I'm a bit confused about an aspect of the pilot study. Maybe someone on this board can clear up the matter for me. I emailed Biomoda and am waiting for a reply. Email pasted below:
Julie,
Thank you again for getting back to me and for answering my question so thoroughly. I do have one more question, though I am not sure that you are the best person to ask. Perhaps if you do not know the answer to my question you could forward it to someone who might.
It is my understanding that the purpose of the CyPath assay is to identify cancerous cells. In the pilot study, two cohorts were used: the cohort with confirmed lung cancer and those at high risk. The purpose of the study was to see how well the CyPath essay could place individuals into their respective cohorts. While the results were encouraging, they weren't overwhelming. I am curious regarding the extent to which the results may be skewed by the CyPath assay's ability to detect cancer cells. Is it not possible that the assay results placed individuals in the cancer cohort who were actually in the high risk cohort precisely because the assay was working as intended, i.e. identifying cancer where other tests were unable to lead to a proper diagnosis? If this were the case, the number of "false positives" would be artificially high, biasing the specificity towards zero.
I am also wondering how the cancerous patients were identified. Were these individuals in an advanced stage of their cancer so that their affliction could be confirmed with 100% certainty, or were they identified using one of the highly unreliable methods which Biomoda is hoping to replace? If these individuals were incorrectly identified as having lung cancer, it could be that the (presumably) superior CyPath is correctly identifying individuals from the cancer cohort as being cancer free, biasing the sensitivity of the assay towards zero.
In all likelihood, those who designed the study took these possibilities into account. But I was struck by the difference between the outcome of the pilot study and that of the prior study which showed that CyPath was able to correctly determine whether or not 27 individuals had cancer with 100% specificity and sensitivity. My above concerns are one possible explanation for these discrepancies. If it doesn't conflict with any proprietary issues, I would greatly appreciate it if you, or better yet someone I involved with the study, could address these concerns.
Thanks again. I eagerly await your reply.
AI
