It's a good question Charles89. I am afraid the answer is too technical for me, but see what I've found below.
When reviewing the template for FDA approval:
"
FDA recommends that the following validation studies be conducted for a SARS-CoV-2 antigen assay: Limit of Detection/Analytical Sensitivity, Cross-reactivity/Analytical Specificity, Microbial Interference, and a Clinical Agreement Study."
I understand from this that Operon carried out validation study for limit of detection and DI was engaged to carry out the remaining studies. Past announcements and looking at IFUs of other tests seem to suggest that.
So in this case, when we look back at the announcements from Oct 2020 regarding Operon's study, the company said that "
the determination of this (LoD) setting will require further development work and validation testing." Then, in Nov 2020 it announced design freeze with the following: "
High Sensitivity – 0.02 ng/ml antigen concentration and positive sample detection in low viral load samples (Ct value of 30 and above)."
If you look at the LoD data of companies on the market now, it varies a lot and some look quite impressive too. But how some can be so sensitive on LoD, yet return bad sensitivity (PPA) results - see Abbott's Binax data?
When you look at the description of each study to see how the LoD is derived, it is understandable that the data will not be consistent. And to that effect, FDA tried to make a comparison of the LoD. Unfortunately, it is not for antigen tests yet. But it gives you a good idea of how much it varies and what is acceptable to FDA. The website also explains a bit about the problem with getting uniform LoD data.
https://www.fda.gov/medical-devices/coronavirus-covid-19-and-medical-devices/sars-cov-2-reference-panel-comparative-dataInterestingly, while compiling the EU antigen tests table (update below), I saw that there have been 2 tests with full clinical trials that return which lower results than reported by the supplier:
For Abbott's Panbio test, the clinical trial returned the following data on LoD:
For SD Biosensor/Roche:
I have no idea how to begin to understand these columns, but what I can gather is that supplier-reported LoD is lower than the actual verified LoD.
Abbott and Roche are the biggest players in EU in terms of antigen tests and the clinical trial data returned embarrassing results.
So, yes, good news that Operon study shows high sensitivity for LoD which we can include in our product description and IFU. And our clinical study data with DI has also returned good sensitivity and specificity data. This makes us competitive on the market, but the clinical trials will matter and somehow I am confident we will do fine. IMO only.
