cancer rates after gene therapy
=biology =medicine
Here's a new
paper on cancer
developing in patients after receiving gene therapy with eli-cel, brand
name Skysona.
What is eli-cel?
Stem cells are isolated and modified by using a
lentiviral vector to add the ABCD1
gene. The stem cells in bone marrow
are killed with chemotherapy, and replaced with the modified stem cells.
How often did cancer develop?
Hematologic cancer developed in 7 of 67 patients after the receipt of eli-cel
Yep, that's about what I expected. How long did it take?
myelodysplastic syndrome (MDS) with unilineage dysplasia in 2 patients at 14 and 26 months; MDS with excess blasts in 3 patients at 28, 42, and 92 months; MDS in 1 patient at 36 months; and acute myeloid leukemia (AML) in 1 patient at 57 months.
Makes sense. What does genetic analysis of the cancer cells indicate?
In the 6 patients with available data, predominant clones contained lentiviral vector insertions at multiple loci, including at either MECOM–EVI1 (MDS and EVI1 complex protein EVI1 [ecotropic virus integration site 1], in 5 patients) or PRDM16 (positive regulatory domain zinc finger protein 16, in 1 patient). Several patients had cytopenias, and most had vector insertions in multiple genes within the same clone; 6 of the 7 patients also had somatic mutations (KRAS, NRAS, WT1, CDKN2A or CDKN2B, or RUNX1), and 1 of the 7 patients had monosomy 7.
Multiple
off-target insertions, different ones for different people. Yep.
Has
this happened with gene therapy attempts before?
Yes, it's been a big
problem with eg γ-retrovirus gene therapy trials. People hoped lentiviruses
would be better because they tend to insert around actively transcribed
genes, while γ-retroviruses tend to insert near transcription start sites.
Why don't we see high cancer rates after DNA viral infections?
We
sort of do: Many cancers are strongly associated with viruses. But eli-cel
is:
- specifically done to stem cells, which are closer to cancer than usual
- applied to cells away from the immune system, which can often detect
infections and kill infected cells early
--:-:-:-:-----------------------------------:-:-:-:--
How about
CRISPR instead of a viral vector?
Casgevy is a
treatment that uses CRISPR on isolated cells. It has serious side effects,
but not as bad as eli-cel. But a lot of people misunderstand what CRISPR
actually does.
CRISPR is a technique that uses the Cas9 enzyme to cut
double-stranded DNA according to added guide RNA. It doesn't insert DNA
itself, but if DNA segments are added that are compatible with the cut
section, they sometimes get inserted by homology-directed
repair.
So,
there are a few obvious issues here:
- Cas9 has to be delivered to cells.
- Sometimes Cas9 cuts the wrong
spot; it depends on the sequences.
- Sometimes the added DNA doesn't get
inserted.
- Double-stranded DNA break repair can have
problems.
Also, DNA
repair is different in different kinds of human cells, and CRISPR doesn't
seem to work as well in non-stem cells.
Currently, physical methods
(microinjection technology, electroporation, and HTVI) are commonly used for
delivering CRISPR/Cas9. But those aren't very practical in live animals, and
also have large effects on cells.
--:-:-:-:-----------------------------------:-:-:-:--
How about mRNA
delivery of CRISPR, then?
Yes, delivering mRNA that codes for Cas9
can work. It might be a better approach for medicine.
That will still
probably be limited to isolated cells in general. In live animals, you run
into a lot of problems with immune system reactions to Cas9; when that's the
desired effect you have mRNA vaccines, which are rather well-known now.
After treatment, any proteins that weren't already being produced can
also cause an immune reaction. Eli-cel hasn't had as many immune system
rejections as transplants, but it's still been a problem. It doesn't work as
well in people with the ABCD1 gene fully deleted instead of nonfunctional,
probably because the fixed protein is similar enough to the
(already-produced) flawed version to not have an immune response.
--:-:-:-:-----------------------------------:-:-:-:--
If Cas9 doesn't
insert genes, how about bridge RNAs with IS110, that fancy RNA-guided
DNA editing technique that was in the news?
That currently only works
for bacteria. The guy who discovered it wants to use AI to modify that
enzyme so it works in humans, but I think that's probably not usable for
eukaryotes. It seems useful for making GM microbes, but plasmid synthesis is
already easier than CRISPR.
--:-:-:-:-----------------------------------:-:-:-:--
OK then, how
about editing a virus so whatever inserts viral DNA is more selective?
Ah, a modified retroviral integrase? A selective one won't evolve
naturally because the success rate of inserting something somewhere will be
lower, but the existence of eg (Cas9 and IS110 and homology-directed repair)
shows such an enzyme should be possible. One of my friends has actually
worked on this a bit, but they're sort of holding back because of concerns
about bioweapons. (They're smarter than me and I'm not qualified to
second-guess their concerns.)
Even if the integrase of a virus can't
be made sequence-specific, there could still be benefits to modifying its
insertion tendencies. If a lentivirus is better than a γ-retrovirus in terms
of where it tends to insert DNA, then maybe you can do better than either.