Clinical Data Software · P-300

The Data Browser
Built for Clinical Trials

Cumba Browser is a high-performance desktop application for viewing, querying, and analyzing tabular clinical and pharmaceutical data - across every format your workflow demands. It runs on your machine; your data never leaves the building.

SAS XPT Dataset-JSON Parquet Excel SAS7BDAT
The Cumba owl - P-300
Cumba Browser - main view
50M+ rows loaded
& Extensive Sorting & Filtering
CDISC standards aware
AI Agent AI always at hand

Built for the one place trust matters most - your data.

The AI controls the tool, not the data

Ask the agent to load, filter, transform, or chart - but every number comes from Cumba, never from the AI. No invented values, no hallucinated results.

Read, never changed

It's a browser - your data is never changed. Edits and conversions always export to a new file; the original stays exactly as it was.

Runs locally, on-premise

Cumba runs on your machine. Your data never leaves the building - no cloud requirement, no data transfer. Built for regulated, privacy-sensitive environments.

Everything a data reviewer needs

Built from the ground up for clinical and pharmaceutical data workflows.

All Clinical Data Formats

Open every format clinical and pharmaceutical workflows depend on - from one application.

SAS7BDAT · XPT · Dataset-JSON · Parquet · RData · Excel · CSV

Lightning Fast & Memory-Efficient

Designed for laptops, scaled for production data.

  • Most 5 GB SAS files fit in less than 500 MB of RAM
  • Sort a 5-million-record file in less than a second

Extensive Filters & Sorting

Slice your data exactly how you need to - and keep the slice.

  • Memorize and save filters
  • Build filters by picking from existing values
  • Filter by CORE or Pinnacle 21 validation findings

Format Overlays

Display SAS-style formats as a value overlay - load them from define.xml, sas7bcat, or a format table.

Smart Number Display

Imprecise floating-point values and SAS missing periods are highlighted on sight.

Find, Reorder & Hide Columns

Effortless column management - even for tables with hundreds of columns.

Full SQL Engine

Run any SQL query against your data - even joining across mixed file formats.

Find Text in Table

Search the entire dataset for a value or pattern, instantly.

Merge Tables

Combine related datasets with first-class CDISC awareness.

  • Merge by key variables
  • Merge SUPP-- domains into the base dataset

Transpose

Pivot a BDS dataset into one column per TESTCD or PARAMCD.

Statistics

Frequency and descriptive statistics - at a finger tip.

Extended Charts Library

17 chart types, from classic bar charts to waterfall plots.

Data Editor

Modify a dataset and export the changes to a new file - the original stays untouched.

Compare Tables

Easy and fast table comparison - value-by-value, key-aware.

Data Converter & Export

Convert single datasets or entire libraries - and export filtered subsets too.

SAS7BDAT · XPT · Dataset-JSON · Parquet · RData · Excel · CSV

CDISC Define-XML Support

First-class CDISC Define-XML - codelists overlay your data and metadata travels with the dataset.

Define-XML 2.0 / 2.1

CDISC CORE & Pinnacle 21 Integration

Read validation reports from both engines and review issues side-by-side with the data they reference.

AI Agent Integrated

A built-in conversational agent with full remote control of the application - ask it to load, filter, transform, or chart.

Remote Backend

Browse data living on remote servers via SSH or HTTP(S) - data and transformations stay where they are while the experience feels local.

OpenStudyBuilder Integration

See every dataset in your OpenStudyBuilder instance, right inside the viewer.

CDISC Library API

Browse the CDISC Library API and use it as the default source for variable attributes.

One viewer, every format

Dataset-JSON
Supports all Dataset-JSON standards (json, ndjson & dsjc)
SAS XPT
Transport format for CDISC submissions
SAS7BDAT
Native SAS dataset format
Parquet
Columnar storage for big data
Excel
XLSX workbooks and sheets
CSV
The good old Comma Separated Value files
RDS / RData
R Data Serialization & R Data

Runs on your machine

A desktop application installed locally, with no cloud dependency. Download the latest release for your platform.

Version 0.5.3.0

8 GB RAM minimum
16 GB RAM recommended

For two decades we built clinical data tools inside pharma. Now we build them for everyone.

The tools people loved to use - because we built them at the front line and used them ourselves. From user, for user.

The Key Features Every Statistical Computing Environment Should Have - and Often Lacks

PhUSE Connect 2025, Hamburg

From Validation Reports to Contextualized Submission Review

PhUSE SDE Utrecht · 10 June 2026