Reader Comments

Post a new comment on this article

Clarification needed: size of tumor and sensitivity

Posted by WillStott on 12 Oct 2010 at 09:39 GMT

We estimated that the tumors we would need to detect to achieve even 50% sensitivity are more than 200 times smaller than the clinically apparent serous cancers typically used to evaluate performance of candidate biomarkers

In the previous paragraph it says that 50% sensitivity requires detection of 1.3cm tumors, but here it says the size of tumors we would need to detect for 50% sensitivity are 200 times smaller than is typically used [today] to evaluate biomarkers. This seems to suggest that the size of tumors used to evaluate biomarkers is 200 x 1.3cm = 260cm!

No competing interests declared.

RE: Clarification needed: size of tumor and sensitivity

pbrown replied to WillStott on 14 Oct 2010 at 15:36 GMT

Thanks for the question. The factor of 200 refers to the volume (thus presumably mass) and not the linear dimensions of the tumor. If we assume that the per-cell rate of net production of any potential biomarker (due to secretion or cell lysis) is independent of size, then the rate of biomarker production is expected to be roughly proportional to the mass of the tumor. A 200-fold difference in diameter corresponds to a 5.85-fold difference in tumor mass (and thus presumably production of a putative biomarker), assuming equal scaling in all dimensions. Sorry that this wasn't stated more clearly in the article.

No competing interests declared.