Personalized Transcriptomic Analyses Identify Unique Signatures That Correlate with Genomic Subtypes in Acute Myeloid Leukemia (AML) Using Explainable Artificial Intelligence

Blood ◽  
2020 ◽  
Vol 136 (Supplement 1) ◽  
pp. 33-34 ◽  
Author(s):  
Yazan Rouphail ◽  
Nathan Radakovich ◽  
Jacob Shreve ◽  
Sudipto Mukherjee ◽  
Babal K. Jha ◽  
...  
Keyword(s):  
Machine Learning ◽  
Board Of Directors ◽  
Research Funding ◽  
State Of The Art ◽  
Data Monitoring Committee ◽  
Biological Pathways ◽  
Machine Learning Algorithms ◽  
Advisory Committees ◽  
Altered Expression ◽  
Individual Level

Background Multi-omic analysis can identify unique signatures that correlate with cancer subtypes. While clinically meaningful molecular subtypes of AML have been defined based on the status of single genes such as NPM1 and FLT3, such categories remain heterogeneous and further work is needed to characterize their genetic and transcriptomic diversity on a truly individualized basis. Further, patients (pts) with NPM1+/FLT3-ITD- AML have a better overall survival compared to patients with NPM1-/FLT3-ITD+, suggesting that these pts could have different transcriptomic signature that impact phenotype, pathophysiology, and outcomes. Many current transcriptome analytic techniques use clustering analysis to aggregate samples and look at relationships on a cohort-wide basis to build transcriptomic signatures that correlate with phenotype or outcome. Such approaches can undermine the heterogeneity of the gene expression in pts with the same signatures. In this study, we took advantage of state of the art machine learning algorithms to identify unique transcriptomic signatures that correlate with AML genomic phenotype. Methods Genomic (whole exome sequencing and targeted deep sequencing) and transcriptomic data from 451 AML pts included in the Beat AML study (publicly available data) were used to build transcriptomic signatures that are specific for AML patients with NPM1+/FLT3-ITD+ compared to NPM1+/FLT3-ITD, and NPM1-/FLT3-ITD-. We chose these AML phenotypes as they have been described extensively and they correlate with clinical outcomes. Results A total of 242 patients (54%) had NPM1-/FLT3-, 35 (8%) were NPM1+/FLT3-, and 47 (10%) were NPM1+/FLT3+. Our algorithm identified 20 genes that are highly specific for NPM1/FLT3ITD phenotype: HOXB-AS3, SCRN1, LMX1B, PCBD1, DNAJC15, HOXA3, NPTXq, RP11-1055B8, ABDH128, HOXB8, SOCS2, HOXB3, HOXB9, MIR503HG, FAM221B, NRP1, NDUFAF3, MEG3, CCDC136, and HIST1H2BC. Interestingly, several of those genes were overexpressed or underexpressed in specific phenotypes. For example, SCRN1, LMX1B, RP11-1055B8, ABDH128, HOXB8, MIR503HG, NRP1 are only overexpressed or underexpressed in patients with NPM1-/FLT3-, while PCBD1, NDUFAF3, FAM221B are overexpressed or underexpressed in pts with NPM1+/FLT3+. These genes affect several important pathways that regulate cell differentiation, proliferation, mitochondrial oxidative phosphorylation, histone modification and lipid metabolism. All these genes had previously been reported as having altered expression in genomic studies of AML, confirming our approach's ability to identify biologically meaningful relationships. Further, our algorithm can provide a personalized explanation of overexpressed and underexpressed genes specific for a given patient, thus identifying targetable pathways for each pt. Figure 1 below shows three pts with the same genotype (NPM1+/FLT3-ITD+) but demonstrate different transcriptomic patterns of overexpression or underexpression that affect different biological pathways. Conclusions We describe the use of a state of the art explainable machine learning approach to define transcriptomic signatures that are specific for individual pts. In addition to correctly distinguishing AML subtype based on specific transcriptomic signatures, our model was able to accurately identify upregulated and downregulated genes that affecte several important biological pathways in AML and can summarize these pathways at an individual level. Such an approach can be used to provide personalized treatment options that can target the activated pathways at an individual level. Disclosures Mukherjee: Partnership for Health Analytic Research, LLC (PHAR, LLC): Honoraria; Novartis: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; EUSA Pharma: Consultancy; Celgene/Acceleron: Membership on an entity's Board of Directors or advisory committees; Bristol Myers Squib: Honoraria; Aplastic Anemia and MDS International Foundation: Honoraria; Celgene: Consultancy, Honoraria, Research Funding. Maciejewski:Alexion, BMS: Speakers Bureau; Novartis, Roche: Consultancy, Honoraria. Sekeres:BMS: Consultancy; Takeda/Millenium: Consultancy; Pfizer: Consultancy. Nazha:Jazz: Research Funding; Incyte: Speakers Bureau; Novartis: Speakers Bureau; MEI: Other: Data monitoring Committee.

scholarly journals Machine Learning Prediction for Complete Response to Hypomethylating Agents with or without Additional Agents in Patients with Newly Diagnosed Myelodysplastic Syndrome

Blood ◽  
2019 ◽  
Vol 134 (Supplement_1) ◽  
pp. 1720-1720
Author(s):  
Koji Sasaki ◽  
Guillermo Montalban Bravo ◽  
Rashmi Kanagal-Shamanna ◽  
Elias Jabbour ◽  
Farhad Ravandi ◽  
...  
Keyword(s):  
Machine Learning ◽  
Board Of Directors ◽  
Research Funding ◽  
Area Under The Curve ◽  
Complete Response ◽  
Learning Model ◽  
Hypomethylating Agents ◽  
Newly Diagnosed ◽  
Advisory Committees ◽  
Machine Learning Model

Background: Myelodysplastic syndrome (MDS) is a heterogeneous malignant myeloid neoplasm of hematopoietic stem cells due to cytogenetic alterations and somatic mutations in genes (DNA methylation, DNA repair, chromatin regulation, RNA splicing, transcription regulation, and signal transduction). Hypomethylating agents (HMA) are the standard of care for MDS, and 40-60% of patients achieved response to HMA. However, the prediction for response is difficult due to the nature of heterogeneity and the context of clinical conditions such as the degree of cytopenias and the dependency on transfusion. Machine learning outperforms conventional statistical models for prediction in statistical competitions. Prediction with machine learning models may predict response in patients with MDS. The aim of this study is to develop a machine learning model for the prediction of complete response (CR) to HMA with or without additional therapeutic agents in patients with newly diagnosed MDS. Methods: From November 2012 to August 2017, we analyzed 435 patients with newly diagnosed MDS who received frontline therapy as follows; azacitidine (AZA) (3-day, 5-day, or 7-day) ± vorinostat ± ipilimumab ± nivolumab; decitabine (DAC) (3-day or 5-day) ± vorinostat; 5-day guadecitabine. Clinical variables, cytogenetic abnormalities, and the presence of genetic mutations by next generation sequencing (NGS) were included for variable selection. The whole cohort was randomly divided into training/validation and test cohorts at an 8:2 ratio. The training/validation cohort was used for 4-fold cross validation. Hyperparameter optimization was performed with Stampede2, which was ranked as the 15th fastest supercomputer at Texas Advanced Computing Center in June 2018. A gradient boosting decision tree-based framework with the LightGBM Python module was used after hyperparameter tuning for the development of the machine learning model with training/validation cohorts. The performance of prediction was assessed with an independent test dataset with the area under the curve. Results: We identified 435 patients with newly diagnosed MDS who enrolled on clinical trials as follows: 33 patients, 5-day AZA; 23, 5-day AZA + vorinostat; 43, 3-day AZA; 20, 5-day AZA + ipilimumab; 19 patients, AZA + nivolumab; 7, AZA + ipilumumab + nivolumab; 114, 5-day DAC; 74, 3-day DAC; 4, DAC + vorinostat; 97, 5-day guadecitabine. In the whole cohort, the median age at diagnosis was 68 years (range, 13.0-90.3); 117 (27%) patients had a history of prior radiation or cytotoxic chemotherapy; the median white blood cell count was 2.9 (×109/L) (range, 0.5-102); median absolute neutrophil count, 1.1 (×109/L) (range, 0.0-55.1); median hemoglobin count, 9.5 (g/dL) (range, 4.7-15.4); median platelet count, 63 (×109/L) (range, 2-881); and median blasts in bone marrow, 8% (range, 0-20). Among 411 evaluable patients for the revised international prognostic scoring system, 15 (4%) had very low risk disease; 42 (10%), low risk; 68 (17%), intermediate risk; 124 (30%), high risk; and 162 (39%), very high risk. Overall, 153 patients (53%) achieved CR. Hyperparameter tuning identified the optimal hyperparameters with colsample by tree of 0.175, learning rate of 0.262, the maximal depth of 2, minimal data in leaf of 29, number of leaves of 11, alpha regularization of 0.010, lambda regularization of 2.085, and subsample of 0.639. On the test cohort with 87 patients, the machine learning model accurately predicted response in 65 patients (75%); 53 non-CR among 56 non-CR (95% accuracy); and 12 CR among 31 CR (39% accuracy). The trend of accuracy improvement by iteration (i.e., the number of decision trees) is shown in Figure 1. The area under the curve was 0.761521 in the test cohort. Conclusion: Our machine learning model with clinical, cytogenetic, and NGS data can predict CR to HMA in patients with newly diagnosed MDS. This approach can identify patients who may benefit from HMA therapy with and without additional agents for response, and can optimize the timing of allogeneic stem cell transplant. Disclosures Sasaki: Otsuka: Honoraria; Pfizer: Consultancy. Jabbour:Takeda: Consultancy, Research Funding; BMS: Consultancy, Research Funding; Adaptive: Consultancy, Research Funding; Amgen: Consultancy, Research Funding; AbbVie: Consultancy, Research Funding; Pfizer: Consultancy, Research Funding; Cyclacel LTD: Research Funding. Ravandi:Cyclacel LTD: Research Funding; Selvita: Research Funding; Menarini Ricerche: Research Funding; Macrogenix: Consultancy, Research Funding; Amgen: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding; Xencor: Consultancy, Research Funding. Kadia:Pfizer: Membership on an entity's Board of Directors or advisory committees, Research Funding; Celgene: Research Funding; Bioline RX: Research Funding; Jazz: Membership on an entity's Board of Directors or advisory committees, Research Funding; AbbVie: Consultancy, Research Funding; BMS: Research Funding; Amgen: Membership on an entity's Board of Directors or advisory committees, Research Funding; Genentech: Membership on an entity's Board of Directors or advisory committees; Pharmacyclics: Membership on an entity's Board of Directors or advisory committees; Takeda: Membership on an entity's Board of Directors or advisory committees. Takahashi:Symbio Pharmaceuticals: Consultancy. DiNardo:syros: Honoraria; jazz: Honoraria; agios: Consultancy, Honoraria; celgene: Consultancy, Honoraria; notable labs: Membership on an entity's Board of Directors or advisory committees; medimmune: Honoraria; abbvie: Consultancy, Honoraria; daiichi sankyo: Honoraria. Cortes:Novartis: Consultancy, Honoraria, Research Funding; Bristol-Myers Squibb: Consultancy, Research Funding; Immunogen: Consultancy, Honoraria, Research Funding; Sun Pharma: Research Funding; Pfizer: Consultancy, Honoraria, Research Funding; Astellas Pharma: Consultancy, Honoraria, Research Funding; Jazz Pharmaceuticals: Consultancy, Research Funding; Merus: Consultancy, Honoraria, Research Funding; Forma Therapeutics: Consultancy, Honoraria, Research Funding; Daiichi Sankyo: Consultancy, Honoraria, Research Funding; BiolineRx: Consultancy; Biopath Holdings: Consultancy, Honoraria; Takeda: Consultancy, Research Funding. Kantarjian:AbbVie: Honoraria, Research Funding; Cyclacel: Research Funding; Pfizer: Honoraria, Research Funding; Astex: Research Funding; Agios: Honoraria, Research Funding; Jazz Pharma: Research Funding; Daiichi-Sankyo: Research Funding; Novartis: Research Funding; Actinium: Honoraria, Membership on an entity's Board of Directors or advisory committees; Immunogen: Research Funding; Takeda: Honoraria; BMS: Research Funding; Ariad: Research Funding; Amgen: Honoraria, Research Funding. Garcia-Manero:Amphivena: Consultancy, Research Funding; Helsinn: Research Funding; Novartis: Research Funding; AbbVie: Research Funding; Celgene: Consultancy, Research Funding; Astex: Consultancy, Research Funding; Onconova: Research Funding; H3 Biomedicine: Research Funding; Merck: Research Funding.

scholarly journals Phase II Trial of Ixazomib and Dexamethasone Versus Ixazomib, Dexamethasone and Lenalidomide, Randomized with NFKB2 Rearrangement. (Proteasome Inhibitor NFKB2 Rearrangement Driven Trial, PINR)

Blood ◽  
2018 ◽  
Vol 132 (Supplement 1) ◽  
pp. 2011-2011
Author(s):  
Leon Bernal-Mizrachi ◽  
Craig E. Cole ◽  
Leonard Heffner ◽  
Ajay K Nooka ◽  
Ravi Vij ◽  
...  
Keyword(s):  
Partial Response ◽  
Board Of Directors ◽  
Interim Analysis ◽  
Proteasome Inhibitor ◽  
Research Funding ◽  
Treated Patient ◽  
Data Monitoring Committee ◽  
Caspase 8 ◽  
Open Label ◽  
Advisory Committees

Abstract Introduction: Proteasome inhibitors (PI) and immunomodulatory drugs have become the backbone of therapy for multiple myeloma (MM). The oral boron-containing selective and reversible proteasome inhibitor ixazomib has shown to induce deep and durable responses (Kumar RK et la, Blood 2016. 128(20):2415-2422). Triplets containing ixazomib, has shown to be more efficacious than doublet regimens in the relapse setting (Moreau P, et al. N Engl J Med 2016. 374:1621-1634).However, to date there is not companion diagnostics capable of predicting PI response. We have recently discovered that MM patients resistant to PI lack of the ankyrin (ANK) and death domain (DD) present in the 3'-end of NFKB2. Loss of NFKB2 3'end frequently resulted from a structural rearrangements. We found that NFKB2-ANK and -DD are crucial at initiating bortezomib's apoptotic signal by facilitating caspase-8 activation. Activation of this caspase resulted from NFKB2 interaction with FADD/caspase-8/p62 (unpublished data). Based on this findings, we design this study to examine the efficacy of NFKB2 break apart FISH to predict the response to the duplet ixazomib and dexamethasone (Id) vs ixazomib, lenalidomide and dexamethasone (IRd) in early relapse MM patients (1-3 lines of therapy). Methods: In this phase 2 biomarker driven open label trial, relapse patients with <4 lines of therapy were randomized to ixazomib 4 mg weekly 3 of 4 weeks and 40 mg weekly dexamethasone vs Id plus 25 mg of lenalidomide daily days 21/28, based on the status of NFKB2 rearrangement in plasma cells. Patients were screened for NFKB2 rearrangement was detected by NFKB2 break apart FISH. Patients without NFKB2 rearrangement received Id and patients with NFKB2 rearrangement were subsequently randomized in a 1:1 fashion to receive Id or IRd. The primary endpoint is to compare the response rate of Id or IRd at 4 cycles according to the rearrangement status of NFKB2 Results: At the moment of interim analysis, 26 patients have achieved 4 cycles of treatment. All treatment groups (NFKB2 FISH [-] Id, n=16, NFKB2 FISH [+] Id, n=6 and IRd, n=4) received a median of 2 prior lines of therapy. A higher ORR was observed in NFKB2 FISH negative treated with Id compared with NFKB2 FISH positive (56.3% CI:29.9%-80.3% vs. 16.7% CI:0.4%-64.1%, P=0.16), including significantly higher rates of ≥very good partial response, ≥ partial response, ≥ minimal response (12%, 37.5%, 6% vs. 0%, 0%, 16%, respectively). ORR for IRd arm is for now 25% CI:0.6%-80.6%. Patients that continue treatment after cycle 4 that achieve minimal response or better improved their depth of response in 6% in Id treated NFKB2 FISH negative patients and 25% of IRd treated NFKB2 FISH negative patients. The most common (≥10%) grade 2/4 include pneumonia 20% (Id treated NFKB2 FISH negative), thrombocytopenia 16% (Id NFKB2 FISH positive), neutropenia 60% (IRd NFKB2 FISH positive. Treatment discontinuations only occurred in 1 Id treated patient (5%). Conclusion: Interim analysis demonstrate a trend higher efficacy of ixazomib with dexamethasone in MM patients with negative NFKB2 break-apart FISH compared to those with a positive test. Efficacy and toxicity of the triplet regimen are comparable to what is seen in Tourmaline 1 trial. This study was registered at www.clinicaltrials.gov as # NCT02765854. Disclosures Bernal-Mizrachi: Takeda Pharmaceutical Company: Research Funding; Kodikaz Therapeutic Solutions: Consultancy, Equity Ownership. Cole:Cancer Support Community myeloma advisory board: Membership on an entity's Board of Directors or advisory committees; University of Michigan: Employment. Heffner:ADC Therapeutics: Research Funding; Kite Pharmaceuticals: Research Funding; Pharmacyclics: Research Funding; Genentech: Research Funding. Nooka:GSK: Consultancy, Membership on an entity's Board of Directors or advisory committees; Takeda: Consultancy, Membership on an entity's Board of Directors or advisory committees; Spectrum Pharmaceuticals: Consultancy, Membership on an entity's Board of Directors or advisory committees; Amgen: Consultancy, Membership on an entity's Board of Directors or advisory committees; BMS: Consultancy, Membership on an entity's Board of Directors or advisory committees; Janssen pharmaceuticals: Consultancy, Membership on an entity's Board of Directors or advisory committees; Celgene: Consultancy, Membership on an entity's Board of Directors or advisory committees; Adaptive technologies: Consultancy, Membership on an entity's Board of Directors or advisory committees. Vij:Celgene: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding; Amgen: Honoraria, Membership on an entity's Board of Directors or advisory committees; Takeda: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding; Jazz Pharmaceuticals: Honoraria, Membership on an entity's Board of Directors or advisory committees; Jansson: Honoraria, Membership on an entity's Board of Directors or advisory committees; Karyopharma: Honoraria, Membership on an entity's Board of Directors or advisory committees; Bristol-Myers Squibb: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding. Kelkar:Empire Genomics: Employment. Lonial:Amgen: Research Funding. Kaufman:Karyopharm: Other: data monitoring committee; Abbvie: Consultancy; Roche: Consultancy; BMS: Consultancy; Janssen: Consultancy.

scholarly journals Central Nervous System Involvement By Mantle Cell Lymphoma

Blood ◽  
2021 ◽  
Vol 138 (Supplement 1) ◽  
pp. 2426-2426
Author(s):  
Nicole McLaughlin ◽  
Jonas Paludo ◽  
Yucai Wang ◽  
David J. Inwards ◽  
Nora Bennani ◽  
...  
Keyword(s):  
Board Of Directors ◽  
Research Funding ◽  
Data Monitoring Committee ◽  
Advisory Board ◽  
Control Group ◽  
Cns Involvement ◽  
Extranodal Involvement ◽  
Advisory Committees ◽  
Scientific Advisory Board ◽  
Scientific Advisory

Abstract Background: While extranodal involvement by mantle cell lymphoma (MCL) is relatively common, involvement of the central nervous system (CNS) is rare (&lt;5% of cases), with limited treatment options. We report the outcomes of 36 patients (pts) with CNS involvement compared to 72 matched control MCL pts without CNS involvement. Methods: MCL pts with CNS involvement seen at Mayo Clinic between 1/1995-9/2020 were identified using the Mayo Data Explorer tool. CNS involvement was defined by tissue biopsy confirmed CNS MCL, CSF analysis demonstrating lymphoma cells, and/or neuroimaging findings compatible with CNS involvement. A 2:1 control group of MCL pts without CNS involvement, matched by age (+/- 2 years) and year of diagnosis (+/- 1 year), was selected among all MCL cases. Medical records were reviewed for baseline characteristics, treatment modalities, and outcomes. Kaplan-Meier method was used for time to event analysis. Wilcoxon test was used to compare continuous variables and Chi square test was used for categorical variables. Results: Out of 1,753 pts with MCL, 36 (2%) had evidence of CNS involvement, including 4 pts with CNS involvement at initial MCL diagnosis. Baseline characteristics of pts with CNS involvement (CNS MCL group) and those without CNS involvement (control group) are shown in Table 1. At MCL diagnosis, non-CNS extranodal involvement was seen in 30 (83%) pts in the CNS MCL group (24 pts with 1 site and 6 pts with ≥ 2 sites), with bone marrow being the most common extranodal site of involvement (n=24, 67%). For the control group, 54 (75%) pts had extranodal involvement (44 pts with 1 site and 10 pts with ≥ 2 sites), and bone marrow was also the most common extranodal site of involvement (n=50, 69%). Notably, advanced stage disease (stage 3-4) was more commonly seen in the CNS MCL group (n=32, 97%) than in the control group (n=59, 83%) (p=0.04) at MCL diagnosis. Blastoid variant was present in a higher proportion of pts in the CNS MCL group (n=11, 31%) compared to the control group (n=8, 11%) (p=0.02). The CNS MCL group also presented with a higher median serum LDH at diagnosis (239 U/L [range 153-1901] vs. 187 U/L [range 124-588], p=0.02), and higher Ki-67 (40% [range 15-100] vs. 30% [range 10-90], p=0.04) compared to the control group. The most common frontline treatment regimen was anthracycline-based therapies (i.e. R-CHOP, Nordic regimen, R-hyperCVAD) for both groups (58% in CNS MCL group and 56% in control group). 14 (39%) pts in the CNS MCL group underwent autologous stem cell transplant in CR1 vs. 31 pts (43%) in the control group. Similar use of rituximab maintenance was seen in both groups (31% in CNS MCL group and 25% in control group). Median total lines of therapy from initial MCL diagnosis was 3 (range 1-9) in CNS MCL group and 2 (range 1-9) in the control group. The median follow-up from MCL diagnosis was 134 months (95% CI:119-163) for the entire cohort. Median OS from MCL diagnosis was 50.3 months (95% CI: 20.9-71.1) for the CNS MCL group compared to 97.1 months (95% CI: 82.6-192.7; p=&lt;0.001) for the control group (Figure 1). Median time from MCL diagnosis to CNS involvement was 25 months (range 0-167). Median OS from CNS involvement was 4.7 months (95% CI: 2.3-6.7). At last follow up, 31 (86%) pts were deceased from the CNS MCL group, compared to 38 (52%) pts in the control group. For the CNS MCL group, the causes of death were CNS lymphoma in 10 (32%) pts, systemic lymphoma in 9 (29%) pts, treatment-related complication in 7 (23%) pts, and other/unknown in 5 (16%) pts. For the control group, the causes of death were systemic lymphoma in 15 (39%) pts, treatment-related in 2 (5%) pts, and other/unknown in 21 (55%) pts. Conclusion: In pts with MCL, CNS involvement is associated with worse outcomes as evident by a shorter median OS from initial MCL diagnosis (50 months vs. 97 months). Involvement of the CNS by lymphoma is an important contributor for the shorter OS as suggested by the median OS of only 5 months from CNS involvement. Advanced stage, blastoid variant, elevated LDH, and elevated Ki67 at MCL diagnosis were features more commonly seen in the CNS MCL cohort. Validation of risk factors at initial MCL diagnosis associated with CNS involvement and exploring the role of CNS prophylaxis are important topics for further investigation. Figure 1 Figure 1. Disclosures Paludo: Karyopharm: Research Funding. Wang: Novartis: Research Funding; TG Therapeutics: Membership on an entity's Board of Directors or advisory committees; MorphoSys: Research Funding; InnoCare: Research Funding; Eli Lilly: Membership on an entity's Board of Directors or advisory committees; LOXO Oncology: Membership on an entity's Board of Directors or advisory committees, Research Funding; Incyte: Membership on an entity's Board of Directors or advisory committees, Research Funding; Genentech: Research Funding. Bennani: Purdue Pharma: Other: Advisory Board; Daichii Sankyo Inc: Other: Advisory Board; Kyowa Kirin: Other: Advisory Board; Vividion: Other: Advisory Board; Kymera: Other: Advisory Board; Verastem: Other: Advisory Board. Nowakowski: Celgene, MorphoSys, Genentech, Selvita, Debiopharm Group, Kite/Gilead: Consultancy, Membership on an entity's Board of Directors or advisory committees; Celgene, NanoString Technologies, MorphoSys: Research Funding. Witzig: Karyopharm Therapeutics, Celgene/BMS, Incyte, Epizyme: Consultancy, Membership on an entity's Board of Directors or advisory committees; Celgene/BMS, Acerta Pharma, Kura Oncology, Acrotech Biopharma, Karyopharm Therapeutics: Research Funding. Habermann: Seagen: Other: Data Monitoring Committee; Tess Therapeutics: Other: Data Monitoring Committee; Incyte: Other: Scientific Advisory Board; Morphosys: Other: Scientific Advisory Board; Loxo Oncology: Other: Scientific Advisory Board; Eli Lilly & Co.,: Other: Scientific Advisor. Ansell: Bristol Myers Squibb, ADC Therapeutics, Seattle Genetics, Regeneron, Affimed, AI Therapeutics, Pfizer, Trillium and Takeda: Research Funding.

scholarly journals Safety and Efficacy of Tenalisib Given in Combination with Romidepsin in Patients with Relapsed/Refractory T-Cell Lymphoma: Final Results from a Phase I/II Open Label Multi-Center Study

Blood ◽  
2021 ◽  
Vol 138 (Supplement 1) ◽  
pp. 1365-1365
Author(s):  
Swaminathan P Iyer ◽  
Auris Huen ◽  
Weiyun Z. Ai ◽  
Deepa Jagadeesh ◽  
Mary Jo Lechowicz ◽  
...  
Keyword(s):  
Family Member ◽  
Board Of Directors ◽  
Research Funding ◽  
Cell Lymphoma ◽  
Data Monitoring Committee ◽  
Data Monitoring ◽  
Open Label ◽  
Advisory Committees ◽  
Monitoring Committee ◽  
Committee Membership

Abstract Background: Tenalisib (RP6530), a highly selective PI3K δ/γ and SIK3 inhibitor has shown promising activity as a single agent in T Cell lymphoma (TCL) and a differentiated safety profile (Huen A et al., Cancers,2020). In vitro studies in TCL cell lines showed synergistic activity when tenalisib was combined with romidepsin. A Phase I/II study of tenalisib in combination with romidepsin was designed to assess safety, pharmacokinetics, and efficacy in patients with relapsed/refractory (R/R) TCL peripheral (PTCL) and cutaneous T cell lymphoma (CTCL) (NCT03770000). Methods: This was a multi-center, open label study. We performed a Phase I, 3+3 dose escalation study to determine the MTD/recommended Phase II dose (RP2D), and a dose expansion study in both the subtypes separately (PTCL and CTCL). Patients received tenalisib at doses ranging from 400-800 mg BID (fasting), orally in combination with romidepsin at doses ranging from 12-14 mg/m 2, intravenously, given on Days 1,8 and 15 of a 28-day cycle. Results: Thirty-three patients (16 PTCL and 17 CTCL) who received more than 1 prior therapy were enrolled in the study; 9 in dose escalation and 24 in dose expansion. Of the 33 patients, 64% were refractory to their last therapy. The median number of prior therapies was 3. Most patients (67%) had stage III/IV disease at time of enrolment. No dose limiting toxicity (DLT) was reported during dose escalation; tenalisib 800 mg BID with romidepsin 14 mg/m 2 (given on Days 1, 8, and 15) was chosen as the RP2D. The most frequent treatment emergent adverse events (TEAEs) were nausea (All: 73% and ≥G3:0%), thrombocytopenia (All:57% and ≥G3:21%), fatigue (All: 54% and ≥G3:6%), AST elevation (All:33% and ≥G3:6%) ALT elevation (All:27% and ≥G3:18%), neutropenia (All: 27% and ≥G3:15%), vomiting (All:27% and ≥G3:0%), decreased appetite (All: 27% and ≥G3:0%). There were no unexpected TEAEs. Among CTCL patients, five related TEAEs led to drug discontinuation were sepsis, ALT elevation, GGT elevation, rash, and dysgeusia. None of the PTCL patients discontinued the study drug due to related TEAEs. Incidences of TEAEs leading to drug interruption (72%) and dose reduction (45%) of any the drugs in the combination were similar in PTCL and CTCL groups. Based on C max and AUC, dose proportional exposure of tenalisib was observed from doses 400-800 mg BID. Co-administration of romidepsin with tenalisib did not significantly alter the PK of either agent. Of the 33 patients enrolled, 27 (12 PTCL and 15 CTCL) who received at least 1 dose of study drug and provided at least 1 post-baseline efficacy assessment were considered evaluable for efficacy as per protocol. The overall response rate (ORR) was of 63%; 7 (26%) patients achieved CR and 10 (37%) patients had PR (Table 1). The median duration of response (DoR) was 5.03 months (range: 2.16 months-Not Reached). In twelve evaluable PTCL patients, the ORR was 75% with 6 CR (50%) and 3 PR (25%). Among 15 evaluable CTCL patients, 8 responded with an ORR of 53.3%, 1 CR (6.7%) and 7 PR (46.7%). The median DoR was 5.03 (range: 2.16 months-Not Reached) for PTCL and 3.8 months (1.9-18.86) for CTCL. Three of the six (50%) PTCL patients with CR were bridged to transplant. Six patients who benefitted with the treatment and completed the protocol were enrolled in an open-label compassionate medication study after Cycle 7 and are being followed up. Conclusions: The combination of tenalisib and romidepsin demonstrates a favorable safety profile and promising anti-tumor activity in patients with R/R TCL. Based on these encouraging results, further development of this combination in PTCL patients in being planned. Figure 1 Figure 1. Disclosures Huen: Rhizen: Research Funding; Elorac: Research Funding; Kyowa Kirin: Research Funding; Tillium: Research Funding; Innate: Research Funding; Galderma: Research Funding; Miragen: Research Funding. Ai: Kymria, Kite, ADC Therapeutics, BeiGene: Consultancy. Feldman: Alexion, AstraZeneca Rare Disease: Honoraria, Other: Study investigator. Alderuccio: ADC Therapeutics: Consultancy, Research Funding; Oncinfo / OncLive: Honoraria; Puma Biotechnology: Other: Family member; Inovio Pharmaceuticals: Other: Family member; Agios Pharmaceuticals: Other: Family member; Forma Therapeutics: Other: Family member. Kuzel: Cardinal Health: Membership on an entity's Board of Directors or advisory committees; Exelixis: Membership on an entity's Board of Directors or advisory committees; Genomic Health: Membership on an entity's Board of Directors or advisory committees; Sanofi-Genzyme Genomic Health Tempus laboratories Bristol Meyers Squibb: Honoraria; Abbvie: Other; Curio Science: Membership on an entity's Board of Directors or advisory committees; AmerisourceBergen Corp: Membership on an entity's Board of Directors or advisory committees; CVS: Membership on an entity's Board of Directors or advisory committees; Tempus Laboratories: Membership on an entity's Board of Directors or advisory committees; Bristol Meyers Squibb: Membership on an entity's Board of Directors or advisory committees; Merck: Other: Data Monitoring Committee Membership; Amgen: Other: Data Monitoring Committee Membership; SeaGen: Other: Data Monitoring Committee Membership; Medpace: Other: Data Monitoring Committee Membership.

scholarly journals Predicting Response to Hypomethylating Agents in Patients with Myelodysplastic Syndromes (MDS) Using Artificial Intelligence (AI)

Blood ◽  
2019 ◽  
Vol 134 (Supplement_1) ◽  
pp. 2089-2089
Author(s):  
Nathan Radakovich ◽  
Mikkael A. Sekeres ◽  
Cameron Beau Hilton ◽  
Sudipto Mukherjee ◽  
Jacob Shreve ◽  
...  
Keyword(s):  
Neural Network ◽  
Sample Size ◽  
Board Of Directors ◽  
Research Funding ◽  
Validation Cohort ◽  
Data Monitoring Committee ◽  
Small Sample ◽  
Patient Specific ◽  
Hypomethylating Agents ◽  
Advisory Committees

Introduction While the hypomethylating agents (HMAs) azacitidine (AZA) and decitabine (DAC) improve cytopenias and prolong survival in MDS patients (pts), response is not guaranteed. Timely identification of non-responders could prevent prolonged exposure to ineffective therapy, thereby reducing toxicities and costs. Currently no widely accepted clinical or genomic models exist to predict response or resistance to HMAs. We developed a clinical model to predict response or resistance to HMA after 90 days of initiating therapy based on changes in blood counts using time series analysis technology similar to the kind used in Apple's Siri or Google Assistant. In the setting of voice recognition, the sequence and context of words determines the meaning of a sentence; similarly, we hypothesized that the pattern of changes in MDS pts' blood counts would predict response or resistance early during treatment. Methods We screened a cohort of 107 pts with MDS (per 2016 WHO criteria) who received HMAs at our institution between February 2005 and July 2013 and had regular CBCs drawn during treatment. Mutations from a panel of 60 genes commonly mutated in myeloid malignancy were included. Responses were assessed after 6 months of therapy per International Working Group (IWG) 2006 criteria. Pts were divided randomly into training (80%) and validation (20%) cohorts. To address the potential for bias due to a small sample size, an oversampling algorithm was used to cluster similar pts based on their CBC data, Revised International Prognostic Scoring System (IPSS-R) score, and % bone marrow blasts at the time of diagnosis. CBC data from the first 90 days of treatment were fed into deep neural network (recurrent neural network) and decision tree algorithms, which were trained to predict whether pts would achieve a response (defined as complete remission (CR), partial remission (PR), or hematologic Improvement (HI)). Area under the curve (AUC) was used to assess model performance. Important features that impact the algorithm's predictions were extracted and plotted. Results 20747 unique data points were used, including CBC, clinical and genomic data. Among 107 pts, 61 (57.0%) received AZA only, 19 (17.8%) DAC only, 4 (3.7%) received both DAC and AZA, and 23 (21.5%) received HMA with an additional agent. Median age was 69 years (range: 37-100 years), and 27 (26.4%) were female. Forty pts (37.4%) were very low/low risk, 32 (29.9%) intermediate, 19 (17.8%) high, and 16 (14.9%) very high risk per IPSS-R. Responses included 23 (22.5%) CR, 2 (1.9%) marrow CR, 4 (3.9%) PR, and 20 (19.6%) HI. The most commonly mutated genes were ASXL1 (17.6%), TET2 (16.7%), SRSF2 (15.7%), SF3B1 (11.8%), RUNX1 (10.8%), STAG2(10.8%), and DNMT3A (10.8%). The median number of mutations per sample was 1 (range, 0-11), and 40 pts (39.2%) had > 3 mutations per sample. When trained using absolute values and changes in CBC values, the model's AUC was 0.95 in the training cohort and 0.83 in the validation cohort. When the cohort was oversampled to 1000 pts, the validation cohort AUC increased to 0.89. Feature extraction algorithms identified increases in MCV and RDW during weeks 2-8 of treatment, increased proportion of lymphocytes, decreased proportion of monocytes, and increased platelet counts during weeks 6-8 as factors favoring response to HMA. The model provides personalized, patient-specific predictions that correlate with blood counts (Figure 1). Conclusions We describe a machine learning model that monitors changes in blood counts during therapy with HMA to predict response or resistance to HMA in MDS pts. Such a model can be used to develop novel trial designs wherein pts predicted to not respond after 90 days of HMA treatment could be assigned to an investigational agent. Conversely, it would help inform the decision to continue HMA therapy in pts predicted to respond. Increasing sample size with oversampling dramatically increased model accuracy; a larger cohort of pts treated at different institutions is currently under development. Disclosures Sekeres: Millenium: Membership on an entity's Board of Directors or advisory committees; Syros: Membership on an entity's Board of Directors or advisory committees; Celgene: Membership on an entity's Board of Directors or advisory committees. Mukherjee:Partnership for Health Analytic Research, LLC (PHAR, LLC): Consultancy; Takeda: Membership on an entity's Board of Directors or advisory committees; Celgene Corporation: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Projects in Knowledge: Honoraria; Novartis: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Pfizer: Honoraria; McGraw Hill Hematology Oncology Board Review: Other: Editor; Bristol-Myers Squibb: Speakers Bureau. Advani:Glycomimetics: Consultancy, Research Funding; Kite Pharmaceuticals: Consultancy; Amgen: Research Funding; Pfizer: Honoraria, Research Funding; Macrogenics: Research Funding; Abbvie: Research Funding. Maciejewski:Alexion: Consultancy; Novartis: Consultancy. Nazha:Novartis: Speakers Bureau; Tolero, Karyopharma: Honoraria; Abbvie: Consultancy; Jazz Pharmacutical: Research Funding; Incyte: Speakers Bureau; Daiichi Sankyo: Consultancy; MEI: Other: Data monitoring Committee.

scholarly journals Transcriptomic Systems Biology Analyses with Buparlisib (BKM120) Pan PI3K Inhibition in Canine B Cell Lymphoma (BCL): Leveraging Comparative Oncology for Cancer Therapeutics and Biomarker Discovery

Blood ◽  
2018 ◽  
Vol 132 (Supplement 1) ◽  
pp. 4114-4114
Author(s):  
Ravi Dashnamoorthy ◽  
Afshin Beheshti ◽  
Sarah Cass ◽  
Athena Kritharis ◽  
Kristine Burgess ◽  
...  
Keyword(s):  
Clinical Study ◽  
Systems Biology ◽  
Cancer Risk ◽  
Board Of Directors ◽  
Differentially Expressed Genes ◽  
Research Funding ◽  
Biological Pathways ◽  
Differentially Expressed ◽  
Advisory Committees ◽  
Key Genes

Abstract Background: The canine is a highly appealing model for cancer research and discovery in part due to comparable histopathological features with humans, a fully intact immune system, similar clinicopathologic features, a more comparable body size and pharmacokinetic properties than the mouse, varied breed-specific incidence rates as well as a shared environment with humans. We and others have shown prominent transcriptomic overlap of human and canine NHL (cNHL) (McDonald T et al. Onctogarget, 2018). PI3K/Akt signaling plays an important role in lymphomagenesis, which is also a promising therapeutic target. However, identification of predictive genetic aberrations of therapeutic efficacy remains elusive. We evaluated the clinical activity of the pan-PI3K inhibitor, buparlisib, in a pilot clinical study in cNHL. Methods :We enrolled and treated 10 dogs with buparlisibwho were diagnosed with BCL in an IRB and IACUC approved clinical study. Cases included 2 treatment naïve and 8 dogs with relapsed disease that had relapsed s/p CHOP (6), L' asparaginase (1) and VELCAP (1) treatment. Pet owners were consented and the study subjects received buparlisib9mg/kg orally for 28 consecutive days. Analysis for tumor response were evaluated on weekly basis through direct tumor measurement or use of x-rays. Post-therapy fine needle aspirates (FNA) were collected on Days 0, 7 and 21 to examine predictors of response to BKM120. RNA from fine need aspirate cells were isolated and the transcriptomic changes were evaluated using Canine Genome 2.0 Affymetrix Array, followed by unbiased systems biology assessment for biological pathways using Gene Set Enrichment Analysis (GSEA) and Ingenuity Pathway Analysis (IPA). We performed unbiased assessment to determine pertinent biological pathways associated with treatment response. The overall impact was used to determine the global effect on tumor progression and cancer risk based on the specific regulation of each gene. A Carcinogenic Risk Score (CRS) was calculated based on these values to determine if there is a promoted risk for cancer (positive value) or inhibitory risk for cancer (negative value) by summing the log2 fold-change values of key genes and subtracting this from the sum of log2 fold-change values of the tumor suppressors when comparing pre-treated to BKM120 treated dogs. Results: Following four weeks of BKM120 treatment, the overall response rate was 30% with 1 complete response lasting 42 days; 2 partial responses lasting 55 and 72 days; 3 stable disease; and 4 progressive disease. Mild treatment related toxicities such elevated blood glucose, thrombocytopenia and anemia, fever, nausea and lethargic symptoms, with no treatment related toxicities in 2 cases were noted. Principal Component Analysis (PCA) and hierrachical clustering analysis of differentially expressed genes show that differentially expressed genes to cluster together in all dogs during post 2 week, indicating a consistent biological activity by BKM120 in all dogs regardless of breed, prior treatment or disease status. Pathway network analysis based on differentially expressed genes predicted activation of upstream regulators associated with tumor suppression including SOX1, SOX3 and GMNN (Week 1) and CEBPA (Week 2). Analysis of "key genes" involved in multiple biological processes appeared to be associated with response of PI3K inhibitortreatment. This included down regulation of CREBBP with a Cancer Risk Score (CRS) of -0.97 and downregulation of VIM, CDH3, WNT3, WNT5B and FGFR2 with a CRS of -2.98 (Fig 1). Conclusion: Results from our pilot study in cNHL showed encouraging clinical responses with a pan-PI3K inhibitor in 3 of 10 dogs. Furthermore, our unbiased characterization of biological pathways revealed that the observed GEP changes associated with tumor suppression and they reduced the risk for cancer progression. Overall, the canine model appears to be particularly attractive model that may be leveraged for the study of clinical and biological responses to novel therapeutic oncologic agents. Disclosures Evens: Bayer: Consultancy; Novartis: Consultancy, Membership on an entity's Board of Directors or advisory committees; Abbvie: Consultancy; Novartis: Consultancy; Acerta: Consultancy; Seattle Genetics, Inc.: Membership on an entity's Board of Directors or advisory committees, Research Funding; Pharmacyclics International DMC: Membership on an entity's Board of Directors or advisory committees; Tesaro: Research Funding; Janssen: Consultancy; Affimed: Consultancy.

scholarly journals Genomic Biomarkers Predict Response/Resistance to Lenalidomide in Non-Del(5q) Myelodysplastic Syndromes

Blood ◽  
2018 ◽  
Vol 132 (Supplement 1) ◽  
pp. 1797-1797 ◽  
Author(s):  
Yazan F. Madanat ◽  
Mikkael A. Sekeres ◽  
Sudipto Mukherjee ◽  
Cassandra M. Hirsch ◽  
Yihong Guan ◽  
...  
Keyword(s):  
Machine Learning ◽  
Board Of Directors ◽  
Association Rules ◽  
Lower Risk ◽  
Research Funding ◽  
Genomic Data ◽  
Advisory Committees ◽  
Cytogenetic Abnormalities ◽  
Myeloid Malignancies ◽  
Very High

Abstract Lenalidomide (Len) is FDA approved for the treatment of patients (pts) with lower-risk, transfusion-dependent myelodysplastic syndromes (MDS) with deletion(5q). It is frequently used in lower-risk pts with non-del(5q) MDS, with a transfusion independence response rate of 27%. Identification of pts who may or may not respond to Len can prevent prolonged exposure to ineffective therapy, avoid toxicities, and decrease unnecessary costs. Clinical or genomic data have limited utility in predicting response/resistance to Len. We developed an unbiased framework to study the association of several mutations/cytogenetic abnormalities in predicting response/resistance to Len in non-del(5q) pts, analogous to Netflix or Amazon's recommender system, in which customers who bought products A and B are likely to buy C: pts who have a molecular/cytogenetic abnormalities in gene A, and B are likely to respond or not respond to Len. Clinical and genomic data from pts with MDS or other myeloid malignancies diagnosed according to 2008 WHO criteria between 02/2004 and 06/2015 were analyzed. Next generation targeted deep sequencing panel of 50 genes that are commonly mutated in MDS and myeloid malignancies was included. Association rules using an apriori algorithm were used to study the relationships among multiple genes/cytogenetic abnormalities and response/resistance to Len. Responses included complete and partial remission and hematologic improvement (CR, PR, HI) per IWG 2006 criteria. Pts with stable disease or progressive disease were considered resistant. Association rules are a machine learning algorithm used to identify the association of variables based on their relationships. Rules with the highest confidence (that an association exists) and highest lift (measuring the strength of the association) were chosen. Of 139 pts treated with Len as monotherapy or in combination for at least 2 cycles included, 118 (85%) had MDS and 21 (15%) had other myeloid malignancies. Median age at diagnosis was 69 years (range 20-90 yrs) and 45% were female. Risk stratification by IPSS-R for MDS pts; 51.5 % had very low/low risk, 19.5% intermediate, and 29% high and very high risk disease. Most pts 100 (73%) had non-del(5q) abnormalities, others (39) had del(5q). Cytogenetic abnormalities for the non-del(5q) cohort included 58 pts with normal karyotype (NK), 19 pts with complex karyotype (CK), 4 pts with trisomy 8, 3 pts with del(7q) abnormalities, and 15 pts with other abnormalities. A total of 108 (79%) pts were treated with Len monotherapy. The median duration of treatment was 6 months (range 2- 66 m). Response rates were 46% (n=46) in the non-del(5q) cohort and 74% (n=29) in del(5q). Association rules identified the following combinations of genomic/cytogenetic abnormalities to predict response to Len in non-del(5q): (DDX41, NK) and (MECOM, KDM6A/KDM6B). The combination of the following abnormalities predicted resistance (ASXL1, TET2, NK), (DNMT3A, SF3B1), (TP53, del(5q)+CK), (STAG2, IDH 1/2, NK), (EZH2, NK), (BCOR/ BCORL1, NK), (JAK2, TET2, NK), (U2AF1, +/- ETV6, NK). [Table 1] Only TP53/CK mutations predicted resistance to Len in del(5q) pts. These associations are present in 39% of pts with non-del(5q), and have a specificity of 77%, with a negative predictive value and sensitivity=100%. The algorithm predicted response/resistance to Len with 82% accuracy. The median OS for non-del(5q) pts was 33.2m [95% CI: 19.9, 40.5]. The median OS for responders was 54.8 compared to 24.7 m for non-responders p=.017. The median OS for rules that predicted response was 70.3 m (95% CI: 70.3-NA). The median OS for pts with del(5q) + CK with a TP53 mutation was 9.8m. Several genomic combinations predicted very poor overall survival, including: (ETV6, U2AF1, NK), (BCOR/ BCORL1, NK), (EZH2, NK) , (JAK2, TET2, NK), with median OS of 10.7 m, 7.6 m, 10.8 m and 7.6 m, respectively. [Figure 1] Genomic biomarkers can identify 39% of non-del(5q) MDS pts who may or may not respond to treatment with very high accuracy. Although these abnormalities are only present in a subset of pts, treatment options for these pts can be tailored, by offering alternative therapies to pts with lower-risk disease who may not respond to Len, and preferentially offering Len to those who are more likely to respond. This study highlights how advanced analytic technologies such as machine learning can translate genomic/clinic data into useful clinical tools. Disclosures Sekeres: Opsona: Membership on an entity's Board of Directors or advisory committees; Celgene: Membership on an entity's Board of Directors or advisory committees; Opsona: Membership on an entity's Board of Directors or advisory committees; Celgene: Membership on an entity's Board of Directors or advisory committees. Gerds:Celgene: Consultancy; Apexx Oncology: Consultancy; CTI Biopharma: Consultancy; Incyte: Consultancy. Carraway:Celgene: Membership on an entity's Board of Directors or advisory committees, Research Funding, Speakers Bureau; Amgen: Membership on an entity's Board of Directors or advisory committees; Novartis: Speakers Bureau; Balaxa: Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; FibroGen: Consultancy; Jazz: Speakers Bureau; Agios: Consultancy, Speakers Bureau. Santini:Novartis: Honoraria; Amgen: Membership on an entity's Board of Directors or advisory committees; Otsuka: Consultancy; Janssen: Honoraria, Membership on an entity's Board of Directors or advisory committees; Celgene: Honoraria, Research Funding; AbbVie: Membership on an entity's Board of Directors or advisory committees. Maciejewski:Ra Pharmaceuticals, Inc: Consultancy; Ra Pharmaceuticals, Inc: Consultancy; Apellis Pharmaceuticals: Consultancy; Apellis Pharmaceuticals: Consultancy; Alexion Pharmaceuticals, Inc.: Consultancy, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Alexion Pharmaceuticals, Inc.: Consultancy, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau. Nazha:MEI: Consultancy.

scholarly journals Follicular Lymphoma Tumor-Cell Transcriptional Programs Associate with Distinct Somatic Alterations and Tumor-Immune Microenvironments

Blood ◽  
2021 ◽  
Vol 138 (Supplement 1) ◽  
pp. 1327-1327
Author(s):  
Jordan E. Krull ◽  
Kerstin Wenzl ◽  
Michelle K. Manske ◽  
Melissa A. Hopper ◽  
Melissa C. Larson ◽  
...  
Keyword(s):  
B Cell ◽  
Board Of Directors ◽  
Copy Number ◽  
Research Funding ◽  
Data Monitoring Committee ◽  
Advisory Board ◽  
Cell Suspensions ◽  
Advisory Committees ◽  
Scientific Advisory Board ◽  
Scientific Advisory

Abstract Background: Follicular lymphoma (FL) exhibits significant clinical, cellular, molecular, and genetic heterogeneity. Our understanding of FL biology and molecular classifications of FL, to date, has largely been driven by pathologic classification (Grade 1-3b), genetic classification (m7-FLIPI), or gene expression profiling (IR-1/2; Huet-23), along with limited studies on small cohorts or targeted panels. In order to further understand the biological underpinnings and complexity of FL, large-scale and integrated whole exome sequencing (WES) and RNA sequencing (RNAseq) studies are needed. Using a highly-annotated cohort of 93 FL tumors with matched RNAseq, WES, and CyTOF data, we have explored the transcriptomic signature of purified FL B cells and identified unique molecular subsets that are defined by distinct pathway activation, immune content, and genomic signatures. Methods: Frozen cell suspensions from 93 untreated FL (Grade 1-3b) patients' tumor biopsies, enrolled in the University of Iowa/Mayo Clinic Lymphoma SPORE, were used for the study. DNA was isolated from whole tumor cell suspensions, and RNA was isolated from both whole tumor and B cell-enriched cell suspensions. RNAseq and WES were performed in the Mayo Clinic Genome Analysis Core. RNAseq and WES data were processed using a standard pipeline and novel driver genes were identified using Chasm+ driver analysis. Copy number variants were identified from WES data using GISTIC 2.0. NMF clustering and single sample gene set testing for B cell lineage and tumor microenvironment (TME) signatures were performed in R using the NMF and singscore packages. Results: Unsupervised clustering of RNAseq data identified three distinct expression programs which separated patient B cell samples into 3 groups: Group 1 (G1, n=20), Group 2 (G2, n=24), Group 3 (G3, n=43). While no clinical attributes were defined by any single group, G1 consisted of cases that had less aggressive characteristics (63% Stage I-II, 79% FLIPI 0-1). To identify unique transcriptional pathways driving the three expression programs, we scored gene level contributions to NMF expression programs and employed gene set enrichment analysis. This revealed significant pathway association with type-I IFN signaling (G1), DNA repair and stress response (G2), and epigenetic modulation (G3) as differentiating programs between the 3 groups (FDR q&lt;0.001). VIPER master regulator activity inferencing revealed that these pathways were likely being controlled by differential activity in NF-kB, IRFs, STAT1, BCL6, and FOXO1. Each program significantly enriched for, but were not defined by, portions of specific germinal center programs, such as pre-memory (G1), light-zone-to-dark-zone transition (G2), and a pre-light-zone intermediate (G3). We next assessed the connection between B cell programs and the tumor microenvironment (TME) using available paired CyTOF data on 67 cases, which revealed an active TME in G1, with an abundance of CD8 T cell and NK cell populations, a wide variety of immune content in G2 that consisted mostly of Tfh and myeloid cells, and a poorly populated immune compartment in G3 compared to G1 and G2. Finally, somatic driver mutations and copy number alterations from WES were identified and associated with the three clusters. The three groups distinguished themselves by significant enrichment of copy number alterations (TNFAIP3-loss , 1q23-gain, 1q32-gain) in G2, while 10q-loss and mutations in BCL2 and chromatin modifiers (KMT2D and CREBBP) enriched in G3. G1, overall, had lower alteration burden and had weak associations with any specific alterations, suggesting an alternative mechanism for driving the G1 program. Conclusion: In this study, we have identified three unique FL tumor B cell groups, defined by specific transcriptional programs. Pathways such as inflammation, DNA damage response, and chromatin modification contribute most to the differences between B cell samples and group membership. Additionally, each program associated with specific genetic events as well as TME composition, highlighting potential drivers of these B cell states. This study improves the understanding of the mechanisms driving FL tumors and motivates further investigation into transcriptional consequences of genetic events as well as potential tumor intrinsic factors that may influence the TME. Figure 1 Figure 1. Disclosures Maurer: BMS: Research Funding; Genentech: Research Funding; Morphosys: Membership on an entity's Board of Directors or advisory committees, Research Funding; Kite Pharma: Membership on an entity's Board of Directors or advisory committees; Pfizer: Membership on an entity's Board of Directors or advisory committees; Nanostring: Research Funding. Rimsza: NanoString Technologies: Other: Fee-for-service contract. Link: MEI: Consultancy; Genentech/Roche: Consultancy, Research Funding; Novartis, Jannsen: Research Funding. Habermann: Tess Therapeutics: Other: Data Monitoring Committee; Seagen: Other: Data Monitoring Committee; Incyte: Other: Scientific Advisory Board; Morphosys: Other: Scientific Advisory Board; Loxo Oncology: Other: Scientific Advisory Board; Eli Lilly & Co.,: Other: Scientific Advisor. Ansell: Bristol Myers Squibb, ADC Therapeutics, Seattle Genetics, Regeneron, Affimed, AI Therapeutics, Pfizer, Trillium and Takeda: Research Funding. King: Celgene/BMS: Research Funding. Cerhan: Genentech: Research Funding; Regeneron Genetics Center: Other: Research Collaboration; Celgene/BMS: Other: Connect Lymphoma Scientific Steering Committee, Research Funding; NanoString: Research Funding. Novak: Celgene/BMS: Research Funding.

scholarly journals Determining Clinical Course of Diffuse Large B-Cell Lymphoma Using Targeted Transcriptome and Machine Learning Algorithms

Blood ◽  
2021 ◽  
Vol 138 (Supplement 1) ◽  
pp. 2395-2395
Author(s):  
Maher Albitar ◽  
Hong Zhang ◽  
Andre H. Goy ◽  
Zijun Xu-Monette ◽  
Govind Bhagat ◽  
...  
Keyword(s):  
Machine Learning ◽  
Overall Survival ◽  
Board Of Directors ◽  
Research Funding ◽  
Safety Monitoring ◽  
Advisory Committees ◽  
Short Survival ◽  
Naive Bayesian ◽  
Safety Monitoring Board ◽  
Naïve Bayesian

Abstract Introduction: Multiple studies have demonstrated that diffuse large B-cell lymphoma (DLBCL) can be divided into subgroups based on their biology. However, these biological subgroups overlap clinically. While R-CHOP (rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone) remains the standard of care for treating patients with DLBCL, predicting which patients will not benefit from such therapy is important so that alternative therapy or clinical trials can be considered. Most of the studies stratifying patients select biomarkers first, then explore how these biomarkers can stratify patients based on outcome. We explored the potential of using machine learning to first group patients with DLBCL based on survival, then isolating the biomarkers necessary for predicting these survival subgroups. Methods: RNA was extracted from tissue paraffin blocks from 379 R-CHOP treated patients with de novo DLBCL, and from 247 patients with extranodal DLBCL. A targeted hybrid capture RNA panel of 1408 genes was used for next generation sequencing (NGS). Sequencing was performed using an Illumina NextSeq 550 System platform. Ten million reads per sample in a single run were required, and the read length was 2 × 150 bp. An expression profile was generated from the sequencing coverage profile of each individual sample using Cufflinks. A machine learning system was developed to classify patients into four groups based on their overall survival. This machine learning approach based on Naïve Bayesian algorithm was also used to discover the relevant subset of genes with which to classify patients into each of the four survival groups. To eliminate the underflow problem commonly associated with the standard Naïve Bayesian classifiers, we applied Geometric Mean Naïve Bayesian (GMNB) as the classifier to predict the survival group for each patient. Results: Using machine learning, patients were first divided into two groups: short survival (S) and long survival (L). To refine this model, we used the same approach and divided the patients in each group into two subgroups, generating four groups: long survival in the long group (LL), short survival in the long group (LS), long survival in the short group (SL), and short survival in the short group (SS). The hazard ratio for this model was 0.174 (confidence interval: 0.120-0.251), and P-value &lt;0.0001. After defining these four groups, a machine learning algorithm was used to discover the biomarkers from the expression data of the 1408 genes from NGS data. To reduce the effects of noise and avoid overfitting, we employed a 12-step cross validation to obtain a robust measure. For an individual gene, a generalized Naïve Bayesian classifier was constructed on the training of one of the 12 subsets and tested on the other 11 testing subsets. This allowed us to limit the prediction process to 60 genes for each separation step. Using the selected biomarkers, we classified the patients in the original set (379 patients) into LL, LS, SL, and SS groups and then evaluated the survival pattern of these groups. As shown in Fig. 1A, the selected biomarkers predicted survival as expected in the overall survival groups prior to biomarker selection. For additional validation of the system, we used the selected biomarkers to classify a completely new set of 247 samples of patients with extranodal DLBCL. As shown in Fig. 1B, these selected biomarkers successfully predicted the overall survival in this group of patients with an HR of 0.530 (confidence interval: 0.234-1.197, P=0.005). This classification correlated with cell of origin classification, TP53 mutation status, MYC expression, and IRF4 expression. However, in a multivariate analysis, only TP53 mutation was independent in predicting prognosis (P=0.005) and age (below or over 60) (P=0.01) along with the survival grouping (P&lt;0.000001). Conclusions: Using a novel machine learning approach with the expression levels of 180 genes, we developed a model that can reliably stratify patients with DLBCL treated with R-CHOP into four survival subgroups. This model can be used to identify patients who may not respond well to R-CHOP to be considered for alternative therapy and clinical trials. Figure 1 Figure 1. Disclosures Hsi: AbbVie Inc, Eli Lilly: Research Funding. Ferreri: Ospedale San Raffaele srl: Patents & Royalties; BMS: Research Funding; Pfizer: Research Funding; Beigene: Research Funding; Hutchison Medipharma: Research Funding; Amgen: Research Funding; Genmab: Research Funding; ADC Therapeutics: Research Funding; Gilead: Membership on an entity's Board of Directors or advisory committees, Research Funding; Novartis: Membership on an entity's Board of Directors or advisory committees, Research Funding; Roche: Membership on an entity's Board of Directors or advisory committees, Research Funding; PletixaPharm: Membership on an entity's Board of Directors or advisory committees; x Incyte: Membership on an entity's Board of Directors or advisory committees; Adienne: Membership on an entity's Board of Directors or advisory committees. Piris: Millenium/Takeda, EUSA, Jansen, NanoString, Kyowa Kirin, Gilead and Celgene.: Membership on an entity's Board of Directors or advisory committees, Speakers Bureau. Winter: BMS: Other: Husband: Data and Safety Monitoring Board; Actinium Pharma: Consultancy; Janssen: Other: Husband: Consultancy; Agios: Other: Husband: Consultancy; Gilead: Other: Husband: Consultancy; Epizyme: Other: Husband: Data and Safety Monitoring Board; Ariad/Takeda: Other: Husband: Data and Safety Monitoring Board; Merck: Consultancy, Honoraria, Research Funding; Novartis: Other: Husband: Consultancy, Data and Safety Monitoring Board; Karyopharm (Curio Science): Honoraria.

scholarly journals A Novel Machine Learning-Derived Molecular Classification Scheme with Prognostic Significance

Blood ◽  
2021 ◽  
Vol 138 (Supplement 1) ◽  
pp. 3666-3666
Author(s):  
Tariq Kewan ◽  
Arda Durmaz ◽  
Hassan Awada ◽  
Carmelo Gurnari ◽  
Waled Bahaj ◽  
...  
Keyword(s):  
Machine Learning ◽  
Board Of Directors ◽  
Research Funding ◽  
Prognostic Significance ◽  
Molecular Classification ◽  
Molecular Architecture ◽  
Review Committee ◽  
Advisory Committees ◽  
Independent Review ◽  
Normal Cytogenetics

Abstract The gold standard for the diagnosis of MDS relies on morphologic alterations hampered by a great deal of subjectivity. Cytogenetic and clinical features allow for clinical classifications predictive of survival. However, with a few exceptions (SF3B1MT, del5q, and certain balanced translocations), neither classic histo-morphology nor prognostic scoring systems (e.g., IPSS-R) are reflective of pathogenic underpinnings. To date supervised analyses of mutational data did not succeed to produce profiles specific or predictive of traditional disease sub-entities. Large cohorts with clinical annotation and a sufficient follow-up allow for innovative biostatistical approaches to subgroup patients according to molecular profiles. Objective operator-independent subcategorization may be congruent with common pathogenic links, rational applications of targeted therapeutics and better prognostications. We hypothesized that machine-learning (ML) strategies used for analysis of mutational/cytogenetic profiles will enable recognition of invariant disease subcategories according to their molecular configurations. Herein, we compiled a meta-analytic database (our cohorts and publicly available sources) of 3,011 MDS (median age 71yrs) and 6,788 pAML/sAML. Results of deep targeted sequencing of a panel of 55 myeloid mutations were collected together with cytogenetics. We then performed unsupervised analysis of MDS and AML patients using Bayesian Latent Class Analysis (BLCA). A consensus matrix was then clustered using Ward's criteria to generate final cluster assignment based on the highest silhouette value. To identify genomic signatures, we used Random Forest classification and extracted mutations with highest global importance indicated by mean decrease in accuracy. Using BLCA we identified 5 unique genomic clusters (GCs) with 3 distinct prognostic outcomes [low risk (LR), intermediate risk (Int), and high risk (HR)] that were validated by survival analysis (Fig.1A,B). The LR group included GC-1 and was characterized by the highest prevalence of normal cytogenetics (100%) and SF3B1 MT (25%) with co-occurring DNMT3A MT (14%), and absence of ASXL1 MT, ETV6 MT, STAG2 MT, TP53 MT, and complex/abnormal cytogenetics. Int group included GC-2 and GC-4. GC-2 was characterized by a higher percentage of abnormal cytogenetics cases than LR group and absence of STAG2 MT, SRSF2 MT, ASXL1 MT, TP53 MT, and normal/complex cytogenetics. GC-4 had the highest frequency of SRSF2 MT (52%) with co-occurring ASXL1 MT (59%), TET2 MT (40%), normal karyotype, and absence of complex/abnormal cytogenetics. Finally, HR included GC-3 and GC-5. GC-3 included ASXL1 MT (67%) with co-occurring SRSF2 MT (47%), TET2 MT (37%), STAG2 MT (22%), and absence of normal cytogenetics. GC- 5 had the highest frequency of -5/del(5q) (50%), -7/del(7q) (43%), -17/del(17p) (16%) and the highest odds of complex karyotype (92%) as well as TP53 MT (48%). Paralleling the genomic ML-based clustering, the clinical relevance of these subgroups was reflected in significantly different survivals [median (95% CI)]: i) GC-1 [69 (59-80)], ii) GC-2 [35 (29-41)], iii) GC-3 [12 (10-16)], GC-4 [27 (22-34)], and GC-5 [9 (7-11)] months (Fig.1C). We then classified the MDS cohort according to the recently published and validated AML GCs (Awada et al Blood 2021) to investigate overlapping genomic features. Overall, 90% of MDS GC-1 and 67% of MDS GC-2 had the same molecular architecture of AML GC-2 and 70% of MDS GC-5 had the same molecular features of AML GC-4. In addition, 98% of MDS GC-3 and 92% of MDS GC-4 had the same features of AML GC-3 (Fig.1D). In sum, we propose a novel objective molecular classification of MDS and related diseases that allows subgrouping of patients according to shared pathogenesis for a better prognostic resolution without errors derived from subjectivity. The model was then internally and externally validated using a cohort of 200 cases. Results of a validation cohort and online URL site of molecular clustering will be presented at the meeting. Figure 1 Figure 1. Disclosures Balasubramanian: Servier Pharmaceuticals: Research Funding. Patel: Alexion: Consultancy, Other: educational talks, Speakers Bureau; Apellis: Consultancy, Other: educational talks, Speakers Bureau. Carraway: Celgene, a Bristol Myers Squibb company: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding, Speakers Bureau; Astex: Other: Independent review committee; Novartis: Honoraria, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Agios: Honoraria, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Takeda: Other: Independent review committee; Bristol Myers Squibb: Honoraria, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Jazz: Honoraria, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; AbbVie: Other: Independent review committee; Stemline: Honoraria, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau. Haferlach: MLL Munich Leukemia Laboratory: Other: Part ownership. Maciejewski: Regeneron: Consultancy; Novartis: Consultancy; Alexion: Consultancy; Bristol Myers Squibb/Celgene: Consultancy.

Sign in / Sign up

Export Citation Format

Share Document