University of Groningen
Next generation sequencing guided molecular diagnostic tests in non-small-cell lung cancer
Wei, Jiacong
DOI:
10.33612/diss.101317239
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below.
Document Version
Publisher's PDF, also known as Version of record
Publication date: 2019
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):
Wei, J. (2019). Next generation sequencing guided molecular diagnostic tests in non-small-cell lung cancer. Rijksuniversiteit Groningen. https://doi.org/10.33612/diss.101317239
Copyright
Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.
6
Summary, discussion and
Chapter 6
Summary
The objective of this thesis was to further develop and improve diagnostic methods for a more comprehensive clinical guidance of patients and to explore the predictive value of EGFR amplifications on tyrosine kinase inhibitors (TKI) response.
In the diagnostic setting, multiple tests including targeted DNA sequencing, FISH and IHC are applied to screen for genomic aberrations relevant for selecting the most optimal therapy. However, the size of the tissue biopsies obtained from lung cancer patients are usually small and therefore can be a limiting factor. In chapter 2, we developed an all‐in‐one RNA‐based NGS assay to simultaneously detect all relevant therapeutic biomarkers. These biomarkers are heterogeneous in nature, including SNVs, INDELs, exon skipping, fusion genes, and gene amplifications, and are currently detected by a combination of different tests. We tested our assay in six lung cancer derived cell lines, five cell lines derived from other cancer types, and 42 lung tumour samples of variable RNA quality, all with known clinically relevant aberrations. For detection of SNVs, INDELs, gene fusions and exon skippings, the assay showed a 100% sensitivity and specificity in FFPE samples that fulfil our proposed inclusion criteria, i.e. a minimal DV200 value of 50 and a minimum of 50K unique reads. Besides, we were able to identify the fusion gene partners, which may have predictive value upon targeted therapies. Currently, there are few methods for simultaneous detection of SNVs, INDELs, exon skippings and gene fusions. Compared with a similar assay, i.e. the TruSight RNA Pan‐Cancer Panel Assay (Illumina, San Diego, CA), our assay is more focused on therapy relevant aberrations with a smaller target region. As the costs for NGS based approaches are for a large part determined by the sequencing costs, a focused panel is more cost effective. Using a limited lung cancer panel as designed by us, 16 samples can be easily pooled even using low capacity sequencers, such as the MiSeq (Illumina, USA). This is appealing especially for large molecular diagnostic pathology centres where a relatively large number of specific cancer types can be collected shortly on a daily basis. For hospitals with a lower number of NGS tests, the iSeq 100 Sequencing System (Illumina, USA), with a sequence capacity for four samples, could be a good alternative. Other commercial panels are listed in Table 1. Some have larger target sizes focusing on genes more broadly associated with cancer while others have a limited set of target genes focusing only on therapeutically relevant markers. For developing countries, these limited panels might be sufficient as also the availability of TKIs is limited. Another advantage of our all‐in‐one assay is that it allows the design of capturing probes close to the hotspot regions, which allows detection of variants, even in poor quality RNA samples such as those obtained from older FFPE tissue blocks. Moreover, our assay has the flexibility of quickly customizable landing probes and adaptation to the latest diagnostic demands. Comparative panels can easily be designed for other malignancies. Overall, considering the accuracy, the costs, as well as the flexibility of our customized panel, our RNA‐based NGS assay has great potential to be applied in routine molecular diagnostic settings. An as yet unexplored application might be the detection of the overexpression as a potential surrogate marker for amplifications since the former one, by theory, is more biologically representative of tumour functionality.
It is recommended to treat NSCLC patients with EGFR sensitizing mutations with TKI drugs. A few studies reported the predictive value of concurrent EGFR amplifications, but currently available data is limited. In chapter 3, we did a retrospective analysis using 3,540 targeted next
SUMMARY, DISCUSSION AND FUTURE PERSPECTIVES
6
Summary
The objective of this thesis was to further develop and improve diagnostic methods for a more comprehensive clinical guidance of patients and to explore the predictive value of EGFR amplifications on tyrosine kinase inhibitors (TKI) response.
In the diagnostic setting, multiple tests including targeted DNA sequencing, FISH and IHC are applied to screen for genomic aberrations relevant for selecting the most optimal therapy. However, the size of the tissue biopsies obtained from lung cancer patients are usually small and therefore can be a limiting factor. In chapter 2, we developed an all‐in‐one RNA‐based NGS assay to simultaneously detect all relevant therapeutic biomarkers. These biomarkers are heterogeneous in nature, including SNVs, INDELs, exon skipping, fusion genes, and gene amplifications, and are currently detected by a combination of different tests. We tested our assay in six lung cancer derived cell lines, five cell lines derived from other cancer types, and 42 lung tumour samples of variable RNA quality, all with known clinically relevant aberrations. For detection of SNVs, INDELs, gene fusions and exon skippings, the assay showed a 100% sensitivity and specificity in FFPE samples that fulfil our proposed inclusion criteria, i.e. a minimal DV200 value of 50 and a minimum of 50K unique reads. Besides, we were able to identify the fusion gene partners, which may have predictive value upon targeted therapies. Currently, there are few methods for simultaneous detection of SNVs, INDELs, exon skippings and gene fusions. Compared with a similar assay, i.e. the TruSight RNA Pan‐Cancer Panel Assay (Illumina, San Diego, CA), our assay is more focused on therapy relevant aberrations with a smaller target region. As the costs for NGS based approaches are for a large part determined by the sequencing costs, a focused panel is more cost effective. Using a limited lung cancer panel as designed by us, 16 samples can be easily pooled even using low capacity sequencers, such as the MiSeq (Illumina, USA). This is appealing especially for large molecular diagnostic pathology centres where a relatively large number of specific cancer types can be collected shortly on a daily basis. For hospitals with a lower number of NGS tests, the iSeq 100 Sequencing System (Illumina, USA), with a sequence capacity for four samples, could be a good alternative. Other commercial panels are listed in Table 1. Some have larger target sizes focusing on genes more broadly associated with cancer while others have a limited set of target genes focusing only on therapeutically relevant markers. For developing countries, these limited panels might be sufficient as also the availability of TKIs is limited. Another advantage of our all‐in‐one assay is that it allows the design of capturing probes close to the hotspot regions, which allows detection of variants, even in poor quality RNA samples such as those obtained from older FFPE tissue blocks. Moreover, our assay has the flexibility of quickly customizable landing probes and adaptation to the latest diagnostic demands. Comparative panels can easily be designed for other malignancies. Overall, considering the accuracy, the costs, as well as the flexibility of our customized panel, our RNA‐based NGS assay has great potential to be applied in routine molecular diagnostic settings. An as yet unexplored application might be the detection of the overexpression as a potential surrogate marker for amplifications since the former one, by theory, is more biologically representative of tumour functionality.
It is recommended to treat NSCLC patients with EGFR sensitizing mutations with TKI drugs. A few studies reported the predictive value of concurrent EGFR amplifications, but currently available data is limited. In chapter 3, we did a retrospective analysis using 3,540 targeted next
generation sequencing (NGS) files that was routinely generated in clinical diagnostic settings. Gene amplifications were analysed by comparing the read counts per amplicon with either internal control amplicons or normal samples as external controls. Amplification status of EGFR was validated by multiplex ligation‐dependent probe amplification (MLPA) with high positive and negative predictive values. Amplification status in MET and ERBB2 were validated using fluorescence in situ hybridization (FISH), with good negative predictive values. However, we could not draw conclusions on positive predictive values due to the low number of FISH positive samples. We focused on advanced stage lung adenocarcinoma patients (n=1,586), of which 146 were reported to have EGFR mutations in the diagnostics. Among the 66 patients from whom we could retrieve clinical data, 49 were treated with first line TKI. Patients with concurrent amplifications had worse overall survival upon TKI treatment as compared to cases without concurrent amplifications using the internal control amplicon approach. This, indicated that
EGFR amplification is a predictor for poor TKI treatment response. A significant differences was
found for both the ratio, being a marker for the magnitude of the amplification, and for the z score, which is a marker of the reliability of the observed amplification. Moreover, a borderline significant interaction was observed for the two parameters and overall survival.
Most NSCLC patients treated with targeted therapy will inevitably develop resistance. Previous studies on crizotinib resistance mechanisms focused on cell lines and tissue samples obtained from resistant tumour sites, without doing a direct comparison to the pre‐treatment samples [1‐ 3]. These studies showed a role of several ALK‐independent bypass mechanisms, including activation of EGFR, KRAS, ERBB and MAPK signalling. However, none of these studies directly compared pre‐ and post‐crizotinib treatment samples, which is required to actually pinpoint the underlying causal resistance‐associated alterations. The strength in chapter 4 is that we compared, matched pre‐ and post‐ treatment biopsies. We focused on the detection of crizotinib‐induced genomic aberrations in re‐biopsies of ALK positive NSCLC patients obtained upon relapse during treatment. Biopsies taken before crizotinib treatment and upon resistance were analysed to identify potential therapy resistant mutations. Starting from a total number of 29 patients, we were able to obtain sufficient material for paired analysis of pre‐ and post‐TKI biopsies for WES for 4 patients. We identified a total of 582 mutations, with 175 mutations being specific for the post‐TKI samples or showing higher variant allele frequencies. An ALK gatekeeper mutation was observed in one of the four patients upon resistance. Based on gene ontology and pathway analysis on the other three patients, we showed an enrichment of mutations in genes associated with epithelial‐mesenchymal transition (EMT) related pathways. This suggested that loss of epithelial differentiation may be an important crizotinib resistance mechanism. Still a limiting factor in our study, and also more general in the literature, is the lack of paired samples, with sufficiently large tissue samples and high tumour percentages.
In recent years, the concept of liquid biopsies using body fluids as diagnostic material is getting more and more attention. In lung cancer, the use of cfDNA to detect therapy‐guiding mutations and known resistance mutations has already been approved. Compared with traditional tissue biopsies, liquid biopsies might be more representative of the heterogeneous lung tumour as compared to regional sampling of the tumour tissue [4]. Moreover, it is a minimal‐invasive procedure. Nevertheless, one should always bear in mind the relative high percentage of false negative results, especially in early disease stages, due to the relatively low signal‐to‐noise ratio
Chapter 6
of NGS. Negative results should be handled with care in a diagnostic setting [5], and may even influence the applicability. In chapter 5, we studied the potential of cell‐free DNA (cfDNA) as a disease biomarker to monitor tumour load after surgical removal of oesophageal squamous cell carcinoma (ESCC). We tried to differentiate between variants present in circulating tumour DNAs (ctDNAs) from the background noise caused by sequencing errors. Somatic variants in pre‐ surgery plasma cfDNA samples were found in 8 of 12 patients. While in post‐surgery plasma cfDNA samples of 14 patients, we found somatic variants only in two patients. One of whom relapsed 12 months after surgery.So far, our study is the most extensive study in stage II and III ESCC focusing on pre‐ and post‐surgery cfDNAs. Although a larger sample size is required for a more solid conclusion, our study is a good example showing the potential of ctDNA as a biomarker of residual disease after surgical resection of ESCC patients. This is of essential importance since biomarkers to predict relapse in ESCC patients after surgery are not yet available. NGS based analyses of liquid biopsies with a broad gene panel will provide information from baseline to further follow up.
Table 1. Curr ent NGS p an els for genomic ab erration d et ecti on . Pane l Nam e Com pan y/Institu te (City, Cou nt ry ) Can ce r Type Gen e Nu m be r Target Libr ar y En richm en t m eth od Re fer en ce s All‐ in ‐on e lun g ass ay UMC G (G ron ing en, th e Neth erland s) Lu ng can cer 21 Ho tspot s + Fu sion s SPET Chap ter 3, th is thesi s Onco mine Therm oFis her (Walth am, USA) Lu ng can cer 23 Ho tspot s + RO S1 Fusion Amplicon ‐ ba se d ht tp s://w ww .th erm ofis her.co m/nl/en/ho me /clini cal/p reclinical ‐co mpan ion ‐dia gno stic ‐ deve lop me nt /on co mine ‐on cology /an aly tical ‐ performanc e‐ on co mine ‐nci ‐m atch ‐trial ‐a ss ay .h tml MSK ‐IMPAC T Mem oria l Sloan Kettering Cancer C enter (Ne w Y ork , USA) Pan can cer 468 Ho tspot s + Fu sion Hy brid cap tu re [6] Fou nd atio nOn e Fou nd atio n Medicine (Cambridg e, USA ) Pan can cer 324 Ho tspot s + Fu sion Hy brid cap tu re ht tp s://w ww .foun da tion me di cine.com /geno mic ‐ tes tin g/ foun da tion ‐on e‐ cdx OmniSeq Ad van ce Roswell Pa rk Canc er Institu te (Buff alo, U SA) Pan can cer 144 Ho tspot s + Fu sion + Amplifi catio ns Amplicon [7] No voFocu s™ NSCLC C Dx T est No vogene ( Tianjin , Chin a) Lu ng can cer 6 Ho tspot s + Fu sion Amplicon ‐ ba se d ht tp s://en.n ovo gene. com/ph arma ‐ ser vic es /tran slation al ‐an d‐ clinica l/no vofo cus ‐ nsclc ‐cdx/ Esse nt ial N GS Pan el Amory Dx (Xia me n, Chin a) Lu ng an d colon can cer 10 Ho tspot s + Fu sion Hy brid cap tu re ht tp ://w w w.amoyd iagn osti cs .com/pro du ctDetail _ 9.ht ml NCC On cop an el Sysm ex (Kobe, Japan ) Pan can cer 114 Ho tspot s + Fu sion + Amplifi catio ns Hy brid cap tu re ht tp s://w ww .mhlw .go.jp /fil e/ 05 ‐Shin gikai ‐ 109010 00 ‐Kenkou ky oku ‐ Sou muka /0000 1797 57.pd f Toda i On coPan el Riken gene sis (T oky o, Japa n) Pan can cer 464 Ho tspot s + Fu sion + Amplifi catio ns Hy brid cap tu re ht tp ://tod aion cop an el.umin.j p/#s ec0 1 SPET: Singl e prim er enrich men t t echno logy
SUMMARY, DISCUSSION AND FUTURE PERSPECTIVES
6
of NGS. Negative results should be handled with care in a diagnostic setting [5], and may even influence the applicability. In chapter 5, we studied the potential of cell‐free DNA (cfDNA) as a disease biomarker to monitor tumour load after surgical removal of oesophageal squamous cell carcinoma (ESCC). We tried to differentiate between variants present in circulating tumour DNAs (ctDNAs) from the background noise caused by sequencing errors. Somatic variants in pre‐ surgery plasma cfDNA samples were found in 8 of 12 patients. While in post‐surgery plasma cfDNA samples of 14 patients, we found somatic variants only in two patients. One of whom relapsed 12 months after surgery.So far, our study is the most extensive study in stage II and III ESCC focusing on pre‐ and post‐surgery cfDNAs. Although a larger sample size is required for a more solid conclusion, our study is a good example showing the potential of ctDNA as a biomarker of residual disease after surgical resection of ESCC patients. This is of essential importance since biomarkers to predict relapse in ESCC patients after surgery are not yet available. NGS based analyses of liquid biopsies with a broad gene panel will provide information from baseline to further follow up.
Table 1. Curr ent NGS p an els for genomic ab erration d et ecti on . Pane l Nam e Com pan y/Institu te (City, Cou nt ry ) Can ce r Type Gen e Nu m be r Target Libr ar y En richm en t m eth od Re fer en ce s All‐ in ‐on e lun g ass ay UMC G (G ron ing en, th e Neth erland s) Lu ng can cer 21 Ho tspot s + Fu sion s SPET Chap ter 3, th is thesi s Onco mine Therm oFis her (Walth am, USA) Lu ng can cer 23 Ho tspot s + RO S1 Fusion Amplicon ‐ ba se d ht tp s://w ww .th erm ofis her.co m/nl/en/ho me /clini cal/p reclinical ‐co mpan ion ‐dia gno stic ‐ deve lop me nt /on co mine ‐on cology /an aly tical ‐ performanc e‐ on co mine ‐nci ‐m atch ‐trial ‐a ss ay .h tml MSK ‐IMPAC T Mem oria l Sloan Kettering Cancer C enter (Ne w Y ork , USA) Pan can cer 468 Ho tspot s + Fu sion Hy brid cap tu re [6] Fou nd atio nOn e Fou nd atio n Medicine (Cambridg e, USA ) Pan can cer 324 Ho tspot s + Fu sion Hy brid cap tu re ht tp s://w ww .foun da tion me di cine.com /geno mic ‐ tes tin g/ foun da tion ‐on e‐ cdx OmniSeq Ad van ce Roswell Pa rk Canc er Institu te (Buff alo, U SA) Pan can cer 144 Ho tspot s + Fu sion + Amplifi catio ns Amplicon [7] No voFocu s™ NSCLC C Dx T est No vogene ( Tianjin , Chin a) Lu ng can cer 6 Ho tspot s + Fu sion Amplicon ‐ ba se d ht tp s://en.n ovo gene. com/ph arma ‐ ser vic es /tran slation al ‐an d‐ clinica l/no vofo cus ‐ nsclc ‐cdx/ Esse nt ial N GS Pan el Amory Dx (Xia me n, Chin a) Lu ng an d colon can cer 10 Ho tspot s + Fu sion Hy brid cap tu re ht tp ://w w w.amoyd iagn osti cs .com/pro du ctDetail _ 9.ht ml NCC On cop an el Sysm ex (Kobe, Japan ) Pan can cer 114 Ho tspot s + Fu sion + Amplifi catio ns Hy brid cap tu re ht tp s://w ww .mhlw .go.jp /fil e/ 05 ‐Shin gikai ‐ 109010 00 ‐Kenkou ky oku ‐ Sou muka /0000 1797 57.pd f Toda i On coPan el Riken gene sis (T oky o, Japa n) Pan can cer 464 Ho tspot s + Fu sion + Amplifi catio ns Hy brid cap tu re ht tp ://tod aion cop an el.umin.j p/#s ec0 1 SPET: Singl e prim er enrich men t t echno logy
Chapter 6
Discussion and Future Perspectives
1. NGS-based Molecular Diagnostic testsIn the last decade, the use of NGS tests for detection of therapy‐guiding aberrations in the molecular diagnostic setting has become an integrated part of routine clinical procedures. The presence of activating variants is considered to be essential to guide therapy for lung cancer patients. For example, the most commonly observed targetable driver mutations in
EGFR are L858R and deletions in exon 19. Less common EGFR aberrations such as EGFR exon
20 insertions are insensitive to traditional EGFR‐TKI. Over the years we have learned the biological features of the intrinsic or primary resistance of these mutations. Therefore, new trials with very specific EGFR‐TKIs targeting EGFR with exon 20 insertions such as TAK788, poziotinib and TAS6417 or higher dose of new EGFR‐TKI such as osimertinib have been started to study clinical benefit.
More recent studies suggest that next to specific drug sensitive mutations (SNVs and INDELs), mutant allele frequency (MAF) of these mutations, as well as copy numbers of the mutant allele, may play a role in predicting tumour response. For example, some studies found that EGFR mutant NSCLC patients with co‐existing EGFR amplifications have a longer PFS than those with only EGFR drug sensitive mutation (16.3 months versus 9.1 months,
p=0.004) [8]. Others showed that NSCLC patients with higher mutant allele frequencies
(MAF) of the EGFR sensitive mutations have better PFS when treated with EGFR TKIs than those with low MAF (12.3 months versus 2 months, p<0.001 using a cut‐off value of 9.5% for L858R; 15 months versus 2.0 months, P<0.001, using a cut‐off value of 4.9% for exon 19 deletion) [9]. A possible explanation might be that a high MAF is an indication of amplification of the allele carrying the driver mutation. Alternatively, a high MAF might be related to relatively large percentage of the tumour cells carrying the specific driver mutation and therefore shows a better response.
Another factor reported to be relevant for response to therapy is the fusion partner in ALK positive patients. The most commonly observed chromosomal rearrangement leads to EML4–ALK fusion genes, albeit with different breakpoint regions in both the EML4 and ALK gene. These fusion genes are also referred to as canonical ALK fusions. Response to ALK TKI has been shown to be better in patients with canonical ALK-EML4 fusions as compared to the rare non‐canonical ALK fusions (20.6 months V.S. 5.4 months, P<0.01) [10]. In another study, patients with an EML4_E20‐ALK_E20 fusion transcript (n=9) were shown to have better PFS to crizotinib compared with other ALK fusion types (34.5 months V.S. 14.3 months, P=0.021) [11], which include EML4_E6‐ALK_E20 (n=20), EML4_E13‐ALK_E20 (n=14) and other rare types (n=6) e.g. EML4_E14‐ALK_E20 and EML4_E18‐ALK_E20. Yet, the published studies were performed using relatively small cohorts (of 20 and 96 patients, respectively) and need to be validated [10,11].
As discussed above, differences in mutant allele frequencies, in fusion gene partners as well as in CNVs, next to the pharmacokinetic properties of the drugs, might be predictive for tumour response under the same targeted therapy. In the future, larger multi‐centre studies are needed to draw solid conclusions on the predictive value of VAF, co‐existing amplifications and the nature of the fusion transcripts.
For lung cancer, amplification of ERBB2 and MET are relevant for therapy choices. At this moment, gene amplifications are usually assessed using FISH. A limited number of cells are counted, to determine the increase in gene copies relative to the copy number of the
SUMMARY, DISCUSSION AND FUTURE PERSPECTIVES
6
Discussion and Future Perspectives
1. NGS-based Molecular Diagnostic testsIn the last decade, the use of NGS tests for detection of therapy‐guiding aberrations in the molecular diagnostic setting has become an integrated part of routine clinical procedures. The presence of activating variants is considered to be essential to guide therapy for lung cancer patients. For example, the most commonly observed targetable driver mutations in
EGFR are L858R and deletions in exon 19. Less common EGFR aberrations such as EGFR exon
20 insertions are insensitive to traditional EGFR‐TKI. Over the years we have learned the biological features of the intrinsic or primary resistance of these mutations. Therefore, new trials with very specific EGFR‐TKIs targeting EGFR with exon 20 insertions such as TAK788, poziotinib and TAS6417 or higher dose of new EGFR‐TKI such as osimertinib have been started to study clinical benefit.
More recent studies suggest that next to specific drug sensitive mutations (SNVs and INDELs), mutant allele frequency (MAF) of these mutations, as well as copy numbers of the mutant allele, may play a role in predicting tumour response. For example, some studies found that EGFR mutant NSCLC patients with co‐existing EGFR amplifications have a longer PFS than those with only EGFR drug sensitive mutation (16.3 months versus 9.1 months,
p=0.004) [8]. Others showed that NSCLC patients with higher mutant allele frequencies
(MAF) of the EGFR sensitive mutations have better PFS when treated with EGFR TKIs than those with low MAF (12.3 months versus 2 months, p<0.001 using a cut‐off value of 9.5% for L858R; 15 months versus 2.0 months, P<0.001, using a cut‐off value of 4.9% for exon 19 deletion) [9]. A possible explanation might be that a high MAF is an indication of amplification of the allele carrying the driver mutation. Alternatively, a high MAF might be related to relatively large percentage of the tumour cells carrying the specific driver mutation and therefore shows a better response.
Another factor reported to be relevant for response to therapy is the fusion partner in ALK positive patients. The most commonly observed chromosomal rearrangement leads to EML4–ALK fusion genes, albeit with different breakpoint regions in both the EML4 and ALK gene. These fusion genes are also referred to as canonical ALK fusions. Response to ALK TKI has been shown to be better in patients with canonical ALK-EML4 fusions as compared to the rare non‐canonical ALK fusions (20.6 months V.S. 5.4 months, P<0.01) [10]. In another study, patients with an EML4_E20‐ALK_E20 fusion transcript (n=9) were shown to have better PFS to crizotinib compared with other ALK fusion types (34.5 months V.S. 14.3 months, P=0.021) [11], which include EML4_E6‐ALK_E20 (n=20), EML4_E13‐ALK_E20 (n=14) and other rare types (n=6) e.g. EML4_E14‐ALK_E20 and EML4_E18‐ALK_E20. Yet, the published studies were performed using relatively small cohorts (of 20 and 96 patients, respectively) and need to be validated [10,11].
As discussed above, differences in mutant allele frequencies, in fusion gene partners as well as in CNVs, next to the pharmacokinetic properties of the drugs, might be predictive for tumour response under the same targeted therapy. In the future, larger multi‐centre studies are needed to draw solid conclusions on the predictive value of VAF, co‐existing amplifications and the nature of the fusion transcripts.
For lung cancer, amplification of ERBB2 and MET are relevant for therapy choices. At this moment, gene amplifications are usually assessed using FISH. A limited number of cells are counted, to determine the increase in gene copies relative to the copy number of the
chromosome (e.g., 20 cells for ERBB2, 50 cells for MET). The result might not be representative of the total heterogeneous tumour mass. In some cases, expression of the potentially amplified gene locus is assessed at the protein level using IHC. However, scoring of staining intensities is subjective and for ERBB2 in lung cancer no international guidelines are available. In theory, detection of amplification using DNA‐based NGS data is more quantitative but influenced by the admixture of normal cells. An alternative approach might be the use of RNA‐based NGS data to determine expression levels, which may serve as an indirect marker for amplification. However, this analysis is also influenced by admixture of normal cells. In the future, better criteria for scoring and coupling of the amplification data to clinical outcome should be performed to evaluate the predictive value of amplification tests in a systematic way. A good example is the study that directly compared the predictive value of ALK FISH and IHC on treatment outcome and showed a better predictive value for IHC based ALK scoring to predict response to ALK [12].
To meet the demand of detecting an increasing number of variants, NGS‐based methods are gradually taking over the traditional one‐by‐one tests, such as FISH and IHC. FISH and IHC are more labour intensive and more dependent on subjective scoring by technicians or pathologist. On the contrary, NGS based assays cover multiple targets and yield information not only for therapy decision, but potentially also for response prediction. Macrodissection of tumour‐rich areas in combination with correcting for tumour cell percentage might enhance the predictive value, when analysed in a well‐defined standardized approach. Diagnostic NGS assays can be categorized into RNA‐based assays, DNA‐based assays, or a combination of both assays using total nucleic acids as input material for two different library preparation approaches in parallel. In lung cancer where targeted therapy choice is guided by activation of driver genes, a comprehensive single RNA based NGS assay, in theory, is the most effective method. Compared with DNA sequencing, RNA sequencing allows quantification of the expression of the mutant allele, which is a requirement for being an oncogenic driver. Moreover, it allows detection of the consequences of intronic variants, such as the exon 14 skipping events in the MET gene. Finally, fusion transcripts can be readily characterized, which is relevant for prediction of response as shown for ALK fusion gene partners. Besides, therapy‐guiding driver genes are expected to have high expression levels especially in the tumour cells, which may reduce the impact of admixture of normal cells and low tumour percentages. A limitation of RNA‐based NGS approaches is that gene amplifications cannot be directly detected. However, gene expression levels, in general, are positively correlated with presence of gene amplifications. This potentially enables the use of RNA expression as an alternative of gene amplification since RNA expression is more relevant with gene function compared with DNA amplification [13]. To explore feasibility of detecting amplifications using RNA‐based NGS needs to be validated in relation to other amplification detection methods and the predictive value should be investigated in comparison to currently available tests.
FFPE tissues are sill the primary material used for diagnostic NGS in lung cancer, because FFPE preserves morphology and enables storage of the tumour sample at room temperature. So, FFPE tissue sections are the first choice for histological assessment, IHC and FISH analysis. However, formaldehyde fixation results in DNA and RNA degradation, base modifications (e.g. alteration from cytosine to uracil and thymine), abasic sites, etc [14]. Assessing good quality frozen tissues for NGS analysis is of no doubt the most optimal solution, yet not routinely available in most hospitals. In addition, the lack of good facilities
Chapter 6
and relatively high costs for storage of frozen tissue samples especially in peripheral hospitals hampers its general implementation. Strategies to improve test outcome on FFPE samples include tests compatible with poor quality DNA / RNA, such as our all‐in‐one transcriptome‐based assay. Additional improvements that can be implemented include molecular tagging of individual DNA/cDNA fragments to increase sensitivity of the assays, limiting the number of cycles used to amplify the sequencing libraries to minimize duplicate reads and maximize the recovery of unique fragments. All of these strategies should be considered in the future.
Besides challenges in the use of FFPE material, another challenge is the relatively long turn‐ around‐time for most NGS‐based tests that take at least one week. Although the costs for NGS have been reduced during the last decade, the total costs for current diagnostic demand are still quite high. This is partly caused by the availability of more and more targeted drugs and the growing need to increase the gene panel. In addition, molecular testing per patient also increases due to the increased knowledge on resistance mechanisms, which requires analysis of multiple re‐biopsies upon resistance to subsequent rounds of TKI treatments. Nevertheless, selection of patients whom are likely to benefit from targeted treatment is essential and will lead to decrease in costs by not treating patients that are unlikely to benefit from specific treatments.
2. Origin of Resistance to TKI Treatment
Mechanisms of acquired or secondary resistance include EGFR point mutations (e.g., T790M, C797S) [15], activation of alternating signalling pathways (e.g. ERBB2, MET, AXL, IGF‐1R, FGFR) [16], activation of downstream pathways (e.g., RAS/RAF/MEK, PI3K/AKT/mTOR, JAK/STAT) [17,18]. Advanced TKIs for patients with resistance associated mutations in EGFR include third generation TKIs, such as osimertinib for T790M [19]. Many others are under evaluation such as TAK788, poziotinib and TAS6417 for EGFR exon 20 resistant mutations, and capmatinib and tepotinib for MET exon 14 skipping. Combination treatments for ERBB2 resistance are introduced.
The mechanisms of ALK‐guided TKI resistance can be categorized as ALK dependent mechanisms caused by mutations in the translocated ALK fusion gene or by ALK amplifications and non‐ALK mechanisms involving activation of alternating signalling and downstream pathways [20]. ALK resistance mutations include 1151Tins, L1152R, C1156Y, F1174L, L1196M, L1198Fb, G1202R, S1206Y, G1269A [21,22]. For many of these so‐called gatekeeper mutations new TKIs have been developed such as lorlatinib. Crizotinib is no longer used as the preferred first‐line drug as alectinib performs better. Alternatively, inhibitors targeting downstream or bypass signalling pathways such as cabozantinib or BLUE‐ 667 targeting MET amplification [23], may be applied upon resistance.
The origin of drug resistance can be explained by selection of pre‐existing low‐frequency resistant clones that are below the detection limits of commonly used DNA/RNA tests [24]. Another explanation is genetic evolution induced by the drug leading to survival of tumour cells with aberrations leading to activation of alternative driver pathways. Both hypotheses were demonstrated in EGFR mutant NSCLC cell lines or in clinical cohorts. In one study, the PC9 lung cancer cell line harbouring an EGFR exon 19 deletion was analysed to unravel the resistance mechanism in vitro [25]. This cell line contained a pre‐existing T790M mutation in a low percentage of the cells. Early resistant clones were all T790M positive and were shown to be derived from the pre‐existing T790M positive cells. The late resistant cells harboured
SUMMARY, DISCUSSION AND FUTURE PERSPECTIVES
6
and relatively high costs for storage of frozen tissue samples especially in peripheral hospitals hampers its general implementation. Strategies to improve test outcome on FFPE samples include tests compatible with poor quality DNA / RNA, such as our all‐in‐one transcriptome‐based assay. Additional improvements that can be implemented include molecular tagging of individual DNA/cDNA fragments to increase sensitivity of the assays, limiting the number of cycles used to amplify the sequencing libraries to minimize duplicate reads and maximize the recovery of unique fragments. All of these strategies should be considered in the future.
Besides challenges in the use of FFPE material, another challenge is the relatively long turn‐ around‐time for most NGS‐based tests that take at least one week. Although the costs for NGS have been reduced during the last decade, the total costs for current diagnostic demand are still quite high. This is partly caused by the availability of more and more targeted drugs and the growing need to increase the gene panel. In addition, molecular testing per patient also increases due to the increased knowledge on resistance mechanisms, which requires analysis of multiple re‐biopsies upon resistance to subsequent rounds of TKI treatments. Nevertheless, selection of patients whom are likely to benefit from targeted treatment is essential and will lead to decrease in costs by not treating patients that are unlikely to benefit from specific treatments.
2. Origin of Resistance to TKI Treatment
Mechanisms of acquired or secondary resistance include EGFR point mutations (e.g., T790M, C797S) [15], activation of alternating signalling pathways (e.g. ERBB2, MET, AXL, IGF‐1R, FGFR) [16], activation of downstream pathways (e.g., RAS/RAF/MEK, PI3K/AKT/mTOR, JAK/STAT) [17,18]. Advanced TKIs for patients with resistance associated mutations in EGFR include third generation TKIs, such as osimertinib for T790M [19]. Many others are under evaluation such as TAK788, poziotinib and TAS6417 for EGFR exon 20 resistant mutations, and capmatinib and tepotinib for MET exon 14 skipping. Combination treatments for ERBB2 resistance are introduced.
The mechanisms of ALK‐guided TKI resistance can be categorized as ALK dependent mechanisms caused by mutations in the translocated ALK fusion gene or by ALK amplifications and non‐ALK mechanisms involving activation of alternating signalling and downstream pathways [20]. ALK resistance mutations include 1151Tins, L1152R, C1156Y, F1174L, L1196M, L1198Fb, G1202R, S1206Y, G1269A [21,22]. For many of these so‐called gatekeeper mutations new TKIs have been developed such as lorlatinib. Crizotinib is no longer used as the preferred first‐line drug as alectinib performs better. Alternatively, inhibitors targeting downstream or bypass signalling pathways such as cabozantinib or BLUE‐ 667 targeting MET amplification [23], may be applied upon resistance.
The origin of drug resistance can be explained by selection of pre‐existing low‐frequency resistant clones that are below the detection limits of commonly used DNA/RNA tests [24]. Another explanation is genetic evolution induced by the drug leading to survival of tumour cells with aberrations leading to activation of alternative driver pathways. Both hypotheses were demonstrated in EGFR mutant NSCLC cell lines or in clinical cohorts. In one study, the PC9 lung cancer cell line harbouring an EGFR exon 19 deletion was analysed to unravel the resistance mechanism in vitro [25]. This cell line contained a pre‐existing T790M mutation in a low percentage of the cells. Early resistant clones were all T790M positive and were shown to be derived from the pre‐existing T790M positive cells. The late resistant cells harboured
T790M mutations or other resistant mechanisms, such as mutations in NRAS, KRAS, BRAF and RET, and amplifications in KRAS [26]. The T790M mutations in these late resistance clones, were shown to be acquired during treatment. This study very‐well exemplified a proof of principle that resistance may originate from rare minor resistant clones or evolve from drug‐tolerant clones. In treatment‐naive NSCLC patients with EGFR sensitizing mutations, pre‐existing EGFR T790M detection rate varied from 1% to 80% in different studies (Table 2). The large differences in the T790M detection rate can be explained by patient selection, but may also partly be explained by differences in sensitivity between the test methods varying from sanger sequencing [27], PCR‐based methods [28‐31], ddPCR [28,32] to NGS [28]. Another factor that makes the high detection rate doubtful is the false positive rate due to the formalin fixation procedure. In one of the studies listed in table 2, no statistical significant difference was found in survival between pre‐treatment T790M positive and negative groups. However, when the patients with T790M were divided into MAF ≥0.5% (n=7) and MAF<0.5% (n=23) and negative subgroups (n=8), the group with T790M MAF ≥0.5% had significantly longer progression free survival compared with the 8 patients without T790M (p=0.0097) and the 23 patients with low MAF (p=0.0019) [30]. In another study, a significant difference was found in survival of T790M mutant and wild type group (9.7 months (95% CI, 6.9–12.9) versus 15.8 months (95% CI, 8.8‐NR) (p<0.0001) [31]. Discriminating between primary resistance or acquired resistance might be important for choice of therapy. Patients with presence of a pre‐existing resistant tumour cells may benefit from T790M targeting TKI even before clinical signs of resistance emerge. In the absence of a pre‐existing resistant clone, monitoring of the patient might allow early identification of a drug resistant clone.
In the future, as more targeted therapies will be applied in the clinic, the resistant patterns to other molecular sub‐types of NSCLC, such as BRAF positive or MET positive patients, need to be unravelled using paired patient samples or patient‐derived cell models. An accurate ultra‐sensitive NGS‐based method will be required to detect potential pre‐existing resistant tumour clones, but this approach is dependent on presence of such clones in the tissue biopsy. Collection of good quality and representative tissue samples for research purposes will increase the change to detect resistance associated variants. Multi‐regional sampling will be required to unveil pre‐existing resistant clones but this might not be feasible due to difficult accessibility of deep anatomic sites and critical severity of patient conditions. Moreover, tissue biopsies are essential to monitor drug resistance mechanisms and guide next‐step treatment strategy. In this respect, liquid biopsies may compensate traditional tissue biopsy to provide a more accurate image of the total spectrum of mutations in the heterogeneous tumour.
3. Liquid biopsies for lung cancer patients
There are a number of successful examples demonstrating detection of driver mutations for targeted therapies using cell free DNA (cfDNA) [33], circulating tumour cell (CTC) [34] and exosome‐derived DNA/RNA [35]. Recently, our group showed marked intra‐tumour heterogeneity in NSCLC, with up to 50% unique somatic mutations at different metastatic sites [36]. This indicates that primary and metastatic tumour sites may show different molecular features. Cell‐free DNA (cfDNA) has the potential advantage of being homogenous and might thus overcome the limitation of a single‐tissue biopsy. Yet, recent studies showed discrepancies between tissue biopsies and matched cfDNA samples, indicating that cfDNA cannot fully represent the heterogeneous tumour [37‐40]. This might be caused by the
Chapter 6
scarcity of tumour derived DNA in the cfDNA fraction. Despite these drawbacks, liquid biopsies are introduced in the clinical diagnostic setting in late stage NSCLC. Genetic analysis of liquid biopsies could be incorporated in a routine setting in case of difficult accessibility of tissue biopsies, requirement of repeated sampling for dynamic monitoring of disease progression, or negative results in tissue biopsies. Whether the parallel NGS tests on both tissue and liquid biopsies will bring more clinical benefits compared with either of the single test needs to be studied in the future. With respect to blood‐based biomarkers, other questions that remain to be answered are: what is the most optimal procedure to obtain liquid biopsies, what is the minimal amount of blood required for a reliable detection of minimal residual disease detection and which detection technique is most sensitive. For detection of fusion genes, both CTCs and exosome‐derived RNAs are two types of materials that have been used. One study showed presence of ALK‐breaks by FISH in CTCs isolated from 18 ALK‐positive patients by using a porous filter approach, with a sensitivity of 100% and concordant results for all cases [41]. In two similar studies, ALK break signals were confirmed in CTCs in all ALK positive patients (n=14 and 21 respectively), using two different microfluid chip based CTC enrichment assays [42,43]. In another study, ALK breaks were found in CTCs of all five ALK positive patients [44]. In late stage lung cancer, it is relatively easy to isolate CTCs for FISH and/or IHC analysis. Performance of molecular analysis on CTCs in a clinical setting using NGS‐based methods needs to be validated using larger patient cohorts in the future. ALK fusions were detected in exosome‐derived RNA by an NGS‐based assay in 9 out of 13 ALK positive and in 26 out of 53 ALK positive NSCLC patients, respectively [45,46]. Thus, exosome‐derived RNA assays to detect ALK fusion genes seem to have a lower sensitivity as compared to the above‐mentioned studies using CTCs. Yet it still holds the advantage that the ALK fusion and ALK resistant mutations can be detected simultaneously. However, the exosome‐based approach is not yet used in the clinic.
In most cases, conventional tissue biopsies remain the first choice for molecular diagnostic testing for lung cancer. However, using tissue biopsy for longitudinal monitoring of treatment response is not always feasible. Benefits of mutation detection using liquid biopsy material has been proven in lung patients under TKI treatment. An affordable as well as reliable method for detection of fusion genes and mutations in RNA isolated from exosomes or the isolation of CTCs followed by FISH should be explored. The most optimal approach in a clinical setting needs to be validated in larger cohorts.
SUMMARY, DISCUSSION AND FUTURE PERSPECTIVES
6
scarcity of tumour derived DNA in the cfDNA fraction. Despite these drawbacks, liquid biopsies are introduced in the clinical diagnostic setting in late stage NSCLC. Genetic analysis of liquid biopsies could be incorporated in a routine setting in case of difficult accessibility of tissue biopsies, requirement of repeated sampling for dynamic monitoring of disease progression, or negative results in tissue biopsies. Whether the parallel NGS tests on both tissue and liquid biopsies will bring more clinical benefits compared with either of the single test needs to be studied in the future. With respect to blood‐based biomarkers, other questions that remain to be answered are: what is the most optimal procedure to obtain liquid biopsies, what is the minimal amount of blood required for a reliable detection of minimal residual disease detection and which detection technique is most sensitive. For detection of fusion genes, both CTCs and exosome‐derived RNAs are two types of materials that have been used. One study showed presence of ALK‐breaks by FISH in CTCs isolated from 18 ALK‐positive patients by using a porous filter approach, with a sensitivity of 100% and concordant results for all cases [41]. In two similar studies, ALK break signals were confirmed in CTCs in all ALK positive patients (n=14 and 21 respectively), using two different microfluid chip based CTC enrichment assays [42,43]. In another study, ALK breaks were found in CTCs of all five ALK positive patients [44]. In late stage lung cancer, it is relatively easy to isolate CTCs for FISH and/or IHC analysis. Performance of molecular analysis on CTCs in a clinical setting using NGS‐based methods needs to be validated using larger patient cohorts in the future. ALK fusions were detected in exosome‐derived RNA by an NGS‐based assay in 9 out of 13 ALK positive and in 26 out of 53 ALK positive NSCLC patients, respectively [45,46]. Thus, exosome‐derived RNA assays to detect ALK fusion genes seem to have a lower sensitivity as compared to the above‐mentioned studies using CTCs. Yet it still holds the advantage that the ALK fusion and ALK resistant mutations can be detected simultaneously. However, the exosome‐based approach is not yet used in the clinic.
In most cases, conventional tissue biopsies remain the first choice for molecular diagnostic testing for lung cancer. However, using tissue biopsy for longitudinal monitoring of treatment response is not always feasible. Benefits of mutation detection using liquid biopsy material has been proven in lung patients under TKI treatment. An affordable as well as reliable method for detection of fusion genes and mutations in RNA isolated from exosomes or the isolation of CTCs followed by FISH should be explored. The most optimal approach in a clinical setting needs to be validated in larger cohorts.
Table 2. Literature review on pre‐existing EGFR T790M mutation in patients with EGFR sensitizing mutations. Method Material Detection of T790M Percentage of
positive patients Ref
Direct DNA Sanger
sequencing frozen or FFPE, no details 1/34 3% [27]
Allele‐specific PCR FFPE 3/394 1% [28] A combination of AS‐PCR, ddPCR and NGS (sensitivity, ∼0.1%) FFPE 2/68 3% [28] MALDI‐TOF MS (0.4‐2.2%) NA 23/73 32% [29]
Colony Hybridization Assay
(sensitivity, 0.01%) Frozen 30/38 79% [30] PNA‐PCR (sensitivity, NA) NA 34/50 68% [31] ddPCR (sensitivity, 0.01– 0.005%) FFPE 299/373 80% [32]
Chapter 6
References
1. Kimura, M.; Endo, H.; Inoue, T.; Nishino, K.; Uchida, J.; Kumagai, T.; Kukita, Y.; Kato, K.; Imamura, F.;
Inoue, M. Analysis of erbb ligand‐induced resistance mechanism to crizotinib by primary culture of lung
adenocarcinoma with eml4‐alk fusion gene. Journal of Thoracic Oncology 2015, 10, 527‐530.
2. Azzato, E.M.; Deshpande, C.; Aikawa, V.; Aggarwal, C.; Alley, E.; Jacobs, B.; Morrissette, J.; Daber, R.
Rare complex mutational profile in an alk inhibitor‐resistant non‐small cell lung cancer. Anticancer research
2015, 35, 3007‐3012.
3. Crystal, A.S.; Shaw, A.T.; Sequist, L.V.; Friboulet, L.; Niederst, M.J.; Lockerman, E.L.; Frias, R.L.; Gainor,
J.F.; Amzallag, A.; Greninger, P. Patient‐derived models of acquired resistance can identify effective drug
combinations for cancer. Science 2014, 346, 1480‐1486.
4. Heitzer, E.; Ulz, P.; Geigl, J.B. Circulating tumour DNA as a liquid biopsy for cancer. Clinical Chemistry
2015, 61, 112‐123.
5. Bernabé, R.; Hickson, N.; Wallace, A.; Blackhall, F.H. What do we need to make circulating tumour DNA
(ctdna) a routine diagnostic test in lung cancer? European journal of cancer 2017, 81, 66‐73.
6. Zehir, A.; Benayed, R.; Shah, R.H.; Syed, A.; Middha, S.; Kim, H.R.; Srinivasan, P.; Gao, J.; Chakravarty,
D.; Devlin, S.M., et al. Mutational landscape of metastatic cancer revealed from prospective clinical sequencing
of 10,000 patients. Nature medicine 2017, 23, 703.
7. Conroy, J.M.; Pabla, S.; Glenn, S.T.; Burgher, B.; Nesline, M.; Papanicolau‐Sengos, A.; Andreas, J.;
Giamo, V.; Lenzo, F.L.; Hyland, F.C.L., et al. Analytical validation of a next‐generation sequencing assay to
monitor immune responses in solid tumour. The Journal of Molecular Diagnostics 2018, 20, 95‐109.
8. Shan, L.; Wang, Z.; Guo, L.; Sun, H.; Qiu, T.; Ling, Y.; Li, W.; Li, L.; Liu, X.; Zheng, B., et al. Concurrence of
egfr amplification and sensitizing mutations indicate a better survival benefit from egfr‐tki therapy in lung
adenocarcinoma patients. Lung cancer 2015, 89, 337‐342.
9. Li, X.; Cai, W.; Yang, G.; Su, C.; Ren, S.; Zhao, C.; Hu, R.; Chen, X.; Gao, G.; Guo, Z., et al. Comprehensive
analysis of egfr‐mutant abundance and its effect on efficacy of egfr tkis in advanced nsclc with egfr mutations.
Journal of Thoracic Oncology 2017, 12, 1388‐1397.
10. Rosenbaum, J.N.; Bloom, R.; Forys, J.T.; Hiken, J.; Armstrong, J.R.; Branson, J.; McNulty, S.; Velu, P.D.;
Pepin, K.; Abel, H., et al. Genomic heterogeneity of alk fusion breakpoints in non‐small‐cell lung cancer. Modern
Pathology 2018, 31, 791.
11. Li, Y.; Zhang, T.; Zhang, J.; Li, W.; Yuan, P.; Xing, P.; Zhang, Z.; Chuai, S.; Li, J.; Ying, J. Response to
crizotinib in advanced alk‐rearranged non‐small cell lung cancers with different alk‐fusion variants. Lung cancer
2018, 118, 128‐133.
12. van der Wekken, A.J.; Pelgrim, R.; 't Hart, N.; Werner, N.; Mastik, M.F.; Hendriks, L.; van der Heijden,
E.H.F.M.; Looijen‐Salamon, M.; de Langen, A.J.; Staal‐van den Brekel, J., et al. Dichotomous alk‐ihc is a better predictor for alk inhibition outcome than traditional alk‐fish in advanced non–small cell lung cancer. Clinical
Cancer Research 2017, 23, 4251‐4258.
13. Jia, Y.; Chen, L.; Jia, Q.; Dou, X.; Xu, N.; Liao, D.J. The well‐accepted notion that gene amplification
contributes to increased expression still remains, after all these years, a reasonable but unproven assumption. J
Carcinog 2016, 15, 3‐3.
14. Do, H.; Dobrovic, A. Sequence artifacts in DNA from formalin‐fixed tissues: Causes and strategies for
minimization. Clinical Chemistry 2015, 61, 64‐71.
15. Yasuda, H.; Park, E.; Yun, C.‐H.; Sng, N.J.; Lucena‐Araujo, A.R.; Yeo, W.‐L.; Huberman, M.S.; Cohen,
D.W.; Nakayama, S.; Ishioka, K., et al. Structural, biochemical, and clinical characterization of epidermal growth
factor receptor (egfr) exon 20 insertion mutations in lung cancer. Science translational medicine 2013, 5,
216ra177‐216ra177.
16. Sequist, L.V.; Waltman, B.A.; Dias‐Santagata, D.; Digumarthy, S.; Turke, A.B.; Fidias, P.; Bergethon, K.;
Shaw, A.T.; Gettinger, S.; Cosper, A.K., et al. Genotypic and histological evolution of lung cancers acquiring
resistance to egfr inhibitors. Science translational medicine 2011, 3, 75ra26‐75ra26.
17. Gao, S.P.; Chang, Q.; Mao, N.; Daly, L.A.; Vogel, R.; Chan, T.; Liu, S.H.; Bournazou, E.; Schori, E.; Zhang,
H., et al. Jak2 inhibition sensitizes resistant egfr‐mutant lung adenocarcinoma to tyrosine kinase inhibitors.
Science Signaling 2016, 9, ra33‐ra33.
18. Ercan, D.; Xu, C.; Yanagita, M.; Monast, C.S.; Pratilas, C.A.; Montero, J.; Butaney, M.; Shimamura, T.;
Sholl, L.; Ivanova, E.V., et al. Reactivation of erk signaling causes resistance to egfr kinase inhibitors. Cancer
SUMMARY, DISCUSSION AND FUTURE PERSPECTIVES
6
References
1. Kimura, M.; Endo, H.; Inoue, T.; Nishino, K.; Uchida, J.; Kumagai, T.; Kukita, Y.; Kato, K.; Imamura, F.;
Inoue, M. Analysis of erbb ligand‐induced resistance mechanism to crizotinib by primary culture of lung
adenocarcinoma with eml4‐alk fusion gene. Journal of Thoracic Oncology 2015, 10, 527‐530.
2. Azzato, E.M.; Deshpande, C.; Aikawa, V.; Aggarwal, C.; Alley, E.; Jacobs, B.; Morrissette, J.; Daber, R.
Rare complex mutational profile in an alk inhibitor‐resistant non‐small cell lung cancer. Anticancer research
2015, 35, 3007‐3012.
3. Crystal, A.S.; Shaw, A.T.; Sequist, L.V.; Friboulet, L.; Niederst, M.J.; Lockerman, E.L.; Frias, R.L.; Gainor,
J.F.; Amzallag, A.; Greninger, P. Patient‐derived models of acquired resistance can identify effective drug
combinations for cancer. Science 2014, 346, 1480‐1486.
4. Heitzer, E.; Ulz, P.; Geigl, J.B. Circulating tumour DNA as a liquid biopsy for cancer. Clinical Chemistry
2015, 61, 112‐123.
5. Bernabé, R.; Hickson, N.; Wallace, A.; Blackhall, F.H. What do we need to make circulating tumour DNA
(ctdna) a routine diagnostic test in lung cancer? European journal of cancer 2017, 81, 66‐73.
6. Zehir, A.; Benayed, R.; Shah, R.H.; Syed, A.; Middha, S.; Kim, H.R.; Srinivasan, P.; Gao, J.; Chakravarty,
D.; Devlin, S.M., et al. Mutational landscape of metastatic cancer revealed from prospective clinical sequencing
of 10,000 patients. Nature medicine 2017, 23, 703.
7. Conroy, J.M.; Pabla, S.; Glenn, S.T.; Burgher, B.; Nesline, M.; Papanicolau‐Sengos, A.; Andreas, J.;
Giamo, V.; Lenzo, F.L.; Hyland, F.C.L., et al. Analytical validation of a next‐generation sequencing assay to
monitor immune responses in solid tumour. The Journal of Molecular Diagnostics 2018, 20, 95‐109.
8. Shan, L.; Wang, Z.; Guo, L.; Sun, H.; Qiu, T.; Ling, Y.; Li, W.; Li, L.; Liu, X.; Zheng, B., et al. Concurrence of
egfr amplification and sensitizing mutations indicate a better survival benefit from egfr‐tki therapy in lung
adenocarcinoma patients. Lung cancer 2015, 89, 337‐342.
9. Li, X.; Cai, W.; Yang, G.; Su, C.; Ren, S.; Zhao, C.; Hu, R.; Chen, X.; Gao, G.; Guo, Z., et al. Comprehensive
analysis of egfr‐mutant abundance and its effect on efficacy of egfr tkis in advanced nsclc with egfr mutations.
Journal of Thoracic Oncology 2017, 12, 1388‐1397.
10. Rosenbaum, J.N.; Bloom, R.; Forys, J.T.; Hiken, J.; Armstrong, J.R.; Branson, J.; McNulty, S.; Velu, P.D.;
Pepin, K.; Abel, H., et al. Genomic heterogeneity of alk fusion breakpoints in non‐small‐cell lung cancer. Modern
Pathology 2018, 31, 791.
11. Li, Y.; Zhang, T.; Zhang, J.; Li, W.; Yuan, P.; Xing, P.; Zhang, Z.; Chuai, S.; Li, J.; Ying, J. Response to
crizotinib in advanced alk‐rearranged non‐small cell lung cancers with different alk‐fusion variants. Lung cancer
2018, 118, 128‐133.
12. van der Wekken, A.J.; Pelgrim, R.; 't Hart, N.; Werner, N.; Mastik, M.F.; Hendriks, L.; van der Heijden,
E.H.F.M.; Looijen‐Salamon, M.; de Langen, A.J.; Staal‐van den Brekel, J., et al. Dichotomous alk‐ihc is a better predictor for alk inhibition outcome than traditional alk‐fish in advanced non–small cell lung cancer. Clinical
Cancer Research 2017, 23, 4251‐4258.
13. Jia, Y.; Chen, L.; Jia, Q.; Dou, X.; Xu, N.; Liao, D.J. The well‐accepted notion that gene amplification
contributes to increased expression still remains, after all these years, a reasonable but unproven assumption. J
Carcinog 2016, 15, 3‐3.
14. Do, H.; Dobrovic, A. Sequence artifacts in DNA from formalin‐fixed tissues: Causes and strategies for
minimization. Clinical Chemistry 2015, 61, 64‐71.
15. Yasuda, H.; Park, E.; Yun, C.‐H.; Sng, N.J.; Lucena‐Araujo, A.R.; Yeo, W.‐L.; Huberman, M.S.; Cohen,
D.W.; Nakayama, S.; Ishioka, K., et al. Structural, biochemical, and clinical characterization of epidermal growth
factor receptor (egfr) exon 20 insertion mutations in lung cancer. Science translational medicine 2013, 5,
216ra177‐216ra177.
16. Sequist, L.V.; Waltman, B.A.; Dias‐Santagata, D.; Digumarthy, S.; Turke, A.B.; Fidias, P.; Bergethon, K.;
Shaw, A.T.; Gettinger, S.; Cosper, A.K., et al. Genotypic and histological evolution of lung cancers acquiring
resistance to egfr inhibitors. Science translational medicine 2011, 3, 75ra26‐75ra26.
17. Gao, S.P.; Chang, Q.; Mao, N.; Daly, L.A.; Vogel, R.; Chan, T.; Liu, S.H.; Bournazou, E.; Schori, E.; Zhang,
H., et al. Jak2 inhibition sensitizes resistant egfr‐mutant lung adenocarcinoma to tyrosine kinase inhibitors.
Science Signaling 2016, 9, ra33‐ra33.
18. Ercan, D.; Xu, C.; Yanagita, M.; Monast, C.S.; Pratilas, C.A.; Montero, J.; Butaney, M.; Shimamura, T.;
Sholl, L.; Ivanova, E.V., et al. Reactivation of erk signaling causes resistance to egfr kinase inhibitors. Cancer
discovery 2012, 2, 934‐947.
19. Jänne, P.A.; Yang, J.C.‐H.; Kim, D.‐W.; Planchard, D.; Ohe, Y.; Ramalingam, S.S.; Ahn, M.‐J.; Kim, S.‐W.;
Su, W.‐C.; Horn, L., et al. Azd9291 in egfr inhibitor–resistant non–small‐cell lung cancer. New England Journal of
Medicine 2015, 372, 1689‐1699.
20. Asao, T.; Takahashi, F.; Takahashi, K. Resistance to molecularly targeted therapy in non‐small‐cell lung
cancer. Respiratory Investigation 2019, 57, 20‐26.
21. Choi, Y.L.; Soda, M.; Yamashita, Y.; Ueno, T.; Takashima, J.; Nakajima, T.; Yatabe, Y.; Takeuchi, K.;
Hamada, T.; Haruta, H., et al. Eml4‐alk mutations in lung cancer that confer resistance to alk inhibitors. New
England Journal of Medicine 2010, 363, 1734‐1739.
22. Gainor, J.F.; Dardaei, L.; Yoda, S.; Friboulet, L.; Leshchiner, I.; Katayama, R.; Dagogo‐Jack, I.; Gadgeel,
S.; Schultz, K.; Singh, M., et al. Molecular mechanisms of resistance to first‐ and second‐generation alk
inhibitors in <em>alk</em>‐rearranged lung cancer. Cancer discovery 2016, 6, 1118‐1133.
23. Reckamp, K.L.; Frankel, P.H.; Mack, P.C.; Gitlitz, B.J.; Ruel, N.; Lara, P.; Li, T.; Koczywas, M.; Gadgeel,
S.M.; Cristea, M.C., et al. Phase ii trial of xl184 (cabozantinib) plus erlotinib in patients (pts) with advanced egfr‐ mutant non‐small cell lung cancer (nsclc) with progressive disease (pd) on epidermal growth factor receptor (egfr) tyrosine kinase inhibitor (tki) therapy: A california cancer consortium phase ii trial (nci 9303). Journal of
Clinical Oncology 2014, 32, 8014‐8014.
24. Oxnard, G.R. The cellular origins of drug resistance in cancer. Nature medicine 2016, 22, 232.
25. Hata, A.N.; Niederst, M.J.; Archibald, H.L.; Gomez‐Caraballo, M.; Siddiqui, F.M.; Mulvey, H.E.; Maruvka,
Y.E.; Ji, F.; Bhang, H.‐e.C.; Krishnamurthy Radhakrishna, V., et al. Tumour cells can follow distinct evolutionary
paths to become resistant to epidermal growth factor receptor inhibition. Nature medicine 2016, 22, 262.
26. Bhang, H.E.; Ruddy, D.A.; Krishnamurthy Radhakrishna, V.; Caushi, J.X.; Zhao, R.; Hims, M.M.; Singh,
A.P.; Kao, I.; Rakiec, D.; Shaw, P., et al. Studying clonal dynamics in response to cancer therapy using high‐
complexity barcoding. Nature medicine 2015, 21, 440‐448.
27. Sequist, L.V.; Martins, R.G.; Spigel, D.; Grunberg, S.M.; Spira, A.; Jänne, P.A.; Joshi, V.A.; McCollum, D.;
Evans, T.L.; Muzikansky, A., et al. First‐line gefitinib in patients with advanced non–small‐cell lung cancer
harboring somatic egfr mutations. Journal of Clinical Oncology 2008, 26, 2442‐2449.
28. Lavdovskaia, E.D.; Iyevleva, A.G.; Sokolenko, A.P.; Mitiushkina, N.V.; Preobrazhenskaya, E.V.; Tiurin,
V.I.; Ivantsov, A.O.; Bizin, I.V.; Stelmakh, L.V.; Moiseyenko, F.V., et al. Egfr t790m mutation in tki‐naïve clinical samples: Frequency, tissue mosaicism, predictive value and awareness on artifacts. Oncology research and
treatment 2018, 41, 634‐642.
29. Su, K.‐Y.; Chen, H.‐Y.; Li, K.‐C.; Kuo, M.‐L.; Yang, J.C.‐H.; Chan, W.‐K.; Ho, B.‐C.; Chang, G.‐C.; Shih, J.‐Y.;
Yu, S.‐L., et al. Pretreatment epidermal growth factor receptor (egfr) t790m mutation predicts shorter egfr tyrosine kinase inhibitor response duration in patients with non–small‐cell lung cancer. Journal of Clinical
Oncology 2012, 30, 433‐440.
30. Fujita, Y.; Suda, K.; Kimura, H.; Matsumoto, K.; Arao, T.; Nagai, T.; Saijo, N.; Yatabe, Y.; Mitsudomi, T.;
Nishio, K. Highly sensitive detection of egfr t790m mutation using colony hybridization predicts favorable
prognosis of patients with lung cancer harboring activating egfr mutation. Journal of Thoracic Oncology 2012, 7,
1640‐1644.
31. Costa, C.; Molina, M.A.; Drozdowskyj, A.; Giménez‐Capitán, A.; Bertran‐Alamillo, J.; Karachaliou, N.;
Gervais, R.; Massuti, B.; Wei, J.; Moran, T., et al. The impact of <em>egfr</em> t790m mutations and <em>bim</em> mrna expression on outcome in patients with <em>egfr</em>‐mutant nsclc treated with
erlotinib or chemotherapy in the randomized phase iii eurtac trial. Clinical Cancer Research 2014, 20, 2001‐
2010.
32. Watanabe, M.; Kawaguchi, T.; Isa, S.; Ando, M.; Tamiya, A.; Kubo, A.; Saka, H.; Takeo, S.; Adachi, H.;
Tagawa, T., et al. Ultra‐sensitive detection of the pretreatment egfr t790m mutation in non‐small cell lung cancer patients with an egfr‐activating mutation using droplet digital pcr. Clinical cancer research : an official
journal of the American Association for Cancer Research 2015, 21, 3552‐3560.
33. Pisapia, P.; Malapelle, U.; Troncone, G. Liquid biopsy and lung cancer. Acta Cytol 2018.
34. Kapeleris, J.; Kulasinghe, A.; Warkiani, M.E.; Vela, I.; Kenny, L.; O'Byrne, K.; Punyadeera, C. The
prognostic role of circulating tumour cells (ctcs) in lung cancer. Frontiers in oncology 2018, 8, 311‐311.
35. Masaoutis, C.; Mihailidou, C.; Tsourouflis, G.; Theocharis, S. Exosomes in lung cancer diagnosis and
treatment. From the translating research into future clinical practice. Biochimie 2018, 151, 27‐36.
36. Saber, A.; Timens, W.; van den Berg, A.; Groen, H.J.M.; Hiltermann, T.J.N.; de Lange, K.; Kok, K.;
Terpstra, M.M. Mutation patterns in small cell and non‐small cell lung cancer patients suggest a different level
Chapter 6
37. Wu, Z.; Yang, Z.; Li, C.S.; Zhao, W.; Liang, Z.X.; Dai, Y.; Zhu, Q.; Miao, K.L.; Cui, D.H.; Chen, L.A.
Differences in the genomic profiles of cell‐free DNA between plasma, sputum, urine, and tumour tissue in
advanced nsclc. Cancer Medicine 2019, 8, 910‐919.
38. Zhang, Y.; Chang, L.; Yang, Y.; Fang, W.; Guan, Y.; Wu, A.; Hong, S.; Zhou, H.; Chen, G.; Chen, X., et al.
Intratumour heterogeneity comparison among different subtypes of non‐small‐cell lung cancer through multi‐
region tissue and matched ctdna sequencing. Molecular cancer 2019, 18, 7.
39. Lam, V.K.; Tran, H.T.; Banks, K.C.; Lanman, R.B.; Rinsurongkawong, W.; Peled, N.; Lewis, J.; Lee, J.J.;
Roth, J.; Roarty, E.B., et al. Targeted tissue and cell‐free tumour DNA sequencing of advanced lung squamous‐
cell carcinoma reveals clinically significant prevalence of actionable alterations. Clinical Lung Cancer 2019, 20,
30‐36.e33.
40. Guo, Q.; Wang, J.; Xiao, J.; Wang, L.; Hu, X.; Yu, W.; Song, G.; Lou, J.; Chen, J. Heterogeneous mutation
pattern in tumour tissue and circulating tumour DNA warrants parallel ngs panel testing. Molecular cancer
2018, 17, 131.
41. Pailler, E.; Adam, J.; Barthelemy, A.; Oulhen, M.; Auger, N.; Valent, A.; Borget, I.; Planchard, D.; Taylor,
M.; Andre, F., et al. Detection of circulating tumour cells harboring a unique alk rearrangement in alk‐positive non‐small‐cell lung cancer. Journal of clinical oncology : official journal of the American Society of Clinical
Oncology 2013, 31, 2273‐2281.
42. Tan, C.L.; Lim, T.H.; Lim, T.K.; Tan, D.S.‐W.; Chua, Y.W.; Ang, M.K.; Pang, B.; Lim, C.T.; Takano, A.; Lim,
A.S.‐T., et al. Concordance of anaplastic lymphoma kinase (alk) gene rearrangements between circulating
tumour cells and tumour in non‐small cell lung cancer. Oncotarget 2016, 7, 23251‐23262.
43. He, W.; Xu, D.; Wang, Z.; Xiang, X.; Tang, B.; Li, S.; Hou, M.; Zhang, Y.; Chen, J.‐F.; Lin, M., et al.
Detecting alk-rearrangement of ctc enriched by nanovelcro chip in advanced nsclc patients. 2016.
44. Ilie, M.; Hofman, V.; Hofman, P.; Long, E.; Marquette, C.H.; Mouroux, J.; Coelle, C.; Mauro, V.; Butori,
C.; Zahaf, K., et al. Alk‐gene rearrangement: A comparative analysis on circulating tumour cells and tumour
tissue from patients with lung adenocarcinoma. Annals of Oncology 2012, 23, 2907‐2913.
45. Rolfo, C.; Laes, J.F.; Reclusa, P.; Valentino, A.; Lienard, M.; Gil‐Bazo, I.; Malapelle, U.; Sirera, R.; Rocco,
D.; Van Meerbeeck, J., et al. P2.01‐093 exo‐alk proof of concept: Exosomal analysis of alk alterations in
advanced nsclc patients: Topic: Targets for treatment prediction. Journal of Thoracic Oncology 2017, 12, S844‐
S845.
46. Rolfo, C.D.; Laes, J.‐F.; Reclusa, P.; Lienard, M.; Gil‐Bazo, I.; Malapelle, U.; Rocco, D.; Meerbeeck, J.V.;
Pauwels, P.; Peeters, M., et al. Liquid biopsy analysis of alk alterations from advanced nsclc patients: The value
