Biotech Values

[poorgradstudent]

SRPT:

On twitter you made reference to the Western Blot still being fairly incompetent. Could you provide more detail on that?

Sure. May get a little esoteric / in-the-weeds, but I suspect readers here are definitely up to it.

The Western blot is a way to use a specific / highly selective antibody to detect your protein of interest amongst a mixture of proteins. Western blots allow you to not only detect the presence of a protein, but also allows you to approximate its molecular weight. In essence, you can tailor conditions to not only detect dystrophin, but also to detect full length dystrophin and smaller / truncated forms. In this case, they are taking patient biopsies over time and trying to determine the presence, and possibly changing content, of their exon-skipped dystrophin protein. So in a hypothetical mixture where you have normal and an exon-skipped / truncated dystrophin, you would putatively see 2 bands relatively close together. Sadly, in many of these kids, they're practically full length dystrophin knockouts so you see no full length dystrophin and would technically see the exon-skipped dystrophin only.

The issue with Western blots is that they depend on a variety of parameters that need to be optimized. If you use a very high abundance of the antibody, it can start detecting other non-specific proteins simply due to the poor conditions, similar to giving someone too high a drug dose and getting off-target effects. Under those cases, your putative detection of the "specific" dystrophin band is compromised as multiple other bands begin to appear. You can see that in SRPT's slide set from today where there are multiple bands at smaller molecular weights other than the "dystrophin" band that they point to as evidence of drug action. The Western blot below (initially linked by andybiotech) is one where another independent group simultaneously detect dystrophin, dysferlin, calpain and sarcoglycan (SG) proteins:



Notice how the dystrophin bands detected with the Dys1 antibody are nice and distinct. The blot is clean and gives confidence that what they are detecting is what we expect. Compare that to the dataset from SRPT using Dys1 and you can see just generally poor technique that, in return, makes me question what it is that they're detecting and the validity of the detection.

Also pointed out on Twitter by Garrett Rhyasen is the long exposure time. For quality Western blots with abundant target protein, the duration should be in the seconds. For low abundance proteins, you're usually in the 5-15 minute range. But even then, conditions should be tailored so that you get specific detection. In SRPT's case, they're doing 30 minute exposures and even then, their dystrophin band of interest isn't the strongest band on their blots. Rather, bands at ~200 kDA and ~40 kDa are strongest. Therefore, if your detection tool for dystrophin is better at detecting other nonspecific bands, what can we make of this piece of evidence? The 30 minute exposure to see dystrophin also suggests that the abundance of the target protein is really really low. That point isn't an indictment per se, but helpful to consider when interpreting the MOA.

An additional point is that these Western data impact the IHC data. The Dys1 antibody is indicated as suitable for IHC by the manufacturer, and it is my assumption that it was also used in these IHC experiments. In this case, IHC is done because it is useful for localization of the protein along the fibers. However, the weakness of IHC is that you don't know which protein is giving you the signal. In other words, if you look at the Dys1 Western blot data, there are multiple non-dystrophin proteins that are giving you signals at various molecular weights. When this is carried over to IHC, the "red" colour comes from the aggregate positioning and abundance of all those individual signals you see on the Western. Therefore, how do we know that the increase in the dystrophin signal in IHC is from dystrophin itself or changes in the abundance of these non-dystrophin proteins that are shown as being nonspecifically detected in the WB? We can't.

If the appraisal of the NDA is going to be based on the 6 MWT results and the FDA thinks those data are strong and stand alone, that's one scenario. But if the FDA is cautious about the 6 MWT and wants molecular data to clinch the decision, then SRPT is doing patients and investors a disservice with this package. Those Western blot data undermine the claim that any dystrophin is produced, even if you give them the benefit of the doubt over the sloppiness of the results.