Citations

PURPOSE Recurrence after curative-intent treatment occurs in 20%-50% of patients with stage II-IV colorectal cancer (CRC), underscoring the need for early detection of minimal residual disease (MRD) using circulating tumor DNA (ctDNA). Here, we examined the pattern of use of a tumor-informed ctDNA assay in CRC MRD monitoring in routine clinical practice at Mayo Clinic, Rochester. METHODS We conducted a retrospective analysis of health records of patients with CRC who had at least one tumor-informed ctDNA assay from May 2019 through July 1, 2022. Recurrence was defined as radiographic evidence of disease. Descriptive characteristics of the cohort, ctDNA results, and subsequent interventions were recorded. RESULTS Of the 120 patients included, the median age at diagnosis was 67 years, 46% were female, and 94% were White. At diagnosis, 10 patients had stage I, 23 stage II, 60 stage III, and 25 stage IV disease. Of 476 ctDNA assays performed, 70% were performed in patients who had recurrent disease most commonly to monitor the effectiveness of therapeutic interventions and 16% resulted in a change in clinical decision making. There were 110 recurrences identified in 62 patients, as some patients experienced more than one recurrence over time. Compared with serum carcinoembryonic antigen levels, ctDNA results correlated better with radiologic imaging. CONCLUSION Routine ctDNA monitoring for MRD detection has been adopted in clinical practice; however, 84% of ctDNA assays performed did not result in a change in clinical management. This suggests the need for further clinical research data to guide routine clinical use of ctDNA MRD testing in CRC.

PURPOSE Multidisciplinary molecular tumor boards (MTBs) decode complex genomic data into clinical recommendations. Although MTBs are well-established in the oncology practice in developed countries, this strategy needs to be better explored in developing countries. Herein, we describe the possible benefits and limitations of the first MTB established in Colombia. METHODS Demographic, clinical, and genomic information was collected between August 2020 and November 2021. By mid-2020, an MTB strategy was created to discuss clinical cases with one or more genomic alterations identified by next-generation sequencing using an open-access virtual platform. We characterized the patient population as benefiting from the recommended treatment option. We assessed the benefits and access to available targeted therapies that have the potential to change clinical management by making recommendations to treating oncologists on the basis of genomic profiling. However, we did not assess the treatment oncologists' compliance with MTB recommendations because they were not intended to replace clinical judgment/standard of care. RESULTS A total of 146 patients were included in the discussions of the MTB. The median age was 59 years, and 59.6% were women. Genomic results prompting a change in therapeutic decisions were obtained in 53.1% of patients (95% CI, 44.9 to 61.3). The most prevalent malignancy was non–small-cell lung cancer (51%). Other malignancies represented 60%, 50%, and 30% of patients with soft-tissue sarcomas, brain tumors, and breast cancer, respectively. CONCLUSION Using an open-access virtual platform, MTBs were feasible in low- and middle-income countries on the basis of the capability to provide the benefits and access to available targeted therapies that are not standard of care. Furthermore, MTB recommendations were made available to the treating oncologist in different locations across Colombia, providing the option to modify clinical management in most of these patients.

Abstract Objective The aim of this study is to determine if supervised machine learning algorithms can accurately predict cancelled and self-intercepted computerized physician order entry in oral and maxillofacial surgery inpatients. Methods Data from Electronic Medical Record included patient demographics, comorbidities, procedures, vital signs, laboratory values, and medication orders were retrospectively collected. Predictor variables included patient demographics, comorbidities, procedures, vital signs, and laboratory values. Outcome of interest is if a medication order was voided or not. Data was cleaned and pro1cessed using Microsoft Excel and Python v3.12. Gradient Boosted Decision Trees, Random Forest, K-Nearest Neighbor, and Naïve Bayes were trained, validated, and tested for accuracy of the prediction of voided medication orders. Results 37,546 medication orders from 1,204 patient admissions over 5 years were used for this study included 3,892 (10.4%) medication orders that were voided. Gradient Boosted Decision Trees, Random Forest, K-Nearest Neighbor, and Naïve Bayes had an Area Under the Receiver Operating Curve of 0.802 802 with 95% CI [0.787, 0.825], 0.746 with 95% CI [0.722, 0.765], 0.685 with 95% CI [0.667, 0.699], and 0.505 with 95% CI [0.489, 0.539], respectively. Area Under the Precision Recall Curve was 0.684 with 95% CI [0.679, 0.702], 0.647 with 95% CI [0.638, 0.664], 0.429 with 95% CI [0.417, 0.434], and 0.551 with 95% CI [0.551, 0.552], respectively. Conclusion Gradient Boosted Decision Trees is the best model of the supervised machine learning algorithms with satisfactory performance in validation cohort for predicting voided Computerized Physician Order Entry in Oral and Maxillofacial Surgery inpatients.