ALBERT

Description: Background and objectives Protocols for acute myeloid leukemia (AML) 1st line patients are centered on the combination of Cytarabine and an anthracycline; Idarubicin (IDA), Daunorubicin (DNR), or Mitoxantrone (MIT). Patients may be treated with IDA, DNR, or MIT depending on the country of residence, because multiple clinical trials have not found significant differences among them. A new Personalized Medicine (PM) test developed by Vivia Biotech based on pharmacological responses in patient samples (ex vivo) is uncovering individual responses to these treatments. Our objective is to explore whether a significant % of individual patients may respond differently to IDA vs DNR vs MIT treatments, in spite that of their “on average” similar response shown by clinical trials. Patients and Methods Multicenter, prospective, non-interventional study of the PETHEMA group for treatment of AML. Bone Marrow (BM) samples were collected at diagnosis for 160 AML patients. Samples were incubated for 48 hours in 96-well plates, each well containing different drugs or drug combinations, each at 8 different concentrations, enabling calculation of dose response curves for each single drug (CYT, IDA, DNR, MIT) and combination used in treatments (CYT-IDA, CYT-DNR, CYT-MIT). Drug response was evaluated as depletion of AML malignant cells in each well after 48 hours incubations. Annexin V-FITC was used to quantify the ability of the drugs to induce apoptosis. Malignant cells were identified with monoclonal antibodies and light scatter properties. 1) We use the whole bone marrow sample, retaining the erythrocyte population and serum proteins, during the entire incubation period; and after 48 h leukocytes are isolated prior to evaluation by flow cytometry. 2) We have pioneered development of a proprietary automated flow cytometry platform called ExviTech. 3) Pharmacological responses are calculated using pharmacokinetic population models. Results Figure left panel shows dose responses for both IDA (red) and DNR (blue) in 125 AML patient samples. Although their average curves (thick red & blue) are similar, the interpatient variability of either drug is quite large. We hypothesized that some patients could show very differential sensitivities to both drugs, as illustrated by the green arrow where a patient sample is resistant to DNR (right shifted dose response curve) but sensitive to IDA (left shifted dose response curve). To identify these cases Figure right panel shows a comparison of the potency IDA vs DNR. Potency is represented by their EC50 (concentration that kills 50% of the cells). Most dots tend to line up, but red dots represent patient samples with a difference in potency between these drugs 〉30%. Repeating this exercise for IDA-MIT and DNR-MIT to cover all alternatives among the 3 anthracyclines identifies 40% of patients samples with 〉30% different potency among IDA-DNR-MIT. Repeating this exercise with the combination treatments CYT-IDA, CYT-DNR, CYT-MIT increases to 58% the population of patients whose samples have a differential sensitivity to these anthracyclines. A fraction of this 57% of patients may benefit in if treatment selection among these 3 treatments were to be aided by this ex vivo testing sensitivities. To identify which fraction would benefit we would need a trial specifically designed. Conclusions This preliminary results show that Vivia's PM test seems able to identify a subset of AML patients who's ex vivo pharmacological response to anthracycline drugs is significantly different. Because this ex vivo test accurately predicts the clinical response to CYT-IDA, if these selective anthracycline ex vivo responses translate to clinical responses, a fraction of this 57% subpopulation could benefit significantly from receiving 1st or 2nd line treatments based on either IDA, DNR, MIT, and their combinations. Hence this approach stands for European integration of treatment protocols, based on ex vivo individual responses data rather than nationality. Disclosures: Primo: Vivia Biotech: Employment. Hernandez-Campo:Vivia Biotech: Employment. Rojas:Vivia Biotech: Employment. Bennett:Vivia Biotech: Employment. Liebana:Vivia Biotech: Employment. Lopez:Vivia Biotech: Employment. Ballesteros:Vivia Biotech: Equity Ownership.

Description: Background and objectives Complete remission (CR) after induction therapy is the first treatment goal in acute myeloid leukemia (AML) patients. The aim of this study is to determine the ability of the Vivia’s novel ex vivo drug sensitivity platform Exvitech analyzing leukemic cell death to predict the CR rates after induction chemotherapy with cytarabine (Ara-C) and idarubicin (Ida) in 1st line AML. Patients and Methods This non-interventional and prospective study included samples from adult patients over 18 years of age diagnosed with de novo AML in Spanish centers from the PETHEMA group. Marrow samples were collected at diagnosis, sent to the Vivia laboratories, and incubated for 48 hours in whole samples in well plates containing Ara-C, Ida, or the combination Ara-C+Ida, each at 8 different concentrations to calculate dose responses. Annexin V-FITC was used to quantify the drug-induced apoptosis. Pharmacological responses are calculated using pharmacokinetic population models. Induction response was assessed according to the Cheson criteria (2003). Patients attaining a CR/CRi were classified as responders. The remaining patients were considered as resistant. Patients dying during induction response assessment were non-evaluable. The correlation was modeled using a generalized additive model with a logit link and a binomial distribution for residuals. Kernel density estimates were then used to plot empirical probability density functions for both groups. Their separation was quantified as the area under the ROC curve and a cut point was selected using the Youden’s criteria to optimize the classification probabilities (sensitivity, specificity). 95% confidence intervals for sampling errors were calculated for all these quantifiers. Results 125 patient samples were used to calculate the dose response curves for Ara-C alone, Ida alone, and the synergism of the Ara-C plus Ida combination. For clinical correlation we used 64 patients with a median age of 55 years (range 31 to 72). Dose responses for Ara-C alone are shown in Figure 1.A; note that for many samples there is a significant number (〉20%) of resistant cells to Ara-C (bracket). This is a strong clinical predictor of resistance because in the patient the drug will never be present at these high doses for 48 h. The second variable that is a good predictor of response is the synergism between these 2 drugs. The generalized additive model identified an algebraic combination of these 2 variables that yielded the best marker to separate both groups of patients. The probability density functions had minimal overlap. The area under the corresponding ROC curve was 0.965 (0.928, 1.000), and the classification probabilities for the optimal cut point (set at 0.414 for the marker), expressed as percentages, were 85% (62.1% to 96.8%) and 86.4% (72.6% to 94.8%) for sensitivity and specificity, respectively. Results are shown in Figure 1.B; Forty-four patients (68.8%) achieved CR after Ida+Ara-C, and the remaining 20 (31.3%) were resistant. Correlations of the PM test are shown in Figure 1.B. Seventeen of the 20 (85%) patients who fail to achieve CR were predicted as resistance in the ex vivo test. Thirty-eight of the 44 patients (86.4%) who achieved CR showed good ex vivo sensitivity to Ida+Ara-C predicting for CR. When the ex vivo test predicted a patient as sensitive it was correct in 38/39 cases (93%), and when it predicted resistant it was correct 17/23 cases (74%). Overall, 45 patients (86%) had an accurate prediction of their response to treatment. Conclusions This study shows that this novel ex vivo pharmacological profile test is able to predict the clinical response to Ida+Ara-C induction. We are increasing the number of patients in this ongoing study, and we are planning a PM Test-adapted Clinical Trial. Disclosures: Martínez: Vivia Biotech: Employment. Ortega:Vivia Biotech: Employment. Primo:Vivia Biotech: Employment. Hernandez-Campo:Vivia Biotech: Employment. Rojas:Vivia Biotech: Employment. Bennett:Vivia Biotech: Employment. Liebana:Vivia Biotech: Employment. Lopez:Vivia Biotech: Employment. Ballesteros:Vivia Biotech: Equity Ownership.

Description: Background To aid in the identification of effective treatments for individual patients, ex vivo assays for detecting cell death inducible by drugs for hematological malignancies have been in development for over 20 years. We have developed a novel approach incorporating 4 key innovations; incubating drugs in whole bone marrow sample without isolating leukocytes, using flow cytometry enables identification of the malignant cells selectively, an automated flow cytometry-based platform (ExviTech) decreases errors and enables full pharmacological characterization, and analyzing the data using pharmacodynamic population models. Aim The purpose of this study is to derive the ex vivo pharmacological profiles across the AML patient population of single drugs and combination treatments as a tool for individualized treatment selection. Patients and Methods Bone-marrow samples from 160 patients diagnosed with AML were sent to Vivia from 24 hospitals across Spain within 24 hrs. The plates were incubated for 48-hours prior to analysis with ExviTech, The percentage of leukemic cell death was determined via labeling with monoclonal antibodies and AnnexinV-FITC. A survival index is computed for each drug, the lower the survival index, the more effective the drug. Dose-response curves of cytarabine, idarubicin, daunorubicine, etoposide, mitoxantrone, fludarabine, clofarabine, and 6-thioguanine were measured in 160 patient samples. The added benefit of combining these drugs into 12 combination treatments was assessed by measuring their synergy in each individual patient. In 39 patients treated with CYT IDA we had clinical data of response, and then we performed a blinded interpretation of this in vitro test by an expert hematologist, to predict the clinical response based in this test result. Results There was a large range of interpatient variability in the response to a single drug and even larger in the synergism between drugs. The Population Pharmacological Profiles for an individual patient is shown on the figure below. The relative drug potency in terms of their percentile ranking within the population is shown in the left panel from 0 (weakest) to 100 (most potent). Green lines represent the individual patient potency relative to the population ranking, with confidence intervals. Third column lists when a drug leaves a significant % of leukemic cells alive, potential resistant clones. The panel on the right side shows the synergism of the drug combinations treatments shown as box-plots at 10-25-75-90% to highlight their distribution. The synergism value for an individual patient in each combination is shown in green, with confidence interval as parallel dotted green lines. This representation of the Pharmacological Profile of an individual patient sample quickly identifies extreme values, when a drug or combination is very sensitive (rightward shift green lines, green boxes) or very resistant (leftward shift green lines, red boxes). This patient showed average sensitivities for most drugs though highly resistant to Clofarabine (red box) that leaves 45% alive. However this patient showed lack of synergism in multiple treatments (right, red boxes). CYT and IDA show average potencies but lack of synergism, suggesting CYT-DAU might be a more efficient treatment. These representations lead to clear guidelines in 〉90% samples, and based on hematologist's interpretation of these guidelines show a clinical correlation with clinical responses to CYT-IDA of 84%. Conclusion We have developed an improved a methodology to measure the pharmacological activity of drugs and drug combinations in AML patient samples as well as modeling their pharmacological behavior. This information may be useful in selecting the optimal treatment for the individual patient, especially relapse/refractory patients in need of therapeutic alternatives. By testing the drugs used in the treatment protocols for AML directly on patient samples, a pharmacological based model has been developed to infer drug resistance or sensitivity, patient by patient. Disclosures: Ballesteros: Vivia Biotech: Equity Ownership. Primo:Vivia Biotech: Employment. Hernandez-Campo:Vivia Biotech: Employment. Rojas:Vivia Biotech: Employment. Liebana:Vivia Biotech: Employment. Lopez:Vivia Biotech: Employment. Iñaki:Vivia Biotech: Consultancy. Bennett:Vivia Biotech: Employment.

Description: Background: We have overcome the limitations of 40 years of ex vivo testing. The aim of this study is to determine the ability of Vivia's novel test to predict the complete remission (CR) rates after induction chemotherapy with cytarabine (Ara-C) and idarubicin (Ida) in 1st line AML. Material and Methods: Bone marrow samples from adult patients diagnosed with de novo AML in Spanish centers from the PETHEMA group were included. Whole marrow samples maintaining their Native Environment were incubated for 48h in well plates containing Ara-C, Ida, or their combination. Pharmacological responses are calculated using population models. Induction response was assessed according to the Cheson criteria (2003). Patients attaining a CR/CRi were classified as responders and the remaining as resistant. Results: 390 patient samples were used to calculate the dose response (DR) curves for Ara-C alone, Ida alone, and their synergism. For clinical correlation we used 142 patients with median 56 years. The strongest clinical predictor was the Area Under the Curve (AUC) of the DR of Ara-C, and the AUC of IDA. The GAM models revealed a significant relationship between the AUC of the concentration-effect curves of both, idarubicin and, particularly, Ara-C, with greater values associated to higher probabilities of post-induction resistance. The fitted Generalized Additive Method predictions of expected values for each patient were in turn related to overall survival when a discrimination value to define positive and negative test results that prioritized specificity over sensitivity was chosen based on equaling positive and negative predictive values (Fig 1A). Prioritizing specificity over sensitivity reflects the higher cost of false positive over false negative decisions: only in very rare instances, an effective treatment would be erroneously negated to a sensitive patient at the expense of overlooking a number of resistant patients. However, the later patients could take their chances on re-induction therapy. While for diagnostics sensitivity and specificity should both be optimized, for Personalized Medicine the positive and negative predictive values should be optimized preferentially because they define the patient response correlation. Fig 1B shows a table illustrating the correlation between clinical outcome (columns) and the test predictions (lines). From a diagnostic criteria (columns), clinically resistant patients (1st column) are not well predicted with a Sensitivity of 51%, while clinically sensitive patients (2nd column) are very well predicted with a Specificity of 94%. From a Precision Medicine criteria (Lines), patients predicted resistant (1st line) and well predicted with 80% positive predictive value, similar to patients predicted sensitive (2nd line) well predicted with 79% Negative Predictive Value. The test does not properly identify 23/142 that are clinically resistant and the test predicts as sensitive (bottom left quadrant right panel). This mismatched subgroup mimics the problems from molecular markers where a resistant clone present in a minority of leukemic cells cannot be detected yet drives the patient response. However, this group mismatch does not prevent a good correlation with the test predicted outcomes. Flow cytometry identified 2 clones in 75% of these 23 samples, and we revised all samples analyzing each of 2 clones separately whenever they were present. Results did not change by this clonal analysis, suggesting flow cytometry may not identify resistant clones. Future improvements of the test adding 16 concentrations to the dose response curves may be able to detect the presence and parameters of these resistant clones driving patient response. Conclusions: This novel test is able to predict the clinical response to Ida + Ara-C induction with overall correlation and predictive values of 80%, higher than ever achieved. It is significantly higher than the current clinical response rate of 66.7%. Thus this novel test may be valuable information to guide 1st line patient therapy. Figure 1. ROC curve and clinical correlation Figure 1. ROC curve and clinical correlation Disclosures Ballesteros: Vivia Biotech: Employment. Cordoba:Celgene: Research Funding. Ramos:GlaxoSmithKline: Honoraria; Janssen-Cilag: Honoraria, Membership on an entity's Board of Directors or advisory committees; Novartis: Consultancy, Honoraria; Celgene Corporation: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding; Amgen: Consultancy, Honoraria. Gaspar:Vivia Biotech: Employment. Gorrochategui:Vivia Biotech: Employment. Rojas:Vivia Biotech: Employment. Gomez:Vivia Biotech: Employment. Hernández:Vivia Biotech: Employment. Robles:Vivia Biotech: Employment.

Description: Cytogenetic analysis is still an important and mandatory component of Acute Myeloid Leukemia (AML) diagnosis and prognosis. Pretreatment cytogenetic and molecular genetic findings are one of the major independent prognostic markers in AML, and they determine chemotherapy response and outcome. However, cytogenetic does not provide alternative treatments when a patient have a high cytogenetic risk, and requires relatively long time until obtaining the results despite the treatment of these patients should begin as soon as possible. The aim of this study is providing data about the utility of a new AML Precision Medicine (PM) Test as a complementary tool to conventional cytogenetic to overcome the main obstacles this later has. For this purpose, AML bone marrow from 111 patients were received at the laboratory 24h from extraction and incubated for 48h in 96-well plates containing single drugs or combinations, representing up to 31 different treatments that are currently given in the clinical practice. The analyses were performed in the automated flow cytometry PharmaFlow platform and the test results can be sent to the hematologists 72h after the extraction of the sample. Pharmacological responses were calculated using pharmacokinetic population models. Induction response was assessed according to the Cheson criteria (2003). Patients attaining a complete remission (CR) or CR with incomplete blood count recovery (CRi) were classified as responders and the remaining as resistant, excluding early deaths. The probability of being resistant or non-responder was modeled using binary logistic generalized additive models (GAM) with Cytarabine (CYT) and Idarubicin (IDA) area under the curve (AUC) data and over the cytogenetic risk (favorable/intermediate/adverse). The empirical ROC curves were calculated for the probabilities of being non-responder from each GAM. Final scores and treatments ranking are based on a therapeutic algorithm that integrates ex vivo activity of single drugs, quantified by the AUC and synergism, referred as α parameter, using a surface interaction model. Clinical and cytogenetic risk data of the patients were monitored and collected. A simple logistic model of the probability of being non-responder over the cytogenetic risk (favorable/intermediate/adverse) explained less variability (29.4%) than the GAM over the AUC values (40.8%) in the subset of 111 patients in whom the cytogenetic risk was informed. Figure 1 shows the results of the clinical correlation of cytogenetics vs PM Test in the cohort of 111 patients analyzed. In both approaches prediction of sensitive patients (Negative Prediction Value, NPV) is better than resistant patients (Positive Predictive Value, PPV), being the PM Test slightly better in predicting the sensitive patients (NPV=93% vs 88%), while the cytogenetics shows a 20% improvement in the prediction of resistant patients (PPV= 76% vs 56% with PM Test). The correlation achieved by the PharmaFlow PM test was 80% that is almost similar than the correlation obtained with the cytogenetic data using the same cut off point (86%). Figure 1 (right) also shows an example of the classification of AML treatments with the PharmaFlow PM Test in a patient sample according to a color scale from higher (green) to lower (red) ex vivo activity. In summary, despite the PharmaFlow PM Test and cytogenetics provide similar information, results from cytogenetic risk are available typically in 10-14 days, and thus after patient treatment, while results from this novel PM Test are available in 48-72h, prior to treatment. Hence, this novel approach provides information to hematologist with higher predictive value than risk factor (deviance explained 40.8% vs 29.4%) and ahead of treatment, and thus represent a valuable in-time prior to treatment decision making. In addition, the PM Test can provide alternative treatments to AML patient in a basis of their ex vivo activity. Disclosures Ballesteros: Vivia Biotech: Employment. Martinez Lopez:Novartis: Research Funding, Speakers Bureau; Celgene: Research Funding, Speakers Bureau; Bristol Myers Squibb: Research Funding, Speakers Bureau; Janssen: Research Funding, Speakers Bureau. Hernandez:Vivia Biotech: Employment. Primo:Vivia Biotech: Employment. Gorrochategui:Vivia Biotech: Employment. Rojas:Vivia Biotech: Employment. Montesinos:Novartis: Research Funding, Speakers Bureau; Daiichi Sankyo: Consultancy, Speakers Bureau.