Background: Progressive supranuclear palsy (PSP) is a rare neurodegenerative disorder that is difficult for primary care physicians to recognize due to its progressive nature and similarities to other neurologic disorders. This case-control study aimed to identify clinical features observed in general practice associated with a subsequent diagnosis of PSP.Methods: We analyzed a de-identified dataset of 152 PSP cases and 3,122 matched controls from electronic medical records of general practices in Germany. We used a random forests algorithm based on machine learning techniques to identify clinical features (medical conditions and treatments received) associated with pre-diagnostic PSP without using an a priori hypothesis. We then assessed the relative effects of the features with the highest importance scores and generated multivariate models using clustered logistic regression analyses to identify a subset of clinical features associated with subsequent PSP diagnosis.Results: Using the random forests approach, we identified 21 clinical features associated with pre-diagnostic PSP (odds ratio ≥2.0 in univariate analyses). From these, we constructed a multivariate model comprising 9 clinical features with ~90% likelihood of identifying a subsequent PSP diagnosis. These features included known PSP symptoms, common misdiagnoses, and 2 novel associations, diabetes mellitus and cerebrovascular disease, which are possible modifiable risk factors for PSP.Conclusion: In this case-control study using data from electronic medical records, we identified 9 clinical features, including 2 previously unknown factors, associated with the pre-diagnostic stage of PSP. These may be used to facilitate recognition of PSP and reduce time to referral by primary care physicians.

INTRODUCTION: The risk of tumor lysis syndrome (TLS) varies depending upon the underlying malignancy, tumor burden, and anti-tumor activity of the treatment administered. With the recent approval of several anti-cancer treatments across hematologic malignancies, mitigation of TLS has been a priority. Understanding the background rates of TLS and factors associated with the risk of TLS will likely aid in TLS mitigation. Current literature documenting background rates of TLS in hematologic malignancies is primarily based on studies involving small sample sizes and/or includes older data. This study aims to address the literature gap by assessing differences in frequency of TLS across hematologic malignancies and evaluating factors associated with the risk of TLS in large real-world populations. METHODS: Retrospective cohort study conducted using MarketScan Commercial, and Supplemental Medicare databases (2012-2016). Data include details on medical and pharmacy utilization for employees and their dependents in employer sponsored health plans in the US. Study includes adults (≥18 years) with ≥2 medical claims with diagnosis for one of the following hematologic malignancies: CLL, CML, MCL, AML, FL, DLBCL, and MM. Patients were classified into 7 sub-cohorts based on the index (initial) malignancy diagnosis. Study cohort was restricted to newly diagnosed patients who initiated malignancy-specific treatments post-diagnosis. The date of treatment initiation defined the index date. Patients were required to have continuous health plan enrollment ≥12 months before index date. TLS event was assessed based on medical claim with following diagnosis codes: ICD-9-CM: 277.88; ICD-10: E88.3x. Patients were followed from index date until TLS event, end of health plan enrollment or end of database, whichever occurred earlier. Time to TLS was calculated from date of treatment initiation until date of TLS event. Demographic characteristics, comorbidities and treatments were assessed for the study cohort. Univariate analysis was conducted to assess differences in patient and treatment characteristics across hematologic malignancies among patients with TLS. RESULTS: Cohort sample size (total n = 10,255), demographic characteristics and baseline comorbidity details are provided in Table 1. Variations in post-diagnosis initial treatments were observed across the 7 sub-cohorts (Table 2). Following treatment initiation the proportion of patients with a medical claim with a diagnosis of TLS was as follows: CLL (2.1%), CML (1.7%), AML (2.1%), MCL (1.4%), FL (0.5%), DLBCL (1.9%), and MM (0.9%). In most patients, the initial medical claim for TLS was observed in an inpatient setting (Figure 1). Median time to development of TLS from treatment initiation across the 7 sub-cohorts was as follows: CLL (12 days), CML (21 days), AML (101 days), MCL (222 days), FL (42 days), DLBCL (17 days), and MM (104 days). The treatments observed during the 60-day period prior to the TLS event are presented in Table 3. Among patients with TLS, age (p-value: 0.015) and treatments received 60 days prior to the TLS event (p-value: 0.001) differed significantly across hematologic malignancies. However, no differences were observed for covariates gender, region and Charlson co-morbidities index score (all p-values >0.05). CONCLUSIONS: This study based on a large-population database provides recent data on differences in the frequency of TLS across hematologic malignancies. Frequency of TLS varied across the 7 sub-cohorts, with higher frequency observed in CLL, CML, AML and DLBCL (range 1.7-2.1%) compared with MCL, FL, MM (range 0.5-1.4%). Majority (89%) of patients were on treatments prior to the TLS event; however, 11% had a TLS diagnosis and received no treatment 60 days prior to the event, indicating that TLS mitigation also needs to be considered in untreated patients. With emerging mono- and combination treatments, this study provides background data along with most common factors associated with risk of TLS, a useful tool to plan for mitigation strategies. Due to the limitations of claims data, the study could not distinguish between laboratory and clinical TLS; also, patients included were from an employer-sponsored plan and therefore likely younger and potentially with fewer comorbidities. Finally, data lacked detailed clinical characteristics these factors should be considered in interpreting the study results. Disclosures Karve: AbbVie: Employment, Equity Ownership. Diegidio:AbbVie: Other: Contractor. Zhang:AbbVie: Employment, Equity Ownership; Merck: Equity Ownership. Sebby:AbbVie: Employment, Equity Ownership. Cerri:AbbVie: Employment, Equity Ownership. Rosenberg:AbbVie: Employment, Equity Ownership.