Identification and Characterization of Immune Checkpoint Inhibitor-Induced Toxicities From Electronic Health Records Using Natural Language Processing

PURPOSE

Immune checkpoint inhibitors (ICIs) have revolutionized cancer treatment, yet their use is associated with immune-related adverse events (irAEs). Estimating the prevalence and patient impact of these irAEs in the real-world data setting is critical for characterizing the benefit/risk profile of ICI therapies beyond the clinical trial population. Diagnosis codes, such as International Classification of Diseases codes, do not comprehensively illustrate a patient's care journey and offer no insight into drug-irAE causality. This study aims to capture the relationship between ICIs and irAEs more accurately by using augmented curation (AC), a natural language processing-based innovation, on unstructured data in electronic health records.

METHODS

In a cohort of 9,290 patients treated with ICIs at Mayo Clinic from 2005 to 2021, we compared the prevalence of irAEs using diagnosis codes and AC models, which classify drug-irAE pairs in clinical notes with implied textual causality. Four illustrative irAEs with high patient impact-myocarditis, encephalitis, pneumonitis, and severe cutaneous adverse reactions, abbreviated as MEPS-were analyzed using corticosteroid administration and ICI discontinuation as proxies of severity.

RESULTS

For MEPS, only 70% (n = 118) of patients found by AC were also identified by diagnosis codes. Using AC models, patients with MEPS received corticosteroids for their respective irAE 82% of the time and permanently discontinued the ICI because of the irAE 35.9% (n = 115) of the time.

CONCLUSION

Overall, AC models enabled more accurate identification and assessment of patient impact of ICI-induced irAEs not found using diagnosis codes, demonstrating a novel and more efficient strategy to assess real-world clinical outcomes in patients treated with ICIs.

Liver disease progression in patients with alpha-1 antitrypsin deficiency and protease inhibitor ZZ genotype with or without lung disease

BACKGROUND

Alpha-1 antitrypsin deficiency (AATD) is caused by mutations in SERPINA1, which encodes alpha-1 antitrypsin, a protease inhibitor (Pi). Individuals with AATD and the homozygous Pi*ZZ genotype have variable risk of progressive liver disease but the influence of comorbid lung disease is poorly understood.

AIMS

To characterise patients with AATD Pi*ZZ and liver disease (AATD-LD-Pi*ZZ) with or without lung disease and describe liver disease-related clinical events longitudinally.

METHODS

This was an observational cohort study of patients in the Mayo Clinic Healthcare System (January 2000-September 2021). Patients were identified using diagnosis codes and natural language processing. Fibrosis stage (F0-F4) was assessed using a hierarchical approach at baseline (90 days before or after the index date) and follow-up. Clinical events associated with liver disease progression were assessed.

RESULTS

AATD-LD-Pi*ZZ patients with lung disease had a longer median time from AATD diagnosis to liver disease diagnosis versus those without lung disease (2.2 vs. 0.2 years, respectively). Compared to those without lung disease, patients with lung disease had a longer time to liver disease-related clinical events (8.5 years and not reached, respectively). AATD-LD-Pi*ZZ patients without lung disease were more likely to undergo liver transplantation compared with those with lung disease.

CONCLUSION

In patients with AATD and lung disease, there is a delay in the diagnosis of comorbid liver disease. Our findings suggest that liver disease may progress more rapidly in patients without comorbid lung disease.

Retrospective study of propionic acidemia using natural language processing in Mayo Clinic electronic health record data

BACKGROUND

Propionic acidemia (PA) is a rare autosomal recessive organic acidemia that classically presents within the first days of life with a metabolic crisis or via newborn screening and is confirmed with laboratory tests. Limited data exist on the natural history of patients with PA describing presentation, treatments, and clinical outcomes.

OBJECTIVE

To retrospectively describe the natural history of patients with PA in a clinical setting from a real-world database using both structured and unstructured electronic health record (EHR) data using novel data extraction techniques in a unique care setting.

DESIGN/METHODS

This retrospective study used EHR data to identify patients with PA seen at the Mayo Clinic. Unstructured clinical text (medical notes, pathology reports) were analyzed using augmented curation natural language processing models to enhance analysis of data extracted by structured data fields (International Classification of Diseases 9th or 10th revision [ICD-9/-10] codes, Current Procedural Terminology [CPT] codes, and medication orders). De-identified health records were also manually reviewed by clinical scientists to ensure data accuracy and completeness. The index date was defined as the patient's date of PA diagnosis at the Mayo Clinic. Results were reported as aggregate descriptive statistics relative to patients' index dates. Complications, therapeutic interventions, laboratory tests, procedures, and hospitalization encounters related to PA were described at and within 6 months of the patient's index date, and from medical history available before the index date.

RESULTS

In total, 13 patients with PA were identified, with visits occurring from 1998 to 2022. Age at diagnosis ranged from birth to 3 years; age at initial evaluation at the Mayo Clinic ranged from 3 days to 28 years. The mean number of Mayo Clinic outpatient visits was 31 (median duration of care, 2 years). PA-related complications were documented in 85% of patients and included nutritional difficulties (46%), metabolic decompensation events (MDEs; 38%), neurologic abnormalities (38%), and cardiomyopathy (7%). One pair of affected siblings had mild symptoms and no complications or MDEs. All 5 patients with a history of MDEs presented with developmental delays. Among patients with MDEs, the mean frequency of outpatient clinical care visits was 10 per year, and 3 patients required inpatient hospitalization (mean duration, 16 days). The incidence of severe complications was higher among patients with MDEs than those without MDEs. Of the patients with MDEs, 2 experienced crises while receiving treatment at the Mayo Clinic, with 9 total MDEs occurring between the 2 patients. Symptoms at presentation included hyperammonemia (78%), fever and/or decreased nutritional intake (67%), hyperglycemia/hypoglycemia (56%), intercurrent upper respiratory infection and/or lethargy (44%), constipation (33%), altered mental status (33%), and cough (33%).

CONCLUSIONS

This study highlights the range and frequency of clinical outcomes experienced by patients with PA and demonstrates the clinical burden of MDEs.

Prognostic significance of cardiac injury in COVID-19 patients with and without coronary artery disease

Objective

COVID-19 is a disease with high mortality, and risk factors for worse clinical outcome have not been well-defined yet. The aim of this study is to delineate the prognostic importance of presence of concomitant cardiac injury on admission in patients with COVID-19.

Methods

For this multi-center retrospective study, data of consecutive patients who were treated for COVID-19 between March 20 - April 20 2020 were collected. Clinical characteristics, laboratory findings and outcomes data were obtained from electronic medical records. In-hospital clinical outcome was compared between patients with and without cardiac injury.

Results

A total of 607 hospitalized patients with COVID-19 were included in the study; the median age was 62.5±14.3 years, and 334 (55%) were male. Cardiac injury was detected in 150 (24.7%) of patients included in the study. Mortality rate was higher in patients with cardiac injury (42% vs. 8%; p<0.01). The frequency of patients who required intensive care unit (ICU) (72% vs.19%), who developed acute kidney injury (AKI) (14% vs. 1%) and acute respiratory distress syndrome (ARDS) (71% vs. 18%) were also higher in patients with cardiac injury. In multivariate analysis, age, coronary artery disease (CAD), elevated CRP levels, and presence of cardiac injury (OR: 10.58, 95% CI: 2.42–46.27; p<0.001) were found to be independent predictors of mortality. In subgroup analysis, including patients free of history of CAD, presence of cardiac injury on admission also predicted mortality (OR: 2.52, 95% CI: 1.17–5.45; p=0.018).

Conclusion

Cardiac injury on admission is associated with worse clinical outcome and higher mortality risk in COVID-19 patients including patients free of previous CAD diagnosis.

Funding Acknowledgement

Type of funding sources: None.

Can disorder act as a chemical pressure? An optical study of the Hubbard model

The optical properties have been studied using the dynamical mean-field theory on a disordered Hubbard model. Despite the fact that disorder turns a metal to an insulator in high dimensional correlated materials, we notice that it can enhance certain metallic behavior as if a chemical pressure is applied to the system resulting in an increase of the effective lattice bandwidth (BW). We study optical properties in such a scenario and compare results with experiments where the BW is changed through isovalent chemical substitution (keeping electron filling unaltered) and obtain remarkable similarities vindicating our claim. We also make the point that these similarities differ from some other forms of BW tuned optical effects.

Clinical profile and outcome of renal tubular disorders in children: A single center experience

Tubular disorders form a significant proportion of pediatric kidney diseases and are an important differential diagnosis of failure to thrive (FTT) in children. Data regarding their outcome is scarce from India. We evaluated the clinical profile of these children and studied the outcome in terms of their growth and renal failure. This is a retrospective longitudinal study of all children with renal tubular disorders attending a tertiary care pediatric nephrology center from 2005 to 2010. Growth and renal outcomes were assessed by Z scores and estimated glomerular filtration rate at diagnosis and. The common disorders encountered were distal renal tubular acidosis (d-RTA) (44%), Bartter-like (Bartter's and Gitelman) syndromes (22%) followed by hereditary Fanconi syndrome (cystinosis and idiopathic Fanconi syndrome) (13%) and few cases of nephrogenic diabetes insipidus, hypophosphatemic rickets and idiopathic hypercalciuria. Male: female ratio was 1.22. The median age at diagnosis was 1.5 (range 0.13-11) years. Growth failure was the presenting feature in 86% of children followed by polyuria (60%) and bone deformities (47%). In 60% of children with hereditary Fanconi syndrome, nephropathic cystinosis was diagnosed, all of whom progressed to stage III chronic kidney disease (CKD) within 3.41 ± 1.42 years. With appropriate therapy, catch-up growth was noted in d-RTA and Bartter syndrome. Renal tubular disorders usually present with FTT. d-RTA is the most common etiology followed by Bartter-like syndrome. Renal function is preserved in all these disorders except for nephropathic cystinosis, who ultimately progressed to CKD. With appropriate and inexpensive therapy, these children do grow well.