If one follows the links in the paper, one can get to it. Springer has made it freely available, in full, for online reading, although it can't be downloaded.
It is here: Integrating genomic features for non-invasive early lung cancer detection.
In order to read the full paper without a subscription or access, one does need to approach it using the newspaper's links and not at the journal's website (where my link leads.)
It is largely about a genetic screening test for lung cancer using genetic biomarkers found in blood.
It does contain the following text in the conclusion suggesting that it might be utilized for other cancers, although the number "50" is not mentioned:
In summary, we have developed an integrated genomic strategy that can detect a significant fraction of early-stage lung cancers using blood plasma. We propose that integration of Lung-CLiP with LDCT or with the analysis of other circulating biomarkers could further improve performance. Additionally, by modifying the at-risk populations considered and incorporating molecular features appropriate for other cancer types, we expect that it could be feasible to develop CLiP methods for a diverse range of malignancies.
This is, regrettably, another case of a journalist grotesquely distorting what a team of scientists is reporting.
= new reply since forum marked as read