New standards have been recently defined and implemented enabling a reliable exchange of pharmacometric models across software tools, and facilitating collaborative drug and disease modeling and simulation (M&S) activities. Among these, the Standard Output (SO) has been proposed as the tool-independent exchange and storage format for typical M&S results. The SO integration within the Drug Disease Model Resource (DDMoRe) interoperability framework (IOF) has already shown its potential for enabling effective data flow across modeling tasks and facilitating information retrieval. Pharmacometric M&S input to study design and decision making frequently requires a variety of methods, potentially using disparate software tools.1-5 There is generally a lack of consistency across these tools in model definition, data format, numerical methods, and output of results. This limits the potential reuse of models and methods across tools by the analyst, who would then need to understand varying model and tool-specific requirements to correctly translate and recode models and data. This lack of standard languages and interoperability across tools also significantly limits the integration of tools into seamless pharmacometric workflow as well as the sharing of existing knowledge through shared code. The DDMoRe project has addressed these shortcomings through the definition, implementation and integration of a set of standards into the so-called IOF.6-9 Among these, the SO has been proposed as the tool-independent format for storing output typically produced from pharmacometric M&S tasks and workflows to guide the model building through assessment of the goodness-of-fit between the model and the dataset and the appropriateness of the underlying model assumptions, as well as to allow the reproducibility of the results. Developing a standardized output format across M&S tools allows better sharing of information, reusable code, and better integration of a wide variety of tools within a single pharmacometric workflow. Although some initiatives aim at improving the standardization of input M&S data (e.g., CDISC, COMBINE, and ISoP Data Standards Working Group), to our knowledge, the SO represents the first standard format proposed for M&S output data. The SO is an XML-based exchange format intended for storage of results in a standardized form. It is complementary to the Pharmacometrics Markup Language (PharmML),8 representing the exchange medium for mathematical and statistical models across various target tools. The SO is based on a hierarchical structure defined to separate independent output information or results obtained from different modeling tasks (e.g., estimation, simulation, and optimal design), whereas standardizing and providing consistent definition of output structure from different estimation methods and tools (e.g., for maximum likelihood and stochastic or sample-based estimation). The standardization is intended to be method specific, rather than software specific. As shown in Figure 1, it consists of seven main sections that comprise specialized SO elements, with varying number of levels. A tree plot of the SO structure is also provided as Appendices S2. The Tool Settings and Raw Results sections are designed to store references to related files. Specifically, the first one keeps the location to files containing tool settings (e.g., task related and graphical) of the performed task. Original data and graphical output files produced by target tools (e.g., WINBUGS, Monolix, NONMEM, PsN, PopED, and PFIM) can be retrieved within the respective elements of the Raw Results section. In the Task Information section, five elements provide the user with information about the modeling step execution: tool message and diagnostic information, execution time, the path pointing to the original output file, number of chains (in case of Markov Chain Monte Carlo (MCMC) execution), and iterations. Warning contents, errors, and program termination are captured here within a dedicated element (). The Estimation section is designed to capture typical results from an estimation step and it is composed of seven main elements. Four elements allow storage of the population parameter estimates and individual parameter estimates along with their uncertainty measures (e.g., covariance, correlations, standard errors, and confidence intervals). Different child-elements are used to account for different techniques (e.g., maximum likelihood estimateion (MLE) and Bayesian), for different measures of central tendency in sampled-based estimation methods and for different variance components (e.g., interindividual/interoccasion variability). In addition, multiple types of residuals (e.g., population residuals, individual weighted residuals, and conditional weighted residuals), and individual predictions can be stored within two separate elements. Finally, the element contains the values of various estimated objective function measures, such as the likelihood, log-likelihood, tool-specific objective function, deviance, and individual contributions to likelihood, as well as the calculated values of the most common information criteria used for model selection purpose (i.e., Akaike information criterion, Bayesian information criteria, and Deviance Information Criterion). To support result's reproducibility, a dedicated section, which we called Model Diagnostic, is designed for storing information for typical model diagnostic plots used to assess the structural model and residual error specification (e.g., individual observations vs. predictions and visual predictive check), and diagnostic plots for individual parameters. This allows the user to use these values in creation of diagnostic plots without further data manipulation. The Estimation and Model Diagnostic sections can store information on parameters at arbitrary variability levels, and are not limited to interindividual variability. In fact, any level below (e.g., occasion) or above (e.g., country or center) the subject level can be considered. Values for relevant random effects are captured using the “columnType” and “level” attributes. In the Simulation section, results from simulations can be captured for each replicate by using separate blocks of child elements. These include the simulated time course of each subject, population and individual parameters, random effects, covariates, dosing records, and references to raw results. This allows prediction of outcomes for a single set of model inputs as well as clinical trial simulations that could incorporate variability in model inputs and parameters. Results from both evaluation and optimization design steps can be stored within the Optimal Design section. Similar to the Simulation section, this was structured to capture results from single as well as multiple evaluation or optimization steps in separate blocks. The Fisher Information Matrix, variance/covariance matrix, parameter values along with their precision, information about the adopted criteria, and performed tests, are some of the output information stored within the defined child elements. The SO has been designed to allow flexibility in populating its elements. Sections or elements can be omitted if these are not available or not appropriate for a given target software. In addition, elements can be populated through the defined XML structure (e.g., as inline data or by referring to external files). Unique data types were defined for each SO element reusing some of the structures in the PharmML schema (e.g., matrices and datasets) and distribution specifications from ProbOnto (e.g., posterior distributions).8, 9 For data stored in a tabular form, the columnType attribute was defined to support column descriptors (e.g., individual parameter, population parameter, and random effect) that improve usability and interpretation of stored values. As an example, the attribute of a column relative to a variability parameter would be specified to clarify whether this reports a variance, standard deviation, covariance, or correlation measure. A detailed description of the structure, features, and XML format specifications can be found in the SO user guide available at https://www.ddmore.foundation/our-standards/ and http://www.pharmml.org/. The user guide of SO specification format version 0.3.1 used in the official IOF public release is also provided as Appendices S1 with copyright permissions. LibSO, a Java library, was developed to provide basic programmatic functionality for creation and validation of SO data XML documents. The library is available via SourceForge within the libPharmML project (https://sourceforge.net/projects/libpharmml.ddmore.p). This library depends on libPharmML and uses the same design pattern. A stand-alone validator, executable via the command line, is also provided (https://sourceforge.net/p/ddmore/libpharmml/wiki/Home/). An alternative library libsoc, developed in C, is also available. It facilitates bindings to multiple languages and an R package based on this library is available on CRAN (https://CRAN.R-project.org/package=libsoc). The R package translates all the data types of the SO into native R data types. As one of the core standards of the DDMoRe IOF, use of the SO has been integrated in the DDMoRe R package, and it has shown its benefit in effectively supporting pharmacometric workflow interactions and information retrieval.10 The SO elements could be populated following estimation and model qualification tasks in PsN, NONMEM, Monolix, and BUGS, and simulation tasks in NONMEM and PsN, thanks to the developed converters and connectors. The SO facilitates interoperability across tools in an analogous way to PharmML, because we can easily pass output information between tools by extracting or converting the information in the SO as input for the next tool. A simplified representation of this workflow, also involving additional tools, is shown in Figure 2. Larizza et al.10 have shown how the DDMoRe standards, including SO, facilitate complex workflow incorporating BUGS and NONMEM to compare methodological approaches. Example workflows available with the DDMoRe IOF show how the SO can be used to update model parameters following estimation and to facilitate model diagnostics in a tool independent way. Examples of this are provided as Appendices S3. SO provides a standard for output information produced in a pharmacometric workflow. It has demonstrated effective data flow across pharmacometric tasks, between software tools, and facilitates information retrieval for postprocessing, reviewing, and reporting. As the other DDMoRe standards, the SO definition is open-source and further development and improvements are expected. Specifically, suggestions from the M&S community to the SO definition are encouraged to enhance collaborative drug and disease modeling based on an efficient exchange and reuse of knowledge between all stakeholders (regulators, academics, and pharmaceutical industries) involved in drug discovery and development. The authors wish to thank the input of many colleagues across the DDMoRe project who have contributed to discussion of and providing training for SO. This study has received support from the Innovative Medicines Initiative Joint Undertaking under grant agreement number 115156, resources of which are composed of financial contributions from the European Union's Seventh Framework Programme (FP7/2007-2013) and EFPIA companies in kind contribution. The DDMoRe project is also financially supported by contributions from Academic and SME partners. As an Associate Editor for CPT: Pharmacometrics & Systems Pharmacology, France Mentré was not involved in the review or decision process for this article. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.

Sharing of resources across organisations to support open innovation is an old idea, but which is being taken up by the scientific community at increasing speed, concerning public sharing in particular. The ability to address new questions or provide more precise answers to old questions through merged information is among the attractive features of sharing. Increased efficiency through reuse, and increased reliability of scientific findings through enhanced transparency, are expected outcomes from sharing. In the field of pharmacometrics, efforts to publicly share data, models and workflow have recently started. Sharing of individual-level longitudinal data for modelling requires solving legal, ethical and proprietary issues similar to many other fields, but there are also pharmacometric-specific aspects regarding data formats, exchange standards, and database properties. Several organisations (CDISC, C-Path, IMI, ISoP) are working to solve these issues and propose standards. There are also a number of initiatives aimed at collecting disease-specific databases - Alzheimer's Disease (ADNI, CAMD), malaria (WWARN), oncology (PDS), Parkinson's Disease (PPMI), tuberculosis (CPTR, TB-PACTS, ReSeqTB) - suitable for drug-disease modelling. Organized sharing of pharmacometric executable model code and associated information has in the past been sparse, but a model repository (DDMoRe Model Repository) intended for the purpose has recently been launched. In addition several other services can facilitate model sharing more generally. Pharmacometric workflows have matured over the last decades and initiatives to more fully capture those applied to analyses are ongoing. In order to maximize both the impact of pharmacometrics and the knowledge extracted from clinical data, the scientific community needs to take ownership of and create opportunities for open innovation.

The lack of a common exchange format for mathematical models in pharmacometrics has been a long-standing problem. Such a format has the potential to increase productivity and analysis quality, simplify the handling of complex workflows, ensure reproducibility of research, and facilitate the reuse of existing model resources. Pharmacometrics Markup Language (PharmML), currently under development by the Drug Disease Model Resources (DDMoRe) consortium, is intended to become an exchange standard in pharmacometrics by providing means to encode models, trial designs, and modeling steps.

Population modeling of tumor size dynamics has recently emerged as an important tool in pharmacometric research. A series of new mixed-effects models have been reported recently, and we present herein a synthetic view of models with published mathematical equations aimed at describing the dynamics of tumor size in cancer patients following anticancer drug treatment. This selection of models will constitute the basis for the Drug Disease Model Resources (DDMoRe) repository for models on oncology.

Introduction: Ruxolitinib is a JAK 1/2 inhibitor approved in patients with myelofibrosis (MF). The safety profile is well characterized with anemia being a common adverse event. The aim of this mathematical modeling analysis is to characterize individual hemoglobin (Hb) dynamics during ruxolitinib treatment, and to establish a dose-response relationship between treatment and occurrence of an anemic event as defined by an on-treatment reduction in Hb to < 8g/dL.

Purpose The pan-Class I PI3K inhibitor buparlisib (BKM120) has shown activity in a range of preclinical cancer models. This first-in-man study was initiated to identify the maximum tolerated dose (MTD) of buparlisib (100 mg/day) and to assess safety and preliminary efficacy. Methods Patients with advanced solid tumors (N = 83) enrolled in a Phase I dose-escalation and -expansion study of single-agent buparlisib. Patients in the dose-expansion arm (n = 43) had tumor samples with PIK3CA and/or PTEN alterations. Results The most common cancers were colorectal (n = 31) and breast cancer (n = 21). Median number of prior antineoplastic regimens was four (range: 1–12). Grade 3/4 adverse events (AEs) included asthenia (12.0 %) and performance status decrease (9.6 %). Treatment-related AEs (all grades) included decreased appetite, diarrhea, nausea (each in 33 % of patients), hyperglycemia (31 %) and rash (29 %). One confirmed partial response (PR; triple-negative breast cancer) and three unconfirmed PRs (parotid gland carcinoma, epithelioid hemangiothelioma, ER + breast cancer) were reported. Tumor molecular status did not predict clinical benefit in the full study cohort, or among the colorectal or breast cancer subpopulations. Pharmacodynamic biomarkers (18F-FDG-PET, C-peptide, pS6) demonstrated dose-dependent changes; however, tumor heterogeneity precluded a clear correlation with clinical benefit. Conclusion Buparlisib was well tolerated up to the 100 mg/day dose and showed preliminary activity in patients with advanced cancers. Future studies in more homogeneous patient populations will evaluate buparlisib in combination with other agents and further investigate the use of predictive biomarkers.

ABSTRACT Background The MTD of single-agent BKM120 was previously declared as 100 mg/day in a Phase I study in patients (pts) with advanced solid tumors (NCT01068483; Bendell et al. 2011). Here we report on the analysis of pharmacodynamic biomarkers from pts in the dose-escalation and dose-expansion arms of the Phase I study. The inhibitory effects of BKM120 were investigated (i) in the context of glucose metabolism regulation, a known PI3K-dependent process, and (ii) in the context of tumor biology, by surveying the phosphorylation of proteins downstream of PI3K. Methods 83 pts received oral daily BKM120. Tumor samples were analyzed for PIK3CA mutation and PTEN expression. Blood C-peptide (glucose metabolism) was measured at Cmax (2–4 hrs post-dose) on Day 1 in 72 pts. Pathway phosphorylation in surrogate tissue was investigated in 51 paired skin samples by correlating changes in pS6 (baseline to Day 28) with the mean dose of BKM120 administered in each pt. In 8 pts where pre/post treatment biopsies could be obtained, pS6, pAKT, p4EBP1 and Ki67 were quantified within limits of tissue availability. Results Blood C-peptide at Cmax on Day 1 increased as a function of the 1st dose administered, with no change after 12.5 mg and mean increase of 72% (range: 56–84%) after 150 mg. Inhibition of pS6 on Day 28 (range: 20–60%) was observed in tumor tissue from 5 out of 8 pts. 4 of these 5 pts also demonstrated a decrease in either pAKT (range: 30–70%) or p4EBP1 (range: 27–35%) with 1 pt displaying a mean 30% decrease in all 3 markers. A decrease in proliferation as measured by Ki67 was observed in 4 pts (range: 13–60%) suggesting a potential biological impact in response to PI3K pathway inhibition. Inhibition of pS6 in skin increased moderately with the mean administered dose of BKM120 (mean inhibition of –28, –37, and –40%, for 12.5–60 mg, >60–90 mg, and >90 mg dose ranges, respectively) suggesting a relationship between treatment dose and the degree of PI3K pathway inhibition. Conclusion This analysis supports the notion that daily BKM120 not only has the ability to induce inhibition of the immediate effector of PI3K (pAKT) but also to downregulate the pathway further downstream (pS6) at MTD. Disclosure F. Eskens: Membership on an advisory board: Participant of LEAD summit meetings (Novartis). E. Di Tomaso: Employee of Novartis. D.W. Sternberg: Employee of Novartis. L. Wang: Employee of Novartis. C. Sarr: Employee of Novartis. Stock Ownership in Novartis Pharmaceuticals. J. Baselga: Novartis Scientific Ad Board (NVPBKM120); other pan‐PI3K inhibitors: Consulting for Exelixis; XL147; Roche‐Genentech Scientific Ad Board; GDC0941. All consulting relationships have CA's in place which conform to the Partners/Harvard COI Policies. All other authors have declared no conflicts of interest.

2543 Background: Fridlyand (2011) retrospectively compared PFS vs. change in tumor burden as a primary endpoint in phase II non-small cell lung cancer (NSCLC) trials to inform phase III decision making and found the use of PFS was superior. Since the classic tumor burden model only uses measurements of target lesions, we investigated whether the model could be strengthened by incorporating new and non-target lesion progression. The ability to use a strong tumor burden model has the benefit of potentially earlier decision making and considerable timeline savings. Methods: We analyzed five phase III trials of combination chemotherapy ± targeted therapies with an OS primary endpoint: 1st, 2nd line NSCLC (ATTRACT-1, -2), 1st, 2nd line colorectal carcinoma (CONFIRM-1, -2), and 2nd line ovarian cancer (EPO906A2303). We applied Cox’s proportional hazards model to OS using the covariates of baseline tumor burden, 1st tumor assessment percentage change from baseline, new lesions, and non-target PD. Results: See table. Conclusions: We show that predictive models for OS should consider new and non-target lesions for PD, as well as target lesion tumor burden, findings independently corroborated by Suzuki (2011). We propose a longitudinal rank-based randomized phase II design, ranking a patient’s risk of death, differentially weighting PDs by type and time of PD, and percentage change in tumor burden. This may be more informative for phase II decision making for phase III trials based on OS, than PFS which only uses time of PD. Further studies with other tumor types and treatment modalities are warranted. [Table: see text]

The phase 3 RECORD-1 trial (NCT00410124) established the efficacy and safety of everolimus in patients with metastatic renal cell carcinoma (mRCC) who progress on sunitinib or sorafenib. In RECORD-1, patients received 10 mg everolimus daily, with dose reduction to 5 mg daily allowed for toxicity. We have developed a model of tumor growth dynamics utilizing serial measurements of the sum of the longest tumor diameters (SLD) from individual RECORD-1 patients to define the dose–response relationship of everolimus.

4602^ Background: The randomized, placebo-controlled phase III trial RECORD-1 (NCT00410124) established mammalian target of rapamycin (mTOR) inhibitor everolimus as an effective therapy for prolonging progression-free survival (PFS) in patients with advanced RCC who had progressed after sunitinib or sorafenib. The 10-mg daily everolimus dose administered in RECORD-1 was based on phase I studies correlating this regimen with constant and near-complete inhibition of mTOR pathway signaling. Dose reduction to 5 mg daily was allowed for toxicity. We developed a mathematical model of tumor growth in RECORD-1 to evaluate the effect of these 2 everolimus doses on growth of target lesions, nontarget lesions, and new metastases. Methods: Tumor growth in all patients with a baseline tumor measurement (n = 407) was described using nonlinear mixed effects modeling. Local radiological data was collected over time on the sum of largest target lesion diameters (SLD), progression status of nontarget lesions, and appearance of new lesions. By fitting a mathematical model for tumor growth to each patient, the impact of everolimus dose on all 3 lesion types was investigated. Results: Everolimus slowed growth of all 3 lesion types versus placebo (P <.0001). For target lesions, a 10-mg dose had a larger effect than a 5-mg dose (P <.0001). No discernable difference between doses was seen for nontarget and new lesions. The model predicts that after 1 year of continuous dosing, the change in SLD of target lesions in the average patient would be 142.1% ± 98.3% on placebo, 22.4% ± 17.2% for a 5-mg dose, and −15.7% ± 11.5% for a 10-mg dose. Conclusions: We developed a dynamic tumor model linking everolimus dosing history with overall tumor time course for each patient from RECORD-1. These tumor growth biomarkers are closer to the primary clinical end point (PFS) than measures of mTOR pathway inhibition, and thus may provide better predictions of trial success. Our analysis demonstrates a significant drug effect on target, nontarget, and new lesions. Furthermore, an everolimus daily dose of 10 mg is more efficacious than 5 mg in reducing growth of target lesions in metastatic RCC.

3056 Background: BEZ235 is a novel pan-class I PI3K/mTOR inhibitor with proven apoptotic effect (Brachman et al 2009), antiangiogenic effect (Schnell et al 2009) and clinical activity especially in patients with PI3K pathway dysregulated tumors (Burris et al, 2010; Rodon et al, submitted for publication). In this study, we attempt to quantify the biologic treatment effects of BEZ235 in order to investigate relationships between exposure to drug and markers of: suppressed proliferation and increased apoptosis in tumor cells; PI3K pathway modulation in surrogate tissue (skin); suppression of angiogenesis pathway components. Methods: Exposure to drug was defined to be area under the concentration curve between dosing times (AUC0-24h) at the day of PK/PD assessment (either day +8 or day +28). PD evaluated markers are listed in the results section. Each PD marker was normalized to represent change relative to its baseline value in the same patient. Each relationship between log exposure and relative change in PD was tested for correlation. Specifically, a least-squares linear regression was performed and a relationship reported if the 95% confidence interval (CI) for the slope (relative PD change per 10-fold AUC increase) excluded zero. Results: For BEZ235, a statistically significant relationship between exposure and change in PD marker was found for FDG-PET uptake, Cpeptide, p-S6 in skin and sVEGFR2; but not for sum of longest diameters (SLD) of target lesions, M30, M65, sVEGFR1, and VEGF. Conclusions: These preliminary exploratory analysis results are consistent with the proposed mechanism of action of BEZ235. The results will be presented for an updated data set. Given the dual mechanism of BEZ235, we intend to compare its PK/PD relationship with that of a TORC1 inhibitor as well as an anti-angiogenic compound. PD marker and day Slope (95% CI) SLD d28 0.00544 (-0.00112, 0.0135) Total SUVmax d28 -0.0318 (-0.0641, -0.00418)* Fasting Cpeptide d8 0.109 (0.027, 0.191)* p-S6 skin d28 -0.0619 (-0.0975, -0.0275)* M30 d8 -0.0225 (-0.0571, 0.0114) M65 d8 -0.0291 (-0.0654, 0.00468) sVEGFR1 d28 -0.0425 (-0.0944, 0.0085) sVEGFR2 day 28 -0.0414 (-0.0574 -0.024)* VEGF day 28 -0.0144 (-0.0693, 0.0434)