by hyperactivation of the MAPK pathway and can perhaps be targeted with BET and MEK
inhibitors, those with loss of the tumor suppressor NF1, which have a higher mutational
burden and present at an older age, and those without mutations in any of these driver genes
[33]. In colon cancer, different genomic subtypes can be recognized with chromosomal
unstable tumors having defects in chromosome segregation, telomere stability, and DNA
damage response and hypermutated tumors usually having a defective DNA mismatch
repair system [34]. The latter respond well to immune checkpoint inhibitors, especially if
they have a high mutational burden, and, therefore, having a hypermutator phenotype is
also a biomarker that can help stratify patients for treatment choice.
Transcriptomic characterization of tumors has also been of great use to identify subtypes
within a cancer type that can be treated specifically to boost therapy efficacy. For example,
breast cancer has been classified into subtypes according to gene expression: Luminal A,
luminal B, ERBB2-overexpressing, normal-like, and basal-like [35]. This classification
takes into account expression of the estrogen receptor (ER+), the progesterone receptor (PR
+), and ERBB2 amplification [35]. Luminal tumors are likely to respond to endocrine
therapy, while those with ERBB2 amplification can be targeted with trastuzumab and
chemotherapy. Treatment of triple-negative breast cancers is more challenging, but
PARP inhibitors and immunotherapy are beginning to be tested in the clinic [36]. Melanoma tumors have also been classified into expression subtypes with potential therapeutic
implications [37]. Therefore, the sequencing of both tumor DNA and RNA can inform
about the biology of cancer and help identify potential therapeutic targets.
As briefly mentioned above, the most useful aspects of tumor sequencing are the
identification of therapeutical targets and the discovery of tumor biomarkers. Apart from
classical such biomarkers like the presence of established driver mutations, another one that
has emerged as an important predictor of response to immunotherapy is tumor mutational
burden (TMB), as some authors suggest that higher TMB predicts favorable outcome to
PD-1/PD-L1 blockade across diverse tumors [38]. Tumors with high TMB are more likely
to respond to this type of treatment because the number of “neoepitopes” is higher, this is,
the amount of novel peptides that arise from tumor-specific mutations and can be
recognized by the immune system [36]. Chromosomal instability, which can also be
detected by sequencing, may also impact therapeutic response.
An exciting development fueled by the ability to identify neoepitopes by nextgeneration sequencing is adoptive T-cell therapy. In this form of treatment, tumor/normal
sample pairs are sequenced in order to identify novel mutations that may be targetable by
the immune system [39]. After these are validated by RNA expression analysis and mass
spectrometry, and are deemed good ligands for HLA molecules by bioinformatic analyses,
then they are co-cultured with tumor-infiltrating lymphocytes (TILs) resected from a tumor
biopsy from the same patient [40]. This methodology then allows the selective expansion
of TILs with a specific reactivity to the target tumor, and can then be introduced back into
the patient.
Another topic that has gained traction in recent years due to the advent of single-cell
transcriptome sequencing is the realization that drug resistance in cancer can be generated
2 Opportunities and Perspectives of NGS Applications in Cancer Research
29
inhibitors, those with loss of the tumor suppressor NF1, which have a higher mutational
burden and present at an older age, and those without mutations in any of these driver genes
[33]. In colon cancer, different genomic subtypes can be recognized with chromosomal
unstable tumors having defects in chromosome segregation, telomere stability, and DNA
damage response and hypermutated tumors usually having a defective DNA mismatch
repair system [34]. The latter respond well to immune checkpoint inhibitors, especially if
they have a high mutational burden, and, therefore, having a hypermutator phenotype is
also a biomarker that can help stratify patients for treatment choice.
Transcriptomic characterization of tumors has also been of great use to identify subtypes
within a cancer type that can be treated specifically to boost therapy efficacy. For example,
breast cancer has been classified into subtypes according to gene expression: Luminal A,
luminal B, ERBB2-overexpressing, normal-like, and basal-like [35]. This classification
takes into account expression of the estrogen receptor (ER+), the progesterone receptor (PR
+), and ERBB2 amplification [35]. Luminal tumors are likely to respond to endocrine
therapy, while those with ERBB2 amplification can be targeted with trastuzumab and
chemotherapy. Treatment of triple-negative breast cancers is more challenging, but
PARP inhibitors and immunotherapy are beginning to be tested in the clinic [36]. Melanoma tumors have also been classified into expression subtypes with potential therapeutic
implications [37]. Therefore, the sequencing of both tumor DNA and RNA can inform
about the biology of cancer and help identify potential therapeutic targets.
As briefly mentioned above, the most useful aspects of tumor sequencing are the
identification of therapeutical targets and the discovery of tumor biomarkers. Apart from
classical such biomarkers like the presence of established driver mutations, another one that
has emerged as an important predictor of response to immunotherapy is tumor mutational
burden (TMB), as some authors suggest that higher TMB predicts favorable outcome to
PD-1/PD-L1 blockade across diverse tumors [38]. Tumors with high TMB are more likely
to respond to this type of treatment because the number of “neoepitopes” is higher, this is,
the amount of novel peptides that arise from tumor-specific mutations and can be
recognized by the immune system [36]. Chromosomal instability, which can also be
detected by sequencing, may also impact therapeutic response.
An exciting development fueled by the ability to identify neoepitopes by nextgeneration sequencing is adoptive T-cell therapy. In this form of treatment, tumor/normal
sample pairs are sequenced in order to identify novel mutations that may be targetable by
the immune system [39]. After these are validated by RNA expression analysis and mass
spectrometry, and are deemed good ligands for HLA molecules by bioinformatic analyses,
then they are co-cultured with tumor-infiltrating lymphocytes (TILs) resected from a tumor
biopsy from the same patient [40]. This methodology then allows the selective expansion
of TILs with a specific reactivity to the target tumor, and can then be introduced back into
the patient.
Another topic that has gained traction in recent years due to the advent of single-cell
transcriptome sequencing is the realization that drug resistance in cancer can be generated
2 Opportunities and Perspectives of NGS Applications in Cancer Research
29