RCONIS has its first birthday!

We cannot believe it! It has already been one year since we launched RCONIS: In July 2024, at the useR! conference in Salzburg, Austria, we had our first photoshooting with the whole team (you can see one of the photos here) and presented our talks for the first time with our RCONIS brand.

Today, we want to celebrate this first birthday with you and show you what we have been busy with in this first year. We are very proud of what we have achieved so far, and we can assure you that much more is to come in the future.

Projects of the first year

Without disclosing any confidential information, we still would like to give you a flavor of the projects we have been working on in the past year.

Building up an internal statistics team

We worked with a clinical research organization to build up their internal statistics team. This included reviewing their business processes, templates and operating procedures, and updating them to the best practices in the industry. We also provided statistical support for trial designs in several projects. Finally, it was rewarding to be part of the recruitment process for the first internal statistician.

Supporting a submission strategy

We supported a biotech company in their submission strategy for a new drug. The trial was already finished, and the company needed to make the right case to the regulatory authorities. We provided statistical support for the submission documents.

Analyzing infectious disease data

We worked with an academic institution to analyze a rich dataset on a specific infectious disease. This involved data cleaning, simple univariate regression modeling, and also more advanced joint modeling of longitudinal and time-to-event data. We provided a detailed report with the results, supporting a publication in a peer-reviewed journal.

Implementing statistical tables and figures in R

We were part of a pharma company’s team to implement statistical tables and figures in R. Based on the internal layouts and expectations, and using open source packages such as tern, mmrm and rbmi, we developed the functions and scripts to create the statistical outputs in R. This contributed to the development of a new open source R package of which you might hear in the near future.

Planning whole clinical development

We have been planning the whole clinical development for a new drug within a biotech company. This includes everything from vendor selection, to trial design, protocol writing, and statistical analysis plans.

Improving open source software

We have been working with several pharma companies to improve open source R packages which they depend on. Some examples are:

rbmi: Added MCMC options for Bayesian multiple imputation, added Monte Carlo standard errors, currently adding additional covariance structures.
RobinCar2: Developed design and prototype for the covariate adjusted survival analysis functionality, currently working on the implementation.
glmmTMB: Fixed a bug in the autoregressive heterogeneous covariance structure such that results now match SAS, currently working on robust sandwich estimator.
mmrm: Fixed a bug leading to very long computation times in the calculation of the robust sandwich estimator, as well as a correlation range problem in the compound symmetry covariance structure.
rpact: We have evolved the package steadily toward more flexible trial designs and smoother user workflows, as part of the service level agreement crowd-funding model. Here are some highlights:
1. Starting with release 4.1.0, negative-binomial count simulations, one-sided test direction control in group-sequential and inverse-normal designs, and flexible result extraction were introduced.
2. Release 4.2.0 then added dose-response support for multi-arm simulations, unequal-variance methods for two-sample mean tests, and a comprehensive installation-qualification workflow. That release also made it easy to save or reset all rpact options and enabled futility boundaries and beta-spending curves in diagnostic plots via simple flags.
3. In the upcoming release 4.2.1, you’ll be able to supply custom efficacy and futility stops, choose pooled or unpooled standard-error estimates for two-sample rate confidence intervals, and get a richer installation qualification report.

Most of these contributions involve a deep dive into the statistical methodology first, before then crafting a solution path and implementing it in the R package. These are fantastic opportunities to help the whole user community with improved software.

Trainings for joint models and MMRMs

We have been working with a pharma company to provide training on joint tumor growth / survival models. More to come on this in the near future, as we are planning to offer this training in other formats as well.

We are also currently working with another organization to provide training on mixed models for repeated measures (MMRM). This will be a one-day face-to-face training at the company’s site, and we are looking forward to it.

Decision making framework for early phase

We are supporting a pharma company in migrating their decision making framework for early phase clinical trials from a Java based graphical user interface with legacy R code to modern R implementations with a Shiny front end. This will allow the company to be compliant regarding IT security and to have a more flexible and maintainable solution in the future.

Pro bono open source contributions

In addition to the projects above, we have been very active in the open source community and giving back:

phase1b: We have been working towards a first complete refactoring of the phase1b R package, mainly via code reviews.
R/Pharma conference: Daniel is again part of the APAC organizing committee of this year’s R/Pharma conference, which will take place in November 2025. Please see rinpharma.com for more information.
openstatsware: Daniel is a co-chair of the openstatsware working group, and is active in the group meetings and projects.
Good software engineering practices for R packages workshop: We have presented the openstatsware workshop in Tokyo in April (see here) and we are currently preparing three additional workshops in the US and Europe for August and October this year.

RPACT Cloud: professional trial design and simulation GUI

RPACT Cloud is a graphical Shiny-based front end for the open-source rpact R package, providing an intuitive GUI for trial design and simulation. A free demo version is available at cloud.rpact.com.

Over the past year, three major pharmaceutical companies have licensed RPACT Cloud for internal installation and use on their own servers. To support these enterprise deployments, we published extensive documentation covering setup on Posit Connect, configuration, and full qualification/validation:

How to Programmatically Clone and Deploy the RPACT Cloud Shiny App to Posit Connect
Installation: Running RPACT Cloud as an R Package
Guide to Environment Variables for RPACT Cloud
RPACT Cloud User Requirements Specification (URS)
RPACT Cloud Test Report: Maps directly to the URS and documents the results of numerous shinytest2 unit tests, providing full transparency on validation outcomes.

Based on the customer feedback, we have continuously improved the RPACT Cloud application, adding new features and enhancing the user experience. This included for example live installation qualification & QA features (including URS and downloadable HTML/PDF reports), session timeout handling, advanced table-rendering controls (rounding, precision toggles), plus UX improvements like legend-position dropdowns and upgraded markdown rendering.

Feedback from our pharma partners and their teams has been overwhelmingly positive, underscoring the robustness, ease of use, and high user acceptance of RPACT Cloud.

What is next?

We are very excited about the future of RCONIS and have many plans for the next year. Here are some of the things we are currently working on:

Gernot has been very busy with the second edition of his book on group sequential and adaptive designs, which is currently in the final stages of editing. The book will be published this year, so stay tuned for more information!
We are launching a new service level agreement crowd-funding for crmPack, the flexible dose escalation R package. 4 companies have already committed, and we are looking for more companies to join. This will allow us to continue the development of crmPack and provide professional support for the users.
We are working on a legal company structure to be able to build the company further. This will allow us to be long-term sustainable and hopefully eventually to grow the responsibilities.

Interested in working with us?

As you can see, we have been active in a wide range of projects in the past year. If any of these ring a bell for you, or if you are interested in any of the topics we have been working on, we would love to hear from you. We are always looking for new projects and opportunities to collaborate. Just drop us an email at and we can have a first informal chat.