In this truly multi-disciplinary effort, now out in @NatureComms, lead author (on the post-doc job market!) @RobertVanderVe1, synthesizes a wide range of data including in vitro #barcoding and in vivo drug assays, #singlecellRNAseq, #mathonco -- something for everyone!
1
5
Using the ALK+ cell line H3122 as the workhorse for a series of #evolution experiments, we started by characterizing the differences in derived resistant lines in terms of collateral response and signalling.
1
1
6
Using selectively neutral DNA barcoding, we found significant diversity in populations of pre-existing ALK targeted therapy TOLERANT populations.
1
1
4
Then a combination of a temporal colony forming assay & stochastic sims helped us determine the distribution (and dynamics) of fitness in individual clones. Here is a stochastic sim from @NaraYoon12 w\ random increasing (levels of 'graduality') fitness & resulting colony growth.
1
1
6
tl;dr, the distribution of fitness is wide - and the initial distribution can not explain the dynamics of outgrowth without concomitant mutations... and importantly THIS CAN NOT be explained by saltatory, one hit mutations to full resistance -- only by a GRADUAL approach.
1
1
7
Comparing evolved resistant cells to a panel of cells w engineered resistance mechanisms: ALK overexp or L1196M muts at low levels alone were insufficient - only a high level of L1196M was able to match the evolved lines -- suggesting the need for more than one aberration.
1
4
Further analysis of the evolved lines revealed a wide variety of combinations existing together conferring resistance -- this makes the idea of a single 'silver bullet' much less likely, except in rare cases.
2
1
7
Enticingly, we also identified sensitivities that exist only in temporally restricted windows en route to full resistance. These 'temporally restricted collateral sensitivities' give hope for evolutionarily rational therapies, but require knowledge of the DYNAMICS of evolution.
1
2
12
The last figure (previous tweet) is awesome for two reasons. first, @AndriyMarusyk snuck some emojis into a figure; and second, it highlights the need for an understanding of an underlying dynamical model. Here, we propose a fitness landscape, but others could be used. (\end)
9
Sort replies: Relevant Recent Liked
Fantastic work @CancerConnector, @AndriyMarusyk & colleagues! Highlighting the importance of measuring *temporal dynamics* in experimental models of drug resistance!
2
9
Congratulations @AndriyMarusyk & @CancerConnector! Beautifully designed study on evolution of resistance to ALK inhibitors!
1
3
Congratulations on a great looking paper, am looking forward to a good read!
2