Background: Drug repositioning is an emerging approach in pharmaceutical research for identifying novel therapeutic potentials for approved drugs and discover therapies for untreated diseases. Due to its time and cost efficiency, drug repositioning plays an instrumental role in optimizing the drug development process compared to the traditional de novo drug discovery process. Advances in the genomics, together with the enormous growth of large-scale publicly available data and the availability of high-performance computing capabilities, have further motivated the development of computational drug repositioning approaches. More recently, the rise of machine learning techniques, together with the availability of powerful computers, has made the area of computational drug repositioning an area of intense activities.Results: In this study, a novel framework SNF-NN based on deep learning is presented, where novel drug-disease interactions are predicted using drug-related similarity information, disease-related similarity information, and known drug-disease interactions. Heterogeneous similarity information related to drugs and disease is fed to the proposed framework in order to predict novel drug-disease interactions. SNF-NN uses similarity selection, similarity network fusion, and a highly tuned novel neural network model to predict new drug-disease interactions. The robustness of SNF-NN is evaluated by comparing its performance with nine baseline machine learning methods. The proposed framework outperforms all baseline methods ( $$AUC-ROC$$AUC-ROC = 0.867, and $$AUC-PR$$AUC-PR =0.876) using stratified 10-fold cross-validation. To further demonstrate the reliability and robustness of SNF-NN, two datasets are used to fairly validate the proposed framework’s performance against seven recent state-of-the-art methods for drug-disease interaction prediction. SNF-NN achieves remarkable performance in stratified 10-fold cross-validation with $$AUC-ROC$$AUC-ROC ranging from 0.879 to 0.931 and $$AUC-PR$$AUC-PR from 0.856 to 0.903. Moreover, the efficiency of SNF-NN is verified by validating predicted unknown drug-disease interactions against clinical trials and published studies.Conclusion: In conclusion, computational drug repositioning research can significantly benefit from integrating similarity measures in heterogeneous networks and deep learning models for predicting novel drug-disease interactions. The data and implementation of SNF-NN are available athttp://pages.cpsc.ucalgary.ca/ tnjarada/snf-nn.php.

Real-world evidence surrounding EGFR positive NSCLC patients in Canada is limited. Administrative databases in Alberta, Canada were used to evaluate EGFR testing and mutation prevalence in de novo metastatic NSCLC, as well as the characteristics, treatment patterns, and outcomes of individuals with Exon 19, L858R and Exon20ins mutations. Between 2013–2019, 2974 individuals underwent EGFR testing, of which 451 (15.2%) were EGFR positive. Among EGFR positive individuals, 221 (49.0%) had an Exon 19 mutation, 159 (35.3%) had an L858R mutation, and 18 (4%) had an Exon20ins mutation. The proportion of individuals who initiated 1L systemic therapy was 89.1% for Exon19, 85.5% for L858R, and 72.2% for Exon20ins carriers. The primary front-line systemic therapy was gefitinib or afatinib monotherapy for individuals with Exon 19 (93.4%) and L858R (94.1%) mutations versus platinum combination therapy for individuals with Exon20ins mutations (61.5%). The Exon20ins cohort had worse median overall survival from initiation of 1L systemic therapy (10.5 months [95% CI: 8.0-not estimable]) than the Exon19 (20.6 months [95% CI: 18.4–24.9]), and L858R cohorts (19.1 months [95% CI: 14.5–23.1]). These findings highlight that Exon20ins mutations represent a rare subset of NSCLC in which treatment options are limited and survival outcomes are worse relative to individuals with more common types of EGFR mutations.

Background: Socioeconomic status (SES) is associated with a range of health outcomes, including cancer diagnosis and survival. However, the evidence for this association is inconsistent between countries with and without single‐payer health care systems. In this study, the relationships between neighborhood‐level income, cancer stage at diagnosis, and cancer‐specific mortality in Alberta, Canada, were evaluated.Methods: The Alberta Cancer Registry was used to identify all primary cancer diagnoses between 2010 and 2020. Average neighborhood income was determined by linking the Canadian census to postal codes and was categorized into quintiles on the basis of income distribution in Alberta. Multivariable multinomial logistic regression was used to model the association between income quintile and stage at diagnosis, and the Fine–Gray proportional subdistribution hazards model was used to estimate the association between SES and cancer‐specific mortality.Results: Out of the 143,818 patients with cancer included in the study, those in lower income quintiles were significantly more likely to be diagnosed at stage III (odds ratio [OR], 1.07; 95% CI [confidence interval], 1.06–1.09) or IV (OR, 1.12; 95% CI, 1.11–1.14) after adjusting for age and sex. Lower income quintiles also had significantly worse cancer‐specific survival for breast, colorectal, liver, lung, non‐Hodgkin lymphoma, oral cavity, pancreas, and prostate cancers.Conclusions: Disparities were observed in cancer outcomes across neighborhood‐level income groups in Alberta, which demonstrates that health inequities by SES exist in countries with single‐payer health care systems. Further research is needed to better understand the underlying causes and to develop strategies to mitigate these disparities.

PURPOSE: The optimal characteristics among patients with breast cancer to recommend neoadjuvant chemotherapy is an active area of clinical research. We developed and compared several approaches to developing prediction models for pathologic complete response (pCR) among patients with breast cancer in Alberta. METHODS: The study included all patients with breast cancer who received neoadjuvant chemotherapy in Alberta between 2012 and 2014 identified from the Alberta Cancer Registry. Patient, tumor, and treatment data were obtained through primary chart review. pCR was defined as no residual invasive tumor at surgical excision in breast or axilla. Two types of prediction models for pCR were built: (1) expert model: variables selected on the basis of oncologists' opinions and (2) data-driven model: variables selected by trained machine. These model types were fit using logistic regression (LR), random forests (RF), and gradient-boosted trees (GBT). We compared the models using area under the receiver operating characteristic curve and integrated calibration index, and internally validated using bootstrap resampling. RESULTS: A total of 363 cases were included in the analyses, of which 86 experienced pCR. The RF and GBT fits yielded higher optimism-corrected area under the receiver operating characteristic curves compared with LR for the expert (RF: 0.70; GBT: 0.69; LR: 0.65) and data-driven models (RF: 0.71; GBT: 0.68; LR: 0.64). The LR fit yielded the lowest integrated calibration indices for the expert (LR: 0.037; GBT: 0.05; RF: 0.10) and data-driven models (LR: 0.026; GBT: 0.06; RF: 0.099). CONCLUSION: Our models demonstrated predictive ability for pCR using routinely collected clinical and demographic variables. We show that machine learning fit methods can be used to optimize models for pCR prediction. We also show that additional variables beyond clinical expertise do not considerably improve predictive ability and may not be of value on the basis of the burden of data collection.