Selective deployment of dynamic telecytology for rapid evaluation of cytology smears: assessment of workflow processes and role of cytopathology fellows as on-site operators.

INTRODUCTION: The deployment of telecytology (TC) requires a substantial investment of financial and human resources. To offset the high demand for rapid on-site evaluation, we performed a limited deployment of dynamic TC and have detailed the workflow processes and the role of trainees. MATERIALS AND METHODS: TC systems were installed in radiology suites with a high volume of cases. Validation was performed using retrospective and prospective cases. Cytotechnologists and cytopathology fellows were the operators of the instrument. TC malignant and benign diagnoses were correlated with the final sign-out diagnoses. RESULTS: Of the 120 cases, 50 (41.6%) were fine needle aspirations and 70 (58.3%) were touch imprint smears of core biopsy specimens. The cytotechnologists were the operators for 34 cases (28.3%) and cytology fellows for 86 cases (71.6%). Adequacy concordance with the final diagnosis was 100% and 98.5% in the retrospective and prospective cases, respectively. In the prospective cases, concordance of TC with the final diagnosis of malignancy was 42 of 45 (93.3%), with 2 of 45 (4.4%) discordant and a downgrade rate of 2.7%. For the benign diagnoses, the concordance was 90%. For the malignant diagnoses, the sensitivity of TC was 97.67% (95% confidence interval [CI], 87.71 to 99.94%; specificity, 81.82%; 95% CI, 48.22% to 97.72%). The positive predictive value was 95.45% (95% CI, 85.69% to 98.66%), the negative predictive value was 90.00% (95% CI, 55.98% to 98.45%), and the accuracy was 94.44% (95% CI, 84.61% to 98.84%). CONCLUSIONS: TC can be deployed in a limited fashion as an option for cytopathologists to offset the high demand for rapid on-site evaluations. Trainee participation in TC service is important for building confidence and honing their cytology skills.

The manifestations of politics in evaluation: An exploratory study across the evaluation process.

Evaluation has been described as a political act. Programs and policies are generated from a political process, and the decision to evaluate and how to use the evaluation are manifestations of the political dynamic. This exploratory study was conducted with practicing evaluators to understand what they view as political situations in the evaluation process and how they responded to these situations. Findings suggest that, in relation to the potential evaluation phases in which each respondent has been involved, evaluations are susceptible to politics when initially attempting to identify stakeholders and when it's time to report the evaluation findings. Evaluators have also developed multiple strategies for dealing with these situations, including finding allies for the evaluation and working to explain the evaluation process and its implications. We hope that this study will help to inform novice and expert evaluators about the various political situations they may encounter in their practice.

The role of titles in enhancing data visualization.

Many in the data visualization and evaluation communities recommend conveying the message or takeaway of the visualization in the visualization's title. This study tested that recommendation by examining how informative or generic titles impact a visualization's visual efficiency, aesthetics, credibility, and the perceived effectiveness of the hypothetical program examined. Furthermore, this study tested how simple or complex graphs, and positive, negative, or mixed results (i.e., valence of the results) affected outcomes. Participants were randomly assigned to one of 12 conditions, representing a 2 (graph: simple or complex) x 2 (title: generic or informative) x 3 (valence: positive, negative, mixed) between-subjects study. The results indicated that informative titles required less mental effort and were viewed as more aesthetically pleasing, but otherwise did not lead to greater accuracy, credibility, or perceived effectiveness. Furthermore, titles did not interact with graph type or the valence of the findings. While the results suggest it is worthwhile to consider adding an informative title to data visualizations as they can reduce mental effort for the viewer, the intended goal of the visualization should be taken into consideration. Considering the goal of the visualization can be a deciding factor of the type of graph and title that will best serve its intended purposes. Overall, this suggests that data visualization recommendations that impact evaluation reporting practices should be scrutinized more closely through research.

Role of cytotechnologists in rapid onsite adequacy assessment of cytology materials for diagnostic workup and specimen allocation for ancillary testing using a standardized protocol.

INTRODUCTION: Data on the performance of cytotechnologists in assessing specimen adequacy of needle core biopsies (NCB) is scant and their role in specimen triaging for ancillary studies have not been well established. MATERIALS AND METHODS: We retrospectively analyzed rapid onsite evaluation (ROSE) performed exclusively by cytotechnologists on 248 NCB and fine-needle aspiration (FNA) specimens. Overall adequacy and accuracy rates were determined by comparing to final diagnosis. We also reviewed the process of specimen allocation for ancillary testing to determine whether specimens were appropriately triaged at the time of ROSE. RESULTS: Of the 248 cases, 222 (89.5%) were touch imprint and 26 (10.5%) were FNA smears. The overall adequacy rate was 73.4% (182 of 248). Concordance for "adequate" interpretation by ROSE with unequivocal malignant or benign diagnoses on final interpretation was 95.6%. The sensitivity, specificity, and accuracy of ROSE for a final "positive for malignancy" were 89.2% (95% CI 83.04% to 93.69%), 43.24% (95% CI 31.77% to 55.28%), and 73.87% (95% CI 67.57% to 55.28%), respectively. Cases with "positive for malignancy" on final diagnosis were "adequate" by ROSE in 89.1% (132 of 148) and "inadequate" in 10.8% (16 of 148), P < 0.0001. Ancillary tests were performed in 168 of 248 (67.7%); the majority were immunohistochemical stains for determining tumor subtype. Predictive biomarkers were performed successfully in 100% of metastatic breast cancers. CONCLUSIONS: Cytotechnologists performed at a high level of competency in providing ROSE and allocating specimens for ancillary testing, which were performed successfully in the majority of cases. Implementation of a standardized protocol for tissue management/prioritization is of paramount importance to maximize tissue preservation and minimize wastage.

A subset of pancreatic adenocarcinomas demonstrates coamplification of topoisomerase IIalpha and HER2/neu: use of immunolabeling and multicolor FISH for potential patient screening andtreatment.

We sought to identify the frequency of amplification of the topoisomerase IIalpha gene (TOP2A) in pancreatic cancer and determine the usefulness of TOP2A immunolabeling in screening for TOP2A and human epidermal growth factor receptor (HER)2/neu amplification. We examined 55 pancreatic adenocarcinoma specimens for TOP2A immunolabeling and identified TOP2A protein expression in all specimens with a nuclear labeling index (NLI; positive nuclei/total nuclei x 100) of 5% to 80%. Normal pancreatic ductal epithelium, proposed to give rise to pancreatic adenocarcinoma, did not demonstrate detectable TOP2A expression. In a subset of specimens selected for fluorescence in situ hybridization analysis of TOP2A and HER2/neu amplification using a recently developed multicolor probe, 7 of 8 lesions with an NLI of 25% or more demonstrated TOP2A amplification, in contrast with 2 of 14 lesions with a TOP2A NLI of less than 25%. In 8 of 9 TOP2A-amplified cases, coamplification of HER2/neu was present, suggesting a potential relationship between TOP2A and HER2/neu in pancreatic adenocarcinoma. We propose that TOP2A immunolabeling be used in conjunction with a newly developed multicolor probe to screen patients with pancreatic adenocarcinoma to determine the best potential therapeutic modalities, such as TOP2A inhibitors, trastuzumab, or both.