Folate Deficiency in an Urban Safety Net Population.

PURPOSE: Since mandatory fortification of grain products with folic acid in the United States in 1998, folate deficiency has become rare. Some have suggested that serum folate levels should be tested rarely in countries with mandatory folic acid fortification, given low rates of deficiency, high cost per deficiency diagnosis, and low rates of supplementation for those diagnosed as deficient. Given persistent racial, ethnic, and socioeconomic disparities in folate deficiency, these suggestions may not apply to all populations. We examine the rate at which serum testing detected folate deficiency in an urban safety net hospital and the characteristics of folate-deficient patients. METHODS: We reviewed the charts of all inpatients and emergency department patients with low serum folate results at a safety net hospital in Boston in 2018. We collected data concerning demographics, social determinants of health, clinical factors, and whether folate supplementation was prescribed. Finally, we performed a cost analysis. RESULTS: Of 1368 patients tested, 76 (5.5%) met criteria for folate deficiency. Overall, 86.8% of these patients were anemic, and 17.1% had macrocytic anemia; 42% were diagnosed with malnutrition. Common social determinants in folate-deficient patients included birth outside of the United States, homelessness, and alcohol use disorder. Of folate-deficient patients, 88% were newly prescribed folic acid supplementation at discharge. The estimated charge per deficient test was $1278. CONCLUSION: Compared with a nearby institution, serum folate testing at our safety net hospital detected deficiency at a higher rate, incurred a lower charge per deficient test, and was more likely to impact management.

Artificial Intelligence for Histology-Based Detection of Microsatellite Instability and Prediction of Response to Immunotherapy in Colorectal Cancer.

Microsatellite instability (MSI) is a molecular marker of deficient DNA mismatch repair (dMMR) that is found in approximately 15% of colorectal cancer (CRC) patients. Testing all CRC patients for MSI/dMMR is recommended as screening for Lynch Syndrome and, more recently, to determine eligibility for immune checkpoint inhibitors in advanced disease. However, universal testing for MSI/dMMR has not been uniformly implemented because of cost and resource limitations. Artificial intelligence has been used to predict MSI/dMMR directly from hematoxylin and eosin (H&E) stained tissue slides. We review the emerging data regarding the utility of machine learning for MSI classification, focusing on CRC. We also provide the clinician with an introduction to image analysis with machine learning and convolutional neural networks. Machine learning can predict MSI/dMMR with high accuracy in high quality, curated datasets. Accuracy can be significantly decreased when applied to cohorts with different ethnic and/or clinical characteristics, or different tissue preparation protocols. Research is ongoing to determine the optimal machine learning methods for predicting MSI, which will need to be compared to current clinical practices, including next-generation sequencing. Predicting response to immunotherapy remains an unmet need.