NEW YORK (GenomeWeb) – A pair of studies published online today in Science is providing a look at the evolution and resulting genetic diversity present in tumors from non-small cell cancer lung cancer sub-types.
"Survival from lung cancer remains devastatingly low with many new targeted treatments making a limited impact on the disease," Charles Swanton, a researcher affiliated with University College London and the Cancer Research UK London Research Institute, said in a statement. "By understanding how it develops we've opened up the disease's evolutionary rulebook in the hope that we can start to predict its next steps."
For one of the studies, Swanton and his team did whole-exome and/or whole-genome sequencing on more than two-dozen regions, representing tumors that had been surgically removed from seven NSCLC patients prior to adjuvant treatment. These individuals included current smokers, former smokers, and never smokers with either lung adenocarcinoma or squamous cell carcinoma NSCLC sub-types.
The researchers then used the multi-region tumor sequences and sequences from each patient's matched normal samples to develop models of lung cancer evolution — a process that involved teasing apart mutations that were ubiquitous from those found in a subset of tumor regions and estimating the proportion of cells with a given mutation in each region.
In particular, they saw signs that early driver mutations — including a range of alterations attributed to cigarette smoke in current and former smokers — sometimes simmer undetected for years before progressing to full-blown lung tumors. As lung cancers evolve, the sequence data suggests, additional mutations start to snowball within tumor sub-clones, leading to multiple, genetically heterogeneous regions within the same tumor.
The team noted that alterations associated with APOBEC cytidine deaminase enzyme activity tend to become increasingly prominent during tumor formation and progression. On the other hand, the accumulation of new mutations linked to smoking carcinogens seemed to wane as tumor formation advanced.
Such intra-patient tumor variation highlights the importance of understanding the stages of lung tumor evolution and finding ways to detect and treat burgeoning tumors prior to the introduction of diverse sub-clones, authors of the study argued, since it is expected to become increasingly difficult to apply targeted treatments effective against the full suite of sub-clones found in a given tumor.
"The regionally separated driver mutations, coupled with the relentless and heterogeneous nature of the genome instability processes, are likely to confound treatment success in NSCLC," they wrote.
For another Science study, University of Texas MD Anderson Cancer Center genomic medicine researcher Andrew Futreal, an honorary faculty member with the Wellcome Trust Sanger Institute, led a team that applied multi-region whole-exome sequencing to four-dozen regions selected from 11 lung adenocarcinoma tumors harboring apparent intra-tumor heterogeneity.
In those samples, researchers identified more than 7,000 verified somatic mutations across the complete sample set, though the number of mutations per tumor varied dramatically.
The analysis indicated that more than three-quarters of the protein-coding mutations in a given sample were shared with other regions from the same tumor, prompting authors involved in that paper to propose that "single-region sequencing may be adequate to identify the majority of known cancer gene mutations in localized lung adenocarcinomas."
Still, the investigators also saw evidence for branched evolution in each of the tumors, leading to sub-clone mutations not found across the complete tumor. Again, mutations affecting cancer driver genes, along with cancer gene amplifications or deletions, appeared to turn up early in the evolutionary process of lung cancer formation, they reported.
Amongst the three patients who went on to have lung cancer relapse, members of that team saw an over-representation of mutations that were sub-clone specific and not shared across tumor regions, hinting that this heterogeneity may herald poor outcomes, though they noted that larger sample sizes are needed to further explore that possibility.
As in the other multi-region sequencing study of NSCLC, authors of the analysis detected a shift in the sorts of mutations occurring in samples from current and former smokers over time, with a preponderance of smoking-related mutation signatures early on in tumor evolution and a rise in mutations not linked to this carcinogenic exposure in later-forming tumor sub-clones.
That group concluded that the "extension of research to epigenetic and phenotypic assessment through regional DNA methylation, chromatin state, and RNA and/or protein expression studies over time and under treatment is needed to fully understand the impact of [intra-tumor heterogeneity] on the biology of the cancer itself and its impact on the clinical phenotype of cancer patients."