Progression from Prediabetes to Diabetes in a Diverse U.S. Population: A Machine Learning Model.

Objective: To date, there are no widely implemented machine learning (ML) models that predict progression from prediabetes to diabetes. Addressing this knowledge gap would aid in identifying at-risk patients within this heterogeneous population who may benefit from targeted treatment and management in order to preserve glucose metabolism and prevent adverse outcomes. The objective of this study was to utilize readily available laboratory data to train and test the performance of ML-based predictive risk models for progression from prediabetes to diabetes. Methods: The study population was composed of laboratory information services data procured from a large U.S. outpatient laboratory network. The retrospective dataset was composed of 15,029 adults over a 5-year period with initial hemoglobin A1C (A1C) values between 5.0% and 6.4%. ML models were developed using random forest survival methods. The ground truth outcome was progression to A1C values indicative of diabetes (i.e., ≥6.5%) within 5 years. Results: The prediabetes risk classifier model accurately predicted A1C ≥6.5% within 5 years and achieved an area under the receiver-operator characteristic curve of 0.87. The most important predictors of progression from prediabetes to diabetes were initial A1C, initial serum glucose, A1C slope, serum glucose slope, initial HDL, HDL slope, age, and sex. Conclusions: Leveraging readily obtainable laboratory data, our ML risk classifier accurately predicts elevation in A1C associated with progression from prediabetes to diabetes. Although prospective studies are warranted, the results support the clinical utility of the model to improve timely recognition, risk stratification, and optimal management for patients with prediabetes.

CKD Progression Prediction in a Diverse US Population: A Machine-Learning Model.

Rationale & Objective: Chronic kidney disease (CKD) is a major cause of morbidity and mortality. To date, there are no widely used machine-learning models that can predict progressive CKD across the entire disease spectrum, including the earliest stages. The objective of this study was to use readily available demographic and laboratory data from Sonic Healthcare USA laboratories to train and test the performance of machine learning-based predictive risk models for CKD progression. Study Design: Retrospective observational study. Setting & Participants: The study population was composed of deidentified laboratory information services data procured from a large US outpatient laboratory network. The retrospective data set included 110,264 adult patients over a 5-year period with initial estimated glomerular filtration rate (eGFR) values between 15-89 mL/min/1.73 m2. Predictors: Patient demographic and laboratory characteristics. Outcomes: Accelerated (ie, >30%) eGFR decline associated with CKD progression within 5 years. Analytical Approach: Machine-learning models were developed using random forest survival methods, with laboratory-based risk factors analyzed as potential predictors of significant eGFR decline. Results: The 7-variable risk classifier model accurately predicted an eGFR decline of >30% within 5 years and achieved an area under the curve receiver-operator characteristic of 0.85. The most important predictor of progressive decline in kidney function was the eGFR slope. Other key contributors to the model included initial eGFR, urine albumin-creatinine ratio, serum albumin (initial and slope), age, and sex. Limitations: The cohort study did not evaluate the role of clinical variables (eg, blood pressure) on the performance of the model. Conclusions: Our progressive CKD classifier accurately predicts significant eGFR decline in patients with early, mid, and advanced disease using readily obtainable laboratory data. Although prospective studies are warranted, our results support the clinical utility of the model to improve timely recognition and optimal management for patients at risk for CKD progression. Plain-Language Summary: Defined by a significant decrease in estimated glomerular filtration rate (eGFR), chronic kidney disease (CKD) progression is strongly associated with kidney failure. However, to date, there are no broadly used resources that can predict this clinically significant event. Using machine-learning techniques on a diverse US population, this cohort study aimed to address this deficiency and found that a 5-year risk prediction model for CKD progression was accurate. The most important predictor of progressive decline in kidney function was the eGFR slope, followed by the urine albumin-creatinine ratio and serum albumin slope. Although further study is warranted, the results showed that a machine-learning model using readily obtainable laboratory information accurately predicts CKD progression, which may inform clinical diagnosis and management for this at-risk population.

Impact of Laboratory Charge Display Within the Electronic Health Record Across an Entire Academic Medical Center: Results of a Randomized Controlled Trial.

OBJECTIVES: To determine the impact of systemwide charge display on laboratory utilization. METHODS: This was a randomized controlled trial with a baseline period and an intervention period. Tests were randomized to a control arm or an active arm. The maximum allowable Medicare reimbursement rate was displayed for tests in the active arm during the intervention period. Total volume of tests in the active arm was compared with those in the control arm. Residents were surveyed before and after the intervention to assess charge awareness. RESULTS: Charge display had no effect on order behavior. This result held for patient type (inpatient vs outpatient) and for insurance category (commercial, government, self-pay). Residents overestimated the charges of tests both before and after the intervention. Many residents failed to notice the charge display in the computerized order entry system. CONCLUSIONS: The impact of charge display depends on context. Charge display is not always effective.

Impact of Continuous Improvement of Laboratory Test Result Comments on Requests for Consultation: A Case Series.

OBJECTIVES: Test interpretation is an important part of the brain-to-brain loop. Poor test comments can lead to calls for explanations and impair test interpretation and possibly patient care. There has been little study on quality improvement of test result comments. The objective of this study was to investigate the impact of improvements in test comments on requests for consultation by clinicians. METHODS: We monitored the volume of requests for consultation regarding test results before and after a test comment was modified. RESULTS: Modifications of test comments eliminated calls for three different tests. Reductions were statistically significant for all tests (P < .007). CONCLUSIONS: Quality improvement activities targeted at test comments can improve service, reduce costs, and improve patient care.

Cutaneous T-cell lymphoma (CTCL): Current practices in blood assessment and the utility of T-cell receptor (TCR)-Vß chain restriction.

BACKGROUND: Accurate quantification of malignant cells in the peripheral blood of patients with cutaneous T-cell lymphoma is important for early detection, prognosis, and monitoring disease burden. OBJECTIVE: We sought to determine the spectrum of current clinical practices; critically evaluate elements of current International Society for Cutaneous Lymphomas (ISCL) B1 and B2 staging criteria; and assess the potential role of T-cell receptor-Vß analysis by flow cytometry. METHODS: We assessed current clinical practices by survey, and performed a retrospective analysis of 161 patients evaluated at Yale (2011-2014) to compare the sensitivity, specificity, positive predictive value, and negative predictive value of parameters for ISCL B2 staging. RESULTS: There was heterogeneity in clinical practices among institutions. ISCL B1 criteria did not capture 5 Yale cohort cases with immunophenotypic abnormalities that later progressed. T-cell receptor-Vß testing was more specific than polymerase chain reaction and aided diagnosis in detecting clonality, but was of limited benefit in quantification of tumor burden. LIMITATIONS: Because of limited follow-up involving a single center, further investigation will be necessary to conclude whether our proposed diagnostic algorithm is of general clinical benefit. CONCLUSION: We propose further study of modified B1 criteria: CD4/CD8 ratio 5 or greater, %CD4(+) CD26(-) 20% or greater, or %CD4(+) CD7(-) 20% or greater, with evidence of clonality. T-cell receptor-Vß testing should be considered in future diagnostic and staging algorithms.

Neoplastic plasma cell aberrant antigen expression patterns and their association with genetic abnormalities.

Little is known about aberrant antigen expression patterns and their association with cytogenetic aberrations in multiple myeloma (MM). We examined the correlation between flow cytometry and florescence in situ hybridization (FISH) in 167 marrow specimens with MM. Gene expression profiling of CD56, CD117, CD52 and CD20 mRNA in plasma cells (PCs) from patients treated on Total Therapy 2 and Total Therapy 3 trials were also evaluated. Higher expression of CD56 and CD117 was associated with hyperdiploidy. High CD52 mRNA expression was associated with c-MAF and FGFR3 subgroups. Higher expression of CD56 mRNA, but lower Kit expression, were noted in association with FGFR3. In contrast, the c-MAF subgroup showed high Kit expression but lacked NCAM mRNA expression. CKS1B amplification showed positive correlation with CD52 (p=0.0065) but negative correlation with CD20 (p=0.0207). These findings indicate that phenotypic differences in MM are associated with distinct genetic subgroups, which potentially has important diagnostic and prognostic value.

A limited plasma cell flow cytometry panel with reflex CD138 immunohistochemistry is an optimal workflow process for evaluating plasma cell neoplasms in bone marrow specimens.

OBJECTIVES: To determine the optimal workflow combination of flow cytometry (FC) and immunohistochemistry tests for efficient and cost-effective evaluation of plasma cell (PC) neoplasms (PCNs) in bone marrow (BM) specimens. METHODS: Various workflow combinations of immunohistochemistry and FC for 4,031 BM specimens worked up for PCNs were compared. Turnaround time (TAT), immunohistochemistry usage, technical charges, and addendum/amendment rates were compared between periods to determine the optimal workflow combination. RESULTS: Five distinct workflow periods were identified, with varying combinations of full or limited FC panels for assessing PC clonality and CD138/κ/λ immunohistochemistry for PC quantification. Replacement of full FC with limited FC was associated with significant reductions in TAT and number of immunostains performed per case. Elimination of immunohistochemistry on cases determined to be polyclonal by FC also resulted in significant reductions in immunohistochemistry usage and significant cost savings. CONCLUSIONS: Assessment of PC clonality using a limited FC panel followed by reflex CD138 immunohistochemistry on cases that are monoclonal by FC provides an optimal and cost-effective workflow for evaluating PCNs in BM samples.

Web-based oil immersion whole slide imaging increases efficiency and clinical team satisfaction in hematopathology tumor board.

BACKGROUND: Whole slide imaging (WSI) is widely used for education and research, but is increasingly being used to streamline clinical workflow. We present our experience with regard to satisfaction and time utilization using oil immersion WSI for presentation of blood/marrow aspirate smears, core biopsies, and tissue sections in hematology/oncology tumor board/treatment planning conferences (TPC). METHODS: Lymph nodes and bone marrow core biopsies were scanned at ×20 magnification and blood/marrow smears at 83X under oil immersion and uploaded to an online library with areas of interest to be displayed annotated digitally via web browser. Pathologist time required to prepare slides for scanning was compared to that required to prepare for microscope projection (MP). Time required to present cases during TPC was also compared. A 10-point evaluation survey was used to assess clinician satisfaction with each presentation method. RESULTS: There was no significant difference in hematopathologist preparation time between WSI and MP. However, presentation time was significantly less for WSI compared to MP as selection and annotation of slides was done prior to TPC with WSI, enabling more efficient use of TPC presentation time. Survey results showed a significant increase in satisfaction by clinical attendees with regard to image quality, efficiency of presentation of pertinent findings, aid in clinical decision-making, and overall satisfaction regarding pathology presentation. A majority of respondents also noted decreased motion sickness with WSI. CONCLUSIONS: Whole slide imaging, particularly with the ability to use oil scanning, provides higher quality images compared to MP and significantly increases clinician satisfaction. WSI streamlines preparation for TPC by permitting prior slide selection, resulting in greater efficiency during TPC presentation.

An analysis of clinical consultation activities in clinical pathology: who requests help and why.

OBJECTIVES: To examine the distribution of callers who made consultation requests and to identify associations between caller categories and consultation topics. METHODS: Review of prospectively collected database of consultations. RESULTS: Direct care personnel made more consultation requests than non-direct care personnel. Consultation topics varied by caller type. Direct care personnel requested more consultations on test interpretation and few consultations on test selection than laboratory personnel. Differences in consultation requests by primary care physicians and specialists were significant. CONCLUSIONS: At our laboratory, consultation requests primarily originate from primary care physicians. Consultation requests vary by caller type.

An analysis of clinical consultation activities in clinical chemistry: implications for transformation and resident training in chemical pathology.

CONTEXT: Clinical consultation is a key role of pathologists. Many have advocated that pathologists expand their consulting activities to improve laboratory utilization. Although many have suggested that residency programs need to provide experience in clinical consultation, little has been written on the nature of consultation or on the methods of training. OBJECTIVE: To characterize the content of consultations and to describe training in consultation in chemical pathology within the residency program at the University of Utah, Salt Lake City. DESIGN: Retrospective review of the consultation database for the period between July 2011 and July 2012. RESULTS: Residents performed an average of 159 consultations a month covering 276 topics during the course of a year. Each topic involved 1 or more specific tests. Eighty percent of the topics received fewer than 12 calls. The most common topics involved virus testing (eg, hepatitis B virus, hepatitis C virus, human immunodeficiency virus). Consultations most often involved test interpretation (53%), selection (38%), and performance characteristics (21%). Twenty-seven percent of consultations involved 2 or more consultation categories (eg, interpretation and performance). CONCLUSIONS: Consultation calls in chemical pathology are widely distributed across topics. Consultations most often involve test interpretation and selection. Methods to assess the effectiveness of consultations and resident teaching should be devised.

Primary intracerebral angiomatoid fibrous histiocytoma: report of a case with a t(12;22)(q13;q12) causing type 1 fusion of the EWS and ATF-1 genes.

Angiomatoid fibrous histiocytoma (AFH) is generally considered a soft tissue sarcoma of low malignant potential that occurs in children/young adults and most frequently affects the extremities. AFH infrequently recurs and rarely metastasizes. AFH has a characteristic histomorphology, and immunohistochemical reactivities for desmin and CD68 have led to myofibroblastic and fibrohistiocytic histogenetic hypotheses, respectively. Although only a limited number of AFH cases have been molecularly characterized, many have demonstrated evidence of an underlying translocation event. Reverse transcription-polymerase chain reaction and fluorescence in situ hybridization studies suggest that chromosomal rearrangement in AFH most frequently involve the EWS, CREB, ATF-1, and FUS genes. We report the first pathologically confirmed case of an AFH presenting as an intracerebral primary in a previously healthy 25-year-old man. Genetic analyses revealed a t(12;22)(q13;q12) and a unique underlying clear cell sarcomalike type 1 EWS/ATF-1 gene fusion.

The role of Janus Kinase 2 V617F mutation in extramedullary hematopoiesis of the spleen in neoplastic myeloid disorders.

Extramedullary hematopoiesis (EMH) in the spleen is a characteristic feature of the chronic myeloproliferative disorders (CMPDs) and various other neoplastic or reactive myeloid conditions. However, the origin of these hematopoietic precursor cells and the molecular mechanisms underlying their development in the spleen is uncertain. The V617F mutation in the Janus Kinase 2 gene (JAK2(V617F)) was recently shown to be frequently and preferentially present in the peripheral blood and bone marrow cells of CMPD patients, and the resulting dysregulation of its downstream targets is important to CMPD pathogenesis. To determine the occurrence and potential role of JAK2(V617F) in splenic EMH cells, we studied splenectomy specimens from 47 patients with significant EMH. JAK2(V617F) was detected by real-time PCR melting curve analysis in 22 specimens, including 11/17 chronic idiopathic myelofibrosis, 7/7 polycythemia vera, 1/1 essential thrombocythemia, 1/3 CMPD unclassifiable, 1/5 chronic myelomonocytic leukemia, 0/5 chronic myelogenous leukemia, 1/3 myelodysplastic syndrome and 0/6 acute myeloblastic leukemia cases, whereas only the JAK2 wild-type allele was detected in the other 25. Nineteen of 20 cases with adequate bone marrow samples available for molecular examination demonstrated concordant JAK2 genotypes. Laser-capture microdissection was then used to enrich the EMH and non-EMH splenic cell fractions, confirming that the mutant alleles specifically originated from the EMH cells. Furthermore, megakaryocytes in the JAK2(V617F)-positive splenectomy specimens expressed higher levels of Bcl-xL, an antiapoptotic protein and downstream target of the JAK2/STAT5 pathway. Thus, JAK2(V617F) is frequently present in splenic EMH cells associated with CMPD, but it is rarely identified in splenic EMH cells associated with other myeloid disorders. Our results indicate that the precursor cells leading to extramedullary hematopoietic expansion in CMPD most likely originate from the transformed bone marrow clone. Also, dysregulation of downstream pathways such as Bcl-xL may be important to CMPD disease pathogenesis in the spleen.

Expression of apoptosis-related antigens, Fas, bcl-2, and p53, and Mib-1 proliferation index in the hypoplastic thymus of DiGeorge syndrome.

Apoptosis, along with cellular proliferation, plays a major role in normal developmental processes and tissue homeostasis. We hypothesized that altered apoptosis-related pathways and/or reduced cell proliferation might play a role in the thymic hypoplasia or aplasia in DiGeorge syndrome (DG). We used immunohistochemistry to evaluate the apoptosis-related antigens Fas (CD95), bcl-2, and p53, as well as Mib-1 proliferation index in the thymuses from six patients with DG. The results were compared with those from the thymuses from six patients with non-DG congenital heart disease. All DG patients (age 32 weeks GA to 4 months) had hypoplastic thymuses ranging from microscopic foci to 2.7 g in weight (expected for age, 4.7 +/- 3.6 g to 10 +/- 6 g). The thymic weights from the patients with non-DG congenital heart disease (age 37 weeks GA to 1 month) ranged from 3 to 5.6 g and were at the lower range of expected weight by age (expected for age, 8.4 +/- 5.6 g to 12 +/- 7 g). All thymuses showed histologic features of stress-induced involution. In both groups, a - 50% Mib-1 proliferation index was found in the cortical thymocytes, whereas <5% Mib-1 labeling was seen in the medullary thymocytes; Fas stained medullary epithelial cells (3+) and cortical epithelial cells (1+); bcl-2 stained medullary thymocytes (3+) and cortical thymocytes (1+); p53 stained less than 1% of nuclei in all sections. No significantly altered Mib-1 proliferation index or expression of Fas, bcl-2, and p53 was observed in the hypoplastic thymuses in DG, compared to these same measures in non-DG. These results suggest that thymic hypoplasia in DG may be mediated by mechanisms other than reduced cellular proliferation and/or altered Fas, bcl-2, and p53 apoptotic pathways.