Data sources and coverage
AGcollectio answers from our company database and from public datasets. The guide on the app's home screen shows a short version of the same list.
Data AGcollectio answers from
- Our company database
- Biotech companies (country, founding year, funding), pipeline assets (modality, indication, stage), partnerships (type, partner, timing, and deal values read from the announcements), companies developing drugs for each target, an index of company platform technologies, pipeline assets by disease along with disease primers (definition, patient numbers, standard of care and unmet needs, with sources), drugs recently approved for each disease, clinical trials linked to companies and assets, and company tags that group companies by technology or modality.
- Clinical trials from ClinicalTrials.gov
- Industry trials registered on ClinicalTrials.gov, the US trial registry: phase, country, therapeutic area, reasons for stopping, eligibility criteria, control arm type, primary endpoint category, and posted results (primary outcome values and arm comparisons, reasons participants did not complete, numbers and rates of serious adverse events, and more).
- EU Clinical Trials Register (EudraCT)
- Trials filed in Europe, mostly up to 2022: member state, commercial or non-commercial sponsor, investigational product, phase and therapeutic area. Matched against the US registry, it shows drugs developed in the US with no European filing, and the reverse.
- EU Clinical Trials Information System (CTIS)
- Trials filed in Europe from 2023 onwards: the review outcome in each member state, commercial or non-commercial sponsor, investigational products and their role (investigational, comparator, placebo, auxiliary) and therapeutic area. CTIS succeeds EudraCT, and we hold it as a separate dataset.
- Conference abstracts
- Trial result figures (response rates, median survival, hazard ratios, adverse event rates and so on) from abstracts presented at the American Society of Clinical Oncology (ASCO), the American Association for Cancer Research (AACR), the American Society of Hematology (ASH), the San Antonio Breast Cancer Symposium (SABCS), the Alzheimer’s Association International Conference (AAIC), the American Diabetes Association (ADA) and others. Figures have been extracted for some abstracts, not all. We do not hold abstract text, summaries or author names, or abstracts from the European Society for Medical Oncology (ESMO).
- Patient numbers and prevalence by disease
- Patient numbers and prevalence for each disease, from Japan’s Patient Survey (Ministry of Health, Labour and Welfare) and from Orphanet, the reference portal for rare diseases and orphan drugs. We do not hold patient numbers (head counts) for countries other than Japan.
- Japanese drug prices
- Official reimbursement prices from Japan’s National Health Insurance drug price list, and whether each product is an originator or a generic, including how far generics have entered each therapeutic class. We do not hold market size or sales forecasts.
- Drug and target master
- A lookup table that links all of the above by drug and target name.
What a discussion can produce
These are not part of a report plan, but you can make them in a discussion.
- Target product profile (TPP) drafts
What we do not hold
We do not hold the following. A request that includes them is not turned away as a whole: the plan is built from the parts we can answer, and a single line in the plan and the report says what was left unanswered.
- Market size and sales forecasts
- US designations that are not made public before approval (such as Breakthrough Therapy and Fast Track designations)
- Japanese designations (orphan drug and SAKIGAKE designations) and US accelerated approval, among others
- Post-marketing adverse event reports
- What was discussed with regulators
- Patent expiry and disputes
Record counts
The counts below were measured on . They move whenever the data is reloaded.
| Biotech companies | 5,609 |
|---|---|
| Pipeline assets Records that only say a company has no pipeline are not counted. | 20,419 |
| Clinical trials (ClinicalTrials.gov) Industry-involved interventional trials of drugs, biologics or gene therapies. The two European registries (EudraCT and CTIS) are held separately and not added to this number. | 111,524 |
| Partnerships since 2020 | 9,791 |
| Company platform technologies Classified against a vocabulary we built by hand. | 2,455 |
| Recorded reasons trials were halted | 14,295 |
| Patient-count and prevalence measurements | 17,446 |
| Conference abstracts (bibliographic records) A count of bibliographic records: title, abstract number, DOI and affiliations. We do not hold abstract text, and result figures have been extracted for some abstracts only. | 50,815 |
How things are counted
- Pipeline counts leave out records that only say a company has no pipeline.
- The two European registries (EudraCT and CTIS) are separate datasets, split by when trials were filed. They are not added together, and not added to the ClinicalTrials.gov count either.
- Partnership types and deal values are read by us from the partnership announcements.
- What each report counted, and how, is set out in the population table at its top and in each section's ⓘ notes (Reading a report).
Showing sources
Tables and charts built on ClinicalTrials.gov data show the source and the date of the data. Sources and terms of use for each dataset are collected on the Data sources page.