Tumor Molecular Profiling Research Articles

An adaptive approach to neoadjuvant therapy to maximize resection rates for pancreatic adenocarcinoma: A phase II trial.

TPS771 Background: Curative treatment for potentially resectable (resectable or borderline resectable) pancreatic adenocarcinoma, despite recent advances, leads to only suboptimal outcomes. Recent US national cooperative group trials have demonstrated the safety and feasibility of neoadjuvant chemotherapy (PMID 33475684, PMID 35834226), and also show that the two frontline regimens, FOLFIRINOX (5-fluorouracil, irinotecan, oxaliplatin) and gemcitabine/nab-paclitaxel (GnP), are comparable in this setting. Finally, ability to undergo resection remains the key driver of cure. Therefore, an approach that selects chemotherapy based on early response assessment, to maximize the probability of resection, is likely to improve curative outcomes. Methods: This is a phase II study intended to maximize the probability of surgical resection for pancreatic cancer. Key eligibility requirements include adult patients with a confirmed histopathologic diagnosis of pancreatic carcinoma or adenocarcinoma, an ECOG PS ≤ 1, resectable or borderline resectable disease by central radiology review, no prior therapy for index pancreatic cancer, and adequate bone marrow, liver and kidney function. Treatment includes 4 doses (~ 2 months total treatment time) of FOLFIRINOX (“Chemo1”). After a re-evaluation, using radiologic response, CA19.9 response, and chemotherapy toxicity, patients will either continue with “Chemo1” for another 4 cycles (8 doses) or switch to GnP (“Chemo2”) which will be administered for 4 cycles (12 doses, for ~4 month total treatment time), followed by surgical resection. Primary outcome will measure the proportion of patients undergoing surgical resection using historical dated compared to 32 patients in a current study with a time frame of 16 months. Using historical data, the expected proportion of patients undergoing resection is ~60%. The goal of this study is to increase this to 80% or higher. With a one-sided of 0.05 and power of 80%, 32 patients will be needed to demonstrate this difference. Key translational correlatives will include serial circulating tumor DNA assessment and extensive tumor molecular profiling. Clinical trial information: NCT04594772 .

Immune-Hot tumor features associated with recurrence in early-stage ovarian clear cell carcinoma.

Ovarian clear cell carcinoma (OCCC) is a distinct histotype of ovarian cancer, which usually presages a worse prognosis upon recurrence. Identifying patients at risk for relapse is an unmet need to improve outcomes. A retrospective cohort analysis of 195 early-stage OCCC patients diagnosed between January 2011 and December 2019 at National Taiwan University Hospital was conducted to identify prognostic factors for recurrence, progression-free survival (PFS) and overall survival (OS). Molecular profiling of tumors was performed in a case-controlled cohort matched for adjuvant therapy for biomarker discovery. Multivariate Cox proportional hazard model revealed that paclitaxel-based chemotherapy was associated with better PFS than nonpaclitaxel chemotherapy (HR=0.19, P=.006). The addition of bevacizumab was associated with better PFS, compared to no bevacizumab (HR=0.09, P=.02). Neither showed significant improvement in OS. Recurrence is associated with an Immune-Hot tumor feature (P=.03), the CTLA-4-high subtype (P=.01) and increased infiltration of immune cells in general. The Immune-Hot feature (HR=3.39, P=.005) and the CTLA-4-high subtype (HR=2.13, P=.059) were associated with worse PFS. Immune-Hot tumor features could prognosticate recurrence in early-stage OCCC.

Radiogenomic association of deep MR imaging features with genomic profiles and clinical characteristics in breast cancer

BackgroundIt has been believed that traditional handcrafted radiomic features extracted from magnetic resonance imaging (MRI) of tumors are normally shallow and low-ordered. Recent advancement in deep learning technology shows that the high-order deep radiomic features extracted automatically from tumor images can capture tumor heterogeneity in a more efficient way. We hypothesize that MRI-based deep radiomic phenotypes have significant associations with molecular profiles of breast cancer tumors. We aim to identify deep radiomic features (DRFs) from MRI, evaluate their significance in predicting breast cancer (BC) clinical characteristics and explore their associations with multi-level genomic factors.MethodsA denoising autoencoder was built to retrospectively extract 4,096 DRFs from 110 BC patients’ MRI. Visualization and clustering were applied to these DRFs. Linear Mixed Effect models were used to test their associations with multi-level genomic features (GFs) (risk genes, gene signatures, and biological pathway activities) extracted from the same patients’ mRNA expression profile. A Least Absolute Shrinkage and Selection Operator model was used to identify the most predictive DRFs for each clinical characteristic (tumor size (T), lymph node metastasis (N), estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER2) status).ResultsThirty-six conventional radiomic features (CRFs) for 87 of the 110 BC patients provided by a previous study were used for comparison. More than 1,000 DRFs were associated with the risk genes, gene signatures, and biological pathways activities (adjusted P-value < 0.05). DRFs produced better performance in predicting T, N, ER, PR, and HER2 status (AUC > 0.9) using DRFs. These DRFs showed significant powers of stratifying patients, linking to relevant biological and clinical characteristics. As a contrast, only eight risk genes were associated with CRFs. The RFs performed worse in predicting clinical characteristics than DRFs.ConclusionsThe deep learning-based auto MRI features perform better in predicting BC clinical characteristics, which are more significantly associated with GFs than traditional semi-auto MRI features. Our radiogenomic approach for identifying MRI-based imaging signatures may pave potential pathways for the discovery of genetic mechanisms regulating specific tumor phenotypes and may enable a more rapid innovation of novel imaging modalities, hence accelerating their translation to personalized medicine.

Pathogenic Variants in Adult-Onset Cancer Predisposition Genes in Pediatric Cancer: Prevalence and Impact on Tumor Molecular Features and Clinical Management.

Clinical genomic sequencing of pediatric tumors is increasingly uncovering pathogenic variants in adult-onset cancer predisposition genes (aoCPG). Nevertheless, it remains poorly understood how often aoCPG variants are of germline origin and whether they influence tumor molecular profiles and/or clinical care. In this study, we examined the prevalence, spectrum, and impacts of aoCPG variants on tumor genomic features and patient management at our institution. This is a retrospective study of 1,018 children with cancer who underwent clinical genomic sequencing of their tumors. Tumor genomic data were queried for pathogenic variants affecting 24 preselected aoCPGs. Available tumor whole-genome sequencing (WGS) data were evaluated for second hit mutations, loss of heterozygosity (LOH), DNA mutational signatures, and homologous recombination deficiency (HRD). Patients whose tumors harbored one or more pathogenic aoCPG variants underwent subsequent germline testing based on hereditary cancer evaluation and family or provider preference. Thirty-three patients (3%) had tumors harboring pathogenic variants affecting one or more aoCPGs. Among 21 tumors with sufficient WGS sequencing data, six (29%) harbored a second hit or LOH affecting the remaining aoCPG allele with four of these six tumors (67%) also exhibiting a DNA mutational signature consistent with the altered aoCPG. Two additional tumors demonstrated HRD, of uncertain relation to the identified aoCPG variant. Twenty-one of 26 patients (81%) completing germline testing were positive for the aoCPG variant in the germline. All germline-positive patients were counseled regarding future cancer risks, surveillance, and risk-reducing measures. No patients had immediate cancer therapy changed due to aoCPG data. AoCPG variants are rare in pediatric tumors; however, many originate in the germline. Almost one third of tumor aoCPG variants examined exhibited a second hit and/or conferred an abnormal DNA mutational profile suggesting a role in tumor formation. aoCPG information aids in cancer risk prediction but is not commonly used to alter the treatment of pediatric cancers.

Essential news for clinical practice—thyroid cancer

SummaryEstablished systemic treatment options for advanced thyroid cancer include the multityrosine kinase inhibitors lenvatinib and sorafenib for radioactive iodine refractory differentiated thyroid cancer (DTC) and vandetanib and cabozantinib for medullary thyroid cancer (MTC). Recently, the COSMIC-311 study resulted in approval of cabozantinib for DTC with progression upon lenvatinib and/or sorafenib; thus, for the first time a specific second-line therapy has been defined for these patients. In addition, the therapeutic landscape of thyroid cancer has been expanded by targeted therapies based on molecular tumor profiles. Selective RET inhibitors such as selpercatinib and pralsetinib show high activity in DTC with RET fusions and MTC with RET mutations, respectively. Further targeted treatment options include NTRK inhibitors for thyroid cancers with NTRK fusions and BRAF-targeted therapy for BRAF V600E-mutated (anaplastic) thyroid cancer.

Molecular profiling of male breast cancer by multigene panel testing: Implications for precision oncology.

Compared with breast cancer (BC) in women, BC in men is a rare disease with genetic and molecular peculiarities. Therapeutic approaches for male BC (MBC) are currently extrapolated from the clinical management of female BC, although the disease does not exactly overlap in males and females. Data on specific molecular biomarkers in MBC are lacking, cutting out male patients from more appropriate therapeutic strategies. Growing evidence indicates that Next Generation Sequencing (NGS) multigene panel testing can be used for the detection of predictive molecular biomarkers, including Tumor Mutational Burden (TMB) and Microsatellite Instability (MSI). In this study, NGS multigene gene panel sequencing, targeting 1.94 Mb of the genome at 523 cancer-relevant genes (TruSight Oncology 500, Illumina), was used to identify and characterize somatic variants, Copy Number Variations (CNVs), TMB and MSI, in 15 Formalin-Fixed Paraffin-Embedded (FFPE) male breast cancer samples. A total of 40 pathogenic variants were detected in 24 genes. All MBC cases harbored at least one pathogenic variant. PIK3CA was the most frequently mutated gene, with six (40.0%) MBCs harboring targetable PIK3CA alterations. CNVs analysis showed copy number gains in 22 genes. No copy number losses were found. Specifically, 13 (86.7%) MBCs showed gene copy number gains. MYC was the most frequently amplified gene with eight (53.3%) MBCs showing a median fold-changes value of 1.9 (range 1.8-3.8). A median TMB value of 4.3 (range 0.8-12.3) mut/Mb was observed, with two (13%) MBCs showing high-TMB. The median percentage of MSI was 2.4% (range 0-17.6%), with two (13%) MBCs showing high-MSI. Overall, these results indicate that NGS multigene panel sequencing can provide a comprehensive molecular tumor profiling in MBC. The identification of targetable molecular alterations in more than 70% of MBCs suggests that the NGS approach may allow for the selection of MBC patients eligible for precision/targeted therapy.

Using publicly available datasets to identify population-based transcriptomic landscape contributing to the aggressiveness of breast cancer in young women.

Introduction: Although the risk of breast cancer increases with advancing age, some regions have larger number of young breast cancer patients (≤45years-old), such as the Middle East, Eastern Asia, and North Africa, with more aggressive and poorly differentiated tumors. We aimed to conduct an in-silico analysis in an attempt to understand the aggressive nature of early-onset breast cancer, and to identify potential drivers of early-onset breast cancer using gene expression profiling datasets in a population-dependent manner. Methods: Functional genomics experiments data were acquired from cBioPortal database for cancer genomics, followed by the stratification of patients based on the age at representation of breast cancer and race. Differential gene expression analysis and gene amplification status analysis were carried out, followed by hub gene, transcription factor, and signalling pathway identification. Results: PAM50 subtype analysis revealed that young patients (≤45years-old) had four-fold more basal tumors and worst progression-free survival (median of 101months), compared with the 45-65years group (median of 168months). Fourteen genes were amplified in more than 14% of patients with an early-onset breast cancer. Interestingly, FREM2, LINC00332, and LINC00366 were exclusively amplified in younger patients. Gene expression data from three different populations (Asian, White, and African) revealed a unique transcriptomic profile of young patients, which was also reflected on the PAM50 subtype analysis. Our data indicates a higher tendency of young African patients to develop basal tumors, while young Asian patients are more prone to developing Luminal A tumors. Most genes that were found to be upregulated in younger patients are involved in important signaling pathways that promote cancer progression and metastasis, such as MAPK pathway, Reelin pathway and the PI3K/Akt pathway. Conclusion: This study provides strong evidence that the molecular profile of tumors derived from young breast cancer patients of different populations is unique and may explain the aggressiveness of these tumors, stressing the need to conduct population- based multi-omic analyses to identify the potential drivers for tumorigenesis and molecular profiles of young breast cancer patients.

Liquid Biopsies: As an Emerging Tool in Cancer Diagnosis and Monitoring Therapies in Metastasis

<bold></bold>Despite the availability of advanced diagnosis and treatment methods, cancer is still a leading cause of morbidity. The most readily and routinely available method for cancer diagnosis is tissue biopsy. Tissue biopsies are generally based on molecular profiling of primary tumors, which can change during the metastasis stage. Therefore, it’s unable to give sufficient information about the tumor. These problems can be overcome by liquid biopsies which is considered as fast, non-invasive, and more sensitive method for cancer diagnosis. It also has applications in early diagnosis and real-time monitoring of tumor prognosis and development. This can be achieved by molecular profiling of circulating tumor cells (CTCs), cell-free DNA (cfDNA), exosomes, microRNA (miRNA), and tumor-educated platelets (TEPs) from biological fluids which are considered as biomarkers. The expression of these biomarkers gives an entire snapshot of both the primary and metastasis stages of cancer. It is very useful to determine different stages of cancer and to modify treatment strategies based on treatment response. Therefore, it opens up an opportunity to use liquid biopsies in the area of precision medicine. This review summarizes in-depth information on liquid biopsies, cancer metastasis, various methods in liquid biopsies, biomarkers such as CTCs, cfDNA, miRNA, exosomes, TEPs, and the role of these biomarkers in monitoring therapy.

Significance of Liquid Biopsy Containing Circulating Tumor DNA (ctDNA) Analysis for Evaluation and Management of Metastatic Cancer

According to the GLOBOCAN 2020 data from WHO, nearly 10 million deaths took place due to cancer thus, cancer became a leading cause of death all over the world. The most common cancer are breast cancer, lung cancer, colon cancer, prostate cancer, and pancreatic cancer. In the last few decades, Tissue Biopsy was considered the major approach to finding a confirmed diagnosis of particular cancer but due to the invasive process, it has become difficult for the clinician to ask for frequent tissue biopsies for the monitoring targeted treatment of the patients and thus clinicians generally assume the potency of the drugs without doing any biopsy analysis and thus there is an increase in the mortality of cancer patients. However, the liquid biopsy technique has changed this protocol of such treatment and greatly revolutionized the field of Clinical Oncology. Liquid biopsy ensures better-targeted treatment and many times help to give a complete cure. Liquid Biopsy is a non-invasive standard procedure to identify and prognose the tumor markers from circulating blood samples. It also offers the ease of sampling continuous screening, monitoring devices, therapeutic regimens, etc. Liquid Biopsy is composed of several components such as Circulating Tumor Cells (CTCs), free Circulating Tumor DNA (ctDNA), exosomes, vesicles, and platelets. The liquid biopsy includes body fluids such as blood, saliva, plasma, cerebrospinal fluid, and urine. Circulating Tumor DNA (ctDNA) in the blood sample of metastatic cancer patients has become the most critical technology to evaluate the progression of metastasis and give a precise treatment for many cancers. ctDNA analysis provides insight into the molecular profiling and biological characteristics of several tumors. This review is mainly aimed at highlighting the important technological development in ctDNA analysis and also to provide the significance of it in monitoring metastatic cancers and providing the best treatment to cure cancer.

Implementation of a Customized Tertiary Analysis Platform for the Reporting of Somatic Variants.

Precision medicine for oncology requires the evaluation of variants identified in molecular profiling of solid tumors and hematologic malignancies. This includes evaluation of pre-analytical and postanalytical quality metrics, variant interpretation, classification, and tiering as outlined in established guidelines, association with clinical significance such as FDA approved drugs and clinical trials, and finally comprehensive reporting. This study documents our experience with the customization and implementation of a software platform that facilitates these requirements for effective reporting of somatic variants.

Tumor PD-L1 expression and molecular profiling are not associated with immune checkpoint inhibitor-induced thyroid dysfunction in advanced NSCLC patients.

Background: Immune-checkpoint inhibitors (ICIs) have revolutionized the treatment of advanced non-small cell lung cancer (NSCLC), however are frequently associated with thyroid immune-related adverse events (IRAEs). We investigated the association between patient characteristics, tumor PD-L1 expression and molecular profile with the development of thyroid IRAEs in NSCLC patients. Methods: Single center, retrospective study including 107 NSCLC patients treated with PD-1/PD-L1 inhibitors from April 2016 to July 2020. All patients were euthyroid at baseline with at least two TSH measurements post-treatment initiation. The primary outcome was the difference in tumor PD-L1 expression in patients who developed any thyroid IRAEs versus those who remained euthyroid. Additional outcomes included development of overt thyroid dysfunction, the association of specific molecular alterations with thyroid IRAEs, and onset of thyroid IRAEs as a function of tumor PD-L1 expression. Results: Overall, 37 (34.6%) patients developed any thyroid dysfunction and 18 (16.8%) developed overt thyroid dysfunction. Tumor PD-L1 staining intensity was not associated with thyroid IRAEs. TP53 mutation was less likely to be associated with any thyroid dysfunction (p < 0.05) and no association was found between EGFR, ROS, ALK or KRAS mutations. There was no association between PD-L1 expression and time to develop thyroid IRAEs. Conclusion: PD-L1 expression is not associated with the development of thyroid dysfunction in advanced NSCLC patients treated with ICIs, suggesting that thyroid IRAEs are unrelated to tumor PD-L1 expression.

A unified mediation analysis framework for integrative cancer proteogenomics with clinical outcomes.

Multilevel molecular profiling of tumors and the integrative analysis with clinical outcomes have enabled a deeper characterization of cancer treatment. Mediation analysis has emerged as a promising statistical tool to identify and quantify the intermediate mechanisms by which a gene affects an outcome. However, existing methods lack a unified approach to handle various types of outcome variables, making them unsuitable for high-throughput molecular profiling data with highly interconnected variables. We develop a general mediation analysis framework for proteogenomic data that include multiple exposures, multivariate mediators on various scales of effects as appropriate for continuous, binary and survival outcomes. Our estimation method avoids imposing constraints on model parameters such as the rare disease assumption, while accommodating multiple exposures and high-dimensional mediators. We compare our approach to other methods in extensive simulation studies at a range of sample sizes, disease prevalence and number of false mediators. Using kidney renal clear cell carcinoma proteogenomic data, we identify genes that are mediated by proteins and the underlying mechanisms on various survival outcomes that capture short- and long-term disease-specific clinical characteristics. Software is made available in an R package (https://github.com/longjp/mediateR). Supplementary data are available at Bioinformatics online.

Leptomeningeal disease in histone-mutant gliomas.

The 2016 WHO classification described a subtype of midline gliomas harboring histone 3 (H3) K27M alterations, and the 2021 edition added a new subtype of hemispheric diffuse gliomas with H3 G34R/V mutations. The incidence and clinical behavior of leptomeningeal disease (LMD) in these patients is not well defined. Retrospective study of patients with H3-altered gliomas diagnosed from 01/2012 to 08/2021; histone mutations were identified through next-generation sequencing (NGS) of tumor biopsy and/or cerebrospinal fluid (CSF). We identified 42 patients harboring H3 mutations (K27M mutations in 33 patients, G34R/V in 8, and both in one). Median age was 21 (4-70); 27 were male. LMD was diagnosed in 21/42 (50%) patients, corresponding to a 3-year cumulative incidence of 44.7% (95% confidence interval (CI): 26.1%-63.4%) for the K27-mutant group and a 1-year cumulative incidence of 37.5% in the G34-mutant group (95% CI: 0.01%-74.4%; no events after 1 year). Median time from tumor diagnosis to LMD was 12.9 months for H3-K27 patients and 5.6 months for H3-G34 patients. H3 mutation was detected in CSF in all patients with LMD who had NGS (8 H3-K27-mutant patients). In the H3-K27-mutant group, modeled risk of death was increased in patients who developed LMD (hazard ratio: 7.37, 95% CI: 2.98-18.23, P < .0001). In our cohort, 50% of patients developed LMD. Although further studies are needed, CSF ctDNA characterization may aid in identifying molecular tumor profiles in glioma patients with LMD, and neuroaxis imaging and CSF NGS should be considered for early LMD detection.

Prevalence of CKIT and PDGFRA Mutation in Gastrointestinal Stromal Tumors among Filipinos

Background: Gastrointestinal stromal tumors (GIST) is defined as specific, typically kit (CD117)-positive and CKIT or platelet-derived growth factor receptor alpha (PDGFRA) mutation-driven mesenchymal tumors that can occur anywhere in the GI tract. GIST diagnosis relies heavily on immunohistomorphology. However, with the advent of molecular testing, the classification, diagnosis and targeted-therapy for gastrointestinal mesenchymal tumors have been greatly improved. In the Philippines, molecular testing is not yet readily available as in other countries. The local molecular profile of gastrointestinal stromal tumors is a point of investigation as treatment may be more tailored to the patients’ needs. Objective: This study aims to determine the prevalence of CKIT and PDGFRA mutations among formalin-fixed and paraffin embedded gastrointestinal stromal tumors and other gastrointestinal mesenchymal tumors in St. Luke’s Medical Center – Quezon City. Methods: A retrospective cross-sectional study of formalin fixed and paraffin embedded tumor samples diagnosed as Gastrointestinal Stromal Tumor from January 1, 2009 to December 31, 2017 will be analyzed for KIT and PDGFRA mutations. Result: The epidemiology of GIST remains constant in that mean age group is the 5th to 6th decade, with equal gender distribution, and stomach followed by small bowel are the most common sites. Mutational analysis of the GISTs showed predominantly KIT Exon 11 (47.83%) followed by CKIT Exon 9 (13.04%) and PDGFRA Exon 18 (10.87%). For KIT Exon 11, deletion is the most common mutations followed by point mutations. No mutation is detected in 47.83% of GISTs. Conclusion:&nbsp;Mutational analysis for CKIT-PDGFRA is warranted among GIST patients, as it may significantly influence treatment protocol in our patients.

&lt;i&gt;MYCN&lt;/i&gt;-amplified stage 1 neuroblastoma: results of a retrospective multicenter study

Neuroblastoma (NB) is considered as a model of risk-adapted therapy for malignant neoplasms in children. The use of a combination of prognostic factors, such as age, stage of the tumor process, histological variant, and molecular genetic profile of a tumor, makes it possible to identify three risk groups characterized by different intensity of therapy and prognosis. However, the stratification of patients with MYCN-amplified stage 1 NB into a certain risk group and, consequently, the choice of therapeutic tactics for patients in this group vary significantly in the protocols of large cooperative NB study groups. The study was approved by the Independent Ethics Committee and the Scientific Council of the Dmitry Rogachev National Medical Research Center of Pediatric Hematology, Oncology and Immunology of Ministry of Healthcare of the Russian Federation. Clinical data were collected on patients with stage 1 NB according to the INSS system and the presence of the MYCN gene amplification, who received treatment in the period from 2013 to 2021, by sending requests and questionnaires to 77 specialized (oncological) medical institutions in the Russian Federation. At the time of information collection, we made an assessement of the extent of therapy and the status of the disease. A total of 7 patients were registered: 3 boys and 4 girls. The median age at diagnosis was 4.8 months (range 1.5–53.7 months). The primary tumor in all patients was located in the adrenal gland. In 4/7 (57.2%) patients, the 1p deletion was detected, in 6/7 (85.7%) patients who underwent an assessment of the 11q status, no deletion was detected. The therapy was carried out according to the modified NB2004 protocol. Induction therapy, high-dose chemotherapy and autologous hematopoietic stem cell transplantation (auto-HSCT) were received by 6/7 (85.7%) patients. Immunotherapy with anti-GD2 monoclonal antibodies was performed in 1 patient (14.2%). At the time of the manuscript preparation, all patients remained alive without events. One patient with severe congenital heart disease (single ventricle, transposition of the great vessels) has not received systemic therapy and is alive without events at the time of the analysis. Patients with MYCN-amplified stage 1 NB included in this study and stratified into the high-risk group according to the NB2004 protocol, had high survival rates without any adverse events during intensive multimodal therapy. However, given the literature data of international cooperative groups on a decrease in the intensity of therapy in this cohort of patients, for example, refusal of auto-HSCT, as well as a high risk of developing long-term side effects of therapy, including secondary malignant tumors, in patients receiving therapy according to protocols for high-risk groups, it is possible to revise the concept of therapy in this category of patients. This requires a complete registration of all cases of NB in the Russian Federation, the implementation of a diagnostic algorithm, including scintigraphy with 123 I-metaiodobenzylguanidine, review of imaging data, histology and molecular cytogenetic studies in national/federal oncological institutions. With the accumulation of more representative data, auto-HSCT is likely to be omitted in patients with MYCN-amplified stage 1 NB with the absence of other unfavorable biological markers of the disease.

Abstract IA009: Tumor organoids of multi-metastatic colorectal cancer: From research tools to treatment decision tools

Abstract Most patients with colorectal cancer (CRC) do not have an “actionable” target based on molecular tumor profiling. Ex vivo drug sensitivity testing of tumor organoids holds promise as a complementary approach in precision oncology. We have established a pre-clinical pharmacogenomics platform for CRC at Oslo University Hospital. A living biobank of 208 organoids of distinct liver lesions from 100 patients treated by hepatic resection for metastatic CRC was generated in the observational phase of the project. Sensitivity testing of custom drug libraries (n=40-47 drugs) and molecular profiling (targeted DNA sequencing, RNA sequencing, multiplex immunohistochemistry) indicated that the organoids recapitulate well-known pharmacogenomic associations, phenotypic tumor features, and clinical treatment responses (n=35 patients published (ref 1,2)). Evaluation of multiple lesions (n=2-6) per patient suggested diversity among patients in the level of inter-metastatic heterogeneity of drug sensitivities, supporting the need for multi-tumor evaluations in a clinical setting. A collection of CRC cell lines (n=107) was screened for sensitivity to more than 500 drugs and this suggested a molecular basis for response to selected targeted agents, including to PARP inhibition in a small subset of the cell lines (ref 3). Combination screens suggested possibility to overcome acquired resistance to dual BRAF and EGFR blockade in BRAFV600E cells (unpublished data). The living biobank serves as a reference for drug nominations in an ongoing phase II umbrella trial of pharmacogenomics-guided treatment in patients with metastatic CRC (EVIDENT; EU Clinical Trials Register no. 2020-003395-41). Several ad hoc patient cases have been screened as a pilot for the trial, and data of clinical benefit from this functional precision oncology strategy will be presented.

Genomic heterogeneity as a barrier to precision oncology in urothelial cancer.

Precision oncology relies on the accurate molecular characterization of individual patients with cancer at the time of treatment initiation. However, tumor molecular profiles are not static, and cancers continually evolve because of ongoing mutagenesis and clonal selection. Here, we performed genomic analyses of primary tumors, metastases, and plasma collected from individual patients to define the concordance of actionable genomic alterations and to identify drivers of metastatic disease progression. We observed a high degree of discordance of actionable genomic alterations, with 23% discordant between primary and metastatic disease sites. Among chromatin-modifying genes, ARID1A mutations, when discordant, were exclusive to the metastatic tumor samples. Our findings indicate that the high degree of lesion-to-lesion genomic heterogeneity may be a barrier to precision oncology approaches for bladder cancer and that circulating tumor DNA profiling may be preferred to tumor sequencing for a subset of patients.

Opportunities and challenges in interpretable deep learning for drug sensitivity prediction of cancer cells.

In precision oncology, therapy stratification is done based on the patients' tumor molecular profile. Modeling and prediction of the drug response for a given tumor molecular type will further improve therapeutic decision-making for cancer patients. Indeed, deep learning methods hold great potential for drug sensitivity prediction, but a major problem is that these models are black box algorithms and do not clarify the mechanisms of action. This puts a limitation on their clinical implementation. To address this concern, many recent studies attempt to overcome these issues by developing interpretable deep learning methods that facilitate the understanding of the logic behind the drug response prediction. In this review, we discuss strengths and limitations of recent approaches, and suggest future directions that could guide further improvement of interpretable deep learning in drug sensitivity prediction in cancer research.

Diagnostic and Prognostic Biomarkers in Renal Clear Cell Carcinoma.

Renal clear cell carcinoma (ccRCC) comprises over 75% of all renal tumors and arises in the epithelial cells of the proximal convoluted tubule. Molecularly ccRCC is characterized by copy number alterations (CNAs) such as the loss of chromosome 3p and VHL inactivation. Additional driver mutations (SETD2, PBRM1, BAP1, and others) promote genomic instability and tumor cell metastasis through the dysregulation of various metabolic and immune-response pathways. Many researchers identified mutation, gene expression, and proteomic signatures for early diagnosis and prognostics for ccRCC. Despite a tremendous influx of data regarding DNA alterations, gene expression, and protein expression, the incorporation of these analyses for diagnosis and prognosis of RCC into the clinical application has not been implemented yet. In this review, we focused on the molecular changes associated with ccRCC development, along with gene expression and protein signatures, to emphasize the utilization of these molecular profiles in clinical practice. These findings, in the context of machine learning and precision medicine, may help to overcome some of the barriers encountered for implementing molecular profiles of tumors into the diagnosis and treatment of ccRCC.

What is confirmed in the treatment of colon cancer?

Colorectal cancer is the second most common cancer diagnosed in Germany and is the third most frequent cause of cancer-related death in both males and females. The majority of colorectal cancers occur via the adenoma-carcinoma sequence of origin. This means that colorectal cancers can be endoscopically detected in premalignant stages and can be curatively treated within the framework of early detection. Screening colonoscopy and, to a lesser extent, fecal occult blood testing, have led to areduction in the colon cancer-related incidence and mortality. The acceptance and the use of screening colonoscopy should therefore be developed further. Treatment strategies for colorectal cancer are based on TNM staging, supplemented by anatomical and histopathological risk features as well as individual patient characteristics and treatment preferences. The molecular tumor profile is increasingly used to complement decision-making in the surgical, adjuvant and palliative treatment of colorectal cancer. Colon and rectal cancer have many similarities; however, they differ in the preoperative, surgical and adjuvant treatment strategies. This article focuses on colon cancer.