After having used the test for research it would appear as if it is very specific. What is one to make of bands that are there but faint? Or what if there are two bands and one is big and bold and the other is faint? Intuitively, I would believe that the starting concentration of the type represented by the bold band is way higher than the type represented by the faint band. Is my intuition correct? Any insights into what that means? Are both HPV types driving the disease process (eg papillomas) or is it just the bold one? Any data? Any intuitions? Any references?
I received this answer privately and was given permission (in fact asked) to share it without attribution.
The package insert for the Linear Array assay defines any visible band as a positive band, so a faint band counts as a positive. The difference in band intensity could be related to viral type abundance but really depends on which bands are being compared. Because this is an end-point assay, it's possible that some virus types amplify with better efficiency than others when found in a mixed infection and it is also likely that overall sensitivity or detection limits may be different for each type. Additionally the probe for HPV52 is cross-reactive with types 33, 35, 58 and in the presence of any of those additional types, it may increase its intensity.
The driver of the disease process is the virus with the most oncogenic potential (but I assume that is only when we are talking about cancer or dysplasia, perhaps not when referring to recurrent respiratory papillomatosis or genital warts), which does not change in the presence or absence of other viruses (and regardless of the band intensity). Additionally, it's important to remember that the resulting disease is clonally derived, originating with a single-type infection (ie. each lesion will be linked to only one HPV type). When multiple viral types are found within an exfoliated sample it signifies multiple different infection sites (and not multiple virus types within single cells). Not all of these will progress to high-grade disease, but rather, the lesions caused by genotypes with the highest associated risk are the most likely.