1710. Profiling Patients with Rare Mucormycosis Infections Using Real-world Data

BACKGROUND: Invasive mucormycosis (IM) is universally fatal if untreated and is a challenge to assess due to its rarity. Diagnosis is difficult and can be missed due to a low index for suspicion. IM prevalence may be increasing with medical advances, especially in neutropenia management, leading to improved survival and expansion of the at-risk patient group. Large administrative databases contain patient-level chart information and may offer a way to describe IM patients in a representative sample of the population. METHODS: A retrospective observational study was conducted using US data from the deidentified Optum Electronic Health Record database between January 2007 and June 2018. Patients with any fungal infection and IM specifically were defined by ICD9 (110–119, 117.7) or ICD10 (B35-49, B46) codes. Descriptive statistics were used to assess demographics, comorbidities, and antifungal agents (AF) prescribed among IM patients with an underlying diagnosis of hematologic malignancy (HM). Restricting to an at-risk population minimized possible false IM coding in the sample. RESULTS: Of the approximately 97 million patients in the database, about 5 million had a fungal infection diagnosis and 5,208 had an IM diagnosis (0.005% overall, 0.11% of fungal infection). Among those with underlying HM (n = 698,187), 641 IM cases were observed (0.09%); of whom, 46% were male, 82% were over 40 years of age, and 77% were in the Midwest region of the United States. They were 83% Caucasian, 7% African American, 2% Asian, and 8% other/unknown race or ethnicity. The mean Charlson Comorbidity Index score was 3 ± 2 and the top comorbidities, aside from malignancy, were diabetes (24%, n = 151), chronic pulmonary disease (22%, n = 141), and renal disease (11%, n = 69). Not all IM patients were treated. There were 376 AF prescriptions, of which 35% were for fluconazole, 28% for posaconazole, and 14% for voriconazole, followed by 7–8% each for isavuconazole and amphotericin formulations. CONCLUSION: A sizable number of IM patients were identified from a large US electronic medical records database. More work is needed to understand the data. Given the significant challenges in prospectively identifying IM patients, a large database may allow for a broader insight into patients at risk and potential predictors of IM. DISCLOSURES: All authors: No reported disclosures.