Fitting a stochastic model of intensive care occupancy to noisy hospitalization time series during the COVID-19 pandemic.

Intensive care occupancy is an important indicator of health care stress that has been used to guide policy decisions during the COVID-19 pandemic. Toward reliable decision-making as a pandemic progresses, estimating the rates at which patients are admitted to and discharged from hospitals and intensive care units (ICUs) is crucial. Since individual-level hospital data are rarely available to modelers in each geographic locality of interest, it is important to develop tools for inferring these rates from publicly available daily numbers of hospital and ICU beds occupied. We develop such an estimation approach based on an immigration-death process that models fluctuations of ICU occupancy. Our flexible framework allows for immigration and death rates to depend on covariates, such as hospital bed occupancy and daily SARS-CoV-2 test positivity rate, which may drive changes in hospital ICU operations. We demonstrate via simulation studies that the proposed method performs well on noisy time series data and apply our statistical framework to hospitalization data from the University of California, Irvine (UCI) Health and Orange County, California. By introducing a likelihood-based framework where immigration and death rates can vary with covariates, we find, through rigorous model selection, that hospitalization and positivity rates are crucial covariates for modeling ICU stay dynamics and validate our per-patient ICU stay estimates using anonymized patient-level UCI hospital data.

A simple model for viral decay dynamics and the distribution of infected cell life spans in SHIV-infected infant rhesus macaques.

The dynamics of HIV viral load following the initiation of antiretroviral therapy is not well-described by simple, single-phase exponential decay. Several mathematical models have been proposed to describe its more complex behavior, the most popular of which is two-phase exponential decay. The underlying assumption in two-phase exponential decay is that there are two classes of infected cells with different lifespans. However, with the exception of CD4+ T cells, there is not a consensus on all of the cell types that can become productively infected, and the fit of the two-phase exponential decay to observed data from SHIV.C.CH505 infected infant rhesus macaques was relatively poor. Therefore, we propose a new model for viral decay, inspired by the Gompertz model where the decay rate itself is a dynamic variable. We modify the Gompertz model to include a linear term that modulates the decay rate. We show that this simple model performs as well as the two-phase exponential decay model on HIV and SIV data sets, and outperforms it for the infant rhesus macaque SHIV.C.CH505 infection data set. We also show that by using a stochastic differential equation formulation, the modified Gompertz model can be interpreted as being driven by a population of infected cells with a continuous distribution of cell lifespans, and estimate this distribution for the SHIV.C.CH505-infected infant rhesus macaques. Thus, we find that the dynamics of viral decay in this model of infant HIV infection and treatment may be explained by a distribution of cell lifespans, rather than two distinct cell types.

Common Secondary Genomic Variants Associated With Advanced Epithelioid Hemangioendothelioma.

Importance: Epithelioid hemangioendothelioma (EHE) is a rare, malignant vascular sarcoma characterized in most cases by a WWTR1-CAMTA1 fusion. The clinical course of EHE exhibits a dual nature. The condition is often indolent but can rapidly grow and metastasize unpredictably. No biomarkers to date are available to predict this phenotype. The hypothesis of this study was that better defining the genomic landscape of EHE using next-generation sequencing could offer additional therapies and insight into clinical outcomes. Objective: To characterize secondary EHE genomic alterations and their association with clinical outcomes. Design, Setting, and Participants: Multicenter, cross-sectional, retrospective study of next-generation sequencing results collected from participants diagnosed with EHE. Data were abstracted between May 1, 2013, and May 31, 2019. This analysis was conducted from January through June 2019. Summary genomic data were provided by commercial genomic testing companies. Main Outcomes and Measures: Presence or absence of secondary pathogenic genomic variants and their association with disease stage and clinical features. Results: A total of 49 participants with EHE were assessed for the presence or absence of secondary genomic variants. Of these, 32 (65.3%) were female; the mean (SD) age at diagnosis was 49.9 (18.3) years (range, 11-81 years). In all, 46 participants (93.9%) had confirmed WWTR1-CAMTA1 fusion; 26 participants (57.1%) exhibited a pathogenic genomic variant secondary to the WWTR1-CAMTA1 fusion; and 9 participants (18.4%) exhibited potentially targetable genomic variants. Commonly altered genes included CDKN2A/B, RB1, APC, and FANCA. Participants older than 45 years at diagnosis had an increased prevalence of secondary genomic variants that was not statistically significant (65.6% vs 38.5%; difference, 27.1%; 95% CI, -3.5% to 58.0%; P = .16) and were more likely to have a clinically targetable variant (28.1% vs 0%; difference, 28.1%; 95% CI, 11.2%-40.2%; P = .03). In 14 participants with clinical data available, those with stage III/IV EHE were more likely to exhibit a secondary pathogenic genomic variant (80% vs 0%; difference, 80%; 95% CI, 55.2%-100%; P = .006). Participants with stage III/IV EHE were diagnosed at an older age (mean [SD] age, 54.6 [14.1] years vs 31.7 [16.0] years; P = .05) and had elevated WWTR1-CAMTA1 fusion expression that was not statistically significant (mean [SD] expression, 677 [706] copies vs 231 [213] copies; P = .20). Conclusions and Relevance: Although EHE exhibits few secondary genomic variants, presence of key secondary variants may be prognostic for aggressive EHE. Further research is needed to confirm this finding and determine whether more intensive upfront treatment is necessary for these patients.

Inhibition of histone deacetylase 2 reduces MDM2 expression and reduces tumor growth in dedifferentiated liposarcoma.

Dedifferentiated liposarcoma (DDLPS) is a highly morbid mesenchymal tumor characterized and driven by genomic amplification of the MDM2 gene. Direct inhibition of MDM2 has shown promise pre-clinically, but has yet to be validated in clinical trials. Early in vitro studies have demonstrated that pan-histone deacetylase (HDAC) inhibition may have anti-MDM2 effects. Here we present in silico, in vitro, and mouse xenograft studies that suggest that specifically targeting HDAC2 reduces MDM2 expression and has anti-tumor affects in DDLPS. Two independent datasets, The Cancer Genome Atlas (TCGA; n = 58) and the Memorial Sloan-Kettering Cancer Center Dataset (MSKCC; n = 63), were used to identify the co-expression between class I HDACs and MDM2, and their clinical impact. HDAC2 was highly co-expressed with MDM2 (TCGA: Spearman's coefficient = 0.29, p = 0.03; MSKCC: Spearman's coefficient = 0.57, p < 0.001). As both a continuous and dichotomous predictor, elevated HDAC2 expression was associated with worsened disease-free survival in the TCGA (Continuous: Hazard-ratio (HR) 1.7; 95% Confidence Interval (95%CI) 0.97-2.9; p = 0.06; Dichotomous: HR 7.1, 95%CI 2.5-19.8, p < 0.001) and distant recurrence-free survival in the MSKCC (Continuous: HR 2.2; 95%CI 1.1-4.8; p = 0.04; Dichotomous: HR 2.8, 95%CI 1.2-6.4, p = 0.02). In vitro, treatment of DDLPS cell lines with the HDAC inhibitors MI-192 (HDAC2/3 inhibitor) or romidepsin (HDAC1/2 inhibitor) reduced MDM2 expression and induced apoptosis. In a murine DDLPS xenograft model, romidepsin reduced tumor growth and lowered tumor MDM2 expression. RNA-sequencing of romidepsin treated mouse tumors demonstrated markers of TP53 reactivation. Taken together, our data supports the hypothesis that targeting HDAC2 may represent a potential strategy to modulate MDM2 expression in DDLPS.

Degree of MDM2 Amplification Affects Clinical Outcomes in Dedifferentiated Liposarcoma.

BACKGROUND: Dedifferentiated liposarcomas (DDLPS) are mesenchymal tumors associated with universally poor response to treatment. Genomic amplification of murine double minute 2 (MDM2) is used as a diagnostic biomarker; however, no established biomarkers exist to guide DDLPS treatment. In the largest study of its kind, we report that the extent of MDM2 amplification, not simply the presence of MDM2 amplification, may be biologically important to the actions of DDLPS. PATIENTS AND METHODS: The distribution of MDM2 amplification in DDLPS was assessed using data from a commercial sequencing laboratory (n = 642) and The Cancer Genome Atlas (n = 57). Data from two retrospective clinical trials (n = 15, n = 16) and one prospective clinical trial (n = 25) were used to test MDM2's utility as a clinical biomarker. in vitro and in vivo assessments were conducted in DDLPS cell lines. RESULTS: Genomic MDM2 amplification follows a highly reproducible log-normal distribution. In patients with DDLPS treated with complete tumor resection, elevated MDM2 was associated with shortened time to recurrence as measured by genomic amplification (p = .003) and mRNA expression (p = .04). In patients requiring systemic therapy, higher MDM2 amplification was associated with reduced overall survival (p = .04). Doxorubicin treatment of DDLPS cells in vitro demonstrated variable sensitivity based on baseline MDM2 levels, and doxorubicin treatment elevated MDM2 expression. In vivo, treatment with doxorubicin followed by an MDM2 inhibitor improved doxorubicin sensitivity. CONCLUSION: MDM2 amplification levels in DDLPS follow a reproducible distribution and are associated with clinical outcomes and drug sensitivity. These results suggest that a prospective study of MDM2 as a predictive biomarker in DDLPS is warranted. IMPLICATIONS FOR PRACTICE: No validated biomarkers exist for treatment selection in dedifferentiated liposarcoma (DDLPS). Although murine double minute 2 (MDM2) is currently used for diagnosis, the clinical relevance of MDM2 amplification has yet to be fully assessed. This study found that MDM2 amplification follows a predictable distribution in DDLPS and correlates with clinical and biological outcomes. These data suggests that MDM2 amplification may be a useful biomarker in DDLPS.