Potential unreliability of ALK variant allele frequency in the efficacy prediction of targeted therapy in NSCLC

Abstract

Anaplastic lymphoma kinase (ALK) is the most common fusion gene involved in non-small cell lung cancer (NSCLC), and remarkable response has been achieved with the use of ALK tyrosine kinase inhibitors (ALK-TKIs). However, the clinical efficacy is highly variable. Pre-existing intratumoral heterogeneity (ITH) has been proven to contribute to the poor treatment response and the resistance to targeted therapies. In this work, we investigated whether the variant allele frequencies (VAFs) of ALK fusions can help assess ITH and predict targeted therapy efficacy. Through the application of next-generation sequencing (NGS), 7.2% (326/4548) of patients were detected to be ALK positive. On the basis of the adjusted VAF (adjVAF, VAF normalization for tumor purity) of four different threshold values (adjVAF < 50%, 40%, 30%, or 20%), the association of ALK subclonality with crizotinib efficacy was assessed. Nonetheless, no statistical association was observed between median progression-free survival (PFS) and ALK subclonality assessed by adjVAF, and a poor correlation of adjVAF with PFS was found among the 85 patients who received first-line crizotinib. Results suggest that the ALK VAF determined by hybrid capture-based NGS is probably unreliable for ITH assessment and targeted therapy efficacy prediction in NSCLC.

Keywords

ALK fusion / next-generation sequencing / fluorescence in situ hybridization / immunohistochemistry / variant allele frequency / intratumoral heterogeneity / targeted therapy

1 Introduction

As the most common fusion gene involved in non-small cell lung cancer (NSCLC), anaplastic lymphoma kinase (ALK) fusions can be found in up to 5% of NSCLCs [1,2]. Several ALK tyrosine kinase inhibitors (ALK-TKIs), including crizotinib, alectinib, brigatinib, ceritinib, and lorlatinib, have been used routinely in clinical practice for the treatment of advanced ALK-positive NSCLC patients and have gained remarkable clinical efficacy [3–7]. Nonetheless, the duration of clinical response is highly variable among patients, and progressive disease (PD) seems inevitable in most patients owing to intrinsic or acquired resistances [8]. Our previous studies have found that several factors, such as EML4-ALK fusion variants and concurrent TP53 mutations in ALK-rearranged tumors, may contribute to the heterogeneous response to ALK-TKIs [9,10]. In addition, pre-existing intratumoral heterogeneity (ITH) has been proven to contribute to the poor treatment response and the resistance to targeted therapies [11], which thus acts as a considerable challenge that should be realized and highlighted in the implementation of precision oncology.

Currently available methods that allow for comprehensive analyses of ITH in tumor tissues include multiregion sequencing and single-cell sequencing [12]. However, these methods are still not practical or technically feasible for all patients, particularly for metastatic NSCLC patients with limited tissue samples. Over the past years, next-generation sequencing (NGS) has been widely used in molecular testing, given that it could determine the alteration status (mutation, fusion, and amplification) of diverse genetic events simultaneously [13]. NGS can also provide the information of variant allele frequency (VAF) for mutations and fusions with genomic DNA, which is calculated through the number of mutant reads divided by the total number of reads at certain genomic position. Biologically, the values of VAF are mainly affected by tumor purity, ITH, and copy number variation in tumor tissues [14]. Thus, a few studies have reported that the adjusted VAF (adjVAF, VAF adjusted for tumor purity) can quantitatively estimate colorectal cancer cells carrying KRAS, NRAS, BRAF, or PIK3CA mutations and lung cancer cells carrying EGFR mutations to a certain extent, suggesting a predictive value for the clinical efficacy of matched targeted therapies [15,16]. Nevertheless, the effect of ALK VAF on reflecting the genomic complexity and predicting the clinical response to ALK-TKIs remains largely unexplored.

In this study, we systematically characterized ALK fusions through NGS, fluorescence in situ hybridization (FISH), or/and immunohistochemistry (IHC) in a large population of Chinese patients with NSCLC. We further investigated the correlation of ALK VAF determined by hybrid capture-based NGS with ALK FISH break-apart ratio and IHC-positively stained cell ratio in the same cases and explored the association of adjVAF with the clinical efficacy of ALK-TKIs. In so doing, we tried to uncover whether ALK VAF has important potential implications for ITH assessment and targeted therapy outcomes.

2 Materials and methods

2.1 Patients and tumor samples

Between March 2015 and May 2021, tumor tissues from newly diagnosed, treatment-naive patients with NSCLC were subjected to our ISO15189-accredited laboratory for molecular testing to assess ALK status (4548 cases by NGS, 4670 cases by FISH, and 12 176 cases by IHC). ALK-positive cases were collected for further analysis. The study was approved by the Institutional Review Board of the Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College. The methods were performed in accordance with approved guidelines, and written informed consent was obtained from all participants.

2.2 DNA extraction

The formalin-fixed and paraffin-embedded (FFPE) tumor tissues with a minimum of 20% tumor purity were sectioned, and genomic DNA was extracted using QIAamp DNA FFPE Tissue Kit, as we previously reported [14]. For the five cases with heterogeneous ALK-IHC staining, tumor tissues were microdissected, and ALK-positive area and ALK-negative area were respectively collected for DNA extraction. DNA quantities were determined using Qubit 3.0 Fluorometer (Thermo Fisher Scientific, Carlsbad, CA, USA).

2.3 NGS

Targeted-capture DNA-based NGS was performed using a 56-gene panel (panel A) to identify ALK fusions and other actionable alterations in 4548 NSCLC patients, as previously reported [17]. Briefly, 100–200 ng of sample DNA was fragmented, followed by ligation to adaptors for library construction. DNA templates were amplified through solid-phase bridge amplification, and the clonal clusters were sequenced on the Illumina NextSeq 550 platform (San Diego, USA). Eighteen ALK-positive cases identified by panel A and the Illumina platform were further tested using another NGS platform. In brief, 100–200 ng of DNA was fragmented, ligated to adaptors, and hybridized to a panel of 283 genes (panel B). DNA nanoballs were generated through rolling circle amplification, deposited on a flow cell, and sequenced on the BGISEQ-500 platform (BGI, Shenzhen, China). ALK fusions were identified when the sequence depth ≥ 1000× and VAF ≥ 1%.

2.4 Adjustment of VAF

Tumor purity was assessed through the corresponding hematoxylin-eosin (HE) slide and corrected through an additional HE slide obtained after sectioning tumor tissues for DNA extraction. As previously reported [18], tumor purity was estimated by two independent pathologists, and final tumor purity was calculated using the average percentage. A third pathologist intervened when more than 10% difference in tumor purity was determined between the estimations of the two pathologists, and final tumor purity was calculated using the average percentage of tumor purity assessed by the three pathologists. To infer the subclonality of ALK fusions, ALK VAF was adjusted for tumor purity (VAF/tumor purity × 100%) to obtain adjVAF. Assuming that ALK fusion only affected one allele, the expected adjVAF should be close to 50% if the ALK event was clonal. Moreover, the presence of ALK fusion only in a subset of tumor cells in a single neoplasm, defined as subclonality, was estimated through four different threshold values of adjVAF (adjVAF < 50%, 40%, 30%, or 20%).

2.5 FISH

FISH was conducted using Vysis LSI ALK Dual Color, Break Apart Rearrangement Probe (Abbott/Vysis, Abbott Park, IL, USA) to detect ALK rearrangements, as previously described [19]. ALK positive was identified when more than 15% of tumor cells with splitting of one or both 5′ and 3′ probe signals or isolated 3′ probe signals were observed. Similar to the tumor purity evaluation, the ALK break-apart ratio was evaluated independently by two experienced pathologists, and a third pathologist was required when more than 10% difference was observed between the evaluations of the two pathologists.

2.6 IHC

As previously described [20], IHC was performed using a fully automated IHC Ventana Benchmark XT stainer (Ventana Medical Systems, Inc., AZ, USA) with Ventana anti-ALK (D5F3) rabbit monoclonal primary antibody. ALK positive was identified when strong granular cytoplasmic staining was observed in any percentage of positive tumor cells. Similar to the tumor purity evaluation, the ALK IHC slides were independently evaluated by two or three expert pathologists to accurately assess the IHC-positively stained cell ratio.

2.7 Evaluation of ALK-TKI efficacy

On the basis of the computed tomography imaging or magnetic resonance imaging every 6–8 weeks, the clinical responses of ALK-TKI (crizotinib) were evaluated following the guidelines of the Response Evaluation Criteria in Solid Tumors version 1.1 [21]. The objective response rate (ORR) was defined as the percentage of patients who achieved a complete response (CR) or partial response (PR), while the disease control rate (DCR) was defined as the percentage of patients who achieved a CR, PR, or stable disease. Progression-free survival (PFS) was determined from the date of targeted therapy to the date of PD, and patients without PD on May 1, 2021 were considered censored.

2.8 Statistical analysis

Chi-square test or Fisher’s exact test was used to determine the association of clinicopathological and molecular features with ALK subclonality. The Kaplan–Meier method was employed to analyze the PFS. Univariate and multivariate analyses were performed by Cox proportional risk regression to define the prognostic factors for PFS. The data were analyzed using the software of SPSS 25.0 (IBM, Chicago, IL), and a two-sided P value < 0.05 was considered statistically significant.

3 Results

3.1 Detection of ALK fusions in NSCLCs

NGS, FISH, and IHC were performed in 4548, 4670, and 12 176 NSCLC cases, respectively, and ALK positive was identified in 326 of the 4548 (7.2%), 359 of the 4670 (7.7%), and 787 of the 12176 (6.5%) cases (Fig.1). During detection, 897 cases were identified ALK positive only by one method (145 cases only by NGS, 201 cases only by FISH, and 551 cases only by IHC). Moreover, 293 cases were concurrently examined using two different platforms (NGS&FISH, IHC&FISH, or NGS&IHC), but discordant results were observed in 16 cases (5.5%) (Table S1). These 16 cases were further tested using a third method to validate ALK status (three different platforms: NGS&FISH&IHC) (Fig.1). These cases were excluded from our further analysis, although five cases were ALK-NGS positive. For the remaining 321 ALK-positive cases identified by NGS, various VAFs (median 19.6%, IQR 11.3%–32.0%) were detected (Fig. S1).

3.2 Association analyses between the clinical outcomes of targeted therapy and ALK subclonality

We further investigated the ALK subclonality of the 321 cases estimated on the basis of adjVAF, which was calculated from VAF normalization for tumor purity in each sample (Fig.2). The ALK adjVAF values were highly variable across patients, with a median value of 49.4% (95% confidence interval [CI]: 43.2%–54.2%, Fig.2). We next assessed the impact of ALK subclonality on the clinical efficacy of ALK-TKI therapy. Among the 85 patients who received first-line crizotinib, the ORR was 70.6% (95% CI: 60.2%–79.2%), the DCR was 91.8% (95% CI: 84.0%–96.0%, Table S2), and the overall median PFS was 12.0 months (95% CI: 11.7–15.4 months, Fig. S2). As shown in Fig.2, we found no association between adjVAF and PFS in first-line crizotinib therapy (Spearman r = −0.094, P = 0.415). Moreover, ALK subclonality was estimated on the basis of four different threshold values of adjVAF (adjVAF < 50%, 40%, 30%, or 20%), and survival analysis was performed in the TKI-treated patients according to these different threshold values. However, no statistical association was observed between median PFS and ALK subclonality (Tab.1). Further univariate and multivariate analyses showed that TP53 mutation was the only factor significantly influencing PFS in patients receiving crizotinib (Table S3).

3.3 Analyses of the correlation of VAF with FISH break-apart ratio and IHC-positively stained cell ratio

We then analyzed the correlation of VAF with the break-apart ratio of FISH and the positive cell ratio of IHC staining in the same sample. Totally, 50 cases were ALK positive, as detected by NGS and FISH. The correlation analysis showed that there was a considerable discrepancy between VAF and break-apart ratio (Spearman r = 0.228, P = 0.112, Fig.3). NGS and IHC determined 126 cases of ALK positive, but no correlation was observed between VAF and IHC-positively stained cell ratio (Spearman r = 0.190, P = 0.083, Fig.3). Eighteen ALK-positive cases identified by NGS were further tested with another NGS platform by using panel B. Interestingly, significant differences in VAF were found between the two NGS assays (panel A vs. panel B, P = 0.020, Fig.3), indicating the unreliability of the VAF determined by NGS.

3.4 ITH revealed by ALK IHC

The median FISH break-apart ratio was 62% (95% CI: 60%–64%, Fig. S3A), and the median IHC-positively stained cell ratio was 100% (Fig. S3B). ITH determined by FISH break-apart ratio may also be inaccurate due to technological factors [22,23]. Thus, IHC-positively stained cell ratio was evaluated to estimate ITH in ALK-positive cases. Only 2.3% (18/778) of the IHC-positive cases showed heterogeneous staining. Of the 18 ALK IHC heterogeneous-expressed cases, 12 had sufficient tissue for DNA sequencing. Five cases whose tumors showed a clear boundary between ALK-positive (a) and ALK-negative (b) areas were microdissected for NGS. As shown in Fig.4 and Table S4, EML4-ALK fusion was detected in area (a) of the five cases. However, no ALK fusion was found in area (b). Instead, ROS1 fusion was identified in one sample, EGFR p.K823Q mutation in another sample, and KRAS p.G12C in the third sample. No actionable alterations were identified in the remaining two cases. Meanwhile, seven cases showed no clear boundary between ALK-positive and ALK-negative regions that were difficult to be separated. The seven cases were further confirmed by NGS. One was found to concurrently harbor EML4-ALK fusion and EGFR p.L858R mutation (Fig. S4A), and the other one concurrently harbored EML4-ALK fusion and FGFR2-TAOK3 fusion (Fig. S4B).

4 Discussion

NGS has been rapidly incorporated into clinical molecular testing in recent years, with gene panels that rely on either an amplicon-based or hybrid capture-based approach. Through the application of hybrid capture-based NGS, multiple ALK fusions with variable fusion partners and VAFs are identified. Our previous studies have proven that the EML4-ALK variants identified by hybrid capture-based NGS using genomic DNA are associated with clinical responses to crizotinib [9], but uncommon ALK fusion variants are unreliable predictors of ALK-TKIs efficacy [17]. In the present study, through the analyses of ALK-positive cases with NGS, FISH, and IHC, we found a poor correlation of VAF with FISH break-apart ratio and IHC-positively stained cell ratio in the same sample. We also determined an inconsiderable association between adjVAF and the efficacy of ALK-TKI.

Several molecular testing modalities, including NGS, FISH, and IHC, have been used in ALK fusion detection of NSCLC [24]. However, each method has its own advantages and disadvantages in the identification of ALK rearrangements. For example, although FISH is a well-established detection method, false-negative results may occur when the separated fluorescence signals are not apart enough, while atypical signals may acquire false-positive interpretation [19,25]. Ventana-D5F3 IHC is a rapid, economical, and simple method for ALK fusion detection. Nonetheless, false-positive signals are sometimes observed in non-adenocarcinoma NSCLC samples with heterogeneous staining, and false-negative results can appear in mucinous adenocarcinoma owing to the influence of extracellular mucin [26]. Hybrid capture-based NGS is suitable for simultaneously identifying multiple driver alterations with genomic DNA, but false-positive or false-negative results may occur due to complex breakpoints at the DNA level or limited tumor cellularity [17,27,28]. In this study, although roughly similar frequencies of ALK-positive cases were detected using NGS, FISH, and IHC in Chinese NSCLC population, 16 of 293 (5.5%) cases showed discordant results when different platforms were used. This suggests the necessity of optimizing the ALK testing strategy with the integrated use of multimodality methods in routine molecular testing.

The VAFs of driver gene mutations, such as EGFR, KRAS, and BRAF, have been reported to play potential roles in the estimation of clonal proportion and the prediction of clinical outcomes with targeted agents [14,29,30]. Here, we analyzed ALK subclonality on the basis of the values of adjVAF. Nonetheless, we found that adjVAF did not correlate with PFS in ALK-TKI therapy, suggesting the limited role of VAF in direct targeted therapy. The discordant results between our work and previous studies are probably due to the differences in the type of alterations detected and molecular detection methods: (1) Previous findings were discussed with a focus on the VAFs of missense and indel mutations, whereas ours mainly focused on gene fusions. (2) PCR-based methods, such as ddPCR [31], BEAMing [32], and amplicon-based NGS [15,33,34], were used to estimate the abundance of mutations in almost all of the previous studies. These methods highly rely on the primers specific for each target. PCR bias, such as allele dropout and off-target primer binding, is the major negative factor that may technically affect the accuracy of VAF. However, hotspot mutations are usually optimized in most available kits to ensure amplification of the mutant and wild-type alleles according to the scaling [35]. (3) The hybrid capture-based NGS analyzes gene mutations and fusions simultaneously through a gene-specific enrichment step with probes complementary to the regions of interest. The probes are designed on the basis of a reference sequence; thus, this method may suffer from off-target capture, especially when variations, including deletion/insertion and rearrangement, are going to be identified [36]. Using this approach, some studies have reported that the VAF of EGFR p.T790M (missense mutation) can predict osimertinib treatment efficacy [37,38], but the VAF of EGFR sensitive mutations (mainly EGFR exon 19 deletion) has minimal potential to predict EGFR-TKI efficacy [39], possibly owing to the technological limitations of the hybrid capture-based method. Similarly, off-target capture may likely occur in rearranged ALK alleles, while DNA fragment, PCR bias, sequence depth, library complexity, and analysis pipelines may have an influence on the accuracy of ALK VAF [40]. Consequently, significant differences in VAF were observed when two different NGS platforms were used on the same samples, strongly suggesting the limitation of hybrid capture-based NGS in capturing and sequencing rearranged ALK alleles. Moreover, there was no statistical association between median PFS and ALK subclonality assessed by adjVAF, and a poor correlation of adjVAF with PFS was found among patients who received crizotinib. These findings further indicate that the ALK VAF determined by hybrid capture-based NGS is mainly related to the technique, not to biology. In line with our results, Bustamante et al. found that NSCLC patients can benefit from targeted treatments based only on the identification of actionable alterations by hybrid capture-based cfDNA NGS, but VAF had no impact on the response [41].

Despite the lack of standards, the ITH estimated by VAF was incredibly high in our study. ALK fusion is a driver alteration that occurs and plays a significant role in tumorigenesis and tumor progression, which is usually present in almost all tumor cells [42]. Thus, ALK subclonality should be a rare event in NSCLC. Actually, the heterogeneity of ALK fusion was observed only in 2.3% of IHC-positive cases. Moreover, we found a considerable discrepancy among the median values of ALK VAF determined by NGS, break-apart ratio determined by FISH, and positively stained cell ratio determined by IHC. Within the same sample, an apparent lack of correlation between VAF and FISH break-apart ratio/IHC-positively stained cell ratio was observed. These results support the conclusion that the cellular heterogeneity of ALK fusions determined by VAF is not reliable.

We previously analyzed an NSCLC case with heterogeneous ALK-IHC staining and identified EML-ALK fusion in the ALK IHC-positive area and MET exon 14 skipping mutation in the IHC-negative area, showing definite intratumor genetic heterogeneity [43]. Thus, we further evaluated ITH on the basis of the IHC results here. Cases whose tumors were harboring heterogeneous ALK-IHC staining were analyzed by NGS, and coexistent driver alterations involving EGFR (n = 2), ROS1 (n = 1), KRAS (n = 1), and FGFR2 (n = 1) were found in these ALK-positive cases. Although rare, these cases demonstrated that coalteration of ALK and other actionable genes can be detected, especially in lung adenocarcinoma with heterogeneous ALK-IHC expression.

The current study has several limitations that should be acknowledged. First, FISH and hybrid capture-based NGS both detect gene rearrangements at the DNA level; performing orthogonal assays capable of validation at the RNA level in ALK IHC-negative but FISH and NGS-positive cases (P11 and P12, Table S1) is better. Second, the tumor purity can be subjective, which may influence the values of adjVAF, although the tumor cell content was assessed strictly by three independent pathologists to obtain a more accurate estimation. Third, mutant allele specific imbalance, including copy number gain or loss of heterozygosity due to the loss of the wild-type allele, may also affect the values of adjVAF, although it is a rare occurrence in ALK locus. In addition, several second-/third-generation ALK-TKIs have been clinically used. Comparison of ALK adjVAF with the clinical outcome of ALK-TKI therapy mainly based on crizotinib may be limited.

Overall, our study demonstrated the potential unreliability of ALK VAF in ITH assessment and ALK-TKI efficacy prediction in NSCLC, possibly owing to the technological limitations of hybrid capture-based NGS. ITH of ALK fusion is a rare event revealed by IHC, but some concurrent actionable alterations could be identified in cases with heterogeneous ALK-IHC expression, suggesting that IHC may be a direct and reliable method for ITH assessment. Additional orthogonal assays, such as NGS, should be performed for ALK IHC heterogeneous-expressed samples to validate ALK fusions and to analyze other concurrent gene alterations.

References

[1] Soda M, Choi YL, Enomoto M, Takada S, Yamashita Y, Ishikawa S, Fujiwara S, Watanabe H, Kurashina K, Hatanaka H, Bando M, Ohno S, Ishikawa Y, Aburatani H, Niki T, Sohara Y, Sugiyama Y, Mano H. Identification of the transforming EML4-ALK fusion gene in non-small-cell lung cancer. Nature 2007; 448(7153): 561–566

[2] Tian HX, Zhang XC, Yang JJ, Guo WB, Chen ZH, Wang Z, Wu YL. Clinical characteristics and sequence complexity of anaplastic lymphoma kinase gene fusions in Chinese lung cancer patients. Lung Cancer 2017; 114: 90–95

[3] Shaw AT, Kim DW, Nakagawa K, Seto T, Crinó L, Ahn MJ, De Pas T, Besse B, Solomon BJ, Blackhall F, Wu YL, Thomas M, O’Byrne KJ, Moro-Sibilot D, Camidge DR, Mok T, Hirsh V, Riely GJ, Iyer S, Tassell V, Polli A, Wilner KD, Jänne PA. Crizotinib versus chemotherapy in advanced ALK-positive lung cancer. N Engl J Med 2013; 368(25): 2385–2394

[4] Camidge DR, Dziadziuszko R, Peters S, Mok T, Noe J, Nowicka M, Gadgeel SM, Cheema P, Pavlakis N, de Marinis F, Cho BC, Zhang L, Moro-Sibilot D, Liu T, Bordogna W, Balas B, Müller B, Shaw AT. Updated efficacy and safety data and impact of the EML4-ALK fusion variant on the efficacy of alectinib in untreated ALK-positive advanced non-small cell lung cancer in the global phase III ALEX study. J Thorac Oncol 2019; 14(7): 1233–1243

[5] Shaw AT, Kim TM, Crinò L, Gridelli C, Kiura K, Liu G, Novello S, Bearz A, Gautschi O, Mok T, Nishio M, Scagliotti G, Spigel DR, Deudon S, Zheng C, Pantano S, Urban P, Massacesi C, Viraswami-Appanna K, Felip E. Ceritinib versus chemotherapy in patients with ALK-rearranged non-small-cell lung cancer previously given chemotherapy and crizotinib (ASCEND-5): a randomised, controlled, open-label, phase 3 trial. Lancet Oncol 2017; 18(7): 874–886

[6] Camidge DR, Kim HR, Ahn MJ, Yang JCH, Han JY, Hochmair MJ, Lee KH, Delmonte A, García Campelo MR, Kim DW, Griesinger F, Felip E, Califano R, Spira A, Gettinger SN, Tiseo M, Lin HM, Gupta N, Hanley MJ, Ni Q, Zhang P, Popat S. Brigatinib versus crizotinib in advanced ALK inhibitor-naive ALK-positive non-small cell lung cancer: second interim analysis of the phase III ALTA-1L trial. J Clin Oncol 2020; 38(31): 3592–3603

[7] Solomon BJ, Besse B, Bauer TM, Felip E, Soo RA, Camidge DR, Chiari R, Bearz A, Lin CC, Gadgeel SM, Riely GJ, Tan EH, Seto T, James LP, Clancy JS, Abbattista A, Martini JF, Chen J, Peltz G, Thurm H, Ou SI, Shaw AT. Lorlatinib in patients with ALK-positive non-small-cell lung cancer: results from a global phase 2 study. Lancet Oncol 2018; 19(12): 1654–1667

[8] Golding B, Luu A, Jones R, Viloria-Petit AM. The function and therapeutic targeting of anaplastic lymphoma kinase (ALK) in non-small cell lung cancer (NSCLC). Mol Cancer 2018; 17(1): 52

[9] 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

[10] Song P, Zhang F, Li Y, Yang G, Li W, Ying J, Gao S. Concomitant TP53 mutations with response to crizotinib treatment in patients with ALK-rearranged non-small-cell lung cancer. Cancer Med 2019; 8(4): 1551–1557

[11] Dagogo-Jack I, Shaw AT. Tumour heterogeneity and resistance to cancer therapies. Nat Rev Clin Oncol 2018; 15(2): 81–94

[12] Lim ZF, Ma PC. Emerging insights of tumor heterogeneity and drug resistance mechanisms in lung cancer targeted therapy. J Hematol Oncol 2019; 12(1): 134

[13] Goodwin S, McPherson JD, McCombie WR. Coming of age: ten years of next-generation sequencing technologies. Nat Rev Genet 2016; 17(6): 333–351

[14] Li W, Qiu T, Ling Y, Gao S, Ying J. Subjecting appropriate lung adenocarcinoma samples to next-generation sequencing-based molecular testing: challenges and possible solutions. Mol Oncol 2018; 12(5): 677–689

[15] Normanno N, Rachiglio AM, Lambiase M, Martinelli E, Fenizia F, Esposito C, Roma C, Troiani T, Rizzi D, Tatangelo F, Botti G, Maiello E, Colucci G, Ciardiello F; CAPRI-GOIM investigators. Heterogeneity of KRAS, NRAS, BRAF and PIK3CA mutations in metastatic colorectal cancer and potential effects on therapy in the CAPRI GOIM trial. Ann Oncol 2015; 26(8): 1710–1714

[16] Gieszer B, Megyesfalvi Z, Dulai V, Papay J, Kovalszky I, Timar J, Fillinger J, Harko T, Pipek O, Teglasi V, Regos E, Papp G, Szallasi Z, Laszlo V, Renyi-Vamos F, Galffy G, Bodor C, Dome B, Moldvay J. EGFR variant allele frequency predicts EGFR-TKI efficacy in lung adenocarcinoma: a multicenter study. Transl Lung Cancer Res 2021; 10(2): 662–674

[17] Li W, Guo L, Liu Y, Dong L, Yang L, Chen L, Liu K, Shao Y, Ying J. Potential unreliability of uncommon ALK, ROS1, and RET genomic breakpoints in predicting the efficacy of targeted therapy in NSCLC. J Thorac Oncol 2021; 16(3): 404–418

[18] Li W, Qiu T, Guo L, Ying J. Major challenges related to tumor biological characteristics in accurate mutation detection of colorectal cancer by next-generation sequencing. Cancer Lett 2017; 410: 92–99

[19] Li W, Zhang J, Guo L, Chuai S, Shan L, Ying J. Combinational analysis of FISH and immunohistochemistry reveals rare genomic events in ALK fusion patterns in NSCLC that responds to crizotinib treatment. J Thorac Oncol 2017; 12(1): 94–101

[20] Ying J, Guo L, Qiu T, Shan L, Ling Y, Liu X, Lu N. Diagnostic value of a novel fully automated immunochemistry assay for detection of ALK rearrangement in primary lung adenocarcinoma. Ann Oncol 2013; 24(10): 2589–2593

[21] Eisenhauer EA, Therasse P, Bogaerts J, Schwartz LH, Sargent D, Ford R, Dancey J, Arbuck S, Gwyther S, Mooney M, Rubinstein L, Shankar L, Dodd L, Kaplan R, Lacombe D, Verweij J. New response evaluation criteria in solid tumours: revised RECIST guideline (version 1.1). Eur J Cancer 2009; 45(2): 228–247

[22] Camidge DR, Theodoro M, Maxson DA, Skokan M, O’Brien T, Lu X, Doebele RC, Barón AE, Varella-Garcia M. Correlations between the percentage of tumor cells showing an anaplastic lymphoma kinase (ALK) gene rearrangement, ALK signal copy number, and response to crizotinib therapy in ALK fluorescence in situ hybridization-positive nonsmall cell lung cancer. Cancer 2012; 118(18): 4486–4494

[23] Camidge DR, Kono SA, Flacco A, Tan AC, Doebele RC, Zhou Q, Crino L, Franklin WA, Varella-Garcia M. Optimizing the detection of lung cancer patients harboring anaplastic lymphoma kinase (ALK) gene rearrangements potentially suitable for ALK inhibitor treatment. Clin Cancer Res 2010; 16(22): 5581–5590

[24] Li W, Zhang J, Wang Z, Li L, Ma J, Zhou X, Wang J, Liang Z, Ying J. Guidelines for clinical practice of ALK fusion detection in non-small-cell lung cancer: a proposal from the Chinese RATICAL study group. Journal of the National Cancer Center 2021; 1(4): 123–131

[25] Ilie MI, Bence C, Hofman V, Long-Mira E, Butori C, Bouhlel L, Lalvée S, Mouroux J, Poudenx M, Otto J, Marquette CH, Hofman P. Discrepancies between FISH and immunohistochemistry for assessment of the ALK status are associated with ALK ‘borderline’-positive rearrangements or a high copy number: a potential major issue for anti-ALK therapeutic strategies. Ann Oncol 2015; 26(1): 238–244

[26] Wang H, Sun L, Sang Y, Yang X, Tian G, Wang Z, Fang J, Sun W, Zhou L, Jia L, Tsao MS, Shi H, Lin D. A study of ALK-positive pulmonary squamous-cell carcinoma: from diagnostic methodologies to clinical efficacy. Lung Cancer 2019; 130: 135–142

[27] Benayed R, Offin M, Mullaney K, Sukhadia P, Rios K, Desmeules P, Ptashkin R, Won H, Chang J, Halpenny D, Schram AM, Rudin CM, Hyman DM, Arcila ME, Berger MF, Zehir A, Kris MG, Drilon A, Ladanyi M. High yield of RNA sequencing for targetable kinase fusions in lung adenocarcinomas with no mitogenic driver alteration detected by DNA sequencing and low tumor mutation burden. Clin Cancer Res 2019; 25(15): 4712–4722

[28] Li W, Wan R, Guo L, Chang G, Jiang D, Meng L, Ying J. Reliability analysis of exonic-breakpoint fusions identified by DNA sequencing for predicting the efficacy of targeted therapy in non-small cell lung cancer. BMC Med 2022; 20(1): 160

[29] Raja R, Kuziora M, Brohawn PZ, Higgs BW, Gupta A, Dennis PA, Ranade K. Early reduction in ctDNA predicts survival in patients with lung and bladder cancer treated with durvalumab. Clin Cancer Res 2018; 24(24): 6212–6222

[30] Li X, Cai W, Yang G, Su C, Ren S, Zhao C, Hu R, Chen X, Gao G, Guo Z, Li W, Zhou C, Hirsch FR. Comprehensive analysis of EGFR-mutant abundance and its effect on efficacy of EGFR TKIs in advanced NSCLC with EGFR mutations. J Thorac Oncol 2017; 12(9): 1388–1397

[31] Beau-Faller M, Pencreach E, Leduc C, Blons H, Merlio JP, Bringuier PP, de Fraipont F, Escande F, Lemoine A, Ouafik L, Denis M, Hofman P, Lacave R, Melaabi S, Langlais A, Missy P, Morin F, Moro-Sibilot D, Barlesi F, Cadranel J; French Cooperative Thoracic Intergroup (IFCT). Independent prognostic value of ultra-sensitive quantification of tumor pre-treatment T790M subclones in EGFR mutated non-small cell lung cancer (NSCLC) treated by first/second generation TKI, depends on variant allele frequency (VAF): results of the French cooperative thoracic intergroup (IFCT) biomarkers France project. Lung Cancer 2020; 140: 19–26

[32] Elez E, Chianese C, Sanz-García E, Martinelli E, Noguerido A, Mancuso FM, Caratù G, Matito J, Grasselli J, Cardone C, Esposito Abate R, Martini G, Santos C, Macarulla T, Argilés G, Capdevila J, Garcia A, Mulet N, Maiello E, Normanno N, Jones F, Tabernero J, Ciardello F, Salazar R, Vivancos A. Impact of circulating tumor DNA mutant allele fraction on prognosis in RAS-mutant metastatic colorectal cancer. Mol Oncol 2019; 13(9): 1827–1835

[33] Friedlaender A, Tsantoulis P, Chevallier M, De Vito C, Addeo A. The impact of variant allele frequency in EGFR mutated NSCLC patients on targeted therapy. Front Oncol 2021; 11: 644472

[34] O’Kane GM, Liu G, Stockley TL, Shabir M, Zhang T, Law JH, Le LW, Sacher A, Shepherd FA, Bradbury PA, Leighl NB. The presence and variant allele fraction of EGFR mutations in ctDNA and development of resistance. Lung Cancer 2019; 131: 86–89

[35] Head SR, Komori HK, LaMere SA, Whisenant T, Van Nieuwerburgh F, Salomon DR, Ordoukhanian P. Library construction for next-generation sequencing: overviews and challenges. Biotechniques 2014; 56(2): 61–64, 66, 68 passim doi:10.2144/000114133

[36] Bruno R, Fontanini G. Next generation sequencing for gene fusion analysis in lung cancer: a literature review. Diagnostics (Basel) 2020; 10(8): 521

[37] Vaclova T, Grazini U, Ward L, O’Neill D, Markovets A, Huang X, Chmielecki J, Hartmaier R, Thress KS, Smith PD, Barrett JC, Downward J, de Bruin EC. Clinical impact of subclonal EGFR T790M mutations in advanced-stage EGFR-mutant non-small-cell lung cancers. Nat Commun 2021; 12(1): 1780

[38] Zheng Q, Hong S, Huang Y, Zhao H, Yang Y, Hou X, Zhao Y, Ma Y, Zhou T, Zhang Y, Fang W, Zhang L. EGFR T790M relative mutation purity predicts osimertinib treatment efficacy in non-small cell lung cancer patients. Clin Transl Med 2020; 9(1): 17

[39] Ai X, Cui J, Zhang J, Chen R, Lin W, Xie C, Liu A, Zhang J, Yang W, Hu X, Hu X, Zhao Q, Rao C, Zang YS, Ning R, Li P, Chang L, Yi X, Lu S. Clonal architecture of EGFR mutation predicts the efficacy of EGFR-tyrosine kinase inhibitors in advanced NSCLC: a prospective multicenter study (NCT03059641). Clin Cancer Res 2021; 27(3): 704–712

[40] Bos MK, Nasserinejad K, Jansen MPHM, Angus L, Atmodimedjo PN, de Jonge E, Dinjens WNM, van Schaik RHN, Del Re M, Dubbink HJ, Sleijfer S, Martens JWM. Comparison of variant allele frequency and number of mutant molecules as units of measurement for circulating tumor DNA. Mol Oncol 2021; 15(1): 57–66

[41] Bustamante Alvarez JG, Janse S, Owen DH, Kiourtsis S, Bertino EM, He K, Carbone DP, Otterson GA. Treatment of non-small-cell lung cancer based on circulating cell-free DNA and impact of variation allele frequency. Clin Lung Cancer 2021; 22(4): e519–e527

[42] Hallberg B, Palmer RH. The role of the ALK receptor in cancer biology. Ann Oncol 2016; 27(Suppl 3): iii4–iii15

[43] Qiu T, Zhang F, Li W, Guo L, Ying J. Concurrent presence of ALK rearrangement and MET mutation in lung adenocarcinoma. J Thorac Oncol 2019; 14(2): e42–e44

RIGHTS & PERMISSIONS

Higher Education Press