Pigs: Large Animal Preclinical Cancer Models.

Pigs are playing an increasingly vital role as translational biomedical models for studying human pathophysiology. The annotation of the pig genome was a huge step forward in translatability of pigs as a biomedical model for various human diseases. Similarities between humans and pigs in terms of anatomy, physiology, genetics, and immunology have allowed pigs to become a comprehensive preclinical model for human diseases. With a diverse range, from craniofacial and ophthalmology to reproduction, wound healing, musculoskeletal, and cancer, pigs have provided a seminal understanding of human pathophysiology. This review focuses on the current research using pigs as preclinical models for cancer research and highlights the strengths and opportunities for studying various human cancers.

Circulating Tumor Cells: How Far Have We Come with Mining These Seeds of Metastasis?

Circulating tumor cells (CTCs) are cancer cells that slough off from the tumor and circulate in the peripheral blood and lymphatic system as micro metastases that eventually results in macro metastases. Through a simple blood draw, sensitive CTC detection from clinical samples has proven to be a useful tool for determining the prognosis of cancer. Recent technological developments now make it possible to detect CTCs reliably and repeatedly from a simple and straightforward blood test. Multicenter trials to assess the clinical value of CTCs have demonstrated the prognostic value of these cancer cells. Studies on CTCs have filled huge knowledge gap in understanding the process of metastasis since their identification in the late 19th century. However, these rare cancer cells have not been regularly used to tailor precision medicine and or identify novel druggable targets. In this review, we have attempted to summarize the milestones of CTC-based research from the time of identification to molecular characterization. Additionally, the need for a paradigm shift in dissecting these seeds of metastasis and the possible future avenues to improve CTC-based discoveries are also discussed.

Benefits and opportunities of the transgenic Oncopig cancer model.

Cancer remains a leading cause of morbidity and mortality, and a paradigm shift is needed to fundamentally revisit drug development efforts. Pigs share close similarities to humans and may serve as an alternative model. Recently, a transgenic 'Oncopig' line has been generated to induce solid tumors with organ specificity, opening the potential of Oncopigs as a platform for developing novel therapeutic regimens.

Red cabbage juice-mediated gut microbiota modulation improves intestinal epithelial homeostasis and ameliorates colitis.

Gut microbiota plays a crucial role in inflammatory bowel disease (IBD) and has therapeutic benefits. Thus, targeting the gut microbiota is a promising therapeutic approach for IBD treatment. We recently found that red cabbage juice (RCJ) ameliorates dextran sulfate sodium (DSS)-induced colitis in mice. However, the underlying mechanisms remain unknown. The current study investigated the modulation of gut microbiota in response to treatment with RCJ to ameliorate the DSS colitis. The initial results demonstrated that mice treated with DSS + RCJ showed increased body weight and decreased diarrhea and blood in feces compared to the DSS alone group. RCJ ameliorated colitis by regulating the intestinal barrier function by reducing the number of apoptotic cells, improving colonic protective mucin, and increasing tight junction protein in RCJ + DSS groups compared to the DSS group. Short-gun metagenomic analysis revealed significant enrichment of short-chain fatty acid (SCFAs)-producing bacteria (Butyrivibrio, Ruminococcaceae, Acetatifactor muris, Rosburia Sp. CAG:303 , Dorea Sp. 5-2) increased PPAR-© activation, leading to repression of the nuclear factor κB (NFκB) signaling pathway, thus decreasing the production of crucial inflammatory cytokines and chemokines in the RCJ + DSS groups compared to the DSS group. Pathway abundance analysis showed an increased abundance of the SCFA pathway, reduced histidine degradation ( Bacteroides sartorii, and Bacteroides caecimuris ), and LCFA production in the RCJ+DSS treated group, suggesting the promotion of good colonic health. Furthermore, increased T-reg (FOXP3+) cells in the colon were due to SCFAs produced by the gut microbiota, which was corroborated by an increase in IL-10, a vital anti-inflammatory cytokine. Thus, our study provides the first evidence that RCJ ameliorates colonic inflammation by modulating the gut microbiota.

Preclinical models to study patient-derived circulating tumor cells and metastasis.

Circulating tumor cells (CTCs) that are detached from the tumor can be precursors of metastasis. The majority of studies focus on enumeration of CTCs from patient blood to predict recurrence and therapy outcomes. Very few studies have managed to expand CTCs to investigate their functional dynamics with respect to genetic changes, tumorigenic potential, and response to drug treatment. A growing amount of evidence based on successful CTC expansion has revealed novel therapeutic targets that are associated with the process of metastasis. In this review, we summarize the successes, challenges, and limitations that collectively contribute to the better understanding of metastasis using patient-derived CTCs as blood-borne seeds of metastasis. The roadblocks and future avenues to move CTC-based scientific discoveries forward are also discussed.

Western diet contributes to the pathogenesis of non-alcoholic steatohepatitis in male mice via remodeling gut microbiota and increasing production of 2-oleoylglycerol.

The interplay between western diet and gut microbiota drives the development of non-alcoholic fatty liver disease and its progression to non-alcoholic steatohepatitis. However, the specific microbial and metabolic mediators contributing to non-alcoholic steatohepatitis remain to be identified. Here, a choline-low high-fat and high-sugar diet, representing a typical western diet, named CL-HFS, successfully induces male mouse non-alcoholic steatohepatitis with some features of the human disease, such as hepatic inflammation, steatosis, and fibrosis. Metataxonomic and metabolomic studies identify Blautia producta and 2-oleoylglycerol as clinically relevant bacterial and metabolic mediators contributing to CL-HFS-induced non-alcoholic steatohepatitis. In vivo studies validate that both Blautia producta and 2-oleoylglycerol promote liver inflammation and hepatic fibrosis in normal diet- or CL-HFS-fed mice. Cellular and molecular studies reveal that the GPR119/TAK1/NF-κB/TGF-ß1 signaling pathway mediates 2-oleoylglycerol-induced macrophage priming and subsequent hepatic stellate cell activation. These findings advance our understanding of non-alcoholic steatohepatitis pathogenesis and provide targets for developing microbiome/metabolite-based therapeutic strategies against non-alcoholic steatohepatitis.

Red Cabbage Juice-Mediated Gut Microbiota Modulation Improves Intestinal Epithelial Homeostasis and Ameliorates Colitis.

Gut microbiota plays a crucial role in inflammatory bowel diseases (IBD) and can potentially prevent IBD through microbial-derived metabolites, making it a promising therapeutic avenue. Recent evidence suggests that despite an unclear underlying mechanism, red cabbage juice (RCJ) alleviates Dextran Sodium Sulfate (DSS)-induced colitis in mice. Thus, the study aims to unravel the molecular mechanism by which RCJ modulates the gut microbiota to alleviate DSS-induced colitis in mice. Using C57BL/6J mice, we evaluated RCJ's protective role in DSS-induced colitis through two cycles of 3% DSS. Mice were daily gavaged with PBS or RCJ until the endpoint, and gut microbiota composition was analyzed via shotgun metagenomics. RCJ treatment significantly improved body weight (p ≤ 0.001), survival in mice (p < 0.001) and reduced disease activity index (DAI) scores. Further, RCJ improved colonic barrier integrity by enhancing the expression of protective colonic mucins (p < 0.001) and tight junction proteins (p ≤ 0.01) in RCJ + DSS-treated mice compared to the DSS group. Shotgun metagenomic analysis revealed an enrichment of short-chain fatty acids (SCFAs)-producing bacteria (p < 0.05), leading to increased Peroxisome Proliferator-Activated Receptor Gamma (PPAR-γ) activation (p ≤ 0.001). This, in turn, resulted in repression of the nuclear factor κB (NFκB) signaling pathway, causing decreased production of inflammatory cytokines and chemokines. Our study demonstrates colitis remission in a DSS-induced mouse model, showcasing RCJ as a potential modulator for gut microbiota and metabolites, with promising implications for IBD prevention and treatment.

Immune-Related Gene Signatures to Predict the Effectiveness of Chemoimmunotherapy in Triple-Negative Breast Cancer Using Exploratory Subgroup Discovery.

Triple-negative breast cancer (TNBC) is an aggressive subtype of breast cancer with limited therapeutic options. Although immunotherapy has shown potential in TNBC patients, clinical studies have only demonstrated a modest response. Therefore, the exploration of immunotherapy in combination with chemotherapy is warranted. In this project we identified immune-related gene signatures for TNBC patients that may explain differences in patients' outcomes after anti-PD-L1+chemotherapy treatment. First, we ran the exploratory subgroup discovery algorithm on the TNBC dataset comprised of 422 patients across 24 studies. Secondly, we narrowed down the search to twelve homogenous subgroups based on tumor mutational burden (TMB, low or high), relapse status (disease-free or recurred), tumor cellularity (high, low and moderate), menopausal status (pre- or post) and tumor stage (I, II and III). For each subgroup we identified a union of the top 10% of genotypic patterns. Furthermore, we employed a multinomial regression model to predict significant genotypic patterns that would be linked to partial remission after anti-PD-L1+chemotherapy treatment. Finally, we uncovered distinct immune cell populations (T-cells, B-cells, Myeloid, NK-cells) for TNBC patients with various treatment outcomes. CD4-Tn-LEF1 and CD4-CXCL13 T-cells were linked to partial remission on anti-PD-L1+chemotherapy treatment. Our informatics pipeline may help to select better responders to chemoimmunotherapy, as well as pinpoint the underlying mechanisms of drug resistance in TNBC patients at single-cell resolution.

TBX21 Methylation as a Potential Regulator of Immune Suppression in CMS1 Subtype Colorectal Cancer.

Cytotoxic T lymphocyte (CTL) infiltration is associated with survival, recurrence, and therapeutic response in colorectal cancer (CRC). Immune checkpoint inhibitor (ICI) therapy, which requires CTLs for response, does not work for most CRC patients. Therefore, it is critical to improve our understanding of immune resistance in this disease. We utilized 2391 CRC patients and 7 omics datasets, integrating clinical and genomic data to determine how DNA methylation may impact survival and CTL function in CRC. Using comprehensive molecular subtype (CMS) 1 patients as reference, we found TBX21 to be the only gene with altered expression and methylation that was associated with CTL infiltration. We found that CMS1 patients with high TBX21 expression and low methylation had a significant survival advantage. To confirm the role of Tbx21 in CTL function, we utilized scRNAseq data, demonstrating the association of TBX21 with markers of enhanced CTL function. Further analysis using pathway enrichment found that the genes TBX21, MX1, and SP140 had altered expression and methylation, suggesting that the TP53/P53 pathway may modify TBX21 methylation to upregulate TBX21 expression. Together, this suggests that targeting epigenetic modification more specifically for therapy and patient stratification may provide improved outcomes in CRC.

Somatic mutation variant analysis in rural, resectable non-small cell lung carcinoma patients.

Rural non-small cell lung cancer (NSCLC) patients do worse, largely related to lack of access to care. In this study, the mutational characteristics and potential for targeted therapy in rural, resectable NSCLC patients using whole exome sequencing (WES) were analyzed. WES was performed on tumor-adjacent normal pairs from rural patients undergoing resection for NSCLC. Sequencing alignment, variant-calling, annotation, and tumor mutational burden (TMB) calculations were performed using standard methods. cBioportal and OncoKB were used for comparisons of mutational frequencies and actionable targets. Thirty-four NSCLC patients underwent WES after surgical resection. The gene most frequently containing somatic variants was TP53. The median number of somatic variants was 188 (Range 11-1056), and median TMB was 3.30 (0.33-18.56) nonsynonymous mutations per Mb. Tumor stage and survival were not associated with number of variants, TMB or TP53 mutational status. Significant concordance among the most common mutations when cross-referenced to cBioportal (R = 0.78, p < 0.0001) was observed. 24% of patients had variants in actionable genes based on OncoKB annotation. In summary, we demonstrate baseline mutational frequency and establish foundations for targeted adjuvant trials in rural NSCLC patients with specific differences. Future studies must ensure to include rural patients to improve NSCLC patient outcomes.

Single Circulating-Tumor-Cell-Targeted Sequencing to Identify Somatic Variants in Liquid Biopsies in Non-Small-Cell Lung Cancer Patients.

Non-small-cell lung cancer (NSCLC) accounts for most cancer-related deaths worldwide. Liquid biopsy by a blood draw to detect circulating tumor cells (CTCs) is a tool for molecular profiling of cancer using single-cell and next-generation sequencing (NGS) technologies. The aim of the study was to identify somatic variants in single CTCs isolated from NSCLC patients by targeted NGS. Thirty-one subjects (20 NSCLC patients, 11 smokers without cancer) were enrolled for blood draws (7.5 mL). CTCs were identified by immunofluorescence, individually retrieved, and DNA-extracted. Targeted NGS was performed to detect somatic variants (single-nucleotide variants (SNVs) and insertions/deletions (Indels)) across 65 oncogenes and tumor suppressor genes. Cancer-associated variants were classified using OncoKB database. NSCLC patients had significantly higher CTC counts than control smokers (p = 0.0132; Mann-Whitney test). Analyzing 23 CTCs and 13 white blood cells across seven patients revealed a total of 644 somatic variants that occurred in all CTCs within the same subject, ranging from 1 to 137 per patient. The highest number of variants detected in ≥1 CTC within a patient was 441. A total of 18/65 (27.7%) genes were highly mutated. Mutations with oncogenic impact were identified in functional domains of seven oncogenes/tumor suppressor genes (NF1, PTCH1, TP53, SMARCB1, SMAD4, KRAS, and ERBB2). Single CTC-targeted NGS detects heterogeneous and shared mutational signatures within and between NSCLC patients. CTC single-cell genomics have potential for integration in NSCLC precision oncology.

Circulating Tumor-Macrophage Fusion Cells and Circulating Tumor Cells Complement Non-Small-Cell Lung Cancer Screening in Patients With Suspicious Lung-RADS 4 Nodules.

PURPOSE: Low-dose computed tomography (LDCT) screening of high-risk patients decreases lung cancer-related mortality. However, high false-positive rates associated with LDCT result in unnecessary interventions. To distinguish non-small-cell lung cancer (NSCLC) from benign nodules, in the present study, we integrated cellular liquid biomarkers in patients with suspicious lung nodules (lung cancer screening reporting and data system [Lung-RADS] 4). METHODS: Prospectively, 7.5 mL of blood was collected from 221 individuals (training set: 90 nonscreened NSCLC patients, 74 high-risk screening patients with no/benign nodules [Lung-RADS 1-3], and 20 never smokers; validation set: 37 patients with suspicious nodules [Lung-RADS 4]). Circulating tumor cells (CTCs), CTC clusters, and tumor-macrophage fusion (TMF) cells were identified by blinded analyses. Screening patients underwent a median of two LDCTs (range, 1-4) with a median surveillance time of 30 (range, 11-50) months. RESULTS: In the validation set of 37 Lung-RADS 4 patients, all circulating cellular biomarker counts (P < .005; Wilcoxon test) and positivity rates were significantly higher in 23 biopsy-proven NSCLC patients (CTCs: 23 of 23 [100%], CTC clusters: 6 of 23 [26.1%], and TMF cells: 15 of 23 [65.2%]) than in 14 patients with biopsy-proven benign nodules (6 of 14 [42.9%], 0 of 14 [0%], and 2 of 14 [14.3%]). On the basis of cutoff values from the training set, logistic regression with receiver operating characteristic and area under the curve analyses demonstrated that CTCs (sensitivity: 0.870, specificity: 1.0, and area under the curve: 0.989) and TMF cells (0.652; 0.880; 0.790) complement LDCT in diagnosing NSCLC in Lung-RADS 4 patients. CONCLUSION: Cellular liquid biomarkers have a potential to complement LDCT interpretation of suspicious Lung-RADS 4 nodules to distinguish NSCLC from benign lung nodules. A future prospective, large-scale, multicenter clinical trial should validate the role of cellular liquid biomarkers in improving diagnostic accuracy in high-risk patients with Lung-RADS 4 nodules.

Tumorigenic circulating tumor cells from xenograft mouse models of non-metastatic NSCLC patients reveal distinct single cell heterogeneity and drug responses.

BACKGROUND: Circulating tumor cells (CTCs) are liquid biopsies that represent micrometastatic disease and may offer unique insights into future recurrences in non-small cell lung cancer (NSCLC). Due to CTC rarity and limited stability, no stable CTC-derived xenograft (CDX) models have ever been generated from non-metastatic NSCLC patients directly. Alternative strategies are needed to molecularly characterize CTCs and means of potential future metastases in this potentially curable patient group. METHODS: Surgically resected NSCLC primary tumor tissues from non-metastatic patients were implanted subcutaneously in immunodeficient mice to establish primary tumor patient-derived xenograft (ptPDX) models. CTCs were isolated as liquid biopsies from the blood of ptPDX mice and re-implanted subcutaneously into naïve immunodeficient mice to generate liquid biopsy CTC-derived xenograft (CDX) tumor models. Single cell RNA sequencing was performed and validated in an external dataset of non-xenografted human NSCLC primary tumor and metastases tissues. Drug response testing in CDX models was performed with standard of care chemotherapy (carboplatin/paclitaxel). Blockade of MYC, which has a known role in drug resistance, was performed with a MYC/MAX dimerization inhibitor (10058-F4). RESULTS: Out of ten ptPDX, two (20%) stable liquid biopsy CDX mouse models were generated. Single cell RNA sequencing analysis revealed an additional regenerative alveolar epithelial type II (AT2)-like cell population in CDX tumors that was also identified in non-xenografted NSCLC patients' metastases tissues. Drug testing using these CDX models revealed different treatment responses to carboplatin/paclitaxel. MYC target genes and c-MYC protein were upregulated in the chemoresistant CDX model, while MYC/MAX dimerization blocking could overcome chemoresistance to carboplatin/paclitaxel. CONCLUSIONS: To overcome the lack of liquid biopsy CDX models from non-metastatic NSCLC patients, CDX models can be generated with CTCs from ptPDX models that were originally established from patients' primary tumors. Single cell analyses can identify distinct drug responses and cell heterogeneities in CDX tumors that can be validated in NSCLC metastases tissues. CDX models deserve further development and study to discover personalized strategies against micrometastases in non-metastatic NSCLC patients.

Nanoliposome C6-Ceramide in combination with anti-CTLA4 antibody improves anti-tumor immunity in hepatocellular cancer.

Combination therapy represents an effective therapeutic approach to overcome hepatocellular cancer (HCC) resistance to immune checkpoint blockade (ICB). Based upon previous work demonstrating that nanoliposome C6-ceramide (LipC6) not only induces HCC apoptosis but also prevents HCC-induced immune tolerance, we now investigate the potential of LipC6 in combination with ICB in HCC treatment. We generated orthotopic HCC-bearing mice, which have typical features in common with human patients, and then treated them with LipC6 in combination with the antibodies (Abs) for programmed cell death protein 1 (PD-1) or cytotoxic T-lymphocyte antigen 4 (CTLA4). The tumor growth was monitored by magnetic resonance imaging (MRI) and the intrahepatic immune profiles were checked by flow cytometry in response to the treatments. Realtime PCR (qPCR) was used to detect the expression of target genes. The results show that LipC6 in combination with anti-CTLA4 Ab, but not anti-PD-1 Ab, significantly slowed tumor growth, enhanced tumor-infiltrating CD8+ T cells, and suppressed tumor-resident CD4+ CD25+ FoxP3+ Tregs. Further molecular investigation indicates that the combinational treatment suppressed transcriptional factor Krüppel-like Factor 2 (KLF2), forkhead box protein P3 (FoxP3), and CTLA4. Our studies suggest that LipC6 in combination with anti-CTLA4 Ab represents a novel therapeutic approach with significant potential in activating anti-HCC immune response and suppressing HCC growth.

Epigenetic Regulation of Cancer Immune Cells.

The epigenetic regulation of immune response involves reversible and heritable changes that do not alter the DNA sequence. Though there have been extensive studies accomplished relating to epigenetic changes in cancer cells, recent focus has been shifted on epigenetic-mediated changes in the immune cells including T cells, Macrophages, Natural Killer cells and anti-tumor immune responses. This review compiles the most relevant and recent literature related to the role of epigenetic mechanisms including DNA methylation and histone modifications in immune cells of wide range of cancers. We also include recent research with respect to role of the most relevant transcription factors that epigenetically control the anti-tumor immune response. Finally, a statement of future direction that promises to look forward for strategies to improve immunotherapy in cancer.

Single-cell RNA sequencing to characterize the response of pancreatic cancer to anti-PD-1 immunotherapy.

Pancreatic cancer (PaC) is resistant to immune checkpoint therapy, but the underlying mechanisms are largely unknown. In this study, we have established four orthotopic PaC murine models with different PaC cell lines by intra-pancreatic inoculation. Therapeutic examinations demonstrate that only tumors induced with Panc02-H7 cells respond to αPD-1 antibody treatment, leading to significantly reduced tumor growth and increased survival in the recipient mice. Transcriptomic profiling at a single-cell resolution characterizes the molecular activity of different cells within tumors. Comparative analysis and validated experiments demonstrate that αPD-1-sensitive and -resistant tumors differently shape the immune landscape in the tumor microenvironment (TME) and markedly altering effector CD8+ T cells and tumor-associated macrophages (TAMs) in their number, frequency, and gene profile. More exhausted effector CD8+ T cells and increased M2-like TAMs with a reduced capacity of antigen presentation are detected in resistant Panc02-formed tumors versus responsive Panc02-H7-formed tumors. Together, our data highlight the correlation of tumor-induced imbalance of macrophages with the fate of tumor-resident effector CD8+ T cells and PaC response to αPD-1 immunotherapy. TAMs as a critical regulator of tumor immunity and immunotherapy contribute to PaC resistance to immune checkpoint blockade.

Clog-free high-throughput microfluidic cell isolation with multifunctional microposts.

Microfluidics have been applied to filtration of rare tumor cells from the blood as liquid biopsies. Processing is highly limited by low flow rates and device clogging due to a single function of fluidic paths. A novel method using multifunctional hybrid functional microposts was developed. A swift by-passing route for non-tumor cells was integrated to prevent very common clogging problems. Performance was characterized using microbeads (10 µm) and human cancer cells that were spiked in human blood. Design-I showed a capture efficiency of 96% for microbeads and 87% for cancer cells at 1 ml/min flow rate. An improved Design-II presented a higher capture efficiency of 100% for microbeads and 96% for cancer cells. Our method of utilizing various microfluidic functions of separation, bypass and capture has successfully guaranteed highly efficient separation of rare cells from biological fluids.

Breast metastases from primary lung cancer: a retrospective case series on clinical, ultrasonographic, and immunohistochemical features.

BACKGROUND: Lung cancer metastases to the breast are less common and consequently have received much less attention in clinical practice. The purpose of this study was to provide a better understanding of clinical, ultrasonographic, and immunohistochemical features of breast metastases from primary lung cancer. METHODS: This retrospective case series included patients with breast metastases from primary lung cancer between January 2012 and December 2020. Clinical features, ultrasonographic characteristics, and immunohistochemical findings were evaluated in this analysis. RESULTS: In all, 7 cases (mean ± standard deviation age: 57.4±8.3 years; range, 49-70 years) were evaluated. The maximum size of breast lesions in 6 cases ranged from 1.2 to 4.5 cm, while 1 case showed a diffused pattern. Ultrasound features of breast metastases from lung cancer were irregular (5/7, 71.4%), indistinct (6/7, 85.7%), hypoechoic (7/7, 100.0%), and parallel (6/7, 85.7%) masses without calcification. Immunohistochemical staining test was positive for thyroid transcription factor 1 (TTF-1) in all patients (7/7, 100.0%), 3 cases (3/5, 60.0%) were negative for p63, 5 cases (5/5, 100.0%) were positive for cytokeratin 7 (CK7), 4 cases (4/5, 80.0%) were positive for napsin A. CONCLUSIONS: The ultrasonographic features of lung metastases to the breast are clinically important to understand. A known history of the primary lung cancer is of great importance when evaluating patients with a breast nodule. The presence of an ipsilateral lung cancer, breast nodule and axillary lymphadenopathy should be considered with pathological and immunohistochemical data to differentiate breast metastases from a primary breast malignancy in this setting.

Current and Prospective Methods for Assessing Anti-Tumor Immunity in Colorectal Cancer.

Colorectal cancer (CRC) remains one of the deadliest malignancies worldwide despite recent progress in treatment strategies. Though immune checkpoint inhibition has proven effective for a number of other tumors, it offers benefits in only a small group of CRC patients with high microsatellite instability. In general, heterogenous cell groups in the tumor microenvironment are considered as the major barrier for unveiling the causes of low immune response. Therefore, deconvolution of cellular components in highly heterogeneous microenvironments is crucial for understanding the immune contexture of cancer. In this review, we assimilate current knowledge and recent studies examining anti-tumor immunity in CRC. We also discuss the utilization of novel immune contexture assessment methods that have not been used in CRC research to date.

Explainable artificial intelligence in high-throughput drug repositioning for subgroup stratifications with interventionable potential.

Enabling precision medicine requires developing robust patient stratification methods as well as drugs tailored to homogeneous subgroups of patients from a heterogeneous population. Developing de novo drugs is expensive and time consuming with an ultimately low FDA approval rate. These limitations make developing new drugs for a small portion of a disease population unfeasible. Therefore, drug repositioning is an essential alternative for developing new drugs for a disease subpopulation. This shows the importance of developing data-driven approaches that find druggable homogeneous subgroups within the disease population and reposition the drugs for these subgroups. In this study, we developed an explainable AI approach for patient stratification and drug repositioning. Contrast pattern mining and network analysis were used to discover homogeneous subgroups within a disease population. For each subgroup, a biomedical network analysis was done to find the drugs that are most relevant to a given subgroup of patients. The set of candidate drugs for each subgroup was ranked using an aggregated drug score assigned to each drug. The proposed method represents a human-in-the-loop framework, where medical experts use the data-driven results to generate hypotheses and obtain insights into potential therapeutic candidates for patients who belong to a subgroup. Colorectal cancer (CRC) was used as a case study. Patients' phenotypic and genotypic data was utilized with a heterogeneous knowledge base because it gives a multi-view perspective for finding new indications for drugs outside of their original use. Our analysis of the top candidate drugs for the subgroups identified by medical experts showed that most of these drugs are cancer-related, and most of them have the potential to be a CRC regimen based on studies in the literature.