 90% experienced a recurrence compared to 46% of the single tumor patients. Of the 1,112 cystoscopies performed in 122 single tumor patients 18% were positive, compared to 33% of the 686 cystoscopies performed in 56 multiple tumor patients. A total of 29 patients had a change in grade, 5 having grade 3 and 24 having grade 2 tumors. Progression to stage T1 occurred in 5 patients and to stage T2 or greater in 3. Of the 36 patients who died, 1 died of obstruction due to bladder cancer. Experimental evidence supports the opinion that the cells of stage Ta, grade 1 tumors are different in several ways from normal urothelium. There are little data to support the use of the term papilloma to describe stage Ta, grade 1 tumors without reservation. The data demonstrate that the tumor diathesis being expressed ceases with time and for unknown reasons. Multiple tumor patients with stage Ta, grade 1 disease might be included in chemotherapy trials only with stratification and a control arm of transurethral resection/fulguration alone.

Intermediate endpoint biomarkers for chemoprevention

The understanding of intermediate endpoint biomarker expression in relation to the sequential events in bladder tumorigenesis establishes a useful approach for evaluating chemopreventive agents. Biomarkers may be genotypic or phenotypic and function as biomarkers of susceptibility, exposure, effect, or disease. This paper reviews several years of research on biomarkers and their use in monitoring chemoprevention therapy. In initial animal experiments, mice were dosed with N-butyl-N-(4-hydroxybutyl)nitrosamine (OH-BBN) while co-administering N-(4-hydroxyphenyl)retinamide (4-HPR). 4-HPR did not statistically reduce tumor incidence, but did affect tumor differentiation and, consequently, nuclear size and DNA ploidy. These results suggest that nuclear size and ploidy may function as intermediate endpoint biomarkers of effect for oncogenesis and that epigenetic as well as genetic mechanisms may be primary in the oncogenic process. Early biomarkers of effect which occur prior to genetic effects or chromosome aberration may portend a higher probability of being modulated by differentiating agents such as retinoids. In vitro studies demonstrated that RPMI-7666 cells cultured with a phorbol ester tumor promoter (12-O-tetradecanoyl-phorbol-13-acetate) could be redifferentiated with 13-cis-retinoic acid and dimethyl sulfoxide (DMSO). F-actin, a cytoskeletal biomarker with a presumed function in the epigenetic mechanisms of carcinogenesis, could also be normalized in HL-60 cells treated with 4-HPR or DMSO. A clinical evaluation of F-actin in patients with varying degrees of risk confirmed the value of F-actin as a differentiating biomarker useful for bladder cancer risk assessment. The clarification of when the phenotypic changes of F-actin occur in the oncogenic process was achieved when a variety of biochemical changes were mapped in the patients with bladder cancer. These studies confirmed that G-actin, a reciprocal form of F-actin, is increased relatively early in bladder cancer oncogenesis when multiple biomarkers are quantitated in the field, adjacent area, and the tumor. Comparison of each individual biomarker's expression from field, adjacent to tumor, and tumor, and subsequent cluster analysis of these biomarkers, indicated that the possible sequence of phenotypic expression of biomarkers in bladder cancer oncogenesis is from G-actin, to p300 antigen, to epidermal growth factor receptor (EGFR), to p185 (neu oncogene product), to DNA aneuploidy and, finally, to visual morphology.(ABSTRACT TRUNCATED AT 400 WORDS)

Identification of a high risk subgroup of grade 1 transitional cell carcinoma using image analysis based deoxyribonucleic acid ploidy analysis of tumor tissue

The use of deoxyribonucleic acid (DNA) cytometry to identify a subset of patients with grade 1, stage Ta or T1 transitional cell carcinoma at high risk for death or recurrence was investigated in a retrospective study using paraffin blocks from 88 low grade transitional cell carcinomas of the bladder with an absorptiometric video-based image analysis system. Tumors were evaluated for ploidy (70 diploid, 16 aneuploid and 2 tetraploid) and the presence of cells with greater than 5C DNA. Survival analysis of 62 patients with adequate followup (15 to 20 years) showed that 43 of 62 (69%) suffered recurrences and 13 (21%) died of bladder cancer. The single most important predictors of death and recurrence were stem line aneuploidy and the presence of cells with greater than 5C DNA, respectively.

Quantitative fluorescence image analysis of deoxyribonucleic acid ploidy in urine from normal children

Quantitative fluorescence image analysis incorporates the 2 diagnostic techniques of cytological analysis with quantitation of deoxyribonucleic acid (DNA). Exfoliated urinary cells are ideal for analysis by this method, which allows the identification of "rare event" abnormal cells. We evaluated the urine from 50 children who had undergone cystoscopy or were catheterized for other reasons. The urine was free of infection by urinalysis. Cytological analysis demonstrated normal or atypical cells in all patients. Of the patients 1 (2%) had greater than 2 of 500 cells analyzed with greater than 5C DNA and 4 (8%) had greater than 2 of 500 cells with greater than 5C double stranded nucleic acid. These data suggest that it may be "normal" for urine to contain "rare event" abnormal cells. The significance of this finding is unclear at present.