Objective Melanoma metastasis to the brain is associated with poor prognosis. We sought to determine patient demographics and primary tumor factors associated with development of brain metastasis (BM) and survival. We also investigated whether the BM detection setting (routine screening vs. symptomatic presentation) affected clinical outcomes. Methods A database of melanoma patients seen from 1999-2015 at our institution was reviewed to identify patients that developed BM. Patients with BM were matched by initial stage with patients who did not develop BM as a control group. Patient demographics, primary tumor characteristics, and clinical outcomes were analyzed. Results 123 patients with BM were matched by initial presenting stage to 237 patients without BM. Characteristics of the primary melanoma tumor associated with BM development included location on the scalp (P=0.030), nodular histologic type (P=0.020) and Breslow depth >4mm (P=0.048), while location on the leg was associated with decreased BM risk (P=0.006). In patients with BM, time to first recurrence for melanomas of the scalp was significantly shorter (10.8 vs. 24.8 months, p=0.007) than non-scalp head and neck tumors. Patient stage, tumor depth, nodular type, and ulceration were also associated with worse clinical outcomes. There were no differences in clinical outcomes between patients whose BM were detected upon routine screening vs. upon symptomatic presentation. Conclusions Factors predictive of developing BM included primary scalp location, nodular type, and depth. In BM patients, scalp location, stage, tumor depth, nodular type and ulceration, but not detection setting, were associated with worse clinical outcomes.
Introduction:Valuable research data is limited in its use when it is unstructured and not stored in discrete meaningful fields. Reports of the bone marrow (BM) aspirate and biopsies performed in patients with suspected or confirmed myeloid neoplasms typically include blood counts, peripheral blood (PB) and BM aspirate/touch preparation differential counts, morphological interpretation of aspirate and core biopsy and ancillary data such as karyotype, fluorescent in situ hybridization (FISH) and molecular mutations. Final BM reports are typically reported in a semi-structured document that are sufficient for a single patient review but inadequate for large scale queries to identify patients with a specific diagnosis or capture important diagnostic data. Manual extraction of these fields is expensive, time consuming and error prone. The aim of this study is to develop a customized algorithm for automated extraction of data from bone marrow biopsy reports and generate a framework that allows us to perform large-scale queries. Methods:We randomly identified 148 patients with a diagnosis of a myeloid neoplasm: chronic myeloid leukemia (n=45), chronic myelomonocytic leukemia (n=54) and acute myeloid leukemia (n=57). Seven patients included in this analysis were initially diagnosed as CMML and subsequently transformed to acute myeloid leukemia. Total number of reports evaluated was 524. Numerical and text diagnostic data were extracted manually from the entire cohort selected, which is considered a gold standard. A customized rule based algorithm was developed for each data attribute using Natural Language Processing (I2E Text Mining platform, Linguamatics Ltd, Cambridge, UK). Numerical data captured included differential counts from peripheral blood, bone marrow aspirate or touch preparation. Diagnostic data was captured as included diagnostic interpretation of peripheral blood smear and bone marrow aspirate. The algorithms for extracting the data were previously trained on a separate cohort. Precision and recall calculated for each data attribute utilizing R programing language and statistical computing environment. The calculation of precision can be defined as an index to measure the accuracy or closeness of a measured value to a known value (also known as positive predictive value). Recall can be defined as a measure of ability to capture all data points of interest (true positive rate or sensitivity). F-measure combines precision and recall as a harmonic mean. Results:Overall accuracy for the data captured was precision n = 0.9117 and recall n =0.7951. Precision and recall values for numerical and text data is reported in Table 1 and Figure 1. Conclusion:Extraction of relevant diagnostic data from unstructured bone marrow biopsy reports through automated approach is feasible and accurate. This method saves time and can be utilized for automated extraction of unstructured pathology reports from patients with different hematologic malignancies. Capturing data and storing in structured formats will allow researchers to perform large-scale queries. At the Huntsman Cancer Institute, this data is stored in easily accessible database and linked to other databases such as tissue banking. This approach will allow physicians and translational researchers to find samples with specific diagnosis or molecular mutation, for example identifying AML patients with mutated FLT3 gene. Data on extraction of karyotype, FISH and molecular mutations is being analyzed for accuracy and will be presented at the meeting. Future work involves identifying and improving accuracy and expanding the algorithms to extract additional fields in bone marrow biopsies and apply these algorithms to other hematologic malignancies. Disclosures Deininger: Blueprint: Consultancy; Pfizer: Consultancy, Membership on an entity's Board of Directors or advisory committees.
Introduction:Chronic myelomonocytic leukemia (CMML) is a genetically heterogeneous myeloid neoplasm characterized by the presence of both dysplastic and proliferative features and highly variable clinical outcome. A CMML specific prognostic system (CPSS) has been developed that stratifies patients in to low, intermediate and high risk groups based on WHO subtype, FAB subtype, transfusion dependent anemia, and karyotype. Somatic mutations and DNA methylation patterns can increase prognostic precision, but fail to explain a large part of the clinical variation, suggesting that additional variables, including comorbidities, may be major determinants of overall survival (OS) in CMML. Methods : We retrospectively identified CMML patients diagnosed between 1996 and 2017 at the Huntsman Cancer Hospital, University of Utah, using ICD codes, tumor registry data and electronic medical records. For all patients a diagnosis of CMML was confirmed based on 2008 WHO diagnostic criteria. Data on comorbidities at the time of diagnosis were obtained by search of electronic medical records using a customized rule based algorithm utilizing linguimatics text mining software (Natural language processing). The comorbidities were scored and categorized as per previously published reports: low, intermediate and high risk groups for MDS Comorbidity Index (MDS-CI) and low, mild, moderate/high (moderate and high included in the same group due to small number of patients) for the Charlson Comorbidity Index (CCI). Continuous variables were transformed into categorical variables, based on cutoffs used in previously published studies. Univariate analysis was performed using the Cox proportional hazards model for categories: MDS-CI (low, intermediate and high) and CCI (low, mild, moderate/high). Other variables analyzed included age (<70 or >70 years), sex (male or female), hemoglobin (<10 gm/dL or >10 gm/dL), platelet count (<100k/uL or >100k/uL), WHO subtype (CMML-0, CMML-1 and CMML-2), FAB subtype (CMML-MD or CMML-MP), karyotype (low, intermediate and high risk) and treatment with hypomethylating agents (yes or no). Kaplan-Meier methods were used for plotting OS. All analysis was performed using R statistical programming software version 3.2.1 (The R Foundation for Statistical Computing, Vienna, Austria). Results shown are censored at the time of allogeneic stem cell transplant. For OS the "Low" category is reference and the p-values are for comparison to this category using the Cox model. Results : We identified 94 patients with confirmed diagnosis of CMML. The median age was 76 (range 33-91 years) and 61 patients were men (65%). Fifty-five (58.5%), 34 (36.2%) and 5 (5.3%) patients were categorized as MDS-CI low, intermediate and high risk respectively. Sixty-two (66%), 26 (27.6%) and 6 (6.4%) were categorized as low, mild and moderate/high CCI risk. Hazard ratios (HR) for MDS-CI risk categories were: intermediate=1.26 (95% CI 0.71 to 2.23; p=0.425) and high risk=2.22 (95% CI 0.86-5.75); p=0.101). HR for CCI risk categories were: mild=1.01 (95% CI 0.56-1.82; p=0.964) moderate/high=4.18 (95% CI 1.57 to 11.10; p=0.004). HR for other variables are shown in Table 1. Kaplan-Meier curve representing the OS of the entire cohort categorized according to CCI and MDS-CI risk categoriesis shown in Figure 1. Estimated median survival for MDS-CI low, intermediate and high is 36, 36, and 23 months respectively. Median survival for CCI-CI low, mild, moderate/high risk categories was 36, 33, and 10 months respectively (Figure 1). Conclusions: High risk CCI and MDS-CI category patients are at markedly higher risk of death, suggesting that co-morbidities are major host-related determinant of OS in CMML. Given the association of clonal hematopoiesis of indeterminate potential (CHIP) with coronary heart disease (Jaiswal et al. N Engl J Med 2017; 377:111-121) and the fact that CHIP genes such as TET2are frequently mutated in CMML, it is conceivable that CMML causally contributes to comorbidities. Somatic mutation data are being collected for inclusion in a multivariate model that will be presented at the conference. Disclosures Shami: JSK Therapeutics: Employment, Equity Ownership, Membership on an entity's Board of Directors or advisory committees; Lone Star Biotherapies: Equity Ownership; Pfizer: Consultancy; Baston Biologics Company: Membership on an entity's Board of Directors or advisory committees. Kovacsovics:Abbvie: Research Funding; Amgen: Honoraria, Research Funding. Deininger:Pfizer: Consultancy, Membership on an entity's Board of Directors or advisory committees; Blueprint: Consultancy.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.