Skip to main content

Clinical Trials

Sangamo’s Gene Therapy Results

Let the arguing commence! That’s my response to the announcement this morning by Sangamo, a company that’s been trying to do gene therapy in the clinic. They’re not a CRISPR outfit, though – they’re using an alternate technique for gene silencing, using zinc finger regions instead. Even at that level, there’s more than one way to go about this. Sangamo is using zinc finger nucleases, while others are trying to get the (broadly similar) TALEN technique to work as a human therapy. Here’s a review from a few years ago comparing these three approaches.

ZFNs and TALENs both rely on taking an active nuclease enzyme domain, ready to break DNA pretty generally, and engineering it to be much more specific through use of particular Cys2-His2 zinc finger regions or TALE DNA-binding domains, respectively. There’s been engineering of the nuclease along the way, too, to improve the chances of specificity being achieved, but the whole effort is a rather long and complicated story. You then allow homologous recombination to fix the break, but in a way that restores the correct sequence, giving you a newly functional gene and protein. Suffice it to say that Sangamo really is the flag bearer for the ZFN method, and they’ve been pursuing this for many years now.

Their trial was aimed at patients with Hunter syndrome (mucopolysaccharoidosis Type II, MPS II). That’s caused by a lack of a specific lysosomal enzyme, IDS (iduronate 2-sulfatase), which causes the glycosaminoglycans (GAGs) heparan sulfate and dermatan sulfate to accumulate in various tissues. That’s really not good, as you’d figure – there are a wide variety of symptoms, depending on just how severely the enzyme’s activity has been impaired. The worst case is early death, but even the best case is pretty bad – longer life, but with a number of problems in different organ systems. There’s a whole range of lysosomal storage diseases in this area that can cause GAG accumulation, and none of them are any kind of good news. As it stands, the only thing for such patients is direct enzyme replacement therapy, which in the case of MPS II definitely helps – but the enzyme, as you’d imagine, does a poor job of crossing the blood-brain barrier and doesn’t really do anything about central nervous system damage.

Sangamo’s site of action would be the liver, which is where many attempts at gene therapy are ending up as a matter of sheer phamacokinetics. But in mouse models of the disease, they’d shown that they could not only get IDS production restored, but that the enzyme produced also seemed to help with CNS effects. The hypothesis is that sustained high levels of enzyme, in a way that’s not feasible with the externally administered variety, allow some of it to actually cross the blood-brain barrier.

There are only two patients that we’re talking about here. But both of them showed significant declines in the heparan sulfate and dermatan sulfate in their urine, which is good news. However, and this is where the arguing really revs up, they could not find any IDS enzyme in the blood of these patients. You’d have kinda thought that you’d see it there (they sure saw it in the animal studies), and a lot of observers/investors sort of lost it at that news, sending SGMO down pretty briskly in the early trading today. The company responded by saying that there had to be new active enzyme in there for the urine numbers to go the way that they did, that their detection method isn’t incredibly sensitive to start with, that there’s another cohort in the trial that’s getting a larger dose of the gene therapy, etc.

Of course, they also said that urinary levels of GAG were an approvable biomarker and that’s how the enzyme replacement therapy was approved, and that appears not to be true. The enzyme (Elaprase, from Shire) showed effects in the amount the treated patients could walk and the trials looked at lung function as well. In the end, you’re going to want to see functional improvements in these patients anyway, but everyone’s looking for an earlier read. The selloff in Sangamo’s shares continues as of this writing, and other gene-editing stocks are dropping as well (and since these are all companies using different gene-editing technologies, that seems a bit like panic if you ask me). But this can’t have been the way that Sangamo wanted this to go. When that next cohort reports, maybe we’ll have a better idea of what’s going on – either that, or the shouting will get even louder. . .

9 comments on “Sangamo’s Gene Therapy Results”

  1. subramanian karur says:

    did they show whether the editing took place?

    1. Derek Lowe says:

      Figure you’d need a liver biopsy for that, plus quite a bit of single-cell sequencing, since it’s unclear (to me, and maybe to them) how many cells were actually affected. . .

  2. Toby says:

    I read that they failed to detect IDS in *trough* samples, so pre-dose, with weekly dosing. I.e. the timepoint where you might expect levels to be lowest. Any insight into why they didn’t look for it at a more PK-optimal timepoint?

    1. Toby says:

      Duh, it was at the trough of the ongoing weekly ERT dosing, not the trough of the gene therapy. Makes perfect sense.

  3. Skeptic says:

    “They’re using an alternate technique for gene silencing”
    Nope: “gene editing”. Or really, since their technique doesn’t control all of the editing process, “targeted DNA cutting”.

  4. Scott says:

    “The selloff in Sangamo’s shares continues as of this writing, and other gene-editing stocks are dropping as well (and since these are all companies using different gene-editing technologies, that seems a bit like panic if you ask me).”

    Please repeat after the Business Major: “The stock market is not a rational actor.”

  5. Barry says:

    Seems like whether one uses Zn-finger nucleases, or TALENs or CRISPR, all these cut techniques converge on homologous recombination to make the splice. And homologous recombination is not error-free.
    Liu’s base-editing seems to be the first gene editing technique that doesn’t rely on this weak link.

  6. johnnyboy says:

    Sangamo have been at this approach for close to 20 years now. They have proven themselves masters of the press release hyping highly promising early results, but not so much at making actual working drugs.

  7. Claus says:

    Maybe liver is not optimal organ for producing a lysosomal enzyme. When a lysosomal enzyme is injected iv most of it will end up in liver due to glycan receptors, so maybe the produced enzyme never gets out of the liver.

Comments are closed.