Search

XLSX file 452K, DSRT Oncology drug collection

PDF file 400K, Supplementary Figures 1-9; Supplementary Tables 4-9 and Supplementary Methods

Acute myeloid leukemia (AML) is an aggressive disease of clonal hematopoietic progenitor cells. Here, we applied ex-vivo drug sensitivity and resistance testing on AML patient cells with 362 emerging and 153 approved cancer drugs together with genomic and transcriptomic profiling to identify and tailor therapies for patients with advanced disease. Ex-vivo testing with freshly isolated patient cells revealed cancer-specific efficacies of approved drugs in 97% of the 164 patient cases, including 47% of the cases with no actionable driver mutations. We identified 142 statistically significant associations between drug responses and somatic mutations, including increased sensitivity to JAK inhibitors in patients with NPM1 mutations. Transcriptomic profiles predicted drug responses better than genomics and helped to identify additional response markers, especially beyond mutations. For example, overexpression of HOX family genes was associated with sensitivity to JAK inhibitors in patients with NPM1 mutation. In a prospective study, we translated the functional drug response and molecular profile data to the clinic and suggested tailored therapy with targeted drugs for 26 relapsed or refractory AML patients. In an observational intervention study, acting on these recommendations resulted in a temporary complete clinical remission or leukemia-free state in 39% of the cases. In summary, we conclude that ex-vivo testing of drugs on patient AML cells i) revealed clinically actionable drug efficacies in almost all AML patients, including patients with no actionable mutations, ii) predicted cases with actionable driver mutations with no pharmacological dependency on the target, and iii) enabled real-time tailoring of therapy with 39% clinical response rate in chemorefractory advanced AML. Taken together, we believe this real-time systems medicine approach could become a powerful strategy for tailoring therapies for individual patients in the future. Citation Format: Disha Malani, Ashwni Kumar, Bhagwan Yadav, Mika Kontro, Swapnil Potdar, Oscar Bruck, Säri Kytölä, Jani Saarela, Samuli Eldfors, Riikka Karjalainen, Muntasir M. Majumder, Imre Västrik, Pekka Ellonen, Matti Kankainen, Minna Suvela, Siv Knappila, Alun Parson, Aino Palva, Pirkko Mattila, Evgeny Kulesskiy, Laura Turunen, Karoliina Laamanen, Elina Lehtinen, Maria Nurmi, Katja Suomi, Astrid Muruimägi, Bjorn T. Gjertsen, Satu Mustjoki, Simon Anders, Maija Wolf, Tero Aittokallio, Krister Wennerberg, Caroline Heckman, Kimmo Porkka, Olli Kallioniemi. Precision systems medicine in acute myeloid leukemia: real-time translation of tailored therapeutic opportunities arising from ex-vivo drug sensitivity testing and molecular profiling [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 458.

We present an individualized systems medicine (ISM) approach to optimize cancer drug therapies one patient at a time. ISM is based on (i) molecular profiling and ex vivo drug sensitivity and resistance testing (DSRT) of patients' cancer cells to 187 oncology drugs, (ii) clinical implementation of therapies predicted to be effective, and (iii) studying consecutive samples from the treated patients to understand the basis of resistance. Here, application of ISM to 28 samples from patients with acute myeloid leukemia (AML) uncovered five major taxonomic drug-response subtypes based on DSRT profiles, some with distinct genomic features (e.g., MLL gene fusions in subgroup IV and FLT3-ITD mutations in subgroup V). Therapy based on DSRT resulted in several clinical responses. After progression under DSRT-guided therapies, AML cells displayed significant clonal evolution and novel genomic changes potentially explaining resistance, whereas ex vivo DSRT data showed resistance to the clinically applied drugs and new vulnerabilities to previously ineffective drugs. Significance: Here, we demonstrate an ISM strategy to optimize safe and effective personalized cancer therapies for individual patients as well as to understand and predict disease evolution and the next line of therapy. This approach could facilitate systematic drug repositioning of approved targeted drugs as well as help to prioritize and de-risk emerging drugs for clinical testing. Cancer Discov; 3(12); 1416–29. ©2013 AACR. See related commentary by Hourigan and Karp, p. 1336 This article is highlighted in the In This Issue feature, p. 1317

Introduction: Conventional cytotoxic chemotherapy regimens for adult acute myeloid leukemia (AML) are effective in curing less than 50% of the patients, and there is a major need for targeted drugs with better anti-cancer selectivity. Here, our aim was to i) identify potential clinically used or emerging cancer drugs by quantitative drug sensitivity and resistance testing (DSRT) of 16 AML cell lines ii) compare the cell line data with results obtained from tested 24 ex vivo AML patient specimens iii) identify genomic correlations potentially explaining drug responsiveness. Methods: The cancer pharmacopeia-wide drug collection is composed of 119 FDA approved and 90 investigational chemical compounds including cytotoxic agents and cell signaling molecule inhibitors. Each drug was tested over a 10,000-fold concentration range and that has generated quantitative five point dose-response curves. AML cells were plated in 384 well plates (where the drugs were pre-printed using an acoustic nano-dispensing technology, Labcyte®) and incubated in standard cell culture conditions. Cell viability was measured by Cell Titer Glow® luminescence assays. Analysis of dose response curves using Dotmatics® software resulted in IC50 values. Moreover, the genomic profiles of the AML cell lines were determined by microarrays and/or next-gen sequencing data for further integration with drug responses. Results: Comprehensive data analysis of 16 AML cell lines indicated that specific targeted drugs were selectively killing AML cells. The data analysis revealed relatively strong responses for MEK inhibitors in most AML cell lines (e.g. refametinib 87%, trametinib 82%, selumetinib 75%) while 21% of ex vivo AML patient samples were sensitive to these MEK inhibitors. In case of rapalog sensitivity, 80% of AML cell lines (e.g. temsirolimus 82%, everolimus 71%, sirolimus 81%) and 25% of ex vivo AML patient cases were responsive to the mTOR inhibitors. The AML cell lines carrying FLT3-ITD mutations were extremely sensitive to FLT3 inhibitors (e.g. quizartinib, lestaurtinib, tandutinib, and sorafenib) but very few responses to FLT3 inhibitors were observed in AML patients carrying an ITD mutation in the FLT3 kinase. Summary: Systematic DSRT profiling of AML cell lines illustrates drug sensitivity patterns to classify the cell lines as sensitive or resistant to specific classes of drugs. mTOR and MEK inhibitors were among the most effective inhibitors for most cell lines and also in some ex vivo patient cases suggesting that these drugs may have potential as therapeutic agents in AML. Also, bioinformatics predictions can be used to identify key synergistic combinations of tested drugs for effective AML therapy. Further integration of molecular profiles and functional responses of AML cell lines will help provide better understanding of drug efficacy based on known genetic background of the disease. Citation Format: Disha Malani, Astrid Murumägi, Tea Pemovska, Bhagwan Yadav, Evgeny Kulesskiy, Jing Tang, John Patrick Mpindi, Maija Wolf, Riikka Karjalainen, Tero Aittokallio, Caroline Heckman, Kimmo Porkka, Krister Wennerberg, Olli Kallioniemi. Identifying AML-specific key targeted drugs using high-throughput drug sensitivity and resistance testing profiles of AML cells. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 5220. doi:10.1158/1538-7445.AM2013-5220

Acute myeloid leukemia (AML) is an aggressive, heterogeneous disease with few options for targeted therapy. Here, we describe a novel translational strategy termed Individualized Systems Medicine (ISM), in which we profile primary AML patient cells functionally, molecularly and clinically to identify novel treatment strategies for patients, monitor and predict disease progression and follow-up therapies, and elucidate drug response and resistance mechanisms. We developed a comprehensive ex vivo drug sensitivity and resistance testing (DSRT) strategy to screen AML patient blast cells ex vivo against a set of 202 conventional chemotherapeutic and targeted approved (n=119) and investigational (n=83) drugs. Quantitative leukemia-selective drug sensitivity scores for each drug were determined by comparing the area under the dose response curve from the patient cells to that of healthy control mononuclear cells. Analysis of consecutive samples from the same patients with DSRT and next-generation sequencing was applied to infer clonal evolution and potential mechanisms of drug response and resistance. Twenty-four samples from 16 recurrent and refractory AML patients were profiled by DSRT, sequencing and proteomic approaches. Several approved and late stage clinical investigated targeted drugs including multi-kinase inhibitors (e.g. dasatinib, sunitinib), TORC1 inhibitors (e.g. temsirolimus), JAK inhibitors (e.g. ruxolitinib) and MEK inhibitors (e.g. trametinib, selumetinib) showed selective leukemic-specific responses in 10-30% of AML samples from patients with recurrent disease. In two refractory AML cases where dasatinib, sunitinib and temsirolimus showed selective responses, the clinical administration of these compounds resulted in complete and partial remission, but was followed by resistance to the applied drugs. Re-sampling and DSRT retesting of cells confirmed diminished sensitivities to the administered drugs, but also indicated new acquired drug sensitivities. Exome and RNA sequencing of the serial samples from both patients revealed diverse subclonal populations characterized by multiple somatic mutations, which were either lost or gained during disease progression and represented drug sensitive or resistant subclones. In conclusion, our results suggest that an ISM strategy based on consecutive cancer sampling, ex vivo DSRT and analysis of clonal evolution could facilitate the rapid design of improved combinatorial therapies for AML. This strategy can also help tailor optimized therapies for patients, and prioritize introduction of new drugs for clinical testing. Citation Format: Krister Wennerberg, Tea Pemovska, Mika Kontro, Bhagwan Yadav, Evgeny Kulesskiy, Henrik Edgren, Samuli Eldfors, Riikka Karjalainen, Naga Poojitha Kota Venkata, Anna Lehto, Muntasir Mamun Majumder, Disha Malani, Astrid Murumägi, Laura Turunen, Jonathan Knowles, Tero Aittokallio, Caroline Heckman, Kimmo Porkka, Olli Kallioniemi. Comprehensive ex vivo drug sensitivity testing combined with in depth molecular profiling of AML patients cells provides individualized treatment strategies and reveals mechanisms of drug resistance. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 65. doi:10.1158/1538-7445.AM2013-65

Identification of signaling pathways that are required for the growth and differentiation block of cells from adult acute myeloid leukemia (AML) is urgently required to facilitate development of novel therapies. Here, we describe an approach to functionally determine molecular drivers of AML by quantitative drug sensitivity and resistance testing (DSRT) of AML blast cells in primary culture ex vivo. The selection of drugs covered the entire cancer pharmacopeia and much of the pipeline of drugs under development in the industry: 120 FDA approved small molecular cancer drugs and 120 emerging drugs, investigational compounds and signal transduction inhibitors. All compounds were tested over a 10,000-fold concentration range to generate quantitative and reliable dose-response data. In addition, whole exome and transcriptome sequencing and phophoproteomic profiling were also performed to derive a comprehensive understanding of the molecular AML-related aberrations on an individual basis. Comparison of 17 AML patient samples and 3 healthy bone marrow control samples based on ex vivo drug responses identified several classes of approved and investigational drugs that showed selective anti-AML activities: mTOR inhibitors (e.g. temsirolomus, everolimus, sirolimus), MEK inhibitors (e.g. AS703026, GSK1120212, RDEA119, selumetinib), tyrosine kinase inhibitors (e.g. dasatinib, ponatinib, sunitinib), Bcl-2 inhibitors (navitoclax) and HSP90 inhibitors (e.g. BIIB021, NVP-AUY922, tanespimycin). In particular, the rapamycin class of mTOR inhibitors and allosteric MEK inhibitors stood out as effective and selective inhibitors in 8/17 (47%) and 9/17 (52%) of the patients, respectively. Simultaneous data from other targeted inhibitors made it possible to dissect the critical steps in signaling and therapeutic efficacy. For example, PI3K and Akt inhibitors were not effective in these patients, suggesting that the mTOR dependency is mediated through a PI3K-Akt-independent pathway. Similarly, the dependency of MEK signaling appears to be through a Ras-Raf-independent pathway since Raf inhibitors were not effective. In conclusion, the DSRT platform allows us to derive quantitative data on the ex vivo drug response profiles of AML cells from individual patients. This information could be used as a diagnostic tool to optimize personalized therapies in the future. Our data demonstrate that mTOR and MEK signaling and the associated inhibitors are the most promising leads for improved AML therapeutics. This analysis also demonstrates gaps in our current understanding of the redundancy of key cancer cell signaling pathways and proves the significant value of data from experimental drug response testing ex vivo. 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 895. doi:1538-7445.AM2012-895

Despite significant advances in characterizing the molecular genetics of AML, the clonal evolution of leukemic cells and the dynamic impact of genomic changes on the development of the disease and progression to drug resistance are not well understood. Here, we applied next-generation sequencing to quantify aberrant tumor subclones carrying specific mutant alleles of key cancer genes and developed a method to extract quantitative high-resolution copy number changes across the genome using exome sequencing data from matching cancer and normal DNA. Serial bone marrow (BM) samples collected from diagnosis to relapse to post-treatment drug resistance in a patient-centric manner made it possible to trace the clonal evolution of AML and to identify variants potentially involved in drug resistance. Exome sequencing from AML blast cells and normal skin biopsies was performed as part of the Finnish Hematology Registry and Biobanking (FHRB) effort. Consecutive paired samples from different patients revealed unique genetic patterns of clonal evolution and cancer progression in each patient. In a pre-resistant sample of one AML M5 patient, we identified four closely spaced insertions in the Wilm's Tumor (WT1) suppressor gene, none of which appear on the same sequence reads. This suggests the presence of multiple distinct leukemic subclones even before treatment resistance and underscores the strong selective advantage conferred by WT1 mutations. After relapse, one of the subclones was lost, and another one significantly increased suggesting that the relapse arose from the expansion of a pre-existing resistant subclone. In this patient, recurrent clones otherwise featured similar copy number changes and the same fusion genes as the primary diagnostic sample. In another AML patient developing recurrence an opposite pattern was observed: The relapsed, drug-resistant cells displayed an enormous increase of small microdeletions compared to the diagnostic, pre-treatment sample, while almost all sequence-level alterations in potential cancer genes were the same between the two samples. This suggests that a distinct type of DNA repair deficiency may have contributed to the drug resistant clone in this patient. Conclusions: Exome sequencing from paired samples of AML cells before and after relapse makes it possible to trace the clonal evolution of the disease and study the impact of therapy both at the level of sequence alterations of key cancer genes and simultaneously at the level of copy number changes inferred from exome sequence data. This analysis has highlighted multiple genomic patterns by which resistance may evolve in vivo during cancer treatment. Refined bioinformatic analysis and interpretation of exome-seq data provides a rich resource to identify genetic biomarkers of drug response and minimal residual disease. 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 5067. doi:1538-7445.AM2012-5067

In order to discover unexpected anti-cancer efficacies of approved and emerging drugs, we established a diagnostic ex vivo drug sensitivity and resistance testing (DSRT) platform covering the entire cancer pharmacopeia as well as emerging anti-cancer compounds. Here, the platform was applied to analyze bone marrow (BM) mononuclear cells from 17 adult acute myeloid leukemia (AML) patients, 3 healthy donors as well as 7 AML cell lines. The DSRT panel covered FDA-approved small molecule oncology drugs (n=120), as well as emerging, investigational and pre-clinical oncology compounds (n=120), such as kinase (e.g. RTKs, checkpoint and mitotic kinases, Raf, MEK, JAKs, mTOR, PI3K), and non-kinase inhibitors (e.g. HSP, Bcl, activin, HDAC, PARP, Hh). To generate dose-response curves, each of the drugs was applied over a 10,000-fold concentration range. In addition, the samples underwent deep molecular profiling including exome- and transcriptome sequencing, as well as phosphoproteomic analysis. DSRT provided consistent and reliable data from ex vivo samples with a high correlation between data from individual healthy BM samples (r=0.93). Bioinformatic processing of the data from AML resulted in several key observations. First, overall drug response profiles of AML blast cells were distinctly different from healthy BM controls suggesting several potential leukemia-selective effects, such as multi-kinase (dasatinib), MEK, and mTOR inhibitors. Second, the overall drug responses from AML cell lines and the patient ex vivo samples showed differences, suggesting that ex vivo testing may reveal cancer-selective effects not previously seen in established cancer cell line panels. Third, the response data from patient samples clustered many drugs consistently into the expected functional classes (such as topoisomerase II inhibitors, MEK inhibitors and rapalogs), whereas other drug classes were more dispersed (such as FLT3 inhibitors with quizartinib clustering away from all other tyrosine kinase inhibitors), suggesting secondary targets playing a key role in drug efficacy. Fourth, analysis of serial samples from patients developing clinical resistance to targeted agents showed striking agreement between the ex-vivo DSRT profiles and clinical responses. In conclusion, comprehensive DSRT platform generated powerful novel insights on AML drug response and may enable individual optimization of therapies, particularly for recurrent leukemias. DSRT will also serve as a powerful hypothesis-generator for clinical trials, particularly for emerging drugs. The ability to correlate ex vivo response profiles for hundreds of drugs in clinical samples with deep molecular profiling data will yield exciting new translational and pharmacogenomic opportunities for cancer therapy. 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 3188. doi:1538-7445.AM2012-3188

Samples from recurrent, treatment-refractory cancers are rarely available, but would be valuable in understanding the molecular drivers of drug resistance. In leukemias, consecutive samples are readily available during treatment. Hence, we explored here the progression of adult acute myeloid leukemias (AML) by serial sampling and by integrating data from multiple platforms. Next-generation exome and RNA sequencing, and phosphoproteomic data were combined with comprehensive 240 cancer drug sensitivity and resistance testing (DSRT) of leukemic blasts ex-vivo before and after clinical relapse. The data were generated in an experimental diagnostic setting, with intent to improve and personalize treatment of patients with recurrent AML. A 54-year old AML-M5 patient with a FLT-3-ITD mutation and a normal karyotype was monitored by serial sampling. The patient was initially refractory to three consecutive high-dose induction treatments and had limited therapy options. AML blasts from the patient were screened with the DSRT platform. Results implied that the blast cells were 710-times more sensitive to temsirolimus and other rapamycin analogs as compared to normal BM cells, and showed a 1100-fold increased sensitivity to dasatinib. Proteomic analysis showed high phosphorylation of several signaling molecules, such as the insulin receptor and mTOR. Sequencing identified WT1 mutations and a NUP98-NSD1 fusion transcript, an infrequent event associated with poor prognosis in AML. Based on the DSRT results, the patient received compassionate off-label treatment with dasatinib, sunitinib and temsirolimus, resulting in a remarkable clinical remission, normalization of blast counts and a rapid recovery of neutrophil counts as a sign of selective elimination of the leukemic cells. The patient relapsed 4 weeks later, and at this point a new DSRT assay was performed, which showed the blast cells to be completely resistant to temsirolimus and less sensitive to dasatinib ex vivo. Consistent with this drug sensitivity profile was a genomic evolution of a distinct AML subclone with new changes, such as NF1 mutation and a microdeletion of the LEF1 gene, which were not observed in the pre-treatment sample. Taken together, we have demonstrated, how molecular profiling and functional ex vivo drug sensitivity and resistance data can be used to individually optimize patient treatment. Remission was achieved in a patient with advanced, treatment-refractory AML. Serial sampling from human AML patients coupled with molecular profiling and drug sensitivity testing may shed light on clonal progression of disease, and the molecular events underlying drug response. 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 4580. doi:1538-7445.AM2012-4580

Chronic myelomonocytic leukemia (CMML) is a rare malignancy characterized by increased peripheral monocytosis and dysplasia in a single- or multilineage fashion. Gene mutations so far reported in CMML include TET2, CBL, NRAS, KRAS, RUNX1 and EZH2 but their pathogenic role and driver status in the disease remains unclear. Altered expression of the microRNA miR-125b has been implicated in the pathogenesis of many types of cancers, including myeloid leukemias and Down syndrome-associated acute myeloid leukemia (DS-AML). In addition, this miRNA has been shown to play an important role in hematopoiesis and the regulation of immune cell response. Here, integration of data from next-generation transcriptome sequencing, exome sequencing and array-CGH in a CMML patient (trisomy 21 by cytogenetics) led to the identification of a novel gene fusion event involving the nuclear receptor interacting protein NRIP1 gene and the open reading frame C21orf34 (both at 21q21 approximately 1 MB apart). The fusion was validated by capillary sequencing and found to involve two copy number transition breaks, inversion of the intervening region and the upregulation of the 3′ end of C21orf34. This intronic region harbors a cluster of three miRNAs: miR-let7c, miR-99a, and miR-125b-2. Based on genomic breakpoint analysis, the gene fusion took place just upstream of miR-125b-2. Consistent with this, only miR-125b-2 was highly expressed in the sample, and was processed to a mature miRNA. By RT-PCR, increased expression of miR-125b-2 was also observed in four other CMML patients and five CML patients when compared to healthy bone marrow controls. In contrast, five AML cases studied showed expression levels similar to or lower than that of controls. Interestingly, one AML patient with trisomy 21 had very high levels of miR-125b-2. We found the NRIP1-C21orf34 fusion only in our index patient and therefore other mechanisms of miRNA deregulation at 21q21 in CMML/CML and AML+21 will also exist. In conclusion, we describe for the first time a fusion gene involving miR-125b-2 in CMML, a previously recognized and well-studied onco-miR, which is known to impact on self-renewal of hematopoietic cell precursors. We also detected overexpression of miR-125b-2 in all CMML samples studied suggesting a key pathogenetic driver gene role for this micro-RNA. The assessment of miR-125b-2 levels could potentially be applied to the diagnosis and follow-up of patients with CMML. 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 3175. doi:1538-7445.AM2012-3175