Search

Background: Over 50% of patients with ovarian cancer (OVCA) succumb to their disease within five years from diagnosis, highlighting the need for new approaches to identify therapeutic options. The PARIS® test is a CLIA certified organoid based high throughput drug sensitivity assay that can be used by oncologists to select personalized treatments. In this study our objectives were to: 1) demonstrate the feasibility of organoid based drug testing in the clinical setting; 2) highlight the clinical utility and the predictive value of the PARIS® test; and 3) identify new therapeutic options for OVCA patients. Methods: 93 patients, with a median age at diagnosis of 55 years (range 25-82) were selected as eligible for the PARIS® test at the discretion of the oncologists. The cohort included 74.1% high grade serous tumors, 6.9% low grade serous tumors, 6.0% clear cell carcinomas, 3.4% granulosa cell tumors, 2.5% Mullerian tumors, 1.7% carcinosarcomas and 2.5% fallopian tumors. 16.1% of patients were treatment naïve and 83.9% received one or more previous lines of treatment. 116 patient derived tumor organoids (PDTOs) were derived from surgical excisions, biopsies, or body fluids and subjected to functional testing with a panel of chemo- and targeted therapies that are FDA-approved or in clinical development. Drug responses were evaluated using a proprietary score ranking drugs from exceptional responses to no response. The median turnaround time from biopsy to report delivery was 19 days. Results: The success rate for deriving OVCA organoids was 92%; 81.0% of samples demonstrated exceptional/good response to at least one drug. PDTO drug sensitivities were highly concordant with both retrospective and prospective clinical response. Drugs that frequently showed exceptional/good responses included WEE1 inhibitors (28.4%), HDAC inhibitors (25.0%), and Pan-DNMT inhibitors (11.2%). Further, 12.9% of cases showed responses to the BTK/EGFR inhibitor ibrutinib and 23.3% to HER2 inhibitors lapatinib and neratinib. Two patients with advanced stage, recurrent low grade serous ovarian cancer showed exceptional and durable clinical responses to PARIS® guided treatments, ibrutinib and lapatinib, respectively. Conclusions: This study demonstrates the utility of organoid based drug testing to identify potentially effective drugs and to guide treatment choice for patients with ovarian cancer. Tumor organoid based personalized treatments lead to extraordinary responses in several patients, including a patient in hospice care who responded to ibrutinib monotherapy and is still alive after 27 months. Citation Format: Amanda M. Prechtl, Marwah Al-Aloosi, Payel Chatterjee, Rachele Rosati, Samantha Velazquez, Lauren Appleyard, Grace Durenberger, Cristina Cordero, Robert Diaz, Hallie Swan, Brady Bernard, Siddhartha Mukherjee, Barbara Goff, Elizabeth Swisher, Heidi Gray, Kalyan Banda, Christopher Kemp, Carla Grandori. Tumor organoid based drug sensitivity testing in ovarian cancer identifies exceptional responders to targeted therapies [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 3399.

Pancreatic ductal adenocarcinoma (PDAC) typically presents as metastatic disease at diagnosis and remains refractory to treatment. Next-generation sequencing efforts have described the genomic landscape, classified molecular subtypes, and confirmed frequent alterations in major driver genes, with coexistent alterations in KRAS and TP53 correlating with the highest metastatic burden and poorest outcomes. However, translating this information to guide therapy remains a challenge. By integrating genomic analysis with an arrayed RNAi druggable genome screen and drug profiling of a KRAS/TP53 mutant PDAC cell line derived from a patient-derived xenograft (PDCL), we identified numerous targetable vulnerabilities that reveal both known and novel functional aspects of pancreatic cancer biology. A dependence on the general transcription and DNA repair factor TFIIH complex, particularly the XPB subunit and the CAK complex (CDK7/CyclinH/MAT1), was identified and further validated utilizing a panel of genomically subtyped KRAS mutant PDCLs. TFIIH function was inhibited with a covalent inhibitor of CDK7/12/13 (THZ1), a CDK7/CDK9 kinase inhibitor (SNS-032), and a covalent inhibitor of XPB (triptolide), which led to disruption of the protein stability of the RNA polymerase II subunit RPB1. Loss of RPB1 following TFIIH inhibition led to downregulation of key transcriptional effectors of KRAS-mutant signaling and negative regulators of apoptosis, including MCL1, XIAP, and CFLAR, initiating caspase-8 dependent apoptosis. All three drugs exhibited synergy in combination with a multivalent TRAIL, effectively reinforcing mitochondrial-mediated apoptosis. These findings present a novel combination therapy, with direct translational implications for current clinical trials on metastatic pancreatic cancer patients. Significance: This study utilizes functional genetic and pharmacological profiling of KRAS-mutant pancreatic adenocarcinoma to identify therapeutic strategies and finds that TFIIH inhibition synergizes with TRAIL to induce apoptosis in KRAS-driven pancreatic cancer.

Kinase inhibitors are a critical tool for cancer treatment, but their efficacy is limited by resistance mechanisms. In thyroid and lung cancer RET gene rearrangements result in constitutive MAPK pathway activation and drive malignancy. While treatment with new selective RET inhibitors has been associated with significant clinical responses, a majority of patients experience a partial response or disease stabilization as their best clinical outcome. Resistance to RET inhibitors has emerged as a clinical problem requiring new treatment strategies. Using a combination of drug screening, proteomic, and biochemical profiling we identified a strategy to overcome adaptive resistance to RET inhibitors in human thyroid cancer cell lines and vertebrate animal models. The identification of alternative signaling pathways that reactivates ERK signaling as a mechanism of resistance to RET inhibitors provides an opportunity to anticipate resistance to selective RET inhibitors and use combination therapy that leads to more significant and durable anti-tumor responses in patients with RET rearranged cancers. Citation Format: Renuka Raman, Jacques Villefranc, Timothy Ullmann, Jessica Thiesmeyer, Viviana Anelli, Chantal Pauli, Rohan Bareja, Kenneth Wha Eng, Princesca Dorsaint, David Wilkes, Shaham Beg, Reid Shaw, Michael Churchill, Naveen Gumpeni, Theresa Scognamiglio, Mark Rubin, Carla Grandori, Juan Mosquera, Juan Mosquera, Juan Mosquera, Noah Dephoure, Andrea Sboner, Olivier Elemento, Yariv Houvras. Uncovering the mechanism of adaptive resistance to RET inhibitors in RET rearranged thyroid cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 1434.

Background: Clinically approved targeted therapies for ovarian cancer patients are currently limited to PARP inhibitors and bevacizumab. To improve treatment outcomes, a precision medicine approach is crucial to match effective drugs to patient-specific genetic features and vulnerabilities. This study aimed to: (1) demonstrate the feasibility of performing high-throughput drug screens on fresh patient organoids using a CLIA-approved assay, (2) assess concordance of responses with genomic and clinical information, and (3) reveal novel biomarkers of response to approved/experimental drugs and insights into ovarian cancer biology. Methods: From 2015 to 2020, 76 ovarian tumor samples were collected from 60 patients. To date, 50 evaluable samples were successfully screened at SEngine Precision Medicine. Drugs tested include a range of chemotherapies and targeted therapies that are FDA-approved or in clinical development, with an average of 61 drugs screened per assay (range: 6-135). Somatic and/or germline DNA sequencing is currently available for 30 samples. Sample collection, screening, and sequencing is ongoing. Results: The cohort included ovarian cancer patients with high grade serous (65%), low grade serous (7%), unknown grade serous (7%), clear cell (7%), carcinosarcoma (5%), stromal (5%), endometrioid (2%) subtypes, and one of unknown pathology. Whole genome mutational analysis of 5 tumor-derived organoids and their original tumors demonstrated a high degree of similarity between the tumor-organoid pairs. We present prospective and retrospective evidence from at least 18 cases that organoid drug screening can accurately predict clinical response to chemotherapy and targeted therapies. We also report a patient with platinum resistant serous carcinoma who responded to ibrutinib treatment after screening identified the drug as having excellent response in this patient's organoids. Three months into treatment, the patient's CA125 level was reduced from 250 to 125U/ml. In samples with available genomic information, we demonstrate high concordance between drug sensitivity and known biomarkers, e.g. 83% of samples with known BRCA1/BRCA2 mutation or high HRDetect showed sensitivity to a PARP inhibitor. Further, organoid screening can identify unique targets for every patient beyond established genomic biomarkers. Subsets of patients responded exceptionally to BET (41%), HDAC (28%), WEE1 (24%), and BTK (11%) inhibitors, indicating potential for these targeted therapies in ovarian cancer. Conclusions: The genomic and histopathological heterogeneity of ovarian cancer points to a need to evolve and prioritize the personalization of treatment. Our data demonstrates the utility of organoid based drug screening to nominate therapeutic options for individual patients with or without known genomic biomarkers. Citation Format: Goldie Lui, Anne Richardson, Payel Chatterjee, Madison Pollastro, Mia Lints, Danielle Peretti, Rachele Rosati, Lauren Appleyard, Grace Durenberger, Robert Diaz, Kay Gurley, Isabella Stork, Adam Whitney, Katannya Kapeli, Hallie Swan, Yasin Memari, Helen Davies, Serena Nik-Zainal, Kalyan Banda, Heidi Gray, Barbara Goff, Elizabeth Swisher, Astrid Margossian, Christopher Kemp, Carla Grandori. Functional drug screening of organoids from ovarian cancer patients demonstrates clinical and genomic concordance and identifies novel therapeutic vulnerabilities [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 534.

The goal of precision medicine is to match the right drug to the right patient. However, every individual cancer carries a unique and complex mosaic of genetic and molecular changes making it difficult to identify the right drug based solely on genomic analysis. We developed a CLIA-certified functional drug assay (PARIS® test) for solid tumors which provides an actionable report of tumor derived organoid sensitivities to targeted, endocrine and chemotherapy agents as a tool for clinical therapeutic decisions. Objectives:1.To establish the concordance between organoid drug sensitivity with well-known genomic or immunohistochemical IHC biomarkers 2.To correlate organoid drug sensitivity with clinical outcomes.Methods: From 2015 to 2020, organoids from 410 tumor samples were subjected to functional testing at SEngine Precision Medicine, including 61 breast tumor samples from 48 patients. Fresh samples of tumor cells from core biopsies, surgical excisions, or from fluids arrived 90%.Conclusions: Organoid based drug testing exhibits strong concordance with genomic or IHC biomarkers and clinical response. In addition, functional testing identifies candidate therapies in patients lacking biomarkers and can nominate variants of unknown significance as candidate biomarkers. This study highlights the utility of functional assays to support clinical decision making in a genetically heterogenous cancer such as breast cancer. Citation Format: Astrid Margossian, Anne Richardson, Madison Pollastro, Michael Churchill, Franz Schaub, Shalini Pereira, Payel Chatterjee, Rachele Rosati, Lauren Appleyard, Grace Durenberger, Alex Federation, G. Adam Whitney, Hallie Swan, Trevor Ainge, Robert Diaz, Natasha Hunter, Eric Gamboa, Chris Kemp, Vijayakrishna Gadi, Carla Grandori. Clinical and genomic correlation of a CLIA certified organoid based functional test in breast cancer patients [abstract]. In: Proceedings of the 2020 San Antonio Breast Cancer Virtual Symposium; 2020 Dec 8-11; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2021;81(4 Suppl):Abstract nr PS4-01.

Introduction: It is difficult to find targeted therapies that work across the genetically diverse Cholangiocarcinoma (CCA) patient population; genomic sequencing alone has not resulted in substantial changes in treatment choices for this rare cancer. We reasoned that CCA patients would benefit from a highly personalized diagnostic test, where patient-derived tumor organoids are tested with a library of targeted drugs in a high-throughput functional assay. Methods: Seventeen patients with CCA were enrolled in an IRB-approved ongoing study from 2017 to 2019; biopsies were collected from surgical resection and shipped on ice overnight to a CLIA-certified laboratory (SEngine Precision Medicine) for organoid derivation and high throughput drug testing. Tumor organoids were challenged with up to 120 FDA-approved and investigational drugs to obtain 6-dose, 3-log response curves. Drug responses were benchmarked against an in-house database of pan-cancer patients to concurrently assess both sensitivity and the uniqueness of each patient's response. Clinical history and genomic information were collected from each patient for retrospective and prospective analysis. Results: 14/19 (74%) samples were successfully derived from 17 patients and screened. Patient age ranged from 36 to 73 (mean=51). The 120-drug library was adapted based on total viable tumor cells per sample, resulting in average of 71 drugs per screen. Top scoring drugs were Ceritinib, MK2206, Belinostat, Taselisib, AZD4547, Everolimus and Poziotinib. In general, the drug sensitivity highlighted the following pathways: AKT, ALK, FGFR, PI3K and EGFR. Drugs were selected per patient and classified as either good or exceptional response and correlated with mutations in known genomic biomarkers in 7 of 12 drug/target pairs. Retrospective analysis demonstrated 100% concordance (n=3) between ex vivo drug resistance and prior treatments. We observed multiple mutation/drug pairs which showed resistance in our functional screen. We achieved a high success rate and clinically relevant turnaround times, from biopsy to final report generation average 17 days (range 9 to 35). Conclusion: This study shows the feasibility of functional testing of organoids derived from CCA patients in a CLIA-certified diagnostic test. While these results correlate well with genomically predicted drug sensitivities, all patients showed additional drug sensitivities beyond those predicted by genomics offering patients additional potential treatment options. This study highlights the importance of functional data in a genetically heterogenous tumor type such as CCA. Each patient represents a mosaic of drug sensitivities, reflecting unique combinations of genetic and epigenetic alterations, and this test can identify the personalized drug combination for them. Citation Format: Astrid Margossian, Franz X. Schaub, Annie Richardson, Michael Churchill, Rachele Rosati, Alexandra Dullea, G Adam Whitney, Hallie Swan, Robert Diaz, Lauren Appleyard, Vijaykrishna Gadi, Ali Zarrinpar, Milind Javle, Christopher Kemp, Shalini Pereira, Carla Grandori. Organoid based functional test to predict personalized treatment in cholangiocarcinoma [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 818.

We hypothesize that new insights into how cancers progress and respond to treatment will come from clinical trials that i) extensively characterize the molecular features of a patient’s cancer; ii) use results to predict drug susceptibilities; iii) treat in accordance with these predictions; and iv) learn from individual patient outcomes to iterate and improve over time. To investigate the feasibility of this type of clinical study, we launched the "Intensive Trial of OMics in Cancer" (ITOMIC) for patients with metastatic triple negative breast cancer (TNBC) (Clinicaltrials.gov ID: NCT01957514). Eligible patients have metastatic TNBC, are platinum-naive, and are scheduled to receive Cisplatin. Biopsies are performed under carefully controlled conditions prior to Cisplatin – starting all subjects on a common treatment path, and uncoupling the time needed for specimen analysis from immediate therapy. Biopsies are repeated upon completion of Cisplatin and following subsequent therapies. A subset of specimens is chosen for whole Exome Sequencing, deep sequencing of a panel of cancer associated genes, and RNA-sequencing. De-identified results are placed on a web-based server for analysis and discussed at a meeting of the ITOMIC tumor board. A report describing results and potential therapies is provided to the subject’s oncologist. Treatment decisions are left to the discretion of the oncologist. If a decision is taken to pursue treatments identified in our report we offer assistance in accessing those treatments. Ten patients have been screened and seven have enrolled. Subjects range in age from 40 to 77 years and all but one has received extensive prior treatment for metastatic TNBC. All seven underwent an initial set of biopsies, targeting between two and five metastatic sites. For most metastatic sites, multiple core needle passes are performed. All subjects tolerated the biopsies well without significant adverse events, and all started treatment with Cisplatin. Three subjects completed Cisplatin and underwent a second round of biopsies. Potential targets for therapy were identified in 5 of the first 6 subjects, and three subjects have received four predicted therapies: 1) a patient with somatic loss of BRCA1 and two linked FGFR2 activating mutations, who was treated first with Veliparib through a single-patient IND and then switched to Ponatinib which produced a partial response; 2) a patient with a novel missense ROS1 mutation treated with crizotinib; and 3) a patient with CYP3A4 copy gain treated with cyclophosphamide. Conclusion: Our early experience indicates that this approach is feasible and may increase the efficiency of learning from patients with advanced cancer. Citation Format: C Anthony Blau, Colin Pritchard, Michael O Dorschner, Sibel Blau, Brigham Mecham, Elisabeth Mahen, VK Gadi, Wayne Monsky, Kimberly Burton, Arturo Ramirez, Jackie Stilwell, Eric Kladjian, Carol Collins, Jeannine S McCune, William S Noble, Julie Gralow, Frank Senecal, Linda Dhaene, Nicole Kuderer, Jennifer Specht, Chaozhong Song, Carla Grandori, Nathan Price, Mary Goldman, Aime Radenbaugh, David Haussler, Jingchun Zhu. Assessing the safety and feasibility of efficient hypothesis testing in patients with metastatic triple negative breast cancer [abstract]. In: Proceedings of the Thirty-Seventh Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2014 Dec 9-13; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2015;75(9 Suppl):Abstract nr P4-08-01.

The pool of targeted therapies is expanding steadily providing novel treatment opportunities. However, the process of expanding indications and including additional cancer types after initial approval is slow. Here we present an in vitro clinical trial platform using the data from the first CLIA-approved Precision Medicine Platform using high-throughput screening (P.A.R.I.S. Assay, SEngine Precision Medicine, Seattle, WA). Briefly, patient-derived tumor cells from biopsies are expanded as living organoids, and functionally evaluated using a high-throughput drug screen. The drug library currently includes FDA approved and experimental drugs. Results are integrated with known genomic information and reported to the clinician to highlight treatment options of an individual patient. Combining this data from multiple patients enables the discovery of novel indications and biomarkers. To date we have acquired data from a wide range of tumor types including hard-to-treat cancers, rare tumor types and tumors from heavily pretreated patients. In addition, we also have collected data from more than 100 patient derived cell lines. Using the data analysis platform developed in R, the drug responses across all patient screens of a given drug can be analyzed to find common indicators of sensitivities. For instance, a BRD4 inhibitor (CPI-203), which was tested on 8 different primary tumor samples, was selectively active in a subset of the patients. Interestingly, most of the responding tumor types are ovarian and breast cancer cases with mutations in BRCA1 or EZH2. However, there are currently no clinical trials investigating the more developed BRD4 inhibitor CPI-0610 for breast or ovarian cancer. Non-responders, on the other hand, are spread between multiple cancer types. The number of patients receiving the P.A.R.I.S. test is steadily increasing and these patient cases contribute to our knowledge base. As this test is used to inform clinicians about potential treatment opportunities, we are also accumulating information on how these patients respond to treatment. Overall, this platform will provide real world data for optimal drug combinations and novel biomarkers to quickly expand indications of existing drugs. Citation Format: Franz X. Schaub, Michael J. Churchill, Hallie A. Swan, Rachele Rosati, Roland M. Watt, Reid C. Shaw, Stephanie A. Murphy, Robert L. Diaz, Shalini C. Pereira, Carla Grandori. Accelerating drug development with a CLIA approved functional test using patient derived organoids [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 582.

Background: Metastatic disease in ovarian cancer is difficult to treat and patients often exhaust standard-of-care regimens. To get a better understanding of potential treatments, genomic data is used in some cases. However, this only points to therapeutic options in a minority of cases highlighting an urgent need to develop assays to identify potential therapies. Current functional tests are limited in the number of tested drugs. Here we present the first CLIA certified high-throughput functional assay employing organoid cultures derived from primary patient specimens to directly aid oncologists for personalized treatment selection (P.A.R.I.S. Assay, SEngine Precision Medicine, Seattle, WA). Experimental Procedures: Organoids are treated for 6 days with a library of 123 clinically relevant drugs. Compounds are evaluated at a multi-dose response curve and ranked by SPM score, which weights both the sensitivity (degree of cell death) and specificity, which compares the patient's tumor cells response to the drug relative to all prior patients. The results are further integrated with genomic data and reported to the clinician to highlight treatment options. SEngine has performed >150 drug screens and established high reproducibility including multiple ovarian cancer cases from either ascites, surgical or core biopsies. Results: Here we present two n-of-one patient studies. Patient 1 is a 48 year old woman with positive family history who was diagnosed with late stage serous ovarian cancer. SEngine generated cancer organoids and performed high-throughput screening with a panel of 123 drugs. Sensitivities to PARP inhibitors prompted SEngine to advise for germline testing which revealed a BRCA1 and TP53 mutation. The cells were also uniquely sensitive to paclitaxel and the combination with carboplatin resulted in remission indicating concordance with clinical data. In addition, highlighting the importance of functional screening, a unique response to a sub-group of EGFR inhibitors was identified, of potential consideration in case of recurrence. The second case also has a family history of ovarian cancer, but no BRCA mutations were detected. Genomic data indicated FGF6 and FGF23 amplification which directly corresponded to a unique sensitivity to one FGFR inhibitor (AZD-4547). The organoids were also resistant to PARP inhibitors, consistent with the absence of BRCA mutations. Impact: We developed a robust ex vivo screening platform to objectively quantify patient specific sensitivity to a panel of 123 oncology drugs. SEngine is compiling a registry capturing clinical data, outcome following the P.A.R.I.S. test as well as genomic data. The power of high throughput technology and organoid isolation will enable the rapid selection of optimal individualized therapies as single agents or in combination. Citation Format: Hallie A. Swan, Rachele Rosati, Caroline Bridgwater, Michael J. Churchill, Roland M. Watt, Reid C. Shaw, Stephanie A. Murphy, Robert L. Diaz, Shalini C. Pereira, Franz X. Schaub, Carla Grandori. Personalized medicine: A CLIA-certified high-throughput drug screening platform for ovarian cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 1619.

In June 2017, SEngine Precision Medicine (Seattle, WA USA) obtained CLIA certification for the first high-complexity and high throughput organotypic chemosensitivity test, P.A.R.I.S., thus opening its doors to oncologists and pharmaceutical companies that may consider personalization of cancer treatments in their practice or in the clinical trial setting. Just as Paris slayed the invulnerable Achilles by aiming specifically at its weaknesses, the PARIS test challenges tumor cells with a library of 137 clinically actionable targeted drugs to identify sensitivity and resistance patterns unique to each individual cancer in an unbiased manner. Here we will present an overview of the approach that has led to CLIA certification, including novel cell culture techniques and analysis metrics. We will also highlight insights gained by comparative analysis across a multitude of cancer samples. SEngine has derived and expanded primary cancer cells from a cohort of over 100 patient derived specimens. Patients include many who have been heavily pre-treated with prior chemotherapy and several with rare cancers, including adrenocortical carcinoma, neuroblastoma, leiomyosarcoma and cholangiocarcinoma. Standard operating procedures have been established for transport of specimens as well as primary cultures in three dimensional organotypic conditions. Our analysis pipeline evaluates the multi-dose response of a given patient's tumor to each drug in our library and compares those responses to all prior patients. This establishes not only functional sensitivity but also the uniqueness of the patient's response. We calculate the SPM score, a proprietary ranking metric weighing both the sensitivity and uniqueness of the response. The results from the P.A.R.I.S. test are highly reproducible (Spearman rank correlation coefficients: 0.95, 0.88 technical and biological replicates respectively). We perform bi-annual proficiency validation of our platform with another CLIA-certified laboratory. Ten samples tested demonstrated greater than 75% concordance between responses. The data obtained are visualized by an in-house application suite (SEngine Medicine APP), consisting of interactive charts, statistical analysis, and reactive reports which can be shared with oncologists and investigators. A cohort of 50 samples was employed to demonstrate sensitivities or resistances concordant with those predicted by the patient's genomic aberrations and/or prior clinical responses.In summary, we envision that the P.A.R.I.S. test will be essential for patient's selection while developing novel targeted drugs, providing crucial functional information much needed to decode the complexity of individual cancer genomics. Citation Format: Hallie A. Swan, Michael J. Churchill, Rachele Rosati, Franz X. Schaub, Reid C. Shaw, Roland M. Watt, Eduardo Mendez, Christopher J. Kemp, Vijayakrishna K. Gadi, Stephanie A. Murphy, Brady Bernard, Robert L. Diaz, Shalini C. Pereira, Carla Grandori. Personalization of cancer treatments with a CLIA-certified high-complexity and high-throughput drug sensitivity test [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 1605.

Goal: Neuroblastoma (NBL) is the second most frequent pediatric solid tumor. We have seen an increase in survival for children with NBL, however treatment still consists of many cycles of chemotherapy and radiation. We reasoned, that in vitro high throughput drug screening (HTDS) of patient derived organoids using a library of targeted therapies, would highlight novel targeted and personalized treatment options. Methods: We optimized an in vitro 3D culture for high-risk NBL samples, initially established as patient derived xenografts (PDX) and screened a total of 10 patient samples with genomic information using two different drug libraries: Informer Set (320 drugs) compiled by the Cancer Target Discovery and Development network which is focused on cancer pathways and SEngine Library (123 drugs) focused on clinically relevant therapies. The screen was performed in a 384 well-format and cell viability was measured. Organoids were characterized by immunohistochemistry and genomic and pathway alterations were confirmed. Results: Here we highlighted unique results from 4 cases. The first sample showed sensitivities to CDK inhibitors in concordance with amplification of CDK4. However, for this patient additional drug sensitivities were not directly linked to a genetic alteration, such as AZD7762 (CHK inhibitor) and MK-1775 (WEE1 inhibitor) indicating sensitivities in the cell-cycle checkpoints, and an exceptional response to Enzalutamide, an androgen receptor antagonist. The second case had unique sensitivities to IGF1R and Insulin Receptor inhibitors (Linsitinib and BMS-754807) indicating a specific addiction to this growth-factors/survival signal. The third sample had multiple sensitivities linked to genomic alterations, such as Dabrafenib, concordant with the presence of a BRAF activating mutation. In addition, sensitivity to Alisertib was detected as previously reported for MYCN amplified NBL, but this was not true for other MYCN amplified cases. Interestingly, this third sample also showed profound sensitivities to PARP inhibitors, most likely related to a mutation in the BRP1 gene a BRCA1 interacting protein gene. The fourth case showed a strong sensitivity to nutlin-3 (MDM2 inhibitor) and Oxaliplatin. Conclusion: A newly established NBL 3D patient organoid culture system coupled with personalized HTDS and genomic analysis offers an effective platform for the discovery of novel personalized therapies and drug combinations tailored to individual patients. In this NBL cohort, we discovered both drugs predicted by the patient-specific genomic alterations, as well as novel patient-specific drug sensitivities without a known genomic marker. More in depth genomic analysis is underway, but epigenetic alterations could explain these exceptional responses. In summary, this investigational study of 10 NBL highlights the need of personalization of targeted therapies in NBL. Citation Format: Rachele Rosati, Hallie A. Swan, Laura Scolaro, Kateryna Krytska, John M. Maris, Franz X. Schaub, Carla Grandori. Patient derived organoids to guide personalized neuroblastoma treatment [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 2625.

Recurrent metastatic breast cancer is difficult to treat and patients often exhaust standard-of-care regimens. P.A.R.I.S. is a CLIA certified high throughput chemosensitivity test developed by SEngine Precision Medicine. (Seattle, WA USA) Tumor cells derived from patient breast tissue are expanded as mammospheres, and challenged with a library of 123 clinically actionable, targeted drugs. These results are integrated with patient genomic data and reported to the oncologist to prioritize treatment options. To date, we have performed over 100 such drug screens using low passage patient-derived tumor cells in 70 independent replicate pairs. Of 18 viable primary breast tumor samples received to date, we have successfully derived organotypic cultures in 13 cases. (72%) Of these, 7 were challenged to the full 123 agent library. Three of the seven cases were HER2 positive and demonstrated clear sensitivity to HER2 kinase inhibitors. Overall, patients can be separated into two groups, characterized by families of drug sensitivities. Patients in group one respond preferentially to inhibitors of signaling pathways such as IGF1R, ALK, FGFR and EGFR. Patients in the second group responded to therapies targeting epigenetic modifiers, cell cycle control and apoptosis. Presented here is a premenopausal female, diagnosed in 2010 as pT2N0M0 (stage IIA), ER+ PR+ HER2+. This patient underwent surgery and received adjuvant chemotherapy, followed by tamoxifen. Upon progression to metastatic disease, biopsy of her recurrent lesion showed loss of ER and PR expression, but remained positive for HER-2 amplification. Sequencing of this biopsy exposed a PIK3CA G106R mutation and amplification of unknown significance. Tumor cells were isolated from ascites, cultured and the P.A.R.I.S. test was performed. Patient derived tumor cells were sensitive to HER2 targeting agents, concordant with the patient's HER2 amplification. Of these, the non-reversible inhibitor neratinib exhibited a significantly enhanced effect and achieved maximum growth inhibition of 80% over the duration of the assay. Two different AKT inhibitors, afuresertib and AZD5363, demonstrated activity in the patient cells consistent with the activating PIK3CA mutation. Also consistent with the loss of ER/PR expression, the patient derived tumor cells no longer responded to anti-estrogen therapies nor to chemotherapies the patient had been refractory to prior: cyclophosphamide, docetaxel, doxorubicin, gemcitabine, and paclitaxel.The PARIS assay is a robust ex vivo screening platform to objectively quantify patient specific sensitivities to a panel of 123 clinically actionable oncology drugs. We show here concordance with both genomic data and prior treatments the patient received in the course of disease however the same approach may be used to successfully identify unique sensitivities in the absence of established biomarkers. Citation Format: Michael J. Churchill, Franz X. Schaub, Hallie A. Swan, Rachele Rosati, Fengting Yan, Reid C. Shaw, Kay E. Gurley, Robert L. Diaz, Shalini C. Pereira, Carla Grandori, Christopher J. Kemp, Vijayakrishna K. Gadi. Patient derived tumor organoids identify actionable targets in heavily pretreated metastatic breast cancer patients [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 1603.

Background: The development of neuroendocrine prostate cancer (NEPC) is one mechanism of treatment resistance to androgen receptor (AR)-targeted therapies for a subset of patients with advanced prostate cancer. This is associated with transition from a prostate adenocarcinoma to small cell/NEPC histology, low AR signaling signaling, and expression of neuroendocrine markers as Chromogranin A (CGHA), Synaphophysin (SYP) and CD56). Patient derived preclinical models recapitulating the NEPC phenotype may be used to address NEPC pathogenesis and test emerging therapeutic targets. Methods: Tumor organoids were developed according to protocols previously described (Gao et al, Cell 2015). Briefly the tissue biopsies (liver and bone biopsy) were washed, enzymatically digested and then seeded in Matrigel (BD) droplets. Organoids were characterized at genomic (WES), RNA and protein level (IHC) to confirm the expression of specific markers. Lentiviral infections were performed using shRNAs against EZH2 to knock down EZH2 in organoids. Organoids were also subcutaneously injected in NSG mice to generate patient derived xenografts (PDXs) for drug treatment in vivo. Results: We developed and characterized two NEPC tumor organoids from tumor biopsies (liver and bone) of two patients both in vitro and in vivo (as PDXs). NEPC tumor organoid models retained the molecular and histological characteristic of their matched patient samples. We successfully manipulated the activity of the histone methyltransferase EZH2 by using a catalytic inhibitor and its expression by infecting organoids with shEZH2. We showed that the absence of EZH2 affects the expression of neuroendocrine-associated programs as stem cell and neuronal pathway. Moreover treatment with EZH2 inhibitor decreased tumor organoids viability and PDXs tumor volume. Drug screening approaches on NEPC organoids were used to discovery novel drug targets and combinations that could potentially benefit NEPC patients. Top single agent hits included previously identified targets such as EZH2, AURKA, as well as novel synergies. Conclusions NEPC patient tumor organoids are clinically relevant tumor models to study the NEPC phenotype in advanced prostate cancer and may be used to elucidate novel drug targets. Citation Format: Loredana Puca, Rohan Bareja, Reid Shaw, Wouter Karthaus, Dong Gao, Chantal Pauli, Juan Miguel Mosquera, Joanna Cyrta, Rachele Rosati, Rema Rao, Andrea Sboner, Carla Grandori, Giorgio Inghirami, Yu Chen, Mark A. Rubin, Himisha Beltran. Patient-derived tumor organoids of neuroendocrine prostate cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 992. doi:10.1158/1538-7445.AM2017-992

Introduction: Identifying molecular aberrations that drive carcinogenesis remains a daunting problem. Uncovering important combinations of such aberrations, e.g. overexpression of a gene in concordance with copy number amplification, pose additional challenges. Publicly available large-scale cell line panels profiled for drug response often do not reflect the true tumor heterogeneity and therefore predictions made solely based on such data may be difficult to translate into therapeutic targets. Integrating predicted associations with drug sensitivity (resistance) on one hand, with siRNA screens with the same drug (patient-derived samples and cell lines) on the other hand, should allow for the identification of therapeutically relevant molecular traits. Methods: We propose a novel Bayesian model for multi-task learning with a gene-wise prior, which accounts for multiple aberrations in a given gene (expression, mutation and copy number). Our method allows for the simultaneous learning of drug response for multiple drugs in a given panel (N_Sanger=138, N_CCLE=24 drugs). Our model aims to identifying sets of important aberrations, which individually occur at low frequencies. To train and evaluate the performance of our model we use the CCLE and Sanger datasets. Further, we use the Cisplatin and Doxorubicin drug responses from the Sanger dataset to rank genome-wise aberrations to find features best predictive of response to the chemotherapeutic agents. We then consider our findings in the context of high-throughput siRNA synthetic lethal screens in head and neck, ovarian and breast cancer with Cisplatin and Doxorubicin in patient-derived samples and cell lines. Results: Our approach improves the quality of drug sensitivity prediction (Pearson correlation in five-fold cross validation) over methods that model individual drug responses (e.g. Elastic Net). It further allows for the subtyping of drugs by target/mechanism of action commonalities, which we recover closely for drugs with known mechanism of action. Our model recovers reasonably the targets of consensus drug interactions (e.g. Erlotinib and EGFR) and in addition identifies other known associations currently undergoing preclinical/clinical validation (i.e. Cisplatin and JAK2 mutation, and RUNX3 expression and Doxorubicin). Integrated analysis of the results from our predictive model in conjunction with the siRNA screens are currently under investigation. Conclusions: We present an integrated approach that combines a novel Bayesian multi-task learning model with high-throughput siRNA screens. Our approach aims to uncover sets of important aberrations and allows for the subtyping of drugs based on similarities in targets and mechanisms of action. We integrate our results with high-throughput RNAi experiments to identify synthetic lethal events in specific therapeutic context. Citation Format: Olga H. Nikolova, Mehmet Gönen, Rodrigo Dienstmann, In Sock Jang, Russell Moser, Silvia Cermelli, Chang Xu, Ryan M. Mitchell, Eduardo Mendez, Carla Grandori, Christopher Kemp, Stephen Friend, Justin Guinney, Adam Margolin. Integrated computational cell-line modeling of drug sensitivity and high-throughput siRNA screening reveals novel molecular biomarkers for conventional chemotherapy. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 5323. doi:10.1158/1538-7445.AM2014-5323

∼ 30 % of Neuroblastoma harbor amplifications of the MYCN oncogene, which is the strongest molecular marker of poor prognosis. However, a subset of high-risk neuroblastoma does not carry alterations of the MYCN gene and the molecular pathways underlying their aggressive behavior remains to be elucidated. We have utilized a functional genomics approach with both genetically defined cell systems, as well as molecularly defined neuroblastoma cell lines, to identify critical survival pathways that distinguish MYCN-amplified versus non-amplified neuroblastoma. The results of high throughput siRNA screens targeting the druggable genome will be presented and discussed in light of public microarray data and validation experiments carried out in our laboratory. These findings revealed that different chromatin-remodeling genes, DNA-repair and developmental pathways differentiate MYCN amplified neuroblastoma versus non-amplified, thus highlighting specific new therapeutic targets associated with MYCN status as well as other molecular markers. In addition to pinpointing novel therapeutic targets for high-risk neuroblastoma, our study also functional defines signaling pathways activated in high-risk neruoblastoma and could be utilized to guide rational therapeutic choices. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 5139. doi:1538-7445.AM2012-5139