Predicting line of therapy transition via similar patient augmentation.

Analyzing large EHR databases to predict cancer progression and treatments has become a hot trend in recent years. An increasing number of modern deep learning models have been proposed to find the milestones of essential patient medical journey characteristics to predict their disease status and give healthcare professionals valuable insights. However, most of the existing methods are lack of consideration for the inter-relationship among different patients. We believe that more valuable information can be extracted, especially when patients with similar disease statuses visit the same doctors. Towards this end, a similar patient augmentation-based approach named SimPA is proposed to enhance the learning of patient representations and further predict lines of therapy transition. Our experiment results on a real-world multiple myeloma dataset show that our proposed approach outperforms state-of-the-art baseline approaches in terms of standard evaluation metrics for classification tasks.

Hypoxia Drives Centrosome Amplification in Cancer Cells via HIF1α-dependent Induction of Polo-Like Kinase 4.

Centrosome amplification (CA) has been implicated in the progression of various cancer types. Although studies have shown that overexpression of PLK4 promotes CA, the effect of tumor microenvironment on polo-like kinase 4 (PLK4) regulation is understudied. The aim of this study was to examine the role of hypoxia in promoting CA via PLK4. We found that hypoxia induced CA via hypoxia-inducible factor-1α (HIF1α). We quantified the prevalence of CA in tumor cell lines and tissue sections from breast cancer, pancreatic ductal adenocarcinoma (PDAC), colorectal cancer, and prostate cancer and found that CA was prevalent in cells with increased HIF1α levels under normoxic conditions. HIF1α levels were correlated with the extent of CA and PLK4 expression in clinical samples. We analyzed the correlation between PLK4 and HIF1A mRNA levels in The Cancer Genome Atlas (TCGA) datasets to evaluate the role of PLK4 and HIF1α in breast cancer and PDAC prognosis. High HIF1A and PLK4 levels in patients with breast cancer and PDAC were associated with poor overall survival. We confirmed PLK4 as a transcriptional target of HIF1α and demonstrated that in PLK4 knockdown cells, hypoxia-mimicking agents did not affect CA and expression of CA-associated proteins, underscoring the necessity of PLK4 in HIF1α-related CA. To further dissect the HIF1α-PLK4 interplay, we used HIF1α-deficient cells overexpressing PLK4 and showed a significant increase in CA compared with HIF1α-deficient cells harboring wild-type PLK4. These findings suggest that HIF1α induces CA by directly upregulating PLK4 and could help us risk-stratify patients and design new therapies for CA-rich cancers. IMPLICATIONS: Hypoxia drives CA in cancer cells by regulating expression of PLK4, uncovering a novel HIF1α/PLK4 axis.

Treatment initiation prediction by EHR mapped PPD tensor based convolutional neural networks boosting algorithm.

Electronic health records contain patient's information that can be used for health analytics tasks such as disease detection, disease progression prediction, patient profiling, etc. Traditional machine learning or deep learning methods treat EHR entities as individual features, and no relationships between them are taken into consideration. We propose to evaluate the relationships between EHR features and map them into Procedures, Prescriptions, and Diagnoses (PPD) tensor data, which can be formatted as images. The mapped images are then fed into deep convolutional networks for local pattern and feature learning. We add this relationship-learning part as a boosting module on a commonly used classical machine learning model. Experiments were performed on a Chronic Lymphocytic Leukemia dataset for treatment initiation prediction. Experimental results show that the proposed approach has better real world modeling performance than the baseline models in terms of prediction precision.

Empirical likelihood-based inferences for median medical cost regression models with censored data.

Recent studies show that medical cost data can be heavily censored and highly skewed, which leads to have more complex cost data analysis. In this paper, we propose influence function and empirical likelihood (EL)-based methods to construct confidence regions for regression parameters in median cost regression models with censored data. We further propose confidence intervals for the median cost with given covariates using the proposed EL-based confidence regions. Simulation studies are conducted to compare the proposed EL-based confidence regions with the existing normal approximation-based confidence regions in terms of coverage probabilities. The new EL-based methods are observed to have better finite sample performances than existing methods particularly when the censoring proportion is high. The new methods are also illustrated through a real data example.

Centrosome amplification: a quantifiable cancer cell trait with prognostic value in solid malignancies.

Numerical and/or structural centrosome amplification (CA) is a hallmark of cancers that is often associated with the aberrant tumor karyotypes and poor clinical outcomes. Mechanistically, CA compromises mitotic fidelity and leads to chromosome instability (CIN), which underlies tumor initiation and progression. Recent technological advances in microscopy and image analysis platforms have enabled better-than-ever detection and quantification of centrosomal aberrancies in cancer. Numerous studies have thenceforth correlated the presence and the degree of CA with indicators of poor prognosis such as higher tumor grade and ability to recur and metastasize. We have pioneered a novel semi-automated pipeline that integrates immunofluorescence confocal microscopy with digital image analysis to yield a quantitative centrosome amplification score (CAS), which is a summation of the severity and frequency of structural and numerical centrosome aberrations in tumor samples. Recent studies in breast cancer show that CA increases across the disease progression continuum, while normal breast tissue exhibited the lowest CA, followed by cancer-adjacent apparently normal, ductal carcinoma in situ and invasive tumors, which showed the highest CA. This finding strengthens the notion that CA could be evolutionarily favored and can promote tumor progression and metastasis. In this review, we discuss the prevalence, extent, and severity of CA in various solid cancer types, the utility of quantifying amplified centrosomes as an independent prognostic marker. We also highlight the clinical feasibility of a CA-based risk score for predicting recurrence, metastasis, and overall prognosis in patients with solid cancers.

Hypoxia-Induced Centrosome Amplification Underlies Aggressive Disease Course in HPV-Negative Oropharyngeal Squamous Cell Carcinomas.

Human papillomavirus-negative (HPV-neg) oropharyngeal squamous cell carcinomas (OPSCCs) are associated with poorer overall survival (OS) compared with HPV-positive (HPV-pos) OPSCCs. The major obstacle in improving outcomes of HPV-neg patients is the lack of robust biomarkers and therapeutic targets. Herein, we investigated the role of centrosome amplification (CA) as a prognostic biomarker in HPV-neg OPSCCs. A quantitative evaluation of CA in clinical specimens of OPSCC revealed that (a) HPV-neg OPSCCs exhibit higher CA compared with HPV-pos OPSCCs, and (b) CA was associated with poor OS, even after adjusting for potentially confounding clinicopathologic variables. Contrastingly, CA was higher in HPV-pos cultured cell lines compared to HPV-neg ones. This divergence in CA phenotypes between clinical specimens and cultured cells can therefore be attributed to an inaccurate recapitulation of the in vivo tumor microenvironment in the cultured cell lines, namely a hypoxic environment. The exposure of HPV-neg OPSCC cultured cells to hypoxia or stabilizing HIF-1α genetically increased CA. Both the 26-gene hypoxia signature as well as the overexpression of HIF-1α positively correlated with increased CA in HPV-neg OPSCCs. In addition, we showed that HIF-1α upregulation is associated with the downregulation of miR-34a, increase in CA and expression of cyclin- D1. Our findings demonstrate that the evaluation of CA may aid in therapeutic decision-making, and CA can serve as a promising therapeutic target for HPV-neg OPSCC patients.

A Quantitative Centrosomal Amplification Score Predicts Local Recurrence of Ductal Carcinoma In Situ.

PURPOSE: The purpose of this study is to predict risk of local recurrence (LR) in ductal carcinoma in situ (DCIS) with a new visualization and quantification approach using centrosome amplification (CA), a cancer cell-specific trait widely associated with aggressiveness. EXPERIMENTAL DESIGN: This first-of-its-kind methodology evaluates the severity and frequency of numerical and structural CA present within DCIS and assigns a quantitative centrosomal amplification score (CAS) to each sample. Analyses were performed in a discovery cohort (DC, n = 133) and a validation cohort (VC, n = 119). RESULTS: DCIS cases with LR exhibited significantly higher CAS than recurrence-free cases. Higher CAS was associated with a greater risk of developing LR (HR, 6.3 and 4.8 for DC and VC, respectively; P < 0.001). CAS remained an independent predictor of relapse-free survival (HR, 7.4 and 4.5 for DC and VC, respectively; P < 0.001) even after accounting for potentially confounding factors [grade, age, comedo necrosis, and radiotherapy (RT)]. Patient stratification using CAS (P < 0.0001) was superior to that by Van Nuys Prognostic Index (VNPI; HR for CAS = 6.2 vs. HR for VNPI = 1.1). Among patients treated with breast-conserving surgery alone, CAS identified patients likely to benefit from adjuvant RT. CONCLUSIONS: CAS predicted 10-year LR risk for patients who underwent surgical management alone and identified patients who may be at low risk of recurrence, and for whom adjuvant RT may not be required. CAS demonstrated the highest concordance among the known prognostic models such as VNPI and clinicopathologic variables such as grade, age, and comedo necrosis.

Influence function-based empirical likelihood for inference of quantile medical costs with censored data.

In this paper, we propose empirical likelihood methods based on influence function and Jackknife techniques to construct confidence intervals for quantile medical costs with censored data. We show that the influence function-based empirical log-likelihood ratio statistic for the quantile medical cost has a standard Chi-square distribution as its asymptotic distribution. Simulation studies are conducted to compare coverage probabilities and interval lengths of the proposed empirical likelihood confidence intervals with the existing normal approximation-based confidence intervals for quantile medical costs. The proposed methods are observed to have better finite-sample performances than existing methods. The new methods are also illustrated through a real example.