The trial followed 25,000 adults from the US and Canada over a year.
Nearly one in a 100 of those tested had a positive result and in 62% of these cancer was later confirmed.
The test correctly ruled out cancer in over 99% of those who tested negative.
When combined with breast, bowel and cervical screening it increased the number of cancers detected overall seven-fold.
Crucially, three quarters of cancers detected were for those which have no screening programme such as ovarian, liver, stomach bladder and pancreas.
The blood test correctly identified the origin of the cancer in 9 out of 10 cases.
These impressive results suggest the blood test could eventually have a major role to play in diagnosing cancer earlier.
Scientists not involved in the research say more evidence is needed to show whether the blood test reduces deaths from cancer.
The topline results are to be released at the European Society for Medical Oncology, external congress in Berlin, but the full details have yet to be published in a peer reviewed journal.
Much will depend on the results of a three-year trial involving 140,000 NHS patients in England, which will be published next year.
The NHS has previously said that if the results are successful, it would extend the tests to a further one million people.