About Cancer Insight
What is Cancer Insight?
An independent, free, educational explorer of cancer research. It shows recent papers, the size and make-up of a cancer’s PubMed literature, treatment literature, registered clinical trials and research coverage, with every count linked to the search that produced it.
Can Cancer Insight tell me which treatment is best?
No. It describes published research; it does not evaluate treatments, diagnose or give medical advice. Discuss personal medical decisions with a qualified clinician.
Who is it for?
Students, researchers, journalists and anyone who wants to understand what research exists on a cancer and where the evidence comes from.
Is the site available in Arabic?
Yes. The whole site is available in Arabic, laid out right to left. You can search with Arabic cancer names; the same English queries are sent to the sources, so the numbers are identical in both languages. Paper titles and abstracts stay in their original language, and exported files stay in English.
Sources and numbers
Where does the research come from?
From NCBI PubMed and PubMed Central (U.S. National Library of Medicine), ClinicalTrials.gov for trials, a third-party research feed (the Cure Cancer With AI research API) for recent records and treatment tags, and Wikimedia Commons for images. Burden figures come from GLOBOCAN 2024 (International Agency for Research on Cancer).
Why does every number link to PubMed?
So you can check it. Each count is a PubMed search; the link runs the identical query on pubmed.ncbi.nlm.nih.gov.
Why is a number slightly different on PubMed?
Cancer Insight caches counts for up to 24 hours and PubMed adds and re-indexes records daily, so small differences are expected.
Why are the papers in “Recent research” all so new?
Because that list is the newest relevant records — a view of what is new, not a sample of the whole literature. Switch the Papers view to PubMed Best Match to see the most relevant records across all years.
Which cancers have treatment tags?
The research feed covers lung, breast, brain, colorectal, prostate, pancreatic, liver, kidney, melanoma, ovarian, cervical, bladder, gastric and esophageal cancers, leukemia and lymphoma. Other cancers are searched directly in PubMed, without treatment tags.
Can I use the data?
Yes. Every analysis is available as JSON or CSV from stable addresses listed on the open data page, together with the searches that produced it. Cite Cancer Insight and the original sources when you reuse it.
Papers and evidence types
What do the evidence types mean?
They come from PubMed publication types: for example randomized controlled trial, observational study, systematic review or case report. They describe what kind of record a paper is, not how good it is.
What does “Design not indexed” mean?
The paper is indexed only as a generic “Journal Article”. Many original studies are indexed this way, so it is not a judgement about the paper.
Why do some papers carry a retraction warning?
PubMed marks them as retracted publications. Read the retraction notice before relying on anything in them. Each cancer profile also counts the cancer’s retracted records.
Can I reuse a paper marked “Free full text”?
Not necessarily. Free to read is not the same as free to reuse; check the licence on the publisher or PMC page.
Treatment literature
If one treatment has more papers or trials, is it better?
No. More records, more trials or more recent records reflect research attention, history and funding — not effectiveness, safety or suitability for anyone.
Why are the example papers on each side of a comparison different?
Each side is a separate PubMed Best Match search. The papers are examples of each literature and are not paired.
Clinical trials
Where do the trial figures come from?
From ClinicalTrials.gov. For each cancer one condition search is sent, every matching study is downloaded, and every figure is counted from that single list — which you can download as CSV to check any number.
Does “completed without results posted” mean results were hidden?
No. These are interventional trials that finished at least 24 months ago and have no results section on ClinicalTrials.gov. Many trials publish in journals instead, and many — such as phase 1 trials or trials outside U.S. law — are not required to post results there.
Can Cancer Insight help me find a trial to join?
It lists open trials with links to their registry pages, but it does not check eligibility or recommend any trial. Discuss trials with your care team; each registry page lists its eligibility criteria and contacts.
Research and burden
What does the research-to-burden ratio mean?
It is a cancer’s share of PubMed research from the last five complete years divided by its share of deaths (or new cases) worldwide in GLOBOCAN 2024. Above 1, its share of research is larger than its share of deaths; below 1, smaller.
Does a low ratio mean a cancer deserves more research?
Not by itself. Burden is one reason among many to study a cancer: survival, curability, how many people live with the disease and scientific opportunity matter too. The ratio describes where attention goes, not how much research each cancer should get.
Why are some cancers missing from the comparison?
GLOBOCAN does not report them as a separate site (sarcoma, neuroblastoma), they are a subtype of a cancer that is listed, or GLOBOCAN defines them differently (skin cancer).
Research Gap Atlas and cancer map
Does a signal prove a research gap?
No. A signal means a design or theme makes up less than half the share it has across oncology. That can reflect biology, epidemiology, terminology or indexing, so it is a prompt for investigation, not proof of a gap.
Why isn’t “children and adolescents” flagged for some cancers?
Cancers that occur mainly in adults are expected to have few pediatric studies, so that measure is shown but not treated as a signal.
Why are treatments never flagged?
Whether a treatment is relevant depends on the biology of each cancer, so a low count cannot be read as a gap.
What do the colours on the cancer map mean?
Each cell is the same ratio as in the atlas. Brown means lower representation than across oncology, blue higher, and the pale neutral colour similar. Every cell shows its value, and only ratios below 0.5 are signals.
What changed from the earlier version?
Earlier versions measured coverage inside a small, recent sample of papers, which could, for example, report zero clinical trials for a cancer with thousands. Method 3.0 compares shares across the whole PubMed index with oncology overall, and keeps a sample only to make the queries inspectable.
Saving and exporting
Where are saved papers stored?
Only in your browser. Clearing browser data, private browsing or another device will not show them, so export anything you want to keep.
What can I export?
CSV files with a provenance header, RIS citations for Zotero, EndNote or Mendeley, PDF reports, the full Gap Atlas evidence as JSON, and trial lists, the burden table and the cancer map as CSV.
Privacy
Is what I search private?
Search terms are sent from Cancer Insight’s server to the research sources listed above and are recorded in Google Analytics events unless your browser sends Global Privacy Control or Do Not Track. Search for cancer types, never personal health details.
Still have a question? Contact Cancer Insight.