Addressing IM resistance - info from this morning's talks

[Guest guest]

Hi, fellow CMLers,

I¹m going to jump forward to some talks that were given today because

they¹re still fresh with me and also really interesting. It was a session

on IM resistance with 6 different speakers, though I only heard 5 of them.

Jeez, these researchers are smart people! I think if I were to start all

over I¹d go into cancer research, just to hang out with folks like this

(also because cancer research is getting more and more interesting. Also

treating cancer patients is more fun than it used to be because there¹s so

much more one can do for them, and the future is even brighter. My

oncologist friend used to be a gloomy lot, for good reason; they seem a lot

happier nowadays!

The first talk was by the ubiquitous Deininger. Unfortunately I got

there a bit late, and the part of his talk I caught didn¹t tell me anything

new, beyond what I¹ve already reported.

Next we heard from Skaggs, from UCLA. He had some really interesting

stuff to say about the many kinase domain mutations that have been

identified to date. Some of these not only confer resistance to IM, but

also greater ³fitness² compared to non-mutated BA (that¹s BCR/ABL, in my

shorthand, remember?). Two mutations, T315I and E255, in particular, confer

growth and/or survival advantages which allow their cells to take over from

the cells with wild-type (un-mutated) BA. T315I and E255 are, by definition,

more transformed, i.e. more cancerous than the wild-type (WT) cells, so

folks unlucky enough to get them have to contend with a double whammy.

Interestingly, these are among the only mutations that show up in patients

with any frequency PRIOR to starting IM ­ which makes perfect sense given

their growth advantages. By contrast, none of the many mutations which show

reduced fitness show up before IM treatment; they undoubtedly occur, but

their host cells die out because they are out-competed by WT cells.

Finally, there¹s a set of mutations with approximately equivalent fitness as

WT cells, and as you might expect, they show up ­ but only occasionally - in

the cells of ³IM naïve² patients. Am I being at all clear here? I hope so,

because this stuff is worth understanding. It¹s pretty cool, actually ­

unless you happen to be unlucky enough to have one of the bad mutations. If

you are, read on, because there¹s hope for you too.

Skaggs mentioned one more thing I had known but forgotten: that IM only

binds BA in its inactive conformation. Not surprisingly therefore, most IM

resistance mutations occur in sites that hold BA in the active conformation,

to which IM can¹t bind. By contrast dasatinib (DS) binds the kinase domain

in both active and inactive states ­ which makes it more potent (it can bind

more of the time), AND explains why DS overcomes many IM resistance

mutations: it can still bind even when these mutations hold the domain

closed/inactive. The only kinase domain mutations affecting DS binding are

those which affect drug contact with BA. This is all very slick, albeit

sinister (and, btw, requires no intelligent designer, for those of you

interested in that debate).

How come some mutations confer greater fitness on their host cells? The

answer in most cases is that these mutations increase the enzymatic potency

of BA ­ that is, they can phosphorylate more downstream enzymes in less

time, so they crank up cell growth and inhibit cell death even better (or

worse, from our point of view) than their wild-type ancestors. Other

mutations don¹t make the kinase more potent, but turn it into a better

³transformer² by becoming a better autophosphorylator ­ but here he kind of

lost me, I¹m afraid.

Phew! OK, on to Copland in Tessa Holyoake¹s lab, on combination therapy

for targeting quiescent cells ­ a matter of great interest to those of us

lucky enough to have reached high-grade molecular responses (a 3 log

reduction or better in qPCR), as most of our remaining cells are presumably

of the quiescent BA+ stem cell variety. Dr. Holyoake¹s lab has previously

shown these cells to be insensitive to IM, as well as to the newer STI¹s,

AMN107 and DS. Why they¹re insensitive is not yet known, but Dr. H¹s lab

and others are hard at work on that. While we wait to find out, it would be

nice to be able to kill these cells though, and that¹s what this study is

about.

Last June, Dr. T et. alia, published a paper showing that one of the

farnesyl transferase inhibitors (FTIs) in combination with IM could

eliminate quiescent stem cells (again, because I¹m a lazy typist I¹ll call

them Qs from now on) better than IM alone; much better, in fact, because it

turns out that IM not only fails to kill Qs, it actually increases their

numbers, presumably by driving some of the dividing stem cells into the

quiescent state.

Dr. H then decided to poke around in the library of FTIs and see if any of

these could do a better job than lornafarnib, the one they tried last year.

Low and behold there is one: BMS 214662 (B662 from now on ­ no way I¹m

typing that long string!). Turns out that in combination with either IM or

DS, this handy little molecule reduces the Qs to nearly zero - at least in

the test tube. Mouse studies to follow, then hopefully human studies.

A couple of other points made here:

€ Some FTI¹s only inhibit cell growth, while others actually kill cells.

B662 is of the latter type, and Dr. H¹s studies show that all by itself it

³reaches down² farther into the stem cell compartment than IM does. That is

to say, B662 can kill some cells that IM cannot, even all on its own.

€ B662 appears to be effective against cancers other than CML, which makes

it MUCH more likely that this compound will be developed because the market

could be substantial.

€ The combo of IM plus B662 kills Q cells better (in the test tube anyway)

than does DS plus B662. This is interesting given that DS is more effective

at CML cell killing generally.

€ They don¹t yet know how B662 kills Qs, but they¹re working on it.

€ They tested B662 on normal white blood cells and found a little toxicity,

but less than from either IM or DS! Of course the question will be whether

the toxicity will be tolerable when these drugs are used together, not just

separately.

Next, Ellen Weisberg talked about combining AMN107 and IM in the treatment

of CML. Though members of the audience questioned some of her results, Dr.

Weisberg appears to have shown that their combined effect is at least

additive, and may actually be synergistic (greater then the sum of the

parts, that is). If this is true, or even if the combo is ONLY as effective

as either agent alone, she argued that combining them may a) suppress

emergence of resistant cells and reduce side effects. This work is early,

but it looks promising.

Copland again, this time looking at why Qs are insensitive to IM, AMN

and DS. There are three possibilities: 1) not enough of the drug is getting

into the cells to inhibit BA; 2) enough is getting in and BA is being

inhibited, but Qs aren¹t dependent on BA for survival; 3) enough is getting

in but BA has these cells have developed resistance mutations.

If (1), then the trick is to get more drug into the cells; if (2), then we

need to develop other strategies to kill these cells (using an FTI, for

example); if (3), then we need to figure out the resistance mechanism and

overcome it.

I don¹t think I can explain the supporting data (I didn¹t understand a lot

of it myself), but I think they¹ve concluded that these drugs are probably

getting into the cells just fine, but that BA is being overexpressed. This

isn¹t the whole story though: if overexpression was the only difference

from regular CML cells, then increasing the drug level should kill the Qs.

However, it doesn¹t, so it must also be true that these cells are not

³addicted² to BA as more mature Phillies are.

A couple more points of interest:

€ As noted above, both IM and DS increase the total number of Q cells, but

listen to this: DS + IM together increase Q cells even more!! It seems to

me that this kind of argues against combo therapy, at least until we find

out how to kill the Q cells.

€ BA and IM increase the number Q cells by 1-2 logs - quite a lot, really.

The last talk is the one I wrote to Tracey about earlier. It was by

Burleigh of SGX Pharmaceuticals on finding compounds that zap both wild type

and and 315 BA CML cells. More about that when I get the time. Good night

for now,

R